Clinical Nephrotoxins: Renal Injury from Drugs and Chemicals

Clinical Nephrotoxins Renal Injury from Drugs and Chemicals Third Edition

Clinical Nephrotoxins Renal Injury from Drugs and Chemicals Third Edition Editors Marc E. DE BROE University of Antwerp, Belgium George A. PORTER Department of Medicine, The Oregon Health Sciences University Portland, Oregon, USA Associate Editors William M. BENNETT Northwest Renal Clinic, Legacy Good Samaritan Hospital Portland, Oregon, USA Gilbert DERAY Groupe Hôspitalier Pitié Salpétrière, Service De Néphrologie Paris, France

Editors Marc E. De Broe University of Antwerp Belgium

George A. Porter Oregon Health Sciences University Portland, Oregon, USA

William M. Bennett Oregon Health Sciences University Portland, Oregon, USA

Gilbert Deray Hopital Pitie-Salpetriere Paris, France

ISBN-13: 978-0-387-84842-6 DOI: 10.1007/978-0-387-84843-3

e-ISBN-13: 978-0-387-84843-3

Library of Congress Control Number: 2008931970  2008 Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Editorial support: Dirk De Weerdt: text processing, correspondence, lay-out, figure design, index, cover design, cover illustration. Contact: [email protected] Printed on acid-free paper springer.com

Clinical Nephrotoxins Renal Injury from Drugs and Chemicals Third Edition

TABLE OF CONTENTS

PREFACE

VIII

LIST OF CONTRIBUTORS

X

GENERAL

1

A1

Clinical relevance

3

George A. Porter

2

Drug-associated acute kidney injury in the intensive care unit

29

Mitchell H. Rosner and Mark D. Okusa

3

Renal handling of drugs and xenobiotics

43

Marc E. De Broe and Françoise Roch-Ramel

4

Pharmacological aspects of nephrotoxicity

73

Marisa D. Covington and Rick G. Schnellmann

5

Pharmacovigilance: from signal to action

85

Hubert G. Leufkens and Antoine C. Egberts

6

Urinary biomarkers and nephrotoxicity

91

William F. Finn and George A. Porter

7

Toxin-induced immunological renal disease

131

Lucette Pelletier, Abdelhadi Saoudi and Gilbert Fournié

8

Cellular mechanisms of nephrotoxicity

155

Istvan Arany, Gur P. Kaushal, Didier Portilla, Judit Megyesi, Peter M. Price and Robert L. Safirstein

9

Animal models for the assessment of acute renal dysfunction and injury

173

Zoltan H. Endre and Charles L. Edelstein

10

Renal cell culture models: contribution to the understanding of nephrotoxic mechanisms

223

Paul Jennings, Christian Koppelstätter, Judith Lechner and Walter Pfaller

11

Use of dialytic therapies for poisoning James F. Winchester, Nikolas Harbord and Donald A. Feinfeld

251

Table of Contents

B 12

SPECIFIC DRUGS Aminoglycosides and vancomycin

265 267

Brian S. Decker and Bruce A. Molitoris

13

Beta-lactam antibiotics

293

Constantin Cojocel

14

Amphotericin B

323

Nathalie K. Zgheib, Blair Capitano and Robert A. Branch

15

Sulfonamides, sulfadiazine, trimethoprim-sulfamethoxazole, pentamidine, pyrimethamine, dapsone, quinolones

353

Gabriel N. Contreras, Cristiane Mocelin Carvalho, Jorge M. Diego, Decio Carvalho and Isabel Espinal

16

Antiviral agents

383

Jeffrey S. Berns, Alden Doyle and Nishaminy Kasbekar

17

Analgesics and 5-aminosalicylic acid

399

Monique M. Elseviers and Marc E. De Broe

18

Non-steroidal anti-inflammatory drugs

419

Ali J. Olyaei, Andrew Whelton, Til Sturmer and George A. Porter

19

Gold salts, D-penicillamine and allopurinol

459

Shiro Ueda and George A. Porter

20

Angiotensin I converting enzyme inhibitors and angiotensin II receptor antagonists

481

Paul E. De Jong

21

Diuretics and alcohol ingestion

495

Sheldon C. Chaffer and Jules B. Puschett

22

Anticancer drugs

511

Corinne Isnard-Bagnis, Vincent Launay-Vacher, Svetlana Karie and Gilbert Deray

23

Anesthetic agents

537

Per-Olof Jarnberg

24

Bisphosphonates and the kidney

547

Jonathan Green

25

Proton pump inhibitors: acute interstitial nephritis and other renal effects

567

Ursula C. Brewster and Mark A. Perazella

26

Oral sodium phosphate bowel purgatives and acute phosphate nephropathy

579

Glen S. Markowitz

27

Illicit drug abuse and renal disease

595

Cheryl L. Kunis, Nidhi Aggarwal and Gerald B. Appel

28

Nephrotoxicity of calcineurin and mTOR inhibitors

617

Emmanuel A. Burdmann and William M. Bennett

29

Immunomodulators: interleukins, interferons, and IV immunoglobulin Joris J. Roelofs, Daniel Abramowicz and Sandrine Florquin

VI

683

Table of Contents

30

Imaging agents

699

Christiane M. Erley, Ihab M. Wahba and George A. Porter

31

Lithium-associated kidney effects

725

Daniel Batlle, Edgar V. Lerma, Parveen Naaz and Santosh Hakkapakki

32

Oxalate

749

Anja Verhulst and Marc E. De Broe

33

Herbal remedies containing aristolochic acid and mushroom nephrotoxicity

579

Frédéric Debelle, Marie-Carmen Muniz-Martinez, Jean-Louis Vanherweghem and Joëlle Nortier

C

ENVIRONMENTAL AND OCCUPATIONAL NEPHROTOXINS

34

Lead nephropathy

771 773

Richard P. Wedeen

35

Cadmium-induced renal effects

785

Gunnar F. Nordberg, Teruhiko Kido and Harry A. Roels

36

Mercury-induced renal effects

811

Bruce A. Fowler, Margaret H. Whittaker and Carl-Gustaf Elinder

37

Organic solvents, silicon-containing compounds and pesticides

827

Patrick C. D’Haese, Monique M. Elseviers, Muhammed Yaqoob and Marc E. De Broe

38

Balkan nephropathy

843

Ljubica Djukanovic and Zoran Radovanovic

39

Nephrotoxins in Africa

859

Charles Swanepoel, Marc Blockman and Joe Talmud

40

Paraphenylene diamine hair dye poisoning

871

Mohamed I. Hamdouk, Mohamed B. Abdelraheem, Ahbab A. Taha, Mohamed Benghanem and Marc E. De Broe

D 41

THE RENAL FAILURE PATIENT Trace metal disturbances in end-stage renal failure patients

881 883

Patrick C. D’Haese

42

Smoking and the kidney

895

Eberhard Ritz

43

Star fruit

901

Miguel Moysés Neto, Ruither O. Carolino, Norberto P. Lopes and Norberto Garcia-Cairasco

44

Drug dosage in renal failure

913

Ali J. Olyaei and William M. Bennett

LIST OF ABBREVIATIONS INDEX

945 951

VII

Preface

A

s with our two previous editions we remained true to our concept of a multi-nationally author book. Our belief remains strong that scientific information is an international commodity whose interpretation and application are significantly influenced by both the cultural and ethnic background of the observer. The opportunity to share in the rich diversity of the international scientific community continues as a fundamental goal of this endeavor. The sharing of intellectual resources fostered by this effort continues to facilitate the advancement of sound science. As the profession develops new and improved methods for treating disease, there has occurred a parallel increase in the recognition of adverse drug reactions. Also, as more of the world industrializes the occurrence of unexpected injury to organisms because of exposure to environmental/industrial toxins gains prominence. Nephrotoxicity is truly a worldwide problem and we recognize this with the addition of several new chapters. As with the two prior editions, drugs/substances were selected for inclusion based on both the frequency of use and current knowledge, thus new additions include: bisphosphonates, proton pump inhibitors, phosphate containing laxatives, oxalate, smoking and the use of star fruit. Similar criteria were used for including environmental/industrial exposure with the addition of trace metals in chronic kidney disease patients. We have also included chapters dedicated to specific circumstances, drugs associated with acute kidney injury in the intensive care unit, plus the use of dialytic therapies for poisoning. The nature of scientific inquiry has remained unchanged through all editions. As stated previously, one approach is the application of Koch’s postulates, aided and abetted by various experimental animal models. Another involves population based epidemiologic associations to identify potentially causal relationships. Each has its advocates and disciples, and each provides valuable information that can be used by the clinician

in better managing his/her patient. However, each technique yields data that must be interpreted with an understanding of the drawbacks and pitfalls inherent in each approach. By enlisting multiple authors for each chapter, plus rigorous editing we hope the final product is a balanced, rationale statement of the field, as it exists today. The statement remains a guiding principle for developing the content of this third edition. As with previous editions we strive to provide a text which is useful, not only to the clinician, but of equal interest to the investigator. The addition of nine new chapters is in response to topics of current interest and we are looking forward to suggestions by the reader ([email protected]). We continue to stress the contribution of cell biology and pathophysiology, believing they provide both a better understanding of toxic injury when known, and a rational direction for therapy and prevention. Since the last edition the application of known risk factors as a means of stratifying acute kidney injury patient outcomes has made a significant contribution to management. With the validation of risk factor stratification the use of preventative techniques is becoming a reality. We continue to include risk factors as a prominent feature with the expectation of a reduction in the incidence of nephrotoxic injury. On a more personal note we confess that without the diligent and tireless polyvalent contribution of Dirk De Weerdt, there would have been no preface for there would have been no book. We also applaud the timely contributions of our authors and their willingness to negotiate compromise when asked. Finally, to our wives Myriam and Marthel, two individuals whose gift of time made this labor possible, we are forever in your debt.

Marc E. DE BROE George A. PORTER Summer 2008

Clinical Nephrotoxins Renal Injury from Drugs and Chemicals Third Edition

Editors Marc E. DE BROE Department of Nephrology and Hypertension, University of Antwerp, Belgium George A. PORTER Department of Medicine, The Oregon Health Sciences University, Portland, Oregon, USA Associate Editors William M. BENNETT Northwest Renal Clinic, Legacy Good Samaritan Hospital, Portland, Oregon, USA Gilbert DERAY Groupe Hôspitalier Pitié Salpétrière, Service De Néphrologie, Paris, France

LIST OF CONTRIBUTORS

ABDELRAHEEM Mohamed B. University of Khartoum, Soba University Hospital Pediatric Nephrology Unit PO Box 8081 Amarat Khartoum Sudan

ARANY Istvan Central Arkansas Veterans Health Care System 4300 West 7th Street 111/LR Little Rock, AR 72205 USA

ABRAMOWICZ Daniel Erasme Hospital Nephrology Department Route de Lennik, 808 1070 Brussels Belgium

BATLLE Daniel C. Northwestern University, Feinberg School of Medicine Division of Nephrology/Hypertension 303 E. Chicago Avenue Chicago, IL 60611 USA

AGGARWAL Nidhi Columbia University College of Physicians and Surgeons Department of Medicine, Division of Nephrology 622 West 168th Street PH 4124, New York, New York 10032 USA.

BENGHANEM Gharbi M. Centre Hospitalier Universitaire Ibn Rochd Service de Néphrologie Quartier Des Hôpitaux Casablanca Morocco

APPEL Gerald B. Columbia University College of Physicians and Surgeons Department of Medicine, Division of Nephrology 622 West 168th Street PH 4124, New York, NY 10032 USA.

BENNETT William M. Legacy Good Samaritan Hospital 1015 NW 22nd Avenue, W004 Portland, OR 97210 USA

Contributors

BERNS Jeffrey S. University of Pennsylvania School of Medicine Renal-Electrolyte and Hypertension Division 3400 Spruce Street, 1 Founders Philadelphia, PA 19104 USA

COJOCEL Constantin Kuwait University Kuwait † 2007

BLOCKMAN Marc Groote Schuur Hospital Division of Pharmacology Cape Town, 7925 Western Cape South Africa

CONTRERAS Gabriel N. University of Miami, Miller School of Medicine Division of Nephrology 1600 N.W. 10th Street Room 7168 (R126) Miami, FL 33136 USA

BRANCH Robert A. Center for Clinical Pharmacology Thermo Fisher Scientific Building, Suite 450 100 Technology Drive Pittsburgh, PA 15219 USA

COVINGTON Marisa D. Medical University of South Carolina 280 Calhoun Street MSC 140 Charleston, SC 29425-1400 USA

BREWSTER Ursula C. Yale University School of Medicine Section of Nephrology FMP 107, 330 Cedar Street (PO Box 208029) New Haven, CT 06520-8029 USA

DEBELLE Frédéric Erasme Hospital, Université Libre de Bruxelles Nephrology Department Route de Lennik 808 1070 Brussels Belgium

BURDMANN Emmanuel A. São José do Rio Preto Medical School (Nephrol Div) Faculdade de Medicina de São José do Rio Preto Av. Brigadeiro Faria Lima 5416 150090-000 São José do Rio Preto SP Brazil

DE BROE Marc E. University of Antwerp Laboratory of Pathophysiology Universiteitsplein 1 B-2610 Wilrijk Belgium

CAPITANO Blair 5934 Elwood Street Pittsburgh, PA 15232 USA

DECKER Brian Indiana University School of Medicine, Division of Nephrology 1001 West 10th Street, OPW 526 Indianapolis, IN 46202-5115 USA

CAROLINO Ruither O. University of Sao Paulo Department of Biochemistry and Immunology Campus Monte Alegre 14048-900 Ribeirao Preto, Sao Paulo Brazil

DE JONG Paul E. Groningen University Medical Center Hanzeplein 1 NL-9713 Groningen Netherlands

CARVALHO Decio University of Miami, Miller School of Medicine Division of Nephrology 1600 N.W. 10th Street Room 7168 (R126) Miami, FL 33136 USA

DERAY Gilbert Groupe Hôspitalier Pitié Salpétrière Service De Néphrologie 83, Boulevard de l’Hôpital F- 75013 Paris France

CHAFFER Sheldon C. The Department of Medicine Division of Nephrology and Hypertension The Texas A&M College of Medicine/Scott & White Clinic Temple, TX 76508 USA

DE WEERDT Dirk L. University of Antwerp Laboratory of Pathophysiology Universiteitsplein 1 B-2610 Wilrijk Belgium

XII

Contributors

D’HAESE Patrick C. University of Antwerp Laboratory of Pathophysiology Universiteitsplein 1 B-2610 Wilrijk Belgium

ERLEY Christiane M. St. Joseph-Krankenhaus Berlin Bäumerplan 24 D-12101 Berlin Germany

DIEGO Jorge M. University of Miami, Miller School of Medicine Division of Nephrology 1600 N.W. 10th Street Room 7168 (R126) Miami, FL 33136 USA

ESPINAL Isabel University of Miami, Miller School of Medicine Division of Nephrology 1600 N.W. 10th Street Room 7168 (R126) Miami, FL 33136 USA

DJUKANOVIC Ljubica Clinical Centre of Serbia Institute of Urology and Nephrology Pasterova 2 11000 Beograd Serbia

FEINFELD Donald A. Beth Israel Medical Center, Division of Nephrology & Hypertension, Baird Hall, 18BH26 350 East 17th Street New York, NY 10003 USA

DOYLE Alden University of Pennsylvania School of Medicine Renal-Electrolyte and Hypertension Division 3400 Spruce Street, 1 Founders Philadelphia, PA 19104 USA

FINN William F. University of North Carolina UNC Kidney Center 7030 Burnett-Womack Bldg CB #7155 Chapel Hill, NC 27599 USA

EDELSTEIN Charles L. University of Colorado Health Sciences Center Renal Box C281, 4200 East 9th Avenue Denver USA 80262 CO

FLORQUIN Sandrine Academic Medical Center, University of Amsterdam Department of Pathology Meibergdreef 9 NL-1105 AZ Amsterdam The Netherlands

EGBERTS Toine C. Department of Pharmacoepidemiology Utrecht Institute for Pharmaceutical Sciences PO Box 90107 Nl-3508 TB Utrecht The Netherlands

FOURNIÉ Gilbert J. INSERM U563, Centre de Physiopathologie Toulouse Purpan, 31024 -Toulouse cedex 3 France

ELINDER Carl Gustaf Department of Renal Medicine Karolinska University Hospital Huddinge & Solna S-141 86 Stockholm Sweden

FOWLER Bruce A. Agency for Toxic Substances and Disease Registry 4770 Buford Hwy NE Atlanta, GA 30341 USA

ELSEVIERS Monique M. University of Antwerp Faculty of Medicine Universiteitsplein 1 B-2610 Wilrijk Belgium

GARCIA-CAIRASCO Norberto University of Sao Paulo, Department of Physiology Faculty of Medicine of Ribeirao Preto, Campus Monte Alegre 14048-900 Ribeirao Preto, Sao Paulo Brazil

ENDRE Zoltan H. Christchurch School of Medicine, University of Otago Department of Medicine Po Box 4345 Christchurch, 8001 Canterbury New Zealand

GREEN Jonathan R. Musculoskeletal Diseases Novartis Institutes for BioMedical Research CH-4002 Basel Switzerland

XIII

Contributors

HAKKAPAKKI Santosh Northwestern University, Feinberg School of Medicine Division of Nephrology/Hypertension 303 E. Chicago Avenue Chicago, IL 60611 USA

KIDO Teruhiko Kanazawa University, Faculty of Medicine School of Health Sciences 5-11-80 Kodatsuno Kanazawa 920-0942 Japan

HAMDOUK Mohamed I. Bahry Renal Centre PO Box 1544 Kartoum North Sudan

KOPPELSTÄTTER Christian Innsbruck Medical University, Department of Internal Medicine, Division of Nephrology Anichstrasse 35 A-6020 Innsbruck Austria

HARBORD Nikolas Beth Israel Medical Center, Division of Nephrology & Hypertension, Baird Hall, 18BH26 350 East 17th Street New York, NY 10003 USA

KUNIS Cheryl L. Columbia University College of Physicians and Surgeons Department of Medicine, Division of Nephrology 622 West 168th Street PH 4124, New York, NY 10032 USA.

ISNARD-BAGNIS Corinne Groupe Hôspitalier Pitié Salpétrière Service De Néphrologie 83, Boulevard de l’Hôpital F- 75013 Paris France

LAUNAY-VAUCHER Vincent Groupe Hôspitalier Pitié Salpétrière Service De Néphrologie 83, Boulevard de l’Hôpital F- 75013 Paris France

JARNBERG Per O. Department of Anesthesiology and Peri-Operative Medicine 3181 S.W. Sam Jackson Pk. Rd., UHS-2 Portland, OR 97239-3098 USA

LECHNER Judith Innsbruck Medical University, Department of Physiology and Medical Physics, Division of Physiology Fritz-Pregl Strasse 3 A-6020 Innsbruck Austria

JENNINGS Paul Innsbruck Medical University, Department of Physiology and Medical Physics, Division of Physiology Fritz-Pregl Strasse 3 A-6020 Innsbruck Austria

LERMA Edgar V. University of Illinois at Chicago College of Medicine/ Associates in Nephrology, S.C. Department of Medicine, Section of Nephrology Chicago, IL 60612 USA

KARIE Svetlana Groupe Hôspitalier Pitié Salpétrière Service De Néphrologie 83, Boulevard de l’Hôpital F- 75013 Paris France

LEUFKENS Hubert G. Utrecht University, Institute for Pharmaceutical Sciences Division of Pharmacoepidemiology & Pharmacotherapy PO Box 80 082 NL-3508 TB Utrecht The Netherlands

KASBEKAR Nishaminy PENN Presbyterian Medical Center, University of Pennsylvania Health System, Department of Pharmacy 51 North 39th Street Philadelphia, PA 19104 USA

LOPES Norberto P. University of Sao Paulo Department of Physics and Chemistry Campus Monte Alegre 14048-900 Ribeirao Preto, Sao Paulo Brazil

KAUSHAL Gur P. Central Arkansas Veterans Health Care System 4300 West 7th Street 111/LR Little Rock, AR 72205 USA

MARKOWITZ Glen Columbia University, College of Physicians and Surgeons, Department of Pathology 630 West 168th Street, VC14-224 New York, NY 10032 USA

XIV

Contributors

MEGYESI Judit Central Arkansas Veterans Health Care System 4300 West 7th Street 111/LR Little Rock, AR 72205 USA

OLYAEI Ali J. Oregon Health Sciences University, Division of Nephrology, Hypertension and Clinical Pharmacology 3181 SW Sam Jackson Park Road Portland, OR 97201 USA

MOCELIN CARVALHO Cristiane University of Miami, Miller School of Medicine Division of Nephrology 1600 N.W. 10th Street Room 7168 (R126) Miami, FL 33136 USA

PELLETIER Lucette INSER U563, Centre de Physiopathologie Toulouse Purpan, 31024 -Toulouse cedex 3 France

MOLITORIS Bruce A. Indiana University School of Medicine Division of Nephrology, FH 115 E 1120 South Drive Indianapolis, IN 46202-5115 USA

PERAZELLA Mark A. Yale University School of Medicine Section of Nephrology, Dept. of Internal Medicine 330 Cedar Street New Haven, CT 06520-8029 USA

MUNIZ-MARTINEZ Marie-Carmen Erasme Hospital, Université Libre de Bruxelles Nephrology Department Route de Lennik 808 1070 Brussels Belgium

PFALLER Walter Innsbruck Medical University, Department of Physiology and Medical Physics, Division of Physiology Fritz-Pregl Strasse 3 A-6020 Innsbruck Austria

NAAZ Parveen Northwestern University, Feinberg School of Medicine Division of Nephrology/Hypertension 303 E. Chicago Avenue Chicago, IL 60611 USA

PORTER George A. Oregon Health Sciences University, Division of Nephrology, Hypertension and Clinical Pharmacology 3181 SW Sam Jackson Park Road Portland, OR 97201 USA

NETO Miguel M. University of Sao Paulo, Division of Nephrology Faculty of Medicine of Ribeirao Preto Campus Monte Alegre 14048-900 Ribeirao Preto, Sao Paulo Brazil

PORTILLA Didier Central Arkansas Veterans Health Care System 4300 West 7th Street 111/LR Little Rock, AR 72205 USA

NORDBERG Gunnar Umea University Environmental Medicine Department of Public Health and Clinical Medicine S-90187 Umea Sweden

PRICE Peter M. Central Arkansas Veterans Health Care System 4300 West 7th Street 111/LR Little Rock, AR 72205 USA

NORTIER Joëlle Erasme Hospital, Université Libre de Bruxelles Nephrology Department Route de Lennik 808 1070 Brussels Belgium

PUSCHETT Jules B. Texas A & M Health Science Center College of Medicine/ Scott & White, Central Texas Veterans Healthcare System Medical Education Building 407 L, 2401 S. 31st St. Temple, TX 76508 USA

OKUSA Mark D. University of Virginia Health System Division of Nephrology Box 133 Charlottesville, VA 22908 USA

RADOVANOVIĆ Zoran Kuwait University, Faculty of Medicine, Department of Community Medicine and Behavioural Sciences POB 24923 13110 Safat Kuwait XV

Contributors

RITZ Eberhard Medizinische Universitatsklinik Heidelberg Department of Internal Medicine, Division Nephrology Im Neuenheimer Feld 162 D-69120 Heidelberg Germany

SWANEPOEL Charles Groote Schuur Hospital Renal Unit (E 13) Observatory Cape Town, 7925 Western Cape South Africa

ROCH-RAMEL Françoise († 2001) Université de Lausanne Switzerland

TAHA Ahbab Bahry Renal Centre Po Box 1544 Khartoum North Sudan

ROELOFS Joris J. Academic Medical Center, University of Amsterdam Department of Pathology Meibergdreef 9 NL-1105 AZ Amsterdam The Netherlands

TALMUD Joe Groote Schuur Hospital Division of Pharmacology Cape Town, 7925 Western Cape South Africa

ROELS Harry A. Industrial Toxicology and Occupational Medicine Unit Université catholique de Louvain Avenue E. Mounier, 53/02 B-1200 Brussels Belgium

UEDA Shiro Chiba University, Department of Drug Information and Communication, Graduate School of Pharmaceutical Sciences 1-8-1 Inohana, Chiba City 260-8675 Japan

ROSNER Mitchell H. University of Virginia Health System Division of Nephrology Charlottesville, VA 22908 USA

VANHERWEGHEM Jean-Louis Erasme Hospital, Université Libre de Bruxelles Nephrology Department Route de Lennik 808 1070 Brussels Belgium

SAFIRSTEIN Robert L. Central Arkansas Veterans Health Care System 4300 West 7th Street 111/LR Little Rock, AR 72205 USA

VERHULST Anja University of Antwerp Laboratory of Pathophysiology Universiteitsplein 1 B-2610 Wilrijk Belgium

SAOUDI Abdelhadi INSERM U563, Centre de Physiopathologie ToulousePurpan Hôpital Purpan BP 3028 31024 Toulouse Cedex France

WAHBA Ihab Oregon Health Sciences University, Division of Nephrology, Hypertension and Clinical Pharmacology 3181 SW Sam Jackson Park Road Portland, OR 97201 USA

SCHNELLMANN Rick G. Medical University of South Carolina 280 Calhoun Street MSC 140 Charleston, SC 29425-1400 USA STURMER Til German Centre for Research on Aging Bergheimer Strasse 20 D-69115 Heidelberg Germany XVI

WEDEEN Richard P. Department of Veterans Affairs New Jersey Health Care System 385 Tremont Ave East Orange, NJ 07018-1095 USA WHELTON Andrew Universal Clinical Research Center Johns Hopkins University 615 North Wolfe Street Baltimore, MD 21205 USA

Contributors

WHITTAKER Margaret Toxservices 1326 18th Street N.W. Washington, DC 20036 USA WINCHESTER James Beth Israel Medical Center, Division of Nephrology & Hypertension, Baird Hall, 18BH26 350 East 17th Street New York, NY 10003 USA YACOOB Muhammed The Royal London Hospital Department of Renal Medicine and Transplantation Whitechapel London E1 1BB UK ZGHEIB Nathalie K. Pharmacology and Therapeutics American University of Beirut PO Box 11-0236 Riad El-Solh / Beirut 1107 2020 Lebanon

XVII

01

Clinical relevance George A. PORTER Oregon Health Sciences University, Portland, Oregon, USA

General incidence and outcome _____________________________________________ 3 Definition Incidence Outcome

3 4 5

Mechanisms of drug induced acute kidney injury _______________________________ 6 Particular features due to specific drugs ______________________________________ 10 Monitoring of renal function _______________________________________________ 12 Populations at risk ________________________________________________________ 13 Genetic/hereditary susceptibility Occupational/environmental exposure Gender Race Nutrition Socio-economic status Age Co-existing chronic diseases Addictive behavior Summary

14 15 16 16 17 17 17 18 19 19

Individual risk factors _____________________________________________________ 20 References ______________________________________________________________ 21

General incidence and outcome

D

rugs are a frequent cause of both in-hospital and community-acquired acute kidney injury (AKI). Nephrotoxic drugs share the spotlight with renal hypoxia as primary etiologic factors for hospital acquired AKI [1,2,3]. With the increasing capacity of the medical community to treat the most serious life-threatening conditions, the in-hospital exposure to nephrotoxic drugs has increased as has the risk of drug-induced AKI, while the expanded drug treatments available for outpatient use is contributing to the rise in community acquired AKI.

Definition Acute kidney injury (AKI) is easily defined as a syndrome characterized by a sudden decrease in GFR accompanied by azotemia [4]. However; the reported incidence of AKI varies depending on a number of independent variables. For example, was the patient population surveyed derived from a community wide database or was it restricted to hospitalized patients? What definition was adopted to designate acute kidney injury (AKI)? The lack of a universally agreed upon definition of acute kidney injury (AKI), makes it difficult to compare clinical reports as to the incidence,

PORTER

severity and outcome. Recently, the Acute Dialysis Quality Initiative [5] has attempted to address this issue involving both nephrologists and critical care physicians in the discussion. Success of this project is critical for it will allow the sharing of information regarding interventions which, in turn, will improve the dismal outcomes that currently exist for patients with acute kidney injury. This dismal outcome is especially true if the patient suffers from the constellation of multiple organ failure which is becoming common place in ICUs. Encouragement comes from the success of the KDOQI classification of Chronic Kidney Disease (CKD) [6] that is being adopted world wide and allows consistent stratification based on glomerular filtration rate (GFR) [7]. Additional variables exist, for example, in-hospital surveys enroll both post-surgical and medical patients, and it is important to isolate the contribution from ICU patients with multi-organ failure! With what precision was the AKI diagnoses established? Were multiple centers involved in providing the information? These are the often unanswered questions that complicate meaningful estimates of the incidence of AKI. In addition, as detailed by Turney et al. [8] and Nash and co-workers [3], significant changes have occurred in both the age of the AKI patients and also the etiologies, with older and sicker patients being admitted for treatment.

Incidence The incidence of in-hospital AKI attributed to drug nephrotoxicity is estimated at between 18 and 40% of cases [9-16], while earlier reports, derived from community-based statistics set the incidence from 0 to 7% of cases [10, 11]; however, more recent series report incidence from 17 to 29% [17,18] which is still lower than the 37% reported by Baraldi et al. [14]. In both series that included hospital acquired AKI [17,18], the contribution of nephrotoxic drugs was slightly higher, 22 and 35%. The recently reported increase in the contribution of nephrotoxic drugs to community acquired AKI is particularly important since the total number of all cause cases of AKI, as derived from community based studies, is 2 to 3.5 times greater that reported from in-hospital statistics [17,19]. This suggested that patients who are hospitalized are either exposed to more nephrotoxic agents and/or they are 4

more vulnerable to the drugs inducing an adverse renal effect. Community acquired AKI is associated with a significantly reduced mortality as compared to hospital acquired AKI [21], (41% vs. 59%, p2%, UOsm 75 years

4

Anemia

3

Diabetes

3

Patient-specific factors Advanced age Diabetes mellitus Impaired renal function Impaired cardiac function Volume depletion Multiple nephrotoxic medications Radiocontrast agent exposure

Contrast media volume

Medication use NSAIDS/Cox-2 inhibitors Aminoglycoside antibiotics Amphotericin B ACE-inhibitors/angiotensin-receptor antagonists Calcineurin inhibitors Chemotherapeutic agents (cisplatin, ifosfamide) Illicit drug use (cocaine) Deliberate or accidental ingestion of toxins (ethylene glycol) Occupational toxins (heavy metals, organic solvents) Herbal remedies (aristolochic acid)

16

57.3%

12.6%

been: severe left ventricular dysfunction (especially that requiring use of an intra-aortic balloon pump), prolonged cardiopulmonary bypass, older age, diabetes mellitus, and pre-existing renal impairment [81-83]. This last factor is perhaps the most important with the risk of AKI requiring dialysis approaching 10-20% in those patients undergoing cardiac surgery with a baseline serum creatinine between 2.0 and 4.0 mg/dL [84]. In patients exposed to radiocontrast agents, the key risk determinants for AKI include: chronic kidney disease stage III or greater (estimated GFR < 60 ml/min), diabetes mellitus, volume depletion, nephrotoxic drug use, preprocedural hemodynamic instability, anemia, congestive heart failure and hypoalbuminemia [85]. The importance of baseline renal function in this setting is exemplified by one registry study that demonstrated an incidence of AKI of 2.5% in patients with mild renal impairment (serum creatinine 1.2 to 1.9 mg/dL), which rose to 30.6% in those patients with more severe renal impairment (serum creatinine > 3.0 mg/dL) [86]. Identification of risk factors has been used to pro34

1 for each 100 ml

Serum creatinine > 1.5 mg/dL or eGFR 40-60 ml/min eGFR 20-40 ml/min eGFR < 20 ml/min Risk score

Risk of contrastinduced nephropathy

4 2 4 6 Risk of dialysis

Adapted from: Mehran R, et al. A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention: development and initial validation. J Am Coll Cardiol 2004; 44: 1393-1399.

duce clinical AKI predictive scoring systems that attempt to better quantify cumulative risk. These scoring systems are most useful in situations where a possible nephrotoxic exposure is to occur at a defined time (such as cardiac surgery or radiographic contrast exposure). They provide a very useful framework to identify patients who are at risk and thus may benefit from renal protective strategies. For example, a scoring system developed at the Cleveland Clinic utilizes 13 pre-operative variables to predict a risk for post-cardiac surgery AKI [87]. Similar scoring systems have been developed by others for cardiac surgery and for other settings such as radiocontrast media exposure [88-90]. An example of one such risk-scoring scheme for contrast-induced nephropathy is shown in Table 3 [90]. These scoring systems attempt to identify a small number of high-risk patients and thus will have good negative predictive power but will often lack positive predictive power. Many of these predictive scoring systems have not been validated across different population groups and thus are limited in their utility. One important factor that limits the determination of risk for AKI is the poor sensitivity of serum

02. Drug-associated acute kidney injury in the intensive care unit

Table 4. General approaches for the prevention of AKI. 1. Avoidance of nephrotoxins Recognition of potential nephrotoxic agents Recognition of high risk patients and clinical settings Avoidance of concomitant use of multiple nephrotoxins Use of lowest dose and for shortest time possible If applicable, monitoring of drug dose Frequent monitoring of renal function Maintain euvolemia 2. Minimization of nosocomial infection 3. Extracellular fluid expansion (maintain good urine output, stable hemodynamics) 4. Avoid agents that impair renal blood flow autoregulation (NSAIDS, ACE inhibitors, ARBs) 5. Pharmacological Interventions – if applicable (Table 5)

Table 5. Examples of specific renal protective strategies. Exposure

Strategy

Radiocontrast agents

IV hydration (normal saline) [95] IV sodium bicarbonate [96] N-acetylcysteine [108, 109] Vitamin C [123] Iso-osmolar contrast [124]

Aminoglycoside antibiotics Once-daily dosing [125] Monitoring of drug levels Tumor lysis (uric acid)

Allopurinol/rasburicase [126] IV hydration/urine alkalinization

Ethylene glycol ingestion

Ethanol/fomepizole [127] Hemodialysis

Rhabdomyolysis

IV hydration/urine alkalinization [128] ± mannitol [129]

6. Use of computer surveillance systems

creatinine values for detection of mild degrees of renal injury. In fact, there is no practical, “real-time” method to provide accurate determination for mild degrees of kidney injury. Oliguria certainly heralds the presence of significant kidney dysfunction, but most causes of AKI are non-oliguric [91]. Thus, a relatively large decrease in glomerular filtration rate (GFR) may be associated with only small changes in the serum creatinine (especially true in those patients with normal baseline renal function). Furthermore, the serum creatinine is influenced by variables such as production rate, muscle mass and the volume of distribution. Thus, a cirrhotic patient who may be malnourished and volume expanded may appear to have a “normal” serum creatinine value when, in fact, there is significant kidney impairment [92]. All of this makes heightened awareness of the clinical setting and risks associated with AKI more important in the early detection of AKI. Careful attention to even small increases in serum creatinine as well as attention to urine abnormalities (presence of granular casts) is critical for the early detection of AKI. It is hoped that sensitive biomarkers of kidney injury may ultimately allow identification of patients at the earliest signs of AKI.

Renal protective strategies Strategies used to prevent AKI can be broadly separated into generalized approaches and those approaches which are more specifically targeted to a particular risk factor (Tables 4 and 5). Certainly improvements in overall ICU care that focus on the

Methotrexate

IV hydration/urine alkalinization [48]

Acyclovir

IV hydration [54]

Calcineurin inhibitors

Monitor drug levels [130] ± calcium-channel blockers [131]

Amphotericin B

Use of lipid formulation [132]

risk factors identified above should reduce the incidence of AKI. In fact, early and aggressive therapy of hemodynamically unstable patients in the emergency department using a combination of IV hydration and pressor agents led to an impressive 88.5% reduction in the incidence of AKI [93]. Thus careful attention to volume status and maximization of cardiac output along with minimization of exposure to nephrotoxic agents should be employed in all at risk patients. Agents that impair the critical autoregulation of renal blood flow such as NSAIDs, ACE inhibitors, angiotensinreceptor antagonists (ARBs) should be avoided. Plasma concentrations of selected nephrotoxic drugs (aminoglycosides, calcineurin inhibitors) should be monitored closely and cumulative dose should be limited. Despite these clear recommendations, Weisbord and co-workers found that 16% of patients who were at clear risk for the development of contrast-induced nephropathy never received pre-procedural IV fluids and 8% of these patients were prescribed NSAIDS or COX-2 inhibitors [19]. One strategy to reduce the incidence of AKI has adopted a computer surveillance system that notifies physicians via e-mail messages whenever a small rise in serum creatinine occurs in their patients who are receiving potential nephrotoxic medications [94]. This notification system led to earlier cessation of offending 35

ROSNER & OKUSA

drugs and a decrease in the incidence of severe AKI from 7.5 to 3.4%.

Specific strategies to reduce the incidence of acute kidney injury Intravenous fluids clearly reduce the risk of AKI across a spectrum of etiologies. For example, in the prevention of contrast-induced nephropathy, one study compared IV hydration with 0.9% saline at 1 ml/kg/ hour beginning 12 hours prior to the study with unrestricted oral fluids. The incidence of AKI (as defined by a 0.5 mg/dL or greater rise in serum creatinine) was 3.7% in the IV hydration group and 34.6% in the oral fluid group [95]. Saline-based therapies may not be as effective as a bicarbonate-based solution in this setting [96], however confirmation will be necessary from other centers. In the setting of sepsis, while IV fluid resuscitation is clearly critical, the optimal form of volume support is not known. Three meta-analyses have compared crystalloid versus colloid solutions with at least no difference or perhaps a slight increase in mortality associated with colloid solutions [97-99]. In a multicenter randomized controlled trial of resuscitation fluids (saline versus albumin), there was no difference between the fluids in 28-day mortality, organ failure, days on renal replacement therapy, days on mechanical ventilation, or hospital days [100]. The Cochrane group concluded that albumin administration in severely ill patients was associated with increased mortality as compared with other IV fluids [101]. Other colloid solutions such as hydroxyethylstarch and gelatin have also been studied and do not seem to have an advantage over crystalloids [102]. In fact, hydroxyethylstarch was associated with a higher risk of AKI than gelatin [62]. In the preoperative setting, the use of IV fluids to “optimize” cardiac performance (as guided by pulmonary artery catheter measurements) has been shown to be beneficial with a reduction in the incidence of AKI from 4.8% to 1.5% in patients undergoing vascular surgery [103]. However, volume expansion to supranormal cardiac indices along with normal mixed venous oxygen saturation had no effect on the incidence of AKI and can not be routinely recommended [104]. In some patients, vasopressor agents are required to maintain hemodynamic stability. Few direct comparisons exist to support one vasopressor over another 36

[105]. However, accumulating evidence supports the use of norepinephrine in patients with septic shock with a retrospective study demonstrating reduced mortality with norepinephrine over other vasopressors [106]. Furthermore, animal data demonstrates that reversal of septic hypotension with norepinephrine leads to increases in renal blood flow [107]. There are no studies that compare the renal outcomes between catecholamine therapy and vasopressin. One renal protective strategy that is often overlooked is the intensive control of blood glucose levels in critically ill patients [107]. Insulin therapy reduced the risk of AKI that required dialysis by 41% in one trial [107]. While the mechanism of this effect is not known, this easily implemented strategy should be considered in all at risk patients. N-acetylcysteine has been widely advocated as a renoprotective agent especially in the setting of radiocontrast media exposure. Several meta-analyses have shown that N-acetylcysteine can reduce the incidence of contrast-induced nephropathy by nearly 50% [108, 109]. However, in other settings such as post-cardiac surgery, N-acetylcysteine has not proved to be of benefit [110]. Furthermore, N-acetylcysteine may be of less benefit in those patients with moderate or severe chronic kidney disease [111]. Many other renal protective strategies have been attempted with poor results. Dopamine at doses between 0.5 to 5.0 ug/kg/minute has been promoted as a therapy to increase renal blood flow, induce natriuresis and diuresis and perhaps increase GFR. However, in multiple settings ranging from sepsis, contrast exposure, and cardiac surgery dopamine has not been shown to be beneficial in preventing AKI (reviewed in 56)[112]. Fenoldopam is a more selective dopamine A-1 agonist that increases renal blood flow to the cortex and outer medulla. A recent meta-analysis of 16 small trials has suggested that there may be a small benefit in reducing the risk of AKI [113]. However, most of the studies in this meta-analysis were underpowered and a larger, randomized clinical trial is required before this therapy can be recommended. Other agents that have been used and have shown no or at best marginal benefits include: atrial natriuretic peptide [114], clonidine [115], calcium channel blockers [116], furosemide [117], inotropic agents [118], growth factors [119, 120], and theophylline [121], as well as numerous others. These failures highlight the critical importance

02. Drug-associated acute kidney injury in the intensive care unit

of nonpharmacological therapies. One controversial strategy is the use of prophylactic dialysis to prevent AKI. This has been evaluated in the setting of high-risk patients undergoing coronary angioplasty [122]. In this study, patients with baseline serum creatinine values > 2 mg/dL were randomized to either IV fluids or IV fluids with hemofiltration that was commenced 4-6 hours prior to the procedure and continued for 18-24 hours after contrast administration. The group receiving extracorporeal therapy had a lower incidence of AKI requiring dialysis, a lower hospital mortality rate. However, the invasiveness and cost of this therapy as well as inherent flaws in the study (difference in total IV hydration, lack of N-acetyl cysteine use, difference in loop diuretic use between groups) prevents this strategy from being used more widely. There are several preventative strategies that are specific to either clinical states (rhabdomyolysis) or

nephrotoxic exposures. These are listed in Table 4 and discussed elsewhere in more detail. In these specific instances, these steps, in addition to the general strategies discussed above, may be employed to reduce the risk of AKI. However, it is critical to realize that these strategies are useful only when applied prophylactically to at-risk patients or are applied very soon after a renal insult. Currently, the best evidence supports the use of non-pharmacological strategies in reducing the risk of AKI. Maintenance of blood pressure, avoidance of nephrotoxins, attention to risk factors and small changes in serum creatinine afford the greatest benefit. In certain specific instances, use of pharmacological agents such as N-acetylcysteine may be of use but more generalized pharmacological approaches to the prevention of AKI have not yet come to fruition. Thus, vigilance and rapid response with conservative measures are warranted in all patients.

References 1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13.

14. 15. 16. 17.

Ronco C, and Bellomo R. 2003. Prevention of acute renal failure in the critically ill. Nephron Clin Pract 93:C13-20. Mehta RL, Pascual MT, Soroko S, Savage BR, Himmelfarb J, Ikizler TA, Paganini EP, and Chertow GM. 2004. Spectrum of acute renal failure in the intensive care unit: the PICARD experience. Kidney Int 66:1613-1621. Uchino S, Kellum JA, Bellomo R, et al. 2005. Acute renal failure in critically ill patients: a multinational, multicenter study. Jama 294:813-818. Silvester W, Bellomo R, and Cole L. 2001. Epidemiology, management, and outcome of severe acute renal failure of critical illness in Australia. Crit Care Med 29:1910-1915. Bellomo R, Ronco C, Kellum JA, Mehta RL, and Palevsky, P. 2004. Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 8:R204-212. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, and Kellum J.. 2006. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care 10:R73. Liano F, Junco E, Pascual J, Madero R, and Verde E. 1998. The spectrum of acute renal failure in the intensive care unit compared with that seen in other settings. The Madrid Acute Renal Failure Study Group. Kidney Int Suppl 66:S16-24. Schetz M, Dasta J, Goldstein S, and Golper T. 2005. Drug-induced acute kidney injury. Curr Opin Crit Care 11:555-565. Evenepoel P. 2004. Acute toxic renal failure. Best Pract Res Clin Anaesthesiol 18:37-52. Choudhury D, and Ahmed Z. 2006. Drug-associated renal dysfunction and injury. Nat Clin Pract Nephrol 2:80-91. Nash K, Hafeez A, and Hou S. 2002. Hospital-acquired renal insufficiency. Am J Kidney Dis 39:930-936. Hou SH, Bushinsky DA, Wish JB, Cohen JJ, and Harrington JT. 1983. Hospital-acquired renal insufficiency: a prospective study. Am J Med 74:243-248. Brivet FG, Kleinknecht DJ, Loirat P, and Landais PJ. 1996. Acute renal failure in intensive care units--causes, outcome, and prognostic factors of hospital mortality; a prospective, multicenter study. French Study Group on Acute Renal Failure. Crit Care Med 24:192-198. Henrich WL. 2005. Nephrotoxicity of several newer agents. Kidney Int Suppl:S107-109. Schlondorff D. 1993. Renal complications of nonsteroidal anti-inflammatory drugs. Kidney Int 44:643-653. Dzeka TN, Townley R, and Arnold JM. 2003. Effects of enalaprilat on venoconstriction to norepinephrine: role of prostaglandins. Cardiovasc Res 59:250-256. Brezis M, and Rosen S. 1995. Hypoxia of the renal medulla--its implications for disease. N Engl J Med 332:647-655.

37

ROSNER & OKUSA

18. 19. 20.

21. 22. 23. 24. 25. 26. 27.

28. 29. 30. 31. 32. 33. 34.

35. 36. 37. 38.

39. 40. 41. 42. 43. 44. 45.

38

Bakris GL, Lass N, Gaber AO, Jones JD, and Burnett JC. 1990 Jan. Radiocontrast medium-induced declines in renal function: a role for oxygen free radicals. American Journal of Physiology 258:F115-120. Weisbord SD, Bruns FJ, Saul MI, and Palevsk, PM. 2004. Provider use of preventive strategies for radiocontrast nephropathy in high-risk patients. Nephron Clin Pract 96:c56-62. Schneider V, Levesque LE, Zhang B, Hutchinson T, and Brophy JM. 2006. Association of selective and conventional nonsteroidal antiinflammatory drugs with acute renal failure: A population-based, nested case-control analysis. Am J Epidemiol 164:881889. Perazella MA. 2003. Drug-induced renal failure: update on new medications and unique mechanisms of nephrotoxicity. Am J Med Sci 325:349-362. Taber SS, and Mueller BA. 2006. Drug-associated renal dysfunction. Crit Care Clin 22:357-374, viii. Zatz R, Dunn BR, Meyer TW, Anderson S, Rennke HG, and Brenner BM. 1986. Prevention of diabetic glomerulosclerosis by pharmacological amelioration of glomeular capillary hypertension. J. Clin Invest 77:1925-1930. Scott LJ, McKeage K, Keam SJ, and Plosker GL. 2003. Tacrolimus: a further update of its use in the management of organ transplantation. Drugs 63:1247-1297. Lamas S. 2005. Cellular mechanisms of vascular injury mediated by calcineurin inhibitors. Kidney Int 68:898-907. Guieu R, Dussol B, Devaux C, Sampol J, Brunet P, Rochat H, Bechis G, and Berland, Y.F. 1998. Interactions between cyclosporine A and adenosine in kidney transplant recipients. Kidney Int 53:200-204. Buchler M, Leibenguth P, Le Guellec C, Carayon A, Watier H, Odoul F, Autret-Lec, E, Lebranchu Y, and Paintaud G. 2004. Relationship between calcineurin inhibition and plasma endothelin concentrations in cyclosporine-A-treated kidney transplant patients. Eur J Clin Pharmacol 60:703-708. First MR, Schroeder TJ, Alexander JW, Stephens GW, Weiskittel P, Myre SA, and Pesce AJ. 1991. Cyclosporine dose reduction by ketoconazole administration in renal transplant recipients. Transplantation 51:365-370. Joannidis M. 2004. Drug-induced renal failure in the ICU. Int J Artif Organs 27:1034-1042. Vittecoq D, Dumitrescu L, Beaufils H, and Deray G. 1997. Fanconi syndrome associated with cidofovir therapy. Antimicrob Agents Chemother 41:1846. Verhelst D, Monge M, Meynard JL, Fouqueray B, Mougenot B, Girard PM, Ronco P, and Rossert J. 2002. Fanconi syndrome and renal failure induced by tenofovir: a first case report. Am J Kidney Dis 40:1331-1333. Ho ES, Lin DC, Mendel DB, and Cihlar T. 2000. Cytotoxicity of antiviral nucleotides adefovir and cidofovir is induced by the expression of human renal organic anion transporter 1. J Am Soc Nephrol 11:383-393. Meier P, Dautheville-Guibal S, Ronco PM, and Rossert J. 2002. Cidofovir-induced end-stage renal failure. Nephrol Dial Transplant 17:148-149. Bertino JS Jr, Booker LA, Franck PA, Jenkins PL, Franck KR, and Nafziger A. 1993. Incidence of and significant risk factors for aminoglycoside-associated nephrotoxicity in patients dosed by using individualized pharmacokinetic monitoring. J Infect Dis 167:173-179. Eisenberg JM, Koffer H, Glick HA, Connell ML, Loss, LE, Talbot, GH, Shusterman NH, and Strom BL. 1987. What is the cost of nephrotoxicity associated with aminoglycosides? Ann Intern Med 107:900-909. Baciewicz AM, Sokos DR, and Cowan RI. 2003. Aminoglycoside-associated nephrotoxicity in the elderly. Ann Pharmacother 37:182-186. Cannella CA, and Wilkinson ST. 2006. Acute renal failure associated with inhaled tobramycin. Am J Health Syst Pharm 63:18581861. Izquierdo MJ, Gomez-Alamillo C, Ortiz F, Calabia ER, Ruiz JC, de Francisco AL, and Arias M. 2006. Acute renal failure associated with use of inhaled tobramycin for treatment of chronic airway colonization with Pseudomonas aeruginosa. Clin Nephrol 66:464467. Paterson DL, Robson JM, and Wagener MM. 1998. Risk factors for toxicity in elderly patients given aminoglycosides once daily. J Gen Intern Med 13:735-739. Sandoval RM, and Molitoris BA. 2004. Gentamicin traffics retrograde through the secretory pathway and is released in the cytosol via the endoplasmic reticulum. Am J Physiol Renal Physiol 286:F617-624. Michel DM, and Kelly CJ. 1998. Acute interstitial nephritis. J Am Soc Nephrol 9:506-515. Toto RD. 1990. Acute tubulointerstitial nephritis. Am J Med Sci 299:392-410. Brewster UC, and Perazella MA. 2007. Acute kidney injury following proton pump inhibitor therapy. Kidney Int 71:589-593. Barakat RK, Singh N, Lal R, Verani RR, Finkel KW, and Foringer JR. 2007. Interstitial nephritis secondary to bevacizumab treatment in metastatic leiomyosarcoma. Ann Pharmacother 41:707-710. Hu D, C H, Okusa M. 2004. Polymorphoneutrophilic Infiltration in Acute Interstitial Nephritis of Sjogren’s Syndrome. The American Journal of the Medical Sciences 237.

02. Drug-associated acute kidney injury in the intensive care unit

46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76.

Markowitz GS, and Perazella MA. 2005. Drug-induced renal failure: a focus on tubulointerstitial disease. Clin Chim Acta 351:3147. Markowitz GS, Appel GB, Fine PL, Fenves AZ, Loon NR, Jagannath S, Kuhn JA, Dratch AD, and D’Agati VD. 2001. Collapsing focal segmental glomerulosclerosis following treatment with high-dose pamidronate. J Am Soc Nephrol 12:1164-1172. Perazella MA. 1999. Crystal-induced acute renal failure. Am J Med 106:459-465. Fogazzi GB. 1996. Crystalluria: a neglected aspect of urinary sediment analysis. Nephrol Dial Transplant 11:379-387. Brigden D, Rosling AE, and Woods NC. 1982. Renal function after acyclovir intravenous injection. Am J Med 73:182-185. Sawyer MH, Webb DE, Balow JE, and Straus SE. 1988. Acyclovir-induced renal failure. Clinical course and histology. Am J Med 84:1067-1071. Keeney RE, Kirk LE, and Bridgen D. 1982. Acyclovir tolerance in humans. Am J Med 73:176-181. Bean B, and Aeppli D. 1985. Adverse effects of high-dose intravenous acyclovir in ambulatory patients with acute herpes zoster. J Infect Dis 151:362-365. Izzedine H, Launay-Vache, V, and Deray G. 2005. Antiviral drug-induced nephrotoxicity. Am J Kidney Dis 45:804-817. Martinez F, Mommeja-Marin H, Estepa-Maurice L, Beaufils H, Boche, M, Daudon M, Deray G, and Katlama C. 1998. Indinavir crystal deposits associated with tubulointerstitial nephropathy. Nephrol Dial Transplant 13:750-753. Reilly RF, Tray K, and Perazella MA. 2001. Indinavir nephropathy revisited: a pattern of insidious renal failure with identifiable risk factors. Am J Kidney Dis 38:E23. Dlott JS, Danielson CF, Blue-Hnidy DE, and McCarthy LJ. 2004. Drug-induced thrombotic thrombocytopenic purpura/hemolytic uremic syndrome: a concise review. Ther Apher Dial 8:102-111. Medina PJ, Sipols JM, and George JN. 2001. Drug-associated thrombotic thrombocytopenic purpura-hemolytic uremic syndrome. Curr Opin Hematol 8:286-293. Pisoni R, Ruggenenti P, and Remuzzi G. 2001. Drug-induced thrombotic microangiopathy: incidence, prevention and management. Drug Saf 24:491-501. Frangie C, Lefaucheur C, Medioni J, Jacquot C, Hill GS, and Nochy D. 2007. Renal thrombotic microangiopathy caused by antiVEGF-antibody treatment for metastatic renal-cell carcinoma. Lancet Oncol 8:177-178. Visweswaran P, Massin EK, and Dubose TD. 1997 Jun. Mannitol-induced acute renal failure. Journal of the American Society of Nephrology 8:1028-1033. Schortgen F, Lacherade JC, Bruneel F, Cattaneo I, Hemery F, Lemaire F, and Brochard L. 2001. Effects of hydroxyethylstarch and gelatin on renal function in severe sepsis: a multicentre randomised study. Lancet 357:911-916. Anderson RJ, Linas SL, Berns AS, Henrich WL, Miller TR, Gabow PA, and Schrier RW. 1977. Nonoliguric acute renal failure. N Engl J Med 296:1134-1138. Corwin HL, Teplick RS, Schreiber MJ, Fang LS, Bonventre JV, and Coggins CH. 1987. Prediction of outcome in acute renal failure. Am J Nephrol 7:8-12. Shusterman N, Strom BL, Murray TG, Morrison G, West SL, and Maislin G. 1987. Risk factors and outcome of hospital-acquired acute renal failure. Clinical epidemiologic study. Am J Med 83:65-71. Rasmussen HH, and Ibels LS. 1982. Acute renal failure. Multivariate analysis of causes and risk factors. Am J Med 73:211-218. SelCuk NY, Odabas AR, Cetinkaya R, Tonbul HZ, and San A. 2000. Frequency and outcome of patients with acute renal failure have more causes than one in etiology. Ren Fail 22:459-464. Behrend T, and Miller SB. 1999. Acute renal failure in the cardiac care unit: etiologies, outcomes, and prognostic factors. Kidney Int 56:238-243. McMurray SD, Luft FC, Maxwell DR, Hamburger RJ, Futty D, Szwed JJ, Lavelle KJ, and Kleit SA. 1978. Prevailing patterns and predictor variables in patients with acute tubular necrosis. Arch Intern Med 138:950-955. Feest TG, Round A, and Hamad S. 1993. Incidence of severe acute renal failure in adults: results of a community based study. Bmj 306:481-483. D’Agati V, and Appel GB. 1997. HIV infection and the kidney. J Am Soc Nephrol 8:138-152. Liano F, Garcia-Martin F, Gallego A, Orte L, Teruel JL, Marcen,R, Matesanz R, and Ortuno J. 1989. Easy and early prognosis in acute tubular necrosis: a forward analysis of 228 cases. Nephron 51:307-313. Morris JA Jr, Mucha P Jr, Ross SE, Moore BF, Hoyt DB, Gentilello L, Landercasper J, Feliciano DV, and Shackford SR. 1991. Acute posttraumatic renal failure: a multicenter perspective. J Trauma 31:1584-1590. Wilkinson AH, and Cohen DJ. 1999. Renal failure in the recipients of nonrenal solid organ transplants. J Am Soc Nephrol 10:11361144. McCullough PA, Wolyn R, Rocher LL, Levin RN, and O’Neill WW. 1997. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med 103:368-375. Rosner MH, and Okusa MD. 2006. Acute kidney injury associated with cardiac surgery. Clin. J. Am. Soc. Nephrol. 1:19-32.

39

ROSNER & OKUSA

77. 78. 79. 80. 81.

82.

83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103.

40

Gines P, Guevara M, Arroyo V, and Rodes J. 2003. Hepatorenal syndrome. Lancet 362:1819-1827. Noel C, Hazzan M, Noel-Walter MP, and Jouet JP. 1998. Renal failure and bone marrow transplantation. Nephrol Dial Transplant 13:2464-2466. Sandhu JS, Sood A, Midha V, Singh AD, Jain D, and Sandhu P. 2000. Non-traumatic rhabdomyolysis with acute renal failure. Ren Fail 22:81-86. McCarthy JT. 1996. Prognosis of patients with acute renal failure in the intensive-care unit: a tale of two eras. Mayo Clin Proc 71:117-126. Mangano CM, Diamondstone LS, Ramsay JG, Aggarwal A, Herskowitz A, and Mangano DT. 1998. Renal dysfunction after myocardial revascularization: risk factors, adverse outcomes, and hospital resource utilization. The Multicenter Study of Perioperative Ischemia Research Group. Ann Intern Med 128:194-203. Abel RM, Buckley MJ, Austen WG, Barnett GO, Beck CH Jr, and Fischer JE. 1976. Etiology, incidence, and prognosis of renal failure following cardiac operations. Results of a prospective analysis of 500 consecutive patients. J Thorac Cardiovasc Surg 71:323333. Anderson LG, Ekroth R, Bratteby LE, Hallhagen S, and Wesslen O. 1993. Acute renal failure after coronary surgery--a study of incidence and risk factors in 2009 consecutive patients. Thorac Cardiovasc Surg 41:237-241. Chertow GM, Lazarus JM, Christiansen CL, Cook EF, Hammermeister KE, Grover F, and Daley J. 1997. Preoperative renal risk stratification. Circulation 95:878-884. McCullough PA, Adam A, Becker CR, Davidson C, Lameire N, Stacul F, and Tumlin J. 2006. Risk prediction of contrast-induced nephropathy. Am J Cardiol 98:27-36. Rital CS, T S, Grill DE, et al. 2002. Incidence and prognostic importance of acute renal failure after pertcutaneous coronary intervention. Circulation:2259 -2264. Thakar CV, Arrigain S, Worley S, Yared, JP, and Paganini EP. 2005. A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol 16:162-168. Fortescue EB, Bates DW, and Chertow GM. 2000. Predicting acute renal failure after coronary bypass surgery: cross-validation of two risk-stratification algorithms. Kidney Int 57:2594-2602. Bartholomew BA, Harjai KJ, Dukkipati S, Boura JA, Yerkey MW, Glazier S, Grines CL, and O’Neill WW. 2004. Impact of nephropathy after percutaneous coronary intervention and a method for risk stratification. Am J Cardiol 93:1515-1519. Mehran R, Aymong ED, Nikolsky E, et al. 2004. A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention: development and initial validation. J Am Coll Cardiol 44:1393-1399. Klahr S, and Miller SB. 1998. Acute oliguria. N Engl J Med 338:671-675. Sherman DS, Fish DN, and Teitelbaum I. 2003. Assessing renal function in cirrhotic patients: problems and pitfalls. Am J Kidney Dis 41:269-278. Bagshaw SM, and Bellomo R. 2007. The influence of volume management on outcome. Curr Opin Crit Care 13:541-548. Rind DM, Safran C, Phillips RS, Wang Q, Calkins DR, Delbanco TL, Bleich HL, and Slack WV. 1994. Effect of computer-based alerts on the treatment and outcomes of hospitalized patients. Arch Intern Med 154:1511-1517. Trivedi HS, Moore H, Nasr S, Aggarwal K, Agrawal A, Goel P, and Hewett J. 2003. A randomized prospective trial to assess the role of saline hydration on the development of contrast nephrotoxicity. Nephron Clin Pract 93:C29-34. Merten GJ, Burgess WP, Gray LV et al. 2004. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. Jama 291:2328-2334. Waikar SS, and Chertow GM. 2000. Crystalloids versus colloids for resuscitation in shock. Curr Opin Nephrol Hypertens 9:501504. Choi PT, Y G, Quinonez LG, et al. 1999. Crystalloids versus colloids in resuscitation: a systematic review. Crit Care Med:200 - 210. Schierhout G, and Roberts I. 1998. Fluid resuscitation with colloid or crystalloid solutions in critically ill patients: a systematic review of randomised trials. Bmj 316:961-964. Finfer S, Bellomo R, Boyce N, French J, Myburgh J, and Norton R. 2004. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 350:2247-2256. Bunn F, Lefebvre C, Li Wan Po A, Li L, Roberts I, and Schierhout G. 2000. Human albumin solution for resuscitation and volume expansion in critically ill patients. The Albumin Reviewers. Cochrane Database Syst Rev:CD001208. Alderson P, Schierhout G, Roberts I, and Bunn F. 2000. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev:CD000567. Berlauk JF, Abrams JH, Gilmour IJ, O’Connor SR, Knighton DR, and Cerra FB. 1991. Preoperative optimization of cardiovascular hemodynamics improves outcome in peripheral vascular surgery. A prospective, randomized clinical trial. Ann Surg 214:289-297; discussion 298-289.

02. Drug-associated acute kidney injury in the intensive care unit

104. Gattinoni L, Brazzi L, Pelosi P, Latini R, Tognoni G, Pesenti A, and Fumagalli R. 1995. A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 333:1025-1032. 105. Myburgh JA. 2006. An appraisal of selection and use of catecholamines in septic shock - old becomes new again. Crit Care Resusc 8:353-360. 106. Martin C, Viviand X, Leone M, and Thirion X. 2000. Effect of norepinephrine on the outcome of septic shock. Crit Care Med 28:27582765. 107. Bellomo R, and Giantomasso DD. 2001. Noradrenaline and the kidney: friends or foes? Crit Care 5:294-298. 108. Meine TJ, and Washam JB. 2004. N-acetylcysteine to prevent contrast nephropathy. Am Heart J 147:440-441. 109. Birck R, Krzossok S, Markowetz F, Schnulle P, van der Woude FJ, and Braun C. 2003. Acetylcysteine for prevention of contrast nephropathy: meta-analysis. Lancet 362:598-603. 110. Burns KE, Chu MW, Novick RJ, Fox SA, Gallo K, Martin CM, Stitt LW, Heidenheim AP, Myers ML, and Moist L. 2005. Perioperative N-acetylcysteine to prevent renal dysfunction in high-risk patients undergoing cabg surgery: a randomized controlled trial. Jama 294:342-350. 111. Fung JW, Szeto CC, Chan WW, et al. 2004. Effect of N-acetylcysteine for prevention of contrast nephropathy in patients with moderate to severe renal insufficiency: a randomized trial. Am J Kidney Dis 43:801-808. 112. Kellum JA, and M Decke, J. 2001 Aug. Use of dopamine in acute renal failure: a meta-analysis. Crit Care Med 29:1526-1531. 113. Landoni G, Biondi-Zoccai GG, Tumlin JA, Bove T, De Luca M, Calabro MG, Ranucci M, and Zangrillo A. 2007. Beneficial impact of fenoldopam in critically ill patients with or at risk for acute renal failure: a meta-analysis of randomized clinical trials. Am J Kidney Dis 49:56-68. 114. Allgren RL, Marbury TC, Rahman SN, et al. 1997. Anaritide in acute tubular necrosis. N.Engl.J.Med. 336:828-834. 115. Kulka PJ, Tryba M, and Zenz M. 1996. Preoperative alpha2-adrenergic receptor agonists prevent the deterioration of renal function after cardiac surgery: results of a randomized, controlled trial. Crit Care Med 24:947-952. 116. Amano J, Suzuki A, Sunamori M, and Tofukuji M. 1995. Effect of calcium antagonist diltiazem on renal function in open heart surgery. Chest 107:1260-1265. 117. Weinstein JM, Heyman S, and Brezis M. 1992. Potential deleterious effect of furosemide in radiocontrast nephropathy. Nephron 62:413-415. 118. Duke GJ, Briedis JH, and Weaver RA. 1994. Renal support in critically ill patients: low-dose dopamine or low-dose dobutamine? Crit Care Med 22:1919-1925. 119. Hirschberg R, Kopple J, Lipsett P, et al. 1999. Multicenter clinical trial of recombinant human insulin-like growth factor I in patients with acute renal failure. Kidney Int 55:2423-2432. 120. Takala J, Ruokonen E, Webster NR, Nielsen MS, Zandstra DF, Vundelinckx G, and Hinds CJ. 1999. Increased mortality associated with growth hormone treatment in critically ill adults. N Engl J Med 341:785-792. 121. Ix JH, McCulloch CE, and Chertow GM. 2004. Theophylline for the prevention of radiocontrast nephropathy: a meta-analysis. Nephrol Dial Transplant 19:2747-2753. 122. Marenzi G, Marana I, Lauri G, Assanelli E, Grazi M, Campodonico J, Trabattoni D, Fabbiocchi F, Montorsi P, and Bartorelli AL. 2003. The prevention of radiocontrast-agent-induced nephropathy by hemofiltration. N Engl J Med 349:1333-1340. 123. Spargias K, Alexopoulos E, Kyrzopoulos S, et al. 2004. Ascorbic acid prevents contrast-mediated nephropathy in patients with renal dysfunction undergoing coronary angiography or intervention. Circulation 110:2837-2842. 124. Liss P, Persson PB, Hansell P, and Lagerqvist B. 2006. Renal failure in 57 925 patients undergoing coronary procedures using isoosmolar or low-osmolar contrast media. Kidney Int 70:1811-1817. 125. Humes HD, Weinberg JM, and Knauss TC. 1982. Clinical and pathophysiologic aspects of aminoglycoside nephrotoxicity. Am J Kidney Dis 2:5-29. 126. Rampello E, Fricia T, and Malaguarnera M. 2006. The management of tumor lysis syndrome. Nat Clin Pract Oncol 3:438-447. 127. Megarbane B, Borron SW, and Baud FJ. 2005. Current recommendations for treatment of severe toxic alcohol poisonings. Intensive Care Med 31:189-195. 128. Ron D, Taitelman U, Michaelson M, Bar-Joseph G, Bursztein S, and Better OS. 1984. Prevention of acute renal failure in traumatic rhabdomyolysis. Arch Intern Med 144:277-280. 129. Homsi E, Barreiro MF, Orlando JM, and Higa EM. 1997. Prophylaxis of acute renal failure in patients with rhabdomyolysis. Ren Fail 19:283-288. 130. Olyaei AJ, de Mattos AM, and Bennett WM. 1999. Immunosuppressant-induced nephropathy: pathophysiology, incidence and management. Drug Saf 21:471-488. 131. Ladefoged SD, and Andersen CB. 1994. Calcium channel blockers in kidney transplantation. Clin Transplant 8:128-133. 132. Deray G. 2002 Feb. Amphotericin B nephrotoxicity. J Antimicrob Chemother 49 Suppl 1:37-41.

41

03

Renal handling of drugs and xenobiotics Marc E. DE BROE1 & Françoise ROCH-RAMEL2 † 1University 2Université

of Antwerp, Belgium de Lausanne, Switzerland

Introduction ____________________________________________________________ 43 Glomerular filtration ______________________________________________________ 46 Renal tubular reabsorption_________________________________________________ 46 Reabsorption by simple diffusion Reabsorption by facilitated mechanisms Endocytosis

46 48 48

Renal tubular secretion of drugs/xenobiotics __________________________________ 49 Transport mechanisms for tubular secretion of organic anions Tubular transport of organic cations ABC transporter family Interactions of xenobiotics/drugs for secretion Interactions between organic anion and organic cation secretion

50 55 59 61 62

Metabolism of drugs/xenobiotics in the kidney ________________________________ 63 References ______________________________________________________________ 64

Introduction

P

harmacology and clinical pharmacology define the desirable and undesirable effects of drugs and xenobiotics whereas pharmacokinetics defines the various processes that are involved in absorption - distribution - elimination of these agents. Needless to say that the former may strongly influence the latter. The kidney and the liver have complementary functions in the elimination of drugs and xenobiotics. Lipophilic non-ionic substances of molecular weight higher than 300-500 dalton and highly bound to proteins appear to be eliminated by the liver, while the

kidney prefers hydrophilic substances of molecular weight smaller than approximately 500 daltons. Metabolism occurs predominantly in the liver, transforming the original substance into more polar and more hydrophilic metabolites, which became dependent on the kidney for elimination. Consequently, the majority of all drugs and xenobiotics in one way or another have to pass through the kidney. In addition to this important “gateway” function of substances, which are not always without side-effects, the kidney itself is particularly sensitive to drugs and xenobiotics. This susceptibility of the kidney to nephrotoxic injury has several reasons (Table 1). Renal blood flow

DE BROE & ROCH-RAMEL

44

Cortex

Outer Stripe

Inner Stripe

Outer Medulla

Ray

in the medullary interstitium can reach values several times higher than tissues elsewhere in the body. Finally during the process of renal excretion, a particular drug may undergo bioactivation resulting in reactive metabolites [2]. The kidney possesses several mechanisms for the renal handling/excretion of drugs and xenobiotics.

Medullary

(25% of the resting cardiac output) exceeds 1000 ml/ Table 1. Vulnerability of the kidney. min = 3.5 ml/g of renal tissue/min. Compared to the Important blood flow (1/4 cardiac output) majority of other tissues, except the brain, this results High metabolic activity Largest endothelial surface by weight in a fifty times higher rate of drug delivery. Multiple enzyme systems The kidney has the greatest endothelial surface Transcellular transport per gram of tissue and possesses the highest capillary Concentration of substances hydrostatic pressure favoring trapping of circulating Protein unbinding antigen and in situ immune complexes formation. Tubular transport and other renal metabolic processes Filtration utilize considerable oxygen and are susceptible to the action of metabolic inhibitors. It is worthwhile to note that the S3-segment Active reabsorption of the proximal tubule has the highest rate of oxygen delivery/oxygen consumption of all functional entities in the body [1]. The kidney Intraluminal metabolisation is the only place where highly protein bound drugs dissociate, traverse the tubular cells and either accumulate within the proximal tubular epithelium Pinocytosis/Endocytosis and/or reach the tubular lumen. An abundance of tubular epithelial enzymes involved in the tubular transport systems can be blocked, particularly in view of the highly Active secretion of anions concentrated solutes in the tubular & cations fluid that may reach urinary/plasma concentration ratios exceeding 1000 in some cases. In the distal part of the nephron, Passive non-ionized urine is concentrated and the likelihood of back diffusion crystalline precipitation increases substantially, particularly if urinary pH favors decreased solubility. As the urinary concentrating process also involves the counRenal metabolism ter-current mechanism, solute concentrations

Inner Medulla

Figure 1. Schematic representation and main localisation along the nephron of the various patterns of drug and xenobiotic handling by the kidney.

03. Renal handling of drugs and xenobiotics

They are listed in Figure 1 and each of them will be briefly discussed in this chapter. Numerous, if not the majority of drugs and xenobiotics, are handled-eliminated at least partly by the kidney. For their elimination by the kidney they use one, or in most cases, two or even more mechanisms (Figure 2). In addition many other polar metabolites are formed by metabolism or conjugates by the liver, which are then excreted by the kidney. The use of various in vitro and in vivo techniques as models in studying drug transport in the kidney and/or renal toxicology is well documented in the literature [3-7]. Each approach possesses its own advantages and disadvantages and all have demonstrated their usefulness and application in renal pharmacokinetics/toxicology. A representative listing of these models, summarizing their most relevant characteristics, is presented in Table 2 [6, 8].

Glomerular filtration Active tubular secretion Passive tubular reabsorption Active reabsorption (carrier-mediated pinocytosis) Intraluminal metabolisation Renal metabolism Osmotic diuretic mannitol Aminoglycosides Organic anions: p-Aminohippurate or benzylpenicilline Phenobarbital Phenylbutazone Oxipurinol Organic cations: Tetraethylammonium Quinine Methadone Paracetamol Nitrofurantoin Choline A few vitamins Imipenem Probenecid

Figure 2. Most drugs and xenobiotics have a renal handling consisting in more than one pattern.

Table 2. In vitro methods for studying drug transport in the kidney Method Stop-flow Isolated perfused kidney Kidney slices

Micropuncture

Advantages Easy to determine net direction of transport. Morphologically identical to kidney in vivo. Can monitor renal function. Easy technique. Good control of experimental conditions without concern for secondary effects due to hemodynamic changes. Can study transepithelial transport in surface portions of proximal and distal tubules. Can study several segments of the nephron

Time resolved twophoton microscopy [8a,b] Proximal tubular Relatively homogenous cell population. suspensions Cultures

Cell lines Primary cultures Vesicles

Long-term storage. Precise control of growing environment. Cell population is relatively homogenous. Cells on filters permit study of bidirectional transport. Easily obtained and subcultured. Closely related to fresh tissue. Origin identified. Transport in apical and basolateral membranes can be studied separately. No metabolism. No intracellular sequestration.

Disadvantages No precise anatomical localization. Short term use. In the process of degeneration. Functional status of tubular lumen not clear. Tissue not homogenous and contains nontubular elements. Diffusion barrier for substrates to nephrons beneath the cut surface. Cannot study deep segments. Expensive methodology Contribution of luminal uptake is dependent on luminal openings and can vary. Short term use. Dedifferentiation. Sterile conditions for culture.

Origin ill-defined. Important dedifferentiation. More difficult to prepare and maintain. Membrane isolation may alter physiological function. Must correct for non-mediated transport.

Adapted from Williams & Rush [6] and Brater et al [8].

45

DE BROE & ROCH-RAMEL

The maturation of renal drug elimination systems occurs at variable rates and may be influenced by a number of factors, including pre-or postnatal exposure to drugs. In addition, the mechanisms of drug uptake and storage in renal tubular cells are subject to maturational changes that may lead to age-related differences in intrarenal accumulation of a drug [8c].

Glomerular filtration One fifth of the renal plasma flow (± 600 ml/min) is filtered at the glomeruli. This filtered fraction indicates that glomerular filtration can account for the plasma clearing of as much as 20% of a non-protein bound substance during one passage through the kidney. The determinants of a drug/xenobiotic to be filtered are protein binding, molecular size and charge, glomerular integrity and the number of filtering nephrons. Glomerular pores (± 75 A in diameter) allow passage of molecules up to the molecular weight of approximately 60, 000 dalton. The vast majority of drugs/xenobiotics are approximately two orders of magnitude smaller than this. For many drugs however, protein binding restricts filtration so that only the unbound fraction can be filtered (e.g. furosemide 95% and NSAID 98% bound to albumin), and in many cases depend on active tubular secretion for renal elimination. Drugs can bind to several serum proteins, however, by far the most important being albumin, followed by a α1-acid glycoprotein, an acute phase reactant. Acidic compounds preferentially bind to albumin [9] whereas for basic compounds binding to α1-acid glycoprotein is more important [10]. Nephrotic syndrome induces two important changes concerning protein binding. Hypoproteinemia causes a decrease in protein binding and the integrity of the glomerulus as a sieve is disrupted in this clinical condition. Drugs and xenobiotics can be carried with albumin into the urine enhancing renal elimination. Hypoproteinemia, however, induces simultaneously an increase in the distribution volume of numerous substances thus lowering their availability for filtration. The overall result on renal elimination being almost unperceptible. Total plasma clearance and distribution volume of furosemide were much larger in analbuminemic rats compared to normals, whereas the urinary excretion was significantly lower. Injecting the albumin/furo46

semide complex markedly decreased the drug distribution volume, promoted diuresis in analbuminemic rats, in contrast to furosemide alone. Injection of the furosemide/albumin complex to furosemide resistant hypoalbuminemic nephrotic patients increased the urine volume. Another factor that may contribute to diuretic resistance in nephrotic patients is the presence of filtered albumin within the tubule lumen. Even when adequate amounts of diuretic are delivered to and secreted by the proximal tubule, much of the diuretic that reaches the lumen in a nephrotic patient will bind to filtered albumin; the protein/diuretic complex may not be effective in inhibiting the Na-K-2Cl pathway [11-14]. In rats with nephrotic syndrome, inhibitors of protein binding (warfarin and sulphisoxazole) restore the potency of furosemide [14]. Uncharged hydrophilic substances prefer glomerular filtration for their renal handling/elimination in contrast to the many ionized organic substances handled by additional nephron mechanisms, such as tubular secretion (e.g. penicillin). Drugs and xenobiotics that have glomerular filtration as their major way of renal elimination will accumulate rapidly during acute or more chronic declines of glomerular filtration. If in addition the therapeutic/ toxic window is narrow, the accumulation will result very quickly in toxic effects (e.g. aminoglycosides).

Renal tubular reabsorption Reabsorption of weak acid and bases is generally passive, but in a few cases reabsorption can occur via facilitated reabsorption by carrier proteins or by endocytosis.

Reabsorption by simple diffusion Passive reabsorption is driven by the progressive reabsorption of tubular fluid along the nephron. To penetrate the membranes of the tubular epithelium, whose main constituents are lipids, compounds should be liposoluble. As ionized compounds are in general hydrophilic, only the undissociated molecules of weak bases and acids will be rapidly reabsorbed by simple diffusion [15]. Consequently determinants for the rate of reabsorption are the pKa of the organic acid or base, the urinary pH, and the liposolubility of the undissociated base or acid. Another important determinant

03. Renal handling of drugs and xenobiotics

Table 3. Drugs and xenobiotics with clinically important urine pH-dependent elimination. Weak acids: increased excretion at luminal pH > 7 Acetazolamide Chlorthiazide Methotrexate (?) Penicillin G Phenobarbital Phenylbutazone Salicylates Sulfonamide derivatives

Weak bases: increased excretion at luminal pH < 5 Amitriptyline Amphetamine Chloroquine Ephedrine Imipramine Phencyclidine (Angel Dust) Quinine Tricyclic antidepressants

is the contact time of the solute with the epithelium. In antidiuresis, this time is prolonged compared to diuresis, and thus passive reabsorption is increased along the whole nephron, as observed for salicylate (Table 3) [16]. Alkaline diuresis will favor the excretion of weak acids (anions) such as salicylate or phenobarbital. Indeed, the more the drug is ionized, the more it is trapped in the tubular lumen and consequently is not reabsorbed, hence eliminated in the urine. This mechanism can play a role in the treatment of severe intoxications. The reverse being true for weak bases (cations) such as methadone. Acidification of the urine facilitates the reabsorption of weak acids and will re-

100% ionized = dissociated

EFFECT OF URINARY pH ON DISSOCIATION OF: ORGANIC WEAK ACIDS

ORGANIC WEAK BASES

e.g. Methadone:

pKa 3.0

pKa 8.3

normal plasma pH value

e.g. Salicylate:

normal plasma pH value

50% 100% non-ionized = undissociated

tard the reabsorption of weak bases. The magnitude of the effect obtained on organic acid excretion by urinary alkalinization will be smaller than that which may be achieved for organic cation excretion by urine acidification. Indeed, the achievable urinary proton concentration is up to three orders of magnitude higher than plasma concentration (pH 4.5 versus 7.4) (Figure 3). At the other end of the pH scale urinary proton concentration cannot exceed a value of one order of magnitude lower than plasma concentration (pH 8.5 versus 7.4) (Figure 3). The effect of urinary pH on the elimination of amphetamine may be better known to abusers of these drugs or particular sport trainers than to clinicians. Since amphetamine is a weak base, alkalizing

3

5

7

9

11

3

5

7

9

11

pH range of urinary pH range of urinary pH 4.5 8.5 4.5 8.5 Ionized drug is trapped in the tubular fluid, not reabsorbed, hence eliminated in the urine.

Figure 3. Effect of urinary pH on dissociation of organic weak acids and organic weak bases. Lipid soluble compounds cross the cellular membranes preferentially in their undissociated form. The ionized form favours trapping and subsequent elimination by the kidney. 47

DE BROE & ROCH-RAMEL

the urine increases the non-ionized amount favoring reabsorption. Amphetamine abusers regularly ingest baking soda to prolong the “high”. Therapeutically, it would be important to acidify the urine of a patient with an overdose of amphetamines or phencyclidine (angel dust) [17]. However, one has to take into account that the extent to which a change in urinary pH alters the rate of total body clearance depends on the contribution of renal clearance to the total body clearance. Weak acids like phenytoin and warfarin which are susceptible to a pH dependent elimination in the urine do not see a substantial effect of change in urinary pH on their total elimination since hepatic metabolism is the more important metabolic pathway [18]. There are examples of weak acids reabsorbed by simple nonionic diffusion which urinary excretion is not influenced by changes in urine pH. It is the case if the pKa is above or close to the upper limit of urine pH, as it is the case for barbital (pKa = 7.8), and a few other barbiturates. Also, if the pKa value is very low, such as it is the case for 2-nitroprobenecid (pKa=1.3), the acid remains mainly unionized in the physiological range of urine pH [15], and its excretion remains independent of tubular urine pH.

Reabsorption by facilitated mechanisms A certain number of drugs and xenobiotics are reabsorbed by facilitated mechanisms. Some organic anions are transported at the apical membrane of proximal tubule by a sodium-cotransport mechanism. It is the case of vitamins, such as ascorbic acid, biotin, panthotenate, nicotinate, and pyridoxine (and its analogues) [19]. Pyrazinoate, a metabolite of pyrazinamide is reabsorbed by a sodium cotransport mechanism [20, 21], as well as by an anion-exchanger [20], which is implicated also in the reabsorption of urate. Oxypurinol, the metabolite of allopurinol might also be reabsorbed by the urate reabsorbing mechanism [22]. M-hydroxybenzoate and morphine-glucuronides are other organic anions reabsorbed by facilitated mechanisms that have yet to be identified [23, 24]. Little is known on the facilitated reabsorption of organic cations. The reabsorption of choline involved a sensitive pathway at the apical membrane [25]. Several peptide-like drugs such as β-lactam antibiotics (ceftibuten, cyclacillin) are substrates of the peptide transporters localized in the brush-border membrane, 48

and are taken up into proximal cells. The peptide transporters mediate an electrogenic H+-coupled cotransport of di- and tri-peptides, which is driven by the proton gradient and the negative transmembrane potential difference [26]. Two homologous peptide transporters have been identified by molecular cloning methods, PEPT1 and PEPT2. In the kidney, PEPT1 was localized to the brush-border membrane of S1 segments of proximal tubule, whereas PEPT2 was localized to the brush-border membrane of S3 segments [27]. Affinity of anionic cephalosporin without a-amino group (ceftibuten) and cyclacillin (aminopenicillin) is greater than that of aminocephalosporin, such as cephalexin, cefadroxil, cephradine. Because of their low affinity for the anionic cephalosporins PEPT1 and PEPT2, should not play a major role in cellular accumulation and potential toxicity of these cephalosporins when given at therapeutic doses. The peptide transporters might, however, be involved in the reabsorption of the nephrotoxin ochratoxin A [28]. The anticancer drug bestatin, and valacyclovir, a non-peptide antiviral agent, are also substrates for the peptide transporters [29]. The angiotensin-converting enzyme inhibitors, quinapril and enalapril, have affinity for the peptide transporters, however it is not known whether they are transported.

Endocytosis One of the mechanisms of active reabsorption is endocytosis. Fluid phase endocytosis consists of the incorporation of fluid and solutes in vesicles formed at the base of the brush border membrane of the proximal tubular cells (Figure 1). A more efficient absorptive endocytosis involves first binding of a drug, such as the cationic aminoglycoside and/or may be cadmium [30, 31], to a carrier (phosphatidylinositol) located in the luminal membrane of the wall of the pinocytotic vesicle occurs followed by endocytosis and lysosomal fusion [32, 33]. Endocytosis is a normal mechanism for protein and insulin reabsorption at the proximal tubule of the kidney. A considerable amount of insulin (50%) is metabolized by the kidney, which may account, at least in part, for the decreased insulin requirement that occurs in diabetic patients with decreased renal function. Furthermore, this uptake process allows highly hydrophilic lipid insoluble drugs such as aminoglyco-

03. Renal handling of drugs and xenobiotics

sides to enter a particular intracellular compartment (lysosomes) without crossing a membrane.

Renal tubular secretion of drugs/xenobiotics Most ionic xenobiotics are secreted by two transport mechanisms, one responsible for organic ion (or “organic acids”) secretion (Table 5), the other for organic cation (or “organic bases”) secretion (Table 6) [34, 35, 35a]. Despite considerable advances in the understanding of basic transport pathways and mechanisms involved in the tubular secretion of organic compounds, there is still relatively little information on the regulation of this transport [35b]. The first step of secretion, transport across the basolateral membrane, of each of the two general mechanisms is performed by several subsystems which may correspond to different carrier molecules, for which substrates of rather unspecific molecular structure may have various affinities [19, 36, 37]. The molecular structure of several isoforms of these transporters has been identified by expression cloning. They are members of a newly identified transporter family, the organic ion transporters, which comprises OAT (organic anion transporter), and OCT (organic cation transporter) isoforms [29, 38, 39]. Our understanding of the organic ion secretory mechanisms derives essentially from investigations on a few transported compounds that are considered representative of other secreted organic ions. For organic anions the classical substrate is p-aminohippurate (PAH) whereas for organic cations classical substrates are tetraethylammonium (TEA) and N1-methylnicotinamide Both classical transport systems are located exclusively in the proximal tubule of the nephron. Several techniques such as visual observations, stop-flow experiments, tubular micropuncture, in vivo and in vitro tubular microperfusions have demonstrated this particular transport capacity of the proximal tubular segment of the nephron [19]. Secretion entails an increase of proximal cell concentration of transported substrates that may become higher than in interstitium, and in some case may result in nephrotoxicity [40-42]. Secretion is not uniform along the proximal tubule, and may differ between superficial and juxtamedullary nephrons. This heterogeneity of secretion, varying between species and substrates, might reflect different

Table 4. Transport and renal drug elimination. Organic anion transport system (Table 5) • organic anion transport family • organic anion transporting polypeptides family Organic cation transport system (Table 6) ABC transporter family (Table 8) • MRP’s: multi-drug resistance associated proteins family • MDR1/P-glycoprotein Breast cancer resistance protein 1 [54a] Multidrug and toxin extrusion transporters family (MATE) [49a,b]

densities of carrier molecules along the tubule [19]. Since the number of carrier molecules is limited, secretion is saturable and subject to competition between substrates. Such competition may thus result in drug interactions some of them being of clinical relevance (see below) [43]. The transport mechanisms of the organic ion transport systems have been characterized at both membrane sides of proximal tubule, mainly by studies in brush-border and basolateral membranes purified from homogenates of renal cortex. Since a detailed review and a critical discussion of the present knowledge in this field was published by Pritchard [44], only the main conclusions are summarized here. Beside these classical, long recognized secretory transport systems, other transport mechanisms are involved in the renal excretion of xenobiotics [45]. They are the basolateral oxalate/sulphate exchanger and the basolateral sodium-dicarboxylate transport system [37]. These transporters were identified by expression cloning, and named SAT1 [46] and NaDC3 [47], respectively. A number of other transporters has been cloned and identified in the renal brush border, but their functional role in the kidney has yet to be defined [45]. Among them are the multidrug resistance-associated proteins MRP and MDR/P-glycoprotein, which are ATP dependent primary active transporters for organic anions and organic cations respectively [48, 49], and recently the MATE family [49a,b]. They stimulate the active efflux from cell to lumen, of various organic ions. OATP1 is another transporter that mediates the apical transport of steroid conjugates and sulphobromophtalein [50], whereas OAT-K1 and OATK2 mediate methotrexate and folate transport [51, 52]. OCTN1 [53] and OCTN2 [54] are apical multispecific organic cation transporters (Table 4). 49

DE BROE & ROCH-RAMEL

Transport mechanisms for tubular secretion of organic anions The transport mechanisms of organic anions have been characterized mainly for PAH (Table 5). Owing to electro-negativity of the cell interior, resulting from Na-K-ATPase activity, a transfer of

negatively charged molecules into cells occurs generally against an electrochemical gradient and requires energy (“active transport”). In contrast, efflux from cell to lumen takes place along a favorable electrochemical gradient and does not necessitate a direct energy supply. Large cell/interstitium concentration gradients, up to 40 in isolated perfused rabbit proximal tubules, can

Table 5. Organic anion transporter (OAT) and organic anion transporting polypeptide (OATP) families (from [34], with permission). Name OAT1 (SLC22A6) Oat1 (Slc22a6) OAT2 (SLC22A7) Oat2 (Slc22a7) OAT3 (SLC22A8)

Human Rat

Human Rat

Human

Oat3 Rat (Slc22a8) OAT4 Human (SLC22A11) Oatpl Rat (Slc21a1)

Oatp3 (SIc21a7)

Rat

OATP-A (SLC21A3)

Human

OATP-B (SLC21A9)

Human

OATP-D Human (SLC21A11) OATP-E Human (SLC21A12) OAT-K1 Rat (Slc21a4) Oat-k2 (Slc21a4)

Rat

Substrates

Inhibitors

PAH, α-KG Drugs: anti-HIV drugs, MTX PAH, α-KG, cAMP, cGMP, folate, ochratoxin A, PGE2, urate Drugs: beta-lactam antibiotics, anti-HIV drugs, MTX PAH, α-KG, CAMP, PGE2, PGF2α Drugs: AZT, MTX PAH, α-KG, PGE2, PGF2α Drugs: NSAIDs, AZT, MTX, zalcitabine

Probenecid, urate Drugs: β-lactam antibiotics, NSAIDs, diuretics Probenecid, glutarate Drugs: β-lactam antibiotics, NSAIDs, diuretics, antidiabetic agents Probenecid, BSP Drugs: β-lactam antibiotics, NSAIDs BSP, cholate Drugs: β-lactam antibiotics, NSAIDs, bumetanide, enalapril, glibenclamide, rifampicin, verapamil Probenecid, BSP, cholate, corticosterone, TEA Drugs: B-lactam antibiotics, diuretics, NSAIDs, quinidine

PAH, CAMP, estrone-S, E217βG, ochratoxin A, PGE2, urate Drugs: AZT, cimetidine, MTX, salicylate PAH, estrone-S, ochratoxin A Drugs: benzylpenicillin, cimetidine PAH, estrone-S, ochratoxin A Drugs: AZT, cimetidine, MTX LTC4, BSP, DNP-SG, aldosterone, cortisol, E217βG, estrone-S, ochratoxin A, thyroid hormones, bile acids Drugs: BQ123, dexamethasone, cardiac glycosides, enalapril. fexofenadine. pravastatin PGE2, DHEA-S, E217βG, estrone-S, LTC4, BSP, thyroid hormones, bile acids Drugs: BQ123, cardiac glycosides, fexofenadine, rocuronium BSP, DHEA-S, E217βG, estrone-S, PGE2, thyroid hormones, ochratoxin A, bile acids Drugs: BQ123, CRC220, chlorambucil, fexofenadine, ouabain, rocuronium Estrone-S, PGE2 Drug: benzylpenicillin Human Estrone-S, PGE2 Drug: benzylpenicillin Estrone-S, PGE2, taurocholate, thyroid hormones Drug: benzylpenicillin DHEA-S, E217βG, estrone-S, folate, ochratoxin A, taurocholate, thyroid hormones Drugs: AZT, MTX DHEA-S, E217βG, estrone-S, PGE2, folate, ochratoxin A, taurocholate, thyroid hormones Drugs: AZT, digoxin, MTX

Probenecid, BSP, cholate, taurocholate Drugs: β-lactam antibiotics, diuretics, AZT, MTX, quinidine Probenecid, BSP, corticosterone Drugs: β-lactam antibiotics, diuretics, NSAIDs Probenecid, steroids Drugs: atorvastatin, furosemide, lovastatin, simvastatin

6’,7’-Dihydroxybergamottin, furanocoumarins in grapefruit juice

Leu-Enkephalin Drugs: anti-HIV drugs, dexamethasone, erythromycin, lovastatin, naloxone, naltrindole, quinidine, verapamil

BSP Probenecid, PAH, BSP, folate Drugs: NSAIDs, furosemide, valproate Probenecid, PAH, BSP, 17β-estradiol Drugs: cardiac glycosides, benzylpenicillin, dexamethasone, furosemide, indomethacin, levofloxacin, prednisolone, valproate

Abbreviations: α -KG, α –ketoglutarate; AZT, azidothymidine; BQ123, cyclo [Trp-Asp-Pro-Val-Leu]; BSP, bromosulfophthalein; DHEA-S, dehydroepiandrosterone-sulfate; DNP-SG, S-(dinitropheny1)-glutathione; E217 β G, estradiol-17 β -D-glucuronide; estrone-S, estrone sulfate; LTC4, leukotriene C4; MTX, methotrexate; PAH, p-aminohippurate; PGE2, PGF2, prostaglandin E2, F2 α; TEA, tetraethylammonium.

50

03. Renal handling of drugs and xenobiotics

capillary BM PAH

pig rabbit

1

PAH

1 rat

tubular cell PAH alpha-ketoglutarate-2

PAH alpha-ketoglutarate-2

XPAH

human bovine

1

PAH Alpha-ketoglutarate-2

lumen

5

PAH

2 Na+ alpha-keto- 3 glutarate-2

5

4

Na+ alphaketoglutarate-2

Figure 4. Tubular transport of the organic anion PAH.

PAH

6

7

3Na+

PAH alphaketoglutarate-2

build up during secretion [44]. However, as only part of the PAH accumulated in the cytoplasm might be free, the concentration gradient of diffusible PAH, between peritubular interstitium and cell might be lower than estimated from total concentration. There exists strong evidence that part of PAH might be sequestered in cytoplasmic organelles [55, 56]. As will be described below, the basolateral transport mechanism, which is the active step in PAH secretion, is identical in all animal species investigated so far, whereas the apical mechanism, which does not require energy, differs between animal species. The left side of Figure 4 shows a section of proximal tubule and the PAH transport mechanisms identified in different mammalian species. On the right side of Figure 4, the mechanisms shown are indirectly implicated in PAH transport, and are common to all species. The transport of PAH in basolateral membranes occurs through an exchange with α-ketoglutarate (mechanism 1). The molecular structure of the PAH/αketoglutarate exchanger has been identified. Different isoforms of this transport protein, denominated OAT1 (organic anion transporter), have been defined [38, 39] in different animal species. The specificity of the PAH/ anion exchanger (OAT1) is high for α-ketoglutarate, the only natural substrate showing affinity for the antiport. This PAH basolateral transport mechanism has been

found in all mammals (including humans) and lower vertebrates investigated so far. The energy source for PAH transport in intact cells is provided by the low intracellular sodium concentration achieved by NaK-ATPase activity (mechanism 2, right part of Figure 4), which expels 3 Na+ from the cell in exchange of 2 K+. This exchange creates the electronegativity of the cell. The transmembrane sodium gradient serves as energy source to drive α-ketoglutarate intracellularly from peritubular interstitium (mechanism 3) as well as from tubular lumen (mechanism 4), since both basal and apical membranes possess an α-ketoglutaratesodium cotransport mechanism. They were identified by molecular cloning and named NaDC3 [47] and NaDC1 [57], respectively. Furthermore, α-ketoglutarate can also be produced by cell metabolism. In the dog, the intracellular concentration of α-ketoglutarate from transport and cell metabolism is about 5-10 times higher than in plasma [58], and is thus not rate limiting for PAH transport. The α-ketoglutarate/PAH exchange at the basolateral membrane (mechanism 1, left side of Figure 4) allows PAH to concentrate intracellularly by a tertiary active transport. Intracellular traffic of secreted anions appears more complex than originally thought, and might proceed through accumulation into cell organelles (black dots on the scheme), implying high local concentrations of the substrate, and involvement of a microtubular network [59, 60]. While the basolateral membrane transport system appears ubiquitous, the mechanisms involved in PAH efflux from cell to lumen differ between animal species [44]. A voltage controlled pathway (mechanism 5) and/or anion exchanger (mechanism 6 and 7) might be implicated. The former, which is present in rabbits, pigs and rats (data are lacking for dogs) [44], is facilitated transport mechanism that, because of electronegativity of the cell, should drive PAH efflux 51

DE BROE & ROCH-RAMEL

from cell to lumen. An anion exchanger, on the other hand, has been identified in rats (mechanism 6), and in dogs (not shown in the figure). This transporter accepts inorganic anions (Cl-, HCO3-, OH-), and several organic anions (X- = lactate, succinate, α-ketoglutarate, etc.) and also uric acid [44]. The respective role of these two transport mechanisms observed in rats is not known. Indeed, in rat proximal tubule in situ, Ulrich did not observe any effect of changes in membrane potential on PAH cellular efflux, and did not observed any stimulation of PAH efflux by anion exchange [61]. The rat and dog anion exchanger which has affinity for urate, is most probably involved in urate reabsorption and might decrease the secretion of PAH by recycling part of it into the cell. In humans, PAH is not transported by a voltage-controlled pathway [62], nor by the anion exchanger that has urate as substrate [62]. The apical transport of PAH is by a PAH/anion exchanger, which accepts only α-ketoglutarate (mechanism 7), as it is the case in basolateral membranes [63]. The same mechanism has been found in bovine [64]. Substrate requirements for the “PAH transport mechanism”. Many studies have been devoted to the characterization of the properties of substrates for the so-called “organic anion secretory mechanism”, by measuring the ability of compounds to compete for PAH transport across the basolateral membrane, the first step in secretion. In particular Ullrich et al. [61], investigated the interaction of all kinds of aliphatic and aromatic molecules on PAH influx in rat proximal tubule cells in situ. The findings have been reviewed by the authors, and will not be detailed here. The main findings [37], which corroborate older data [36], are that the molecular structure of the transport substrates is rather unspecific, and that substrate affinity depends on the acidity and hydrophobicity of the substrate [37]. These authors demonstrated that, unexpectedly, the ionization of the substrate is not a prerequisite for interaction with the transporter [65]. In general, most anionic xenobiotics are secreted by the PAH secretory mechanism (Table 4), and their secretion can be inhibited by probenecid. Substrates with high affinity are monovalent anions, containing a hydrophobic domain with a minimal length of about 4 A° (benzoyl derivatives). Ullrich et al. also characterized the substrate characteristics for the oxalate/sulphate transport mechanism, and the sodium-dicarboxylate cotransport mechanism, in 52

situ in rat proximal tubules [37]. The oxalate/sulfate exchanger (SAT1) and the sodium-dicarboxylate transporter (NaDC3) have a much narrower substrate specificity than the PAH transporter (OAT1), which is the major basolateral organic anion transporter, and represents the classical “organic anion secretory mechanisms”. Bisphosphonates [66], and in particular alendronate [67] might be secreted by SAT1. Fewer studies have been devoted to the characterization of substrate requirement for the apical PAH transport. The general substrate characteristics appear similar to that of PAH basolateral transport, i.e. hydrophobicity and acidity, and lack of molecular structure requirement [68, 69]. The molecular biology of the OAT transporter family Several isoforms of OAT1 (rat, human, mice OAT1) have been cloned and their functional properties examined in different cultured cell systems and Xenopus oocytes. Detailed molecular structures and functional characteristics have been recently reviewed [29, 38, 39] (Figure 5). OAT1 belongs to a subgroup of a newly identified transporter family, the organic ion transporter family, which comprises other OAT isoforms, OAT2, and OAT3. These proteins possess a common structural feature, i.e. 12 putative transmembrane domains, with large hydrophobic loops between the first and second, and the sixth and seventh transmembrane domains. The organic cation transporters, OCT and OCTN are members of the same organic ion transporter family [39]. Human and rat OAT1 has been localized exclusively to the basolateral membrane of S2 segments of proximal tubules, and when transfected in cell systems such as Xenopus oocytes or epithelial cultured cells, OAT1 has the ability to transport a wide variety of organic anions which are known to be secreted in vivo. Transport is through an α-ketoglutarate/organic anion exchanger, which is dependent on the presence of chloride in the extracellular medium [70]. These are the same requirements than for transport trough the basolateral membrane of proximal tubules. The rat, human, or flounder OAT1 isoforms were demonstrated to transport PGE2, cAMP, cGMP, α-ketoglutarate, estradiol 17β-D-glucuronide, nonsteroidal anti-inflammatory drugs (salicylate, acetylsalicylate, indomethacin, etc.), antiviral drugs (azidothymidine, acyclovir, etc.), diuretic (thiazide, bumetanide, ethacrynic acid, tienilic acid, and the nephrotoxin ochratoxin A [38, 39, 71-74].

03. Renal handling of drugs and xenobiotics

capillary BM Na+

tubular cell

lumen

metabolism

Na+ dicarboxylate

OAT1

+

+

OAT2

X– OAT-K1

ATP ADP

-3 mV

+ organic anion

OAT3 MRP1

Na- and alpha-ketoglutarate dependent PAH transporter

OAT1

OAT-K2 MRP2/cMOAT

-70 mV

0 mV

Na- and alpha-ketoglutarate independent PAH transporter

Figure 5. Mechanisms of organic anion transport in renal tubular cells. Cellular uptake of organic anions across basolateral membranes (BM) is mediated by OAT1 (1), which is an organic anion/dicarboxylate exchanger, and by OAT2 (2) and OAT3 (3). Anionic drug conjugates with glutathione may be extruded from cells into blood by MRPI (4). Exit of cellular organic anions across brush border membranes (BBM) is mediated by unidentified transmembrane potential-dependent organic anion transporter (5) and organic anion/anion (X–) exchanger (6). Bidirectional transport of hydrophobic anions such as methotrexate and folic acid in the brush-border membranes is mediated by OAT-K1 (7) OAT-K2 (8) may also participate in tubular reabsorption and/or secretion of hydrophobic anions such as bile acids, methotrexate. and prostaglandin E2. MRP2/ cMOAT (9) may contribute to tubular secretion of anionic conjugates of hydrophobic compounds. Adapted from [29].

The different OAT1 isoforms have some differences in substrate affinities, which might correspond to species differences in transport. For example, urate is transported by the rat rOAT1 [75], but not by the flounder fOAT1 [76] and the human hOAT1 [77]. In human, urate is not secreted by the PAH transport mechanism [78]. This observation gives support to the role of hOAT1 in PAH secretion. Methotrexate and PGE2 are transported by rat rOAT1 [75, 79], but they have no affinity for the human hOAT1 [80]. Probenecid is not transported by the rat and the flounder transporters despite its binding affinity [75, 79]. Because probenecid, PGE2 and methotrexate are secreted in human and rat, this lack of transport suggests that other OAT isoforms or transport proteins are involved in their secretion.

methotrexate and folate transporter methotrexate, folate and other hydrophobic anion transporter multidrug resistance transporter

Recently, OAT2 and OAT3, two OAT1 isoforms, have been identified. OAT2 mRNA is predominantly expressed in the liver, and weakly in the kidney. In contrast OAT3 mRNA is expressed in the kidney, as well as in the liver and the brain [80a]. The substrate spectrum of OAT2 and OAT3 is diverse like that of OAT1 [81], but in contrast to OAT1, transport is not dependent on α-ketoglutarate, and a concentration gradient is sufficient to allow transport. The nephron distribution, and the membrane localization of OAT2 and OAT3 have not been established. Rat rOAT3 mediates PAH transport [82] estrone sulfate, ochratoxin A. Substrates for human hOAT3 are still to be defined [77]. There exists clear evidences that OAT1 plays a major role in PAH and other organic anion secretion. It is localized in the basolateral membrane of proximal tubule, it has the transport characteristics of basolateral membrane transport, i.e. it is an organic anion/α-ketoglutarate exchanger, and transport is dependent on the presence of chloride in the extracellular medium. The recent observation that the expression of rat OAT1 is strongly increased at birth is compatible with the fact that the PAH secretory system develops post-natal [83]. Molecular identification of apical putative PAH transporters. NPT1 and MPR2 are two organic anion transporters which have been identified by molecular biology techniques, and which, in the kidney, were localized to the apical membrane of proximal tubule. They might be involved in organic anion secretion. 53

DE BROE & ROCH-RAMEL

Uchino et al. [84] cloned and characterized an apical PAH transporter isolated from human kidney, named NPT1. NPT1 was first identified as a low affinity sodium-dependent phosphate transporter, later it was characterized as an organic anion transporter. In human embryonic kidney cells transfected with NPT1, PAH, urate, benzyl penicillin, faropenem, estradiol-βglucuronide are transported, and PAH uptake can be inhibited by various organic anions. NPT1 does not function as an organic anion exchanger, and thus is not the PAH/organic anion exchanger observed in rat and dog brush border membrane vesicles. Rabbit NPT1, a homologous of human NPT1, was demonstrated to mediate electrogenic penicillin transport [85], thus, NPT1 might be the PAH voltage sensitive pathway observed in rat and rabbit brush-border membrane vesicles. Further studies should confirm this hypothesis. Another putative apical PAH transporter is the ATP-dependent export pump MRP2, a multidrug resistance protein isoform characterized by its apical localization in polarized cells such as hepatocytes [49]. In the kidney, MRP2 has been localized to the apical membrane of human and rat proximal tubule [86]. Substrates are anionic conjugates with glutathione (leukotriene C4) or glucuronide (estradiol-17 β-Dglucuronide), as well as non-conjugated substrates such as probenecid, sulfinpyrazone, indomethacin, furosemide and penicillin [87]. Recently two research groups demonstrated that PAH and ochratoxin are transported substrates [88, 89]. MRP2 thus might contribute to the efflux of PAH and other organic anions at the apical membrane. MPR1 another member of the ATP-dependent export pumps that is associated with multidrug resistance in cancer cell and is expressed in a few renal tubular segments, but not in the proximal tubule [90]. Both NPT1 and MRP2 appear to be involved in the apical efflux of organic anions, the second membrane step in secretion. Transport through NPT1 occurs down an electrochemical gradient, whereas MRP2 transport is primarily active. Recently P. Smeets et al demonstrated that in addition to MRP2, the classical ATP-dependent PAH transporter, there is another PAH transporter MRP4 with an even higher affinity for PAH compared to MRP2, and is expressed at higher levels in the kidney [90a].

54

Molecular identification of organic anion transporters without affinity for PAH A number of transport molecules have been cloned from different tissues and identified in the renal proximal tubule, which do not transport PAH, but may contribute to the apical efflux of organic secretion [45]. OATP1. The S3 segment of proximal tubule expresses OATP1, an organic anion transporter cloned from rat liver, which transport bile acid, bromosulfophtalein, and conjugated and unconjugated steroid hormones, in a sodium independent manner. Although hepatic OATP1 is expressed in the basolateral membrane (blood side) of hepatocytes, in the kidney it is located in the apical membrane. In the rat renal OATP1 mRNA, but not the hepatic one, is strongly regulated by androgens and to a lesser extent by estrogens. OATP1 might play a role in the renal excretion of estrogens [50]. A homolog of OATP1, OATP3 was isolated from a rat retina and found to be expressed specifically in the retina and in the kidney. It transports taurocholate as well as thyroid hormone (T3 and T4) [91]. A homologous transporter, OATP2, a liver specific transporter, is not expressed in the kidney [91]. OAT-K1 and OAT-K2. These transporters are two homologous organic anion transporters specific to the kidney, which have been identified by molecular cloning strategy [51, 52]. In rats, OAT-K1 was localized in the apical membrane of straight proximal tubules. When expressed in cultured renal epithelial cells, OAT-K1 mediates both uptake and efflux of methotrexate through the apical membrane, and appears to be specific for methotrexate and folate [51]. Non-steroid anti-inflammatory drugs (indomethacin, ketoprofen, ibuprofen, flufenamate, phenylbutazone) inhibit methotrexate OAT-K1 mediated uptake, but are not transported themselves. OAT-K1 appears to be a site for methotrexate and non-steroidal anti-inflammatory drugs interaction [92]. In rats, OAT-K2, as OAT-K1, was localized in the apical membrane of straight proximal tubule [52]. When transfected in cultured epithelial cells, it mediates not only the apical transport of methotrexate and folate but also that of taurocholate and prostaglandin E2. In cis-inhibition studies, steroids, bile acid analogs, and cardiac glycosides were shown to have a high affinity for OAT-K2, suggesting that it participates to the apical transport of hydrophobic anionic compounds in the kidney [52].

03. Renal handling of drugs and xenobiotics

Conclusions The molecular identification of various organic anion transport proteins, and the characterization of their transport mechanisms in various cell systems, gives an insight in the complexity of the renal secretion of organic anions (Figure 5). Among these numerous transport systems characterized at the molecular level, only OAT1 and OAT3 have a clearly established role, being the most likely candidates of the PAH secretory mechanism. The identification of the apical transporter for the PAH secretory mechanism remains to be established. However, in contrast to a main basolateral transporter, several apical organic anion transporters appear to facilitate the transport of the various substrates accumulated in the proximal cells by OAT1. The respective role of the apical transporters, need to be demonstrated in situ. In vivo models, such as transgenic mice, will allow the elucidation of the physiological and pharmacological roles of these transport proteins. The proximal tubule, the nephron segment which suffers the greatest damage during renal ischemia, is essentially aerobic, with little or no glycolytic capacities in adult life, relying on Krebs cycle intermediates, including the OAT counterion, alpha-ketoglutarate. Thus, along with cellular metabolism and sodiumdicarboxylate cotransporter activity (carbon substrate influx), OAT activity (carbon substrate efflux) might be a key determinant of the metabolic health of the proximal tubule. This hypothesis suggests that circulating OAT substrates might increase the susceptibility of the proximal tubule to ischemic injury, because they are exchanged by basolateral OATs for intracellular carbon substrate. Concomitant with the transport of nephrotoxic OAT substrates (e.g. cephaloridine, cidofovir, ochratoxin) into the proximal tubular cell from blood, there is an equimolar loss of dicarboxylate, which may additionally compromise the metabolic integrity of the cell at the very time when noxious substances are increasing withing it. It is possible to imagine a vicious cycle leading to increased proximal tubule injury in such a setting [52a].

Tubular transport of organic cations

Owing to electro-negativity of cell interior, a transfer of positively charged molecules from peritubular interstitium into cells occurs along a favorable electrochemical gradient and does not require energy. In contrast, energy is necessary for the efflux from cell to lumen which takes place against the electropositivity of the lumen. The situation is opposite to that of organic anions for which the active step is the basolateral transport. The mechanisms involved in tubular secretion of organic cations are schematically summarized in Figure 6. Transport of organic cations at the basolateral membrane occurs by a voltage sensitive pathway (mechanism 1), which was described for N1-methylnicotinamide, TEA and/or procainamide in rats, dogs and rabbits. Because of the electronegativity of the cell this facilitated pathway drives organic cations into cells. In rabbits, an organic cation exchanger has also been observed (mechanism 2), the role of which in tubular secretion is unclear. As is described below, the molecular structure of a few isoforms of an organic cation transporter (OCT) has been defined, some of which might be the basolateral transporter of proximal tubule. The Nernst equation predicts that because of the cell electronegativity, passive facilitated diffusion should allow a concentration ratio cell water/external medium approximating 10 to 15 at steady state. In isolated unperfused proximal tubules from rabbits, ratios exceeding 100 for TEA, have been measured [44] and one can wonder if another mechanism exists, for example a cation exchanger, as demonstrated in rabbits (mechanism 2), but which has generally not been observed in rats and dogs [44], which might be implicated in basolateral uptake. However, as reported for anions, capillary BM TEA+

lumen

TEA+

1

TEA+

3 2

Y+

2K+

H+

TEA+ H+

4 Na+

5 Transport mechanisms for organic cations have been investigated not only for the classical substrates, TEA and N1-methylnicotinamide, but also for a few other organic cations, mainly drugs (Table 6).

tubular cell

3Na+

Figure 6. Model of the organic cation tetraethylammonium transport in proximal tubule. 55

DE BROE & ROCH-RAMEL

Table 6. Organic cation transporter (OCT) families (from [34], with permission). Name OCT1 (SLC22A1)

Human

Substrates MPP+, TEA Drugs: acylclovir, ganciclovir

Oct1 (Slc22a1)

Rat

OCT2 (SLC22A2)

Human

Oct2 (Slc22a2)

Rat

OCT3 (SLC22A3)

Human

Oct3 (Slc22a3)

Rat

MPP+, TEA, guanidine

OCTN1 (SLC22A4)

Human

Octn1 (Slc22a4) OCTN2 (SLC22A5)

Rat

TEA, MPP+, L-carnitine, acetyl-Lcarnitine Drugs: pyrilamine, quinidine, verapamil TEA, MPP+

Octn2 (Slc22a5)

Human

Rat

TEA, MPP+, NMN, monoamine neurotransmitters Drugs: AZT, cimetidine, cladribine, cytambine, D-tubocurarine TEA, MPPf, NMN, agmatine, monoamine neurotransmitters Drugs: amantadine, memantine TEA, MPP, adrenaline, agmatine, creatinine, monoamine neurotransmitters Drugs: amantadine, cimetidine, memantine MPP+, guanidine, monoamine neurotransmitters Drugs: cimetidine, tyramine

TEA, MPP+, L, D-carnitine, acetyl-1carnitine, betaine, choline, cysteine, lysine, methionine Drugs: pyrilamine, quinidine, valproate, verapamil L-carnitine. TEA

Inhibitors Choline, matinine, corticosterone, desipramine, dopamine, β-estradiol, nicotine, NMN, progesterone Drugs: anti-HIV drugs, acebutolol, amantadine, cimetidine, clonidine, disopyramide, midazolam, procainamide, prazosin, quinine, quinidine, vecuronium, verapamil Corticosterone, guanidine, histamine, nicotine, o-methylisoprenaline Drugs: clonidine, desipramine, mepiperphenidol, procainamide, reserpine, quinine, quinidine Corticosterone, o-methylisoprenaline, progesterone, SKF550 Drugs: despramine, mepiperphenidol, phenoxybenzamine, procainamide, quinine Corticosterone, dexoycorticosterone, β-estradiol, NMN, progesterone, monoamine neurotransmitters Drugs: cimetidine, cisplatin, procainamide, quinine

Corticosterone, β-estradiol, MPTP, o-methylisoprenaline, progesterone, SKF550 Drugs: clonidine, desipramine, imipramine, phenoxybenzamine, prazosin, procainamide Monoamine neurotransmitters, corticosterone, dexoycorticosterone, β-estradiol, NMN, progesterone, testosterone Drugs: amphetamine, cimetidine, clonidine, desipramine, methamphetamine D-carnitine, nicotine Drugs: cephaloridine, cimetidine, procainamide, quinine DMA, nicotine Drugs: cirnetidine, desipramine, imipramine, procainamide, verapamil Aldosterone, corticosterone, MPTP, nicotine Drugs: cephalosporin antibiotics, cimetidine, clonidine, desipramine, emetine, procainamide, pyrilamine, quinine

MPTP, nicotine Drugs: cephalosporin antibiotics, cimetidine, clonidine, despramine, procainamide

Abbreviations: AZT, azidothymidine; MPF+, 1-methyl-4-phenylpyridinium; MPTP, l-methyl-4-phenyl-l,2,3,6-tetrahydropyridine; NMN, N1-methylnicotinamide; SKF550, (9fluorenyl)-N-methyl-β-chloroethylamine; TEA, tehaethylammonium.

there is evidence that part of the TEA accumulated into cells is bound to cytoplasmic organelles and that only part of TEA is freely diffusible [44, 93]. It has also been demonstrated that endosomal membrane vesicles isolated from rat renal cortex can accumulate TEA by an ATP-dependent mechanism [93]. It is conceivable that the favorable transmembrane potential is the principal or single driving force required for cellular uptake. The efflux from cell to lumen is the active step of organic cation secretion transport being against the transmembrane potential. This active transport occurs through an exchange with protons (mechanism 56

3), maintained by the proton concentration gradient resulting from the Na+/H+ exchange at the same membrane (mechanism 4), mechanism energized itself by the low Na+ concentration resulting from the Na-K -ATPase activity (mechanism 5). Thus, in all species investigated so far (rats, dogs, rabbits [44], pig [94], humans [95]), TEA and N1-methylnicotinamide were demonstrated to be transported in brush-border membrane vesicles by an electroneutral organic cation exchange system where one organic cation molecule is transported against one proton. Amiloride, cimetidine, morphine, procainamide, disopyramide, gentamicin

03. Renal handling of drugs and xenobiotics

and verapamil [96] are also transported by such an electroneutral proton exchanger, in rats and/or rabbits. Larger and more hydrophobic compounds (quinidine, quinine, d-tubocurarine, vecuronium) are inhibitors of organic cation transport, but are not transported by the proton-organic cation exchanger. They are transported by another transport mechanism, probably by the MDR1/P-glycoprotein (see below). Many studies have been performed to characterize the requirements for a substrate to be transported by the “organic cation transport mechanism” [97-100]. As for organic anions, the molecular structure of substrates is rather unspecific. Hydrophobicity and basicity are the general characteristics of substrates, but their ionization is not a prerequisite for interacting with the basolateral carrier [65]. Similar properties were found in brush border membrane [97-100]. Although the ratio of basolateral to apical membrane affinities may vary with substrates and animal species [97]. More than one organic cation/proton exchanger appears to be involved in the transport of organic cations through the apical membrane, and substrates show overlapping affinities for these different exchangers [101]. Thus, proton-stimulated guanidine uptake by brush border membrane vesicles is only minimally inhibited by TEA, N1-methylnicotinamide and choline, whereas amiloride, clonidine, imipramine and harmaline are more potent inhibitors [101]. There are also species differences. Cephalexin, for instance, shows an affinity for the N1-methylnicotinamide or TEA transporter in human brush border membranes [95], while in rats it has no affinity for the TEA transport mechanism. At present the molecular identity of the organic cation/H+ remains unknown, although, as discussed below, two organic cation transporters were recently identified in the kidney, which might be the apical transporters of organic cations. Molecular Identification of putative basolateral organic cation transporter belonging to the OCT family Expression cloning allowed the identification of several isoforms of a polyspecific organic cation transporter OCT. The molecular biology of these various OCTs have been described in detail by Koepsell et al. [108], Zhang et al. [109] and Burckardt and Wolf [76]. After the cloning of the first organic cation transporter (rOCT1) isolated from a rat kidney [110], a number of homologous cation transporters have been identified

[108]. When expressed in various cell systems, OCT1 and OCT2 isoforms demonstrate a broad substrate affinities and a voltage dependent transport. These transport characteristics made them candidates to be the organic cation basolateral transporter of proximal tubule [29, 108, 111]. The HIV protease inhibitors, indinavir, nelfinavir, ritonavir, saquinavir inhibit TEA transport by hOCT1 but they are probably not transported [112]. Inhibitor potency for OCT1 and OCT2 varies with species [108, 109]. In general human hOCT1 interacts with the n-tetraalkylammonium compounds with a lower affinity than that of rats, mice, or rabbits [113]. Among OCT isoforms, rat rOCT1, rOCT2, rOCT3, human hOCT2 and hOCT3, and mice mOCT3 are involved in the renal transport of organic cations. In rats, rOCT1 and rOCT2 are expressed primarily in the kidney, and are localized in the basolateral membrane of proximal tubule [114, 115]. Both probably play a role in organic cation secretion [111]. The expression level of rOCT2 mRNA and protein in males is much higher than in females, which correlates with the higher transport of TEA in male basolateral membrane vesicles and cortical slices [116]. Because no gender differences were observed for rOCT1 expression in the kidney, rOCT2 and not rOCT1 might represent the main renal organic cation transporter in rats. Another isoform, rOCT3, which transport TEA and guanidine, is expressed in many organs including the kidney. However, because its tubular localization is still unknown, its functional role remains to be defined [117]. In the mice mOCT3 mRNA was found to be expressed in the proximal and distal tubule, but the membrane localization is unknown. In human hOCT1 is expressed in the liver and not in the kidney, whereas hOCT2 is present predominantly in the kidney. However hOCT2, being restricted to the distal convoluted tubule, does not represent the organic cation secretory transporter in human [29](Figure 7). Human OCT3 is expressed in the kidney and also in other organs, its nephron localization has not been determined [117]. There are discrepancies between Gründeman et al. [118] and Wu et al. [117] concerning the substrate affinities for hOCT3. Wu et al demonstrated a broad substrate affinities for hOCT3, which transports various organic cations including, TEA, clonidine, imipramine, procainamide, endogenous amine, etc., whereas Gründeman et al. concluded that hOAT3 is limited to the transport of endogenous or57

DE BROE & ROCH-RAMEL

capillary BM

tubular cell

lumen Na+

H+ ATP ADP

organic cation OCT1

–

H+

–

OCT2 OCT3 ATP ADP -3 mV

-70 mV

P-glycoprotein 0 mV

Figure 7. Mechanisms of organic cation transport in renal tubular cells. Cellular uptake of organic cations across the basolateral membranes (BM) is mediated primarily by membrane potential-dependent organic cation transporters such as OCT1 (1) and OCT2 (2). OCT3 (3) may contribute in part to the cellular uptake of organic cations. Exit of cellular organic cations across brush border membranes (BBM) is mediated principally by unidentified H+/organic cation antiporter (4). P-glycoprotein (5) is involved in tubular secretion of hydrophobic drugs such as digoxin, anticancer agents, and some immunosuppressants (cyclosporine and tacrolimus). Adapted from [29].

ganic cations, such as dopamine, histamine, and that it does not transport TEA. At present human OCT3 is the only transporter isoform that has been implicated in renal transport of organic cations. Its presence in the basolateral membrane of proximal tubule must be demonstrated before one can conclude it is the human basolateral secretory transporter of organic cations. In conclusion, substantial evidence exist that rOCT1 and rOCT2 are involved in the secretion of organic cations in rat proximal tubule. More precise localization of transporters are needed before determining the role of rOCT3 in rats, and hOCT3 in human. Molecular identification of putative apical organic cation transporters A few transport mechanisms have been identified in the kidney by expression cloning, which might be involved in the apical step of organic cation secretion, although their function in situ has not been established. OCTN1, OCTN2. Two organic cation transporters, OCTN1 and OCTN2, were identified in the kidney and other organs of rats, mice, rabbits and human, by 58

their homology to the basolateral transporter OCT [53, 54, 119]. When expressed in human embryonic kidney cells and Xenopus oocytes, human OCTN1 mediates the transport of TEA in a pH dependent manner, transport being higher at neutral or alkaline pH than at acidic pH. The transport of TEA was observed to be bidirectional and inhibited by various organic cations, such as choline, clonidine, cimetidine, quinidine, verapamil, etc., and by zwitterionic compounds such as L-carnitine, cephaloridine, levoflaxin. The transport of a few of these inhibitors, quinidine, verapamil, and L-carnitine, was demonstrated [53]. In summary, OCTN1 is a multispecific, bidirectional and pH-dependent organic cation transporter, which is probably energized by a proton antiport mechanism. Although its subcellular localization in the kidney is unknown, the functional characteristics of OCTN1 suggest that it might be involved in the apical step of organic cation secretion. In the kidney, OCTN2 is expressed predominantly in cells of proximal and distal tubules, as well as in glomeruli. OCTN2 has the same functional characteristics than OCTN1, but the substrate affinities for the transporter differ [54]. Human, rat and mouse OCTN2 has the additional peculiarity of transporting L-carnitine and other zwitterions such as cephalosporins that contain quaternary nitrogen, in a-sodium dependent manner [54, 120, 121]. Site directed mutagenesis experiments provided evidence that the transport sites for organic cations and for carnitine are distinct [122]. OCTN2 thus might play a role in organic cation secretion, and in the reabsorption of carnitine by a sodium carnitine cotransport. In vivo cephaloridine was reported to increase the fractional excretion of carnitine by interfering with its reabsorption [120]. The possibility exists that this type of cephalosporin might be inefficient in patients with primary carnitine deficiency, that are receiving carnitine supplementation, because of competition for carnitine transport [120]. The anionic cephalosporins are not substrates for OCTN2, but they are substrates for the peptides transporters PEPT1 and PEPT2 [45] (Table 7). Conversely, the cephalosporins, which have affinity for OCTN2, are not substrate of the peptide transporters [120]. In recent years polymorphisms of genes encoding proteins involved in the metabolism and subsequent

03. Renal handling of drugs and xenobiotics

Table 7. Peptide transporter (PEPT) nucleoside transporter families. Name PEPT1 (SLC15A1) Pept1 (Slc15a1) PEPT2 (SLC15A2) Pept2 (SLC15a2) CNT1 (SLC28A1) Cnt1 (SIc28a1) CNT2 (SLC28A2) Cnt2 (Slc28a2)

Human Rat Human Rat Human Rat Human Rat

Substrates Glycylsarcosine, di-, tripetides Drugs: β-lactam antibiotics, cyclacillin, valacyclovir Glycylsarcosine, di-, tripetides Drugs: bestatin, β-lactam antibiotics, Glycylsarcosine, ALA Drugs: bestatin, cephalexin, valacyclovir Glycylsarcosine Drug: valacyclovir Adenosine, thymidine, uridine, Drugs: AZT, zalcitabine Adenosine, thymidine, uridine Drug: AZT Adenosine, uridine, inosine, thymidine Drugs: cladribine, didanosine Adenosine, guanosine, inosine, thymidine, uridine Drug: didanosine

Inhibitors Valine, pentaglycine Drugs: β-lactam antibiotics, bestatin, certain ACE inhibitors Drugs: β-lactam antibiotics, certain ACE inhibitors, tolbutamide, chlorpropamide Drugs: β-lactam antibiotics Drugs: β-lactam antibiotics, bestatin, chlorpropamide, glibenclamide, tolbutamide

Drugs: cytarabine, floxidine, gemcitabine, idoxuridine, zalcitabine

Abbreviations: ACE, angiotensin converting enzyme; ALA, delta-aminolevulinic acid; AZT, azidothymidine

renal and/or extrarenal elimination of xenobiotics have been shown to correlate with drug sensitivity. Gain of function of an OCT relevant for drug elimination will decrease plasma levels and may prevent appropriate therapeutic effects at standard dosage. A loss-of-function polymorphism may lead to increased toxicity in affected individuals. Activation of protein kinase C leads to strong stimulation of rOCT1 expressed in human embryonic kidney cells. Protein kinase C does not only increase the maximal transport rate but it also alters the relative selectivity of the carrier. Cation transporter isoforms do not only differ in substrate affinities but also in regulation. More research correlating polymorphisms of genes encoding transport-regulating kinases with drug elimination is needed [120a].

ABC transporter family Multidrug resistance-associated protein transporters (MRPs) The MRP family is a subgroup of the ATP-binding cassette (ABC) transporters superfamily (Table 8). It comprises 13 members (ABCC1 to ABCC13) named MRP (1 to 9), CRTR (cystic fibrosis transmembrane conductance regulator - ABCC7), and SUR1 or 2 (sulphonylurea receptors (ABCC8 and 9). MRP mRNA is retrived from several tissues including the kidney and the transporter is located at the basolateral membrane of Henle’s loop and collecting duct cells [120b,c]. MRP1 carries in an ATP-dependent manner different substrates among which are found several conjugated

derivatives, sulfates, and GSH. Carrying some nonconjugated drugs necessitates an exchange with GSH. As a result, drug resistance mediated by MRP1 may be counterbalanced by GSH synthesis inhibition. In the kidney, MRP2 (ABCC2) contributes to the detoxification of drugs and both endogenous and exogenous compounds, mainly under their conjugated form. It has been located at the brush border membrane of S1, S2 and S3 segments of proximal tubular cells [120d,e]. In the kidney, but not in the liver, 8 days following cisplatin administration MRP expression is increased. Subtotal nephrectomy induced a 200% increase in MRP2 mRNA in the remaining kidney. In the kidney, MRP3 (ABCC3) is expressed at the basolateral membrane of distal renal tubular cells and carries glucuroconjugated compounds and other molecules from the internal tubular cell into the blood. MRP3 was shown to confer cellular drug resistance to etoposide, tenoposide and vincristine [120f,g]. MRP4 (ABCC4) mRNA has been detected in the kidney, at the brush-border membrane of proximal tubular cells. It enhances cell resistance to some antiviral agents such as adefovir and zidovudine. It also seems to play an important role in antiviral drugs renal excretion. MRP4 is also the transporter for cyclic AMP and GMP through an ATP-dependent system and it constitutes the elective excretion pathway for cyclic nucleotides in renal epithelial cells. MRP5 (ABCC5) is widely expressed in the organism, including the kidney. It is located at the basolateral 59

DE BROE & ROCH-RAMEL

Table 8. ABC transporter family (from [34], with permission). Name MRP1 (ABCC1)

Human

MRP1 (Abcc1) MRP2 (ABCC2)

Mouse

Mrp2 (Abcc2)

Rat

MRP3 (ABCC3) Mrp3 (Abcc3) MRP4 (ABCC4) MRP5 (ABCC5) MRP6 (ABCC6) MDR1 (ABCB1)

Human

mdr1a/ mdr1b (Abcb1)

Human

Rat Human Human Human Human

Rat/ Mouse

Substrates LTC4, bilirubin-glucuronide, glutathione conjugates, GSH, PAH, fluo-3, calcein Drugs: etoposide-glucuronide, S-(ethacrynic acid)glutathione, MTX LTC4, calcein, APA-SG Drugs: daunorubicin, vincristine LTC4, E217βG, bilirubin-glucuronide, glutathione conjugates, GSH, PAH, ochratoxin A, fluo-3 Drugs: anti-HIV drugs, benzbromarone, furosemide, indomethacin, MTX, vinblastine LTC4, LTD4, E217βG, anionic glucuronide conjugates, bilirubin-glucuronide, BSP, endothelin-1, fluo-3, folate, GSH, GSSG Drugs: cefpiramide, cefhiaxone, indomethacin, irinotecan and SN-38, MTX, pravastatin LTC4, DNP-SG, E217βG, folate, glycocholate Drug: MTX LTC4, , bile acids Drugs: E3040-glucuronide, MTX E217βG, cAMP, cGMP Drugs: adefovir, AZTMP, MTX DNP-SG, CAMP, cGMP, GSH Drugs: adefovir, 6-MP LTC4, NEM-SG Drug: BQ123 E217βG, calcein, fluo-3, rhodamine 123 Drugs: cardiac glycosides, anti-HIV drugs, anticancer agents, verapamil

Inhibitors Probenecid, ochratoxin A Drugs: benzbromarone, CSA, S-(decyl)-glutathione, indomethacin, MK571, sufinpyrazone, valspodar GSSG Drugs: arsenate, genistein, MK571 Probenecid, BSP Drugs: CSA, glibenclamide, MK571

Probenecid Drugs: CSA, glibenclamide, MK571

Drugs: benzbromarone, MK571 Anionic glucuronide / GSH conjugates Probenecid, anionic glucuronide conjugates Drugs: benzbromarone, sildenafil, trequinsin, zaprinast Probenecid Drugs: benzbromarone, sildenafil, trequinsin, zaprinast Probenecid Drugs: benzbromarone, indomethacin Progesterone Drugs: amiodarone, amitriptyline, chlorpromazine, diltiazem, dipyridamole, elacridar, fluphenazine, fucidin, lovastatin, mefloquine, phenothiazines, pimozide, propafenone, propranolol, quinine, quinidine, reserpine, simvastatin, spironolactone, staurosporin, tamoxifen, trifluoperazine, triflupromazine, valspodar

Rhodamine 123 Drugs: anti-HIV drugs, CSA, dexamethasone, digoxin, doxorubicin, fexofenadine, ivermectin, verapamil, vinblastine

Abbreviations: APA-SG, azidophenacyl-S-glutathione; AZTMP, azidothymidine monophosphate; BQ123, (cyclo [Trp-Asp-Pro-Val-Leu]); BSP, bromosulfophthalein; CSA, cyclosporine A; DNP-SG, S-(dinitrophenyl)-glutathione; E217βG, estradiol-17β-D-glucuronide; GSH, reduced glutathione; GSSG, oxidized glutathione; LTC4/LTD4, leukotriene C4/D4; MK571, 3-[3-[2-(7-chloroquinolin-2-yl)vinyl]phenyl]-(2-dimethylcarbamoyl-ethylsulfanyl) methylsulfanyl] propionic acid; 6-MP, 6-mercaptopurine; MTX, methotrexate; NEM-SG, N-ethylmaleimide glutathione; PAH, p-aminohippurate.

membrane and carries GSH. MRP6 is expressed in the kidney, at the basolateral membrane of proximal tubular cells. It has been suggested that the loss of functional MRP6 in the kidney and the liver could induced the phenotype observed in Pseudoxanthoma elasticum patients. Mutidrug transporters/P-glycoprotein (MDR or Pg) The apical membrane of proximal tubules is particularly rich in MDR-glycoprotein (“multidrug transporter”), a membrane ATPase that mediates the active efflux of a wide variety of drugs across the 60

plasma membrane of several cell types. This property explains the resistance of some cancer cells to hydrophobic cationic drugs [102]. It was demonstrated that MDR/P-glycoprotein can extrude many organic compounds (e.g. vinblastine, vincristine, colchicine, cyclosporine analogues) from renal proximal cell [103105]. P-glycoprotein transport mechanism differs from the proton/organic cation exchanger since it does not transport TEA [94, 105], but the more lipophilic substrate, and vinblastine, a substrate of MDR/P-glycoprotein, is not exchanged against protons in pig brush border membrane vesicles [93].

03. Renal handling of drugs and xenobiotics

MDR/P-glycoprotein, which transports organic cations is the analogous of MRP2, which transport lipophilic organic anions. Both are responsible for the multidrug resistances of cells to anticancer drugs. Since some substrates of the P-glycoprotein system are also transported by the proton/organic cation exchanger, it is often difficult to clearly distinguish between the two systems at the functional level. Compounds transported by both the MDR/P-glycoprotein and the organic cation transporter, include daunomycin, colchicine, verapamil, quinidine and vinblastine [106]. On the other hand, secretion of digoxin, which is not an organic cation, is restricted to P-glycoprotein only [107] (Table 8). In the kidney, P-glycoprotein is constitutively expressed on the brush border of the proximal tubular cells and on the distal tubule [107a] and it has been suggested that P-glycoprotein may be instrumental in cyclosporine A (CsA) nephrotoxicity. CsA is a substrate of P-glycoprotein [107b] and variations in expression and/function of P-glycoprotein could lead to accumulation of CsA, along with other cytotoxic agents, within the tubular cell. An inverse relationship between CsA deposits in renal tissue and the level of P-glycoprotein expression in proximal tubular cells in animal models, suggesting that the normal P-glycoprotein response mat be defective in patients suspectible to CsA-related nephrotoxicity, leading to retention of excess amounts of CsA in the cells [107c,d]. ABCB1 polymorphism in kidney allograft donors, which is associated with decreased expression of P-glycoprotein in renal tissue, has ben shown an independent risk factor for the development of CsA-related nephrotoxicity [107e]. These findings suggest that factors that modulate Pglycoprotein-expression may have an impact on CsArelated nephrotoxicity by causing an accumulation of CsA within the renal cells. The new immunosuppressive agent sirolimus is also a P-glycoprotein substrate [107f], although perceived as a non-nephrotoxic drug, reducing renal function when given concomitantly with CsA [107g]. Recent studies have shown that administration of sirolimus around the time of renal injury can exacerbate the injury and delay repair, an effect that may be due to a potent antiproliferative effect of sirolimus on tubular cells [107h]. Using human renal epithelial cells in primary culture it was shown that sirolimus inhibits the P-glycoprotein-mediated efflux and cellular concentration

of CsA, explaining at least partly the exacerbation of CsA nephrotoxicity of sirolimus. Effects of protein binding on organic ion secretion It is generally recognized that the tubular secretory rate is proportional to the concentration of free drug or xenobiotic [123-126], and that plasma albumin binding is not rate limiting for tubular secretion of organic anions with high affinity for the transport system [19, 36], because the dissociation rate of the organic anion/ albumin complex is much faster than the transtubular transit time [19, 127]. Such is the case for hydrochlorothiazide [36]. On the other hand, organic anions with lower affinity for the transporter (e.g. phenol red) have a reduced secretion when bound to plasma proteins [124, 128]. Although the secretion of furosemide in the perfused isolated rat kidney can be delayed by the addition of albumin to the perfusate [123], secretion in humans does not appears to be limited by protein binding. Thus, in spite of a binding of more than 95% to plasma proteins, the urinary clearance (uncorrected for plasma protein binding) of furosemide in therapeutic doses is somewhat higher than inulin clearance [36, 129, 130]. Because of the high protein binding of furosemide its filtration rate is negligible and its diuretic effect, which is related to its luminal concentration in the thick ascending limb of Henle’s loop, depends on its tubular secretion. Hence, inhibition of furosemide secretion by probenecid inhibits its diuretic effect [131].

Interactions of xenobiotics/drugs for secretion Probenecid, which was first developed to delay penicillin excretion, is now generally used (besides its use as a uricosuric) to inhibit secretion of organic anions [131a]. Thus, it is generally considered that a compound whose secretion or transport across the proximal basolateral membrane is inhibited by probenecid is a substrate of the organic anion secretory mechanism. Probenecid has also been used as a tool to investigate the role of cellular accumulation of xenobiotics in nephrotoxicity. Inhibition of basolateral uptake of cephalosporins, such as cephaloridine and cephaloglycin, by probenecid, can prevent their cellular toxicity. These cephalosporins have a low extrusion rate through the apical membrane, resulting in a rather high concentration, which is a major contributing factor to their nephrotoxicity. However, it is worth noting that 61

DE BROE & ROCH-RAMEL

cell accumulation is necessary but not sufficient for cytotoxicity, as shown by cephalexin that has a low nephrotoxic potential despite marked cortical accumulation [40] (see also chapter 9). The nephrotoxicity of cisplatin is reduced in humans [132], mice [133] and dogs [134] by co-administration of probenecid, suggesting that cisplatin is transported by the PAH transport system. It has been proposed that platinum, like other nephrotoxic metal ions such as mercury and potassium dichromate, are taken up by tubular cells as sulphydryl conjugate through a probenecid-sensitive pathway [133]. However, cisplatin might also be transported by the organic cation transport system, since quinidine, cimetidine and ranitidine inhibited its net secretion flux in the dog kidney [134]. In human, methotrexate is largely cleared unchanged from the body by renal excretion through glomerular filtration and tubular secretion. Rises in serum methotrexate levels accompanied by life-threatening increases in methotrexate toxicity can occur if aspirin, salicylates or non-steroidal anti-inflammatory drugs are given concurrently. The increased methotrexate toxicity observed by concomitant administration of ibuprofen [135], salicylates [135], or flurbiprofen [136] might be in part the result of interaction at the basolateral membrane [137], resulting in a decrease in methotrexate renal excretion. Excretion of digoxin is primarily renal, by glomerular filtration and tubular secretion and reabsorption. Competition studies have shown that the “classic” anion or cation transport systems are not involved. The secretory process as studied in vivo and in a renal epithelium in vitro, may be carried out by the apical membrane P-glycoprotein. It is well known clinically, that several drugs (most notably quinidine, verapamil, nifedipine, propofenone, spironolactone, and amiodarone) reduce the renal (tubular) clearance of digoxin and increase the plasma concentration and toxic risks of the cardiac glycoside [63, 138, 139]. Accordingly, these interactions may be explained by a competition at the secretory step controlled by P-glycoprotein at the luminal membrane. Such a possibility has received experimental support for several of these compounds [139-143] Concurrent use of drugs that reduce renal blood flow in patients with renin-angiotensin prostaglandin dependent renal perfusion (e.g. NSAID), that are weak 62

organic acids competing for tubular secretion [144] and/or nephrotoxic (cisplatin) can delay drug excretion [145] and lead to severe myelosuppression. Interactions of cimetidine and other H2-receptor antagonists with the renal secretion of several drugs have been repeatedly described, and comprehensively listed [146]. Thus, cimetidine inhibits renal secretion of procainamide in humans and prolongs its elimination half-life [147, 148]. Similar inhibitory effects have been shown on creatinine, ranitidine and many other cationic compounds [149].

Interactions between organic anion and organic cation secretion Clinical significant interactions occur only when the affected transporter represents the major pathway for the overall elimination. Because of the involvement of multiple renal processes (i.e. filtration, tubular secretion, tubular reabsorption) in renal drug handling and the functional redundancy of some renal drug transporters, severe clinical drug-drug interactions at the renal level seem to be not very common. Clinical relevance of renal drug-drug interactions needs to be evaluated in the context of efficacy and safety profile of the affected drug. In vitro transporter interaction screening during preclinical development should be performed for all drug candidates that have a narrow therapeutic window [131a]. The lack of strict structural requirements for substrates in organic anion and cation transport systems, the prominent role of substrate hydrophobicity in the interaction with both classes of carriers, and the ability of non-ionized substrates to interact with the transporters, are all factors explaining that some substrates might be transported by both transport systems [37]. For example, the renal excretion of cimetidine and famotidine, two organic cations, is reduced by probenecid [150, 151]. In vitro also, cimetidine uptake by brush border membrane vesicles is inhibited by probenecid or furosemide, and cimetidine in turn can inhibit PAH uptake, demonstrating the existence of some link between organic anion and cation transport [152-155]. Such observations appear to overturn the dogma of distinct transport systems for organic ions. Some compounds have chemical characteristics (“zwitterions”), which account for their particular substrate behavior: creatinine [156], amino-cephalosporins (e.g. cephalori-

03. Renal handling of drugs and xenobiotics

dine) and gyrase inhibitors [157] bear both positive and negative charges, and are therefore “bisubstrates” [37]. Cimetidine has affinity for the organic cation basolateral transporter through its imidazole group, while hydrophobicity of the molecule and electronegativity of the cyanoguanidine group explain the affinity of the drug for the organic anion transporter [158]. Famotidine and ranitidine have a guanidine group and a nucleophilic side-chain accounting for the affinity for both transport systems [158]. Clonidine and pilocarpine are other imidazole derivatives interacting with both transporters [158]. Zidovudine secretion in rats might also proceed through both transport mechanisms [159], though the anion transporter appears to predominate [160, 161]. As reported above, cisplatin also appears to be transported by both transport systems. Many other compounds interact at the basolateral membrane with the PAH and the organic cation transport systems as was demonstrated by the systematic studies by Ullrich et al. [37].

Metabolism of drugs/ xenobiotics in the kidney Metabolic transformation is the biological conversion of a drug to another chemical form, occurring mainly in the liver, although many other tissues, among them the kidney are also capable of drug metabolism. Microsomal enzymes are responsible for oxidation, acetylation, conjugation (acylglucuronidation, N-glucuronidation, glycination) hydrolysis of drugs and xenobiotics. The usual result of this enzymatic conversion is drug metabolites, which are more polar, and less lipid soluble than the parent compound and consequently favoring renal excretion. The same enzymatic pathways for drug metabolism present in the liver are also found in the kidney, although the specific activity of these pathways in the kidney is substantially lower than those in the liver [2, 162]. In contrast to the liver, the metabolic pathways in the kidney are not uniformly distributed throughout the kidney, they are localized to specific nephron segments (Table 4). Examples of drug metabolism by the isolated perfused kidney are oxidation of bumetanide [163], acetylation of sulphisoxazole [164], conjugation of salicylic acid [165], and esterolysis of enalapril to enalaprilat [166]. The role of renal enzyme systems involved in the metabolism of drugs and their potential nephrotoxicity

is well documented in the case of analgesic mixtures containing acetylsalicylic acid, acetaminophen and/ or phenacetin combined with addicting compounds such as caffeine and codeine [167]. The kidney can metabolize acetaminophen to glucuronyl and sulphate conjugates but also to an arylating intermediate via the cytochrome P-450 mixed function oxidase system [168, 169]. The intra-renal distribution of this enzyme system explains the proximal tubular localization of acute acetaminophen toxicity [170]. Several observations in the Fischer rat suggests that this acute renal toxicity is mediated through the cytochrome P-450 mechanism [168]. Renal metabolism of isoproterenol [171], bumetanide [163], cimetidine [172] and N-methylnicotinamide [173] has been reported. Renal metabolites may have different mode of excretion [174], and may be more nephrotoxic than the original substance [175]. Renal glucuronidation may be substantial as in the case of morphine [176]. Xenobiotic glucuronidation can proceed by linkage through an ether or an ester bound. The latter process is called “acyl-glucuronide” characterized by instability under physiological conditions such that the glucuronide can deconjugate back to the parent compound (futile cycle). In patients with normal renal function, acyl-glucuronides are readily eliminated in the urine. In patients with renal insufficiency, the conjugate accumulates in plasma where it can spontaneously hydrolyse to reform the parent compound. This phenomenon, demonstrated for clofibrate [177, 178] diflunisal [179, 180] and some NSAID [181, 182], leads to a paradox in which a drug may accumulate in patients with renal insufficiency even through negligible amounts of parent drug are eliminated in the urine of patients with normal renal function. The main role of the kidney in the process of drug metabolism consists in the excretion of the many, more or less pharmacologically active metabolites formed in the liver [8]. Needles to say that renal insufficiency may result in the accumulation of metabolites and, if pharmacological active, may result in serious side effects/ toxicity [33]. Renal metabolism of drug-xenobiotics and its contribution to elimination has been inadequately explored so that clinical implications are for the most part inferred from animal models or speculative. The impact of knowledge of renal handling on drugs and xenobiotics on their clinical use is clearly demonstrated with the aminoglycosides (chapter 12). 63

DE BROE & ROCH-RAMEL

References 1. 2. 3. 4. 5. 6. 7. 8. 8a. 8b. 8c. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

64

Brezis M, Rosen S, Silva P, Epstein FH. Renal ischemia: a new perspective. Kidney Int 1984; 26: 374-383. Rush GF, Smith JH, Newton JF, Hook JB. Chemically induced nephrotoxicity: role of metabolic activation. CRC Crit Rev Toxicol 1984; 13: 99-160. Hook JB, Hewitt WR. Toxic responses of the kidney. In: Casarett & Doull’s Toxicology. The basic science of poisons, 3rd edition. Klaassen CD, Amdur MO, Doull J (editors). Macmillan, New York 1986; p. 310-329. Tay LK, Bregman CL, Masters BG, Williams PD. Effects of cis-diamminedichloroplatinum (II) on rabbit kidney in vivo and on rabbit renal proximal tubule cells in culture. Cancer Res 1988; 48: 2538-2543. Williams PD. The application of renal cells in culture in studying drug-induced nephrotoxicity. In vitro 1989; 25: 800-805. Williams PD, Rush GF. An evaluation of in vitro models for assessing nephrotoxicity. In: In vitro toxicity testing. Frazier JM (editor). Marcel Dekker Inc, New York/Basel/Hong Kong 1992; p. 85-110. Kastner S, Soose M, Stolte H. Human kidney in in vitro pharmaco-toxicology. In: Human cells in in vitro pharmaco-toxicology: present status within Europe. Rogiers V, Sonck W, Shephard E, Vercruysse A (editors). VUBPress, Brussels 1993; p. 197-238. Brater DC, Sokal PP, Hall SD, McKinney TD. Disposition and dose requirements of drugs in renal insufficiency. In: The kidney: physiology and pathophysiology. Seldin DW, Giebisch G (editors). Raven Press Ltd, New York 1992; p. 3671-3695. Hörbelt M, Wotzlaw C, Sutton TA, Molitoris BA, Philipp T, Kribben A, Fandrey J, Pietruck F, Pietruck F. Organic cation transport in the rat kidney in vivo visualized by time-resolved two-photon microscopy. Kidney Int 2007; 72: 422-429. Koepsell H, Koepsell H. In vivo two-photon fluorescence microscopy opens a new area for investigation of the excretion of cationic drugs in the kidney. Kidney Int 2007; 72: 387-388. John EG & Guignard JP. Development of renal excretion of drugs during ontogeny. In: Neonatal and fetal medicine. Polin RA & Fow WW Eds. Saunders WB, Philadelphia. pp 188-193, 1997. Kragh-Hansen U. Molecular aspects of ligand binding to serum albumin. Pharmacol Rev 1981; 33: 17-53. Routledge PA. The plasma protein binding of basic drugs. Brit J Clin Pharmacol 1986; 22: 499-506. Green TP, Mirkin BL. Resistance of proteinuric rats to furosemide: urinary drug protein bindings as a determinant of drug effect. Life Sci 1980; 26: 623-630. Green TP, Mirkin BL. Furosemide disposition in normal and proteinuric rats: urinary drug protein bindings as a determinant of drug excretion. J Pharmacol Exp Ther 1981; 218: 122-127. Kirchner KA, Voelker JR, Brater DC. Intratubular albumin blunts the response to furosemide - a mechanism for diuretic resistance in the nephrotic syndrome. J Pharmacol Exp Ther 1990; 252: 1097-101. Kirchner KA, Voelker JR, Brater DC. Binding inhibitors restore furosemide potency in tubule fluid containing albumin. Kidney Int 1991; 40: 418-424. Weiner IM. Organic acids and bases and uric acid. The kidney: physiology and pathophysiology. Seldin DW, Giebisch G (editors). Raven Press Ltd, New York 1985; p. 1703-1724. Roch-Ramel F, Roth L, Arnow J, Weiner LM. Salicylate excretion in the rat; free flow micropuncture experiments. J Pharmacol Exp Ther 1978; 207: 737-747. Brater DC. Drugs and the kidney: renal contribution to handling of drugs. In: Handbook of renal therapeutics. Martínez-Maldonado M (editor). Plenum Medical Book Company, New York/London 1983; p. 191-206. Goodman LS, Gilman A, Rall TW, Nies AS, Taylor P (editors). The pharmacological basis of therapeutics. Pergamon Press, New York 1990. Roch-Ramel F, Besseghir K, Murer H. Renal excretion and tubular transport of organic anions and cations. In: Handbook of physiology, section 8: renal physiology. Windhager EE (editor). Oxford University Press, New York/Oxford 1992; p. 2189-2262. Guggino SE, Aronson PS. Paradoxical effects of pyrazinoate and nicotinate on urate transport in dog renal microvillus membranes. J Clin Invest 1985; 76: 543-547. Bessighir K, Roch-Ramel F. Pyrazinoate transport in the isolated perfused rabbit proximal tubule. Pflügers Arch 1986; 407: 643648. Elion GB, Yü TF, Hitchings GH. Renal clearance of oxipurinol, the chief metabolite of allopurinol. Am J Med 1968; 45: 69-77. May DG, Weiner IM. The renal mechanisms for the excretion of m-hydroxybenzoic acid in cebus monkeys: relationship to urate transport. J Pharmacol Exp Ther 1971; 176: 407-417. Van Crugten JT, Sallustio BC, Nation RL, Somogyi A. Renal tubular transport of morphine, morphine-6-glucuronide, and morphine-3-glucuronide in the isolated perfused rat kidney. Drug Metab Disp 1991; 19: 1087-1092. Wright SH, Wunz TM, Wunz TP. A choline transporter in renal brush-border membrane vesicles - energetics and structural specificity. J Membrane Biol 1992; 126: 51-65.

03. Renal handling of drugs and xenobiotics

26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 35a. 35b. 36. 37. 38. 39. 40. 41.

42. 43. 44. 45. 46. 47. 48. 49. 49a.

49b. 50. 51. 52.

Daniel H, Herget M. Cellular and molecular mechanisms of renal peptide transport. (review). Am J Physiol 1997; 273: F1-F8. Shen H, Smith DE, Yang T, Huang YG, Schnermann JB, Brosius FCI. Localization of PEPT1 nand PEPT2 proton-coupled oligopeptide cotransporter mRNA and protein in rat kidney. Am J Physiol 1999; 276: F658-F665. Keppler D, Leier I, Jedlitschky G. Transport of glutathione conjugates and glucuronides by the multidrug resistance proteins MRP1 and MRP2 (review). Biol Chem 1997; 378: 787-791. Inui K, Masuda S, Saito H. Cellular and molecular aspects of drug transport in the kidney. Kidney Int 2000; 58: 944-958. Foulkes EC. Transport of heavy metals by the kidney. Toxicology Letters 1990; 33: 29-31. Templeton DM. Cadmium uptake by cells of renal origin. J Biol Chem 1990; 265: 21764-21770. De Broe ME, Paulus GJ, Verpooten GA, Roels F, Buyssens N, Wedeen RP, Van Hoof F, Tulkens PM. Early effects of gentamicin, tobramycin and amikacin on the human kidney. A prospective, comparative study. Kidney Int 1984; 25: 643-652. Laurent G, Tulkens PM. Aminoglycoside nephrotoxicity: cellular and molecular aspects. ISI Atlas of Science - Pharmacology 1987; 1: 40-44. Lee W and Kim RB. Transporters and renal drug elimination. Annu Rev Pharmacol Toxicol 2004; 44: 137-66. Besseghir K, Roch-Ramel F. Renal excretion of drugs and other xenobiotics. Renal Physiol 1987; 10: 221-241. Dresser, M. J., Leabman, M. K., and Giacomini, K. M. Transporters involved in the elimination of drugs in the kidney: Organic anion transporters and organic cation transporters [Review]. Journal of Pharmaceutical Sciences 90(4), 397-421. 2001. Berkhin, E. B. and Humphreys, M. H. Regulation of renal tubular secretion of organic compounds [Review]. Kidney International 59, 17-30. 2001. Møller JV, Sheikh MI. Renal organic anion transport system: pharmacological, physiological, and biochemical aspects. Pharmacol Rev 1983; 34: 315-358. Ullrich KJ. Specificity of transporters for organic anions and organic cations in the kidney. Biochim Biophys Acta 1994; 1197: 45-62. Burckhardt G, Wolff NA. Structure of renal organic anion and cation transporters (review). Am J Physiol Renal Physiol 2000; 278(6): F853-F866 Sekine T, Cha SH, Endou H. The multispecific organic anion transporter (OAT) family (review). Pflügers Arch 2000; 440(3): 337350. Tune BM. Renal tubular transport and nephrotoxicity of β-lactam antibiotics: structure-activity relationships. Miner Electrolyte Metab 1994; 20: 221-231. Dantzler WH, Evans KK, Groves CE, Welborn JR, North J, Stevens JL, Wright SH. Relation of cysteine conjugate nephrotoxicity to transport by the basolateral organic anion transport system in isolated S2 segments of rabbit proximal renal tubules. J Pharm Exp Ther 1998; 286: 52-60. Berndt WO. The role of transport in chemical nephrotoxicity (review). Tox Pathol 1998; 26: 52-57. Kosoglou T, Vlasses PH. Drug interactions involving renal transport mechanisms: an overview. CICP Ann Pharmacother 1989; 23: 116-22. Pritchard JB, Miller DS. Mechanisms mediating renal secretion of organic anions and cations. Physiol Rev 1993; 73: 765-796. Roch-Ramel F. Renal transport of organic anions (review). Curr Opin Nephrol Hypertens 1998; 7(5): 517-524. Karniski LP, Lotscher M, Fucentese M, Hilfiker H, Biber J, Murer H. Immunolocalization of sat-1 sulfate/oxalate/bicarbonate anion exchanger in the rat kidney. Am J Physiol 1998; 275(1 Pt 2): F79-F87. Wang H, Fei YJ, Kekuda R, Yang-Feng TL, Devoe D, Leibach FH, Prasad PD, Ganapathy V. Structure, function, and genomic organization of human Na+-dependent high-affinity dicarboxylate transporter. Am J Physiol 2000; 278: C1019-C1030. Klein I, Sarkadi B, Varadi A. An inventory of the human ABC proteins (review). Biochim Biophys Acta 1999; 1461: 237-262. Borst P, Evers R, Kool M, Wijnholds J. A family of drug transporters: the multidrug resistance-associated proteins (review). J Nat Cancer Inst 2000; 92: 1295-1302. Masuda S, Terada T, Yonezawa A, Tanihara Y, Kishimoto K, Katsura T, Ogawa O, and Inui K. Identification and functional characterization of a new human kidney–specific H+/organic cation antiporter, kidney-specific multidrug and toxin extrusion 2. J Am Soc Nephrol 2006; 17: 2127-2135. Omote H, Hlasa M, Matsumo T et al. The MATE proteins as fundamental transporters of metabolic and xenobiotic organic cations. Trends Pharmacol Sci 2006; 27: 587-593. Lu R, Kanai N, Bao Y, Wolkoff AW, Schuster VL. Regulation of renal OATP mrna expression by testosterone. Am J Physiol 1996; 270(2 Pt 2): F332-F337. Masuda S, Ibaramoto K, Takeuchi A, Saito H, Hashimoto Y, Inui KI. Cloning and functional characterization of a new multispecific organic anion transporter, OAT-K2, in rat kidney. Mol Pharmacol. 1999; 55(4): 743-752 Masuda S, Takeuchi A, Saito H, Hashimoto Y, Inui K. Functional analysis of rat renal organic anion transporter OAT-K1: bidirectional methotrexate transport in apical membrane. FEBS Letters 1999; 459: 128-132.

65

DE BROE & ROCH-RAMEL

52a. Eraly SA, Blantz RC, Bhatnagar V, Nigam SK. Novel aspects of renal organic anion transporters. Curr Opin Nephrol Hypertens 2003; 12: 551-558. 53. Yabuuchi H, Tamai I, Nezu J, Sakamoto K, Oku A, Shimane M, Sai Y, Tsuji A. Novel membrane transporter OCTN1 mediates multispecific, bidirectional, and pH-dependent transport of organic cations. J Pharmacol Exp Ther 1999; 289: 768-773. 54. Wu X, Huang W, Prasad PD, Seth P, Rajan DP, Leibach FH, Chen JW, Conway SJ, Ganapathy V. Functional characteristics and tissue distribution pattern of organic cation transporter 2 (OCTN2), an organic cation/carnitine transporter. J Pharmacol Exp Ther 1999; 290: 1482-1492. 54a. Huls M, Brown CD, Windass AS, Sayer R, van den Heuvel JJ, Heemskerk S, Russel FG, Masereeuw R. The breast cancer resistance protein transporter ABCG2 is expressed in the human kidney proximal tubule apical membrane. Kidney Int 2007; 73: 220-225. 55. Miller DS, Stewart DE, Pritchard JB. Intracellular compartmentation of organic anions within renal cells. Am J Physiol 1993; 264: R882-R890. 56. Masereeuw R, vandenBergh EJ, Bindels RJ. Characterization of fluorescein transport in isolated proximal tubular cells of the rat: evidence for mitochondrial accumulation. J Pharmacol Exp Ther 1994; 269: 1261-1267. 57. Pajor AM. Sequence and functional characterization of a renal sodium/dicarboxylate cotransporter. J Biol Chem 1995; 270(11): 5779-5785. 58. Lemieux G, Junco E, Perez R, Lemieux C, Allignet E. The metabolic response of the kidney to acute sodium lactate alkalosis. Can J Physiol Pharmacol 1985; 63: 687-692. 59. Miller DS, Barnes DM, Pritchard JB. Confocal microscopic analysis of fluorescein compartmentation within crab urinary bladder cells. Am J Physiol 1994; 267: R16-R25. 60. Miller DS, Pritchard JB. Nocodazole inhibition of organic anion secretion in teleost renal proximal tubules. Am J Physiol 1994; 267: R695-R704. 61. Ullrich KJ, Rumrich G. Luminal transport step of para-aminohippurate (PAH) - transport from pah-loaded proximal tubular cells into the tubular lumen of the rat kidney in vivo. Pflügers Arch 1997; 433(6): 735-743 62. Roch-Ramel F, Werner D, Guisan B. Urate transport in brush-border membrane of human kidney. Am J Physiol 1994; 266: F797F805. 63. Roch-Ramel F, Guisan B, Schild L. Indirect coupling of urate and p-aminohippurate transport to sodium in human brush-border membrane vesicles. Am J Physiol 1996; 270: F165-F272. 64. Schmitt C, Burckhardt G. p-Aminohippurate/2-oxoglutarate exchange in bovine renal brush-border and basolateral membrane vesicles. Pflügers Arch 1993; 423: 280-290. 65. Ullrich KJ, Rumrich G. Renal contraluminal transport systems for organic anions (p-aminohippurate, p-aminohippurate) and organic cations (N1-methyl-nicotinamide, NMeN) do not see the degree of substrate ionization. Pflügers Arch 1992; 421: 286288. 66. Troehler U, Bonjour JP, Fleisch H. Renal secretion of diphosphonates in rats. Kidney Int 1975; 8: 6-13. 67. Lin JH, Chen IW, Deluna FA, Hichens M. Renal handling of alendronate in rats. An uncharacterized renal transport system. Drug Metab Dispos 1992; 20: 608-613. 68. Russel FGM, Vermeulen WG. Effect of substituted benzoylglycines (hippurates) and phenylacetylglycines on p-aminohippurate transport in dog renal membrane vesicles. Pharmaceut Res 1994; 11: 1829-1833. 69. Russel FGM, Heijn M, De Laet RC, Van Ginneken CAM. Effect of substituted benzoates on p-aminohippurate transport in dog renal membrane vesicles. Naunyn-Schmiedeberg’s Arch Pharmacol 1991; 343: 102-107. 70. Hosoyamada M, Sekine T, Kanai Y, Endou H. Molecular cloning and functional expression of a multispecific organic anion transporter from human kidney. Am J Physiol 1999; 276: F122-F128. 71. Van Aubel RAMH, Masereeuw R, Russel FGM. Molecular pharmacology of renal organic anion transporters (review). Am J Physiol Renal Physiol 2000; 279(2): F216-F232. 72. Uwai Y, Saito H, Hashimoto Y, Inui KI. Interaction and transport of thiazide diuretics, loop diuretics, and acetazolamide via rat renal organic anion transporter rOAT1. J Pharmacol Exp Ther 2000; 295: 261-265. 73. Wada S, Tsuda M, Sekine T, Cha SH, Kimura M, Kanai Y, Endou H. Rat multispecific organic anion transporter 1 (rOAT1) transports zidovudine, acyclovir, and other antiviral nucleoside analogs. J Pharmacol Exp Ther 2000; 294: 844-849. 74. Cihlar T, Lin DC, Pritchard JB, Fuller MD, Mendel DB, Sweet DH. The antiviral nucleotide analogs cidofovir and adefovir are novel substrates for human and rat renal organic anion transporter 1. Mol Pharmacol 1999; 56: 570-580. 75. Sekine T, Watanabe N, Hosoyamada M, Kanai Y, Endou H. Expression cloning and characterization of a novel multispecific organic anion transporter. J Biol Chem 1997; 272: 18526-18529. 76. Burckhardt BC, Wolff NA, Burckhardt G. Electrophysiologic characterization of an organic anion transporter cloned from winter flounder kidney (fROAT). J Am Soc Nephrol 2000; 11(1): 9-17.

66

03. Renal handling of drugs and xenobiotics

77.

Race J E, Grassl SM, Williams WJ, Holtzman EJ. Molecular cloning and characterization of two novel human renal organic anion transporters (hOAT1 and hOAT3). Biochem Biophys Res Comm 1999; 255: 508-514. 78. Boner G, Steele TH. Relationship of urate and p-aminohippurate secretion in man. Am J Physiol 19973; 225: 100-104. 79. Uwai Y, Okuda M, Takami K, Hashimoto Y, Inui K. Functional characterization of the rat multispecific organic anion transporter OAT1 mediating basolateral uptake of anionic drugs in the kidney. FEBS Letters 1998; 438: 321-324. 80. Lu R, Chan BS, Schuster VL. Cloning of the human kidney PAH transporter: narrow substrate specificity and regulation by protein kinase C. Am J Physiol 1999; 276: F295-303. 80a. Cha, S. H., Sekine, T., Fukushima, J., Kanai, Y., Kobayashi, Y., Goya, T., and Endou, H. Identification and characterization of human organic anion transporter 3 expressing predominantly in the kidney. Molecular Pharmacology 59(5), 1277-1286. 2001. 81. Sekine T, Cha S, Kanai Y, Endou H. Molecular biology of multispecific organic anion transporter family (OAT family). Clin Exp Nephrol 1999; 3: 237-243. 82. Kusuhara H, Sekine T, Utsunomiya-Tate N, Tsuda M, Kojima R, Cha SH, Sugiyama Y, Kanai Y, Endou H. Molecular cloning and characterization of a new multispecific organic action transporter from rat brain. J Biol Chem 1999; 274: 13675-13680. 83. Nakajima N, Sekine T, Cha SH, Tojo A, Hosoyamada M, Kanai Y, Yan K, Awa S, Endou H. Developmental changes in multispecific organic anion transporter 1 expression in the rat kidney. Kidney Int 2000; 57: 1608-1616. 84. Uchino H, Tamai I, Yamashita K, Minemoto Y, Sai Y, Yabuuchi H, Miyamoto K, Takeda E, Tsuji A. p-aminohippuric acid transport at renal apical membrane mediated by human inorganic phosphate transporter NPT1. Biochem Biophys Res Comm 2000; 270: 254-259. 85. Busch AE, Schuster A, Waldegger S, Wagner CA, Zempel G, Broer S, Biber J, Murer H, Lang F. Expression of a renal type I sodium/ phosphate transporter (NaPi-1) induces a conductance in Xenopus oocytes permeable for organic and inorganic anions. Proc Natl Acad Sci USA 1996; 93(11): 5347-5351. 86. Schaub TP, Kartenbeck J, Konig J, Spring H, Dorsam J, Staehler G, Storkel S, Thon WF, Keppler D. Expression of the MRP2 geneencoded conjugate export pump in human kidney proximal tubules and in renal cell carcinoma. J Am Soc Nephrol 1999; 10: 1159-1169. 87. Bakos E, Evers R, Sinko E, Varadi A, Borst P, Sarkadi B. Interactions of the human multidrug resistance proteins MRP1 and MRP2 with organic anions. Mol Pharmacol 2000; 57: 760-768. 88. Leier I, Hummel-Eisenbeiss J, Cui YH, Keppler D. ATP-dependent para-aminohippurate transport by apical multidrug resistance protein MRP2. Kidney Int 2000; 57: 1636-1642. 89. Van Aubel RAMH, Peters JGP, Masereeuw R, van Os CH, Russel FGM. Multidrug resistance protein Mrp2 mediates ATP-dependent transport of classic renal organic anion p-aminohippurate. Am J Physiol Renal Physiol 2000; 279(4): F713-F717. 90. Peng KC, Cluzeaud F, Bens M, Van Huyen JP, Wioland MA, Lacave R, Vandewalle A. Tissue and cell distribution of the multidrug resistance-associated protein (MRP) in mouse intestine and kidney. J Histochem Cytochem 1999; 47: 757-768. 90a. Smeets PH, van Aubel RA, Wouterse AC, van den Heuvel JJ, and Russel FG. Contribution of multidrug resistance protein 2 (MRP2/ ABCC2) to the renal excretion of p-aminohippurate (PAH) and identification of MRP4 (ABCC4) as a novel PAH transporter. J Am Soc Nephrol 2004; 15: 2828-2835. 91. Abe T, Kakyo M, Sakagami H, Tokui T, Nishio T, Tanemoto M, Nomura H, Hebert SC, Matsuno S, Kondo H, Yawo H. Molecular characterization and tissue distribution of a new organic anion transporter subtype (oatp3) that transports thyroid hormones and taurocholate and comparison with oatp2. J Biol Chem 1998; 273: 22395-22401. 92. Masuda S, Saito H, Inui KI. Interactions of nonsteroidal anti-inflammatory drugs with rat renal organic anion transporter, OAT-K1. J Pharmacol Exp Ther 1997; 283: 1039-1042. 93. Pritchard JB, Sykes DB, Walden R, Miller DS. ATP-dependent transport of tetraethylammonium by endosomes isolated from rat renal cortex. Am J Physiol 1994; 266: F966-F976. 94. Dutt A, Heath LA, Nelson JA. P-glycoprotein and organic cation secretion by the mammalian kidney. J Pharmacol Exp Ther 1994; 269: 1254-1260. 95. Ott RJ, Hui AC, Yuan G, Giacomini KM. Organic cation transport in human renal brush-border membrane vesicles. Am J Physiol 1991; 261: F443-F451. 96. Takahashi Y, Itoh T, Kobayashi M, Sugawara M, Saitoh H, Iseki K, Miyazaki K, Miyazaki S, Takada M, Kawashima Y. The transport mechanism of an organic cation, disopyramide, by brush-border membranes: comparison between renal cortex and small intestine of the rat. J Pharm Pharmacol 1993; 45: 419-424. 97. Wright SH, Wunz TM, Wunz TP. Structure and interaction of inhibitors with the tetraethylammonium/H+ exchanger of rabbit renal brush border membranes. Pflügers Arch 1992; 429: 313-324. 98. David C, Rumrich G, Ullrich KJ. Luminal transport system for H+/organic cations in the rat proximal tubule. Kinetics, dependence on pH; specificity as compared with the contraluminal organic cation-transport system. Pflüugers Arch 1995; 430(4): 477-492.

67

DE BROE & ROCH-RAMEL

99.

Wright S H, Wunz TM. Influence of substrate structure on turnover of the organic cation/H+ exchanger of the renal luminal membrane. Pflügers Arch 1998; 436: 469-477. 100. Wright SH, Wunz TM. Influence of substrate structure on substrate binding to the renal organic cation H+ exchanger. Pflügers Arch 1999; 437: 603-610. 101. Miyamoto Y, Tiruppathi C, Ganapathy V, Leibach FH. Multiple transport systems for organic cations in renal brush-border membranes vesicles. Am J Physiol 1989; 256: F540-F548. 102. Ford JM, Hait WN. Pharmacology of drugs that alter multidrug resistance in cancer. Pharmacol Rev 1990; 42: 156-199. 103. Speeg KV, Maldonado AL, Liaci J, Muirhead D. Effect of cyclosporine on colchicine secretion by the kidney multidrug transporter studied in vivo. J Pharmacol Exp Ther 1992; 261: 50-55. 104. De Lannoy IAM, Mandin RS, Silverman M. Renal secretion of vinblastine, vincristine and colchicine in vivo. J Pharmacol Exp Ther 1994; 268: 388-395. 105. Schramm U, Fricker G, Wenger R, Miller DS. P-glycoprotein-mediated secretion of a fluorescent cyclosporin analogue by teleost renal proximal tubules. Am J Physiol 1995; 268: F46-F52. 106. Horio M, Pastan I, Gottesma MM, Handler JS. Transepithelial transport of vinblastine by kidney-derived cell lines. Application of a new kinetic model to estimate in situ Km of the pump. Biochim Biophys Acta 1990; 1027: 116-122. 107. Koren G. Clinical pharmacokinetic significance of the renal tubular secretion of digoxin. Clin Pharmacokin 1987; 13: 334-343. 107a. Koziolek MJ, Riess R, Geiger H et al. Expression of multidrug resistance P-glycoprotein in kidney allografts from cyclosporine A-treated patients. Kidney Int 2001; 60: 156-166. 107b. Saeki T, Ueda K, Tanigawara Y et 01. Human P-glycoprotein transports cyclosporin A and FK506. J Biol Chem 1993; 268: 60776080. 107c. Del Moral RG, Olrno A, Osuna A et a/. Role of P-glycoprotein in chronic cyclosporine nephrotoxicity and its relationship to intrarenal angiotensin II deposits. Transplant Proc 1998; 30: 2014-2016. 107d. Garcia del Moral R, O’Valle F, Andujar M et al. Relationship between P-glycoprotein expression and cyclosporin A in kidney. An immunohistological and cell culture study. Am J Path01 1995: 146: 398-408. 107e. Hauser IA, Schaeffeler E, Gauer S et al. ABCBl genotype of the donor but not of the recipient is a major risk factor for cyclosporlnerelated nephrotoxicity after renal transplantation. J Am Soc Nephrol 2005; 16: 1501-1511. 107f. Miller DS, Fricker G, Drewe J. P-Glycoprotein-mediated transport of a fluorescent rapamycin derivative in renal proximal tubule. J Pharmacol Exp Ther 1997; 282: 440-444. 107g. Kahan BD. Efficacy of sirolimus compared with azathioprine for reduction of acute renal allograft rejection: a randomised multicentre study. The Rapamune US Study Group. Lancet 2000; 356: 194-202. 107h. Pallet N, Thervet E, Le Corre D et al. Rapamycin inhibits human renal epithelial cell proliferation: effect on cyclin D3 mRNA expression and stability. Kidney Int 2005; 67: 2422-2433. 108. Koepsell H, Gorboulev V, Arndt P. Molecular pharmacology of organic cation transporters in kidney (review). J Membr Biol 1999; 167: 103-117. 109. Zhang L, Brett CM, Giacomini KM. Role of organic cation transporters in drug absorption and elimination (review). Annual Rev Pharmacol Toxicol 1998; 38: 431-460. 110. Grundemann D, Gorboulev V, Gambaryan S, Veyhl M, Koepsell H. Drug excretion mediated by a new prototype of polyspecific transporter. Nature 1994; 372: 549-552. 111. Okuda M, Urakami Y, Saito H, Inui K. Molecular mechanisms of organic cation transport in OCT2-expressing Xenopus oocytes. Biochim Biophys Acta 1999; 1417: 224-231. 112. Zhang L, Gorset W, Washington CB, Blaschke TF, Kroetz DL, Giacomini KM. Interactions of HIV protease inhibitors with a human organic cation transporter in a mammalian expression system. Drug Metab Dispos 2000; 28: 329-334. 113. Dresser MJ, Gray AT, Giacomini KM. Kinetic and selectivity differences between rodent, rabbit, and human organic cation transporters (OCT1). J Pharmacol Exp Ther 2000; 292: 1146-1152. 114. Karbach U, Kricke J, Meyer-Wentrup F, Gorboulev V, Volk C, Loffing-Cueni D, Kaissling B, Bachmann S, Koepsell H. Localization of organic cation transporters OCT1 and OCT2 in rat kidney. Am J Physiol Renal Physiol 2000; 279(4): F679-F687. 115. Sugawara-Yokoo M, Urakami Y, Koyama H, Fujikura K, Masuda S, Saito H, Naruse T, Inui K, Takata K. Differential localization of organic cation transporters rOCT1 and rOCT2 in the basolateral membrane of rat kidney proximal tubules. Histochem Cell Biol 2000; 114: 175-180. 116. Urakami Y, Nakamura N, Takahashi K, Okuda M, Saito H, Hashimoto Y, Inui K. Gender differences in expression of organic cation transporter OCT2 rat kidney. FEBS Letters 1999; 461: 339-342. 117. Wu XA, Huang W, Ganapathy ME, Wang HP, Kekuda R, Conway SJ, Leibach FH, Ganapathy V. Structure, function, and regional distribution of the organic cation transporter OCT3 in the kidney. Am J Physiol Renal Physiol 2000; 279: F449-F458.

68

03. Renal handling of drugs and xenobiotics

118. Grundemann D, Liebich G, Kiefer N, Koster S, Schomig E. Selective substrates for non-neuronal monoamine transporters. Mol Pharmacol 1999; 56(1): 1-10. 119. Tamai I, Yabuuchi H, Nezu J, Sai Y, Oku A, Shimane M, Tsuji A. Cloning and characterization of a novel human ph-dependent organic cation transporter, OCTN1. FEBS Letters 1997; 419: 107-111. 120. Ganapathy ME, Huang W, Rajan DP, Carter AL, Sugawara M, Iseki K, Leibach FH, Ganapathy V. β-lactam antibiotics as substrates for OCTN2, an organic cation/carnitine transporter. J Biol Chem 2000; 275: 1699-1707. 120a. Lang F. Regulating renal drug elimination? J Am Soc Nephrol 2005; 16: 1535-1536. 120b. Zaman GJ, Versantvoort CH, Smit JJ, et al. Analysis of the expression of MRP, the gene for a new putative transmembrane drug transporter, in human multidrug resistant lung cancer cell lines. Cancer Res 1993; 53: 1747-1750. 120c. Wijnholds J, Scheffer GJ, van der Valk M, van der Valk P, Beijnen JH, Scheper RJ, Borst P. Multidrug resistance protein I protects the oropharyngeal mucosal layer and the testicular tubules against drug-induced damage. J Exp Med 1998; 188: 797-808. 120d. Peng KC, Cluzeaud F, Bens M et al. Tissue and cell distribution of the multidrug resistance-associated protein (MRP) on mouse intestine and kidney. J Histochem Cytochem 1999; 47: 757-767. 120e. Schaub TP, Kartenbeck J, Konig J, Vogel O, Witzgall R, Kriz W, Keppler D. Expression of the conjugate export pump encoded by the mrp2 gene in the apical membrane of kidney proximal tubules. J Am Soc Nephrol 1997; 8: 1213-1221. 120f. Kool M, van der Linden M, de Haas M, et al. MRP3, an organic anion transporter able to transport anti-cancer drugs. Proc Natl Acad Sci USA 1999; 96: 6914-6919. 120g. Konig J, Rost D, Cui Y, Keppler D. Characterization of the human multidrug resistance protein isoform MRP3 localized to the basolateral hepatocyte membrance. Hepatology 1999; 29: 1156-1163. 121. Wagner CA, Lukewille U, Kaltenbach S, Moschen I, Broer A, Risler T, Broer S, Lang F. Functional and pharmacological characterization of human Na+-carnitine cotransporter hOCTN2. Am J Physiol Renal Physiol 2000; 279: F584-F591. 122. Seth P, Wu X, Huang W, Leibach FH, Ganapathy V. Mutations in novel organic cation transporter (OCTN2), an organic cation/carnitine transporter, with differential effects on the organic cation transport function and the carnitine transport function. J Biol Chem 1999; 274: 33388-33392. 123. Bowman RH. Renal secretion of [35S] furosemide and its depression by albumin binding. Am J Physiol 1975; 229: 93-98. 124. Koschier FJ, Acara M. Transport of 2, 4, 5-trichlorophenoxyacetate in the isolated, perfused rat kidney. J Pharmacol Exp Ther 1979; 208: 287-293. 125. Webb DE, Edwards RM, Grantham JJ. Dependence of proximal tubule p-aminohippurate secretion on serum proteins and metabolic substrates. Am J Physiol 1986; 251: F619-F626. 126. Grantham JJ, Kennedy J, Cowley B. Tubule urate and p-aminohippurate transport: sensitivity and specificity of serum protein inhibition. Am J Physiol 1987; 252: F683-F690. 127. Rodrigues de Miranda JF, Hilbers CW. A nuclear magnetic resonance study of the kinetics of the binding of the renal contrast medium acetrizoate to albumin. Mol Pharmacol 1976; 12: 279-290. 128. Ochwadt BK, Pitts RF. Disparity between phenol red and diodrast clearance in the dog. Am J Physiol 1956; 187: 318-322. 129. Honari J, Blair AD, Cutler RE. Effects of probenecid on furosemide kinetics in man. Clin Pharmacol Ther 1977; 22: 395-401. 130. Homeida M, Roberts C, Branch RA. Influence of probenecid and spironolactone on furosemide kinetics in man. Clin Pharmacol Ther 1977; 22: 402-409. 131. Brater DC. Pharmacodynamic considerations in the use of diuretics. Annual Rev Pharmacol Toxicol 1983; 23: 45-62. 131a. Li M, Anderson GD, Wang J. Drug-drug interactions involving membrane transporters in the human kideny. Expert Opin Drug Metab Toxicol 2006; 2: 505-532. 132. Jacobs C, Kaubisch S, Halsey J, Lum BL, Gosland M, Coleman CN, Sikic BI. The use of probenecid as a chemoprotector against cisplatin nephrotoxicity. Cancer 1991; 67: 1518-1524. 133. Ban M, Hettich D, Huguet N. Nephrotoxicity mechanism of cis-plantinum (II) diamine dichloride in mice. Toxicol Letters 1994; 71: 161-168. 134. Klein J, Bentur Y, Cheung D, Moselhy G, Koren G. Renal handling of cisplatin: interactions with organic anionas and cations in the dog. Clin Inv Med 1991; 14: 388-394. 135. Tracy TS, Krohn K, Jones DR, Bradly JD, Hall SD, Brates DC. The effects of a xalicylate, ibuprofen, and naproxen on the disposition in patients with rheumatoid arthritis. Eur J Clin Pharmacol 1992; 42: 121-125. 136. Frenia ML, Long KS. Methotrexate and nonsteroidal antiinflammatory drug interactions. Ann Pharmacother 1992; 26: 234-237. 137. Nierenberg DW. Competitive inhibition of methotrexate accumulation in rabbit kidney slices by nonsteroidal anti-inflammatory drugs. J Pharmacol Exp Ther 1983; 226: 1-6. 138. Fromm MF, Kim RB, Stein CM, Wilkinson GR, Roden DM. Inhibition of P-glycoprotein-mediated drug transport - A unifying mechanism to explain the interaction between digoxin and quinidine. Circulation 1999; 99: 552-557.

69

DE BROE & ROCH-RAMEL

139. Verschraagen M, Koks CHW, Schellens JHM, Beijnen JH. P-glycoprotein system as a determinant of drug interactions: the case of digoxin-verapamil. Pharmacol Res 1999; 40: 301-306. 140. De Lannoy IAM, Koren G, Klein J, Charuk J, Silverman M. Cyclosporin and quinidine inhibition of renal digoxin excretion - evidence for luminal secretion of digoxin. Am J Physiol 1992; 263(4 Pt 2):F613-F622. 141. De Lannoy IAM, Mandin RS, Silverman M. Renal Secretion of Vinblastine, Vincristine and Colchicine in vivo. J Pharmacol Exp Ther 1994; 268: 388-395. 142. Okamura N, Hirai M, Tanigawara Y, Tanaka K, Yasuhara M, Ueda K, Komano T, Hori R. Digoxin-cyclosporin A interaction: modulation of the multidrug transporter P-glycoprotein in the kidney. J Pharmacol Exp Ther 1993; 266: 1614-1619. 143. Woodland C, Verjee Z, Giesbrecht E, Koren G, Ito S. The digoxin-propafenone interaction - characterization of a mechanism using renal tubular cell monolayers. J Pharmacol Exp Ther 1997; 283: 39-45. 144. Nierenberg DW. Drug inhibition of penicillin tubular secretion: concoidance between in vitro and clinical findings. J Pharmacol Exp Ther 1987; 240: 712-716. 145. Dupuis L, Loren G, Shore A, Silverman ED, Laxer RM. Methotrexate non-steroidal anti-inflammatory drug interaction in children with arthritis. J Rheumat 1990; 17: 1469-1473. 146. Banditt P, Meyer FP, Walther H. Influence of cimetidine on the pharmacokinetics of other drugs. Pharmazie 1990; 45: 11-16. 147. Somogyi A, McLean A, Heinzow B. Cimetidine-procainamide pharmacokinetic interaction in man: evidence of competition for tubular secretion of basic drugs. Eur J Clin Pharmacol 1983; 25: 339-345. 148. Christian CJ, Meredith CG, Speeg KJ. Cimetidine inhibits renal procainamide clearance. Clin Pharmacol Ther 1984; 36: 221-227. 149. Van Crugten J, Bochner F, Keal J, Somogyi A. Selectivety of the cimetidine-induced alterations in the renal handling of organic substrates in humans. Studies with anionic, cationic and zwitterionic drugs. J Pharmacol Exp Ther 1986; 236: 481-487. 150. Gisclon LG, Boyd RA, Williams RL, Giacomini KM. The effect of probenecid on the renal elimination of cimetidine. Clin Pharmacol Exp Ther 1989; 45: 444-452. 151. Inotsume N, Nishimua M, Nakano M, Fujiyama S, Sato T. The inhibitory effect of probenecid on renal excretion of famotidine in young, healthy volunteers. J Clin Pharmacol 1990; 30: 50-56. 152. Hsyu PH, Gisclon LG, Hui AC, Giacomini KM. Interactions of organic anions with the organic cation transporter in renal BBMV. Am J Physiol 1988; 254: F56-F61. 153. Ott RJ, Hui C, Giacomine KM. Mechanisms of interactions between organic anions and the organic cation transporter in renal brush border membrane vesicles. Biochem Pharmacol 1990; 40: 659-661. 154. McKinney TD, Myers P, Speeg JKV. Cimetidine secretion by rabbit renal tubules in vitro. Am J Physiol 1981; 241: F69-F76. 155. Brandle E, Greven J. Transport of cimetidine across the basolateral membrane of rabbit kidney proximal tubules: interaction with organic anions. Pharmacology 1992; 45: 231-240. 156. Schück O. Tubular secretion of creatinine and its plasma concentration. Int J Clin Pharmacol Ther Toxicol 1990; 28: 127-128. 157. Sokol PP. Effect of DQ-2556, a new cephalosporin, on organic ion transport in renal plasma membrane vesicles from the dog, rabbit and rat. J Pharmacol Exp Ther 1990; 255: 436-441. 158. Ullrich KJ, Rumrich G, David C, Fritzsch G. Bisubstrates-substances that interact with renal contraluminal organic anion and organic cation transport systems. A. amines, piperidines, piperazines, azepines, pyridines, quinolines, imidazoles, thiazoles, guanidines and hydrazines. Pflügers Arch 1993; 425: 280-299. 159. Chatton JY, Odone M, Besseghir K, Roch-Ramel F. Renal secretion of 3’-azido-3’-deoxythymidine by the rat. J Pharmacol Exp Ther 1990; 255: 140-145. 160. Griffiths DA, Hall SD, Sokol PP. Interaction of 3’-azido-3’-deoxythymidine with organic ion transport in rat renal basolateral membrane vesicles. J Pharmacol Exp Ther 1991; 257: 149-155. 161. Griffiths DA, Hall SD, Sokol PP. Effect of 3’-azido-3’-deoxythymidine (AZT) on organic ion transport in rat renal brush border membrane vesicles. J Pharmacol Exp Ther 1992; 260: 128-133. 162. Anders MW. Metabolism of drugs by the kidney. Kidney Int 1980; 18: 636-647. 163. Bekersky I, Popick A. Metabolism of bumetanide by the isolated perfused rat kidney. Drug Metab Dispos 1983; 11: 512-513. 164. Bekersky I, Colburn WA. Acetylation of sulfisoxazole by the isolated perfused rat kidney. J Pharm Sci 1980; 69: 1359. 165. Bekersky I, Colburn WA, Fishman L, Kaplan SA. Metabolism of salicylic acid in the isolated perfused rat kidney: interconversion of salicyluric and salicylic acids. Drug Metab Dispos 1980; 8: 319-324. 166. DeLannoy IAM, Nespeca R, Pang KS. Renal handling of enalapril and enalaprilat: studies in the isolated red blood cell-perfused rat kidney. J Pharmacol Exp Ther 1989; 251: 1211-1222. 167. Elseviers MM, De Broe ME. Epidemiology of analgesic nephropathy. J Nephrol 1992; 5: 94-98. 168. McMurtry RJ, Snodgrass WR, Mitchell JR. Renal necrosis glutathione depletion and covalent binding after acetaminophen. J Toxicol Appl Pharmacol 1978; 46: 87-100.

70

03. Renal handling of drugs and xenobiotics

169. Mudge GH, Gemborys MW, Duggin GG. Covalent binding of metabolites of acetaminophen to kidney protein and depletion of renal glutathione. J Pharmacol Exp Ther 1978: 206: 218-226. 170. Kleinman JG, Breitenfield RV, Roth DA. Acute renal failure associated with acetaminophen ingestion: report of a case and review of the literature. Clin Nephrol 1980; 14: 201-205. 171. Szefler SJ, Acara M. Isoproterenol excretion and metabolism in the isolated perfused rat kidney. J Pharmacol Exp Ther 1979; 210: 295-300. 172. Rennick B, Ziemniak J, Smith I, Taylor M, Acara M. Tubular transport and metabolism of cimetidine in chicken kidneys. J Pharmacol Exp Ther 1984; 228: 387-392. 173. Bessighir K, Pearce LB, Rennick B. Renal tubular transport and metabolism of organic cations by the rabbit. Am J Physiol 1981; 241: F308-F314. 174. Watrous WM, May DG, Fujimoto JM. Mechanism of the renal tubular transport of morphine and morphine ethereal sulfate in the chicken. J Pharmacol Exp Ther 1970; 172: 224-229. 175. Hook JB, MacCormack KM, Kluwe WM. Biochemical mechanisms of nephrotoxicity. In: Reviews in biochemical toxicology, vol 1. Hodgson, Bend, Philpot (editors). Elsevier/North-Holland, New York 1979; p. 53-78. 176. Jacqz E, Ward S, Johnson R, Schenker S, Gerkens J, Branch RA. Extrahepatic glucuronidation of morphine in the dog. Drug Metab Dispos 1986; 14: 627-630. 177. Meffin PJ, Zilm DM, Veendendaal JR. Reduced clofibric acid clearance in renal dysfunction is due to a futile cycle. J Pharmacol Exp Ther 1983; 227: 732-738. 178. Gugler R, Kurten JW, Jensen CJ, Klehr U, Hartlapp J. Clofigrate disposition in renal failure and acute and chronic liver disease. Eur J Clin Pharmacol 1979; 15: 341-347. 179. Verbeeck R, Tjandramaga TB, Mullie A, Verbesselt R, Verberckmoes R, De Schepper PJ. Biotransformation of diflunisal and renal excretion of its glucoronides in renal insufficiency. Brit J Clin Pharmacol 1979; 7: 273-282. 180. Eriksson LO, Wahlin-Boll E, Odar-Cederlof I, Lindholm L, Melander A. Influence of renal failure, rheumatoid arthritis and old age on the pharmacokinetics of diflunisal. Eur J Clin Pharmacol 1989; 36: 165-174. 181. Aronoff GR, Ozawa T, DeSante KA, Nash JF, Ridolfo AS. Benoxaprofen kinetics in renal impairment. Clin Phamacol Ther 1982; 32: 190-194. 182. Advenier C, Roux A, Gobert C, Massias P, Varaquaux O, Flouvat B. Pharmacokinetics of ketoprofen in the elderly. Brit J Clin Pharmacol 1983; 16: 65-70.

71

04

Pharmacological aspects of nephrotoxicity Marisa D. COVINGTON and Rick G. SCHNELLMANN Medical University of South Carolina, Charleston, SC, USA

Introduction ____________________________________________________________ 73 Glomerulus _____________________________________________________________ 74 Proximal tubule __________________________________________________________ 75 Distal nephron___________________________________________________________ 76 Cellular injury ___________________________________________________________ 77 Renal repair and regeneration ______________________________________________ 78 References ______________________________________________________________ 79

Introduction

B

ecause the kidney is vital to total body homeostasis, a toxic insult to the kidney can have profound effects – an insult of sufficient severity can permanently damage renal tissue, necessitating chronic dialysis or kidney transplantation. Such susceptibility to various toxicants is due to several functional properties of the kidney. First, the kidney receives approximately one-quarter of the total body blood flow to support renal function, including glomerular filtration, permitting the delivery of high levels of toxicants. The absorption of water and solutes along the nephron concentrates the tubular fluid, thereby exposing tubular epithelial cells to greater concentrations of toxicants.

The high metabolic rate and work load of renal cells increases its susceptibility to toxicants. Furthermore, the kidney possesses biotransformation enzymes that can result in formation of toxic metabolites and reactive intermediates which can damage renal macromolecules. Because the nephron has specialized transporters for reabsorption and excretion, toxicants can enter and accumulate within renal cells, leading to nephrotoxicity. Finally, the unique functions of the varied segments along the nephron impart different susceptibilities to toxicants in the kidney, complicating the potential toxicities and subsequent renal damage via a variety of mechanisms. In this chapter, we will review some of these sites and mechanisms of nephrotoxicity.

COVINGTON & SCHNELLMANN

Glomerulus The glomerulus, a specialized capillary bed composed of endothelial cells, is the filtering unit of the kidney. The glomerular capillary wall forms both a charge- and size-selective barrier that prevents passage of plasma proteins and results in the formation of an ultrafiltrate. Because the glomerulus is the first structure encountered in the nephron, it is the initial site of toxicant exposure in the kidney. Nephrons are functionally integrated and as a result, toxicantinduced damage to the glomerulus not only impairs glomerular function, but also affects the function of the entire nephron. Toxicants can decrease the glomerular filtration rate (GFR) by increasing afferent arteriolar resistance, resulting in a decrease in hydrostatic pressure. Toxicants can also decrease the glomerular surface area available for filtration by decreasing the size and number of endothelial fenestrae or altering the number of anionic charges on the glomerular structural elements, allowing passage and eventual urinary excretion of polyanionic and high-molecular weight proteins. For example, puromycin aminonucleoside exposure results in a loss of membrane anionic charges, permitting the passage of negatively charged proteins through the glomerulus, and resulting in proteinuria [1, 2]. See Figure 1 for other examples. Chemically induced glomerular damage also can occur without significant loss of glomerular structural integrity. For example, mild renal ischemia and reperfusion results in formation of reactive oxygen species (ROS), proteinuria, and loss of charged glomerular structures with no apparent change in morphology [3]. Toxicants such as cyclosporine A and amphotericin B directly decrease renal circulation through injury of renal vessels and decrease GFR [4, 5]. Similarly, gentamicin interacts with anionic sites on the endothelial cells to decrease GFR and renal blood flow [6]. This diminished blood flow also decreases the delivery of oxygen and other critical metabolites to the tubules, further enhancing nephrotoxicity. These drugs create glomerular renal dysfunction with few morphological alterations in the glomeruli. An important class of filtered molecules, soluble immune complexes, are generated after antibody responses to antigens which can be derived from drugs 74

PROXIMAL TUBULE

GLOMERULUS Puromycin Cyclosporine A Penicillamine Heavy metals Hydrocarbons Colloidal gold/ gold salts NSAIDs Mitomycin C

Cisplatin Acyclic nucleoside phosphonates Heavy metals Aminoglycosides 5-Aminosalicylic acid Indinavir Chloroform Acetaminophen

DISTAL NEPHRON Amphotericin B Cisplatin Analgesics

VASOCONSTRICTION Amphotericin B Cyclosporine A NSAIDs Radiocontrast agents Aminoglycosides

COLLECTING DUCT LOOP of HENLE

Figure 1. Nephrotoxic targets along the nephron.

and toxicants. Although soluble immune complexes are not always associated with pathology, some complexes can be deposited in the glomerulus and can subsequently activate complement, initiating a sequence of inflammatory events which may include recruitment of inflammatory cells, release of inflammatory mediators and enzymes, and destruction of glomerular structures. Macrophages and neutrophils are observed in the glomeruli in membranous glomerulonephritis, and the release of cytokines and ROS contribute to glomerular injury [7]. Recently, the activation and secretion of calpains, calcium-activated cysteine proteases, have been shown to participate in the development of immune glomerular injury [8]. A chemical may adhere to a native protein to produce an antigen and elicit an antibody response. For example, colloidal gold and gold salts, which are used to treat rheumatoid arthritis, induce membranous nephropathy with numerous electron-dense deposits

04. Pharmacological aspects of nephrotoxicity

on glomerular basement membranes [9]. Penicillamine, a drug used in chelation therapy and rheumatoid arthritis, produced similar immunoglobulin- and complement-containing deposits in glomerular basement membranes. Similarly, glomerulonephritis was observed after exposure to heavy metals, hydrocarbons, and captopril [10-13].

Proximal tubule Proximal tubular injury is the most common toxicant-induced renal injury (see Figure 1 for examples). The proximal renal tubules are vulnerable to direct toxic effects of chemicals because of their absorption and secretion functions. Proximal tubular cells contain transporters for organic anions and cations, low-molecular weight proteins, glutathione (GSH) conjugates, and metals, which can result in the accumulation of chemicals and subsequent toxicity. Amphotericin B can bind to low-density lipoproteins and be internalized through low-density lipoprotein receptors [14]. Acyclic nucleoside phosphonates, such as cidofovir, adefovir, and tenofovir, are transported by the organic anion transporter-1 [15]. Although the inherent functionality of the kidney often means that toxicant concentrations are high in renal cells, the nephrotoxic effects are actually dependent upon the intrinsic reactivity with cellular and molecular targets. Absorption and secretion in the nephron are functions of high energy demand; thus, these cells have elevated rates of oxidative metabolism. Therefore, chemicals that directly or indirectly disturb renal cell energy metabolism will result in cell injury and consequent renal dysfunction. For instance, heavy metals such as mercuric chloride alter mitochondrial function and morphology prior to tubular necrosis [16, 17]. Mitochondrial dysfunction is also observed with exposures to lead, aminoglycosides, and cephalosporins [17, 18]. Drugs that injure proximal tubules include 5-aminosalicylic acid (5-ASA), which is used to treat inflammatory bowel disease, and adefovir, a nucleoside reverse transcriptase inhibitor. Although the exact mechanism is unknown, 5-ASA has been shown to cause renal damage due to the uncoupling of oxidative phosphorylation and inhibition of prostaglandin synthesis [19]. Such a mechanism of action is thought to be similar to toxicities of other salicylates [20]. Adefovir has been reported to induce acute tubu-

lar necrosis and produce severe structural alterations in proximal tubular mitochondria [21]. Toxicity is thought to be mediated by direct effects on mitochondrial DNA replication, including the synthesis of cytochrome C oxidase [22], which is an important respiratory chain enzyme in the mitochondria, and its inhibition leads to mitochondrial dysfunction and loss of ATP. As the glomerular filtrate proceeds down the tubule, the filtrate becomes increasingly concentrated and the pH of the filtrate becomes more acidic. Therefore chemicals with pH-dependent solubility have the potential to precipitate and cause tubular obstruction, resulting in local interstitial inflammation, granuloma formation, and fibrosis. Indinavir is a protease inhibitor used in the treatment of immunodeficiency virus that has been reported to cause renal toxicity. The solubility of indinavir is pH- and flow-dependent, and some patients treated with indinavir form urinary crystals that obstruct tubules, leading to inflammation or granuloma formation, resulting in renal failure [23]. Cytochrome P-450 and cysteine conjugate β-lyse are primarily localized in the proximal tubules, and these enzymes also contribute to the susceptibility of the proximal tubule to toxicant injury. Specifically, widely used industrial solvents such as chloroform produce tubular nephrotoxicity via cytochrome P450 activation, and haloalkanes and haloalkenes (e.g. trichloroethylene) are rendered toxic by cysteine conjugate β-lyse activation [24, 24a]. In addition, overdoses of acetaminophen (APAP) cause nephrotoxicity that is characterized by proximal tubular necrosis [25]. APAP undergoes cytochrome P-450-mediated activation to produce a toxic electrophile, N-acetyl-p-benzoquinoneimine (NAPQI) [25a]. Although NAPQI is extremely reactive, it is detoxified by conjugation with reduced GSH unless NAPQI is formed in excess of the cellular capacity for GSH conjugation. The excess NAPQI is available to bind to critical cellular proteins and to induce oxidative stress, resulting in disruption of cellular homeostasis and tubular injury [26]. It is now recognized that GSH conjugation plays a critical role in chemical-induced nephrotoxicity and, in many cases, may be responsible for selective targeting and damage to the kidney versus other organ systems (Table 1). The capacity of the kidney to process and activate GSH conjugates is extensive and results in the release of toxic species within the proximal tubular cells. For example, haloalkenes and quinones are con75

COVINGTON & SCHNELLMANN

Table 1. Chemicals that produce nephrotoxicity through their glutathione/cysteine conjugates. •

cisplatin

•

acetaminophen

•

sevoflurane

•

hydroquinone

•

bromohydroquinone

•

cadmium

•

mercury

•

trichloroethylene

•

tetrafluoroethylene

•

hexachlorobutadiene

O C X

S

NH

CH2

COOH

CH3 NH

C

CH2

CH2

O

CH

COOH

NH2

?-GT O C X

S

NH

CH2

COOH

CH3 NH2 Aminodipeptidase COOH C

jugated to GSH in the liver, circulate to the kidney, and are metabolized by γ-glutamyl transpeptidase (γ–GT) to produce the cysteinyl-glycine conjugate (Figure 2) [27, 28]. The cysteinyl-glycine conjugate is further metabolized extracellularly by aminodipeptidases to cysteine conjugates [29]. The cysteine conjugates are then transported into the proximal tubule cells, where they are further metabolized into highly reactive thiols by cysteine-S-conjugate β-lyase [30]. The reactive thiols bind to cellular macromolecules, ultimately triggering cell death. In addition, the cysteine conjugates initiate oxidative stress and lipid peroxidation [31, 31a]. Similarly, the nephrotoxicity of quinone-GSH conjugates arises from their ability to undergo redox cycling [27]. Although cisplatin is not a substrate for γ–GT or cysteine-S-conjugate β-lyase, it has been shown to form GSH conjugates spontaneously in solution [32]. Cisplatin-GSH conjugates may be important in targeting cisplatin to the kidney and its resulting nephrotoxicity because γ–GT is necessary for the toxicity of the cisplatin conjugates, suggesting that metabolism of cisplatin in proximal tubule cells is required for nephrotoxicity [33, 34]. Furthermore, in vivo studies support the hypothesis that formation of a cisplatin-GSH conjugate is an essential component of nephrotoxicity of cisplatin [35-37]. Finally, the nephrotoxicity of inorganic mercury (i.e. mercuric chloride) has been shown to be the result of a GSH or cysteine conjugate. In a series of experiments, Zalups has shown that mercuric chloride in the blood is conjugated to cysteine and GSH as mono- or di-substituted conjugates and to serum proteins [38]. 76

X

S

CH3 NH2 Cysteine-S-conjugate Beta-lyase

X

S

Figure 2. Activation of glutathione-conjugates to reactive thiols. Halogenated alkenes form glutathione-S-conjugates and are metabolized to nephrotoxins via this pathway. This pathway results in the production of unstable reactive thiols, which are toxic. The X represents the alkene. Adapted from Townsend et al., 2003 [34].

Uptake of these cysteine and GSH conjugates occurs across the apical membrane through Na+-dependent and -independent amino acid transporters and across the basolateral membrane through an amino acid transporter and/or the organic anion transporter (OAT1). These examples provide strong evidence that the nephrotoxicity of numerous xenobiotics is dependent on GSH conjugation-mediated delivery to the kidney.

Distal nephron Although most nephrotoxicity occurs in the proximal part of the nephron, some chemicals damage distal structures. The function of these structures facilitates their vulnerability to toxicants. For instance, the loop of Henle is critical to the process of urinary concentration and therefore utilizes relatively high rates of Na+, K+-ATPase activity and oxygen demand. This, and the fact that oxygen supply to the medulla is minimally sufficient to meet physiological needs, contributes to

04. Pharmacological aspects of nephrotoxicity

the susceptibility of the loop of Henle to hypoxic injury. For example, amphotericin B increases the tubular work load in the loop of Henle, intensifying hypoxic injury [39]. The final regulation of urinary volume and composition occurs in the distal tubule and collecting duct. Water permeability of the medullary collecting duct is controlled by hypertonicity and the action of antidiuretic hormone (ADH). Chemicals that increase medullary blood flow or interfere with ADH synthesis, secretion, or action will impair the concentration of urine. Drugs that have been associated with distal nephron injury impair the concentrating ability in the thick ascending limb and/or the collecting duct resulting in an ADH-resistant polyuria. The distal tubular epithelial cells are tightly bound, forming a strong barrier. Amphotericin B inhibits reabsorption in the distal nephron through its ability to form transmembrane pores and disrupt membrane permeability [40]. Cisplatin also induces polyuria , but the mechanism is not completely understood; although [41] Safirstein and Deray suggested that the polyuria arises through a vasopressin and prostaglandin inhibitor pathway. The renal papilla is the target of analgesic abuse or the excessive ingestion of analgesics, which are often mixed with caffeine or alcohol, and results in papillary necrosis and chronic renal failure [42, 43]. Analgesics also can inhibit the vasodilatory effects of prostaglandin, predisposing the renal papillae with its already tenuous blood-supply to further ischemia and damage [44]. Because there is a high papillary concentration of toxicant, direct cellular insults of toxicants would be detrimental to the renal papilla. This is true for analgesics that cause injury by covalently binding to cells and causing oxidative damage [44].

Cellular injury The nature and the intensity of toxicant-induced insults to the kidney determine the severity of renal damage. Most nephrotoxic chemicals target renal epithelial cells and produce cell death, which is thought to occur by apoptosis or oncosis, also known as necrotic cell death [45]. Apoptosis is a tightly controlled process in which cell death is executed through the activation of specific signaling pathways and is characterized morphologically as membrane blebbing, cell shrinking, nuclear condensation, and chromatin aggregation.

Neighboring cells and macrophages rapidly digest these cellular fragments, or apoptotic bodies, without inducing inflammation or damage [46]. Apoptosis is the favored and controlled method of cell death and is vital for many processes such as organogenesis, normal cellular turnover, and the deletion of potentially neoplastic cells [47]. Apoptosis is initiated after numerous cellular insults and may proceed via an intrinsic (mitochondrial) or extrinsic (death receptor-mediated) pathway. Extrinsic apoptosis is initiated through ligand binding to one of a variety of death receptors. Tumor necrosis factor alpha (TNF-α) and FasL-induced apoptosis have been thoroughly evaluated in renal cells, and in fact, both cytokines are produced by renal epithelia and by infiltrating leukocytes [48, 49]. Nearly all renal cell types express receptors for both TNF-α and FasL, but vary in their sensitivity to these cytokines [50, 51]. Ligand binding of these lethal signaling molecules induces death receptor oligomerization, activation of caspase-8, and the downstream activation of effector caspases [49]. Apoptosis may be initiated at the level of the mitochondria or as a result of damage to an organelle, such as the endoplasmic reticulum or the nucleus. Initiation of the intrinsic or mitochondrial-mediated apoptotic program begins with the release of cytochrome c, which may be enhanced or inhibited by members of the Bcl-2 family of proteins (Bax, Bak, Bid, Bcl-2, BclxL) [49]. Cytochrome c then recruits other adaptor proteins including APAF-1 and caspase 9, thereby forming what is known as the apoptosome. This cell death complex goes on to activate the effector caspase, caspase 3 [50], resulting in cell death. Although the initiation of intrinsic and extrinsic apoptotic pathways are different, they both play a role in toxicant-induced apoptotic cell death. Oncosis/necrosis is characterized by organelle and cell swelling, cell rupture, and release of intracellular contents, which initiates an inflammatory response that is not observed in apoptosis. It is common for some toxicants to cause apoptosis at low concentrations and oncosis at high concentrations. Because apoptosis is an ATP-dependent process, nephrotoxicants that target the mitochondria and/or induce a decreased ATP predominantly cause oncosis rather than apoptosis [51, 52]. If cellular ATP levels are low and the mitochondrial membrane potential is quickly lost, then oncosis 77

COVINGTON & SCHNELLMANN

occurs. A rapid influx of Ca2+ into the mitochondria causes rupture of the inner and outer mitochondrial membranes resulting in a rapid loss of the mitochondrial membrane potential [53]. In contrast, if the loss of membrane potential is slow, and ATP levels are maintained, apoptosis is favored. Toxicants that reduce ATP disrupt cell volume, ion concentrations, and cell polarity. Disruption of cell volume and ion homeostasis occurs by toxicant interaction with the plasma membrane increasing ion permeability or by attenuating energy production. ATP depletion results in a decrease in Na+, K+-ATPase activity, resulting in cell swelling, and ultimately cell rupture [54, 55]. The tubular epithelia are polarized cells with specific transporters on the apical and basolateral domains. When a toxicant causes ATP depletion there is a dissociation of the Na+, K+-ATPase from the actin cytoskeleton and a redistribution from the basolateral to apical domain in the renal proximal tubule cells [56]. The loss of polarity of the cells disrupts the adhesion complexes and loss of cell-to-cell contact that facilitates further renal damage.

cells and those originating from the bone marrow in animal models of acute renal failure [61-63]. However, the significance of these findings is still unclear, and more research in this area is needed. In order for quiescent tubular cells of the injured nephron to carry out regeneration, gene upregulation, protein synthesis, and cell cycle entry is required. Therefore, growth factors are thought to be crucial in regenerating tubule cells, although the precise growth factors involved and their regulation are unknown [64, 65]. However, the epidermal growth factor receptor plays an important role in regulating de-differentiation, proliferation, and migration [66, 67]. There is a significant increase in cellular proliferation by surviving proximal tubular cells after renal injury in both animal models [68] and human cases of acute tubular necrosis [43] as measured by PCNA

Toxicant Injury

Oncosis Apoptosis

Renal repair and regeneration Although research has focused on the cellular events of nephrotoxicity, less emphasis has been placed on the mechanism of renal cell repair and regeneration after a toxic insult. Knowledge of post-injury repair/ regeneration will facilitate development of new therapeutics to promote renal recovery. As a result of toxicant-induced renal injury, renal epithelial dysfunction is typically characterized by the loss of cellular apical/ basal polarity, cytoskeletal redistribution, severe ATP depletion, mitochondrial dysfunction, impaired solute transport, and decreased ion pump activity including the Na+/K+ ATPase (Figure 3) [57, 58]. Depending on the extent of damage to the renal epithelium, cells die via necrosis or apoptosis. The remaining tubular cells survive in a sublethally injured state and undergo the complex process of regenerating the destroyed renal parenchyma [59, 60]. The standard hypothesis concerning renal cell regeneration is that sublethally injured and/or uninjured tubular cells restore cellular function, de-differentiate, proliferate, migrate, and finally re-differentiate to restore morphologic and physiologic function to the damaged nephron. Investigators have reported the presence of both endogenous renal stem 78

Sublethal injury

Regeneration

Cellular repair Proliferation

Migration

Figure 3. Proposed mechanism of renal cell repair and regeneration. Healthy renal epithelia are differentiated, quiescent columnar epithelia. After injury, numerous renal cells die via necrosis and apoptosis depending on the level of insult. However a few cells are sublethally injured and lose cell polarity and many physiological functions. These cells can either initiate the repair process immediately or dedifferentiate into mesenchymal-like cells. Sublethally injured epithelial cells begin to migrate and proliferate to fill in denuded regions of the tubular lumen. The epithelial cells finally redifferentiate back into quiescent tubular cells and regain their polarity and physiological functions.

04. Pharmacological aspects of nephrotoxicity

staining and incorporation of [3H]thymidine into nuclear DNA. Proliferating cells resemble mesenchymal cells with flattened cell bodies, loss of a brush border, and the de-differentiated expression of embryonic proteins such as vimentin and neural cell adhesion molecule [68-70]. In many ways, dedifferentiated renal epithelial cells reiterate the cellular ontogeny of renal organogenesis. Recently more and more arguments are collected indicating that regeneration by surviving tubular epithelial cells is the predominant mechanism of repair after ischemic tubular injury in the adult mammalian kidney [71]. While the endogenous growth factors (including paracrine and autorcrine) responsible for the proliferative phase of renal cell regeneration have not been identified, numerous studies have demonstrated that exogenously administered growth factors such as epidermal growth factor (EGF), insulin-like growth factor 1 (IGF-1), hepatocyte growth factor (HGF), fibroblast growth factor (FGF), heparin-binding EGF (HB-EGF), bone morphogenic protein-7 (BMP-7), and transforming growth factor (TGF)-β promote cellular proliferation in vitro and enhance renal recovery after ischemia/reperfusion injury [72- 78]. While quite effective in animal models, only one of these growth factors has been evaluated in humans (IGF-1). In one clinical trial, treatment with recombinant IGF-1 did not improve renal function in subjects with comorbidities [79] while the acute renal failure was not achieved in the other clinical trial [80]. It should be noted, however,

that growth factor regimens for humans would be acute in nature, to avoid potential adverse effects caused by growth factor-mediated overproduction of cells or by the stimulation of occult neoplastic cells. Recent studies have provided evidence that repair of sublethally-injured renal tubular cells requires functional attachment of the cells to the basement membrane. It is thought that integrin ligation to collagen IV elicits signal transduction events in injured renal tubular cells that stimulate cell survival and are critical for the repair of physiological functions such as polarity and Na+-transport [60, 81, 82]. The molecular mechanisms driving this return of function are not completely understood at the present, and greater understanding of the signaling pathways responsible for renal repair is needed before improved therapies can be developed for patients in the setting of postischemic renal injury. It should be noted that nephrotoxicants can further cause renal damage by inhibiting cellular repair and regeneration, delaying or completely inhibiting recovery of normal renal function. Cisplatin and aminoglycosides have been reported to inhibit renal regeneration in vivo [83, 84]. In addition, mercury chloride, fumonisin B1, and haloalkene cysteine conjugate inhibits the proliferation and migration of renal tubular cells [85]. Therefore, nephrotoxicants are capable of inhibiting the normal renal regeneration process which further contributes to renal dysfunction.

References 1. 2. 3. 4. 5. 6. 7.

Alcorn D and Ryan GB. Distribution of anionic groups in the glomerular capillary wall in rat nephrotoxic nephritis and aminonucleoside nephrosis. Pathology 13: 37-50, 1981. Kerjaschki D, Vernillo AT, and Farquhar MG. Reduced sialylation of podocalyxin--the major sialoprotein of the rat kidney glomerulus in aminonucleoside nephrosis. The American Journal of Pathology 118: 343-349, 1985. Andersson M, Nilsson UA, Hjalmarsson C, Haraldsson B, and Nystrom Sorensson J. Mild renal ischemia-reperfusion reduces charge and size selectivity of the glomerular barrier. Am J Physiol Renal Physiol 292(6): F1802-9, 2007. Perico N, Remuzzi A, Imberti O, Cavallotti D, Bertani T, and Remuzzi G. Morphometrical analysis of glomerular changes induced by cyclosporine in the rat. Am J Kidney Dis 17: 537-543, 1991. Deray G. Amphotericin B nephrotoxicity. J Antimicrob Chemother 49 Suppl 1: 37-41, 2002. Abbate GF, Alagia I, Giaquinto E, Giordano B, Leonessa V, and Altucci P. Preclinical comparative evaluation of aminoglycosides. Chemioterapia 3: 378-384, 1984. Budisavljevic MN, Hodge L, Barber K, Fulmer JR, Durazo-Arvizu RA, Self SE, Kuhlmann M, Raymond JR, and Greene EL. Oxidative stress in the pathogenesis of experimental mesangial proliferative glomerulonephritis. Am J Physiol Renal Physiol 285: F11381148, 2003.

79

COVINGTON & SCHNELLMANN

8.

Peltier J, Bellocq A, Perez J, Doublier S, Dubois YC, Haymann JP, Camussi G, and Baud L. Calpain activation and secretion promote glomerular injury in experimental glomerulonephritis: evidence from calpastatin-transgenic mice. J Am Soc Nephrol 17: 34153423, 2006. 9. Davies DJ, Dowling J, and Xipell JM. Gold nephropathy. Pathology 9: 281-288, 1977. 10. Guery JC, Druet E, Glotz D, Hirsch F, Mandet C, De Heer E, and Druet P. Specificity and cross-reactive idiotypes of anti-glomerular basement membrane autoantibodies in HgCl2-induced autoimmune glomerulonephritis. European Journal of Immunology 20: 93-100, 1990. 11. Koren G. The nephrotoxic potential of drugs and chemicals. Pharmacological basis and clinical relevance. Medical Toxicology and adverse drug experience 4: 59-72, 1989. 12. Ponticelli C and Pezzagno G. Hydrocarbons and glomerulonephritis--is the definite answer in? Nephrol Dial Transplant 9: 15391540, 1994. 13. Textor SC, Gephardt GN, Bravo EL, Tarazi RC, Fouad FM, Tubbs R, and McMahon JT. Membranous glomerulopathy associated with captopril therapy. The American Journal of Medicine 74: 705-712, 1983. 14. Wali RK and Henrich WL. Recent developments in toxic nephropathy. Current Opinion in Nephrology and Hypertension 11: 155-163, 2002. 15. Ho ES, Lin DC, Mendel DB, and Cihlar T. Cytotoxicity of antiviral nucleotides adefovir and cidofovir is induced by the expression of human renal organic anion transporter 1. J Am Soc Nephrol 11: 383-393, 2000. 16. Weinberg JM, Harding PG, and Humes HD. Mitochondrial bioenergetics during the initiation of mercuric chloride-induced renal injury. II. Functional alterations of renal cortical mitochondria isolated after mercuric chloride treatment. The Journal of Biological Chemistry 257: 68-74, 1982. 17. Fillastre JP, Moulin B, and Josse S. Aetiology of nephrotoxic damage to the renal interstitium and tubuli. Toxicology Letters 46: 45-54, 1989. 18. Fowler BA, Kimmel CA, Woods JS, McConnell EE, and Grant LD. Chronic low-level lead toxicity in the rat. III. An integrated assessment of long-term toxicity with special reference to the kidney. Toxicology and Applied Pharmacology 56: 59-77, 1980. 19. Pardi DS, Tremaine WJ, Sandborn WJ, and McCarthy JT. Renal and urologic complications of inflammatory bowel disease. Am J Gastroenterol 93: 504-514, 1998. 20. Thuluvath PJ, Ninkovic M, Calam J, and Anderson M. Mesalazine induced interstitial nephritis. Gut 35: 1493-1496, 1994. 21. Kahn J, Lagakos S, Wulfsohn M, Cherng D, Miller M, Cherrington J, Hardy D, Beall G, Cooper R, Murphy R, Basgoz N, Ng E, Deeks S, Winslow D, Toole JJ, and Coakley D. Efficacy and safety of adefovir dipivoxil with antiretroviral therapy: a randomized controlled trial. The American Journal of Medicine 282: 2305-2312, 1999. 22. Tanji N, Tanji K, Kambham N, Markowitz GS, Bell A, and D’Agati V D. Adefovir nephrotoxicity: possible role of mitochondrial DNA depletion. Human Pathology 32: 734-740, 2001. 23. Sarcletti M, Petter A, Romani N, Lhotta K, Konig P, Maier H, and Zangerle R. Pyuria in patients treated with indinavir is associated with renal dysfunction. Clinical Nephrology 54: 261-270, 2000. 24. Monks TJ and Lau SS. The pharmacology and toxicology of polyphenolic-glutathione conjugates. Annual Review of pPharmacology and Toxicology 38: 229-255, 1998. 24a. Chen JC, Stevens JL, Trifillis AL, and Jones TW. Renal cysteine conjugate beta-lyase-mediated toxicity studied with primary cultures of human proximal tubular cells. Toxicology and Applied Pharmacology 103: 463-473, 1990. 25. Kleinman JG, Breitenfield RV, and Roth DA. Acute renal failure associated with acetaminophen ingestion: report of a case and review of the literature. Clinical Nephrology 14: 201-205, 1980. 25a. Dahlin DC, Miwa GT, Lu AY, and Nelson SD. N-acetyl-p-benzoquinone imine: a cytochrome P-450-mediated oxidation product of acetaminophen. Proceedings of the National Academy of Sciences of the United States of America 81: 1327-1331, 1984. 26. Cohen SD and Khairallah EA. Selective protein arylation and acetaminophen-induced hepatotoxicity. Drug Metabolism Review 29: 59-77, 1998. 27. Bolton JL, Trush MA, Penning TM, Dryhurst G, and Monks TJ. Role of quinones in toxicology. Chemical Research in Toxicology 13: 135-160, 2000. 28. Dekant W, Vamvakas S, and Anders MW. Formation and fate of nephrotoxic and cytotoxic glutathione S-conjugates: cysteine conjugate beta-lyase pathway. Advances in pharmacology (San Diego, Calif ) 27: 115-162, 1994. 29. Hughey RP, Rankin BB, Elce JS, and Curthoys NP. Specificity of a particulate rat renal peptidase and its localization along with other enzymes of mercapturic acid synthesis. Archives of biochemistry and biophysics 186: 211-217, 1978. 30. Wolfgang GH, Gandolfi AJ, Stevens JL, and Brendel K. N-acetyl S-(1,2-dichlorovinyl)-L-cysteine produces a similar toxicity to S(1,2-dichlorovinyl)-L-cysteine in rabbit renal slices: differential transport and metabolism. Toxicology and Applied Pharmacology 101: 205-219, 1989.

80

04. Pharmacological aspects of nephrotoxicity

31.

Groves CE, Lock EA, and Schnellmann RG. Role of lipid peroxidation in renal proximal tubule cell death induced by haloalkene cysteine conjugates. Toxicology and Applied Pharmacology 107: 54-62, 1991. 31a. Chen Q, Jones TW, Brown PC, and Stevens JL. The mechanism of cysteine conjugate cytotoxicity in renal epithelial cells. Covalent binding leads to thiol depletion and lipid peroxidation. The Journal of Biological Chemistry 265: 21603-21611, 1990. 32. Bernareggi A, Torti L, Facino RM, Carini M, Depta G, Casetta B, Farrell N, Spadacini S, Ceserani R, and Tognella S. Characterization of cisplatin-glutathione adducts by liquid chromatography-mass spectrometry. Evidence for their formation in vitro but not in vivo after concomitant administration of cisplatin and glutathione to rats anc cancer patients. Journal of Chromatography 669: 247-263, 1995. 33. Townsend DM and Hanigan MH. Inhibition of gamma-glutamyl transpeptidase or cysteine S-conjugate beta-lyase activity blocks the nephrotoxicity of cisplatin in mice. The Journal of Pharmacology and Experimental Therapeutics 300: 142-148, 2002. 34. Townsend DM, Deng M, Zhang L, Lapus MG, and Hanigan MH. Metabolism of Cisplatin to a nephrotoxin in proximal tubule cells. J Am Soc Nephrol 14: 1-10, 2003. 35. Mayer RD, Lee KE, and Cockett AT. Inhibition of cisplatin-induced nephrotoxicity in rats by buthionine sulfoximine, a glutathione synthesis inhibitor. Cancer Chemotherapy and Pharmacology 20: 207-210, 1987. 36. Mistry P, Lee C, and McBrien DC. Intracellular metabolites of cisplatin in the rat kidney. Cancer Chemotherapy and Pharmacology 24: 73-79, 1989. 37. Mayer RD, Lee KE, and Cockett AT. Improved use of buthionine sulfoximine to prevent cisplatin nephrotoxicity in rats. Journal of Cancer Research and Clinical Oncology 115: 418-422, 1989. 38. Zalups, R.K. and Diamond, G.L., Nephrotoxicity of Metals, In:Toxicology of the Kidney, Eds. J.B. Tarloff and L.H. Lash, CRC Press, 3rd edition, 2005, 937-993. 39. Brezis M, Rosen S, Silva P, and Epstein FH. Selective vulnerability of the medullary thick ascending limb to anoxia in the isolated perfused rat kidney. The Journal of Clinical Investigation 73: 182-190, 1984. 40. Bernardo JF, Murakami S, Branch RA, and Sabra R. Potassium depletion potentiates amphotericin-B-induced toxicity to renal tubules. Nephron 70: 235-241, 1995. 41. Safirstein R, Deray G. Anticancer: Cisplatin/carboplatin. In: DeBroe ME, Porter GA, Bennett WM, Verpooten GA (eds) Clinical Nephrotoxins. Renal injury from drugs and chemicals. Kluwer Academic, Dordrecht, 261–271, 1998. 42. De Broe ME, Elseviers MM. Analgesic nephropathy. N Engl J Med 1998; 338: 446-452. 43. Nadasdy T, Laszik Z, Blick KE, Johnson DL, Burst-Singer K, Nast C, Cohen AH, Ormos J, and Silva FG. Human acute tubular necrosis: a lectin and immunohistochemical study. Human Pathology 26: 230-239, 1995. 44. Silva GA, Costa LM, Brito FC, Miranda AL, Barreiro EJ, and Fraga CA. New class of potent antinociceptive and antiplatelet 10Hphenothiazine-1-acylhydrazone derivatives. Bioorganic & Medicinal Chemistry 12: 3149-3158, 2004. 45. Levin S, Bucci TJ, Cohen SM, Fix AS, Hardisty JF, LeGrand EK, Maronpot RR, and Trump BF. The nomenclature of cell death: recommendations of an ad hoc Committee of the Society of Toxicologic Pathologists. Toxicologic Pathology 27: 484-490, 1999. 46. Savill J. Apoptosis in resolution of inflammation. Kidney & Blood pressure research 23: 173-174, 2000. 47. Rich T, Allen RL, and Wyllie AH. Defying death after DNA damage. Nature 407: 777-783, 2000. 48. Ortiz A, Lorz C, Catalan MP, Danoff TM, Yamasaki Y, Egido J, and Neilson EG. Expression of apoptosis regulatory proteins in tubular epithelium stressed in culture or following acute renal failure. Kidney International 57: 969-981, 2000. 49. Ortiz A, Justo P, Catalan MP, Sanz AB, Lorz C, and Egido J. Apoptotic cell death in renal injury: the rationale for intervention. Current Drug Targets 2: 181-192, 2002. 50. Bonegio R and Lieberthal W. Role of apoptosis in the pathogenesis of acute renal failure. Current Opinion in Nephrology and Hypertension 11: 301-308, 2002. 50. Meldrum KK, Meldrum DR, Hile KL, Yerkes EB, Ayala A, Cain MP, Rink RC, Casale AJ, and Kaefer MA. p38 MAPK mediates renal tubular cell TNF-alpha production and TNF-alpha-dependent apoptosis during simulated ischemia. American Journal of Physiology 281: C563-570, 2001. 51. Kern JC and Kehrer JP. Acrolein-induced cell death: a caspase-influenced decision between apoptosis and oncosis/necrosis. Chemico-Biological Interactions 139: 79-95, 2002. 51. Lorz C, Ortiz A, Justo P, Gonzalez-Cuadrado S, Duque N, Gomez-Guerrero C, and Egido J. Proapoptotic Fas ligand is expressed by normal kidney tubular epithelium and injured glomeruli. J Am Soc Nephrol 11: 1266-1277, 2000. 52. Lash LH, Hueni SE, Putt DA. Apoptosis, necrosis, and cell proliferation induced by S-(1,2-dichlorovinyl)-L-cysteine in primary cultures of human proximal tubular cells. Toxicology and Applied Pharmacology 177: 1-16, 2001. 53. Lemasters JJ, Qian T, Elmore SP, Trost LC, Nishimura Y, Herman B, Bradham CA, Brenner DA, and Nieminen AL. Confocal microscopy of the mitochondrial permeability transition in necrotic cell killing, apoptosis and autophagy. BioFactors (Oxford, England) 8: 283-285, 1998.

81

COVINGTON & SCHNELLMANN

54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64.

65. 66. 67. 68.

69.

70. 71. 72. 73. 74.

75. 76. 77. 78.

82

Miller GW and Schnellmann RG. A novel low-affinity strychnine binding site on renal proximal tubules: role in toxic cell death. Life Sciences 53: 1203-1209, 1993. Miller GW and Schnellmann RG. Inhibitors of renal chloride transport do not block toxicant-induced chloride influx in the proximal tubule. Toxicology Letters 76: 179-184, 1995. Molitoris BA, Leiser J, and Wagner MC. Role of the actin cytoskeleton in ischemia-induced cell injury and repair. Pediatric Nephrology (Berlin, Germany) 11: 761-767, 1997. Liu KD. Molecular mechanisms of recovery from acute renal failure. Critical Care Medicine 31:S572-81, 2003. Bush KT, Keller SH, and Nigam SK. Genesis and reversal of the ischemic phenotype in epithelial cells. The Journal of Clinical Investigation 106: 621-626, 2000. Bonventre JV. Dedifferentiation and proliferation of surviving epithelial cells in acute renal failure. J Am Soc Nephrol 14 Suppl 1: S55-61, 2003. Nony PA and Schnellmann RG. Mechanisms of renal cell repair and regeneration after acute renal failure. The Journal of Pharmacology and Experimental Therapeutics 304: 905-912, 2003. Lin F. Stem cells in kidney regeneration following acute renal injury. Pediatric Research 59: 74R-78R, 2006. Patschan D, Plotkin M, and Goligorsky MS. Therapeutic use of stem and endothelial progenitor cells in acute renal injury: ca ira. Current Opinion in Pharmacology 6: 176-183, 2006. Bussolati B, Bruno S, Grange C, Buttiglieri S, Deregibus MC, Cantino D, and Camussi G. Isolation of renal progenitor cells from adult human kidney. The American Journal of Pathology 166: 545-555, 2005. Kawaida K, Matsumoto K, Shimazu H, and Nakamura T. Hepatocyte growth factor prevents acute renal failure and accelerates renal regeneration in mice. Proceedings of the National Academy of Sciences of the United States of America 91: 4357-4361, 1994. Ichimura T, Maier JA, Maciag T, Zhang G, and Stevens JL. FGF-1 in normal and regenerating kidney: expression in mononuclear, interstitial, and regenerating epithelial cells. The American Journal of Physiology 269: F653-662, 1995. Wang Z, Zhang B, Wang M, Carr BI. Cdc25A and ERK interaction: EGFR-independent ERK activation by a protein phosphatase Cdc25A inhibitor, compound 5. Journal of Cell Physiology 204:437-44, 2005. Zhuang S, Dang Y, and Schnellmann RG. Requirement of the epidermal growth factor receptor in renal epithelial cell proliferation and migration. Am J Physiol Renal Physiol 287: F365-372, 2004. Witzgall R, Brown D, Schwarz C, and Bonventre JV. Localization of proliferating cell nuclear antigen, vimentin, c-Fos, and clusterin in the postischemic kidney. Evidence for a heterogenous genetic response among nephron segments, and a large pool of mitotically active and dedifferentiated cells. The Journal of Clinical Investigation 93: 2175-2188, 1994. Zhuang S, Yan Y, Han J, and Schnellmann RG. p38 kinase-mediated transactivation of the epidermal growth factor receptor is required for dedifferentiation of renal epithelial cells after oxidant injury. The Journal of Biological Chemistry 280: 21036-21042, 2005. Abbate M, Brown D, and Bonventre JV. Expression of NCAM recapitulates tubulogenic development in kidneys recovering from acute ischemia. The American Journal of Physiology 277: F454-463, 1999. Humphreys BD, Valerius MT, Kobayashi A, Mugford JW, Soeung S, Duffield JS, McMahon AP, Bonventre JV. Intrinsic epithelial cells repair the kidney after injury. Cell Stem Cell 2008 2: 284-291. Franquesa M, Alperovich G, Herrero-Fresneda I, Lloberas N, Bolanos N, Fillat C, Rama I, Cruzado JM, Grinyo JM, and Torras J. Direct electrotransfer of hHGF gene into kidney ameliorates ischemic acute renal failure. Gene Therapy 12: 1551-1558, 2005. Harris RC. Growth factors and cytokines in acute renal failure. Advances in Renal Replacement Therapy 4: 43-53, 1997. Humes HD, Cieslinski DA, Coimbra TM, Messana JM, Galvao C. Epidermal growth factor enhances renal tubule cell regeneration and repair and accelerates the recovery of renal function in postischemic acute renal failure. Journal of Clinical Investigations 84: 1757-61, 1989. Ichimura T, Finch PW, Zhang G, Kan M, Stevens JL. Induction of FGF-7 after kidney damage: a possible paracrine mechanism for tubule repair. American Journal of Physiology 271: F967-76, 1996. Igawa T, Kanda S, Kanetake H, Saitoh Y, Ichihara A, Tomita Y, and Nakamura T. Hepatocyte growth factor is a potent mitogen for cultured rabbit renal tubular epithelial cells. Biochemical and Biophysical Research Communications 174: 831-838, 1991. Miller SB, Martin DR, Kissane J, and Hammerman MR. Insulin-like growth factor I accelerates recovery from ischemic acute tubular necrosis in the rat. Proceedings of the National Academy of Sciences of the United States of America 89: 11876-11880, 1992. Zeisberg M, Shah AA, and Kalluri R. Bone morphogenic protein-7 induces mesenchymal to epithelial transition in adult renal fibroblasts and facilitates regeneration of injured kidney. The Journal of Biological Chemistry 280: 8094-8100, 2005.

04. Pharmacological aspects of nephrotoxicity

79.

80. 81. 82. 83. 84.

85.

Hirschberg R, Kopple J, Lipsett P, Benjamin E, Minei J, Albertson T, Munger M, Metzler M, Zaloga G, Murray M, Lowry S, Conger J, McKeown W, O’Shea M, Baughman R, Wood K, Haupt M, Kaiser R, Simms H, Warnock D, Summer W, Hintz R, Myers B, Haenftling K, Capra W, and et al. Multicenter clinical trial of recombinant human insulin-like growth factor I in patients with acute renal failure. Kidney International 55: 2423-2432, 1999. Franklin SC, Moulton M, Sicard GA, Hammerman MR, and Miller SB. Insulin-like growth factor I preserves renal function postoperatively. The American Journal of Physiology 272: F257-259, 1997. Nowak G, Carter CA, and Schnellmann RG. Ascorbic acid promotes recovery of cellular functions following toxicant-induced injury. Toxicology and Applied Pharmacology 167: 37-45, 2000. Nony PA and Schnellmann RG. Interactions between collagen IV and collagen-binding integrins in renal cell repair after sublethal injury. Molecular Pharmacology 60: 1226-1234, 2001. Kovacs CJ, Braunschweiger PG, Schenken LL, and Burholt DR. Proliferative defects in renal and intestinal epithelium after cisdichlorodiammine platinum (II). British Journal of Cancer 45: 286-294, 1982. Leonard I, Zanen J, Nonclercq D, Toubeau G, Heuson-Stiennon JA, Beckers JF, Falmagne P, Schaudies RP, and Laurent G. Modification of immunoreactive EGF and EGF receptor after acute tubular necrosis induced by tobramycin or cisplatin. Renal Failure 16: 583-608, 1994. Counts RS, Nowak G, Wyatt RD, and Schnellmann RG. Nephrotoxicant inhibition of renal proximal tubule cell regeneration. The American Journal of Physiology 269: F274-281, 1995.

83

05

Pharmacovigilance: from signal to action Hubert G. LEUFKENS and Antoine C. EGBERTS Utrecht Institute for Pharmaceutical Sciences and University Medical Centre, Utrecht, The Netherlands

Introduction ____________________________________________________________ 85 Pharmacovigilance reflects a continuum _____________________________________ 86 To report or not to report __________________________________________________ 86 Methodology of pharmacovigilance _________________________________________ 87 Final thoughts and integration______________________________________________ 89 References ______________________________________________________________ 90

Introduction

F

rom the very beginning of modern pharmacotherapy there has been the challenge of identifying drug-induced unintended effects as soon and as comprehensive as possible. Any suspicions of an expected or a new problem with a medicine should be well reported, signalled and evaluated. Despite extensive testing of medicines before they are approved for marketing, unexpected and/or rare adverse drug reactions may occur when the medicine is used in normal daily practice. Moreover, also in case that a possible drug-induced problem is already known from the pharmacology of the medicine, e.g. so-called type A effects, it is important to quantify this risk (e.g. in terms of absolute risk, number needed to harm and risk factors), and to put into context when the product is extensively used in clinical practice. This context may include possible strategies for risk management, tailoring the treatment scenario to the individual patient in terms of choice of the medicine, dose and duration, genotyping, consideration of alternative treatments, and so on. Pre-marketing findings regarding safety of medicines are commonly based on the experience of

only a few hundreds to thousand people at a maximum, who have been treated in controlled randomised trials. These trials have important limitations in terms of that they [a] usually include rather homogeneous populations (no elderly patients with other diseases, no impaired renal or liver function, etc), [b] they are too small to detect very rare events, [c] they are usual too short to detect long-term effects, [d] they are unable to predict the real world of clinical practice. A complicating factor is that individual medicinal products are increasingly prone to extensive public debate and societal uncertainty about the safety of medicines in general. Public debate acts as a twosided sword: attention may work out well because it sensitises patients and health care professionals to be vigilant and to report any observed event or problem related to the use of a medicine. On the other side there is the risk of differential over- and under reporting, very often resulting in biased estimates of the possible drug-induced risk. One may question whether it is still feasible to elucidate and unravel a possible drug exposure-outcome relationship in an independent, science and clinical relevance based fashion, when the debate is shaking the public and economic press,

LEUFKENS & EGBERTS

as we could witness with the COX-2 inhibitors (e.g. cardiovascular and gastrointestinal risk), the statins (e.g. myopathy and rhabdomyolysis) or the glitazones (e.g. cardiovascular, fracture risk).

Pharmacovigilance reflects a continuum The characterization of the full safety profile of a medicine is a dynamic continuum that never ends as long as the drug is on the market. Pharmacovigilance has been defined by WHO as ‘the science and activities relating to the detection, assessment, understanding and prevention of adverse effects or any other possible drug-related problems’. The context of the prescribing and usage environment may change over time leading to variations of the risk of drug-drug interactions and of adverse effects related to the patient susceptible genotype. A sudden increase in the rate of spontaneous reports always needs to be analysed in the context of these dynamics. There is increasing evidence that drug safety is a function of both molecular features and the prescribing and usage context leading to three scenarios for drug-induced risk: • Safety issues primarily related to drug specific characteristics, e.g. type A (pharmacological) and type B (idiosyncratic) adverse reactions. • Safety issues primarily related to patient specific characteristics e.g. underlying disease, severity of the condition or susceptible genotype, including type B (idiosyncratic) adverse reactions. • Safety issues primarily related to errors in the prescribing, dispensing and patient usage process, e.g. prescribing-induced interactions, non-compliance with drug labelling, problems as a consequence of usage errors. There may be some overlap between type B and risks related to patient characteristics, where the later category represents events with a known patient-related mechanism of action. Usually the mechanism of type B drug events is unknown or speculative. However as science and new insights evolve, we see a shift from type B to type A. Examples include for instance severe sensitivity reactions due to the use of abacavir, an HIV drug. This risk was already known from the beginning of clinical drug development and was classified at that time as type B reaction. Interestingly, now we have more insight in the underlying HLA-driven mechanism of this adverse effect, it is becoming more 86

and more a type A reaction. Another issue that needs careful consideration is that currently a number of complex safety issues cover possible drug-induced problems very close to the indication of the drug, e.g. glitazones for the treatment of diabetes and cardiovascular ADRs or antidepressants and suicide risk in children and adolescents. These cases require comprehensive methods to unravel any causality of drug and event.

To report or not to report The basis of pharmacovigilance lies in careful watching, cross-patient thinking and the prepared mind that everything that happens in the course of a disease may be of relevance to evaluate treatment outcomes, both beneficial and adverse. Spontaneous reports represent essentially a behavioural dimension of pharmacovigilance, as doctors may be reluctant to report because they think they are too busy, they feel not responsible, or they are afraid of being held accountable or liable for any harm experienced by the patient. As the majority of medicines (between 80-90%) in most countries are prescribed in the community, an essential target of pharmacovigilance is the primary care setting. Primary care physicians are crucial to identify and communicate drug-related events with their colleagues and the competent authorities. But also in hospitals, medical specialists are in the position to link unwanted health effects to drug usage, particularly when it comes to new, specialised products, e.g. oncology drugs, immunosuppressives. Reporting of unexpected and/or adverse events to the relevant pharmacovigilance units in hospitals or to the authorities is an important responsibility of health professionals. Without these reports no functional pharmacovigilance system could exist. Because reporting is such an explicit human activity is carries in it all the risks of selective reporting, underor overreporting, and so on. Its power, however, lies in the nation/worldwide collection of suspicions about adverse drug reactions enabling early detection of possible drug hazards far more early than an individual professional ever could. Pharmacovigilance is essentially based on ‘numbers count’, although one should be careful for biased surges in reports. The literature is full of experiences where spontaneous reports were affected, both qualitative and quantitative, by publicity

05. Pharmacovigilance: from signal to action

in the media, by publications in the medical press, or by regulatory action. In some countries there are obligatory systems of reporting of possible drug-induced problems endorsed by formal legislative systems. Many countries don’t have these, and so far there is not full agreement on which of these is the best.

Methodology of pharmacovigilance History shows that, with all the limitations, the dedicated and watchful doctor has remained the backbone for signalling any possible drug-induced problem. The act of prescribing of a medicine cannot be separated from the responsibility to follow the patient over time and to evaluate the treatment response, including any unintended effects, as detailed and prompt as possible. In virtually all countries there is some kind of a system of collecting spontaneous reports of (possible) adverse drug reactions from physicians, pharmacists, industry, nurses, and increasingly acknowledged, also from patients. Spontaneous reports from physicians heavily rely very much on careful observation and recognition of any relevant change in the clinical condition of a patient given the use of a medicine. Because such spontaneous reports can be random noise or a real signal of a clinically relevant drug-induced problem, physicians need to be trained to have an open mind towards the unexpected. Medicines may have unforeseen side effects when used in patients with multiple morbidities, impaired organ function or used in an inappropriate fashion. Determining the signal-noise ratio is a key activity in pharmacovigilance and requires a close collaboration between health professionals, patients, academia, regulatory authorities and the industry. There is ample literature on the various methods applied in pharmacovigilance. Although there are many differences in the way the methods are implemented or adjusted for specific purpose, the underlying principles are virtually the same. Pharmacovigilance is a cyclic process of [a] signal detection, [b] strengthen them by careful analysis of background rates of the particular types of events, looking at series of reports, characteristics of the individual patients, and underlying diseases of the patients exposed to the medicine, and finally [3] follow-up of the signal in formal pharmacoepidemiological studies. In addition, mechanistic studies are important since they provide clues towards prevention, but have also shown to be a trigger for

drug innovation. While in the detection phase there is a strong focus on qualitative issues related to a signal, e.g. quality of the report, causality assessment, the more quantitative methods are key in the strengthening, and particularly in the follow-up phase. The cyclic nature of pharmacovigilance is reflected is a constant learning loop from report-signal-data to (if needed) regulatory action-communication to patients and health professional, and back consequently back into the drug innovation process. A first, and essential, step in signal detection of drug-induced risk is the proactive and systematic collection of spontaneous reports. After quality control of the data and exclusion of obvious unlikely associated cases from these, the question arises what the numbers say. Are 5 reports enough to raise a signal, do we need 10, 25 or more? As discussed before, a crucial limitation to detect drug-safety signals through spontaneous reports is the frequent lack of valid exposure data. How to cope with this? These are all valid questions and need to be addressed in the context of the question is the observed number more or less then expected. So we need to create an estimate of ‘what could be expected’, in other words what delivers a valid signal? Overall there are two approaches to tackle this: 1. Calculation of a frequency (or rate) of the number of reports per 1000 patients, prescriptions, DDD/ 1000p/day or another available denominator. From there several external comparisons can be made with similar frequencies based on number of cases in an unexposed population, also coined as baseline risk or background frequency, or the number of cases in a population exposed to another medicine from the same therapeutic category. This approach is only possible when reliable denominator data are present. Moreover the head-to-head comparisons require limited under-, over- or selective reporting in the two frequencies of observed possible induced drug problems. Example: A study in 1,219 patients of the ATHENA (AIDS Therapy Evaluation National Centre) cohort of patients infected with HIV receiving antiretroviral therapy in the Netherlands showed a frequency of urological symptoms (including nephrolithiasis, renal colic, flank pain, hematuria, renal insufficiency, or nephropathy) of 8.3 per 100 treatment-years for indinavir compared to 0.8 per 100 treatment-years for other HIV protease inhibi87

LEUFKENS & EGBERTS

tors. 8.3 versus 0.8 represents a clear signal, and also a quantitative strengthening of earlier reports of indinavir-induced nephrotoxicty. 2. An alternative approach is an internal comparison within all collected reports assuming again limited under-, over- or selective reporting. Within all reports related to, for instance nephrotoxicity, a distribution of the different drug exposures is made. Consequently the question is addressed whether this distribution is different (disproportional) when compared with the distribution of drug exposure in all the other or a sample of all other reports. Disproportionality, evaluating more or less then expected, is a key concept in signal detection. Over the years several measures of disproportionality have been developed with all their inherent pros and contras. The most frequently used is the so-called reporting odds-ratio (ROR). Example: In follow-up of the receipt of 7 reports acute interstitial nephritis (AIN) by the Netherlands Pharmacovigilance Centre Lareb, the databank of World Health Organisation Collaborating Centre for International Drug Monitoring in Uppsala, Sweden (containing about 3.7 million spontaneous reports from more than 80 countries worldwide) was searched for cases of AIN. A total of 150 AIN cases with recorded proton-pump inhibitors (PPI) use was found. The proportionality of PPI use within the AIN cases was compared to the same in the rest of all 3.7 million reports, resulting in a ROR of 9.4 for omeprazole.

covigilance an important public health tool for the benefit of patient’s health. When we look at the international picture of pharmacovigilance over the last two decades (Figure 1) we may identify three important ‘waves’ of learning. In the mid-80s, there was growing awareness about flawed comparisons of spontaneous reports when just looking at the crude numbers without comparisons over equivalent periods of the marketing life cycles of the drugs compared. A key paper from this period showed in the case of piroxicam and the risk of gastrointestinal bleeding, perforation, and ulcer, that crude rates of spontaneous reports changed dramatically after adjustments for the heterogeneity in the underlying reporting rates over time. The importance of this learning wave was the acknowledgement of the limitations of spontaneous reports, but not to neglect them as they carry critical information items for pharmacovigilance. The second wave of pharmacovigilance learning in the early and mid 90s coincides with the evolution of the science of pharmacoepidemiology with a strong emphasis on exposure ascertainment as a crucial factor to evaluate drug-induced effects in a valid fashion. When a surge in reports on possible druginduced risk is observed, the proper question should be raised whether the drug is bringing the problem to the patient or the patient the problem to the drug? Very often drugs are selectively prescribed to patients with a risky profile resulting in a higher likelihood of drug-induced risk. This process of so-called drug ‘channeling’, also coined as confounding by indication, is important to understand and to factor in appropriated risk assessment of medicines. This concept has

In the strengthening and follow-up phase of pharmacovigilance we can identify a broad array of approaches including Prescription Event Monitoring 3. Proactive risk management, (PEM), also applied successfully ‘landscaping’ drug use, genetics for signal detection, observational pharmacoepidemiological studies 2. Exposure correlates (e.g. channeling, confounding in automated databases (e.g. cohort, by indication), disproportionality methods case-control and variations), and 1. Beyond counting crude numbers prospective randomised clinical of spontaneous ADR reports trials. Finally we should add what all the acquired evidence of a valid signal means for regulatory action 1987 2007 and informing prescribers and the public. It’s the combination of all these activities that Figure 1. Two decades of learning in pharmacovigilance. carry the potential, and the need, to make pharma88

05. Pharmacovigilance: from signal to action

also been linked to the fact that aggressive marketing by the pharmaceutical industry can ‘kill’ important medicinal products when drugs are selectively used in high-risk patients or in an inappropriate, off-label fashion. Essentially, from a methodological point of view, channelling underpins the notion that drug prescribing is virtually never a random activity, thereby implying important challenges for studying these in an observational fashion in an unbiased way, particularly when it comes to study differential drug-induced harm related to individual drugs within drug classes. In the same period we see also a strong development in pharmacovigilance on statistical methods to support evaluation of disproportionality of rates of spontaneous reports. We now enjoy the fruits of these important methodological developments in quantifying possible signals of drug-induced harm. In the third wave of learning in pharmacovigilance is rightly reflected by the term ‘proactive’. There has been growing awareness among regulators, industry and other stakeholders that ‘wait and see’ is not the way we should continue in protecting individual patients and the public from unintended drug effects. This has resulted in for instance the development of comprehensive programmes for Risk Management Plans to be submitted by pharmaceutical companies as part of the dossier of new medicinal products at the level of the European regulatory system. But also in other regulatory hemispheres there is ample attention for proactive approaches for identifying, and evaluating drug-induced harm as soon as possible, including adequate risk minimisation measures, e.g. information to prescribers, precautions to be taken by patients, and the like. An important signature of these Risk Management Plans is the need to fill the gap between the first signal of drug-induced harm and scientific proof of the risk, followed by regulatory and communication action. Regulatory decision-making and timely action are often hampered by the lack of reliable data on the evidence of the risk when such evidence has to be collected in a retrospective fashion. Another feature of this third wave of pharmacovigilance learning is the growing notion that populations exposed to certain drugs carry specific baseline risks. There is increasing evidence that the likelihood of the majority of the problems we face in pharmacovigilance is in certain patients more at risk than others. Therefore a critical part of pharmacovigilance is seen

in characterizing, also coined ‘landscaping’, the patient population in order to identify patients and patterns of drug usage susceptible to increased risk. As part of this, pharmacogenetic biomarkers are increasingly considered as important tools to identify proactively possible non-responders in terms of safety to drug therapy. We have already pointed on the example of abacavir and more applications of pharmacogenetics are established or underway. The abacavir case is the first example of European regulatory including pharmacogenetic screening as an integral part of the drug label.

Final thoughts and integration Pharmacovigilance has become an essential part of public health and pharmaceutical innovation. After new medicinal products have been approved for usage in normal clinical practice the real practice-based benefit-risk balance should be established. In Table 1 a number of well-known cases of drug-induced nephrotoxicity and their pharmacovigilance commonalities are listed. All the five cases show ample variety with respect to signal and exposure factors, the presence of denominator data and how confounding or effect modification might be a issue to evaluate possible drugrelated risk on the renal system. This array shows the importance of integrative thinking and well-developed knowledge about the possibilities and limitations of certain approaches, from the historical case of analgesic (including phenacetin) nephropathy towards the most recent findings on gadolinium based contrast agents and nephrogenic systemic fibrosis (NSF). The prepared mind of the doctor and the awareness that reporting is always important, of course in case of newly introduced medicinal products but also with old products, makes pharmacovigilance a typical partnering activity in health care. There is no single party that can do all the work. There is no single approach that suits the solution of all problems. By its very nature, the renal system carries a highrisk profile for drug-induced toxicity. Mechanistic and pathophysiologic thinking remains critical for a better understanding of observed harm and prediction of possible future harm. This requires at least valid data on signals, exposure, denominator and confounders. When ever possible, proactive and prospective design of the data collection is preferable, if not in many cases the only way to get reliable answers. Current 89

LEUFKENS & EGBERTS

international thinking on pharmacovigilance is in line with these methodological considerations, making the

future of this important field in medicine and public health promising for the benefit of patients.

Table 1. Cases of drug-induced nephrotoxicity and their pharmacovigilance commonalities Signal factors

Exposure factors

Denominator data

Confounding factors, effect modifiers

Phenacetin

Spontaneous reports, time gap signal and use

OTC, combined with other analgesics

Poor

Co-medications, disease severity, protopatic bias

Protease inhibitors

Already known from RCTs

Combined with other drugs, dosing

Good quality, large cohorts

Previous treatment, body mass, climate

Statins

Spontaneous reports, public media effects

Shift to high potency use, class effect?

Good quality

Drug channelling, selective prescribing

Contrast agents

Problem not signalled by prescriber/radiologist

Timing of exposure, class effect?

Poor

Co-morbidity, confounding by marketing

Cyclosporine

Already known from RCTs

Dose/duration of use, long-term effects

Reasonable quality

Confounding by renal transplant indication

References 1. 2.

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

90

Bruin ML de, van Puijenbroek EP, Egberts AC, Hoes AW, Leufkens HG. Non-sedating antihistamine drugs and cardiac arrhythmias, biased risk estimates from spontaneous reporting systems? Br J Clin Pharmacol. 2002; 53: 370-4. Dieleman JP, Sturkenboom MC, Jambroes M, Gyssens IC, Weverling GJ, ten Veen JH, Schrey G, Reiss P, Stricker BH; Athena Study Group. Risk factors for urological symptoms in a cohort of users of the HIV protease inhibitor indinavir sulfate: the ATHENA cohort. Arch Intern Med 2002; 162: 1493-501. Egberts TC, Smulders M, de Koning FH, Meyboom RH, Leufkens HG. Can adverse drug reactions be detected earlier? A comparison of reports by patients and professionals. Br Med J 1996; 313: 530-1. Härmark L, van der Wiel HE, de Groot MC, van Grootheest AC. Proton pump inhibitor-induced acute interstitial nephritis. Br J Clin Pharmacol 2007; 64: 819-23. Ingelman-Sundberg M. Pharmacogenomic biomarkers for prediction of severe adverse drug reactions. N Engl J Med 2008; 358: 637-9. Leufkens HG, Urquhart J. Variability in patterns of drug usage. J Pharm Pharmacol 1994; 46 Suppl 1: 433-7. Maitland-van der Zee AH, de Boer A, Leufkens HG. The interface between pharmacoepidemiology and pharmacogenetics. Eur J Pharmacol 2000; 410: 121-130. Meyboom RH, Edwards IR. Rosuvastatin and the statin wars--the way to peace. Lancet 2004; 364: 1997-9. Meyboom RH, Lindquist M, Egberts AC, Edwards IR. Signal selection and follow-up in pharmacovigilance. Drug Saf 2002 ;25: 459-65. Puijenbroek EP van, Bate A, Leufkens HG, Lindquist M, Orre R, Egberts AC. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf 2002; 11:3-10. Rossi AC, Hsu JP, Faich GA. Ulcerogenicity of piroxicam: an analysis of spontaneously reported data. Br Med J 1987; 294:14750. Stricker BH, Psaty BM. Detection, verification, and quantification of adverse drug reactions. Br Med J 2004; 329(7456): 44-7. Thomsen HS, Marckmann P, Logager VB. Nephrogenic systemic fibrosis (NSF): a late adverse reaction to some of the gadolinium based contrast agents. Cancer Imaging 2007; 7: 130-7. Waller PC, Evans SJ. A model for the future conduct of pharmacovigilance. Pharmacoepidemiol Drug Saf 2003; 12:17-29. WHO. The importance of pharmacovigilance. Safety monitoring of medicinal products. Geneva: World Health Organization, 2002. Xiaoqing G, Chike N. How to prevent, recognize, and threat drug-induced nephrotoxicity. Cleveland Clin J Med 2002: 69: 28912.

06

Urinary biomarkers and nephrotoxicity William F. FINN1 and George A. PORTER2 1University 2Oregon

of North Carolina, Chapel Hill, North Carolina, USA Health Sciences University, Portland, Oregon, USA

Introduction ____________________________________________________________ 92 Categories of biomarkers __________________________________________________ 92 Biomarkers of exposure Biomarkers of effect Biomarkers of susceptibility

93 93 95

Urinalysis _______________________________________________________________ 96 Test strip screening Urine microscopy

96 96

Blood urea nitrogen concentration and urea clearance __________________________ 97 Serum creatinine concentration _____________________________________________ 97 Creatinine clearance ______________________________________________________ 98 Glomerular filtration rate __________________________________________________ 98 Renal blood flow ________________________________________________________ 100 Tubular function ________________________________________________________ 100 Proteinuria _____________________________________________________________ 101 High-molecular weight proteinuria Low-molecular weight proteinuria

103 105

Enzymuria _____________________________________________________________ 107 Renal papillary antigen ___________________________________________________ 110 Cytokines ______________________________________________________________ 110 Cell adhesion molecules __________________________________________________ 112 Miscellaneous biomarkers ________________________________________________ 115 References _____________________________________________________________ 117

FINN & PORTER

Introduction

T

here are a number of definitions of the term “biomarker”. In general, they have in common three components: [1] that they are objectively measured indicators of specific anatomic, physiologic, biochemical, or molecular events; [2] that thay are associated with normal biological processes or accompany the onset, progression and/or severity of specific pathological or toxic conditions and [3] are that they are useful for measuring the progress of injury, disease or the effects of therapeutic intervention. For example, according to the National Institutes of Health (NIH) working group, a biomarker is a characteristic that is objectively measured as an indicator of normal biological processes, pathogenic processes, or a pharmacological response to a therapeutic intervention [1]. The types of biomarkers and the purposes served vary to some extent depending on the population beng observed. For public health purposes, the requirements of useful biomarkers to protect from injurious xenobiotic exposure are three-fold: firstly, to achieve the earliest identification of the potential for health impairment; secondly, to gain insight into the mechanism(s) responsible for any adverse impact on the health of individuals or specific populations at risk; and thirdly, to help assess the effects of interventions designed to minimize the short and longterm consequences of the initial injury. Important requirments for biomarker development are a detailed understanding of biochemical pathways involved in nephrotoxicity, minimal invasiveness and capacity to screen large atrisk populations. Those involved in individual health assessment are concerned with the early detection of specific organ kidney injury. With regard to acute kidney injury (AKI), biomarkers may serve several additional purposes. That is, they may determine AKI subtypes (prerenal, intrinisic renal, or postrenal), identify the etiology of AKI (ischemia, toxins, sepsis, or a combination), differentiate AKI from other forms of acute kidney disease (urinary tract infections, glomerulonephritis, interstitial nephritis), predict the AKI severity (risk stratification for prognostication as well as guide to therapy), monitor the course of AKI, and monitor the response to AKI interventions. For chronic kidney disease (CKD), they provide both evidence and severity of exposure and may be used to assess response to removal of offend92

ing toxin. The pharmaceutical industry has specific interest in the development and utilization of biomarkers for evaluating and predicting the safety of drug candidates during the process of their development. In drug trials, biomarkers have been proposed for use in efficacy determination and patient population stratification, in deducing pharmacokinetic-pharmacodynamic relationships and in safety monitoring [2]. These different phases of drug development involve different functional categories of biomarkers and often involve the patterns of several biomarkers - rather than a change in a single biomarker. The effort to identify reliable biomarkers often involves the interaction of several disciplines such as genetics and epigenetics, genomics, proteomics, metabonomics and assay development [3].

Categories of biomarkers There have been a number of attempts to formally categorize biologic markers of renal injury in order to achieve a uniform and consistent approach. This have included biomarkers related to a spcific physiologic parameter, such as markers of renal blood flow, glomerular filtration rate, or tubular function; and the chemical nature of the biomarker, such as growth factors, enzymes, adhesion molecules, inflammatory cytokines,etc. One additional classification attempts to define sequential changes in the appearance of one or more biomarkers as renal injury, either acute or chronic, progresses from the initial insult to clinical disease and includes four overlapping stages during that process [4]. These stages consider the nature and magnitude of the initial insult, its relationship to a biologically injurious stimulus, the presence of early biologic effects and eventually on alterations in the structure and/or function of the kidney. At each point along this line, individual susceptibility - which is also subject to various external factors - determines whether or not the process progresses to the development of clinical renal impairment (Figure 1). In this schema, biomarkers are considered to fall in the three general designations. These include biomarkers of exposures, biomarkers of effect, and biomarkers of susceptibility. Each of these types of biomarkers has specific and relevant applications to the understanding of renal injury and disease. Specific and sensitive bi-

06. Urinary biomarkers and nephrotoxicity

Exposure

Internal dose

Biologically effective dose

Early biologic effect

Altered structure/ function

Clinical disease

Susceptibility

Figure 1. Simplified flow chart of classes of biologic markers (indicated by boxes). Solid lines indicate progression, if it occurs to the next class of marker. Dashed lines indicate that individual susceptibility influences the rates of progression, as do other variables. Biologic markers represent a continuum of changes, and the classification of change might not always be distinct. (adapted from Committee on Biological Markers of the National Research Council, USA, 1987)

omarkers constitute the missing link in the continuum of exposure to toxins and susceptibility, disease development and possible therapeutic intervention [5].

Biomarkers of exposure Biomarkers of exposure are of greatest utility when monitoring exposure to xenobiotics, that is, various chemicals, drugs, and pollutants not naturally present in the body. A biomarker of exposure is more formally defined as “an exogenous substance or its metabolite(s) or the product of the interaction between a xenobiotic agent or other injurious stimulus and the target molecule or cell that is measured within a compartment of an organism” [4]. With regard to xenobiotics, a marker of external exposure is simply the amount of the xenobiotic to which a person is subjected, whereas a marker of internal exposure is the amount of a substance absorbed into the body. Markers of internal exposure are a more accurate means of estimating exposure than are markers of external exposure and require the analysis of biological samples. Biomarkers of exposure are particularly important in toxicology because they are an indicator of internal dose, or the amount of chemical exposure that has resulted in absorption into the body. Biomarkers of exposure to xenobiotics causing nephrotoxicity may take one of several forms. The measurement of blood or tissue levels of drugs known to have adverse effects on the kidney, such as cyclosporine, aminoglycoside antibiotics, or lithium, is a standard practice. The awareness of the total amount of drug administered is frequently important when considering amphotericin, analgesics, and cisplatin nephrotoxicity. More difficulty is encountered with the determination of the body burden of a toxicant, although under certain circumstances

such a value is necessary to determine the health effects of exposure to heavy metals such as cadmium and lead, and some analgesics. Ideally, biomarkers of exposure should have a direct and quantitative relationship to the xenobiotics’ biologically effective dose. This term refers to the internal dose of xenobiotic that produces a predictable biologic effect. To gain an understanding of the biologically effective dose, several facts are required (Table 1). These include the knowledge of the amount of xenobiotic which is present in the external environment, its route of entry and the extent of absorption, distribution and accumulation within the body, the target cell or receptor site of the xenobiotic, the route and extent of its metabolism, the modification of the effective dose by associated metabolic, physiologic and pathologic conditions, and finally the pathways of elimination.

Biomarkers of effect A biomarker of effect is defined as “a measurable alteration of an endogenous component within an organism that, depending on magnitude, can be recognized as a potential or established health impairment or disease” [4]. Markers of effect represent points on Table 1. Determinants of the biologically effective dose of a xenobiotic. Amount in external environment Route of entry Extent of absorption, distribution and accumulation Target cell or receptor site Modification by associated conditions Route and extent of metabolism Pathways of elimination

93

FINN & PORTER

a continuum of health impairment and may be measured qualitatively or quantitatively. Early responses to exposure may include changes in the function of target tissues or responses in organs or tissues such as chromosomal damage, mutations of critical target genes, or altered hormone status. Biomarkers of effect are classified according to their impact on health status. The utility of a biomarker of effect may range from enabling prediction of future health impairment to confirming the presence of clinical disease. The biomarker may either be an indirect manifestation of a disease process or may be a direct result of impaired organ function. An example of an indirect marker of xenobiotic-induced renal disease is the elevated level of red cell content of either delta amino-levulinic acid dehydrase or free erythrocyte protoporphyrin in patients with lead nephrotoxicity whereas direct urinary markers of lead nephrotoxicity are capable of defining the presence of both glomerular and tubular involvement [6]. Direct examples of biomarkers of effect are dependent upon the nature of the disease process itself. To mention a few, the presence of small amounts of albumin in the urine of patients with diabetes mellitus is an early warning sign of diabetic nephropathy with an increased cardiovascular risk and impaired renal prognosis [7]. Microalbuminuria may also be found in individuals chronically exposed to cigarette smoke accompanied by elevated serum cadmium and lead levels [8]. The appearance in the urine of abnormal amounts of low molecular weight proteins such as β-2 microglobulin (β2-m) and/or retinol binding protein have been useful in the detection and stratification of workers with industrial exposure to various heavy metals [9,10]. Abnormal patterns of urinary electrolyte excretion and impaired acidification have long been recognized in patients with amphotericin-induced renal injury [11]. Structural lesions within the kidneys may be found in certain electrolyte depletion syndromes such as in the case of the prolonged use of potassium-depleting diuretics [12]. Patients with either acute or chronic renal failure may present with many and varied manifestations of uremia. In these patients, the application of biomarkers of effect to detect clinical disease in its earliest stages is of great importance. Table 2 contains a list of various groups of xenobiotics associated with acute or chronic renal disease. Among the occupational and environmental xe94

Table 2. Xenobiotics associated with renal disease. Occupational and environmental xenobiotics Organic solvents Heavy metals Pesticides Recreational drugs Heroin Cocaine Diagnostic and therapeutic agents Antibacterial agents Antiviral agents Antifungal agents Antineoplastic agents Immunosuppressive agents Non-steroidal anti-inflammatory drugs Osmotic agents Radiographic contrast material Natural toxic compounds Aflotoxins Hemolytic agents and myotoxins

nobiotics associated with AKI are specific substances such as toluene (organic solvents), lead (heavy metals) and chlordane (pesticides). Of the diagnostic and therapeutic agents, the aminoglycosides (antibacterials), acyclovir (antivirals), amphotericin (antifungals), cisplatin (chemotherapeutic agents), cyclosporine (immunosuppressives), and contrast agents stand out. Nonselective non-steroidal anti-inflammatory drugs (NSAIDs) inhibit both cyclooxygenase (COX)-1 and COX-2 are some of the most commonly used medications worldwide. Along with the selective COX-2, these drugs continue to be associated with AKI [13,14]. Notably, high cumulative NSAID exposure is associated with an increased risk for rapid CKD progression in the elderly [15]. The parenteral administration of high doses of certain polyols (mannitol, sorbitol), sugars (glucose, fructose, sucrose, lactose), polysaccharides (inulin), and other products (e.g. radiocontrast agents) may be associated with renal injury marked by vacuolation and subsequent swelling of renal tubular epithelial cells - the so-called “resorptive vacuolation”[16]. An increasing concern is the renal dysfunction associated with the use of heroin and cocaine (recreational drugs) [17]. Some agents such as arsine may trigger a severe hemolytic reaction, causing hemoglobinuria and subsequent acute renal failure. Others may lead to the destruction of striated muscle, and myoglobinuria

06. Urinary biomarkers and nephrotoxicity

leading to AKI. In both cases, the consequent “pigment nephropathy” is not an uncommon cause of acute kidney injury. In sum, the most prevalent mechanisms of drug-induced acute kidney injury are vasoconstriction, altered intraglomerular hemodynamics, tubular cell toxicity due to medullary hypoxia, interstitial nephritis, crystal deposition, thrombotic microangiopathy, and osmotic nephrosis [18].

Biomarkers of susceptibility A biomarker of susceptibility can be defined as “an indicator of an inherent or acquired limitation of an organism to respond to the challenge of exposure to a specific xenobiotic substance” [4]. These markers indicate differences in individuals or populations that affect the body’s response to xenobiotic exposure. They may include variations in the balance between enzymes that detoxify or enhance the toxicity of chemicals, genetic differences in the capacity of cells to recover from injury, inherited genetic defects that increase the risk of cancer. Perhaps the most important susceptibility marker and one quite specific to the kidney is the presence or absence of underlying CKD. It has become apparent that individuals with CKD are at increased risk for the development of more severe injury in response to either nephrotoxic or ischemic events and that the susceptibility is inversely related to the severity of the underlying renal impairment. While it is understood that the kidneys play a major role in the excretion of drugs and has the capacity to metabolize endogenous and exogenous compounds, CKD decreases the ability of the kidney to metabolize Table 3. Some factors influencing nephrotoxicity. Urine flow rate Urine pH Renal blood flow

drugs. Less understood is the fact that patients with CKD have a decrease in the nonrenal clearance of multiple drugs resulting in prolonged retention of either the unmodified toxin or its metabolic toxic residue. CKD affects the metabolism of drugs by inhibiting key enzymatic systems in the liver, intestine and kidney [19]. It should be appreciated that a major research goal is to link markers of exposure with markers of effect. Unfortunately, for the vast majority of patients with suspected toxic renal injury the precise knowledge of the offending agent is speculative and not measurable by current techniques. As a result, more is known about the risk factors associated with an adverse health effect than is known about the parameters of exposure (Table 3). Translating this concept of progressive appearance of biomarkers from exposure to disease into actual practice remains a challenge. The dual aspects of renal function, i.e., filtration/elimination and reabsorption/secretion, assure that no single test or measure can define global renal function. Furthermore, the substantial metabolic and endocrine functions of the kidney are not considered in the classical techniques used to analyze renal function. This has led to the use of a separate category of tests designed to serve as markers of renal dysfunction or injury (Table 4). Also, considerable attention has been directed to the immunological responses that follow xenobiotic exposure. Finally, as the mechanisms responsible for cell injury, Table 4. New parameters and techniques applicable to monitor nephrotoxicity. Parameters

Clearance of lithium, H2O High pressure liquid and N-methylnicotinamide chromatography metabolites Enzymes and antigens

Fluorimetric and luminometric immunoassays

Microproteins

2-Dimensional electrophoresis Immunoblotting techniques Nephelometry, turbidimetry

DNA, mRNA

Southern blotting Pulse field electrophoresis Northern blotting Restricted fragment length analysis

In vivo imaging

Nuclear magnetic resonance spectroscopy

Sodium balance Pre-existing disease Other drug therapy Tolerance Pharmacokinetic factors

Techniques

Microsomal enzyme activity Dosage and route of administration

In vitro imaging

Electron probe analysis

Duration of exposure

Surface markers

Cell sorting

95

FINN & PORTER

death and regeneration become more apparent, a new and promising set of biomarkers is emerging.

Urinalysis Test strip screening The examination of the urine using qualitative test strip provides an estimate of glucose, pH, hemoglobin, protein, specific gravity and a number of other substances including ketones, bilirubin, urobilinogen, leukocytes and nitrate. The degree of sophistication has progressively increased to the extent that reading of test strips with reflectometers is possible. There is a good probability that urines negative by dipstick for protein, blood, leukocytes, nitrates, glucose and ketones will be negative on microscopic examination, with only 5.3% having any abnormality. However, urines positive for one or more of these findings may not correlate well with the microscopic findings due to a number of false positive and false negative by dipsticks for red cells and leukocytes. Sensitivities for dipsticks have been reported to be 75.3% and 81.0% and specificities were 88.6% and 64.3% for red cells and leukocytes, respectively [20]. It is recommended that microscopic analysis be limited to urines in which the dipstick is abnormal. Other limitations have been identified. For example, patients with microalbuminuria or tubular proteinuria are not detected by current test strip methods. Immunological techniques, which enable the determination of specific protein molecules, may make such detection possible [21].

sample should be used for analysis. Pyuria: The normal number of white blood cells in the concentrated, resuspended urine sample does not exceed 1 to 2 per high- powered field. In patients with pyelonephritis or nephrotoxic interstitial nephritis, neutrophils may be found whereas with allergic interstitial nephritis, eosinophils may appear. Macrophages and lymphocytes can be found in the urine of some patients with glomerulonephritis and be useful in monitoring the activity of the disease [23] Tubular epithelial cells The appearance in the urine of epithelial cells is most likely a result of tubular injury. These cells may be present alone or in casts and be indicative of either acute or chronic tubulointerstitial nephritis. Since casts may dissolve in alkaline urine, an acid urine sample is preferred for analysis.

Urine microscopy

Eosinophiluria The finding of eosinophils in the urine with the use of Hansel’s stain has been suggested to be useful in establishing the diagnosis of acute interstitial nephritis [24]. However, the positive predictive value in screening samples may be too low, and the number of false positives and negatives in selected groups may be too high for eosinophiluria to stand alone in making the diagnosis of acute interstitial nephritis [25] c. Urinary macrophages: The presence of macrophages in the urine of patients with glomerulonephritis reflects the pathological events in the kidney. Urinary macrophage counts increase in patients with proliferative GN, especially in the presence of active glomerular injury [26].

The microscopic examination of the urine sediment provides enhanced diagnostic efficiency. Hematuria: The normal number of erythrocytes in resuspended urine sediment is no more than 1 to 2 per high-powered field. When an abnormal number of erythrocytes are present it is necessary to distinguish between their origin being renal or non-renal. The simultaneous presence in the urine of casts and protein favor a renal origin. With phase-contrast microscopy, a high percentage of dysmorphic erythrocytes support a renal source of hematuria [22]. The urine should be examined immediately after voiding. Since erythrocytes may be lysed in low specific gravity urine, a concentrated

Visceral epithelial cells The loss of glomerular visceral epithelial cells (podocytes) has been associated with the development of glomerular sclerosis and loss of renal function. The majority of urinary podocytes are viable, although apoptosis occurs in about one-half of the cells. Patients with active glomerular disease excrete more podocytes/mg creatinine than do healthy controls and patients with quiescent disease. It appears that the difference in growth behavior between healthy controls and subjects with active glomerular disease suggests that in active disease viable podocytes detach from the glomerular tuft due to local environmental factors

96

06. Urinary biomarkers and nephrotoxicity

rather than defects in the podocytes per se, whereas in healthy individuals, mostly senescent podocytes are shed [27]. While examination of the urinary sediment has traditionally been used to discriminate the severity of acute kidney injury and to differentiate pre-renal azotemia from established AKI or ATN, the value of this approach is imperfect. Examination of the urinary sediment may have value in critically ill patients, in particular when there is suspicion of systemic vasculitis the detection of dysmorphic red blood cells or red blood cell casts may have important diagnostic, prognostic and therapeutic value [28]. In addition, new technological evolutions have enabled creative diagnostic approaches in urinalysis. Urinary flow cytometry and automated microscopic pattern recognition are two new techniques that are characterised by a much lower imprecision and a higher throughput as compared to conventional microscopy of the urine sediment. Automated urinary test strip analysis offers analytical, clinical, and labour costsaving advantages [29,30]. Despite these advances, for borderline results, there is no substitute for a urinalysis performed by an experienced nephrologist [31].

Blood urea nitrogen concentration and urea clearance Urea is quantitatively the most important solute excreted by the kidney and was the first organic solute detected in the blood of patients with kidney failure [32]. Yet it is a poor marker of uremic illness. Furthermore, the blood urea nitrogen (BUN) is not a

satisfactory measurement of the glomerular filtration rate. The use of urea to estimate GFR, however, is problematic due to the numerous extra-renal factors that influence its endogenous production and renal clearance, independent of GFR. First, the rate of urea production is not constant. Urea can be grossly modified by a high protein intake, critical illness (i.e. sepsis, burns, trauma), gastrointestinal hemorrhage, or drug therapy such as use of corticosteroids or tetracycline. Conversely, patients with chronic liver disease and low protein intake can have lower urea levels without noticeable changes in GFR. Second, the rate of renal clearance of urea is not constant. An estimated 40–50% of filtered urea is passively reabsorbed by proximal renal tubular cells. Moreover, in states of decreased effective circulating volume (i.e. volume depletion, low cardiac output), there is enhanced reabsorption of sodium and water in the proximal renal tubular cells along with a corresponding increase in urea reabsorption. Consequently, the serum urea concentration may increase out of proportion with changes in SCr and be underrepresentative of GFR. In addition, the urea clearance (Curea) is proportional to the urine flow rate. For example, at low and high rates of urine flow, the minimal and maximal values of the Curea may vary from 30% to 60% of the glomerular filtration rate. This occurs because various tubular segments are permeable to urea and allow passive reabsorption to occur under conditions of antidiuresis. The fractional excretion of urea (FEurea) is calculated as [(urine urea/plasma urea)/(urine creatinine/plasma creatinine) x 100]. A low FEurea may be used as an index of decreased renal perfusion [33,34].

Serum creatinine concentration Table 5. Reference ranges for the Scr taking into consideration differences in age and gender. Age/Sex

Scr mg/dL

Scr μmol/L

0-7d

0.6 - 1.1

53.0 - 97.2

8 d - 1 mo

0.3 - 0.7

26.5 - 61.9

1 mo - 2 yr

0.3 - 0.6

26.5 - 53.0

3 - 4 yr

0.3 - 0.7

26.5 - 61.9

5 - 9 yr

0.4 - 0.9

35.4 - 79.6

10 - 17 yr Male

0.5 - 1.1

44.2 - 97.2

10 - 17 yr Female

0.4 - 1.0

35.4 - 88.4

18 yr+ Male

0.8 - 1.4

70.7 - 123.8

18 yr+ Female

0.7 - 1.1

61.9 - 97.2

Serum creatinine is an amino acid compound derived from the metabolism of creatine in skeletal muscle and from dietary meat intake [35]. The serum creatinine concentration (Scr) is a commonly used marker for the estimation of adequate renal function due to the fact that it is released into the plasma at a relatively constant rate, is freely filtered by the glomerulus, and is not metabolized nor reabsorbed by the kidney. Various ‘reference ranges’ for the Scr take into consideration differences in age and gender (Table 5), but fail to consider other variables such as race, body weight and muscle mass. As a result, a Scr within the ‘reference 97

FINN & PORTER

range’ cannot be considered to be a priori evidence of ‘normal’ renal function. For example, an estimated 10–40% of creatinine clearance occurs by tubular secretion of Scr into the urine [36]. For subjects with chronic kidney disease (CKD) there is a considerable lack of precision in accepting the Scr. SCr values may not show significant increases until approximately 50% of kidney function is lost. For individuals with glomerular filtration rate greater than 30 ml/min, the 95% confidence interval for Scr is ± 22%, whereas it is ± 13% in patients with glomerular filtration rate less than 30 ml/min [37]. The actual Scr may be increased or decreased independent of changes in the glomerular filtration rate by inhibiting or stimulating renal tubular secretion. For example, trimethoprim and/or trimethoprim/sulfamethoxazole have been demonstrated to cause a 15 to 35% increase in Scr due to an inhibition of tubular secretion [38]. The free radical scavenger, N-acetylcysteine, appears to facilitate tubular secretion in volunteers with normal renal function as judged by a fall in the Scr without a change in cystatin C levels [39], although the situation may be different in patients with stage 3 CKD [40]. In addition to trimethoprim, cimetidine and salicylates also produce elevations in the SCr by altering the normal elimination pathways of creatinine. Phenacemide has been reported to increase creatinine elimination [41]. Differences in analytical techniques may also account for variation in the reported Scr [42,43]. For example, overestimation of the Scr may occur because of interference from substances other than creatinine (“noncreatinine chromogens”), such as proteins and ketoacids, and high levels of bilirubin or glucose to cause false elevations of the Scr. Several drugs have been reported to interfere with SCr results obtained with both the Jaffé-based and enzymatic analytical assay systems by producing assay interference. When Scr samples are calibrated in a single reference laboratory, noncalibrated Scr values were greater than standardardized creatinine values [44]. The National Kidney Foundation’s current practice guidelines recommend standardization of serum creatinine assay calibration to increase assay accuracy. This will result in a lower range of values being considered normal and will result in higher calculated glomerular filtration rates and creatinine clearance [45]. Recently, it has been reported that serum levels tryptophan glycoconjugate might 98

replace inulin clearance in the clinical setting [45a]. However, because of the technical difficulty in measuring tryptophan glycoconjugate [45b] it is unlikely to gain wide-spread acceptance. Lastly, it should be mentioned that the SCr does not depict real-time changes in GFR that occur with acute reductions in kidney function. Rather, SCr requires time to accumulate prior to being detected as abnormal, thus leading to a potential delay in the diagnosis of AKI.

Creatinine clearance The endogenous creatinine clearance (Ccr) gives an acceptable estimate of the glomerular filtration rate and is the most widely used method in clinical practice for routine purposes. However, in normal individuals, the majority of measurements tend to yield values of Ccr that exceed the actual glomerular filtration rate by a substantial amount, owing to the fact that there is a small but significant amount of creatinine which appears in the urine as a result of tubular secretion. This problem is accentuated when the glomerular filtration rate declines. Ccr measurements may be twofold higher than the actual glomerular filtration rate because of continued tubular secretion of creatinine at a time when the rate of filtration is severely curtailed. Indeed, the amount of secreted creatinine varies inversely with the glomerular filtration rate [46,47]. An alternative method is the determination of creatinine clearance (CCr) after oral administration of cimetidine. This drug blocks tubular secretion of creatinine and CCr measured under these conditions is reported to be nearly identical to GFR in mild or severe renal failure. It has been suggestd that the measurement of CCr (without cimetidine) is an anachronism and should be abandoned [48].

Glomerular filtration rate Estimated glomerular filtration rate An alternative to the measurement of the Ccr is the use of either nomograms or formulae to estimate the glomerular filtration rate. The two most widely used equations are the Cockcroft-Gault and the Modification of Diet in Renal Disease (MDRD) study equations [46,47].

06. Urinary biomarkers and nephrotoxicity

Cockroft and Gault, 1976 [49]: Males: Ccr (ml/min) = (140 - age in years) × (weight in kg) × 1.73 72 × serum creatinine (mg/100 ml) × body surface area (kg/m2) Females = males × 0.85

Modification of Diet in Renal Disease (MDRD): Original MDRD Study equation [50]: Estimated GFR (mL/min/1.73 m2) =

Table 6. Stages of CKD based on eGRF as proposed by KDOQI guidelines of NKF. Stage of kidney disease

GFR

Stage 1

> 90 mL/min/1.73 M2 and structural abnormalities

Stage 2

60-89 mL/min/1.73 M2

Stage 3

30-59 mL/min/1.73 M2

Stage 4

15-29 mL/min/1.73 M2

Stage 5

60ml/min/1.73m3. The variation in calculated Ccr is particularly true in the elderly or others with large decreases in muscle mass, in patients with liver disease, and individuals ingesting a high-protein diet or those receiving parenteral nutrition containing amino acid solutions. Absolute variation is also more evident at higher estimated GFR. Measured glomerular filtration rate Any substance used to measure glomerular filtration rate should be metabolically intact, freely filtered

through the glomerular capillary wall, and be neither secreted nor reabsorbed by the tubules. Accurate plasma and urine quantitation also should be easily achievable. In addition to inulin, several compounds are useful for the measurement of glomerular filtration rate. These include the urologic contrast media, eg. diatrizoate, iohexol, other useful compounds include: 57cocyanocobalamine, 51Cr-ethylenediaminetetraacetic acid (EDTA) or sodium 125I iodothalamate and 99mTc-diethylenetriaminepentaacetic acid (DTPA). All provide reliable measurement of glomerular filtration rate [53]. Inulin is a polymer of fructose and is an ideal glomerular filtration rate marker because it is freely filtered and neither reabsorbed or secreted by the tubules. It has been widely used as a research tool but because of a number of technical difficulties, it is rarely used in clinical settings. Isotopic methods offer a high level of reliability but the impracticality of using these methods in a clinical setting makes them unsuitable for routine use. On the other hand, Iohexol is a convenient, reliable technique for measuring GFR and has the same precision as 125I-iodothalamate [54]. As an alternative to the standard clearance techniques which involve the collection of urine over a known period of time plus maintaining a constant plasma level of an appropriate marker, the glomerular filtration rate can be calculated from the rate of disappearance from the plasma of any tracer, where: Clearance =

Injected dose Area under plasma concentration curve

Additional techniques to obtain more reliable estimates of glomerular filtration rate without resorting to steady-state infusions involve the plotting of the declining plasma level of radio isotopic agents [55] or 99

FINN & PORTER

non-radioactive iodinated contrast agents [56] if they are cleared by glomerular filtration. The glomerular filtration rate as measured with iohexol shows excellent agreement with the values obtained using inulin and chelates throughout a wide range of kidney function. [53, 57,58]. As noted previously, in healthy adults, the endogenous creatinine clearance tends to exceed the “true” GFR as determined by inulin or iohexol clearances [59]. The “renal reserve” is determined by measuring the percentage increase in baseline dglomerular filtration rate following ingestion of a high protein meal. The failure of the glomerular filtration rate to increase in response to such a challenge suggests that underlying chronic disease and nephron atrophy has been masked by hypertrophy of other nephrons so that overall renal function seems to be well maintained. Renal function is preserved with aging in healthy subjects at the expense of elimination of the ‘renal reserve” [60]. In this regard, the GFR, whether estimated or measured, is a reliable marker of susceptibity to both ischemic and nephrotixic injury – particularly when the value falls below 60 mL/min. Indeed, the susceptibility of the kidney to superimposed acute injury markedly increases as renal function declines.

Renal blood flow If a marker is extracted from the blood exclusively by the kidney resulting in a renal venous concentration of 0% (i.e. the arterio-venous extraction fraction is 100%), then the calculated value of the clearance of the marker (Cx) is equal to renal plasma flow. In practice, a compound, such as para-amino hippurate (PAH) with an extraction fraction of about 87%, is used. To acknowledge the fact that there is discrepancy between the PAH clearance and renal plasma flow, the term effective renal plasma flow is used when the extraction factor is not measured. In sum, renal plasma flow = effective renal plasma flow + extraction factor and renal blood flow = effective renal plasma flow + the hematocrit. A decrease in the PAH clearance might be due to either an actual decline in renal plasma flow or a decrease in the extraction factor of PAH. The latter occurs when the tubular secretion of PAH in proximal tubules is impaired due to tubular disease or the presence of substances, which compete with transcellular 100

PAH transport. Thus, the PAH clearance cannot be considered a reliable measure of renal plasma flow, unless the extraction factor of PAH is measured simultaneously. This requires that a sample of renal venous blood be obtained.

Tubular function The identification of a reliable and convenient method for the estimation of the reabsorptive and secretory capacity of the kidney has proven to be a considerable challenge to Nephrology. This is not unexpected when one considers the complex and integrated functions contributed by the various tubular segments to insure proper composition of bladder urine. General estimates of integrated tubular function include the capacity of the kidneys to concentrate or dilute the urine in response to water deprivation or administration; the ability to excrete an administered acid load; and the precision with which sodium balance is maintained. But lacking is a technique for assessing tubular function, which rivals the measurement of glomerular filtration rate. Specific gravity and osmolality The urinary specific gravity and osmolality are indicators of the ability of the kidney to concentrate and dilute the urine. The urinary specific gravity depends upon the size and weight of urinary solutes. The normal range is 1.003 to 1.025 whereas the possible range is 1.001 to 1.040. Osmolarity indicates the total number of solute particles per kilogram of urine water. The normal range is from 150 to 900 mosm/kg with a possible range from 50 to 1200 mosm/kg. The discovery of four major water channels in the kidney, namely aquaporins (AQP) 1, 2, 3 and 4, has allowed a substantial increase in our understanding of renal water regulation. The renal aquaporin water channels are involved in the urinary dilution and concentrating defects in cardiac failure, cirrhosis, syndrome of inappropriate hormone secretion, pregnancy, hypothyroidism, isolated glucocorticoid deficiency, isolated mineralocorticoid deficiency, primary polydipsia, acquired and genetic nephrogenic diabetes insipidus [61,62]. pH

A hydrogen ion concentration gradient of 1 to 1000 may be established across tubular cell membranes of

06. Urinary biomarkers and nephrotoxicity

the kidney. Since the pH is the negative logarithm of the hydrogen ion concentration, this translated into a decrease from the normal plasma pH value of 7.4 to the minimal urine pH of 4.4. The pH of urine is dependent on the time of day, the prandial state, diet, health status, and medications. Urinary pH exhibits a diurnal variation with decreased pH values at night and in the early morning (most acidic towards midnight) followed by increasing pH values upon awakening. Urine tends to become alkaline immediately after a meal because of a phenomenon known as the alkaline tide and gradually becomes acidic between meals. A high protein diet is associated with acidic urine, and a vegetarian diet typically produces more alkaline urine because of bicarbonate formation from fruits, especially citrus, and vegetables. Bacterial contamination of urine with microorganisms that split urea may yield urinary pH values > 8.0 because of bacterial decomposition of urea to ammonia. The pH values of specimens stored at -20 degrees C are relatively stable, whereas pH results > 9 develop when urine samples are stored at room temperature or higher. Degradation of nitrogenous urine analytes is most likely responsible for the noted increases in pH. [63]. Lithium clearance The study of renal segmental tubular sodium handling by measurement of exogenous or endogenous lithium clearance has been a source of valuable information about in-vivo alterations of tubular sodium and water transport in humans The lithium clearance is a used to estimate the amount of sodium and water delivery from the pars recta of the proximal tubule into the descending limb of the loop of Henle [64]. This information may be helpful in the assessment of the state of hydration. The method is based on several assumptions the most important of which are that lithium reabsorption parallels sodium and water along the entire proximal tubule; that lithium is neither reabsorbed ‘in measurable amounts beyond the pars recta of the proximal tubule; nor is it secreted by the tubular cells [65]. This technique is based on the principle that, while sodium and water are reabsorbed at several sites along the nephron, the lithium ion is taken up almost exclusively at proximal tubular sites, so that the amount of lithium escaping reabsorption at this level is quantitatively excreted in the urine. As lithium in the proximal tubule is transported by the same systems

driving sodium and water, the parallel measurement of lithium, sodium and creatinine clearance may provide reasonably accurate and complete information as to the occurrence of abnormalities in sodium and water handling at different sites along the nephron. LiCl has been used to evaluate salt and water handling in cirrhotic patients and found increased sodium reabsorption in the distal tubule accounts for the salt retention that characterizes this clinical condition [66]. Increased proximal sodium re-absorption is associated with the metabolic syndrome (MS) in white men and women. This relationship is not seen in people of African or South Asian origin, despite a greater degree of insulin resistance [67]. It appears that an alteration of renal tubular sodium handling is an important feature of MS, involving an increased rate of proximal sodium and water reabsorption with a modification of the normal pressure–natriuresis relationship [68]. Patients with ascites showed a positive correlation between lithium fractional excretion and glomerular filtration rate (r = 0.64, P < 0.05). Reduction in renal perfusion, increased filtration fraction, and Tubular-Glomerular Feedback derangement, as found in decompensated patients, are indicative of prevalent postglomerular arteriolar vasoconstriction, with ensuing stimulation of proximal tubular sodium reabsorption [69].

Proteinuria The glomerular wall contains three layers: endothelial cells, basement membrane, and epithelial cells. Under normal circumstances, the glomerular filtration barrier restricts the transfer of high molecular weight proteins from plasma to the nephron lumen while allowing the filtration of small molecules. Much of the selectivity of filtration occurs in the basement membrane, where the barrier excludes proteins on the basis of both their size and their charge. Uncharged molecules pass through the basement membrane more readily than negatively charged proteins of a similar size. In certain pathologic states, the permselectivity of the filtration barrier changes allowing high molecular weight proteins to appear in the urine. These proteins undergo pinocytotic reabsorption in the proximal tubule creating cytoplasmic vesicles that then fuse with primary lysosomes to form secondary lysosomes. In this final form the proteins are hydrolyzed to amino acids, which are delivered into the blood stream. In 101

FINN & PORTER

Table 7. Classification of proteinuria according to site of origin. Plasma proteins Kidney-derived proteins Proteins from the urogenital tract Proteins released from tissue outside the urogenital tract Pregnancy associated proteins Tumor-derived proteins

102

proteins > 50 kD proteins < 50 kD

Normal

Glomerular

Tubular

Overflow

urine

contrast, under normal conditions a finite amount of low molecular weight proteins are filtered which then undergo reabsorption by proximal tubular cells. Exopeptidases situated on the brush border membrane are responsible for splitting peptides up to a molecular weight of 10,000 daltons. Following metabolic conversion, reabsorption of the amino acids or dipeptides occurs by specific sodium-dependent carriers [70]. When the reabsorptive capacity of the proximal tubular epithelium is disrupted, various low molecular weight proteins escape reabsorption and can be measured in the urine. Thus, the distinction between so-called “glomerular” proteinuria and “tubular” proteinuria is based on both the quantity and quality of the proteins measured in the urine [71] (Figure 2). A recent refinement of this differentiation in protein selectivity has been the “so-called” urine protein expert system [72,73]. This expert system, which includes total protein, albumin, β1-microglobulin, IgG, α2-microglobulin, NAG and creatinine, has proven to be more discriminatory in providing correct clinical diagnoses which are histologically confirmed, as compared to human expert diagnosis. Another approach to differentiating glomerular from tubular disease involves analyzing urinary proteins with the SDS-PAGE system that separates various urinary protein species. In a recent report, Lau and Woo [74] found an excellent correlation between SDS-PAGE prediction and findings on renal biopsy. In general, proteins in the urine may be classified into six main categories according to their origin (Table 7). Measuring urinary protein excretion has been simplified by the introduction of the urine protein to creatinine ratio [UP:Ucr] [75,76]. Although random spot UP:Ucr ratio predicts actual 24 h protein excretion with reasonable accuracy in patients with lower levels of protein excretion but is unreliable in patients with high protein excretion and should not be used in the

Figure 2. Three kinds of proteinuria.

clinical setting unless 24 h urine collection is unavailable [77]. None-the-less, the use of spot urine UP:Ucr ratio is useful as a tool in screening and monitoring proteinuria [78, 78a,78b]. In addition to serving as markers of renal dysfunction, it is now evident that the filtration of abnormal amounts and/or types of proteins influences the progression of renal disease by promoting secondary injury to tubular epithelial cells and interstitial structures. Proteinuria itself has been proposed to contribute to progressive renal injury inflammation [79,80]. For example, the upregulation of various cytokines in tubular epithelial cells may contribute to the development of interstitial fibrosis and cell cycle activation leading to tubular cell proliferation and/or apoptosis [81-83]. Albumin can increase AngII production and in turn upregulate TGF-β receptor expression [84]. Other filtered components of the urine in proteinuric states, such as oxidized proteins, appear to be more potent in inducing direct injury of tubular epithelial cells and activating proinflammatory and fibrotic chemokines and cytokines. Complement and various lipoproteins are also present in the urine in proteinuric disease states and can activate reactive oxygen species [85,86]. Proteinuria may thus alter tubule cell function directly, potentially contributing to a more profibrotic phenotype, and also augment interstitial inflammation, in particular by macrophages. Proteinuria may activate many profibrotic pathways through its ability to increase NF-kB, and also by other pathways. These include, for instance, complement synthesis occurring

06. Urinary biomarkers and nephrotoxicity

in renal tubules [87]. Proteomics is the study of protein expression in a tissue or biological fluid. Comparison of protein patterns in biological fluids between healthy individuals and patients with disease is increasingly being used both to discover biological markers of disease and to identify biochemical processes important in disease pathogenesis. Currently available tests for urine proteins measure either the total level of urine protein or the presence of a single protein species. Emerging proteomic technologies allow simultaneous examination of the patterns of multiple urinary proteins and their correlation with individual diagnoses, response to treatment or prognosis [88,88a]. The application of proteomic methods and informatic analysis have been used to identify patterns of urine proteins that are characteristic of the nephritic syndrome resulting from FSGS, lupus nephritis, membranous nephropathy, or diabetic nephropathy. These data showed that diseases that cause nephrotic syndrome change glomerular protein permeability in characteristic patterns. The fingerprint of urine protein charge forms identifies the glomerular disease [89].

High-molecular weight proteinuria The appearance in the urine of serum proteins with a molecular weight (MW) in excess of 40,000 to 50,000 daltons is an early marker of glomerular damage. The commonly measured high molecular weight

proteinuria includes: albumin (Mr 69,000), transferrin (Mr 77,000) and IgG (Mr 146,000) (Table 8). Albumin Albumin is quantitatively the major urinary protein derived from plasma. Its average concentration in normal urine is at least 50 times higher than most other low molecular weight proteins. Albumin’s molecular size (molecular radius: 3.6 nm) and strong negative charge, effectively retarded filtration at the glomerular barrier since the vast majority of pores perforating the glomerular filtration barrier have a radius of 2.9–3.1 nm. The loss of negative charges causes the effective small pore radius to increase to approximately 4.5 nm, which allows the passage of albumin. The small amounts of albumin that ordinarily escape into the glomerular filtrate are reabsorbed by the proximal tubule with a presumed efficiency of 99%. Transient and totally reversible increases in the albumin excretion may be observed in various “physiologic” situations that induce increases in the glomerular filtration rate such as heavy exercise, fever, or assuming an orthostatic position. A 24 hour unine collection showing an albumin excretion at a rate of 20 to 200 μg/min or a urinary concentration of 30 to 300 mg/L measured on at least two occasions is refeerred to as microalbuminuria. Urinary albumin levels above these values are called "macroalbuminuria", or sometimes just albuminuria. To compensate for the variablility in urine concentra-

Table 8. Characteristics of urinary proteins. Protein

Abbv.

Molecular weight kD

GSC*

Normal urinary excretion mg/mmol creatinine

Normal plasma levels mg/L

β2-microglobulin

β2-m

11.6

18.3 (4.4-32.2)

< 0.05

1.3

Retinol-binding protein (free)

RBP

21

13.6 (5.1-22.1)

cephalexin and other 3methyl-cephalosporins [97]. It appears that there is a partial correlation between β-lactam acylation potency and their nephrotoxicity which ranked as following: cephaloglycin > cephaloridine > cephalothin > cefazolin > cefaclor > penicillins, cephalexin, ceftazidime and cefotaxime [26, 67]. Moreover, cefaclor which appears to have high acylation potential has low renal toxicity. Further, cephaloridine with moderate acylating activity is one of the most nephrotoxic cephalosporins. It was speculated that this discrepancy may be due to the presence of the cationic nitrogen group near to the carboxyl group of cephaloridine; this could limit cephaloridine access to the anionic targets and thus requiring high intracellular concentration to induce nephrotoxicity [67]. Whereas experimental evidence supports the concept that some β-lactams may induce acylation of mitochondrial substrate carries, little is known about the functional consequences of acylation of other cellular proteins [67].

Mitochondrial dysfunction It has been suggested that mitochondrial injury may mediate, at least in part, the nephrotoxicity of some β-lactams [67]. Mitochondrial respiration with and uptake of succinate after exposure to toxic doses of cephaloridine, cephaloglycin, or imipenem [98] showed significant reduction of both functions. Cephalexin did not affect either the mitochondrial uptake or respiration with succinate. Depressed mi306

tochondrial respiration secondary to acylation of the mitochondrial transporter for succinate appears to be implicated in renal toxicity caused by cephalosporins and carbapenems [98]. The organic anion fluorescein accumulates in mitochondria of renal proximal tubular cells [99, 100]. Valproate, indometacin, and salicylate induced a significant inhibition of fluorescein [101]. However, cephaloglycin and cephaloridine did not inhibited the fluorescein uptake. This is contrast with the results of previous studies in which an activation of the mitochondrial transporter was described [56]. This discrepancy between the results of these studies may be explained by the involvement of other carrier systems and/or species differences. Using t-butyl hydroperoxide as a model hydroperoxide, the temporal sequence of cellular events leading to renal proximal tubular cell death was determined [102]. The results of the in vitro studies using rabbit isolated tubule suspensions showed that lipid peroxidation and glutathione oxidation are the initial events in t-butyl hydroperoxide-induced toxicity followed by mitochondrial dysfunction and cell death [102]. The temporal sequence of cellular events causing functional impairment and cell death was determined after exposure of rat renal cortical slices or suspensions of rabbit renal cortical tubules to cephaloridine [37, 38]. The results of these studies indicate that GSH depletion and lipid peroxidation are initial events, which precede mitochondrial dysfunction, impairment of the cellular uptake of organic ions and cell death. Moreover, supplementation of GSH to the incubation medium containing renal cortical microsomes significantly reduced cephaloridine-induced lipid peroxidation within the first 3 minutes after onset of incubation [36].

Glutathione and glutathione transferases Reduced glutathione is the most important nonprotein thiol present in animal cells [103]. Most of the intracellular GSH is found in the cytosol. However, a minor mitochondrial pool of GSH contributes to the total cellular pool of glutathione [102, 104, 105]. GSH transferases are inducible enzymes with overlapping substrate specificity [73]. They are also found in renal cells as cytosolic enzymes or as membrane-bound microsomal transferases. GSH conjugates are usually less toxic than their parent compounds and are readily excreted in the bile and in the urine as their correspond-

13. Beta-lactam antibiotics

ing mercapturic acids. However, evidence is accumulating that GSH conjugates and/or their corresponding cysteine conjugates are nephrotoxic [106, 107]. Moreover, intracellular accumulation and cytochrome P450 catalyzed bioactivation of β-lactams such as cephaloridine overwhelms of the GSH redox cycle by inhibiting glutathione reductase activity [35, 56], depletion of GSH and accumulation of GSSG [35, 42, 49, 56]. Most of GSSG formed is subsequently reduced by glutathione reductase and GSH is regenerated with concomitant oxidation (consumption) of NADPH to NADP+ [104]. Depletion of GSH by cephaloridine [34, 35, 56], cephaloglycin and imipenem [56]) was accompanied by a significant rise of GSSG concentration of the renal cortex. GSH depletion in renal cortex was dosedependent and was greatest in rabbits, intermediate in rats and least in mice [42]. This pattern is consistent with the species susceptibility to cephaloridine nephrotoxicity. Further in vitro studies [36-38] using kidney slices and renal proximal tubule suspension were aimed at establishing the temporal sequence of biochemical events leading to cell death. The results of these studies showed that GSH depletion and lipid peroxidation were the earliest measurable events (0.25 to 1.5 hours) occurring after exposure of the renal tissue to cephaloridine [36-38]. These results correlate with the in vivo studies where a significant GSH depletion was measured 1 h after treating animals with cephaloridine, cephaloglycin or imipenem [42, 56]. Modulation of GSH level in cells (inhibition or stimulation) prior to treatment with different compounds affects the cellular response and drug toxicity [42,104,109]. Pretreatment of mice with buthionine sulfoximine enhanced peroxidative injury and trichloroethylene-mediated nephrotoxicity [109]. Similarly, diethylmaleate significantly depleted GSH in the rat renal cortex and potentiated cephaloridine-induced nephrotoxicity [42]. GSH synthesis may be stimulated by the drug oxothiazolidine-4-carboxylate (OTZ). After uptake in the cell, OTZ is enzymatically decarboxylated to yield cysteine, which is then used to synthesize GSH and thus increasing cellular GSH levels [110]. Another way of increasing cellular and tissue GSH levels is by use of GSH esters. The ester group attached to GSH facilitates penetration through the cell membrane inside the cell, where esterases hydrolyze the ester group to

yield free GSH.

Reactive oxygen species and lipid peroxidation It has been shown that the renal bioactivation of xenobiotics such as the herbicides paraquat and diquat [10, 111, 112], and of β-lactams such as cephaloridine and cefsulodin [10, 40, 41] or the antitumor agent adriamycin [113, 114] can induce the generation of reactive oxygen species (oxidative stress) which can be involved in alterations of the structure and functions of cell membranes, cytoskeletal injury, mutagenicity, carcinogenicity, and cell necrosis [115-117]. Reactive oxygen species Although the mechanism(s) of β-lactam-induced nephrotoxicity is not fully elucidated, there is growing evidence that for some of the β-lactams, oxidative stress plays a pivotal role in the chain of events leading to nephrotoxicity and cell death [10, 34, 40]. The univalent reduction during redox cycling of compounds such as paraquat or cephaloridine, after exposure to renal microsomes, leads to production of the superoxide anion radical (Figure 5) [10, 40, 112]. Recent in vitro studies utilizing renal microsomes demonstrate that cephaloridine-induced reactive oxygen species readily oxidized porphyrinogens to porphyrin [118]. Results of in vivo studies in rats show that treatment with cephaloridine (10-500 mg/kg) produced a dosedependent increase in urine concentration of the total porphyrin levels [118]. These results support cephaloridine-induced production of reactive oxygen species, in vivo. Pyridinium ring containing cephalosporins such as cephaloridine, cefsulodine and ceftazidime as well as other β-lactams such as mezlocillin and aztreonam, which do not contain a pyridinium ring, also induce superoxide production in the presence of rat renal microsomes and NADPH [10]. The capacity to generate, and the amount of superoxide produced by a in vitro renal microsome system is dependent on the molecular structure of the specific β-lactam. Superoxide production is a function of exposure time and β-lactam concentration (Figure 5). The rank order of the magnitude of superoxide production by β-lactams in vitro is as follows: cephaloridine > cefsulodin > mezlocillin > aztreonam > ceftazidime > cefotaxime [10]. The magnitude of renal damage caused by oxy307

COJOCEL

gen reactive species will also be influenced by the presence or absence of transition metals. Addition of FeCl2 to a renal microsomes system increased cephaloridine-induced peroxidation of membrane lipids in a concentration-dependent manner [36]. These data are relevant to in vivo conditions where the availability of physiological concentrations of iron is critical. Ferritin, which is present at the subcellular level in the cytosol and endoplasmatic reticulum, appears to be the source for ferric iron in vivo [119].

Figure 5. Concentration-dependent production of superoxide induced by paraquat (Œ), cephaloridine (z) and ceftazidime (). 308

Superoxide generated by xanthine oxidase or in the redox cycling of paraquat can cause the reductive release of F3+ from ferritin, a process that is dependent on the activity of microsomal NADPH-cytochrome P-450 reductase [119]. Iron appears to be an essential component in the formation of reactive species such as superoxide and hydroxyl radical via redox cycling of cephaloridine. Addition of EDTA or of the specific iron chelator desferrioxamine to an incubation system containing renal cortex microsomes and cephaloridine depressed cephaloridine-induced peroxidation of microsomal lipids significantly; EDTA showed a weaker effect than desferrioxamine at equimolar concentrations. By chelating F3+ preferentially [120], desferrioxamine reduced the availability of F2+ produced by the iron redox cycle and decreased cephaloridine-stimulated peroxidation of membrane lipids [36, 37]. Previous studies have shown that renal cortical microsomes are able to catalyze the reduction of cephaloridine in the presence of NADPH with subsequent formation of superoxide and hydrogen peroxide [40]. The divalent reduction of oxygen or the univalent reduction of superoxide yields non-radical species that are protonated at physiological pH to give hydrogen peroxide in a concentration-dependent manner (Figure 6). Hydrogen peroxide, which is a long-lived and membrane permeable species can diffuse and cause injury of cell macromolecules at considerable distances from its generation site. Beta-lactam-induced generation of superoxide and hydrogen peroxide triggers formation of further highly

Figure 6. Concentration-dependent hydrogen peroxide production induced by cephaloridine: 3-24 mmol/l (z), cephaloridine and catalase: 60 μg/ml (S), cephaloridine and catalase: 120 μg/ml (‰), cefotaxime: 3-24 mmol/l (‰).

13. Beta-lactam antibiotics

reactive and cytotoxic oxygen species such as hydroxyl radical. Hydroxyl radical can further contribute in the presence of iron salts, to the decomposition of hydrogen peroxide and formation of additional reactive oxygen species such as singlet oxygen [40]. Kohda and Gemba [120a] have assessed the participation of reactive oxygen species (ROS) generation on cephaloridine nephrotoxicity in rats. Based on chemiluminescence and protein kinase C activity, they detected enhanced ROS generation in mitochondria at both 1.5 and 3.5 hours after cephaloridine administration which preceded histologic damage as assessed by electromicroscopy. They speculate that enhanced PKC and subsequent ROS generation precede changes in plasma parameters and histologic changes characteristic of cephaloridine toxicity. In addition these same authors recently reported a renoprotective effect of a serum thymic factor, FTS, when rats were administered nephrotoxic doses of cephaloridine [120b]. Beta-lactam induced lipid peroxidation Free radical chain reactions, which occur during lipid peroxidation, lead to formation of lipid hydroperoxides that decompose to several types of secondary free radicals and a large number of secondary reactive compounds, such as aldehydes, all resulting in the destruction of cellular membranes and other cytotoxic responses. Under in vivo conditions, liver and kidney microsomal NADPH-cytochrome P-450 reductases are also able of initiating peroxidation reactions resulting in the breakdown of polyunsaturated fatty acids to short-chain products. Uncontrolled, these peroxidation reactions can cause disorganization of membrane

structure, leading to the inactivation of membrane-associated enzymes, membrane leakage and cell death. Activated oxygen species resulting from bioactivation of paraquat and β-lactams react with polyunsaturated fatty acids to cause peroxidation of the cell membrane lipids and subsequent nephrotoxicity [10, 34, 37, 39, 40, 56]. Among the more stabile end products of lipid peroxidation are compound such as malondialdehyde (MDA), ethane, pentane, and hydroxy-trans-neonal. Generation of conjugated dienes, MDA and pentane have been frequently used to demonstrate in vivo induction of β-lactam peroxidative damage in the kidney [34, 37, 56, 109]. Because NADPH-cytochrome P-450 reductase activity is highest in the cortex [77,121] and medullary microsomes lack cytochrome P450 [121], renal cortical tissue was used to investigate peroxidative injury caused by β-lactam accumulation in the kidney. Renal cortical microsomes, slices, tubule and cell suspensions, primary cultured renal cells and established kidney cell lines were exposed to β-lactams with the aim to investigate the subcellular mechanism of the nephrotoxic injury. Studies conducted with renal cortical slices from pig, rabbit and rat revealed that slices from rabbit and rat renal cortex are more susceptible to β-lactam induced peroxidative injury [43]. Comparing the peroxidative potential of cephalosporins of different generations revealed that not only first-generation cephalosporins, but also second-generation cephalosporins such as cefotiam and third- and fourthgeneration cephalosporins (Table 2) can produce a significant increase of lipid peroxidation measured as MDA production [10,26,49].

Table 2. Malondialdehyde (MDA) content and gluconeogenesis as a function of the cephalosporin concentration in the incubation medium. MDA (nmol/h/g tissue) Cephalosporin (mg/ml)

Gluconeogenesis (μmol/h/g tissue)

0

1.25

2.5

5

10

0

1.25

2.5

5

10

Cephaloridine

36.0 ± 6.2

48.4 ± 5.5

65.5* ± 3.1

96.4* ± 2.0

111.2* ± 2.5

26.7 ± 1.7

18.8* ± 2.1

16.2* ± 1.3

5.4* ± 1.9

4.4* ± 1.2

Ceftazidime

38.0 ± 3.6

41.5 ± 3.4

44.4* ± 1.4

48.6* ± 0.4

57.4* ± 2.3

26.9 ± 2.4

28.8 ± 1.3

8.2* ± 3.3

13.8* ± 2.8

2.8* ± 0.7

Cefpirome

41.2 ± 3.1

46.5 ± 4.1

45.2 ± 1.2

47.3 ± 4.0

58.3 * ± 3.1

25.6 ± 3.3

25.7 ± 1.9

21.2 ± 2.9

18.9 ± 3.9

23.0 ± 3.5

Cefotaxime

41.3 ± 2.9

42.8 ± 2.4

41.3 ± 1.9

46.6 ± 3.6

55.4 * ± 4.2

25.5 ± 1.6

25.3 ± 2.7

24.3 ± 4.1

25.7 ± 1.6

26.8 ± 1.3

Data represent mean ± SD from at least 4 rats.* Values are significant at P< 0.05.

309

COJOCEL

Exposure of renal cortical microsomes or primary renal epithelial culture cells to different type of antibiotics led to a significant increase in production of superoxide and MDA after cephaloridine and mezlocillin [10, 40] but not after gentamicin [40, 122]. Significant increase in the cephaloridine-induced MDA generation was manifest in the proximal tubule suspensions while incubation of distal tubules with cephaloridine failed to increase MDA production tubule cell toxicity [36, 41, 50]. Exposure of rabbit and rat isolated proximal tubules or rat renal cortical slices to cephaloridine caused a time- and concentration-dependent generation of MDA [37, 38, 40, 41]. Inhibition of cephaloridine uptake into kidney slices [40] or isolated proximal tubules by 1.0 and 2.0 mM probenecid reduced MDA production in a concentration-dependent manner [39]. These results provide indirect evidence that biochemical processes leading to MDA production do not occur in the incubation medium but within the cortical cells after an obligatory uptake process across the cell membrane. Furthermore, pretreatment of rats with 60 mg/kg cobalt chloride significantly decreased cephaloridine-induced lipid peroxidation in renal cortical slices [31]. Addition of FeCl2 to the incubation medium of renal cortical microsomes caused a significant stimulation of the cephaloridine-induced lipid peroxidation [36, 37]. Collectively, these results are indicative of the cytochrome P450 involvement in the intracellular bioactivation of cephaloridine and its subsequent peroxidative and nephrotoxic action [123]. However, it appears that β-lactams are nephrotoxic through more than one molecular mechanism. Dr Cojocel’s final publication defined the use of +cynidanol-3 and Vitamin E to protect the kidney against cephalosporin induced lipid peroxidation [123a]. Together with his co-authors he reported that, when animals were pretreated with either vitamin E or cyanidanol E , the cephalosporin-induced lipid peroxidation was significantly reduced and the renal cortical PAH uptake improved indicating a renoprotective effect against cephalosporin toxicity. Protection by antioxidants and radical scavengers Under normal physiological conditions the liver and the kidney cells appear to possess adequate defense mechanisms against lipid peroxidation. The most crucial intracellular components of the antiperoxidant defense system are glutathione and the glutathione310

dependent enzymes. The use of the detoxifying enzymes superoxide dismutase and catalase to suppress formation of superoxide and hydrogen peroxide, respectively, as well as specific radical scavengers for the hydroxyl radical and singlet oxygen such as mannitol, (+)-cyanidanol-3, thiourea, sodium benzoate, N-acetyl tryptophan and histidine, effectively decreased paraquat- or cephaloridine-induced peroxidation of microsomal lipids in vitro [15, 40, 41]. The chelation of iron should inhibit the production of hydroxyl radical and therefore mitigate the lipid peroxidation. Deferoxamine, a specific iron chelator, significantly inhibited peroxidation and protects against nephrotoxicity [36, 37]. Moreover, nonspecific antioxidants such as vitamin E, N,N’diphenyl-phenylenediamine, promethazine, probucol or reduced glutathione significantly depressed cephaloridine-induced peroxidation of lipids in renal cortical slices and microsomes [37, 40, 41, 124]. Intracellular signaling pathways of cAMP and protein kinase C (PKC) have been reported to modulate cephaloridineinduced free radicals and nephrotoxicity. [72, 125]. Phorbol myristate acetate (PMA) enhancement of cephaloridine-induced lipid peroxidation and cell injury was blocked by a PKC inhibitor [71].

Alterations of cellular biochemical processes Various β-lactam antibiotics such as cephalosporins and guanylureido penicillins may cause nonimmunologic nephrotoxic effects. The elucidation of the precise biochemical mechanisms involved in nephrotoxicity of β-lactams is of obvious importance for their rational and efficient utilization in the clinical management of infectious disease and for development of future cephalosporins.

Renal transport systems For the zwitterion cephaloridine (CPH) a quantitative correlation between CPH-concentration and the degree of nephrotoxicity has been found [126]. CPH is taken up from blood into the proximal tubule cells and it was assumed that CPH uptake across the basolateral membrane occurs by the transport systems for PAH [127, 128]. However, it was also shown that zwitterionic β-lactams such as CPH can interact with

13. Beta-lactam antibiotics

the cation transport systems [82]. Because rats treated with CPH had altered protein composition and enzymatic activities of membranes from endoplasmic reticulum membranes [74, 75] and since intracellular CPH accumulation and nephrotoxicity was ascribed to relative impermeability of the luminal membrane for CPH [129], the effects of CPHtreatment on transport systems located in the brush

Figure 7. Time-dependent uptake of PAH, L-alanine and glucose by renal brush border membrane vesicles. (O), control rats; (P), rats treated with cephaloridine (1200 mg/kg for 3 days).

border membrane were investigated [77]. The uptake of D-glucose into renal brush border vesicles (BBMV) from control rats is a Na+-dependent transport process which demonstrates an overshoot phenomenon. After treatment of rats with CPH, the D-glucose transport into renal BBMV shows neither a Na+-dependency nor an overshoot phenomenon (Figure 7). Furthermore, the equilibrium values for D-glucose uptake were reduced to 35% of controls in these studies. Similar results were obtained with BBMV from small intestine after treatment of animals with the anticancer drug mitomycin C [130, 131] but the equilibrium uptake values for D-glucose remained unchanged. The effect of CPH-treatment upon the Na+-dependent transport of the amino acid L-alanine was investigated [75]. The results of these studies showed that Na+-dependent transport of L-alanine was also reduced by the treatment with CPH and the overshoot phenomenon completely eliminated (Figure 7). In contrast to D-glucose, the equilibrium uptake values for L-alanine remained unchanged. Weinberg and colleagues [132] found that alanine and glycine can be protective against injury associated with increases in cytosolic free Ca2+, reactive oxygen species, ATP depletion, and Na-K-ATPase inhibition in isolated kidney tubule cells in culture. Thus, the cephaloridineinduced decrease of alanine transport at the luminal cell membrane would diminish the cell defense ability against the toxic injuries caused by oxygen reactive species resulting from intracellular bioactivation of accumulated cephaloridine. The carrier-mediated uptake of p-aminohippuric acid (PAH) into BBMV (Figure 7) and PAH accumulation by renal cortical slices [69,77] were also significantly reduced by CPH treatment (1200 mg/kg/d for 3d). Furthermore, the transport of other cephalosporins across the renal brush border membrane is also affected by CPH-treatment; the uptake of cephalexin and cefotiam into BBMV was greatly reduced whereas the uptake of CPH remained unaffected [77]. Secretion of cephalosporins across the brush border membrane is assumed to occur by the PAH-system as well as by the organic cation/H+-antiporter [127, 133]. Reabsorption of many cephalosporins is performed by the dipeptide transport system [69, 133]. The unaffected uptake of CPH into BBMV from CPH-treated rats indicates that CPH is transported by a system different from the 311

COJOCEL

dipeptide transporter. This is in agreement with results of other studies [82, 127] indicating that CPH interacts with transport systems for organic cations and anions in the brush border membrane. The similar uptake values for CPH in renal BBMV from untreated and CPH-treated rats do not support the previous hypothesis that the brush border membrane is impermeable for CPH [129]. Since cephalexin is transported by the dipeptide transport system [75, 133], the question arose whether or not the reduction of cephalexin transport activity following CPH treatment could be caused by either reduction in the number of transport sites or an impairment of the transport system for β-lactam antibiotics and dipeptides [77]. Using photoaffinity labeling, two membrane polypeptides of brush border membrane of molecular weight of 130,000 and 95,000 were identified as constituents of the dipeptide transport system [77]. The results of this study demonstrated that CPH-treatment of rats greatly reduced the photoaffinity labeling of the binding protein for β-lactam antibiotics and dipeptides with apparent molecular weight 130000. The labeling of the polypeptide of 95000 molecular weight was almost completely depressed [77]. The decrease in labeling intensity of the putative dipeptide transporter is suggestive for a reduction in the number of transport sites following CPH treatment. These results provide further evidence to elucidate the biochemical mechanism by which cephaloridine-induced oxidative injuries alter cell membrane permeability. Recent reports from Endou laboratory in Japan [133a,b,c] have provided insight as to the transport of cephalosporin antibiotics into proximal tubular cells. Both rat and human organic anion transporters (OAT) E have been shown to be involved in the transcellular transport. In particular, human OAT-1 and OAT-3 mediate the basal uptake of cephalosporins from the plasma, while OAT-4 is responsible for the apical transfer. Based in the difference in Ki values between hOAT-4 and hOAT1, Takeda et al [133b] speculate that hOAT-4 limits the efflux of cephaloridine and contributes to it mechanism of nephrotoxicity.

Gluconeogenesis Gluconeogenesis is an important metabolic function of the kidney [134]. Renal cortical slices from nive rats exposed to cephalosporins in vitro or renal 312

cortical slices from animals treated with cephaloridine showed a time- and dose-dependent decrease of renal gluconeogenesis [26, 37, 49]. Glucose synthesis occurred in the proximal but not in the distal tubule suspensions [36]. Inhibition gluconeogenesis within 5 minutes of drug treatment may be an early event in cephaloridine-induced renal toxicity occurring prior to the onset of lipid peroxidation in renal cortical slices [135]. However, decreased gluconeogenesis should not cause cellular necrosis. Interestingly, antioxidants used to protect against cephaloridine-induced inhibition of organic ion accumulation do not block inhibition of gluconeogenesis by cephaloridine [135]. Cephaloridine-induced decrease in gluconeogenesis has been shown to be related to a simultaneous inhibition of the microsomal bound enzyme glucose-6-phosphatase activity in the renal cortex [135]. In contrast, the activity of another rate-limiting enzyme of gluconeogenesis, fructose-1,6-diphosphatase, was not inhibited by cephaloridine [12].

Renal lipid metabolism and protein degradation Penicillin treatment of rabbit neonates (90,000 IU for 2d) altered lipid metabolism in vivo by significantly increasing serum concentration of non-esterified fatty acids and decreasing renal triglyceride content [136]. It appears that penicillin was either decreasing the utilization of non-esterified fatty acids or increasing release. The decrease of renal triglyceride content could be the result of the inhibition of the triglyceride synthesis or penicillin might have increased the utilization of this substrate. Cephaloridine contains a quaternary nitrogen, exists as a zwitterion under physiological conditions and has structural similarities with carnitine. Proximal tubule cells are the internal sites of carnitine acylation [137]. Cephalosporin and carbapenem antibiotics inhibit carnitine tubular reabsorption [68, 138] and mitochondrial uptake of acylcarnitine leading to massive acylcarnitinuria [67]. Newer β-lactam such as cefepime and cefoselis, which possess a quaternary nitrogen as does carnitine, may also inhibit carnitine tubular reabsorption [88]. In order to be metabolized, long-chain fatty acids must first undergo conjugation to carnitine for transport by the acylcarnitine-carnitine carrier across the mitochondrial inner membrane [139]. Short-chain fatty

13. Beta-lactam antibiotics

acids enter the mitochondria through monocarboxylic acid transporters [139]. Studies were carried out to assess the effects cephaloridine, cephaloglycin and cephalexin on the mitochondrial oxidative metabolism of fatty acids such as butyrate and palmitate [67]. The results of these studies showed significant inhibition of palmitoylcarnitine-mediate respiration by cephaloridine in vitro, whereas cephaloglycin, which lacks structural homology with carnitine, caused a greater inhibition of the mitochondrial transport and oxidation of butyrate than cephaloridine. It is possible that the mitochondrial uptake of butyrate was not affected by cephaloridine maybe because the pyridinyl nitrogen hinders its attack on the monocarboxylate receptors. Cephalexin induced only mild in vitro toxicity to the mitochondrial uptake and oxidation of butyrate and palmitate [67]. Cephaloridine effect on the intracellular renal protein degradation was investigated using the labeled low molecular weight protein, 125I-lysozyme. Treatment of rats with cephaloridine for 5 days was followed by administration of 125I-lysozyme one hour prior sacrifice. Release of trichloroacetic acid (TCA) soluble radioactivity into incubation medium from renal cortical slices was used to quantify lysosomal degradation of lysozyme [141]. The results of these experiments showed that cephaloridine caused a dose-dependent decrease of intracellular protein degradation thus impairing the renal metabolism of endogenous and exogenous peptides and proteins taken up by the renal cells.

Clinical toxicity of beta-lactam antibiotics Cephalosporins are the 16th most frequent cause of adverse drug reactions in hospitals in the United States [141a]. Their importance as a cause of nephrotoxicity was recently confirmed by meta-analysis [141b]. Usually β-lactam induced adverse reactions are readily recognized by the clinician. On the other hand, the relationship between antimicrobial activity and the development of a drug-initiated adverse effect can be very subtle and elude the most astute clinician. If a β-lactam is uniquely advantageous for a patient, a carefully controlled rechallenge can be considered to more precisely identify a cause-effect relationship. With appropriate clinical management renal failure caused by β-lactams is often reversible. Identification

and elimination of the risk factors associated with βlactam nephrotoxicity is essential to the prevention of nephrotoxicity. Of these factors, correction of volume depletion and/or congestive heart failure and reversing diminished renal perfusion are of primary importance. While fluid resuscitation can limit the renal damage caused by nephrotoxic β-lactams, there is a risk of overhydration if renal failure develops. Monitoring of serum drug concentration should be helpful to confirm β-lactam-induced renal toxicity, especially when drug interactions are involved.

Interaction with other nephrotoxic drugs Beta-lactam induced renal toxicity can results from their use in monotherapy or when used in combination with other nephrotoxic drugs such as aminoglycosides, amphotericin B, cisplatin, cyclosporine, furosemide, ifosfamide, vancomycin and nephrotoxic β-lactams. While the risk of nephrotoxic injury from monotherapy with β-lactams is relatively low, this risk is substantially increased when multiple drug combinations are required. Benzylpenicillin and ureidopenicillins such as piperacillin and mezlocillin appear to have a little or no nephrotoxic potential when administered alone or in combination with other drugs. Rats treated with piperacillin (1600 mg/kg) and furosemide (100 mg/kg) have elevation blood urea nitrogen (BUN) and creatinine concentration, and mild histologic degeneration of the proximal tubules. These alterations were similar to those observed in rats treated with furosemide alone [142]. The combination of cephalothin with an aminoglycoside was more nephrotoxic than methicillin plus aminoglycoside [143]. There is good evidence that concurrent administration of cephalothin and gentamicin are additive nephrotoxins in humans, especially in patients over 60 years of age as wells as in rabbits [144], and renal injuries are intensified in the presence of mild renal ischemia or endotoxemia [108]. The results of prospective randomized comparative studies of the combination mezlocillin/cefotaxime versus gentamicin/cefoxitin showed that the concurrent administration of mezlocillin/cefotaxime has low renal toxicity and can be recommended for the rational and empirical treatment of serious systemic infections [145]. 313

COJOCEL

Results from animal studies indicate that while furosemide enhanced cephaloridine nephrotoxicity no increased renal toxicity was observed by combining of piperacillin with furosemide [142]. Latamoxef and flomoxef may decrease nephrotoxicity of vancomycin by inhibiting its uptake into the kidney [146, 147]. The results of a retrospective study including renal transplant patients indicate that aztreonam can be safely administered with cyclosporine [148]. Combination therapy with ampicillin/aztreonam in neonates showed a lower renal toxicity than in the group with concurrent administration of oxacillin/amikacin [149].

New lactam antibiotics There is a continuous need for new antibiotics to overcome resistance. However, in the case of β-lactams there is a need to inhibit β-lactamase enzymes, which hydrolyze, and thereby inactivate β-lactam antibiotics. Novel tricyclic carbapenems (trinems) and 2-naphtylcarbapenems have broad spectrum and showed potent activities against gram-negative bacteria [150, 151] including methicillin-resistant Staphylococcus aureus (MRSA). The γ-lactams may be less susceptible to degradation by hydrolases. A number of compounds containing the γ-lactam (pyrrolidin-2-one) moiety show interesting biological and pharmaceutical activities. Some novel monocyclic thienyl γ-lactams are reported to show moderate to high antibacterial activity against gram-positive and gram-negative bacteria [152].

Prevention of clinical toxicity of beta-lactam antibiotics Adverse drug effects represent a major source of morbidity and mortality and must be considered in the differential diagnosis for patients who are experiencing new medical problems or whose clinical status is worsening. Familiarity with β-lactam induced adverse reactions can improve antibiotic selection and reduce adverse events. Before antibiotic therapy is started, the potential benefits and the possible adverse effects should be investigated in light of each patient’s situation. Prevention should be considered in the first place, but if adverse events do occur, they must be recognized and corrected promptly. The most important approach to decreasing βlactam nephrotoxicity is judicious use of these drugs. 314

If a β-lactam is uniquely advantageous for a patient, a carefully controlled rechallenge can be considered to more precisely identify a cause-effect relationship. When β-lactams are used in neonates, accurate determination of the dosage is required, especially for compounds with low therapeutic index and in patients with renal failure. Occurrence of acute renal failure from β-lactam treatment may be prevented by early treatment of serious infections, together with maintenance of hemodynamic stability, renal perfusion, and urinary solute excretion. The β-lactam induced renal failure has a time course comparable to acute tubular necrosis of other origins. While there is no firm evidence that dialysis will speed up renal recovery, clinical stability and good nutrition are likely to improve recovery, as it is also the case with other types of renal failure. Concomitant administration of piperacillin and cephaloridine to rabbits resulted in a significant protective effect against cephaloridine nephrotoxicity [153]. Cephaloridine nephrotoxicity can be prevented by administration of other cephalosporins or penicillins that produce little or no reduction of the cortical concentration of cephaloridine [154]. However, ceftriaxone protects against tobramycin nephrotoxicity by reducing the intracortical accumulation of tobramycin [155]. Combination of tobramycin with latamoxef protects the rat kidney from tobramycin nephrotoxicity, and the protective effect may be partially due to suppression of intrarenal accumulation of tobramycin by latamoxef. This suppression of nephrotoxicity is roughly dependent on the latamoxef dosage [81, 156]. Methimazole (1-methylimidazole-2-thiol) protects against cephaloridine-induced nephrotoxicity when was given 30 min prior cephaloridine administration to rats [157]. Furthermore, cephaloridine transport and accumulation in the kidney was not affected by methimazole [157]. Comparison of cephaloridine-induced nephrotoxicity in normoglycemic and diabetic rats showed lower renal toxicity in diabetic rats than normoglycemic rats. This is apparently due to the fact that the diabetic renal tissue accumulated less cephaloridine than the tissue from normoglycemic rats [158]. Acknowledgements Supported by Kuwait University.

13. Beta-lactam antibiotics

References 1. 2. 3. 4. 5. 6. 7. 8. 8a. 8b. 9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Fleming A. On the antibacterial action of cultures of Penicillium with special reference to their use in the isolation of B.influemzae. Brit J Exper Pathol 1929; 10: 228-236. Abraham EP, Newton GGF. The structure of cephalosporin C. Biochem J 1961; 79: 377-393. Nagarajan R, Boeck LD, Gorman M, Hamil RC, Higgens CE, Hoehn MM, Stark WM, Whitney, JG. β-lactam antibiotics from Streptomyces. J Am Chem Soc 1971; 93: 2308-2310. Page MG. B-Lactamase inhibitors. Drug Resist Updat 2000; 3(2): 109-125. Cooper RD. The carbacephems: a new β-lactam antibiotic class. Am J Med 1992; 92(6A): 2S-6S. Tally FP, Jacobus NV, Gorbach SL. In vitro activity of thienamycin. Antimicrob Agents Chemother 1978; 14 (3): 436-438. Sader HS, Gales AC. Emerging strategies in infectious diseases: new carbapenem and trinem antibacterial agents. Drugs 2001; 61(5): 553-564. Cimarusti CM, and Sykes RB. Monocyclic β-lactam antibiotics. Med Re. Rev 1984; 4(1): 1-24. Jacoby GA and Munoz-Price LS. The new β-lactamases. New Eng J Med 2005;352:380-391 Zhang Q, Matsumura Y, Teratani T, Yoshimoto S, Mineno T, Nakagawa K, Nagahama M, Kuwata S, Takeda H. The application of an institutional clinical data warehouse to the assessment of adverse drug reactions (ADRs). Methods Inf Med 2007;46:516-522 Yousif T, Pooyeh S, Hannemann J, Baumann J, Tauber R, Baumann K. Nephrotoxic and peroxidative potential of meropenem and imipenem/cilastatin in rat and human renal cortical slices. Inter. J Clin Pharmacol Therap 1999; 10: 475-486. Duwe J, Cojocel C, Baumann K. Effect of paraquat-like substances and cephalosporins on accumulation of p-aminohippurate and tetraethylammonium in rat renal cortical slices, and on lipid peroxidation in rat renal microsomes and cortical alices. In: Nephrotoxicity. Mechanisms, early diagnosis, and therapeutic management. Bach PH, Gregg NJ, Wilks MF, Delacruz L (editors). Marcel Dekker Inc, New York 1991; p. 13-17. Tune BM. The nephrotoxicity of cephalosporin antibiotics-structure-activity relationships. Comm Toxicol 1986; 1(2): 145-170. Goldstein RS, Smith PF, Tarloff, JB, Contardi L, Rush GF, Hook JB. Biochemical mechanisms of cephaloridine nephrotoxicity. Life Sci 1988; 42(19): 1809-1916. Kaloyanides GJ. Metabolic interactions between drugs and renal tubulo-interstitial cells. Role in nephrotoxicity. Kidney Int 1991; 39: 531-540. Schnellmann RG, Griner RD. Mitochondrial mechanisms of tubular injury. In: Mechanisms of injury in renal disease and toxicity. Goldstein RS, Editor.CRC Press Inc, Boca Raton, Florida, USA 1994; p. 247-265. Tomita M, Okuyama T, Watanabe S, Watanabe H. Quantitation of the hydroxyl radical adducts of salicylic acid by micellar electrokinetic capillary chromatography: oxidizing species formed by a Fenton reaction. Arch Toxicol 1994; 68: 428 - 433. Tomita M., and Okuyama T. Effect of paraquat on the malondialdehyde level in rat liver microsomes (in vitro). Arch. Toxicol 1994; 68: 187-192. Tune BM, Hsu C-H, Fravert D. Cephalothin and carbacephem nephrotoxicity. Role of tubular cell uptake and acylating potential. Biochem Pharmacol 1996; 51: 557-561. Roselle GA, Clyne DH, Kaufman CA. Carbenicillin nephrotoxicity. South Med J 1978; 71(1): 84-86. Elis D, Fried WA, Yunis EJ, Blau EB. Acute interstitial nephritis in children: a report of 13 cases and review of the literature. Pediatrics 1981; 67(6): 862-870. Pommer W, Offerman G, Schultze G, Krause PH, Molzhan M. Acute interstitial nephritis caused by drugs. Dtsch Med Wochenschr 1983; 108(20): 783-788. Browning MC, Tune BM. Reactivity and binding of β-lactam antibiotics in rabbit renal cortex. J Pharmacol. Exp Ther 1983; 226(3): 640-644. Geller RJ, Chevalier RL, Spyker DA. Acute amoxicillin nephrotoxicity following an overdose. J. Toxicol Clin Toxicol 1986; 24(2): 175-182. Schellie SF, Groshong T. Acute interstitial nephritis following amoxicillin overdose. Mo Med 1999; 96(6): 209-211. Hedstrom SA, Hybbinette CH. Nephrotoxicity in isoxazolylpenicillin prophylaxis in hip surgery. Acta Orthop Scand 1988; 59(2): 144-147. Dunn GL. Ceftizoxime and other third-generation cephalosporins: Structure activity relationships. J Antimicrob Chemother 1982; 10: 1-10. Cojocel C, Goettsche U., Toelle K-L., Baumann, K. Nephrotoxic potential of first-, second-, and third-generation cephalosporins. Arch Toxicol 1988; 62: 458-464. Fillastre JP, Kleinknecht D. Acute renal failure associated with cephalosporin therapy. Am Heart J 1975; 89(6): 809-810. Silverblatt F, Turck M, Bulger R. Nephrotoxicitydue to cephaloridine: a light- and electron-microscopy study in rabbits. J Infect Dis 1970; 122(1): 33-44.

315

COJOCEL

29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56.

316

Yamada Y. Studies on the mechanisms of renal damage induced by nephrotoxic compounds. Nippon Hoigaku Zasshi 1995; 49(6): 447-457. Tune BM. Effect of organic transport inhibitors on renal uptake and proximal tubular toxicity of cephaloridine. J Pharmacol Exper Therap 1972; 181: 250-256. Cojocel C, Laeschke KH, Insellman G, Baumann K. Inhibition of cephaloridine-induced lipid peroxidation. Toxicology 1985; 35(4): 295-305. Takeda M, Tojo A, Sekine T, Hosoyamada M, Kanai Y, Endou H. Role of organic anion transporter 1 (OAT1) in cephaloridin (CER)induced nephrotoxicity. Kidney Int 1999; 56: 2128-2136. Jariyawat S, Sekine T, Takeda M, Apiwattanakul N, Kanai Y, Sophasan S, Endou H. The interaction and transport of β-lactam antibiotics with the cloned rat renal organic anion transporter 1. J Pharmacol Exp Ther1999; 290 (2): 672-677. Kuo C-H, Maita, K, Sleight SD, Hook JB. Lipid peroxidation: a possible mechanism of cephaloridine-induced nephrotoxicity. Toxicol Appl Pharmacol 1983; 67: 78-88. Tune B, Fravert D, Hsu C-Y. Oxidative and mitochondrial toxic effects of cephalosporin antibiotics in the kidney. A comparative study of cephaloridine and cephloglycine. Biochem Pharmacol 1989; 38(3): 795-802. Cojocel C, Mayer D. Protection by glutathione and antioxidants against cephaloridine-induced lipid peroxidation. Biochem. Pharmacol (Life Sci Adv) 1991; 10: 41-51. Goldstein RS, Pasino DA, Hewit WR, Hook JB. Biochemical mechanisms of cephaloridine nephrotoxicity: time and concentration dependence of peroxidative injury. Toxicol Appl Pharmacol 1986; 83: 261-270. Rush GF, Ponsler GD. Cephaloridine-induced biochemical changes and cytotoxicity in suspensions of rabbit isolated proximal tubules. Toxicol Appl Pharmacol 1991; 109: 314-326. Rush GF, Heim RA, Ponsler GD, Engelhardt J. Cephaloridine-induced renal pathological and biochemical changes in female rabbits and isolated proximal tubules in suspension. Toxicol Pathol 1992; 20(2): 155-168. Cojocel C, Hannemann J, Baumann K. Cephaloridine-induced lipid peroxidation initiated by reactive oxygen species as a possible mechanism of cephaloridine nephrotoxicity. Biochim Biophys Acta 1985; 834: 402–410. Suzuki Y, Sudo J. Lipid peroxidation and generation of oxygen radicals induced by cephaloridine im renal cortical microsomes of rats. Jpn J Pharmacol 1990; 52(2): 233-243. Kuo C-H, Hook JB. Depletion of renal gluthathione content and nephrotoxicity of cephaloridine in rabits, rats, and mice. Toxicol Appl Pharmacol 1982; 63: 292-302. Cojocel C, Inselmann G, Laeschke KE, Baumann K. Species differences in cephalosporin-induced lipid peroxidation. Drugs Exptl Clin Res 1984; 10(11): 781-784. Kays SE, Crowell WA, Johnson MA. Cephaloridine nephrotoxicity is potentiated by selenium deficiency but not by copper deficiency in rats. J Nutr 1992; 122: 1232-1241. Alitalo R, Ruutu M, Valtonen V, Lehtonen T, Pentkäinen PJ. Hyoprothrombinaemia and bleeding during administration of cefamandole and cefoperazone. Ann Clin Res 1985; 17: 116-119. Wachsmuth ED. Nephrotoxicity of cefotiam (CGP 14221/E) in rats and rabbits. Arch Toxicol 1996; 48: 135-156. Norrby SR, Burman LA, Linderhol H, Trollfors B. Ceftazidime: pharmakokinetics in patients and effects on renal function. J Antimicrob Chemother 1982; 10(3): 199-206. Alestig K, Trollfors B, Andersson R, Oolaison L, Suurkula M, Norrby SR. Ceftazidime and renal function. J Antimicrob Chemother 1984; 13(2): 177-181. Cojocel C. Biochemical aspects of the renal tolerance for cefpirome and other cephalosporins. rzneimittelforschung 1990; 40(10):1140-1144. Suzuki H, Imamura K, Yoshida T, Shibata M, Matsuzawa T, Ozaki H, Sakai T, Shiobara Y, Tachibana A, Yanko K. Animal pharmacokinetis and toxicology of cefotetan - a new cephamycin antibiotic. J Antimicrob Chemother 1983; 11(Suppl): 179-183. Thieme RE, Caldwell SA, Lum GM. Acute interstitial nephritis associated with loracarbef therapy. J Pediatr 1995; 127(6): 9971000. Birnbaum J, Kahan FM, Kropp H, Macdonald JS. Carbapenems, a new class of β-lactam antibiotics: discovery and development of imipenem/cilastatin. Am J Med 1985; 78(Suppl 6A): 3-21. Kropp H, Sundelof JG, Hajdu R, Kahan FM. Metabolism of thienamycin and related carbapenem antibiotics by the renal dipeptidase, dehydropeptidase-1. Antimicrob Ag Chemother 1982; 22: 62-70. Cunha BA. Meropenem in elderly and renally impaired patients. Int J Antimicrob Agents 1988; 10: 107-117. Asbel LE, Levison ME. Cephalosporins, carbapenems, and monobactams. Infect Dis Clin North Am 2000; 14(2): 435-437. Tune BM., Fravert D, Hsu C-Y. Thienamycin nephrotoxicity. Mitochondrial injury and oxidative effects of imipenem in the rabbit kidney. Biochem Pharmacol 1989; 38(21): 3779-3783.

13. Beta-lactam antibiotics

57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67.

68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81.

82. 83.

Hirouchi Y, Naganuma H, Kawahara Y, Okada R, Kammiya A, Inui K, Hori R. Preventive effect of ßmipron on nephrotoxicity and uptake of carbapenems in rabbit renal cortex. Jpn J Pharmacol 1994; 66(1): 1-6. Cunha BA. Aztreonam. Urology 1993; 41(3): 249-258. Moore RD, Lerener SA, Levine, DP. Nephrotoxicity and ototoxicity of aztreonam versus aminoglycoside therapy. J Infect Dis 1992; 165(4): 683-688. Chartrand SA. Safety and toxicity profile of aztreonam. Pediatr Infect Dis J 1989; 8(Suppl 9): 120-123. Ohya S, Yamazaki M, Sugawara S. Effect of 7 α substitution of cephems on their β-lactamase stability and affinity for penicillinbinding proteins in Morganella morganii. Antimicrob Agents Chemother 1983; 23(4): 522-525. Patel IH, Kaplan SA. Phrmacokinetic profile of ceftriaxone in man. Am J Med 1984; 77(4C): 17-25 Sattler FR, Weitekamp MR, Ballard JO. Potential for bleeding with the new β-lactam antibiotics. Ann Intern Med 1986; 105(6): 924-931. Goto K, Oda M, Saitoh H, Nishida M, Takada M. Effect of side chains including N-methyl-tetrazole-thiol group of β-lactam antibiotics on transport in cultured kidney epithelial cells LLC-PK1. Biol Pharm Bull 1988; 21(10): 113-116. Williams PD, Laska DA, Tay LK, Hottendorf, GH. Comparative toxicities of cephalosporin antibiotics in a rabbit kidney cell line (LLC-RK1). Antimicrob Agents Chemother 1988; 32(3): 314-318. Viotte G, Morin GP, Godin M, Fillastre JP. Chamges in the renal function of rats treated with cefoxitin and a comparison with other cephlosporins and gentamicin. J Antimicrob Chemother 1981; 7(5): 537-550 Tune BM. The renal toxicity of β-lactam antibiotics: mechanism and clinical impications. In: Clinical nephrotoxins – renal injury from drugs and chemicals. 1st Ed.. De Broe ME, Porter GA, Bennett WM, Verpooten GA (editors). Kluwer Academic Publ, Dordrecht, 1998; p. 121-134. Topham JC, Murgatroyd LB, Johns DV, Goonetilleke UR, Wright J. Safety evaluation of meropenem in animals: studies on the kidney. J Antimicrob Chemother 1989; 24(Suppl A): 287-386. Harrison MP, Jones DV, Pickford RJ, Wilson ID. ß-Hydroxybutyrate: a urinary marker of imipenem induced nephrotoxicity in the cynomolgus monkey detected by high field spectroscopy. Biochem Pharmacol 1991; 41(12): 2045-2049. Nouda H, Matsumura H, Tanio T, Sunagawa M. Structural feature of carbapenem compounds or nephrotoxicity: effect of C-2 side chain. J Antibiotics 1996; 49(6): 603-606. Kohda Y, Gemba M. Modulation by cyclic AMP and phorbol myristate acetate of cephaloridine-induced injury in rat renal cortical slices. Jpn J Pharmacol 2001; 85(1): 54-59. Cojocel C, Maita K, Pasino DA, Kuo, C-H, Hook J B. Metabolic heterogeneity of proximal and distal tubules. Life Sci 1983; 33: 855861. Lock EA. Renal drug-metabolizing enzymes in experimental animals and humans. In: Mechanisms of injury in renal disease and toxicity. Goldstein R S (editor). CRC Press Inc, Boca Raton, Florida 1994; p. 173-206. Cojocel C, Kramer W, Mayer D. Depletion of cytochrome P450 and alterations in activities of drug metabolizing enzymes induced by cephaloridine in rat kidney cortex. Biochemical Pharmacol 1988; 37(19): 3781-3785. Kramer W, Cojocel C, Mayer D. Specific alterations of rat renal microsomal proteins induced by cephaloridine. Biochem Pharmacol 1988; 37 (21): 4135-4140. Olivier MF, Dutertre-Catella H, Thevenin M, Martin C, Warnet JM, Claude JR. Increased reduced glutathione and glutathione Stransferase activity in chronic cephaloridine nephrotoxicity studies in the rat. Drug Chem Toxicol 1990; 13(1-2): 209-219. Kramer W, Cojocel C, Mayer D. Effect of cephaloridine treatment on transport systems of the rat renal brush border membrane. Biochem Pharmacol (Life Sci Adv) 1990; 9: 127-133. Fry M, Plummer DT. The interaction of cephaloridine with model membrane systems and rat kidney lysosomes. Chem Biol Interact 1979; 25(1): 113-124. Ngaha EO. Further studies on the in vivo effect of cephaloridine on the stability of rat kidney lysosomes. Biochem Pharmacol 1982; 31(10): 1843-1847. Yamagouchi A, Hiruma R, Sawai T. Phospholipid bilayer oermeability of β-lactam antibiotics. J Antibiot (Tokyo) 1982; 35(12): 1692-1696. Kojima R, Ito M, Suzuki Y. Studies on the nephrotoxicity of aminoglycoside antibiotics and protection from these effects (4). Effects of tobramycin alone and in combination with latamoxef on the stability of rat kidney lysosomal membranes. Jpn J Pharmacol 1987; 43(1): 73-80. Ullrich KJ, Rumrich G, Kloss S. Contraluminal organic anion and cation transport in the proximal renal tubule: V. Interaction with sulfamoyl- and phemoxy diuretics, and with β-lactam antibiotics. Kidney Int 1989; 36(1): 78-88. Van Aubel RAMH, Masereeuw R, Russsel FGM. Molecular pharmacology of renal organic anion transporters. Am J Physiol Renal Physiol 2000; 279(2): F216-F232.

317

COJOCEL

84. 85.

86. 87.

88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111.

318

Williams PD, Hitchcock MJ, Hottendorf GH. Effect of cephalosporins on organic ion transport in renal membrane vesicles from the rat and rabbit kidney cortex. Res Commun Chem Pathol Pharmacol 1985; 47(3): 357-371. Ullrich KJ, Rumrich G, Wieland T, Dekant W. Contraluminal para-aminohippurate (PAH) transport in the proximal tubule of the rat kidney. VI. Specificity: amino acids, their N-methyl-, Nacetyl- and N-benzoylderivatives; glutathione- and cysteine conjugates, di- and oligopeptides. Pflügers Arch 1989; 415(3): 342-350. Atkinson RM, Curie LP, Prat PAH, Sharpe HM, Tomich EG. Acute toxicity of cephaloridine, an antibiotic derived from cephalosporin C. Toxicol Appl Pharmacol 1966; 8(3): 398-406. Schaub TP, Kartenbeck J, König J, Spring H, Dorsam J, Sthaeler G, Störkel S, Thon WF, Keppler D. Expression of the MRP2 geneencoded conjugate export pump in human kidney proximal tubule and in renal carcinoma. J Am Soc Nephrol 1999; 10(6): 1159-1169. Ganapathy ME, Huang W, Rajan DP, Carter AL, Sugawara M, Iseki K, Leibach FH, Ganapathy V. β-lactam antibiotics as substrates for OCTN2, an organic cation/carnitine transporter. J Biol Chem 2000; 275(3): 1699-1707. Terada T, Saito H, Mukai M, Inui K-I. Recognition of β-lactam antibiotics by rat peptide transportrers, PEPT1 and PEPT2 in LLC-Pk1 cells. Am J Physiol Renal Physiol 1997; 273(42): F706-F711. Daniel H, Herget M. Cellular and molecular mechanisms of renal peptide transport. Am J Physiol Renal Physiol 1987; 273:F1F8. Hori R, Ishikawa Y, Takano M, Okano T, Kitazawa S, Inui K. The interaction of cephalosporin antibiotics with renal cortex of rats: accumulation to cortical slices and binding to purified plasma membranes. Biochem Pharmacol 1982; 31(13): 2267-2272. McMurty RJ, Mitchell JR. Renal and hepatic necrosis after metabolic activation of 2-sustituted furans and thiophenes, including furosemide and cephaloridine. Toxicol Appl Pharmacol 1977; 42: 285-300. Spurling NW, Harcourt RA, Hyde JJ. An evaluation of the safety of cefuroxime axetil during six months oral administration to beagle dogs. J. Toxicol Sci 1986; 11(4): 237-277. Johns RN. A review of cephalosporin metabolism: a lesson to be learned for future chemotherapy. Diagn. Microbial Infect Dis 1989; 12(1): 25-32. Marone P, Navarra A, Monzillo V, Traverso A. Antibacterial activity of combined cefotaxime and desacetyl-cefotaxime against aerobic gram-negative bacilli. Drugs Exp Clin Res 1990; 16(12): 629-633. Hottendorf GH, Laska DA, Williams PD, Ford SM. Role of desacetylation in the detoxification of cephalothin in renal cells in culture. J Toxicol Environ Health 1987; 22(1): 101-111. Indelicato JM, Dinner A, Peters LR, Wilham WL. Hydrolysis of 3-chloro-3-cephems. Intramolecular nucleophilic attack in cefaclor. J Med Chem 1977; 20: 961-963 Tune BM, Hsu C-Y. The renal mitochondrial toxicity of β-lactam antibiotics: In vitro effects of cephaloglycin and imipenem. J Amer Soc Nephrol 1990; 1(5): 815-821. Masereeuw R, van den Bergh EJ, Bindels RJ, Russel Fg. Characterization of fluorescein transport in isolated proximal tubular cells of the rat: Evidence for mitochondrial accumulation. J Pharmacol Exp Ther 1994; 269: 1261-1267. Miller DS, Stewart DE, Pitchard JB. Inracellular compartimentation of organic anions within renal cells. Am J Physiol 1993; 264: R882-R890. Terlouw SA, Tanriseven O, Russel FGM, Masereeuw R. Metabolite anion carriers mediate the uptake of the anion drug fluorescein in renal cortical mitochondria. J Pharmacol Exp Ther 2000; 293(3): 968-973. Schnellmann R G, Gilchrist S M, Mandel L J. Intracellular distribution and depletion of glutathione in rabbit renal proximal tubules. Kidney Int 1988; 34: 229-233. Meister A, Anderson M A. Glutathione. Annu Rev Biochem 1985; 52: 711-760. Reed D J. Regulation of reductive processes by glutathione. Biochem Pharcol 1986; 35(1): 7-13. Moldeus P, Quanguang J. Importance of the glutathione cycle in drug metabolism. Pharmacol Ther 1987; 33: 37-40. Elfarra AA, Anders MW. Renal processing of glutathione conjugates. Role in nephrotoxicity. Biochem Pharmacol 1984; 33(23): 3729-3732. Beuter W, Cojocel C, Muller W, Donaubauer HH, Mayer D. Peroxidative damage and nephrotoxicity of dichlorovinylcysteine in mice. J Appl Toxicol. 1989; 9(3): 181-186. Tune BM, Hsu C-Y, Fravert D. Mechanisms of bacterial endotoxin-cephaloridine toxic synergy and the protective effects of saline infusion in the rabbit kidney. J Pharmacol Exp Ther 1988; 244(2): 520-525. Cojocel C, Beuter W, Müller W, Maye, D. Lipid peroxidation: a possible mechanism of trichloroethylene-induced nephrotoxicity. Toxicology 1989; 55: 131-141. Mitchell JB, Biaglow JE, Russo A. Role of glutathione and other endogenous thiols in radiation protection. Pharmacol Ther 1988; 39(1-3): 269-274. Bus JS, Gibson JE. Paraquat: model for oxidant-initiated toxicity. Environ Health Persp 1984; 55: 37-46.

13. Beta-lactam antibiotics

112. Tomita M. Comparison of one-electron reduction activity against the bipyridylium herbicides, paraquat and diquat, in microsomal and mitochondrial fractions of liver, lung and kidney (in vitro). Biochemical Pharmacol 1991; 42(2): 303-309. 113. Gianni L, Zweier JL, Levy A, Mayers CE. Characterisation of the cycle of iron-mediated electron transfer from adriamycin to molecular oxygen. J Biol Chem 1985; 260(11): 6820- 6826. 114. Mimnaugh EG. Potentiation by reduced glutathione of adriamycin-stimulated lipid peroxidation in kidny microsomes. Biochem Pharmacol 1986; 35(23): 4337-4339. 115. Fridovich I. Biological effects of the superoxide radical. Arch Biochem Biophys 1986; 247: 1-11. 116. Halliwell B, Gutteridge JMC. Oxygen free radicals and iron in relation to biology and medicine: some problems and concepts. Arch Biochem Biophys 1986; 246: 501-514. 117. Li W, Zhao Y, Chou IN. Paraquat-induced cytoskeletal injury in cultured cells. Toxicol Appl Pharmacol 1987; 91: 96-106. 118. Miller DM, Woods JS. Urinary porphyrins as biological indicators of oxidative stress in the kidney. Interaction of mercury and cephaloridine. Biochem Pharmacol 1993; 46 (12): 2235-2241. 119. Thomas CE, Aust SD. Reductive release of iron from ferritin by cation free radicals of paraquat and other bipyridils. J Biol Chem 1986; 261(28): 13064-13070. 120. Minotti G, Aust SD. The role of iron in the initiation of lipid peroxidation. Chem Phys Lipids 1987; 44(2-4): 191-208. 120a. Kohda Y, Gemba M. Enhancement of protein kinase C activity and chemiluminescence intensity in mitochondria isolated from the kidney cortex of rats treated with cephaloridine. Biochem Pharmacol 2002;64:543-549. 120b.Kohda Y, Matsunaga Y, Katsuya Y, Yoshiko K, Aways A, Gemba M. Protective effect of serum thymic factor, FTS, on cepholridineinduced nephrotoxicity in rats. Biol Pharm Bull 2005;28:2087-2091. 121. Zenser TV, Mattammal MB, Davis BB. Differential distribution of the mixed-function oxidase activities in rabbit kidney. J Pharmacol Exp Ther 1978; 207(3): 719-725. 122. Swann JD, Acosta D. Failure of gentamicin to elevate cellular malondialdehyde content or increase generation of intracellular reactive oxygen species in primary cultures of renal cortical epithelial cells. Biochem Pharmacol 1990; 40(7): 1523-1526. 123. Tune BM, Kuo CH, Hook JB, Hsu CY, Fravert D. Effects of piperonyl butoxide on cephalosporin nephrotoxicity in the rabbit. An effect on cephaloridine transport. J Pharmacol Exp Ther 1983; 224(3): 520-524. 123a. Cojocel C, Tolle K-L, El-Hajj H, Baumann K. Protection against cephalosporin-induced lipid peroxidation and nephrotoxicity by (+)-cyanidanol-3 and vitamin E. Brazilian J Med Biol Res. 2007;40:867-875. 124. Roy K, Saha A, De K, Sengupta C. Ceftriaxone induced lipid peroxidation and its inhibition with various antioxidants: Part II. Evaluation of glutathione and probucol as antioxidants. Acta Pol Pharm 2000; 57(5): 385-390. 125. Kawai Y, Gemba M. Amelioration by cAMP of cephaloridine-induced injury in the porcine kidney cell line LLC-PK1. Jpn J Pharmacol 1996; 72(1): 67-70. 126. Tune BM, Wu KY, Longerbeam DF, Kempson RL. Transport and toxicity of cephaloridine in the kidney. Effect of furosemide, paminohippurate and saline diuresis. J Pharmacol Exp Ther 1977; 202(2): 472-478. 127. Kasher JS, Holohan PD, Ross CR. Na+ gradient-dependent p-aminohippurate (PAH) transport in rat basolateral membrane vesicles. J Pharmacol Exp Ther 1983; 227(1): 122-129. 128. Tune BM, Fernholt M. Relationship between cephaloridine and p-aminohippurate transport in the kidney. Am J Physiol 1973; 225(5): 1114-1117. 129. Tune BM, Fernholt M, Schwartz A. Mechanism of cephaloridine transport in the kidney. J Pharmacol Exp Ther 1974; 191(2): 311317. 130. Mizuno M, Hamaura T, Hashida M, Sezaki H. Changes in D-glucose uptake by brush-border vesicles from small intestine of rats treated with mitomycin C. Biochem Pharmacol 1986; 35(7): 1153-1158. 131. Mizuno M, Yoshino H, Hashida M, Sezaki H. Decreased transport of D-glucose and L-alanine across brush-border membrane vesicles from small intestine of rats treated with mitomycin C. Biochim Biophys Acta 1987; 902(1): 93-100. 132. Weinberg JM, Venkatachalam MA, Roeser NF, Davis JA, Varani J, Johnson KJ. Amino acid protection of cultured kidney tubule cells against calcium ionophore-induced lethal cell injury. Lab Invest 1991; 65: 671-678. 133. Inui K, Saito H, Takano M, Okano T, Kitazawa S, Hori R. Enzyme activities and sodium-dependent active D-glucose transport in apical membrane vesicles isolated from kidney epithelial cell line (LLC-PK1). Biochim Biophys Acta 1984; 769(2): 514-518. 133a. Jung KY, Takeda M, Shimoda M, Narikawa S, TojoA, Kim DK, Chairoungdua A, Choi BK, Kusuhara H, SugiyamaY, Sekine T, Endou H. Involvement of rat organic anion transporter 3 (rOAT3) in cephaloridine-induced nephrotoxicity: in comparion with rOAT1. Life Sci 2002;70:1861-1874 133b. Takeda M, Babu E, Narikawa S, Endou H. Interaction of human organic anion transporters with various cephalosporin antibiotics. Eur J Pharmacol 2002;438:137-142.

319

COJOCEL

133c. Khamdang S, Takeda M, Babu E, Noshiro R, Onozato ML, Tojo A, Enomoto A, Huang XL, Narikawa S, Anzai N, Piyachaturawat P, Endou H. Interaction of human and rat organic anion transporter 2 with various cephalosporin antibiotics. Eur J Pharmacol 2003;465:1-7. 134. Cersosimo E, Garlick P, Ferretti J. Renal substrate metabolism and gluconeogenesis during hypoglycemia in humans. Diabetes 2000; 49: 1186-1193. 135. Goldstein RS, Contardi LR, Pasino DA,Hook JB. Mechanisms mediating cephaloridine inhibition of renal gluconeogenesis. ToxicolApplPharmacol 1987; 87: 297-305. 136. Stroo WE, Hook JB. Dissociation of renal organic anion transport from renal lipid metabolism. I. Endogenous nonesterified fatty acids (NEFA) as determinants of transport. J Pharmacol Exp Ther 1983; 227(1): 55-59. 137. Wagner S, Deufel T, Guder WG. Carnitine metabolism in isolated rat kidney cortex tubules. Biol Chem Hoppe Seyler 1986; 367(1): 75-79. 138. Arrigoni-Martelli E, Caso V. Carnitine protects mitochondria and removes toxic acyls from xenobiotics. Drugs Exp Clin Res 2001; 27(10): 27-49. 139. Bieber LL, Farel SS. Carnitine acyl transferases. Enzymes 1983; 16: 624-644. 140. Ramsay RR, Gandour RD, van der Leij FR. Molecular enzymology of carnitine transfer and transport. Biochem Biophys Acta 2001; 1546(1): 21-43. 141. Cojocel C, Smith JH, Maita K, Sleight SD, Hook JB. Renal protein degradation: a biochemical target of specific nephrotoxicants. Fund Appl Toxicol 1983; 3: 278-284. 141a. Bond CA and Raehl CL. Adverse drug reactions in United States hospitals. Pharmacother 2006;26:601-608 141b. Falagas ME, Matthaiou DK, Karveli EA, Peppas G. Meta-analysis: randomized controlled trials of clindamycin/aminoglycoside vs. β-lactam monotherapy for the treatment of intra-abdominal infections. Aliment Pharmacol Ther. 2007;25:537-556. 142. Hori R, Shimakura M, Aramata Y, Kizawa K, Nozawa I, Takahata M, Minami S. Nephrotoxicity of piperacillin combined with furosemide in rats. Jpn J Antibiot 2000; 53(8): 582-591. 143. Wade JC, Smith CR, Petty BG, Lipski JJ, Conrad G, Ellner J, Lietman PS. Nephrotoxicity of piperacillin combined with furosemide in rats. Jpn J Antibiot 2000; 53(8): 582-591. 144. Bendirdjian JP, Prime DJ, Browning MC, Hsu CY, Tune BM. Additive nephrotoxicity of cephalosporins and aminoglycosides in the rabbit. J Pharmacol Exp Ther 1981; 218(3): 681-685. 145. Kosmidis J, Daikos GK. Prospective randomized comparative studies of mezlocillin/cefotaxime vs. gentamicin/cefoxitin. J Antimicrob Chemother 1983; 11(Suppl C): 91-95. 146. Itoh F, Sato K, Harauchi T, Hirata M, Mizushima Y. Modification of vancomycin nephrotoxicity by other antibiotics in rats. Jpn J Antibiot 1995; 48(3): 380-388. 147. Toyoguchi T, Nakagawa Y. Nephrotoxicity and drug interaction of vancomycin (2). Nippon Yakurigaku Zasshi 1996; 107(5): 225235. 148. Reitbroeck RC, Hoitsma AJ, Koene RA. Aztreonam can be safely used with cyclosporin without aggravating nephrotoxicity. Transpl Int 1989; 2(4): 232-234. 149. Fanos V, Musap M, Verlato G, Plebani M, Padovani EM. Evaluation of the antibiotic-induced nephrotoxicity in preterm neonates by determining urinary a1-microglobulin. Pediatr Nephrol 1996; 10: 645-647. 150. Greenlee ML, DiNinno F, Herrmann JJ, Jaworsky C, Muthard DA, Salzmann TN. 2-Naphthylcarbapenems: broad spectrum antibiotics with enhanced potency against MRSA. Bioorg Med Chem Lett 1999; 9(19): 2893-2896. 151. Kanno O, Shimoji Y, Ohya S, Kawamoto I. Synthesis and biological evaluation of novel tricyclic carbapenems (trinems). J Antib (Tokyo) 2000; 53(4): 404-414. 152. Kar GK, Roy BC, Adhikari SD, Ray JK, Brahama NK. Synthesis of some thieno γ lactam monocyclic acids with high antibacterial activity: a new look at an old molecular system. Bioorg Med Chem 1998; 6(12): 2397-2402. 153. Hayashi T, Watanabe Y, Kumano K, Kitayama R, Yasuda T, Saikawa I, Katahira J, Kumada T, Shimizu K. Protective effect of piperacillin against nephrotoxicity of cephaloridine and gentamicin in animals. Antimicrob Agents Chemother 1988; 32(6): 912-918. 154. Tune BM, Browning MC, Hsu CY, Fravert D. Prevention of cephalosporin nephrotoxicity by other cephalosporins and by penicillins without significant inhibition of renal cortical uptake. J Infect Dis 1982; 145(2): 174-180. 155. Beauchamp D, Theriault G, Grenier L, Gourde P, Perron S, Bergeron Y, Fontaine L, Bergeron MG. Ceftriaxone protects against tobramycin nephrotoxicity. Antimicrob Agents Chemother 1994; 38(4): 750-756. 156. Kojima R, Ito M, Suzuki Y. Studies on the nephrotoxicity of aminoglycoside antibiotics and protection from these effects (3). Protective effect of latamoxef against tobramycin nephrotoxicity and its protective mechanism. Jpn J Pharmacol 1986; 42(3): 397-404. 157. Sausen PJ, Elfarra AA, Cooley AJ. Methimazole protection of rats against chemically induced kidney damage in vivo. J Pharmacol Exp Ther 1992; 260(1): 393-401.

320

13. Beta-lactam antibiotics

158. Valentovic M, Ball JG, Rogers BA, Meadows MK, Harmon RC, Moles J. Cephaloridine in vitro toxicity and accumulation in renal slices from normoglycemic and diabetic rats. Fundam Appl Toxicol 1997; 38(2): 184-190.

321

14

Amphotericin B Nathalie K. ZGHEIB1, Blair CAPITANO1 and Robert A. BRANCH1 1University

of Pittsburgh, Pittsburgh, Pennsylvania, USA

Introduction ___________________________________________________________ 324 Clinical manifestations of nephrotoxicity ____________________________________ 324 Incidence and risk factors Urinary concentration defects Electrolyte disturbances Renal tubular acidosis

325 326 326 327

Pathological findings ____________________________________________________ 327 Mechanisms of nephrotoxicity _____________________________________________ 328 Effects on cell membranes Effects on physiological parameters: whole animal studies

328 330

Measures to reduce nephrotoxicity _________________________________________ 332 Salt supplementation Continuous Infusion Other formulations: lipid formulation of amphotericin B Alternative agents

333 334 335 341

Clinical use_____________________________________________________________ 342 General underlying condition of the patient Amphotericin B administration Concomitant use of nephrotoxic drugs Potassium and magnesium supplementation Follow-up

342 342 343 343 344

Conclusion _____________________________________________________________ 344 References _____________________________________________________________ 344

ZGHEIB, CAPITANO & BRANCH

Introduction

I

n recent years, systemic mycoses have become a prominent cause of disease particularly in severely ill and immunocompromised patients. The factors contributing to the increased prevalence of fungal infections are related to larger number of patients with underlying immunosuppression, for example the acquired immunodeficiency syndrome (AIDS), more aggressive cancer chemotherapy, increase in transplantation, greater number of other immunocompromised patients, and more frequent use of prosthetic devices [1]. There have been a number of recent surveys, which illustrate the extent of this problem. The Center for Disease Control reported that among 51 USA hospitals, candidiasis was the eighth most common infection, accounting for 5% of the isolates [1, 2]. This value can be considerably higher in certain specific patient groups. The National Cancer Institute estimated that 43% of patients dying with acute leukemia had systemic fungal infection at autopsy [3]. In patients with AIDS, the most common fungal infection is oropharyngeal candidiasis. However, in these patients, the fungal infection with the highest mortality rate is cryptococcosis. It is evident that systemic fungal infection is an important consideration in the treatment of a severely ill, immunosuppressed patients [4]. Amphotericin B (AmB) has remained a mainstay of therapy for serious fungal infections since its introduction in 1956, owing to its broad spectrum of reliable activity and lack of availability of equally efficacious alternative agents [5]. The usefulness of this agent, however, is limited by the frequent occurrence of several acute and chronic adverse effects that often necessitate changes in, or premature discontinuation of, therapy. These include fever, chills, nausea, vomiting, anorexia, headache, bronchospasm, hypotension, anaphylaxis, and bone marrow suppression. The most limiting adverse effect, however, is nephrotoxicity [611]. Several novel antifungal agents, found to be equally efficacious and less toxic as compared to AmB in clinical trials, have been introduced over the past several years. Thus, the role of AmB as the “gold standard” in the treatment of serious fungal infections is likely to be challenged and re-defined in the next decade [12]. AmB is a member of the polyene macrolide class of antibiotics. The molecule consists of a large macrolide lactone ring of 37 carbon atoms, one side of which is 324

composed of a rigid lipophilic chain of seven conjugated double bonds, and the opposite side of a similar number of hydrophilic hydroxylated carbon atoms (Figure 1). Thus, the molecule is amphipathic, and this feature of its structure is believed to be important in its mechanism of action [13]. The major action of AmB is believed to be on the cell membrane of fungal and mammalian cells. It is generally accepted that the drug binds to sterols in the cell membrane and induces formation of aqueous pores, which result in impairment of barrier function and loss of protons and cations from the cell. At low concentrations, the increased permeability is restricted to small molecules or cations such as sodium and potassium. At higher concentrations or after prolonged exposure, other cell constituents are lost and this leads to metabolic disruption and even cell death [13]. The cellular events that follow this membrane effect are complex and depend on a variety of factors, such as the growth phase of the cells, the dose, and the mode of AmB administration [14]. Some studies suggest that cell mortality is not simply a consequence of changes in permeability of membranes, and that formation of active oxygen species may play a role in the lytic or lethal actions of AmB [15, 16].

Clinical manifestations of nephrotoxicity The most restrictive adverse effect associated with AmB therapy is its potential to induce nephrotoxicity, manifested as disturbances in both glomerular and tubular function. The clinical manifestations usually include azotemia, renal tubular acidosis, decreased concentrating ability of the kidney, and electrolyte disturbances such as urinary potassium wasting leading to hypokalemia, and magnesium wasting to result in hypomagnesemia [17].

C O N H

Figure 1. Chemical structure of amphotericin B.

14. Amphotericin B

Incidence and risk factors

100

% with normal renal function

with salt

The incidence of AmB-induced renal impairment is highly variable depending on the definition of nephrotoxicity and upon the presence of underlying risk factors. Following AmB introduction, a survey of 56 patients treated between 1956 and 1963 confirmed that 93% of patients developed values of BUN exceeding 200 mg/L, and 83% had serum creatinine levels greater than 15 mg/L [18]. A more recent report indicates that in almost every patient treated with AmB, the glomerular filtration rate (GFR) falls approximately 40% within the first 2 to 3 weeks of therapy, stabilizes at 20% to 60% of normal and remains at this level throughout the course of treatment [19]. Clements and Peacock [8] reported an incidence for azotemia of 60% in a retrospective analysis conducted between 1984 and 1987. In general, the incidence of azotemia due to AmB in the literature ranges between 50-90%. This variability may reflect various factors, among which are the definition used for nephrotoxicity, the dose of AmB, concomitant administration of other nephrotoxic agents, and the presence or absence of other proposed risk factors. In general, azotemia is transient and limited to the duration of therapy; renal function usually returns to pretreatment levels after discontinuation of the drug. In many cases, cessation of therapy for a few days allows renal function to recover enough to permit administration of the full course of therapy. In rare cases, however, permanent renal damage persists after cessation of therapy. The relationship of the cumulative dose of AmB to the development of nephrotoxicity is controversial. Earlier studies suggested that greater cumulative doses of AmB (e.g. 3-4 g) were associated with a greater risk of nephrotoxicity [20]. This implies that the likelihood of a rise in the serum creatinine concentration increases in proportion to the length of therapy. However, we observed patients who developed azotemia at doses ranging from 100 mg to 1.5 g, with no significant increase in frequency as the cumulative dose increased. Our experience indicated that the frequency of nephrotoxicity did not increase with extended therapy over this dose range (Figure 2) [21]. With larger cumulative doses, renal impairment may be irreversible, as reported by Winn [22] who found persistent renal impairment in 88% of patients who had received cumulative doses exceeding 5 g.

80

60 without salt 40

20

0

0

7

14 Time

21 28 (days)

35

42

Figure 2. Estimated proportion of patients retaining normal renal function during therapy with amphotericin B. Patients received amphotericin B with (dotted line, n=17) or without (solid line, n=21) parenteral salt supplementation due to coadministration of ticarcillin. (Used with permission from [21])

The clinical utility of a risk factor such as cumulative dose to identify patients at risk of nephrotoxicity is limited since knowledge of this parameter is unavailable prior to the commencement of treatment. The rate of infusion of AmB and the frequency of dosing have also been found to impact the likelihood of nephrotoxicity in that use continuous infusion or administration of drug on alternate days reduces the incidence of nephrotoxicity [23, 24]. Several clinical assessments have identified additional characteristics that may be used to identify patients at increased risk for AmB nephrotoxicity such as abnormal baseline renal function, dehydration, older age [10, 20], diuretic use, pre-existing atherosclerosis, diabetes and heart failure [25]. In a multivariate risk factor assessment, Luber et al. [9] demonstrated that the AmB nephrotoxicity incidence varied with the definition used for renal impairment. Among 178 patients a change in creatinine of >46 μmol/L over baseline occurred in 50%; a doubling of creatinine over baseline in 49%; a change in creatinine of >92 μmol/L in 29%; a doubling and/or a change in creatinine of >92 μmol/L in 49%; and an increase in creatinine to >230 μmol/L in 8%. In this study, nephrotoxicity was associated with a greater cumulative dose of AmB and concomitant nephrotoxic drugs for all definitions. In those patients who experienced severe nephrotoxicity (creatinine increased to >230 μmol/L), 325

ZGHEIB, CAPITANO & BRANCH

concomitant cyclosporine therapy was the most significant risk factor (odds ratio 18.8, P= 0.022). Harbarth and colleagues conducted a 9 year retrospective epidemiological assessment of AmB nephrotoxicity in 494 patients who received ≥ 2 doses in attempt to clarify discrepancies in the literature regarding the incidence and influence of specific risk factors, including AmB dose and duration. The results yielded 5 categorical risk factors for nephrotoxicity including an average daily dose of ≥ 35 mg, male sex, weight ≥ 90 kg, chronic renal disease and use of amikacin or cyclosporine. The incidence of moderate to severe nephrotoxicity, defined as doubling of the serum creatinine level to ≥ 2.0 mg/dl, was 4% (6/137) in patients with none of these risk factors, 8% (14/181) in those with 1 risk factor and 29% (21/117) in those with ≥ 3 risk factors. Based on these data, the authors proposed that patients identified with ≥ 2 of identified risk factors or a “risk score” of ≥2 to be candidates for alternative therapy. A limitation of this study was lack of information as to whether patients received hydration or sodium load prior to AmB administration, a method known to reduce nephrotoxicity [11]. Girmania and colleagues tested the utility of Harbath’s risk score in 46 consecutive patients who received AmB at 1 mg/kg over 2 hours with hydration of at least 1 L/m2 containing at least 1 L of 0.9% saline solution. The rate of moderate to severe nephrotoxicity was 0% (0/12) in patients with 1 risk factor, 6.3% in patients with 2 (1/16) and 11.1% (2/18) in patients with 3 risk factors. The overall rate of AmB induced nephrotoxicity in this study was lower than that observed in the previous study (13% vs. 28%, respectively). Although the population was small, the results found the risk scores to be predictive of patients at risk of moderate to severe nephrotoxicity. The authors concluded that adequate hydration and sodium loading should be employed prior to AmB administration to reduce the incidence of nephrotoxicity [26]. Consistent with previous data, a retrospective analysis of a homogenous population of 69 bone marrow or peripheral blood stem cell transplant recipients with multiple myeloma revealed baseline estimated creatinine clearance, use of cyclosporine or receipt of multiple nephrotoxic agents within 30 days of starting AmB to be significantly predictive of nephrotoxicity in this patient population [27]. Wasan and colleagues have conducted a series of studies evaluating the association of AmB with 326

plasma lipoproteins demonstrating that less damage to renal cells is evoked when the drug associates with high density lipoproteins (HDL) as compared to low density proteins (LDL). Their work continues in order to determine whether measurement of plasma HDL and LDL concentrations may be used to identify patients more likely to experience AmB nephrotoxicity. Although very interesting, at this time, no clinically applicable conclusions may be extrapolated from the data [28-31].

Urinary concentration defects Many studies have shown that AmB can induce a loss of concentrating ability of the kidney [18, 32, 33]. This abnormality is almost invariably present and occurs early (1-2 weeks) in the course of therapy. The impairment in concentrating ability probably reflects direct tubular toxicity since it occurs in the absence of a decrease in GFR, and is temporally unrelated to azotemia. Barbour et al. [34] reported a study of 3 patients whose inability to concentrate the urine was associated with a defect in free water reabsorption even under maximal stimuli, and concluded that a tubular functional abnormality was induced because of the failure of the vasopressin response in the medullary collecting tubule.

Electrolyte disturbances Electrolyte disorders secondary to renal wasting of potassium and magnesium are commonly encountered adverse effects in patients receiving AmB [17, 35]. Although hypokalemia has been emphasized in prior studies, its impact on patient management and on the course of other manifestation of AmB nephrotoxicity has not been well examined. In addition to its known systemic effects (muscle weakness, fatigue, cramps, rhabdomyolysis and myoglobinuria), potassium depletion may alter renal function causing further impairment of concentrating ability, urinary acidification, renal insufficiency and abnormal sodium reabsorption [36]. It is conceivable that these effects may influence or contribute to the nephrotoxicity of AmB. Approximately 75% of patients develop hypokalemia during the course of treatment with AmB [37] However, a need for potassium supplementation to maintain a normal plasma level of potassium can be

14. Amphotericin B

regarded as an objective parameter of a potassium losing diathesis. Using this criterion, the incidence is as high as 90% or more [8, 38]. The maintenance of normokalemia requires up to 300 mEq of potassium chloride replacement a day. These patients are often severely ill and unable to tolerate oral supplementation, so prolonged (6-7 hours) of administration of large intravenous doses of potassium chloride with appropriate and careful monitoring may be necessary. The logistics of such continuous intravenous maintenance infusions can create problems in timely administration [8]. Some investigators consider hypokalemia a dosedependent response, although the mechanism of urinary potassium wasting is unclear. A recent study has shown that AmB affects sodium flux in both the distal and transverse human colon, suggesting a change in sodium/potassium exchange to result in potassium loss [39]. Selective renal distal tubular epithelial toxicity seems to be, at least in part, responsible for the profound potassium wasting. The magnitude of urinary potassium loss increases in the presence of a high sodium chloride intake. If potassium depletion is allowed to occur with AmB, a vicious circle is created which further enhances the tubular toxicity and contributes to overall changes in renal function [40]. Magnesium wasting has also been reported as a consequence of AmB administration [41, 42]. A negative magnesium balance probably occurs in all patients, but clinically relevant magnesium depletion only occurs when the urinary loss is high and not replaced. In the study by Barton et al [41] the lowest serum level and the largest fractional excretion of magnesium were observed by the fourth week of AmB therapy, after a cumulative dose of approximately 500 mg. This abnormality was fully reversible, evidenced by the normal serum and urinary magnesium levels measured 1 year after discontinuation of therapy. As in the case of potassium depletion, if magnesium depletion is evaluated by measurement of magnesium balance rather than by the serum level, the incidence of magnesium depletion is high. In a recent study, a marked change in the urinary excretion of magnesium occurred after a cumulative AmB dose of only 150 mg, suggesting some degree of magnesium depletion, although serum magnesium levels remained in the low normal range [38]. The mechanisms for the observed AmB induced renal magnesium wasting remain unclear. Increased

urinary excretion of magnesium, despite a reduced filtered load, suggests a tubular defect in magnesium reabsorption [41]. When magnesium and potassium wasting occur concomitantly, potassium replacement may not be successful unless magnesium deficiency is corrected first.

Renal tubular acidosis Chronic features of renal tubular acidosis can be anticipated in patients receiving total doses of AmB of 0.5-1 g or more, and are generally reversible after therapy is discontinued [5]. In our experience this is one of the earlier manifestations of tubular toxicity, since all patients developed an acidification defect in response to an acid load after 2 weeks of therapy and a cumulative dose of 300 mg of AmB [38]. This defect appears to be a specific tubular effect of AmB, since impairment in acid secretion has been demonstrated in the isolated turtle bladder and attributed to increased passive permeability of the luminal membrane to hydrogen ions [43, 44], plus the impaired excretion of titratable acids is greater than can be accounted for by depression of GFR [45, 46]. It is also thought that distal renal tubular acidosis is a contributing pathogenic mechanism for urinary losses of potassium and magnesium [37, 45, 46].

Pathological findings Despite the almost universal changes in renal function, histological changes associated with AmB therapy are minimal and occur in both glomerulus and renal tubule. Tubular damage primarily involves the distal convoluted tubule and the ascending limb of the loop of Henle [41]. Morphologic changes include fragmentation and thickening of basement membranes, necrosis and vacuolization of distal tubular epithelium and nephrocalcinosis [18, 32, 46, 47]. Glomerular changes include calcific foci, along with hypercellularity and vacuolization of smooth-muscle cells in small arteries and arterioles [33, 46]. In studies conducted in rats, cortical changes associated with AmB were restricted to the medullary ray, an area that is vulnerable to hypoxia, and consisted of focal rupture and calcification of the thick ascending limbs [48]. Calcification was also detected in the macula densa, an area rich in oxygen. Administration of AmB to salt depleted rats 327

ZGHEIB, CAPITANO & BRANCH

Figure 3. Distal tubulus of rat kidney after 14 days of AmB administration. Electron microscopy shows intratubular casts and debris, loss of brush border, tubular cell vacuolization, and protrusion of cells into the tubulus lumen. Magnification 3600x. With permission from [49].

resulted in an extension of these changes to the area rich in vascular tissue between the medullary rays and to atrophic changes in the thick ascending limb in the inner stripe [49] (Figure 3).

Mechanisms of nephrotoxicity Before mechanisms can be proposed to account for renal cell injury, the possible sites of nephron involvement should be identified based upon structural and functional changes [50]. AmB is known to cause acute renal vasoconstriction and to preferentially damage the distal tubular epithelium, but the exact mechanisms mediating its nephrotoxicity have not been clearly defined. The initial event is thought to involve binding of AmB to membrane sterols in the renal vasculature and epithelial cells causing an alteration in membrane permeability. This interaction may then trigger other cellular events that result in activation of second messenger systems, release of mediators or activation of renal homeostatic mechanisms. It is, therefore, possible that the membrane effect per se is not the sole factor that determines the extent of change in renal function.

328

Effects on cell membranes It is generally accepted that AmB-induced injury to cells is due to its binding to sterols in the cell membrane: ergosterol in the case of fungal cells and cholesterol in mammalian cells [14]. This binding is more avid to ergosterol than to cholesterol, which explains AmB’s relatively selective toxicity to fungal cells [51, 52]. In the early 1960s, studies showed that polyene antibiotics induced changes in cellular permeability that resulted in the leakage of important cellular constituents, followed by lysis and death [53-57] it was also discovered that the toxic effect of the drugs on cells was dependent on the presence of sterols in the cell membranes, and that addition of sterols to the growth media of certain fungi prevented the polyene-induced inhibition of growth and permeability changes [53, 58, 59]. This increased permeability has been documented in both artificial and natural membranes [60]. It has been proposed that AmB, acting as a pseudophospholipid, interacts with sterol molecules to cause formation of aqueous pores, which consist of an annulus of polyene and sterol, in which the hydrophilic region of the drug molecule faces the interior of the pore (Figure 4) [6062]. Among the documented effects of AmB on living tissues are increased permeability of the toad urinary

14. Amphotericin B

LIPID

WATER

phospholipid

CHOLESTEROL

C20-C33 heptaene segment

AMPHOTERICIN B

PORE

AMPHOTERICIN B C20-C33 heptaene segment

CHOLESTEROL C O N H

phospholipid

LIPID

Figure 4. Proposed partial model for the AmB-induced pore in the cell membrane. The drug acts as a counterfeit phospholipid, with the C15 hydroxyl, C16 carboxyl, and C19 mycosamine groups situated at the membrane-water interface, and the C1 to C14 and C20 to C33 chains aligned in parallel within the membrane. The heptaene chain will seek a hydrophobic environment while the hydroxyl groups will seek a hydrophilic environment. Thus, a cylindrical pore will be formed, the inner wall of which consists of the hydroxyl-substituted carbon chains of the AmB molecules, and outer wall of which is formed by the heptaene chains of the molecules and by sterol nuclei. (Used with permission from [60]).

bladder to urea, potassium and chloride ions [63-65], of erythrocytes and liposomes to potassium ions [66, 67], and of erythrocytes to sodium and chloride ions [68, 69]. It also alters the permeability of the turtle bladder and of purified renal brush border membrane vesicles to sodium and hydrogen ions [43, 70-72]. Considering the renal tubular effects of AmB observed in clinical practice, it is reasonable to suggest that part or all of these effects may be explained by a

direct effect on tubular cell membranes. In support of this suggestion is the in vivo finding that while AmB binds to sterols in most tissues, the highest level documented is in the kidney [73]. Furthermore, binding of AmB to the cell membrane appears to be necessary for its toxic effect, since inhibition of sterol synthesis by ketoconazole reduces the binding of AmB as well as the permeability changes induced by AmB in a parallel fashion [71, 74]. In agreement with these suggestions is the finding of increased tubular permeability to inulin in vivo in rats following acute or chronic administration of AmB, resulting in back-leak of inulin [75]. Further evidence to support a direct toxic effect of AmB on renal cells is the demonstration of increased apoptosis in proximal tubular and medullary interstitial cell lines [76]. The occurrence of apoptosis has also been confirmed in an in-vivo model in rats in which AmB administration also caused hypokalemia, loss of concentrating ability and dehydration. Interestingly, prevention of apoptosis by recombinant human insulin growth factor-1 (rhIGF-1) ameliorated the tubular toxicity indicating the importance of apoptosis in AmB-induced renal tubular toxicity process. A possible mechanism for this action is suggested by a recent study that has demonstrated that AmB exposure increases cellular ceramide as well as sphingomyelin levels in proximal tubular cells [77]. It is noteworthy that ceramide has been postulated to play a role in inducing apoptosis in several cell types [78]. Although the role of these changes in nephrotoxicity is still uncertain, these findings suggest that the interaction of AmB with cell membranes is not limited to a physicochemical interaction with sterols leading to pore formation and changes in permeability, but may also involve other complex effects that lead to alteration in production or function of membrane associated signaling molecules. An alternative postulated mechanism of AmB induced cell damage involves oxidative stress with the formation of free radical intermediates [15, 16, 79]. Evidence against this hypothesis has been provided by recent studies that evaluated the anti- or pro-oxidant effects of AmB by examining its effects on phospholipid pattern in aortic smooth muscle cells [80] as well as on lipid-peroxidation of cis-Parinaric acid in liposomes [81]. These studies provided evidence for an antioxidant role for AmB rather than a pro-oxidant role and suggesting that oxidative stress is not involved in AmB-induced toxicity. 329

ZGHEIB, CAPITANO & BRANCH

In addition, alternative factors may modulate the direct cellular toxicity of AmB. For example, maintaining kidney epithelial cells in an acidic environment enhances the permeability changes induced by AmB in an irreversible fashion [82]. This suggests that the low pH characteristic of the distal tubule makes those cells more vulnerable to the toxic effects of AmB, and may explain the protective effect of alkalinizing agents [45].

Effects on physiological parameters: whole animal studies Acute studies (Single dose) Infusions of AmB, intravenously or into the renal artery, induce short-term reduction in renal blood flow (RBF) and GFR, and an increase in renal vascular resistance, in both rats and dogs [83-85]. The effects of short term infusions of AmB on the renal microcirculation in rats revealed that the single nephron GFR was decreased by 2 mechanisms (Table 1): 1) a decrease in single nephron plasma flow, due to vasoconstriction of the afferent and efferent arterioles, and 2) a reduction in the glomerular capillary ultrafiltration coefficient (Kf), an effect probably mediated by mesangial cell contraction [86]. Previous micropuncture studies demonstrated a similar vasoconstriction of the afferent arteriole but also an increased permeability of the tubular epithelium to inulin [75]. Thus, the reduction in GFR after acute AmB infusions can be attributed to contraction of afferent smooth muscle cells, efferent smooth muscle cells and glomerular mesangial cells, as well as increased tubular permeability with back-leak

into the interstitial space. The mechanisms responsible for the contractile responses to AmB have not been identified. Theoretically, the drug can act either directly on the vascular smooth muscle or through release of secondary mediators. A large number of studies have examined putative indirect mechanisms of action. Those studies have revealed that neither renal denervation nor angiotensin II receptor blockade prevent the renal vasoconstriction or the reduction in GFR [87, 88]. Although Cutaia et al [89] demonstrated a toxic effect of AmB on endothelial cells, endothelin does not appear to be involved in the acute responses to AmB [88, 90] and reduced nitric oxide synthesis, consequent to endothelial injury is not involved in modulation of AmB-induced renal vasoconstriction [88]. It has also been suggested that activation of tubuloglomerular feedback (TGF) may play a role in the acute renal effects of this compound. That hypothesis suggested that the tubular toxicity of AmB resulted in impaired reabsorption of sodium and chloride ions by the proximal tubule, which increased distal tubular delivery of these ions, thus activating TGF [91]. Indirect evidence in support of a role for TGF was derived from studies which demonstrated inhibition of the acute renal effects of AmB by physiological and pharmacological interventions that also blocked TGF, namely, salt loading, and administration of furosemide, theophylline or calcium channel blockers [84-86, 92-99]. Finally, some studies suggested a protective effect of pentoxiphylline, a vascular decongestant and antagonist of tumor necrosis factor-alpha (TNF) and interleukin-1alpha, against AmB-induced acute and

Table 1. Effect of amphotericin B infusions (0.05 mg/kg/min i.a.) on systemic glomerular hemodynamic parameters.

Mean arterial pressure (mmHg)

Before

After

p-value

114 ± 5

117 ± 4

NS

Renal plasma flow (ml/min)

4.69 ± 0.35

2.82 ± 049

< 0.025

Glomerular filtration rate (ml/min)

1.03 ± 0.65

0.70 ± 0.09

< 0.005

Single nephron glomerular filtration rate (ml/min)

35.5 ± 2.2

22.8 ± 2.8

< 0.0005

Plasma flow (ml/min)

142 ± 12

89 ± 14

< 0.005

Single nephron filtration fraction

0.26 ± 0.03

0.27 ± 0.04

NS

dyn.sec.cm-5)

1.91 ± 0.17

3.95 ± 0.38

< 0.01

Efferent arteriolar resistance (1010 dyn.sec.cm-5)

1.30 ± 0.10

2.08 ± 0.12

< 0.01

0.043 ± 0.008

0.032 ± 0.009

< 0.005

Afferent arteriolar resistance

(1010

Glomerular capillary ultrafiltration coefficient [n/(sec.mmHg)]

330

14. Amphotericin B

Chronic studies (Multidose) Animal models of chronic nephrotoxicity have

300

A

(6) (11) (7)

200

(7)

100 (5)

Rise in intracellular calcium (nM)

chronic nephrotoxicity, suggesting a role for these factors in the renal effects of the drug [100, 101]. More recent studies provided evidence against a role for TGF in acute AmB nephrotoxicity. In contrast to its inhibition of TGF activity, theophylline prevented the acute renal responses to AmB by a mechanism unrelated to adenosine receptor antagonism [102]. Furthermore, micropuncture studies revealed that the AmB-induced reduction in single nephron GFR was the same irrespective of whether TGF was active or interrupted (by measuring GFR from distal and proximal tubular collections, respectively) [96]. The latter study also showed that distal tubular chloride ion concentrations were not increased by AmB, which indicated that the signal for TGF was unchanged. A direct effect of AmB on cell function was suggested by in vitro experiments, which demonstrated a vasoconstrictor action of AmB in perfused afferent arterioles and isometrically contracting rings of rabbit aorta or renal artery, effects which were prevented in Ca++-free medium and by Ca++ channel blockers or theophylline [103]. Thus, AmB-induced reductions in RBF or GFR are not secondary to activation of TGF, but due to its direct vasoconstrictor effect. A role for thromboxane A2 has also been suggested based upon partial inhibition of the AmB-induced vasoconstriction and reduction in GFR by pretreatment with ibuprofen or a thromboxane receptor antagonist [104]. The results obtained in isolated vessels are consistent with findings in cultured glomerular mesangial cells where AmB caused a concentration-dependent increase in intracellular calcium levels ([Ca++]i), an effect almost completely inhibited when either Ca++ or Na+ ions were omitted from the cell medium (Figure 5) [105]. Diltiazem (20 μM) also suppressed the AmBinduced rise in [Ca2+]i. These results indicated that the reduction in Kf observed in vivo was most likely due to contraction of mesangial cells. Thus, the contractile effects of AmB in the nephron are probably due to a direct interaction with cell membranes, leading to formation of pores. One possibility is that these pores allow entry of Na+ ions into the cells along the electrochemical gradient leading to depolarization-induced opening of voltage-dependent calcium channels and contraction.

0

300

0.05

(6)

0.1 0.5 1 Amphotericin B conc. (µM)

5

B

200

100

*

*

*

Dilt

-Na

0 Control

-Ca Group

Figure 5. Concentration dependent increase in intracellular calcium levels in cultured glomerular mesangial cells (A) and its inhibition by removal of Ca++ and Na+ ions from the medium (-Ca and -Na, respectively), and by addition of 20 μM diltiazem (Dilt) (B). (Used with permission from [105]).

also shown that certain interventions can modify the nephrotoxicity of AmB. Rats co-treated with sodium bicarbonate sustain smaller reductions in GFR compared with rats treated with AmB alone for 3 weeks [45]. Oral NaCl supplementation also attenuates the decrease in GFR and the elevation in renovascular resistance induced by daily administration of AmB over 3 weeks [96, 98]. In addition, renal impairment following a 7-day course of AmB in rats was less severe when theophylline was co-administered [99]. These interventions also attenuate the acute renal responses to AmB, suggesting that similar mechanisms contrib331

ZGHEIB, CAPITANO & BRANCH

ute to its chronic nephrotoxicity. Similar logic would suggest that salt supplementation and theophylline are protecting the kidney by a mechanism unrelated to TGF. It is, however, possible that the latter does contribute to AmB nephrotoxicity but only at later stages of therapy, when severe damage to the tubules may have taken place. Interestingly, the protection by salt loading is associated with lower concentrations of drug in the kidney despite similar concentrations in plasma and liver tissue [96]. This raises an alternative possible mechanism of protection by salt loading involving a pharmacokinetic interaction with AmB, which limits its uptake into the kidney. Studies on the effects of chronic administration of calcium channel blockers on chronic AmB induced nephrotoxicity have been discordant. Nifedipine does not offer a significant protective effect [106], but diltiazem blunts the increase in serum creatinine and the decreases in GFR and RPF [97]. It is possible that these differences relate to the heterogeneity of calcium channels and the differential activity of calcium channel blockers on them. However, no specific studies have addressed this question. The co-administration of 5-flucytosine with AmB, which is commonly used clinically to obtain a synergistic antifungal effect, protects against acute and chronic nephrotoxicity [107]. The mechanisms by which flucytosine influences the renal response to AmB are not clear but may relate to (i) its administration in 0.9% NaCl, which itself is protective, (ii) a renal vasodilator effect of flucytosine that antagonizes AmB-induced vasoconstriction, and (iii) reduction in renal uptake of AmB [107]. Cell death induced by AmB in the medullary thick ascending limb is prevented by ouabain [108]. A reasonable explanation for this observation is that ouabain, by inhibiting transport, decreases the oxygen demand of an area of the nephron that already has a limited oxygen supply. This is consistent with the observation that AmB exhibits preferential damage to the medullary ray, an area that is vulnerable to hypoxic injury [48]. It is also conceivable that AmB-induced renal vasoconstriction and ischemia to this section of the nephron enhances cell death produced by a direct toxic action. Thus, any maneuver that improves renal perfusion, or decreases oxygen demand, would be expected to be protective. This may explain the salutary effect of salt loading, theophylline, calcium channel 332

blockers, pentoxiphylline, dopamine or dopamine pro-drugs such as fenoldopam on AmB nephrotoxicity [102-104]. All these interventions can be expected to improve renal perfusion. Furosemide protection could be explained on a basis that it not only inhibits solute transport in the thick ascending limb to reduce oxygen demand, but also enhances renal perfusion to increase oxygen supply.

Measures to reduce nephrotoxicity Despite being considered one of the most toxic antimicrobial drugs in use today [109], AmB remains a primary choice for the treatment of otherwise uniformly fatal systemic fungal infections [4, 5]. Consequently, it will remain in use despite the predictable occurrence of severe systemic and renal toxicity. Therefore, therapeutic interventions that decrease AmB toxicity assume critical importance. Among the early interventions that were examined was the administration of mannitol. Studies suggested a protective effect of mannitol in dogs and renal transplant recipients [110112]. Unfortunately, these were either case-reports or poorly controlled since later reports failed to detect any protective effect of mannitol in dogs, and ascribed the protective findings to the lower doses of AmB used [113]. In addition, a small controlled trial of mannitol co-administration in humans failed to document any beneficial effect [33]. A more recent intervention is the use of spironolactone which has been suggested as adjuvant therapy with AmB, a combination that prevented hypokalemia in neutropenic patients, but this has not been pursued [114]. Theoretically, any of the protective interventions mentioned in the previous section may be applicable to a clinical setting, but few have actually been studied, in some instances, because there are practical limitations to their use. For example, the duration of protection conferred by furosemide is brief, being confined to the time furosemide is present in the renal tubule. Furosemide would exacerbate electrolyte imbalance by causing sodium and potassium depletion, which, if not adequately monitored and replaced, would be expected to potentiate AmB-induced nephrotoxicity. Furthermore, none of the advocated drug interventions are innocuous. Of all the alternatives, manipulation of sodium status or of the method of administration offer simple interventions that can be readily and usually

14. Amphotericin B

safely be introduced into clinical practice [91]. The use of continuous infusion or administration of drug on alternate days has been found to reduce the incidence of nephrotoxicity [23, 24, 115]. An alternative approach is to use a lipid formulation of amphotericin B (LFAB) or other antifungal agents.

Salt supplementation The demonstration of a renal protective effect of salt loading on AmB-induced nephrotoxicity in animal models has provided a rational basis to evaluate this simple intervention in patients. Clinical evidence supporting the ability of sodium loading to attenuate AmB-induced nephrotoxicity is derived from three sources: case reports, retrospective studies and prospective studies. One of the earliest case reports was by Butler and colleagues who reported a patient in whom a low sodium diet (9 mEq/d) exacerbated renal dysfunction, increased urinary sodium loss, and caused postural hypotension [18]. Administration of supplemental oral sodium chloride promptly reversed the defect within 12 hours. These abnormalities were confirmed on rechallenge during treatment, but were absent 13 months after completion of AmB therapy. In a subsequent study, 5 patients were receiving AmB in clinical situations where salt-conserving states could be identified. These included dietary salt restriction, vomiting, diuretic therapy, Addison’s disease and cirrhosis with ascites. In each patient sustained increases in BUN and serum creatinine levels were observed within 6 to 12 days after starting AmB [116]. Four to 12 days after liberalization of dietary sodium intake, administering intravenous saline, and/or discontinuation of diuretic therapy, renal function improved in all patients. Improvement was sustained and the full course of AmB was successfully completed after a brief interruption (range: 1 to 5 days), without permanent renal impairment. A retrospective study revealed that only two of 17 patients (12%) receiving ticarcillin (with its obligatory sodium load of 150 mEq/day) had a nephrotoxic response to AmB, compared with 14 of 21 patients (67%) not receiving ticarcillin (Figure 2) [21]. Anecdotally, withdrawal of ticarcillin in patients continuing to receive AmB led to deterioration of renal function over a one-week period. In a companion study, the benefit of routine intravenous saline (1 L of 0.9%

saline) was assessed prospectively in leukemic patients receiving a 28-day course of AmB for persistent fever of unknown origin. Only two of 20 patients (10%) developed mild renal dysfunction, which, however, did not necessitate interruption of therapy. Furthermore, a recent medical record review for 573 extremely low birth weight infants born at Michigan State University revealed that combining conventional AmB with adequate hydration and higher amounts of sodium (>4 mEq/kg/day) is renoprotective [117]. The question of the influence of salt supplementation was addressed using a prospective, randomized, placebo controlled trials of the influence of salt supplementation on the course of renal function during therapy with AmB [38, 118]. In Llanos et al’s study, AmB administration was preceded by 1 liter of either 0.9% saline or 5% dextrose in water administered i.v. over 4 hours. While the mean serum creatinine increased over time in the dextrose group, it remained unchanged in the saline group. Similarly, creatinine clearance decreased in the dextrose group, but remained unchanged in the saline group. The beneficial effect of salt loading however occurred at the expense of greater hypokalemia, since the saline group required significantly higher amounts of potassium supplementation to maintain a normal serum K level (Figure 6). Therefore, based upon these studies, it is reasonable to recommend routine salt supplementation with administration of AmB, with special attention being paid to maintaining potassium balance [38]. More recently, Mayer and coworkers [119] have expanded the notion that early introduction of potassium and magnesium supplements in equal amounts to that lost through the kidneys results in decreased incidence of AmB infusion-related side effects and decreased frequency of an increase in serum creatinine. This study provided further evidence to support that sufficient hydration and timely supplementation of sodium, potassium and magnesium reduces the renal toxicity of AmB. This has also been supported by further recent evidence from Oto et al in patients with febrile neutropenia and fungal infections [120]. Interestingly, a prospective trial has shown that oral rehydration solution (ORS) is more effective in preventing hypokalemia when compared to intravenous saline solution (SS) [121].

333

ZGHEIB, CAPITANO & BRANCH

A 5

*

Serum K (mmol/L)

4 3

* *

* *

* *

*

*

*

*

*

*

*

*

2

*

*

*

*

NaCl

1

Dextrose

0 0

100 250 400 550 700 850 1000 1150 1300 1450

B

K supplement (mmol/day)

120

*

NaCl

100

Dextrose

80

*

*

*

* *

60 40 20 0 0

100 250 400 550 700 850 1000 1150 1300 1450 Amphotericin B cumulative dose (mg)

Figure 6. Serum K levels (A) and K supplements given to maintain serum K levels at or above 3 mmol/l (B) in patients receiving AmB with either 1 liter of 0.9% NaCl (solid line) or 1 liter of 5% dextrose in water (dotted line). Notice the difficulty in maintaining serum K levels despite significantly higher amounts of supplements in the former group. *: P7000 pills

2.8

0.7

ASPIRIN

Sandler Pommer Morlans Perneger Fored Ibanez van der Woude Kurth Curhan

PARACETAMOL

25.9

2.2

1

Sandler Pommer Morlans Perneger Fored Ibanez van der Woude Kurth Curhan

0.1

23.1 6.10

1.4 1.26

0.1

C

34.7

2.79

1

10

100

10

100

Odd’s Ratio (95% CI)

Figure 1. Overview of epidemiological studies investigating the renal risk of analgesic consumption. A. Description of methodological details used in the included studies. B. Presentation of the overall risk (odds ratio with 95% confidence interval) associated to the consumption of ‘any analgesic’ exceeding the mentioned dose. C. Presentation of the odds ratios with 95% confidence interval published in the included epidemiological studies focussing separately on the ingredients aspirin and paracetamol. 401

ELSEVIERS & DE BROE

Table 1. Sources of bias in the epidemiological studies. Selection bias

Information bias

Indication or protopathic bias

Ingredient bias

Dose bias

McCredie et al, Australia, 1982 [33]

yes

no

yes

yes

no

Murray et al, USA, 1983 [34]

yes

no

yes

yes

yes

Sandler et al, USA, 1989+1991 [16,17]

no

yes

yes

yes

no

Pommer et al, West Berlin, 1989 [36]

no

no

yes

no

no

Morlans et al, Barcelona,1990 [18]

yes

no

no

yes

yes

Perneger et al, USA,1994 [20]

yes

yes

yes

yes

no

Fored et al, Sweden, 2001 [21]

no

no

no

yes

yes

Ibanez et al, Barcelona, 2005 [37]

yes

no

no

yes

yes

Van der Woude et al, Austria, Germany, 2007 [38]

yes

no

no

no

yes

Dubach et al, Switzerland, 1983 [39]

no

yes

yes

yes

no

Elseviers and De Broe, Belgium, 1995 [40]

no

no

yes

no

no

Kurth et al, USA, 2004 [41]

no

yes

yes

yes

no

Curhan et al, USA, 2004 [42]

no

yes

yes

yes

no

Case-control studies

Prospective controlled cohort studies

Observational cohort studies

Selection bias = random selection of controls failed or the chosen control population is biased. Information bias = methods used to obtain information about analgesic consumption were doubtful. Indication (protopathic) bias = failure to control for analgesic intake preceding the development of renal failure. Ingredient bias = failure to entangle the use of particular ingredients either as single analgesic or as one of the ingredients of analgesic mixtures. Dose bias = definition of analgesic use far below the amount consumed by patients with analgesic nephropathy.

remarkably similar. The included observational cohort studies aimed primarily to investigate the health status of a large cohort of US male physicians and US female nurses during a follow-up period of 14 and 11 years respectively [41, 42]. In a retrospective analysis, subjects who developed renal failure were compared with controls without renal failure with regard to analgesic consumption. Nephrotoxicity of different kinds of analgesic mixtures In the majority of the early analgesic nephropathy reports, phenacetin was singled out as the nephrotoxic culprit on the basis of association and circumstantial evidence. Nearly all patients initially reported had taken large amounts of analgesic mixtures containing phenacetin. Prescott [14], was the first to evaluate the nephrotoxic role of phenacetin and other analgesics. He stated that in the past insufficient attention had been given to the possible nephrotoxicity of the other analgesics invariably taken with phenacetin, and that the common belief that phenacetin is the primary cause of analgesic nephropathy can be challenged on many counts. He argued that numerous chronic toxicity stud402

ies in animals with phenacetin have failed to produce renal papillary necrosis, that the removal of phenacetin in some countries has not been followed by the expected fall in mortality from analgesic nephropathy and that analgesic nephropathy has a poor prognosis if phenacetin is discontinued but other analgesics are abused further. The withdrawal of phenacetin from analgesic mixtures in Western Europe and the United States, gave rise to question the nephrotoxic potency of the different kinds of products without phenacetin, available on the market [10]. The nephrotoxic potency of the newer analgesic mixtures could be demonstrated using different kinds of epidemiological observations [46]. First, the published case-control studies could confirm the nephrotoxic potency of analgesic mixtures and the different substances worked-up in these mixtures (Figure 1). Interpretation of the presented odds ratios per substance however, remains difficult since they were seriously influenced by the additional effect of other substances invariably taken together. Most casecontrol studies suffered from ingredient bias and were not able to evaluate the nephrotoxic effect of different

17. Analgesics and 5-aminosalicylic acid

combinations separately (Table 1). The only study that carefully avoid this bias, used too low doses to achieve trustful risk estimates [38]. Only Pommer attempted to entangle thoroughly the influence of different substances worked up in analgesic mixtures, showing an increased risk for phenacetin, paracetamol and phenazone containing analgesic mixtures controlled for the use of other combinations [36]. In the prospective controlled cohort studies, the study design and the limited number of cases with renal failure did not allow to study the nephrotoxic effect of different substances used [39, 40]. Moreover, a cohort of 226 patients with a clear diagnosis of analgesic nephropathy was investigated regarding their analgesic consumption. Patients were recruited within the framework of diagnostic criteria studies in Belgium (n=130) and eleven other European countries (n=96) [50, 51]. In all patients, analgesic nephropathy was diagnosed using the same validated renal imaging criteria with high diagnostic performance [51, 52]. In all included patients, the history of abuse was documented by the same methodology using the same structured questionnaire accompanied by a color picture book showing the analgesics with a high sales volume in each particular country. Results clearly showed that analgesic nephropathy was associated with the abuse of different kind of analgesic mixtures mostly containing phenacetin. However, 46 out of the 226 patients never consumed phenacetin-containing analgesics. Their documented analgesic nephropathy was associated with the abuse of the following combinations: aspirin and acetaminophen, aspirin and a pyrazolone, acetaminophen and a pyrazolone, and two pyrazolones all of which were combined with caffeine, codeine or both. Additionally, the minimal analgesic consumption for developing analgesic nephropathy could be defined as a daily consumption for at least five years. None of the subjects with a daily use of analgesic mixtures for less than 5 years (n=16) or those with a weekly but not a daily consumption for more than 5 years (n=19) met the renal imaging criteria of analgesic nephropathy [53]. Furtheron, a broad range of other clinical and epidemiological observations is in support with the previous results. For single analgesics, abuse is only poorly documented and the nephrotoxic potency of single analgesics can be considered as minimal. Even in patients with rheumatoid arthritis in which high

dose salicylate therapy was the mainstay of treatment, analgesic nephropathy seldom developed [14]. For single analgesics combined with caffeine/codeine, the example of Sweden is of particular interest. Although, Sweden has a high sales volume of this type of analgesics (40% of the total volume), prevalence of analgesic nephropathy remained at the low level of 1-2% during the last decade [54]. Moreover, it is of interest to note that in countries with a low prevalence of analgesic nephropathy such as Sweden and France, analgesic mixtures containing two analgesic substances combined with caffeine/codeine are not available (Sweden) or not sold (France), despite the fact that in both countries the total volume of analgesics sold is higher than in Belgium. Nephrotoxicity of single analgesics The most important clinical question remains to evaluate the nephrotoxic potential of different analgesic ingredients, when used as single substance. Particularly, the potential nephrotoxicity of paracetamol used as single analgesic remains a matter of debate. Case-control studies as well as observational cohort studies have controversial results, with 6 studies showing an increased risk and 3 studies that did not (Figure 1). Since most studies are not able to distinguish between paracetamol used as single analgesic and in combinations, presented risk ratios do not answer the question of nephrotoxicity when used as single analgesic (ingredient bias, see Table 1). The renal safety of aspirin used as single ingredient is easier to evaluate. From the seven case-control studies, only 3 showed an increased risk. All 3 suffered from the same ingredient bias as previously mentioned for paracetamol. In contrast however, both observational studies reported a robust, slightly decreased, odds ratio for the use of aspirin (Figure 1). In both studies, calculated odds ratio’s were based on hundreds of regular users of aspirin [41, 42]. Quantification of the problem Detailed information concerning the extent of the problem of analgesic nephropathy is limited, particularly for recent years. National annual data were collected in Australia/New Zealand by the Australian and New Zealand Dialysis and Transplant Registry (ANZDATA) [51] and in the United States by the United States Renal Data System (USRDS) [52]. In Eu403

ELSEVIERS & DE BROE

rope, the registration system of the European Dialysis and Transplant Association (EDTA) [53] published regularly incidence and prevalence data of analgesic nephropathy for all European countries in the past. Australian incidence rates showed a significant decline after the restriction of over-the-counter analgesic sales in 1979. During the 1970’s, Australia had the highest incidence rate in the world (up to 22%). The incidence declined to 15% in 1985 and to 11% in 1990 [58, 59]. In recent years the incidence remained at a level of 4%, decreasing earlier and faster among younger patients [55]. In Flanders, a region with welldocumented high incidence of analgesic nephropathy in end stage renal failure patients, the incidence fell from 17% in the mid eighties to 3-4% in recent years (Figure 2). In the United States the national prevalence of analgesic nephropathy is not well documented. In the 1980’s, local studies showed incidences, ranging from 1.7 in Philadelphia and 2.8% in Washington D.C. to 10% in Northwest North Carolina [34, 60,61]. According the USRDS annual data report, incidence of analgesic nephropathy remained at the very low level of 0.2% for patients starting renal replacement therapy in the last decade (USRDS 2005) [56]. In Canada 2.5% of dialysis patients had analgesic nephropathy in 1976. The recent prevalence can be expected to be low [62]. In South Africa in the early eighties, 33% of the white patients starting chronic renal replacement therapy in Durban were diagnosed with analgesic nephropathy [63]. In Kuala Lumpur, Malaysia, 8% of the 180 dialysis patients had consumed excessive quantities of analgesics and in 4% signs of renal papillary necrosis were observed [64]. More recently high analgesic abuse of 7-10% in rural areas was reported [65]. Incidence of analgesic nephropathy in ESRD population of Thailand is however unknown. On the other hand, Central and Eastern European countries were confronted in the 1990s with an increasing incidence of the disease partly due to the increasing number of older patients accepted for renal replacement therapy. In 1992, Matousovic et al. [66] measured an incidence rate of 9.1% of analgesic nephropathy in the Czech and Slovak Republics using renal imaging criteria [67]. The same methodology was used in Hungary where an incidence up to 13% was noted in 1996 [68]. In contrast, in the southwest region of Poland not any case of analgesic nephropathy could be identified 404

in the period 1991-1992. The investigators concluded, however, that a reassessment of the incidence after 5-10 years should be mandatory because in the early 1990’s only 40% of (younger) ESRD patients received dialysis treatment [69].

Pathophysiology The exact pathophysiological mechanism(s) of analgesic nephropathy is unknown. The disease is characterized by capillary sclerosis of the vessels of the renal pelvis and ureteral mucosa, renal papillary necrosis and calcification, interstitial infiltration fibrosis, progressive cortical atrophy next to zones with hypertrophy of the remaining nephrons, aspecific glomerular changes. The main pathological lesion strongly indicates the more distal parts of the nephron as the predilected target for analgesic toxicity (Figure 3). The potentiating effect of aspirin with both phenacetin and acetaminophen may be related to two factors: • Acetaminophen undergoes oxidative metabolism by prostaglandin H synthase to reactive quinoneimine that is conjugated to glutathione. If acetaminophen is present alone, there is sufficient glutathione generated in the papillae to detoxify the reactive intermediate. However, if acetaminophen is ingested with aspirin, the aspirin is converted to salicylate, which becomes highly concentrated and depletes glutathione in both the cortex and papillae of the kidney. With the cellular glutathione depleted, the reactive metabolite of acetaminophen then produces lipid peroxides and arylation of tissue proteins, ultimately resulting in necrosis of the papillae [9, 70]. • Aspirin and NSAID suppress prostaglandin production by inhibiting cyclooxygenase enzymes. Renal blood flow, particularly within the renal medulla, is highly dependent upon systemic and local production of vasodilatory prostaglandins. Thus, this region, in the setting of combined aspirin and NSAID use, is more prone to ischemic damage. Loss of proteoglycans and glycosaminoglycans, essential constituents of medullary matrix may occur.

Clinical aspects Clinically, analgesic nephropathy is characterized

17. Analgesics and 5-aminosalicylic acid

Phenacetin withdrawn: 1967 from Vincent® powders 1975 from Bex® powders

34%

1979-80: Mixed analgesic legislation

28%

20%

9% 8% 8% 6% 7% 6% 6% 6% 5% 4% 3% 3%

15%

16% 18% 14% 15% 13% 15% 11% 13%

20% 20%

23%

25%

24%

25%

22%

23% 24% 25%

30%

10%

77

70

1925

1750

85

2000 100

96

1700

97

1600

78

96 1371

1617

95

1560

96

1357

31 1108

3.0% 2006

38 1088

35

2004

47 1053

1152

2003

40 1036

2.8%

2002

48 944

3.5%

2001

47 881

2005

2000

41 882

4.5%

1998 62 821

3.9%

1997 63 769

5.1%

1996 69 723

5.3%

1995 65 703

4.6%

9.5% 6.4% 8.2% 5.3% 7.6%

9.2%

1994 73 652

92-93

90-91

88-89

86-87

FLANDERS New AN Number of patients Total new ESRF

6.0%

6.5%

11.2% 7.9%

7.3%

10.5%

11.0%

10.0% 82-83

80-81

78-79

76-77

74-75

Year

2001: Codeine containing analgesic mixtures on prescription

Belgium (10 million inh.) Flanders (6 million inh.)

13.5%

14.3%

8.0% 72-73

0%

1988: Free sales of phenacetin-containing analgesics banned 1988: written, signed request mandatory for obtaining analgesic mixtures 1988-91: most 2+ analgesics changed into 1+

4.0%

10%

5%

13.0%

17.3% 11.5%

15%

70-71

Incidence of analgesic nephropathy

20%

15.7%

Phenacetin withdrawn: 1972 from Witte Kruis® powders 1981 from Mann® powders

84-85

B

1371

Total new ESKD

1127 124

New AN Number of patients

992 119

Year

1111 100

1967

0%

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

5%

1999

Incidence of analgesic nephropathy

Australia

29% 30%

35%

28% 28%

A

Figure 2. . Incidence of analgesic nephropathy (AN) in Australia 1971-2005. B. Incidence of AN in Belgium 1970-1998 and in Flanders 1994-2006. At the bottom, absolute numbers of new AN. 405

ELSEVIERS & DE BROE

Synergistic toxicity of analgesics in the renal inner medulla

Phenacetin MFO

Centrally acting dependence producing drugs

Aspirin

Cyt p450

Paracetamol

Salicylate

Renal papillary concentration

Renal papillary concentration

Prostaglandin synthase

Glutathione depletion

Caffeine ± 50 mg

Codeine 10-30 mg

N-acetyl-p-benzoquinoneimine

Arylation of renal papillary protein + oxidative stress

Analgesic Nephropathy

Renal papillary necrosis

Figure 3. Synergistic toxicity of analgesics in the renal inner medulla and centrally acting dependence-producing drugs leading to analgesic nephropathy. Paracetamol undergoes oxidative metabolism by prostaglandin H synthase to reactive quinoneimine that is conjugated to glutathione. If paracetamol is present alone, there is sufficient glutathione generated in the papillae to detoxify the reactive intermedicate. If the paracetamol is ingested with aspirin, the aspirin is converted to salicylate and salicylate becomes highly concentrated in both the cortex and papillae of the kidney. Salicylate is a potent depletory of glutathione. With the cellular glutathione depleted, the reactive metabolite of paracetamol then produces lipid peroxides and arylatyion of tissue proteins, ultimately resulting in necrosis of the papillae. [Reproduced with permission from refs. 5 and 9].

by its slow and stealthy progression. Most analgesic nephropathy patients only attended the outpatient nephrology clinic when renal failure reached a chronic and advanced stage. End-stage renal failure due to analgesic abuse was observed after consumption for approximately 20 years and most patients with analgesic nephropathy entered renal replacement therapy in their fifth-sixth decade. An increased occurrence of anemia and an increased risk of developing vascular diseases and ischemic heart disease are mentioned in patients with analgesic nephropathy [71]. Gastrointestinal manifestations occur in more than half of analgesic abusers and 406

particularly gastric ulcerations are frequently reported [72, 73]. Psychological and psychiatric manifestations are common in analgesic abusers and this is reflected in the frequency of associated addictive habits such as smoking, alcoholism and the excessive use of psychotropic drugs. Also the prematurely aged appearance of these patients has been emphasized. These observations pointed to the fact that analgesic nephropathy is part of a much wider syndrome called “the analgesic syndrome” [30, 74, 75]. Moreover, in 1965 a first publication from Sweden drew attention to the increasing incidence of tumoral degeneration of the kidney and the urinary tract observed in analgesic abusers [76-78]. Additional case reports were published in Switzerland, Australia, and Belgium [79-81]. Although, the risk for developing tumors of the urinary tract after the abuse of different kinds of analgesics is not clearly established, the abuse of phenacetin containing products showed a four-to tenfold increased risk [84]. The tumors generally become apparent after 15 to 25 years of analgesic abuse [82], usually but not always in patients with clinically evident analgesic nephropathy [83]. Most patients are still taking the drug at the time of diagnosis, but clinically evident disease can first become apparent several years after cessation of analgesic intake and even after renal transplantation has been performed [82]. It is presumed that the induction of malignancy results from the intrarenal accumulation of N-hydroxylated phenacetin metabolites that have potent alkylating action [83]. Because of urinary concentration, the highest concentration of these metabolites will be in the renal medulla, ureters, and bladder, possibly explaining the predisposition to carcinogenesis at these sites. The major presenting symptom of urinary tract malignancy in analgesic nephropathy is microscopic or gross hematuria. Thus, continued monitoring is essential, and new hematuria should be evaluated with urinary cytology, and, if indicated, cystoscopy with retrograde pyelography [78]. It may also be prudent to obtain yearly urine cytology for the first several years if analgesics are discontinued or indefinitely if drug intake persists. The incidence of urothelial carcinoma after renal transplantation in patients with analgesic nephropathy is comparable to the general incidence of up to 10% of urothelial carcinomas in end-stage renal failure patients with analgesic nephropathy. Removal of the native kidneys prior to renal transplantation has

17. Analgesics and 5-aminosalicylic acid

B. Diagnostic criteria used

A

Renal Size

Right kidney RA A

RV

RA Left kidney A

SP

B B Decreased: A + B 5

Bumpy Contours

C

Normal kidney

Moderate renal failure

Belgian female, age 62 y, Scr 1.8 mg/dl. Abuse: 20 y mixture of pyrazolone derivatives

End-stage renal failure

Belgian female, age 59y, ESKD. Abuse: 8 y mixture of pyrazolone derivatives Abuse: 26 y mixture of aspirin + paracetamol

Figure 4. Diagnostic criteria of analgesic nephropathy. A. Macroscopic aspect of an analgesic nephropathy kidney from an ESKD patient. B. Diagnostic criteria used. C. CT scans without contrast material of subject with normal kidneys, and patients with analgesic nephropathy with CKD3 and ESKD.

also been suggested, but the efficacy of this regimen has not been proven [82].

Diagnosis Until recently, the diagnosis of analgesic nephropathy was difficult to obtain. The disease is associated with a large number of mainly aspecific clinical symptoms [80]. Renal papillary necrosis, considered as the hallmark of analgesic nephropathy, can only be directly demonstrated by autopsy, after nephrectomy or in the exceptional case of a patient eliminating a papilla [85]. In a large part of cases, the diagnosis was mainly based on a documented history of abuse after a process of exclusion of other causes of renal failure. Since, furthermore, several authors [86, 87] have noted that a clear history of abuse is difficult to obtain, the need for diagnostic criteria with a well-defined performance

became mandatory. In Belgium, a prospective controlled multicentre study started in 1988, aiming to select diagnostic criteria for analgesic nephropathy with well-defined performance in patients with end-stage renal failure. In a cohort of 60 analgesic abusers and 188 controls, all starting renal replacement therapy, a large number of clinical, laboratory and radiological signs reported to be associated with analgesic nephropathy were tested. It was found that renal imaging investigations (sonography and tomography) demonstrating a decrease in length of both kidneys combined with either bumpy contours or signs of renal papillary necrosis were the only ones which showed a high sensitivity and specificity for diagnosing the disease. Other signs frequently mentioned such as hypertension, anemia, sterile pyuria and bacteriuria showed low sensitivity and/or specificity [51]. 407

ELSEVIERS & DE BROE

In a separate study, the diagnostic value of CT scan without contrast media was compared to the previously used renal imaging techniques (sonography and tomography) (Figure 4). A cohort of 40 analgesic abusers (= daily use of mixtures during at least 5 years) and 40 controls, all end-stage renal failure patients without a clear renal diagnosis were investigated with sonography, tomography and CT scan without contrast, searching for the renal imaging signs of analgesic nephropathy. Using CT scan the renal size and contour could be evaluated with comparable results while this technique scored better for the detection of papillary calcifications (Figure 4) [50, 67]. In an additional controlled study, the diagnostic performance of CT scan in patients with incipient/ moderate renal failure was studied. In a cohort of 53 analgesic abusers with a serum creatinine between 1.5 and 4 mg/dl and in the absence of a clear renal diagnosis, a CT scan was performed. It was found that the renal image of analgesic nephropathy on CT scan in an early stage of renal failure is comparable with the observations made in end-stage renal failure patients (Figure 4). Especially the demonstration of bilateral papillary calcifications showed a high sensitivity of 92% with a specificity of 100% for the early diagnosis of analgesic nephropathy (Figure 4) [50]. The diagnostic value of CT-scan in the case of AN in ESKD patients was validated using all CT-scan documents (N=67) performed within the framework of the ANNE study in seven renal units. The renal imaging criteria were validated by a radiologist not involved in the process of data collection, and without knowledge of a possible history of abuse. His validation consisted of a blind re-examination of the CT-scans, accepting or rejecting the diagnosis of AN based on the observation of a decrease in renal volume plus either bumpy contours and or papillary calcifications. Afterwards, these results were compared with the history of AN use/abuse. This blind re-examination resulted in a comparable number of patients in whom the diagnosis of AN was accepted or rejected. The overall accordance of 94% was obtained between the original examination and the blind re-examination [50]. Several studies using this validated diagnostic test showed either the absence or low prevalence of AN [69], others confirmed the underestimation of AN in their country showing that within the cohort of patients with “unknown aetiology”, or chronic interstitial ne408

phritis a substantial number of patients with AN were detected [66, 68]. A recent study in ESKD patients (NANS-study) in the US evaluated the value of the non-contrast-enhanced computerized tomography as diagnostic test for AN [88]. It turned out that in contrast to previous studies [66, 68, 69], the sensitivity of the non contrast CT-scan for the detection of analgesic associated kidney injury was too low to be used as a test system. A specificity of more than 95% was found, being comparable with the earlier reports [51]. This overall result is not surprising since the low prevalence of AN (clearly below 5%) found in the US precludes clinically relevant sensitivity of the test.

Prevention Analgesic nephropathy is one of the few renal diseases currently suitable for primary prevention. Informative campaigns focused on the population at risk did not solve the problem of analgesic nephropathy. In Belgium, it was clearly demonstrated that in most abusers, sustained analgesic consumption was no longer related to a physical complaint but analgesics were mainly taken for their mood-altering capacities. Most analgesic abusers admitted to having been informed of the health risks related to long-standing analgesic abuse and even if renal impairment occurred, only a part of the cases stopped their analgesic abuse [89]. Also the withdrawal of phenacetin from analgesic mixtures did not solve the problem. When phenacetin was withdrawn from most analgesic mixtures in Australia (1970’s), no decline in the occurrence of analgesic nephropathy could be observed [80, 85]. A declining incidence rate was only observed after restriction of the over-the-counter sales of all analgesic mixtures in 1979-1980 [59, 55, 90]. Some countries in Europe, particularly Sweden, have succeeded in controlling the disease after legislative measures were taken. As early as the sixties, Sweden elaborated legislation that only a few years later became very effective. The legislation was simple and clear: all analgesic containing, even the slightest dose of phenacetin became prescription limited. This resulted in a prescription status of almost all combined analgesics, hence a dramatic drop in their sale. In spite of the substantial total increase of consumption of single analgesics between 1980-1990, analgesic nephropathy belongs nowadays to the his-

17. Analgesics and 5-aminosalicylic acid

tory of medicine-nephrology (< 1% of Swedish dialysis population) [54]. In contrast, in many other European countries, no effective legislative measurements were taken. In Belgium, Germany and Switzerland, the pharmaceutical industry spontaneously removed phenacetin from their products. Phenacetin was replaced by another analgesic substance such as pyrazolone maintaining a high volume of analgesic mixtures still containing two or more analgesic substances. In Belgium, the Ministry of Health decided in 1988 that when obtaining analgesic mixtures in the pharmacy users had to sign a request and received an information sheet warning for possible renal consequences of extensive analgesic consumption. This resulted obviously in a fall of the sale/consumption of analgesic mixtures. Although these measures were only effective during one year, their long-term and indirect effect was more important. After 1988, several pharmaceutical companies modified their analgesic products, resulting in a reduction of the mixtures from two analgesic components to one analgesic plus caffeine and/or codeine. Analgesic nephropathy gained recognition in recent years in several Central and Eastern European countries. Abuse of analgesic mixtures is also reported in several third world countries without any knowledge about the extent of the problem of analgesic nephropathy. Moreover, in many countries there are no legislative limitations to introduce analgesic mixtures containing two analgesic substances combined with caffeine/codeine onto the market. In view of prevention, it would be advisable to obtain legislative measures worldwide in order to limit the over-the-counter availability of all analgesics containing two analgesic components plus caffeine/codeine. This is formally asked in Europe [8] as well as in the United States [7] by a large group of investigators active in the field.

5-Aminosalicylic acid Epidemiological observations Case reports The association between the use of 5-aminosalicylic acid (5-ASA) and the development of chronic tubulointerstitial nephritis in patients with inflammatory bowel disease (IBD) gained recognition in the 1990s by

the publication of several case reports [26, 28, 92-99] (Figure 5). Reported cases are summarized in Table 2. The disease was more prevalent in males with a male/ female ratio of 16:3. The age of reported cases ranged from 14 to 45 years. In contrast with analgesic nephropathy where renal lesions were only observed after several years of analgesic abuse, interstitial nephritis associated to 5-ASA was already observed during the first year of treatment in 8 out of 19 reported cases. Most cases started 5-ASA therapy with a documented normal renal function. Complete recovery upon arrest of the drug however, was only observed in 6 out of 19 published cases and dependent on the degree of renal damage at diagnosis. In a recent review article, case reports of IBD patients showing renal disease associated with 5-ASA treatment, increased to a total of 46 reported cases [100]. Retrospective study A retrospective study was performed aiming to obtain more insight in the frequency of this disease [27]. Nephrologists of Belgium, France and the Netherlands were asked to report all cases of inflammatory bowel disease (IBD) showing signs of renal impairment associated or not with 5-ASA therapy. Questionnaires were completed and returned by 71 nephrologists. Among them, 44 reported that they had no such cases. The remaining 27 nephrologists sent detailed information on 40 cases of IBD with renal failure. Among 40 reported cases 26 used 5-ASA including 15 with biopsy proven interstitial nephritis (Figure 5). It is worthwhile to notice that on the one hand a few cases with chronic tubulointerstitial nephritis never used 5-ASA and on the other hand some cases with 5-ASA therapy showed renal failure diagnosed as glomerulonephritis or amyloidosis. A recent investigation performed by a written questionnaire sent to all gastro-enterologists and nephrologists in the United Kingdom confirmed these observations. Retrospectively, a total of 202 cases of 5-ASA nephrotoxicity were identified during the preceeding 10 years. On drug withdrawal, complete renal recovery was only observed in 25% of the patients [101]. Prospective studies Until now, the risk ratio of 5-ASA associated renal failure in patients with inflammatory bowel disease is not known. It will be rather difficult to obtain more 409

ELSEVIERS & DE BROE

Table 2. Published case reports of interstitial nephritis in patients with inflammatory bowel disease using 5-ASA. Reference

Sex

Age

von Mühlendal [92] Henning [92] Ruf-Ballauf [94] Mehta [95]

m m m m

14 31 45 29

5 42 7 5

Masson [96] Thulavath [97]

m m m m f m m m m m m f m

26 28 24 42 37 25 30 34 31 43 24 30 24

18 26 36 20 5 26 5 8 42 28 22 3 23

Smilde [98]

World [26]

Stolear [28]

5-ASA use duration (months)

Creatinine clearance (ml/min) onset therapy lowest level ? 31* normal 17 80* 33 normal 38 normal normal 121* 90 51 78 116 73 ? ? ? ? 104*

Follow-up upon arrest 5-ASA (months)

55* 45 47 18 25 14 54 13 33 25 17 12 54

complete no no partial complete partial no no no no no no partial

? 12 36 5 2 12 27 44 27

*:calculated creatinine clearance

B

C

A Serum Creatinine (mg/dl) 12 10.6

C.P. 10

man born 19.01.1971

8

7.3

is

y re

l na

bio

ps

y

4.0

3.9

3.8 Potassium (mEq/L):

3.3

32 mg/day

2

3.3

01 8/ /0

8/ 00 /0 05

16

5/ 99 /0

22

3/ 94

2/ 94

/0

/0 23

02

3/ 92 /0

0/ 91 15

/1 03

HEMODIALYSIS

Methylprednisolon

/1 02 1/9 /1 4 22 2/9 /1 4 31 2/9 /1 4 2 06 /94 /0 1/ 95 01 /0 5/ 96 01 /1 2/ 96

HEMODIALYSIS

3x500 mg/day orally

0

16 mg/day

08

Pentasa®

2.8

2.6

1.1

3/ 05

ps

4.3

/0

bio

4.2

15

re

l na

5.4

5.3

4.9

8/ 02

d

os

/0

IBD

4

n iag

03

6

Figure 5. Case report of nephrotoxicity of 5-aminosalicylic acid (5-ASA) in inflammatory bowel disease. A. Evolution of renal failure. B. First renal biopsy. C. Second renal biopsy. Note the important cellular infiltration in both biopsies. Normal aspect of glomeruli. B-C: H&E staining, orig. magn. x350.

insight in this risk. Drug usage in these patients is irregular and acute episodes of inflammation result in either increasing drug regimen and/or increasing number of drugs prescribed, including some with known nephrotoxic potential. Furthermore, the kidney is an extra-intestinal target of the disease as shown in 410

Table 2. In recent years several attempts were made to measure early signs of renal impairment in patients with IBD treated with 5-ASA. Schreiber et al. [102] investigated 223 IBD patients using sensitive markers of glomerular and tubular dysfunction. Patients

17. Analgesics and 5-aminosalicylic acid

of clinical nephrotoxicity of only one in 4000 patients/ year [101].

Pathophysiology This particular form of chronic tubulo-interstitial nephritis is characterized by an important cellular infiltration of the interstitium with macrophages, T-cells but also B-cells. Furthermore, after arrest of the drug, there is improvement of the renal function in some cases [26, 93]. In those in which there is a delayed diagnosis of renal damage, recovery of renal function does not occur. Instead, several of those patients needed one or another form of renal replacement therapy. An important aspect of this type of toxic nephropathy is the documented persistence of the inflammation of the renal interstitium even several months after arrest of drug intake [28]. The molecular structure of 5-ASA is very close of that of salicylic acid, phenacetin and aminophenol, drugs with a well-documented nephrotoxic potential (Figure 6). In rats, it is demonstrated that after a single intravenous injection of 5-ASA, at doses of 1.4, 2.8, 5.7 mM per kg body weight (high pharmacological doses), necrosis of the proximal convoluted tubules and papillary necrosis developed [107]. The mechanism of renal damage caused by 5-ASA may be analogue to that of salicylates by inducing hypoxia of renal tissues either by uncoupling oxidative phosphorylation in renal mitochondria, by inhibiting the synthesis of renal prostaglandins, or by rendering the kidney susceptible to oxidative damage by a reduc-

HOOC

COOH

mesalazine (5-aminosalicylic acid)

H5C2O

phenacetin

NHCOCH3

salicylic acid

OH

OH

NH2

OH

receiving high amounts of 5-ASA showed an increased prevalence of tubular proteinuria. He concluded that the possibility exists that high doses of 5-ASA may be associated with proximal tubular proteinuria but that his study design was not able to dissect the possible impact of chronic inflammation on the development of renal impairment. In contrast, K.R. Herlinger et al. [103] performed an investigation on 95 IBD patients carefully assessing disease activity. They concluded that tubular proteinuria occurred in the majority of IBD patients and was related to disease activity rather than to 5-ASA treatment. Their observations were confirmed by A.C. Poulou et al [104] in a prospective study of 86 IBD patients, showing that the observed microproteinuria was mainly associated with IBD activity but not affected by 5-ASA A European prospective study aiming to register all IBD patients with renal impairment and to control for a possible association with 5-ASA therapy was performed [105, 106]. During a one-year observation period, gastroenterologists of Belgium, France, Italy, Republic of Macedonia and Yugoslavia registered 1529 patients with IBD seen at their outpatient clinic. At the start of the study a questionnaire was filled in focused on medical and drug history. Additional data were collected at baseline, after 6 months and after 12 months, including activity of IBD, actual medication and results of the serum creatinine determination. Only 34 patients (2.2%) showed at least once a decreased creatinine clearance. Consecutive decreased creatinine clearances were observed in 13 patients (0.9%). Dehydration due to low body mass combined with active IBD was the main reason for an intermittent decrease in renal function in most of these patients. Particularly, the observation of 5-ASA therapy in 5 patients with sustained renal impairment of unknown origin is suggestive for a possible etiological role of 5-ASA. Comparing patients with and without renal impairment, the presence of a stoma revealed the highest increased risk. The number of renal impairment cases observed in this prospective study is highly comparable with the estimations made by World et al. 5 years ago. They stated that the available evidence suggested that renal impairment of any severity may occur in up to one in 100 patients, but clinically significant interstitial nephritis occurs in less than one in 500 patients [26]. The more recent experience of 5-ASA nephrotoxicity in the United Kingdom led to an estimated incidence

NH2

p-aminophenol

Figure 6. Molecular structure of 5-aminosalicylic acid, salicylic acid, phenacetin and p-aminophenol. 411

ELSEVIERS & DE BROE

ing renal glutathione concentration after inhibition of the pentose phosphate shunt. 5-ASA is taken up by the gastrointestinal tract, particularly in the acetylated form and eliminated as such in the urine. The colon is the predilected place for this acetylation since in the small bowel there is a lack of the responsible bacterial flora. Hence, there is a limited readily absorption of 5-ASA as such in the small bowel. How far this may form a rationale for a possible difference in nephrotoxicity for the different preparations remains to be determined. Indeed, experimental evidence has shown that free 5-ASA is more nephrotoxic than the acetylated form [108, 109].

Clinical aspects A typical case report is shown in Figure 5. An association between the use of 5-ASA in patients with chronic inflammatory bowel disease and the development of a particular type of chronic tubulo-interstitial nephritis is difficult to interpret since renal involvement in chronic inflammatory bowel disease may be an extra-intestinal manifestation of the underlying disease [110]. Extra-intestinal manifestations of chronic inflammatory bowel disease are well recognized. The most frequent renal complications are oxalate stones and their consequences such as pyelonephritis, hydronephrosis and on the long-term amyloidosis [111, 112]. As for many drugs, reversible acute interstitial nephritis has been described [90]. Glomerulonephritis may be associated with chronic inflammatory bowel disease and has a heterogeneous expression [113]. Minimal change glomerulonephritis, membranous, membranoproliferative, focal glomerulosclerosis, and proliferative crescentic glomerulonephritis have been described and a summary of these case reports is available in the paper of Wilcox et al. [114]. In almost half of these cases, there was no relationship with drug intake such as sulphasalazine or 5-ASA. That 5-ASA seems to be implicated in the generation/development/maintenance of this particular reaction at the level of the kidney however, is supported by a large number of case reports appearing in recent

412

literature of patients with IBD using 5-ASA as the only medication, the improvement at least partial of the impaired renal function arrest of the drug and a worsening after resuming 5-ASA use [87, 100, 101].

Prevention The efficacy of 5-ASA as first-line treatment for IBD is clearly documented and generally accepted [115, 116]. Preventive measures need to be taken into consideration however, in order to avoid nephrotoxic adverse effects. Although the incidence and risk ratio’s of 5-ASA associated chronic tubulo-interstitial nephritis are not well known, the link established by case reports and the demonstration that recovery of renal function was observed only in patients with limited renal damage necessitates preventive measures [26]. The experience in the United Kingdom confirmed that the improvement of renal function for patients with nephrotoxicity treated for less than one year was significantly better than those on treatment for much longer [101]. Patients receiving 5-ASA should be screened regularly in order to detect signs of renal impairment. It is suggested that serum creatinine concentration should be measured each month for the first 3 months of treatment, three monthly for the remainder of the first year and annually thereafter [26]. The use of concurrent immunosuppressive therapy may necessitate extension to the period of intensive monitoring. Moreover, it is shown that IBD patients with a stoma and patients with extreme dehydration are more susceptible to develop renal impairment (Table 3).

Table 3. Risk factors for renal impairment in inflammatory bowel disease patients. Risk ratio (95% CI) Stoma

6.2 (1.8-20.9)

Male sex

3.1 (1.1-8.6)

Duration of IBD symptoms (weeks)

1.06 (1.01-1.12)

17. Analgesics and 5-aminosalicylic acid

References 1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.

Spuehler O, Zollinger HU. Die chronisch-interstitielle Nephritis. Zeitschr Klin Medizin 1953; 151: 1–50. Gloor F. Die doppelseitige chronische nichtobstruktive interstitielle Nephritis. Ergebnisse der allgemeinen Pathologie und Pathologischen Anatomie 1961; 41: 63–207. Burry AF, de Jersey P, Weedon D. Phenacetin and renal papillary necrosis: results of a prospective autopsy investigation. Med J Australia 1966; 53: 873–879. Kincaid-Smith P. Pathogenesis of the renal lesion associated with the abuse of analgesics. Lancet 1967; i: 859–862. Kincaid-Smith P, Nanra RS. Lithium-induced and analgesic-induced renal diseases. In: Diseases of the kidney, 5th edition. Schrier RW, Gottschalk CW (editors). Little Brown, Boston 1993; p. 1099-1129. Mihatsch MJ, Torhorst J, Amsler B, Zollinger HU. Capillarosclerosis of the lower urinary tract in analgesic (phenacetin) abuse. Virchows Archiv A. Pathol Anat Histol 1978; 381: 41–47. Henrich WL, Agodoa LE, Barrett B, Bennett WM, Blantz RC, Buckalew VM Jr, D’Agati VD, De Broe ME, Duggin GG, Eknoyan G. Analgesics and the kidney: summary and recommendations to the Scientific Advisory Board of the National Kidney Foundation from an Ad Hoc Committee of the National Kidney Foundation. Am J Kidney Dis 1996; 27(1): 162-165. De Broe ME, Elseviers MM, Bengtsson U, Mihatsch MJ, Molzahn M, Pommer W, Ritz E, Schwarz A. Analgesic nephropathy. Nephrol Dial Transplant 1996; 11: 2407-2408. Duggin GG. Combination analgesic-induced kidney disease: the Australian experience. Am J Kidney Dis 1996; 28 (suppl 1): S39S47. De Broe ME, Elseviers MM. Analgesic nephropathy. N Engl J Med 1998; 338: 446-452. Haschek H (editor). International symposium on problems of phenacetin abuse. Egermann, Vienna 1973. Gsell O. Nephropathie durch Analgetika. Ergebnisse der inneren Medizin und Kinderheilkunde 1974; 35: 68–175. Carro-Ciampi G. Phenacetin abuse: a review. Toxicology 1978; 10: 311–339. Prescott LF. Analgesic nephropathy: a reassessment of the role of phenacetin and other analgesics. Drugs 1982; 23: 75-149. Kincaid-Smith P. Effects of non-narcotic analgesics on the kidney. Drugs 1986; 42(4): 109-128. Sandler DP, Smith JC, Weinberg CR, Buckalew VM Jr, Dennis VW, Blythe WB, Burgess WP. Analgesic use and chronic renal disease. N Engl J Med 1989; 320(19): 1238-1243. Sandler DP, Burr FR, Weinberg CS. Nonsteroidal anti-inflammatory drugs and the risk for chronic renal disease. Ann Int Med 1991; 115(3): 165-172. Morlans M, Laporte JR, Vidal X, Cabeza D, Stolley PD. End-stage renal disease and non-narcotic analgesics: a case-control study. Br J Clin Pharmacol 1990; 30(5): 717-723. Pommer W. Clinical presentation of analgesic-induced nephropathy. In: Analgesic and NSAID-induced kidney disease. Stewart JH (editor). Oxford University Press, Oxford 1993; p. 108-118. Perneger TV, Whelton PK, Klag MJ. Risk of kidney failure associated with the use of acetaminophen, aspirin, and nonsteroidal antiinflammatory drugs. N Engl J Med 1994; 331: 1675–1679. Fored CM, Ejerblad E, Lindblad P, Fryzek JP, Dickman PW, Signobello LB, Lipworth L, Elinder CG, Blot WJ, McLaughlin JK, Zack MM, Nyrén O. Acetaminophen, aspirin, and chronic renal failure. N Engl J Med 2001; 345: 1801-1808. Larsen K, Moller CE. A renal lesion caused by abuse of phenacetin. Acta Med Scand 1959; 164: 53-71. Jacobs LA, Morris JG. Renal papillary necrosis and the abuse of phenacetin. Med J Australia 1962; 2(14): 531-538. Grimlund K. Phenacetin and renal damage at a Swedish factory. Acta Med Scan 1963; 174 (suppl 405): 1-26. Azad Khan AK, iris J, Truelove SC. An experiment to determine the active therapeutic moiety of sulphasalazine. Lancet 1977; 1: 892-895. World MJ, Stevens PE, Ashton MA, Rainford DJ. Mesalazine-associated interstitial nephritis. Nephrol Dial Transplant 1996; 11: 614-621. De Broe ME, Stolear JC, Nouwen EJ, Elseviers MM. 5-Aminosalicylic acid (5-ASA) and chronic tubulo-interstitial nephritis in patients with chronic inflammatory bowel disease: is there a link ? Nephrol Dial Transplant 1997; 12: 1839-1841. Stolear JC, Georges B, Gossun A, Ros Ph, Nouwen EJ, De Broe ME. Néphrite interstitiëlle associée à la Mésalazine. Confrontations anatomo-cliniques en pathologie renale et vasculaire; Hôpital Tenon, Paris, 10 Janvier 1996. Murray RM. Analgesic nephropathy: removal of phenacetin from proprietary analgesics. Brit Med J 1972; 4: 131-132. Nanra RS, Stuart-Taylor J, De Leon AH, White KH. Analgesic nephropathy: etiology, clinical syndrome, and clinicopathologic correlation in Australia. Kidney Int 1978; 13: 79-92. Schwarz A, Kunzendorf U, Keller F, Offerman G. Progression of renal failure in analgesic-associated nephropathy. Nephron 1989; 53: 244-249.

413

ELSEVIERS & DE BROE

32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50.

51.

52.

53. 54. 55. 56.

57.

414

Hauser AC, Derfler K, Balcke P. Progression of renal insufficiency in analgesic nephropathy: Impact of continuous drug abuse. J Clin Epidemiol 1991; 44: 53-56. McCredie M, Stewart JH, Mahony JF. Is phenacetin responsible for analgesic nephropathy in New South Wales ? Clin Nephrol 1982; 17(3): 134-140. Murray TG, Stolley PD, Anthony JC, Schinnar R, Hepler-Smith E, Jeffreys JL. Epidemiologic study of regular analgesic use and end-stage renal disease. Arch Intern Med 1983; 143: 1687-1693. Pommer W, Glaeske G, Molzahn M. The analgesic problem in the Federal Republic of Germany: analgesic consumption, frequency of analgesic nephropathy and regional differences. Clin Nephrol 1986; 26: 273-278. Pommer W, Bronder E, Greiser E, Helmert U, Jesdinsky HJ, Klimpel A, Borner K, Mohlzahn M. Regular analgesic intake and the risk of end-stage renal failure. Am J Nephrol 1989; 9: 403-412. Ibáñez L, Morlans M, Vidal X, Martínez MJ, Laporte JR. Case-control study of regular analgesic and nonsteroidal anti-inflammatory use and end-stage renal disease. Kidney Int 2005; 67(6): 2393-8. van der Woude FJ, Heinemann LA, Graf H, Lewis M, Moehner S, Assmann A, Kuhl-Habich D. Analgesics use and ESRD in younger age: a case-control study. BMC Nephrol 2007; 8(1): 15. Dubach UC, Rosner B, Pfister E. Epidemiologic study of abuse of analgesics containing phenacetin. Renal morbidity and mortality (1968-1979). N Engl J Med 1983; 308: 357-362. Elseviers MM, De Broe ME. A long-term prospective controlled study of analgesic abuse in Belgium. Kidney Int 1995; 48: 19121919. Kurth T, Glynn RJ, Walker AM, Rexrode KM, Buring JE, Stampfer MJ, Hennekens CH, Gaziano JM. Analgesic use and change in kidney function in apparently healthy men. Am J Kidney Dis 2003; 42(2): 234-44. Curhan GC, Knight EL, Rosner B, Hankinson SE, Stampfer MJ. Lifetime nonnarcotic analgesic use and decline in renal function in women. Arch Intern Med 2004; 164(14):1519-24. McLaughlin JK, Lipworth L, Chow W-H, Blot WJ. Analgesic use and chronic renal failure: A critical review of the epidemiologic literature. Kidney Int 1998; 54: 679-686. Delzell E, Shapiro S. A review of epidemiologic studies of nonnarcotic analgesics and chronic renal disease. Medicine 1998; 77(2): 102-121. Buckalew VM, Schey HM. Renal disease from habitual antipyretic analgesic consumption: an assessment of the epidemiologic evidence. Medicine 1986, 11: 291-303. Elseviers M.M., De Broe M.E. Analgesic nephropathy: Is it caused by multi-analgesic abuse or single substance use ? Drug Safety 1999; 20(1): 15-24. Gates TN, Temple AR. Analgesic use and chronic renal disease. N Engl J Med 1989; 321: 1125. McCredie M. Analgesic use and chronic renal disease. N Engl J Med 1989; 321: 1125. Elseviers MM, De Broe ME. Epidemiology of analgesic nephropathy. J Nephrol 1992; 5: 94-98. Elseviers MM, De Schepper A, Corthouts R, Bosmans JL, Cosyn L, Lins RL, Lornoy W, Matthys E, Roose R, Van Caesbroeck D, Waller I, Horackova M, Schwarz A, Svrcek P, Bonucchi D, Franek E, Morlans M, De Broe ME. High diagnostic performance of CT scan for analgesic nephropathy in patients with incipient to severe renal failure. Kidney Int 1995; 48: 1316-1323. Elseviers MM, Bosteels V, Cambier P, De Paepe M, Godon JP, Lins R, Lornoy W, Matthys E, Moeremans C, Roose R, Theelen B, Van Caesbroeck D, Verbanck J, De Broe ME. Diagnostic criteria of analgesic nephropathy in patients with end-stage renal failure: results of the Belgian study. Nephrol Dial Transplant 1992; 7: 479-486. Elseviers MM, Waller I, Nenov D, Levora J, Matousovic K, Tanquerel T, Pommer W, Schwarz A, Keller E, Thieler H, Köhler H, Lemoniatou H, Cresseri D, Bonucchi D, Fiocchi O, Jordans J, Franek E, Silva FJ, Fernandez Ruiz EJ, Morlans M, Hellström M, Wauters JP, Felle D, Cassi HV, De Broe ME. Evaluation of diagnostic criteria for analgesic nephropathy in patients with end-stage renal failure: results of the ANNE study. Nephrol Dial Transplant 1995; 10: 808-814. Elseviers MM, De Broe ME. Combination analgesic involvement in the pathogenesis of analgesic nephropathy. Am J Kidney Dis 1996; 28(6): 958-962 (letter to the editor). Noels LM, Elseviers MM, De Broe ME. Impact of legislative measures on the sales of analgesics and the subsequent prevalenceof analgesic nephropathy: a comparative study in France, Sweden and Belgium. Nephrol Dial Transplant 1995; 10: 167-174. Chang SH, Mathew TH, McDonald SP. Analgesic Nephropathy and Renal Replacement Therapy in Australia: Trends, Comorbidities and Outcomes. Clin J Am Soc Nephrol 2008; 3: 768-776. U.S. Renal Data System, USRDS 2007 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2007. Wing AJ, Brunner FP, Geerlings W, Broyer M, Brynger H, Fassbinder W, Rissoni G, Selwood NH, Tufveson G. Contribution of toxic nephropathies to end-stage renal failure in Europe: a report from the EDTA-ERA Registry. Toxicol Lett 1989; 46: 281-292.

17. Analgesics and 5-aminosalicylic acid

58. 59. 60. 61. 62. 63. 64. 65.

66.

67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78.

79. 80. 81.

82. 83. 84.

Nanra RS. Analgesic nephropathy in the 1990’s: an Australian perspective. Kidney Int 1993; 42 (suppl 44): 86-92. Stewart JH, McCredie M, Disney APS, Mathew TH. Trends in incidence of end-stage of end-stage renal failure in Australia, 19721991. Nephrol Dial Transpl 1994; 9: 1377-1382. McAnally JF, Winchester JF, Schreiner GE. Analgesic nephropathy. An uncommon cause of end-stage renal disease. Arch Intern Med 1983; 143: 1897-1899. Gonwas TA, Hamilton RW, Buckalew VM. Chronic renal failure and end-stage renal disease in Northwest North Carolina. Importance of analgesic-associated nephropathy. Arch Intern Med 1981; 141: 462-465. Wilson DR, Gault MH. Declining incidence of analgesic nephropathy in Canada. Can Med Assoc J 1982; 127: 500-503. Seedat YK, Naicker S, Rawat R, Parsoo I. Racial differences in the causes of end-stage renal failure in Natal. S Afr Med J 1984; 65: 956-958. Segasothy M, Suleimam AB, Puvaneswary M, Rohana A. Paracetamol: a cause for analgesic nephropathy and end-stage renal disease. Nephron 1988; 50: 50-54. Kantachuvesiri S, Kaojarern S, Kitayaporn D, Phanichphant S, Sumethkul V, Wananukul W, Pongskul C, Arkaravichien W, Jankriengkri S, Chanchairujira T. Risk factors between analgesic use and chronic nephropathy in Thailand. Southeast Asian J Trop Med Public Health 1996; 27(2): 350-355. Matousovic K, Elseviers MM, Devecka D, Horackova M, Turek T, De Broe ME, and the Group of Czech and Slovak Nephrologists. Incidence of analgesic nephropathy among patients undergoing renal replacement therapy in Czech republic and Slovak republic. Nephrol Dial Transplant 1996; 11: 1048-1051. Elseviers MM, De Broe ME. Is analgesic nephropathy still a problem in Belgium ? Nephrol Dial Transplant 1988; 2: 143-149. Matyus J, Ujhelyi L, Karpati I, Ben T, Kakuk G. Increase in the incidence of analgesic nephropathy in Hungary. Nephrol Dial Transplant 1997; 12(8): 1774-1775. Franek E, Kokot F, Grzeszczak W, Gajos L. Is analgesic nephropathy a problem in the South-West Region of Poland ? Nephron 1996; 72: 353 (letter to the editor). Bennett WM, De Broe ME. Analgesic nephropathy – a preventable renal disease. N Engl J Med 1989; 320: 1269-1271. Mihatsch MJ, Staehelin HB, Musfeld D, Perret E, Oberholzer M. Phenacetin-Abusus: Kardiovaskuläre Risikofactoren. Nieren Hochdruck 1983; 3: 83-92. Ivey KJ. Gastrointestinal intolerance and bleeding with non-narcotic analgesics. Drugs 1986; 32(suppl 4): 71-89. Laporte J-R, Carne X, Vidal X, Moreno V, Juan J. Upper gastrointestinal bleeding in relation to previous use of analgesics and non-steroidal anti-inflammatory drugs. Lancet 1991; 337: 85-89. Mihatsch MJ. Empfohlene Massnahmen zur Bekampfung des Analgetikamisbrauchs. In: Das Analgetikasyndrom. Mihatsch MJ (editor). George Theme Verlag, Stuttgart 1986; p. 134-135. Stewart JH. The analgesic syndrome. In: Analgesic and NSAID-induced kidney disease. Stewart JH (editor). Oxford University Press, Oxford 1993; p. 58-66. Hultengren N, Lagergren C, Ljungqvist A. Carcinoma of the renal pelvis in renal papillary necrosis. Acta Chir Scand 1965; 130: 314-320. Bengtsson U, Johansen S, Angervall L. Malignancies of the urinary tract and their relation to analgesic abuse. Kidney Int 1978; 13: 107-113. Kliem V, Thon W, Krautzig S, Kolditz M, Behrend M, Pichlmayr R, Koch KM, Frei U, Brunkhorst R. High mortality from urothelial carcinoma despite regular tumor screening in patients with analgesic nephropathy after renal transplantation. Transpl Int 1996; 9: 231-235. Mihatsch MJ, Manz T, Knüsli C, Hofer HO, Rist M, Guetg R, Rutishauser G, Zollinger HU. Phenacetinabusus III. Maligne Harnwegtumoren bei Phenacetinabusus in Basel 1963-1977. Schweiz Med Wschr 1980; 110: 255-264. McCredie M, Ford JM, Stuart Taylor J, Stewart JH. Analgesics and cancer of the pelvis in New South Wales. Cancer 1982; 49: 26172625. Lornoy W, Becaus I, Morelle V, Fonteyne E. Néphropathies chroniques par abus d’analgésiques: fréquence, pathogenèse, aspects cliniques, développement de tumeurs malignes. In: Séminaires d’uro-néphrologie. Chatelain C (editor). Masson, Paris 1988; p. 99-105. Blohme I, Johansson S. Renal pelvic neoplasms and atypical urothelium in patients with end-stage analgesic nephropathy. Kidney Int 1981; 20(5): 671-675. McCredie M, Stewart JH, Carter JJ, Turner J, Mahony JF. Phenacetin and papillary necrosis: independent risk factors for renal pelvic cancer. Kidney Int 1986; 30(1): 81-84. Kincaid-Smith P. Renal toxicity of non-narcotic analgesics. At-risk patients and prescribing applications. Med Toxicol 1986; 1 (suppl 1): 14–22.

415

ELSEVIERS & DE BROE

85. 86. 87. 88.

89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104.

105. 106.

107. 108. 109. 110. 111.

112. 113. 114.

416

Gloor FJ. Changing concepts in pathogenesis and morphology of analgesic nephropathy as seen in Europe. Kidney Int 1978; 13: 27-33. Murray RM. Genesis of analgesic nephropathy in the United Kingdom. Kidney Int 1978; 13: 50-57. Schwarz A, Faber U, Borner K, Keller F, Offerman G, Molzahn M. Reliability of drug history in analgesic users. Lancet 1984; 2:11631164. Henrich WL, Clark RL, Kelly JP, Buckalew VM, Fenves A, Finn WF, Shapiro JI, Kimmel PL, Eggers P, Agodoa LE, Porter GA, Shapiro S, Toto R, Anderson T, Cupples LA, Kaufman DW. Non-contrast-enhanced computerized tomography and analgesic-related kidney disease: report of the national analgesic nephropathy study. J Am Soc Nephrol 2006; 17(5):1472-80. Elseviers MM, De Broe ME. Analgesic nephropathy in Belgium is related to the sales of particular analgesic mixtures. Nephrol Dial Transplant 1994; 9: 41-46. Kincaid-Smith P. Analgesic nephropathy in Australia. Contrib Nephrol 1979; 16: 57-64. Nanra RS, Kincaid-Smith P. Experimental evidence for nephrotoxicity of analgesics. In: Analgesic and NSAID-induced kidney disease. Stewart JH (editor). Oxford University Press, Oxford 1993, p. 17-31 von Mühlendahl KE. Nephritis durch 5-aminosolicylsäure. Deutsch Med Wschr 1989; 114: 236. Henning HV, Meinhold J. Chronische interstitielle nephritis durch 5-aminosalicylsäure ? Deutsch Med Wsch 1989; 114: 1091. Ruf-Ballauf W, Hofstädter F, Krentz K. Acute interstitielle nephritis durch 5-aminosalicylsäure ? Internist 1989; 30: 262-264. Mehta RP. Acute interstitial nephritis due to 5-aminosalicylic acid. Can Med Assoc J 1990; 143(10): 1031-1032. Masson EA, Rodhes JM. Mesalazine associated nephrogenic diabetes insipidus presenting as weight loss. Gut 1992; 33: 563-4. Thuluvath PJ, Ninkovic M, Calam J, Anderson M. Mesalazine induced interstitial nephritis. Gut 1994; 35: 1493-1496. Smilde TJ, van Liebergen FJHM, Koolen MI, Gerlag PGG, Assmann KJM, Berden JHM. Tubulo-interstitiële nefritis door mesalazine (5-ASA)-preparaten. Ned Tijdschr Geneesk 1994; 138: 2557-2561 Calvino J, Romero R, Pintos E, Losada E, Novoa D, Güimil D, Mardaras J, Sanchez-Guisande D. Mesalazine-associated tubulointerstitial nephritis in inflammatory bowel disease. Clin Nephrol 1998; 49(4): 265-257. Gisbert JP, Gonzalez-Lama Y, Maté J. 5-Aminosalicylates and renal function in inflammatory bowel disease: a systematic review. Inflamm Bowel Dis 2007; 13(5): 629-638. Muller AF, Stevens PE, Mcintyre AS, Ellison H, Logans RF. Experience of 5-aminosalicylate nephrotoxicity in the United Kingdom. Aliment Pharmacol Ther 2005; 21: 1217-1224. Schreiber S, Hämling J, Zehnter E, Howaldt S, Daerr W, Raedler A, Kruis W. Renal tubular dysfunction in patients with inflammatory bowel disease treated with aminosalicylate. Gut 1997; 40: 761-766. Herrlinger KR, Noftz MK, Fellermann K, Schmidt K, Steinhoff J, Stange EF. Minimal renal dysfunction in inflammatory bowel disease is related to disease activity but not to 5-ASA use. Aliment Pharmacol Ther 2001; 15: 363-369. Poulou AC, Goumas KE, Dandakis DC, Tyrmpas I, Panagiotaki M, Georgouli A, Soutos D, Archimandritis A. Microproteinuria in patients with inflammatory bowel disease: is it associated with the disease activity or the treatment with 5-aminosalicylic acid? World J Gastroenterol 2006; 12(5): 739-746 Elseviers M.M., De Broe M.E., the 5-ASA-Study Group. Risk of chronic interstitial nephritis in patients with inflammatory bowel disease treated with 5-ASA therapy. J Am Soc Nephrol 11: 128A, 2000. Elseviers MM, D’Haens G, Lerebours E, Plane C, Stolear JC, Riegler G, Capasso G, Van Outryve M, Mishevska-Mukaetova P, Djuranovic S, Pelckmans P, De Broe ME, for the 5-ASA study group. Renal impairment in patients with inflammatory bowel disease: association with aminosalicylate therapy? Clin Nephrol 2004; 61(2): 83-91. Calder IC, Funder CC, Green CR, Ham KN, Tange JD. Nephrotoxic lesions from 5-aminosalicylic acid. Brit Med J 1972; 1: 152-154. Calder IC, Funder CC, Green CR, Ham KN, Tange JD. Comparative toxicity of aspirin and phenacetin derivatives. Brit Med J 1971; 4: 518-521. Dwarakanath AD, Michael J, Allan RN. Sulphasalazine induced renal failure. Gut 1992; 33: 1006-1007 Izzedine H, Simon J, Piette AM, Lucsko M, Baumelou A, Charitanski D, Kernaonet E, Baglin AC, Deray G, Beaufils H. Primary chronic interstitial nephritis in Crohn’s disease. Gastroenterology 2002; 123(5): 1436-1440. Glickman RM. Inflammatory bowel disease (ulcerative colitis and Crohn’s disease). In: Harrison’s Principles of Internal Medicine (13th edition, volume 2). Isselbacher KJ, Braunwald E, Wilson JD, Martin JB, Fauci AS, Kasper DL (editors). McGraw-Hill, In, New York, p. 1403-1417. Greenstein AJ, Janowitz HD, Sachar DB. The extra-intestinal complications of Crohn’s disease and ulcerative colitis: a study of 700 patients. Medicine 1976; 55(4): 401-412. Moayyedi P, Fletcher S, Harnder P, Axon ATR, Brownjohn A. Mesangiocapillary glomerulonephritis associated with ulcerative colitis: case report of two patients. Nephrol Dial Transplant 1995; 10: 1923-1924. Wilcox GM, Aretz HT, Roy MA, Roche JK. Glomerulonephritis associated with inflammatory bowel disease. Gastroenterology 1990; 98: 786-791.

17. Analgesics and 5-aminosalicylic acid

115. Järnerot G. New salicylates as maintenance treatment in ulcerative colitis. Gut 1994; 35: 1155-1158. 116. Clemett D, Markham A. Prolonged-release mesalazine: a review of its therapeutic potential in ulcerative colitis and Crohn’s disease. Drugs 2000; 59(4): 929-956.

417

18

Non-steroidal anti-inflammatory drugs Ali J. OLYAEI1, Andrew WHELTON2, Til STURMER3 and George A. PORTER1 2Universal

1Oregon Health Science University, Portland, Oregon, USA Clinical Research Center and Johns Hopkins University, Baltimore, Maryland, USA 3German Centre for Research on Ageing, Heidelberg, Germany

Introduction ___________________________________________________________ 419 Prostaglandins and renal function __________________________________________ 420 The prostaglandin pathway Renal prostanoid receptors

420 421

Cyclooxygenase isoforms _________________________________________________ 422 Factors regulating isoform expression Distribution within the kidney

422 422

Mechanism of action of NSAIDs ____________________________________________ 422 Renal syndromes associated with NSAIDs ____________________________________ 423 Acute deterioration of renal function Salt and water retention The concept of “renal sparing” NSAIDs Nephrotic syndrome with interstitial nephritis Chronic renal failure/papillary necrosis Other NSAID-induced renal syndromes

424 428 430 431 432 434

Renal effects of COX-2 inhibitors ___________________________________________ 435 Effects on renal function: GFR/urinary sodium excretion Incidence of adverse cardio-renal events

435 436

Concurrent use of an oral synthetic prostaglandin analog with a NSAID ___________ 444 Conclusions and future challenges _________________________________________ 445 References _____________________________________________________________ 449

Introduction

T

he discovery and commercialization of asprin over 100 years ago, and the introduction of other non-steroidal anti-inflammatory drugs (NSAIDs) have had a profound impact on the practice of medicine and the treatment of the inflammatory conditions. Widespread access and over-the-counter availability of these agents has lead to the impression

that these drugs are safe and relatively void of toxicity. NSAID use can pose substantial risks to patients, especially when used chronically. Gastrointestinal (GI) complications associated with NSAID use are the most common serious adverse drug reaction reported in the United States. Additionally, aspirin is extensively used as an anti-platelet agent, as well as an analgesic agent. Aspirin, as well as other non-specific NSAID’s have a demonstrated risk of gastrointestinal hemorrhage.

OLYAEI, WHELTON, STURMER & PORTER

NSAIDs are frequently used to treat chronic inflammatory conditions and for the amelioration of acute and chronic pain. Unfortunately, to report a reliable numerical frequency to the renal functional disorders induced by non-steroidal anti-inflammatory drugs (NSAID) is next to impossible. This is due, in part, to the heterogeneity of the individuals who consume these agents and the variability in social customs that strongly influence the per capita ingestion of analgesic-anti-inflammatory drugs. Nonetheless, in most unselected populations in developed countries who seek care from their family physicians, approximately 1-3% of persons ingesting a NSAID will manifest one of a variety of renal functional abnormalities typically requiring physician intervention [1-5]. Although this percentage is relatively low, the number of “at risk” individuals are very high because of the current use-profile of NSAIDs and their availability either by prescription or as over-the-counter medications. In view of the enormous number of patients consuming these compounds, the frequency with which patients expected to develop some variety of renal functional abnormality is substantial. Over 30 billion tablets of non-steroidal anti-inflammatory drugs (NSAID) were dispensed in the United States in 2000; approximately 16% represent prescriptions for NSAIDs [1]. One in seven inhabitants of the North American (~ 50 million) is likely to be treated with an NSAID for a rheumatologic disorder in any given year [3]. In 2004 nearly 112 million prescriptions were written for NSAIDs of which almost half, 50+ million were for COX-2 inhibitors [3A]. These compounds enjoy a remarkable benefit/risk ratio when used in the treatment of acute self-limited pain syndromes. Unfortunately, when taken for prolonged periods of time, either by the elderly or individuals with certain co-morbid conditions, the frequency of adverse reactions rises dramatically. The NSAID-induced abnormalities of renal function, in descending order of clinical frequency, are (i) fluid and electrolyte disturbances; (ii) destabilization of controlled hypertension (iii) decompensated congestive heart failure; (iv) acute deterioration of renal function; (v) nephrotic syndrome with interstitial nephritis; and (vi) chronic renal failure/papillary necrosis [1, 3-5]. Most of the renal abnormalities that are clinically encountered as a result of NSAIDs can be attributed to the inhibitory action of these compounds upon pros420

taglandin production within the kidney. Hence, a brief overview of the influence of prostaglandins on renal function will be presented, followed by an analysis of the pathophysiologic mechanisms involved in the induction of renal disturbances, the clinical manifestations of these abnormalities, the patient risk factors involved and the preventive approaches to NSAID related renal syndromes.

Prostaglandins and renal function The prostaglandin pathway Renal prostaglandins serve a critical role in regulating both glomerular hemodynamics and tubular function [6]. For this process to occur, an intact arachidonic acid cascade is crucial. Prostaglandins are derived from deacylated arachidonic acid derived from cell membranes (Figure 1). The cellular release of arachidonic acid is controlled by a variety of vasoactive hormones including: norepinephrine, angiotensin, bradykinin and vasopressin [7, 8]. Once released, cyclooxygenase [COX-1 and -2] facilitates the addition of molecular oxygen to arachidonic acid creating endoperoxide PGG2. The key role that COX’s occupies in the cascade revolves around the regulation of the rate and amount of prostaglandin precursors that is converted to prostacyclin, prostaglandin and thromboxane (Figure 1). Prostaglandins are ubiquitous substances that influence renal function along with the function of other body systems [6, 8]. Prostaglandins are local hormones or ‘autocoids’ because they act in a paracrine or autocrine fashion. Biologic activity is characteristically limited to their site of production and interaction with the associated prostanoid receptors (Figure 1), the latter being responsible for activating the cellular response mechanisms. Because of the short circulatory half-life of prostaglandins, they are without significant systemic effect. In addition, prostaglandins are not stored in tissue but, rather, are synthesized on demand. Arachidonic acid can also be metabolized to a variety of mediators, depending on the cell type. For example, lipoxygenase catalyzes the production of leukotrienes, and mixed-function oxygenases catalyze the production of epoxyeicosatrienoic acids. Collectively, these oxygenated metabolites may play a critical role in NSAID-induced nephrotic syndrome by shunting arachidonic acid metabolism from prostaglandins to

18. Non-steroidal anti-inflammatory drugs

lipoxygenase products, a shift that favors production of eicosanoid, an endogenous product that increase capillary permeability [9]. Prostaglandins act as autocoids at either cortical and medullary sites of renal production [10]. Prostaglandins produced in the renal cortex modulate vascular resistance [RVR] and renin secretion, while those produced in the medulla have a major influence on salt and water balance. The major prostaglandins with renal action include: PGE2, PGI2 and TxA2. PGE2, produced in the greatest amounts, is found in both tubular and interstitial cells. Prostaglandins undergo rapid local metabolized to inactive products by a 15-prostaglandin dehydrogenase [7].

Renal prostanoid receptors Four EP receptor sub-types have been identified (table 1) [11]. Since prostaglandins are autocoids with a short half-life, interaction with specific EP receptors

Mem Diverse physical, chemical, inflammatory and mitogenic stimuli

within the nephron activates the biologic effect of PGE2. Three of the four E-prostanoid receptors, EP2, EP3, and EP4, exert their biologic effect by the coupling of G proteins to cAMP, whereas, EP1 receptor action is coupled by increasing intracellular calcium. The existence of EP2 receptor in the kidney remains to be confirmed. Breyer et al. [11] has recently reviewed the distribution of the EP receptors known to exist in the kidney. EP4, IP, and possible EP2 are located in the glomerular area. IP and EP4 probably mediate afferent arteriolar dilatation, while EP4 is involved with renin release. EP3 in the mTAL is thought to modify intense active Cl transport and reduce NaCl reabsorption. EP1, EP3, EP4 coexist in the both the cortical collecting duct (CCD) and medullary collecting duct (MCD). EP3 inhibits basolateral water reabsorption, while EP1 inhibits basolateral Na reabsorption. EP4, which is located at both the luminal and basolateral cell surfaces, stimulates water reabsorption. The relative expression

brane p ho

spholipids

Phospholipase A2

Arachidonic acid

Prostaglandin G/H synthase 1 (cyclooxygenase-1)

Prostaglandin G/H synthase 2 (cyclooxygenase-2)

Prostaglandin G2

COX

Prostaglandin G2

Prostaglandin H2

HOX

Prostaglandin H2

Tissue-specific isomerases

Prostanoids: Receptors:

Prostacyclin

Thromoboxane A2 Prostaglandin D2 Prostaglandin E2 Prostaglandin F2a

IP

TPa, TPb

DP1, DP2

Endothelium, kidney, platelets, brain

Platelets, vascular smoothmuscle cells, macrophages, kidney

Mast cells, brain, airways

EP1, EP2, EP3, EP4

FPa, FPb

Brain, kidney, Uterus, airways, vascular smooth- vascular smoothmuscle cells, muscle cells, platelets eye

Figure 1. Arachidonic acid is cleaved from membrane phospholipids by the action of phospholipase A2. The liberated arachidonic acid is then acted upon by prostaglandin G/H synthase to produce the unstable intermediate PGH2. PGH2 is converted to the multiple prostanoids shown by tissue specific isomerases. The resulting prostanoids then activate cell-membrane receptor which couple G proteins leading to the terminal effect designated in each of the boxes (with permission from [8]). 421

OLYAEI, WHELTON, STURMER & PORTER

Table 1. E-prostanoid (EP) receptor characteristics. E-receptor EP1

Function Contracts

Signal IP3/DAG/PKC

EP2

Relaxes

↑ cAMP

EP3

Contracts

↓ cAMP

EP4

Dilates

↑ cAMP

mRNA CD/musc. Mucosa Utreus Arteries CD/cTal Stomach Kidney Bladder

Adapted from Breyer et al. [14]

of each of the receptors at the various intrarenal sites will determine the extent of the biologic modulation induced by local prostaglandin production.

Cyclooxygenase isoforms Factors regulating isoform expression Two isoforms of human cyclooxygenase (COX-1 and -2), possessing similar molecular weights (70kDa) have been cloned, sequenced, and identified as being expressed in various human cells and tissues but possessing different mechanisms of regulation [6, 12]. COX-1 has been referred to as a constitutive enzyme, responsible for maintenance of normal cellular processes such as platelet function, protection of the gastrointestinal mucosa, and renal function under conditions of hemodynamic stress or decreased renal perfusion [13]. COX-2 was initially thought to be solely an inducible enzyme, activated to mediate the inflammatory response and pain perception [14]. It is now recognized that COX-2 also plays a constitutive role within the kidney [15, 16] although its specific functions have yet to be fully characterized. Nonetheless, COX-2 appears to play an important role in regulating renal salt and water homeostasis and renal hemodynamics and is induced during the inflammatory response thus contributing to the development of interstitial fibrosis [17-19]. In general, inhibition of COX-2 probably accounts for many of the desired therapeutic abilities of a NSAID while inhibition of COX-1 explains many of the undesirable gastrointestinal side effects of a NSAID. The relative proportion of COX-1/COX-2 inhibition exhibited by both non-selective and selective NSAID has become an important clinical issue since the selective inhibition of the COX-2 isoforms offers 422

the opportunity for a drug that possesses exclusive anti-arthritic therapeutic benefit without the drawback of gastric and renal side effects. Recently, FitzGerald and Patrono [8] have summarized the in vitro COX-1/ COX-2 IC-50 inhibitory actions of a variety of NSAIDs. While the whole-blood assay for COX inhibitory action has improved the predictability of a similar result in human applications, the ultimate test remains the clinical trial. In this regard several large efficacy and safety trials have been conducted using the COX-2 inhibitors, celecoxib, rofecoxib and valdecoxib, and these are reviewed in detail later in this chapter. The results of whole-blood assays identify indomethacin as the most potent inhibitor of cyclooxygenase-1, being 60 times more potent against this isoform than against cyclooxygenase-2 [20]. Aspirin was 166 times more active against cyclooxygenase-1 than cyclooxygenase-2, but was less potent than indomethacin on each of the isoforms. Acetaminophen was only a weak inhibitor of both isoforms. Some of the NSAIDs were virtually equally potent in their effects upon cyclooxygenase-1 and cyclooxygenase-2 (ibuprofen and naproxen). Of the once available COX-2 inhibitors, rofecoxib and valdecoxib are the most potent inhibitor of cyclooxygenase-2 and also demonstrated the greatest selectivity for cyclooxygenase-2 inhibition [21].

Distribution within the kidney Distribution of the COX-2 isoform in the adult human kidney is based upon in-situ hybridization and immunolocation studies [11]. COX-2 has been detected in both the macula densa and medullary interstitial cells in patients with Bartter’s syndrome and congestive heart failure [22] as well as in elderly patients. COX-1, in addition to being expressed in the glomerulus, is constitutively expressed in both the cortical and medullary collecting ducts [15, 16] (Figure 2). The exact role of the duel expression of both COX isoforms in the medullary collecting duct remains to be elucidated.

Mechanism of action of NSAIDs All NSAIDs act by inhibiting COX and thereby preventing prostaglandin synthesis[5]. The interaction between aspirin and cyclooxygenase (acetylation) is irreversible, whereas with other NSAIDs this binding is

18. Non-steroidal anti-inflammatory drugs

reversible. Traditional NSAIDs are non-selective blockers of both the COX-1 and COX-2 isoforms, whereas celecoxib, rofecoxib, lumaricoxib and valdecoxib are specific inhibitors of COX-2 [21]. The kidney is a frequent target of adverse effects from NSAIDs use [1-5, 9]. Much of this relates to the pharmacological action of NSAIDs in the presence of a stimulated endogenous prostaglandin system. NSAIDs therapeutic action derives from the 70-95% inhibition of the key regulatory enzyme COX. This inhibition has a profound effect on renal function since it eliminates the possible production of compensatory prostaglandins. This is especially true for the hemodynamically stressed individual where compensatory prostaglandin production acts to preserve renal function in the face of a systemic reduction in blood flow. Renal blood flow [RBF] is regulated by changes in RVR, which ultimately is determined by the balance between the amount of vasodilatory PGE2 and PGI2 and vasoconstrictive vasopeptides, e.g. TxA2, angiotensin II, endothelin [4, 5]. Glomerular filtration rate (GFR) also responds to these prostaglandins, increasing with PGE2 and PGI2 and declining with TxA2. Because of the reduction in RVR, which follows vasodilatation, prostaglandins can directly influence renin secretion, with PGI2, PGE2 increasing it, and TxA2 either without effect or decreasing it. Medullary salt and water regulation are strongly influenced by PGE2, which has both a natriuretic and diuretic action, while PGI2 action is limited to natriuresis [23].

Renal prostaglandin production is minimal during non-stress conditions, and thus do not play a significant role in the maintenance of renal function under normal conditions. However, their production and release are substantially increased during hemodynamic instability being called forth to preserve both glomerular perfusion and tubular function [1]. A reduction in effective blood volume initiates secretion of the various vasoconstrictive peptides, which can initiate arachidonic acid release from the membrane (Figure 1). If, during such a stimulated state, NSAIDs are administered a marked reduction in production of vasodilatory prostaglandins PGE2 and PGI2 will result in a predictable imbalance causing a decreased renal perfusion and an increased tubular sodium reabsorption. Interruption of PG’s production by NSAIDs is manifested by a variety of renal syndromes [1-3, 3-5, 24].

Renal syndromes associated with NSAIDs Several renal syndromes can complicate NSAID use [1-3, 3-5, 24]. Generally, individuals who have normal renal function and are properly hydrated, are not at risk for developing adverse renal effects [1]. NSAIDinduced deterioration in renal function depends on the specific drug, the dose and duration of pharmacologic effect and the state of health of the recipient [25]. Patients who have prostaglandin-dependent states associated with co-morbid diseases, such as high renin states or chronic renal insufficiency, are especially susceptible

COX-1

Renal medulla

COX-2

Glomerulus

Vasa recta

Figure 2. Localization of COX-1 and COX-2 immunoreactive protein in adult and fetal human kidney (reproduced with permission from [15]. 423

OLYAEI, WHELTON, STURMER & PORTER

to NSAID-induced renal toxicities. Renal prostaglandins, by initiating counterregulatory vasodilation, are crucial in maintaining perfusion in 1) individuals with parenchymal renal disease and renal impairment, and 2) when circulating volume is decreased, such as in dehydrated patients or in individuals with a decrease in their “effective” circulating volume such as CHF or significant liver disease associated with ascites [1, 3]. The renal syndromes associated with NSAIDs can be predicted based upon inhibition of COX, which modifies the compensatory actions of prostaglandins. These modifications lead to a fall in both RBF and GFR with concomitant abnormal water and electrolyte excretion [24]. In addition, nephrotic syndrome, papillary necrosis and chronic tubulo-interstitial disease can complicate NSAIDs use [2]. These syndromes are summarized on table 2.

Acute deterioration of renal function NSAID-induced acute renal deterioration occurs in the setting of severe vasoconstrictive renal ischemia and can be attributed to interruption of the delicate balance between hormonally mediated pressor mechanisms and prostaglandin-associated vasodilatory effects (Figure 3). During NSAID inhibition of renal

prostaglandin synthesis, unopposed vasoconstriction occurs by eliminating crucial counter-regulatory vasodilation [4, 5]. Similar to traditional NSAIDs, all COX-2 specific agents, celecoxib, rofecoxib, lumaricoxib and valdecoxib have been shown to reduce renal prostaglandin synthesis [19, 26-28]. High-risk individuals (Table 3) can develop AKI within days of starting traditional NSAID therapy. Fortunately, the incidence of such an event is low, ranging from 0.5% to 1.0% of patients [3]. There is an apparent association between the relatively rapid onset of AKI and ingestion of NSAIDs with short half-lives (e.g. ibuprofen) [29]. In a crossover study, involving 11 days of active treatment, renal decompensation appeared within a few days of initiation of ibuprofen therapy, whereas no evidence of AKI was reported from NSAIDs with prolonged half-lives (e.g. sulindac and piroxicam) [29]. Although NSAIDs do not reduce glomerular filtration in normal individuals [30, 31], they are capable of induce acute renal decompensation in “at risk” patients with various renal and extrarenal clinical conditions that cause a decrease in blood perfusion to the kidney (Table 3). Renal prostaglandins play an important role in the maintenance of homeostasis in these patients, so disruption of counterregulatory mechanisms can produce clinically impor-

Table 2. Effects of NSAIDs on the kidney (adapted from Whelton [125]). Syndrome

Mechanism

Risk factors

Treatment

Sodium retention and edema

↓ PG, ↓ RBF, ↓ GFR, ↑ chloride absorption

NSAID use, hepatic disease, renal disease, HTN, DM, diuretic use, circulatory compromise, dehydration, advanced age

Discontinue NSAID

Hyperkalemia

↓ PG, ↓ RAA axis activity, ↓ K+ delivery to renal tubule

Renal disease, CHF, type 2 DM, multiple myeloma, use of K+ supplements, K+ sparing diuretics, ACE inhibitors

Discontinue NSAID, avoid indomethacin in patients at risk

Acute renal failure

↓ PG, ↓ RBF, ↓ GFR Hemodynamic disruption

CHF, renal disease, hepatic disease, diuretic use, advanced age, dehydration, SLE, shock, sepsis, hyperreninemia, hyperaldosteronemia

Discontinue NSAID, support with dialysis and steroids, if needed

Proteinuria/ Interstitial nephritis*

↑ recruitment and activation of lymphocytes, likely through leukotriene formation, affecting glomerular and peritubular permeability

Fenoprofen use, possibly female gender, advanced age

Discontinue NSAID, support with dialysis and steroids, if needed

Renal papillary necrosis*

Direct toxicity ↓ PG

Massive NSAID ingestion Dehydration

Discontinue NSAID Rehydrate

Abbreviations: NSAID = nonsteroidal anti-inflammatory drugs, PG = prostaglandin, RBF = renal blood flow, GFR = glomerular filtration rate, HTN = hypertension, DM = diabetes mellitus, K+ = potassium, RAA = renin-angiotensin- aldosterone, CHF = congestive heart failure, ACE = angiotensin-converting enzyme, SLE = systemic lupus erythematosis. *: distinctly unusual

424

18. Non-steroidal anti-inflammatory drugs

At-risk patients for NSAID-induced acute renal failure

­ Renin-angiotensin axis ­ Angiotensin II

­ Adrenergic nervous system ­ Catecholamines

Table 3. “At risk” patients for NSAID-induced acute renal failure. Severe heart disease (congestive heart failure) Severe liver disease (cirrhosis) Nephrotic syndrome (low oncotic pressure) Chronic renal disease

Renal vasoconstriction ¯ Renal function

"Normalized" renal function Inhibition by NSAID Compensatory vasodilation induced by renal prostaglandin synthesis

Figure 3. Mechanism by which NSAIDs disrupt the compensatory vasodilation response of renal prostaglandins to vasoconstrictor hormones in patients with prerenal conditions.

tant and even severe deterioration in renal function [4, 5]. Typically, the addition of a NSAID increases the risk of hemodynamically mediated ischemic damage to the kidney by removing the protective effects of vasodilatory prostaglandins and allowing unopposed vasoconstriction. In the hemodynamically stressed patient a reduction in effective arterial blood volume initiates a neuroendocrine cascade, which has both renal and extrarenal consequences that require intact prostaglandin production [3]. Fortunately, the AKI usually reverses once the NSAID is stopped but in the high risk patient, can occur with any COX inhibitor. Epidemiology and Incidence Recently, Perez Gutthann et al. [32] evaluated the incidence of NSAID-induced AKI from a populationbased, case control study using data from Canada. Over 200, 000 health plan members were included because they had filled at least 1 prescription for NSAIDs during the 5-year interval. The crude incidence for AKI requiring hospitalization was 1.7/100, 000 persons. Current NSAID use increased risk of AKI 4 fold, a risk that equaled the risk associated with other known nephrotoxins, e.g. aminoglycosides, contrast media. The risk of AKI was especially high

Elderly population (age 80 or >) Dehydration (protracted - several days)

during the first month of NSAID therapy and a direct dose relationship was observed. While the incidence of AKI is low as compared to other clinical settings, the outcome is serious since nearly half the patients died. Confirmation of the increased incidence of AKI during the first 30 days of NSAID use comes from the recent publication of Schneider et al [32A]. Using a population-based, nested case-control analysis of 121, 722 new NSAID users older then 65 years of age, they identified 4228 cases of AKI and 84540 age and followup time matched controls. Current users (30 days from index date) had an adjusted rate ratio of 2.05 (CL95 1.61-2.60). The risk declined with continued use. In addition they calculated the AKI risk for specific NSAID’s. For conventional NSAID’s:2.30 (1.60-3.32); Rofecoxib: 2.31(1.73-3.08); Naproxen: 2.42 (1.52-3.85); Celecoxib: 1.54(1.14-2.09). In addition, the average incidence of AKI was 1.48 cases/100 person-years and remained stable during the study period. AKI cases were more likely to be male and have hypertension, diabetes and preexisting renal disease. Patients using more than one NSAID during the 30 day interval had a doubling of the risk ratio, 4.65 (2.31-9.37). They also noted a dose dependence with regard to AKI when higher doses of Rofecoxib, Naproxen and Celecoxib were used. The authors conclude “Compared with the other NSAIDs, celecoxib tended to have a better renal safety profile, particularily at a dose of 200 mg/day or less.” Finally, this same group of authors has published on the incidence of NASID associated acute myocardial infarction in the same cohort (162) and found that the incidence of AKI was 50% higher. The adverse effect of multiple NSAIDs on renal outcomes was confirmed by the recent report of Clinard et al [32B] who found that the Odd Ratio for adverse drug reactions involving the kidney, liver or GI tract increase between 50 and 100%. For AKI this increased from 3.2 (CL95 2.5-4.1) to 4.8 (2.6-8.8). 425

OLYAEI, WHELTON, STURMER & PORTER

Evans et al. [33] conducted a case-control study involving a population base of 420, 600 individuals searching for any relationship between AKI and NSAID ingestions. AKI was confirmed by analysis of individual hospital charts. These authors found that the risk of AKI was doubled for patients who ingested NSAIDs within 90 days of hospitalization for AKI and also for patients taking therapeutic doses of Aspirin. Interestingly, they could not identify any interaction between NSAID use in patients with chronic renal failure and subsequent hospitalization for AKI. Griffin et al. [34] reported a nested, case-control study involving 1799 AKI cases compared to 9899 controls. Chart survey was used to confirm both the diagnosis of AKI and the use of NSAIDs. 18.1% of the AKI patients were NSAID-users compared to only 11.3% of the controls (OR 1.58; 1.34-1.86). They estimated that NSAID-use was associated with 25 excess hospitalizations/10, 000 years of use. The NSAIDs with odds ratio that were significantly correlated with AKI included ibuprofen, piroxicam, fenoprofen, and indomethacin. Based on these finding and the results of three other case-control studies, they concluded that the odds ratio significantly favored a direct relationship between NSAID ingestion and AKI, especially in the elderly. The more recent case-control study is that reported by Huerta et al [34A] using the General Practice Research Database from the United Kingdom. They confirmed that current use of NSAID increased the relative risk of AKI by 3.2 fold over non-users. They also identified significant increase in the relative risk of AKI when NSAID were combined with either diuretics (RR 11.6 (CL95 4.2-32.2)) or calcium channel blockers (RR 7.8 (3.0-20.5)). In addition, they provided analysis of individual NSAID relative risk. Diclofenac 3.12 (1.38-7.05), Ibuprofen 2.64 (1.01-6.88), Meloxicam 8.05 (1.98-32.81), Naproxen 2.98 (0.62-14.21). In patients with congestive heart failure the combination of current use of NSAIDs increased their RR from 2.82 (1.05-7.57) to 7.63 (2.7-21.56), while for patients with hypertension the RR increased from 2.09 (0.87-5.02) to 6.12 (2.54-14.78). It is noteworthy that the incidence of AKI in this study, 1.1 cases/10, 000, is 100 times less than that reported by Schneider et al [32A] using a population based approach. To summarize, the risk characteristics, based upon these epidemiological studies include: 1) Patients tak426

ing NSAIDs develop AKI 2 to 4 times more frequently than non-users; 2) AKI is more common within the first month of starting NSAIDs; 3) Elderly patients and patients taking ibuprofen, diclofenac, naproxen, meloxicam, piroxicam, fenoprofen, indomethacin, and rofecoxib are at greater risk of developing AKI. 4) Patient taking more that one NSAID simultaneously are at greater risk of AKI, as are patients taking diuretics and calcium channel blockers in concert with their NSAID. However, these case control studies have methodological limitations. Limitations include: confounding by indication for the drug use, the bias introduced by difficulty in establishing the time-order of exposure, and, in some cases, the bias introduced by recall. While Evans and co-workers [33] failed to demonstrate an increased risk of NSAID-induced AKI in patients with pre-existing renal impairment, Schneider et al [32A] did report pre-existing renal failure as a risk factor and should prompt a more focused study of this selected risk category. Clinical features of NSAID-induced acute kidney injury At onset, NSAID-induced renal impairment is of moderate severity and is characterized by increasing blood urea nitrogen, serum creatinine, serum potassium, and weight with variable decrease in urinary output. With early detection and drug discontinuation, this form of NSAID-induced acute kidney injury is usually reversible over 2-7 days. Indomethacin-induced acute kidney injury may take longer to reverse following drug discontinuation, but reversal is the rule [29]. If NSAID-induced renal failure is not recognized early, severe morbid consequences occur. Continued NSAID therapy in the setting of deteriorating renal function may advance rapidly to the point wherein dialysis support is needed [35]. While this profound level of renal failure is often designated as “acute tubular necrosis”, it often is only the extreme end of the spectrum of a hemodynamic insult and probably does not deserve identification as a separate clinical entity. Fortunately, even this profound level of renal functional impairment will nonetheless recover several days to weeks after discontinuation of the NSAID. A possible relationship between the parenteral administration of the NSAID, ketorolac, and AKI has been evaluated in a multi-center study by Feldman et al. [36]. These authors found no difference in the frequency of AKI for patients receiving either ketorolac

18. Non-steroidal anti-inflammatory drugs

or morphine sulfate during the first 5 post-operative days; however, a significant, preferential increase in AKI frequency occurred when ketorolac treatment was extended beyond 5 days. Clinical risk factors for acute kidney injury A tabulation of patients at risk for NSAID induced AKI is presented on Table 3. Thus, conditions associated with reduced RBF, e.g. CHF, cirrhosis, shock, and volume contraction, triggers pressor responses via adrenergic and renin-angiotensin pathways that is referred to as the neuro-endocrine cascade. The risk of NSAID-induced acute renal deterioration is greatest in patients with liver disease, pre-existing renal impairment including the nephrotic syndrome, cardiac failure, volume contraction due to protracted intercurrent dehydration or diuretic therapy, and old age. For example, NSAID-induced renal decompensation has been well documented in patients with cirrhosis, particularly when ascites is present [7]. This sensitivity can be traced to increased urinary excretion of prostaglandin E2, prostacyclin metabolites, and thromboxane A2 in these patients [37, 38]. An analogous prostaglandin dependent renal function exists in patients with underlying congestive heart failure [39], nephrotic syndrome [40, 41], or lupus nephritis [42, 43]. A drug-drug interaction to be aware of is that combining NSAIDs with triameterene which significantly increases the risk of AKI [44]. Patients with chronic renal impairment because of diminished renal prostaglandin production may also be at increased risk of NSAID-induced renal failure. NSAID-induced acute kidney injury has been documented in patients with asymptomatic, but mild chronic renal failure, defined as a recruitment serum creatinine between 133 μmol/L and 265 μmol/L (1.5 and 3.0 mg/dl) [45]. Baseline excretion of urinary prostaglandin E2 and 6-keto-prostaglandin F1α was quantitatively lower in the individuals who developed NSAID-induced renal decompensation than in those who did not. Upon initiation of ibuprofen, urinary prostaglandin excretion fell in all patients, but trough concentrations were quantitatively lower in the subset of patients who experienced acute kidney injury. Volume contraction due to diuretic therapy or an intercurrent disease represents another important risk factor for the development of NSAID-induced acute deterioration of renal function [1, 3, 33]. Elderly pa-

tients are also at increased risk. It is estimated that, in the absence of other disease entities, the age of 80 years or greater is an independent risk factor since 50% of the population at age 80 have a 50% loss of glomerular function primarily as a result of the progression of arteriolonephrosclerosis [46]. To summarize patient a risk of NSAID-induced AKI. Frequency will be greater in patient populations with restricted renal blood flow, e.g. CHF, cirrhosis, nephrotic syndrome, shock. However, for absolute numbers, the elderly are probably most at risk since they are the primary group who take NSAIDs for relieve rheumatic complaints [3]. Hyperkalemia/Hyporenin-Hypoaldosteronism sydrome Hyperkalemia is an unusual complication of NSAID ingestion, presumably because of the multiplicity of factors that are capable of maintaining potassium homeostasis, even in the absence of prostaglandins. However, NSAID-induced hyperkalemia can occurs in up to 46% of high-risk individuals, but is reversible upon cessation of therapy [47]. Patients at risk to develop hyperkalemia include those with: pre-existing renal impairment [48, 49], cardiac failure [50], diabetes [50], multiple myeloma [51], concomitant potassium supplementation [52], potassium-sparing diuretic therapy [53] or taking an angiotensin converting enzyme inhibitor [3]. The interaction of NSAIDs with ACE inhibitors is an important and common form of drug-drug interaction. In particular, this interaction must be recognized when an arthritic patient, who is receiving long term NSAID treatment, develops hypertension that requiring drug therapy. If, in addition, the patient has mild renal impairment, our experience suggests that a baseline serum creatinine value of 180 μmol/L or greater (2.0 mg/dl or >) at least doubles their risk for NSAID related acute deterioration of renal function. In this clinical situation the angiotensin converting enzyme (ACE) inhibitor-NSAID drug combination should be avoided because of the potential for the development of both hyperkalemia and acute renal injury [3, 45]. The general interaction of NSAIDs with antihypertensive drugs will be addressed later in this chapter. Indomethacin appears to be the NSAID most frequently associated with the development of hyperkalemia, including patients without apparent risk factors [54]. In addition to the known effects of NSAIDs on potassium delivery to the distal tubule and their inhibi427

OLYAEI, WHELTON, STURMER & PORTER

tion of the renin-angiotensin and aldosterone pathways, indomethacin may have a direct effect to limit cellular uptake of potassium [55]. As noted above, hyperkalemia often complicates the NSAID-induced acute renal deterioration. However, the severity of hyperkalemia can be disproportionate to the degree of renal impairment. Tan et al. [56] have reported a patient who had a serum potassium level of 6.2 mEq/L in spite of only mildly abnormal renal function. In this patient, plasma renin and aldosterone levels were suppressed and failed to respond to furosemide or postural changes. Urinary prostaglandin E2 was also suppressed. Discontinuation of indomethacin resulted in normalization of potassium, prostaglandin E2, and a rebound of renin and aldosterone. The COX-2 inhibitor, Celecoxib, appears to have little effect on serum potassium, even in patients receiving diuretic therapy [57-59] and similarly, there does not appear to be any significant effect of rofecoxib on serum potassium. In conclusion, hyperkalemia associated with the use of traditional NSAIDs or the coxibs becomes a clinical risk in individuals with significantly decreased renal function and/or in those with the combination of decreased renal function and use of an ACE inhibitor. Pharmacodynamic and pharmacokinetic relationships in NSAID-induced acute kidney injury NSAID-induced acute renal decompensation is a pharmacologically predictable phenomenon that possesses a dose-dependent component. In a triplecrossover study of 12 females with mild renal failure, ibuprofen (800 mg three times daily) was discontinued in 3 patients after 8 days because of worsening renal function (> 133 μmol/L - > 1.5 mg/dl increase in serum creatinine) or hyperkalemia (potassium > 6 mmol/ml). When these 3 patients were rechallenged at a 50% lower dose of ibuprofen, two developed evidence of acute renal deterioration [45]. An additional important finding from this study was the time of onset of acute renal decompensation. Ibuprofen-induced renal failure occurred rapidly (within 8 days), but piroxicam and sulindac were not associated with any deterioration of renal function during the 11-day treatment period [39]. A pharmacokinetic analysis of the drugs used in these patients suggested the following: Ibuprofen, which has a short elimination half-life, reached maximum serum concen428

trations quickly; in contrast, piroxicam and sulindac have longer half-lives and continued to accumulate throughout the treatment period. These findings are consistent with basic pharmacologic principles and suggest that NSAIDs having short elimination half-lives will reach steady-state and exert maximum pharmacologic effects before this occurs with NSAIDs having longer half-lives.

Salt and water retention Electrolyte abnormalities NSAID, by inhibiting both cortical and medullary prostaglandin production, cause a variety of electrolyte abnormalities include sodium, potassium and water retention [24, 60, 61]. While sodium retention is usually transient with escape after several days, occasionally a patient will develop significant edema [62]. Water retention secondary to NSAIDs is manifest as hyponatremia [63] and occurs when the basal prostaglandin antagonism of antidiuretic hormone is removed, allowing unmodified water reabsorption in the collecting duct of the kidney. When this action is coupled with the NSAID-induced enhanced sodium chloride reabsorption in the thick ascending limb of Henle, free water clearance is virtually eliminated causing even more profound hyponatremia. Edema Formation Edema due to NSAIDs induced sodium and fluid retention usually occurs in susceptible individuals within the first week of therapy. Furthermore, these effects are reversible when the drug is discontinued. Clinically evident peripheral edema occurs in up to 5% of patients [3], likely as a result of decreased renal blood flow, possible redistribution of intrarenal blood flow, and increased reabsorption of sodium chloride in the thick ascending loop of Henle. In elderly patients this increased sodium chloride reabsorption coupled with increased water reabsorption is more likely to result in the edema. Diuretics and NSAIDs The renal saluretic response to loop diuretics is partially dependent on intact intrarenal prostaglandin production in the thick ascending loop of Henle. The decrease in the response to loop diuretics is mediated both by removing the inhibition of sodium chloride

18. Non-steroidal anti-inflammatory drugs

comparing the effect of selective COX2 inhibitors vs. reabsorption and an increase in renal medullary blood flow causing a reduction in renal concentrating capac- non-specific NSAIDs on blood pressure. They included 19 randomized control trials and used weighted mean ity. The net result is that the concurrent use of a NSAID differences and the Der Simonian and Laird method may blunt the diuresis induced by loop diuretics. of pooled results to obtain relative risk of developing For the practicing physician, this interaction is not hypertension. COX-2 inhibitors were associated with of major clinical importance since either increasing a non-significantly higher RR of 1.61 (CL95 0.91-2, 84) the diuretic dose or, if possible, discontinuation of the NSAID will permit reinstitution of the desired compared to placebo, and a non-significantly higher diuretic response. In patients who are well controlled RR of 1.25 (0.87-1.78) when compared to non-selective on a stable regimen of chronic loop diuretics use, the NSAIDs. In a recent review of analgesics and hyperintercurrent need for long term use of an NSAID will tension, Graziano [66B] concluded that while acute typically lead to increasing the dosage of the loop changes in BP following the introduction of NSAID agent, or the addition of a diuretic that acts in the therapy seem well established, the long term impact distal nephron. on NSAIDs on blood pressure is less certain. While Thiazide diuretics do not stimulate or require pros- NSAIDs antagonized the antihypertensive action of taglandins to produce their desired effect and they do b-blockers =ace-inhibitors > vasodilators > diuretics not directly interact with NSAIDs. The magnitude of = calcium channel blockers, significant alteration of increased risk of NSAID-induced AKI with concomi- body weight, daily urinary sodium excretion, creatitant triamterene cannot be estimated based on sporadic nine clearance, plasma renin activity, or the urinary case reports [44]. excretion of either PGE2 or 6-keto-PGF1 were absent. Thus, in hypertensive patients, especially the elderly, NSAIDs will interfere with antihypertensive treatment Antihypertensive drugs and NSAIDs Four recent reports have provided insight regard- especially if β-blockers, ACE inhibitors or angiotensin receptor blockers [67] are the principle drugs used. ing the interaction between antihypertensive therapy and NSAIDs. The first is a case control study by Gur- The interaction between NSAIDs and antihypertensive medications is likely due to the fact that certain witz et al. [64] involving 9411 medicare patients and antihypertensive medications exert a substantial part examined the frequency with which antihypertensive of their therapeutic effect via prostaglandin-mediated therapy was required following initiation of NSAID mechanisms [68]. Calcium channel blockers are not therapy. Based on odds ratio, NSAID users were nearly dependent on the prostaglandin pathway however, b70% more likely to require antihypertensive drugs and this requirement correlated with the NSAID dose. blockers, vasodilators, and ACE inhibitors seem to be particularly affected by NSAID therapy [4]. The lack The need for antihypertensive treatment was evident within the first 3 months of NSAID administration. of an interaction between chronic NSAID-treatment and the anti-hypertensive action of CCB’s has been Two study of interest are both meta-analysis of NSAIDs confirmed by a large prospective study [69]. Regarding effect on blood pressure. Pope et al. [65] included 54 short-term studies encompassing 1324 patients, 92% the likelihood of developing new onset hypertension with COX2 inhibitors vs. NSAIDs, Wang et al [69A]) being hypertensive. The adverse influence of NSAIDs conducted a cohort study using secondary data from on blood pressure (3.5 mm Hg – 6.2 mm Hg increase) was limited to hypertensive patient taking indometh- GE Centricity Electronic Medical Record database involving a sample of 51444 patients of whom 17148 acin, naproxen or piroxicam. These authors could not eliminate the confounding effect of dietary sodium. were receiving celecoxib and 34296 were receiving non-specific NSAIDs. Relative to non-specific NSAIDs The meta-analysis reported by Johnson et al. [66] users, celecoxib users had a similar rate of post-expoincluded 50 clinical trials encompassing 771 patients only 80% of who were hypertensive. NSAID adminis- sure hypertension with the hazard rate of 1.013 (CL95 tration resulted in a mean increase in blood pressure 0.862-1.190). Thus there was no difference in the risk of of 5 mmHg in the hypertensive patients, but no sig- developing new onset hypertension when comparing nificant increase in the normotensive patients. A more specific COX2 inhibitors with traditional NSAIDs. recent meta-analysis was conducted by Aw et al [66A] 429

OLYAEI, WHELTON, STURMER & PORTER

Proposed mechanism of blood pressure destabilization Prostaglandins, in concert with nitric oxide, act as a renal vasodilator-natriuretic system [70] whose action is to offset the vasoconstrictive-sodium retaining effects of the renin-angiotensin system. Because of these interactions, significant destabilization of blood pressure control can occur during systemic administration of NSAIDs. PGE2 and PGI2 possess both prohypertensive and antihypertensive actions on blood pressure [68]. The prohypertensive actions involve increasing renin release and raising cardiac output. The antihypertensive action includes vasodilatation, reversing vasopeptide-induced vasoconstriction and inducing a negative sodium balance. Recent evidence has identified a decline in nitric oxide availability in both elderly [71] and hypertensive patients [72]. In addition, the plasma nitric oxide response to alterations in dietary sodium intake is distinctly abnormal in elderly salt-sensitive hypertensive patients [73]. When these observations are combined with the recent studies of Perinotto et al. [74], which confirmed that endogenous prostaglandin will counteract the renal actions of endogenous angiotensin II in the face of NO inhibition, a mechanistic explanation for NSAIDs-induced hypertension can be formulated. The speculation involves the following: The decline in NO production in elderly, hypertensive patients puts additional requirement on the endogenous renal PG to counteract the intrinsic action of the RAS. When NSAIDs are given, the vasodilator-natriuretic action of PG is removed and thus the RAS is unopposed leading to destabilization of BP control. The interaction between sodium intake, blood pressure and NSAIDs has been studied by Mulkerrin et al. [75]. These authors measured the change in blood pressure and sodium excretion in five young normotensive females and five elderly females to an intravenous saline load before and after 1800 mg of ibuprofen was given for 3 days. Saline loading induced a consistent 25mmHg rise in systolic pressure with or without ibuprofen in the elderly patients, while ibuprofen alone caused a 14mmHg rise from baseline in the elderly patients before saline. The naturiesis associated with saline loading in both groups was significantly blunted in the elderly after treatment with ibuprofen. They concluded that aging increases the susceptibility to salt retention and hypertension from NSAIDs. This may well be due to unmasking the diminished activity 430

of nitrous oxide synthetase, which characterizes elderly patients who are salt sensitive. Alam et al. [76] used chronic salt loading to examine the interaction between blood pressure, salt and NSAIDs. Thirty-one healthy individuals, age 60 or more, were enrolled in a randomized, placebo-controlled, crossover study. Patients were stratified as to normotensive or isolated systolic hypertension based on their blood pressure response after 6 weeks of a controlled 150 mEq/d sodium diet. Crossover involved a two-week interval receiving either low sodium (90 mEq/d) diet or high sodium diet (240 mEq/d) diet and placebo or indomethacin 75 mg/d. For all patients, high salt diet was associated with a 6mmHg rise in systolic pressure and 3 mmHg in diastolic pressure. Indomethacin administration increased systolic but not diastolic pressure. High salt diet and indomethacin had an additive effect on blood pressure, but failed to demonstrate any interaction. Indomethacin significantly elevated the blood pressure in normotensive individuals but did not in patients classified as saltsensitive. They concluded that salt-sensitive patients with isolated systolic hypertension were resistant to the pressor effect of indomethacin but normotensive elderly patients were not. The duration and magnitude of salt loading between these two studies may account for the different conclusion reached by each set of authors.

The concept of “renal sparing” NSAIDs While all NSAIDs have the potential for inducing renal failure, there has been speculation of quantitative differences among the individual NSAIDs. Sulindac was thought to be renal sparing, possibly because of its unusual metabolic pathway [29, 77-79]. The parent compound, sulindac sulphoxide, is an inactive prodrug, which undergoes hepatic metabolism to sulindac sulphide, the metabolite responsible for its anti-inflammatory activity. Sulindac sulphoxide is also metabolized to a much lesser extent to an inactive metabolite, sulindac sulphone. It was hypothesized that, within the human kidney, sulindac sulphide was reversibly oxidized to the inactive parent compound, sulindac sulphoxide, with the result that renal prostaglandin production was not perturbed [29, 78]. In clinical studies, urinary prostaglandin levels and renal effects were unchanged in patients with normal

18. Non-steroidal anti-inflammatory drugs

renal function [29, 30] and patients with proteinuria [79]. However, the duration of sulindac exposure in these studies may not have been sufficient to allow the full pharmacologic effect of sulindac. Also, NSAID-induced changes may not have been detectable because of the presence of only very mild renal impairment or the absence of co-existing renal failure in this study [80]. Longer courses of sulindac in patients with slightly more severe renal impairment have been associated with statistically significant reductions in urinary prostaglandins [45, 80] and GFR [81]. The ability of sulindac to inhibit prostaglandin synthesis and impair renal function has been confirmed in a different high-risk group, namely patients with hepatic cirrhosis and ascites [82]. We have also identified the development of profound acute kidney injury in risk prone patients who received sulindac for several days to weeks. Collectively, these studies suggest caution in accepting any NSAID as being “renal sparing”.

Nephrotic syndrome with interstitial nephritis This rare complication of NSAID use may develop at any time during treatment, but typically occurs months to years after therapy has been initiated, and generally resolves upon discontinuation of therapy [1, 83]. Fenoprofen, on a per capita use basis, has been associated with interstitial nephritis more frequently than other traditional NSAIDs [83, 84]. To date, there have been three reports of coxib-induced acute interstitial nephritis [85-87]. All were biopsy proven and cleared after stopping the coxib. Clinical presentation The features of this NSAID-induced renal syndrome are somewhat variable. The patient may experience edema, oliguria, and clinical signs indicative of significant proteinuria [88, 89]. Systemic signs of allergic interstitial nephritis such as fever, drug rash, peripheral eosinophilia, and eosinophiluria are typically absent. The urine sediment contains microscopic hematuria and cellular elements reported as pyuria [9, 89]. In a recent discussion of NSAID-induced acute interstitial nephritis, Rossert [83] confirmed that proteinuria, usually in the nephrotic range, occurs in 70% of cases [84]. The occurrence of acute kidney injury

parallels the nephrotic syndrome. For patients without the nephritic syndrome the functional extent of renal deterioration can range from minimal to requiring hemodialysis. The onset of NSAID-induced nephrotic syndrome is usually delayed, having a mean time of onset of 5.4 months after initiation of NSAID therapy and ranging from 2 weeks to 18 months [9, 88]. NSAIDinduced nephrotic syndrome is usually reversible between 1 month and 1 year after discontinuation of NSAID therapy. During the recovery period, some 20% of patients require dialysis. Steroids have been used empirically, but it is not certain that they hasten recovery. If proteinuria is not significantly reduced within two weeks of discontinuation of the putative NSAID, we recommend a standard 2-month trial of corticosteroids as would be employed in an adult patient with idiopathic minimal change glomerulonephritis. While pyuria and eosinophiluria develop in ~40% of patients who present with nephrotic syndrome, gross hematuria occurs in less than 10% of patients [83]. Histologic features of NSAID-induced nephrotic syndrome NSAID-induced acute interstitial nephritis is a recognized cause of AKI [17], the frequency of which appears to be increasing [90]. In a recent series reported by Schwarz [90] of 64 biopsy-proven cases of acute interstitial nephritis, 85% were drug induced. The responsible drugs included: antibiotics, analgesics, NSAIDs and diuretics. Characteristically, the histology of this form of NSAID-induced nephrotic syndrome consists of minimal change glomerulonephritis with tubulointerstitial nephritis. This is an unusual combination of findings and, when noted in the clinical setting of protracted NSAID use, is virtually pathognomic of NSAID-related nephrotic syndrome. Nephrotic syndrome without apparent interstitial disease has been reported in a handful of patients taking fenoprofen, sulindac, or diclofenac. Conversely, interstitial disease without nephrosis has been reported in a few patients, but this may, possibly, represent allergic interstitial nephritis [89]. In spite of the nephrotic range proteinuria, the most impressive histopathologic findings in NSAID-induced nephrotic syndrome involve the interstitium and tubules [91]. A focally, diffuse inflammatory infiltrate can be found around the proximal and distal tubules. While this infiltrate consists primarily of cytotoxic T lymphocytes, it also contains other T cells, some B 431

OLYAEI, WHELTON, STURMER & PORTER

cells, and plasma cells [92]. Changes in the glomeruli in these patients were minimal and resembled those of classic minimal change glomerulonephritis with marked epithelial-foot process fusion. These findings are consistent with reports by other investigators [8, 35, 36, 93]. Of the 14 cases of biopsy proven drug-induced allergic nephritis reported by Shibasaki et al. [94], 4 were ascribed to NSAIDs; while in the series reported by Schwarz et al. [90], 16 of 68 biopsies were ascribed to NSAIDs. While 3 of the patients had taken the offending agent for less than 1 week, the forth had received aspirin for 3 months. All presented with oliguric renal failure without systemic signs of rash or fever. Positive 67Ga scintigrams were obtained in both patients in whom it was performed. In 3 of the 4 patients serum creatinine returned to normal range at follow-up. The authors conclude that 67Ga scintigram combined with a lymphocyte stimulation test can confirm a diagnosis of suspect drug-induced allergic nephritis without resorting to a renal biopsy. NSAID induced nephrotic syndrome is suspected of being immunologically mediated and idiosyncratic. It has a distinct presentation when compared to that ascribed to acute interstitial nephritis. The nephrotic syndrome is not associated with hemodynamically stressed patients. Recently Radford et al. [95] published a retrospective study of NSAIDs induced membranous nephropathy using the Mayo Clinic biopsy registry. They reported that >10% of biopsy proven membranous glomerulonephritis [stage I/II] was attributable to NSAIDs. They summarized the clinical features of NSAID-induced nephrotic syndrome as having no consistent clinical predisposition, with a median duration of 43 weeks of drug ingestion, and nephrotic range proteinuria that was present for 4 hours) had a significantly greater prevalence of impaired renal function. Interestingly, diuretics were associated with a significant incidence of renal impairment, OR 3.5 (95% CI 1.6 to 7.6), but no additional interaction with NSAIDs could be identified [100]. No significant interaction between ACE inhibitors and NSAIDs was evident from their data. The authors concluded that elderly patients taking long half-life NSAIDs are at increased risk for impaired renal function. While the information to date is suggestive of an association between high dose and/or long duration NSAID use and ESRD, additional epidemiological studies are needed [99]. Three observational studies using national cohorts have examimed the risk of deteriorating renal function with chronic NSAID exposure. Two of the reports involvedata from the Physician Health Study [99A, 99B] while the third involved the nurses from the National Health Survey [99C]. After adjusting for confounding risk factors none of the studies identified an increase risk of loss of renal function with moderate NSAID intake. More recently Ibanez et al [99D] conducted a 2 year case control study regarding the relative risk of ESRD associated with NSAIDs. They reported a Odd Ratio of 1.22 (CL95 0.89-1.66) for the risk of ESRD associated with NSAIDs. Calvo-Alen et al. [101] evaluated creatinine clearance, osmolar clearance, free water clearance, sodium excretion and urinalysis in 104 arthritic patients whose treatment with NSAIDs exceeded 2 years compared to 123 health controls. The major abnormal finding was restrict to impaired renal concentrating capacity

in the arthritic patients as manifested by a decreased osmolar clearance, increased free water clearance and a decreased urinary density. Compared to controls, no significant differences in either sodium excretion or creatinine clearance were recorded. However, Murray et al. [25] determined the incidence and risk factors for ibuprofen-associated renal impairment by analyzing 1908 computerized patient records. Multivariable analysis of the 343 patient records with renal impairment identified: age, prior renal insufficiency, coronary artery disease, male gender, elevated systolic blood pressure and diuretic use as risk factors. Only two subsets of at risk patients, age > 65 and coronary artery disease, were at greater risk to develop renal insufficiency when compared to acetaminophen. The observation by Schwarz et al. [90] are germane to the influence of NSAID-induced AKI on the development of chronic renal failure. While NSAIDs accounted for only 20% of the cases of acute interstitial nephritis [90], nearly 2 out of every 3 patients from the NSAID subgroup was found to have permanent renal impairment at follow-up which represented the greatest frequency of any of the drug-induced acute interstitial nephritis. Papillary necrosis In a prospective study by Segasothy et al. [102] conducted over 11 years, IVP confirmed NSAID-induced papillary necrosis was reported to occur in 27% of heavy analgesic users. In over half of the cases (55%), the offending analgesic was excess NSAIDs consumption more often a single type rather than multiple agents. In over 80% of the cases the NSAID was prescribed for an arthritic condition, with male: female ratio of 1.9:1. Coexisting additive behavior was rare in the patients include in this study. Because of the wide differences in relative risk noted in these limited studies, plus questions that have been raised as to their validity [9], a precise risk cannot be stated. Papillary necrosis is the least common type of NSAIDinduced renal toxicity, but unlike the other types, it is irreversible. Volume depleted patients who ingest large quantities of NSAIDs may be at higher risk for developing papillary necrosis and parenchymal damage is permanent [103-105]. Its cause is likely a combination of decreased renal papillary perfusion and excessive papillary parenchymal NSAID and NSAID-metabolite concentrations. 433

OLYAEI, WHELTON, STURMER & PORTER

Definition and differentiation of acute versus chronic papillary necrosis lesion By definition, papillary necrosis represents the development of irreversible damage within the parenchyma of the renal papillae. The papillae of the kidney contain the tip portions of the long loops of Henle, together with the terminal portions of the collecting duct complexes, which open in to the minor calyces. The minor calyces of the kidneys representing the first location in the upper renal outflow tract into which urine is collected before it travels into the renal pelvis and into the urinary bladder via the ureters. The mechanism of NSAID-induced acute papillary necrosis is often not clear and the causative role of the NSAID in question may be difficult to delineate because of the presence of confounding factors such as underlying disease, urinary tract infection, and/or concomitant medications. Selected NSAIDs may exert a direct toxic effect on renal papillae and may become highly concentrated in the medullary-papillary region of the kidney. Aspirin depletes cellular glutathione, which would otherwise neutralize the acetaminophen metabolite, N-acetyl-benzo-quinoneimine. Without glutathione, this highly reactive metabolite could lead to cell death [106]. Prostaglandin inhibition may also play a role [9]. Medullary ischemia, a possible precipitating factor in development of papillary necrosis, results from NSAID-induced reduction of blood into the renal medulla in experimental models [107, 108]. The development of acute papillary necrosis, as a consequence of the use of a single NSAID, at recommended dosing levels, is an extremely rare event. In preclinical studies, nearly all of the NSAIDs produced papillary necrosis in experimental animal models. Although, as already identified, clinical toxicity is exceedingly rare it has been reported for ibuprofen [103], phenylbutazone [109, 110], fenoprofen [105], and mefenamic acid [104] and, according to prescribing information, several other NSAIDs. The chronic progression of events that lead to NSAID/analgesic related papillary necrosis are well known since the days of the first descriptions of chronic combined analgesics abuse nephropathy and the subsequent extensive investigations which defined the consequences of chronic (5-20 years) exposure of the kidney to high doses of analgesic combinations such as salicylate and acetaminophen (the metabolite of phenacetin) often with the addition of caffeine [106]. 434

Fortunately, the incidence of this form of chronic analgesic abuse nephropathy has diminished because of a better understanding of the drugs involved, patient education, and in some countries thanks to efficient regulatory measures. The topic of chronic papillary necrosis related to analgesic-NSAID mixtures is reviewed in detail elsewhere in the text and will not be further discussed here. The clinical circumstances that lead to chronic “analgesic abuse” nephropathy [111] are quite distinct to the rare occurrence of acute papillary necrosis associated with exposure of the patient to a single NSAID and often with only a short period of drug exposure. In these acute circumstances, the patient will typically present clinically with gross hematuria and may have flank pain suggestive of ureteric obstruction consequent to the passage of a sloughed papilla.

Other NSAID-induced renal syndromes Phenylbutazone, suprofen, and benoxaprofen produce unique renal syndromes that are of historic interest. Fortunately, the use of phenylbutazone use has diminished because of the availability of safer drugs, and suprofen and benoxaprofen have been voluntarily removed from the market. Two mechanisms responsible for phenylbutazoneinduced acute oligo-anuric renal failure include: 1) inhibition of uric acid reabsorption, leading to hyperuricosuria and, ultimately, bilateral ureteral obstruction due to uric acid stones [112]; 2) an idiosyncratic reaction has been reported that results in acute tubular injury without uric acid precipitation [113]. Suprofen-induced AKI is characterized by acute flank and/or abdominal pain. In series of 16 patients, Hart et al. [114] described that the mean peak serum creatinine was 3.6 mg/dl (range: 2 to 8 mg/dl), which returned to normal limits at follow-up. Suprofen is know to have uricosuric activity leading Hart and colleagues [114] to suggest that this renal syndrome may have resulted from ureteral or tubular precipitation of uric acid. Benoxaprofen, an NSAID with a very long halflife, was removed from the market in the early 1980s because of severe hepatic toxicity that occasionally resulted in death; however, renal failure was a contributing factor. Risk factors for benoxaprofen-induced toxicity were old age, concomitant diuretic therapy,

18. Non-steroidal anti-inflammatory drugs

and likely excessive drug administration.

Renal effects of COX-2 inhibitors Vane published the seminal work on the mechanism of action of aspirin-like drugs in the early 1970’s [115]. Since that time, the goal in NSAID research has been to formulate agents with increased potency and limited toxicity. The elderly comprise the majority of patients who use high doses of NSAIDs for their analgesic and anti-inflammatory effects. However, the gastrointestinal toxic effects of the traditional NSAIDs and underlying disease states, such as hypertension and congestive heart failure (CHF), may preclude their use. Hence, these agents must be used cautiously in this population. The most recent advance in NSAID pharmacology are agents that specifically block the cyclooxygenase2 (COX-2) isoform while sparing the effect of COX-1 related activities [116] (Figure 4). These drugs have been designated by the WHO as a new pharmacology category of NSAIDs, namely the ‘coxibs’ [117]. By blocking COX-2, the intent is to spare toxicity in organs such as the gut and kidney, thereby increase their utility, especially in elderly patients. All of the currently available COX-2 specific inhibitors, i.e. celecoxib, rofecoxib and valdecoxib, have established their safety advantage with respect to clinically important reductions in gastrointestinal toxicity and platelet-sparing characteristics [28, 116-123]. Bleeding complications as seen with aspirin and traditional NSAIDs have essentially been eliminated. However, the clinical impact of the coxibs upon renal and cardiovascular function is an area of evolving information, especially now that it is known that the COX-2 isoenzyme is expressed within the human kidney [15, 17, 124]. The nephrotoxic effects of traditional NSAIDs are well recognized and have been the subject of extensive reviews [4, 125].

weight, blood pressure and the urinary metabolites of thromboxane. Patients were randomized to either rofecoxib, indomethacin or placebo while receiving a isocaloric diet containing 200 mEq of sodium. Both NSAIDs induced a significant, but transient, decrease in sodium excretion during the first 72 hours of ingestion. Following this sodium excretion returned to pretreatment levels despite continued administration of the NSAIDs. Only indomethacin caused a significant reduction in GFR after 14 days of treatment. Body weight and blood pressure did not change significantly for any of the treatment groups. Inhibition of platelet thromboxane synthesis was limited to indomethacin treated patients, while both rofecoxib and indomethacin were associated with a significant reduction in urinary excretion of the prostacyclin metabolite, 2, 3-dinor-6-keto prostaglandin F1 . Because of the later finding, the possibility of a prothrombotic state resulting from the administration of coxibs was suggested by these authors. The basis for the speculation is as follows: Activation of platelet aggregation is thromboxane dependent and under the control of the COX-1 isoform. Production of prostacyclin by vascular endothelium is a COX-2 dependent step. By inhibiting the production of prostacyclin, a anti-platelet aggregation factor, this would leave thromboxane mediated platelet aggregation unopposed and could result in a prothrombotic state. However, the anti-platelet aggregation action of endothelin would be unaffected during COX-2 inhibition [8]. Physiologic stimulus Inhibition by NSAID

COX-1 constitutive Stomach Intestine

Kidney Platelets

Endothelium

Effects on renal function: GFR/ urinary sodium excretion The effect of COX-2 specific inhibitors on renal function, including sodium excretion, has been assessed in prostaglandin dependent patients. Catella-Lawson et al. [26] enrolled 36 healthy elderly patients for her study, which evaluated not only sodium excretion and glomerular filtration rates, but also changes in body

Inflammatory stimuli

PGE2

TxA2

PGI2

Physiologic functions

COX-2 inducible Inflammatory sites (macrophages, synoviocytes)

Inflamma•tory PGs Proteases O2

Inflammation

Figure 4. Different functions of COX-1 and COX-2 in prostag435

OLYAEI, WHELTON, STURMER & PORTER

Rossat et al. [126] conducted their renal assessment of COX-2 inhibition using health male volunteers rendered prostaglandin-dependent by a combination of low salt diet and administration of a loop diuretic. Their trial consisted of a parallel, randomized study involving giving either celecoxib, 200 mg bid or 400 mg bid, naproxen 500 mg bid or placebo for seven days. Blood pressure, renal hemodynamics, urinary salt and water excretion were measured before and 3 hours after ingestion of the test drug. The urinary excretion of sodium, potassium, lithium, and water were significantly decreased on both day one and day seven at the second and third hour after administration of either celecoxib or naproxen. Accumulative sodium excretion was significantly reduced during the first 3 days of NSAID dosing, but then subjects escaped from the effect. Glomerular filtration rates were transiently decreased following the 800 mg dose of celecoxib on day one, and the naproxen dose on day seven. These same authors confirmed that lack of an effect of celecoxib on platelet thromboxine synthesis. They concluded that COX-2 inhibition in salt-depleted subjected induced retention of sodium and potassium. Whelton et al. [19] enrolled 29 healthy elderly individuals in a single blind, randomized, cross-over study to determine the effects of celecoxib on prostaglandin dependent renal function. Either celecoxib or naproxen was given for 10 days followed by a 7 day washout period and then 10 days of the alternate drug. During the first 5 days, celecoxib 200 mg bid was given, then the dose was increased to 400 mg bid for the final 5 days of the trial. Naproxen dose as 500 mg bid throughout the 10 days. Only the –7.5 ml/min/1.73 m2 decrease in GFR on day 6 of naproxen proved to be significant. Transient sodium retention was noted with both celecoxib and naproxen treatments, returning to baseline within the first 3 days. Both NSAIDs caused a significant reduction in urinary PGE2 and 6-keto-PGF-1 throughout the 10 days of administration. The authors concluded that, like conventional NSAIDs, celecoxib effects the urinary excretion of both sodium and prostaglandin E2. However, in elderly patients, unlike conventional NSAIDs, celecoxib spares renal hemodynamics. Swan et al. [127] conducted a multi-center that involved both a randomized, single-dose crossover study and a randomized, parallel group, multidose study involving elderly, salt-depleted subjects. The single dose study involved 15 subjects who where crossed 436

over between rofecoxib 250 mg or indomethicin 75 mg. For the multidose trial 60 subjects received either rofecoxib 12.5 or 25 mg/d, indomethacin 150 mg/d, or placebo for 6 days, with measurement performed during the last 6 hours of study day 6. Peak GFR, measured by either inulin or iothalamate clearance, fell by nearly 40% following acute administration of either rofecoxib or indomethacin. For the multidose trial, the reduction in GFR, while still significant, was less than 10%. While sodium excretion was reduced by both drug following acute administration, only 12.5 mg rofecoxib was associated with significant sodium retention after 6 days of drug. These author concluded that the effects of rofecoxib on renal function resembled nonselective NSAIDs and that COX-2 plays an important role in human renal function. Collectively, these studies suggest that COX-2 plays a dominate role in the regulation of salt and water excretion in prostaglandin dependent patient, while the role of COX-1 seems to involve the regulation of renal hemodynamics, including GFR. The Swan et al. [127] study suggests that COX-2 may also play a role in regulating GFR; however, the combination of elderly patients who are salt depleted may have provided a more severe hemodynamic stress than was present in the other three studies.

Incidence of adverse cardio-renal events Serum electrolytes and creatinine In the recent 8000-patient celecoxib long-term arthritis safety study [118], significantly more patients receiving traditional NSAIDs (ibuprofen or diclofenac) experienced clinically significant elevations in serum creatinine and/or serum urea nitrogen levels when compared to celecoxib. This was confirmed in a followup study using the same data base [118A]. In patients defined as having pre-existing uremia, when these patients received either diclofenac or ibuprofen, the had significantly greater increases in serum creatinine than patients receiving celecoxib. In an equally large gastrointestinal safety trial with rofecoxib, the incidence of adverse effects related to renal function for rofecoxib was similar to naproxen (1.2% versus 0.9%, respectively) [119]. When rofecoxib and celecoxib were directly evaluated in elderly hypertensive OA patients who manifested “normal” serum creatinine at the time of study recruitment, the overall incidence of clinically

18. Non-steroidal anti-inflammatory drugs

significant increases in serum creatinine, blood urea nitrogen, and serum potassium was 1.5% for both agents [128]. In post-marketing surveillance, AKI has been reported for both coxib compounds. Uniformly, this complication has been reported in patients with significant pre-existing renal impairment (serum creatinine ≥ 3.0 mg/dl [250 mmol/L] prior to coxib treatment). Details of 4 cases of AKI associated with COX-2 specific inhibitor use have been reported in the literature [129, 130]. In each of these cases, creatinine clearances returned to baseline after cessation of COX-2 specific inhibitor therapy. Ahmad et al [131], reported 264 cases of renal failure due to either celecoxib or rofecoxib based on voluntary reported submitted to the FDA AER system. 122 cases occurred with celecoxib and 142 with rofecoxib. Hypertension, diabetes, congestive heart failure and renal insufficiency were shared risk factors for both drugs. However, concomitant use of diuretics, ACE inhibitors and other NSAID’s occurred more frequently in patients with renal failure attributed to rofecoxib. No correlation with dose was evident. In of the 122 cases of celecoxib associated renal failure initial renal function was normal, while in 12 of 142 cases of rofecoxib initial renal function was reported to be normal. Zhao et al [132] compared the renal-related adverse drug reactions between rofecoxib and celecoxib as reported to WHO Safety Monitoring Center. The center uses a statistical parameter, e.g. information component (IC), from a Bayesian confidence propagation neural network method to calculate each drug-ADR combination. When the IC values for rofecoxib were compared to celecoxib, an statistically significant adverse renal impact of rofecoxib was present for: water retention (R 1.97 vs. C 1.18, p10 per high power field are present. For patients whose disease has improved but who developed proteinuria of between 300 to 1, 000 mg/day but without other renal abnormality, they suggest continuing the drug cautiously at a reduced dose with close monitoring. If the proteinuria exceeds 2 g/day or the glomerular filtration rate falls, the drug should be discontinued immediately [65]. Manthorpe et al. reported a successful one year treatment with auranofin (6 mg/day) in 7 rheumatoid arthritis patients with previous proteinuria associated with parenterally injected gold salts [66].

Prediction, prevention and monitoring of development of gold nephropathy To predict the adverse effects of gold, the association with HLA antigen has been studied [27, 28, 67-69]. A genetic predisposition to gold toxicity was first suggested by Panayi et al. [67]. Wooley et al. [68] investigated the possible relation between HLA antigens and toxicity of D-penicillamine and sodium aurothiomalate in rheumatoid arthritis patients. Nineteen of 24 patients in whom proteinuria developed were positive for HLAB8 and DRw3 antigens. Furthermore, all 13 episodes of proteinuria exceeding 2 g/day occurred in patients with DRw3. Several investigators confirmed the association between gold-induced proteinuria and DR3 [27-30] and B8 [30], but others were unable to confirm it [70]. Conversely, DR3 patients tended to exhibit a better therapeutic response to sodium aurothiomalate than patients with DR4 [28]. DR4 and/or DR2 positive patients may have some degree of protection against gold toxicity [28, 29]. Given the uncertainty about HLA 463

UEDA & PORTER

types and toxic reactions, together with the suggestion that patients with DR3 respond better than the more numerous DR4, and taking into account the cost involved, any suggestion of using HLA typing as a guide to therapy seems premature [71]. While Van Riel et al. [72] reported the predictive value of serum IgA for gold toxicity, the study of Ostuni et al., involving a larger population, concluded that the monitoring of serum IgA was not useful in predicting gold toxicity [73]. Recently, Ayesh et al. [74] reported the predictive efficacy of the prior measurement of sulphoxidation capacity. A patient with poor sulphoxidation capacity had a nine-fold greater risk of developing gold-induced adverse reactions including nephropathy. Hopefully this will be confirmed by prospective studies involving various races and a large population. To date, there is no confirmed method for predicting gold toxicity including nephropathy, thus it is essential to monitor patients closely for any appearance of nephropathy. However, Shah et al [74a] have evaluated the association between gold ADR’s (thrombocytopenia or proteinuria) and HLA-DR3 status. Based on a cohort of 41 patients they concluded that patients with nodular disease were more likely to develop ARDs (51.3% vs. 25.6%, OR= 3.0, p=0.02 and also more likely to be HLADR3 positive (41.2% vs. 17.6%, OR= 3.0, p= 0.045. The authors suggest that nodular patients with HLA-DR3 should not receive parenteral gold as their primary treatment for RA. The decline in the number of reports of parenterally administered gold-induced nephropathy may indicate that the dose of gold salts used per injection is decreased and intervals between injections are being extended to prevent adverse reactions. Furthermore, introduction of methotrexate therapy, along with several biological agents, for rheumatoid arthritis, has contributed to decreased reliance on gold salts. However, intriguing reports using nanotechnology gold in treatment of malignancies has renewed interest in gold as a therapeutic agent [74b, c].

Auranofin nephropathy Auranofin, a unique gold compound, has been available for clinical use for 25 years after it proved to be one of the most potent oral antiarthritic com-

464

pounds among alkylphosphine gold coordination complexes [75]. Initial clinical studies suggested that this compound was therapeutically active when taken by mouth, with no renal adverse effects in any of the 32 patients studied [5-7]. Subsequently, the therapeutic benefits and toxicity of auranofin have been evaluated [24, 76], compared with placebo [9, 77, 78], sodium aurothiomalate [8-10, 79], and D-penicillamine [8082]. The incidence of proteinuria in a world-wide trial was 3% for auranofin [10, 24]. The risk of developing proteinuria with auranofin therapy is significantly less than with parenteral gold [9, 24], or D-penicillamine [82]. Histopathological findings in renal biopsy specimens from patients with moderate to heavy proteinuria are consistent with the membranous nephropathy similar to injectable gold nephropathy [33, 83, 84]. Heuer et al. [10] reported a total of 3, 475 rheumatoid arthritis patients receiving auranofin therapy in 27 countries. Proteinuria developed in 3% of the patients, resulting in drug withdrawal in 0.9%, compared with 4% proteinuria in patients receiving injectable gold, with 0.8% being withdrawn. Katz et al. [24] evaluated proteinuria in 1800 rheumatoid arthritis patients given chrysotherapy. Three percent (41 cases) of 1283 auranofin-treated patients had an abnormal 24-hour urine protein level: 15 had mild (0.15 to 1 g/day), 17 had moderate (1 to 3.5 g/day), and 9 had heavy (>3.5 g/day) proteinuria. Permanent renal impairment did not occur in any patient. In 36 patients with long-term follow-up after drug withdrawal, proteinuria cleared in 31 patients within 1 week to 24 months. Seven of 8 patients who were rechallenged once the proteinuria had cleared were able to continue treatment without recurrent episodes [24]. Pathogenic mechanism of auranofin-induced nephropathy resemble those of parenteral gold-induced nephropathy. The reason for the reduced risk of proteinuria with auranofin compared to parenteral gold salts is not known. However, differences in the pharmacokinetics of the two types of gold preparations may be important. In rats treated with auranofin or sodium aurothiomalate for one year, renal gold concentrations were 33 times higher with the latter formulation [85]. Renal elimination of an orally administered dose of auranofin in human is less than 15%, compared with greater than 70% for parenterally administered sodium aurothiomalate [86].

19. Gold salts, D-penicillamine and allopurinol

D-penicillamine Introduction D-penicillamine is so named because it was first isolated as an amine, from the degradation products of penicillin by Abraham et al [87]. Later studies showed the characteristic chemical behavior of D-penicillamine which involves three types of reactions, formation of disulphide links, formation of thiazolidine rings, and formation of metal complexes and chelates [67]. It was first used in 1956 in the treatment of Wilson’s disease [88]. D-penicillamine has since been used in the treatment of many diseases, such as cystinuria [89], rheumatoid arthritis [90-92], systemic sclerosis [93], primary biliary cirrhosis [94], heavy metal poisoning due to lead [95], cadmium [96], and mercury [97], and hyperviscosity syndrome [99]. In rheumatoid arthritis, D-penicillamine has been widely accepted as an effective second line treatment. Despite of its effectiveness, it causes many adverse effects, such as skin rashes [99, 100], taste abnormalities [100, 101], hepatic dysfunction [102-104], gastrointestinal toxicity [99, 105], proteinuria [100, 106], hematuria [107, 108], thrombocytopenia [92, 109], aplastic anemia [110], lupus-like syndrome [111, 112], Goodpasture’s-like pulmonary renal syndrome [113115], vasculitis [116, 117], myasthenia gravis [118-122], polymyositis [123, 124], and dermatomyositis [125]. One or more of these adverse reactions was recorded in nearly 60% of patients treated with D-penicillamine [100, 126-129]. Among these adverse reactions, nephropathy developed in patients with proteinuria, hematuria, lupus-like syndrome, Goodpasture’s-like pulmonary renal syndrome, and vasculitis.

Proteinuria Proteinuria, including nephritic syndrome, is the commonest manifestation of nephropathy, reported as occurring in between 2 and 32% of patients [100, 101, 109, 124, 126-130]. The risk of proteinuria is increased at higher doses [100, 131-133], in patients with HLA B8 and/or DRW3 antigens [68], and in patients with previous gold toxicity [134, 135]. However, others have not confirmed the relationship to the drug dosage [136], duration of therapy [137], or HLA antigens [70]. In the majority of patients, proteinuria is accompanied by

microscopic hematuria [100, 127]. The peak incidence of proteinuria occurs in the second six months of treatment, but it may develop at any time from 6 weeks to 74 months [107, 101, 138]. Proteinuria may be persistent or may slowly progress to nephrotic syndrome if therapy is continued. Up to 1/3 of the patients with significant proteinuria progress to nephrotic syndrome if therapy is continued [106]. Renal function is normal to minimal impairment in patients with isolated proteinuria.

Histopathology Histopathological examination of renal biopsy specimens from the patients with isolated proteinuria due to D-penicillamine shows predominant membranous glomerulopathy [139-141]. Electron microscopy of renal tissue usually demonstrates subepithelial electron dense deposits and fusion of epithelial foot processes [139-141]. The deposits on the epithelial side of the glomerular basement membrane appear to be slowly covered and later incorporated into the basement membrane. With time the deposits become fainter and move towards the endothelial side of the basement membrane [142]. Immunofluorescent study may demonstrate granular deposits of IgG and C3 in the capillary wall. These changes in glomerular histology can persist for at least a year after the withdrawal of the drug [139]. Sellars et al. [143] reviewed the renal biopsies of 30 patients with rheumatoid arthritis and clinical evidence of renal disease. They reported all 9 patients with membranous glomerulonephritis but only 6 of 13 with mesangial change had received Dpenicillamine or gold. Besides membranous glomerulonephritis, there are reports of minimal change glomerulonephritis [144, 145], mild mesangioproliferative glomerulonephritis without crescent [110, 142, 146], or IgM nephropathy [147, 148] associated with D-penicillamine induced proteinuria.

Therapy and prognosis of proteinuria Proteinuria usually resolves slowly after withdrawal of the drug. Hall et al. [149] reported a long-term study of 33 patients with rheumatoid arthritis who developed proteinuria during treatment with D-penicillamine. Of these, fourteen patients developed proteinuria within 6 months after the start of treatment and 27 within 12 months. When treatment was stopped, the proteinuria 465

UEDA & PORTER

reached a median peak of 4.2 g/day (range 0.3-15 g/ day) at one month (range 0-7 months) before resolving spontaneously by six months in 12 patients, 12 months in 21, and 21 months in all. In all their patients whose nephropathy was due to D-penicillamine the proteinuria resolved completely when the drug was withdrawn; renal function did not deteriorate, and corticosteroids were unnecessary [149]. Jaffe [150] reported that reintroduction of D-penicillamine in patients with drug induced proteinuria, starting with a daily dose of 250 mg, was usually followed by a return of proteinuria at about the same time and at about the same cumulative dose as on the first occasion. However, Hill et al. [133] reported successful reintroduction and continuation for a minimum of 13 months in 5 rheumatoid arthritis patients who developed proteinuria during the first course of the drug. They instituted the “go slow, go low” method of Jaffe [151], starting with a daily dose of 50 mg and increasing by monthly increment of 50 mg to a maintenance dose of 150 mg daily. The dose was held at 150 mg/day for 4 months and thereafter increased by 50 mg at 3-months intervals if disease remained active. Proteinuria did not recur, and improvement of disease was shown in all 5 patients [133]. Howard-lock et al. [65] advocated withholding D-penicillamine if there is (1) proteinuria of 2+ on the dipstick, (2) persistent (longer than 3 weeks) proteinuria of 1+ (3) if there are red cell casts, white cell casts, or hyaline casts present, or (4) if red cells > 10 per high power field are present. For patients whose disease has improved but who developed proteinuria between 300 to 1, 000 mg/day, but without other renal abnormality, they suggest the continued use of the drug cautiously at a reduced dose with close monitoring. If proteinuria exceeds 2 g/day or the glomerular filtration rate falls, the drug should be discontinued immediately.

Goodpasture’s-like syndrome Besides the benign proteinuria mentioned above, proliferative glomerulonephritis with fulminant renal failure has also occurred with D-penicillamine therapy. One is Goodpasture’s-like syndrome, which is characterized by pulmonary hemorrhage and rapidly progressive glomerulonephritis. Goodpasture’s -like syndrome associated D-penicillamine treatment has been reported in patients with Wilson’s disease [113], rheumatoid arthritis [114, 115, 152, 153], primary bil466

iary cirrhosis [154], and progressive systemic sclerosis [155]. D-penicillamine was given for at least 7 months (range: 7-84 months), and at a daily dose higher than 750 mg (range: 750-2, 000 mg) preceding the onset of symptoms. Pulmonary X-rays showed bilateral extensive infiltrates in all 10 cases. Lung hemorrhage was the principle cause of death in 3 cases [113]. The histopathology of renal specimens usually showed proliferative glomerulonephritis with crescent formation in 30 to 100% of the glomeruli. Direct immunofluorescent study failed to show linear IgG deposition along the glomerular basement membrane, but granular deposition of IgG and/or C3 were present along the glomerular capillary walls in 5 of 6 patients. Subepithelial electron dense deposits were observed in 3 of 4 patients tested. Circulating anti-glomerular basement membrane antibody was not detected in any of the cases tested. In Brown Norway rats, the administration of D-penicillamine induced antinuclear antibodies and significantly high concentrations of immune complexes. Iin these animals there was no granular deposition of IgG, but linear deposition of IgG along the glomerular basement membrane. IgG eluted from diseased kidneys bound both in vitro and in vivo to the kidney basement membrane [156]. HLA-DR2 antigen was absent in the 2 cases where HLA phenotype was determined, whereas there is a strong association between HLA-DR2 and antibody-mediated Good-pasture’s syndrome [157]. Anti-nuclear antibodies have been detected both before [115, 156] and after initiation of the drug [152, 115]. Although this syndrome is potentially life-threatening, aggressive treatment with plasmapheresis, steroids, immunosuppressive drugs such as azathioprine and cyclophosphamide, and mechanical ventilation with PEEP may be life saving [113, 152-155]. Derk and Jimenez [155a] recently reviewed the case for Goodpasture-like syndrome occurring in systemic sclerosis patients treated with penicillamine. Basically they describe rapidly progressive glomerulonephritis without anti-GBM antibiodies, but with linear or granular glomerular deposits. While they raise the possibility of pausi-immune GN this could not be confirmed since ANCA was not tested in their patient. Despite their conclusions, the case report by Bienaime et al ([155b] makes a compelling case for penicillamine induced ANCA associated RPGN in a patient with Wilson Disease. Since Wilson Disease has never been associated with pauci-immune GN, the

19. Gold salts, D-penicillamine and allopurinol

authors are confident that penicillamine is capable of inducing ANCA associated RPGN, To date all patients with suspected pauci-immune GN have tested positive for anti-MPO antibodies. Since this observation was limited to a patient with Wilson Disease, it remains to be confirmed that patients with systemic sclerosis or rheumatoid arthritis receiving penicillamine who develop the Goodpasture-like syndrome described as a complication of penicillamine treatment should be classified as pauci-immune GN with anti-MPO antibodies.

Renal vasculitis Extracapillary glomerulonephritis with renal vasculitis is also been reported as a rare complication of D-penicillamine therapy [117, 126, 156]. Necrosis of interlobular arteries with glomerular crescent [117] and necrotic and occluded periglomerular arterioles [156] have been reported. Aggressive treatment with pulse steroid, anticoagulants, and antiplatelet agents may be beneficial. The two patients with renal vasculitis, whose outcome was known, died from bacterial infection within ten months after the onset of the disease [117, 156]. These cases mostly likely represent pauci-immune GN as reviewed in the preceding paragraph.

mucocutaneous reactions to chrysotherapy. Manifestations included pleurisy in 5 of 6 patients, rashes in 3, and nephritis in 2. LE cells were present in 5 patients, anti nuclear antibodies in all 6, anti-double-strand DNA in 3, 3 were Coomb’s test positive, and low C4 complement in 5 of the 6 [111]. Results of a renal biopsy from a patient with nephritis showed diffuse endocapillary proliferative glomerulonephritis with focal crescent formation and vasculitis. Electron microscopy showed scattered subendothelial deposits, and immunofluorescent study revealed granular deposition of IgG, IgM, C3 complement and C1q. The patient was successfully treated with prednisolone and azathioprine [112]. Ntoso et al. [156] reported penicillamine-induced rapidly progressive glomerulonephritis in two patients with progressive systemic sclerosis. Anti nuclear antibodies, anti-Sm antibody, and Coomb’s antibodies were positive in both patients. Renal biopsies from the two patients demonstrated a diffuse, predominantly extracapillary, proliferative glomerulonephritis with crescents and focal necrosis, and by immunofluorescence, focal areas of IgG, C3, and fibrinogen were observed in areas of glomerular necrosis. Subendothelial and mesangial deposits were observed by electron microscopy. Both patients responded to pulse methylprednisolone and subsequent daily steroids [156].

Systemic lupus erythematosus syndrome A drug-induced systemic lupus erythematosus (SLE) with proliferative glomerulonephritis has also been described in patients treated with D-penicillamine [111, 157]. Systemic lupus erythematosus syndrome is induced in approximately 2% of patients treated with D-penicillamine [112, 158]. Unlike other forms of druginduced systemic lupus erythematosus, anti-doublestrand DNA antibodies and/or hypocomplementemia are seen in D-penicillamine-induced systemic lupus erythematosus syndrome [111, 156]. Nephropathy is rare in D-penicillamine-induced systemic lupus erythematosus syndrome [111]. Walshe [112] reported that 8 patients developed the serological change of systemic lupus erythematosus of 120 patients with Wilson’s disease treated with D-penicillamine, but none of them showed nephropathy. Chalmers [111] reported 6 rheumatoid arthritis patients with D-penicillamine-induced systemic lupus erythematosus syndrome. All patients had previous

Pathogenesis of D-penicillamineinduced nephropathy Deposition of immune complexes in the glomerular basement membrane may play an important role in the pathogenesis of D-penicillamine-induced nephropathy, such as isolated proteinuria, Goodpasture’s-like syndrome, and nephritis associated with D-penicillamineinduced systemic lupus erythematosus rheumatoid arthritis syndrome. Immunofluorescent study show predominantly granular deposition of IgG and/or C3, and electron microscopy revealed subepithelial or subendothelial electron dense deposits. In rheumatoid arthritis patients, D-penicillamine alters the circulating immune complexes [159]. D-penicillamine has the capacity to convert large complexes into small ones in vitro and there has been speculation that similar mechanisms in vivo could explain the deposition of complexes and renal damage [160]. Small immune complexes deposit in the glomeruli easier than big ones. 467

UEDA & PORTER

In addition to penicillamine nephropathy, other side effects of the drug may be related to the widespread deposition of immune complexes (Figure 3). Dense, granular immunoglobulin deposits have been identified at the epidermodermal junction in 4 rheumatoid arthritis patients who developed toxic reactions, such as severe rashes, thrombocytopenia, aplastic anemia, and proteinuria. Three of 4 penicillamine-induced systemic lupus erythematosus syndrome patients had similar findings on skin biopsy [161]. Besides immune complex deposition, autoantibodies against several autoantigens are frequently detected in patients treated with D-penicillamine, leading to autoimmune diseases. The exact mechanism by which this drug induces autoimmunity remains to be investigated. It may directly stimulate oligoclonal B cell activity, upset the balance between T cell subsets, or alter antigens by hapten formation. D-penicillamine can bind with various proteins, and may change the antigenicity of these proteins as a hapten. However, to date, no evidence for the presence of penicillamine in renal immune deposits has been reported. Nagata et al. [162] reported that D-penicillamine can act as a hapten for specific T cells when presented on the surface of appropriate stimulator cells, and suggested that the adverse immunological side effects of this drug in patients may have a pathogenesis similar to graft-versus-host reaction. The possibility of ANCA associated vasculitis, as described in the case report of Bienaime et al [155b], raises an alternate explanation for the pathogenesis of penicillamine-induced vasculitis. Since all of the ANCA associated GN due to penicillamine have had anti-MPO antibodies, this suggests that an interaction with MPO is critical in triggering the ANCA induced vasculitis. However, it is not clear that penicillamine induces autoimmunity, thus the exact mechanism remains to be elucidated, although both humoral and cellular immunity are thought to play significant roles [162a].

Prediction and monitoring of development of D-penicillamine nephropathy To predict D-penicillamine side effects, the association between side effects and various factors, such as HLA antigens [68, 70, 128, 130, 163, 164], autoantibodies [165, 166], and previous gold toxicity [101, 138, 167, 168] has been studied. Wooley et al. [68] investigated 468

D-PENICILLAMINE NEPHROPATHY D-penicillamine Immunoenhancement

Fragmentation

Immune complex

Deposition GBM

Figure 3. An illustration of the pathogenesis of D-penicillamine induced nephropathy.

the possible interaction between HLA antigens and toxicity of D-penicillamine and sodium aurothiomalate in rheumatoid arthritis patients. Nineteen of 24 patients in whom proteinuria developed were positive for HLA-B8 and DRw3 antigens. Furthermore, all 13 episodes of proteinuria exceeding 2 g/day occurred in patients with DRw3 [68]. There is also a strong association between idiopathic membrane nephropathy and HLA-DRw3, B8 and B18 [169]. Other investigators have confirmed the association between D-penicillamineinduced proteinuria and DR3 [128, 130, 164] and B8 [70, 128, 130]. However, other investigators could not confirm a significant association between D-penicillamine proteinuria and HLA-DR3 [70, 170]. In addition to HLA antigens, Emery et al. [163] emphasized the sulphoxidation status of patients as a new predictor of outcome of drug toxicity. Moutsopoulos et al. [165, 166] reported that antiRo (SSA) positive Greek rheumatoid arthritis patients experienced a significantly high frequency of side effects from D-penicillamine. Despite their dissimilar chemical structures, the thiol compounds, sodium aurothiomalate and D-penicillamine, have remarkably similar clinical effects, and this similarity extends to the incidence and type of adverse effects [138, 167]. Several investigators have noted the association between prior gold nephropathy and D-penicillamine. Billingsley and Stevens reported the significant correlation of D-penicillamine-induced proteinuria to a previous history of

19. Gold salts, D-penicillamine and allopurinol

gold nephropathy [134]. Patients with gold-induced proteinuria are at a higher risk for the development of proteinuria during D-penicillamine therapy (p 1.5 469

UEDA & PORTER

mg/dL) in 65% and a decreased eGRF in 73%. CKD III was present in 47%, CKD IV in 20% and CKD V in 5%. Combination therapy with Colchicine and NSAIDs was used in over 80% of the patients with renal failure and gout. Only 27% of the patients admitted with gouty arthritis were receiving allopurinol prophylaxis. In patients receiving allopurinol prophylaxis as outpatient treatment, one quarter do not have their serum creatinine monitored [188c].

Histopathology Histopathological examination of renal biopsy or autopsy specimens revealed renal vasculitis [181], focal segmental glomerulonephritis [184], and acute interstitial nephritis [185, 187, 189, 190]. Jarzobski et al. [181] reported a case of the hypersensitivity type of vasculitis with fibrinoid necrosis and eosinophilic reaction, involving multiple organs, especially the kidney, resulting in uremia and death. Boyer et al. [191] also reported 3 cases of the same type including the efficacy of prednisolone in treating this type of disease. Kantor et al. [182] reported a case of glomerulonephritis associated with allopurinol-hypersensitivity. Linear deposition of IgG and complement along the glomerular basement membrane were demonstrated, and a necrotizing, hemorrhagic pneumonitis was also reported. However, no circulating anti-glomerular basement membrane antibody was detected. Acute interstitial nephritis has also been reported associated with by the administration of allopurinol [185, 187, 189, 190]. Gelbart et al. [185] reported a case of allopurinolinduced interstitial nephritis with extensive infiltration of lymphocytes, plasma cells and tubular damage. No immunoglobulins, complement, or fibrin were evident in the tubular basement membrane. This patient also had other typical symptoms of hypersensitivity reactions. Grussendrof et al. [187] also reported a case of acute interstitial nephritis with circulating anti-tubular basement membrane antibody and granular C3 deposition on the tubular basement membrane. The interstitium was diffusely widened, edematous and infiltrated with lymphocytes, plasma cells, histiocytes and numerous eosinophils. The nephritis was induced by controlled re-exposure to allopurinol in a patient who had two successive severe hypersensitivity reactions to this drug. More recently, Morel et al [191a] reported a case of allopurinol hypersensitivity reac470

tion with renal failure on admission, in which skin manifestations and renal failure recurred after initial recovery. The case was considered unique due to the presence of an ANA titer or 1:2000 on admission. Treatment consisted of intravenous and oral steroids with residual renal impairment after 7 months of therapy. A renal biopsy, at the time of recurrence, yielded deposits of C3 complement in the vessel walls. A previous skin biopsy on admission yielded leukocytoclastic, non-specific vasculitis. The authors concluded that the “findings suggested the participation of ribonucleotide alterations in the pathophysiology of allopurinol hypersensitivity syndrome”.

Pathogenesis The pathogenesis of nephropathy associated with allopurinol-induced hypersensitivity reactions is unclear. However, pathogenic role of the immune reactions against allopurinol or its metabolites has not been excluded. Emmerson et al. [192] studied the lymphocyte reactivities to allopurinol and its active metabolite, oxypurinol, in 9 patients with previous documented adverse reactions to allopurinol. They suggested that some adverse reactions to allopurinol represented delayed type hypersensitivity to oxypurinol, but not to allopurinol. Allopurinol is oxidized by xanthine oxidase to oxypurinol, which is also an inhibitor of the enzyme (Figure 4). Allopurinol plasma half life is less than 2 hours due to rapid renal clearance and oxidation to oxypurinol [193]. Oxypurinol, because of its reabsorbance by the renal tubules, has a plasma half-life of 18 to 30 hours. The clearance of oxypurinol is diminished in renal insufficiency [194]. In addition, thiazide diuretics might be expected to cause accumulation of oxypurinol since its renal handling is similar to that of uric acid [195]. Hypersensitivity syndrome has been found to occur most frequently when allopurinol is given with thiazides or in patients with renal insufficiency [184, 188]. The immune reactions to oxypurinol may play an important role in the pathogenesis of the syndrome, including being dose dependent. The serum concentration of oxypurinol has been monitored to prevent adverse reactions [195, 196]. Recommended plasma oxypurinol concentrations are below 100 μmol/L [196]. Several authors [195, 196] reported that no adverse reactions have occurred in patients with lower plasma oxypurinol levels; how-

19. Gold salts, D-penicillamine and allopurinol

ALLOPURINOL NEPHROPATHY O2 Allopurinol

active

(T1/2 < 2 h)

Oxypurinol (T1/2 18-30h) Xanthine oxidase

in renal insufficiency: T1/2 ­­

Figure 4. Suggestion of reactions leading to allopurinol nephropathy.

ever, hypersensitivity syndrome occasionally develops in patients with a therapeutic plasma oxypurinol concentration [197]. In addition to plasma oxypurinol concentration, other factors probably contribute to the development of the syndrome. Human herpes virus 6 (HHV 6) infection is recently attracted a great deal of attention as a possible cause of drug-induced hypersensitivity. Suzuki et al reported a case of allopurinol-induced hypersensitivity syndrome with dramatically increased anti-HHV 6 IgG antibodies. They also demonstrated the presence of HHV 6 in the skin of this patient using a polymerase chain reaction and in situ hybridization [198]. Thus, drug-induced hypersensitivity syndrome may not be a simple allergic reaction to drug. Further investigations regarding the relation of HHV 6 infection and drug-induced hypersensitivity syndrome may provide insight to the pathogenesis of allopurinol-induced hypersensitivity syndrome.

Therapy, prognosis, and prevention Withdrawal of the drug and the prolonged administration of systemic steroids are beneficial for the hypersensitivity syndrome with renal involvement. Initial dose of steroid should be 1 to 2 mg/kg/day of methylprednisolone, with careful gradual tapering of steroids required in the majority of patients. The recovery time ranged from 1 week to 11 months. Mortality from this syndrome is high, with twenty-one of 80 patients died as a result of the syndrome [188]. In fulminant cases, such as acute renal failure complicating toxic epidermal necrosis or Stevens-Johnson syndrome, methylprednisolone ‘pulse’ therapy might be

beneficial. Patients with HHV 6 infection also require prednisolone therapy. To prevent unnecessary morbidity and mortality due to the allopurinol hypersensitivity, Singer et al. [188] recommended the indications for allopurinol as follow: 1) tophaceous gout; 2) major uric acid overproduction (urinary excretion of more than 900 mg of uric acid/day on a diet with rigid purine restriction); 3) frequent gouty attacks unresponsive to prophylactic colchicines, when uricosuric agents cannot be used due to intolerance, lack of efficacy, renal insufficiency, or poor patient compliance; 4) recurrent uric acid renal calculi; 5) recurrent calcium oxalate renal calculi when associated with hyperuricosuria; or 6) prevention of acute urate nephropathy in patients receiving cytotoxic therapy for malignancies. The tumor lysis syndrome (TLS) has come under increased scrutiny with the more aggressive chemotherapeutic management of both hemopoeitic and solid tumor malignancies. Because of the massive release of purine nuceloties, pretreatment with allopurinol often is inadequate to control the hyperuricemia and acute uric acid nephropathy develops [198a]. To overcome this deficiency of allopurinol protection, febuxostat, a more prowerful xanthine oxidase inhibitor has been developed. However, in criticizing a recent clinical trial comparing febuxostat with allopurinol [198b], Gelber [198c] wrote “caution, however, needs to be exercise in as much as the reported frequency of adverse events leading to discontinuation of the drug occurred two and three times as often in the low-dose and high-dose febuxostat group, respectively, as in the allopurinol group”. Rasburicase is a urate oxidase that converts uric acid to allantoin which is much more soluble thus precluding acute urate nephropathy in TLS [198d, h]. A combination of rasburicase and allopurinol has been successfully used in preventing hyperuricemia of TLS [198i]. While Rasburicase has been shown to be successful in preventing the hyperuricemia of TLS [198e, f, g] it should not be given to patients with G6PD deficiency, methemaglobinemia and history of anaphalaxis [198e]. Also, concern has been raised about the high immunogenicity of rasburicase and native uricase since antibodies to the drug occurred in 14% of patient in a clinical trial of TLS [198a]. There is disagreement regarding the value of allopurinol treatment for asymptomatic hyperuricemia, uncomplicated gout, and acute gouty attacks which Singer et al [188] consider counter- indicated while Kelley [199] advised 471

UEDA & PORTER

allopurinol therapy for asymptomatic hyperuricemia, but only when it is truly severe (serum uric acid level > 13 mg/dl and 24-hour urine excretion > 1, 100 mg). For the treatment of acute hyperuricemia with renal insufficiency Ronco and coworkers [198g] present compelling data supporting the use of rasburicase. The allopurinol hypersensitivity syndrome occurs most frequently when the drug is given with diuretics or in patients with renal insufficiency. CKD patients on allopurinol therapy should be closely monitored especially within

the first several weeks after initiating administration of the drug. If the AHS develops, allopurinol should be withdrawn but may be reintroduced using a gradually increasing dosage schedule [200]. Finally, patients at high risk for developing hyperuricemia should start the therapy with lower dose of allopurinol. Indications for chronic treatment of symptomatic gout in high risk patients using either rasburicase or febuxostat needs to be confirmed by clinical trials.

References 1.

Lande K. Die gunstige Beeinflussung schleichender Dauerinfekte durch Solganal. Munchen Med Wochenshr 1927; 74: 11321134. 2. The Research Sub-Committee of the Empire Rheumatism Council. Gold therapy in rheumatoid arthritis. Final report of a multicenter controlled trial. Ann Rheum Dis 1961; 20: 315-333. 3. Penneys NS, Eaglestein WH, Frost P. Management of pemphigus with gold compounds. Arch Dermatol 1976; 112: 185-187. 4. Muranaka M, Miyamoto T, Shida T, Kabe J, Makino S, Okumura H, Takeda K, Suzuki S, Horiuchi Y. Gold salt in the treatment of bronchial asthma - a double blind study. Ann Allergu 1978; 40: 132-137. 5. Finkelstein AE, Walz DT, Batista V, Mizrazi M, Roisman F, Misher A. Auranofin – new oral gold compound for treatment of rheumatoid arthritis. Ann Rheum Dis 1976; 35: 251-257. 6. Berglof FE, Berglof K, Walz DT. Auranofin: an oral chrysotherapeutic agent for the treatment of rheumatoid arthritis. J Rheumatol 1978; 5: 68-74. 7. Weisman MH, Hannifin DH. Management of rheumatoid arthritis with oral gold. Arthritis Rheum 1979; 22: 922-925. 8. Davis P, Menard H, Thompson J, Harth M, Beaudet F. One-year comparative study of gold sodium thiomalate and auranofin in the treatment of rheumatoid arthritis. J Rheumatol 1985; 12: 60-67. 9. Ward JR, Williams HJ, Egger MJ, Reading JC, Boyce E, Altz-Smith M, Samuelson CO, Willkens RF, Solsky MA, Hayes SP, Blocka KL, Weinstein A, Meenan RF, Guttadauria M, Kaplan SB, Klippel J. Comparison of auranofin, gold sodium thiomalate and placebo in the treatment of rheumatoid arthritis: A controlled clinical trial. Arthritis Rheum 1983; 26: 1303-1315. 10. Heuer MA, Pietrusko RG, Morris RW, Scheffler BJ. An analysis of worldwide safety experience with auranofin. J Rheumatol 1985; 12: 695-699. 10a. O’Dell JR, Petersen K, Leff R, Palmer W, Schned E, Blakely K, Haire C, Fernandez A. Etanercept in combination with sulfasalaazine, hydroxychloroquine, or gold in the treatment of rheumatoid arthritis. J Rheumatol 2006;33:213-218. 10b. Rau R, Schleusser B, Herborn G, Karger T. Longterm combination therapy of refractory and destructive rheumatoid arthritis with methotrexate (MTX) and intramuscular gold or other disease modifying antirheumatic drugs compare to MTX monotherapy. J Rheumatol 1998;25:1485-1492. 11. Hartfall SJ, Garland HG, Goldie W. Gold treatment of arthritis. A review of 900 cases. Lancet 1937; 9: 838-842. 12. Gibbons RB. Complication of chrysotherapy. A review of recent studies. Arch Intern Med 1979; 139: 343-346. 13. Penneys NS, Ackerman AB, Gottlieb NL. Gold dermatitis. Arch Dermatol 1979; 109: 372-376. 14. Benson WG, Moore N, Tugwell P, D, Souza M, Singal DP. HLA antigens and toxic reactions to sodium aurothiomalate in patients with rheumatoid arthritis. J Rheumatol 1984; 11: 358-361. 15. Adachi JD, Bensen WG, Singal DP, Powers PJ. Gold-induced thrombocytopenia: platelet associated IgG and HLA typing in three patients. J Rheumatol 1984; 11: 355-357. 16. Gottlieb NL, Gray RG. Diagnosis and management of adverse reaction from gold compounds. J Analytic Toxicol 1978; 2: 173. 17. Baldwin JL, Storb R, Thomas ED, Mannik M. Bone marrow transplantation in patients with gold-induced marrow aplasia. Arthritis Rheum 1977; 20: 1043-1048. 18. Gibson J, McGirr EE, York J, Kronenberg H. Aplastic anemia in association with gold therapy for rheumatoid arthritis. Aust NZ J Med 1983; 13: 130-135. 19. Winterbauer RH, Wilske KR, Wheelis RF. Diffuse pulmonary injury associated with gold treatment. N Engl J Med 1976; 294: 919921.

472

19. Gold salts, D-penicillamine and allopurinol

20.

Partanen J, van Assendelft AHW, Koskimies S, Forsberg S, Hakala M, Ilonen J. Patients with rheumatoid arthritis and fold-induced pneumonitis express two high-risk major histocompatibility complex patterns. Chest 1987; 92: 277-281. 21. Fam AG, Paton TW, Shamess CJ, Lewis AJ. Fulminant colitis complicating gold therapy. J Rheumatol 1980; 7: 479-485. 22. Silverberg DS, Kidd EG, Shnitka TK, Ulan RA. Gold nephropathy. A clinical and pathological study. Arthritis Rheum 1970; 13: 812825. 23. Kean WF, Anastassiades TP. Long-term chrysotherapy. Incidence of toxicity and efficacy during sequential time periods. Arthritis Rheum 1979; 22: 495-501. 24. Katz WA, Blodgett RC Jr, Pietrusko RG. Proteinuria in gold treated rheumatoid arthritis. Ann Intern Med 1984; 101: 176-179. 25. Brock H, Han MT. Gold nephrosis. N Engl J Med 1966; 274: 210-211. 26. Furst DE, Levine S, Srinivasan R, Metzger L, Bangert R, Paulus E. A double-blind trial of high versus conventional dosages of gold salts for rheumatoid arthritis. Arthritis Rheum 1977; 20: 1473-1480. 27. Gran JT, Husby G, Thorsby E. HLA DR antigens and gold toxicity. Ann Rheum Dis 1983; 42: 63-66. 28. Bensen WG, Moore N, Tugwell P, D Souza M, Singal DP. HLA antigens and toxic reactions to sodium aurothiomalate in patients with rheumatoid arthritis. J Rheumatol 1984; 11: 358-361. 29. Barger BO, Acton RT, Koopman WJ, Alarcon GS. DR antigens and gold toxicity in white rheumatoid patients. Arthritis Rheum 1984; 27: 601-605. 30. Hakala M, van Assendelef AHW, Ilonen J, Jalava S, Tiilikainen A. Association of different HLA antigens with various toxic effects of gold salts in rheumatoid arthritis. Ann Rheum Dis 1986; 45: 177-182. 31. Vaamaonde CA, Hunt FR. The nephritic syndrome as a complication of gold therapy. Arthritis Rheum 1970; 13: 826-834. 32. Francis KL, Jenis EH, Jensen GE, Calcagno PL. God-associated nephropathy. Arch Pathol Lab Med 1984; 108: 234-238. 33. Hall CL, Fothergill NJ, Blackwell MM, Harrison PR, MacKenzie JC, Maclver AG. The natural course of gold nephropathy: long term study of 21 patients. Br Med J 1987; 295: 745-748. 34. Tornroth T, Skrifvars B. Gold nephropathy. Prototype of membranous glomerulonephritis. Am J Med 1974; 75: 573-590 34a. Hall CL. Gold nephropathy. Nephron 1988;50:265-272. 34b. Hall CL, The natural course of gold and penicillamine nephropathy. A longterm study of 54 patients. Adv. Exp Med Biol 1989;252:247-256/ 34c. Klinkhoff AV, Teufel A. Reinstitution of gold after gold induced proteinuria. J Rheumatol 1997;24:1277-1279. 35. Lee JC, Dushinkin M, Eyring EJ, Engleman EP, Hopper JR. Renal lesions associated with gold therapy. Light and electron microscopic studies. Arthritis Rheum 1965; 8: 1-13. 36. Mattson JC. Glomerular diseases as a complication of gold therapy. Lab Invest 1974; 30: 383. 37. Katz A, Little AH. Gold nephropathy. An immunopathologic study. Arch Pathol 1973; 96: 133-136. 38. Watanabe I, Whitter FC, Moore J, Cuppage FE. Gold nephropathy. Ultrastructural fluorescent and microanalytic study of two human cases. Arch Path Lab Med 1976; 100: 632-635. 39. Davies DJ, Dowling J, Xipell JM. Gold nephropathy. Pathology 1977; 9: 281-288. 40. Skrifvars BV, Tornroth TS, Tallqvist GN. Gold-induced immune complex nephritis in seronegative rheumatoid arthritis. Ann Rheum Dis 1977; 36: 549-556. 41. Wolters J, Frederik P, van Rie H, Zeppenfeldt E. Minimal change nephropathy during gold treatment. A case with unusual histopathological and immunopathological features. Netherland J Med 1987; 31: 234-240. 42. Skirfvars B, Tornroth T, Flalck HM. Granulomatous glomerulonephritis in a patient with rheumatoid arthritis treated with gold salts. Scand J Rheumatol 1979; 8: 234-240. 43. Cramer CR, Hagler HK, Silva FG, Eigenbrodt EH, Meltzer JI, Pirani CL. Chronic interstitial nephritis associated with gold nephropathy. Arch Path Lab Med 1983; 107: 258-263. 44. Tubbs RR, Valenzuela R, McGorman LJ, Pohl MA, Barenberg S. Gold nephropathy. New Eng J Med 1977; 296: 1413-1414. 45. Yarom R, Stein H, Peter PD, Slavin S, Hall TA. Nephrotoxic effect of parenteral and intraarticular gold. Ultrastructural and electron microprobe examination of clinical and experimental materials. Arthritis Rheum 1975; 99: 36-43. 46. Iesato K, Mori Y, Ueda S, Wakashin Y, Wakashin M, Matsui N, Inoue S, Okuda K. Renal tubular dysfunction as a complication of gold therapy in patients with rheumatoid arthritis. Clin Nephrol 1982; 17: 46-52. 47. Ueda S, Wakashin M, Wakashin Y, Yoshida H, Iesato K, Mori T, Mori Y, Akikusa B, Okuda K. Experimental gold nephropathy in guinea pigs: Detection of autoantibodies to renal tubular antigens. Kidney Int 1986; 29: 539-548. 48. Nagi AH, Alexander F, Barabas AZ. Gold nephropathy in rats. Light and electron microscopic studies. Exp Molec Path 1971; 15: 354-362. 49. Eiseman JL, Ribas JL, Knight E, Alvares AP. Acute nephropathy induced by gold sodium thiomalate: Alterations in renal heme metabolism and morphology. Toxicol Appl Pharmacol 1987; 91: 193-203.

473

UEDA & PORTER

50.

Bretza J, Wells I, Novey HS. Association of IgE antibodies to sodium aurothiomalate and adverse reactions to chrysotherapy for rheumatoid arthritis. Am J Med 1983; 74: 945-950. 51. Davis P, Ezeoke A, Munro J, Hobbs JR, Hughes GRV. Immunological studies on the mechanism of gold hypersensitivity reactions. Br Med J 1973; 3: 676-678. 52. Denman EJ, Denman AM. The lymphocyte transformation test and gold hypersensitivity. Ann Rheum Dis 1968; 27: 582-588. 53. Derot M, Kahn J, Mazalton A, Peyrafort J. Fatal anuric nephritis with associated chrysocyanosis following gold therapy. Bull Soc Med Hop Paris 1954; 70: 234-239. 54. Palosuo T, Provast TT, Milgrom F. Gold nephropathy. Serologic data suggesting an immune complex disease. Clin Exp Immunol 1976; 25: 311-318. 55. Skirfvars B. Hypothesis for the pathogenesis of sodium aurothiomalate induced immune complex nephritis. Scand J Rheumatol 1979; 8: 113-118. 56. Jessop JD, Vernon-Roberts B, Harris J. Effects of gold salts and prednisolone on inflammatory cells. I. Phagocytic activity of macrophages and polymorphs in inflammatory exudates studied by a ‘skin-window’ technique in rheumatoid and control patients. Ann Rheum Dis 1973; 32: 294-300. 57. Vernon-Roberts B, Jessop JD, Dore J. Effects of gold salts and prednisolone on inflammatory cells. II. Suppression of inflammation and phagocytosis in the rat. Ann Rheum Dis 1973; 32: 301-307. 58. Panush RS. Effects of certain antirheumatic drugs on normal human peripheral blood lymphocytes. Inhibition of mitogen- and antigen- stimulated incorporation of tritiated thymidine. Arthritis Rheum 1976; 19: 907-917. 59. Harth M, Stiller CR, Sinclair C, Evans J, McGirr D, Zuberi R. Effects of a gold salt on lymphocyte responses. Clin Exp Immunol 1977; 27: 357-364. 60. Measel W. Effects of gold on the immune response of mice. Infect Immunity 1975; 11: 350-354. 61. Ueda S, Wakashin M, Wakashin Y, Yoshida H, Azemoto R, Iesato K, Mori T, Mori Y, Ogawa M, Okuda K. Autoimmune interstitial nephritis in inbred mice. Analysis of mouse tubular basement membrane antigen and genetic control of immune response to it. Am J Pathol 1988; 132: 304-318. 62. Ueda S, Wakashin Y, Yoshida H, Mori T, Mori Y, Azemoto R, Ogawa M, Kato I, Wakashin M. Gold nephropathy - effect of gold on immune response to renal tubular basement membrane (TBM) antigen in mice. In: Nephrotoxicity. In vitro to in vivo, animals to man. Bach BH, Lock EA, editors. Plenum, New York 1989; p. 139-145. 63. Lorber A, Kunishima D, Simon T. Selective inhibition of T suppressor lymphocytes during chrysotherapy [Abstract]. Clin Pharmacol Ther 1983; 303: 300-302. 64. Newton O, Swinburn WR, Swinson DR. Proteinuria with gold nephropathy: When should gold be permanently stopped? Br J Rheumatol 1983; 22: 11-17. 65. Howard-Lock HE, Lock CJL, Mewa A, Kean WF. D-penicillamine; Chemistry and clinical use in rheumatic disease. Sem Arthritis Rheum 1986; 15: 261-281. 66. Manthorpe R, Bendixen G. Auranofin in rheumatoid arthritis: Use in patients with side-effect or lack of effect to gold sodium thiomalate or gold thioglucose and/or D-penicillamine. Scand J Rheumatol 1988; 17: 401-405. 67. Panayi GS, Wooley P, Batchelor JR. Genetic basis of R.A. HLA antigen, disease manifestations and toxic reactions to drugs. Br Med J 1978; 11: 1326-1328. 68. Wooly PH, Griffen J, Panayi GS, Batchelor JR, Welsh KI, Gibson TJ. HLA-DR antigens and toxic reaction to sodium aurothiomalate and D-penicillamine in patients with rheumatoid arthritis. N Engl J Med 1980; 303: 300-302. 69. Bardin T, Legrand L, Naveau B, Marcelli-Barge A, Debeyre N, Lathrop GM, Poirier JC, Schmid M, Ryckewaert A, Dryll A. HLA antigens and seronegative rheumatoid arthritis. Ann Rheum Dis 1985; 44: 50-53. 70. Dequeker J, Van Wanghe P, Verdickt W. A systematic survey of HLA-A, B, C, and D antigens and drug toxicity in rheumatoid arthritis. J Rheumatol 1984; 11: 282-286. 71. Ford PM. HLA antigens and drug toxicity in rheumatoid arthritis. J Rheumatol 1984; 11: 269-261. 72. Van Riel PL, Van de Putte LB, Gribnau FW, de Waal RM. Serum IgA and gold induced toxic effects in patients with rheumatoid arthritis. Arch Intern Med 1984; 144: 1401-1403. 73. Ostuni PA, Simioni M, Marson P, Travaglia P, Volante D, Gambari PF. Serum IgA and gold toxicity in rheumatoid arthritis: lack of predicting value. Clin Exp Rheumatol 1986; 4: 359-362. 74. Ayesh R, Mitchell SC, Waring RH, Withrington RH, Seifert MH, Smith RL. Sodium aurothiomalate toxicity and sulphoxidation capacity in rheumatoid arthritis patients. Br J Rheumatol 1987; 26: 197-201. 74a. Shah P, Griffith SM, Shadforth MF, Fisher J, Dawes PT, Poulton KV, Thomson W, Ollier WE, Mattey DL. Can gold ltherapy be used more safely in rheumatoid arthritis? Adverse drug reactions are more likely in patients with nodular disease, independent of HLA-DR3 status. J Rheumatol 2004;31:1903-1905.

474

19. Gold salts, D-penicillamine and allopurinol

74b. Pissuwan D, Vlaenzuela SM, Cortie MB. Therapeutic possibilities of plasmonicall heated gold nanoparticles. Trends in Biotech. 2006;24:62-67 74c. Zharov VP, Galitovskaya EN, Johnson C, Kelly T. Synergistic enhancement of selective nanophotothermolysis with gold nanoclusters: Potential for cancer therapy. Lasers in Surg Med. 2005;37:219-226. 75. Blodgett RC Jr, Heuer MA, Pietrusko RG. Auranofin: a unique oral chrysotherapeutic agent. Semin Arthritis Rheum 1984; 13: 255-273. 76. Bandilla K, Gross D, Gross W, Herrlinger J-D, Kriegel W, Muller W, Siegmeth W, Tausch G, Thumb N, Wagenhauser J. Oral gold therapy with auranofin (SK&F39162). J Rheumatol 1982; 9 (Suppl 8): 154-159. 77. Katz W, Alexander S, Bland J, Blechman W, Bluhm GB, Bonebrake Ra, Falbo A, Greenwald RA, Hartman S, Hobbs T, Indenbaum S, Lergier JE, Lanier BG, Lightfoot RW, Phelps P, Sheon RP, Torretti D, Wenger ME, Wilske K. The efficacy and safety of auranofin compared to placebo in rheumatoid arthritis. J Rheumatol 1982; 9 (Suppl 8): 173-178. 78. Wenger ME, Alexander S, Bland JH, Blechman WJ. Auranofin versus placebo in the treatment of rheumatoid arthritis. Am J Med 1983; 75: 123-127. 79. Menard HA, Beaudet F, Davis O, Harth M, Percy JS, Russell AS, Thompson M. Gold therapy in rheumatoid arthritis. Interim report of the Canadian multicentre prospective trial comparing sodium aurothiomalate and auranofin. J Rheumatol 1982; 9: 179-183. 80. Felix-Davies DD, Stewart AM, Wilkinson BR, Bateman JR, Delamere JP. A 12-month comparative trial of auranofin and D-penicillamine in rheumatoid arthritis. Am J Med 1983; 75(6A): 138-141. 81. Manthrope R, Horbov S, Sylvest J, Vinterberg H. Auranofin versus penicillamine in rheumatoid arthritis. One-year results from a prospective clinical investigation. Scand J Rheumatol 1986; 15: 13-22. 82. Hochberg MC. Auranofin or D-penicillamine in the treatment of rheumatoid arthritis. Ann Intern Med 1986; 105: 528-535. 83. Revach M, Freed RL, Ehrlich GE. Reversible proteinuria as a complication of oral gold therapy [Letter]. Arthritis Rheum 1979; 22: 1417-1418. 84. Plaza JJ, Herrero G, Barat A, Loutaif JJ, Hernando L, Vallado P, Oliva H. Membranous glomerulonephritis as a complication of oral gold therapy [Letter]. Ann Intern Med 1982; 97: 563-564. 85. Gottlieb NL. Comparative pharmacokinetics of parenteral and oral gold compounds. J Rheumatol 1982; 9 (Suppl 8): 99-109. 86. Blocka K. Auranofin versus injectable gold: Comparison of pharmacokinetic properties. Am J Med 1983; 75(6A): 114-122. 87. Abraham EP, Chain E, Baker W, Robinson R. Penicillamine, a characteristic degradiation product of penicillin. Nature 1943; 151: 107. 88. Walshe JM: Penicillamine. A new oral therapy for Wilson’s disease. Am J Med 1956; 2: 487-495. 89. Drawhall JC, Scowen EF, Watts RWE. Effects of penicillamine on cystinuria. Br Med J 1963; 1: 588-590. 90. Jaffe IA. Intra-articular dissociation of the rheumatoid factor. J Lab Clin Med 1962; 60: 409-421. 91. Jaffe IA. The effect of penicillamine in the laboratory parameters in rheumatoid arthritis. Arthritis Rheum 1965; 8: 1064-1078. 92. Multicenter trial group. Controlled trial of D-penicillamine in severe rheumatoid arthritis. Lancet 1973; 1: 275-280. 93. Steen VD, Medsger TA, Rodnan GP. D-penicillamine therapy in progressive systemic sclerosis (scleroderma). A retrospective analysis. Ann Intern Med 1982; 97: 652-659. 94. Jain S, Scherer PJ, Samourian S. A controlled trial of D-penicillamine therapy in primary biliary cirrhosis. Lancet 1977; 1: 831834. 95. Goldberg Am Smith JA, Lochhead AC. Treatment of lead poisoning with oral penicillamine. Br Med J 1963; 1: 1270-1275. 96. Freeman HC, Huq F, Stevens GN: Metal binding by D-penicillamine: Crystal structure of D-penicillamine cadmium (II) hydrate. J Chem Soc Chem Commun 1976; 90-91. 97. Tamir M, Boustein B, Behar M, Chwat M. Mercury poisoning from an unsuspected source. Br J Ind Med 1968; 21: 299-303. 98. Deutsch MF, Morton JI. Dissociation of human serum macroglobulins. Science 1957; 125: 600-601. 99. Kean WF, Dwosh JL, Anastassiades TP, Ford PM, Kelly HG. The toxicity pattern of D-penicillamine therapy. A guide to the use in RA. Arthritis Rheum 1980; 23: 158-165. 100. Stein HB, Patterson AC, Offer RC, Atkins CJ, Teufel A, Robinson HS. Adverse effects of D-penicillamine in rheumatoid arthritis. Ann Intern Med 1980; 92: 24-29. 101. Kean WF, Lock CJL, Howard-Lock HE, Buchanan WW. Prior gold therapy does not influence the adverse effects of D-penicillamine in rheumatoid arthritis. Arthritis Rheum 1982; 25: 917-922. 102. Wollheim FA, Lindstrom CG. Liver abnormalities in penicillamine treated patients with rheumatoid arthritis. Scand J Rheumatol 1979; 28 (Suppl):100-107. 103. Barzilai D, Dickstein G, Enat R, Bassan H, Lichtig C, Gellei B. Cholestatic jaundice caused by D-penicillamine. Ann Rheum Dis 1978; 37: 98-100. 104. Multz CV. Cholestatic hepatitis caused by penicillamine. JAMA 1981; 246: 674-675. 105. Hickling P, Fuller J. Penicillamine causing acute colitis. Br Med J 1979; 2: 367.

475

UEDA & PORTER

106. Crawhall JC. Proteinuria in D-penicillamine-treated rheumatoid arthritis. J Rheumatol 1981; 8: 161-163. 107. Davison AM, Day AT, Golding JR, Thomson D. Effect of penicillamine on the kidney. Proc R Soc Med 1977; 70: 109-113. 108. Barraclough D, Cunningham TJ, Muirden KD. Microscopic haematuria in patients with rheumatoid arthritis on D-penicillamine. Aust NZ J Med 1981; 11: 706-708. 109. Singh G, Fries JF, Williams CA, Zatarain E, Spitz P, Bloch DA. Toxicity profiles of disease modifying antirheumatic drugs in rheumatoid arthritis. J Rheumatol 1991;18: 188-194. 110. Weiss AS, Markenson JA, Weiss MS, Kammerer WH. Toxicity of D-penicillamine in rheumatoid arthritis. A report of 63 patients including two with aplastic anemia and one with the nephritic syndrome. Am J Med 1978; 64: 114-120. 111. Chalmers A, Thompson D, Stein HE, Reid G, Patterson HC. Systemic Lupus Erythematosus during penicillamine therapy for rheumatoid arthritis. Ann Intern Med 1982; 97: 659-663. 112. Walshe JM. Penicillamine and the SLE syndrome. J Rheumatol 1981; 8 (Suppl 7): 155-160. 113. Sternlieb I, Bennet B, Scheinberg IH. D-penicillamine induced Goodpasture’s syndrome in Wilson’s disease. Ann Intern Med 1975; 82: 673. 114. Gibson T, Barry HC, Ogg C. Goodpastures’ syndrome and D-penicillamine. Ann Intern Med 1976; 84: 100. 115. Gavaghan TE, McNaught PJ, Ralston M, Hayes JM. Penicillamine-induced ‘Goodpasture’s syndrome’; successful treatment of a fulminant case. Aust NZ J Med 1981; 11: 261-265. 116. Jaffe IA. The treatment of rheumatoid arthritis and necrotizing vasculitis with penicillamine. Arthritis Rheum 1970; 13: 436443. 117. Banfi G, Imbasciati E, Guerra L, Mihatsch MJ, Ponticelli C. Extracapillary glomerulonephritis with necrotizing vasculitis in D-penicillamine treated patients with rheumatoid arthritis. Nephron 1983; 33: 56-60. 118. Torres CF, Griggs RC, Baum J, Penn AS. Penicillamine induced myasthenia gravis in progressive sclerosis. Arthritis Rheum 1980; 23: 505-508. 119. Dawkins RL, Garlepp MJ, McDonald BL, Williamson J. Myasthenia gravis and D-penicillamine. J Rheumatol 1981; 8 (Suppl 7): 169-174. 120. Vincent A, Newsom-Davis J. Acetylcholine receptor antibody characteristic in myasthenia gravis. Patients with penicillamineinduced myasthenia or idiopathic myasthenia of recent onset. Clin Exp Immunol 1982; 49: 266-272. 121. Bever CT, Chang HW, Penn AS, Jaffe IA, Bock E. Penicillamine induced myasthenia gravis. Effects of penicillamine on acetylcholine receptor. Neurology 1982; 32: 1077-1082. 122. Bocanegra T, Espinoza LR, Vassey FB, Germain BF. Myasthenia gravis in penicillamine treatment of RA. JAMA 1980; 244: 18221823. 123. Cucher BG, Goldman AL. D-penicillamine induced polymyositis in RA. Ann Intern Med 1976; 85: 615-618. 124. Schreider PL, Peters HA, Dahl DS. Polymyositis and penicillamine. Arch Neurol 1972; 27: 456-457. 125. Wojnarowska F. Dermatomyositis induced by penicillamine. J R Soc Med 1980; 73: 884. 126. Halverson PB, Kozin F, Bernhard GC, Goldman AL. Toxicity of penicillamine. A serious limitation to therapy in rheumatoid arthritis. JAMA 1978; 240: 1870-1871. 127. Steen VD, Blair S, Medsger TA. The toxicity of D-penicillamine in systemic sclerosis. Ann Intern Med 1986; 104: 699-705. 128. Moens HJB, Ament BJW, Feltkamp BW, Korst JK. Longterm followup of treatment with D-penicillamine to rheumatoid arthritis: effectivity and toxicity in relation to HLA antigens. J Rheumatol 1987; 14: 1115-1119. 129. Shiokawa Y, Horiuchi Y, Honma M, Kageyama T, Okada T, Azuma T. Clinical evaluation of D-penicillamine by multicentric double blind comparative study in chronic rheumatoid arthritis. Arthritis Rheum 1977; 20: 1464-1472. 130. Stockman A, Zilko PJ, Major GAC, Tait BD, Property DN, Mathews JD, Hannah MC, McCluskey J, Muirden KD. Genetic markers in rheumatoid arthritis-Relationship of toxicity from D-penicillamine. J Rheumatol 1986; 13: 269-273. 131. Hill HFH, Hill AGS, Day AT, Brown RM, Golding JR, Lyle WH. Maintenance dose of penicillamine in rheumatoid arthritis: A comparison between standard and a response-related flexible regimen. Ann Rheum Dis 1979; 38: 429-433. 132. Williams HJ, Ward JR, Reading JC, Egger MJ, Grandone JT, Samuelson CO, Furst DE, Sullivan JM, Watson MA, Guttadauria M, Cathcart ES, Kaplan SB, Halla JT, Weinstein A, Plotz PH. Low dose D-penicillamine treatment of R.A. A controlled double blind clinical trial. Arthritis Rheum 1983; 26: 581-592. 133. Hill H, Hill A, Davison M. Resumption of treatment with penicillamine after proteinuria. Ann Rheum Dis 1979; 38: 229-231. 134. Billingsley LM, Stevens MB. The relationship between D-penicillamine-induced proteinuria and prior gold nephropathy. Johns Hopkins Med J 1981; 148: 64-67. 135. Halla JT, Cassidy J, Hardin JG. Sequential gold and penicillamine therapy in rheumatoid arthritis. Am J Med 1982; 72: 423-426. 136. Ross JH, McGinty F, Brewer DG. Penicillamine nephropathy. Nephron 1980; 26: 184-186. 137. Dische FE, Swinson DR, Hamilton EBD, Parsons V. Immunopathology of penicillamine-induced glomerular disease. J Rheumatol 1976; 3: 145-154.

476

19. Gold salts, D-penicillamine and allopurinol

138. Halla JT, Cassidy J, Hardin JG. Sequential gold and penicillamine therapy in rheumatoid arthritis. Am J Med 1982; 72: 423-426. 139. Bacon PA, Tribe CR, MacKenzie JC, Jones JV, Cumming RH, Amer B. Penicillamine nephropathy in rheumatoid arthritis. Q J Med 1976; 45: 661-684. 140. Jaffe IA, Treser G, Suzuki Y, Ehrenreich T. Nephropathy induced by D-penicillamine. Ann Intern Med 1968; 69: 549-556. 141. Swinson DR, Hamilton EBD, Dische FE. Penicillamine nephropathy [letter]. Ann Rheum Dis 1975; 34: 544. 142. Neild GH, Gartner HV, Bohle A. Penicillamine induced membranous nephritis. Scand J Rheumatol 1979; 28: 79-90. 143. Sellars L, Siamopoulos K, Wilkinson R, Leohapand T, Morley AR. Renal biopsy appearances in rheumatoid disease. Clin Nephrol 1983; 20: 114-120. 144. Herve JP, Leguy P, Cledes J, Leroy JP. Nephrotic syndrome with minimal glomerular lesions during treatment with D-penicillamine. Nouv Press Med 1980; 9: 2847. 145. Kitazawa K, Ohseto Y, Kawazumi H, Saito K, Shibata T, Uchida J, Yonekura M, Schiwachi S, Ito S, Sugisaki T. D-penicillamine induced lipoid nephrosis in a patient with scleroderma. Ryumachi 1982; 22: 397-403. 146. Hayslett JP, Bensch KG, Kashgarian MZ, Rosenberg LE. Focal glomerulitis due to penicillamine. Lab Invest 1968; 19; 376-381. 147. Rehan A, Johnson K. IgM Nephropathy associated with penicillamine. Am J Nephrol 1986; 6: 71-74. 148. Liao K, Huang CC, Ho HH, Wu CH, Hsueh S. IgM nephropathy after D-penicillamine therapy: Report of a case. J Formosan Med Assoc 1987; 86: 906-909. 149. Hall CL. Jawad S, Harrison PR, MacKenzie JC, Bacon PA, Klouda PT, Maclver AG. Natural course of penicillamine nephropathy: A long-term study of 33 patients. Br Med J 1988; 296: 1083-1086. 150. Jaffe IA. D-penicillamine. Bull Rheum Dis 1977-8; 28: 948-952. 151. Jaffe IA. Penicillamine in rheumatoid disease with particular reference to the rheumatic factor. Postgrad Med J 1968 (suppl): 34-40. 152. Swainson CP, Thompson D, Short AIK, Winney RJ. Plasma exchange in the successful treatment of drug-induced renal disease. Nephron 1982; 30: 244-249. 153. Peces R, Rivera JR, Arboleya LR, Lopez-Larrea C, Alvarez J. Goodpasture’s syndrome in a patient receiving penicillamine and carbimazole. Nephron 1985; 45: 316-320. 154. Matloff DS, Kaplan MM. D-penicillamine-induced Goodpasture’s-like syndrome in primary biliary cirrhosis: successful treatment with plasmapheresis and immunosuppressives. Gastroenterology 1980; 78: 1046-1049. 155. Devogelaer J-P, Pirson Y, Vandenbroucke J-M, Cosyns J-P, Brichard S, Deuxchaisnes CN. D-penicillamine induced crescentic glomerulonephritis: Report and review of the literature. J Rheumatol 1987; 14: 1036-1041. 155a. Derk CT & Jimenez SA. Goodpasture-like syndrome induced by d-penicillamine in a patient with systemic sclerosis: Report and review of the literature. J. Rheumatol 2003;30:1616-1620. 155b. Bienaime G, Clerbaux G, Plaisier E, Mougenot B. Ronco P, Rougier JP. D-penicillamine-induced ANCA-associated crescentic glomerulonephritis in Wilson disease. Am J Kid Dis 2007;50:821-825. 156. Falck HM, Tornooth T, Kock B, Wegelius O. Fatal renal vasculitis with minimal change glomerulonephritis complicating treatment with penicillamine. Acta Med Scand 1979; 205: 133-138. 157. Ntoso KA, Tomaszewski JE, Jimenez SA, Neilson DG. Penicillamine-induced rapidly progressive glomerulonephritis in patients with progressive systemic sclerosis: Successful treatment of two patients and review of the literature. Am J Kidney Dis 1986; 8: 159-163. 158. Camus J-P, Homberg JC, Crouzet J, Mery C, Delrieu F, Massias P, Abuaf N. Autoantibody formation in D-penicillamine treated rheumatoid arthritis. J Rheumatol 1981; 7 (suppl): 80-83. 159. Mohammed I, Barraclough D, Holborrow EJ, Ansell BM. Effect of D-penicillamine therapy on circulating immune complexes in rheumatoid arthritis. Ann Rheum Dis 1976; 35: 458-462. 160. Aarden LA. Pathogencity of immune complexes. In: Non-articular forms of rheumatoid arthritis. Feltkamp TEW, editor. Stafleu’s Scientific Publishing Co, Leyden 1977; 15. 161. Kirby JD, Dieppe PA, Huskisson EC, Smith B. D-penicillamine and immune complex deposition. Ann Rheum Dis 1979; 38: 344346. 162. Nagata N, Hurtenbach U, Gleichmann E. Specific sensitization of Lyt-1+2-stereoisomer. J Immunol 1986; 136: 136-142. 162a. Doyle MK and Cuellar ML. Drug-induced vasculitis. Expert Opin Drug Saf. 2003;2:401-409. 163. Emery P, Panayi GS, Huston G, Welsh KI, Mitchell SC, Shah RR, Idle JR, Smith RL, Waring RH. E-penicillamine induced toxicity in rheumatoid arthritis: the role of sulphoxidation status and HLA-DR3. J Rheumatol 1984; 11: 626-632. 164. Speerstra F, Reekers P, van de Putte LBA, Vandenbroucke JP, Rasker JJ, de Rooij DJRAM. HLA-DR antigens and proteinuria induced by aurothioglucose and D-penicillamine in patients with rheumatoid arthritis. J Rheumatol 1983; 10: 948-953.

477

UEDA & PORTER

165. Moutsopoulos HM, Skopouli FN, Sarras AK, Tsampoulas C, Mavridis AK, Constantopoulos SH, Maddison PJ. Anti-Ro(SSA) positive rheumatoid arthritis: a clinicoserological group of patients with high incidence of D-penicillamine side effects. Ann Rheum Dis 1985; 44: 215-219. 166. Moutsopoulos HM, Giotaki H, Maddison PJ, Mavridis AK, Drosos AA, Skopouli FN. Antibodies to cellular antigens in Greek patients with autoimmune diseases: anti-Ro(SSA) antibody a possible marker of D-penicillamine intolerance. Ann Rheum Dis 1984; 43: 285-287. 167. Dodd MJ, Griffiths ID, Thomson M. Adverse reaction to D-penicillamine after gold toxicity. Br Med J 1980; 26: 1498-1500. 168. Webley M, Coomes EN. Is penicillamine therapy in RA. influenced by previous gold ? Br Med J 1978; 2: 91. 169. Klouda PT, Manos J, Acheson EJ, Dyer PA, Goldby FS, Harris R, Lawler W, Mallick NP, Williams G. Strong association between idiopathic membranous nephropathy and HLA-DRW3. Lancet 1979; 2: 770-771. 170. Swiss Federal Commission for the rheumatic diseases, Subcommission for Research. HLA-DR antigens in rheumatoid arthritis –A Swiss collaborative study; final report. Rheumatol Int 1986; 6: 89-92. 170a. Schiff MH and Whelton A. Renal toxicity associated with disease-modifying antirheumatic drugs used for the treatment of rheumatoid arthritis. Semin Arthritis Rheum 2000; 30:196-208. 171. Rundles RW, Wyngaarden JB, Hitchings GH, Elison B, Silberman HR. Effects of a xanthine oxidase inhibitor on thiopurine metabolism, hyperuricemia, and gout. Trans Assoc Am Physicians 1963; 76: 126-140. 172. Rundles RW, Metz EN, Silberman HR. Allopurinol in the treatment of gout. Ann Intern Med 1966; 64: 229-258. 173. Auerbach R, Orentrich N. Alopecia and ichthyosis secondary to allopurinol. Arch Dermatol 1968; 98: 104. 174. Boston Collaborative Drug Surveillance Program. Allopurinol and cytotoxic drugs. JAMA 1974; 227: 1036-1040. 175. Pinnas G. Possible association between macular lesions and allopurinol. Arch Ophthalmol 1968; 79: 786-787. 176. Korting HC, Lesch R. Acute cholangitis after allopurinol treatment. Lancet 1978; 1: 275-276. 117. Boyer TD, Sun N, Reynolds TB. Allopurinol-hypersensitivity. Vasculitis and liver damage. West J Med 1977; 126: 143-147. 178. Raper R, Barnes P, Ibels L, Lunzer M, Lauer C. Fulminant hepatic failure due to allopurinol. Aust NZ J Med 1984; 14: 63-65. 179. Bailey RR, Neale TJ, Lynn KL. Allopurinol-associated arteritis. Lancet 1976; 2: 907. 180. Greene ML, Fujimoto WY, Seegmiller JE. Urinary xanthine stones, a rare complication of allopurinol therapy. N Eng J Med 1969; 280: 426-427. 180a. LaRosa C, McMullen L, Bakdash S, Ellis D, Krishnamurti L, Wu HY, Mortiz ML. Acute renal failure from xanthine nephropathy during management of acute leukemia. Pediatr Nephrol 2007;22:132-135 180b. Rebentisch G, Stolz s, Muche J. Xanthinuria with xantine lithiasis in a patient with Lesch-Nyhan syndrome under allopurinol therapy. Aktuelle Urol. 2004;215-221. 181. Jarzobski J, Ferry J, Womboldt D, Fitch DM, Egan JD: Vasculitis with allopurinol therapy. Am Heart J 1970; 79: 116-121. 182. Kantor GC. Toxic epidermal necrolysis, azotemia, and death after allopurinol therapy. JAMA 1970; 212: 478-479. 183. Mills RM. Severe hypersensitivity reactions associated with allopurinol. JAMA 1971; 216: 799-802. 184. Young JC, Boswell RB, Niles AS: Severe allopurinol hypersensitivity. Arch Intern Med 1974; 134: 553-558. 185. Gelbart DC, Weinstein AB, Fajardo CF. Allopurinol-induced interstitial nephritis. Ann Intern Med 1977; 86: 196-198. 186. Lupton GP, Odom RB. The allopurinol hypersensitivity syndrome. J Am Acad Dermatol 1979; 1(4): 365-374. 187. Grussendorf M, Andrassy K, Waldherr R, Ritz E. Systemic hypersensitivity to allopurinol with acute interstitial nephritis. Am J Nephrol 1981; 1: 195-109. 188. Singer JZ, Wallace SL. The allopurinol hypersensitivity syndrome. Unnecessary morbidity and mortality. Arthritis Rheum 1986; 29: 82-87. 188a. Tausche AK, Aringer M, Schroeder HE, Bornstein SR, Wunderlich C, Wozel G. The janus faces of allopurinol – Allopurinol Hypersensitivity Syndrome. Am J Med. 2008;121:e3-e4. 188b. Petersel D, Schlesinger N. Treatment of acute gout in hospitalized patients. J. Rheumatol 2007;34:1566-1568. 188c. Raebel MA, McClure DL, Simon SR, Chan KA, Feldstein AC. Gunter MJ, Lafata JE, Harrold L, Nelson WW, Platt R. Frequency of serum creatinine monitoring during allopurinol therapy in ambulatory patients. Ann Pharmacother 2006;40:386-391. 189. McMenamie RA, Davies LM, Crasswell OW. Drug induced interstitial nephritis, hepatitis and exfoliative dermatitis. Aust NZ J Med 1976; 6: 583-587. 190. McKendrick MW, Geddes AM. Allopurinol hypersensitivity. Br J Med 1979; i: 988. 191. Boyer TD, Sun N, Reynolds TB. Allopurinol-hypersensitivity. Vasculitis and liver damage. West J Med 1977; 126: 143-147. 191a. Morel D, Cuez S, Merville P, Deminiere C, Tamisier JM, Potaux L. Recurrent renal failure associated with hypersensitivity to allopurinol. Nephrol Dial Transplant 1999;14:780-781. 192. Emmerson BT, Hazelton RA, Frazer IH. Some adverse reactions to allopurinol may be mediated by lymphocyte reactivity to oxypurinol. Arthritis Rheum 1988; 31: 436-440.

478

19. Gold salts, D-penicillamine and allopurinol

193. Elion GB, Kovensky A, Hitchings GH. Metabolic studies of allopurinol, an inhibitory of xanthine oxidase. Biochem Pharmacol 1965; 15: 863-880. 194. Elion GB, Yu TF, Gutman AB, Hitchings GH. Renal clearance of oxypurinol, the chief metabolite of allopurinol. Am J Med 1968; 45: 69-77. 195. Hande KR, Noone RM, Stone WJ. Severe allopurinol toxicity. Description and guidelines for prevention in patients with renal insufficiency. Am J Med 1984; 76: 47-56. 196. Simmonds HA, Cameron JS, Morris GS, Davies PM. Allopurinol in renal failure and the tumor lysis syndrome. Clin Chim Acta 1986; 31: 189-195. 197. Puig JP, Casas EA, Ramos TH, Michan AA, Mateos FA. Plasma oxypurinol concentration in a patient with allopurinol hypersensitivity. J Rheumatol 1989; 16: 842-844. 198. Suzuki Y, Inagi R, Aono T, Yamanishi K, Shihara T. Human herpesvirus 6 infrction a risk factor for the development of severe druginduced hypersensitivity syndrome. Arch Dermatol 1998; 134: 1108-1112. 198a. Tiu RV, Mountantonakis SE, Dunbar AJ, Schreiber MJ Jr. Tumor lysis syndrome. Semin Thromb Hemost. 2007;33:397-407. 198b. Becker MA, Schumacher R Jr., Wortmann RL, MacDonald PA, Eustace D, Palo WA, Streit J, Joseph-Ridge N. Febuxostat compared with allopurinol in patients with hyperuricemia and gout. N Eng J Med 2005;353:2450-2461. 198c. Gelber AC. Febuxostat versus allopurinol for gout. N Engl J Med 2006;354:1532-1533. 198d. Stamp L, Searle M, O’Donnell J, Chapman P. Gout in solid organ transplantation; A challenging clinical problem. Drugs 2005;65:25932611. 198e. Sood AR, Burry LD, Cheng DKF. Clarifying the role of rasburicase in tumor lysis syndrome. Pharmacother 2007;27:11-=121. 198f. Bessmertny O, Robitaille LM, Cairo MS. Rasburicase: a new approach for preventing and/or treating tumor lysis syndrome. Current Pharmaceutical Design. 2005;11:4177-4185. 198g. Ronco C, Inguaggiato P, Bordoni V, DeCal M, Bonello M, Andrikos E, Assuman Y, Rattanarat R, Bellomo R. Rasburicase therapy in acute hyperuricemia and renal dysfunction. Contrib Nephrol 2005;147:115-123. 198h. Suh WM, Wainberg ZA, deVos S, Cohen AH, Kurtz I, Nguyen MK. Acute lymphoblastic leukemia presenting as acute renal failure. Naature Clin Pract. Nephrology 2007;3:106-110. 198i. DeAngelis S, Noce A, DiRenzo L, Cianci R, Naticchia A, Giarrizzo GR, Giordano F, Tozzo C, Splendiani G, DeLorenzo A. Is rasburicase an effective alternative to allopurinol for the management of hyperuricemia in renal failure patients/ A double –blind randomized study. Euro Rev Med Pharmacol Sci. 2007;11:179-184. 199. Kelley WN. Gout and related disorders of purine metabolism. In: Textbook of Rheumatology (1st edition). Kelly WN, Harris Jr ED, Sledge CB, editors. WB Saunders, Philadelphia 1981; p. 1397-1437. 200. Fam AG. Gout in the elderly. Drug & Aging. 1999;13:229-243.

479

20

Angiotensin I converting enzyme inhibitors and angiotensin II receptor antagonists Paul E. DE JONG University of Groningen, The Netherlands

Introduction ___________________________________________________________ 481 Captopril-associated membranous glomerulopathy ___________________________ 482 Angiotensin I converting enzyme inhibitor-induced acute interstitial nephritis ______ 482 Angiotensin I converting enzyme inhibitor-induced fall in GFR ___________________ 483 Renal artery stenosis Congestive heart failure Renal failure Risk for combined treatment

485 486 487 488

Angiotensin I converting enzyme inhibitor-induced fetal nephrotoxicity ___________ 488 Lessons to be learned from these side effects _________________________________ 488 Angiotensin I converting enzyme inhibition renography Antiproteinuric effects and renal function preservation

489 489

Summary ______________________________________________________________ 491 References _____________________________________________________________ 491

Introduction

O

ver the last decade, the treatment of hypertension has changed dramatically from the concept of stepped care, advocated in the 1970’s, to the more individualized care preferred nowadays. This phenomenon was largely due to the recent development of new classes of antihypertensives, which made it possible to adequately lower blood pressure in most patients with only one or two antihypertensive drugs, thus avoiding the need for a combination of multiple drugs. One of these new drug classes, the angiotensin I converting enzyme inhibitors (ACEI), drew a lot of attention since these were aimed at inhibiting the formation of angiotensin II, a hormone thought to be involved in the origin of systemic hypertension. M.E. De Broe & G.A. Porter (Editors). Clinical Nephrotoxins, 3rd Edition, 481-494

However, some major concerns appeared to restrict the widespread use of these drugs, including both renal histological changes such as a membranous glomerulopathy and an acute interstitial nephritis associated with ACEI, and functional changes such as an ACEI-induced fall in glomerular filtration rate (GFR) in some specified risk groups. Interestingly, although this fall in GFR was initially a reason for concern, after further studies that increased our understanding of the causes of this fall, some possible clinical uses of this phenomenon were recognized. Among these was the use of ACEI to improve the diagnostic armamentarium for renovascular hypertension, to treat urinary protein leakage in patients with the nephrotic syndrome, and most importantly, to preserve renal function in patients with progressively declining renal function.

DE JONG

In this chapter we first will discuss the undesirable aspects of these effects of ACEI and will show how most of these effects may be prevented by cautious use of the agents. Since the mechanisms of the ACEI-induced membranous glomerulopathy and interstitial nephritis are different from those causing the fall in GFR, we will discuss each separately.

Captopril-associated membranous glomerulopathy Proteinuria in association with membranous glomerulopathy has been described during the use of captopril [1-3]. Because of the similar pattern of these side effects to that of other agents containing a sulfhydryl group, like penicillamine, it was suspected that the sulfhydryl moiety of the captopril molecule was involved in the genesis of these effects, either by a direct toxic action or by an immunological mechanism [4]. It was feared that this would seriously limit the use of captopril and future sulfhydryl-containing angiotensin I converting enzyme inhibitors [5]. However, Lewis et al. reported that only 1.1% of 4878 patients treated with captopril exhibited increased proteinuria, with an incidence of nephrotic range proteinuria (more than 3 grams per 24 hours) of 0.8%. Analysis of these cases revealed that more than half of these patients had preexisting renal disease, and many were taking doses in excess of 450 mg per day [6]. Furthermore, in studies in which doses of captopril of 37.5 to 150 mg per day were used, no increased incidence of proteinuria was detected as compared to placebo [7]. Lewis even questioned whether the occurrence of proteinuria during ACEI is related to the sulfhydryl moiety of captopril, since proteinuria has also been demonstrated during treatment with the non-sulfhydryl containing enalapril [8]. However, no data are available on biopsy-documented membranous glomerulopathy in relation to enalapril or other ACEI. The causal role of captopril in the pathogenesis of membranous glomerulopathy has been questioned by the finding of glomerular abnormalities suggestive for membranous glomerulopathy in biopsies of hypertensive patients that had not received the ACEI. In both captopril- and non-captopril-treated patients, spherical dense bodies were found within the glomerular capillary wall with vascular and mesangial deposits of immunoglobulins and C3 that in the early reports 482

were suspected to represent a captopril-induced membranous glomerulopathy [6, 9, 10]. Taken together, data lead us to conclude that proteinuria due to membranous glomerulopathy during captopril treatment seems to be restricted to patients with pre-existing renal disease who use high doses of the drug.

Angiotensin I converting enzyme inhibitorinduced acute interstitial nephritis Acute interstitial nephritis during treatment with an ACEI has been observed in very few instances. Luderer et al. described a patient with skin rash, Coombs positive hemolytic anemia, eosinophilia, and acute kidney injury with eosinophiluria seven weeks after the start of captopril (300 mg per day). An allergic interstitial nephritis was suspected, but unfortunately no renal biopsy was performed and the patient moreover also received furosemide and aspirin [11]. Renal function improved after discontinuation of captopril. Cahan described two patients, one with a biopsy-proven acute eosinophilic interstitial nephritis (together with a membranous glomerulopathy) and the other with chronic interstitial nephritis during treatment with captopril [12]. Since again, both of these patients were also receiving furosemide, the development of interstitial nephritis could not definitely be attributed to captopril. In both patients the nephrotic range proteinuria persisted despite discontinuation of captopril and treatment with prednisolone [12]. Four other cases of acute interstitial nephritis with eosinophils have been described, mostly after usual doses of captopril (50125 mg) given for a few days or weeks [13-16]. In one patient renal interstitial granulomas were also found [15]. In these cases renal function improved promptly after discontinuation of the drug. Another case report described a hypertensive patient presenting with a generalized maculopapular rash after three weeks of captopril therapy [17]. Eosinophilia was present without eosinophiluria. The renal biopsy showed acute tubular necrosis, however without evidence of allergic interstitial nephritis. Renal function improved promptly after discontinuation of captopril. Although a rash and eosinophilia have also been described during enalapril treatment, no data are available on the occurrence of acute interstitial nephritis in patients on enalapril. Moreover, in one of the above-mentioned case reports, captopril rechallenge, but not enalapril,

20. Angiotensin I converting enzyme inhibitors

caused renal functional deterioration [14]. Finally, functional tubular changes have also been described. Renal glycosuria, either with [18] or without [19] a fall in GFR has been found during treatment with captopril. In both cases the abnormality disappeared after withdrawal of the drug. Thus far, no reports have been published on membranous glomerulopathy or acute interstitial nephritis in relation to the use of angiotensin II receptor antagonists. Whether this is due to the relatively short experience with these agents, or the fact that these ACEIinduced side effects are specific for ACEI and thus not related to the interference in the renin angiotensin system in general, cannot be concluded as yet.

Angiotensin I converting enzyme inhibitor-induced fall in GFR In order to understand the ACEI-induced fall in GFR, it is important to begin with a basic understanding of the physiological role of the renin-angiotensin system in the regulation of renal hemodynamics (Figure 1). When renal perfusion pressure drops, renin is released into the plasma and lymph by the juxtaglomerular cells of the kidney. This enzyme cleaves angiotensinogen to form angiotensin I, which is further cleaved by converting enzyme to form angiotensin II, the primary effector molecule in this system. AngioAngiotensinogen

tensin II participates in GFR regulation in at least two ways. First, angiotensin II increases arterial pressure, directly and acutely by causing vasoconstriction, and indirectly and more chronically by increasing body fluid volumes through stimulation of renal sodium retention (both indirectly via aldosterone and through a direct effect on the tubules), as well as by stimulating thirst. Second, angiotensin II preferentially constricts the efferent arteriole, thus helping to preserve glomerular capillary hydrostatic pressure and, consequently, GFR. Although the renin-angiotensin system is now known to be much more complicated than originally thought, including the likelihood that it serves paracrine and autocrine functions as well as endocrine functions, the simplified description above still holds true. As shown in figure 1, angiotensin I converting enzyme or kininase II also interferes in the breakdown of bradykinins, which may contribute to the vasodilation of ACE-inhibitors. Under conditions in which arterial pressure or body fluid volumes are sensed as subnormal, the renin-angiotensin system will be activated and plasma renin activity and angiotensin II levels will be elevated. These conditions include dietary sodium restriction or sodium depletion (such as during diuretic therapy), renal artery stenosis, and congestive heart failure. In each case, fluid and sodium will be retained until the pressure and volume are again sensed as normal. Note Kininogen

Pre-kallikrein

Renin Kallikrein Angiotensin I

Angiotensin converting enzyme Kininase II

Angiotensin II

Increased aldosterone release Potentiation of sympathetic activity Increased Ca2+ current

Activated factor XII

Bradykinin

Inactive peptide Arachidonic acid

Prostaglandins Cough?

VASOCONSTRICTION

VASODILATION

Figure 1. Inhibition of the angiotensin converting enzyme, or kininase II. 483

DE JONG

Normal glomerulus

Autoregulation Afferent arteriole

Perfusion pressure reduced but still within autoregulatory range

Perfusion pressure ¯¯¯ (prerenal azotemia)

Autoregulation PGE2: VD

Autoregulation PGE2: VD

Efferent arteriole

Local angiotensin II: VC

Local angiotensin II: VC

Sympathetic activity angiotensin II: VC

Figure 2. Renal hemodynamics in normal and hypoperfusion conditions. PGE2= prostaglandin E2; VD= vasodilatation; VC= vasoconstriction.

that it is possible for the pressure and/or volume to actually be greater than normal but sensed as normal or subnormal, as in the case of congestive heart failure or renovascular hypertension. In conditions in which the renin-angiotensin system is activated, this system becomes especially important in maintenance of GFR, and the kidney becomes more sensitive to the effects of blockade with an ACEI or angiotensin II receptor antagonist. The phenomenon of constant GFR and plasma flow during changes in renal arterial pressure is known as autoregulation, which under normal conditions is not renin angiotensin dependent (Figure 2). In case of renal arterial pressures below a certain value (about 80 mmHg), the renin-angiotensin system becomes involved. Its importance in this response has been demonstrated by Hall et al., who showed that intrarenal infusion of an angiotensin II antagonist impaired GFR autoregulation but not renal blood flow autoregulation, and that the impairment was more pronounced in sodium-depleted dogs [20]. During angiotensin II receptor antagonism, filtration fraction and efferent arteriolar resistance progressively fell at all renal arterial pressures below control. These investigators also showed that administration of captopril to sodium-depleted dogs impaired autoregulation of GFR but not of renal blood flow when renal perfusion pressure was reduced [21]. Both GFR and renal blood flow were returned to control values when angiotensin II was infused during captopril administration and aortic constriction. Calculated afferent and efferent resistances suggested that an angiotensin-stimulated increase in efferent resistance is important for efficient autoregulation of GFR 484

when renal arterial pressure is clearly reduced. Thus, these investigators have provided strong evidence that an intact renin-angiotensin system is required for maintenance of GFR when renal perfusion pressure falls, and that angiotensin II participates in this GFR autoregulation by preferentially constricting the efferent arteriole. Although generally accepted to be true, it is actually not known whether the renal hemodynamic effects of ACEI are necessarily due to blockade of the reninangiotensin system. Acute kidney injury has not been seen after administration of other antihypertensive agents that do not interfere with the renin-angiotensin system, suggesting that it is blockade of this system, which is responsible for the acute kidney injury. However, angiotensin-converting enzyme is identical to kininase II, the enzyme responsible for degradation of kinins, so that administration of ACEI causes a buildup of vasodilator kinins (e.g. bradykinin) as well as depletion of angiotensin II. Thus, an excess of vasodilator kinins could theoretically contribute to the fall in GFR during ACEI. However, the finding of Hall et al. that an angiotensin receptor antagonist has similar effects to captopril or renin depletion on GFR autoregulation [20], and that the effects of captopril can be reversed by an infusion of angiotensin II [21], would suggest that changes in the kinin system play a minor role, if any, in the effects of ACEI on renal hemodynamics. In patients in whom the renin-angiotensin system is activated, one would expect that efferent arteriolar resistance is maintained at least in part by circulating and/or intrarenal angiotensin II. If angiotensin II preferentially constricts the efferent vessels, then admin-

20. Angiotensin I converting enzyme inhibitors

istration of an ACEI should preferentially dilate these vessels, thus causing a fall in glomerular hydrostatic pressure and a fall in GFR. This would be expected to occur even if renal perfusion pressure was unchanged. Moreover, a captopril-induced fall in systemic arterial pressure (and therefore renal artery pressure), together with impairment of GFR autoregulatory capability, would further contribute to a reduction in GFR. These predictions from a basic understanding of the physiology of the renin-angiotensin system are upheld by clinical findings. Specifically, in some pathophysiological conditions in which maintenance of GFR is highly dependent on an angiotensin II-mediated efferent vasoconstriction, ACEI may result in an acute and pronounced fall in GFR. This is true for patients with bilateral renal artery stenosis or renal artery stenosis in a solitary kidney, for patients with congestive heart failure, and for patients with severe renal failure, especially when they are volume depleted. We will discuss the effect of ACEI in these three patient groups separately.

Renal artery stenosis Shortly after the introduction of ACEI in clinical practice, attention was given to the acute and severe fall in GFR that may be encountered with these drugs in patients with bilateral renal artery stenosis and artery stenosis of a solitary kidney, the latter for example in patients with a renal allograft. In the first report to document such a GFR decline it was suggested to be due to a direct nephrotoxicity of the ACEI [22]. It soon became clear, however, that the fall in filtration was the consequence of renal ischemia, possibly related to the fall in blood pressure and thus in perfusion pressure in the post-stenotic kidney. In one of the earliest reports it was shown that not only captopril but also minoxidil caused GFR to decrease in a patient with a transplant renal artery stenosis [23], suggesting that it was the fall in blood pressure itself, which caused the reduced GFR. However, in other studies it was found that GFR decreased only during treatment with captopril and enalapril [24] whereas a fall in blood pressure during sodium nitroprusside [25] or minoxidil [26, 27], which do not directly interfere with the renin-angiotensin system, did not result in a decline in GFR. Furthermore, studies from Anderson et al. indicated that during infusion of

an ACEI in the renal artery, in doses low enough not to cause any systemic blood pressure effect, an efferent vasodilation occurs resulting in a lowering of filtration pressure in the post-stenotic kidney [28]. The same authors showed that the recovery of GFR that is normally observed after the induction of a renal artery stenosis in dogs is prevented by enalapril treatment [29]. Thus it appears that it is not the fall in systemic blood pressure per se that causes GFR to decrease after captopril in a renal artery stenosis patient. Whatever the precise mechanism may be, the fall in intraglomerular capillary pressure in the post-stenotic kidney may lead to a severe decrease in GFR, even up to total loss of filtration. This is reflected by an acute rise in serum creatinine in patients with bilateral stenosis or stenosis in a solitary kidney [24, 30]. With the contralateral kidney intact however, changes in overall GFR tend to be small and variable, due to compensation by the non-stenotic kidney. Wenting et al., in their elegant study [31] showed not only that captopril greatly reduced the extraction ratio of sodium iodohippurate and iothalamate in the poststenotic kidney in half of the patients with unilateral renal artery stenosis, but also that such a fall in GFR could easily be detected on renal scintigraphy with 99mTc-diethylenetriaminepenta-acetic acid (DTPA), whereas DTPA-uptake had not diminished in the kidney of patients with essential hypertension (Figure 3). Since in patients with a unilateral renal artery stenosis a fall in filtration in the poststenotic kidney may not be detected by a rise in serum creatinine, they concluded that radioisotope renography thus should be performed after beginning captopril treatment in patients with renal artery stenosis [31]. In case of a fall in tracer uptake after captopril the drug should be withdrawn and the physician should aim at a curative approach if possible. As predicted by our understanding of the basic physiology, the fall in filtration after ACEI in a patient with renal artery stenosis is dependent upon the prevailing sodium status of the patient [27, 32-34]. The critical role of sodium balance in this fall in GFR during ACEI has been nicely documented in a case report by Hricik [34], who showed that GFR decreased more markedly in a patient with a transplant renal artery stenosis when captopril was given in a sodium depleted as compared to a sodium replete situation (Table 1). Moreover, Andreucci et al. reported that intravenous infusion of saline could reverse the fall in creatinine 485

DE JONG

clearance or rise in serum creatinine during captopril administration [35]. Since the fall in GFR after ACEI is reversible immediately after withdrawal of the drug [27, 30-32, 36], some authors concluded that its use in patients with renal artery stenosis is relatively safe [37, 38]. 180

Mean arterial pressure (mmHg)

170 160 150 140 130 120

Congestive heart failure

110

A fall in GFR may also be encountered if ACEI are given to patients with congestive heart failure. In a double blind study, Cleland found GFR to decrease from 53 to 48 ml/min (although not a statistically significant fall) during long-term treatment with captopril in 14 patients with congestive heart failure [42]. This fall in renal function, however, is not observed in all patients [43], and may also be different during the different stages of treatment [44]. Packer et al. showed that creatinine clearance worsened only in one third of the patients with severe chronic heart failure during treatment with captopril or enalapril, whereas creatinine clearance remained stable or improved in the other two thirds of the patients during ACEI [43]. The patients that demonstrated worsening of renal function had a lower central venous pressure and used more diuretics prior to the start of the ACEI. They exhibited a greater

100 90 60

DTPA uptake (%)

50

40

30

20

10

0

Group 1 Group 2 (affected kidney) Renal artery stenosis

(right kidney) Essential hypertension

Figure 3. Effect of long-term captopril 150 mg daily on blood pressure and single kidney uptake of 99mTc-DTPA in 14 patients with unilateral renal artery stenosis and 17 patients with essential hypertension. Note that DTPA uptake diminished impressively in half of the poststenotic kidneys in patients with renovascular hypertension, whereas it did not change in another half of the post-stenotic kidneys and in the kidneys of essential hypertensive patients. Reproduced with permission from [31]. 486

However, both animal data and human experience suggests that after continued treatment with an ACEI, atrophy of the stenosed kidney (Figure 4) [39] or complete obstruction of the artery with ischemia of the poststenotic kidney may occur [40, 41]. Whether these effects are directly related to the ACEI or to the natural history of the disease has yet to be established. Another argument against using ACEI in patients with renovascular hypertension, is the fact that in a number of these patients therapy (either transluminal angioplasty, stenting or operative procedures) that will directly address the primary problem is frequently available. The fact that renal function does not worsen during treatment with an ACEI should be evaluated against an expected improvement in renal function after correction of the stenosis.

Table 1. The effect of sodium intake on the renal response to captopril in a patient with a transplant renal artery stenosis. Captopril:

Sodium deplete – +

Sodium replete – +

73.6

73.9

75.2

76.8

Blood pressure (mmHg)

150/96

121/71

130/75

131/83

Serum creatinine (mg/dl)

1.5

3.6

1.6

1.6

Inulin clearance (ml/min)

73

37

62

53

Body weight (kg)

20. Angiotensin I converting enzyme inhibitors

ics in heart failure and their alteration by ACEI.

2.5 Non-clipped Clipped

Weight (g)

2.0

1.5

1.0

*** 0.5

0.0 No treatment

Elanapril

Minoxidil

Figure 4. Weight of the clipped and nonclipped kidneys of two-kidney one-clip rats with Goldblatt´s hypertension after 12 months of no treatment, enalapril, or minoxidil treatment. *** p1g/m2] methotrexate followed by leucoverin in the treatment of head and neck carcinomas and osteosarcoma has led to a more widespread use of this therapy in patients with these and other tumors. Pharmacology Methotrexate is readily filtered by the kidneys, and renal clearance is influence by both tubular secretion [139, 140, 141, 142] and tubular reabsorption [142]. Intravenous administration of methotrexate 140-350 mg/kg [1gr/m2], the urinary concentration of methotrexate and 7-hydroxymethotrexate may exceed the aqueous solubility of these agents at urinary pH [148]. This hypothesis is supported by ability of urinary alkalinization and hydration to decrease the incidence and severity of methotrexate-induced nephrotoxicity [149]. Direct tubular toxicity [148, 150] and decreased tubular filtration [149, 151] may also be components of methotrexate-induced nephrotoxicity. Toxicity is enhanced when patients have been treated previously with cisplatin, or concomitantly with other nephrotoxic drugs [nonsteroidal antiinflammatory drugs] [152]. An enhanced toxicity was indeed observed when administered concomitantly with another highly protein bound agent such as ketoprofen [152] but the mechanism of this interaction is unclear. If anything this interaction should actually have resulted in an increased clearance of the drug. It has been suggested that the decrease in renal clearance of methotrexate observed after concomitant nonsteroidal anti-inflammatory drug treatment could be explained by either a competitive inhibition of methotrexate tubular secretion or inhibition of renal prostaglandins secretion inducing altered glomerular filtration rate in the setting of pre renal volume contraction [153]. When high-doses methotrexate are administered, glomerular filtration rate falls in a majority of patients in a rapid and dose related fashion [154, 155]. Patients should be euvolemic prior to receiving treatment with methotrexate. In addition, adjunctive hydration and urinary alkalinization should be included in therapy for patients receiving dosages equal to or exceeding 50 mg/m2. Due to the significant renal clearance of methotrexate and the risk of increased toxicity associated with increased concentrations of methotrexate over time, dosing of this agent should be done relative to renal function. It is noteworthy that a significant clearance of methotrexate can be achieved with high-flux dialyz-

ers. Serum methotrexate levels can be successfully lowered in patients with methotrexate-induced acute renal failure by charcoal hemoperfusion and sequential hemodialysis [156, 157].

Gemcitabine Several therapeutic agents can promote the development of Thrombotic MicroAngiopathy (TMA), which is characterized clinically by microangiopathic hemolytic anemia, thrombocytopenia, and various ischemic end organ manifestations. Pathologically, vessel wall thickening, endothelial cell swelling, intraluminal platelet thrombi, and microvascular occlusion are noted [158]. Some forms of TMA are dominated by renal impairment and are usually referred to as haemolytic uremic syndrome (HUS); others show predominant central nervous system involvement and are referred to as thrombotic thrombocytopenic purpura (TTP). Gemcitabine is a nucleoside analog that was released in 1996 for the treatment of unresectable pancreatic cancer. Subsequently, it has been employed successfully to treat several malignancies including bladder cancer, non-small cell lung cancer, ovarian cancer, and breast cancer. Initial reports by Flombaum and colleagues of TMA associated with gemcitabine were rare [159] and the manufacturer estimated an incidence of 0.015% according to adverse events reporting through 1997 [160]. Potential underreporting is possible, especially from spontaneous sources, but when compared with the incidence rates ranging from 2.6-13.0% cited in the literature for either malignancyinduced or chemotherapy-induced HUS [160, 161], the incidence of HUS associated with gemcitabine is relatively rare. However, anecdotal experience and review of the literature suggested that TMA occurs more commonly after gemcitabine exposure than initially believed. One retrospective study determined that gemcitabine-associated TMA was more common than previously reported [162]. A cumulative incidence of 0.31% was calculated in this study. Eight of 12 affected patients (67%) required treatment with dialysis. Acute onset or worsening hypertension occurred in 7 (58%) of twelve affected patients. The median duration of treatment with gemcitabine was 5.8 months (range = 3.8 to 13.1 months). Renal dysfunction was universal, while hematuria was present in more than two thirds of patients. Signs and symptoms of HUS developed 521

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

within 1 to 2 months of the last gemcitabine treatment in all patients. Two of 12 patients (17%) had previously received treatment with a mitomycin-containing regimen [160]. The clinical course of TMA, which developed after chronic therapy was dose-dependent. More recently, to characterize the clinical presentation of gemcitabine nephrotoxicity, the medical records of 29 patients were reviewed by Flombaum et al. [163]. Twenty-eight patients were evaluated for new-onset renal disease and one for microangiopathic hemolytic anemia. Median age was 64 years (50-81). The typical presentation consisted of rising serum urea and creatinine levels occurring over a period of weeks to months, in association with new or worsening hypertension (26 patients). Thrombocytopenia and microangiopathic hemolytic anemia of varying degrees were present in all patients. Twenty-three out of 26 patients have a low or undetectable haptoglobin level. LDH level was elevated in all Pts. Schistocytes were present in 21 of the 24 patients who had their blood smears reviewed. The median cumulative dose of gemcitabine was 22gm/m2 (4-81) given over 77 months (1.75-34). Prior chemotherapy with mitomycin C (9 patients), especially if given in close proximity (4 patients), may be synergistic and was particularly severe and appeared within a short period after gemcitabine administration. Full or partial recovery of renal function occurred in 19 patients (in 2 patients, after requiring dialysis for 3 and 20 months). Seven patients progressed to end-stage renal disease and 3 patients developed chronic renal failure, but did not need dialysis. Microhematuria and proteinuria was present in 27 patients. Oedema and chronic heart failure was present in 21 and 7 patients respectively. A high index of suspicion is essential and discontinuation of gemcitabine alone often improved the outcome. Review of literature suggests that cancer-associated HUS usually occurs during widespread metastatic disease or poorly controlled carcinomas, whereas chemotherapy-associated HUS is more common when the patient is in disease remission or has minimal tumor burden [164, 165]. However, the discrimination is not always clear. Murgo [165] attempted to distinguish the characteristics of malignancy-induced and chemotherapy-induced HUS and identified several features to separate the two while Gordon and Kwaan [164] showed that there are more similarities than differences. Some researchers suggest the level of serum 522

factors such as TNF-α, IL-1, and IL-6 as well as von Willebrand factor (vWF) antigen and low molecular weight vWF multimers may be used to distinguish between malignancy-associated HUS and chemotherapy-associated HUS [166, 167]. However, such studies remain experimental and are not readily available in the majority of community settings. Ancillary treatments or antidotes that effectively prevent or reduce the severity of gemcitabine associated HUS have not be identified. Treatment modalities employed for TMA include immunocomplex removal (plasmapheresis, immunoadsorption, hemodialysis, or exchange transfusion), antiplatelet/anticoagulant therapies (antiplatelet drugs, heparin, prostacyclin, or splenectomy), immunosuppressive therapies (corticosteroids, vincristine, or azathioprine), and fresh frozen plasma transfusion [160, 168, 169]. Most of such modalities of treatment are safe and quite effective, in particular if performed in a specialized setting. Despite the availability of these treatments, the outcome with TMA is poor with a high mortality. Mortality rates rage from 10 to 40% in the majority of series [168, 169] to as high as 60-70% in others [161, 170]. Such a high mortality rate approaching 50% is not surprising because the majority of these patients had advanced disease. Since this adverse effect is frequently associated with a poor outcome, patients should be monitored for signs and symptoms of HUS for 3 months following completion of treatment with gemcitabine. Upon recognition of gemcitabine-associated TMA, the drug must be discontinued. Depending on the timing of diagnosis, there may be full renal recovery with early recognition. Late diagnosis is associated with chronic kidney disease, development of end stage renal disease requiring dialysis, and death due to progressive disease.

Antitumor antibiotics Mitomycin Mitomycin C does not need dosage adjustment in the presence of impaired renal function since less than 20% of the dose appears in the urine [171, 172, 173]. Potentially life-threatening hemolytic uremic syndrome is an adverse event that occurs with Mitomycin C therapy [174.]. Hematological findings include anemia, thrombocytopenia and the presence of schizocytes

22. Anticancer drugs

on peripheral blood smear. Acute renal failure in these settings may be associated with proteinuria and microscopic hematuria [175, 176, 177, 178, 179]. The onset of signs and symptoms associated with renal impairment generally occurs 6-17 months following the initiation of treatment with this agent [180]. Corticoids and plasma exchanges have been associated with drastic improvement of the renal parameters [181]. The mechanism of Mitomycin induced nephrotoxicity is unknown. To prevent the occurrence of this side effect, the maximum cumulative dose should be 40 mg/m2. Severe hemolytic uremic syndrome can also be seen with different other anticancer drugs [5-fluoro-uracile, vincristine, cis-platin, bleomycin, adriamycin] [182].

Immunotherapy Interleukin 2 Recombinant Interleukin 2 has opened a new approach in the treatment of renal cell carcinoma [183, 184]. Interleukin 2 alone, or combined with interferon alpha and lymphokine activated killer cells is now used at a high dosage in patients with advanced melanoma or renal cancer to induce regression of solid tumor and metastasis [185, 186, 187]. Renal toxicity is the main limiting side effect of Interleukin 2 administration, more important in aging and subjects with altered renal function and often leads physicians to discontinue or reduce interleukin-2 doses. Clinical studies document a reversible syndrome of hypotension, oliguria, fluid retention, azotemia, and a very low urinary excretion of sodium [188, 189]. It is associated to the so-called vascular leak syndromedescribed in experimental studies as a primary increase in the vascular permeability with consequent shifting of proteinaceous intravascular fluid into the interstitium of many organs, hypoalbuminemia and reduction of the intravascular volume [190]. Therefore, acute renal failure after interleukin-2 administration has initially been considered as secondary to the systemic alterations associated with this treatment. Further studies have suggested that in cancer patients with interleukin-2 continuous infusion, renal failure may occur even in stable hemodynamic conditions [191]. Acute interstitial nephritis characterized by parenchymal infiltration with T lymphocytes was also reported [192] in few patients. It has been suggested

Figure 3. Mitomycin C induced thrombotic microangiopathy. Glomeruli show swelling and detatchment of endothelial cells and luminal occlusion. The arterioles and arteries show intimal cellular swelling and hyperplasia and fibrin deposition. Masson’s trichrome staining, magn. x325.

that acute tubular nephritis could be the result of a cytotoxic lymphocyte-mediated reaction induced by the interleukin-2 treatment [193]. In human, the pathophysiology of interleukin-2 induced renal dysfunction is still poorly understood. Interleukin-2 may act directly on the vascular tonus and endothelial integrity or may stimulate generation of other cytokines, which in turn would increase vascular permeability. Occurrence of an intrinsic renal lesion has been suggested by Shalmi et al. [194] who showed that glomerular filtration rate is altered in 90% of the patients [mean decrease of 43%] whereas renal plasma flow decreases is only slightly altered [mean decrease 5%] in 50% of the patients. Since interleukin-2 induced rate of response in patients with metastatic melanoma or renal cell cancer is schedule and dose dependent, and because renal toxicity is the main cause of treatment discontinuation, more studies are warranted to elucidate the observed nephrotoxicity.

Alpha-Interferon Pharmacology Alpha-interferon, a 165 amino acid glycoprotein, is effective in the treatment of viral hepatitis C and B, and also myeloma, melanoma, and renal carcinoma. Little is known about the renal metabolism of alpha-interferon 523

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

despite extensive studies in experimental animals. In patients with normal renal function, a serum peak level is obtained 8 hours after a subcutaneous injection of 3 106 units of alpha-interferon. Terminal elimination halflife ranges from 4 to 16 hours and after 24 to 48 hours, the interferon molecule is not detected anymore in the serum [195]. Alpha-interferon urinary level is undetectable. Some authors have suggested that, despite the lack of urinary excretion, the kidney could play a role in alpha-interferon metabolism [196]. Indeed, as far as hepatitis C treatment is concerned, dialysis patients often show a better response to therapy than non-dialysis patients do but at the expense of increased side effects. This better efficacy is associated with lower tolerance in this population. This raises the question of pharmacokinetics modifications. Few studies documented that clearance kinetics of interferon alpha in patients with chronic renal failure is about twice as low as in patients with normal renal function [197]. Indeed interferon is filtered by the glomeruli and largely absorbed and catabolized within tubular cells [198]. Main side effects are dose-dependent chills, fatigue and gastrointestinal disturbances. Rarely seizures, encephalopathy and strokes have been reported [199]. Although there has been considerable experience with interferons in clinical trials during the past 20 years, acute renal failure as a side effect of interferon treatment has been rarely reported. In 1976, Gresser et al. [200] described experimental lesions induced in the kidney by interferon in mice. Glomerular nephropathy was observed, either hyalynosis or rapidly progressive glomerulonephritis [201]. In human, while proteinuria has been noted in up to 15 to 20% patients treated with interferon [202], acute renal failure syndrome has rarely been observed [203], [204]. Nephrotic syndrome was present in some cases [205] and the histopathology was described as a combination of acute interstitial nephrotoxicity and minimal change nephropathy. This pattern is similar to that seen with renal injury from non-steroidal anti-inflammatory drugs and ampicillin [206]. A pathogenic role for cellular immunity being enhanced by interferon therapy has been therefore suspected. The overall incidence of alpha-interferon acute renal toxicity was recently evaluated as below 5% in patients treated for myeloproliferative syndrome [207]. Acute renal insufficiency as a complication of gamma interferon treatment has been reported anecdotally. 524

Anti-VEGF agents Those latter years have seen the emergence of new anticancer molecules with novel pharmacological mechanisms named « targeted therapies ». They differ from conventional chemotherapy and radiotherapy in the way that they demonstrate a high specificity towards their target. Among those new drugs, the inhibitors of angiogenesis are the most developed. Their therapeutic targets include the vascular endothelial growth factor (VEGF), its circulating form or its receptors. Indeed, bevacizumab has been marketed a few years ago. It is a humanized monoclonal antibody targeted to VEGF, or other molecules such as VEGFTrap. Other classes of anti-angiogenic agents include the inhibitors of tyrosine kinase sunitinib, sorafenib, AG013736, or valatinib, which specifically inhibit the tyrosine kinases from the intracellular domain of VEGF receptor. Most common renal effects of those drugs, which are besides that well-tolerated, were hypertension and proteinuria, witth some differences in their histological presentation.

Bevacizumab In a Phase III study, Hurwitz et al. [208] report an incidence of grade III hypertension of 11% in patients receiving bevacizumab and chemotherapy as compared to 2.3% in patients receiving the same regimen without bevacizumab. The incidence of all grades hypertension was 22.4% in that study and no episode of hypertensive crisis or death secondary to hypertension have been observed. In 2003, Kabbinavar had already reported that 19 out of 104 patients had presented hypertension, 47% of them having a history of hypertension before the introduction of the treatment [209]. The incidence of all grades hypertension was 28% in this study following administration of bevacizumab 10 mg/kg/dose. This incidence was dose-related since it was 11% in the group of patients receiving bevacizumab 5 mg/kg/ dose. Similarly, the Summary of Product Characteristics (SmPC) of bevacizumab, Avastin® mentions that an elevated incidence of hypertension has been notes in patients receiving bevacizumab in combination with 5-FU (60 to 67%) as compared to those patients who did not received bevacizumab (43%). The same observation was noted regarding severe hypertension: 7 to 10% in bevacizumab-treated patients versus 2% in

22. Anticancer drugs

non-treated patients [210]. The mechanism of this hypertension is unclear. One hypothesis would be the inhibition of VEGF. Indeed, VEGF stimulates Nitric Oxide production (NO) and thus acts as a vasodilator. Its inhibition could thus lead to vasoconstriction and elevation of blood pressure. In fact, bevacizumab plays a key-role in angiogenesis and hemodynamics through inhibiting VEGF. Some animals and human studies suggest that an appropriate, balanced expression of VEGF is mandatory to maintain the structure and functions of the renal glomerulus. For instance, hyper-expression or under-expression of VEGF may lead to the development of a glomerulopathy. In one study, an elevated concentration of the soluble VEGF receptor R1 is a predictor of pre-eclampsia, associated with proteinuria and hypertension [211]. Animal studies have shown that VEGF is expressed in podocytes and that its receptors are present in endothelial glomerular cells [212] and in another study, the authors demonstrated that under-expression or heterozygous expression of VEGF in podocytes induces endotheliosis with hyaline deposits, clinically accompanied by a nephritic syndrome resembling pre-eclampsia [213]. On the other hand, hyper-expression of VEGF in podocytes may also induce proteinuria, secondary to a collapsing of the glomerulus [214]. Those results demonstrate that an appropriate expression of VEGF is essential in maintaining the function and structure of the renal glomerulus [215]. There are very few histological data from renal biopsies in cancer patients treated with anti-VEGF agents and the precise anatomopathological profile of the proteinuria observed remains unclear. Furthermore, proteinuria may also be due to the elevation of intraglomerular pressure secondary to hypertension, at least partly, or related to other nephrotoxic drugs frequently used in cancer patients. In the article from Miller [216], the incidence of proteinuria was 33.9% in patients receiving bevacizumab and pamidronate alone induced proteinuria in 18.5% of cases. Those two drugs in combination may thus add their nephrotoxic effects, bevacizumab through a proteinuria, sometimes severe, and pamidronate through a collapsing focal and segmental glomerular sclerosis or other glomerulopathies [217]. Bevacizumab-induced renal effects must not favour neglecting other possible causes of renal toxicity. For instance, in two patients with grade 3 proteinuria

reported in the literature for whom a renal biopsy was performed, authors show lesions of focal and segmental glomerulo sclerosis due to co-administration of pamidronate for 2 years in one [216] a cryoglobulinemic glomerulonephritis for the second [218]. Among anti-VEGF agents, hypertension and proteinuria have also been reported with sunitib [219], sorafenib [220], AG13736 [221], and VEGF-Trap [222].

Sunitib The most frequent adverse events encountered under treatment with sunitib are hypertension (18 to 28%) and asthenia (42 to 47%). In two Phase II studies in patients with metastatic renal carcinoma, hypertension has been noted in 48 patients out of 169 (28%), with 6% presenting with a grade 3 hypertension versus 1% of patients receiving placebo. No grade 4 hypertension has been reported and dosage reduction or withdrawal of treatment allowed blood pressure control [219, 223]. The mean time before apparition of hypertension was 131 days [7 – 316] after bevacizumab first dose. Moreover, 12% of those patients from this study had elevation of their serum creatinine.

Sorafenib Ratain et al. report the occurrence of hypertension in 86 patients out of 202 (43%) with 62 (31%) presenting a grade ¾ in a population of metastatic renal carcinoma patients treated with sorafenib for 12 weeks. Forty-six percent of those patients necessitated antihypertensive therapy and none of them died [224]. In another study [220], 75% of 20 metastatic renal carcinoma patients exhibited an increase in their systolic blood pressure of more than 10 mmHg. In 12 patients, this increase was superior to 20 mmHg after 3 weeks under sorafenib therapy. Diastolic blood pressure was also increased, by more than 9.3 mmHg under treatment. In another study in 28 patients presenting solid tumours of different types, hypertension was observed in 5 patients after 3 to 4 weeks of sorafenib treatment [225].

Other anti-VEGF agents Thirty-three percent of 17 patients treated with AG013736 developed hypertension. In one patient, treatment had to be resumed because of this side effect, 525

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

and 6 of them (12%) presented a grade ¾ hypertension [221]. Data are even more scanty with VEGF-Trap but it seems that this drug behaves the same way with regard to this issue, as showed in ref [222].

Treatment of anti-VEGF agent-induced renal effects Optimal treatment of hypertension and proteinuria induced by anti-VEG agents necessitates a clear understanding of their physiopathology. In fact, the association of hypertension with bevacizumab may result from the pharmacological effect of the drug resulting in the blockade of the VEGF pathway. Indeed, the intravenous administration of VEGF in rats induces a dose-dependent hypotension with reflex tachycardia. This manifestation is linked to an interaction with NO [226] and the induction of prostacyclin synthesis [227]. As demonstrated in the VIVA study [228], in rats, hypotension is a limiting and dose-dependent side-effect of VEGF IV infusions. Angiotensin converting enzyme (ACE) inhibitors or angiotensin II receptor antagonists (ARBs) are usually used to reduce proteinuria which occurs during hypertensive episodes in patients with type 2 diabetes or glomerulopathies. There are no evidence to date that such therapies may be efficient in reducing proteinuria and hypertension secondary to bevacizumab treatment. Furthermore, angiotensin II may increase VEGF secretion in podocytes, as observed in renal cells from mice [229]. As a result, a treatment with either an ACE inhibitors or an ARBs may be tried in such patients, keeping in mind that there are no clinical evidence suggesting their potential efficacy. However, because decreasing blood pressure is a crucial issue to allow maintenance of anti-VEGF therapy, anti-hypertensive therapies may be used with antihypertensive drugs from those classes or calcium inhibitors, beta blockers, or diuretics. None of those classes may be recommended to date. However, in one Phase II study in patients receiving bevacizumab for the treatment of advanced carcinoïd tumours, blood pressure was not in the target in 50% of the patients receiving nifedipine, and in 78% of the patients receiving other antihypertensive therapies [230]. A strict monitoring of blood pressure is thus mandatory in those patients, before treatment is started and every 2 weeks thereafter, and more frequently if hypertension is diagnosed [210]. Proteinuria may be easily monitored with the determination of the pro526

tein/creatin ratio on a urine spot, which is an easy and reliable method in this indication [231].

Radiation nephritis Radiation nephropathy is defined as renal injury caused by ionizing radiation. Number of cases increases steadily and parallels the increase of bone marrow transplant procedures using total body irradiation [232]. Radiation nephritis is dose dependent [233]. Doses traditionally associated with radiation nephropathy are above 2000 cGy. Fractionation, time and chemotherapy may change the time course and severity of radiation-induced nephropathy. Increasing tolerance is observed with increasing fractionation, probably because it allows repair of sublethal damage during the time between fractional doses. Therefore, chronic nephropathy can be prevented by kidney shielding or, alternatively, by fractionating doses. Previous cytotoxic chemotherapy, radiocontrast agents, antibiotics potentiate the effects of ionizing radiation [234]. Radiation nephropathy can present in several forms. An acute form is usually seen within a year after radiation and presents with hypertension, anemia and edema. A more insidious chronic form presents primarily with diminished glomerular filtration, hypertension and occasionally proteinuria. When present, associated accelerated hypertension can promote renal failure. Some patients may develop hypertension within several years after radiation but no azotemia. In a subset of patients, mild proteinuria may be the only feature of chronic renal disease [235]. Interstitial fibrosis is the common pathologic finding in patients with chronic radiation nephritis. Morphologic studies of radiation nephropathy have documented injury to blood vessels, glomeruli, tubular epithelium and interstitium. Recent ultrastructural studies indicate that glomerular endothelium is an early site of visible injury [236] with endothelial disruption and leukocyte adherence. Later, tubular degeneration and atrophy occur. The second pathophysiologic hypothesis holds vascular injury as the main initial event [237] which helps understand the hypertension occurring in radiation nephritis but does not account for the glomerular lesions. Even though radiation nephropathy has been known for a long time, first therapeutic attempts were only made recently when it has been documented

22. Anticancer drugs

that ACE inhibitors could attenuate in animals the response to renal irradiation over the short term [238] and slow down the decline in renal function even after the onset of renal injury [239]. Same positive effects were observed with angiotensin II antagonists whereas other antihypertensive agents (hydrochlorothiazide or verapamil) were ineffective. These data point to a role

for the renin-angiotensin system in the pathogenesis of radiation nephropathy but clinical data are scarce and no long-term benefit of ACE inhibitors has been yet established. Acknowledgements Figures 1-3 by courtesy of Dr. Hélène Beaufils.

References 1. 2.

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

Safirstein RL. Anticancer drugs. In Clinical Nephrotoxins, Debroe ME, Porter GA, Bennett WM, Verpoten GA. [Editors]. Kluwer Academic Publ. Dordrecht, 1998: 261-271. Meijer S, Mulder NH, Sleijfer DT, de Jong PE, Sluiter WJ, Schraffordt Koops H, van der Hem GK. Nephrotoxicity of cis-diamminedichloride platinum [CDDP] during remission-induction and maintenance chemotherapy of the testicular carcinoma. Cancer Chemother Pharmacol 1982; 8: 27-30. Brock PR, Koliouska DE, Baratt TM, Yeomans E, Pritchard J. Partial reversibility of cisplatin nephrotoxicity in children. J Pediatric 1991; 118: 531-4. Yuhas JM, Culo F. Selective inhibition of the nephrotoxicity of cis-dichlorodiamine platinum by WR 272l without altering its antitumor properties. Cancer Treat Rep 1980; 64: 57-64. Borch RF, Pleasants ME. Inhibition of cis-platinum nephrotoxicity by diethyldithiocarbamate rescue in a rat renal model. Proc Natl Acad Sci USA 1979; 76: 6611-4. Berry J-P, Pauwells C, Tlouzeau S, Lespinats G. Effect of Selenium in combination with cis-diamminedichloroplatinum [II] in the treatment of murine fibrosarcoma. Cancer Res 1984; 44: 2864-8. Citkovic E, Hayes DM, Golbey RB, Krakoff IH. Cisplatin-nephrotoxicity: diethyldithiocarbomate, WR2721 or just water. J Clin Oncol 1991; 9: 707-9. Litterst CL, Torres IJ, Guarino AM. Plasma levels and organ distribution of platinum in the rat, dog, and dog fish following intravenous administration of cis-DDP[II]. J Clin Hemat Oncol 1977; 7: 169-78. Safirstein R, Daye M, Miller P, Guttenplan J. Renal disposition and metabolism of liganded platinum: implications to toxicity. Fed Proc 1980; 40: 651A Safirstein R, Miller P, Guttenplan JB. Uptake and metabolism of cisplatin by rat kidney. Kidney Int 1984; 25: 753-8. Jacobs C, Kalman SM, Tretton M, Weiner MW. Renal handling of cis-diammine dichloroplatinum [II]. Cancer Treat Rep 1980; 64: 1223-6. Levi J, Jacobs C, Kalman S, McTigue M, Weiner MWJ. Mechanism of cis-platinum nephrotoxicity I. Effects on sulfhydryl groups in rat kidneys. Pharmacol Exp Ther 1980; 213: 545-50. Safirstein R, Winston J, Moel D, Dikman S, Guttenplan J. Cisplatin nephrotoxicity insights into mechanism. Int J Andrology 1987; 10: 325-46. Safirstein RL, Daye M, Guttenplan JB. Mutagenic activity and identification of excreted platinum in human and rat urine and rat plasma after administration of cisplatin. Cancer Letters 1983; 18: 329-38. Bradley LJN, Yacema KJ, Lippard SJ, Essigmann JM. Mutagenicity and genotoxicity of the major DNA adduct of the antitumor drug cis-diamminedichloroplatinum [II]. Biochemistry 1993; 32: 982-8. Greene MH. Is cisplatin a human carcinogen ? J Natl Cancer Inst 1992; 84: 306-12. Winston JA, Safirstein R. Reduced renal blood flow in early cisplatin-induced acute renal failure in the rat. Am J Physiol. 1985, 249: F490-496. Hutchinson FN, Perez EA, Gandara DR, Lawrence HJ, Kaysen G. Renal salt-wasting in patients treated with cisplatin. Ann Intern Med. 1988, 108:21-5. Schilsky Rl, Anderson T. Hypomagnesemia and renal magnesium wasting in patients receiving cis-diamminedichloroplatinum II. Ann Intern Med 1979; 90: 929-31. Schilsky RL, Barlock A, Ozols RF. Persistent hypomagnesemia following cisplatin chemotherapy for testicular cancer. Cancer Treat Reports 1982; 66: 1767-9. Mavichak v, Wong NLM, Quamme GA, Magil AB, Sutton RAL, Dirks JH. Studies on the pathogenesis of cisplatin-induced hypomagnesemia in rats. Kidney Int 1985; 28: 914-21. Swainson CP, Colls BM, Fitzharris BM. Cisplatinium and distal renal tubule toxicity. N Z Med J, 1985, 98:375-378.

527

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

23. 24. 25. 26. 27. 28. 29. 30. 31. 32.

33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.

45. 46. 47. 48. 49. 50.

528

Safirstein R, Winston J, Moel D, Dikman S, Guttenplan J. Cisplatin nephrotoxicity insights into mechanism. Int J Andrology 1987; 10: 325-46. Schor N, Ichikawa I, Rennke HG, Troy JL, Brenner BM. Pathophysiology of altered glomerular function in aminoglycoside-treated rats. Kidney Int 1981; 19: 288-92. Gordon JA, Gattone VH. Mitochondrial alterations in cisplatin-induced acute renal failure. Am J Physiol 1986; 250: F991-8. Brady HR, Kone BC, Stroniski ME, Zeidel ML, Giebisch G, Gullans SR. Mitochondrial injury: an early event in cisplatin-toxicity to renal proximal tubules. Am J Physiol 1990; 258: F1181-7. Harder HC, Rosenberg B. Inhibitory effects of anti-tumor platinum compounds on DNA, RNA, and protein synthesis in mammalian cells in vitro. Int J Cancer 1970; 6: 207-16. Howle JA, Gale GR. Cis-dichlorodiammine platinum II persistent and selective inhibition of deoxyribonucleic acid synthesis in vitro. Biochem Pharmacol 1970; 19: 2757-62. Munchausen LL, Rahn RO. Biological and chemical effects of cis-dichlorodiammineplatinum [II] on DNA. Cancer Chemother Rep 1975; 59: 643-6. Roberts JJ, Prascoe JM. Cross linking of complementary strands of DNA in mammalian cells by antitumor platinum compounds. Nature 1972; 235: 282-4. Zwelling LA, Anderson T, Kohn KW. DNA-protein and DNA interstrand cross-linking by cis- and trans platinum [II] Diamminedichloride in Ll2lO mouse leukemia cells and relation to cytotoxicity. Cancer Res 1979; 39: 365-9. Fraval HNA, Rawlings CJ, Roberts JJ. Increased sensitivity of UV repair deficient human cells to DNA bound platinum products which unlike thymine dimers are not recognized by an endonuclease extracted from Micrococcus leuteus. Mutat Res 1978; 51: 12l-32. Safirstein R, Zelent AZ, Gordon R. Cisplatin-nephrotoxicity: new insights into mechanisms. In: Hacker MP, Lazo JS, Tritton TR, editors. Organ directed toxicities of anticancer drugs. Boston: Martinus Nijhof, 1988: 172-89. Madias NE, Harrington JT. Platinium nephrotoxicity. Am J Med 1978; 65: 307-14. Goldstein RS, Mayor GH. Minireview: the nephrotoxicity of cisplatin. Life Sci 1983; 32: 685-90. Heidemann HTH, Gerkens JF, Jackson EK, Branch RA. Attenuation of cisplatin-induced nephrotoxicity in the rat by high salt diet, furosemide and acetazolamide. Naunyn-Schmiedeberg’s Arch Pharmacol 1985; 329: 201-5. Hayes DM, Cvitkovic E, Golbey RB, Scheiner E, Helson L, Krakoff IA. High dose cisplatinium diammine dichloride: amelioration of renal toxicity by mannitol diuresis. Cancer 1977; 39: 1372-81. tark JJ, Howell SB. Nephrotoxicity of cisplatinium [II]-dichlorodiammine. Clin Pharmacol Ther 1978; 23: 461-6. Einhorn LH, Donohue J. Cis diammine dichloroplatinium, uniblastine and gleomycine combination chemotherapy in disseminated testicular cancer. Ann Intern Med. 1977. 87:293. Cvitkovic B, Spaulding J, Bethune V, Martin J, Whitemore WF. Improvement of cis-dichlorodiammineplatinum therapeutic index in an animal model. Cancer 1977; 39: 1357-61. Pera MF Jr, Zook BC, Harder HC. Effects of mannitol or furosemide diuresis on the nephrotoxicity and physiological disposition of cis-dichlorodiammine platinum [II] in rats. Cancer Res 1979; 29: 1269-78. Litterst CL. Alterations in the toxicity of cis- dichlorodiammine platinum II and in tissue localization of platinum as a function of NaCl concentration in the vehicle of administration. Appl Pharmacol 1981; 61: 99-108. Lehane D, Winston A, Gray R, Daskal Y. The effect of diuretic pretreatment on clinical morphological and ultrastructural cis-platinium induced nephrotoxicity. Int J Radiation Oncology Biol Phys 1979; 5: 1393-9. All-Sarraf M, Fletcher W, Oishi N, Pugh R, Hewlett JS, Balducci L, McCracken J, Padilla F. Cisplatin hydration with and without mannitol diuresis in refractory disseminated malignant melanoma: a southwest oncology group study. Cancer Treat Rep 1982; 66: 31-5. Salem P, Khalyl M, Jabboury K, Hashimi L. Cis-diamminedichloroplatinium [II] by a 5-day continuous infusion. Cancer 1984; 53: 837-40. Haas A, Anderson L, Lad T. The influence of aminoglycosides on the nephrotoxicity of cis-diammine-chloroplatinium in cancer patients. J Infect Dis 1983; 147: 363. Gonzales-Vitale JC, Hayes DM, Cvitkovic E, Sternberg SS. Acute renal failure after cis-dichlorodiammine-platinium [II] and gentamicin-cephalotin therapies. Cancer Treat Rep 62: 693 Jongejan HTM, Provost AP, Molenar JC. Potentiation of cis-diamminedichloroplatinium nephrotoxicity by amikacin in rats. Cancer Chemother Pharmacol 1988; 22: 178-80. Bregman CL, Williams PD. Comparative nephrotoxicity of carboplatin and cisplatin in combination with actinomycin. Cancer Chemother Pharmacol 1986. Hirosawa A, Niitani H, Hayashibara K, Tsubo IE. Effects of sodium thiosulfate in combination therapy of cis-dichlorodiammine platinium and vindesine. Cancer Chemother Pharmacol 1989; 23: 255-8.

22. Anticancer drugs

51. 52. 53. 54. 55. 56.

57. 58. 59. 60. 61.

62. 63. 64. 65. 66. 67. 68.

69. 70. 71. 72. 73. 74. 75. 76. 77.

Howell S, Pfeifle C, Wung W, Olshen R, Lucas W, Yon J, Green M. Intraperitoneal cisplatin with systemic thiosulfate protection. Ann Intern Med 1982; 97: 845-51. Cubeddu LX, Hoffmann IS, Fuenmayor NT, Finn AL. Efficacity of odansetron and the role of serotonin in cisplatin induced nausea and vomiting. N Engl J Med 1990; 322[12]: 810-5. Boven E, Van der Vijgh WJF, Nauta MM, Schlüper HMM, Pinedo HM. Comparative activity and distribution studies of five platinium analogues in nude mice bearing human ovarian carcinoma xenografts. Cancer Research, 45:86-90, 1985 Gormley PE, Bull JM, LeRoy AF, Cysyk R. Kinetics of cis-dichlorodiammineplatinum. Clin Pharmacol Ther. 1979 Mar;25[3]:351-7 Siddik ZH, Jones M, Boxall FE, Harrap KR. Comparative distribution and excretion of carboplatin and cisplatin in mice Cancer Chemother Pharmacol. 1988;21[1]:19-24 Sharma H, Thatcher N, Baer J, Zaki A, Smith A, McAucliffe CA, Crowther D, Owens S, Fox BW. Blood clearance of radioactively labeled cis-diammine 1, 1-cyclobutane dicarboxylate platinum [II] [CBDCA] in cancer patients. Cancer Chemother Pharmacol. 1983;11[1]:5-7 Elferink F, van der Vijgh WJ, Klein I, Pinedo HM. Interaction of cisplatin and carboplatin with sodium thiosulfate: reaction rates and protein binding. Clin Chem. 1986 Apr;32[4]:641-5 Harland SJ, Newell DR, Siddik ZH, Chadwick R, Calvert AH, Harrap KR. Pharmacokinetics of cis-diammine-1, 1-cyclobutane dicarboxylate platinum[II] in patients with normal and impaired renal function. Cancer Res. 1984 Apr;44[4]:1693-7 Koeller JM, Trump DL, Tutsch KD, Earhart RH, Davis TE, Tormey DC. Phase I clinical trial and pharmacokinetics of carboplatin [NSC 241240] by single monthly 30-minute infusion. Cancer. 1986 Jan 15;57[2]:222-5 Trump DL, Elson P, Brodovsky H, Vogl SE. Carboplatin in advanced, refractory germ cell neoplasms: a phase II Eastern Cooperative Oncology Group Study. Cancer Treat Rep. 1987 Oct;71[10]:989-90 Egorin MJ, Van Echo DA, Tipping SJ, Olman EA, Whitacre MY, Thompson BW, Aisner J. Pharmacokinetics and dosage reduction of cis-diammine[1, 1-cyclobutanedicarboxylato]platinum in patients with impaired renal function. Cancer Res. 1984 Nov;44[11]:54328 Gaver RC, George AM, Deeb G. In vitro stability, plasma protein binding and blood cell partitioning of 14C-carboplatin. Cancer Chemother Pharmacol. 1987;20[4]:271-6 Harland SJ, Newell DR, Siddik ZH, Chadwick R, Calvert AH, Harrap KR. Pharmacokinetics of cis-diammine-1, 1-cyclobutane dicarboxylate platinum[II] in patients with normal and impaired renal function. Cancer Res. 1984 Apr;44[4]:1693-7 Calvert AH, Newell DR, Gumbrell LA, O’Reilly S, Burnell M, Boxall FE, Siddik ZH, Judson IR, Gore ME, Wiltshaw E. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol. 1989 Nov;7[11]:1748-56 Martinez F, Deray G, Dubois M, Beaufils H, Jacquiaud C, Bourbouze R, Benhmida M, Jaudon MC, Jacobs C. Comparative nephrotoxicity of carboplatin and cisplatin in euvolemic and dehydrated rats. Anticancer Drugs 1993; 4: 85-90. Levine BS, Henry MC, Port CD, Richter WR, Urbanek MA. Nephrotoxic potential of cis-diamminedichloroplatinum and four analogs in male Fischer 344 rats. J Natl Cancer Inst. 1981 Jul;67[1]:201-6 Canetta R, Rozencweig M, Carter SK. Carboplatin: the clinical spectrum to date. Cancer Treat Rev. 1985 Sep;12 Suppl A:125-36 Egorin MJ, Van Echo DA, Tipping SJ, Olman EA, Whitacre MY, Thompson BW, Aisner J. Pharmacokinetics and dosage reduction of cis-diammine[1, 1-cyclobutanedicarboxylato]platinum in patients with impaired renal function. Cancer Res. 1984 Nov;44[11]:54328 Reed E, Jacob J. Carboplatin and renal dysfunction. Ann Intern Med. 1989 Mar 1;110[5]:409 Gore ME, Calvert AH, Smith LE. High dose carboplatin in the treatment of lung cancer and mesothelioma: a phase I dose escalation study. Eur J Cancer Clin Oncol. 1987 Sep;23[9]:1391-7 Sleijfer DT, Smit EF, Meijer S, Mulder NH, Postmus PE. Acute and cumulative effects of carboplatin on renal function. Br J Cancer. 1989 Jul;60[1]:116-20 Schuler U, Ehninger G and Wagner T. Repeated high_dose cyclophosphamide administration in bone marrow transplantation: exposure to activated metabolites. Cancer Chemother Pharmacol. 20:248-252, 1987 Bagley CM, Bostick FW and De Wita. Clinical pharmacology of cyclophosphamide. Cancer Res. 88226-233, 1978 Struck RF, Kirk M, C, Mellet LB, el Dareer S and Hill DL. Urinary metabolites of the antitumor agent cyclophosphamide. Mol Pharmacol 7:519-529, 1971 Dooley JS, James CA, Rogers HJ and Stuart-Harris. Biliary elimination of cyclophosphamide in man. Cancer Chemother Pharmacol, 9:26-29, 1982 Juma FD, Rogers HJ, Trounce JR. Pharmacokinetics of cyclophosphamide and alkylating activity in man after intravenous and oral administration. Br J Clin Pharmacol. 8:209-217, 1979 Mouridsen HT, and Jacobsen E. Pharmacokinetics of cyclophosphamide in renal failure. Acta Pharmacol Toxicol 86:409-414, 1975

529

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108.

530

Brock N, Pohl J, and stekarJ. Studies of the urotoxicity of oxazophosphorine cytostatics and its prevention. Eur J Cancer. 17:595607, 1981 Droller MJ, Sral R and Santos G. Prevention of cyclophosphamide-induced hemorrhagic cystitis. Urology, 20:256-258, 1982 Link H, Neef V, Niethammer D and Wilms K. Prophylaxis of haemorrhagic cystatis due to cyclophosphamide-conditioning for bone marrow transplantation. Blut, 48:329-330, 1981 Ann Intern Med 106:79-94, 1987 Philips FS, Sternberg SS, Cronin AP, and Vidal PM. Cyclophosphamide and urinary bladder toxicity. Cancer Res. 21:1577-1589, 1961 Bode U, Seif SM, and Levin AS. Studies on the antidiuretic effect of cyclophosphamide: vasopressin release and sodium excretion. Med Pediatr Oncol 8: 295-303, 1980 De Fronzo RA, Colvin OM, Braine H, Robertson GL, and Davis PJ. Cyclophosphamide and the kidney. Cancer 33:483-491, 1974 Bode U, Seif SM, Levin AS. Studies on the antidiuretic effect of cyclophosphamide: vasopressin release and sodium excretion. Med Pediatr Oncol, 8:295-303, 1980 Skinner R, Sharkey IM, Pearson ADJ, et al. Ifosfamide, Mesna and nephrotoxicity in children. J Clin Oncol 11:173-190, 1993. Kaijser G, Bejnen J, Bult A, and Underberg W. Ifosfamide metabolism and pharmacokinetics. Anticancer Res 14:517-532, 1994 Boos J, Welslau U, Ritter J, Blaschke G and Schellong G. Urinary excretion of the enantiomers of Ifosfamide and its inactive metabolites in children. Cancer Chemother Pharmacol 28:455-460, 1991 Mohrmann M, Pauli A, Ritzer M, Schonfeld B, Seifert B and Brandis M. Inhibition of sodium-dependent transport systems in LLCPK1 cells by metabolites of ifosfamide. Renal Physiol Biochem 15, 289-301, 1992 Mohrmann M, Ansorge S, Schmich U, Schonfeld B, and Brandis M. Toxicity of Ifosfamide, cyclophosphamide and their metabolites in tubular cells in culture. Pediatr Nephrol 8:157-163, 1994 Zamlauski-Tucker M, Morris M, and Springate: Ifosfamide metabolite chloroacetaldehyde causes Fanconi syndrome in the perfused rat kidney. Toxicol App Pharmacol 129:170-175, 1994 Allen LM, Creaven PJ, and Nelson RL. Studies on the human pharmacokinetics of Ifosfamide [NSC-109724]. Cancer Treat Rep, 60:451-458, 1976 Springate J, Zamlauski-Tucker M, Lu H, and ChanKK. Renal Clearance of Ifosfamide. Drug Metab and Dispos, 29, 9:1081-1082, 1997 Suarez A, McDowell H, Niaudet P, Comoy E, and Flamant F. Long term follow up of Ifosfamide renal toxicity in children treated for malignant mesenchymal tumors: an international society of pediatric oncology report. J Clin Oncol, 9: 2177-2182, 1991 Pratt CB, Meyer WH, Jenkins JJ, Avery L, McKay CP, Wyatt RJ, Hancock ML. Ifosfamide, Fanconi’s syndrome and reickets. J Clin Oncol 9:1495-1499, 1991 Moncrieff M, and Foot A. Fanconi syndrome after Ifosfamide. Cancer Chemother Pharmacol, 23:121-122, 1989 Skinner R, Sharkey IM, Pearson ADJ and Craft AW. Ifosamide, mesna and nephrotoxicity in children. J Clin Oncol. 11:173-190, 1993 Brock N, Pohl J and Stekar J. Studies on urotoxicity of oxazophosphorine cytostatics and its prevention. Eur J Cancer 17:595-607, 1981 Van Dyck JJ, Falkson HC, Van der Merwe AM, and Falkson G. Unexpected toxicity in patients treated with ifosfamide. Cancer Res 32: 921-924, 1972 Falkson G, Van Dick JJ, Stapelberg R and Falkson HC. Mesnum as protector against kidney and bladder toxicity with high-dose ifosfamide treatment. Cancer Chemother Pharmacol 9:81-84, 1982 Stuart-Harris R, Harper PG, Kaye SB, and Wiltsaw E. High-dose ifosfamide by infusion with mesna in advanced soft tissue sarcoma. Cancer Treat Rep 10[suppA]:163-164, 1988 Newbury-Ecob RA, Noble VW, Barbor PRH. Ifosfamide-induced Fanconi syndrome. Lancet 2, 1328, 1989 Patterson WP, Khojasteh A. Ifosfamide-induced renal tubular defects. Cancer 63:649651, 1989 Stuart-Harris RC, Harper PJ, Parson CA and coll. High-dose alkylation therapy using Ifosfamide infusion with Mesna in the treatment of adult soft-tissue sarcoma. Cancer Chemother Pharmacol, 11:69-74, 1983 Willemse PHB, De Jong PE, Elema JD, Mulder NH. Severe renal failure following high-dose Ifosfamide and Mesna. Cancer Chemother Pharmacol, 23:329-330, 1989 Elias AD, Eder JP, Shea T, Begg C, Frei E III, Antman K. High-dose Ifosfamide with mesna uroprotection: a phase I study. J Clin Oncol, 8: 170-178, 1990 Rougier P, Mitry E. Digestion; Chemotherapy in the treatment of neuroendocrine malignant tumors, 62 Suppl 1:73-8, 2000 Bhuyan BK, Kuentzel SL, Gray LG, Fraser TJ, Wallach D and Neil GL. Tissue distribution of streptozotocin [NCS-85998]. Cancer Chemother Rep. 58:157-165, 1974

22. Anticancer drugs

109. Adolphe AB, Glasofer ED, Troetel WM, Ziegenfuss J, Stambaugh JE, Weiss AJ, and Manthei RW. Fate of streptozotocin [NSC-85998] in patients with advanced cancer. Cancer Chemother Rep. 59:547-556, 1975 110. Schein PS, O’Connell MJ, Blom J, Hubbard S, Magrath IT, Bergevin P, Wiernik JL, Ziegler PH, and DeVita VT. Clinical antitumor activity and toxicity of streptozotocin [NSC-85998]. Cancer 84:993-1000, 1974 111. Myerowitz RL, Sartiano GP, and Cavallo T. Nephrotoxic and cytoproliferative effects of streptozotocin. Cancer 88:1550-1555, 1976 112. Perry DJ, and Weiss RB. Nephrotoxicity of streptozotocin. Ann Int Med 96:122, 1982 113. Sadoff L. Nephrotoxicity of Streptozotocin [NSC-85998]. Cancer Chemother. Rep 54:457-459, 1970 114. Moertel CG, Reitemeir RJ, Schutt AJ, and Hahn RG. Phase II study of streptozotocin [NSC-85998] in the treatment of advanced gastro-intestinal cancer. Cancer Chemother Rep 55:303-307, 1971 115. Broder LE, and Carter SK. Pancreatic islet cell carcinoma. II Results of therapy with streptozotocin in 52 patients. Ann Intern Med, 79:108-118, 1973 116. Loftus L, Cuppage FE, Hoogstraten B. Clinical and pathological effects of streptozotocin. J Lab Clin Med. 84:407-413, 1974 117. Myerowitz RL, Sartiano GP, and Cavallo T. Nephrotoxic and cytoproliferative effects of streptozotocin. Report of a patient with multiple hormone-secreting islet cell carcinoma. Cancer, 38:1550-1555, 1976 118. Anderson T, Schein PS, McMenamin MG, and Gooney DA. Streptozotocin diabetes. Correlation with extend of depression of pancreatic islet nicotinamide adenine dinucleotide. J Clin Invest 54:672-677, 1974 119. Schein PS and Loftus S. Streptozotocin: depression of mouse liver pyridine nucleotides. Cancer Res 28:1501-1506, 1968 120. Horton L, Fox C, Corrin B and Sönksen PH. Streptozotocin-induced renal tumors in rat.Br J Cancer. 36:692-699, 1977 121. Rakieten N, Gordon BS, Cooney DA, Davis RD and Schein PS. Renal tumorigenesis action of streptozotocin [NSC-85998] in rats. Cancer Chemother Rep. 52:563-567, 1968 122. Dumontet C, Thieblemont C, Espinouse D, Bouafia F, Hequet O, Salles G, Coiffier B. A prospective study of intensive induction therapy with high-dose consolidation in patients with aggressive non-Hodgkin’s lymphoma and two or three adverse prognostic factors. Leukemia. 2000 Dec;14[12]:2159-65 123. Dillman RO, Soori G, Wiemann MC, Schulof R, Dobbs TW, Ruben RH, DePriest CB, Church C. Phase II trial of subcutaneous interleukin-2, subcutaneous interferon-alpha, intravenous combination chemotherapy, and oral tamoxifen in the treatment of metastatic melanoma: final results of cancer biotherapy research group 94-11. Cancer Biother Radiopharm. 2000 Oct;15[5]:48794 124. Nieder C, Grosu AL, Molls M. A comparison of treatment results for recurrent malignant gliomas. Cancer Treat Rev. 2000 Dec;26[6]:397-409 125. Engelhard HH. The role of interstitial BCNU chemotherapy in the treatment of malignant glioma. Surg Neurol. 2000 May;53[5]:45864 126. Spitzer G, Dickie KA, Litam Y et al. High-dose combination of chemotherapy with autologous bone marrow transplantation in adult solid tumors. Cancer, 45:3075-3085, 1980 127. Olivero VT, Toxicology and pharmacology of the nitrosoureas. Cancer Chemother rep. 4[3]:13-20, 1973 128. Wheeler GP, A review of studies on the mechanism of action of tne nitrosoureas. In [Sartorelli AC ed.] Cancer Chemotherapy, volume 30, Washingtoon, DC:American Chemical Society.pp87-119, 1976 129. Oliverio VT. Toxicology and pharmacology of the nitrosoureas. Cancer Chemother Rep part 3 4:13-20, 1973 130. De Vita VT, Carbone PP, Owens AH, Gold GL, Krant MJ, and Edmonson J. Clinical trials with 1, 3-bis [2-chloroethyl]-1-nitrosourea, NSC 409962. Cancer Res 25:1876-1881, 1965 131. Bagley CM, Bostick FW, DeVita VT. Clinical pharmacology of cyclophosphamide. Cancer Res. 1973 Feb;33[2]:226-33 132. Goupil A, Baglin A, Clavel B, Verger C, Fritel D. Chronic renal failure after CCNU treatment [author’s transl]. Nouv Presse Med. 1980 Nov 1;9[41]:3069-70 133. Harmon WE, Cohen HJ, Schneeberger EE, Grupe WE Chronic renal failure in children treated with methyl CCNU. N Engl J Med. 1979 May 24;300[21]:1200-3 134. Nichols WC, Moertel CG. Nephrotoxicity of methyl CCNU. N Engl J Med. 1979 Nov 22;301[21]:1181 135. Schacht RG, Feiner HD, Gallo GR, Lieberman A, Baldwin DS. Nephrotoxicity of nitrosoureas. Cancer. 1981 Sep 15;48[6]:1328-34 136. Silver HKB, and Morton DL. CCNU nephrotoxicity following sustained remission in oat cell carcinoma. Cancer Trea Rep, 63:226227, 1979 137. Berglund J, Progressive renal insuffisancy after CCNU therapy. Läkartidningen 77:1760, 1980 138. Schacht RG, Feiner HD, Gallo GR, Liebermann A and Baldwin DS. Nephrotoxicity of Nitrosoureas. Cancer 48:1328-1334, 1981 139. Liegler DG, Henderson ES, Hahn MA and Oliverio VT. 1989. The effect of organic acids on renal clearance of methotrexate on man. Clin Pharmacol Ther. 10.849-857.

531

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

140. Patel OD, Morgenthaler FR, Khazei AM, Grimaidi R and Watkins E. 1969. Methetrexate excretion patterns and renal toxicity. Arch Surg 98:305-308. 141. Balis FM, Savich JL, and Bleyer WA. 1983. Pharmacokinetics of oral methotrexate in children. Cancer Res. 43:2342-2345. 142. Shen DD and Azarnoff DL. 1978. Clinical pharmacokinetics of methotrexate. Clin Pharmacokinet. 8:1-18. 143. Breithaupt H and Kuenzlen E. 1982. Pharmacokinetics of methotrexate and 7-hydroxymethotrexate following infusions of highdose methotrexate. Cancer Treat Rep. 66:1733-1741. 144. Winograd B, Lippens RJJ, Oosterbaan MJM, Dirks MJM, Vree TB and van der Kleijn E. 1986. Renal excretion and pharmacokinetics of methotrexate and 7-hyfroxymethotrexate following a 24 hours high dose infusion of methotrexate in children. Eur J Clin Pharmacol. 30:231-238. 145. Jacobs SA, Stoller RG, Chabner BA, and Johns DG. 1977. Dose-dependent metabolism of methotrexate in man and Rhesus monkey. Cancer Treat Rep, 61:651-658. 146. Pearson ADJ. Mills S, Amineddine HA, Long DR, Craft AW and Chessels JM. 1987. Pharmacokinetics of oral and intramuscular methotrexate in children with acute lymphoblastic leukaemia. Cancer Chemother Pharmacol 20:243-247. 147. Sonneveld P, Schultz FW, Nooter K and Hahlen K. 1986. Pharmacokinetics of methotrexate and 7-hydroxymethotrexate in plasma and bone marrow of children receiving, low-doses oral methotrexate. Cancer Chemother Pharmacol 18:111-118. 148. Rieselbach RF and Garnick MB. 1988. Renal diseases indiced by antineoplastic agents. In Schrier RW, Gottschalk CW. [eds], Diseases of the kidney [Fourth edition], Boston, MA Little Brownand Company.pp 1275-1299. 149. Pittman SW, and Frei E. 1977. Weekly methotrexate-calcium leucovorin rescue: effect of alkalinization on nephrotoxicity; pharmacokinetics in the CNS, and use in CNS non Hodgkin’s lymphoma. Cancer Treat Rep. 61:695-701. 150. Glode LM, Pitman SW and Ensminger WD. 1979. A phase I study of high dose of aminopterin with leucoverin rescue in patients with advanced metastatic tumors. Cancer Res. 39:3707-3714. 151. Link DA, Fosberg MT and Ingelfinger JR. 1976. Renal toxicity of high-dose methotrexate. Pediatr Res. 10:455 [abstr]. 152. Thyss A, Milano G, Kubar J, Namer M, Schneider M. Clinical and pharmacokinetic evidence of a life-threatening interaction between methotrexate and ketoprofen. Lancet, 1986, I:256-258. 153. Kremer JM and Hamilton RA. The effects of nonsteroidal antiinflammatory drugs on methotrexate [MTX]pharmacokinetics: Impairment of renal clearance of MTX at weekly maintenance doses but not at 7.5 mg. J Rheumatol. 1995. 22:2072-2277. 154. Pitman SW, Parker LM, Tattersall MNH, Jaffe N, Frei E. Clinical trial of high-dose methotrexate with citrovorum factor-toxicologic and therapeutic observations. Cancer Chemother Rep. 1975. 6:43-49. 155. Abelson HT, Fosburg MT, Beardsley GP. Methotrexate-induced renal impairment. J Clin Oncol 1983. 1:208-216. 156. Molina R, Fabian C, Cowley B. Use of charcoal hemoperfusion wih sequential hemodialysis to reduce serum methotrexate levels in a patient with acute renal insufficiency. Am J Med. 1987. 82:350-352. 157. Relling MV, Stapelton BF, Ochs J, Jones DB, Meyer W, Wainer IW, Crom WR, McKay CP and Evans WE. Removal of methotrexate, leucoverin and their metabolites by combined hemodialysis and hemoperfusion. Cancer, 62.884-888, 1988. 158. Remuzzi G, Galbusera M, Salvadori M, et al. Bilateral nephrectomy stopped disease progression in plasma-resistant hemolytic uremic syndrome with neurological signs and coma. Kidney Int 49: 282-6, 1996. 159. Flombaum CD, Mouradian JA, Casper ES, et al. Thrombotic microangiopathy as a complication of long-term therapy with gemcitabine. Am J Kidney Dis 33: 555-62, 1999. 160. Fung MC, Storniolo AM, Nguyen B, et al. A review of hemolytic uremic syndrome in patients treated with gemcitabine therapy. Cancer 85: 2023-32, 1999. 161. Sheldon R, Slaughter D. A syndrome of microangiopathic hemolytic anemia, renal impairment, and pulmonary edema in chemotherapy-treated patients with adenocarcinoma. Cancer 58: 1428-36, 1986. 162. Humphreys BD, Soiffer RJ, Magee CC. Renal failure associated with cancer and its treatment: an update. J Am Soc Nephrol 16: 151-61, 2005. 163. Flombaum CD. Gemcitabine Nephrotoxicity and the Hemolytic Uremic Syndrome (HUS): Report of 29 Cases from a Single Institution. J Am Soc Nephrol, 2005, 777°. Poster SA-PO1008. 164. Gordon LI, Kwaan HC. Cancer- and drug-associated thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Semin Hematol 34: 140-7, 1997. 165. Murgo AJ. Thrombotic microangiopathy in the cancer patient including those induced by chemotherapeutic agents. Semin Hematol 24: 161-77, 1987. 166. Monteagudo J, Pereira A, Roig S, et al. Investigation of plasma von Willebrand factor and circulating platelet aggregating activity in mitomycin C-related hemolytic-uremic syndrome. Am J Hematol 33: 46-9, 1990. 167. Zeigler ZR, Rosenfeld CS, Andrews DF 3rd, et al. Plasma von Willebrand Factor Antigen (vWF:AG) and thrombomodulin (TM) levels in Adult Thrombotic Thrombocytopenic Purpura/Hemolytic Uremic Syndromes (TTP/HUS) and bone marrow transplantassociated thrombotic microangiopathy (BMT-TM). Am J Hematol 53: 213-20, 1996.

532

22. Anticancer drugs

168. Hollenbeck M, Kutkuhn B, Aul C, et al. Haemolytic-uraemic syndrome and thrombotic-thrombocytopenic purpura in adults: clinical findings and prognostic factors for death and end-stage renal disease. Nephrol Dial Transplant 13: 76-81, 1998. 169. Sens YA, Miorin LA, Silva HG, et al. Acute renal failure due to hemolytic uremic syndrome in adult patients. Ren Fail 19: 279-82, 1997. 170. Mergenthaler HG, Binsack T, Wilmanns W. Carcinoma-associated hemolytic-uremic syndrome in a patient receiving 5-fluorouracil-adriamycin-mitomycin C combination chemotherapy. Oncology 45: 11-4, 1988. 171. van Hazel GA, Scott M, Rubin J, Moertel CG, Eagan RT, O’Connell MJ and Kovach JS. Pharmacokinetics of mitomycin C in patients receiving the drug alone or in combination. Cancer Treat Rep, 67:805-810, 1982 172. Schilder RB, Young JD, Ratanatharathorn V, Karanes C and Baker LH. Clinical pharmacokinetics of high-dose mitomycin C. Cancer Chemother Pharmacol, 13:186-190, 1984 173. Erlichman C, Rauth AM, Battistella R and Fine S. Mitomycin C pharmacokinetics in patients with recurrent or metastatic colorectal carcinoma. Can J Physiol Pharmacol, 65:407-411, 1987 174. Lelesne JB, Rothschild N, Erickson B, Koreu S, Sisk R, Keller J, Arbus M, Wooley P, Chiazze L, Schein PS, and Neefe JR. Cancer-associated hemolytic-uremic syndrome: analysis of 85 cases from the national registry. J Clin Oncol, 7:781-789, 1989 175. Cantrell JE, Phillips TM, and Schein PS. Carcinoma-associated hemolytic-uremic syndrome: a complication of mitomycin chemotherapy. J Clin Oncol, 3:723-734, 1985 176. Hanna WT, Krauss S, Regester RF and Murphy WM. Renal disease after Mitomycin C therapy. Cancer, 48:2583-2588, 1981 177. Gulati SC, Sordillo P, Kempin S, Reich L, Magill GB, Scheiner E and Clarkson B. Microangiopathic hemolytic anemia observed after treatment of epidermoid carcinoma with mitomycin C and 5-fluorouracil. Cancer, 45:2252-2257, 1980 178. Bayle F, Vialtel P, Bastrenta F, Swiercz P, Dechelette E, Pin I, Carpentier P, Mousseau M, Schaerer R, Couderc P, Cordonnier D. [Thrombotic microangiopathy and chronic renal failure caused by mitomycin C in 4 women treated for breast cancer]. Nouv Presse Med. 1982 Jun 26;11[30]:2300-1 179. Hanna WT, Krauss S, Rolland FR, Murphy WM. Renal disease after mitomycin C therapy. Cancer, 1981, 48:2583-88. 180. LiuK, Mittelmann A, Sproul RR, and Elias EG. Renal toxicity in man treated with mitomycin C Cancer, 28:1314-1320, 1971 181. Lyman NW, Michaelson R, Viscuso RL, Winn R, Mulgaonakar S, Jacobs MG. Mitomycin-induced hemolytic-uremic syndrome. Successful treatment with corticosteroids and intense plasma Arch Intern Med. 1983 Aug;143[8]:1617-8. 182. Jackson AM, Rose BD, Graff LG, Jacobs JB, Schwartz JH, Strauss GM, Yang JP, Rudnick MR, Elfenbein IB, Narins RG. Thrombotic microangiopathy and renal failure associated with antineoplastic chemotherapy. Ann Intern Med. 1984 Jul;101[1]:41-4. 183. Rosenberg SA, Lotze MT, Muul LM, Leitman S, Chang AE, Ettinghause SE, Matory YL, Skibber JM, Shiloni E, Vetto JT, Seipp CA, Simpson CG, Reichert CM: Observations on the systemic administration of autologous lymphokine-activated killer cells and recombinant interleukin-2 to patients with metastatic cancer. N Eng J Med, 1985, 313:1485-1491. 184. Law TM, Motzer RJ, Gluck L et al. Phase III randomized trial of interleukine-2 with or without lymphokine activated killer cells in the treatment of patients with advanced renal cell carcinoma. Cancer, 1995, 76: 824-832. 185. Rozenberg SA, Lotze MT et al. A progress report on the treatment of 157 patients with advanced cancer using lymphokine activated killer cells and interleukine 2 or high dose interleukine 2 alone. N ehg J Med. 1987, 316, 889-897. 186. Thompson JA, Shulman KL, Benyunes MC, Lindgren CG, Collins C, Lange PH, Bush WH Jr, Benz LA, Fefer A. Prolonged continuous intravenous infusion interleukin-2 and lymphokine activated killer cell therapy for metastatic renal cell carcinoma. J Clin Oncol 1992, 10:960-968. 187. Guleria AS, Yang JC, Topalian SL, et al. Renal dysfunction associated with the administration of high-dose interleukin-2 in 199 consecutive patients with metastatic melanoma or renal cell carcinoma. J Clin Oncol. 1994, 12:2714-2722. 188. Webb DE, Austin HA, Belldegrun A, Vaughan E, Linehan WM, Rosenberg SA. Metabolic and renal effects of Interleukin-2 immunotherapy for metastatic cancer. Clin Nephrol. 1988, 30 [3]:141-145. 189. Philip T, Mercatello A, Negrier S, Philip I, Rebattu P, Kaemmerlin P, Gaspard M, Tognier E, Combaret V, Bijman JT, et al. Interleukine2 with and without LAK cells in renal cell carcinoma: the Lyon first-year experience with 20 patients. Cancer Treat Rev. 1989, 13 [suppA]:91-104. 190. Rosenstein M, Ettinghausen SE, Rosenberg SA: Extravasation of intravascular fluid wediated by the systemic administration of recombinant interleukin-2. J Immunol. 1986, 137:1735-1742. 191. Ponce P, Cruz J, Travassos J, Moreira P, Oliveira J, Melo-Gomes E, Gouveia J. Renal toxicity mediated by continuous infusion of recombinant interleukin-2. Nephron, 1993, 64[1]:114-118. 192. Feinfeld DA, D’Agati Y, Dutcher JP, Werfel SB, Lynn RI, Wiernik PH. Interstitial nephritis in a patient receiving adoptive immunotherapy with recombinant interleukin-2 and lymphokine-activated killer cells. Am J Nephrol. 1991, 11[6], 489-492. 193. Valsveld LT, van de Wiel-van Kamenade E, de Boer AJ, Sein JJ, Galle MP, Krediet RT, Mellief CJ, Rankin EM, Hekman A, Figdor CG. Possible role for cytotoxic lymphocytes in the pathogenesis of acute intertitial nephritis after recombinant interleukin-2 treatment for renal cell cancer. Cancer Immunol Immunother. 1993, 36[3]:210-213.

533

ISNARD-BAGNIS, LAUNAY-VACHER, KARIE & DERAY

194. Shalmi CL, Dutcher JP, Feinfeld DA, Chun KJ, Saleemi KR, Freeman LM, Lynn RI, and Wiernik PH. Acute renal dysfunction during interleukin-2 treatment: suggestion of an intrinsic renal lesion. J Clin Oncol, 1990, 8[11]: 1839-1846. 195. Wills RJ, Clinical pharmacokinetics of interferons. Clin Pharmacokinet, 1990, 19:390-399. 196. Grimbert P, Deray G, Lebon P, Faucher C, Malingre P, Raymond F, Petitclerc T, and Jacobs C. Pharmacokinetics of recombinant leukocyte interferon in a patient on continuous ambulatory peritoneal dialysis. Am J Nephrol, 1995, 15:175(2). 197. Rostaing L, Chatelut E, Payen JL, Izopet J, Thalamas C, Ton-That H, Pascal JP, Durand D, and Canal P. Pharmacokinetics of alpha IFN-2b in chronic hepatitis C virus patients undergoing hemodialysis or with normal renal function. Clinical implications. J Am Soc Nephrol. 1998, 9:2344-2348. 198. Bocci V, Pacini A, Muscettola M, Paulesu L, Pessina GP, santiano M, Viano AI. Renal filtration, absorption and catabolism of human alpha interferon. J Interferon Res, 1981, 3: 347-352 199. Bailly F, Mattei A, SiAhmed SN, Trepo C. Uncommon side-effects of interferon. J Viral Hep, 1997, 4 [suppl 1]: 89-94. 200. Gresser I, Maury M, Tovey M, Morel Maroger L, Pontillon F. Progressive glomerulonephritis in mice treated with interferon preparation at birth. Nature. 1976. 263:420-422. 201. Morel-Maroger L, Sloper JC, Vinter J, Woodrow D, Gresser I. An ultrastructural study of the development of nephritis in mice treated with interferon in the neonatal period. Lab Invest. 1978 Nov;39(5):513-22. 202. Quesada J et al. Clinical toxicity of interferons in cancer patients: a review. J Clin Oncol. 1986, 4:234. 203. Quesada JR, Talpaz M, Rios A, Kurzrock R, Gutterman JU. Clinical toxicity of interferons in cancer patients: a review. J Clin Oncol. 1986 4(2):234-43. 204. Miranda-Guardiola F, Fdez-Llama P, Badia J, Botey A, Estruch R, Darnell A, Rozman C and Revert L. Acute renal failure associated with alpha-interferon therapy for chronic hepatitis B. Nephrol. Dial Transplant, 1995, 10:1441-1443. 205. Nair S, Ernstoff M, Bahnson R, Arthur S, Johnston J, Downs M, Neuhart J, and Kirkwood J. Interferon-induced reversible acute renal failure with nephrotic syndrome. Urology, 1992, 39 2), 169-172. 206. Averbuch SD, Austin HA, Sherwin SA, Antonovych T, Bunn PA, Longo DL. Acute interstitial nephritis with the nephrotic syndrome following recombinant leukocyte a interferon therapy for mycosis fungoides. N Engl J Med. 1984 5;310(1):32-5. 207. Kurschel E, Metz-Kurschel U, Niederle N, Aulbert E. Investigations on the subclinical and clinical nephrotoxicity of interferon alpha IIB in patients with myeloproliferative syndromes. Renal Failure, 91:13:87-93. 208. Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, Berlin J, Baron A, Griffing S, Holmgren E, Ferrara N, Fyfe G, Rogers B, Ross R, Kabbinavar F.. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med. 2004; 350: 2335-42 209. Kabbinavar, F, Hurwitz, HI, Fehrenbacher, L. Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol, 21:60-5, 2003 210. Genentech: Avastin. labeling Text, FDA Approval Date: September 2005 211. Levine, RJ, Maynard, SE, Qian, C. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med, 350:672-83, 2004 212. Simon, M, Grone, HJ, Johren, O. Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and in adult kidney. Am J Physiol, 268:F240-50, 1995. 213. Kincaid-Smith P. The renal lesion of preeclampsia revisited. Am J Kidney Dis. 1991 Feb;17(2):144-8. 214. Eremina, V, Sood, M, Haigh, J. Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases. J Clin Invest, 111:707-16, 2003. 215. Mathieson, PW: How much VEGF do you need? J Am Soc Nephrol, 17:602-3, 2006. 216. Miller, KD, Chap, LI, Holmes, FA. Randomized phase III trial of capecitabine compared with bevacizumab plus capecitabine in patients with previously treated metastatic breast cancer. J Clin Oncol, 23:792-9, 2005. 217. Markowitz, GS, Appel, GB, Fine, PL. Collapsing focal segmental glomerulosclerosis following treatment with high-dose pamidronate. J Am Soc Nephrol, 12:1164-72, 2001. 218. Johnson, DH, Fehrenbacher, L, Novotny, WF. Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small cell lung cancer. J Clin Oncol, 22:2184-91, 2004. 219. Motzer RJ, Michaelson MD, Redman BG, Hudes GR, Wilding G, Figlin RA, Ginsberg MS, Kim ST, Baum CM, DePrimo SE, Li JZ, Bello CL, Theuer CP, George DJ, Rini BI. Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. J Clin Oncol. 2006 Jan 1;24(1):1624. 220. Veronese ML, Mosenkis A, Flaherty KT, Gallagher M, Stevenson JP, Townsend RR, O’Dwyer PJ. Mechanisms of hypertension associated with BAY 43-9006. J Clin Oncol. 2006 Mar 20;24(9):1363-9. 221. Rini B, Rixe O, Bukowski R, et al. (2005) AG-013736, a multitarget tyrosine kinase receptor inhibitor, demonstrates anti-tumor activity in a phase 2 study of cytokine-refractory, metastatic renal cell cancer (RCC). ASCO Proc (abstract 4509).

534

22. Anticancer drugs

222. Nguyen QD, Shah SM, Hafiz G, Quinlan E, Sung J, Chu K, Cedarbaum JM, Campochiaro PA; CLEAR-AMD 1 Study Group. A phase I trial of an IV-administered vascular endothelial growth factor trap for treatment in patients with choroidal neovascularization due to age-related macular degeneration. Ophthalmology. 2006 Sep;113(9):1522.e1-1522.e14. 223. Motzer RJ, Rini BI, Bukowski RM, Curti BD, George DJ, Hudes GR, Redman BG, Margolin KA, Merchan JR, Wilding G, Ginsberg MS, Bacik J, Kim ST, Baum CM, Michaelson MD. Sunitinib in patients with metastatic renal cell carcinoma. JAMA. 2006 Jun 7;295(21):2516-24. 224. Ratain MJ, Eisen T, Stadler WM, Flaherty KT, Kaye SB, Rosner GL, Gore M, Desai AA, Patnaik A, Xiong HQ, Rowinsky E, Abbruzzese JL, Xia C, Simantov R, Schwartz B, O’Dwyer PJ. Phase II placebo-controlled randomized discontinuation trial of sorafenib in patients with metastatic renal cell carcinoma. J Clin Oncol. 2006 Jun 1;24(16):2505-12. 225. Faivre S, Delbaldo C, Vera K, Robert C, Lozahic S, Lassau N, Bello C, Deprimo S, Brega N, Massimini G, Armand JP, Scigalla P, Raymond E. Safety, pharmacokinetic, and antitumor activity of SU11248, a novel oral multitarget tyrosine kinase inhibitor, in patients with cancer. J Clin Oncol. 2006 Jan 1;24(1):25-35. 226. Yang, R, Thomas, GR, Bunting, S. Effects of vascular endothelial growth factor on hemodynamics and cardiac performance. J Cardiovasc Pharmacol, 27:838-44, 1996. 227. Neagoe, PE, Lemieux, C ? Sirois, MG. Vascular endothelial growth factor (VEGF)-A165-induced prostacyclin synthesis requires the activation of VEGF receptor-1 and -2 heterodimer. J Biol Chem, 280:9904-12, 2005. 228. Henry, TD, Annex, BH, McKendall, GR. The VIVA trial. Vascular endothelial growth factor in Ischemia for Vascular Angiogenesis. Circulation, 107:1359-65, 2003. 229. Chen, S, Lee, JS, Iglesias-de la Cruz, MC. Angiotensin II stimulates alpha3 (IV) collagen production in mouse podocytes via TGFbeta and VEGF signalling: implications for diabetic glomerulopathy. Nephrol Dial Transplant, 20:1320-8, 2005. 230. Mares, JE, Worah, S, Mathew, SV. Increased rates of hypertension (HTN) among patients with advanced carcinoid treated with bevacizumab. the 41st Annual Meeting of the American Society of Clinical Oncology (ASCO).Abstract No: 4087, 2005. 231. Ginsberg, JM, Chang, BS, Matarese, RA. Use of single voided urine samples to estimate quantitative proteinuria. N Engl J Med, 309:1543-6, 1983. 232. Cohen E. Radiation nephropathy after bone marrow transplantation. Kidney International, 2000, 58:903-918. 233. Redd BL. Radiation nephritis: review, case report and animal study. Roentgenol. 1960, 83:88. 234. Phillips TL, Wharam MD, Margolis LW. Modification of radiation injury to normal tissues by chemotherapeutic agents. Cancer, 1975, 35:1678-1684. 235. Kelly CJ and Neilson EG. Intertitial diseases. In Brenner & Rector’s, Ed by Barry B. Brenner, WB Saunders Company, Phildelphia, 1655-1679, Vth edition, 1996. 236. Jaenke RS, Robbins MEC, Bywaters T, Whitehouse E, Rezvani M, Hopewell JW. Capillary endothelium: target site of renal radiation injury. Lab Invest 1993, 68:396-405. 237. Rubin P, Casarett GW. Clinical Radiation pathology. Philadelphia, Saunders, 1968, pp 293-333. 238. Robbins MEC, Hopewell JW: Physiological factors effecting renal radiation tolerance: A guide to the treatment of late effects. Br J Cancer, 1986, 53:265-267. 239. Cohen EP, Fish BL, Moulder JE: Treatment of radiation nephropathy with captopril. Radiat Res. 1992, 132: 346-350.

535

23

Anesthetic agents Per-Olof JARNBERG Oregon Health & Science University, Portland, Oregon, USA

Introduction ___________________________________________________________ 537 Comparative renal pharmacology of inhaled and injectable anesthetic agents ______ 538 Inhaled anesthetic agents Injectable anesthetic agents

538 538

Metabolism of inhaled anesthetic agents ____________________________________ 538 Halothane Enflurane Isoflurane Sevoflurane Desflurane

539 539 539 539 540

Mechanisms of fluoride toxicity ____________________________________________ 540 Fluoride elimination

541

Considerations in pediatric patients ________________________________________ 542 Clinical implications _____________________________________________________ 542 References _____________________________________________________________ 542

Introduction

R

enal function impairment remains a common event in connection with anesthesia and surgery. Severe perioperative renal dysfunction (SCr > 6 mg/dL, CrCl ≤ 15 ml/min) accounts for one half of all patients requiring acute dialysis [1] and is associated with a mortality in excess of 50% [2]. Mild to moderate renal function impairment is surprisingly common after surgery. In a group of 278 patients undergoing non-emergency general, vascular, or gynecological surgery, 65 of the patients developed an increase in serum creatinine levels ≥ 20% within the first six postoperative days [3]. Thirty-two patients had increases that were sustained for more than 48 hours. For half of these patients, creatinine clearance had not returned to baseline

levels by the time of discharge. In most cases, the perioperative changes in renal function are not due to the anesthetic agent itself, although some volatile anesthetics have nephrotoxic potential due to direct toxicity of their metabolites. Instead, postoperative renal failure is more commonly multifactorial. Risk factors include: 1) preexisting renal or cardiac disease, 2) the type of surgical procedure, 3) occurrence of rhabdomyolysis or hemolysis, 4) adverse hemodynamic events, 5) inappropriate fluid management, and 6) concurrent administration of potentially nephrotoxic substances such as radiographic contrast agents, aminoglycoside antibiotics, and cyclosporine. Such risk factors usually play a more important role than the anesthetic agent in the development of postoperative renal dysfunction [4].

JARNBERG

Comparative renal pharmacology of inhaled and injectable anesthetic agents Inhaled anesthetic agents Since Pringle et al. described oliguria during ether anesthesia in 1905, many investigators have focused on the effects of anesthesia on renal function [5]. All general anesthetics have significant but reversible effects on renal function. These effects are mediated directly by changes in renal vascular resistance, renal blood flow, glomerular filtration rate, and renal tubular function, or indirectly by changes in cardiovascular function and neuroendocrine activity. Modern inhaled anesthetic agents – halothane, enflurane, isoflurane, desflurane, and sevoflurane – decrease glomerular filtration rate, sodium excretion, and urine output [6-10]. Studies of the response of renal blood flow to these agents have yielded conflicting results. Initially, investigators using clearance techniques concluded that halothane and enflurane reduce renal blood flow [9, 10]. In later studies, direct measurement techniques indicated that clinical doses of inhaled agents decrease renal vascular resistance thus maintaining renal blood flow when perfusion pressure decreases during anesthesia [11-15]. These changes are transient and usually return to normal in the immediate postoperative period. In one study, even prolonged hypotension to a mean arterial pressure of 60 mm Hg induced with isoflurane was not associated with persistent derangement of renal function postoperatively [16].

Injectable anesthetic agents Sodium thiopental does not alter renal blood flow although glomerular filtration rate and urine output are moderately affected [17]. The same is true for opioids such as morphine [18] and fentanyl [19, 20] and the more recently introduced I.V. agent propofol [21]. The effects of these drugs on renal function are transient. There is no evidence that injectable anesthetic agents are associated with direct nephrotoxicity. In addition to the effects of anesthetic agents themselves, other intraoperative interventions may also influence renal function. The initiation of mechanical ventilation, especially with the application of positive end-expiratory pressure, is associated with decreases in 538

sodium excretion and urine output [22-24]. Decreased cardiac output, increased sympathetic outflow and release of renin and decreased release of atrial natriuretic peptide have all been implicated as being responsible for these changes [25, 26]. In summary, virtually all anesthetic agents and techniques are associated with reductions in glomerular filtration rate and urine output. These changes are usually readily reversed in the immediate postoperative period and represent the net effect of complex interactions between direct actions of the anesthetics on the kidney and indirect changes in cardiac output, blood pressure, and neuroendocrine function.

Metabolism of inhaled anesthetic agents Toxic effects of biodegradation products from anesthetic agents may also directly influence renal function. On rare occasions renal failure will result. Modern inhalation anesthetics are fluorinated to reduce flammability. Initially, these inhaled agents were believed to be biochemically inert. Over the past 30 years, however, research findings have demonstrated that not only are inhaled anesthetics metabolized in vivo [27], but their metabolites are also responsible for both acute and chronic toxicities [28, 29]. Therefore, the use of some anesthetics has been discontinued, including methoxyflurane because of its nephrotoxicity; and other anesthetics are more selectively used, e.g. halothane due to a rare incidence of liver toxicity. Studies have also provided the impetus to develop new agents – isoflurane and desflurane – with properties that lower their toxic potential. The result has been improved safety, but there is room for further improvement as our insight into toxicological mechanisms expands. Metabolism of inhaled anesthetics usually begins with oxidation and is carried out by cytochrome P-450 enzymes located in the microsomes of the liver and the kidneys [30, 31]. Under certain circumstances, some agents, such as halothane, might also undergo reduction. In addition to their primary metabolism, some anesthetics, sevoflurane for instance, also undergo phase II conjugation reactions prior to excretion. The cytochrome P-450 enzyme system is comprised of multiple isoenzymes that are inducible to varying degrees [32]. These two characteristics are major determinants of metabolic pathways and rates. Induction can be caused by exposure to one or more of a large

23. Anesthetic agents

variety of compounds, including ethanol, phenobarbital, cimetidine, phenytoin, isoniazid, and some volatile anesthetics. Both transcriptional and translational processes are stimulated by the inducer to produce cytochrome P-450 [33, 34]. Expression of the various isoenzymes depends not only on induction, but also on such factors as sex, obesity, fasting, and diabetes. For example in Streptozotocin- induced diabetes in rats, P-4502E increases several fold, causing an enhanceds enflurane and isoflurane metabolism [35]. Most halogenated anesthetics are similar in composition; however, they vary greatly in their rate and pathway of metabolism. Minor alterations in configuration can be associated with major changes in metabolism. Also, their degree of lipid solubility, which governs the drug’s access to and duration at metabolizing enzyme sites, is important in determining metabolic rate and the amount of drug that is biotransformed.

difluoroacetic acid, fluoride ions are also released in sufficient quantity to merit some concern about renal function [44]. Plasma inorganic fluoride concentrations after clinical enflurane anesthesia are usually in the 15-25 μM range [8, 9, 44]. Longer procedures [45] and obesity [46] are associated with higher postanesthetic fluoride levels. A study of surgical patients with preanesthetic chronic consumption of enzyme-inducing drugs such as phenobarbital, phenytoin, diazepam, and ethanol did not reveal increased plasma fluoride levels compared to untreated patients [47]. In contrast, about 50% of surgical patients on chronic isoniazid therapy demonstrated significantly elevated plasma fluoride concentrations after enflurane anesthesia [48]. Enflurane is the only modern inhaled anesthetic that may be linked to fluoride-induced renal failure in a very limited number of cases [49, 50], but this has not yet been proven.

Halothane

Isoflurane

Halothane (CF3CHBrCl), the first of the modern halogenated volatile anesthetics, was introduced into clinical practice in 1956. It is normally metabolized in an oxidative pathway forming bromide ions and trifluoroacetic acid, neither of which has potential for tissue toxicity [36, 37]. Reductive metabolism of halothane takes place during low oxygen tension states in the liver [38]. This pathway has been linked to halothane-induced liver necrosis through production of free radicals that bind to cellular macromolecules [39, 40]. Reductive metabolism is also associated with production of fluoride ions [41], although the quantities produced are too small to have nephrotoxic importance. The extent of halothane metabolism has been reported to be 17-20% of an administered dose [36]. Oxidation to trifluoroacetic acid is the principal pathway of halothane metabolism, and no fluoride ions are released. Therefore, fluoride-induced renal toxicity is not a concern with halothane.

Isoflurane (CHF2OCH2ClCF3) is an isomer of enflurane and has been in clinical use for about twenty years. It has a very low degree of defluorination [51]. Approximately 0.2-0.4% of a given dose is metabolized. Fluoride levels in humans after isoflurane anesthesia peak at 4-6 μM, which represents only a modest rise over basal fluoride levels. Although enzyme induction increases defluorination somewhat, it is not associated with plasma fluoride concentrations of clinical significance [52, 53].

Enflurane Enflurane (CHF2OF2CHClF) has been in clinical use for the last three decades. It is metabolized to a much lesser degree than halothane. Approximately 2-3% of a given dose undergoes biodegradation [42, 43]. Although the chief metabolite is difluoromethoxy-

Sevoflurane Sevoflurane [(CF3)2CHOCH2F] was first used in Japan and introduced into American clinical practice in 1995. Sevoflurane is defluorinated to approximately the same extent as enflurane. In fact, initial studies reported that plasma levels of fluoride associated with sevoflurane anesthesia are comparable to those seen after enflurane administration [54, 55]. More recent studies, however, report that plasma fluoride concentrations often rise above 50 mM [56, 57]. Due to sevoflurane’s low blood/gas solubility, only limited quantities build up during anesthesia; and, as a result, fluoride levels fall very quickly after termination of anesthesia. In vivo, defluorination in rats is increased by pretreatment with phenobarbital [58]. 539

JARNBERG

Numerous studies have addressed the same issues raised with enflurane regarding fluoride production and nephrotoxic potential including fluoride levels after prolonged exposure [56, 57, 59], urine concentrating ability [57, 59-61], the effect of obesity [57, 60], and the effect of preexisting renal function impairment [62, 63]. The findings demonstrated that sevoflurane has little or no potential for fluoride-induced nephrotoxicity (For further information see section on mechanisms of fluoride toxicity). Sevoflurane undergoes degradation in the presence of soda lime and barium hydroxide lime, both of which are used in modern anesthesia machines for CO2 absorption. The chief degradation product is fluoromethyl-2,2-difluoro-1(trifluoromethyl)-vinyl ether, also called compound A [64]. In an anesthesia circle circuit using the above absorbents, compound A concentrations correlate directly with sevoflurane concentrations and absorbent temperature, and inversely with the inflow rate of fresh gas [65]. Increasing inflow rates reduces compound A concentrations by decreasing the rebreathing of gas that has passed through the absorbent, thereby decreasing the amount of CO2 that reaches the absorbent. The amount of CO2 absorbed determines the temperature of the absorbent, since CO2 absorption is an exothermic reaction [65, 66]. Compound A is nephrotoxic in rats at thresholds estimated at 180 ppm/hour [67]. Renal toxicity is characterized histologically by proximal tubular cell degeneration and necrosis in the corticomedullar region of the kidney and biochemically by proteinuria, glucosuria, and enzymuria (NAG and α-GST) with increased serum creatinine and BUN concentrations occurring with severe toxicity [67-70]. In humans, there seems to be a dose-dependent association between compound A exposure and the appearance of urinary biomarkers such as albumin, glucose, and the enzymes NAG or α-GST. These findings appear in studies when the compound A exposure exceeds 160 ppm/hour [71-74], while they are absent in studies with lower compound A exposure [75-77]. In all studies associated with higher exposure of compound A, the urinary markers are transient, lasting 3-5 days with total normalization within one week. There is no correlation between serum creatinine and the urinary markers. In summary, the available information indicates that sevoflurane anesthesia is nontoxic to the kidney 540

as long as exposure to compound A is kept below 150 ppm/hour. However, there are significant questions regarding the potential for compound A to cause renal injury: Are larger doses than 160 ppm/hour harmful? Do they cause histologically detectable tissue damage? Is there a cumulative effect of repeated exposures? Are particular patients more prone to injury? The concerns and questions surrounding the degradation of sevoflurane by CO2 absorbents to toxic compounds would disappear if the use of absorbents that minimally degrade sevoflurane became standard [78-80]. Such absorbents exist, and they do not contain sodium or potassium hydroxide. An example is Amsorb® (Armstong Medica Ltd., Coleraine, Northern Ireland). It is completely inert when brought into contact with sevoflurane [81]. This absorbent is widely used in Europe and is commercially available in the US. Medisorb®, Spherasorb®, Loflosorb®, and Superia® are other examples of CO2 absorbents which produce little or no sevoflurane degradation. [81a].

Desflurane Desflurane (CHF2OCHFCF3) has been in clinical use in the US for more than a decade. It has very low lipid solubility [82] and is highly resistant to metabolism and to degradation in soda lime [83]. Data from studies in rats and humans suggest that desflurane is not toxic to the liver or kidney [84-86]. Serum inorganic fluoride concentrations do not rise above background levels even after prolonged exposure [87, 88]. Since desflurane has a boiling point of 23.5°C, it requires a special vaporizer to ensure a stable output.

Mechanisms of fluoride toxicity For more than forty years, the potential for nephrotoxicity, particularly when fluoride induced, has influenced every aspect of the development of new inhaled anesthetics. This concern is based on the experience with methoxyflurane, which was introduced in the US in 1960 [89]. The exact mechanism(s) responsible for fluoride nephrotoxicity have not been defined. The fluoride ion interferes with normal cell function on several levels. Fluoride inhibits several cellular enzyme systems and diminishes tissue respiration and anaerobic glycolysis [90]. The lethal dose of sodium fluoride in humans is approximately 5 g [90]. In the kidney,

23. Anesthetic agents

fluoride interferes with the transport of sodium in the proximal convoluted tubule. It also inhibits adenylate cyclase in the collecting tubules and diminishes the action of antidiuretic hormone. Experimental evidence in rats indicates that the chloride-dependent pump in the thick ascending part of Henle’s loop is also inhibited [91]. In cultures of human collecting duct cells, exposure to fluoride ions inhibits Na-K-ATPase and causes morphologic changes in mitochondria [92]. In 1966, renal failure was reported in 13 of 41 patients receiving methoxyflurane anesthesia for abdominal surgery [93]. The cause was later associated with methoxyflurane metabolism and increased plasma fluoride levels [94]. Methoxyflurane undergoes oxidative metabolism by cytochrome P450, and inorganic fluoride ions are released [95]. The clinical manifestations of this process consist of vasopressin-resistant polyuria, hypernatremia, hyperosmolality, and azotemia. The degree of nephrotoxicity is positively correlated with plasma fluoride levels. Subclinical toxicity occurs at fluoride levels of 50-80 μM, while fluoride concentrations of 90-120 μM are associated with established renal failure that becomes severe when levels reach 150-175 μM [96]. When animals are injected with inorganic fluoride, the changes in renal function are similar to those seen after the administration of methoxyflurane. The dose of sodium fluoride required to cause nephrotoxicity, however, results in much higher fluoride levels (>400 μM) than those seen after methoxyflurane anesthesia [97, 98]. Despite this observation, the conclusion was drawn that nephrotoxicity from methoxyflurane must be caused by metabolically-released inorganic fluoride ions. This hypothesis was subsequently generalized to include all fluoride-containing volatile anesthetics, and 50 μM of fluoride was considered the nephrotoxic threshold. Plasma fluoride levels exceeding 50 μM or even 100 μM following administration of sevoflurane are not associated with renal damage. This lack of correlation between peak serum fluoride levels and nephrotoxicity led one investigator to suggest that intrarenal production of fluoride is more important in the etiology of nephrotoxicity than the blood levels resulting from hepatic production of fluoride [99]. Indeed, during the last decade, we have learned that hepatic defluorination and blood transport of fluoride to the kidney is not the mechanism responsible for volatile anesthetic nephrotoxicity and that neither plasma fluoride concentrations greater than 50 μM nor

the duration of fluoride increase have implications for renal toxicity [99-101]. Based on the seminal work of Evan Kharasch, it now seems clear that nephrotoxicity of inhaled anesthetics is agent-specific (methoxyflurane) and caused by an organic methoxyflurane metabolite in combination with fluoride, rather than by metabolic fluoride generation alone [102, 103]. Methoxyflurane is metabolized by two different pathways [95, 104]. Oxidative dechlorination of the chloromethyl carbon produces 2,2-difluoro-2-methoxyacetic acid (MDFA). Oxidative O-demethylation of the methoxy group results in formation of fluoride and dichloroacetic acid (DCAA). Experiments in rats revealed no functional or histologic signs of nephrotoxicity when either MDFA or DCAA was administered intraperitoneally. Fluoride in combination with DCAA, but not with MDFA, resulted in significant dose-dependent histologic (necrosis) and functional renal injury [103]. Fluoride administered alone in varying doses resulting in a total 4-day urine recovery of fluoride equal or greater than that after methoxyflurane anesthesia caused reduced urine osmolality and significant diuresis at the highest dose [103]. These findings may explain why increased fluoride formation from methoxyflurane, but not other anesthetics, is associated with nephrotoxicity and may have implications for the importance of volatile anesthetic defluorination, future development of inhaled anesthetic agents, and the laboratory methods used to evaluate potential toxicity.

Fluoride elimination Fluoride is removed from plasma by urinary excretion [105] and uptake into calcified tissues [106]. Normally, each mechanism represents about 50% of the removal [107]. Renal fluoride excretion begins with glomerular filtration which is followed by variable tubular reabsorption. The tubular reabsorption is influenced by tubular fluid flow rate [108] and urinary pH [109, 110]. Manipulation of urinary pH in patients undergoing a standard enflurane anesthetic resulted in plasma fluoride levels that were 50% lower in patients with alkaline urine than in patients with acidic urine [111]. Bone uptake may also influence plasma fluoride concentrations. Studies in rats have demonstrated that metabolic acidosis increases the rate of bone resorption 541

JARNBERG

while metabolic alkalosis increases the rate of osseous accretion [112].

Considerations in pediatric patients Renal function is markedly diminished in neonates because of low perfusion pressure and immature glomerular and tubular function. Nearly complete maturation of glomerular filtration and tubular function occurs by approximately 20 weeks after birth in term infants but is delayed in premature infants. Complete maturation of renal function occurs by approximately two years of age [113, 114]. The ability to excrete potentially nephrotoxic degradation products associated with anesthesia may, therefore, be impaired in neonates and small children. Halothane and sevoflurane are commonly used for inhaled induction of anesthesia in children because they do not have a noxious smell. These drugs and isoflurane or desflurane are then used to maintain anesthesia, according to the preference of the anesthesiologist. Enflurane is rarely used today because it irritates the airway [115]. Therefore, of the inhaled agents currently used in pediatric patients, only sevoflurane has nephrotoxic potential. In two studies of children undergoing sevoflurane anesthesia, mean plasma fluoride levels were 15.8 μM and 21.5 μM in ages 1-12 years and 3 months-7 years, respectively [116, 117]. The latter study also reported on compound A levels in the breathing system. Maximum

inspired concentration was 5.4 ± 4.4 ppm (mean ± SD), and maximum expired concentration was 3.7 ± 2.7 ppm. There were no changes in serum creatinine values from samples obtained 24 hours postanesthesia compared with the control. These limited studies give no reason for concern about an increased risk for nephrotoxicity from sevoflurane in the pediatric population.

Clinical implications Sevoflurane is the only volatile anesthetic that has nephrotoxic potential due to biodegradation by the CO2 absorbents currently used in anesthesia circuits. Sevoflurane has been used in patients with moderate renal function impairment ( average CrCl 30-32 ml/ min, CKD3/4) without worsening of renal failure [62, 63]. To minimize the risks with compound A formation from sevoflurane, it seems prudent to follow the Food and Drug Administration’s (FDA) recommendations. The FDA warns against administration of sevoflurane at fresh gas flows < 1 L/min. Fresh gas flows at 1-2 L/ min are limited to a total of 2 MAC hours after which the recommended flow rate is 2 L or more. However, since its introduction to the US in 1995, sevoflurane has been given to tens of millions of patients without a single report of nephrotoxicity [118]. Isoflurane and desflurane do have no known nephrotoxic properties and are excellent choices for anesthetizing patients with preexisting renal disease.

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

542

Kasiske BL, Kjellstrand CM. Perioperative management of patients with chronic renal failure and postoperative acute renal failure. Urol Clin North Am 1983;10(1):35-50. Metnitz PG, Krenn CG, Steltzer H, Lang T, Ploder J, Lenz K, et al. Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients. Crit Care Med 2002;30(9):2051-8. Charlson ME, MacKenzie CR, Gold JP, Shires GT. Postoperative changes in serum creatinine. When do they occur and how much is important? Ann Surg 1989;209(3):328-33. Coggins C, Fang LS-T. Acute renal failure associated with antibiotics, anesthetic agents, and radiographic contrats agents. In: Brenner BM, Lazarus, JM, editor. Acute Renal Failure: Churchill Livingstone; 1988. p. 295-352. Pringle H MR, Pringle S. Clinical effects of ether anaesthesia on renal activity. Brit Med J 1905;2:542-545. Blackmore WP EK, Wiegand OF, Lipsey R. Renal and cardiovascular effects of halothane. Anesthesiology 1960;27:793-803. Mazze RI, Cousins MJ, Barr GA. Renal effects and metabolism of isoflurane in man. Anesthesiology 1974;40(6):536-42. Cousins MJ, Greenstein LR, Hitt BA, Mazze RI. Metabolism and renal effects of enflurane in man. Anesthesiology 1976;44(1):4453. Jarnberg PO, Ekstrand J, Irestedt L, Santesson J. Renal function and fluoride formation and excretion during enflurane anaesthesia. Acta Anaesthesiol Scand 1979;23(5):444-52.

23. Anesthetic agents

10. 11. 12. 13. 14. 15.

16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

29. 30. 31. 32. 33. 34. 35. 36. 37.

Deutsch S, Goldberg M, Stephen GW, Wu WH. Effects of halothane anesthesia on renal function in normal man. Anesthesiology 1966;27(6):793-804. Bastron RD, Pyne JL, Inagaki M. Halothane-induced renal vasodilation. Anesthesiology 1979;50(2):126-31. Priano LL. Effect of halothane on renal hemodynamics during normovolemia and acute hemorrhagic hypovolemia. Anesthesiology 1985;63(4):357-63. Gelman S, Fowler KC, Smith LR. Regional blood flow during isoflurane and halothane anesthesia. Anesth Analg 1984;63(6):55765. Jarnberg PO, Marrone, B, Priano, LL. Enflurane preserves renal blood flow. Anesthesiology 1990;73:A572. Hysing ES, Chelly JE, Doursout MF, Merin RG. Comparative effects of halothane, enflurane, and isoflurane at equihypotensive doses on cardiac performance and coronary and renal blood flows in chronically instrumented dogs. Anesthesiology 1992;76(6):97984. Lessard MR, Trepanier CA. Renal function and hemodynamics during prolonged isoflurane-induced hypotension in humans. Anesthesiology 1991;74(5):860-5. Cho KW KS, Koh GY, Seul KH, Kim HJ, Song HS. Renalm and renin effects of sodium thiopental in rabbits. Ren Physiol 1987;10:261271. Bidwai AV, Stanley TH, Bloomer HA, Blatnick RA. Effects of anesthetic doses of morphine on renal function in the dog. Anesth Analg 1975;54(3):357-60. Hunter JM, Jones RS, Utting JE. Effect of anaesthesia with nitrous oxide in oxygen and fentanyl on renal function in the artificially ventilated dog. Br J Anaesth 1980;52(3):343-8. Priano LL. Effects of high-dose fentanyl on renal haemodynamics in conscious dogs. Can Anaesth Soc J 1983;30(1):10-8. Jarnberg PO, Leyden P, Woods L. Renal blood flow is maintained during propofol anesthesia in dogs. Anesthesiology 1992;77: A655. Baratz RA, Philbin DM, Patterson RW. Plasma antidiuretic hormone and urinary output during continuous positive-pressure breathing in dogs. Anesthesiology 1971;34(6):510-3. Hall SV, Johnson EE, Hedley-Whyte J. Renal hemodynamics and function with continuous positive-pressure ventilation in dogs. Anesthesiology 1974;41(5):452-61. Jarnberg PO, de Villota ED, Eklund J, Granberg PO. Effects of positive end-expiratory pressure on renal function. Acta Anaesthesiol Scand 1978;22(5):508-14. Kharasch ED, Yeo KT, Kenny MA, Buffington CW. Atrial natriuretic factor may mediate the renal effects of PEEP ventilation. Anesthesiology 1988;69(6):862-9. Andrivet P, Adnot S, Brun-Buisson C, Chabrier PE, Darmon JY, Braquet P, et al. Involvement of ANF in the acute antidiuresis during PEEP ventilation. J Appl Physiol 1988;65(5):1967-74. Van Dyke R CM, Poznak AV. Metabolism of volatile anesthetics: 1. Conversion in vivo of several anesthetics to 14 CO2 and chloride. Biochem Pharmacol 1964;13:1239-1248. Subcommittee on the National Halothane Study of the Committee on Anesthesia. National Academy of Sciences-National Research Council. Summary of the national halothane study: possible association halothane anesthesia and postoperative necrosis. JAMA 1966;197:775-83. Mazze RI, Trudell JR, Cousins MJ. Methoxyflurane metabolism and renal dysfunction: clinical correlation in man. Anesthesiology 1971;35(3):247-52. Thummel KE, Kharasch ED, Podoll T, Kunze K. Human liver microsomal enflurane defluorination catalyzed by cytochrome P-450 2E1. Drug Metab Dispos 1993;21(2):350-7. Pacifici GM, Viani A, Franchi M, Gervasi PG, Longo V, Di Simplicio P, et al. Profile of drug-metabolizing enzymes in the cortex and medulla of the human kidney. Pharmacology 1989;39(5):299-308. Nebert DW, Adesnik M, Coon MJ, Estabrook RW, Gonzalez FJ, Guengerich FP, et al. The P450 gene superfamily: recommended nomenclature. DNA 1987;6(1):1-11. Waxman DJ. Interactions of hepatic cytochromes P-450 with steroid hormones. Regioselectivity and stereospecificity of steroid metabolism and hormonal regulation of rat P-450 enzyme expression. Biochem Pharmacol 1988;37(1):71-84. Conney AH. Pharmacological implications of microsomal enzyme induction. Pharmacol Rev 1967;19(3):317-66. Pantuck EJ, Pantuck CB, Conney AH. Effect of streptozotocin-induced diabetes in the rat on the metabolism of fluorinated volatile anesthetics. Anesthesiology 1987;66(1):24-8. Rehder K, Forbes J, Alter H, Hessler O, Stier A. Halothane biotransformation in man: a quantitative study. Anesthesiology 1967;28(4):711-5. Cohen EN, Trudell JR, Edmunds HN, Watson E. Urinary metabolites of halothane in man. Anesthesiology 1975;43(4):392-401.

543

JARNBERG

38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66.

544

Lind RC, Gandolfi AJ, Sipes IG, Brown BR, Jr., Waters SJ. Oxygen concentrations required for reductive defluorination of halothane by rat hepatic microsomes. Anesth Analg 1986;65(8):835-9. Wood CL, Gandolfi AJ, Van Dyke RA. Lipid binding of a halothane metabolite. Relationship to lipid peroxidation in vitro. Drug Metab Dispos 1976;4(4):305-13. Trudell JR, Bosterling B, Trevor AJ. Reductive metabolism of halothane by human and rabbit cytochrome P-450. Binding of 1chloro-2,2,2-trifluoroethyl radical to phospholipids. Mol Pharmacol 1982;21(3):710-7. Van Dyke RA, Gandolf AJ. Anaerobic release of fluoride from halothane. Relationship to the binding of halothane metabolites to hepatic cellular constituents. Drug Metab Dispos 1976;4(1):40-4. Chase RE, Holaday DA, Fiserova-Bergerova V, Saidman LJ, Mack FE. The biotransformation of ethrane in man. Anesthesiology 1971;35(3):262-7. Burke TR, Jr., Branchflower RV, Lees DE, Pohl LR. Mechanism of defluorination of enflurane. Identification of an organic metabolite in rat and man. Drug Metab Dispos 1981;9(1):19-24. Maduska AL. Serum inorganic fluoride levels in patients receiving enflurane anesthesia. Anesth Analg 1974;53(3):351-3. Mazze RI, Calverley RK, Smith NT. Inorganic fluoride nephrotoxicity: prolonged enflurane and halothane anesthesia in volunteers. Anesthesiology 1977;46(4):265-71. Rice SA, Fish KJ. Anesthetic metabolism and renal function in obese and nonobese Fischer 344 rats following enflurane or isoflurane anesthesia. Anesthesiology 1986;65(1):28-34. Dooley JR, Mazze RI, Rice SA, Borel JD. Is enflurane defluorination inducible in man? Anesthesiology 1979;50(3):213-7. Mazze RI, Woodruff RE, Heerdt ME. Isoniazid-induced enflurane defluorination in humans. Anesthesiology 1982;57(1):5-8. Loehning RW, Mazze RI. Possible nephrotoxicity from enflurane in a patient with severe renal disease. Anesthesiology 1974;40(2):203-5. Eichhorn JH, Hedley-Whyte J, Steinman TI, Kaufmann JM, Laasbert LH. Renal failure following enflurane anesthesia. Anesthesiology 1976;45(5):557-60. Holaday DA, Fiserova-Bergerova V, Latto IP, Zumbiel MA. Resistance of isoflurane to biotransformation in man. Anesthesiology 1975;43(3):325-32. Mazze RI, Hitt BA, Cousins MJ. Effect of enzyme induction with phenobarbital on the in vivo and in vitro defluorination of isoflurane and methoxyflurane. J Pharmacol Exp Ther 1974;190(3):523-9. Rice SA, Talcott RE. Effects of isoniazid treatment on selected hepatic mixed-function oxidases. Drug Metab Dispos 1979;7(5):2602. Holaday DA, Smith FR. Clinical characteristics and biotransformation of sevoflurane in healthy human volunteers. Anesthesiology 1981;54(2):100-6. Fujii K, Morio, M, Kikuchi,. Pharmacokinetic study in excretion of inorganic fluoride, a metabolite of sevoflurane. Hiroshima J Med Sci 1987;36:89-94. Kobayashi Y, Ochiai R, Takeda J, Sekiguchi H, Fukushima K. Serum and urinary inorganic fluoride concentrations after prolonged inhalation of sevoflurane in humans. Anesth Analg 1992;74(5):753-7. Frink EJ, Jr., Malan TP, Jr., Isner RJ, Brown EA, Morgan SE, Brown BR, Jr. Renal concentrating function with prolonged sevoflurane or enflurane anesthesia in volunteers. Anesthesiology 1994;80(5):1019-25. Cook TL, Beppu WJ, Hitt BA, Kosek JC, Mazze RI. A comparison of renal effects and metabolism of sevoflurane and methoxyflurane in enzyme-induced rats. Anesth Analg 1975;54(6):829-35. Nishiyama T, Hirasaki A. Effects of sevoflurane anaesthesia on renal function--duration of administration and area under the curve and rate of decrease of serum inorganic fluoride. Eur J Anaesthesiol 1995;12(5):477-82. Frink EJ, Jr., Malan TP, Jr., Brown EA, Morgan S, Brown BR, Jr. Plasma inorganic fluoride levels with sevoflurane anesthesia in morbidly obese and nonobese patients. Anesth Analg 1993;76(6):1333-7. Higuchi H, Arimura S, Sumikura H, Satoh T, Kanno M. Urine concentrating ability after prolonged sevoflurane anaesthesia. Br J Anaesth 1994;73(2):239-40. Conzen PF, Nuscheler M, Melotte A, Verhaegen M, Leupolt T, Van Aken H, et al. Renal function and serum fluoride concentrations in patients with stable renal insufficiency after anesthesia with sevoflurane or enflurane. Anesth Analg 1995;81(3):569-75. Tsukamoto N, Hirabayashi Y, Shimizu R, Mitsuhata H. The effects of sevoflurane and isoflurane anesthesia on renal tubular function in patients with moderately impaired renal function. Anesth Analg 1996;82(5):909-13. Hanaki C, Fujii K, Morio M, Tashima T. Decomposition of sevoflurane by sodalime. Hiroshima J Med Sci 1987;36(1):61-7. Fang ZX, Eger EI, 2nd. Factors affecting the concentration of compound A resulting from the degradation of sevoflurane by soda lime and Baralyme in a standard anesthetic circuit. Anesth Analg 1995;81(3):564-8. Bito H, Ikeda K. Effect of total flow rate on the concentration of degradation products generated by reaction between sevoflurane and soda lime. Br J Anaesth 1995;74(6):667-9.

23. Anesthetic agents

67.

Gonsowski CT, Laster MJ, Eger EI, 2nd, Ferrell LD, Kerschmann RL. Toxicity of compound A in rats. Effect of a 3-hour administration. Anesthesiology 1994;80(3):556-65. 68. Keller KA, Callan C, Prokocimer P, Delgado-Herrera L, Friedman MB, Hoffman GM, et al. Inhalation toxicity study of a haloalkene degradant of sevoflurane, Compound A (PIFE), in Sprague-Dawley rats. Anesthesiology 1995;83(6):1220-32. 69. Morio M, Fujii K, Satoh N, Imai M, Kawakami U, Mizuno T, et al. Reaction of sevoflurane and its degradation products with soda lime. Toxicity of the byproducts. Anesthesiology 1992;77(6):1155-64. 70. Kharasch ED, Thorning D, Garton K, Hankins DC, Kilty CG. Role of renal cysteine conjugate beta-lyase in the mechanism of compound A nephrotoxicity in rats. Anesthesiology 1997;86(1):160-71. 71. Goldberg ME, Cantillo J, Gratz I, Deal E, Vekeman D, McDougall R, et al. Dose of compound A, not sevoflurane, determines changes in the biochemical markers of renal injury in healthy volunteers. Anesth Analg 1999;88(2):437-45. 72. Higuchi H, Sumita S, Wada H, Ura T, Ikemoto T, Nakai T, et al. Effects of sevoflurane and isoflurane on renal function and on possible markers of nephrotoxicity. Anesthesiology 1998;89(2):307-22. 73. Eger EI, 2nd, Koblin DD, Bowland T, Ionescu P, Laster MJ, Fang Z, et al. Nephrotoxicity of sevoflurane versus desflurane anesthesia in volunteers. Anesth Analg 1997;84(1):160-8. 74. Eger EI, 2nd, Gong D, Koblin DD, Bowland T, Ionescu P, Laster MJ, et al. Dose-related biochemical markers of renal injury after sevoflurane versus desflurane anesthesia in volunteers. Anesth Analg 1997;85(5):1154-63. 75. Bito H, Ikeuchi Y, Ikeda K. Effects of low-flow sevoflurane anesthesia on renal function: comparison with high-flow sevoflurane anesthesia and low-flow isoflurane anesthesia. Anesthesiology 1997;86(6):1231-7. 76. Kharasch ED, Frink EJ, Jr., Zager R, Bowdle TA, Artru A, Nogami WM. Assessment of low-flow sevoflurane and isoflurane effects on renal function using sensitive markers of tubular toxicity. Anesthesiology 1997;86(6):1238-53. 77. Ebert TJ, Messana LD, Uhrich TD, Staacke TS. Absence of renal and hepatic toxicity after four hours of 1.25 minimum alveolar anesthetic concentration sevoflurane anesthesia in volunteers. Anesth Analg 1998;86(3):662-7. 78. Forster H, Dudziak R. [Causes for the reaction between dry soda lime and halogenated inhalation anesthetics]. Anaesthesist 1997;46(12):1054-63. 79. Forster H, Warnken UH, Asskali F. [Various reactions of sevoflurane with the individual components of soda lime]. Anaesthesist 1997;46(12):1071-5. 80. Neumann MA, Laster MJ, Weiskopf RB, Gong DH, Dudziak R, Forster H, et al. The elimination of sodium and potassium hydroxides from desiccated soda lime diminishes degradation of desflurane to carbon monoxide and sevoflurane to compound A but does not compromise carbon dioxide absorption. Anesth Analg 1999;89(3):768-73. 81. Murray JM, Renfrew CW, Bedi A, McCrystal CB, Jones DS, Fee JP. Amsorb: a new carbon dioxide absorbent for use in anesthetic breathing systems. Anesthesiology 1999;91(5):1342-8. 81a. Keijzer C, Perez, RSGM, De Lange, JJ. Compound A and carbon monoxide production from sevoflurane and seven different types of carbon dioxide absorbent in a patient model Acta Anaesth Scand 2007;51(1):31-37. 82. Koblin DD, Eger EI, 2nd, Johnson BH, Konopka K, Waskell L. I-653 resists degradation in rats. Anesth Analg 1988;67(6):534-8. 83. Eger EI, 3rd. Stability of I-653 in soda lime. Anesth Analg 1987;66(10):983-5. 84. Eger EI, 2nd, Johnson BH, Strum DP, Ferrell LD. Studies of the toxicity of I-653, halothane, and isoflurane in enzyme-induced, hypoxic rats. Anesth Analg 1987;66(12):1227-9. 85. Eger EI, 2nd, Johnson BH, Ferrell LD. Comparison of the toxicity of I-653 and isoflurane in rats: a test of the effect of repeated anesthesia and use of dry soda lime. Anesth Analg 1987;66(12):1230-3. 86. Jones RM, Koblin DD, Cashman JN, Eger EI, 2nd, Johnson BH, Damask MC. Biotransformation and hepato-renal function in volunteers after exposure to desflurane (I-653). Br J Anaesth 1990;64(4):482-7. 87. Koblin DD, Weiskopf RB, Holmes MA, Konopka K, Rampil IJ, Eger EI, 2nd, et al. Metabolism of I-653 and isoflurane in swine. Anesth Analg 1989;68(2):147-9. 88. Sutton TS, Koblin DD, Gruenke LD, Weiskopf RB, Rampil IJ, Waskell L, et al. Fluoride metabolites after prolonged exposure of volunteers and patients to desflurane. Anesth Analg 1991;73(2):180-5. 89. Artusio JF, Jr., Van Poznak A, Hunt RE, Tiers RM, Alexander M. A clinical evaluation of methoxyflurane in man. Anesthesiology 1960;21:512-7. 90. Haynes RJ. Agents affecting calcification. In: Gilman AF RT, Nies AS, Taylor P, editor. Goodman and Gilman’s The Pharmacological Basis of Therapeutics. 8th ed: Pergamon Press; 1990. p. 1518-1522. 91. Roman RJ, Carter JR, North WC, Kauker ML. Renal tubular site of action of fluoride in Fischer 344 rats. Anesthesiology 1977;46(4):2604. 92. Cittanova ML, Lelongt B, Verpont MC, Geniteau-Legendre M, Wahbe F, Prie D, et al. Fluoride ion toxicity in human kidney collecting duct cells. Anesthesiology 1996;84(2):428-35.

545

JARNBERG

93. 94. 95. 96. 97. 98. 99. 100. 101. 102.

103.

104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118.

546

Crandell WB, Pappas SG, Macdonald A. Nephrotoxicity associated with methoxyflurane anesthesia. Anesthesiology 1966;27(5):591607. Mazze RI, Shue GL, Jackson SH. Renal dysfunction associated with methoxyflurane anesthesia. A randomized, prospective clinical evaluation. Jama 1971;216(2):278-88. Holaday DA, Rudofsky S, Treuhaft PS. The metabolic degradation of methoxyflurane in man. Anesthesiology 1970;33(6):58993. Cousins MJ, Mazze RI. Methoxyflurane nephrotoxicity. A study of dose response in man. Jama 1973;225(13):1611-6. Cousins MJ, Mazze RI, Kosek JC, Hitt BA, Love FV. The etiology of methoxyflurane nephrotoxicity. J Pharmacol Exp Ther 1974;190(3):530-41. Frascino JA. Effect of inorganic fluoride on the renal concentrating mechanism. Possible nephrotoxicity in man. J Lab Clin Med 1972;79(2):192-203. Kharasch ED, Hankins DC, Thummel KE. Human kidney methoxyflurane and sevoflurane metabolism. Intrarenal fluoride production as a possible mechanism of methoxyflurane nephrotoxicity. Anesthesiology 1995;82(3):689-99. Brown BR, Jr. Shibboleths and jigsaw puzzles. The fluoride nephrotoxicity enigma. Anesthesiology 1995;82(3):607-8. Mazze RI, Jamison R. Renal effects of sevoflurane. Anesthesiology 1995;83(3):443-5. Kharasch ED, Schroeder JL, Liggitt HD, Park SB, Whittington D, Sheffels P. New insights into the mechanism of methoxyflurane nephrotoxicity and implications for anesthetic development (part 1): Identification of the nephrotoxic metabolic pathway. Anesthesiology 2006;105(4):726-36. Kharasch ED, Schroeder JL, Liggitt HD, Ensign D, Whittington D. New insights into the mechanism of methoxyflurane nephrotoxicity and implications for anesthetic development (part 2): Identification of nephrotoxic metabolites. Anesthesiology 2006;105(4):737-45. Selinsky BS, Perlman ME, London RE. In vivo nuclear magnetic resonance studies of hepatic methoxyflurane metabolism. II. A reevaluation of hepatic metabolic pathways. Mol Pharmacol 1988;33(5):567-73. Chen PS, Jr., Gardner DE, Hodge HC, O’Brien JA, Smith FA. Renal clearance of fluoride. Proc Soc Exp Biol Med 1956;92(4):87983. Carlson CH, Armstrong WD, Singer L. Distribution and excretion of radiofluoride in the human. Proc Soc Exp Biol Med 1960;104:2359. Hosking DJ, Chamberlain MJ. Studies in man with 18 F. Clin Sci 1972;42(2):153-61. Ekstrand J, Ehrnebo M, Boreus LO. Fluoride bioavailability after intravenous and oral administration: importance of renal clearance and urine flow. Clin Pharmacol Ther 1978;23(3):329-37. Whitford GM, Pashley DH, Stringer GI. Fluoride renal clearance: a pH-dependent event. Am J Physiol 1976;230(2):527-32. Ekstrand J, Ehrnebo M, Whitford GM, Jarnberg PO. Fluoride pharmacokinetics during acid-base balance changes in man. Eur J Clin Pharmacol 1980;18(2):189-94. Jarnberg PO, Ekstrand J, Irestedt L. Renal fluoride excretion and plasma fluoride levels during and after enflurane anesthesia are dependent on urinary pH. Anesthesiology 1981;54(1):48-52. Barzel US, Jowsey J. The effects of chronic acid and alkali administration on bone turnover in adult rats. Clin Sci 1969;36(3):51724. Leake RD, Trygstad CW. Glomerular filtration rate during the period of adaptation to extrauterine life. Pediatr Res 1977;11(9 Pt 1):959-62. Jose PA, Fildes RD, Gomez RA, Chevalier RL, Robillard JE. Neonatal renal function and physiology. Curr Opin Pediatr 1994;6(2):1727. Fisher DM, Robinson S, Brett CM, Perin G, Gregory GA. Comparison of enflurane, halothane, and isoflurane for diagnostic and therapeutic procedures in children with malignancies. Anesthesiology 1985;63(6):647-50. Lerman J, Sikich N, Kleinman S, Yentis S. The pharmacology of sevoflurane in infants and children. Anesthesiology 1994;80(4):81424. Frink EJ, Jr., Green WB, Jr., Brown EA, Malcomson M, Hammond LC, Valencia FG, et al. Compound A concentrations during sevoflurane anesthesia in children. Anesthesiology 1996;84(3):566-71. Bedford RF, Ives HE. The renal safety of sevoflurane. Anesth Analg 2000;90(3):505-8.

24

Bisphosphonates and the kidney Jonathan GREEN Novartis Institutes for BioMedical Research, Basel, Switzerland

Introduction ___________________________________________________________ 547 Pharmacokinetics and renal transport of bisphosphonates ______________________ 549 Preclinical renal toxicity __________________________________________________ 552 Clinical renal toxicity _____________________________________________________ 554 Etidronate, clodronate and tiludronate Pamidronate Zoledronic acid Ibandronate

554 555 555 557

Histopathology _________________________________________________________ 558 Pamidronate Zoledronic acid Alendronate

558 559 560

Conclusion _____________________________________________________________ 562 Acknowledgement ______________________________________________________ 562 References _____________________________________________________________ 563

Introduction

B

one remodelling is a continuous and dynamic process that normally involves the coordinated interplay among 3 types of cells: the bone resorptive osteoclasts, the bone-forming osteoblasts, and osteocytes, which are thought to act as mechano-transducers in bone [1]. The process becomes unbalanced in the elderly, in patients with benign bone disease [2], and in patients with primary bone lesions from multiple myeloma or bone metastases from solid tumours [3, 4]. Bisphosphonates are synthetic analogues of pyrophosphate—a naturally occurring modulator of mineralisation found in plasma, urine, and bone. They inhibit osteoclast-mediated bone resorption through several mechanisms, including inhibition of osteoclastogenesis, disruption of intracellular vesicular

trafficking, and induction of osteoclast apoptosis, as well as indirectly via effects on osteoblasts [5]. Bisphosphonates are transported through the bloodstream and are deposited at sites of active bone remodelling, where they bind avidly to the mineralised bone matrix via the bisphosphonate moiety [5]. During bone resorption, bisphosphonates are internalised by osteoclasts, wherein they mediate their antiresorptive effects [5]. Therefore, bisphosphonates have provided increasing clinical utility in the management of patients with pathologies associated with perturbations in bone metabolism [3, 4, 6]. Several generations of bisphosphonates have been developed for the treatment of bone disease, each with different affinities for their cellular targets and increasing clinical efficacy (Figure 1). First-generation bisphosphonates (e.g., etidronate and clodronate)

GREEN

are relatively simple pyrophosphate analogues that lack a nitrogen atom and induce apoptosis after they are metabolised by osteoclasts into nonhydrolyzable, cytotoxic analogues of adenosine triphosphate (ATP) [5]. The addition of a nitrogen-containing side chain to the bisphosphonate backbone resulted in a class of compounds that inhibited a key enzyme in the mevalonate pathway, farnesyl pyrophosphate synthase (FPPS), and greatly increased their potency as inhibitors of osteoclast-mediated osteolysis [5, 7]. Early nitrogen-containing bisphosphonates, including pamidronate, alendronate, and ibandronate, contain an aliphatic side chain with a single nitrogen atom. Of the newest-generation bisphosphonates, risedronate contains a heterocyclic side chain with 1 nitrogen atom, and zoledronic acid contains 2 nitrogen atoms in an imidazole ring, giving these molecules increased potency and a higher affinity for the intracellular target enzyme, FPPS [5, 7, 8]. The presence of a nitrogen-containing side chain facilitates interaction with the catalytic site of FPPS, an enzyme in the metabolic pathway that is required for the production of the isoprenoid lipids farnesyl diphosphate and geranylgeranyl diphosphate, essential metabolites for posttranslational protein prenylation [5, 8]. Inhibiting the prenylation of guanosine triphosphate-binding proteins such as Ras, Rho, and Rac disrupts the normal cellular signal transduction that is required for osteoclast function and survival [5]. Bisphosphonates are widely used for the prevention and treatment of osteopenia and osteoporosis and for the reduction of skeletal complications in patients with malignant bone disease. Several oral bisphosphonates, including alendronate, risedronate, and ibandronate, are approved worldwide for the treatment of osteoporosis in postmenopausal women, as are intravenous (i.v.) formulations of ibandronate (3 mg quarterly) and zoledronic acid (5 mg annually). Several i.v. bisphosphonates are available for the treatment of the skeletal complications that frequently occur in malignant disease, such as hypercalcaemia of malignancy (HCM), multiple myeloma, and bone metastases associated with solid tumours. Pamidronate is approved worldwide for the treatment of HCM, multiple myeloma, and breast cancer bone metastases. Although not registered for oncology indications in the United States, i.v. ibandronate is widely available elsewhere for HCM and breast cancer bone metastases, 548

and as an oral formulation for breast cancer bone metastases. Clodronate, which also is not registered in the United States, is approved in many other countries for the treatment of HCM (oral and i.v. formulations) and osteolysis due to malignancy (oral formulation). Zoledronic acid, by contrast, has the broadest label of any i.v. bisphosphonate for use in HCM, multiple myeloma, and bone metastases associated with any solid tumour (e.g. breast, prostate, lung, kidney, thyroid, head and neck). There are class effects specific to each generation of bisphosphonate. As a result, the therapeutic index is different for each agent. Oral bisphosphonates have been associated with upper gastrointestinal disorders including dysphagia, esophagitis, esophageal ulcers, and gastric ulcers [9]. These effects often limit the long-term compliance with, and therefore the efficacy of, these agents [10]. Patients have reported mild to moderate acute flu-like symptoms after the initial i.v. infusions of nitrogen-containing bisphosphonates [9]. The use of i.v. bisphosphonates is associated with an increased risk of adverse renal events, especially with the intensive dosing regimens used in patients with cancer who, in any case, are more likely to have impaired renal function before the initiation of bisphosphonate therapy and may be receiving other nephrotoxic drugs. The renal tolerability profile of several i.v. bisphosphonates has been investigated in randomised clinical trials, and it appears to be dependent not only on molecular structure but also on the dosing regimen (Table 1) [11-26]. Occasional incidents of renal adverse events with oral alendronate have also been reported but not to the extent of i.v. bisphosphonates [27, 28]. Intravenous bisphosphonates should not be used in patients with renal failure to avoid further impairment of renal function. Therefore, to ensure renal safety, measurement of serum creatinine is recommended before administering i.v. bisphosphonates to patients with mild to moderately impaired renal function, and the dosing regimen should be adjusted appropriately in accordance with the prescribing information for each drug [29-34]. In patients with normal renal function at the initiation of therapy, renal adverse events have typically been infrequent, mild, and transient. Overall, the benefits of i.v. bisphosphonate therapy in the oncology setting far outweigh the risk of adverse events. This chapter provides a detailed review of the pharmacokinetics and renal safety profiles of i.v.

24. Bisphosphonates and the kidney

Figure 1. Chemical structure of bisphosphonates. For consistency all compounds are shown as the dissociated anions. Similarly, zoledronate is used in this figure, although the registered generic name of zoledronic acid is used throughout the text.

bisphosphonates based on published preclinical and clinical data.

Pharmacokinetics and renal transport of bisphosphonates Bisphosphonates are transported through the bloodstream bound to plasma proteins; however, the proportion of bound bisphosphonate varies according to agent. For example, the proportions of clodronate, alendronate, and risedronate that bind to human plasma proteins are approximately 40%, 78%, and 24%, respectively [35, 36]. Notable differences exist in

the binding of specific bisphosphonates to the plasma proteins of different species: alendronate is highly bound to rat plasma protein (96%) but less so to dog (45%), monkey (62%), or human (78%) plasma [35]. Moreover, binding can vary with experimental conditions, making the direct comparison of the plasma protein binding of different bisphosphonates difficult. In a head-to-head comparison of ibandronate and zoledronic acid, protein binding was investigated in plasma from 3 different species under controlled experimental conditions [37]. No significant differences were observed between the binding of ibandronate and zoledronic acid to human, dog, or rat plasma at 549

GREEN

Table 1. Renal tolerability profiles of intravenous bisphosphonates reported in randomised clinical trials. Publication/ Study drug Gucalp [11] EHDP PAM Purohit [12] CLOD PAM Atula [13] CLOD PAM

Disease Study design Patients randomised, N HCM Randomised, EHDP, n = 35; double blind, PAM, n = 30 comparative HCM Randomised, CLOD, n = 21; double-blind, PAM, n = 20 comparative HCM Randomised, CLOD low, n = 10 double blind, CLOD high, n = 21; comparative PAM, n = 20

Gucalp [14] PAM

HCM

Major [15] PAM ZOL Berenson [16] PAM

HCM MM

Hortobagyi [17] BC PAM

Randomised, double-blind, placebocontrolled Randomised, double blind, comparative Randomised, double-blind, placebocontrolled Randomised, double-blind, placebocontrolled Randomised, double-blind, placebocontrolled Randomised, double-blind, placebocontrolled Randomised, double-blind, placebocontrolled Randomised, double-blind, placebocontrolled Randomised, double blind, comparative

Theriault [18] PAM

BC

Rosen [19] ZOL

LC, OST

Saad [20] ZOL

PC

Kohno [21] ZOL

BC

Rosen [22] ZOL PAM

MM, BC

Reid [23] ZOL RIS

PDB

Randomised, double-blind, comparative

Black [24] ZOL

PMO

Ralston [25] IBN

HCM

Body [26] IBN

BC

Randomised, double-blind, placebocontrolled Randomised, double-blind, dose escalation Randomised, double-blind, placebocontrolled

Regimen

Measure of renal impairment EHDP 7.5mg/kg/day i.v. 2h Serum creatinine for 3 days; increase PAM 60 mg i.v. 24h once ≥ 0.5 mg/dl CLOD 1500 i.v. 4h once; Serum creatinine PAM 90 mg i.v. 4h once increase

Renal safety results EHDP, 5.7%; PAM, 13.3% CLOD, 23.8%; PAM, 0%

CLOD 900 mg i.v. 4h once; CLOD 1500 mg i.v. 4h once; PAM 90 mg i.v. 4h once PAM 60 mg i.v. 4h once; PAM 60 mg i.v. 24h once; PLA i.v.

Serum creatinine increase

Slight decrease in both CLOD and PAM groups

Serum creatinine increase ≥ 0.5 mg/dl

PAM 4h, 13%; PAM 24h, 13%; PLA, 4.3%

PAM, n = 86; ZOL low, n = 90; ZOL high, n = 99 PAM, n = 205; PLA, n = 187

PAM 90 mg i.v. 2h once; ZOL 4mg i.v. 5 min once; ZOL 8 mg i.v. 5 min once PAM 90 mg i.v. 4h every 4 wks for 21 months

Serum creatinine PAM, 4.0% increase, grade 3/4 ZOL low, 2.3% ZOL high, 5.6% Serum creatinine No difference increase ≥ 1 mg/dl between PAM and PLA groups

PAM, n = 185; PLA, n = 197

PAM 90 mg i.v. 2h every 3-4 wks for 24 months

Serum chemistry, clinical AEs

No evidence of increased renal AEs with PAM*

PAM, n = 182; PLA, n = 189

PAM 90 mg i.v. 2h every 4 wks for 24 months

Serum chemistry, clinical AEs in > 10% patients

No evidence of increased renal AEs with PAM

ZOL low, n = 257; ZOL high, n = 266†; PLA, n = 250

ZOL 4 mg i.v. 5-15 min; ZOL 8/4 mg i.v. 5-15 min every 3 wks for 9 months

Notable serum creatinine increase**

ZOL low, 10.9%; ZOL high, 12.7%; PLA 90 mg, 6.7%§

ZOL low, n = 214; ZOL high, n = 221†; PLA, n = 208

ZOL 4 mg i.v. 5-15 min; ZOL 8/4 mg i.v. 5-15 min every 3 wks for 15 months

Notable serum creatinine increase||

ZOL low, 15.2%; ZOL high, 20.7%; PLA, 11.5%¶

ZOL, n = 114; PLA, n = 114

ZOL 4 mg i.v. 15 min every 4 wks for12 months

Notable serum creatinine increase**

No evidence of decreased renal function with ZOL

ZOL 4 mg, n = 564; ZOL 8/4 mg, n = 526†; PAM 90 mg, n = 558 ZOL, n = 182; RIS, n = 175

ZOL or PAM every 3-4 wks for 24 months

Grade 3/4 serum creatinine increases after 25 months Serum creatinine and urinary protein measured 9-11 days after dosing Serum creatinine increase > 0.5 mg/dl at days 9-11 postinfusion Serum creatinine

ZOL 4mg, 0.4%; ZOL 8/4 mg, 2.7%; PAM 90 mg, 1.9%‡

PAM 4h, n = 23; PAM 24h, n = 23; PLA, n = 23

ZOL 5 mg i.v. 15 min once; RIS 30 mg/day oral for 60 days

ZOL, n = 3,889; PLA, n = 3,876

ZOL 5 mg i.v. 15 min once a year for 3 years

IBN low, n = 45; IBN mid, n = 44; IBN high, n = 42

IBN 2 mg i.v. 2h; IBN 4 mg i.v. 2h; IBN 6 mg i.v. 2h

IBN low, n = 154; IBN high, n = 154; PLA, n = 158

IBN 2 mg i.v. bolus; IBN 6 mg i.v. 1-2h every 3-4 wks for 15-24 months

Increased creatinine levels (300 mM)

No significant difference between ZOL and RIS groups ZOL, 1.3%; PLA, 0.4%+ No renal toxicity attributable to IBN IBN low, 0.7%; IBN high, 2.6%; PLA, 1.3%

AE = Adverse event; BC = Breast cancer; CLOD = Clodronate; EHDP = Etidronate; HCM = Hypercalcaemia of malignancy; IBN = Ibandronate; i.v. = Intravenous; LC = Lung cancer; MM = Multiple myeloma; OST = Other solid tumours (renal, head and neck, thyroid, other); PDB = Paget’s disease of bone; PAM = Pamidronate; PC = Prostate cancer; PLA = Placebo; PMO = Postmenopausal osteoporosis; RIS = Risedronate; ZOL = Zoledronic acid. *1 patient with preexisting glomerulonephritis developed renal failure, possibly related to study drug. (continued on next page)

550

24. Bisphosphonates and the kidney

clinically relevant concentrations (Table 2) [37]. The interaction between bisphosphonates and plasma proteins is not well understood, but it is known to be influenced by both calcium and iron ions [38]. The addition of calcium to human plasma in an in vitro study increased the proportion of plasma proteinbound versus free pamidronate, whereas the addition of calcium chelators reduced the proportion of drug bound. Similarly, the addition of ferric ions to plasma increased the proportion of plasma protein-bound versus free pamidronate [38]. It has been reported that 40% to 60% of bisphosphonate that reaches the systemic circulation rapidly binds to bone and remains there for a long time; the terminal half-life of alendronate in the human skeleton has been calculated to be approximately 10.5 years [39]. Bisphosphonates are preferentially deposited on exposed mineral at sites in the bone where turnover is high, especially the growth plates, trabecular bone, and sites of injury, infection, or metastasis. Bisphosphonates are metabolically stable, so the remaining unbound bisphosphonate fraction is rapidly eliminated, unchanged, by the kidneys [29-34, 36]. Pharmacokinetic studies of i.v. clodronate and pamidronate in subjects with different degrees of renal insufficiency have demonstrated that renal clearance of bisphosphonate is markedly compromised with a declining glomerular filtration rate, resulting in an increased area under the serum drug concentration-time curve (AUC0-∞) [40, 41]. Table 2. Bisphosphonate binding to 3 individual human plasma samples at pH 7.4 in vitro [37]. Concentration

2 ng/ml 20 ng/ml 200 ng/ml 2,000 ng/ml

Zoledronic acid

40 ± 2

35 ± 2

28 ± 0

23 ± 2

Ibandronate

31 ± 13

26 ± 1

24 ± 1

21 ± 1

Data are expressed as mean percent binding ± standard deviation.

In an open-label pharmacokinetic study of zoledronic acid (4 mg/month for 3 months) in 19 patients with malignant bone disease and varying levels of renal function, renal clearance was consistently lower in patients with impaired renal function [42]. Correspondingly, the 24-hour cumulative urinary excretion of zoledronic acid in patients with renal impairment was lower than in patients with normal renal function. However, in this relatively small study, none of the differences between groups was statistically significant. Although definitive proof is lacking, there appears to be an active transport mechanism for the elimination of bisphosphonates through the kidney, distinct from the known renal transport systems for organic anions and cations and EDTA [43-45]. In an early study with conscious rats, the clearance of etidronate and clodronate was found to be higher than the glomerular filtration rate by a factor of about 1.5, indicating net tubular secretion of both drugs [43]. Further studies in rats demonstrated that renal excretion of the nitrogen-containing bisphosphonate alendronate is concentration- and dose-dependent and saturable, implying secretion by an active-transport mechanism [44]. Moreover, alendronate clearance could be inhibited by the concomitant addition of the non-nitrogen–containing bisphosphonate etidronate in a dose-dependent manner, suggesting competition between the 2 compounds for an uncharacterised renal transport system. Using a luminal stop-flow tubular microperfusion technique with rat kidney, Ullrich et al. found that only etidronate and clodronate had moderate affinity for the renal sulphate transporter, whereas 5 other nitrogen-containing bisphosphonates had only low or no affinity for any of the contraluminal anion transporters [45]. Although the mechanistic details of bisphosphonate handling by the kidney remain largely unknown, attempts have been made to determine the location

(Footnote Table 1 continued) **Serum creatinine increase ≥ 0.5 mg/dl from baseline for patients with normal baseline serum creatinine (< 1.4 mg/dl), or increase ≥ 1.0 mg/dl from baseline for patients with baseline serum creatinine above normal (≥ 1.4 mg/dl), or any increase ≥ 2 times the baseline value. †To ensure renal safety during the study, the dose was reduced from 8 mg to 4 mg, the infusion volume was increased from 50 ml to 100 ml, and the infusion time from 5 min to 15 min. ‡No significant difference in renal toxicity between ZOL 4 mg/100 ml infused over 15 min and PAM 90 mg/250 ml infused over 2 hours, (risk ratio, 1.057; P = 0.839). §After protocol amendment, elevated serum creatinine was not statistically significant between the 4-mg ZOL group and the PLA group (hazard ratio = 1.57; P = 0.228). ||Serum creatinine increases of ≥ 0.5 mg/dl (if baseline value was < 1.4 mg/dl) or ≥ 1.0 mg/dl (if baseline value was ≥ 1.4 mg/dl). ¶Compared to patients who received placebo, patients treated with ZOL 4 mg or 8/4 mg had a comparable risk for renal toxicity, relative risk ratio of 1.07 (P = 0.882) and 1.76 (P = 0.165), respectively. +Transient increase, resolved before the next infusion; no significant difference at 3 years between ZOL and PLA groups for serum creatinine level or creatinine clearance.

551

GREEN

of the rate-determining steps. In an in vivo rat study with alendronate, Kino et al. determined the influx and efflux at the basolateral membrane, sequestration of the drug at the brush-border membrane, and net secretion via the renal tubules [46]. The data indicated saturable transport from renal tubular cells into the luminal duct, suggesting the presence of a transport system on the renal brush-border membrane. Furthermore, the active transport mechanism appeared to be dependent on bisphosphonate plasma concentration, with the uptake process being the rate-determining step of renal secretion at low plasma concentrations. Increasing plasma concentrations had only a minimal effect on the uptake clearance to renal tubular cells, and secretion from tubular cells into the luminal duct could be saturated. It was concluded that dose-dependent transport mechanisms on both the basolateral and brush-border membranes of renal tubular cells are involved in the renal secretion of alendronate [46]. Clearly, further studies are still required to identify the transporter molecules responsible for the renal clearance of bisphosphonates.

Preclinical renal toxicity Although the kidney has been identified as a major target organ for all bisphosphonates at the high doses used in preclinical toxicity studies, a clear therapeutic window still exists, enabling bisphosphonates to be safely administered in general clinical use for the inhibition of bone resorption. However, there are differences in the therapeutic indices between individual agents and in their pharmacokinetic and pharmacodynamic profiles in patients with both normal and impaired renal function. Comparative preclinical studies have provided additional insight into differences in the renal safety profiles of the commonly used bisphosphonates. In a 25week preclinical rat study [47], renal effects were compared for ibandronate (1 mg/kg) and zoledronic acid (1 mg/kg or 3 mg/kg), with both drugs given either as a single-dose i.v. injection or repeatedly every 3 weeks for 6 months. Both the single and intermittent doses of ibandronate 1 mg/kg resulted in similar incidences of proximal tubular degeneration and single-cell necrosis, without any increase of histopathologic damage (Figure 2) [47]. By contrast, degeneration and single-cell necrosis were observed following intermittent dosing 552

Figure 2. Histopathologic findings in the kidney of rats after single or intermittent dosing of ibandronate or zoledronic acid. PCT = Proximal convoluted tubules. (Reprinted with permission from [47], Copyright Elsevier 2003)

of zoledronic acid 1 mg/kg but not after a single i.v. infusion of the same dose. The incidence and severity of damage to the proximal tubules increased with intermittent dosing of zoledronic acid 3 mg/kg. Tubular atrophy and degenerative changes in the outer medulla were associated with this dosing regimen of zoledronic acid but not with the single dose. Hypertrophy and hyperplasia of the collecting ducts and distal tubules were observed following intermittent dosing of both drugs but not after a single dose. These dosing regimens were selected to produce minimally nephrotoxic effects in animals, but they do not reflect the dosing schedules used in the clinical setting [47]. Zoledronic acid is several-fold more potent than ibandronate as an inhibitor of bone resorption [5, 7]; therefore, the doses used in this study were not pharmacologically equivalent. As pointed out by Hirschberg [48], the flaws in the design of this study and in the interpretation of the data make it difficult to draw any clinically relevant conclusions from these experiments. Another study in rabbits has compared the efficacy and safety of pamidronate (1 mg/kg in 20 ml saline infused over 2 hours) and zoledronic acid (0.1 mg/kg in 20 ml saline infused over 20 minutes). Renal toxicity was identified histologically in 14 of 20 kidneys of pamidronate-treated rabbits but was not detected in the 20 kidneys of rabbits infused with zoledronic acid [49]. Because zoledronic acid is a much more potent inhibitor of bone resorption, this agent could be administered at one tenth of the pamidronate dose and yet still achieve superior therapeutic efficacy without

24. Bisphosphonates and the kidney

evidence of renal toxicity. In a similar comparative study in rats that were infused with very high doses (1.5-50 mg/kg) of zoledronic acid or pamidronate (~1, 000-fold greater than the therapeutic dose), the dose required to increase serum urea by 100% from baseline was approximately 4 times higher for zoledronic acid than for pamidronate [50]. Furthermore, on subcutaneous injection of 1 mg/kg 9 times over the course of 14 days, zoledronic acid had no effect on the cumulative urinary excretion of a marker enzyme of renal damage, malate dehydrogenase, whereas pamidronate resulted in a 2-fold increase [50]. Again taking into account the different antiresorptive potencies of the 2 compounds, the therapeutic index for zoledronic acid was determined to be 7-fold greater than that of pamidronate. In view of the subsequent clinical data, although these preclinical assays may have some value for early compound screening, they do not appear to be predictive of renal tolerability in humans. Data from a comparative, preclinical study of single doses of ibandronate (1-20 mg/kg i.v.), zoledronic acid (1-10 mg/kg i.v.), and clodronate (400 mg/kg intraperitoneally), indicated that the proximal tubules were the primary target for renal toxicity as assessed by clinical biochemistry and renal histopathology at 1 and/or 4 days postdosing [51]. Tubular degeneration and single-cell necrosis of the proximal convoluted tubules were observed for all 3 agents at 4 days after dosing, although the severity, type, and location of renal damage differed between compounds. Although high doses of zoledronic acid (10 mg/kg) resulted in granular proteinaceous deposits in the lumen of distal tubules, there was no evidence from X-ray microanalysis of any precipitation of bisphosphonate or formation of aggregates in the kidney. Overall, there were no significant changes in serum biochemical parameters and urinary enzymes in bisphosphonate-treated animals compared with controls [51]. Similarly, in both rats and mice, renal toxicity has been observed with i.v. administration of pamidronate [52]. Renal effects, including renal tubular necrosis and enzymuria, were most prominent when animals received doses of pamidronate (≥10 mg/kg) several-fold greater than those used clinically (~0.5-1.5 mg/kg) [52]. The preclinical toxicity of alendronate has been studied at doses comparable with those used clinically with acute (single-dose) and chronic (repeated-dose) dosing regimens [53]. The most common lethal toxicity

observed in mice and rats after oral administration of alendronate was primarily related to gastrointestinal irritation, and no lethality was reported in dogs receiving high oral doses of the drug. Nephrotoxicity was reported in rats and dogs that received high doses of alendronate relative to the clinical doses used in the treatment of osteoporosis (>10 times). Chronic administration of alendronate (at least 0.1 mg/kg/d i.v. for 5 weeks) resulted in microscopic renal damage characterised by very slight to slight focal nephritis in dogs, although no corresponding changes in serum biochemical markers of renal impairment were observed [53]. In further follow-up studies, lower doses of alendronate (0.01 and 0.05 mg/kg/d i.v.) did not cause renal lesions. Similar results were reported in subsequent long-term studies in young dogs receiving oral alendronate at doses of 0.5, 2, or 8 mg/kg/d [53]. No renal lesions were reported at 27 weeks, and only a low incidence of chronic nephritis was seen at 53 weeks with the highest dose. Overall, high doses of alendronate (1 mg/kg/d i.v. or 8 mg/kg/d orally) may cause minimal renal lesions in dogs with no resulting impairment in renal function [53]. These data are consistent with the clinical experience accrued from the extensive use of oral alendronate for the treatment of osteoporosis, which shows minimal risk of renal impairment. The nephrotoxicity of the experimental bisphosphonate cimadronate (also known as incadronate or YM-175) has also been investigated in dose-escalation studies in rats and dogs. Histopathologic signs of renal toxicity were observed in rats administered i.v. doses of cimadronate ≥0.62 mg/kg/d for 30 days; however, these effects were transient and disappeared after a 30-day recovery period [54]. By contrast, no renal toxicity was observed in rats treated for 26 weeks with weekly i.v. doses of 0.31-1.25 mg/kg/week. In corresponding studies in beagles, 2 dogs were killed in extremis due to renal failure at days 4 and 7 after a single cimadronate dose of 10 mg/kg i.v., whereas there were no drug-related findings at the lower doses of 0.3-3 mg/kg [55]. Animals treated for 30 days with cimadronate at a dose of 1 mg/kg/d i.v. exhibited nephropathy characterised by cortical tubular necrosis or degeneration with tubular dilation and basophilia, as well as slight interstitial nephritis, resulting in the death of 1 animal on day 16. These findings were not observed at lower doses (0.03-0.3 mg/kg/week). No 553

GREEN

histopathologic changes were observed in the kidney when the compound was administered for 26 weeks at doses of up to 1.25 mg/kg/week [55]. Overall, these findings with cimadronate are consistent with those observed for other bisphosphonates and suggest that renal toxicity, typically proximal tubular degeneration, is most probably a class effect associated with the mechanism of action of these compounds (i.e. inhibition of the enzyme FPPS in the mevalonate pathway). Nevertheless, an acceptable therapeutic window clearly exists for both oral and i.v. bisphosphonates, resulting in the efficacious inhibition of bone resorption with good overall tolerability.

Clinical renal toxicity Similar to the data from preclinical studies, the tolerability profile of the different bisphosphonates in clinical use is not uniform and is dependent on the dosing regimen as well as the patient population. The extensive use of daily or weekly oral bisphosphonates to treat osteoporosis is primarily associated with gastrointestinal adverse effects including abdominal pain, dyspepsia, nausea, and esophagitis [9]. Renal tolerability has not emerged as an issue in this setting and will not be discussed further [9, 56-58]. For patients with metastatic bone disease, a monthly dosing schedule with i.v. bisphosphonate is established as the standard of care, but some compounds are also available in certain countries as oral formulations for oncology indications. Intravenous bisphosphonates are associated with mild to moderate flu-like symptoms in a significant proportion of patients, predominantly after the initial infusion [9], and adverse effects on renal function may also occur infrequently [9]. Consistent with the preclinical observations, all i.v. bisphosphonates have the potential to affect renal function in clinical use; however, renal adverse effects appear to be dependent on the baseline renal status of the patient and the dosing regimen. Although the incidence of renal adverse events is infrequent when i.v. bisphosphonates are administered at their recommended doses and infusion rates, monitoring of renal function is advisable in all patients receiving i.v. bisphosphonate therapy [29-34]. Attempts to compare the renal safety profiles of different bisphosphonates are confounded by limited head-to-head comparisons between different bisphosphonates. Although limited 554

in number, double-blind, Phase III trials of bisphosphonates in comparison with placebo or active comparator provide the most stringent means to assess the relative effects of specific bisphosphonates on renal tolerability in a controlled setting.

Etidronate, clodronate and tiludronate Early tolerability studies with the first-generation i.v. bisphosphonates etidronate and clodronate initially identified transient renal effects in patients receiving bisphosphonate therapy for HCM [59-69]. Fatal cases of renal failure were reported after the infusion of high doses of etidronate or clodronate in patients with breast cancer or multiple myeloma [60]. In an early randomised, comparative study of etidronate (3 infusions of 7.5 mg/kg/d for 3 consecutive days), clodronate (600-mg single i.v. infusion), and pamidronate (30-mg single i.v. infusion) in 48 patients with HCM, the latter compound was found to be the most potent at lowering serum calcium and also had the most rapid onset and the longest duration of response [61]. Deterioration of renal function was not observed in patients receiving clodronate or pamidronate; however, significant renal impairment developed in 1 etidronate-treated patient. In further studies with etidronate in patients with HCM, transient increases in serum creatinine were observed in some studies [11, 62, 64, 65]. Overall, in small studies of patients receiving i.v. etidronate for HCM, reports of creatinine elevation ranged from 0% [63] to 13% [64], depending on infusion rate, duration of treatment, and dose [59-65]. Similarly, in trials with a wide range of clodronate i.v. dosing regimens for the treatment of HCM, rates of serum creatinine elevation ranged from 0% to 24% [12, 13, 59, 61, 62, 66-69]. The clinical utility of another first-generation, nonnitrogen–containing bisphosphonate, tiludronate, has been limited because of concern over renal adverse effects observed in early trials. In a dose-finding study for the treatment of HCM, 19 patients received i.v. drug followed by oral maintenance therapy [70]. Three patients had elevated serum creatinine levels after i.v. drug administration of 4.5 or 6.0 mg/kg, 1 of whom developed acute renal insufficiency and subsequently died, most probably due to tiludronate, although renal infection and allopurinol therapy could have played a contributory role. While receiving oral therapy with either 400 or 800

24. Bisphosphonates and the kidney

mg/d, 5 other patients also presented with elevated serum creatinine levels. From the results of this study, it was concluded that, in comparison with the nitrogencontaining bis-phosphonates, tiludronate could not be recommended for the treatment of HCM because of the need for high, potentially nephrotoxic, i.v. doses [70].

Pamidronate In comparison with the results obtained with earlier bisphosphonates, clinical trials with the nitrogen-containing bisphosphonate pamidronate in more than 1, 300 patients with cancer have reported greater efficacy without a concomitant increase in renal toxicity. In 3 large, randomised, double-blind, Phase III trials comparing pamidronate 90 mg against placebo in patients with multiple myeloma or breast cancer, renal safety was similar for both treatment groups on the basis of serum chemistry and clinical adverse events (Table 1) [16-18]. In 1 of the studies in patients with malignant bone disease from breast cancer, 1 patient in the pamidronate group discontinued treatment because of renal failure; however, this patient had a history of glomerulonephritis [17]. Similarly, in patients with bone metastases from multiple myeloma, a similar incidence of elevated serum creatinine values (≥1 mg/dl) above baseline was observed in both the pamidronate and placebo treatment groups [16]. However, the administration of pamidronate at doses higher than the recommended 90 mg/month to patients with multiple myeloma has been associated with nephrotic proteinuria, which reversed in the majority of cases after dose reduction or discontinuation [71].

Zoledronic acid In small Phase I trials, bolus doses of zoledronic acid of up to 16 mg infused over 30 to 60 seconds appeared safe [72], but, subsequently, the infusion time for the 4-mg dose was extended first to 5 and then to 15 minutes, and the infusion volume was increased to 100 ml to ensure renal safety in routine clinical use. Moreover, further investigations with the 8-mg dose were discontinued because there was no evidence of increased efficacy versus 4 mg, and renal tolerability clearly decreased [19, 20, 22]. For the acute treatment of HCM, a single 5-minute infusion of zoledronic acid (4 or 8 mg) was investigated

in 2 identical, randomised, double-blind, Phase III trials against a single 2-hour infusion of pamidronate 90 mg [15]. The complete response rates by day 10 were 88.4%, 86.7%, and 69.7% for zoledronic acid 4 mg, zoledronic acid 8 mg, and pamidronate 90 mg, respectively, whereas the corresponding grade 3/4 increases in serum creatinine were 2.3%, 5.2%, and 4%. The long-term safety and efficacy of zoledronic acid in the oncology setting were investigated in 4 randomised, double-blind clinical trials involving more than 3, 000 patients with multiple myeloma, breast cancer, prostate cancer, and lung cancer or other solid tumours [19-22]. Zoledronic acid has demonstrated an acceptable renal safety profile compared with placebo in 3 long-term, placebo-controlled trials. In patients with prostate cancer who completed the 24-month study (n=122), there was no significant difference in time to first serum creatinine increase between patients who received 4 mg zoledronic acid compared with those who received placebo (P=0.752), and the hazard ratio (HR) of experiencing an elevation in serum creatinine was similar between the 2 groups (HR=1.14) [20, 73]. Similarly, in a randomised, Phase III, double-blind, placebo-controlled trial in 773 patients with lung cancer or other solid tumours, including renal cell carcinoma, there was a slight trend towards increased serum creatinine in patients receiving a 15-minute infusion of 4 mg zoledronic acid compared with those receiving placebo (HR=1.57); however, it was not statistically significant (P=0.228) [19]. Notably, in the subset of patients (n=33) with renal cell carcinoma who were assessed for safety, there was no significant difference in rate of renal adverse events between patients who received zoledronic acid (2/18) and those who received placebo (3/15) [74]. A Japanese study compared the efficacy and safety of zoledronic acid 4 mg, administered as a 15-minute infusion every month for 1 year, versus placebo in 228 women with bone metastases from breast cancer [21]. Zoledronic acid reduced skeletal-related events by 39% and was well tolerated with a safety profile similar to that of placebo. Only 1 patient in the zoledronic acid group had a notable serum creatinine increase (2.0 mg/dl) from a baseline of 1.3 mg/dl compared with 7 patients in the placebo group. Moreover, no patient treated with zoledronic acid developed a grade 3 or 4 serum creatinine increase according to the National Cancer Institute common toxicity criteria, whereas 555

GREEN

Figure 3. Kaplan-Meier estimates of time to first notable serum creatinine increase in patients with multiple myeloma or breast cancer with bone metastases receiving 4 mg zoledronic acid or 90 mg pamidronate and Andersen-Gill multiple event analysis of the risk of elevated serum creatinine between treatment groups. *After start of study drug. (Reprinted with permission from [75])

1 patient in the placebo group had a grade 3 serum creatinine elevation [21]. In the multicentre, Phase III comparative trial in patients with ≥1 bone lesion from breast cancer or multiple myeloma (N=1, 648), there were no significant differences in renal safety profiles between patients given a 15-minute infusion of 4 mg zoledronic acid and those given a 2-hour infusion of 90 mg pamidronate (Table 1) [22]. Kaplan-Meier estimates demonstrated that there were no significant differences in time to first notable serum creatinine increase between treatment groups (HR=1.057; P=0.839; Figure 3) [75]. Zoledronic acid has also been investigated in the prevention of cancer treatment-induced bone loss in 401 premenopausal women receiving adjuvant endocrine therapy for hormone-responsive breast cancer in a randomised, open-label, Phase III clinical trial [76]. In this study, patients received tamoxifen and goserelin with or without zoledronic acid (4 mg i.v. every 6 months) versus anastrozole and goserelin with or without zoledronic acid (4 mg i.v. every 6 months) for 3 years. The combination of zoledronic acid with endocrine therapy was well tolerated and was not associated with changes in renal function in this patient population. Over 3 years, 2, 904 serum creatinine measurements were taken, the mean serum creatinine level was 0.78 ± 0.17 mg/dl, and no patient had serum creatinine levels that exceeded 1.5 times the upper limit of normal [76]. 556

Outside of the controlled clinical trial setting, the renal tolerability of zoledronic acid has also been assessed in routine practice at a single cancer centre. In a retrospective analysis of 446 patients with malignant bone disease who received a total of 3, 115 doses of zoledronic acid (median, 4 doses; range, 1-28 doses) over 2 years, renal deterioration was reported in 9.4% of patients (median rise in creatinine level, 1.0 mg/dl; range, 0.5-4.4 mg/dl) [77]. Eight patients discontinued zoledronic acid therapy because of renal deterioration; however, no patient required dialysis, and no patient died as a result of renal dysfunction [77]. In a retrospective analysis of spontaneous adverse event reports encompassing more than 430, 000 patients who had received zoledronic acid between August 2001 and March 2003, only 72 cases of renal failure were identified by the US Food and Drug Administration [78, 79]. It should be noted, however, that patients with risk factors for renal deterioration, including advanced cancer, previous bisphosphonate exposure, and use of nonsteroidal anti-inflammatory medications, may have contributed to the progression of renal failure [79]. Because of the potentially serious nature of this adverse event, it is recommended to monitor renal function in patients with cancer before each infusion of zoledronic acid, provide adequate hydration, and modify or discontinue treatment if renal complications occur [30, 78, 79]. The renal tolerability of zoledronic acid has also

24. Bisphosphonates and the kidney

been investigated in benign bone disease, using a dose of 5 mg infused over 15 minutes, either once for the treatment of Paget’s disease or repeated annually for 3 years in postmenopausal osteoporosis. In 2 identical, randomised, double-blind, controlled trials with 357 patients suffering from Paget’s disease, patients received either one 15-minute infusion of 5 mg zoledronic acid or 60 days of oral risedronate (30 mg/d) [23]. Zoledronic acid produced a significantly greater and quicker therapeutic response compared with risedronate in this patient population (P0.5 mg/dl) in the serum creatinine level compared with 0.4% in the placebo group (P=0.001). However, within 30 days, the levels in greater than 85% of patients returned to within 0.5 mg/dl of preinfusion values, and the remainder reached this level before the next annual infusion. At 3 years, there was no significant difference in either serum creatinine levels or creatinine clearance between the groups, indicating no cumulative effect on renal function [24].

Ibandronate A Phase III, randomised, clinical trial investigated the efficacy and renal safety of i.v. ibandronate 6 mg (n=154) compared with placebo (n=158) infused over 1-2 hours every 3-4 weeks for up to 2 years in patients with malignant bone disease from breast cancer [26]. Ibandronate 6 mg significantly reduced skeletal events associated with metastatic disease (P CYP3A4

CYP2C19 > CYP3A4

14-25% renal inactive metabolites 1.5 mg/dL) urine collections are extended to at least 3 days. [4]. The CaNa2EDTA lead mobilization test has proved useful in detecting excessive lead absorption. It permitted an unbiased assessment of the consequences of lead absorption at a time when blood leads up to 80 μg/dL were deemed acceptable. The validity of arbitrary standards for the CaNa2EDTA lead mobilization test is, however, unclear since recent epidemiologic studies demonstrate an adverse impact of levels of blood lead far lower than previously considered dangerous. Setting an “acceptable” level of urinary lead excretion during the CaNa2EDTA test is, therefore, problematic. The same caveat applies to the lead mobilization test performed with the oral chelating agent, succimer. Despite these considerations the blood lead concentration remains the “gold standard” for assessing lead exposure. Adverse effects of lead on blood pressure and kidney function have been found at blood lead levels as low as 2 μg/dL [5]. There thus appears to be no threshold below which adverse effects of lead are not found. From the point of view of protecting health, there is no safe level for blood lead, chelatable lead, or bone lead. Bone lead measured by non-invasive K-XRF is particularly useful for assessing cumulative lead absorption in population studies. However, the relationship of XRF to the CaNa2EDTA mobilization test or the blood lead index over the full range of possible exposure situations is unknown. The fraction of urinary lead coming from bone following the administration of chelators presumably varies with the time, duration

34. Lead nephropathy

and level of recent and past exposure as well as with physiologic factors modulating bone remodeling and renal function. Under steady-state conditions (absent ongoing heavy exposure) the CaNa2EDTA mobilization test correlates well with direct chemical measurement of lead in transiliac bone biopsies [6, 7]. Because lead in bone has a biologic half-life of years, compared to a biologic half-life of lead in blood of weeks [3], bone more closely reflects cumulative lead stores. In groups heavily exposed to lead over many years, blood lead correlates well with bone lead [7, 8]. Although the blood lead concentration reflects absorption of both organic and inorganic lead, the clinical symptoms of organic lead absorption (e.g. tetraethyl lead gasoline additive) are of rapid onset and primarily cerebral. Colic, peripheral neuropathy, and anemia, characteristic symptoms of inorganic lead poisoning, are absent. Chelation therapy, highly effective for inorganic lead poisoning, is ineffective in organic lead poisoning [9]. The distinctive hallucinogenic effects induced by massive absorption of tetraethyl lead was dramatized when DuPont’s Chambers Works in Deepwater, New Jersey, became known as the “House of Butterflies” shortly after manufacture of the antiknock gasoline additive began in 1923 [10]. Differentiating the residual cognitive defects induced by lead from those induced by organic compounds is challenging [11]. Neither the acute Fanconi syndrome (aminoaciduria, phosphaturia, and glycosuria without hyperglycemia) nor chronic interstitial nephritis have been described as a consequence of tetraethyl lead exposure [9]. A longitudinal study of tetraethyl lead workers, however, showed a positive correlation between blood pressure and both blood and tibial lead [12].

Acute lead nephropathy In children with lead encephalopathy, proximal tubule reabsorptive defects characterized by the Fanconi syndrome have been observed [13]. The Fanconi syndrome appears when blood lead levels approach 150 μg/dL. It is rapidly reversed by chelation therapy designed to treat the far more dangerous lead encephalopathy. The proximal tubule reabsorptive defect can regularly be induced experimentally in rats fed dietary lead [14]. In both children and experimental animals, acute lead nephropathy is consistently associated with acid-fast intranuclear inclusions in proximal tubule

epithelial cells [14]. The intranuclear inclusion bodies consist of a lead–protein complex and may be seen in tubular epithelial cells in the urinary sediment during acute poisoning [15]. Lead-containing intranuclear inclusions have also been observed in liver, neural tissue, and osteoclasts. Acute poisoning is associated with morphologic and functional defects in tubular epithelial cell mitochondria.

Chronic lead nephropathy The phrase chronic lead nephropathy refers to the slowly progressive interstitial nephritis occurring infrequently in adults following prolonged exposure to lead and manifested by a reduced glomerular filtration rate (GFR), and meager proteinuria. It is frequently associated with hypertension, and gout. Interstitial nephritis following symptomatic lead poisoning was described in the nineteen century and was widely identified among symptomatic lead workers, and consumers of contaminated illegal whiskey (“moonshiners”) in the twentieth century [1]. Recognition of the adverse effects of lead in asymptomatic individuals depended on the development of chemical methods for measuring lead in blood in the twentieth century. Identifying the effects of lead when blood levels are below 20 μg/dL had to await large epidemiologic studies in community populations as opposed to clinically or occupationally defined groups. Epidemiologic studies demonstrated the statistical significance of small alterations in blood pressure and renal function due to low-level lead, alterations that have major public health implications. Lead may contribute to the finding that almost four million Americans have both elevated creatinines and hypertension [16]. Occupational lead nephropathy has developed after as little as 3 years of intense occupational exposure [4] Analysis of death certificates of 601 men employed at the Bunker Hill Lead Mine and Smelter in Kellogg, Idaho, up to 1977 indicated a twofold-increased risk of dying from chronic renal disease [17]. The increased risk approached fourfold after 20 years of occupational exposure. Although most frequently recognized in lead workers after decades of occupational exposure, chronic lead nephropathy was originally described among young adults in Australia who sustained acute childhood lead poisoning [18]. Sporadic case reports of lead nephropathy arising from unusual accidental 775

WEDEEN

exposure such as geophagia [19] or Asian folk remedies and cosmetics continue to appear in the medical literature [1]. Lead-induced chronic interstitial nephritis in the absence of symptomatic lead poisoning was first described among American workers [4, 20], and in U.S. Armed Service veterans suffering from renal failure that had been initially attributed to gout or essential hypertension [21, 22]. The contribution of excessive lead absorption to hypertension and interstitial nephritis was indicated by urinary excretion of more than 600 μg of lead during the CaNa2EDTA lead-mobilization test performed after renal failure was apparent. Medical histories obtained from these men were misleading with respect to prior lead exposure; patient recall frequently contradicted the objective evidence of the chelation test. In the 1960s and 1970s the arbitrary cut off of 600 μg/3days for lead excretion during the CaNa2EDTA test proved useful for identifying excessive lead exposure groups in Australia, the United States, Spain and Italy [4, 6, 18, 23]. Studies using this cut off identified the relatively frequent appearance of renal insufficiency with hypertension and/or gout at high exposure levels (CaNa2EDTA lead mobilization test > 600μg/3d). Detecting the lower prevalence of adverse effects at low body lead burdens required studies of asymptomatic populations with lower exposure [24, 25]. “Queensland nephritis” appears to represent the transition from the proximal tubule reabsorptive defects of acute lead poisoning in children, to the chronic interstitial nephritis of adults [18]. Lead-induced interstitial nephritis was first recognized among young adults in Queensland, Australia, who were lead poisoned as children in the 1890s through the 1920s. The evolution of acute lead nephropathy to chronic interstitial nephritis has been produced in experimental animals but was only recently reported in follow-up studies of Americans adults exposed in childhood. In an early follow-up study of untreated childhood lead poisoning, diagnostic criteria for both lead poisoning and renal disease were unacceptably vague [26]. A 50year follow-up of untreated lead-poisoned children in the United States found evidence of an increased prevalence of renal disease [27]. Chronic lead nephropathy from moonshine came to medical attention because of the dramatic symptoms of acute lead poisoning. Lead colic and anemia were 776

associated with reduced GFR, which often improved following chelation therapy. Transient renal failure, apparently the result of renal vasoconstriction [28], was superimposed on structural renal damage that appeared to be less responsive to chelating agents. This therapeutic response in pre-azotemic lead nephropathy may reflect reversal of functional impairment rather than reversal of established interstitial nephritis. Acute reductions in GFR and acute elevations of blood pressure may be mediated by the blood lead concentration whereas long-term effects such as interstitial nephritis and sustained hypertension may be determined by cumulative lead absorption [12, 29]. This difference in the short-term (days) effects of lead on GFR and blood pressure compared to the effects of long-term (years) exposure corresponds to the difference between Fanconi syndrome following brief acute exposure and the delayed development of chronic interstitial nephritis following prolonged exposure. The acute and chronic effects of lead on the kidney may have different pathogenetic mechanisms. Epidemiologic evidence suggests that modest azotemia is significantly more prevalent among lead-exposed workers than among nonexposed counterparts, presumably owing to both morphologic changes mediated by cumulative exposure and functional changes mediated by current blood lead levels [30]. Chronic lead nephropathy in moonshiners, more often than not, is accompanied by gout and hypertension, in accord with 19th century descriptions of plumbism and reports from Australia [1]. A statistically significant odds ratio of 2.4 has been reported for moonshine consumption and end-stage renal disease, suggesting a causal association with lead in the absence of symptomatic lead poisoning [31]. Renal biopsies in chronic lead nephropathy show nonspecific tubular atrophy and interstitial fibrosis with minimal inflammatory response as well as mitochondrial swelling, loss of cristae, and increased lysosomal dense bodies within proximal tubule cells [4, 18]. (Figure 1) Intra-renal arteriolar changes indistinguishable from nephrosclerosis are found, often in the absence of clinical hypertension [4]. The appearance of arteriolar nephrosclerosis before hypertension develops and the relatively short duration of hypertension before renal failure supervenes suggest that the initial renal injury from lead may be in the microvascular endothelium

34. Lead nephropathy

Figure 1. Tubular atrophy and interstitial fibrosis in a case of chronic lead nephropathy. H&E staining, orig. magn. x300.

[30, 32]. This view is consistent with the possibility that the acute effects of lead on blood pressure are mediated by the current blood lead concentration whereas the long-term effects are mediated by endothelial injury resulting from cumulative lead absorption. Intranuclear inclusion bodies are often absent when the renal disease is long-standing and advanced or following the administration of chelating agents. Clumped chromatin and nuclear invaginations of cytoplasmic contents may be found even in the absence of intranuclear inclusions. Morphologic alterations are minimal in glomeruli until the reduction in GFR is advanced The functional changes in chronic lead nephropathy appear to be less specific than those observed in acute poisoning. As in other forms of interstitial nephritis, proteinuria and glycosuria are initially absent. In contrast to cadmium nephropathy, the excretion of a large array of urinary marker proteins such as retinal binding protein, lysozyme, and ß2-microglobulin [33, 34] is not increased in the absence of a reduced GFR. The increase in urinary N-acetyl-ß-D-glucosaminidase (NAG) with increasing blood lead concentrations reflects the proximal tubule dysfunction seen in acute lead nephropathy rather than the chronic interstitial nephritis associated with occupational lead exposure [35, 36]. NAG excretion correlates positively with the blood lead concentration but not with the bone lead concentration [37]. Eicosanoid excretion in lead workers is similar to that in patients with essential

hypertension [33, 36]. In contrast to the reabsorptive defect of acute lead nephropathy, saturnine gout is characterized by renal retention of uric acid [18]. The clearance (CPAH) and maximal secretion rate (TmPAH) for p-aminohippurate (PAH) have been found to be variable in patients with occupational lead nephropathy.

Saturnine gout Hyperuricemia and gout are common among individuals with excessive exposure to lead, apparently the result of decreased excretion and increased production of uric acid. Although hyperuricemia invariably accompanies azotemia, gout is uncommon in patients with renal failure except in those with lead nephropathy. Half of uremic patients with lead nephropathy have clinical gout [18] but in the absence of renal failure, gout cannot usually be attributed to lead despite coexisting hypertension [23, 37]. There is substantial evidence that renal failure in gout is often secondary to overt or unsuspected lead poisoning. In Queensland, Australia, as many as 80% of gout patients with renal failure had elevated EDTA lead-mobilization tests [18]. In New Jersey, chelatable lead was found to be significantly greater among gout patients with renal failure than among gout patients with normal renal function [21]. Because patients with comparable renal failure owing to known causes other 777

WEDEEN

Lead nephropathy does not account for renal failthan lead show no increase in chelatable lead, the excessive mobilizable lead in these gout patients ap- ure in all hypertensives with kidney disease any more than it accounts for renal failure in all gout patients pears to be the cause rather than consequence of their with kidney disease. The heavy metal may, however, renal failure. Measurement of lead levels in transiliac contribute to the association of gout with hypertension, bone biopsy specimens from patients with end-stage as well as to the variable incidence of renal failure in renal disease confirms the finding that renal failure each of these conditions. per se does not cause increased mobilizable lead or An etiologic role for lead in hypertension is supbone lead [6, 7]. The conclusion that lead absorption causes renal damage rather than the reverse is sup- ported by epidemiologic studies in populations with mean blood leads > 10 μg/dL, but with exposure too ported by longitudinal studies [4, 6]. Unrecognized low to produce symptomatic lead poisoning. The Seclead poisoning, therefore, may explain the occurrence ond National Health and Nutrition Examination Surof renal failure in some gout patients who have neither vey (NHANES II) performed between 1976 and 1980 urinary calculi nor intratubular uric acid deposition included blood lead and blood pressure measurements disease. Similarly, overt lead poisoning may explain the protean manifestations of gout in past centuries, in almost 10,000 non-institutionalized Americans aged irregular gout, as well as the almost forgotten associa- 6 months to 74 years [41] Correlation between blood tion of gout with (lead-laden) wine [1]. Sporadic con- lead and blood pressure was robust even when both measurements were within the (at the time) accepted tamination of alcoholic drinks with lead throughout history may have been responsible for irregular gout “normal” range [42, 43]. Similar observations have that terminated in cerebral disease (e.g., uremia, stroke, been made in studies performed throughout the world, although non-statistically significant findings in small lead encephalopathy). studies have also been reported. Lead-induced hyperuricemia may contribute to chronic lead nephropathy. Uric acid per se induces endothelial cell injury, renal microvascular disease, Low-level exposure and hypertension, at least in part mediated by oxidative stress [38]. Independent of uric acid, reactive Epidemiologic evidence that low-level lead absorpoxygen species induced by lead have been implicated tion (blood leads < 10 μg/dL) increases blood pressure in endothelial cell injury, increased vascular reactivity, and decreases renal function has been obtained from and the production of hypertension in humans and studies undertaken after lead was removed from gasoexperimental animals [39]. line as blood lead levels were falling in the United States population as a whole. Analysis of data on over 15,000 Americans from 1988 to 1994 in NHANES III showed Hypertension that hypertensives had significantly higher blood leads (4.21 vs 3.30 μg/dL) and a higher frequency of elevated The association between lead and hypertension creatinines (11.5 vs 1.8%) than non-hypertensives [44]. has been a subject of controversy since the first use Cross-sectional studies showed a significant positive of the sphygmomanometer. The early view that renal association between low-level lead exposure and seinjury induced by lead causes hypertension has gained rum creatinine [40, 45]. Major adverse consequences increasing support. The duration of hypertension in of hypertension, mortality from cardiovascular disease, patients with lead nephropathy tends to be shorter than myocardial infarction, and stroke, have been found to that in hypertensives without renal failure, suggesting correlate positively with blood lead levels above [45] that lead-induced renal vascular damage precedes, and therefore causes the hypertension. This view is consist- and below 10 μg/dL [5]. Even as the mean blood lead in the US population fell below 2 μg/dL, those in the ent with the finding that creatinine clearance decreases with increasing blood lead in the general population, highest lead quartile were 2.72 time more likely to have an effect that is independent of blood pressure [40]. chronic kidney disease than those in the lowest blood lead quartile [47]. Mortality data show that death from hypertensive The adverse effect of low-level lead exposure on cardiovascular disease is more frequent among lead renal function is supported by longitudinal observaworkers than the general population [17] 778

34. Lead nephropathy

tions in non-occupationally exposed populations. Among 509 randomly selected men in the Department of Veterans Affairs Normative Aging Study who had a mean blood lead of 9.9 μg/dL, Kim et al. found that blood lead correlated positively and significantly with serum creatinine [48]. In 1171 of these veterans (mean blood lead of 6.3 μg/dL), tibia, patella and blood lead levels were significantly higher in those who developed hypertension than in those who did not [49]. In another subset of the Normative Aging Study, Thais et al. reported that the rate of progression of renal failure was 17.6- and 12.8-fold greater in diabetics in the highest tibial and blood lead quartiles, respectively, than in nondiabetics indicating high vulnerability to the adverse effects of low-level lead in diabetics [50]. Similarly, increased sensitivity to the additional adverse effects of low-level lead exposure on kidney function was observed among hypertensives [49]. Among 964 subjects (mean blood lead 3.5 μg/dL) blood (but not tibial) lead was significantly correlated with systolic and diastolic blood pressure but not with hypertension [29]. Tibia lead correlated with hypertension suggesting that current blood lead levels influence blood pressure but cumulative absorption influences the development of sustained hypertension. Cross-sectional studies of normal populations also show adverse effects of exposure to lead at blood lead levels < 10μg/dL. Tibial lead (but not blood or patella lead) was a significant predictor of hypertension and systolic pressure in the normotensive range in the Normative Aging Study (mean blood lead 6.09 μg/dL) [51]. In contrast, patella lead, but not tibia lead, was found to be a significant predictor of hypertension in nurses [52]. In one study, the impact of low level lead exposure assessed by blood lead on blood pressure was statistically significant in blacks but not in whites [53]. The impact of cumulative lead absorption on blood pressure may be ameliorated by high dietary calcium intake (> 800 mg/day), and further modulated by polymorphisms in the vitamin D receptor gene [54, 55]. A study in pregnant women with a geometric mean blood lead of 1.9 μg/dL found increased bone leads were associated with an increased risk of hypertension [56]. Diastolic and systolic blood pressure were significantly and positively associated with blood lead concentration. The major portion of the effect was found with blood leads < 5 μg/dL. Blood lead < 10 μg/dL also was a risk factor for postpartum hyper-

tension among women in Tehran [57]. A compelling case is therefore emerging indicating adverse effects on blood pressure and the kidneys at blood lead levels < 5 μg/dL in diverse populations [58].

Causality and environmental exposure The variability of the positive correlations found between blood pressure or GFR with biomarkers of lead absorption (see above) has been used as an argument to deter preventive action. Such hesitancy may in part derive from a fundamental misunderstanding of the scientific rationale (i.e.Bradford Hill’s considerations) for determining causality in environmental disease [59] By convention, statistical analysis favors accepting the null hypothesis; finding no significant difference between groups. The methodology results in finding false negatives more readily than false positives. Measurement error and human variability tend to support the null hypothesis. Because control of renal function and blood pressure is multifactorial, the causal contribution of lead is difficult to isolate. A number of biomarkers (blood, tibial, and patella lead), and a variety of populations differing by age, gender, race, and level of exposure are examined. Systolic and diastolic pressures are assessed separately and may be analyzed both as continuous or dichotomous variables. Kidney function is assessed by the serum creatinine concentration or empirical adjustments of the creatinine to estimate GFR. Large populations are required to achieve statistical significance amidst the noise of the multifactorial causality and the imprecision of outcome measures. Inconsistent results and weak correlations are, therefore, expected as smaller and smaller outcome effects are evaluated. Consequently, the finding of statistical significance for some but not all of the biomarkers of lead absorption does not nullify the importance of the positive associations. On the contrary, statistical inference is stacked against finding false positives and therefore may underestimate real associations. The failure to find statistical significance does not have weight equal to the finding of significance. Despite the inevitable persistence of uncertainty, the obligation to recognize the importance of statistically significant findings, and to undertake preventive action, remains. Although some controversy exists about the magnitude of the dose-response relationship, there is a grow779

WEDEEN

ing consensus that lead contributes to hypertension in the general population, particularly in the presence of renal dysfunction. Lead may also contribute to the disproportionate representation of black men with hypertensive nephrosclerosis and diabetic nephropathy in end-stage renal disease programs in the United States [60]. The observation reported from Ja-Liang Lin’s laboratory in Taiwan that chelation therapy improves renal function in renal failure patients with low body lead stores (CaNa2EDTA lead mobilization tests < 80 μg/3d) reinforces the conclusion that unrecognized low-level lead absorption contributes to renal failure due to other causes [25]. Lin’s laboratory reported that blood lead levels correlate with the rate of fall of GFR in patients with diabetic nephropathy (serum creatinine range 1.7-3.9 mg/dl) such that an increase in blood lead of 1 μg/dL predicted a reduction in GFR of 0.56 ml/min/1.73 m2 over 1 year of observation before chelation therapy (mean low blood lead group 5.9 μg/dL, N=15; mean high blood lead 7.5 μg/dL, N=15) [61]. These data reinforce the observation that low-level lead absorption accelerates the reduction in GFR in diabetic nephropathy made in Boston [50]. Following 2 years of chelation therapy (averaging a total of 7.0 g CaNa2EDTA) the GFR increased an average of 6 ml/min/1.73 m2 compared to a decrease of 1.4 ml/min/1.73 m2 in the untreated group [61]. Additional observations from Lin’s laboratory raise the possibility that CaNa2EDTA may improve GFR in all patients with reduced GFR. They treated 32 nondiabetic patients (mean creatinine 2.1 mg/dL; mean blood lead 5.3 μg/dL) with CaNa2EDTA, 4-13 g IV, over two years. The treated group had an increase in GFR averaging 3.4 ml/min while untreated controls (N=32) had a decrease in GFR of 1.0 ml/min [62]. The effect of comparable chelation on an important control group of renal failure patients with virtually no lead absorption was, unfortunately, not studied. Although these findings from Taiwan need to be confirmed in other laboratories, they raise the possibility that the beneficial effect of CaNa2EDTA may be unrelated to urinary lead excretion. The salutary results might, for example, be due to a non-specific antioxidant effect of CaNa2EDTA that increases GFR.

780

Treatment Lead nephropathy is important because it is one of the few renal diseases that is preventable. Moreover, lead-induced acute renal dysfunction can sometimes be reversed by chelation therapy [19, 28, 63]. The salutary effect of chelation therapy appears to be on the acute reduction in GFR and the acute elevation of blood pressure associated with elevated blood lead concentration rather than on the long-term effects of cumulative exposure associated with endothelial dysfunction, hypertension, and interstitial nephritis. There is no evidence that such therapy reverses established interstitial nephritis. The partial remission achieved among moonshiners and lead workers appears to represent reversal of the physiologic effects of acute poisoning superimposed on chronic lead nephropathy. No improvement in renal function has been observed once advanced interstitial nephritis is present and the steady-state serum creatinine concentration exceeds about 3 mg/dL [64]. Despite the effectiveness of chelation therapy in increasing the rate of lead excretion, the most appropriate treatment for asymptomatic excessive lead absorption is preventing further exposure. Elderly males store about 500 mg of lead in their bones in the absence of unusual exposure while occupational exposure may result in several grams stored in bones. Chelation therapy briefly increases the rate of removal of lead from the body, but, in the long run, the negative balance established by preventing further exposure is far more effective in reducing the body burden. The unstimulated daily excretion of lead ranges from a few micrograms in those without unusual exposure to hundreds of micrograms per day in those with heavy exposure. Chelation therapy increases lead excretion 10 or 20 fold for a few days, but in total cannot match the negative balance due to unstimulated daily excretion occurring over decades when intake of lead approaches zero. Prevention of lead intake is therefore far more effective than chelation therapy in asymptomatic individuals. However, chelation therapy is justified in the face of symptomatic lead poisoning or when blood levels exceed about 80 μg/dL because of the danger of lead encephalopathy.

34. Lead nephropathy

In summary, chelation therapy is justified in cases of symptomatic lead poisoning or when the blood lead exceeds about 80 μg/dL. When no symptom end-point

is clearly defined, chelation for blood lead < 80 μg/dL is not usually justified.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

Wedeen RP. Poison in the Pot: the Legacy of Lead. Carbondale, Ill: Southern Illinois University Press, 1984. Lancereaux E. Saturnisme chronique avec acces de goutte et Arthritis uratiques. Compt Rend Soc de Biol (Paris) 1872; 2: 99106. Rabinowitz MB. Kinetic analysis of lead metabolism in healthy humans. J Clin Invest 1976; 58: 260-270. Wedeen RP, Maesaka JK, Weiner B, Lipat GA, Lyons MM, Vitale LF. Joselow MM. Occupational lead nephropathy. Am J Med 1975; 59: 630 641. Menke A, Munter P, Batuman V, Silbergeld EK, Guallar E. Blood lead below 0.48 βmol/L (10 ug/dL) and mortality among US adults. Circulation 2006; 114: 1388-1394. Sanchez-Fructuoso AI, Torralbo A, Arroyo M, Luque,\ M, Ruilope LM, Santos JL, Cruycyera A, Barrientes A. Occult lead intoxication as a cause of hypertension and renal failure. Nephrol Dial Transplant 1996; 11:1775-178. Van de Vyver FL, D’Haese PC, Visier WJ, Elseviers MM, Knippenberg LI, Lamberts LW, Wedeen, RP, De Broe ME. Bone lead in dialysis patients. Kidney Int 1988; 33: 601-607. Inglis JA, Henderson DA, Emmerson BT. The pathology and pathogenesis of chronic lead nephropathy occurring in Queensland. J Path 1978; 124: 65-76. Grandjean P. Biological effects of organolead compounds. Boca Raton, FL: CRC Press, 1984. Wedeen RP. The politics of lead. In: Sheehan HE, Wedeen RP, eds. Toxic circles: environmental hazards from the workplace into community. New Brunswick: Rutgers University, 1993:168-200. Fiedler, N., Weisel, C., Lynch, R., Kelly-McNeil, K., Wedeen, R., Jones, K., Udasin, I., Ohman-Strickland, P., and Gochfeld, M. Cognitive effects of chronic exposure to lead and solvents. Am. J. Indust. Med. 2003; 4: 413-423. Glenn BS, Stewart WF, Links JM, Todd AC, Schwartz BS. The longitudinal association of lead and blood pressure. Epidemiology 2003; 14: 30-36. Chisolm JJ, Harrison HC, Eberlern WR, Harrison HE. Aminoaciduria, hypophosphatemia and rickets in lead poisoning. Am J Dis Child 1955; 89: 159-68. Goyer RA, Wilson MH. Lead-induced inclusion bodies. Results of ethylenediaminetetraacetic acid treatment. Lab Invest 1975; 32: 149-56. Schumann GB, Lerner SI, Weiss MA, Gawronski L, Lohiya GK. Inclusion-bearing cells in industrial workers exposed to lead. Am J Clin Pathol 1980; 74: 192-6. Coresh J, Wei GL, McQuillan G, Brancati FL, Levey AS, Jones C, Klag MJ. Prevelence of high blood pressure and elevated serum creatinine level in the United States. 2001; 161; 1206-1216. Selevan SG, Landrigan PJ, Stern FB, Jones JH. Mortality of lead smelter workers. Am J Epidemiol 1985; 122: 673-83. Emmerson BT. Chronic lead nephropathy. The diagnostic use of calcium EDTA and the association with gout. Aust Med J 1963; 12: 310-324. Wedeen RP, Mallik DK, Batuman V, Bogden JD. Geophagic lead nephropathy: A case report. Environ Res 1978; 17: 409-415. Wedeen RP, Mallik DK, Batuman V. Detection and treatment of occupational lead nephropathy. Arch Intern Med 1979; 139: 5357. Batuman, V., Maesaka, J.K., Haddad, B., Tepper, E., Landy, E. and Wedeen, R.P. The role of lead in gout nephropathy. N. Engl. J. Med. 1981; 304: 520 523. Batuman V, Landy E, Maesaka JK, Wedeen RP. Contribution of lead to hypertension with renal failure. N Engl J Med 1983; 309: 17-21. Colleoni N, D’Aminco G. Chronic lead accumulation as a possible cause of renal failure in gouty patients. Nephron 1986: 44; 3235. Behringer D, Craswell P., Mohl C. Stoeppler M, Tritz E. Urinary lead excretion in uremic patients. Nephron 1986: 42 323-329. Lin J-A, Ho H-H, Yu C-C. Chelation therapy for patients with elevated body lead burden and progressive renal insufficiency. A randomized controlled trial. Ann Intern Med 1999; 130:7-13. Tepper L. Renal function subsequent to childhood plumbism. Arch Environ Health 1963; 7: 76-85.

781

WEDEEN

27. 28. 29. 30. 31. 32. 33.

34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53.

782

Hu H. A 50-year follow-up of childhood plumbism: hypertension, renal function, and hemoglobin levels among survivors. Am J Dis Child 1991; 145:681-687. Lilis R, Garvrilescu N, Nestroescu R, Dumitriu C, Roventa A. Nephropathy in chronic lead poisoning. Br J Ind Med 1968;25:196202. Martin D, Glass TA, Bandeen-Roche K, Todd AC, Shi W, Schwartz BS. Association of blood lead and tibia lead with blood pressure and hypertension in a community sample of older adults. Am J Epidemiol 2006; 163: 467-478. Chai S S, Webb RC. Effects of lead on vascular reactivity. Environ Health Perspect 1988; 78: 85-91. Steenland NK, Thun Mj, Furguson W, Port FK. Occupational and other exposures associated with male end stage renal disease: a case/control study. Am J Public Health 1990; 80: 153-159. Kopp SJ, Barron JT, Tow JP. Cardiovascular actions of lead and relationship to hypertension: a review. Scand J Work Environ Health 1985; 11: 15-23. Cardenas A, Roels H, Bernard AM, Barbon R, Buchet JP, Lauwerys RR, Rosello J, Ramis I, Mutti A, Franchini I, Fels LM, Stolte H, De Broe ME, Nuyts GD, Taylor SA, Price RG. Markers of early renal changes induced by industrial pollutants. II. Application to workers exposed to lead. Brit J Ind Med 1993; 50: 28-36. Mueller PW, Paschal PC, Hammel RR, Klincewitz SL, McNeil ML, Speirto B, Steinberg KK. Chronic effects in three studies of men and women occupationally exposed to cadmium. Arch Environ Contam Toxicol 1992; 23: 125-136. Endo G, Horiguchi S, Kiyota I. Urinary N-acetyl-ß-D-glucosaminidase activity in lead-exposed workers. J Appl Toxicol 1990; 10: 235-238. Wedeen RP, Udasin I., Fiedler N, D’Haese P, De Broe ME, Gelpi E, Jones KW, Gochfeld M. Urinary biomarkers as indicators of renal disease. Renal Failure. 1999; 21: 241-249. Peitzman SJ, Bodison W, Ellis I. Moonshine drinking among hypertensive veterans in Philadelphia. Arch Intern Med 1985; 145: 632-634. Nakagawa t, Kang D-H, Sanchez-Lozada LG, Srinivas TR, Sautin Y, Ejaz AA, Segal M, Johnson RJ. Unearthing uric acid: An ancient factor with recently found significance in renal cardiovascular disease. Kidney International 2006; 69:1722-1735. Vaziri ND, Sica DA. Lead-induced hypertension: Role of oxidative stress. Curreent Hypertension Reports 2004; 6: 314-320. Staessen J, Lauwerys RR, Buchet J-P, Bulpitt CJ, Rondia D, Vanenterghem Y, Amery A. Impairment of renal function with increasing blood lead concentration in the general population. The Cadmibel study group. N Engl J Med 1994; 327: 151-156. Mahaffey KR, Annest JL, Roberts J, Murphy RS. National estimates of blood lead levels: United States, 1976–1980. Association with selected demographic and socioeconomic factors. N Engl J Med 1982; 307: 573-579. Harlan WR. The relationship of blood lead levels to blood pressure in the U.S. population. Environ Health Perspect 1988; 78: 9-13. Pirkle JL, Schwartz J, Landia JR, Harlan WR. The relationship between blood lead levels and blood pressure and its cardiovascular risk implications. Am J Epidemiol 1985; 121: 246-258. Munter P, He J, Vupputuri S, Coresh J, Batuman V. Blood lead and chronic kidney disease in the general United States population: Results from NHANES III. Kidney Int 2003; 63: 1044-1050. Payton M, Hu H, Sparrow D, Young JB, Landsberg L, Weiss S. Relation between blood lead and urinary biogenic amines in community-exposed men. Am J Epidemiology. 1993; 138(10): 815-825. Lustberg M, Silbergeld E. Blood lead levels and mortality. Arch Intern Med 2002; 162: 2443-2449. Munter P, Menke A, DeSalvo KB, Rabito FA, Batuman V. Continued decline in blood lead levels among adults in the United States. Arch Intern Med 2006; 165: 2155-2161. Kim R, Rotnitski A, Parrow D, Weiss ST, Wager C, Hu H. A longitudinal study of low-level lead exposure and impairment of renal function. The Normative Aging Study. JAMA 1996; 275: 1177-1176. Hu H, Aro A, Payton M, Korrick S, Sparrow D, Weiss ST. The relationship of bone and blood lead to hypertension. The Normative Aging Study. JAMA 1996; 275: 1171-1176. Thais S-W, Korrick S, Schwartz J, Amasirlwardena C, Aro A, Sparrow D, Hu H. Lead diabetes, hypertension, and renal function: Normative Aging Study. Environ Health Perspect 2004; 112: 1178-1182. Cheng Y, Schwartz J, Sparrow D, Aro A, Weiss, ST, Hu H. Bone lead and bone lead levels in relation to baseline blood pressure and the prospective development of hypertension: The Normative Aging Study. Am J Epidemiology. 2001; 153(2): 164-171. Korrick SA, Hunter DJ, Rotnitzky A, Hu H, Speizer FE. Lead and hypertension in a sample of middle-aged women. Am J Public Health. 1999; 89(3): 330-335. Vupputuri S, He J, Munter P, Bazzano LA, Whelton PK, Batuman V. Blood lead level is associated with elevated blood pressure in blacks. Hypertension 2003; 41: 463-468.

34. Lead nephropathy

54.

55. 56. 57.

58. 59. 60. 61. 62. 63. 64.

Byung-Kook L, Lee G-S, Stewert WF, Kyu-Dong A, Simon D, Kelsey KT, Todd AC, Schwartz BS. Associations of blood pressure and hypertension with lead dose measures and polymorphisms in the vitamin D receptor and β-aminolevulinic acid dehydratase genes. Environ Health Perspectives 2001; 109: 383-389. Elmarsafawy SF, Nitin J, Schwartz J, Sparrow D, Nie H, Hu H. Dietary calcium as a potential modifier of the relationship between lead burden to blood pressure. Epidemiology 2006; 17(5): 531-7. Rothenberg S, Khondrashov V, Manalo M, Jiang J, Cuellar R, Garcia M, et al. Increases in hypertension and blood pressure during pregnancy with increased bone lead levels. Am J Epidemiology 2002; 156(12):1079-1087 Vigeh M, Ghazizadeh S, Yokayama K, Sakai T, Mazaheri M, Morita Y, Beheshti S, Kitamura F, Araki S. Relationship between increased blood lead and preganacy hypertension in women without occupational lead exposure in Tehran, Iran. Arch Environ Health 2004; 59: 70-75. Ekong EB, Jaar BJ, Weaver VM. Lead-related nephrotoxicity: A review of the epidemiologic evidence. Kidney Int 2006; 70: 20742084. Legator MS, Morris DL. What did Sir Bradford Hill really say? Arch Environ Health 2003; 58: 718-720. Wedeen RP. Blood lead levels, dietary calcium, and hypertension. Ann Intern Med 1985; 102:403. Lin J-L, Lin-Tan D-T, Yu C-C, Li Y-J, Huang Y-Y, li K-L. Environmental exposure to lead and progressive diabetic nephropathy in patients with type II diabetes. Kidney International 2006; 69: 2049-2056. Lin J-L, Lin-Tan D-T, Hsu K-H, Yu C-C. Environmental lead exposure and progression of chronic renal diseases in patients without diabetes. N Eng J Med 2003; 348: 277-286. Sehnert KW, Claque AF, Cheraskin E. The improvement in renal function following EDTA chelation and multi-vitamin-trace mineral therapy: a study in creatinine clearance. Med Hypoth 1984; 15: 301-304. Germain MJ, Braden GL, Fitzgibbons JP. Failure of chelation therapy in lead nephropathy. Arch Intern Med 1984; 144: 24192420.

783

35

Cadmium-induced renal effects Gunnar F. NORDBERG1, Teruhiko KIDO2 and Harry A. ROELS3 1Umea

University, Umea, Sweden, University, Kodatsuno, Kanazawa, Japan 3Université catholique de Louvain, Brussels, Belgium 2Kanazawa

Introduction ___________________________________________________________ 785 Exposure ______________________________________________________________ 785 Toxicokinetics

786

Toxic effects of cadmium _________________________________________________ 787 Acute toxicity Long-term exposure Reproductive toxicity Carcinogenicity

787 788 788 788

Nephrotoxicity _________________________________________________________ 788 Biomarkers of exposure and internal dose in humans __________________________ 789 Human nephrotoxicity by cadmium exposure ________________________________ 789 Sweden Japan Belgium Other countries

789 790 796 803

References ____________________________________________________________ 804

Introduction

T

he first observations on adverse renal effects of cadmium (Cd) exposure in humans were made by Friberg in the late 1940s [1]. He reported a high prevalence of proteinuria (65% using the nitric acid test and 81% using the trichloroacetic acid test) in Cd-exposed workers. In Japan, an unusual disease named “itai-itai byo” or “ouch-ouch disease” was reported in 1955 [2]. This disease is characterized clinically by bone and kidney damage. In 1968, the Japanese Ministry of Health and Welfare concluded that itai-itai disease was caused by

chronic Cd poisoning [3]. The kidneys are particularly affected by Cd following long-term exposure [4, 5]. Studies of workers chronically exposed to air borne Cd report renal effects as well as respiratory effects though less frequently. Therefore, the kidneys are considered the critical target organ for Cd in the general population as well as occupationally exposed population.

Exposure Low concentrations of the element Cd occur naturally in the environment. Human exposure in the gen-

NORDBERG, KIDO & ROELS

eral environment occurs mainly from ingested foods. Concentrations of Cd in food items from areas without industrial contamination are summarized in Table 1. For basic food items such as rice, potatoes and wheat, Cd concentrations usually are lower than 0.1 mg/kg, while higher concentrations occur naturally in certain meats or shellfish. The daily dietary intake of Cd has been estimated to be 10-20 μg in several countries of the European Union and in several studies from the USA [3, 5]. In areas contaminated by emissions from industrial activities much higher daily oral intakes have occurred with amounts up to 200-1800 μg in people living in such areas in Japan and China [3, 6, 7]. Cadmium can also occur as an aerosol in air. While inhalation of ambient air usually does not contribute significantly to the daily intake of Cd, cigarette smoking does. The content of Cd often is 1-2 μg per cigarette. Based on data concerning the Cd content of cigarettes, it has been estimated that smoking of 20 cigarettes per day results in a daily inhalation of 2-4 μg [3]. Since approximately 50% may be absorbed, this can result in an uptake of 1-2 μg of Cd per day. Occupational exposure in the Cd-related industries can be associated with the inhalation of considerable amounts of Cd. In the 1950s, before the health hazards of Cd were recognized, Cd concentrations in the air of the working environment were sometimes high, i.e. in the order of milligrams per m3. In recent years, concentrations in industrial air have been reduced to 2-50 μg/m3, with higher values being reported in some exceptional cases. Examples of Cd-related industrial activities include: manufacturing of alkaline (nickel-Cd) batteries, smelting operations involving copper/zinc-Cd ores or alloys, production of Cd-based pigments, soldering with silver-Cd containing solder and welding in Cd-containing materials. In several EU countries (e.g. Sweden) certain uses of Cd such as its use in pigments, in electroplating and in soldering have been banned.

Toxicokinetics Uptake Inhalation of airborne Cd leads to variable uptake depending on size and solubility of particles. The systemic uptake of aerosolized Cd with a particle size of 10 μm has been estimated to be about 7%, while the uptake following inhalation of a particle size of 0.1 μm 786

Table 1. Concentrations of cadmium in different foodstuffs*. Food Beef meat Beef kidney Fish meat (other than crab) Oysters Wheat grains Rice (non-contaminated areas) Milk Potatoes

Mean mg/kg wet weight 0.005-0.02 0.2-1.3 0.004-0.1 0.1-4.7 0.005-0.08 0.008-0.13 0.00017-0.002 0.01-0.06

*From Nordberg et al. [5].

may be as high as 50% [8]. After oral ingestion, systemic uptake has been reported to be 1-6% in animal experiments. Factors that have been shown to influence oral uptake are dose and composition of the diet. In humans, the systemic uptake usually varies between 3 and 10% of the oral intake. In individuals with depleted body iron stores, uptake may be as high as 20% or even higher [5, 8]. Transport and distribution Figure 1 represents the uptake and transfer of Cd to the kidney. Following uptake, Cd is primarily bound in serum to albumin, the form in which it is transported to the various body pools. Cadmium bound to albumin (which is the dominating form in plasma shortly after uptake) is taken up primarily by the liver where it accumulates, and is dissociated. Released Cd-ions induce the synthesis of metallothionein which results in an increasing proportion of liver Cd being bound to metallothionein. The uptake of albumin-bound Cd by liver cells may be mediated by albumin receptors on the sinusoidal surfaces of hepatocytes [9]. In long-term chronic exposure a slow release of Cd-metallothionein from liver to blood occurs. During the phase when plasma Cd is bound to albumin, there is only limited uptake of Cd in the kidney. A long time after a single exposure or in a long-term chronic exposure situation, a considerable proportion of plasma Cd is bound to metallothionein. The Cd-metallothionein complex, because of its small molecular size, is filtered by the glomerular membrane and is efficiently taken up by renal tubular cells. Moreover, metallothionein-bound Cd is taken up more efficiently by renal cells of Cdexposed animals than by cells from non-exposed animals [10]. After entering renal tubule cells via

35. Cadmium-induced renal effects

Figure 1. Pathways of cadmium uptake and interaction with target sites in the kidney.

pinocytosis [11] or, as shown by Bernard et al. [12], by adsorptive endocytosis the Cd-metallothionein complex is catabolized in lysosomes releasing toxic Cd ions [13]. The balance between metallothioneinbound and non-metallothionein-bound Cd in the cell is considered of importance for the expression of toxicity. Non-metallothionein Cd in renal cells induces de novo synthesis (Figure 1). This process may account for the long biological half-life of Cd in the kidney where the element may be retained 10-20 years [5, 8]. Such a long biological half-life explains why Cd continues to accumulate in humans up to 50 years of age, reflecting the historical intake from the environment. Cadmium does not readily pass the blood-brain barrier, the blood-testis barrier or the placental barrier, but it accumulates in the placenta of animals and humans [14]. In humans, placental transfer of Cd seems to be very low and Cd is found to accumulate in the placenta [15, 16]. Excretion of cadmium The daily elimination of Cd (0.01-0.02 % of the body burden per day) via urine and feces is small as would be expected from the element’s long biological half-life [5, 8]. This implies that there is an age-related accumulation of Cd in the body and the increased urinary excretion of Cd with age is due to the increasing body burden. While this interrelationship has been documented in humans on a group basis, there exists a large variation among individuals. Cadmium is also excreted in the feces, but the majority of fecal Cd consists of the unabsorbed fraction of the metal passing through the gastrointestinal tract. The fecal content is, often a good indicator of dietary Cd intake since 90% or more of the ingested amount is unabsorbed and

eliminated via feces. True fecal elimination of the body burden of Cd is difficult to study in humans due to the preponderance of unabsorbed Cd. Data from animal experiments indicate that fecal elimination is dependent both on dose and body burden. Thus, in long-term low-level exposures, the fecal excretion may be largely related to body burden [5, 13]. The daily fecal content of Cd in persons whose exposure is limited to the general environment, is approximately 50 times higher than the urinary excretion. Mathematical models of cadmium toxicokinetics A mathematical model of long-term toxicokinetics in humans has been developed [17, 18]. Subsequently, a more detailed description of Cd toxicokinetics was formulated considering additional events that modify the behaviour of Cd in humans and includes relationships between levels in urine, blood and major organs [19, 20]. The kidney and particularly the cortex, is considered the critical target tissue for Cd and its accumulation is of decisive importance for risk assessment. In longterm exposures (life-long) either a simple one-compartment model or a multi-compartment model predicts that 1/3 to 1/2 of the total body burden accumulates in the kidney and that the concentration of Cd in the kidney cortex is 1.25 times higher than the average concentration in the whole kidney [5, 8].

Toxic effects of cadmium Acute toxicity Acute effects of excess Cd in the diets or drinking water of humans (more than 15 mg Cd/kg) involve vomiting and diarrhoea [21]. Acute inhalation of high 787

NORDBERG, KIDO & ROELS

concentrations of Cd (about 5 mg/m3 or higher) causes effects on the lungs in the form of pneumonitis and may be lethal [3, 5].

Long-term exposure Pulmonary toxicity may occur after long-term exposure to inhaled Cd. In such situations emphysema and other chronic pulmonary effects have been observed both in animals and in humans. Respiratory effects of Cd have not been recorded in the general population [3, 5].

Reproductive toxicity It is well known that the injection of Cd in experimental animals induces testicular necrosis in males and placental necrosis in pregnant females. Whether such effects may also occur after long-term dietary or environmental exposure in animals or humans is still a matter of discussion [5, 22, 23]. In humans placental transfer of Cd is limited (see above). A protective role of metallothionein in both human placenta and pregnant rats exposed to Cd may explain the lack of an effect on birth weights of children from Cd-exposed female Cd battery workers [23, 24].

Carcinogenicity Cadmium has been reported to induce cancer in animals at the site of injection. Respiratory cancers may occur after inhalation of Cd compounds [25]. There is also epidemiological evidence of an association between Cd exposure and cancer in occupational groups such as smelter and battery workers [26]. Both prostate and lung cancers have been reported to occur in increased frequency. The International Agency for Research on Cancer (IARC) [25] concluded that there was sufficient evidence supporting the carcinogenicity of Cd, although methodological problems in the interpretation of the studies have been recognized [26, 27]. The overall standardized mortality and incidence ratios of all malignant neoplasms among persons in Japan previously exposed to environmental Cd, were not significantly increased [28]. Some studies performed after the IARC assessment, have not given support for carcinogenicity of Cd [29] while other studies have given such support [30, 31]. 788

Nephrotoxicity It has long been recognized that Cd exposure either after inhalation or ingestion, can give rise to nephrotoxicity in humans and that this effect is usually considered to be the earliest and most important health effect [32]. In this regard, the dominating effect was recognized early and consisted primarily of injury to the renal tubules inducing a proteinuria characterized by the excretion of low molecular weight (LMW) plasma proteins. As noted previously, in long-term exposures to Cd, both in experimental animals and in humans there is continuous accumulation of Cd in liver and kidneys. Nephrotoxicity in animal experiments usually does not develop until the concentration of Cd in the renal cortex is in the range of 100-400 μg/g wet weight. Increased concentrations of urinary LMW proteins were found in ±10% of a study population of industrial workers, having a Cd concentration in the kidney cortex of about 200 μg/g as assessed by in vivo neutron activation analysis [33-35]. Reports from Belgium indicate that, in workers with a urinary Cd excretion lower than 10 μg/g creatinine, renal effects may occur [36, 37], whilst concentrations of urinary Cd as low as 2 to 4 μg/g creatinine have been associated with an increased prevalence of various indicators of renal tubular dysfunction in the general population [38]. This concentration in urine corresponds to a renal cortical concentration varying between 50-100 μg/g wet weight [29]. Although renal cortical concentrations greater than 50 μg/g wet weight may be accompanied with mild effects on the renal tubules in humans who have been exposed to Cd for a long time, it has been demonstrated in animal models that renal tubular injury can occur following injection of Cd-metallothionein at concentrations in renal cortex as low as 10-20 μg/g wet weight [39, 40], whereas in animals with long-term exposure concentrations of 100 μg/g wet weight or higher are required [3]. The explanation for this discrepancy is most likely due to differences in metallothionein induction that may occur in these two situations. Indeed, in the long-term exposure situation, ample time will be available for the induction of a protective level of metallothionein synthesis, whereas this will not be the case of acute exposure after Cd-metallothionein injection. The acute injection delivers a bolus dose of this

35. Cadmium-induced renal effects

complex to the renal tubule where it is metabolized in the lysosomes and toxic Cd ions are released [11]. This non-metallothionein bound Cd interacts with sensitive sites in the renal cells like enzymes and high molecular weight membrane proteins, particularly in the basolateral membrane and causes cellular damage with related changes in cellular calcium balance (Figure 1) [41-43]. In addition to the protective effect of metallothionein, stress proteins may also participate in this protection [42, 44]. Cadmium is the most potent inducer of metallothionein synthesis. The detailed mechanisms underlying the induction, regulation and toxicological role in various tissues of metallothionein remain to be elucidated, but the metallothionein genes have been identified [45] and the toxicological role has been reviewed [46].

Biomarkers of exposure and internal dose in humans Cadmium levels in blood are generally recognised as a biomarker of recent exposure to cadmium. It can also be used as biomarker of cumulative internal dose and accumulation of cadmium, but only when there is long-term (decade long) continuous exposure, for example in subsistence farmers consuming their own crops. Cadmium levels in urine are a widely recognised biomarker of cumulative internal dose, kidney and body burden of Cd. Dose-response relationships between urinary Cd and occurrence of kidney effects are described in the subsequent sections of this chapter “Sweden”, ”Japan”, “Belgium”, and “Other countries”. Reports concerning metallothionein in plasma and urine of Cd-exposed persons are limited [47-49]. This is at least in part due to the fact that the accurate measurement of Cd and metallothionein levels in plasma appears to be difficult [50]. The concentration of metallothionein in urine and blood has to be measured using the Onosaka saturation method, radio-immunoassay (RIA) or enzyme-linked immunosorbent assay (ELISA). The detection limits in human serum and urine for metallothionein by RIA is 1 pg [50]. For ELISA the detection limits are higher. Normal values range between 0.01-1 ng/ml for serum and between 1-10 ng/ml for urine. Metallothionein concentrations in Cd-exposed workers are reported to vary between 2-11 ng/ml in plasma and 2-155 ng/ml in urine [47].

Human nephrotoxicity by cadmium exposure Sweden As mentioned in the introduction to this chapter, chronic Cd poisoning with proteinuria resulting from occupational exposure was identified in Sweden in the late 1940s by Friberg [1]. Subsequently, it was shown that the proteinuria is of the tubular type and that the LMW proteins that were plasma proteins were not reabsorbed because of tubular damage [51, 52]. As a result of the discovery of the role of metallothionein in the toxicology of Cd [47, 53, 54] and the recognition of metallothionein binding as an explanation of the long biological half-life of Cd, interest was focused on the long-term intake of Cd via food. Kjellström et al. [55] assessed the temporal evolution of Cd in Swedish wheat, sampled from 1880 to 1970, and found a statistically significant time-dependent increase. The possibility of a risk for renal dysfunction and disease as a result of long-term dietary Cd intake was recognized and the relationship between occupational Cd exposure, renal accumulation of the element and tubular proteinuria was established [56]. Based on animal and human studies investigating long-term exposure to Cd from food and inhalation, a critical concentration of Cd in renal cortex was related to the risk of developing renal effects. These estimates were published in extensive reviews and evaluation volumes [13, 17, 56-58]. They were confirmed and/or partly revised according to epidemiological data from Japan, Belgium, and China (see the respective separate sections) and later on summarized in reviews by Järup et al. [29] and Nordberg et al. [5]. It was concluded that a small increase (less than one percent above background) in the prevalence of tubular dysfunction is expected to occur at renal cortical Cd concentrations exceeding 50 μg/g wet weight. The corresponding level of urinary Cd was estimated at 2.5 μg/g creatinine. These levels distinguishing apparent thresholds in renal cortex and urine are valid for persons not simultaneously exposed to other toxic substances and not suffering diabetes [5] or other conditions with increased risks of renal disease (see following text “Other countries”). In an epidemiological study reporting data from subjects previously exposed to higher levels of Cd in Sweden [59] a relationship between current urinary Cd levels 789

NORDBERG, KIDO & ROELS

and increased excretion of tubular proteins was demonstrated. Relationships between urinary Cd and occurrence of osteoporosis have also been established [60]. Since the slightly increased urinary Cd levels observed in these studies resulted from past exposures whilst recent exposure to Cd most probably was considerably lower, it is difficult to estimate from these data at what levels of cumulative exposures and urinary Cd, one might expect increased proteinuria and osteoporosis. Studies in another Swedish area (Skane) reported a statistically significant increase of age-adjusted urinary β2-microglobulin clearance in relation to urinary Cd at urinary Cd levels below 1 μg/g creatinine, but there were no statistically significant increases in other indicators of renal dysfunction [61]. Another study among women in the same area [62], however reported statistically significant increases in urinary NAG (Nacetyl-β-D-glucosaminidase) and protein HC (α1-microglobulin) in those women having urinary Cd of 0.8 μg/g creatinine compared to those having 0.48 μg/g creatinine. Interactions with diabetes were indicated. These studies give support to the notion that tubular proteinuria/enzymuria might already be induced at lower cumulative exposures than those giving rise to 2.5 μg/g creatinine, particularly among diabetics (see also section “Other countries”).

Japan Clinical features of itai-itai disease The main features of itai-itai disease are osteomalacia and osteoporosis [2]. The patients usually have several fractures that are caused by events as trivial as coughing. They suffer from severe pain when sleeping or even breathing. Compression fractures in the spine resulting in skeletal deformity and eventually shortening of the stature may occur. Patients also develop a duck-like gait and progressive difficulties in walking. While most of the itai-itai patients are postmenopausal women with several pregnancies no hereditary factors have been identified. X-ray findings include marked decalcification and the presence of “Looser’s zones” localized at areas where pressure causes pain. In severe cases, multiple pathological fractures are found. Skeletal deformities are frequently observed in pelvic bones, costae, and thoracic and lumbar vertebrae. Blood chemistry showed an increase in serum alkaline phosphatase and decreases in serum inorganic phosphorus 790

and calcium, while urinalysis revealed proteinuria, glucosuria, and aminoaciduria. The urinary protein excretion is characterized by the so-called ‘tubular protein pattern’ consisting of mainly LMW proteins such as β2-microglobulin, retinol-binding protein, and lysosomal enzymes. The aminoaciduria of the patient is of the “generalized aminoaciduria” type. The Cd content in urine is remarkably high. Increased excretion of calcium is also noticed. The principal pathological changes in bones are similar to the combined findings of osteomalacia and osteoporosis. Nearly 60% of 75 autopsied itai-itai disease patients had some degree of osteomalacia. All of them had severe to extreme osteoporosis [63]. Although the kidney is contracted, there is no obvious change in the glomeruli. The tubuli however, show a marked atrophy and degeneration. By the end of March, 2006, 188 inhabitants living in the Jinzu River basin had been diagnosed with itai-itai disease and 2 were still alive [64]. Renal effects by cadmium exposure The typical Cd-induced proteinuria reported by Butler and Flynn resembles that of acquired Fanconi syndrome [65] and mainly consists of LMW proteins derived from the plasma [66]. β2-microglobulin excretion is considered as one of the best indicators of early Cd-induced nephropathy since serum concentrations are stable and analysis of β2-microglobulin using radio-, latex-, or ELISA-assays is sensitive and accurate [67]. Unlike β2-microglobulin, urinary α1-microglobulin is stable at pH down to 4.5 [17]. It can be analyzed using commercially available ELISA assays. A significant correlation has been reported between α1-microglobulin and β2-microglobulin in the urine of Cd-exposed subjects [17, 68]. The urinary excretion of metallothionein parallels urinary Cd and evidences early renal dysfunction as indicated by increased excretion of either β2-microglobulin or α1-microglobulin. Based on these results, the urinary excretion of metallothionein reflects not only the level of Cd exposure but also any renal dysfunction caused by long-term Cd exposure [18, 29]. Enzymes of higher molecular weight, which preclude filtration, enter the urine from renal proximal tubuli. They are also indicators of Cd-induced renal damage, which confirm renal tubular damage even in clinical states where the overproduction of LMW

35. Cadmium-induced renal effects

proteins in blood occurs. Of all the urinary enzymes, N-acetyl-β-D-glucosaminidase is the most widely studied and used indicator of renal tubular damage. An increased urinary N-acetyl-β-D-glucosaminidase activity has been documented in Cd-exposed subjects [69]. However, the N-acetyl-β-D-glucosaminidase activity in urine of itai-itai patients was less than twice that of the controls, while β2-microglobulin levels were more than 100-fold those of the controls [70, 71]. This suggests that urinary N-acetyl-β-D-glucosaminidase activity decreases when renal tubular epithelia destruction becomes so severe that the cells can no longer excrete the enzyme into the urine. N-acetyl-β-D-glucosaminidase is probably a better marker for the acute effects or initial stage of chronic effects. Urinary trehalase activity in inhabitants of Cd-polluted areas was significantly higher than in the reference area [72]. Intestinal-type alkaline phosphatase is specifically located in the S3-segment of the proximal tubuli [73]. Urinary intestinal-type alkaline phosphatase activity is significantly higher in the Cd-exposed subjects than in the non-exposed subjects [74]. The relationship between β2-microglobulin and intestinal-type alkaline phosphatase can be fit to a fourth-order mathematical function. The β2-microglobulin level corresponding to the inflexion point of intestinal-type alkaline phosphatase activity is smaller than that for N-acetyl-β-D-

glucosaminidase. This result supports the contention that intestinal-type alkaline phosphatase is more useful for detecting renal tubular damage in the early stage of Cd exposure. Higher molecular weight (HMW) proteins such as albumin or mucoproteins are also excreted by the Cd-exposed subjects [75]. Urinary levels of various indicators of Cd exposure assessed in subjects living near the Kakehashi River basin (one of the Cd-polluted areas in Japan) and nonexposed subjects are compared in Table 2. Some causal relations among various urinary indices were identified using path analysis method. Cadmium-induced renal dysfunction develops in the following order: Cd exposure → increased β2microglobulin and/or metallothionein → increased excretion of amino-nitrogen and/or total protein → increased excretion of glucose [76]. A decline of the creatinine clearance was also evident during the early stage of renal dysfunction and a significant correlation between tubular reabsorption of phosphate and glomerular filtration rate was reported in subjects exposed to Cd [77, 78]. These results provide evidence of Cd-induced glomerular dysfunction. However, histopathological examination revealed that while the glomeruli were relatively well maintained in number and size, renal tubuli were markedly damaged, resulting in obstruction of the lumen [79]. The mechanism responsible for the changes in glomerular

Table 2. Proteinuria and urinary cadmium in cadmium-exposed and non-exposed subjects. Cadmium-exposed subjects β2-microglobulin (μg/g creatinine) α1−microglobulin (μg/g creatinine) N-acetyl-β-D-glucosaminidase (U/g creatinine) Human intestinal alkaline phosphatase (IU/g creatinine) Mucoprotein (mg/g creatinine) Albumin (mg/g creatinine) Total protein (mg/g creatinine) Cadmium (μg/g creatinine)

Sex M F M/F

N 67 102 27

M F M F M F M F M F M F

39 36 18 22 67 102 67 102 67 102 67 102

Mean 7116 10934 18880 51.1 43.9 4.62 4.74 228.6 309.2 93.6 140.0 185.6 251.7 7.5 10.1

Non-exposed subjects

S.D. 6.38** 5.11** 5.5**

N 26 55 10

mean 141 174 352

S.D. 387 3.47 4.2

2.45** 2.21* 2.07** 2.99** 1.82** 1.84** 4.79** 3.60** 3.62** 2.73** 1.82** 1.74**

22 26 18 22 26 55 26 55 26 55 26 55

25.3 27.2 1.26 1.82 75.9 81.5 29.3 31.7 68.3 73.4 2.5 4.0

1.60 1.88 1.75 2.28 1.75 1.90 2.47 2.80 1.81 2.01 1.58 1.45

Mean S.D.: Geometric mean and geometric standard deviation.**: Significant difference from control (p 40 years and in women >30 years old were significantly higher than those of younger ages, whilst levels of subjects >50 years were significantly lower than those of subjects aged >60 years [87]. The epidemiological study reported by the Japan Environment Agency in 1989 failed to detect any renal tubular dysfunction among 7,196 persons in the Cd non-polluted areas, while in 202 persons among 13,570 (1.5%) of the Cd-polluted areas, proximal renal tubular dysfunction was seen [88]. A follow-up survey on 2,101 inhabitants (1,566 men and 535 women), who participated in a 1967-health survey and had resided in their actual rural community since birth, was conducted to determine the influence of environmental Cd exposure on the mortality of the general population in the Jinzu River basin. The Cox hazard ratios for males and females exposed to Cd concentration in rice >0.30 mg/kg were 1.42 and 1.10, respectively. Especially, this value is statistically significant in men. Since the mean Cd concentration in rice was closely related to the development of renal injury, the Cd-induced renal injury is believed to be the factor underlying the increased mortality observed [89]. Relationship between cadmium-induced renal and bone effects Itai-itai disease is considered the most advanced stage of chronic Cd intoxication. Cadmium-induced bone effects are also suggested to occur in the more advanced stage. Originally, attention was focused on osteomalacia in the diagnosis of this disease. Recent studies, however, showed that osteopenia, a main characteristic of osteoporosis, can be detected in the early stage of chronic Cd intoxication. Bone density was analyzed in 28 women with itaiitai disease, 92 men and 114 women with Cd-induced renal dysfunction and 44 men and 66 women living in non-polluted areas using a microdensitometer [90]. To assess the degree of bone density by microdensitom-

35. Cadmium-induced renal effects

etry, an X-ray of the hands along with an aluminum step-wedge was obtained and the bone density was measured at the middle site of the metacarpal bone 11 [91]. The values of indices for both cortical width and bone mineral content were significantly lower in itaiitai disease patients than the Cd-exposed subjects. The Cd-exposed women also showed a decrease in bone density compared with the non-exposed subjects. A significant decrease in bone density was also observed between Cd-exposed men and the non-exposed subjects, although the difference was not as distinct as in women. In other Cd-polluted areas such as the Jinzu River basin or Tsushima Island, a decrease in bone density in Cd-exposed subjects has also been reported using the same method [92, 93]. The relationship between the bone density and renal dysfunction was studied in 85 female inhabitants of the Cd-polluted Jinzu River basin aged 55 to 71 years who had various concentrations of β2-microglobulin in urine [92]. A significant negative correlation between the urinary β2-microglobulin level and indicators of microdensitometry was found. In a study involving 203 Cd-exposed subjects with renal dysfunction and 80 non-exposed subjects an association was observed between Cd-induced renal dysfunction and osteopenia [94]. The relationship between biological parameters such as urinary β2-microglobulin and serum creatinine, calcium and phosphorus, and microdensitometric indices were analyzed using multivariate analysis. Age, urinary β2-microglobulin, and serum creatinine were significantly associated with

indices of osteopenia in Cd-exposed men. In contrast, age showed the most significant association with the microdensitometric parameters in women of both groups. However, only in Cd-exposed women did urinary β2-microglobulin levels significantly correlate with indices of microdensitometry. Using ultrasonic equipment, bone density was measured in 35 Cd-exposed and 68 non-exposed subjects [95]. The bone density was significantly decreased in Cd-exposed subjects as compared to the non-exposed subjects. Values obtained with this method (which is considered to be safer since it lacks radiation exposure) showed a significant correlation with those measured by microdensitometry. Bone-G1a protein ((osteocalcin) is rapidly emerging as a clinically important diagnostic parameter of bone pathology since bone-G1a protein appears to be a highly specific marker of osteoblast function and is expressed during bone formation. Serum levels of bone-G1a protein were evaluated in 76 Cd-exposed subjects with renal tubular dysfunction and 133 nonexposed subjects [96]. Serum bone-G1a protein levels were higher in Cd-exposed subjects than in the nonexposed subjects. In 29 Cd-exposed men, bone-G1a protein, % tubular reabsorption of phosphorus (TRP) and base excess were found to show significant associations with the microdensitometry indicators. In 42 Cd-exposed women, parathyroid hormone, age, blood Cd and bone-G1a protein significantly correlated with the microdensitometric indicators. Only serum boneG1a protein showed a significant correlation in both

Table 3. Prevalence (%) of abnormal urinary findings in cadmium-exposed and non-exposed subjects. Age: Male N Glucose ≥20 mg/dl with protein ≥5 mg/dl Amino acids ≥300 mg/g creatinine β2-microglobulin ≥1000 μg/g creatinine Metallothionein ≥638 μg/g creatinine Female N Glucose ≥20 mg/dl with protein ≥5mg/dl Amino acids ≥300 mg/g creatinine β2-microglobulin ≥1000 μg/g creatinine Metallothionein ≥693 μg/g creatinine

50-59

Cadmium-exposed subjects 60-69 70-79 80Total

600 1.3 0.0 4.8 1.5

494 265 2.6 4.2 1.6 3.0 13.0** 28.7 6.5 7.5

713 0.6 5.9 4.9* 4.5

591 1.9 7.8 17.1* 10.2

340 7.1 10.6 36.5** 10.9

50-59

Non-exposed subjects 60-69 70-79 80-

Total

65 7.7 9.2 52.3 6.2

1424 2.6 1.8 14.3** 4.6

62 4.8 1.6 0 4.9

38 0 2.6 0 0

26 0 0 26.9 0

7 0 0 14.3 0

133 2.3 1.5 6.0 2.3

110 20.0 23.6* 61.8* 16.5

1754 3.5 8.6** 18.7** 8.4*

64 0 1.6 0 0

49 4.1 2.0 6.1 10.2

34 0 2.9 5.9 0

14 7.1 0 21.4 0

161 1.9 1.9 5.0 3.1

*: Significant difference from control (p1000 μg/g creatinine) of both sexes were higher than those of the general Japanese population, whereas the cumulative survival curves were lower than those of the urinary β2-microglobulin negative group [109]. A significant association was also found between urinary β2-microglobulin and mortality, using a Cox’s proportional hazards model. In multiple comparisons using four indices of renal dysfunction (i.e. urinary β2-microglobulin, protein, glucose and amino acid), urinary protein and β2-microglobulin in women and urinary protein in men were the most contributory factors to the mortality rates [110]. Data from a 7-year follow-up study in another Cdpolluted area (Nagasaki) showed that, in both men and women, serum β2-microglobulin and creatinine, as well as urinary total protein and β2-microglobulin were significantly related to mortality independent of age as assessed by the Cox’s proportional hazards model [111]. In advanced cases, the excess mortality of subjects with Cd-induced renal tubular dysfunction might, to some extent, be ascribed to a reduction in GFR. In conclusion, these results suggest that the prognosis of subjects with Cd-induced renal dysfunction is unfavourable. The mortality rate tended to become higher as the severity of renal dysfunction progressed. Moreover, an isolated increase in urinary β2-microglobulin is an important factor in assessing the prognosis of a person’s mild proximal tubular dysfunction.

Belgium Cadmium is an important occupational and environmental pollutant in Belgium. This is mainly due to the long-standing commercial production of this metal 796

as a “by-product” of the mining and refining of zinc ores, which contain minor quantities of Cd (0.1–0.3 %). Because of the presence of zinc/lead ores, non-ferrous metallurgy workshops developed in the Meuse–Vesdre Valley near Liège as early as the 18th century. After Canon Jean-Jacques Dony discovered in Liège a coalbased thermic process to extract zinc from zinc blende (ZnS), an industrial revolution occurred in the zinc metallurgy from the 1850s on. This industry expanded rapidly in the Liège area concomitantly with coal mining and iron and steel works. After the 1st world war, increasing amounts of imported zinc ores were refined using the DONY-process, which in 1972 was replaced by electrolytic zinc refinery. Dust and waste from the primary zinc industry constituted the bulk of the basic material for the production of Cd using thermic refinery processes. After the 2nd world war, however, the heavy industrial activity in the Meuse Valley basin declined which resulted in a progressive shut down of non-ferrous industries of which all activity ceased in the early 1980s. In 1888, a similar non-ferrous metallurgic activity developed in the northeast of Belgium, a rural region near the Dutch border (Noorderkempen), where during the 20th century several primary zinc smelters and Cd refineries were in operation. Thermic processes, such as the horizontal retort zinc furnace for reduction of zinc calcine with coal at 1100-1300°C (DONY-process), were widely used in zinc refineries. As the boiling point of Cd (765°C) is much lower and the technology to recover Cd fume/dust from zinc furnaces was not very efficient, thermic processes were one of the main causes of the large scale dispersion of Cd in this rural area comprising about 300 km2. For instance, the Lommel-Overpelt smelters refined 250 metric tonnes of Cd in 1950 whereby 340 kg Cd/day were emitted in the air. Technological improvements raised the Cd production to 300 metric tonnes in 1970 while the atmospheric Cd losses, though still high, had dropped to 200 kg Cd/ day. By 1974, the ore-roasting/electrolysis-based zinc refining process had replaced the zinc furnaces and in the 1980s the electrolytic Cd refinery process, used since 1935 in the Balen smelter, was generalized and the re-melting and casting ovens modernized. Hence, for 600 metric tonnes of Cd produced in 1989 only 0.35 kg Cd/day was lost in the air. In 1992, the two remaining smelters in the Noorderkempen produced 1000 metric tonnes of Cd whereby barely 0.04 kg Cd/day escaped

35. Cadmium-induced renal effects

in the atmosphere [112]. Belgium has always been a major Cd producer in Europe and by 1997 its share of the Cd production in the European Union was 20%. To date, one high-performing big zinc smelter remains in the Noorderkempen, however, its Cd refinery facility shut down in the spring of 2002. Although cases of acute Cd intoxication were first recorded in 1858 in Belgium (domestic servants polishing silverware with Cd carbonate) [113], it should be pointed out that before 1970 systematic epidemiological studies had never been conducted in Belgium to assess health risks of Cd exposure in the industrial setting or the general population. For historical reasons, however, it is interesting to mention the 1953-report of the occupational physician of the Balen plant (Vieille Montagne at that time) dealing with clinical observations made in a group of 30 workers who were exposed to fume and dust of Cd in the Cd production facility of this plant [114]. In 10 workers with less than 2 years of exposure, a slight reticulocytosis was seen and the urinary Cd concentrations ranged 10-20 μg Cd/L as assessed by the dithizon method. In 8 workers with 2-8 years of exposure, the characteristic yellow dental Cd line was noticed together with a reticulocytosis exceeding 2%, a mild hyperchromic anemia, and a urinary Cd varying from 20 to 90 μg Cd/L. The same observations were made in twelve workers with 8 to 30 years of exposure and urinary Cd levels usually of 60 μg Cd/L, but interestingly in seven of them a proteinuria was found which displayed the same LMW protein characteristics as the Cd proteinuria already described by Friberg in 1948 [115].

Occupational exposure to cadmium in Belgium Critical cadmium concentration in kidney and urine In the early 1970s, Lauwerys et al. conducted the first cross-sectional epidemiological survey ever in Belgian factories. The workers (31 women, 49 men) were exposed to Cd dust and fume and were recruited from an electronic workshop, a nickel-Cd storage battery factory, and a Cd producing plant [116]. At the time of the study, the average airborne Cd ranged from 31 to 134 μg Cd/m3 (total dust), which was below the American Conference of Governmental Industrial Hygienists’ (ACGIH) threshold limit value (TLV) being 200 μg Cd/m3 in 1972. The kidney was found more sensitive to Cd exposure than the lung. Proteinuria

showed abnormal electrophoretic patterns of LMW and/or HMW proteins in 4/27 male workers with less than 20 years of exposure and in 15/22 with more than 20 years of exposure. Moreover, on the basis of the correlation between total proteinuria and cadmium concentration in urine (CdU), it was suggested that the risk of renal damage would be low when CdU is kept below 15 μg Cd/g creatinine [116]. In addition, blood Cd was found to reflect current exposure to Cd, whereas Cd in urine would reflect body burden of Cd when industrial exposure is low to moderate, but it would reflect current exposure when industrial exposure is high [116]. Subsequent studies in male workers of two Cd refineries confirmed previous findings of other investigators [117-119], in that prolonged Cd exposure is usually characterized by microproteinuria due to impairment of the tubular reabsorption of plasma-derived LMW proteins, e.g. β2-microglobulin and retinol-binding protein [120]. An isolated glomerular effect with increased permeability of HMW proteins, e.g. albumin and transferrin, was less commonly found [120, 121]. To obtain a reliable and direct estimate of the critical body burden of Cd in relation to Cd nephropathy, the Cd concentrations of liver and left kidney were determined in 1978 in about 300 male workers from two Cd refineries using in vivo neutron activation analysis, and the urinary β2-microglobulin concentration was measured as well. A dose-response relation between liver Cd and prevalence of increased β2-microglobulinuria was found, indicating an increased prevalence (>5%) of abnormal β2-microglobulinuria when hepatic Cd was exceeding 30 μg Cd/g wet weight (Table 4). Unlike liver Cd, renal Cd was found to drop in workers with abnormal urinary β2-microglobuline concentrations and a concomitant steep rise in urinary Cd excretion may be seen. This study established that abnormal β2-microglobulinuria is likely to occur when Cd in the renal cortex or in the urine exceeds the critical concentrations of 216 μg Cd/g wet weight and 10.8 μg Cd/g creatinine respectively [34, 35]. Predictive significance of tubular proteinuria Further research on occupational Cd nephropathy aimed at a better understanding of the predictive value of Cd-induced microproteinuria and explored underlying features of early glomerular impairment seen in a few Cd-exposed workers. A retrospective examination 797

NORDBERG, KIDO & ROELS

Table 4. Dose-response relation between cadmium concentration in liver and abnormal β2-microglobulinuria in a group of 148 workers from two zinc/cadmium smelters in Belgium. Cadmium in livera (μg/g) 10-19 20-29 30-39 40-49 50-59 60-69 70-160

Number of workers

54 27 28 18 8 5 8

Prevalence of abnormal β2microglobulinuriab n % 0 0 1 4 3 11 3 17 2 25 2 40 8 100

Mean β2-microglobulinuria in workers with abnormal values (mg/g creatinine) 7.30 0.28 1.42 7.00 4.89 6.45

Adapted from Roels et al. [34]. aCadmium concentration measured in vivo by neutron activation analysis. bβ2-microglobulinuria was considered abnormal when exceeding 0.20 mg/g creatinine.

of serum creatinine, total proteinuria, aminoaciduria, albuminuria, and microproteinuria (retinol-binding protein and β2-microglobulin) was carried out in a group of nineteen workers (40 to 60 years old) with 16 to 42 years of occupational Cd exposure. These renal markers were measured on average 1.2 years before and 4.2 years after removal from exposure and showed in this group of workers that the Cd-induced nephropathy was not reversible when exposure ceased, that in particular the microproteinuria exacerbated, and that serum creatinine tended to increase [122]. A few workers turned into end-stage renal insufficiency (unpublished data). To better assess the health significance of Cd-induced microproteinuria in male Cd workers three studies were carried out. The first study was a 5-year prospective study conducted in 23 Cd workers removed from exposure because of the discovery of microproteinuria [123]. They were exposed for 25 years on average and at the time of the first examination the mean age of the group was 59 (46-68 years). The mean±SEM CdU in the subjects amounted to 22.2±2.9 μg Cd/L, the geometric means of urinary retinol-binding protein and β2-microglobulin were 1.57 and 1.77 mg/L respectively, whilst serum creatinine was normal (0.30 mg/g creatinine, and whose geometric mean (range) of CdU was 4.7 (2.1-8.8) and 11.1 (5.8-21.7) μg Cd/g creatinine respectively. The subjects in both groups aged on average 55 years (50 to 64 years) and all had a normal serum creatinine (0.30 mg/g creatinine) corroborated our earlier finding, namely that the risk of abnormal microproteinuria may dramatically increase when CdU regularly exceeds 10 μg Cd/g creatinine. However, when reduction of Cd exposure took place at the time β2-microglobulinuria did not exceed 0.30 mg/g creatinine, the risk of developing tubular dysfunction at a later stage was low, even in cases with historical CdU values occasionally >10 but always 9.8 mmol/24h) suggested a 10% prevalence of hypercalciuria when CdU exceeded 1.9 μg Cd/24h [38]. Hypercalciuria should be considered an early adverse tubulotoxic effect, because it may exacerbate the development of osteoporosis, especially in the elderly. A prospective study from 1992-1995 (median follow-up of 6.6 years) in the above-mentioned Cadmibel subcohort from the rural area showed for a two-fold increase in urinary Cd a significant (p 25

U-Hg (nmol/mmol creatinine)

Figure 4. Box-plots showing the relation between U-Hg and U-NAG in the exposed group (10th, 25th, 50th, 75th, 90th percentiles, and outliers indicated). (reproduced from Langworth [121], with permission)

35. Mercury-induced renal effects

N-acetyl-β-D-glucosaminidase (Figure 4). No significant correlation was evident for other renal parameters: U-albumin, U-orosomucoid, U-β2-microglobulin, Ucopper, S-creatinine, and S-β2-microglobulin. Studies on chlor-alkali workers in Scandinavia [122-124] have reported minimal and apparently reversible renal effects from mercury exposures in this occupational group as evaluated by urinary excretion of NAG, albumin and titers of autoantibodies. These investigators noted that a small number of susceptible individuals may exist and that selenium status appears to have a major effect on urinary NAG excretion [124]. There are no reports of human nephrotoxicity caused by release of mercury from amalgam fillings. This is supported by experimental data from ten humans where standard measurements of renal function (glomerular filtrate rate, urinary albumin excretion, β2-microglobulin, N-acetyl-β-D-glucosaminidase) were monitored before and 60 days after the removal of mercury amalgam fillings [125]. Bellinger et al [126] conducted a proper randomized controlled clinical trial in 534 children aged 6 to 10 years to find out if mercury released from amalgam fillings could give rise to any neuropsychological or renal (glomerular) effects. The number of amalgam resorted surfaces over five years was 14.6 in the amalgam group (n=267) and none in the composite group (n=267). Urinary excretion of mercury was slightly higher in the amalgam group; 0.9 μg and 0.6 Hg/g creatinine in amalgam and composite respectively but there was no difference in the urinary albumin excretion, and likewise no difference in a battery of neuropsychological tests. In an environmental epidemiological study from the UK [127] calculated airborne exposure to mercury and mortality in ‘Nephritis, nephritic syndrome and nephrosis’ have been significantly associated. However, as most individuals in the UK and other developed countries have access to Renal Replacement Therapy (RRT) nowadays and die from mostly other causes than nephritis and nephritic syndrome, the associations seen between modeled air levels of mercury and mortality in kidney disease should be regarded as ‘suggestive’ at the most.

Treatment Treatments currently available for mercurial poisoning in humans involve the use of thiol-based chelating agents such as British Anti-Lewisite (BAL), penicillamine [60] and more recently, agents such as 2, 3 dimercaptopropane-1 sulphonate (DMPS) [128] or 2, 3 dimercaptosuccinic acid (DMSA). Chelation is the formation of a metal ion complex in which the metal ion is association with a charged or uncharged electron donor. Studies by Bluhm et al. [129] compared the efficacy of D-penicillamine with dimercaptosuccinic acid (DMSA), and demonstrated that DMSA was able to increase the excretion of mercury to a greater extent than D-penicillamine. Studies in humans demonstrate that chelation therapy successfully lowers body burden of mercury and increases urinary mercury levels [130, 131]. However, the impact of chelation on long-term outcome of parenteral mercury exposure remains uncharacterized [132]. Standard dose regimens for the above chelators are as follows: penicillamine given with paradoxen at doses of 500 mg P.O., every 6 hours for 5 days; DMPS given at total a dose of 250 mg I.M. or I.V./4 hours on day 1, 6 hours on day 2, and on 6 to 8 hours on day 3 for the remaining course; DMSA given at 10 mg/kg P.O. every 8 hours for 5 days [133]. DMPS given IV at a dose of 50 mg/day to a 22-yearold man who had injected about 8 g of elemental mercury dramatically increased the urinary excretion of mercury but was – due to the relative small amount of mercury chelated and excreted daily – unable to eliminate the total load of mercury efficiently and it was not until residual mercury droplets were surgically extirpated, after three years, that the blood levels of mercury went down and the remaining symptoms of mercury poisoning (mainly tremor) disappeared [134]. Although DMPS has been used in Europe for the past 25 years (under the names Unithiol and Dimaval), it is not widely used in the United States because it is not approved as a drug.

821

FOWLER, WHITTAKER & ELINDER

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

28. 29. 30. 31.

822

United States Food and Drug Administration. Mercury compounds in drugs and foods. Center for Biologics Evaluation and Research (CBER). 2001. United States Geological Service (U.S.G.S.). U.S. Geological Survey, Mineral Commodity Summaries (Mercury). 2002. United States Geological Service (U.S.G.S.). U.S. Geological Survey, Mineral Commodity Summaries (Mercury). 1999. Chevalier, P. Mercury. Canadian Minerals Yearbook 1999. Natural Resources Canada. 2000. Byrne L. Brazil’s mercury poisoning disaster. Brit Med J 1992; 304: 1397. Schutte NP, Knight AL, Jahn O. Mercury and its compounds. In: Dickerson OB, Horovath EP, Zenz C, editors. Occupational medicine, 3rd ed. St Louis: Mosby-Year Book Inc, 1994: 549-57. National Research Council (NRC). Toxicological effects of methylmercury. 2000. Sällsten G, Barregård L, Österberg T. Tandgnissling hos amalgambärare – en orsak till hög kvicksilverutsöndring ? (Tooth grinding among wearers of amalgam fillings – a cause of high mercury release ?). Läkartidningen (Sweden) 1991; 88: 232-3. Geijersstam E, Sandborgh-Englund G, Jonsson F, Ekstrand J. Mercury uptake and kinetics after ingestion of dental amalgam. J Dent Res 2001; 80:1793-6. International Agency for Research on Cancer (IARC). IARC monograph on Beryllium, Cadmium, Mercury, and Exposures in the Glass Manufacturing Industry. Volume 58. Lyon, France. 1994: 239-345. United States Environmental Protection Agency (U.S. EPA). Mercury update: impact on fish advisories. 2001. Office of Water. Grandjean P, Weihe P, White RF, Debes F, Araki S, Yokoyama K, Murata K, Sorensen N, Dahl R, Jorgensen PJ. Cognitive deficit in 7-year old children with prenatal exposure to methylmercury. Neurotoxicol. Teratol. 1997; 19: 417-28. United States Food and Drug Administration (U.S. FDA). Methyl mercury levels in commercial seafood species. 2000. Center for Food Safety and Applied Nutrition. Office of Seafood. World Health Organization. Methylmercury. Environmental Health Criteria 101. World Health Organization, Geneva 1990. United States Environmental Protection Agency (U.S. EPA). Mercury study report to Congress. 1997. Office of Air Quality Planning and Standards and Office of Research and Development. Campbell D, Gonzales M, Sullivan JB. 1992. Mercury. In, Sullivan, J.B. & Krieger, G.R., eds, Hazardous Materials Toxicology. 824-33. Clinical Principles of Environmental Health, Baltimore:Williams and Wilkins. Elinder C-G, Friberg L, Nordberg GF, Kjellström T, Oberdoerster G. Biological monitoring of metals. Chemical safety monographs. International Programme on Chemical Safety. WHO/EHG/94.2 1994: 1-80. Sällsten G, Barregård L, Järvholm B. Mercury in the Swedish chloralkali industri - an evaluation of the exposure and preventive measures over 40 years. Ann Occup Hyg 1990; 34: 205-14. Gutiérrez F, Leon L. Elemental mercury embolism to the lung. New England Journal of Medicine 2000; 342(24): 1791. Diner B, Brenner B. Toxicity, mercury. eMed J. 2001; 2:1-13. Available at: www.emedicine.com/EMERG/topic813.htm. World Health Organization. Inorganic mercury. Environmental Health Criteria 118. World Health Organization, Geneva 1991. Hursh JB, Clarkson TW, Cherian MG, Vostal J, Mallie RV. Clearance of mercury (Hg-197 and Hg-203) vapor inhaled by human subjects. Arch Env Health 1976; 31: 302-9. Barregård L, Ställsten G, Schutz A, Attwell R, Skerfving S, Jarvholm B. Kinetics of mercury in blood and urine after brief occupational exposure. Arch Env Health 1992; 47: 176-84. Morrow PE, Gibb FR, Johnson L. Clearance of insoluble dust from the lower respiratory tract. Health Phys 1964; 10: 543-55. Clarkson TW. Human toxicity of mercury. The Journal of Trace Elements in Experimental Medicine 1998; 11: 303-17. Rahola T, Hattula T, Korolainen A, Miettinen JK. Elimination of free and protein-bound ionic mercury in man. Ann Clin Res 1973; 5: 214-19. Hall LL, Allen PV, Fisher HL, et al. 1994. The kinetics of intravenously-administered inorganic mercury in humans. In: Kinetic Models of Trace Elements and Mineral Metabolism During Development, K.M.S. Subramanian and M.E. Wastney, Ed. CRC Press, Boca Raton, FL. p. 1-21. Carmichael NG, Fowler BA. Effects of separate and combined chronic mercuric chloride and sodium selenite administration in rats: histologic, ultrastructural and x-ray microanalytical studies of liver and kidney. J Env Pathol Toxicol 1979; 3: 399-412. Aslamkhan AG, Han YH, Yang XP, et al.Human renal anion transporter 1 dependent uptake and toxicity of mercuric thiol conjugates in Madin-Darby canine kidney cells. Mol Pharmacol 2003; 63(3):590-6. Lash LH, Hueni SE, Putt DA, et al. Role of organic anion and amino acid carriers in transport of inorganic mercury in rat renal basolateral membrane vesicles: influence of compensatory renal growth. Toxicol Sci 2005; 88(2):630-44. Zalups RK, Ahmad S. Handling of cysteine S-conjugates of methylmercury in MDCK cells expressing human OAT1. Kidney Int 2005; 68(4):1684-99

35. Mercury-induced renal effects

32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.

60.

Aberg B, Ekman L, Falk R, Greitz U, Persson G, Snihs J. Metabolism of methyl mercury (Hg) compounds in man: Excretion and distribution. Arch Environ. Health 1969; 19: 478-84. Miettinen JK. 1973. Absorption and elimination of dietary (Hg++) and methyl mercury in man. In: Mercury, Mercurial, and Mercaptans, M.W. Miller and T.W. Clarkson, Ed. Springfield, IL. p. 233-43. Fowler BA, Lucier GW, Mushak P. Phenobarbital protection against methylmercury nephrotoxicity. Proc Soc Exp Biol Med 1975; 149: 75-9. Vahter M, Mottet NK, Friberg L, Lind B, Shen DD, Burbacher T. Speciation of mercury in the primate blood and brain following long-term exposure to methyl mercury. Toxicol Appl Pharmacol 1994; 124(2): 221-29. Fowler BA. The morphologic effects of dieldrin and methyl mercuric chloride on pars recta segments of rat kidney proximal tubules. Am J Pathol 1972; 69: 163-74. Yasutake A, Hirayama K, Inouye M. Sex difference in acute renal dysfunction induced by methyl mercury in mice. Renal Failure 1990; 12: 233-40. Fowler BA, Brown HW, Lucier GW, Krigman MR. The effect of chronic oral methylmercury exposure on the lysosome system of rat kidney. Morphometric and biochemical studies. Lab Invest 1975; 32: 313-22. Zalups RK. Molecular interactions with mercury in the kidney. Pharmacological Reviews. 2000; 52(1):113-143. Clarkson TW. Mercury. J. Am. Coll. Toxicol. 1989; 8(7): 1291-96. Hu H. Exposure to metals. Occupational and Environmental Medicine 2000; 27(4):983-96. Roels HA, Hoet P, Lison D. Usefulness of biomarkers of exposure to inorganic mercury, lead, or cadmium in controlling occupational and environmental risks of nephrotoxicity. Renal Failure 1999; 21(3 & 4): 251-62. Satoh H. Occupational and environmental toxicology of mercury and its compounds. Industrial Health 2000; 38: 153-64. Grandjean P, Weihe P, White RF, Debes F: Cognitive performance of children prenatally exposed to “safe” levels of methylmercury. Environ Res 1998; 77(2): 165-72. Legrand M, Passos CJ, Mergler D, et al. Biomonitoring of mercury exposure with single human hair strand. Environ Sci Technol 2005; 39(12):4594-8. Legrand M, Lam R, Passos CJ, et al. Analysis of mercury in sequential micrometer segments of single hair strands of fish-eaters 2007; 41(2):593-8. Xue F, Holzman C, Rahbar MH, et al. Maternal fish consumption, mercury levels, and risk of preterm delivery. Environ Health Perspect 2007; 115(1):42-7. Ohno T, Sakamoto M, Kurosawa T, et al. Total mercury levels in hair, toenail, and urine among women free from occupational exposure and their relations to renal tubular function. Environ Res 2007; 103(2):191-7. Byrne AR, Kosta L. Simultaneous neutron-activation determination of selenium and mercury in biological samples by volatilization. Talanta. 1974; 21: 1083-90. WHO (World Health Organization). 1976. Environmental Health Criteria: Mercury. World Health Organization, Geneva, Switzerland, 121. Hatch WR, Ott WL. Determination of sub-microgram quantities of mercury by atomic absorption spectrophotometry. Anal Chem 1968; 40(14): 2085-87. Magos L, Clarkson TW. Atomic absorption and determination of total, inorganic, and organic mercury in blood. J AOAC 1972; 55(5): 966-71. Von Burg R, Rustam H. Electrophysiological investigations of methyl mercury intoxication in humans: Evaluation of peripheral nerve by conduction velocity and electromyography. Electroenceph Clin Neurophysiol 1974; 37: 381-92. Cappon CJ, Smith JC. A simple and rapid procedure for the gas-chromatographic determination of methylmercury in biological samples. Bull Environ Contam. Toxicol 1978; 19(5): 600-7. Marsh, DO, Clarkson TW, Cox C, Amin-Zaki L, Al-Tikriti S. Fetal methyl mercury poisoning: Relationship between concentration in single strands of maternal hair and child effects. Arch Neurol 1987; 44: 1017-22. Mason HJ, Hindell P, Williams NR. Biological monitoring and exposure to mercury. Occup Med 2001; 51(1): 2-11. Kulig, K. A tragic reminder about organic mercury. The New England Journal of Medicine. 1998; 338(23): 1692-94. Goldman L, Shannon MW, The Committee on Environmental Health. Technical Report: Mercury in the Environment: Implications for Pediatricians. Pediatrics 2001; 108(1): 197-205. Litovitz TL, Klein-Schwartz W, White S, Cobaugh DJ, Youniss J, Drab A., Benson BE. The 1999 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance Systems. The American Journal of Emergency Medicine 1999; 18(5): 517-74. Berlin M. Mercury. In: Friberg L, Nordberg GF, Vouk VB, editors. Handbook on the toxicology of metals, Vol II. Amsterdam: Elsevier, 1986: 387-445.

823

FOWLER, WHITTAKER & ELINDER

61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86.

87. 88. 89.

824

Smith RG, Vorwald AJ, Patil LS, Mooney TF. Effects of exposure to mercury in the manufacture of chlorine. Am Ind Hyg Ass J 1970; 31: 687-700. Smith PJ, Langolf GD, Goldberg J. Effects of occupational exposure to elemental mercury on short term memory. Brit J Ind Med 1983; 40: 413-9. Roels H, Gennart JP, Lauwerys R, Buchet JP, Malchaire J, Bernard A. Surveillance of workers exposed to mercury vapour: validation of a previously proposed biological threshold limit value for mercury concentration in urine. Am J Ind Med 1985; 7: 45-71. Langworth S, Almkvist O, Söderman E, Wikström BO. Effects of occupational exposure to mercury vapour on the central nervous system. Brit J Ind Med 1992; 49: 545-55. Clarkson TW. Mercury - an element of mystery. New Eng J Med 1990; 323: 1137-9. Curtis HA, Ferguson SD, Kell RL, Samuel AH. Mercury as a health hazard. Arch Dis Childhood 1987; 62: 293-5. Langworth S, Bjorkman L, Elinder CG, Jarup L, Savlin P. Multidisciplinary examination of patients with illness attributed to dental fillings. J Oral Rehabil 2002; 29:705-13. Brownawell AM, Berent S, et al. ”The potential adverse health effects of dental amalgam.” Toxicol Rev 2005; 24: 1-10. Bates MN. ”Mercury amalgam dental fillings: an epidemiologic assessment.” Int J Hyg Environ Health 2006; 209: 309-16 Riddle M, Gardner F, Beswick I, Filshie I. The nephrotic syndrome complicating mercurial diuretic therapy. Brit Med J 1958; 1: 1274-7. Cameron JS, Trounce R. Membranous glomerulonephritis and the nephrotic syndrome appearing during mersalyl therapy. Guy’s Hosp Report 1965; 114: 101-7. Silbergeld E, Devine PJ. Mercury- are we studying the right endpoints and mechanisms. Fuel Processing Technology 2000; 65-66: 35-42. Langworth S, Elinder CG, Sundqvist KG: Minor effects of low exposure to inorganic mercury on the human immune system. Scand J Work Environ Health 1993; 19(6): 405-13. Langworth S, Elinder CG, Sundquist KG, Vesterberg O: Renal and immunological effects of occupational exposure to inorganic mercury. Br J Ind Med 1992; 49(6): 394-401. Ellingsen DG, Efskind J, Berg KJ, Gaarder PI, Thomassen Y. Renal and immunologic markers for chloralkali workers with low exposure to mercury vapor. Scand J Work Environ Health 2000; 26(5): 427-35. De Broe M. Renal toxicity from environmental toxins, drugs, and contrast agents. In: Essential Atlas of Nephrology. Lippincott, Williams & Wilkins, 2001. Hua J, Pelletier L, Berlin M, Druet P. Autoimmune glomerulonephritis induced by mercury vapour exposure in the Brown Norway rat. Toxicology 1993; 79: 119-29. Gritzka TL, Trump BF. Renal tubular lesions caused by mercuric chloride: electron microscopic observations. Am J Pathol 1968; 52: 1225-78. Ganote CE, Reimer KA, Jennings RB. Acute mercuric chloride nephrotoxicity: an electron microscopic and metabolic study. Lab Invest 1975; 31: 633-47. Kempson SA, Ellis BG, Price RG. Changes in rat renal cortex, isolated plasma membranes, and urinary enzymes following the injection of mercuric chloride. Chem Biol Interact 1977; 18: 217-34. Trump BF, Berezesky IK, Sato T, Laiho KU, Phelps PC, DeClaris N. Cell calcium, cell injury and cell death. Environ Health Perspect 1984; 54: 281-7. Trump BF, Berezesky IK, Smith MW, Phelps PC, Eliget KA. The relationship between cellular ion deregulation and acute and chronic toxicity. Toxicol Appl Pharmacol 1989; 97: 6-22. Smith MW, Ambudkar IS, Phelps PC, Regec AL, Trump BF. HgCl2-induced changes in cytosolic Ca2+ of cultured rabbit renal tubular cells. Biochim Biophys Acta 1987; 931: 130-42. Nielsen JB, Andersen HR, Andersen O, Starklint H. Mercuric chloride-induced kidney damage in mice: time course and effect of dose. J Toxicol Environ Health 1991; 34: 469-83. Tarabova B, Kurejova M, Sulova Z, et al. Inorganic mercury and methylmercury inhibit the Cav3.1 channel expressed in human embryonic kidney 293 cells by different mechanisms. J Pharmacol Exp Ther 2006; 317(1):418-27. Wang Y, Bollard ME, Nicholson JK, et al. Exploration of the direct metabolic effects of mercury ll chloride on the kidney of SpragueDawley rats using high-resolution magic angle spinning 1H NMR spectroscopy of intact tissue and pattern recognition. J Pharm Biomed Anal 2006; 40(2):375-81. Brandao R, Santos FW, Farina M, et al. Antioxidants and metallothionein levels in mercury-treated mice. Cell Biol Toxicol 2006; 22(6):429-38. Brkljacic J, Perisic T, Dundjerski J, et al. Interaction of rat renal glucocorticoid receptor with Hsp90 and Hsp70 upon stress provoked by mercury 2007; 27(1):43-50. Woods JS, Fowler BA. Renal porphyrinuria during chronic methyl mercury exposure. J Lab Clin Med 1977; 90: 266-72.

35. Mercury-induced renal effects

90. 91. 92. 93. 94.

95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118.

Hultman P, Bell LJ, Eneström S, Pollard KM. Murine susceptibility to mercury. I. Autoantibody profiles and systemic immune deposits in inbred, congenic, and intra-H-2 recombinant strains. Clin Immunol Immunopathol 1992; 65: 98-109. Hultman P, Eneström S. Mercury induced antinuclear antibodies in mice: characterization and correlation with renal immune complex deposits. Clin Exp Immunol 1988; 71: 269-74. Hultman P, Eneström S. Mercury induced b-cell activation and antinuclear antibodies in mice. J Clin Lab Immunol 1989; 28: 14350. Bernard AM, Collette C, Lauwerys R. Renal effects of in utero exposure to mercuric chloride in rats. Arch Toxicol 1992; 66: 50813. Bellon B, Capron M, Druet E, Verroust M-CV, Sapin C, Girard JF, Foidart JM, Mathieu P, Druet P. Mercuric chloride induced autoimmune disease in Brown-Norway rats: sequential search for anti-basement membrane antibodies and circulating immune complexes. Eur J Clin Invest 1982; 12: 127-33. Esnault VLM, Mathieson PW, Thiru S, Olveira DBG, Lockwood M. Autoantibodies to myeloperoxidase in Brown Norway rats treated with mercuric chloride. Lab Invest 1992; 67: 114-20. Havarinasab, S, Bjorn, E, Ekstrand, J, et al. Dose and Hg species determine the T-helper cell activation in murine autoimmunity. Toxicology 2007; 229(1-2):23-32. Klein R, Herman SP, Bullock BC, Talley FA. Early functional and pathological changes in rat kidney during methyl mercury intoxication. Arch Pathol 1976; 96: 83-90. Fowler BA. Ultrastructural evidence for nephropathy induced by long-term exposure to small amounts of methylmercury. Science 1972; 175: 780-1. Gage JC. The toxicity of alkyl- and arylmercury salts. Biochem Pharmacol 1961; 8: 77. dos Santos AP, Mateus ML, Carvalho CM, et al. Biomarkers of exposure and effect as indicators of the interference of selenomethionine on methylmercury toxicity. Toxicol Lett 2007; 169(2):121-8. Troen P, Seymour A, Kaufman SA, Katz KH. Mercuric bichloride poisoning. N Engl J Med 1951; 244: 459-63. Wands JR, Weiss SW, Yardley JH, Maddrey WC. Chronic inorganic mercury poisoning due to laxative abuse. Am J Med 1987; 57: 92-101. Munck O, Nissen NI. Development of nephrotic syndrome during treatment with mercurial diuretics. Acta Med Scand 1956; 153: 397-413. Burston J, Darmady EM, Stranack F. Nephrosis due to mercurial diuretics. Brit Med J 1958; 1: 1277-8. Becker CG, Becker EL, Maher JF, Schreiner GE. Nephrotic syndrome after contact with mercury. Arch Intern Med 1962; 83: 17886. Preedy LRK, Russel DS. Acute salt depletion associated with the nephrotic syndrome. Development during the treatment with a mercurial diuretic. Lancet 1953; 2: 1181-4. Williams NE, Bridge HG. Nephrotic syndrome after the application of mercury ointment. Lancet 1958; 2: 602. Wilson VK, Thomson ML, Holzel A. Mercury nephrosis in young children. Brit Med J 1952; 1: 358-60. Langworth S, Bjorkman L, Elinder CG, Jarup L, Savlin P. Multidisciplinary examination of patients with illness attributed to dental fillings. J Oral Rehabil 2002; 29:705-13. Barr RD, Rees PH, Cordy PE, Kungu A, Woodger BA, Cameron HM. Nephrotic syndrome in adult African in Nairobi. Brit Med J 1972; 2:131-4. Oliviera DBG, Foster G, Savill J, Syme PD, Taylor A. Membranous nephropathy caused by mercury-containing skin lightning cream. Postgr Med J 1987; 63: 303-4. Weldon MM, Smolinski MS, Maroufi A, Hasty BW, Gilliss DL, Boulanger LL, Balluz LS, Dutton RJ. Mercury poisoning associated with a Mexican beauty cream. West J Med 2000; 173:15-18. Tang HL, Chu KH, et al. “Minimal change disease following exposure to mercury-containing skin lightening cream.” Hong Kong Med J 2006; 12(4): 316-8. Agner E, Jans H. Mercury poisoning and nephrotic syndrome in two siblings. Lancet 1978; 2: 951. Friberg L, Hammarström S, Nyström A. Kidney injury after chronic exposure to inorganic mercury. Arch Ind Hyg Occup Med 1953; 8: 149-53. Kazantzis G, Schiller KFR, Asscher AW, Drew RG. Albuminuria and the nephrotic syndrome following exposure to mercury and its compounds. Quart J Med 1962; 31: 403-19. Aymaz S, Gross O, et al. “Membranous nephropathy from exposure to mercury in the fluorescent-tube-recycling industry.” Nephrol Dial Transplant 2001; 16(11): 2253-5. Buchet JP, Roels H, Bernard A, Lauwerys R. Assessment of renal function of workers exposed to inorganic lead, cadmium or mercury vapor. J Occup Med 1980; 22: 741-50.

825

FOWLER, WHITTAKER & ELINDER

119. Stonard MD, Chater BV, Duffield DP, Nevitt AL, O’Sullivan JJ, Steel GT. An evaluation of renal function in workers occupationally exposed to mercury vapor. Int Arch Environ Health 1983; 52: 177-89. 120. Langworth S, Elinder CG, Sundquist KG, Vesterberg O. Renal and immunological effects of occupational exposure to inorganic mercury. Br J Ind Med 1992; 49: 394-401. 121. Langworth S. Early effects of occupational and environmental exposure to inorganic mercury (dissertation). Karolinska Institute Stockholm, Sweden 1992. 122. Ellingsen DG, Barregard L, Gaarder PI, et al. Assessment of renal dysfunction in workers previously exposed to mercury vapour at a chloralkali plant. Br J Ind Med 1993; 50(10):881-7. 123. Barregard L, Enestrom S, Ljunghuse O, et al. A study of autoantibodies and circulating immune complexes in mercury-exposed chloralkali workers. Int Arch Occup Environ Health 1997; 70(2):101-6. 124. Efskind J, Ellingsen DG, Hartman A, et al. Renal function of chloralkali workers after the cessation of exposure to mercury vapor. Scan J Work Environ Health 2006; 32(3):241-9. 125. Sandborgh-Englund G, Nygren, AT, Ekstrand J, Elinder CG. 1996. No evidence of renal toxicity from amalgam fillings. Am J Physiol 271(4 Pt 2):R941-945. 126. Bellinger DC, Trachtenberg F, et al. ”Neuropsychological and renal effects of dental amalgam in children: a randomized clinical trial.” JAMA 2006; 295(15): 1775-83. 127. Hodgson S, Nieuwenhuijsen MJ, et al. “Kidney disease mortality and environmental exposure to mercury.” Am J Epidemiol 2007; 165(1): 72-7. 128. Sällsten G, Barregård L, Schutz A. Clearance half-life of mercury in urine after the cessation of long term occupational exposure: influence of a chelating agent (DMPS) on excretion of mercury in urine. Brit J Occupational Environ Med 1994; 51: 337-42. 129. Bluhm RE, Bobbit RG, Welch LW, Wood AJ. Elemental mercury vapour toxicity, treatment and prognosis after acute, intensive exposure in chloralkali plant workers. Part I: History, neuropsychological findings and chelator effects. Human Exp Toxicol 1992; 11: 201-10. 130. Gonzalez-Ramirez D, Zuniga-Charles M, Narro-Juarez A, Molina-Recio Y, Hurlbut KM, Dart RC, Aposhian HV. DMPS (2, 3-dimercaptopropane-1-sulfonate, dimaval) decreases the body burden of mercury in humans exposed to mercurous chloride. J Pharmacol Exp Ther 1998; 287(1):8-12. 131. Garza-Ocanas L, Torres-Alanis O, Pineyro-Lopez A. Urinary mercury in twelve cases of cutaneous mercurous chloride (calomel) exposure: effect of sodium 2, 3-dimercaptopropane-1-sulfonate (DMPS) therapy. J Toxicol Clin Toxicol 1997; 35(6): 653-5. 132. McFee RB, Caraccio TR. Intravenous mercury injection and ingestion: clinical manifestations and management. J Toxicol Clin Toxicol 2001; 39(7): 733-8. 133. McFee RB. and Caraccio TR. “Intravenous mercury injection and ingestion: clinical manifestations and management.” J Toxicol Clin Toxicol 2001; 39(7): 733-8. 134. Winker R., Schaffer AW, et al. „Health consequences of an intravenous injection of metallic mercury.“ Int Arch Occup Environ Health 2002; 75(8): 581-6.

826

37

Organic solvents, silicon-containing compounds and pesticides Patrick C. D’HAESE1, Monique M. ELSEVIERS1, Muhammed YAQOOB2 and Marc E. DE BROE1 1University 2The

of Antwerp, Belgium Royal London Hospital, London, UK

Summary ______________________________________________________________ 827 Introduction ___________________________________________________________ 828 Organic solvents ________________________________________________________ 828 Hydrocarbons: what’s in a name? Exposure to organic solvents Nephrotoxicity of organic solvents

828 828 829

Silicon containing compounds_____________________________________________ 832 Silicon: occurrence, uses and essential chemistry Exposure to silicon containing compounds Nephrotoxicity of silicon containing compounds

832 832 832

Pesticides______________________________________________________________ 836 Conclusion_____________________________________________________________ 837 References _____________________________________________________________ 837

Summary

E

nvironmental/industrial exposure to heavy metals, light hydrocarbons, pesticides and siliconcontaining compounds all have been associated with the development and/or progression of renal failure. Exposure to heavy metals, more particularly lead, cadmium and mercury has been linked with the development of acute or chronic renal failure. The current literature also contains a growing body of information linking solvent exposure with renal injury. To what extent exposure to environmental/occupational

contaminants such as pesticides play either a causal or contributive role in the development of chronic renal failure is less clear. Reported observations suggest either a primary or secondary role of silicon-containing compounds in the development of anti-neutrophil cytoplasmic antibody-positive rapidly progressive glomerulonephritis and Wegener’s granulomatosis. Such observations deserve further confirmation as do studies suggesting a particular sensitivity of the diabetic kidney for the damaging effects of certain occupational exposures.

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

Introduction

Organic solvents

Despite the overwhelming amount of information dealing with the nephrotoxic effects of particular environmental/occupational exposures that have appeared in the literature, to date no data have been reported on the incidence of renal diseases resulting from exposure to particular toxins and chemicals. In view of this it is worth noting that recent data (year 2006) from the Belgian Society of Nephrology indicate that the cause of the disease is not known in up to 11% of cases of endstage renal disease. This allows us to suggest that the impact of exposure to environmental and occupational toxins on the development of renal disease probably is more important than generally assumed. Hence, better information as to the impact of such exposure is of paramount importance because it can lead to both primary and secondary prevention, a rather exceptional privilege in nephrology. With the exception of lead [1], the failure to demonstrate an etiological role for other potential “environmental/occupational nephrotoxins” in the development or progression of renal disease is due, in fact, to the lack of well elaborated clinical or epidemiological studies. In the search for a role for such exposure, the following questions need to be answered: (i) does occupational/environmental exposure to a potential nephrotoxic substance play a direct etiological role in the induction of a particular renal disease, (ii) does the exposure correlate with an increased risk for the progression of renal damage already present in patients with glomerulonephritis, diabetic nephropathy, hypertensive renal disease etc. (iii) do both possibilities have to be considered concomitantly or separately? Some interesting observations have been published recently. While some authors confirm the role of previously identified risk factors others have, based on some experimental evidence, put forward an additional number of potential occupational/environmental nephrotoxins. Also a contributive role for specific occupational exposures, such as organic solvents, on the progression of diabetic nephropathy been suggested. Finally, studies reporting a striking association between exposure to silicon containing compounds and the occurrence of Wegener’s granulomatosis may contribute to a better insight in the pathogenesis of that disease.

Hydrocarbons: what’s in a name?

828

The term “hydrocarbon” refers to any aliphatic, alicyclic, aromatic, halogenated and oxygenated hydrocarbons, glycols and organic solvents. Exposure may occur either via abuse or in during various industrial processes or household activities. Halogenated hydrocarbons (carbon tetrachloride, chloroform) are contained in cleaning agents, insecticides, plastics, degreasers, paint removers, household cleaners. Aromatic hydrocarbons are additives in glues and plastics while the aliphatic compounds occur in fuels. The oxygenated hydrocarbons include alcohols, ketones and ethers and are mostly contained in paint removers, varnishes and glues. Glycols (e.g. ethylene glycol, diethylene glycol, dioxane, glycerol) are used in household and industry. Solvents of abuse are e.g. toluene and xylene. As can be appreciated, solvents possess a wide variety of chemical and physical properties. Because of this diversity there are many different health effects associated with excessive exposure to solvents. While acute renal failure has been documented following exposure to halogenated hydrocarbons [2], glycols [3] and aromatic hydrocarbons, those attributed to light petroleum hydrocarbon exposure are restricted to isolated clinical case reports [4]. More important, but less well proven, is the role of organic solvents in the development or progression of glomerulonephritis or other types of renal diseases.

Exposure to organic solvents Exposure to organic solvents can occur either through inhalation, skin and/or mouth contact. For most solvents, inhalation is considered the most important route of exposure. Once inhaled, the solvent vapors directly irritate the upper respiratory tract (nose, throat and bronchial tubes) and the lungs. Solvent vapors are easily absorbed from the lungs into the bloodstream and are distributed to other parts of the body to produce additional toxic effects. Solvents can also be absorbed through the skin and thus be distributed to various organs. Although not a common route of entry, mouth contact with contaminated hands, food and cigarettes may provide solvents entry into the body and the bloodstream via the digestive system.

37. Organic solvents, silicon-containing compounds and pesticides

The route of entry of any solvent depends, to a certain extent, on the chemical group involved. Thus, alcohols enter the body through inhalation, skin absorption, and ingestion. Aromatic hydrocarbons are readily absorbed through the skin, whilst chlorinated hydrocarbons vaporize, presenting an inhalation hazard. Glycols are water-soluble and glycol ethers and several ketones are absorbed through the skin; an exposure route which can be more serious than inhalation.

Nephrotoxicity of organic solvents Epidemiological studies Sprecace [5] was the first to suggest an association between gasoline exposure and the pulmonary renal presentation of “idiopathic pulmonary hemosiderosis”, more commonly known as Goodpasture’s syndrome. Following this observation several cross-sectional [6, 7-14] and case-control [15-29] studies investigating the relation between renal impairment and occupational hydrocarbon exposure have been published. Cross-sectional studies [6, 7-14] mainly involve the determination of a few up to 23 urinary markers of early tubular or glomerular changes/dysfunction in individuals chronically exposed to organic solvents with various compositions. In these studies renal effects were defined as early subclinical effects. Overt clinical problems have not been reported. In a critical literature review on cross-sectional epidemiological studies of gasoline associated glomerulonephritis Churchill et al. [30] concluded that based upon study design and execution only the study by Ravnskov et al. [18] made a compelling case for a causal association. Furthermore they judged that neither a cohort analytical study nor randomized clinical trial hold a feasible approach to confirm a suspected association. Hence, additional case-control studies are recommended [30]. From 1975 on, several case-control studies [15-29] investigating possible nephrotoxic effects of occupational exposure to solvents have been reported. Although in general the reported findings are highly suggestive of a relation between hydrocarbon exposure and glomerulonephritis, the applied methodology and statistical power have been criticized. These shortcomings are summarized in two excellent reviews by Churchill et al. [30] and Angell [31] who identify four areas of methodological weaknesses: (i) inappropriate control groups, (ii) use of unblinded interviewers, (iii) no

consideration of recall bias and (iv) failure to define a credible measure of the degree and duration of solvent exposure. Moreover most of these studies suffer from small sample size, equivocal case definition and lack of information on important covariates. [32]. In addition, epidemiological studies should consider the magnitude of the observed effect and weigh it against the “biological plausibility”. It must be noted that experimental models are not available which possess the genetic and/or environmental factors that make specific individuals susceptible to solvent nephropathy. Based on epidemiological studies the relation between hydrocarbon exposure and glomerulonephritis seems to be well established by both case-control and cross-sectional studies. However, at present it is unclear which solvents are associated with which type of glomerulonephritis. The studies by Stengel et al. [29] and Porro et al. [25] suggest that the risk is highest for IgA nephropathy and that the possible role of oxygenated solvents in the development of this particular renal disease should be further investigated. Yaqoob et al. [26] found risk factors of 15.5, 5.3, 2.0 for respectively aliphatic, halogenated (greasing/degreasing agents) and aromatic and oxygenated (glue/paints) compounds. Furthermore, they demonstrated a direct correlation between the intensity of hydrocarbon exposure and the appearance of (early) markers of renal dysfunction such as serum creatinine, proteinuria, urinary N-acetyl-β-D-glucosaminidase, leucine aminopeptidase, and γ-glutamyl transferase [6]. An accelerated progression of glomerulonephritis has been reported in patients with intense and continued solvent exposure [33, 34]. A cohort study has investigated the contributive role of solvent exposure in the progression of primary glomerulonephritis [6]. Yaqoob et al. found that progressive renal failure was associated with a greater exposure to organic solvents when compared to individuals presenting with stable or improving renal function. Moreover patients whose occupational solvent exposure continued following the diagnosis of glomerulonephritis, presented with heavy proteinuria and more severe hypertension. In two recent reviews, Ravnskov [35, 36] considered both the hypothesis of a direct casual effect of solvent exposure and the hypothesis that the exposure worsens renal function separately. Results from 14 cross-sectional, 18 case-control studies, 2 cohort studies and 15 experiments on laboratory animals and 2 on humans, 829

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

together with many case reports satisfied all but one (lack of specificity) of Hill’s criteria for both hypotheses prompting the author to conclude that early elimination of the exposure may prevent the progress of renal failure in many patients. Aside from glomerulonephritis, the impact of solvent exposure in other renal diagnoses needs to be explored. Indeed, it is of particular note that all these studies are limited to the former type of renal disease while the role of hydrocarbons in the other renal diagnoses such as diabetic nephropathy should be considered [27, 28]. Interestingly in this context is the recent observation by Nuyts et al. [28] in a group of patients with diabetic nephropathy where hydrocarbon exposure was found in 39% of the patients with that particular type of renal disease. This corroborates the results of Yaqoob et al. [27] who found higher exposure scores to hydrocarbons in patients with incipient (odds ratio 4.0) and overt (odds ratio 5.8) diabetic nephropathy as compared to diabetic individuals without clinical evidence of nephropathy. In contrast to the above, data from a recent large nation-wide case-control study by Fored et al. [32] in which 926 incident cases in a pre-uremic stage (serum creatinine: men >3.4 mg/dl; women >2.8 mg/dl) and 998 control subjects were included the overall risk for chronic renal failure among subjects ever exposed to organic solvents was virtually identical to that among never-exposed individuals (odds ratio 1.01; 95% CI: 0.81 to 1.25). Also there were no dose-response relationships observed for lifetime cumulative solvent exposure, average dose, or exposure frequency or duration. Moreover, the absence of association pertained to all subgroups of chronic renal failure: glomerulonephritis (odds ratio 0.96; 95% CI 0.68 to 1.34), diabetic nephropathy (odds ratio 1.02; 95% CI 0.74 to 1.41), renal vascular disease (odds ratio 1.16; 95% CI 0.76 to 1.75), and other types of chronic renal failure (OR 0.92; 95% CI 0.66 to 1.27). Pathology and mechanism(s) of solvent induced nephrotoxicity Whereas acute renal failure has been documented following exposure to halogenated hydrocarbons [2], glycols [3] and aromatic hydrocarbons, episodes attributed to exposure to light hydrocarbons are restricted to isolated clinical case reports [4]. More important, and less well proven is the role of organic solvents in the 830

development or progression of glomerulonephritis or other types of renal diseases. One of the portals for entry of volatile hydrocarbons is the lung. Lipophilic hydrocarbons rapidly penetrate the lipid membranes thus gaining intracellular access. The link between pulmonary and renal lesions is believed to result from the antigenic similarity shared by the basement membranes of the alveolus and the glomerulus. The immunodominant or epitope is located within the glomerular non-collagenous domain of type IV collagen. It has been proposed that organic solvents or other environmental agents may expose the otherwise cryptic Goodpasture antigen (type IV collagen α 3 chain) to the immune response system in susceptible individuals [37, 38]. The major pathologic presentation of solvent associated nephropathy is that of anti-glomerular basement membrane disease [39] but epimembraneous and subacute proliferative glomerulonephritis have also been demonstrated. In addition, Narvarte et al. [40] reported on a patient with ulcerative colitis in which chronic interstitial nephritis developed that later was attributed to long-term exposure to organic solvents. The histological severity of tubulointerstitial damage in primary glomerular disorders appears to correlate with severity of renal impairment and can predict the future outcome of renal disease [41]. Recent data correlating solvent exposure with morphological parameters of tubulointerstitial damage in 59 patients with biopsy-proven primary glomerulonephritis showed that solvent exposure correlated significantly with relative interstitial volume and serum creatinine. Solvent exposure, relative interstitial volume, degree of interstitial fibrosis and magnitude of chronic inflammatory cellular infiltrate in the renal cortex at the time of renal biopsy were higher in these glomerulonephritic patients developing progressive renal failure as compared to those presenting a stable or improving renal function [6]. Since, in solvent associated nephropathy, the renal injury is insidious its accurate detection/diagnosis remains an intriguing challenge. Indeed, to be of clinical value methods of detection must be sensitive, quantitative, and correlate with the usual parameters of renal impairment. Measurements of enzymuria, proteinuria and specific tubular antigens have all been proposed. However, until now there is no consensus on their diagnostic sensitivity, specificity and

37. Organic solvents, silicon-containing compounds and pesticides

predictive value [12, 42-45]. At present albuminuria, compatible with altered membrane permeability [45], turns out to be the most consistent renal abnormality in solvent-associated nephropathy. Indeed, in a recent analysis of the available literature evaluating relevant cross-sectional studies were evaluated and a series of markers were analyzed with respect to their suitability as biomarker for renal damage, an increased albumin excretion was observed more frequently in groups of workers exposed to various solvents (like toluene, styrene, aliphatic/aromatic hydrocarbon mixtures, tetrachloroethene, mixtures of chlorinated hydrocarbons) as compared to controls whilst no clear pattern emerged for the other markers [46]. The issue of coexisting solvent-associated tubular damage is more controversial. While a urinary increase in tubular derived enzymes has been reported by some authors [12, 43, 44], others have failed to detect any change using either β2-microglobulin or retinol binding protein excretion [13, 43, 45]. The mechanism underlying solvent-induced glomerulopathy remains speculative. Possible pathways have been proposed by Roy et al. [4] (Figure 1). Conceptually it is proposed that when a genetically sensitive individual is exposed to environmental hydrocarbons, any or all three of the pathways could induce a hypersensitive reaction leading to glomerulonephritis. Glomerulonephritis appears to be mainly an immune-mediated disease and some solvents are found to act as immunosuppressants [47-48]. Experimentally, solvent exposure results in glomerular and tubulointerstitial injury [50] which can be explained since membranous glomerulonephritis can be induced by administration of proximal tubular brush border antigens [51], thus suggesting that solvent exposure may induce a low-grade tubular injury. This tubular injury could provoke local autoimmunity by releasing tubular or basement membrane antigens (antibodies to proximal tubular changes, laminin, Goodpasture’s antigens) with activation or damage of the underlying endothelium resulting in the induction of glomerulonephritis. Alternatively potential glomerulotoxic immune factors may arise independently of solvent exposure. Also, the immunosuppressant action of solvents may facilitate the deposition of these mediators of immune damage in renal tissue.

Experimental studies Several animal models have been used for studying the nephrotoxic effects of solvent exposure. Using rats exposed to petroleum vapors Klavis and Drommer [52] demonstrated renal lesions similar to those noted in Goodpasture’s syndrome. In another study 60% of rats fed N,N’-diacetylbenzidine [53] had an increased blood urea nitrogen level. The N,N’-diacetylbenzidine induced glomerulonephritis was characterized by rapid crescent formation, fragmentation of the capsular basement membrane and early obliterative glomerulosclerosis. The site of action of N,N’-diacetylbenzidine appeared to be localized at Bowman’s capsule and was not dependent on either deposition of fibrin or coagulative mechanisms [54]. Zimmerman and Norbach [55] demonstrated mesangial proliferative glomerulonephritis with focal glomerulosclerosis after long-term administration of carbon tetrachloride to rats. Although the pathogenesis of the glomerular lesion was not clear, glomerular deposits of antigenantibody complexes were not observed. In addition to the glomerular lesions the same workers also noted tubulointerstitial damage in a similar experiment [55]. Exposing LLC-PK1 cells to either toluene or p-xylene resulted in reduced cell viability and increased DNA fragmentations which might indicate that long-term Hydrocarbon exposure (environmental)

genetically predisposed individual

Direct tubular toxicity

Immunosuppression

Altered membrane permeability

? Type II and III hypersensitivity reactions

Glomerulonephritis

Figure 1. Possible mechanism of hydrocarbon associated glomerulonephritis (adapted with permission from Roy et al. [4]). See also Chapter 7 by Pelletier et al. 831

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

exposure to organic solvents may be associated with proximal tubule cell apoptosis [56].

Silicon containing compounds Silicon: occurrence, uses and essential chemistry Silicon (Si) is the second most abundant element in the earth’s crust, contributing around 28%. Silicon acts as a nonmetal in its chemical behavior but its electrical and physical properties are those of a semimetal. Crystalline silicon is a gray, lustrous solid. The chemistry of silicon is dominated by compounds that contain the silicon-oxygen (Si-O) linkage. The element is used in ceramic industries and for the fabrication of semiconductors. Silicon-based polymers (silicones: polymeric chains containing alternately linked silicon and oxygen atoms) have wide application in industry as well as for clinical and pharmaceutical purposes. In the literature the nomenclature used to describe the various silicon containing compounds is confusing. In nature silicon does not occur as the free element; rather it is either found as silicon dioxide (SiO2), the so-called silica, in an enormous variety of silicates or in its carbide form i.e. carborundum (SiC). The soil water or the so-called ‘soil solution’ [57] contains silicon as silicic acid (H4SiO4). In the form of silicic acid silicon is readily absorbed by plants and all soil grown plants contain it as an appreciable but variable fraction of the dry matter [57]. Particularly the hulls of grains and the macrohairs of a number of grasses may contain high concentrations of the element (up to 10% of the plant’s dry weight).

there are multiple industries in which exposure may occur. However, in modern, industrialized societies, due to the extensive enforcement of occupational health standards, exposure from well recognized sources such as mining and quarrying activities, sandblasting, stonecutting, ceramics, glass, abrasives etc. … is well controlled, while other sources such as cosmetics, electrical and electronic machinery, grain dust and cotton or wool textiles are less well recognized.

Nephrotoxicity of silicon containing compounds

Epidemiology Silicon toxicity is virtually limited to occupational exposure to silicon compounds e.g. miners, sandblasters, bricklayers, pottery workers in which inhalation of the compounds has been associated with the diseases of the lung. The later being evidenced by nodule formation and acute silicosis, mixed dust fibrosis and diatomite pneumoconiosis. Knowledge regarding renal injury and the development of anti-neutrophil cytoplasmic antibody (ANCA) positivity associated with silica exposure is rudimentary being limited to epidemiological observations. Moreover, information about the health significance of the occupational exposure to other silicon containing compounds apart from silica and crystalline silicates is lacking. During the past decade a number of case reports have described the occurrence of different forms of renal disease in patients exposed to silica [58-66]. However, only a few reports concerned subjects exposed to silica but without silicosis. Most of the cases demonstrated renal lesions compatible with rapidly progressive glomerulonephritis with a necrotizing component present in most cases. Crescent formation Exposure to silicon containing compounds was described in a patient with proliferative glomerulonephritis [59] and three individuals with IgA neDue to the element’s abundance in nature, human phropathy [60]. beings are exposed to relatively large but variable Renal lesions observed after silica exposure have amounts of silicon through food, drinking water and been associated with ANCA positivity suggesting a dust. In the human body, however, the element is only pathogenetic role of ANCA [64-67]. Other autoimmune present in trace amounts. The prolonged inhalation of crystalline silica dust is associated with silicosis. Amor- manifestations have been reported in a cohort of 50 workers after occupational exposure to a scouring powphous silica is considered much less pathogenic than der mainly containing silica [68]. Systemic symptoms crystalline forms. In occupational settings, the main concern regarding exposure to silicon containing com- were present in 32 of these subjects including Sjögren’s syndrome (n=6), systemic lupus erythematosus (n=3), pounds with inhalation of silica and silicate-containing mineral dust. Since silica is such an abundant mineral, “overlap” syndrome (n=5) and with undifferentiated 832

37. Organic solvents, silicon-containing compounds and pesticides

findings (n=13) not meeting the criteria for a defined disease (Table 1). In most patients renal disease occurred after a long latency period. In the reports where the information is available, renal symptoms occurred 3 to 27 years after silica exposure. Data of a retrospective cohort study including 2412 white male gold miners which had been working underground for at least 1 year between 1940 and 1965 showed an elevated relative risk for non-systemic end-stage renal disease (i.e. glomerulonephritis or interstitial nephritis) of 4.22 (95% CI: 1.54-9.19) increasing to 7.70 (95% CI: 1.59-22.48) among workers with 10 or more years of employment [69] when compared to the incidence of end-stage renal disease in the US population. These data could be confirmed in a more recent cohort study in which 2670 men employed before 1980 for 3 years or more in North-American sand-producing plants. Al-

though evaluation of death from renal disease was not the primary objective of their study the total number of deaths from nephritis or nephrosis was 16 against 7.6 (SMR 212, P = 0.002) from state/provincial rates, with the excess present only in workers employed for 10 years or more [70]. Increased levels of early markers of renal dysfunction have been demonstrated even in currently exposed workers [71, 72]. The cross-sectional observations in workers exposed to silica confirmed signs of renal impairment in patients with silicosis [71, 73] as well as in workers exposed to silica dust for less than 2 years and without lung injury [74]. In a cross-sectional study by Boujemaa et al. [72], who evaluated early indicators of renal dysfunction in silicotic workers (n=116), recorded a delay after cessation of exposure up to 30 years (mean 23 years). The

Table 1. Observations in silica exposed workers. Cross sectional studies Reference Exposed workers

Non-exposed workers

Lung

Ng et al. 1992 33 drillers/crushers in [71] granite quarries current exposure Boujemaa et 116 underground miners al. 1994 [72] past exposure

19 age-matched nonexposed workers

Silicosis (7)

61 age-matched general population

Silicosis

Hotz et al. 1995 [74]

86 age-matched nonexposed workers

No silicosis

Non-exposed workers / /

Disease Malignant neoplasms Respiratory cancer Renal disease

Standardized mortality ratio 108.3 112.1 /

US population

Renal disease

161.0

Non-exposed workers

Disease

/

Systemic illness (32) - Sjörgen (6) - Systemic sclerosis (5) - Overlap syndrome (5) - Systemic lupus erythematosis (3) - Undifferentiated findings (13)

US population

Non-systemic end-stage renal disease - Glomerulonephritis or interstitial nephritis Standard incidence ratio 4.22 (1.54-9.19)

86 workers in quartzite rock quarry current exposure

Mortality studies Reference Exposed workers Marsh et al. 16661 man-made mineral 1985 [75] fiber workers Goldsmith Man-made mineral fiber 1993 [76] workers Steenland et 4626 workers in sand al. 2001 [77] industry Cohort studies Reference Exposed workers Prospective Sanchez50 scouring powder Roman 1993 producing factory [68]

Retrospective 2412 silica exposed gold Calvert et al. miners 1997 [69]

Early markers of renal dysfunction: increased compared to controls Albumin α-1-microglobulin ß-N-acetyl-glucosaminidase Albumin Retinol-binding protein ß-N-acetyl-glucosaminidase Albumin Transferrin Retinol-binding protein ß-N-acetyl-glucosaminidase

(continued on next page) 833

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

Table 1 (continued) Case control studies Reference Cases Steenland et 325 end-stage renal failure al. 1990 [78] patients Gregorini et 16 patients with ANCAal. 1993 [79] positive rapidly progressive glomerulonephritis Nuyts et al. 272 renal failure patients 1995 [28]

Controls 325 age-matched general population 32 age-matched other renal failure patients

Occupational exposure Silica Brick and foundry Silica dust

OR (95% CI) 1.67 (1.02-2.74) 1.92 (1.06-3.46) 14.0 (1.7-113.8)

272 age-matched general population

2.51 (1.37-4.60) 2.96 (1.24-7.04)

Nuyts et al. 1995 [80]

16 patients with Wegener’s granulomatosis

32 age-matched general population

Duna et al. 1998 [84]

101 patients with Wegener’s granulomatosis

54 ‘Healthy’ gendermatched patients from medical clinics

Silicon containing compounds Grain dust Silica Silicon containing compounds Construction & farm workers

5.0 (1.4-11.6) 6.5 (1.3-13.5)

31 cases of biopsy proven vasculitis* 65 ANCA-SVV patients with pauci-immune necrotizing GN Flores-Suarez 76 ANCA-positive primary et al. 2003 systemic vasculitis [86] Lane et al., 75 patienst with primary 2003 [83] systemic vasculitis Beaudreuil et 60 ANCA-positive patients al., 2005 [67]

58 age/sex residencematched controls 65 other renal patients matched for age, gender and race 159 healthy, age-matched patients

Silica

OR not reported Construction 31% cases versus 19% controls Farm 36% versus 22% controls 2.4 (p=0.04)

Silica dust

4.4 (1.4-14.4)

Dusty areas Silica

3.1 (1.5-6.8) 3.2 (1.1-9.2)

220 age-matched hospital patients 120 age and gendermatched hospital patients

31 ANCA-positive vasculitis with renal and lung involvement 129 ANCA-SVV with renal biopsy proven glomerular involvement

30 age, gender and residence-matched office workers 109 healthy age, gender, state-matched controls

Silica Agricultural silica High Silica Medium silica Low silica Silica & asbestos

3.0 (1.0-8.4) 4.4 (1.1-18.1) 6.9 (1.3-35.1) 2.3 (0.6-8.2) 0.8 (0.1-3.9) 22% (13% silica and 9% asbestos) versus 0% in controls (P < 0.05) 1.9 (1.0-3.5) 2.5 (1.1-5.4)

Stratta et al. 2001 [81] Hogan et al. 2001 [82]

Rihova et al., 2005 [87] Hogan et al., 2007 [85]

Silica Crop harvesting

OR = Odds ratio; ANCA = anti-neutrophil cytoplasmic antibody ; CI = Confidence interval; SVV = small vessel vasculitis; GN = glomerulonephritis *18 pauci-immune crescentic glomerulonephritis, 9 microscopic polyangitis, 4 Wegener’s granulomatosis (Adapted with permission from De Broe et al. [117])

silicotic subjects excreted, on average, slightly higher amounts of albumin, retinol binding protein and Nacetyl-β-D-glucosaminidase [71, 72, 74]. A survey of the literature [28, 58-65, 71, 72, 74-80] indicates that the most frequent exposure to silicon involves exposure to silica and silicates mainly in their crystalline forms. Health risks associated with the exposure to other silicon containing compounds were reported in the mortality study of 16.661 manmade mineral fiber workers employed during 1945 to 1963 at one of 17 U.S. manufacturing plants [75]. Fiber exposure in the plants producing fibrous glass or mineral wool, or both, was associated with increased 834

standardized mortality ratios for overall mortality as well as for mortality from nephritis and nephrosis. Further evidence of the nephrotoxic role of these and other kinds of silicon containing compounds was reported by Goldsmith and Goldsmith [76]. They argued that in California an increased mortality from diseases of the urinary system was observed for farmers and farm workers. More recently, Steenland et al. [77] examined renal disease morbidity and mortality as well as arthritis mortality in a cohort of 4.626 silica-exposed workers in the industrial sand industry (an industry previously unstudied). Comparison of the cohort with the US population revealed an excess mortality ratio

37. Organic solvents, silicon-containing compounds and pesticides

from chronic renal disease of 1.61 [95% CI = 1.13-2.22]. Linking of the cohort with the US registry of end-stage renal disease for the years 1977-1996 demonstrated an excess of end-stage renal disease incidence (standardized incidence ratio: 1.97, 95% CI: 1.25-2.96), which was highest for glomerulonephritis (3.85, 95% CI: 1.55-7.93) and increased with increasing cumulative exposure. The most firmly based epidemiological observations are derived from recently published case-control studies [78-80]. Two studies, based on a large sample size, retrospectively examined occupational exposures of renal failure patients. Amongst others, an increased odds ratio’s for silicon-containing compounds was also observed [28, 78]. Nuyts et al. [28] were the first to demonstrate an increased risk for the exposure to grain dust that potentially may contain high amounts of silicon, an observation that later on has been confirmed by others [80-82]. Other studies only [79, 80] focused on a small sample of patients with rapidly progressive glomerulonephritis and the specific exposure to silicon containing compounds. Gregorini et al. [79] selected only ANCA positive patients and Nuyts et al. [80] investigated patients with Wegener granulomatosis, 80% of who were ANCA positive. Studying a group of 31 cases of biopsy proven vasculitis (18 pauci-immune crescentic glomerulonephritis, 9 microscopic polyangitis, 4 Wegener granulomatosis) Stratta et al. [81] also found an increased odds ratio (2.4) for exposure to silica whilst no other significant association with a series of other exposures could be found. Hogan et al. [82] studying 65 patients with ANCA-associated smallvessel vasculitis (all of them having biopsy-proven pauci-immune crescentic glomerulonephritis) also demonstrated the odd’s ratio of silica dust exposure in the development of the disease to be 4.4 times greater as compared to control subjects. In contrast to an increased risk for the development of ANCA-associated small-vessel vasculitis, exposure to silica could not be associated with systemic lupus erythematosus [82]. In a more recent studies these and other groups confirmed the association between silica exposure and onset of biopsy-proven glomerulonephritis resulting from ANCA-associated small vessel vasculitis (Table 1) [67, 83-87]. Pathology and mechanism(s) of silicon-induced nephrotoxicity Data presented above are highly indicative for

an association between silica and renal disease. The underlying pathophysiological mechanisms, however, are far from clear. At least two mechanisms have been proposed. A direct nephrotoxic effect of silicon has been suggested by Hauglustaine et al. [88]. Recently, Hotz et al. [74] reported on subclinical renal effects as indicated by an increased excretion of albumin, transferrin, retinol binding protein and N-acetyl-βD-glucosaminidase following short time (less then 2 years) exposure to silica in non-silicotic workers. In a recent review on the association between renal disease and silica exposure Kallenberg [89] suggested that the tubular dysfunction observed in silica workers resulted from a direct nephrotoxic effect of the silicon compound. Experimentally, the nephrotoxic potential of silica has been demonstrated in the dog [90]. The exact mechanism responsible for the nephrotoxic effect of silicon remains to be elucidated although membrane damage possibly related to oxidant generation [89] or inhibition of superoxide dismutase activity [91] might be rational explanations. Based on reports on lung toxicity related to the chemical, morphological and surface characteristics of the various silicon compounds, it is not known yet to which extent these exhibit direct toxic effects at the level of the kidney [90, 92]. A second possible mechanism consists in the interaction of the inhaled silicon compounds with the cell membrane particularly that of macrophages. Once engulfed a series of events may ensue resulting in an important inflammatory reaction at the alveolar level. In addition silica particles have been shown to induce rupture of phagosomes of macrophages [93] with the release of lysosomal enzymes such as proteinase 3 or myeloperoxidase the antigens of ANCA into the microenvironment that in turn may be followed by the genration of the autoantibodies. Recently, increasing interest has been raised in the role of apoptosis in the induction of autoimmunity [94]. There is growing evidence that apoptotic antigens are the natural targets for many autoantibodies [95]. The possibility that silica, in vitro, may induce apoptosis of monocytes or macrophages and possibly neutrophils may represent an alternative mechanism that is operative in the induction of ANCA-associated vasculitis [95, 96]. Surface expression of ANCA antigens proteinase 3 and myeloperoxidase have been demonstrated during apoptosis of neutrophils [95, 97]. Therefore, ANCA’s 835

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

may bind to their target antigen on apoptotic cells and via an Fc-dependent bridging, the antibodies may amplify the release of cytokines, oxygen radicals, and lysosomal enzymes. To which extent the generated ANCA’s are responsible for initiating vasculitis, or may increase or even perpetuate vasculitis remains to be determined. Since ANCA (i) may directly activate neutrophils in vitro, (ii) may damage endothelial cells expressing the proteinase-3 antigen, (iii) are capable of inducing in vitro adherence of neutrophils to endothelial cells, (iv) block (c-ANCA) the inactivation of proteinase-3 by α-1 antitrypsin, a pathophysiological role may be suggested. Based on experimental studies it has been suggested that silicates may stimulate lymphocytes via a T-cell receptor Vβ-specific T-cell activation pathway resulting in the production of autoantibodies or autoimmune diseases [94, 98, 99]. In this context it must be noted that not only ANCA’s but also other autoantibodies such as antinuclear antibodies and rheumatoid factors frequently occur in workers heavily exposed to silicon-containing compounds [68, 77]. An intriguing observation made from case-control studies remains the controversy that exists between silica exposure and the development of a particular renal disease. Indeed, in a recent case-control study on occupational risk factors for chronic renal failure, Nuyts et al. [28] demonstrated exposure to silicon containing compounds to be related to the development of virtually all diagnostic groups of chronic renal failure. On the other hand, in two other studies [79, 80] silicon exposure was linked to a significantly higher relative risk for the development of ANCA-associated rapidly progressive glomerulonephritis or Wegener disease. These observations might indicate that silicon-containing compounds may act as a contributive as well as a causative factor in the development of renal disease. A similar observation has also been made in subjects taking analgesics. Here, besides the development of the so-called analgesic nephropathy identifiable with high accuracy by the visualization of renal papillary necrosis [100], analgesic abuse also seems to accelerate the development and evolution of the other types of renal diseases [101]. In a recent study it was demonstrated that acetaminophen and aspirin exhibit exacerbating effects on the development of all types of chronic renal failure [102]. 836

Pesticides The information linking environmental/occupational exposure to pesticides (including herbicides/ fungicides/insecticides) is confined to some case reports and an occasional retrospective review on occupational exposure and acute renal failure [103]. Serious exposure to pesticides is usually accidental although suicidal ingestion’s have occurred [103, 104]. Since many of these compounds have both commercially and domestic application exposures usually occur when proper protective precautions are ignored. Usually the acute renal failure following pesticide poisoning turns out to be multifactorial. For example, poisoning by the now banned pesticide Lindane® caused both acute volume depletion [105] and rhabdomyolysis [106], either of which could account for the subsequent acute renal failure. Other examples of multifactorial causes of acute renal failure due to pesticide exposure are reviewed by Abuelo [103]. Due to the increased litigation based on premise of product liability a renewed interest in the renal effects of herbicides, fungicides, pesticides, and insecticides has been noted during the last years. However, because of the lack of a valuable experimental animal model the current knowledge of the pathophysiologic mechanisms of the pesticide-induced renal injury is highly limited. The possibility that these pesticides act similar to that of light hydrocarbons is worthwhile to be considered, however, at the present is at is still highly speculative. Little is also known of the long term renal effects of chronic low dose exposure to pesticides. Chronic exposure to the now banned dichlorodiphenyltrichloroethane (DDT), a lipophilic compound with prolonged body fat retention, has been associated with renal injury [105]. Insights in the renal handling of 2,4-dichlorophenoxyacetic acid have contributed to a better knowledge of the extent of occupational exposure to this widely used herbicide [107-111]. Recently, Kancir et al. [112] reported on a case of oliguric acute renal failure complicated by profound and recurrent hypocalcemia, severe hyperphosphatemia, and inappropriately high urinary sodium concentrations following exposure to this compound. In the studies of Manninen et al. [111] the peak herbicide concentration which was noted during the first 12 hours post exposure turned out

37. Organic solvents, silicon-containing compounds and pesticides

to be associated with an increased excretion of both sodium and potassium. In in-vitro experiments the uptake of either 2,4-dichlorophenoxyacetic acid or 2,4,5-trichlorophenoxyacetic acid via a proximal tubule organic acid transport system was demonstrated in both rat and rabbit renal cortical slices [113]. Based on these experiments it was suggested that once 2,4dichlorophenoxyacetic acid is secreted into the proximal tubule, it is probably non-reabsorbable an acts to bind intraluminal sodium and potassium. This, in turn, induces electrolyte depletion which could cause the rhabdomyolysis and severe hypocalcemia and hyperphosphatemia observed by Kancir et al. in the above mentioned study [112]. Lindane® [106], diquat® [104], copper sulphate [114] and paraphenylene diamine [115, see also Chapter 40] all have been reported to induce rhabdomyolysis and acute renal failure. Recently, Talbot et al. [116] reported the poisoning of 93 patients with the glyphosphate-surfactant herbicide (Round-up®). In ten patients (14%) manifest renal abnormalities were noted which was accompanied by a nearly uniform increase in serum creatinine (>180 μM/L) and oliguria in 3 patients. Based on their own investigations and those of Japanese workers, the

authors concluded [116] that in 50% of the cases in which exposure to glyphosphate-surfactant herbicide was reported renal failure resulted.

Conclusion Recent literature clearly points to a role for exposure to solvents in the development or progression, or both, of chronic renal failure. With regard to longterm exposure to pesticides no clear-cut evidence for a linkage with renal disease has been presented to date. Furthermore, a number of observations of the past two years supports the prveviously suggested primary or secondary role of substances such as silicon-containing compounds in the development of ANCA-associated rapidly progressive glomerulonephritis or Wegener’s granulomatosis as well as an increased susceptibility of the diabetic kidney to the toxic effects of particular occupational pollutants. Further experimental and clinical studies are required to gain insight in the underlying mechanisms by which the environmental/occupational contaminants exert their toxic action at the level of the kidney.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Croll DG. The incidence of chronic nephritis amongst young people in Queensland. Med J Aust 1929; 2: 144-145. Pederen LM. Biological studies in human exposure to and poisoning with organic solvents. Pharmacol Toxicol 1987; 3: 1-38. Porter GA. The treatment of ethylene glycol poisoning simplified. N Eng J Med 1988; 319(2): 109-110. Roy AT, Brautbar N, Lee DBN. Hydrocarbons and renal failure. Nephron 1991; 58: 385-392. Sprecace GA. Idiopathic pulmonary hemosiderosis. Am Rev Resp Dis 1963; 8: 330-337. Yaqoob M, Bell GM, Stevenson A, Mason H, Percy DF. Renal impairment with hydrocarbon exposure. Quart J Med 1993; 86: 165174. Krusell L, Nielsen HK, Baelum J, Lundqvist G, Omland O, Vaeth M, Husted SE, Mogensen CE, Geday E. Renal effects of chronic exposure to organic solvents: a clinically controlled trial. Acta Med Scand 1985; 218: 323-327. Enarson DA, Chan-Yeung M, Embree V, Wang R, Schulzer M. Occupational exposure to chlorophenates. Scand J Work Environ Health 1986; 12: 144-148. Ng TP, Ong SG, Lam WK, Jones MG, Cheung CK, Ond CN. Urinary levels of proteins and metabolites in workers exposed to toluene. Int Arch Occup Environ Health 1990; 62: 43-46. Hotz P, Pilliod J, Bernard A, Berode M, Rey F, Mazzocato C, Guillemin M, Boillat MA. Hydrocarbon exposure, hypertension and kidney function tests. Int Arch Occup Environ Health 1990; 62: 501-508. Lauwerys R, Herbrand J, Buchet JP, Bernard A, Gaussin J. Health surveillance of workers exposed to tetrachloroethylene in drycleaning shops. Int Arch Occup Environ Health 1983; 52: 69-77. Mutti A, Lucertini S, Falzoi M, Cavatorta A, Franchini I. Organic solvents and chronic glomerulonephritis: a cross-sectional study with negative findings for aliphatic and alicyclic C5-C7 hydrocarbons. J Appl Toxicol 1981; 1(4): 224-226. Viau C, Bernard A, Lauwerys R, Buchet JP, Quaeghebeur L, Cornu L, Cornu M, Phillips SC, Mutti A, Lucertini S, Francini I. A crosssectional survey of kidney function in refinery employees. Am J Ind Med 1987; 11: 177-187.

837

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.

33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

838

Mutti A, Alinovi R, Bergamaschi E, Giagini C, Cavazinni S, Franchini I, Lauwerys RR, Bernard AM, Roels H, Gelpi E, Rosello J, Ramis I, Price RG, Taylor SA, De Broe ME, Nuyts GD, Stolte H, Fels LM, Herbort C. Nephropathies and exposure to perchloroethylene in dry cleaners. Lancet 1992; 340: 189-193. Zimmerman SW, Groehler K, Beirne GJ. Hydrocarbon exposure and chronic glomerulonephritis. Lancet 1975; 2: 199-201. Lagrue G, Kamalodine T, Hirbec G et al. Rôle de l’inhalation de substances toxiques dans la genèse des glomérulonéphritis. Nouv Press Med 1977; 6: 3609-3613. Ravnskov U. Exposure to organic solvents - a missing link in poststreptococcal glomerulonephritis ? Acta Med Scand 1978; 203: 351-356. Ravnskov U, Forsberg B, Skerfving S. Glomerulonephritis and exposure to organic solvents: a case control study. Acta Med Scan 1979; 205: 575-579. Finn R, Fennerty AG, Ahmad R. Hydrocarbon exposure and glomerulonephritis. Clin Nephrol 1980; 14(4): 173-175. Van der Laan G. Chronic glomerulonephritis and organic solvents: a case control study. Int Arch Occup Environ Health 1980; 47: 1-8. Ravnskov U, Lundström S, Nordén A. Hydrocarbon exposure and glomerulonephritis evidence from patients’ occupations. Lancet 1983; 2: 1214-1216. Bell GM, Gordon ACH, Lee P, Doig A, MacDonald MK, Thomson D, Anderton JL, Robson JS. Proliferative glomerulonephritis and exposure to organic solvents. Nephron 1985; 40: 161-165. Harrison DJ, Thomson D, MacDonald MK. Membranous glomerulonephritis. J Clin Pathol 1986; 39: 167-171. Harrington JM, Whitby H, Gray CN, Reid FJ, Aw TC, Waterhouse JA. Renal disease and occupational exposure to organic solvents: a case referent approach. Brit J Ind Med 1989; 46: 643-650. Porro A, Lomonte C, Coratelli P, Passavanti G, Maria Ferri G, Assennato G. Chronic glomerulonephritis and exposure to solvents: a case referent study. Brit J Ind Med 1992; 49: 738-742. Yaqoob M, Bell GM, Percy D, Finn R. Primary glomerulonephritis and hydrocarbon exposure: a case control study and literature review. Quart J Med 1992; 301: 409-418. Yaqoob M, Patrick AW, McClelland P, Stevenson A, Mason H, Percy DF, White MC, Bell GM. Occupational hydrocarbon exposure in diabetic nephropathy. Diabetic Med 1994; 11(8): 789-793. Nuyts GD, Van Vlem E, Thys J, De Leersnijder D, D’Haese PC, Elseviers MM, De Broe ME. New occupational risk factors for chronic renal failure. Lancet 1995; 346: 7-11. Stengel B, Cénée S, Limasset JC, Protois JC, Marcelli A, Brochard P, Hémon D. Organic solvent exposure may increase the risk of glomerular nephropathies with chronic renal failure. Int J Epidem 1995; 24: 427-434. Churchill DN, Fine A, Gault MH. Association between hydrocarbon exposure and glomerulonephritis. An appraisal of the evidence. Nephron 1983; 33: 169-172. Angell M. The interpretation of epidemiologic studies. N Engl J Med 1990; 323: 823-825. Fored CM, Nise G, Ejerblad E, Fryzek JP, Lindblad P, McLaughlin JK, Elinder C-G, Nyrén O. Absence of Association between Organic Solvent Exposure and Risk of Chronic Renal Failure: A Nationwide Population-Based Case-Control Study. J Am Soc Nephrol 2004; 15:180-186. Bell GM, Doig D, Thompson D, Anderton JL, Robson JS. End-stage renal disease associated with occupational exposure into organic solvents. Proc EDTA-ERA 1985; 22: 725-729. Ravnskov U. Influence of hydrocarbon exposure on the course of glomerulonephritis. Nephron 1986; 42: 156-160. Ravnskov U. Hydrocarbon exposure may cause glomerulonephritis and worsen renal function: evidence based on Hill’s criteria for causality. Q J Med 2000; 93: 551-556. Ravnskov U. Hydrocarbons may worsen renal function in glomerulonephritis: A meta-analysis of the case-control studies. Am J Ind Med 2000; 37: 599-506. Salant DJ. Immunopathogenesis of crescentic glomerulonephritis and lung purpura. Kidney Int 1987; 32: 408-425. Kalluri R, Gattone VH 2nd, Noelken ME, Hudson BG. The α 3 chain of type IV collagen induces autoimmune Goodpasture syndrome. Proc Natl Acad Sci USA 1994; 91(13): 6201-6205. Daniell WE, Couser WG, Rosenstock L. Occupational solvent exposure and glomerulonephritis. A case report and review of the literature. JAMA 1988; 259: 2280-2283. Narvarte J, Sabe SR, Ramirez G. Occupational exposure to organic solvents causing chronic tubulointerstitial nephritis. Arch Intern Med 1989; 149: 154-158. Cameron JS. Tubular and interstitial factors in the progression of glomerulonephritis. Pediatr Nephrol 1992; 6: 292-303. Meyer BR, Fischbein A, Rosenman K, Lerman Y, Drayer DE, Reidenberg MM. Increased urinary enzyme excretion in workers exposed to nephrotoxic chemicals. Am J Med 1984; 76: 989-998.

37. Organic solvents, silicon-containing compounds and pesticides

43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72.

Askergen A. Studies in kidney function in subjects exposed to organic solvents. III. Excretion of cells in the urine. Acta Med Scand 1981; 210: 103-108. Francini I, Cavatorta A, Folzoi M Lucertini S, Mutti A. Early indicators of renal damage in workers exposed to organic solvents. Int Arch Occup Environ Health 1983; 52: 1-9. Lauwerys R, Bernard A, Viau C, Buchet JP. Kidney disorders and hemotoxicity from organic solvent exposure. Scan J Work Environ Health 1985; II (suppl 1): 84-90. Voss JU, Roller M, Brinkmann E, Mangelsdorf I. Nephrotoxicity of organic solvents: biomarkers for early detection. Int Arch Occup Environm Health 2005; 78: 475-485. Ravnskov U. Possible mechanism of hydrocarbon-associated glomerulonephritis. Clin Nephrol 1985; 23: 294-298. Bekesi JG, Holland JF, Anderson HA, Fischbein AS, Rom W, Wolff MS, Selikoff IJ. Lymphocyte function of Michigan dairy farmers exposed to polybrominated biphenyls. Science 1978; 199: 1207-1209. Peters DK, Lachmann PJ. Immunity deficiency in the pathogenesis of glomerulonephritis. Lancet 1974; 1: 58-61. Shibita S, Yokoyama M. Nephritogenic glycoproteins. Nephron 1990; 55: 152-158. Glassock RJ, Lachmann PJ, Watson JI, Dixon FJ. Autologous immune complex nephritis induced by renal tubular antigens. J Exp Med 1986; 127: 573-587. Klavis G, Drommer W. Goodpasture’s syndrome and the effects of benzene. Arch Toxicol 1970; 26: 40-55. Harman JW, Miller EC, Miller JA. Chronic glomerulonephritis and nephrotic syndrome induced in rats by N, N’-diacetylbenzidine. Am J Pathol 1952; 28: 529. Harman JW. Chronic glomerulonephritis and nephrotic syndrome induced in rats with N, N’-diacetylbenzidine. J Pathol 1970; 104: 119-128. Zimmerman SW, Norbach DH. Nephrotoxic effects of long term carbon tetrachloride administration in rats. Arch Pathol Lab Med 1980; 104: 94-99. Al-Ghamdi SS, Raftery MJ, YAqoob MM. Toluene and p-xylene induced LLC-PK1 apoptosis. Drug Chem Toxicol 2004; 27(4): 425432. Epstein E. The anomaly of silicon in plant biology. Proc Natl Acad Sci 1994; 91: 11-17. Slavin RE, Swedo JL, Brandes D, Gonzalez-Vitale JC, Osornio-Vargas A. Extrapulmonary silicosis: a clinical, morphologic, and ultrastructural study. Hum Pathol 1985; 16: 393-412. Osorio AM, Thun MJ, Novak RF, Van Cura EJ, Avner ED. Silica and glomerulonephritis: case report and review of the literature. Am J Kidney Dis 1987; 9(3): 224-230. Bonnin A, Mousson C, Justrabo E, Tanter Y, Chalopin JM, Rifle G. Silicosis associated with crescentic IgA mesangial nephropathy. Nephron 1987; 47: 229-230. Sherson D, Jorgensen F. Rapidly progressive crescenteric glomerulonephritis in a sandblaster with silicosis. Brit J Ind Med 1989; 46: 675-676. Arnalich F, Lahoz C, Picazo ML, Monerero A, Arribas JR, Martinez Ara J, Vazques JJ. Polyarteritis nodosa and necrotizing glomerulonephritis assciated with long-standing silicosis. Nephron 1989; 51: 544-547. Dracon M, Noel C, Wallaert B, Dequiedt P, Lelievre G, Tacquet A. Glomérulonéphrites rapidement progressives chez les mineurs de charbon pneumoconiotiques. Néphrologie 1990; 11: 61-65. Gregorini G, Tira P, Lancini L, Bozzola G, Maiorca R. Silica exposure as a risk factor in the development of necrotising and crescentic glomerulonephritis with anti-mieloperoxidase antibodies. Nephrol Dial Transplant 1991; 6: 773 (abstract). Neyer U, Wöss E, Neuweiler J. Wegener’s granulomatosis associated with silicosis. Nephrol Dial Transplant 1994; 9: 559-561. Nakajima H, Miyazaki M, Imai N, Yokokawa T, Yamamoto S. A case of silicosis with MPO-ANCA-associated glomerulonephritis and alveolar hemorrhage. Nippon Jinzo Gakkai Shi 2001; 43: 351-356. Beaudreuil, S, Lasfargues G, Lauériere, Ghoul ZE, Fourquet F, Longuet C, Halimi J-M, Nivet H, Büchler M. Occupatiopal exposure in ANCA-positive patients: A case-control study. Kidney Int 2005; 67: 1961-1966. Sanchez-Roman J, Wichmann I, Salaberri J, Varela JM, Nunez-Roldan A. Multiple clinical and biological autoimmune manifestations in 50 workers after occupational exposure to silica. Ann Rheum Dis 1993; 52: 534-538. Calvert GM, Steenland K, Palu S. End-stage renal disease among silica-exposed gold miners. A new method for assessing incidence among epidemiologic cohorts. JAMA 1997; 277: 1219-1223. McDonald AD, McDonald DC, Rando RJ, Hughes JM, Well H. Cohort mortality study of North American industrial sand workers. I. Mortality from lung cancer, silicosis and other causes. Ann Occup Hyg 2001; 45: 193-199. Ng TP, Ng YL, Lee HS, Chia KS, Ong HY. A study of silica nephrotoxicity in exposed silicotic and non-silicotic workers. Brit J Ind Med 1992, 49: 35-37. Boujemaa W, Lauwerys R, Bernard A. Early indicators of renal dysfunction in silicotic workers. Scand J Work Environ Health 1994; 20: 188-191.

839

D’HAESE, ELSEVIERS, YAQOOB & DE BROE

73. 74. 75. 76. 77. 78. 79.

80.

81.

82. 83. 84. 85.

86. 87. 88. 89. 90. 91. 92. 93. 94. 95.

96. 97.

840

Rosenman KD, Moore-Fuller M, Reilly MJ. Kidney disease and silicosis. Nephron 2000; 85: 14-19. Hotz P, Gonzalez-Lorenzo J, Siles E, Trujillano G, Lauwerys R, Bernard A. Subclinical signs of kidney dysfunction following short exposure to silica in the absence of silicosis. Nephron 1995; 70: 438-442. Marsh GM, Enterline PE, Stone RA, Henderson VL. Mortality among a cohort of U.S. manmade mineral fiber workers: 1985 follow-up. J Occup Med 1985; 32: 594-604. Goldsmith JR, Goldsmith DF. Fiberglass or silica exposure and increased nephritis or end-stage renal disease. Am J Ind Med 1993; 23: 873-881. Steenland K, Sanderson W, Calvert GM. Kidney disease and arthritis in a cohort study of workers exposed to silica. Epidemiology 2001; 12: 405-412 Steenland NK, Thun MJ, Ferguson CW, Port FK. Occupational and other exposures associated with male end-stage renal disease: a case/control study. Am J Public Health 1990; 80: 153-159. Gregorini G, Ferioli A, Donato F, Tira P, Morassi L, Tardanico R, Lancini L, Maiorca R. Association between silica exposure and necrotizing crescentic glomerulonephritis with P-ANCA and anti-MPO antibodies: a hospital-based case-control study. Adv Exp Med Biol 1993; 336: 435-439. Nuyts GD, Van Vlem E, De Vos A, Daelemans RA, Rorive G, Elseviers MM, Schurgers M, Segaert M, D’Haese PC, De Broe ME. Wegener granulomatosis is associated to exposure to silicon compounds: a case-control study. Nephrol Dial Transplant 1995; 10: 1162-1165. Stratta P, Messuerotti A, Canavese C, Coen M, Luccoli L, Bussolati B, Giordia L, Malavenda P, Cacciabu M, Bugiani M, Bo M, Ventura M, Camussi G, Fubini B. The role of metals in autoimmune vasculitis: epidemiological and pathogenic study. Sci Total Environ 2001; 270: 179-190. Hogan SL, Satterly KK, Dooley MA, Nachman PH, Jennette JC, Falk RJ. Silica exposure in anti-neutrophil cytoplasmic autoantibodyassociated glomerulonephritis and lupus nephritis. J Am Soc Nephrol 2001; 12: 134-142. Lane SE, Watts RA, Bentham G, Innes NJ, Scott DG. Are environmental factors important in primary systemic vasculitis? A casecontrol study. Arthritis Rheum 2003; 48: 814-823. Duna GF, Cotch MF, Galperin C, Hoffman DB, Hoffman GS. Wegener’s granulomatosis: role of environmental exposures. Clin Exp Rheumatol 998; 16(6): 669-674. Hogan SL, Cooper GS, Savitz DA, Nylander-French LA, Parks CG, Chin H, Jennette CE, Lionakin S, Jennette JC, Falk RJ. Association of silica exposure with anti-neutrophil cytoplasmic autoantibody small-vessel vasculitis: a population-based, case-control study. Clin J Am Soc Nephrol 2007; 2(2): 290-299. Flores-Suarez LF, Contreras I, Brise CA, Tilde N, Villa AR. Environmental risk factors important in primary systemic vasculitides. Kidney Blood Press Res 2003; 26: 249-302. Rihova Z, Maixnerova, D, Jancova E, Pelclova D, Bartukova J, Fenclova Z, Vankova Z, Reiterova J, Merta M, Rysava R, Tesar V. Silica and asbestos exposure in ANCA-associated vasculitis with pulmonary involvement. Ren Fail 2005; 27(5): 605-8. Hauglustaine, Van Damme B, Daenens P, Michielsen P. Silicon nephropathy: a possible occupational hazard. Nephron 1980; 26: 219-224. Kallenberg CGM. Renal disease-another effect of silica exposure. Nephrol Dial Transplant 1995; 10: 1117-1119 (Editorial Comment). Warheit DB, Driscoll KE, Oberdoerster G, Walker C, Kuschner M, Hesterberg TW. Symposium overview. Contemporary issues in fiber toxicology. Fund Appl Toxicol 1995; 25: 171-183. Schainkin-Kerstenbaum R, Adler AJ, Berlyne GM. Inhibition of superoxide dismutase activity by silicon. J Trace Elem Electrolytes Health Dis 1990; 4: 97-99. Fubini B, Bolis V, Giamell E, Volante M. Chemical functionalities at the broken fiber surface relatable to free radical production. In: Mechanisms in fibre carcinogenisis. Brown RC, editor. Plenum Press, New York 1991, p. -. Fubini B. The possible role of surface chemistry in the toxicity of inhaled fibres. In: Fiber toxicology. Warheit DB (editor). Academic Press, San Diego 1993; p. 229-257. Tervaert JWC, Stegeman CA, Kallenberg CGM. Silicon exposure and vasculitis. Current Opinion Rheumatol 1998; 10: 12-17. Gilligan HM, Bredy B, Brady HR, Hebert MJ, Slayter HS, Xu Y, Rauch J, Shia MA, Koh JS, Levine JS. Antineutrophil cytoplasmatic autoantibodies interact with primary granule constituents on the surface of apoptotic neutrophils in the absence of neutrophil priming. J Exp Med 1996; 184: 2231-2241. Iyer R, Hamilton RF, Li L, Holian A. Silica-induced apoptosis mediated via scavenger receptor in human alveolar macrophages. Toxicol Appl Pharmacol 1996; 141: 84-92. Yang JJ, Tuttle R, Taylor JG, Jennette JC, Falk RJ. ANCA target antigens are not on the surface of normal neutrophils but are present on both primed and apoptotic neutrophils. J Am Soc Nephrol 1997; 8; 546 (abstract).

37. Organic solvents, silicon-containing compounds and pesticides

98. 99.

100.

101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117.

Ueki A, Yamaguchi M, Ueki H, Watanabe Y, Ohsawa G, Kinuguwa K, Kawakami Y, Hyodoh F. Polyclonal human T-cell activation by silicate in vitro. Immunology 1994; 82: 332-335. Shiffenbauer J, Johnson H, Soos J. Superantigens in autoimmunity: their role as etiologic and therapeutic agents. In: Superantigens: Molecular biology, immunology, and relevance to human disease. Leung DYM, Huber BT, Schlievert PM (editors). Marcel Dekker, New York 1997; p. 525-549. Elseviers MM, De Schepper A, Corthouts R, Bosmans JL, Cosyn L, Lins RL, Lornoy W, Matthys E, Roose R, Van Caesbroeck D, Waller I, Horackova M, Schwarz A, Svrcek P, Bonucchi D, Franek E, Morlans M, De Broe ME. High diagnostic performance of CT scan for analgesic nephropathy in patients with incipient to severe renal failure. Kidney Int 1995; 48: 1316-1323. Perneger TV, Whelton PK, Klag MJ. Risk of kidney failure associated with the use of acetaminophen, aspirin and nonsteroidal antiinlammatory drugs. New Engl J Med 1994; 331: 1675-1679. Fored CM, Ejerblad E, Lindblad P, Fryzek JP, Dickman PN, Signorello LB, Lipworth L, Elinder CG, Blot WJ, McLaughlin JK, Zack MM, Nyrén O. N Engl J Med 2001; 345: 1801-1808. Abuelo JG. Renal failure caused by chemicals, foods, plants, animal venoms, and misuse of drugs. Arch Intern Med 1990; 150: 505-510. Vanholder R, Colardyn F, De Reuck J, Praet M, Lameire N, Ringoir S. Diquat intoxication: report of two cases and review of the literature. Am J Med 1981; 70: 1267-1271. Poisoning: toxicology, symptoms, treatments. Arena JM, Drew RH, editors. Charles C Thomas Publ, Springfield Ill 1983; p. 177179. Munk ZM, Nantel A. Acute lindane poisoning with development of muscle necrosis. Can Med Assoc J 1977; 117: 1050-1052. Nash RG, Kearney PC, Maitlen JC, Sell CR, Fertig SN. Agricultural applicators exposure to 2, 4-Dichlorophenoxyacetic acid. In: Pesticide residues and exposure. Plimmer JR, editor. ACS symposium series #182, 1982, Chap 10: 119-132. Kolmodin-Hedman B, Hoglund S, Akerblom M. Studies on phenoxy acid herbicides. I. Field study. Occupational exposure to phenoxy acid herbicides (MCPA, dichlorprop, mecoprop and 2, 4-D) in agriculture. Arch Toxicol 1983; 54(4): 257-265. Kolmodin-Hedman B, Hoglund S, Swensson A, Akerblom M. Studies on phenoxy acid herbicides. II. Oral and dermal uptake and elimination in urine of MCPA in humans. Arch Toxicol 1983; 54(4): 267-273. Libich S, To JC, Frank R, Sirons GJ. Occupational exposure of herbicide applicators to herbicides used along electric power transmission line right-of-way. Am Ind Hyg Assoc J 1984; 45: 56-62. Manninen A, Kangas J, Klen T, Savolainen H. Exposure of finnish farm workers to phenoxy acid herbicides. Arch Envir Contam Toxicol 1986; 15: 107-111. Kancir CB, Andersen C, Olesen AS. Marked hypocalcemia in a fatal poisoning with chlorinated phenoxy acid derivatives. Clin Toxicol 1988; 26: 257-264. Berndt WO, Koschier F. In vitro uptake of 2, 4-dichlorophenoxyacetic acid (2, 4-D) and 2, 4, 5-trichlorophenoxyacetic acid (2, 4, 5-T) by renal cortical tissue of rabbits and rats. Toxicol Appl Pharmacol 1973; 26: 559-570. Chugh KS, Singhal PC, Nath IVS, Pareek SK, Ubroi HS, Sarkar AK. Acute renal failure due to non-traumatic rhabdomyolysis. Postgrad Med J 1979; 55: 386-392. Averbukh Z, Modai D, Leonov V, Weissgarten J. Rhabdomyolysis and acute renal failure induced by paraphenylenediamine. Hum Toxicol 1989; 8: 345-348. Talbot AR, Shiaw MH, Huang JS, Yang SF, Goo TS, Wang SH, Chen CL, Sanford TR. Acute poisoning with a glyphosate-surfactant herbicide (‘Round-up’): a review of 93 cases. Hum Exper Toxicol 1991; 10: 1-8. De Broe ME, D’Haese PC, Nuyts GD, Elseviers MM. Occupational renal diseases. Current Opinion Nephrol Hypert 1996; 5: 114121.

841

38

Balkan nephropathy Ljubica DJUKANOVIĆ1 and Zoran RADOVANOVIĆ2 1Clinical 2Faculty

Centre of Serbia, Beograd, Serbia of Medicine, Kuwait University, Kuwait

Introduction ___________________________________________________________ 844 Epidemiological features _________________________________________________ 844 Distribution and frequency Demographic data Chronological characteristics Other epidemiological characteristics Overview of the descriptive epidemiological research

844 845 845 845 846

Etiology _______________________________________________________________ 846 Genetic factors Biological agents and their products Agents from the inanimate environment Overview of the etiological research

846 846 847 848

Pathomorphological Changes _____________________________________________ 848 Macroscopic features Morphological studies of renal changes in post mortem material Optic microscopic, immunofluorescent and electron microscopic studies of renal biopsies Overview of morphological studies

848 848 849 849

Clinical features, diagnostics and treatment __________________________________ 850 Clinical picture and course Laboratory findings Imaging methods Diagnosis Prevention and treatment Overview of clinical and laboratory studies

850 850 852 852 853 853

References _____________________________________________________________ 854

DJUKANOVIĆ & RADOVANOVIĆ

Introduction

B

alkan (or endemic) nephropathy is a chronic tubulointerstitial disease of unknown, presumably exotoxic etiology. It has been shown to exist only in some parts of the southeastern Europe. While there have been many meetings and papers [1, 2] concerning both cause and treatment of Balkan nephropathy, sociopolitical turmoil, including wars, and economical hardship prevented any meaningful research on the problem during the 1990’s. Thus, despite numerous proceedings and a large number of publications on the subject, many features of Balkan nephropathy, its etiology and natural history in particular, remained nearly as mysterious as when described in the mid-fifties. Meetings organized by international organizations [3-7] had a key role in informing the international scientific community on the disease. A recent source of information is a bilingual (in English and Serbian) monograph published in 2000 [8].

Figure 1. Medical geography of Balkan nephropathy.

844

Epidemiological features Distribution and frequency Though exclusive geographical restriction of the agent(s) of Balkan nephropathy is not very likely, the disease has been diagnosed only among people living (or those who used to live) in more or less well defined areas of the Balkans. Along with Bulgaria and Romania, three republics of the former Yugoslavia have been affected: Bosnia, Croatia and Serbia, including Kosovo (Figure 1). As recently summarized [9], the affected territory has a shape of a rhomboid. Its longer diameter spreads over 500 km (from the Vratza municipality in Bulgaria to villages west of Slavonski Brod in Croatia), while its transversal diameter has about 300 km (from endemic foci in eastern Romania to Vitina municipality in Kosovo). The disease affects individuals who live (or used to live) in rural environment. There are spared households even in the most affected areas, leading

38. Balkan nephropathy

to frequently cited remarks on mosaic distribution of the disease. Topography of the terrain differs between endemic regions. All 14 endemic villages in Croatia are located in a single lowland municipality, at an altitude of about 100 m, while Bosnian foci are found up to 130 m. About 90% of all endemic settlements in Serbia are also situated at a low altitude, below 200 m. [10], either in large plains, river valleys or, much more seldom, in hilly regions. There have been no studies of medical geography of Balkan nephropathy in Romania for the last 40 years and endemic localities are yet to be determined [11], but the original findings pointed out to hilly areas, with endemic villages laying at the bottom of valleys eroded by flooding, at an altitude of 200-300 m. The endemic region in Bulgaria was described as mountainous or semi-mountainous, without any relationship between endemicity and altitude. Hydrogeological features [12] and lack of floods differentiate (at least some) Bulgarian foci from other typical endemic regions. Controversial data on the frequency of Balkan nephropathy were mainly result of methodological shortcomings [9]. A main obstacle was the operational definition of the disease, leading to huge differences in estimated prevalence rates. The highest ever recorded average annual incidence rate was 16.6 per 1000 in Cakonica, Bulgaria. The average cause-specific mortality rate from Balkan nephropathy over 15 years in one of the most affected Serbian foci was 3.3 per 1000 [13].

Demographic data During initial Balkan nephropathy research, patients were frequently in their thirties [14], and it was widely accepted that azotemia usually affects the age group 30-50 [3]. Later an apparent shift towards the older ages occurred, with most identified patients being above the age of 60 [9]. The diagnosis of clinical forms before the age of 20 was rare and never independently confirmed. Despite occasional statements on laboratory and bioptic abnormalities in the first decade of life among clinically healthy children from endemic areas, no follow up study ever showed that these children developed subsequently kidney disorder. Both genders are similarly affected, especially considering mortality. As explained in details elsewhere [9], higher prevalence rates among women reported by

some authors appears to be a consequence of unreliable diagnostic criteria. A vast majority of experts believe that a link exist between agricultural activity and exposure to the agent(s) of Balkan nephropathy. There is also near consensus on the absence of ethnic and/or religious differences as a risk of developing the disease. The most convincing data come from Croatia, where the large group of Ukrainians who settled a century ago had the same odds of being affected as the indigenous population. The first generation immigrants developed Balkan nephropathy usually a couple of decades after moving into an endemic region. Initial studies of affected households showed a low standard of living, poor hygienic level, and insufficient nutrition. However, socio-economic factors, including living conditions and well water quality, did not differ between contiguous affected and non-affected households or between endemic and neighboring nonendemic villages.

Chronological characteristics The initial description of Balkan nephropathy emanated from Bulgaria [15, 16] and Serbia [17-19]. By 1957, the disease was recognized in Bosnia and Croatia and by 1958 in Romania [20]. On retrospect, it was not a newly emerging condition but rather recognition of an already existing endemic process, a previous epidemic wave seems having occurred in the early forties. Unfortunately, attempts to trace the disease prior to World War 2 are speculative, due to the absence of reliable data and a high frequency of the competing causes of morbidity, notably tuberculosis and malaria. As for the secular trend, two facts, common to all endemic areas, are crucial in assessing any future dynamics of the disease. These are, an apparent shift of the age distribution of the incidence towards the older age groups, and a much longer natural history of the condition compared to previous data. Consequently, it suggests that the intensity of exposure diminished (if still present at all).

Other epidemiological characteristics Clustering of cases within a household is one of the most conspicuous features of the disease. It is generally 845

DJUKANOVIĆ & RADOVANOVIĆ

agreed that the disease affects both blood related and non-blood related family members. The “phenomenon of simultaneous deaths” (dying of parents and their children within a short interval) was also observed. Between 1/3 and 1/2 of patients with Balkan nephropathy develop urothelial tumors [21]. An exceptionally high frequency of these tumors was also observed in the general population of endemic regions [22]. When initially studied, the attributive risk of developing upper urothelial tumors in inhabitants of endemic foci amounted to several dozen or even to as much as 100-200. There is no evidence that domestic and/or wild animals in endemic regions develop a similar condition.

Overview of the descriptive epidemiological research There is general agreement on the following descriptive-epidemiological characteristics of Balkan nephropathy [9]: The disease is known to exist only in some parts of the southeastern Europe, with Central Serbia as the most affected region. Balkan nephropathy does not spread beyond its already defined foci; the disease is distributed mosaically: non-endemic villages exist in the most affected regions, and there are spared families and households in the most affected settlements. Clustering of cases in families and households has been described. Children and adolescents are spared of clinical disease. Incidence is proportional to age, except for the oldest age groups. There are no major sex differences in the cause-specific mortality rates. The excess risk of developing transitional cell urothelial tumors was expressed by two- or even threedigit numbers. The large majority of researchers support the following statements [9]: autochthonous urban population is spared; rural way of life, i.e., agricultural activity is needed for exposure to the agent(s). Separation from an endemic focus early may prevent the disease, while immigration to an endemic area provides risk of disease development, providing that the exposure was sufficient. Prevalence of the disease has been stable over many years, but now appears to decline in most affected settlements. Incidence rates are shifting towards the older age groups, and the clinical course is much more protracted suggesting a less intensive contact with the agent(s) and, consequently, possible future spontaneous disappearance of Balkan nephropathy. 846

Etiology Genetic factors The most elaborate and, seemingly consistent, hypotheses regarding etiology initially came from proponents of heredity as an explanation of the disease occurrence. These authors assumed that the risk of developing the disease was restricted only to specific, ethnically distinct, population groups, irrespectively of their place of birth and residence history. Wider acceptance of these hypotheses was hampered by the different perception of descriptive epidemiology of Balkan nephropathy by a majority of researchers on the topic. A specific chromosome marker (3q25) in Balkan nephropathy patients from Bulgaria was identified, and this isolated finding was used to support arguments in favor of a crucial role of genetic factors [24-26]. More recently, the same authors acknowledged that environment is also important [27]. Some aberrations of the X chromosome have been reported, but they resembled changes occurring after exposure to ochratoxin A [28]. Major anomalies of urinary organs allegedly occurred in a high percentage of otherwise healthy children from affected households. However, such a finding has never been replicated. Genetic epidemiological approach suggest two possibilities, either polygenic type of inheritance with an insufficient expression of the main gene [29], or monofactorial model with a crucial role of a single gene of incomplete penetrance [30]. In both cases, contributing environmental factor is needed. There is no evidence supporting an immunological mechanism in Balkan nephropathy.

Biological agents and their products Unspecified viral particles [31], an unidentified cythopathogenic agent, serially propagated slow viruses [32], and an unknown virus associated with foci of natural infection [33] have been mentioned in the context of Balkan nephropathy etiology. Several specific viruses, notably West Nile [34], coronavirus [35], and papova virus [36], were also suggested as a causative agent. A common feature of all these hypotheses was unimpressive supporting evidence and ignorance of

38. Balkan nephropathy

basic epidemiological features of the disease, in particular its absence of spreading [37]. Bacteria received particular attention in initial stages of the Balkan nephropathy research but their possible etiological importance has been unanimously considered as ruled out [2]. Protozoa have never attracted any attention. Toxic fungal products were until recently the principle and prime potential culprits. Most efforts have concentrated on ochratoxin A, a mycotoxin responsible for porcine (swine) nephropathy [38]. The substance is found in endemic foci but it is also present in neighboring non-endemic areas, and the differences are not statistically significant [39, 40]. Still, the consistent isolating of ochratoxin A in greater frequency and higher concentrations from food and sera samples obtained from endemic, compared to control villages, offered some arguments in favor of this hypothesis. Association of ochratoxin A with chronic interstitial nephropathy in Tunisia [41] and its relation to renal tumors [42] provides additional support for the idea of the etiological role of this mycotoxin. Other fungal toxins, as zerealenone, citrinin [43] and aflatoxin were also isolated in endemic foci. Experimental models suggested that a combination of mycotoxins, rather than a single one, might be involved in the etiology of Balkan nephropathy [44]. Aristolochic acid and its salts, originated from a weed, Aristolochia clematitis, have toxic and carcinogenic effects to the kidneys and urothelium [45], respectively. Ivic [46] postulated that this plant may be a cause of Balkan nephropathy, but failed to provide convincing evidence from field surveys. Evidence that A. clematitis played a central role in the etiology of Chinese herb nephropathy [47-49], a condition similar to Balkan nephropathy, initiated a second look at this previously abandoned hypothesis and it gained a lot of weight by recent data on the association between DNA adduct formation derived from AA, mutation pattern and tumour development in BEN [50] (see also chapter 33). No local practice in terms of the use of teas or folk medicine could have been implicated. No one has ever studied flora of the local wells.

Agents from the inanimate environment Chronologically, lead poisoning was first offered as

an explanation for the occurrence of Balkan nephropathy [17-19]. The idea on lead-contaminated flour led to abandonment of water mills in a part of Central Serbia. This energetic public health action had no impact on the disease frequency. Effects of non-occupational exposure to cadmium [51], itai-itai disease in particular [52, 53], were occasionally compared with kidney damage seen in Balkan nephropathy patients. In spite of some resembling features, the idea of a common etiology between cadmium nephropathy (including itai-itai disease) and Balkan nephropathy was refuted [52, 54]. Many other metals, including radioactive ones such as uranium [55], were also suggested as possible causative agents of the disease. Results were non-convincing and non-reproducible. Inability to identify a single toxic effect of any metal or metalloid as a cause of Balkan nephropathy led researchers to two alternative approaches. First, deficiency, rather than abundance of such a chemical element was proposed [56], with selenium as the most likely candidate [57]. Second, attention was paid to a combined adverse effect of several elements. Synergism of uranium and some other elements, none of which exceeding maximal allowed levels, was proposed [58]. It was also noted that criteria used in occupational medicine (exposure only during working hours) have been applied to an ecological problem (constant exposure) and that concentrations of lead or cadmium within formally acceptable level, combined with other factors, such as selenium deficiency, might lead to the disease [58]. All these suggestions remained speculative. As for non-metals, there were attempts to relate Balkan nephropathy to silicon [59-62]. However, when affected and non-affected households were compared, there was even an inverse relationship between the silica content and endemicity. On one occasion, small differences in silica content happened to reach the level of statistical significance but the association was explained as a result of confounding variables [63]. Common hydrogeological characteristics of endemic foci [12] and inverse relationship between altitude of wells and disease frequency in a longitudinal (cohort) study [63], pointed to potable water as a vehicle of the agent(s). However, none of the already mentioned or several dozen other non-organic substances were associated with the disease [64]. Organics in water have been investigated and pro847

DJUKANOVIĆ & RADOVANOVIĆ

vided some interesting data [65]. Except for nitrites [66], chemically unstable substances have not been studied. Wells associated with the disease were reported as situated on alkali soil [67], but the finding was restricted to a single endemic area and never reproduced. Based on chronological data, it is clear that no pesticides, fertilizers or chemicals introduced during the last few decades may be blamed for the occurrence of Balkan nephropathy. Except for exposure to agricultural activities, no occupation, habit (e.g., smoking, alcohol consumption), or hobby (e.g., hunting, fishing) might have been shown to precede the disease onset.

Overview of the etiological research Genetic factors may play a role in different individual risk of developing Balkan nephropathy, upper urothelial tumors, both diseases or none of them [68]. However, epidemiological data indicate that one or more external, environmental factors are crucial for the occurrence of both Balkan nephropathy and excessive frequency of these tumors in endemic areas. Among biological agents and their products, the candidates for etiological agents are mycotoxins and, much more probably, toxic plants, notably Aristolochia clematitis. The possible role of viruses is very unlikely, indeed. As for inanimate environment, there is no chemical element that has been consistently detected in higher concentrations in biological material of Balkan nephropathy patients and/or their environment, as compared to the controls. However, though unlikely, insufficiency of an essential element has not been completely ruled out. Speculations on a combination of vaguely defined environmental factors have never been substantiated by facts.

A [75], Aristolochia clematitis [46], and recently with Chinese herbs [47, 49].

Macroscopic features Before introduction of hemodialysis in the treatment of chronic renal patients, the kidneys of patients who died of Balkan nephropathy used to be the smallest seen at post mortem examinations, weighing 14.8-80 g each (Figure 2A) the difference between the left and right kidneys being small (5-20 g) [74, 76-78]. Surface of the kidneys is smooth, occasionally wavy but never granulated or roughly nodular. The section shows markedly narrowed cortex, pyramid and Bertin’s columns are fairly well preserved, and corticomedular border is well differentiated. Papillary necrosis of the pyramids has not been found. Small, papillary, usually multiple tumors of the renal pelvis and ureters are also one of the characteristic findings (Figure 2B]. In post-mortem studies tumors were reported in 8-50% of cases [74, 79].

Morphological studies of renal changes in post mortem material Diffuse fibrosis of cortical interstitium and tubular atrophy may be observed along in the absence of sig-

A

Pathomorphological Changes Balkan nephropathy is non-destructive and noninflammatory tubulointerstitial renal disease [69]. The changes are non-specific and in the chronic, sclerotic phase they may be quite similar to changes observed in other chronic interstitial diseases such as analgesic nephropathy [70], vascular nephrosclerosis [69] cyclosporine-induced nephropathy [71], radiation nephritis [72, 73] and aging [72], intoxication with silicate, cadmium, lead, uranium [74], mycotoxin ochratoxin 848

B

Figure 2. A. Macroscopic appearance of the right kidney weighing 35 g in a case of BN, surface if smooth, occasionally mildly wrinkled (Autopsy, a man aged 48). B. Multifocal papillary tumor of the right ureter (autopsy).

38. Balkan nephropathy

nificant cellular interstitial infiltration. In contrast to the cortex, Bertin’s columns are less markedly affected. Even in severe tubulointerstitial cortical changes, glomeruli are well preserved, partially collapsed, and subsequently subjected to focal or generalized sclerosis mainly collapsing. Glomeruli in Bertin’s column are occasionally compensatorily enlarged. Pyramids are preserved or less markedly affected [74, 76, 79]. Blood vessels, arcuate or interlobular arteries, and arterioles are affected in the form of intimal sclerosis and thickening of lamina elastica interna. In addition, the blood vessels are compressed and torsioned [79]. In cases of tumors of the renal pelvis and ureters morphological signs of pyelonephritis are often found [74].

Optic microscopic, immunofluorescent and electron microscopic studies of renal biopsies In oligosymptomatic clinical cases, rare disseminated foci of interstitial fibrosis and tubular atrophy with preserved glomeruli are seen. These changes have no special predilection of distribution and are not inflammatory. They tend to be triangular, with the base oriented toward the renal surface [69, 79]. In cases with initial renal failure, the fields of acellular interstitial fibrosis are larger and even diffuse. The striking atrophic process observed in Balkan nephropathy suggests that apoptosis may play a role in this disease. In this context it is of interest that Savin et al. observed an increased apoptosis to proliferation ratio at the level of the tubuli [80]. The glomeruli are usually affected by generalized [80%] or segmental sclerosis [10%]/ and only in 8% hyalinosis is recorded. Double contour glomerular basement membrane was recorded in 22% of the cases. In 2.7-6% of cases fetal-like glomeruli can be seen in the kidneys, while glomerular hypercellularity was recorded in 4% [72, 79, 81]. The most interesting changes are recognized in pre- and postglomerular blood vessels. In about 50% of cases PAS positive proteins are deposited in vas afferens walls in a focal segmental or circumferent manner in the form of droplets, bands or granules [72, 79, 81]. Interlobular capillaries are filled with thick proteinaceous substances that are also deposited below the capillary endothelium and may even be found free in the interstitium. These changes are described as capil-

lary sclerosis [70]. Although renal vascular changes in Balkan nephropathy have been pointed out as very important, they are not specific and can be encountered in other renal diseases. Ferluga et al. [72] and Sindjić [79] commented on their similarities with cyclosporine induced nephrotoxic changes. Immunofluorescence revealed irregular and scarce deposits of C3, fibrin and IgM, and occasionally IgA, C1q and C4, mainly on the vascular walls, Bowman’s capsule and some sclerotic glomeruli [69, 82, 83]. Electron microscopic findings are either normal or correspond to degenerative and sclerotic changes. While some authors describe virus-like particles [35, 84, 85], others point out that such particles were not found [72, 73]. Despite these findings, some authors described Balkan nephropathy as a form of glomerulonephritis [81, 86]. However, the lack of reliable evidence supporting glomerulonephritis has lead to it being discarded [73, 77] and abandoned even by its advocates [87]. Optic microscopic, immunofluorescent and electron microscopic studies of renal biopsies in children aged 5-15 from affected families in endemic regions failed to detect any Balkan nephropathy related changes [79].

Overview of morphological studies It is generally agreed that the morphological changes of Balkan nephropathy are not specific and correspond to non-destructive, non-inflammatory kidney disease accompanied by marked changes on the blood vessels in both early and late stages of the disease, interstitial, multifocal fibrous expansion and severe tubular atrophy mainly in the upper cortex [69, 72, 73, 79, 81]. Changes on kidneys arterioles have been described suggesting that the changes in early stage of the disease may be responsible for the development of multifocal, ischemic, vascular nephrosclerosis encountered in chronic stages of the disease [69, 72]. On the other hand, close similarity of Balkan nephropathy with analgesic and cyclosporin-induced nephropathy has been recognized [71, 72, 79]. All this leads to a suggestion that Balkan nephropathy develops following a model of toxic nephropathy, targeting primarily the vascular endothelium where the tubular epithelium is affected either directly or indirectly due to accompanying ischemia. 849

DJUKANOVIĆ & RADOVANOVIĆ

Clinical features, diagnostics and treatment Clinical picture and course Balkan nephropathy is a chronic tubulointerstitial disease with occult, insidious onset, usually progressing slowly with no apparent signs of symptoms. After a long asymptomatic period, the disease is manifested as chronic renal failure. Less commonly blunt lumbar pain or renal colic may develop or, occasionally, dysuric symptoms induced by urinary tract infection. If hematuria exists, urothelial tumor should be suspected. In an advanced case polyuria and nocturia are present due to impaired concentrating ability of the kidneys. The disease is tolerated well and the patients preserve their working ability until advanced stages of renal failure [18, 76, 88, 89]. Objective examination reveals characteristic skin tan of Balkan nephropathy patients: a pale yellow with copperish glow on the cheeks has been recognized since the augural reports on the disease [18, 88]. Besides, xantochromia of the palms and soles is also frequently observed. In the advanced phase of the disease physical examination detects signs of chronic renal failure [19]. Patients with Balkan nephropathy do not suffer from edema, and their blood pressure is usually described as normal [18, 88-90]. Recently, several studies reported a higher prevalence of hypertension even in offspring of Balkan nephropathy families [91, 92, 93]. As Balkan nephropathy is characterized with slow asymptomatic course, most authors identify two main stages of the disease: the first, asymptomatic (latent, subclinical) and second, manifest (symptomatic). The latter is usually subdivided into the stage without renal failure (early, compensated Balkan nephropathy, with no azotemia) and chronic renal failure (decompensated Balkan nephropathy, uremia) [19, 88, 89]. An important feature of Balkan nephropathy is its association with a high incidence of tumors of the renal pelvis and ureters, but not urinary bladder tumors [22, 94, 95]. However, the difference between the incidence of upper urothelial tumors in endemic and non-endemic regions diminished in the last decades. In the sixties and seventies the incidence of these tumors was reported to be several dozen times higher in endemic than in non-endemic regions, while in the last decades this difference almost disappeared [21, 22, 94, 96, 97]. 850

Upper urothelial tumors of patients originating from the region with Balkan nephropathy differ from tumors identified in patients from other regions in their similar incidence in both sexes, bilateral occurrence, and more common association with chronic renal failure [95].

Laboratory findings Appearance and urine color are unchanged in most patients with Balkan nephropathy. Urine sediment is usually scarce, while microhematuria or leukocyturia are usually associated with the occurrence of tumors or urinary tract infection [88, 89]. Bacteriological studies usually reveal sterile urine, but in 8.3-31.8% significant bacteriuria was confirmed and considered as superimposed urinary tract infection [88, 89]. Proteinuria is a common finding in patients with Balkan nephropathy [18, 88]. It is usually intermittent, less than 1 g per day and it becomes permanent in advanced renal failure [98]. Although proteinuria is one of the criteria for diagnosis of Balkan nephropathy, it has been reported in healthy members of endemic families [30, 98, 99]. Tubular proteinuria is the most common and increased excretion of low-molecular weight proteins such as β2 -microglobulins, lysozyme, ribonuclease, light chains of immunoglobulin, retinolbinding protein has been reported [100-105]. Beside tubular proteinuria, smaller numbers of patients manifest mixed proteinuria, while in patients with renal failure, glomerular proteinuria may be encountered [103, 105]. Anemia has been noted in patients with Balkan nephropathy in early studies [18] and described as normocytic and normochromic or mildly hypochromic [88]. It has been suggested that anemia occurs earlier in the course of the disease progression than is the case in other renal diseases and that it precedes azotemia [88, 106]. However, recent studies have failed to substantiate this claim [98, 107]. Also, there is no evidence that anemia in Balkan nephropathy differs from anemia accompanying other renal diseases in either features [107] or rate of deterioration in the progression of renal failure [108]. Nevertheless, anemia in Balkan nephropathy patients treated with hemodialysis is more severe than in patients with other renal diseases [108]. The leukocyte count in the peripheral blood of patients with Balkan nephropathy is normal and without

Figure 3. Interstitial fibrosis and tubular atrophy; glomerulus with mild mesangial hypercellularity and another with incomplete hyalinosis. PAS, x120.

Figure 4. Interlobar artery showing intimal fibrosis. PAS, x240.

Figure 5. Reduced number of tubules; fibrotic interstitium; few infiltrating cells. Masson’s trichrome, x120.

Figure 6. Extensive interstitial scarring associated with severe arterio and arteriolosclerosis. PAS, x120.

DJUKANOVIĆ & RADOVANOVIĆ

pathological changes in the differential count and bone marrow [19, 88]. Investigation of renal function in patients with endemic nephropathy has revealed tubular dysfunctions in the earliest stage of the disease: renal glycosuria, increased uric acid and amino acid excretion [101], as well as increased excretion of low molecular weight proteins [104]. Significantly higher activity of cellular enzymes in the urine and increased urinary excretion of Tamm-Horsfall protein was described in patients with Balkan nephropathy, as well as in healthy members of endemic families [109]. Findings of a distal tubular disorders (impaired urinary acidification, impaired urine concentrating ability) were described in earlier studies [88, 89] but could not be confirmed in studies conducted in larger groups of patients with normal or mildly impaired glomerular filtration rate [98, 110]. The occurrence of certain disorders of the tubular function recorded in the course of chronic renal failure (increased natriuria, phosphaturia) can be considered as the result of kidney adaptation to the lost nephron mass, instead of Balkan nephropathy properties [110]. The immunological studies have failed to indicate that immune disorders participate in the pathogenesis of Balkan nephropathy, with some of detected changes having been attributed to advanced renal failure [111].

Imaging methods Different methods of kidney imaging have shown that Balkan nephropathy patients with chronic renal failure have symmetrically shrunken kidneys with smooth surface and no calcifications [90]. The time at which the shrinking occurs remains to be determined. While some authors suggest that the size of the kidneys remains normal in patients in the latent phase of the disease and with normal renal function, others report cases of shrunken kidneys in patients in an early phase with normal glomerular filtration rate, and it was even proposed that the disease was characterized with primarily small kidneys [98, 110, 112]. As ultrasound became a standard imaging method in the evaluation of kidney dimensions, several recent studies that used this method showed diminished kidney length and cortex width in members of Balkan nephropathy families with normal kidney function [113, 114]. Besides, 852

significantly shorter kidney length, as well as higher protein, albumin and b2-microglobulin excretion was found among offspring with a maternal history of Balkan endemic nephropathy (BEN), not a paternal one [113] . Excretory urography does not reveal changes in the pyelocaliceal system, except in cases with secondary infection or urothelial tumors. Radionuclide methods have shown that renal plasma flow impairment is the first sign of the early phase. Glomerular and tubular functions correspond to the severity of the disease.

Diagnosis The most commonly used criteria for the diagnosis of Balkan nephropathy are still those proposed by Danilović [106]. They include: 1) farmers living in the endemic villages, (2) familial history positive for Balkan nephropathy, (3) mild proteinuria, (4) low specific gravity of the urine, (5) anemia, (6) retention of nitrogen compounds in the blood (urea > 50 mg/dl, creatinine > 1.5 mg/dl) and (7) symmetrically shrunken kidneys. Using these criteria, Danilović suggested classification of patients in field studies into the following groups: 1. potential, a group with intermittent proteinuria, those that fulfill at least the first three criteria, 2. suspected patients, that in addition to the first three fulfill at least one of the remaining three criteria, 3. affected patients, that fulfill at least 5 out of 6 criteria, 4. decompensated patients that fulfill at least 5 out of 6 criteria and have urea values >150 mg% and manifested signs of uremia. Analysis of these criteria leads to the conclusion that they enable detection only of patients with overt disease, and the criteria are not sufficiently specific to enable a reliable diagnosis. Therefore, numerous studies have been focused on developing sufficiently sensitive and specific criteria to enable diagnosis in the early phase. Although markers of tubular disorders, particularly tubular proteinuria, may be used as sufficiently specific diagnostic criteria, so far not a single clinical or laboratory finding is considered pathognomonic for Balkan nephropathy when differentiate it from other, specially, tubulointerstitial diseases. The diagnosis of Balkan nephropathy is now established according to the first two criteria (residence in

38. Balkan nephropathy

endemic village and positive family history) suggested by Danilović [106], presence of tubular proteinuria and ruling out other renal diseases. Histopathological analysis makes the diagnosis of Balkan nephropathy significantly easier [72, 79], and it is considered indispensable in classifying the following groups of patients with urinary abnormalities suggestive of endemic nephropathy: 1. Patients from families that were not previously been affected with endemic nephropathy, but live in an endemic village, 2. In cases of nephropathy of unknown etiology in villages close to endemic foci, 3. In immigrants to endemic regions and in emigrants from these regions [111]. Differential diagnosis of Balkan nephropathy should include all chronic, slowly progressive renal diseases, primarily chronic tubulointerstitial diseases. Although no specific indicators of Balkan nephropathy have been recognized, epidemiological data, familial history as well as clinical characteristics of the disease enable differential diagnosis. Thus, shrunken kidneys with smooth surface are characteristic of Balkan nephropathy and they differentiate it from analgesic nephropathy, pyelonephritis or reflux nephropathy that are characterized by shrunken kidneys with uneven surface. Pyelocaliceal system of the kidneys remains unaffected in patients with Balkan nephropathy, unlike the characteristic changes observed in pyelonephritis or obstructive nephropathy. Absence of papillary necrosis/calcifications also enables differentiation of Balkan nephropathy from analgesic, obstructive, reflux nephropathy [110, 115]. Recently similarity of Balkan nephropathy and nephropathy induced by Chinese herbs used in slimming diets have been suggested [48]. Nevertheless, Chinese herb nephropathy is rapid progressive tubulointerstitial diseases with pronounced fibrosis and progression towards end-stage renal disease within few years, clearly different from the protracted clinical course of Balkan nephropathy.

Prevention and treatment Balkan nephropathy is a disease of unknown etiopathogenesis, so that recommendations regarding effective prevention are not possible. Efforts have been made to improve the living conditions, bring high

quality drinking water to endemic villages and undertake other hygienic measures. Treatment is planned according to the stage of the disease. In principle, the treatment involves the measures for slowing down deterioration of renal function and those applied in chronic renal failure [116]. End-stage renal disease is treated with dialysis and kidney transplantation. Hypertension and cardiovascular diseases affect the Balkan nephropathy patients less frequently, so they tolerate hemodialysis rather well compared to patients with other renal diseases. The Balkan nephropathy patients on long-term hemodialysis frequently develop upper urothelial or urinary bladder carcinoma. Although the number of reported cases with kidney transplant is small, neither specific post-transplantation problems nor disease recurrency on the transplanted kidney have been described. However, recent studies indicated that patients with Balkan nephropathy are at increased risk for the development of upper urothelial tumors in both native and transplanted kidneys [117].

Overview of clinical and laboratory studies Balkan nephropathy is a chronic tubulointerstitial disease with insidious occult onset progressing without symptoms. Agreement as to how to define the early asymptomatic phase of the disease is lacking, since no specific indicators for the diagnosis have been recognized. The diagnosis is established according to epidemiological criteria (farmers in endemic villages, familial history positive for endemic nephropathy), presence of tubular proteinuria, findings of symmetrically shrunk kidneys with smooth surface, without calcifications and ruling out of other renal disease. Renal biopsy may make the diagnosis easier, although the changes are non-specific. One of the important features of Balkan nephropathy is its association to high incidence of tumors of the renal pelvic and ureters, comparable to analgesic nephropathy (se chapter 17) and aristolochic acid nephropathy (see chapter 33). So far, laboratory studies have failed to detect any disorder as a specific marker for early detection of the disease or a reliable indicator for differential diagnosis. Laboratory studies have confirmed that Balkan nephropathy is a tubulointerstitial disease so that tubular disorders precede impairment of glomerular 853

DJUKANOVIĆ & RADOVANOVIĆ

filtration. Although anemia is one of the criteria for the diagnosis of the disease, it has not been evidenced that pathogenesis and features of this anemia differ from that observed in other chronic renal diseases. It is only more severe in end-stage Balkan nephropathy patients

than in patients with other kidney diseases. Acknowledgement Photos of all tissue specimens were provided by Professor Jovan Dimitrijević.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

854

Astrug A (editor). Endemic Nephropathy – Bibliography 1956-1970, Sofia: Ministry of Public Health; 1971. Radovanović Z. Aetiology of Balkan nephropathy: A reappraisal after 30 years. Eur J Epidemiol 1989; 5:372-7. World Health Organization. The “Endemic Nephropathy” of South-Eastern Europe. Bull Wld Hlth Org 1965; 32(3):431-448. Wolstenholme GEW, Knight J (editors). The Balkan Nephropathy. Ciba Foundation Study Group No. 30. London: Churchill; 1967. World Health Organization. WHO Meeting of Investigators on Endemic Nephropathy, Belgrade and Lazarevac, November 1974, NCD/WP/74. Castegnaro M, Plestina R, Dirheimer G, Chernozemsky IN, Bartsch H (editors). Mycotoxins, Endemic Nephropathy and Urinary Tract Tumours. IARC Scientific Publications No. 115. Lyon: International Agency for Research on Cancer; 1991. Hall P, Batuman V (guest editors). Kidney Int 1991; 40: Suppl 34 (Proceedings of the IUPAB meetings, Belgrade, 1989 and 1990). Radovanović Z, Sinđić M, Polenaković M, Djukanović Lj, Petronić V. Endemic Nephropathy. Belgrade: Office for Textbooks and Teaching Aids, 2000. Radovanović Z. Epidemiology and aetiology of endemic nephropathy. In: Radovanović Z, Sinđić M, Polenaković M, Djukanović Lj, Petronić V. Endemic Nephropathy. Belgrade: Office for Textbooks and Teaching Aids, 2000. Radovanović Z. Topographical distribution of the Balkan endemic nephropathy in Serbia (Yugoslavia). Trop geogr Med 1979; 31:185-189. Gluhovschi G, Stefanovic V, Dimitrov Tz et al. Endemic Balkan Nephropathy. Helicon, Timisoara, 1994. Radovanović Z, Perić J. Hydrogeological characteristics of endemic nephropathy foci. Public Health (Lond.) 1979; 93:76-81. Radovanović Z, Danilović V, Velimirović D et al. Beta2-microglobulinuria as a predictor of death in a population exposed to Balkan endemic nephropathy. Kidney Int 1991; 40(S34):32-34. Danilović V. Effet des recherches des néphrites endemiques en Yougoslavie. Arch. Union Méd. Balkan. 1964; 2(1): 31-40. Tančev I, Evstatijev P, Dorosiev D, Penceva Z, Cvetkov G. Proučvanija na nefritite v’v Vracanska okolija. S’vr. med. 1956; 7(9): 1429. Ivanov L. Proučvanija v’rhu etiologijata na b’brežnite zaboljavanija v’v Vracanska okolija. S’vr. med. 1956; 7(9): 30-34. Danilović V, Djurišić M, Mokranjac M, Stojimirović B, Živojinović J, Stojaković P. Porodična oboljenja bubrega u selu Šopic izazvana hroničnom intoksikacijom olovom. Srp. arh. celok. lek. 1957; 85(10): 1115-25. Danilović V, Djurišić M, Mokranjac M, Stojimirović B, Živojinović J, Stojaković P. Néphrites chroniques provoquées par l’intoxication au plomb par voie digestive (farine). Presse méd. 1957; 65(90): 2039-40. Danilović V. Chronic nephritis due to ingestion of lead-contaminated flour. Brit. Med. J. 1958; 1: 27-8. Milojčić B. Epidemic of chronic nephritis of unknown aetiology. Brit. Med. J. 1960; i:244-5. Radovanović Z, Janković S, Jevremović I. Incidence of tumors of urinary organs in a focus of Balkan endemic nephropathy. Kidney Int 1991; 40(S34): 75-76. Petković S, Mutavdžić M, Petronić V, Marković V. Les tumeurs du bassinet et de l’uretere. Recherches cliniques et étiologique. J. Urol. Néphrol. 1971; 77(6): 429-39. Radovanović Z, Krajinović S, Petković S, Hall PW. Papillary transitional cell tumours, Balkan nephropathy, and beta2-microglobulin. The Lancet 1981; 2:689. Toncheva D, Dimitrov Ts, Tzoneva M. Cytogenetic studies in Balkan endemic nephropathy. Nephron 1988; 48:18-21. Toncheva D, Gergov T, Tzoneva M, Bouchakliev Z. Spontaneous and induced chromosome aberrations in Balkan endemic nephropathy. Kidney Int. 1991; 40 (Suppl. 34): 97-101. Toncheva D, Dimitrov T. Genetic predisposition to Balkan endemic nephropathy. Nephron 1996; 72(4):564-9. Toncheva D, Dimitrov T, Stojanova S. Etiology of Balkan endemic nephropathy: a multifactorial disease? Eur J Epidemiol. 1998 Jun;14(4):389-94.

38. Balkan nephropathy

28.

29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.

45. 46. 47.

48. 49. 50.

51. 52. 53. 54. 55.

Manolova Y, Manolov G, Parvanova L, Petkova-Bocharova T, Castegnaro M, Chernozemsky IN. Induction of characteristic chromosomal aberrations, particularly X-trisomy, in cultured human lymphocytes treated by ochratoxin A, a mycotoxin implicated in Balkan endemic nephropathy. Mutat. Res. 1990; 231(2):143-9. Tucić N, Marinković D. The inheritance of liability of Balkan nephropathy. Genetika 1979; 11(3): 213-20. Banjević D, Tucić N. The monofactorial model for inheritance of liability to disease and its implications for relatives at risk. Genetika 1979;11(3): 221-30. Georgescu L, Litvac B, Manescu N, Petrovici A, Schnjartzkopf A, Zosin C. Particules virales dans le rein de la nephropathie endemique Balcanique. Sem. Hôp. Paris 1970; 46:3526. Draganescu N, Antipa C, Girjabu E, Melencu M. Endemic Balkan nephropathy, a slow virus disease? Rev. Roum. Med.-Virol. 1983; 34(1): 3-6. Birtašević B, Vuković B, Drndarević D et al. Endemska (Balkanska) nefropatija kao prirodno-žarišna infekcija? Vojnosanit pregl 1983; 40: 319-24. Kraus N. Studien über die endemische Nephrozirrhose der Balkanhalbinsel. Zeitschrift für ärztliche Fortbildung 1966; 60:82934. Apostolov K, Spasić P, Bojanić N. Evidence of a viral etiology in endemic (Balkan) nephropathy. Lancet 1975; ii: 1271-3. Nastac E, Stoian M, Hozoc M, Iosipenco M, Melencu M. Further data on the prevalence of serum antibodies to papova viruses (BK and SV40) in subjects from the Romanian area with Balkan endemic nephropathy. Rev. Roum. Med.-Virol. 1984; 35(1):65-7. Radovanović Z. Epidemiological evidence on Balkan nephropathy as a viral disease. Medical Hypotheses 1987; 22:171-5. Krogh P. Causal association of mycotoxic nephropathy. Acta Path Microbiol Scand Sect A (Suppl) 1978; 269:1-28. Petkova-Bocharova T, Castegnaro M. Ochratoxin A contamination of cereals in an area of high incidence of Balkan endemic nephropathy in Bulgaria. Food Addit Contam 1985; 2: 267-70. Cvetnić Z, Pepeljnjak S. Ochratoxinogenicity of Aspergillus ochraceus strains from nephropathic and non-nephropathic areas in Yugoslavia. Mycopathologia 1990; 110(2): 93-9. Maaroufi K, Achour A, Betbeder AM et al. Foodstuffs and human blood contamination by the mycotoxin ochratoxin A: correlation with chronic interstitial nephropathy in Tunisia. Arch. Toxicol. 1995; 69:552-8. Maaroufi K, Pfohl-Leszkowicz A, Achour A et al. Ochratoxin A genotoxicity, relation to renal tumors. Arch Inst Pasteur Tunis 1994; 71(1-2):21-31. Yeulet SE, Mantle PG, Rudge MS, Greig JB. Nephrotoxicity of Penicillium aurantiogriseum, a possible factor in the aetiology of Balkan endemic nephropathy. Mycopathologia 1988; 102: 21-30. Bach PH, Gregg NJ, Delacruz L. Relevance of a rat model of papillary necrosis and upper urothelial carcinoma in understanding the role of ochratoxin A in Balkan endemic nephropathy and its associated carcinoma. Food Chem Toxicol. 1992; 30(3): 20511. Mengs U, Lang W, Poch JA. The carcinogenic action of aristolochic acid in rats. Arch. Toxicol. 1982; 51:107-19. Ivić M. Problem etiologije endemske nefropatije. Acta Fac Med Naiss 1970; 1:29-38. Vanherweghem JL, Depierreux M, Tielemans C, Abramowicz D, Dratwa M, Jadoul M, Richard C, Vandervelde D, Verbeelen D, Vanhaelen-Fastre R, et al.Rapidly progressive interstitial renal fibrosis in young women: association with slimming regimen including Chinese herbs. Lancet. 1993;341(8842):387-91. Cosyns JP, Jadoul M, Squifflet JP et al. Chinese herbs nephropathy: a clue to Balkan endemic nephropathy? Kidny Int. 1994; 45(6):1680-8. Cosyns JP, Dehoux JP, Guiot Y, Goebbels RM, Robert A, Bernard AM, van Ypersele de Strihou C. Chronic aristolochic acid toxicity in rabbits: a model of Chinese herbs nephropathy? Kidney Int. 2001; 59(6):2164-73 Grollman AP, Shibutani S, Moriya M, Miller F, Wu L, Moll U, Suzuki N, Fernandes A, Rosenquist T, Medverec Z, Jakovina K, Brdar B, Slade N, Turesky RJ, Goodenough AK, Rieger R, Vukelić M, Jelaković B. (2007) Aristolochic acid and the etiology of endemic (Balkan) nephropathy. PNAS 2007; 104: 12129-12134. Nogawa K, Kobayashi E, Honda R, Ishizaki A, Kawano S, Matsuda H. Renal dysfunctions of inhabitants in a cadmium-polluted area. Environ Res 1980; 23(1): 13-23. Danilović V (editor). Bulletin, T. LXIX, Classe des sciences medicales No. 12. Beograd: Academie Serbe des Sciences et des Arts, 1980. Frieberg L, Elinder C-G, Kjellstrom T, Nordberg GF (editors). Cadmium and Health: A Toxicological and Epidemiological Appraisal. Boca Raton, CRC Press, 1986. Piscator M, Kostial K, Plestina R. Urinary excretion of cadmium and zink among women with Balkan endemic nephropathy. Trace Elements Med. 1984; 1:134-8. Ignjatović B. Prilog proučavanju etiologije hroničnog nefrita. Narodno zdravlje 1962; 18: 329-34.

855

DJUKANOVIĆ & RADOVANOVIĆ

56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74.

75.

76. 77. 78. 79. 80. 81. 82. 83. 84.

856

Maksimović Z, Radovanović Z. Balkan endemic nephropathy in Yugoslavia and geochemical studies. In: Hemphill DD, ed. Trace Substances in Environmental Health (18). Univ. Missouri, 1984; p. 230-6. Maksimović Z. Selenium deficiency and Balkan endemic nephropathy. Kidney Int 1991; 40 (suppl. 34):S12-14. Wedeen RP. Environmental renal disease: Lead, cadmium and Balkan endemic nephropathy. Kidney Int 1991; 40 (suppl. 34): S4-8. Duančić V. Essai d’explication de la manifestation de l’affection rénale de masse å Slavonski Kobas. Biol. glasnik 1960; 3:43-50. Marković B, Lebedev S. Role étiopathogénique des silicates érosifs dans la néphrite endémique. Presse méd. 1965; 73(8):4014. Marković B, Arambašić M. Experimental chronic interstitial nephritis compared with endemic human nephropathy. J. Pathol. 1971; 103: 35-40. Goldsmith JR, Goldsmith DF. Fiberglass or silica exposure and increased nephritis or ESRD (end-stage renal disease). Am J Ind Med 1993; 23:873-81. Radovanović Z, Marković-Denić Lj, Marinković J, Jevremović I, Janković S. Well water characteristics and the Balkan nephropathy. Nephron 1991; 57:52-54. Radovanović Z, Edmunds WM. Groundwater chemistry and the incidence of Balkan endemic nephropathy. Environ Geochem Health 1991; 13(2):43-9. Finkelman RB, Feder GL, Orem WH, Radovanovic Z. Relationship between low-rank coal deposits and Balkan endemic nephropathy. AGID News, 1991; 65: 23. Radovanović Z, Stevanović D. Different forms of mineral nitrogen in drinking water and the Balkan nephropathy. Arch Environm Contam Toxicol 1988; 17: 813-5. Filipović A, Abinun A, Nijemčević S, Leković G, Popović N. Ispitivanje ovisnosti pojave endemske nefropatije i sastava zemljišta iz kojeg stanovništvo uzima vodu za piće u opštini Modriča. Med arhiv 1977; 31(5): 277-84. Radovanović Z, Velimirović D, Naumović T. Upper urothelial tumours and the Balkan nephropathy - Inference from the study of a family pedigree. Eur J Cancer 1990; 26:391-2. Sindjić M. Pathomorphology of the kidneys in endemic nephropathy. Academie Serbe des Scientes et des Arts Bulletin 1980; 69: 127-140. Zollinger H U, Mihatsch M J, Torhost J. Nephropathy due to analgesics. Morphological aspects. XIII Inter. Congr. of Inter. Acad. of Pathology, Paris, Abstracts, 1980; 17. Mihatsch MJ, Thuel G, Ryffel N. Morphology of cyclosporin nephropathy. Prog Allergy 1986; 38: 447-465. Ferluga D, Hvala A, Vizjak A, Trnavcevic S, Halibasic A. Renal function, protein excretion and pathology of Balkan endemic nephropathy. III Light and electron micorscopy studies. Kidney Int 1991; 40 (Suppl 34): S54 - S67. Hall PW III, Dammin GJ Balkan nephropathy. In: Renal pathology with clinical and functional correlations. Fisher G, Brenner BM (editors). J.B. Lippincott Co., Philadelphia, 1989; 913-924. Sindjić M. Resultats des recheches patho-anatomiques, faites jusqu’a present, concernant la nephropathie endemique.In: I Symposium sur la nephropathie endemique. 1973 Beograd. Danilovic V (redacteur). Academie Serbe des Scientces et des Arts, Beograd 1973; 187-202. Krogh P, Hald B, Pleština R, Čeović S. Enviromental Ochratoxin A and Balkan (endemic) nephropathy: Evidence for support of a casual relationship. In: Endemic (Balkan) Nephropathy, Proc. 4th Symp., Nis 1979. Eds. S. Strahinjic & V. Stefanovic. Inst. Nephr. Haemod., Nis 1981, 35-43. Radonić M, Radošević Z, Županić V. Endemic nephropathy in Yugoslavia. In: The Kidney. Mostby S, David E (editors). Wiliams & Wilkins Co., Baltimore, 1966; 503-522. Zosin L et al. Aspects anatomo-cliniques de la néphropathie endémique balkanique. Sém Hôp, 1966; 42(3/1):194-201. Hall PW et al. Investigation of chronic endemic nephropathy in Yugoslavia. II Renal Pathology. Am J Med 1965; 39(2): 210-217. Sindjić M. Morphological changes in kidneys affected by endemic nephropathy. In: Radovanović Z, Sindjić M, Polenaković M, Djukanović Lj, Petronić V. Endemic nephropathy. Belgrade: Office for Textbooks and Teaching Aids, 2000. Savin M, Bumbaširević V, Djukanović Lj, Petronić V. The significance of apoptosis for early diagnosis of Balkan nephropathy. Nephrol Dial Transplant 2001; 16(Suppl 6): 30-32. Doichinov D, Todorov D, Dinev I, Milev B. Biopsy specimen examination during the early stage of endemic nephropathy. In: Endemic Nephropathy. Proc. 2nd Int. Symp. on End. Nephr., Sofia 1972, Bulg. Acad. Sc. Press, Sofia 1974, 189-194. Polenaković M, Hrizoho D. Endemska nefropatija – imunofluorescentna mikroskopska ispitivanja. In: II Simpozijum o endemskoj nefropatiji, Beograd-Lazarevac 1977. Ur. V. Danilovic, SANU, Beograd 1979, 137-150. Vizjak A., Trnačević S., Ferluga D., Halilbašić A. Renal function, protein excretion and pathology of Balkan endemic nephropathy. IV-Immunohistology. Kidney Int 1991; 40 (Suppl 34) 68-74. Georgescu L, Litvac B, Diosi P, Plavosin L, Herzog G. Viruses in endemic (Balkan) nephropathy. Lancet 1976; i: 1086.

38. Balkan nephropathy

85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111.

Uzelac-Keserovic B, Vasic D, Ikonomovski J, Bojanic N, Apostolov K. Isolation of a coronavirus from urinary tract tumours of endemic Balkannephropathy patients. Nephron 200;86:93-4 Pantić VR, Suša S. Ultrastructure of nephrons in kidney of endemic nephropathy patients. In: Endemic (Balkan) Nephropathy, Proc. 4th Symp., Nis 1979. Strahinjic S & Stefanovic V (editors). Inst Nephr Haemod, Nis 1981; 123-132. Doitchinov D. Immunohistochemical studies in Balkan endemic nephropathy. In: Endemic (Balkan) Nephropathy, Proc. 4th Symp., Nis 1979. Strahinjic S & Stefanovic V (editors). Inst. Nephr. Haemod., Nis 1981, 265. Puchlev A, Popov N, Astrug A, Dotchev D. Clinical studies on endemic nephropathy in Bulgaria. In: Int. Symp. on End. Nephr., Sofia, 1963, Bulg. Acad. Sc. Press, Sofia 1965; 17-24. Bruckner I., Zosin C ., Lazarescu R, et al. A clinical study of nephropathy of an endemic character in the people’s Republic of Rumania. In: Int. Symp. on Endemic Nephropathy, Sofia 1963. Bulg. Acad. Sc. Press, Sofia 1965; p. 25-35. Djukanović Lj, Velimirović D, Sindjić M. Balkan Nephropathy. In: Clinical Nephrotoxins. Renal Injury from Drugs and Chemicals. De Broe ME, Porter GA, Bennett WM, Verpoolen GA (editors) Kluwer Academic Publishers. Dordrecht 1988; 426-436. Arsenovic A, Bukvic D, Trbojevic S, Maric I, Djukanovic L. Detection of renal dysfunctions in family members of patients with Balkan Endemic Nephropathy. Am J Nephrol 2005; 25: 50–54. Dimitrov P, Tsolova S, Georgieva R, Bozhilova D, V Simeonov V, Bonev A, Karmaus W. Clinical markers in adult offspring of families with and without Balkan endemic nephropathy. Kidney Int 2006; 69: 723–729. Bukvić D, Marić I, Arsenović A, Janković S, Djukanović Lj. The prevalence of Balkan endemic nephropathy has not changed since 1971 in the Kolubara region in Serbia. Kidney Blood Press Res 2007;30:117-123 Petković S. Epidemiology and treatment of renal pelvic and ureeral tumours. J Urol 1971; 114: 858. Petronić V. Tumors of the upper urothelium and endemic nephropathy. In: Endemic nephropathy. Radovanović Z, Sindjić M, Polenaković M, Djukanović Lj, Petronić V. Office for Textbooks and Teaching Aids, Beograd, 2000. Markovic N, Ignjatovic I, Cukuranovic R, Petrovic B, Kocic B, Stefanovic V. Decreasing incidence of urothelial cancer in a Balkan endemic nephropathy region in Serbia. A surgery based study from 1969 to 1998. Pathol Biol (Paris). 2005;53: 26-9. Dragicevic D, Djokic M, Pekmezovic T, Micic S, Hadzi-Djokic J, Vuksanovic A, Simic T. Survival of patients with transitional cell carcinoma of the ureter and renal pelvis in Balkan endemic nephropathy and non-endemic areas of Serbia. BJU Int. 2007; Trnačević S, Halilbšić A, Ferluga D et al. Renal function, protein excretion and pathology of Balkan endemic nephropathy. I. Renal function. Kidney Int 1991; 40 (Suppl 34): S49-S51. Čvorišćec D, Radonić M, Čeović S, Aleraj B. Characteristics of proteinuria in endemic nephropathy. J Clin Chem Clin Biochem 1983; 21: 569-571. Bruckner I, Stoica G, Serban M. Studies on urinary proteins. In: The Balkan Nephropathy. Wolstenholme GEW & Knight J (editors). Ciba Foundation Study Group No. 30. Churchill, London 1967; p. 15-20. Hall PW, III, Piscator M, Vasiljević M, Popović N. Renal function studies inindividuals with the tubular proteinuria of endemic Balkan nephropathy. Q J Med 1972; 41: 385 - 393. Radošević Z, Traeger J, Radonić M, Manuel Y, Revillard J P. Etude elctrophoretique de la proteinurie de 31 sujet crates vivant en pays de nephropathie endemique. J Urol Nephrol 1968; 74: 703-10. Bruckner I, Stoica Gh. Urinary proteins in renal diseases. II. Alpha2-microglobulins in tubular protienuria with special reference to endemic nephropathy. Rev Roum Med Med Int 1978; 16: 2-8. Hall PW, Vasiljević M. Beta2-microglobulin excretion as an index of renal tubular disorders with special reference to endemic Balkan nephropathy. J Lab Clin Med 1973; 81: 897-904. Raičević S, Trnačević S, Hranisavljević J, Vučelić D. Renal function, protein excretion, and pathology of Balkan endemic nephropathy. II. Protein excretion. Kidney Int 1991; 40(Suppl. 34): S-52-56. Danilović V. Endemic nephropathy in Yugoslavia. In: Endemic (Balkan) Nephropathy, Proc. of the 4th Symp., Ni{ 1979. Strahinji} S & Stefanovi} V (editors). Inst. Nephr. Haemod., Nis 1981; 1-5. Pavlović -Kentera V, Djukanović Lj, Clemons G K, Trbojević S, Dimković N, Slavković A. Anaemia in Balkan endemic nephropathy. Kidney Int 1991; 40 (Suppl. 34): S46-S48. Pavlovic-Kentera V, Clemons G K, Trbojevic S, Dimkovic N, Djukanovic Lj. Erythropoietin and anemia in the progression of Balkan endemic nephropathy and other renal diseases. Nephron 1990; 54: 139-143. Čvorišćec D, Stavljenić A, Radonić M. Relationship between tubular and Tamm-Horsfall proteinuria in Balkan endemic nephropathy. Nephron 1998; 42: 152-155. Djukanović Lj. Laboratory studies. In: Endemic nephropathy. Radovanović Z, Sindjić M, Polenaković M, Djukanović Lj, Petronić V (editors). Office for Textbooks and Teaching Aids, Beograd 2000. Polenakovic M, Stefanovic V. Balkan Nephropathy. In: Oxford Textbook of Clinical Nephrology, Oxford medical Publications, Sec. ed. Davison A., Cameron JS, Grunfeld JP, Kerr DNS, Ritz E and Winearls Ch (editors). Oxford, Oxford University Press, 1998; 12021210.

857

DJUKANOVIĆ & RADOVANOVIĆ

112. Hrisoho DT. Balkan Endemic Nephropathy. Study of pathogenesis based on genetic factors. In: Endemic (Balkan) Nephropathy, Proc. 4th Symp., Nis 1979. Eds. S. Strahinjic & V. Stefanovic. Inst. Nephr. Haemod., Nis 1979; 1981; p. 147-154. 113. Dimitrov P, Tsolova S, Georgieva R, Bozhilova D, V Simeonov V, Bonev A, Karmaus W: Clinical markers in adult offspring of families with and without Balkan endemic nephropathy. Kidney Int 69: 723–729, 2006. 114. Djukanovic L, Bukvic D, Maric I. Creatinine clearance and kidney size in Balkan endemic nephropathy patients. Clin Nephrol 61:384-6, 2004. 115. Elseviers MM, De Schepper A, Corthouts R et al. High diagnostic performance of CT scan for analgesic nephropathy in patients with incipient to severe renal failure. Kidney Int 1995; 48: 1316-1323. 116. Polenaković M, Djukanović Lj. Clinical picture, diagnostics and treatment of endemic nephropathy. In: Endemic nephropathy. Radovanović Z, Sindjić M, Polenaković M, Djukanović Lj, Petronić V (editors). Belgrade: Office for Textbooks and Teaching Aids, 2000. 117. Basic-Jukic N, Hrsak-Puljic I, Kes P et al. Renal transplantation in patients with Balkan endemic nephropathy. Transplant Proc 2007; 39:1432-5.

858

39

Nephrotoxins in Africa Charles SWANEPOEL1, Marc BLOCKMAN2 and Joe TALMUD2 Divisions of 1Nephrology and 2Pharmacology, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa

Introduction ___________________________________________________________ 859 Potassium dichromate ___________________________________________________ 861 Cresols ________________________________________________________________ 862 Cantharidin ____________________________________________________________ 862 Ox-eye daisy or Impila (Callilepsis Laureola) __________________________________ 863 Cape aloe _____________________________________________________________ 863 Senecio _______________________________________________________________ 864 Mercury _______________________________________________________________ 864 Paraphenylene diamine (hair dye) __________________________________________ 865 Sodium bromate _______________________________________________________ 865 Copper Sulphate ________________________________________________________ 865 Paraquat ______________________________________________________________ 866 Crystal metamphetamine and methylenedioxymethamphetamine _______________ 866 Ethylene glycol _________________________________________________________ 867 Violet tree _____________________________________________________________ 868 Discussion _____________________________________________________________ 868 References _____________________________________________________________ 869

and in most instances deny both the consultation, and subsequent ingestion of prescribed herbal therapy oxin-induced acute kidney injury (AKI) is a com- [2]. This handicaps the planning of management, mon cause of morbidity and mortality in Africa. particularly as some toxins have multi-system effects, e.g. acute kidney injury accompanied by hepatitis and However reports in the medical literature are colitis, as occurs in dichromate poisoning [3]. While the limited because in the majority of cases, identification majority of patients admitted with poisoning have been of the precise toxin is missing [1]. A majority of the toxin induced acute kidney injury prescribed by traditional healers, approximately 12% of the patients have obtained their medications from in South Africa follows a visit to the traditional diviner (the “sangoma”). This often results in a conspiracy of “African” shops (equivalent of a western–style chemist) [1]. It is not always the diviner who is responsible silence; the patients are reluctant to admit such a visit

Introduction

T

SWANEPOEL, BLOCKMAN & TALMUD

for the prescribing these toxins, but rather the patients who buy medicines without completely understanding their content. De Smet [4] and others [5] have advocated the need to disseminate knowledge about the risks and benefits of herbal and alternative medicines. Such information would allow ‘ingestors’ of such medicines the knowledge to decide whether or not to consume herbal concoctions. The ingestion of alternative medicines for the improvement of well-being is a global problem. This probably reflects, in part, the dissatisfaction many patients express concerning western style medical practice. Larrey [6] points out that the trend in the use of herbal medicines is growing due to a belief that natural products are both good and innocuous when compared with western style medicines. De Smet [4] summarized this global problem using a series of selected case reports. In the summary he described the intake of herbal tea (contained the toxic pyrolizidine alkaloids) leading to hepatotoxicity and death. Another example was the use of azarcon in Mexico (lead tetroxide) causing severe lead poisoning with resultant seizures, encephalopathy and death in a three-year old child. Furthermore, he presented another tragic case of a woman, whom, despite repeated warnings, had continued to eat raw dried rattlesnake meat, contaminated with Salmonella Arizona. She succumbed from sepsis. He concludes that there is a strong placebo effect derived from the ritual of taking herbal medicines and this entices many to try alternative treatments. Psychosomatic complaints may benefit from this ritual and where health resources are restricted – as in South Africa - may save the State millions of rands in health costs! Chan [7] supports this view and mentions that in many developing countries, traditional methods of treatment (as opposed to the conventional western style prescription methods) are the only affordable and available forms of health care for the majority of the population. We are all aware of the substantial benefits patients have derived from the use of botanical derivatives to treat medical conditions (digitalis comes to mind immediately). The clinical results with feverfew, which has benefits as an anti-migraine agent, is but one example [4]. However, the acceptability of these plant extracts arose only after safety and efficacy was assured. An example is research conducted by the Chinese on 860

the leaves of Artemisia annua and the discovery of the anti-malarial artemisinin. While the anti-febrile effects of Artemisia annua herb have been recognized in China since the 4th century AD [8], it was only in 1972 that the research into the anti-malarial properties began. A starting point would be to assemble a catalogue of safe herbal remedies, which the traditional healers could use for their patients. Watt and Breyer-Brandwijk [9] published such a catalogue and listed the local names of numerous medicinal plants. The poisonous ones were also identified in their publication. However, ensuring that this information is easily accessible to the traditional healers is challenging. In addition to the need for a revised, updated and expanded version, the document must be written in the language of the traditional healer and be user friendly. Bye and Dutton [2] have researched the culture of the Zulu people (concentrated mainly in the KwaZulu/ Natal region) and the use of traditional remedies. The Zulu believe that disease is a reflection of disharmony between an afflicted person and his/her ancestors. The sangoma (diviner) diagnoses the problem by consulting with the spirits and thus identifies the source of the disharmony. The inyanga prepares and dispenses the herbal treatment required to dispel the disharmony and in so doing hopes to cure the affliction. Although this work concentrated on the Zulu population [2], there exists this common thread of belief throughout Africa that, during times of illness amongst the blacks, there is a need for ancestral placation. Therefore, the administration of toxic herbal substances (or chemical substances e.g. mixtures of solutions with battery acid and others with dichromates) is intrinsic to the whole African continent. The lack of good toxicological services in a large part of the African continent is a major contributor to the inability to identify most of the culprit toxins. Another major problem is that the registration of herbalists has not been uniform, which can lead to a situation where ignorant persons are dispensing substances of which they, and anyone else for that matter, know little. When herbal remedies are recommended, there are no checks and balances in the treatment protocols. Unhappily, fatalities as a result of herbal use, are a major problem in infants including the death of healthy babies. Similar to western-type medical practice, charlatans are encountered amongst the sangomas and inyangas. These quacks are usually ignorant of

39. Nephrotoxins in Africa

the safer, tried and tested traditional remedies. They may prescribe the hard-core, more toxic substances. An example of such an occurrence is a case of cresol poisoning discussed below. This chapter will attempt to outline the extent of the problem in South Africa. We have confined our comments to toxins that have a major impact on renal function – although some (if not most!) toxins have secondary effects with consequent kidney failure. We have also incorporated a description of toxins, which are taken for cosmetic and suicidal purposes, but also by accident and for recreational reasons (and therefore the sangoma and inyanga is definitely innocent in this instance).

Potassium dichromate The excellent work done by Wood et al [3] in describing the extent of the toxicity of potassium dichromate has been of great educational value. Potassium dichromate is the principal active ingredient in purgative solutions; the indications for its use are broad and may encompass any complaint. The substance is toxic in the hexavalent state but metallic chromium is inert. It is used in the leather industry for tanning, as an industrial cleaning agent and in electroplating. It is a bright yellow crystalline substance in its natural state. When taken orally or rectally it is irritant to the mucosa and can cause acute tubular necrosis, hepatitis and colitis. The toxic hexavalent chromium becomes rapidly bound to tissue (in the trivalent form). Therefore clinical measures to reduce absorption must be administered immediately in order to have an effect. When inhaled, it causes chronic bronchitis, interstitial pneumonitis and fibrosis. The indications for which dichromates may be prescribed are numerous. Figure 1 is an example of a purgative, the contents of which, on analysis was found to have a mixture of dichromates and faeces. This mixture was given to a patient to relieve mild constipation. The patient initially denied that she had consulted a sangoma but later, after she had undergone dialysis for two weeks followed by complete recovered from her acute tubular necrosis (ATN), brought in the medicine bottle and admitted that she had taken the substance orally on prescription of a healer. This had produced severe diarrhea and dehydration with subsequent ATN. Note (Figure 1) the healer has used western-style prescription methods

including standard abbreviations such as t.d.s. and p.o. Furthermore note that the indications for use of the medicine range from “blood” disorders to diarrhea to libido problems! Wood has presented several cases of known dichromate poisoning [3]. In six of the seven cases, the patients were able to produce the ingested compound for analysis and dichromate proved to be the principal active ingredient. All his patients had blood levels of chromium well in the toxic range. They all required dialysis support and in one tissue was obtained at post-mortem examination. This patient died from sepsis and massive gastrointestinal hemorrhage. It is noteworthy that the liver had the highest chromium content followed by the kidney and this was found 26 days after the initial presentation! The clinical presentation of these 7 cases was that of a spectrum involving the kidneys with GIT manifestations in all, as well as hepatic failure in two. Sepsis and hemorrhage were secondary manifestation of mucosal damage and in one case, rectal perforation (probably traumatic enema) was an additional finding.

Figure 1. Medicine bottle containing dichromates. Note indications for use. By courtesy of Prof Wood. 861

SWANEPOEL, BLOCKMAN & TALMUD

Treatment consists of early hemodialysis to remove a larger amount of chromium (due to the rapid tissue binding). There is no role for chelating agents. Otherwise, treatment for dichromate poisoning is entirely supportive. Fortunately, the renal failure is reversible. The chromium is taken up by the kidneys and produces a nephrogram on straight abdominal X-ray (Figure 2). It is intriguing to know that dichromate use is common (personal communication with a sangoma), but the disease is uncommon. Therefore there are unknown mechanisms at play which render the dichromate poison less toxic. It may be related to dose of dichromate, accompanying ingredients in the sangoma preparation or an individual’s idiosyncratic metabolism.

Cresols Transcutaneously absorbed substances may lead to severe toxic systemic effects. Substances, which fall into this category, include the phenols and the closely related compounds, cresols. The commercially available Jeyes fluid is a cresol. Jeyes fluid ingestion and administration by enema are well-documented forms of poisoning in South Africa. Berg et al. (personal communication) are the first in South Africa to describe the development of ATN following the transcutaneous absorption of a cresol. The case was a woman who may have presented to a healer (Berg cannot be sure that it was not a charlatan that was consulted) complaining of nausea. She was “painted” with a solution of Jeyes fluid. Whilst being painted she immediately lost consciousness and was taken to a nearby hospital. In addition to being deeply comatose, superficial chemical burns where noted on admission. Her vital signs were normal including a normal blood pressure. She regained consciousness after approximately 3 hours; her urine was bloody and she became oliguric. It was thought that she had aspirated and she was subsequently treated for pneumonia (confirmed on X-ray and located to the right upper lobe). Examination of the urine revealed macroscopic blood and granular casts. Her serum biochemistry was abnormal with hyperkalemia and an elevated urea and creatinine. She ultimately required dialysis but recovered sufficient renal function within 3 days to allow cessation of dialysis therapy. In view of the close association between the exposure to the Jeyes 862

fluid and the development of ATN, a renal biopsy was not performed and a causal role was assumed. It appears that cresols are absorbed across intact skin [10]. Once absorbed, phenols are widely distributed throughout the body and are toxic to various cell types. Green reports on a one-year old child who died 4 hours following the accidental application of a phenol solution to his head. At post-mortem examination, the presence of the phenols in the internal organs was detectable by the typical odor of phenol [10]. ATN was also documented histologically. Bruce et al. document 2 cases of cresol poisoning and their resultant deaths [11].

Cantharidin Is commonly known as Spanish fly and is derived from blister beetles. Karras speculates that it may be a more common cause of morbidity than is recognized [12]. It is used as a sexual stimulant and is an ingredient in some wart removal remedies [13]. Poisoning is noteworthy for its dramatic effect on the gastrointestinal and urinary tracts, as well as occasionally inducing cardiac abnormalities and seizures [12, 13]. The patient

Figure 2. Straight abdominal XRay illustrating a nephrogram (and dilated colon because of a concurrent colitis).

39. Nephrotoxins in Africa

rectal bleeding. The injury in some cases extended to involve the whole colon. Of the enema ingredients, potassium dichromate was prominent. Therefore this is a ubiquitous toxic substance used widely throughout South Africa. Other ingredients were vinegar, caustic Ox-eye daisy or Impila (Callilepsis Laureola) soda and dettol. The severe cases were complicated by renal failure. Obviously – with the exception of the Seedat, in 1978 [14], concluded that the common- dichromates, which we know are directly nephrotoxic est medical causes of acute kidney injury in the east - the cause of the renal impairment was multifactorial. Sepsis probably played a major role in the pathogenesis of South Africa were toxins. The toxins were mainly of the acute kidney injury. herbal in nature and the composition of the majority Males predominate in the group with herbal inwas unknown. However the best studied toxin from duced AKI. This observation is supported by a report this area is the impila [15]. from Zimbabwe describing the pattern of poisoning The impila (which means health in Zulu) bears a from traditional medicines in that country [20]. Apparroot similar to a sweet potato. It is harvested in winter ently men resort to taking muthi because it is perceived and stored after drying and crushing. It is boiled in as being manly and also because they have easier access water for 30 min and thereafter administered either to the sangomas than do women. rectally or orally [2]. It is a multi-purpose muthi There is an interesting report [21], which described (medicine) and is given for general health, impotence a male who presented in AKI after drinking one and HIV symptoms. It is also believed that if the root spoonful of a sangoma prescription. He had visited is buried near a person’s home, then it will intercept the sangoma with the complaint of vomiting and any evil directed towards that household [2]. abdominal cramps. He required 2 months of dialysis Impila causes massive centrilobular liver necrosis and recovery was incomplete. A sample of the ingested with hypoglycemia and liver failure. It also causes acute tubular necrosis [16, 17]. Atractyloside is a com- compound was found to contain Cape Aloe. This was a ponent in the root of the impila and it is this substance, surprise finding, since the aloe is considered to be safe. which has been demonstrated to cause acute tubular Therefore one must be cautious in interpreting this as the agent responsible for the AKI, since there may necrosis in rats [17]. have been other substances present in the concoction, Watson described 50 black children who had died following the administration of this toxin [16]. which could not be detected. An unanswered question here is what was the cause of the vomiting and cramps, Post-mortem examination was conducted in all cases which led the man to consult the sangoma in the first confirming the diagnosis of impila poisoning. No place? Did that have any deleterious, causal effect on common trend was noted in the clinical presentation renal function? of these children. It was concluded that hypoglycemia Aloes occur throughout the world. The genus and evidence of hepatic and renal dysfunction, were Aloe includes herbs, shrubs and trees. The leaves are strong indicators of impila poisoning. used for the preparation of medicine or cosmetics [9, A substantial experience of toxin induced renal 22, 23]. failure has been gained at the Chris Hani Baragwanath Hospital, Soweto. This 3000-bed teaching hospital serves approximately 4 million people from Soweto. Cape aloe Once again patients often visit traditional healers, usually prior to, or instead of consultation with a mediThis is a common species of aloe and is derived cal doctor [1, 18]. A study done at the hospital [Katz from Aloe ferox. It is reported to be the most exten- personal communication] revealed that 13% of cases sively used plant substance as an herbal remedy in of AKI were caused by herbal toxins. South Africa [9]. The aloe is also identified as one of Segal and others reported on ritual–enema-induced the most commonly used herbal propriety products colitis [19]. Their report incorporates 11 patients where [22]. It is not considered toxic. Therefore the case [21] the clinical hallmarks of the injury were peritonitis and discussed above, must have had other additives that may present with massive hematemesis and hematuria. The kidney is often involved with ATN and glomerular damage. Treatment is supportive and includes dialysis when indicated.

863

SWANEPOEL, BLOCKMAN & TALMUD

ringe has also been used [3]. The funnel-shape of the were not measurable or the “dose” may have been too cow horn makes it easy to use this form of treatment high. Or, as must always be considered, did the original disease for which help was sought from the sangoma, – however we must remember the report from Segal [19] in which the complications of rectal perforation not play a role in causing the AKI [31]? and colitis are ascribed to the instruments and methods To support its safety, Van Wyk [23] mentions the used to administer treatments rectally. The hollowed medicinal uses of the Cape aloe. The yellow juice from out reed is only used in children. Here the prescribed the leaves is dried and a small crystal (the size twice solution is first aspirated into the hollow reed and the that of a match head) of the dried substance is taken blackened tip is inserted into the rectum. The sangoma orally as a laxative. Its use as a laxative has also been will then blow through the hollow reed forcing the important from a commercial point of view. The export market has been a valuable source of revenue for SA. herbal medicine into the rectum of the child. It may also be used for arthritis, but at a much smaller dose than that required for a catharsis. Van Wyk men- Mercury tions that eczema, hypertension and stress have also been included in the list of indications for this product. Barr in 1972, discussed the nephrotic syndrome One is uncertain as to how these indications were ar- in adult Africans in Nairobi [25]. In this report he rived at and whether there is any substantive evidence showed that young, English-speaking women, with of efficacy, with the use of aloes, in the treatment of the nephrotic syndrome, were in the majority. They these conditions. were able to separate these patients from the rest by the The active purgative ingredient in the aloe is called cosmetics that they used. In fact, more specifically, by barbaloin. The barbaloin is a prodrug and once in the the habit of applying skin lightning creams. On further colon is it converted to the active substance, aloe-emo- analysis, it was found that they had used creams condin anthrone [22, 23]. The conversion to active drug taining amino-mercuric chloride. Analysis of the urine is facilitated by the colonic flora. The laxative action revealed high levels of mercury. After cessation of the results from the inhibition of colonic Na–K-ATPase mercury containing creams, the urinary mercury levels with the resultant increase in the water content of the rapidly fell to normal. This study was of interest in that colon. only 12% of the biopsies obtained were diagnosed as membranous nephropathy. The majority (50%) had minimal change disease. Senecio The mean duration of use of the creams before presentation with leg edema, was 13 months. The Rose described the toxicity of the senecio plant remission rate was 50% in those with minimal change in 1972 [24]. It is the most common plant species to contain the pyrolizidine alkaloids. Toxicity includes hepatic necrosis and later intrahepatic veno-occlusion. A major secondary component is ATN. There are over 50 species of senecio plants in the south east of South Africa. The plants are used extensively as enemas and purgatives. Rose mentions that, despite the deaths resulting from the use of these plants, the local inhabitants are not aware of the danger these plants pose to their well-being. Figure 3 is a photograph of a cow horn and segment of hollowed out reed. These objects in the photograph have actually been used in sangoma/inyanga treatment procedures. They were obtained from a sangoma practicing in Cape Town. They are the standard instruments used by the sangomas to administer the various Figure 3. Cow horn and hollow reid used by the traditional healers for the administration of herbal enemas. herbal remedies, via the rectum. The Higginson’s sy864

39. Nephrotoxins in Africa

disease after withdrawal of the creams. Human exposure is either to mercury vapor or methyl mercury compounds [26]. See also chapter 36. Both of these forms of mercury can lead to kidney involvement with nephrotic range proteinuria. The effect on the kidney is suggested to be on the basis of mercury-stimulated T lymphocytes [26]. These T lymphocytes produce damaging antibodies to the basement membrane with consequent heavy proteinuria. The damage may manifest as membranous nephropathy with the nephrotic syndrome [27, 28] or as minimal change disease [25]. Of importance, there are no case reports of nephrotoxicity resulting from exposure to mercury from amalgam tooth fillings [29]. There is no specific treatment for mercury poisoning of the kidneys but removal of the source of the metal is important. This maneuver may result in spontaneous improvement in 50% of cases [25]. Brown, in a study from Malawi, described the failure to improve in 2 out of 6 patients with membranous nephropathy who were known to have used skin creams [28]. This occurred despite removal from exposure to the mercury as well as the administration of steroids.

Paraphenylene diamine (hair dye) This dye, paraphenylenediamine, when mixed with henna, blackens the hair in a very short time. The substance is a common cause of ATN in the Sudan [30]. It is also toxic to the heart and liver. It is absorbed through the skin but individuals have ingested the dye in suicide attempts. Within 3-4 hours after ingestion they develop angioneurotic oedema soon followed by renal failure. Renal biopsy shows the typical features of acute tubular necrosis. See chapter 40.

Sodium bromate This compound is a constituent of hair waving/curling lotions used in the hairdressing salon for perms. A male patient was admitted to our institution with a history of the intentional ingestion of the “American Look” hair preparation. It contained sodium bromate. He presented 2 days after ingestion with vomiting and deafness. There was no accompanying history of renal impairment, no past history of hypertension and he was not on regular medications. Progressive oliguric renal failure ensued, necessitating hemodialysis. His Hb

fell from 7.9 to 2.4 over the course of 5 days and hemolysis was diagnosed. The clinical features in keeping with bromate toxicity were rapid onset of sensorineural deafness and acute kidney injury with progression to anuria [31]. The hemolysis stopped spontaneously and he eventually recovered renal function but remained totally deaf. He did not develop a peripheral neuropathy which is another toxic effect of bromate, as is hemolysis. The mechanisms are unknown. Bromate is rapidly absorbed from the GIT and within 15 minutes maximum plasma levels are achieved. It is converted to bromide in liver and kidney by glutathione and excreted as bromate and predominantly bromide in urine. The potential toxicity is related to direct renal, cochlea and haematological damage. The exact mechanism of toxicity is unsure but is thought to be due to free radical formation. Ototoxicity, leading to sensorineural deafness may occur within 4 – 16 hrs of exposure [32]. Kidney histology shows epithelial separation in the proximal tubules under light microscopy. Electron microscopy confirms this separation but shows an intact basement membrane, in keeping with tubular necrosis [33]. The glomeruli are unaffected. There has been a report suggesting that rapid removal of bromates, by hemodialysis, prevents the occurrence of irreversible hearing loss and post ATN renal dysfunction [34].

Copper Sulphate Copper sulphate is a readily available chemical substance in Africa. It is widely used commercially in painting and the leather industry. The containers in which it is sold, clearly state “poisonous”. The bright blue colour makes it attractive for children to want to taste and adults to ingest, either following on from an inyanga preparation or for suicidal purposes. There are reports of a suicide attempt using copper sulphate intravenously [35, 36]. The biggest reported series of poisoning comes from India [37, 38]. In Africa anecdotes are commonly mentioned but not substantiated in published reports. The toxic effects on the kidneys occur via the induction of acute intravascular hemolysis with the development of acute tubular necrosis. One of the cases mentioned above, who administered the copper sulphate intravenously, had a kidney biopsy 8 weeks following on from presentation and this showed 865

SWANEPOEL, BLOCKMAN & TALMUD

chronic tubulo-interstitial nephritis [36]. Hemodialysis is ineffective in removing copper as it rapidly enters red blood cells (it is also taken up by the liver where it is incorporated into ceruloplasmin). This copper-protein structure circulates until eventually metabolized and is excreted in bile. As for Wilson’s disease, chelation treatment is recommended; Oldenquist and Salem (35) successfully used EDTA infusions. Takeda and his group [39] reported the successful use of chelation therapy (dimercaprol and penicillamine) together with hemoperfusion and hemodiafiltration in a patient with cupric sulphate intoxication.

Paraquat This is a dipyridilium compound and is used as a herbicide. The common methods of paraquat poisoning are either accidental or with suicide intent. Mortality rates are high. The incidence however has decreased in South Africa when compared with the early 1980s (personal observation). It was particularly prevalent in the farming communities. Accidental poisoning resulting from drinking the solution from a softdrink bottle, in which the farmers had stored the poison, was not unusual. The major toxic effect is on the lungs. The poison causes pulmonary oedema initially, followed by the rapid (within days) development of pulmonary fibrosis, respiratory failure and in most cases by this stage, death [40]. Renal failure usually ensues due to tubular necrosis (direct nephrotoxic effect and from shock and superimposed sepsis). There have been reports of paraquat-induced Fanconi syndrome [41] with severe hypophosphataemia and tubular necrosis on biopsy. The postulated mechanism was an effect of paraquat on the sodium-phosphate transporter in the proximal tubules. The patient made a full recovery after 23 days hospitalization. Intravenous phosphate was administered. An interesting report comes from Bairaktari and others [42]. In this article mention is made of nuclear magnetic resonance spectroscopy of urine from 2 patients who ingested paraquat intentionally. The investigators were able to confirm that the paraquat damage was to the pars recta of the proximal tubule. This may explain the pathogenesis of the development of the Fanconi syndrome [41]. Whilst hemodialysis does remove paraquat, it is far less efficient than charcoal hemoperfusion, which 866

is the treatment of choice. 30% Fuller’s earth (a diatomaceous earth) must be administered orally as soon as possible to act as a sorbent of the paraquat in the gut after oral ingestion. Cathartics are administered simultaneously. The policy at our institution is to start hemoperfusion and administer Fuller’s earth as soon as the patient arrives in the hospital and not to wait for the results of paraquat blood levels. This practice followed from the knowledge that prognosis is related to the plasma paraquat level and the duration of such levels [40]. An article from Korea [43], which reported on 147 patients following paraquat ingestion, showed that the mortality rate was high at 44.2%. The authors were able to show that certain laboratory parameters could predict poor outcome. It was suggested that with the aid of these parameters, unhelpful and invasive (hemoperfusion) treatments can be avoided (and by implication, the patients are beyond help). The parameters used as prognosticators were abnormal liver enzymes, renal dysfunction, metabolic acidosis and abnormal urine analysis. They were supported in this observation when Yamaguchi et al found – after examining laboratory data from 160 paraguat poisoning patients – that renal function and acid-base imbalance were useful in judging prognosis [44].

Crystal metamphetamine (“Tik”) and methylenedioxymethamphetamine (MDMA, Ecstasy) These amphetamines are used mostly as a recreational substance in “rave” parties and clubs. Crystal methamphetamine is also known as Ice, Straws, Globes and Tik (called tik, because of a clicking sound when heated), is cheap (price ranges from R30-R60) and is easily available in the streets of Cape Town. The crystals are large and are smoked from a heated light bulb from which the base and element have been removed. Crank and Speed are the same substance but are in powder form (smaller granules than Tik). Tik is structurally related to noradrenaline. It has an indirect sympathomimetric effect; it blocks the presynaptic reuptake of dopamine and noradrenaline. The increased catecholamine activity causes intense systemic vasoconstriction and the stimulant effects are longlasting (up to 12 hours). There is an accompanying “rush

39. Nephrotoxins in Africa

both “tik” and “ecstasy” [51]. A renal biopsy confirmed ‘causing a state of high self-esteem and agitation. the presence of fibrinoid necrosis involving arterioles A 21 year old male presented to the emergency and small arteries. The lumina were occluded by room complaining of vomiting, diarrhea, headache intimal thickening. The glomeruli contained necrotisand visual disturbances. There was a strong history of recreational drug abuse and included mostly Tik, ing lesions. The hepatitis serology was negative in but also ecstasy, cocaine and cannabis. He occasion- this patient. The association with Hepatitis B and C and drug abuse (intravenous use with contaminated ally used alcohol. His blood pressure was recorded at 220/140 mmHg in both arms. The examination find- needles) is well known; Hepatitis C may give rise to a mesangiocapillary glomerulonephritis (MCGN) with ings of note were on fundoscopy and included retinal cryoglobulinaemia. Methamphetamine–induced acute infarcts, bleeds and papilloedema. ECG evidence of hypertension was present. Blood biochemistry re- interstitial nephritis has also been described [52]. Hyperpyrexia and hyponatraemia are other complications vealed normal electrolytes but an elevated urea of 59 encountered in cases of amphetamine abuse [53, 54]. mmol/L and a creatinine at a high of 2394 umol/L. He The hyponatraemia results from water intoxication or had a markedly elevated serum creatinine kinase (in keeping with rhabdomyolysis) HBV, HCV and HIV inappropriate ADH secretion. serology were negative. An ultrasound examination of his kidneys showed them to be equal and normal Ethylene glycol sized but reported that they were highly echogenic. A renal biopsy confirmed advanced interstitial fibroEthylene glycol (EG) is the constituent found in all sis and tubular fall-out. There were areas of tubular anti-freeze products. It is commonly used for suicidal necrosis and all glomeruli seen were sclerosed. The intent, or taken by mistake or drank in the belief that vessels showed hypertensive changes. A CT scan of it is alcohol. Ethylene glycol is toxic to the kidneys, the brain found multiple low densities in both grey producing acute tubular necrosis and nephrocalcinosis. and white matter in keeping with hypertensive dam- The early impressions were that the metabolic product age. His blood pressure was very easily controlled on – oxalate – was responsible for the renal impairment. two agents and dialysis. The patient became dialysis However work on the proximal tubular segments of dependent. The presentation could have been that of the mouse kidney, has shown that glycoaldehyde and end-stage glomerulonephritis; however there was no glyoxylate are responsible for EG nephrotoxicity [55]. preceding history or report of a clinical examination This pathogenetic mechanism is via the depletion of prior to his admission to confirm this. The ease of blood ATP. Studies from Prague have reiterated the known pressure control once the substance abuse was halted, fact that concurrent ethanol use with the EG – or the leads one to believe that the hypertension was possibly administration of ethanol soon after EG intake – imconsequent on Tik abuse. He additionally used ecstasy, proves the outcome [56-57]. Recovery is expected and which may have played a synergistic role, with Tik supportive dialysis is recommended when necessary. inhalation, in the pathogenesis of the hypertension However the lethal blood level of EG is 2 g/L, resulting and kidney disease. in multi-organ failure [57]. Steenkamp and Stewart have published an excellent article, providing guidance on the use of analytic Recreational drugs cause a spectrum of methods, to examine the constituents of plants [58]. glomerular, interstitial and vascular diseases Some of the constituents so analysed have not been The effects of amphetamines on the kidney are mainly acute tubular necrosis on the basis of rhab- shown to produce toxicity in humans. Clearly they have the potential to produce adverse effects, but as domyolysis (with myoglobinuria) and a disseminated intravascular coagulopathy. But, malignant hyperten- we know with dichromates, administration of the toxin does not always produce adverse effects in humans. sion and the resultant effects on the kidneys, must Most of what follows is derived from this article [58]. always be a consideration in the differential diagnosis Europe was stunned with the outbreak of Chinese of renal failure [45-50]. These effects are likely to be herb nephropathy. There have been no reports from chronic and irreversible. Bingham et al reported a case of necrotising vasculopathy after the ingestion of Africa and - because of the superb epidemiological 867

SWANEPOEL, BLOCKMAN & TALMUD

work done to find the culprit and to advertise its toxicity - it may well turn out to be a toxin that does not reach Africa. Aristocholic acid (AA) is a naturally occurring carcinogen and nephrotoxin; it induces a chronic relentlessly progressive tubulo-interstitial nephritis ( [59]. The toxic effects caused a number of Belgian women to become dialysis dependent. They took a Chinese herb, contaminated by AA, in the belief that the herb was a safe slimming agent. Steroids slow the progressive nature of the disease [60]. Yams (make up the everyday diet for many in Africa. It is therefore surprising to learn that they contain a toxic substance called dioscorine(, which has convulsive properties and causes hepatic and renal failure [61, 62]. The toxicity is related to the incomplete preparation of the food. Khat leaf ( (Catha Edulis) ( chewing is common in East Africa and the Yemen. It produces renal toxicity in rabbits [63]. The leaves contain S-Cathinone (which is metabolised to norephidrine and norpseudoephidrine [64]. No reports of human cases of nephrotoxic adverse effects can be found.

Violet tree Violet tree ( (Wild Wisteria) (has produced poisoning in the Congo. The roots of this plant when taken orally or intravaginally, in search for a cure for dysmenorrhoea, can produce renal ischaemia and death [9] A strychnine-(like substance has been found in the root; additionally a high concentration of methyl salicylate (is found in the oils from the root [9, 58].

Discussion The study and identification of all the herbal medicines in Africa will be an important contribution to the well-being of the majority of the population. Collaboration with the sangomas is ongoing but limits in the financial capabilities of organizations make it a slow process of accumulating information. The benefits from such collaboration are not limited to defining a therapeutic option, but also will provide education on the culture behind the sangoma/inyanga influence. The existence of “secret” formulations handed

868

down in families of sangomas/inyangas is still another stumbling block in the analysis of traditional medicines. The belief exists among these traditional healers, that once the “secret” is known, the “medicine” will then lose its power. This must be counteracted by evidence of efficacy. There is much distrust of traditional healers. This is perhaps justified in some instances; it is particularly relevant when no formal register of traditional healers is in place. Traditional healers are therefore not held accountable. In turn, Savage and Hutchings [65] point out the failings of the western-style doctors. Many symptoms are unfairly ascribed to the herbal treatment administered. The original disease for which the aid of the sangoma is sought, is often ignored. This situation was described as “an aloof attitude of mild contempt” by Savage and Hutchings [65]. Only once trust is established will the necessary knowledge be made available for the benefit of all those who prescribe medicines. Steenkamp and Stewart provide guidance for analysis of suspected toxins [58]. Unfortunately, the costs of such analyses will not be borne by the South African government and only privately-funded or/and university endeavours will enable any meaningful investigation into the ingredients of presumed toxic concoctions. Ongoing education against the dangers of substance abuse must continue at school level and in the home. The widespread use of Tik and others has lead to the situation where any individual, admitted to the emergency room, with severe hypertension, must be questioned about the use of amphetamines. Industrial chemicals must be safeguarded from the general public and information – on the dangers of these substances – advertised in the media for the population to be adequately informed.

Acknowledgements We are indebted to the contribution made by the co-authors of this chapter in the 2nd Ed of Clinical Nephrotoxins. This edition has built on their contribution and the work has been enlarged and updated.

39. Nephrotoxins in Africa

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

Joubert PH. Poisoning admissions of Black South Africans. Clin Toxicology 1990; 28(1): 85-94. Bye SN, Dutton MF. The inappropriate use of traditional medicines in South Africa. J Ethnopharmacol 1991; 34: 253-259. Wood R, Mills PB, Knobel GJ, Hurlow WE, Stokol JM. Acute dichromate poisoning after use of traditional purgatives. A report of seven cases. S Afr Med J 1990; 77: 640-642. De Smet PAGM. Is there any danger in using traditional remedies? J Ethnopharmacol 1991; 32: 43-50. Perharic L, Shaw D, Colbridge M, House I, Leon C, Murray V. Toxicological problems resulting from exposure to traditional remedies and food supplements. Drugs Safety 1994; 11(4): 284-294. Larrey D. Hepatotoxicity of herbal remedies. J of Hepatol 1997; S26: 47-51. Chan TYK. Monitoring the safety of herbal medicines. Drug Safety 1997; 17(4): 209-214. De Smet PAGM. The role of plant-derived drugs and herbal medicines in healthcare. Drugs 1997; 54(6): 801-840. Watt JM, Breyer-Brandwijk MG. The medicinal and poisonous plants of Southern and Eastern Africa. Livingstone, Edinburgh, 1962. Green MA. A household remedy misused – fatal cresol poisoning following cutaneous absorption (a case report). Med Sci Law 1975; 15(1): 65-66. Bruce AM, Smith H, Watson AA. Cresol poisoning. Med Sci Law 1976; 16(3): 171-176. Karras DJ, Farrell SE, Harrigan RA, Henretig FM, Gealt L. poisoning from “Spanish fly” (cantharidin) 1996; 14(5): 478-483. Till JS, Majmudar BN. Canthanridin poisoning. South Med J. 1981; 74(4): 444-447. Seedat YK. Acute renal failure among blacks and Indians in South Africa. S Afr Med J 1978; 54: 427-431. Bhoola KDN. A clinicopathological and biochemical study of the toxicity of Callilepsis laureola (impila). MD thesis. University of Natal 1983. Watson AR, Coovadia HM, Bhoola KD. The clinical syndrome of Impila (Callilepis laureola) poisoning in children. S Afr Med J 1979; 55: 290-293. Wainwright J, Schonland MM, Candy HA. Toxicity of Callilepsis laureola. S Afr Med J 1977; 52: 313-315. Buchanan N, Cane RD. Poisoning associated with witchdoctor attendance. S Afr Med J 1976; 50: 1138-1140. Segal I, Tim LO, Hamilton DG, Lawson HH, Solomon A, Kalk F, Cooke SAR. Ritual –enema-induced colitis. Dis Col Rect 1979; 22(5): 195-199. Kasilo OMJ and Nhachi CFB. The pattern of poisoning from traditional medicines in urban Zimbabwe. S Afr Med J 1992; 82: 187188. Luyckx VA, Ballantine R, Claeys M, Cuyckens F, Van den Heuvel H, Cimanga RK, Vlietinck AJ, De Broe ME, Katz IJ. Herbal remedy - associated acute renal failure secondary to Cape aloes (Case report). Am J Kidney Dis 2002; 39(3): E13 (www.ajkd.org). Iwu MM. Handbook of African Medicinal Plants. CRC press 1993. Van Wyk B-E, Van Oudtshoorn B, Gericke N. Medicinal plans of South Africa, 1st edition. Briza publications 1997. Rose EF. Senecio species: toxic plants used as food and medicines in the Transkei. S Afr Med J 1972;1039-1043. Barr RD, Rees PH, Cordy PE, Kungu A, Woodger BA, Cameron HM. Nephrotic syndrome in adult Africans in Nairobi. B Med J 1972; 2: 131-134. Clarkson TW. Mercury-an element of mystery. N Engl Med J 1990; 323(16): 1137. Oliviera DBG, Foster G, Savill J, Syme PD, Taylor A. Membranous nephropathy caused by mercury-containing skin lightening cream. Postgr Med J 1987; 63: 303-304. Brown KGE, Abrahams C, Meyers AM. The nephrotic syndrome in Malawian Blacks. S Afr Med J 1977; 55: 275-278. Sandborgh-Englund G, Nygren AT, Ekstrand L, Elinder CG. No evidence of renal toxicity from amalgam fillings. Am J Physiol 1996; 271(4 pt 2): R941-R945. Suleiman SM, Homeida M, Aboud OI. Paraphenylenediamine induced acute tubular necrosis following hair-dye ingestion. Hum Toxicol 1983; 2(4): 633-635. Sashiyama H, Irie Y, Ohtake Y, Nakajima K, Yoshida H, Sakai T, Okuda K. Acute renal failure and hearing loss die to sodium bromate poisoning: a case report and review of the literature. Clin Nephrol 2002; 58(6): 455-457. Campbell KC. Bromate-induced ototoxicity. Toxicology 2006; 221(2-3):205 – 211 Watanabe T, Abe T, Satoh M, Oda Y, Takada T, Yanagihara T. Two children with bromate intoxication due to ingestion of the second preparation for permanent hair waving. Acta Paediatr Jpn 1992;34(6):601-605. Uchida HA, Sugiyama H, Kanehisa S, Harada K, Fujiwara K, Ono T, Yamakido M, Makino H. An elderly patient with severe acute renal failure due to sodium bromate intoxication. Intern Med 2006; 45(3):151-154. Oldenquist G and Salem M. Parenteral copper sulphate poisoning causing acute renal failure. Nephrol Dial Transplant 1999;14:441443

869

SWANEPOEL, BLOCKMAN & TALMUD

36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65.

870

Bhowmik D, Mathur R, Bhargava Y, Dinda AK, Agarwal SK, Tiwari SC, Das SC. Chronic interstitial nephritis following parenteral copper sulphate poisoning. Ren Fail 2001;23(5):731-735 Chugh KS, Sharma BK, Singhal PC, Das KC, Datta BN. Acute renal failure following copper sulphate intoxication. Postgrad Med J 1977;53(615):18-23. Chugh KS, Singal PC, Sharma BK, Makakur AC, Pal Y, Datta BN, Das KC. Acute renal failure due to intravascular hemolysis in the North Indian patients. Am J Med Sci 1977; 274(2):139-146. Takeda T, Yukioka T, Shimazaki S. Cupric sulphate intoxication with rhabdomyolysis, treated with chelating agents and blood purification. Intern Med 2000;39(3):253-255. Proudfoot AT, Stewart MS, Levitt T, Widdop B. Paraquat poisoning: significance of plasma-paraquat concentrations. Lancet 1979;2:330-332. Gil HW, Yang JO, Lee EY, Hong SY. Case report. Paraquat-induced Fanconi syndrome.Nephrology 2005;10:430-432. Bairaktari E, Katapodis K, Siamopoulos KC, Tsolas O. Paraquat-induced renal injury studied by 1H nuclear magnetic resonance spectroscopy of urine. Clinical Chemistry 1998;44(6):1256-1261. Hong S-Y, Yang D-H, Hwang K-Y. Associations between laboratory parameters and outcome of paraquat poisoning. Toxicology Letters 2000;118:53-59 (www.elsevier.com/locate/toxlet) Yamaguchi H, Shigehito S, Watanabe S, Naito H. Pre-embarkment prognostication for acute paraquat poisoning. Hum Exp Toxicol 1990;9:381-384. Karch SB, Stephens BG, Ho CH. Methamphetamine-related deaths in San Francisco: demographic, pathologic, and toxicologic profiles. J Forensic Sci 1999;44:359–368. Ruttenber AJ, McAnally HB, Wetli CV. Cocaine-associated rhabdomyolysis and excited delirium: different stages of the same syndrome. Am J Forensic Med Pathol 1999;20:120–127. Richards JR. Rhabdomyolysis and drugs of abuse. J Emerg Med 2000;19:51–56. Kendrick WC, Hull AR, Knochel JP. Rhabdomyolysis and shock after intravenous amphetamine administration. Ann Intern Med 1977;86:381–387. Richards JR, Johnson EB, Stark RW, Derlet RW. Meth-amphetamine abuse and rhabdomyolysis in the ED: a 5-year study. Am J Emerg Med 1999;17:681–68 Fahal IH, Sallomi DF, Yaqoob M, Bell GM. Acute renal failure after ecstasy. BMJ 1992; 305(6844):29. Bingham C, Beaman M, Nicholls AJ, Anthony PP. Necrotising renal vasculopathy resulting in chronic renal failure after ingestion of methampethamine and 3, 4- methylenedioxymethamphetamine (‘ecstasy’). Nephrol Dial Transplant 1998;13:2654-2655. Foley RJ, Kapatkin K, Verani R, Weinman EJ. Amphetamine-induced acute renal failure. South Med J 1984;77:258-259. White SR. Amphetamine toxicity. Semin Respir Crit Care Med 2002;23(1):27-36. Milroy CM, Clark JC, Forrest AR. Pathology of deaths associated with “ecstasy” and “eve” misuse. J Clin Pathol 1996;49(2):149153. Poldelsky V, Johnson A, Wright S, Rosa VD, Zager RA. Ethylene glycol-mediated tubular injury; identification of critical metabolites in injury pathways. Am J Kidney Dis 2001;38(2):339-348. Krenova M, Pelclova D. Course of intoxications due to concurrent ethylene glycol and ethanolic ingestion. Przegl Lek. 2005;62(6):508-510 Krenova M, Pelclova D, Navratil T, Merta M. Experiences of the Czech toxicological information centre with ethylene glycol poisoning. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2005;149(2):473-475. Steenkamp V and Stewart MJ. Nephrotoxicity associated with exposure to plant toxins, with particular reference to Africa. Ther Drug Monit 2005;27(8):270-277. Vankerweghem JL, Depierreux M, Tielemans C, et al. Rapidly progressive interstitial renal fibrosis in young women; association with slimming regimen including Chinese herbs. Lancet 1993;41:387-391. Vanherweghem JL, Abromowicz D, Tielemans C, Depierreux M. Effects of steroids on the progression of renal failure in chronic interstitial renal fibrosis: a pilot study in Chinese herbs nephropathy. Am J Kidney Dis 1996;27:209-215. Neuwinger HD. African ethnobotany. Poisons and drugs. London:Chapman and Hall 1996. Bevan CWL and Hirst J. A convulsant alkaloid of Discorea Dumetorum. 1958;25:103. Al-Mamary A, Al-Habori M, Al-Aghbari AM. Investigation into the toxicological effects of Catha Edulis: a short term study in animals. Phytother.Res. 2002;16:127-132. Brenneisen R, Geisshuslers S, Schorno X. Metabolism of cathinone to (-) norephedrine and (-) norpseudoephedrine. J Pharm Pharmacol 1986;38:298-300. Savage A, Hutchings A. Poisoned by herbs. Br Med J 1987; 295; 1650-1651.

40

Paraphenylene diamine hair dye poisoning Mohamed I. HAMDOUK1, Mohamed B. ABDELRAHEEM2, Ahbab A. TAHA1, Mohamed BENGHANEM3 and Marc E. DE BROE4 1Bahry

Renal Centre, Khartoum, Sudan University Hospital, University of Khartoum, Sudan 3Ibn Rochd University Hospital Center, Casablanca, Morocco 4University of Antwerp, Belgium 2Soba

Introduction ___________________________________________________________ 871 Paraphenylene diamine characteristics ______________________________________ 872 Pharmaco-toxicology ____________________________________________________ 872 Clinical presentation _____________________________________________________ 873 Epidemiology Clinical features and systemic toxicity

873 875

Diagnosis _____________________________________________________________ 876 Treatment _____________________________________________________________ 877 References _____________________________________________________________ 877

Introduction

S

ince 1883, paraphenylene diamine (PPD) has traditionally been used for dyeing (dark color) hair in Europe [1-2] as a fresh preparation mixed with hydrogen peroxide (H2O2) [3]. In Sudan PPD is used by women to color their hair and as a body dye when added to henna (Lawasonia alba). Henna on it’s own need to be applied two or three times for several hours to give the desired color (dark red or black). In contrast this can be achieved with one single application in less than one hour by adding PPD to the henna. The toxicity of PPD, when added to henna occurs through skin

absorption. This toxicity can not be attributed to henna, (L. Alba) which is a herb used for cosmetic purposes and also used in folk medicine for the treatment of some skin diseases, as an antiinflammatory, antipyretic and analgesic agent [4-5]. In Morocco, Takaout beldia indicates a non-toxic vegetable product extracted from the gallnut of Tamaris Orientalis (Figure 1). This non-toxic substance is highly appreciated by women for its hair dyeing properties. Its rarefaction resulted in the use of paraphenylene diamine as substitute under the name of Takaout Erroumia. If nowadays accidental ingestion is exceptional, its use for suicidal attempt in young women in Morocco and Tunisia is increasing [6].

HAMDOUK, ABDELRAHEEM, TAHA, BENGHANEM & DE BROE

Figure 1. Natural Takaout beldia (Morocco).

Figure 2. PPD from the local market.

For the last seventy-five years the dermatological effects of PPD are well studied and as a result of these studies the European Union restricted the concentration of PPD in hair dye formulations to a maximum of 6% [3]. Systemic intoxication is not well studied although there are many studies of its mutagenicity and carcinogenicity [7-8]. Its ingestion is responsible for a respiratory (cervico-facial edema), muscular (rhabdomyolysis) and a renal syndrome (acute kidney injury due to hypovolemia and myoglobulinuria). The respiratory syndrome mainly determines its prognosis.

Paraphenylene diamine characteristics Paraphenylene diamine (PPD) [C6H4 (NH2) 2] is an aromatic amine not found in nature. It is a derivative of paranitroanaline and it is available in the form of white crystals when pure and rapidly turns to brown when exposed to air [9]. Paraphenylene diamine has a molecular weight of 108 Dalton; its boiling point is 267°C and melting point 140°C. It is soluble in ethanol, ether, benzene, chloroform, and acetone and with agitation in water [10]. Rinne and Zinke first prepared PPD in 1874 by reducing 1,4 dinitrobenzene with tin and hydrochloric acid. Now it is produced commercially by reducing 1amino-4-nitrobenzene by (1) iron and hydrochloric acid or (2) iron, ammonium polysulphide and hydrogen or (3) iron and ferrous chloride [2]. PPD is used in a variety of industrial products. Along with its derivatives, it has important antioxidant actions – used in the manufacturing of synthetic and natural rubbers, petroleum products, cellulose ethers and alfalfa meals [2]. PPD also has commercial application as photographic developers and in a variety of antioxidants and is also used in dyeing furs and 872

for printing of cellulosic textile materials. PPD hydrochloride has been used as an analytical reagent in the testing of blood, hydrogen sulphide, amyl alcohol and milk [11]. PPD is used in hair dye formulations and can produce a variety of shades depending on the formulation. The concentration of PPD in hair dye formulation range from 0.20% in golden blond dyes to 3.75% in black hair dyes. Exact concentrations of PPD in different formulations are not known because most hair dye formulations are proprietary. For safety reasons, different occupational health authorities, in the countries where PPD is produced, have developed standards regarding the degree of air contamination. It has been stated that employees’ exposure to PPD should not exceed 0.1 mg/m2 in the working atmosphere in any eight-hour work shift of forty-hour week. This is the maximum allowable concentration in Germany, Japan and UK [9].

Pharmaco-toxicology PPD has two modes of reactions by which it has a biological effect: Oxidation: PPD gives benzoquinone imines as a result of oxidation. The imines react rapidly with the couplers (another chemical material in the formulation) and/or an oxidized PPD to produce indo dyes. The most frequent couplers are 2, 4-diaminoanisole (blue forming coupler), resorcinol (green brown), metaminophenol (magenta/brown) and 1-naphtol (purple blue color). The most commonly used oxidant is hydrogen peroxide. Free ammonia is present to promote the oxidation reaction and the pH of the mixture on the dyed area is about 9.5 [3]. Deamination: Deamination has been suggested as a

40. Paraphenylene diamine hair dye poisoning

mode of action of PPD, which results in the production of aniline, which may contribute in part to the toxic effects of the compound [12]. PPD induces one of the most severe edema both in humans and animal studies. The edema appears to be grossly specific and selectively localized in the head and neck. It was suggested that the toxic effect of the PPD might be produced by the conversion of the PPD on mucus surfaces to its oxidation product quinondimine, which is responsible for intense local irritation [13]. Some authors believed that PPD toxicity is due to some effect either on the blood colloids or on vascular permeability [15]. Also it was believed that the PPD toxicity is due to altered vascular permeability and involvement of the parasympathetic nervous system [13]. Deamination and formation of analine is claimed to be responsible in part for the toxic symptoms [12]. These different views as to the cause of PPD edema appear to be due to the fact that the exact number and nature of the oxidation products is not known [14]. At high concentrations and after a long period of exposure PPD produces cell death. This effect together with lipid peroxidation can be the cause of the production of superoxide and hydrogen peroxide by the autooxidation of PPD [15]. It was proved that at non-toxic doses, PPD induces intercellular adhesion molecule-1 (ICAM-1) expression on the keratinocytes [16]. These results were consistent with the view that oxidative stress may be an essential part of the pre-immunological phase in the induction of the allergic contact dermatitis by PPD [16]. PPD can cause methemoglobinemia by oxidation of the ferrous form (Fe2+) of hemoglobin to the ferric (Fe3+) form. Aniline, nitrobenzene, phenacetin and other nitro and amino organic compounds are powerful methemoglobin formers. From the studies of the intracutaneous sensitization of guinea pigs using PPD, hydroquinone, quinhydrone and benzoquinone it has been suggested that benzoquinone formation plays an important role in the allergic action of PPD [15]. Studies in rats demonstrate that subcutaneous administration of 3 mg of the PPD hydrochloride induces skeletal muscle lesions in the form of rhabdomyolysis with infiltration of inflammatory cells, necrosis and accumulation of neutral lipids and dilatation of sacroplasmic reticulum [17]. In rats, teratogenicity was studied by testing four

commercially available hair dye formulations containing 1, 2, 3 and 4% PPD and several aromatic amine derivatives among their constituents [18]. No abnormal foetal effects were noted, except with the formulation containing 2% PPD, which induced skeletal deformities [18]. Experimental studies in guinea pigs when dermally exposed to PPD revealed that, PPD is absorbed through the skin into the serum and excreted in the urine. There was an increase in malondialdehyde (MDA), which indicates lipid peroxidation, suggesting that increased free radical formation is responsible for the histopathologically tissue damage in the kidney, liver and skin [15]. The increase in histamine level in the blood is a sign of hypersensitive reaction associated with increased permeability of the Mast cells [15]. There were also increased activities of the cytoplasmic enzymes AST and ALT and that of tyrosinase, observed in skin following repeated exposure to PPD. This indicates a metabolic disturbance in amino acid metabolism, which may be responsible for the epidermal thickening and erythematous changes [15]. There are many reports about the dark coloration of urine after topical application of commercial hair dye formulations containing PPD. It was shown that PPD is excreted in urine after topical application [19]. It is believed that the darkening of urine was caused by oxidizing agents and was taken as evidence of the excretion of unchanged PPD [19]. It was found that the LD50 of PPD was 250 mg/kg bw in rabbits and 100 mg/kg/BW in cats. The subcutaneous LD50 was found to be 170 mg/kg bw in rats, 200 mg/kg bw in rabbits and 100 mg/kg bw in dogs. The intraperitoneal LD50 was found to be 37 mg/kg bw in rats [3]. The lethal dose for humans was estimated to be 10 grams of pure PPD [6].

Clinical presentation Epidemiology In Sudan, PPD in its pure form (90-99%) is available in the local markets and there are no restrictions for its use or trade (Figure 2). The major problem of PPD toxicity results from the ingestion of the compound accidentally, in suicidal or homicidal attempts. However, there are some reported cases of severe intoxication after topical application of 873

HAMDOUK, ABDELRAHEEM, TAHA, BENGHANEM & DE BROE

Figure 3. Intoxication due to massive topical use (Sudan).

Figure 4. Accidental PPD intoxication in a child.

the pure PPD mixed with henna or for dyeing hair [20] (Figure 3). In a recent study, PPD intoxication due to inhalation was seen in 2.7% [21]. Samples of the PPD collected from the local market were found to have a purity of 97% when analyzed [6]. A survey of suicidal attempts in Khartoum, the capital of Sudan, in the period 1987-1990 revealed a number of 264 cases, with an age range between 10 to 30 years. In 35% of these cases PPD was used [22]. In reported series of 24 patients who presented with PPD intoxication and were admitted to Om Durman hospital in Sudan within a period of 12 months, twelve patients took the PPD intentionally and eight of them died [23]. Over a period of 2 years a series of 18 cases were reported in Khartoum North Hospital and there were two babies among them aged eighteen months, 70% were suicidal attempts. The mortality rate in this series was 22% [6]. A number of 150 cases with PPD intoxication had been admitted to the renal unit in Khartoum Teaching Hospital from 1985 to 1995. Sixty percent of them developed ARF requiring dialysis [24]. Recent statistics from the ENT teaching hospital in Khartoum from 1995 to 2005 showed that the total number of patients admitted with PPD intoxication was 3159 patient with an average of 287.1 per year. The common age group affected was 15 - 24 years (52%), this was shown also in other studies [21]. There was a predominance of females 80.7%, and the majority of cases 87% were due to suicidal attempts. The average mortality rate over 10 years was 10.6% peaking up to 27% in 1995 and declining to 5.5% in 2005 which reflects better care. A 10 year review of acute PPD intoxication during

1989 to 1999 from Wad Medani (Main Gazira State hospital-Sudan), revealed 122 cases. 93.4% was due to suicidal intent and 3.3% was due to accidental and homicidal equally, 90% of the cases were females. The mortality rate was 22.1% [25]. The experience of PPD intoxication in children is even worse. A reported series of 31 Sudanese children between 1984 and 1989, all children presented with acute and severe angioneurotic oedema, 15 required tracheotomy. ARF was reported in 5 children and the mortality rate was 41% and most children died within the first 24 hours. [27]. In another report of Wad Medani teaching hospital 2.6% from their series were children and the common cause of poisoning was accidental [25] (Figure 4). Statistics from the ENT teaching hospital in Khartoum from 1995 to 2005 showed that of 3159 patient admitted with PPD intoxication 568.6 (18%) were children below the age of 14 years. In a report from the poison control centre of Morocco, PPD intoxication was reported in 43 (11.5%) children below the age of 15 years. [21] In Morocco, intoxication with PPD is a major health problem. A reported series of 171 cases of PPD poisoning admitted to the medical resuscitation service in Ibn Roshd hospital between January 1994 and October 1997. In this series, there were 5 men and 166 women, with a mean age around 26 years. Twenty four percent of the patients developed severe ARF and 55 deaths (38.7%) were observed in this study [26]. In 90% of the cases PPD was ingested in the context of a suicidal attempt. The amount ingested varied between 3 and 15 grams. Recent evolution of the problem from the Poison

874

40. Paraphenylene diamine hair dye poisoning

control centre of Morocco (1992 – 2002) reported 374 cases. There were female predominance 77%, the majority of poisoning was intentional 78.1% and the younger population 15-25 years accounted for 54.3%. The mortality rate remained high 21.1% [21] Cases with PPD poisoning were reported in the UK, France, Israel, Japan and other countries [28-32].

Clinical features and systemic toxicity Acute systemic toxicity Cases reported with systemic toxicity of PPD had shown various clinical manifestations as well as biochemical and histological changes, the intoxication represents 30% of the intensive care admittance. It was the second reason for hospitalization in the intensive care unit of the Casablanca University hospital in 1999 and the first reason for admission in the emergency unit (Portes Médicales) of Rabat University hospital in 2003. Acute poisoning by PPD ingestion is ranked amongst the most frequent causes of suicidal poisoning requiring hospitalization in Morocco [21]. Patients with acute poisoning have a characteristic presentation of painless swelling of the face and neck with bulging eyes, a swollen dry hard protruding tongue and chocolate brown colour of the urine (25). The onset of symptoms usually occurs within hours of ingestion or contact with the dye. The frequency of clinical manifestations seen in Sudanese and Moroccan experience is seen in Tables 1 and 2 [25-26].

Renal damage induced by chemicals is well known. Renal lesions associated with PPD intoxication received much attention because most of the clinical investigators reported renal failure [6, 34-35]. Experimental studies in mice exposed to PPD showed no histological changes in the kidneys [37]. However, evidence of severe nephrotoxicity has been reported in humans [25-28]. Histological changes typical of acute tubular necrosis have been also reported [39]. A case report of systemic vasculitis and crescentic glomerulonephritis has been published in patients chronically exposed to henna containing PPD [40]. In a prospective study performed in Khartoum Kidney Dialysis Centre and Sheffield Kidney Institute 19 renal biopsies out of a series of 23 patients with severe (39%), moderate (35%) and mild intoxication (26%) were studied under light microscopy. Glomerular injury observed in 94% of the biopsies in the form of hypercellularity, membranous proliferation, glomerular swelling, and capsular drop and accentuated lobular architecture [41]. Tubular lesions were found in 78.9% of the studied samples. Different epithelial necrosis is the most common lesion observed (78.9%) while tubular atrophy had been found in (15.8%) of the studied samples. Interstitial lesions were observed in 16 samples from the studied biopsies (84.2%). Focal inflammation (neutrophils and eosinophils) was the most common injury (47.3%). No vascular injury was observed in all of the studied biopsies [41].

Nephrotoxicity The kidneys are particularly vulnerable to effects of noxious agents because of their high perfusion rate.

Dermatological manifestations PPD is a top listed allergen [42-43]. It is well known to cause irritation and dermatitis when conveyed to the skin of susceptible people [12, 30]. Erythematous

Table 1. Frequency of clinical symptoms observed in 171 patients with PPD intoxication in Morocco.

Table 2. Presenting symptoms and signs in 122 patients with PPD intoxication in Sudan.

Clinical symptoms

Percentage

Oedema

94%

Acute respiratory insufficiency - Trachial intubations (72%) - Tracheostomy (21%)

56%

Signs of rhabdomyolysis

88%

Gastrointestinal symptoms (abdominal pain)

53%

Oliguric acute renal failure

32%

Presenting signs and symptoms

No. of Patients

Percentage

Angioneurotic edema and stridor

50

41%

Dark discoloration of urine

122

100%

Flaccid paraplagia

51

42%

Convulsions

5

4%

Cranial nerve palsies (Bulbar)

10

8%

Abdominal pains

20

17%

875

HAMDOUK, ABDELRAHEEM, TAHA, BENGHANEM & DE BROE

urticarial papules, plaques and target lesions (erythema multiform like eruptions) were described. These skin manifestations occur as a result of an allergic contact dermatitis, which generally manifest as an eczematous rash [44]. Cardiovascular system In many reports of PPD toxicity cardiac arrest was the main cause of death. In these cases cardiac arrest is attributed to arrhythmia (Figure 5). Most notably ventricular tachyarrhythmia including ventricular fibrillation has been the major feature of PPD cardiac toxicity [31-32]. Cases of myocardial infarction associated with cardiac rhabdomyolysis have been reported [45]. Respiratory system On admission the first clinical presentation consists mainly of edema, which is of sudden onset and localized to the cervico-facial region [33]. Dyspnoea, tachypnoea and asphyxia with chest pain following acute PPD poisoning have been reported in a number of studies [6] [12] [28]. PPD was proved to be the cause of asthmatic attacks in the sensitive individuals [46]. A case of Goodpasture’s syndrome was reported to be induced by exposure to PPD [47]. Extrinsic allergic alveolitis also has been reported [48]. Ophthalmic effects In animal study it was reported that 89% of the mice fed PPD developed lenticular changes indicating that PPD has cataractogenous effects, which are related to the duration, amount and individual sensitivity [49]. It was concluded that PPD is potentially toxic to human lens. Exophthalmia and permanent blindness due to optic nerve atrophy following PPD poisoning were reported [6]. Using a patch test to determine PPD phototoxicity, it was proved that PPD could cause a phototoxic reaction and photoallergy [50- 51]. Hepatotoxicity Subacute toxic hepatitis due to PPD poisoning was early reported with post-mortem small fibrosed liver and fibrous adhesions [33].There was a case report of liver enlargement with progressive neurological symptoms followed by death [52].On other series tender palpable livers were found,lobular inflammation of variable degree,and mild in sinusoids and portal tract inflammation, individual cell necrosis and inflamma876

tory cells around and a granulomatous appearance in sinusoids[41].Radiological examination showed congested liver with prominent hepatic veins and low echogenicity[41]. The histopathological findings of the livers of the sacrificed animals showed signs of focal and early degenerative changes in hepatocytes, along with mild fatty changes. There was a moderate congestion of sinusoids and a focal granulomatous reaction with occasional Langerhans type giant cells [17].. In contrast others reported that there were no hepatic changes seen in their patients [6]. Neuromuscular toxicity In animal studies it was proved that PPD has a toxic effect on the parasympathetic nerves [14]. In humans, neurotoxicity consist of mental status alterations ranging from drowsiness to coma were reported. Also flaccid paraparesis has been published [6] [25]. Foot drop, palatopharyngeal and laryngeal paralysis were also reported (25). Rhabdomyolysis following intoxication with PPD has been reported [28-30]. Skeletal muscle biopsy of patients showed scattered coagulation necrosis and inflammatory cellular infiltration [30]. Psychiatric The psychiatric manifestations were studied in 50 patients with PDD poisoning in Wad Medani teaching hospital, 34 (70%) patients showed significant psychiatric manifestations, 24 of them showed depressive disorder while 11 showed conversion disorder. The patients who completed suicide and 62.1% who attempted suicide were from the depressive group (53) Chronic systemic toxicity Repeated and prolonged exposure to PPD is believed to increase the risk of non-Hodgkin’s lymphomas and multiple myeloma and cancer of the bladder [34–35]. Hair dye formulations containing PPD was incriminated in the increased risk of systemic lupus erythematosus (SLE) and breast cancer; however other studies, showed that there is no significant relationship [36-37]. Aplastic anemia due to PPD exposure also has been reported [38].

Diagnosis Determination of PPD has a great value in the diag-

40. Paraphenylene diamine hair dye poisoning

Figure 56. Severe PPD intoxication in a female showing tongue swelling, neck swelling and tracheostomy.

Figure 5. Complete heart block (Pace Maker) following PPD intoxication.

nosis; follow up of the treatment and also for medicolegal purposes. PPD was detected first in the urine of experimental animals [19] and in urine of humans by Yagi and colleagues, using thin layer chromatography [6]. Determination of PPD in the serum is not mentioned in the literature.

Treatment There is no specific antidote for the PPD. The early challenge threatening the patient’s life is asphyxia due to edema of the upper respiratory tract and the airways. Tracheostomy is a life saving measurement in this condition [6] (Figure 6). Nasotracheal intubation was proven also to be effective [29]. Vascular refilling is installed promptly in order to

prevent as much as possible the development of acute kidney injury. Acute kidney injury (ARF) was found to be the second life threatening effect. Hemodialysis had been used as a method of treatment with variable success [27-29]. On the other hand, peritoneal dialysis was used in the treatment of the ARF due to PPD toxicity in other reports [39]. Symptoms related to PPD poisoning seem to be due to histamine release; the use of antihistamines was suggested [54]. Intensive medical treatment by steroids and chlorpheneramine maleate was given to all patients together with prophylactic penicillin in one report [6]. Pethidine was given for relief of muscle pain in another report [30]. Numerous questions concerning PPD remain: physiopathological mechanisms of neurological myocardial and renal damage induced by the toxin, availability of an antidote and the extraction by hemodialysis/hemoperfusion.

References 1. 2. 3. 4. 5. 6. 7.

Proger B, Jacobson P, Shmidt P and Stern D. Beilsteins handbouch der organischen chemie. In: IARC monograph on evaluation of the carcinogenic risk of chemical to man. 1977; 16: 127. Thirtle JR. Phenylene diamines and toluen diamines. In: Encyclopedia of chemical technology (2nd edition, volume 15). Kirk RE, Othmer OE (editors). John Wiley and Sons, New York 1968, p. 216-224. Burnett C, Goldenthal EI, Harris SB, Wazeter FX, Strausburg J, Kapp R, Voelker R. Teratology and percutaneous toxicity studies on hair dyes. J Toxicol Environ Health 1976; 1(6):1027-1040. Gupta S, Gupta SS. Anti-inflammatory and antihyaluronidase activity of (Lwasonia alba). Indian Pharmacol 1985; 10: 90-91. Singh S, Modi NT, Saifi AGQ. Anti-inflammatory activity of (Lwasonia inermis). Curr Sci1982; 51: 470-471. Yagi H, El hindi AM, Khalil SI. Acute poisoning from hair dye. East African J 1991; 68: 404-411. Searl CE, Havenden DG, Venih S, Gyde OHB. Carcinogenicity and mutagenicity test of some hair colourents and consituents. Nature 1975; 225: 506-509.

877

HAMDOUK, ABDELRAHEEM, TAHA, BENGHANEM & DE BROE

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

878

Nishioka H. The examination of cancer causing property of pigmented cosmetics through microbiol test. Fourth meeting of Environmental Mutagen Society of Japan, Kyoto 1975, September. 27. IARC Working Group 1987. IARC Monographs on the evaluation of the carcinogenic risk of chemical to man. IARC publication. Grasseli JG (editor). CRC Atlas of Spectral Data and Physical Constants for Organic Compounds. Chemcial Rubger Co., Cleveland, Ohio 1973, p. B-223. Windholz M, Budavaria S. Merck index 1983. Published by Merck and Co. Inc. Rathway J. USA, p. 214-227. Nott HW. Systemic poisoning by hair dye Brit Med J 1924; 1: 421-422. Tainter ML, Hanzlik PJ. The mechanism of edema productionby paraphenylene diamine. J Pharmacol Exp Ther 1924; 24: 179211. Tainter ML, James M, Vandeventer W. Comparative edemic actions of ortho meta and para-phenylene diamines in different species. Arch int Pharmacodyn 1929; 36: 152-162. Mathur AK, Gupta BN, Muther N, Singh A, Shukla IJ, Ravi Shanker. Biochemical and histopathological changes following dermal exposure to paraphenylene diamine in guinea-pigs. J Applied Toxicology 1990; 10: 383-386. Picardo M, Zompella C, Marchese C, De-Luca C, Faggioni A, Schmidt RJ, Santucci B. Paraphenylene diamine a control allergen, induces oxidative stress and ICAM-1 expression in human keratinocytes. Brit J Dermatol 1992; 126: 450-455. Mascres C, Jasmin G. Alterations de la fibre musculaire induites chez la rat par la p-phenylene diamine. Path Biol (Paris) 1975; 23: 193-199. Burnett C, Loehr R, Corbett J. Dominant lethal mutagenicity study on hair dyes. J Toxicol Envirom Hlth 1977; 2: 657-662. Kiese M, Raucher E. The absorption of P-toluene diamine through human skin hair dye. Toxicol Appl Pharmacol1968; 13: 325327. D’Arcy PF. Fatalities with the use of a henna dye. Pharmacy international 1982; 3: 217-218. Ayoub Filali, Ilham Semlali, Valeria Ottaviano, Canmelo Fumari, Danillo Corradini, Rachida Soulaymani. A retrospective study of Acute systemic poisoning of paraphenylene Diamine (Occidental Takawt). Afr. J. Trad. CAM (2006) 3 (1): 142 – 149 Abdelkarim EE, Ali HM, Harron DWG and Ali KhM. Suicide attempt with Paraphenylene diamine (PPDA) dye in Sudan. Health Services Journal of the Eastern Mediterranean Region. WHO 1992; 6, 2: 44-48. El-Ansary EH, Ahmed MEK, Adam SE. Systemic toxicity of paraphenylene diamine. Lancet 1983; 1: 1341-1342. Salma MS, Mirghani F, Mohamed EN, et al. Poisoning with hair dye containing Paraphenylene diamine: ten years experience. Saudi J Kidney Dis Transplant 1995; 6: 286-289. Othman Mohamed Mustafa. Acute Poisoning with Hair dye Containing Paraphenylene Diamine – The Gazira Experience. Journal of the Arab Board of Specialization 2001: 3 (2) Charra B, Menebhil L, Bensalama A, Mottaoukkil S. Systemic toxicity of paraphenylene diamine. Works of the Sixth Congress of the Arab Society of Nephrology and Renal Transplantation. February 21-24, 2000- Marrakech, Morocco. M. Sir Hashim, Y.O.Hamza, B.Yahia, F.M.Khogali, G.I.Sulieman. Poisoning from henna dye and para-phenylenediamine mixtures in children in Khartoum. Annals of Tropical Paediatrics 1992: 12 3-6 Averbukh Z, Modai D, Leonov Y, Weissgarten J, Lewinsohn G, Fucs L, Golik A, Rosenmann E. Rhabdomyolysis and acute renal failure induced by paraphenylenediamine. Hum Toxicol 1989; 8(5): 345-348. Bourquia A, Jabrane AJ, Ramdani B, Zaid D. Systemic toxicity of paraphenylene diamine 4 cases. Press Med 1988; 17: 17981800. Saito K, Murai T, Yabe K, Hara M, Watanabe H, Hurkawa T. Rhabdomyolisis due to paraphenylene diamine (hair dye). Report of an autopsy case. Nippon-Hoigaku-Zasshi 1990; 44: 469-474. Lifshits M, Yagoubsky P, Sofer S. Fatal paraphenylene diamine (hair dye) intoxication in child resembling Ludwig’s angina. J Toxicol Clin Toxicol 1993; 31: 653-656. Ashraf W, Dawling S, Farrow LJ. Systemic paraphenylene diamine (PPD) poisoning a case report and review. Hum Exp Toxicol 1994; 13: 167-170. Israels MGG, Manch MB, William S. Systemic poisoning by paraphenylene diamine with report of fatal case. Lancet 1934; 1: 505510. Pearce N, Bethwaite P. Increasing incidence of non-Hodgkin’s lymphoma: occupational and environmental factors. Cancer Res 1992; 52: 5496-5500. Brown LM, Evert GD, Burmeister LF, Blair A. Hair dye use and multiple myeloma in white men. Am J Public Health 1992; 82: 16731674. Petri M, Allbritton J. Hair product use in systemic lupus erythematosus. A case control study. Arthritis Rheum 1992; 35: 625629. Koening KL, Pasternack BS, Shore RE, Strax P. Hair dye use and breast cancer: a case control study among screening participants. Am J Epidemiol 1991; 133: 985-995.

40. Paraphenylene diamine hair dye poisoning

38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54.

Hopkins JE, Manoharan A. Severe aplastic anaemia following the use of hair dye: report of two cases and review of literature. Postgrad Med J 1985; 61: 1003-1005. Suliman SM, Homeida M, Aboud OI. Paraphenylene diamine induced acute tubular necrosis following hair dye ingestion. Human Toxicol 1983; 2: 633-635. Brown JH, Conway B, Hill CM. Chronic renal failure associated with topical application of paraphenylene diamine. Brit Med J 1987; 294: 155-156. Hamdouk M. PPD nephrotoxicity. In: Paraphenylene diamine (Hair Dye) Acute Systemic toxicity. Thesis for MMed Sci in Nephrology, Sheffield Kidney Institute, UK. 2001, p. 34-47. Burnett C, Corbett JF. The Chemistry and Toxicology of Hair Dyes. Academic press, New York 1977, p. 203-224. Seidnari S, Di-Nardo A, Motolese A, Pincelli C. Erythema multiforme associated with contact sensitization description of six clinical cases. J Ital Dermato Venerol 1990; 125: 35-40. Nethercott JR, Macpherson M, Choi BC, Nixon B. Contact dermatitis in hairdressers. Contact Dermatitis 1986; 14: 73-79. Ababou A, Ababou K, Mosadik A, Lazreq C, Sbihi A. Myocardial rhabdomyolysis following paraphenylene diamine poisoning. Ann Fr Anesth Reanim 2000; 19: 105-107. Bardana EJ, Andrach RH. Occupotional asthma secondary to low molecular weight agents used in the plastic and resin industries. Eur J Respr Dis 1983; 64: 241-251. Bernis P, Homles J, Quoidbach A, Mahier PH, Bouvy P. Remission of Goodpasture’s Syndrome after withdrawal of an unusual toxic. Clinical Nephrol 1985; 23: 312-317. Wang GH, Jiang XF, Luo WC. Report of two cases of extrinsic allergic alveolitis and review of the relevant literature. Chung Hua Chich Chih 1994; 1: 376-8, 384. Jain IS, Jain GC, Kaul RL, Dhir SP. Catractogenous effect of hair dye. A clinical and experimental study. Ann Ophtalmol 1979; 11: 686-687. Kroons S. Standard photopatch testing with waxter, para-aminobenzonic acid, potassium chromate and balsam of Peru. Contact Dermatitis 1983; 9: 5-9. Horio T. Allergic and photoalergic dermatitis from diphenhydramine. Arch Dermatol 1976; 112: 1124-1126. Davison C. Paraphenylene diamine with changes in the central nervous system. Arch Neural psych 1943, 49: 254-56 Diaa E. El Gaili, Osman M. Mustafa. Psychiatric Manifestations in Hair Dye poisoning. Journal of the Arab Board of Medical Specializations 2001; 3: 2 Gleason MN, Glosselin RE, Hodge HC. Clinical Toxicology of commercial Products. Williams and Wilkins Co., 1963, p. 93.

879

41

Trace metal disturbances in end-stage renal failure patients Patrick C. D’HAESE Laboratory of Pathophysiology, University of Antwerp, Belgium

Introduction ___________________________________________________________ 883 Sources and mechanisms of trace element disturbances in uremia _______________ 883 Trace metal accumulation/toxicity in uremic and dialysis patients ________________ 886 References _____________________________________________________________ 890

Introduction

T

he association between metal exposure and renal failure can be approached from two points of view. On the one hand environmental/industrial exposure to heavy metals, more particularly, lead, cadmium and mercury and other inorganic substances such as silicon has been linked to a reduced renal function and/or the development of acute or chronic renal failure [1]. This issue has been dealt with in other chapters of this book. On the other hand patients with chronic renal failure, especially those treated by dialysis are at an increased risk for trace element disturbances (Figure 1). Indeed in these subjects the reduced renal function, the presence of proteinuria, metabolic alterations associated with renal insufficiency, the dialysis treatment, medication etc. all may contribute to either accumulation or deficiency of trace metals. With regard to aluminum intensive research on the element’s toxic effects has been performed in the past. Recently, new metal-containing medications have been introduced of which the potential toxic effects should be considered and put in a justified context.

Sources and mechanisms of trace element disturbances in uremia Trace metal disturbances may be due to the uremia per se. Indeed, as the urinary excretion route is an important pathway of elimination of many trace elements, i.e. silicon, strontium, aluminum, ... impairment of the kidney will be an important determinant of their accumulation, whilst in the presence of a reabsorptive defect a number of trace elements, especially those that are reabsorbed because of their essential role, be lost resulting in a deficient state. The presence of proteinuria may reasonably result in losses of protein bound elements. It has also been shown also that residual renal function may importantly alter the accumulation and hence toxic effects of aluminum [2]. In uremia translocation of a particular metal from one tissue to another may also occur. As an example, under normal circumstances the kidney is an important target organ for cadmium. In chronic renal failure however, possibly as a consequence of a reduction in binding proteins (e.g. metallothionein), the concentration of cadmium in this tissue decreases to extremely low levels which

D’HAESE

however, goes along with an increased concentration in other tissues such as the liver because of the failure of the diseased kidney to excrete the element [3]. Since the biosynthesis of the active 1α,25-(OH)2VitaminD3 compound is significantly reduced in renal failure, the uptake of elements which follow the calcium; i.e. vitamin D mediated pathway for their gastrointestinal absorption such as e.g. strontium [4] or lead [5] may be altered. On the other hand has an increased gastrointestinal absorption of particular elements (e.g. aluminum, lanthanum) in uremia, as compared to health, been attributed to a possible effect of atrophic alterations of the intestinal mucosa; a situation frequently observed in chronic renal failure [6]. In addition may medication related to the uremic state lead to important trace element accumulation. In the past this has clearly been established for aluminum resulting from the use of aluminum hydroxide as a phosphate binding agent. As aluminum-based phosphate binders may be contaminated with other elements, e.g. strontium the possibility for a simultaneous accumulation of different elements has been suggested [7]. Strontium is mainly eliminated by the kidney and has been associated with bone mineralization defects when present at high concentrations. In view of this the use of strontium ranelate for the treatment and preven-

tion of osteoporosis should be avoided in patients with a GFR below 30 ml/min [8,9]. Till now, evidence has been presented that in contrast to aluminum hydroxide which is mainly eliminated via the kidney, lanthanum carbonate, which recently has been introduced as an alternative phosphate binding agent, does not pose dialysis patients at an increased risk for accumulation as the element is eliminated via the bile (Figure 2) [10]. Erythropoietin when used to correct the patients’ anemia may lead to a relative iron deficiency, or indirectly to iron overload [11]. To which extent and by which mechanism erythropoietin also affects the status of other elements as suggested for silicon, zinc, nickel and manganese [12] needs further confirmation. Various reports in the literature have indicated that solutions for parenteral nutrition and albumin replacement fluids may contain non-negligible amounts of aluminum, chromium and nickel which may accumulate in the body when administered to patients with impaired renal function [13,14]. In various recent papers an association between gadolinium-based contrast agents and the development of nephrogenic systemic fibrosis has been reported in patients with reduced renal function [15]. In patients with uremia, trace element disturbances may also occur by the dialysis treatment per se. Indeed,

Figure 1: Overview of trace elements and metals that may either be nephrotoxic after environmental or occupational exposure or of whom the concentration may be disturbed in patients already having chronic renal failure. 884

41. Trace metal disturbances in end-stage renal failure patients

according to the concentration gradient between the ultrafiltrable amount of a particular element in serum and its concentration in the dialysis fluid, some trace elements may be removed whereas others present as contaminants in the dialysis solution will be transferred to the patients. Due to the hemodialyis treatment each patient is exposed to 15000-30000 liters of dialysis fluid/year. Hence, the minor dialysate contamination with a given element may already result in its distinct accumulation in those subjects. Serious acute and chronic intoxications as well as metal deficiencies have been reported [16,17]. In hemodialysis the dialysis fluids are prepared from the tap water which may contain considerable amounts of trace metals. In the absence of adequate water treatment procedures it must be considered the main source of trace metal dialysate contamination. Some domestic tap water contains aluminum in high concentrations either naturally or as a result of the addition of the element as a flocculant to the water basins, a procedure which is part of the water purification process and has led to an acute, fatal intoxication of a considerable number of patients in a Portuguese dialysis center (see also below) [17,18]. Worth noting is that concentrations of particular elements in tap water may vary seasonally, e.g. silicon, or even on a day-to-day

basis, e.g. aluminum. Trace elements can adequately be removed during water treatment, provided that in addition to softening and deionization the water is treated by reverse osmosis (RO). Carrying out these procedures however, does not necessarily imply the total removal of the elements from the final dialysis fluid [17-19]. Aside from the water treatment procedure The use of commercial salts used to prepare the final dialysis fluid may also be a cumbersome issue. Here besides aluminum, identification of other elements that might be responsible for pathological effects in dialysis patients, e.g. strontium, is warranted [20]. In line with this statement, Padovese et al. [21] demonstrated that dialysis fluids used for either continuous ambulatory peritoneal dialysis (CAPD), hemodialysis and hemofiltration may contain trace metals in various concentrations depending on the chemical composition of the salts used to prepare the final dialysis fluids. They demonstrated that for a series of trace elements including gold, barium, gallium, thallium, vanadium, nickel, chromium, etc. the weekly exposure via the dialysis fluid appeared to be 50- to 12,000-fold higher than the corresponding estimated amount absorbed via the diet. In this context are the recent findings from multicenter surveys demonstrating the addition of concentrates to result in high silicon or strontium

Figure 2: Comparison of the aluminum and lanthanum metabolism. 885

D’HAESE

containing dialysates of particular interest [19,20]. Whether the type of the dialysate, i.e. bicarbonate vs. acetate importantly influences the trace metal concentration and subsequent trace metal disturbances is not yet clear. Findings by Schrooten et al. [20] point to a higher risk for strontium accumulation when acetatebased concentrates are used. It might be anticipated that under certain conditions trace elements used in plasticizers or alloys such as mercury, iron, cadmium, tin and chromium could be introduced into the dialysis fluid also and thus transferred to the patient resulting in either acute or chronic intoxication. The driving force for the transfer of trace elements during dialysis is the gradient between their concentration in the dialysis fluid and the free diffusible fraction in the blood compartment. As a consequence, with highly protein bound elements an even low concentration of these substances in the dialysis fluid may promptly result in a transfer of the element across the dialysis membrane. Hence, in dialysis patients having a serum aluminum level around 20 μg/L (0.74 μmol/L) and of which 80-90% is protein bound, there might already be a transfer to the systemic circulation at aluminum levels in the dialysis fluid as low as 5 μg/L (0.19 μmol/L). In contrast to this, low dialysate levels of a number of elements; e.g. boron [22], fluoride [23], selenium [24], vanadium [25], may result in an adequate intra-dialytic removal of these components from the blood compartment which in turn may result in deficiency of some essential elements, e.g. selenium [24]. Compared to patients treated by hemodialysis, data on trace element accumulation/deficiency in CAPD are rather scanty and less documented. In CAPD the total volume of dialysis fluid patients get in contact with during treatment is much lower. Hence, the potential amount of trace metals that can be exchanged during treatment is much less than in hemodialysis. Therefore, CAPD patients must be considered at a lower risk for trace element accumulation/toxicity as compared to hemodialysis; a statement which is also reflected by the data presented of Padovese et al. [21] comparing the estimated exposure of forty-four trace elements in both patient groups. Distinct differences have been noted in the trace metal content between CAPD and hemodialysis fluids. Also the trace metal content of CAPD fluids may greatly differ between each other, which has been ascribed to the wide range of trace 886

metal concentration in the salts added to prepare the final dialysis fluid [21,26]. To which extent techniques such as on-line hemodiafiltration and biofiltration or sorbent charcoalbased ultrafiltrate regeneration may alter trace metal levels in chronic renal failure patients is not yet clear and can only be evaluated by long-term longitudinal monitoring.

Trace metal accumulation/toxicity in uremic and dialysis patients Aluminum Aluminum is without any doubt the most intensively studied trace metal in dialysis patients and its harmful effects have been well documented during the last decades. So, aluminum overload has been implicated in the pathogenesis of several clinical disorders of the musculoskeletal, central nervous, and hematologic systems [27]. Due to the introduction of adequate water treatment systems and the (partial) withdrawal of aluminum-containing phosphate binders together with the establishment of regular monitoring programs, chronic, caricatural aluminum overload is nowadays rarely seen. At the present day the issue has switched towards more subtle disorders at the level of the parathyroid gland function and bone turnover, resistance to erythropoietin therapy and anemia [28]. It should be noted however, that the risk for, mainly acute, intoxications not be neglected as several outbreaks of (sub)acute fatal intoxications have been reported during the last 2 decades [29] even with criminal prosecution of physicians for gross negligence [30]. Furthermore, aluminum accumulation remains a threat for patients dialyzed in centers from countries which do not have always at their disposal adequate systems for water treatment, aluminum-free phosphate binders or concentrates, or where regular monitoring programs have not been organized yet. In these countries elevated serum aluminum levels (> 20 μg/L, > 0.74 μmol/L) are still frequently observed [29]. In this context it is worth mentioning that the Renal Association (RA) standard states that although it is now generally acknowledged that aluminiumrelated bone disease is a diminishing problem and water treatment facilities in hemodialysis units have considerably improved, the serum aluminum concentration should be measured every three months in all

41. Trace metal disturbances in end-stage renal failure patients

patients on hemodialysis and all peritoneal dialsysis patients receiving oral aluminium hydroxide, whilst the water aluminum content should be tested on a monthly base. The KDOQI guidelines are slightly less stringent than the RA guidelines, with the recommendation that serum aluminium should be measured at least yearly and every 3 months in patients receiving aluminium-containing medications [31]. Aside from regular monitoring it has been advised recently that products and drugs destinated for chronic renal failure aptients should have their composition re-evaluated in order to contain only components with negligible aluminum contamination [32]. Recent research on the speciation and protein binding characteristics of aluminum has led to a better insight in the mechanisms underlying the element’s tissue distribution and toxicity [33, 34]. In this context, the provocative hypothesis linking the widespread use of erythropoietin to the increasing prevalence of adynamic bone disease is of particular interest [35]. Lanthanum Lanthanum carbonate, together with sevelamer hydrochloride belongs to the new generation of calcium-free phosphate binders that can control hyperphosphatemia without adding to the patients calcium load. Lanthanum is only minimally absorbed (0.00127 ± 0.00080% in healthy humans; being 2 to 3 orders of magnitude lower than the values reported for aluminium) [36] and serum lanthanum levels > 2 μg/L (> 0.014 μmol/L) are only rarely seen during long-term treatment in dialysis patients [37], Available bone biopsy data in dialysis patients treated for up to 4 years with lanthanum carbonate indicate low-level bone deposition, the highest concentration ever measured in any patient being 9.4 μg/g (0.067 μmol/L). Unlike aluminium, no direct effects of lanthanum on bone have been reported so far in any clinical or experimental setting [37-39]. Ultrastructural localization demonstrated a heterogeneous distribution of lanthanum in the bone of rats and man, which was independent of the underlying type of renal osteodystrophy [40]. The exclusive presence of lanthanum in the bile and in the lysosomes of the liver cell is consistent with excretion of lanthanum by the liver via the transferrin receptor-endosomal-lysosomal-bile canaliculus pathway [41]. Clinical studies of up to 4 years have not

disclosed any hepatotoxic effect of the drug in patients treated with this phosphate binder. Brain lanthanum levels did not exceed 10 ng/g [0.072 nmol/g] in the brain of rats fed orally during 20 weeks or after intravenous administration of high doses of lanthanum over 4 weeks [42]. No neurological effects have been observed during long-term treatment in dialysis patients [43]. Further studies unraveling the speciation of lanthanum in biological fluids will contribute to a further understanding of its metabolism and kinetics. Strictly spoken, there is no risk for dialysis patients to be prone to increased uptake of the element via the use of dialysis fluids or parenteral solutions. Strontium Strontium levels are increased in plasma of renal failure patients. In dialysis patients the accumulation of the element has been reported to be strongly centre and country-dependent and values up to 50 times those noted in subjects with normal renal function have been reported within the latter population [20]. In addition to the renal failure, accumulation of the element turned out to be due to the use of strontiumcontaminated dialysis fluids secondary to the addition of contaminated acetate-based concentrates. To which other factors such as age, medication, treatment modalities etc ... also contribute to the increased levels is not yet fully understood. The distribution of the element is similar to that of calcium which means that 99% of the body burden is deposited in bone [44]. Within the dialysis population, bone strontium levels were found to be significantly higher in subjects with osteomalacia as compared to this presenting the other types of renal osteodystrophy [45]. A causal, dose-dependent role of strontium in the development of this bone disease has been established in a chronic renal failure ratmodel [46,47]. Moreover the bone osteomalacic lesions were found to be reversible after withdrawal of strontium [9,48]. Silicon Serum silicon levels correlate with the degree of renal failure [49] and once enrolled in a dialysis program patients not only accumulate the element via the oral intake of high silicon drinking water but especially by the use of silicon contaminated dialysis fluids [19,50,51]. Contamination of the dialysis fluid 887

D’HAESE

was found to be due to either the use of RO membranes that insufficiently retain the element during water treatment or by the addition of concentrates containing high amounts of silicon [19]. The clinical significance of increased silicon levels is not yet understood. Parry et al. [52] suggested silicon and aluminum to interact with each other. In their dialysis patients high serum silicon levels were associated with low serum aluminum concentrations inferring either a reduced intestinal absorption -which had been postulated earlier in subjects with normal renal function [53] - or an increased removal of aluminum through dialysis by the formation of the aluminum-silicon complex. To which extent increased silicon levels in the dialysis population may inhibit superoxide dismutase activity [54], favors dextran deposition by linking polysaccharide chains [55, 56] or gives yield to the development of a so-called ‘silicon-related syndrome’, expressed as by painful, nodular skin eruptions and aberrant hair growth and characterized as perforating folliculitis on skin biopsy [57], is not yet fully understood. Selenium Blood selenium levels in dialysis patients are frequently lower than in controls [58]. This is either due to a deficient dietary intake of the element in dialysis patients and/or losses through dialysis membranes [24]. Selenium is known to play a critical role in the glutathione peroxidase activity; an enzyme that protects membrane lipids and other cellular and extracellular components from oxidative damage [58] and an aggravation of oxidative stress during dialysis resulting from low serum serum selenium levels (in combination with decreased serum copper and zinc) has been reported in children with end-stage renal disease as compared to healthy subjects [59]. In view of this oral supplementation of the element to dialysis patients has been recommended and resulted in a significant rise in glutathione peroxidase activity [24, 60]. Selenium supplementation to hemodialysis patients resulted in a progressive increase in T-cell response to phytohemoagglutin and in delayed-type hypersensitivity in the absence of severe side-effects [60]. Chromium Chromium may enter the body via the dialysis 888

fluid [61]. We (data not published) as well as others [62] noted serum chromium concentrations to be increased up to twenty-fold those observed in subjects with normal renal function. In its hexavalent state, the element may act as a carcinogenic substance. Whether the increased levels in the dialysis population are of clinical significance is not yet clear, nor is it elucidated if the increased serum chromium levels in these patients are accompanied by an increased body burden of the element. In view of the latter, it is worth mentioning that in a previous study of our group, in which we assessed the bone trace element content in end-stage renal failure, bone chromium levels in dialysis patients were significantly increased as compared to those noted in subjects with intact renal function. The accumulation of the element in bone, however, could not be associated with the development of a particular type of renal osteodystrophy [63]. Zinc Although there are still some discrepancies in the literature regarding zinc levels in dialysis patients, most studies have found decreased levels of the element in serum [64,65] and muscles whereas the levels in bone [63] and other tissues seem to be normal or even increased suggesting translocation of the element in uremia. The dialysis treatment itself seems to have little or no effect on the serum zinc concentrations. Zinc deficiency in uremic patients has been associated with anorexia, disturbances in taste and sexual performance [66] whereas decreased plasma zinc seem to correlate with erythrocyte superoxide dismutase levels [67]. As evaluated by Türk et al. [68], zinc supplementation did not have any effect on the restoration of immune parameters or enhancement of the antibody response to multivalent influenza vaccine in hemodialysis patients. On the other hand however, has zinc supplementation been reported an effective means of improving serum levels of zinc and cholesterol in the hemodialysis patient [69]. Copper Copper levels in serum of CAPD patients tend in general to be lower than those noted in the presence of a normal function. As for zinc however, its deficiency seems not to be due to the dialysis treatment itself. Here, loss of the element as the ceruloplasmin compound into the peritoneum has been suggested

41. Trace metal disturbances in end-stage renal failure patients

[70]. The effect of copper deficiency is not fully understood. The element is required for lysil oxidase activity which is necessary for cross-linking of collagen. Its deficiency has been associated with growth retardation and anemia. Also has in dialysis patients a correlation been demonstrated between serum copper levels and superoxide dismutase activity. Iron Whereas until a decade ago, the development of iron overload secondary due to blood transfusions to correct for the patients’ anemia, was one of the major problems in dialysis patients, the issue has now switched towards a relative iron deficiency with the introduction of erythropoietin. It has been demonstrated that relative iron depletion may favor the binding of aluminum to transferrin leading to a preferential uptake of the element in transferrin receptor expressing tissues such as e.g. the parathyroid gland which in turn may lead to a reduced parathyroid secretion and hence the development of adynamic bone disease [33]. On the other hand however, may iron overload, by preventing the binding of aluminum to transferrin, result in an increased deposition of the non-protein bound aluminium at the bone calcification front [71]. Gadolinium Free gadolinium, a rare earth that is mainly eliminated by the kidney (97%) is toxic to tissues and therefore unsafe in human use. The trivalent gadolinium is very close to the divalent calcium ion as reflected in size, bonding, coordination and donor atom preference [72]. Gadolinium-containing compounds however, are widely used as contrast agents and when used for this application the element is sequestered by binding to a chelate. These stable complexes have long been thought to be safe, even in patients with impaired renal function. Recently however, exposure to gadoliniumcontaining contrast agents has been associated with the development of ‘Nephrogenic Systemic Fibrosis’ (NSF). Initially named ‘Nephrogenic Fibrosing Dermopathy’ (NFD) in 2001, NSF is a recently characterized systemic fibrosing disorder occurring in patients with underlying renal disease. This condition principally leads to skin thickening and hardening and may induce joint

immobility and inability to walk and potentially leads to death when several organ systems are involved. Since its recognition in 1997 and the first description in 2000, more than 215 cases have been reported worldwide [73,74]. Clusters of NSF were associated to an exposure to gadolinium containing contrast agents during magnetic resonance imaging [75]. Gadolinium has been detected in skin tissue of patients with NSF [76]. Gadodiamide, a linear gadolinium chelate appears to be particularly at risk as reflected by an odds ratio of 32.5 to acquire NSF when exposed to this particular compound [73,77]. During renal failure, gadodiamide accumulation may explain the development of NSF. Evidence for gadolinium to play an etiological role in the development of the disease is supported by the notion that before the introduction of gadoliniumcontaining contrast media, NSF has not been known [73]. Regulatory decisions have been taken to contraindicate gadodiamide in patients with severe renal impairment [78]. To date it is uncertain whether other gadolinium-based contrast media are involved in some of the affected patients, but this is a matter of current evaluation [79]. See also chapter 30. Metal-metal interactions It should be noted that in the majority of the above mentioned studies, metal-induced renal injury was considered as if exposure occurred to only one metal at a time. In reality it is clear that environmental and occupational exposure may involve several metals at the same time and in varying concentrations [34]. It has been shown that with combined exposure various metals may interact with each other and that one metal may alter the potential toxicity of another in either a beneficial or deleterious way. As an example, whilst arsenic has been shown to worsen cadmium-induced nephrotoxicity, data from experimental studies have shown that selenium may protect against the renal effects induced by cadmium [52]. Other studies have shown that the iron status may alter the toxic effects of aluminium at the level of the bone and the parathyroid gland [53,54], whilst in a recent increased lead accumulation was associated with disturbances in the concentration of a number of essential trace elements [55].

889

D’HAESE

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14. 15. 16. 17. 18.

19.

20. 21.

22. 23. 24. 25.

890

Nuyts GD, Van Vlem E, Thys J, De Leersnijder D, D’Haese PC, Elseviers MM, De Broe ME. Defined and undefined occupational risk factors of chronic renal failure. Lancet 1995; 346: 7-11. Altmann P, Butter KC, Plowman D, Chaput de Saintonge DM, Cunningham J, Marsh FP. Residual renal function in hemodialysis patients may protect against hyperaluminemia. Kidney Int 1989; 32: 710-13. Smythe WR, Alfrey AC, Craswell PW, Crouch CA, Ibels LS, Kubo H, Nunnelly LL, Rudolph H. Trace metal abnormalities in chronic uremia. Ann Intern Med 1982; 96: 302-310. Blumsohn A, Morris B, Eastell R. Stable strontium absorption as a measure of intestinal calcium absorption: comparison with the double-radiotracer calcium absorption test. Clin Science 1994; 87: 363-368. Barton JC, Conrad ME, Harrison L, Nuby S. Effects of calcium on the absorption and retention of lead. J Lab Clin Med 1978; 91: 366-376. Ittel TH. Determinants of gastrointestinal absorption and distribution of aluminium in health and uremia. Nephrol Dial Transplant 1993; 8 (Suppl 1): 17-24. Canavese C, Pacitti M, Salomone M, Santoro A, Strata P, Mangiarotti G, Talarico S, Sabbioni E, Pietra R, Vercellone A. Strontium overload in uremic patients on regular dialytic treatment. Trans Am Soc Artif Inter Organs 1986: 32: 120-122. Lehmann G, Hein G, Wolf G. The osteoporosis patient with renal insufficiency: what has to be taken into account in the selection and administration of antiosteoporosis medication? Z Rheumatol 2006; 65: 380-382. Oste L, Bervoets AR, Behets GJ, Dams G, Marijnissen RL, Geryl H, Lamberts LV, Verberckmoes SC, Van Hoof VO, De Broe ME, D’Haese PC. Time-evolution and reversibility of strontium-induced osteomalacia in chronic renal failure rats. Kidney Int 2005; 67: 920-930. Damment SJP, Pennick M. Systemic lanthanum is excreted in the bile of rats. Toxicol Lett. 2007; 171: 69-77. Singh H, Reed J, Noble S, Cangiano JL, Van Wyck DB. Effect of intravenous iron sucrose in peritoneal dialysis patients who receive erythropoiesis-stimulating agents for anemia: A randomized, controlled trial. Clin J Am Soc Nephrol 2006; 1: 475-482. Hosokawa S, Yoshida O 1993 Effects of erythropoietin on trace elements in patients with chronic renal failure undergoing hemodialysis. Nephron 1993; 65: 414-417. Gammelgaard B, Sandberg E. Aluminium and nickel in human albumin solutions. J Trace Elem Electrolytes Health Dis 1989; 3: 39-42. D‘Haese PC, Lamberts LV, Liang L, Boone LP, Van Waeleghem, De Broe ME. Contribution of parenteral and dialysate solutions to the aluminum accumulation in dialysis patients. Blood Purif 1990; 8: 359-362. Reilly RF. Risk for nephrogenic systemic fibrosis with gadoteriol (ProHance) in patients who are on long-term hemodialysis. Clin J Am Soc Nephrol 2008; 3: 747-751. D’Haese PC, De Broe ME. Adequacy of dialysis: trace elements in dialysis fluids. Nephrol Dial Transplant 1996; 11 (Suppl 2): 9297. Simoes J, Barata JD, D’Haese PC, de Broe ME. Cela n’arrive qu’aux autres. Nephrol Dial Transplant 1994; 9: 67-68. Barata JD, D’Haese PC, Pires C, Lamberts LV, Simoes J, De Broe ME. Low-dose (5mg/kg) desferrioxamine treatment in acutely aluminum-intoxicated hemodialysis patients using two drug administration schedules. Nephrol Dial Transplant 1996; 11: 125132. D’Haese PC, Shaheen FA, Huraib SO, Djukanovic L, Polenakovic MH, Spasovski G, Shikole A, Schurgers ML, Daneels RF, Lamberts LV, Van Landeghem GF, De Broe ME. Increased silicon levels in dialysis patients due to the high silicon content in the drinking water, inadequate water treatment procedures, and concentrate contamination: A multicentre study. Nephrol Dial Transplant 1995; 10: 1838-1844. Schrooten I, Elseviers MM, Lamberts LV, De Broe ME, D’Haese PC. Increased serum strontium levels in dialysis patients. An epidemiological survey. Kidney Int 1999; 56: 1888-1892. Padovese P, Gallieni M, Brancaccio D, Pietra R, Fortaner S, Sabbioni E, Minoia C, Markakis K, Berlin A. Trace elements in dialysis fluids and assessment of the exposure of patients on regular hemodialysis, hemofiltration and continous ambulatory peritoneal dialysis. Nephron 1992; 61: 442-448. Mauras Y, KS Ang KS, P Simon P, B Tessier B, F Cartier F, P Allain P. Increase in blood plasma levels of boron and strontium in hemodialyzed patients. Clin Chim Acta 1986; 156: 315-320. Chaleil D, Simon P, Tessier B, Cartier F, Allain P. Blood plasma fluoride in haemodialysed patients. Clin Chim Acta 1986; 156: 105108. Saint-Georges MD, DJ Bonnefont DJ, BA Bourely BA, M-CT Jaudon M-CT, P Cereze P, P Chaumeil P, C Gard C, CL D’Auzac CL. Correction of selenium deficiency in hemodialyzed patients. Kidney Int 1989; 36 (suppl 27): 274-277. Hosokawa S, Yoshida O. Serum vanadium levels in chronic hemodialysis patients. Nephron 1993; 64: 388-394.

41. Trace metal disturbances in end-stage renal failure patients

26. 27. 28. 29. 30. 31. 32. 33.

34. 35. 36. 37.

38. 39. 40. 41.

42. 43.

44. 45. 46. 47. 48. 49. 50. 51. 52.

Wallaeys B, Cornelis R, Mees L, Lameire N. Trace elements in serum, packed cells, and dialysate of CAPD patients. Kidney Int 1986; 30: 599-604. De Broe ME, Coburn JW (eds). Aluminum and renal failure. Dordrecht: Kluwer Acad Publ, 1989. De Broe ME, Drüeke TB, Ritz E. Consensus Conference: diagnosis and treatment of aluminum overload in end-stage renal failure patients. Nephrol Dial Transplant 1993; 8 (Suppl 1): 1-4. D’Haese PC, De Broe ME. Trace elements in dialysis patients. In: Hörl WE, Koch KM, Lindsay RM, Ronco C, Winchester JF (eds). Replacement of Renal Function by Dialysis. Dordrecht: Kluwer Acad Publ, 2004 pp 903-925. Berend K, Knoops GJ, De Wolff FA. Prosecution after an outbreak of subacute aluminum intoxication in a hemodialysis center. Leg Med 2004; 6: 1-10. Lamb EJ, Hodsman A, van Schalkwyk D, Ansell D, Warwick G. Serum calcium, phosphate, parathyroid hormone, albumin, aluminium and cholesterol achievement on replacement therapy (Chapter 9). Nephrol Dial Transplant 2007; 22 (suppl 7): 105-118. Bohrer D, Bertagnolli DC, de Oliveira SMR, do Nascimento PC, de Carvalho LM, Pomblum SG. Drugs as a hidden source of aluminium for chronic renal patients. Nephrol Dial Transplant 2007; 22: 605-611. Smans KA, D’ Haese PC, Van Landeghem GF, Andries LJ, Lamberts LV, Hendy GN, De Broe ME. Transferrin-mediated uptake of aluminium by human parathyroid cells results in reduced parathyroid hormone secretion. Nephrol Dial Transplant 2000; 15: 1328-1336. Van Landeghem GF, De Broe ME, D’Haese PC. Al and Si: Their speciation, distribution, and toxicity - A critical review. Clin Biochem 1998; 31: 385-397. Smans KA, Van Landeghem GF, D’Haese PC, Couttenye MM, De Broe ME. Is there a link between erythropietin therapy and adynamic bone disease? Nephrol Dial Transplant 1996; 11: 1248-1249. Pennick M, Dennis K, Damment SJ. Absolute bioavailability and disposition of lanthanum in healthy human subjects administered lanthanum carbonate. J Clin Pharmacol 2006; 46: 738-746. D’Haese PC, Spasovski GB, Sikole A, Hutchison A, Freemont TJ, Sulkova S, Swanepoel C, Pejanovic S, Djukanovic L, Balducci A, Coen G, Sulowicz W, Ferreira A, Torres A, Curic S, Popovic M, Dimkovic N, De Broe ME. A multicenter study on the effects of lanthanum carbonate (Fosrenol) and calcium carbonate on renal bone disease in dialysis patients. Kidney Int 2003; 85 (Suppl): 73-78. Persy VP, Behets GJ, Bervoets AR, De Broe ME, D’Haese PC. Lanthanum: a safe phosphate binder. Semin Dial 2006; 19:195-199. Behets GJ, Dams G, Vercauteren SR, Damment SJ, Bouillon R, De Broe ME, D‘Haese PC. Does the phosphate binder lanthanum carbonate affect bone in rats with chronic renal failure? J Am Soc Nephrol 2004; 15: 2219-2228. Behets GJ, Verberckmoes SC, Oste L, Bervoets AR, Salomé M, Cox AG, Denton J, De Broe ME, D’Haese PC. Localization of lanthanum in bone of chronic renal failure rats after oral dosing with lanthanum carbonate. Kidney Int 2005; 67: 1830-1836. Yang Z, Schryvers D, Roels F, D’Haese PC, De Broe ME. Demonstration of lanthanum in liver cells by energy-dispersive X-ray spectroscopy, electron energy loss spectroscopy and high-resolution transmission electron microscopy. J Microsc 2006; 223: 133-139. Bervoets ARJ, Yang Z, Schryvers D, Roels F, Behets GJ, Verberckmoes SC, Dauwe S, Blust R, De Broe ME, D’Haese PC. Transhepatocellular transport and increased gastro-intestinal absorption of lanthanum in chronic renal failure. Kidney Int (Submitted). Altmann P, Barnett ME, Finn WF; SPD405-307 Lanthanum Carbonate Study Group. Cognitive function in Stage 5 chronic kidney disease patients on hemodialysis: no adverse effects of lanthanum carbonate compared with standard phosphate-binder therapy. Kidney Int 2007; 71: 252-259. Cabrera WE, Schrooten I, De Broe ME, D’Haese PC. Strontium and bone. J Bone Miner Res 1999; 14: 661-668. D’Haese PC, Schrooten I, Goodman WG, Cabrera WE, Lamberts LV, Elseviers MM, Couttenye M-M, De Broe ME. Increased bone strontium levels in hemodialysis patients with osteomalacia. Kidney Int 2000; 57: 1107-1114. Schrooten I, Cabrera W, Dauwe S, Lamberts LV, Marynisen R, Dorriné W, Goodman WG, De Broe ME, D’Haese PC. Strontium causes osteomalacia in chronic renal failure rats. Kidney Int 1998; 54: 448-456. Schrooten I, Behets GJ, Cabrera WE, Vercauteren SR, Lamberts LV, Verberckmoes SC, Bervoets AJ, Dams G, Goodman WG, De Broe ME, D’Haese PC. Dose-dependent effects of strontium on bone of chronic renal failure rats. Kidney Int 2003; 63: 927-935. Oste L, Verberckmoes SC, Behets GJ, Dams G, Bervoets AR, Van Hoof VO, Bohic S, Drakopoulos M, De Broe ME, D’Haese PC. Strontium incorporates at sites critical for bone mineralization in rats with renal failure. X-Ray Spectrometry 2007; 36: 42-44. Gitelman HJ, Alderman F, Perry SJ. Renal handling of silicon in normals and in patients with renal insufficiency. Kidney Int 1992; 42: 957-959. Dobbie JW, Smith MJB Urinary and serum silicon in normal and uremic individuals. Ciba Found Symp 1986; 121: 194-213. Gitelman HJ, Alderman FR, Perry SJ. Silicon accumulation in dialysis patients. Am J Kidney Dis 1992; 19: 140-143. Parry R, Plowman D, Delves HT, Roberts NB, Birchall JD, Bellia JP, Davenport A, Ahmad R, Fahal I, Altmann. Silicon and aluminum interactions in haemodialysis patients. Nephrol Dial Transplant 1998; 13: 1759-1762.

891

D’HAESE

53. 54. 55. 56.

57. 58. 59. 60. 61. 62. 63. 64. 65.

66. 67. 68.

69. 70. 71. 72. 73. 74. 75. 76. 77.

78.

892

Edwardson JA, Moore PB, Ferrier IN, Lilley JS, Newton GWA, Barker J, Templar J, Day JP. Effect of silicon on gastrointestinal absorption of aluminium. Lancet 1993; 342: 211-212. Shainkin-Kestenbaum R, Adler AJ, Berlyne GM. Inhibition of superoxide dismutase activity by silicon. J Trace Elem Electrolytes Health Dis 1990; 4: 97-99. Bergonzi G, Paties C, Vassallo G, Zangrandi A, Poisetti PG, Ballocchi S, Fontana F, Scarpioni L. Dextran deposits in tissues of patients undergoing hemodialysis. Nephrol Dial Transplant 1990; 5: 54-58. D’Haese PC, Van Landeghem GF, De Broe ME. Increased silicon levels in dialysis patients due to the high silicon content in the drinking water, inadequate water treatment procedures, and concentrate contamination: A multicentre study. Nephrol Dial Transplant 1996; 11: 1188-1189 (Reply Letter to the Editor). Saldanha LF, Gonick HC, Rodriguez HJ, Marmelzat JA, Repique EV, Marcus CL. Silicon related syndrome in dialysis patients. Nephron 1997; 77: 48-56. Hsieh Y-Y, Shen W-S, Lee L-Y, Wu T-L, Ning H-C, Sun C-F. Long-term changes in trace elements in patients undergoing chronic hemodialysis. Biol Trace Elem Res 2006; 109: 115-121. Zwołińska D, Grzeszczak W, Szczepańska M, Kiliś-Pstrusińska K, Szprynger K. Lipid peroxidation and antioxidant enzymes in children on maintenance dialysis. Pediatr Nephrol 2006; 21: 705-710. Bonomini M, Forster S, De Risio F, Rychly J, Nebe B, Manfrini V, Klinkmann H, Albertozzi A. Effects of selenium supplementation on immune parameters in chronic uraemic patients on haemodialysis. Neprol Dial Transplant 1995; 10: 1654-1661. Rukgauer M, Schmitt Y, Schneider H, Krusejarres JD. Trace element balance in patients undergoing hemodialysis. Trace Elem & Electrolytes 1994; 11: 155-168. Henderson I, Leung A, Halls D, Fell G, Dobbie J, Kennedy AC. Hyperchromiumaemia in chronic dialysis patients. Proc EDTA-ERA 1985; 22: 270-275. D’Haese PC, Couttenye M-M, Lamberts LV, Elseviers MM, Goodman WG, Schrooten I, Cabrera W, De Broe ME. Al, Fe, Pb, Cd, Cu, Zn, Cr, Mg, Sr and Ca content in bone of end-stage renal failure patients. Clin Chem 1999; 45: 1548-1556. Esfahani ST, Hamidian MR, Madani A, Ataei N, Mohseni P, Roudbari M, Marzieh Haddadi. Serum zinc and copper levels in children with chronic renal failure. Pediatr Nephrol 2006; 21: 1153-1156. Navarro-Alarcon M, Reyes-Pérez A, Lopez-Garcia H, Palomares-Bayo M, Olalla-Herrera M, Lopez-Martinez MC. Longitudinal study of serum zinc and copper levels in hemodialysis patients and their relation to biochemical markers. Biol Trace Elem Res; 2006; 113: 209-222. Mahajan SK, Bowersox EM, Rye DL, Abu-Hamdam DK, Prasad AS, McDonald FD, Biersack KL . Factors underlying abnormal zinc metabolism in uremia. Kidney Int 1989; 36 (Suppl): 269-273. Richard MJ, Arnud J, Jurkovitz C, Hachache T, Meftahi H, Laporte F, Foret M, Favier A, Cordonnier D. Trace element lipid peroxidation abnormalities in patients with chronic renal failure. Nephron 1991; 57: 10-15. Türk S, Bozfakioglu S, Ecder ST, Kahraman T, Gürel N, Erkoç R, Aysuna N, Türkmen A, Bekiroglu N, Ark E. Effects of zinc supplementation on the immune system and on antibody response to multivalent influenza vaccine in hemodialysis patients. Int J Artif Organs 1998; 21: 274-278. Chevalier CA, Liepa G, Murphy MD, Suneson J, Vanbeber AD, Gorman MA, Cochran C. The effects of zinc supplementation on serum zinc and cholesterol concentrations in hemodialysis patients. J Ren Nutr 2002; 12: 183-189. Thomson NP, Stevens BJ, Humphery TJ, Atkins RC. Comparison of trace elements in peritoneal dialysis, hemodialysis and uremia. Kidney Int 1983; 23: 9-14. Van Landeghem GF, D’Haese PC, Lamberts LV, Djukanovic L, Pejanovic S, Goodman WG, De Broe ME. Low serum aluminum values in dialysis patients with increased bone aluminum levels. Clin Nephrol 1998; 50: 69-76. Evans C (ed). Biochemistry of lanthanides. Plenum Press New York 1990. Rosenkranz AR, Grobner T, Mayer GJ. Conventional or gadolinium containing contrast media: the choice between acute renal failure or nephrogenic systemic fibrosis. Wien klin Wochenschr 2007, 119: 271-275. Evenepoel P, Zeegers M, Segaert S, Claes K, Kuypers D, Maes B, Flamen P, Fransis S, Vanrenterghem Y. Nephrogenic fibrosing dermopathy: a novel, disabling disorder in patients with renal failure. Nephrol Dial Transplant 2004, 19: 469-473. Grobner T, Prischl. Gadolinium and nephrogenic systemic fibrosis. Kidney Int 2007; 72: 260-264. Boyd AS, Zic JA, Abraham JL. Gadolinium deposition in nephrogenic fibrosing dermopathy. J Am Acad Dermatol 2007; 56: 2730. Marckmann P, Skov L, Rossen K, Dupont A, Brimnes Damholt M, Goya Heaf J, Thomsen HS. Nephrogenic systemic fibrosis: suspected causative role of gadodiamide used for contrast-enhanced magnetic resonance imaging. J Am Soc Nephrol 2006; 17: 2359 - 2362. Andréjak M, Thuillier D, Lok C, Gras-Champel V. Fibrose systémique néphrogénique et produits à base de gadolinium: données disponibles début 2007 Nephrogenic Systemic Fibrosis and Gadolinium-based Contrast Media. Thérapie 2007; 62: 169-172.

41. Trace metal disturbances in end-stage renal failure patients

79. 80. 81. 82. 83.

Thomsen HS, (ed) Contrast Media. Safety Issues and ESUR Guidelines. Springer: Berlin, Heidelberg, New York, 2006. Liu J, liu Y, Habeebu SM, Waalkes MP, Klaassen CD. Chronic combined exposure to cadmium and arsenic exacerbates nephrotoxicity, particularly in metallothionein-I/II null mice. Toxicology 2000, 147:157-166. El-Sharaky AS, Newairy AA, Badreldeen MM, Eweda SM, Sheweita SA. Protective role of selenium against renal toxicity induced by cadmium in rats. Toxicology 2007; 235: 185-93. Van Landeghem GF, D’ Haese PC, Lamberts LV, De Broe ME. Competition of iron and aluminium for transferrin: the molecular basis for aluminium deposition in iron-overloaded dialysis patients? Exp Nephrol 1997; 5: 239-245. Ahamed M, Singh S, Behari JR, Kumar A, Siddiqui MKJ. Interaction of lead with some essential trace metals in the blood of anemic children from Lucknow, India. Clin Chim Acta 2007; 377: 92-97.

893

42

Smoking and the kidney Eberhard RITZ Ruperto-Carola-University, Heidelberg, Germany

Introduction ___________________________________________________________ 895 Acute effects of smoking on the kidney _____________________________________ 895 Smoking and renal disease ________________________________________________ 896 Conclusion _____________________________________________________________ 898 References _____________________________________________________________ 898

Introduction

S

moking, mostly cigarette smoking, is one of the most important modifiable renal risk factor. In contrast to the long known potential of cigarette smoking to promote carcinogenesis, lung disease and cardiovascular events, even in the renal community the renal risk has only recently attracted attention [1], although the renal risk conferred by smoking had been known to diabetologists for more than 2 decades [2].

Acute effects of smoking on the kidney As early as in 1907 Hesse had described in his doctoral thesis the transient increase of blood pressure and heart rate during cigarette smoking [3]. Nevertheless it had long been claimed that there was no excess hypertension in smokers [4], presumably hypertension is masked because the body weight of smokers is low. Cryer et al [5] documented marked sympathetic activation during cigarette smoking and the release of blood pressure active hormones such as AVP, aldosterone, cortisone and ACTH. The effect of smoking on blood

pressure outlasts the period of smoking. In a controlled study Ritz documented higher night time blood pressure on a day when occasional smokers smoked compared to a day when they did not smoke (Figure 1) [6]. She found that this was accompanied by commensurate changes in heart rate. The effects of smoking are complex as illustrated by the findings that smoking also increases arterial stiffness [7] and causes a curious “reversed” office hypertension, i.e. normal office blood pressure in patients with high home blood pressure measurements [8]. The effects of smoking on renal hemodynamics are pronounced. We showed that smoking caused an acute increase in circulating epinephrine, in heart rate and in blood pressure; this is accompanied by a decrease in the filtration fraction with a significant increase of renal vascular resistance [9]. This renal hemodynamic pattern could be reproduced in healthy volunteers by chewing a nicotine containing gum, suggesting that nicotine is the main culprit. Interestingly in patients with IgA glomerulonephritis smoking failed to consistently reduce the filtration fraction; on average filtration fraction remained unchanged and a transient increase

RITZ

Figure 1. Systolic and diastolic blood pressure as well as heart rate in healthy occasional smokers – comparison of a smoking day with a non-smoking day [6].

in urinary albumin excretion was seen, consistent with (but not proof for) acute glomerular hypertension. Plasma renin activity in the circulation did not increase, but even this is inappropriate to the blood pressure increase. Nevertheless the renin-angiotensin system may play a role in the genesis of the acute hemodynamic changes in the kidney, since they were abrogated by ß-1 selective blockers [10]. In animal experiments we produced severe of proteinuria, pronounced glomerulosclerosis and marked interstitial fibrosis when we applied an acetone extract of cigarette smoke to the oral mucosa of subtotally nephrectomized rats [11]. Despite these acute alterations of glomerular function and morphology, in the long run the damage from smoking seems to be mainly mediated by damage to the renal vasculature. Halimi [12] found that acutely cigarette smoking increased the excretion of cGMP in the urine, pointing to compensatory vasodilatation in response to nicotin mediated vasoconstriction. In contrast in long-term observations Gambaro [13] found increased endothelin-1 concentrations in cigarette smokers and this was associated with increased renovascular resistance. The concept of a primary vascular problem in the kidneys of smokers is in line with the observation of Lhotta [14]: he examined renal biopsies of patients with primary renal disease. In smokers he found more severe myointimal hyperplasia and arteriolar hyalinosis. The concept of primarily vascular damage is also in line with our recent observation [15] that diabetic patients with microalbuminuria/proteinuria had a more rapid increase of serum creatinine concen896

tration with time compared to non smokers despite no difference in urinary protein excretion.

Smoking and renal disease Smoking and microalbuminuria In early studies on microalbuminuria smoking was identified as a powerful predictor of microalbuminuria [16, 17]. This has recently been confirmed and extended in large population based studies. In the PREVEND study Pinto-Sietsma [18] found that current smokers had an adjusted relative risk of microalbuminuria of 1.65, former smokers a RR of 1.27, but heavy smokers > 20 cigarettes per day a RR of 1.96. This finding was confirmed in the ARIC study where the odds ratio for current smokers was 2.33 and in former smokers 1.58 [19]. In Okinawa Tozawa found in a prospective follow up study that the relative risk to develop proteinuria was 1.32 in a prospective follow up study [20]. Atherosclerotic renal artery stenosis Many investigators identified smoking as a strong risk factor for atherosclerotic renal artery stenosis; Hadj-Abdelkader [21] found that 70-80% of patients with this diagnosis were smokers and frequently these stenoses are bilateral [22] and frequently the source of cholesterol microembolism [23]. Ischemic nephropathy A selective decrease of renal plasma flow (RBF) with no decrease in glomerular filtration rate in smokers

42. Smoking and the kidney

[24] is most likely the result of smoking induced endothelial cell dysfunction and possibly also the result of increased production of endothelin 1 [13]. Smoking specific glomerulopathy (including “idiopathic nodular glomerulosclerosis”) Markowitz described “idiopathic nodular glomerulosclerosis”, a distinct entity linked to hypertension and smoking [25]. Even in the absence of hypertension a more recent series [26] in nondiabetic heavy smokers showed as smoking specific glomerular damage – apart from nodular glomerulosclerosis – a broader spectrum with segmental or focal glomerulosclerosis, glomerular ischemia, interstitial fibrosis and tubular atrophy as well ad arterial sclerosis and hyalinosis. Electron microscopy showed capillary basement thickening and reduplication. Diabetic nephropathy Since the seminal description of Christiansen [2] cigarette smoking had been known to diabetologists as a risk factor for the onset and progression of microangiopathy. Cigarette smoking diabetics have a higher risk to develop microalbuminuria, a greater rate of progression to proteinuria and a greater risk to experience an elevation of serum creatinine. This has been confirmed in numerous studies [27-32]. In diabetic patients with overt diabetic nephropathy and elevated serum creatinine the measured rate of loss of creatinine clearance was twice as high in smokers compared to non-smokers [33]. For obvious reasons there are no controlled prospective studies on whether cessation of smoking attenuates the rate of loss of GFR, but the observation of Sawicki is telling [32]: in type I diabetic patients diabetic nephropathy progressed in 11% of non-smokers, in 33% of ex-smokers and in 53% of current smokers, suggestive of ( but not proof for) a beneficial effect of cessation of smoking. It is of interest that several recent studies confirm that smoking increases not only the risk to develop diabetic nephropathy, but also the risk to develop type 2 diabetes as found in the nurses health study [34] and recently confirmed by other studies [35].

dominant polycystic kidney disease [37]; but it had remained uncertain, whether smoking affected the immune response and – in view of the two hit hypothesis to explain the generation of renal cysts in ADPKD – whether the higher renal risk of smoking in ADPKD patients is not explained by DNA mutations resulting from smoking. In a retrospective case control study, however, examining both patients with inflammatory (IgA glomerulonephritis) and non-inflammatory (ADPKD) primary renal disease, Orth et al. [38] found an increased odds ratio of progression to end stage renal disease in patients with 5–15 pack years [odds ratio 3.5 (1.3–9.6)] and in patients with > 15 pack years [5.8 (2.0–17.0; p < 0.001)]. Interestingly the increased odds ratio was found only in patients who were not on ACE inhibitors (Table 1). ACE inhibitors do not provide complete protection, however. In type 2 diabetics Chuahirun found that despite treatment with ACE inhibitors reaching target blood pressure values, serum creatinine had risen to significantly higher values in smokers (1.78 ± 0.2 mg/dl) compared to non-smokers (1.32 ± 0.01) during a 61 months follow up [39]. Chronic kidney disease and end-stage renal disease The adverse effect of smoking on renal function in patients without primary renal disease has also been well documented. Smoking was the most powerful predictor of progression in patients with severe essential hypertension in the study of Regalado [40] and, as reflected by an increase of serum creatinine > 3 mg/dl, also in the study of Bleyer [41]. In a large prospective population sample, primarily a cancer research project (CLUE study), Haroun [42] noted that smoking accounted for no less than 30% of the “attributable risk” of chronic kidney disease (CKD), defined as serum creatinine > 2 mg/dl, and this was particularly true in the elderly.

Table 1. Smoking – ESRF in men (n=144) with primary renal disease (retrospective case-control study) [27]. Pack-years

Nondiabetic primary renal disease There had been some past reports that the risk of progression was higher in patients with lupus nephritis who smoked [36] and in patients with autosomal

Odds ratio (95%-CI)

p-value

0-5

1.0

5-15

3.5 (1.3-9.6)

0.017

-

>15

5.8 (2.0-17.0)

0.001

897

RITZ

In a nationwide study Ejerblad [43] found a relative risk of major serum creatinine (>3.4 mg/dl) elevation of 1.51 for smokers > 20 cigarettes/day and a risk of 1.52 for smokers with > 30 packyears. A similar magnitude of risk was noted by Ishani [44] in the MRFIT study (RR1.84). Based on NHANES data 1976–1980 Stengel [45] found in individuals smoking 1–20 cigarettes/day a relative risk of 1.2 and in individuals smoking > 20 cigarettes/day a relative risk of 2.3 for endstage renal disease. Renal transplants The adverse effect of cigarette smoking on renal function was not seen only in the kidneys of patients

with primary renal disease but also in renal allograft recipients [46]. The hazard ratio of graft loss censored for patient death was 1.48. Sung even found a relative risk of 2.3 for patient s with a history of pretransplant smoking [47].

Conclusion It emerges from the above that smoking is a major renal risk factor, causing onset and progression of renal disease. The magnitude of the risk is comparable to that conferred by blood pressure elevation and proteinuria. It is a modifiable risk factor [48] but the attention which the medical community devotes to this risk factor is far from what it should be.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

898

Orth SR, Ritz E, Schrier RW: The renal risks of smoking. Kidney Int 51:1669-1677, 1997. Christiansen JS: Cigarette smoking and prevalence of microangiopathy in juvenile-onset insulin-dependent diabetes mellitus. Diabetes Care 1:146-149, 1978. Hesse E: Einfluss des Rauchens auf den Kreislauf. Deutsches Archiv klinische Medizin 89:565-575, 1907. Green MS, Jucha E, Luz Y: Blood pressure in smokers and nonsmokers: epidemiologic findings. Am Heart J 111:932-940, 1986. Cryer PE, Haymond MW, Santiago JV, Shah SD: Norepinephrine and epinephrine release and adrenergic mediation of smokingassociated hemodynamic and metabolic events. N Engl J Med 295:573-577, 1976 Ritz S: Einfluss des Rauchens auf Blutdruck und Herzfrequenz bei Probanden mit und ohne primäre Nierenerkrankung. Inauguraldissertation (Medizinische Fakultät der Ruprecht-Karls-Universität Heidelberg), 2003. Yasmin, Falzone R, Brown MJ: Determinants of arterial stiffness in offspring of families with essential hypertension. Am J Hypertens 17:292-298, 2004. Narkiewicz K, Kjeldsen SE, Hedner T: Is smoking a causative factor of hypertension? Blood Press 14:69-71, 2005. Ritz E, Benck U, Franek E, Keller C, Seyfarth M, Clorius J: Effects of smoking on renal hemodynamics in healthy volunteers and in patients with glomerular disease. J Am Soc Nephrol 9:1798-1804, 1998. Benck U, Clorius JH, Zuna I, Ritz E: Renal hemodynamic changes during smoking: effects of adrenoreceptor blockade. Eur J Clin Invest 29:1010-1018, 1999. Odoni G, Ogata H, Viedt C, Amann K, Ritz E, Orth SR: Cigarette smoke condensate aggravates renal injury in the renal ablation model. Kidney Int 61:2090-2098, 2002. Halimi JM, Giraudeau B, Vol S, Caces E, Nivet H, Lebranchu Y, Tichet J: Effects of current smoking and smoking discontinuation on renal function and proteinuria in the general population. Kidney Int 58:1285-1292, 2000. Gambaro G, Verlato F, Budakovic A, Casara D, Saladini G, Del Prete D, Bertaglia G, Masiero M, Checchetto S, Baggio B: Renal impairment in chronic cigarette smokers. J Am Soc Nephrol 9:562-567, 1998. Lhotta K, Rumpelt HJ, Konig P, Mayer G, Kronenberg F: Cigarette smoking and vascular pathology in renal biopsies. Kidney Int 61:648-654, 2002. Orth SR, Schroeder T, Ritz E, Ferrari P: Effects of smoking on renal function in patients with type 1 and type 2 diabetes mellitus. Nephrol Dial Transplant 20:2414-2419, 2005. Mimran A, Ribstein J, DuCailar G, Halimi JM: Albuminuria in normals and essential hypertension. J Diabetes Complications 8:150156, 1994. Hörner D, Fliser D, Klimm HP, Ritz E: Albuminuria in normotensive and hypertensive individuals attending offices of general practitioners. J Hypertens 14:655-660, 1996. Pinto-Sietsma SJ, Mulder J, Janssen WM, Hillege HL, de Zeeuw D, de Jong PE: Smoking is related to albuminuria and abnormal renal function in nondiabetic persons. Ann Intern Med 133:585-591, 2000.

42. Smoking and the kidney

19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38.

39. 40. 41. 42. 43. 44.

Hogan S.H.,Colindres R.E., Cai J., Coresh J.: Association of Smoking with Albuminuria in a cross-sectional Probability Sample of US Adults. J.Am.Soc.Nephrol. 12:209a, 2001. Tozawa M, Iseki K, Iseki C, Oshiro S, Ikemiya Y, Takishita S: Influence of smoking and obesity on the development of proteinuria. Kidney Int 62:956-962, 2002. Hadj-Abdelkader M, Alphonse JC, Boyer L, Younes H, Deteix P: [Smoking and atheromatous stenosis of the renal arteries]. Arch Mal Coeur Vaiss 94:925-927, 2001. Shurrab AE, Mamtora H, O’Donoghue D, Waldek S, Kalra PA: Increasing the diagnostic yield of renal angiography for the diagnosis of atheromatous renovascular disease. Br J Radiol 74:213-218, 2001. Scolari F, Tardanico R, Zani R, Pola A, Viola BF, Movilli E, Maiorca R: Cholesterol crystal embolism: A recognizable cause of renal disease. Am J Kidney Dis 36:1089-1109, 2000. Baggio B, Budakovic A, Casara D, Gambaro G, Saladini G, Piccoli A, Verlato F: Renal involvement in subjects with peripheral atherosclerosis. J Nephrol 14:286-292, 2001. Markowitz GS, Lin J, Valeri AM, Avila C, Nasr SH, D’Agati VD: Idiopathic nodular glomerulosclerosis is a distinct clinicopathologic entity linked to hypertension and smoking. Hum Pathol 33:826-835, 2002. Liang KV, Greene EL, Oei LS, Lewin M, Lager D, Sethi S: Nodular glomerulosclerosis: renal lesions in chronic smokers mimic chronic thrombotic microangiopathy and hypertensive lesions. Am J Kidney Dis 49:552-559, 2007. Telmer S, Christiansen JS, Andersen AR, Nerup J, Deckert T: Smoking habits and prevalence of clinical diabetic microangiopathy in insulin-dependent diabetics. Acta Med Scand 215:63-68, 1984. Rossing P, Hougaard P, Parving HH: Risk factors for development of incipient and overt diabetic nephropathy in type 1 diabetic patients: a 10-year prospective observational study. Diabetes Care 25:859-864, 2002. Scott LJ, Warram JH, Hanna LS, Laffel LM, Ryan L, Krolewski AS: A nonlinear effect of hyperglycemia and current cigarette smoking are major determinants of the onset of microalbuminuria in type 1 diabetes. Diabetes 50:2842-2849, 2001. Chase HP, Garg SK, Marshall G, Berg CL, Harris S, Jackson WE, Hamman RE: Cigarette smoking increases the risk of albuminuria among subjects with type I diabetes. Jama 265:614-617, 1991. Bangstad HJ, Osterby R, Rudberg S, Hartmann A, Brabrand K, Hanssen KF: Kidney function and glomerulopathy over 8 years in young patients with Type I (insulin-dependent) diabetes mellitus and microalbuminuria. Diabetologia 45:253-261, 2002. Sawicki PT, Didjurgeit U, Muhlhauser I, Bender R, Heinemann L, Berger M: Smoking is associated with progression of diabetic nephropathy. Diabetes Care 17:126-131, 1994. Biesenbach G, Grafinger P, Janko O, Zazgornik J: Influence of cigarette-smoking on the progression of clinical diabetic nephropathy in type 2 diabetic patients. Clin Nephrol 48:146-150, 1997. Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, Willett WC: Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med 345:790-797, 2001. Patja K, Jousilahti P, Hu G, Valle T, Qiao Q, Tuomilehto J: Effects of smoking, obesity and physical activity on the risk of type 2 diabetes in middle-aged Finnish men and women. J Intern Med 258:356-362, 2005. Ward MM, Studenski S: Clinical prognostic factors in lupus nephritis. The importance of hypertension and smoking. Arch Intern Med 152:2082-2088, 1992. Chapman AB, Johnson AM, Gabow PA, Schrier RW: Overt proteinuria and microalbuminuria in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 5:1349-1354, 1994. Orth SR, Stockmann A, Conradt C, Ritz E, Ferro M, Kreusser W, Piccoli G, Rambausek M, Roccatello D, Schafer K, Sieberth HG, Wanner C, Watschinger B, Zucchelli P: Smoking as a risk factor for end-stage renal failure in men with primary renal disease. Kidney Int 54:926-931, 1998. Chuahirun T, Wesson DE: Cigarette smoking predicts faster progression of type 2 established diabetic nephropathy despite ACE inhibition. Am J Kidney Dis 39:376-382, 2002. Regalado M, Yang S, Wesson DE: Cigarette smoking is associated with augmented progression of renal insufficiency in severe essential hypertension. Am J Kidney Dis 35:687-694, 2000. Bleyer AJ, Shemanski LR, Burke GL, Hansen KJ, Appel RG: Tobacco, hypertension, and vascular disease: risk factors for renal functional decline in an older population. Kidney Int 57:2072-2079, 2000. Haroun MK, Jaar BG, Hoffman SC, Comstock GW, Klag MJ, Coresh J: Risk factors for chronic kidney disease: a prospective study of 23,534 men and women in washington county, Maryland. J Am Soc Nephrol 14:2934-2941, 2003. Ejerblad E, Fored CM, Lindblad P, Fryzek J, Dickman PW, Elinder CG, McLaughlin JK, Nyren O: Association between smoking and chronic renal failure in a nationwide population-based case-control study. J Am Soc Nephrol 15:2178-2185, 2004. Ishani A, Grandits GA, Grimm RH, Svendsen KH, Collins AJ, Prineas RJ, Neaton JD: Association of single measurements of dipstick proteinuria, estimated glomerular filtration rate, and hematocrit with 25-year incidence of end-stage renal disease in the multiple risk factor intervention trial. J Am Soc Nephrol 17:1444-1452, 2006.

899

RITZ

45. 46.

47. 48.

900

Stengel B, Tarver-Carr ME, Powe NR, Eberhardt MS, Brancati FL: Lifestyle factors, obesity and the risk of chronic kidney disease. Epidemiology 14:479-487, 2003. Woo YM, McLean D, Kavanagh D, Ward L, Aitken S, Miller GJ, Egan P, Hughes K, Clark L, Carswell K, Morris ST, Northridge DB, Rodger RS, Jardine AG: The influence of pre-operative electrocardiographic abnormalities and cardiovascular risk factors on patient and graft survival following renal transplantation. J Nephrol 15:380-386, 2002. Sung RS, Althoen M, Howell TA, Ojo AO, Merion RM: Excess risk of renal allograft loss associated with cigarette smoking. Transplantation 71:1752-1757, 2001. Remuzzi G, Schieppati A, Ruggenenti P: Clinical practice. Nephropathy in patients with type 2 diabetes. N Engl J Med 346:11451151, 2002.

43

Star fruit Miguel M. NETO1, Ruither O. CAROLINO2, Norberto P. LOPES4 and Norberto GARCIA-CAIRASCO3 Departments of 1Nephrology, 2Biochemistry and Immunology, Faculty of Medicine, 4Physics and Chemistry, Faculty of Pharmacy, University of São Paulo, Ribeirão Preto, SP, Brazil

3Physiology,

Introduction ___________________________________________________________ 901 Clinical manifestations of neurotoxicity______________________________________ 902 Treatment and outcome: a literature review __________________________________ 904 Treatment and outcome summary _________________________________________ 905 Clinical manifestations of nephrotoxicity ____________________________________ 906 Mechanisms of neurotoxicity ______________________________________________ 907 Molecular and biochemical aspects Neurobiological aspects

907 908

Conclusions ___________________________________________________________ 910 References _____________________________________________________________ 911

Introduction

T

he carambola or star fruit belongs to the Oxalidaceae family, species Averrhoa carambola. Slices cut in cross-section have the form of a star (Figure 1). It is believed to have originated in Ceylon and the Moluccas but it has been cultivated in southeast Asia and Malaysia for many centuries. It is commonly grown in some provinces in southern China, in Taiwan and India. It is rather popular in the Philippines and Queensland, Australia and moderately so in some of the South Pacific Islands, particularly Tahiti, New Caledonia and Netherlands New Guinea, and in Guam and Hawaii. There are some subspecies in the Caribbean Islands, in Central America and in tropical West Africa. It is also common in Brazil, where it is served as fresh beverage, in natura, or as an industrialized juice, as it is also served throughout the world. It was introduced into Southern Florida before 1887 and the

fruits are also available in many European countries and Canada. It is widely used in restaurants for decorative purposes. In India, ripe fruit is administered to halt hemorrhages and to relieve bleeding hemorrhoids. In Brazil, star fruit is recommended as a diuretic for kidney and bladder complaints [1, 2]. There are 2 distinct classes of carambola: the smaller, very sour type, richly flavored, with more oxalic acid and the larger, so-called “sweet type”, mild flavored, rather bland, with less oxalic acid. The oxalic content of ripe carambolas could reach to an average of 0.5 g per 100 mL of juice, the acid being mostly in the free state. Physicians should be informed of this because there are some individuals who may be adversely affected by ingestion of even small amounts of oxalic acid or oxalates. The acid types of carambola have been used to clean and polish metal, especially brass, as they dissolve tarnish and rust [1].

NETO, CAROLINO, LOPES & GARCIA-CAIRASCO

Clinical manifestations of neurotoxicity

Figure 1. Averrhoa carambola

The first description of a patient intoxication outbreak was reported by Martin LC et al in 1993 [3]. They described an outbreak of intractable hiccups in patients on a regular program of hemodialysis. Of the 10 patients, 8 developed intractable hiccups. The other 2 patients who did not present hiccups ingested the fruit immediately before the hemodialysis sessions. Treatment with lidocaine, chlorpromazine, metoclopramide, flunitrazepam was useless. All patients with hiccups improved only after they were submitted to hemodialysis sessions. The authors did not describe any signs of neurological involvement such as behavioral disturbances or mental confusion. The hiccups had been seen as a curiosity and hallmark of star fruit intoxication and not as a threat till 1998. In that year, our group described 6 patients under dialysis that presented hiccups with neurological disturbances after eating star fruit: one patient died with convulsions, others presented mental confusion, psychomotor agitation, insomnia and 5 patients improved after a few hemodialysis sessions [4]. Most of the patients (5 patients) had intractable hiccups as their first symptoms. 902

In 2000 Chang JM et al [5] reported 19 patients with renal failure under dialysis and one also with renal failure but in supportive treatment who after star fruit ingestion developed signs and symptoms of intoxication such as: hiccups, mental confusion, paresis, muscular weakness and convulsions. Eight of those patients died (including the patient in supportive treatment) After those reports many other reports confirmed these findings and are summarized in Table 1 [3-19]. The most common symptom of star fruit intoxication is persistent and intractable hiccups, observed in almost all 88 patients reported in Table 1. After star fruit ingestion, hiccups start in a variable fashion from 2.5 hours to 14 hours (average 4.6 h) [5] or half to 10 hours (average 2 to 3 hours) [13]. In most of the intoxication cases, attempts to treat hiccups with chlorpromazine and metoclopramide were unsuccessful. The amount of fruit ingested varied in the literature from small pieces to large amounts as 500 mL of juice in 1 to 3 days [13]. Even a small amount of star fruit can cause severe neurological complications and death. According to Chang JM et al [5] presenting symptoms were predominantly neuromuscular. Report of limb numbness were noted in 75%, persistent hiccups in 60%, disturbed consciousness of various degrees in 50%, decreased muscle power in 35%, dyspnea in 25%. Eight of the 10 patients with abnormal consciousness died (80%). According to Neto MM et al [13] persistent hiccups were noted in 93.7%, vomiting in 68.7%, disturbed consciousness of variable degrees (mental confusion that progressed to coma in some cases, psychomotor agitation) in 65.6%, decreased muscle power, limb numbness, paresis, insomnia and paresthesias) in 40.6%, seizures in 21.8% and hemodynamic instability (hypotension and shock) in 9.3%. Seven of the 32 patients who presented seizures and severe consciouness disturbances died. Star fruit intoxication can be classified into three levels according to signs and symptoms that might provide a useful guideline for institution of proper treatment: mild, moderate and severe (Table 2). Certain cases of mild intoxication progress to a severe level if patients are not treated and the velocity of progression is extremely variable, depending on the characteristics of each patient. In some cases this progression happens with less than 24 hours after star fruit ingestion [5, 13, 9, 10, 15, 17, 18, 19]. Therefore,

43. Star fruit

Table 1. Reports of star fruit intoxication. Reports Martin LC et al, 1993 [3] Neto MM et al, 1998 [4] Chang JM et al, 2000 [5]

Number of cases 08 06 20

Wang JL et al, 2000 [6] Lo KY et al, 2001 [7] Ho MP et al 2001[8] Wu CC et al, 2002 [9] Yap HJ et al, 2002 [10]

02 01 02 02 03

Chan YL et al, 2002 [11] Tse KC et al, 2003 [12]

01 06

Neto MM et al, 2003 [13]

32*

Neto MM et al 2004 [14] Chang CH et al 2004[15] Hung SW et al 2004[16] Tsai MH et al 2005[17] Chen LL et al 2005 [18] Wu MY et al 2007[19]

04 01 01 02 01 02

Total

88

Treatment before intoxication HD: 7 PD: 1 HD: 4 PD: 2 HD: 15 PD: 4 Supportive: 1 Supportive: 2 PD Suppportive: 2 Supportive: 2 HDI: 1 Supportive: 2 PD: 1 PD: 4 HD: 1 Supportive: 1 HD: 20 PD: 8 Supportive: 4 Supportive: 4 HD Supportive Supportive: 2 Supportive Supportive: 1 HD: 1 HD: 46 PD: 19 Supportive: 23

Symptoms described Hiccups, insomnia Hiccups, vomiting, insomnia, consciousness disturbance, seizures Hiccups, consciousness disturbance, numbness of limbs, decreased muscle power, skin paresthesia, seizures Hiccups, consciousness disturbance, seizures Hiccups, insomnia Hiccups, vomiting, consciousness disturbance, seizures Hiccups, consciousness disturbance Hiccups, insomnia, vomiting, consciousness disturbance, seizures Hiccups, consciousness disturbances, seizures Hiccups, vomiting, consciousness disturbances, hyperkalemia Hiccups, vomiting, consciousness disturbances, seizures Hiccups, vomiting, diarrhea, consciousness disturbances, seizures Vomiting, consciousness disturbances, seizures Hiccups, vomiting, diarrhea, numbness of the lower limbs, seizures Hiccups, nausea, consciousness disturbances, seizures Hiccups, consciousness disturbances, Hiccups, vomiting, consciousness disturbances, seizures

HD: Hemodialysis, PD: peritoneal dialysis, supportive: not yet on a dialysis program * Included 6 patients of Ref 4.

Table 2. Clinical levels of star fruit intoxication. Intoxication level Mild

Moderate

Severe

Signs and symptoms • Hiccups • Vomiting • Insomnia • Psychomotor agitation • Numbness, paresthesias and decreased muscle power of the limbs • Mild mental confusion • Moderate to severe mental confusion progressing to coma • Seizures progressing to status epilepticus • Hemodynamic instability progressing to hypotension and shock

any patient with renal failure (stages 3 to 5) with a suspected star fruit intoxication should not be discharged and should be observed very closely. The severe cases may be difficult to diagnose promptly, since the symptoms mimic either strokes (brain stem strokes) or may even resemble “metabolic” or uremic disturbances [9, 13, 15, 16, 19]. Due

to suspicion of organic neurologic problems, many patients with severe intoxication had cerebrospinal fluid collected, and were submitted to CT or MRI. No specifical findings were detected in any of this examinations according to review of the literature. A single exception has been the report by Chan YL et al (11) in which the authors claim that star fruit poisoning can induce hyperintense lesions at the thalami and right temporo-occipital cortex revealed by single voxel proton MR spectroscopy. According to Martin LC et al [3] from 10 patients that ingested the fruit only 8 developed hiccups, the other 2 ate the fruits before their hemodialysis sessions and did not present any symptom. Although star fruit has enriched potassium content, hyperkalemia has not been suggested as causing of death in reported cases [5, 13]. Most of the patients that were reported in literature were on a regular program of peritoneal dialysis (PD) or hemodialysis (HD). From 88 described patients, 23 patients were without need of regular dialysis or they 903

NETO, CAROLINO, LOPES & GARCIA-CAIRASCO

did not know that they have kidney problems (Table 1). Their creatinine levels ranged from 2.3 mg/dL to 20.5 mg/dL. These patients developed signs and symptoms varying from mild to severe levels of intoxication. Seizures are present in 30% of patients with star fruit intoxication [17], and most patients have convulsive [6, 8, 10, 11, 13, 14, 17, 18, 19] or non-convulsive [16] status epilepticus. The mortality rate of patients with seizures occurring after star fruit intoxication (severe intoxication) is significantly higher than of patients without seizures [13, 17]. Phenytoin, midazolam, diazepam and phenobarbital seem to have little or no effect on the control of persistent seizures provoked by star fruit toxicity. However, significant clinical improvement of seizure was demonstrated in one patient after the use of profofol [20]. Sometimes there is a poor correlation with the degree of underlying renal function and the symptoms, while more severe symptoms may develop in those patients with pre-dialyzed conditions as opposed to those with end stage renal disease [17]. These variations of symptoms among individuals might be explained by individual biological responses, genetic factors, patient age, the amount of toxin content in each fruit, various star fruit subespecies, and the detoxification, excretion, or both, of the toxin from the blood stream [7, 13].

Treatment and outcome: a literature review According to the revised literature there are 88 patients described, 46 previously on hemodialysis (HD), 19 on peritoneal dialysis (PD) and 23 without previous dialytic treatment (supportive treatment). These patients were described in 16 reports in the medical literature with their symptoms and treatment outcomes [3-19]. In one of the first observations, Martin LC et al [3] reported that hemodialysis eliminated all symptoms, although all of them presented clinical pictures of mild intoxication (hiccups and vomiting). None of those patients had mental confusion or seizures. In another report 6 patients who were described by Neto et al [4], one patient died and PD was offered as the only treatment. The other five received conventional hemodialysis and improved without sequelae (mild and moderate levels of intoxication, and in one case the patient had severe mental confusion without seizures or hemodynamic instability). Unfortunately there is no neurological 904

follow up of any kind in any of the studies in order to look for eventual neurological sequelae in survivors after what we may call the hemodialysis rescue. Chang et al [5] in a retrospective study of 20 patients described that among 10 patients with abnormal consciousness 8 died despite additional emergent hemodialysis. They did not specify the length of dialysis and the time this emergent dialysis was performed after ingestion. The patients who died had an earlier onset of symptoms (average 4.6 hours after ingestion) than survivors (average 8.8 hours after ingestion). In their series of 32 patients Neto et al [13] showed that seven patients died after intoxication episodes. The main characteristics of the patients who died were convulsive activities in 6 and severe mental confusion in all 7 patients, while 2 of them presented hemodynamic instability (hypotension and shock). Most of the patients who died were treated by peritoneal dialysis or did not receive any other kind of treatment. The other 25 patients improved without sequelae and they were treated either by conventional hemodialysis, daily hemodialysis (6 to 8 hours duration) or even by continuous methods of dialysis. A few patients were treated by peritoneal dialysis. Complete recovery time in these 25 patients ranged from 1 to 12 days (mean 4.4 days and median 4.0 days). Tse KC et al [12] presented 6 cases with mild or moderate intoxication. Three patients were treated with intensive hemodialysis and improved, 3 were treated with peritoneal dialysis and also improved. Most patients responded after 2-3 days of treatment. Neto MM et al [14] presented 4 cases, 3 with mild intoxication and 1 with severe intoxication. Patients with mild intoxication recovered spontaneously and were not dialysed. The patient with severe intoxication and end-stage renal disease was submitted to daily hemodialysis of 6 to 10 hours duration and woke up after 10 days and 80 hours of hemodialysis (extended daily dialysis - EDD). This patient was discharged after 20 days of hospitalization and was enrolled in a program of regular hemodialysis. Wang JL et al [6] reported 2 cases of patients with severe intoxication who died despite hemodialysis but the time of hemodialysis initiation after ingestion is not known. Lo KY et al [7] reported one patient with mild intoxication who improved maintaining previous treatment (continuous ambulatory peritoneal dialysis - CAPD).

43. Star fruit

Ho MP et al [8] described 2 patients, one with moderate and other with severe intoxication who improved after hemodialysis. The patient with moderate intoxication recovered with 2 sessions of HD and the other recovered after 14 days of HD. Wu CC et al [9] described 2 cases: one with mild intoxication that recovered with only one 6 hour HD session and a severe case who improved with HD after 5 days. Hemodialysis was performed at the same day when the patients were admitted to the hospital. Yap HJ et al [10] presented 3 cases: one with mild intoxication that improved after HD in 2 days; the second patient presented with mild intoxication that progressed to moderate intoxication after one week without treatment, and improved after 2 days of HD; the third patient who had severe intoxication, was treated with 2 sessions of plasmapheresis without improvement and received HD later. He died after 25 days after admission. Chan YL et al [11] described one patient with severe intoxication who was submitted to HD 2 days after intoxication and died 7 weeks later with worsening of consciousness disturbances. Hung SW et al [15] reported a severe case of a patient with intoxication who died. A continuous method of renal replacement therapy was started on the fourth day of intoxication. Chang CH et al [16] reported one case of severe intoxication that improved with HD in 11 days. Tsai MH et al [17] reported 2 cases of severe intoxication. Both patients were submitted to hemodialysis, one, 2 days and the other, 3 days after admission, and both died after 23 days and 7 days of treatment respectively. Chen LL et al [18] reported one severe case of a patient who started hemodialysis 3 days after admission. His state of consciousness did not modify with hemodialysis. Charcoal hemoperfusion was performed during 6 hours and his consciousness improved progressively. About 24 hours after the 6 hours session of hemoperfusion, his consciousness returned to normal without subsequent mental confusion. Wu MY et al [19] reported 2 severe cases of intoxication. One patient ingested 2 fruits, arriving to the hospital with persistent hiccups, nausea and vomiting. During the next 24 hours, agitation, subsequent drowsy consciousness developed. He was intubated and seizures occurred with a generalized tonic-clonic

pattern, evolving to status epilepticus. Hemodialysis was performed in the first hospital day. On the second hospital day hemoperfusion was performed due to persistent comatose state after hemodialysis. The condition improved after 20 hours of hemoperfusion. The other patient ingested 1 star fruit in the afternoon and at the same night she presented with hiccups, agitation, bizarre behaviour and mental confusion. Due to deterioration of consciousness and respiratory distress, the trachea was intubated. On the first day the patient underwent a hemodialysis session and remained comatose. A neurologist suspected of a brain stem stroke. Daily dialysis was arranged for 2 days, but she remained comatose. Charcoal hemoperfusion was performed on the third hospital day for 8 hours and consciousness recovered 16 hours after the hemoperfusion session on the fourth hospital day. The patient was weaned off a ventilator on the fifth hospital day.

Treatment and outcome summary Seven patients who were in supportive treatment (without need for dialysis) at the time of star fruit ingestion had mild intoxication presenting hiccups or diarrhea. Six patients improved without dialysis. Time to improve was up to 24 hours in 4 patients, 5 days in another one, and there is no information in one patient. One patient improved after IPD (intermittent peritoneal dialysis)[13, 14]. Peritoneal dialysis was not an efficient method of treatment although 1 patient with signs and symptoms of moderate intoxication and 2 with mild intoxication changed from CAPD to IPD (intermittent peritoneal dialysis) and improved [13]. Two patients that remained in CAPD also improved [7, 12]. In one case [13] patient presented diplopia that continued for 6 weeks after improvement of the acute intoxication episode. Patients with severe intoxication did not benefit from peritoneal dialysis treatment [13]. Hemodialysis was an efficient method in 30 reported cases especially if initiated early, together with aggressive supportive care including mechanical ventilation in some severe cases [8, 9, 10, 12, 13, 14, 16]. Many patients presented rebound effects after dialysis, with symptoms starting a few hours after the end of the dialytic procedures. These rebound effects included persistence of hiccups or worsening of consciousness disturbances [13]. Interestingly, 2 surviving patients with severe intoxication presenting with seizures 905

NETO, CAROLINO, LOPES & GARCIA-CAIRASCO

and hemodynamic instability, were given continuous replacement therapy as first choice treatment [13]. The recovery time in these 2 cases was 8 and 12 days. Another patient treated with EDD woke up after 10 days of treatment (14). In other 14 reported patients hemodialysis was not effective and patients died despite treatment [5, 6, 10, 11, 15, 17]. In 8 cases reported by Chang JM et al [5] the study was retrospective and we do not know the emergent dialysis starting time and neither the dialysis dose. We also do not have this information in the 2 cases reported by Wang JL et al [6]. Yap HJ et al [10]. describes one severe case of intoxication submitted twice to plasma exchange without improvement. Hemodialysis was started later and the patient died 25 days later due to pneumonia and septic shock without improving consciousness disturbance. In the other 4 reported cases hemodialysis was started within 2 or more days [10, 11, 15, 17]. Early recognition of star fruit intoxication and prompt and properly treatment with hemodialysis seem to be an important factor affecting the survival of patients. In severe cases, prolonged coma duration may be associated with increased morbidity and mortality [10, 11, 13, 14, 17]. Hemoperfusion was used for the first time as an option of treatment in a severe case by Chen LL et al [18] and patient consciousness returned to normal without subsequent mental confusion. Wu MY et al [19] submitted 2 patients with severe intoxication to 20 hours and 8 hours of hemoperfusion and also had good and fast improvement of the intoxication condition. In this 3 cases, 2 hemodialysis sessions in the first described patient, 1 hemodialysis session and 2 daily hemodialysis session in the patients of the second report failed to counterbalance neurotoxicity; however consciousness improved dramatically after hemoperfusion [19]. A dramatic decrease in comatose time and rapid weaning from the ventilator may help reduce morbidity and mortality [19]. Hemoperfusion seems to be a promising kind of treatment to severe cases of star fruit intoxication. However this issue requires further analysis with large trials [19].

Clinical manifestations of nephrotoxicity Chen CL et al [21] described 2 patients who developed nausea, vomiting, abdominal pain and lumbar pain and presented acute renal failure due to acute in906

terstitial nephritis after the ingestion of great amounts of star fruit juice. Both patients were submitted to hemodialysis, and kidney histology showed typical alterations due to acute oxalate nephropathy such as intraluminal and intraepithelial deposition of colorless oxalate crystals with a pattern of birefringence including all colours of the rainbow under polarized light. These crystals also appear blue in hematoxylin-eosin stain and black in von Kossa´s stain. In both cases renal function recovered in 4 weeks. One of the patients ingested 1600 mL of the juice and the other 3000 mL in a short time interval. The authors determined the oxalate content of sour carambola juice ingested by the patients. The results showed the oxalate contents of star fruit juice ingested in both cases; in one case were 820 mg/dL and in the other case 308 mg/dL. The estimated amounts of ingested oxalate were 13.1 g in one case and 9.2 g for the other case. There are 2 types of star fruit, sour and sweet. The sour type contains more oxalate than the sweet type [1]. Oxalic acid and its soluble salts are poisonous to humans and animals, whereas insoluble salts of calcium and magnesium oxalate are not. Oxalates ingested by humans may be precipitated by calcium as an insoluble complex, which then is excreted in feces [21, 22, 23]. In both cases reported by Chien et al [21], patients ingested sour carambola juice on an empty stomach so that the protective effect of calcium and magnesium in food was not present. The dehydration state may have contributed to the development of carambola- associated acute nephropathy. The authors do not report any concomitant neurological signs or symptoms [21]. Fang HC et al [24] in their report intended to establish a connection between star fruit and acute oxalate nephropathy. They administered star fruit juice, 4 mL/100 g of body weight, in male Sprague-Dawley rats of 180 to 200 g, with an oxalate concentration of 2.4g/dL, approximately 1 g/kg. The authors established a strong relationship between star fruit and acute oxalate nephropathy. This relationship was found only in the experimental group under both fasting and water deprivation conditions. The acute interstititial nephritis after oxalate overload may be due to calcium oxalate crystals inducing obstructive effect, nephrocalcinosis and also by inducing apoptosis of renal epithelial cells [22, 23, 25]. Recently, Niticharoenpong K et al [26] reported a patient with underlying chronic renal disease, who

43. Star fruit

developed a rapid increase in serum creatinine and oxalate nephropathy after chronic ingestion of star fruit juice without overt neurotoxicity. Urinalysis of this patient revealed numerous crystals, consistent with oxalate crystals. A kidney biopsy was performed and light microscopy showed crystals deposition consistent with oxalate crystals. The patient had been given star fruit by his family for over 3 years. The exact quantity of star fruit juice varied, and could not be ascertained with certainty. Although there are not many other cases of star fruit oxalate nephrotoxicity described, it seems reasonable to avoid consumption of large amounts of star fruit juice especially on an empty stomach and chronic consumption in patients with underlying chronic renal disease.

Mechanisms of neurotoxicity Molecular and biochemical aspects Despite the recent reports of neurotoxicity of star fruit in uremic patients, the first one on star fruit toxicity was described by Muir and Lam [27] in 1980. They related toxic effects of star fruit extract after its intraperitoneal administration in mice. This extract induced seizures and death. In 1993, Martin LC et al. reported the outcome of intractable hiccups in uremic patients associated to star fruit consume [3]. However, the first report of neurotoxicity in human beings was performed by Moyses-Neto M et al (4). Subsequently, many similar case reports were described, but only in 2002 an etiological agent was proposed; oxalic acid would be the star fruit neurotoxin [28]. In fact, oxalic acid (Figure 2), a substance associated to intoxications, is found in other vegetables. The oxalic acid poisoning is well exemplified in the Sorrel poisoning. Sorrel refers to two species of plants belonging to the genus Rumex; R.acetosa L. and R. acetosella L., which present high content of oxalate. Sorrel poisoning has been known to occur in man, horse and sheep and is associated to ingestion of these plants . Succinctly, the symptoms of Sorrel poisoning in human is due to the sequestering of serum calcium by the oxalic acid to form insoluble calcium oxalate. This reduction in available calcium leads to violent muscular stimulation with convulsions and collapse, associated to derangement of the blood-clotting mechanism. Post-mortem,

Figure 2. Examples of chemical constituents presenting on star fruit juice.

calcium oxalate crystals are found in the renal tubules and in others tissues. The kidney shows cloudy swelling, hyaline-degeneration and tubular sclerosis. The mouth, esophagus and stomach show corrosive effects, and cerebral edema is commonly found [29]. Hence, oxalic acid was considered a logical putative neurotoxin of star fruit. In fact, this hypothesis was recently supported by a study that described the star fruit juice or oxalate solution administered by gavage could evoke seizures (convulsion associated to electroencephalographic recordings showing generalized spike-waves in parietal and frontal lobes) and death in nephrectomized rats. Furthermore, neither seizure nor death was reported when calcium-treated star fruit juice was administered. These observations are similar to those made after the administration of star fruit juice or oxalate solution administration to sham-operated rats [30]. These data showed the oxalate neurotoxicity in this animal model and gave support for the involvement of oxalate in the intoxication by star fruit in uremic patients.The hipothesis above is not supported by the fact that foods presenting comparable or higher oxalic acid content (i.e. rhubarb and spinach, respectively) [31]; do not induce 907

NETO, CAROLINO, LOPES & GARCIA-CAIRASCO

similar intoxication in uremic patients. Considering this, star fruit may also accumulate secondary natural metabolites that may act as a potent toxin for mammals as occur in other vegetal species (32). Phytochemical investigation of the star fruit juice revealed the occurrence of a large number of volative terpenoids [33] that can not be correlated with the observed neurotoxic effects. In addition to the volatile constituents the star fruit juice has also several carotenoids and flavonoids and some of them are shown in Figure 3. These compounds exhibit potent antioxidant activity [34, 35, 36, 37, 38, 39] and may preserve other unstable structures against the oxidative reaction induced by O2 during the juice preparation. These facts are well known in current chemistry literature and had encouraged the development of several hyphenated techniques and analytical methodologies for the identification of antioxidant polyenes and vitamins [40, 41, 42, 43]. Some neurotoxic compounds are small molecules and some of them contain functional groups that can react with O2. Considering this hypothesis the presence of the carotenoids and flavonoids may preserve the toxin chemical structure. Recently, Carolino RO et al isolated and partially characterized a neurotoxic fraction from star fruit applying soft purification steps. This fraction was referred to as AcTx and was able to evoke seizures in mice and rats [44]. Initial insights into the mechanism of action of convulsant activity of AcTx were performed focused on GABA (γ-aminobutyric acid) and glutamate neurotransmitter systems, since it is well-established that an imbalance between these systems may lead to hyperexcitability, provoking seizures [45, 46]. The release and uptake of GABA and glutamate were assayed in synaptosomes. This is a well-recognized model for studying neurotransmitter nerve-terminal-related mechanisms since it retains all machinery for the uptake, storage, release of neurotransmitter, and ionic conductance, while being sufficiently simple and homogeneous for meaningful biochemical studies [47]. In that preparation, AcTx was not able to alter significantly both GABA and glutamate release and re-uptake. The AcTx ability to bind to GABA and glutamate receptors was also evaluated, and we demonstrated it was able to bind only to GABA receptors. Preliminary chemical studies on AcTx content indicated that it was free of oxalic acid and proteinogenic amino acids. In addition, AcTx is a small molecule, with molecular weight less than 500, 908

compatible with renal excretion [44]. These data indicate that star fruit contains at least one other neurotoxin in addition to oxalic acid.

Neurobiological aspects When behavioral recordings are coupled with electroencephalography, in a digital format, the so-called Video-EEG, allows to prove, in freely moving animals, the behavioral and EEG effects after star fruit ingestion or after local application in specific brain regions of either, the crude or the purified toxin. In the first case the hypothesis that experimentally uremic animals, induced by HgCl2, a known model of renal failure [48], will reproduce the star fruit intoxication effects found in the patients can be tested (see above). In the second case, the hypothesis that the crude or purified toxin per se will be able to induce behavioral and EEG activity compatible with brain hyperexcitability, possibly associated to seizures is tested. As a positive effect, the latter experimental protocol (with not relationships with renal alterations) will even validate the potential of this neurotoxin as a new tool in the neuroscience field. Currently, rats are implanted with electrodes and cannulas in their cortices following known stereotaxic coordinates according to the Atlas of Paxinos and Watson [49]. Control electroencephalograms (EEG) (baseline) are recorded prior to (controls) and after vehicle (0.9% saline solution) injections, for 30 min prior to AcTx application. The Video-EEG recordings are made using specific equipment [50] and only animals that had the electrodes and cannulas in the right position, confirmed by histology, are used (Figure 4C). Control behavior and EEGs were examined in the basal situation that is, in the absence of vehicle or AcTx microinjections. Usually, animals explore the cage and, being awake, display typical desynchronized, high frequency-low amplitude EEG activity. Subsequent microinjections of 1μL of the 0.9% saline vehicle did not modify the EEG, nor induce any behavioral alterations of the animals (Figure 3A; two upper recordings). Animals (n=3) submitted to cortical microinjection of 170 mM AcTx presented strong progression of epileptiform EEG activity from 10 to 240 min (Figure 3) [44] . The continued phenomenon observed over such a long period of time in both cases is called status epilepticus, a mimetic of the clinical situation. This is

43. Star fruit

Figure 3. Evolution of epileptiform activity in rat 02 after cortical microinjection of AcTx (170 μM). A: Control and saline recordings express typical desynchronized EEG in the waking state. Ten minutes after AcTx injection, the EEG activity recording indicates subtle baseline alterations with poly-spikes that thereafter, from 20 to 240 min, evolved into a sustained electrographic status epilepticus. Note that at 30 min EEG recording activity was of opposite polarity when compared for example, to the EEG recording at 240 min. B: Observe the very weak EEG epileptiform activity following cortical injection of 17 μM/1μL AcTx, in comparison with the strong EEG epileptiform activity shown after cortical injection of 170 μM/1μL of the toxin. Notice also, as shown in A, that in the second half of the recording period (at around 140 min), there occurred a clear-cut inversion of EEG polarity. Reprinted with permission from Carolino et al [44].

a protocol interesting for the evaluation of long term effects of the star fruit intoxication, and allows us to look for chronic behavioral, EEG and structural or cellular alterations in these animals. Behavioral and EEG effects of a cortical AcTx (170 μM/1 μL) microinjection are shown in Figure 4. A preliminary characterization of the video-EEG after crude star fruit juice had been given to animals

Figure 4. Behavioral and EEG effects of a cortical AcTx (170 μM/1μL) microinjection. A: Digitalized behavior sequence (16 frames captured in a video-EEG setup). Aligned frames allow the detection of subtle behavioral alterations such as forelimb (white rectangles) and head (white circles) and myoclonic activity. B: Observe the EEG window with hypersynchronous epileptiform activity coinciding with video frames ranging from the 1st to the 16th in A. C: Cellular Nissl staining showing cortical localization of chemitrodes used for AcTx microinjection. Reprinted with permission from Carolino et al [44].

bearing induced acute renal failure [51]. In addition to that, Video-EEG recordings following cortical administration of AcTx showed behavioral changes, including partial limbic seizures (forelimb and head myoclonus), evolving to a status epilepticus, accompanied by sustained cortical EEG epileptiform activity, particularly after the 170 μM AcTx injection (Figures 3 and 4). Star fruit juice also induced seizures when applied to cortical areas, showing that convulsant activity is present in crude star fruit extracts. The present data confirm the excitatory profile of AcTx and star fruit extracts. The progressive and sustained EEG epileptiform activity induced by AcTx is a characteristic of known 909

NETO, CAROLINO, LOPES & GARCIA-CAIRASCO

excitatory convulsants, although with potentially different mechanisms of action, such as pilocarpine and kainic acid [52]. In the case of patients, the only report in which the authors claim that star fruit intoxication induce brain lesions (in this case thalamic and cortical) is the one from Chan YL et al [11]. Unfortunately there is no neurological follow up of the eventual neurological sequelae that patients who survive will display in their future lives after the treatment’s rescue. Based on the characteristics of star fruit intoxications and of the isolated neurotoxic fraction (AcTx), we postulated a hypothesis on how the star fruit ingestion would induce intoxication in uremic patients. Initially, the fact that only uremic patients are involved in star fruit intoxications [4, 5, 13] indicate that the neurotoxic substance should be filtered by the kidneys and eliminated in urine, so the star fruit toxin would be hydrosoluble and must have low molecular weight. Actually, AcTx is hydrosoluble and has a low molecular weight (less than 500). Therefore, after star fruit ingestion, the neurotoxin would be absorbed in digestive system, and its plasmatic concentration would increase while its renal excretion would initiate. Conversely, in uremic patients, the renal excretion is impaired or absent, so the plasmatic concentration of star fruit neurotoxin would increase until a serum level which could significantly cross the blood-brain-barrier. In the central nervous system, the star fruit neurotoxin would probably bind to GABAergic receptors inducing excitotoxicity, which would culminate in seizures and, probably, other symptoms of the star fruit intoxication. It is interesting to notice that the characterization of the cellular and molecular mechanisms of the star fruit intoxication needs to pass through a group of different experimental protocols among them in vivo and the in vitro bioassays. The correlation between in vivo and in vitro models is then more complex that we should think it is [53]. Thus, the particular case of synaptosomes and GABA and glutamate release and re-uptake, shows neurochemical dynamics associated to star fruit intoxication mechanisms in a preparation which consists of isolated synaptic terminals (44). However, additional studies are needed with brain slices from control brains treated with the AcTx and even the use of ex vivo models (in vitro bioassays from tissue after in vivo experiments), for example, in our case, brain slices from treated animals. 910

Conclusions All observations in our reports and reports from others show that star fruit intoxication may be harmful and even life threatening in patients with renal failure on supportive or dialytic treatment. Hiccups and vomiting, which are common symptoms, could be used as an indication of star fruit intoxication in renal patients presenting with neurological and consciousness disturbances that have no apparent cause. Hemodialysis, especially on a daily basis is an effective treatment for star fruit intoxication in the majority of cases, if started earlier, according to the literature. In severe cases, continuous methods of replacement may provide a superior initial procedure. In those severe cases that do not respond to these methods, hemoperfusion could be an effective and fast method of treatment. Peritoneal dialysis is useless as a treatment, especially when consciousness disorders ensue. Although there are not many described cases of star fruit oxalate nephrotoxicity, it seems reasonable to avoid consumption of large amounts of star fruit juice especially on an empty stomach and chronic consumption in patients with underlying chronic renal disease. The star fruit toxin seems to be a small molecule (molecular weight less than 500), differing from oxalate and common amino acids. It is a potent neurotoxin able to induce seizures and death. Preliminary assays, in synaptosomes, indicated that star fruit neurotoxin binds to GABA receptors. Acknowledgements We thank Prof. Dr. Joaquim Coutinho Netto, from the Biochemistry Department, Faculty of Medicine of Ribeirão Preto, University of São Paulo (FMRP-USP), for the investigations of biochemical aspects of the star fruit neurotoxin. To Flavio Del Vecchio for the video-EEG recordings as well as to José Antonio Cortes Oliveira for the behavioral evaluations, both from the Neurophysiology and Experimental Neuroethological Laboratory, Department of Physiology, FMRP-USP. We also thank FAPESP-Brazil (proc. 2003/00873-2), Cinapce-FAPESP (Proc 05/56447-7) and CNPq-Brazil (proc 473448/2003-3) for their financial support to the star fruit projects held at the Neurophysiology and Experimental Neuroethology Laboratory at the FMRP-USP.

43. Star fruit

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Morton JF. Fruits of warm climates. Flair Books, Miami, FL, 1987:pp125-128. Margen S. The Wellness Encyclopedia of Food and Nutrition. Health Letter Assoc. New York, NY; 1992:pp271-272. Martin LC, Caramori JST, Barretti P, Soares VA. Soluço intratável desencadeado por ingestão de carambola (averrhoa carambola) em portadores de insuficiência renal crônica. J Bras Nefrol 1993; 15:92-94. Neto MM, Robl F, Netto JC. Intoxication by star fruit (averrhoa carambola) in six dialysis patients? ( Preliminary report). Nephrol Dial Transpl 1998;13:570-572. Chang JM, Hwang SJ, Kuo HT, Tsai JC, Guh JY, Chen HC, Tsai JH, Lai YH. Fatal outcome after ingestion of star fruit (averrhoa carambola) in uremic patients.Am J Kidney Dis 2000; 35:189-193. Wang JL, Cheng CH, Wu MJ, Chen CH, D, Shu KH. (In Chinese) Status epilepticus in two patients with chronic renal failure after ingestion of star fruit. Kidney Dialysis 2000; 8:166-169. Lo KL, Tong GMW, Wong PN, Mak SK, Wong AKM. Persistent hiccup in a continuous ambulatory peritoneal dialysis patient following ingestion of star fruit. Hong Kong Journal of Nephrology 2001; 3(1):45-46. Ho MP, Pai MF, Fan CD, Chang KS. Neurological Manifestations Induced by Star fruit (Averrhoa carambola) Intoxication in Patients with Chronic Renal Failure: Two Cases Report. J Taiwan Emerg Med 2001; 3:2-7 Wu CW, Denq JC, Tsai WS, Lin SH. Star fruit-induced neurotoxicity in two patients with chronic renal failure. J Med Sci 2002; 22(2). Yap HJ, Chen YC, Fang JT, Huang CC. Star fruit: a neglected bu serious fruit intoxicant in chronic renal failure. Dialysis & Transplantation. 2002; 31:564. Chan YL, Leung CB, Yeung DKW. Phosporus and single voxel proton MR spectroscopy and diffusion-weighted imaging in a case of star fruit poisoning. Am J Neuroradiol 2002; 23: 1557-1560. Tse KC, Yip PS, Lam MF, Choy BY, Li FK, Lui SL, Lo WK, Chan TM. Star fruit intoxication in uraemic patients: case series and review of the literature. Internal Medicine Journal 2003; 33:314-316. Neto MM, Costa JAC, Garcia–Cairasco N, Netto JC, Nakagawa B, Dantas M. Intoxication by star fruit (averrhoa carambola) in 32 uraemic patients: treatment and outcome. Nephrol Dial Transpl 2003; 18: 120-125. Moyses Neto M, Nardim MEP, Vieira-Neto OM, Vannuchi MTI, Raspanti EO. Intoxicação por carambola (averrhoa carambola) em quatro pacientes renais crônicos pré-dialíticos e revisão da literatura. J Bras Nefrol 2003; 26:228-232 Hung SW, Lin ACM, Chong CF, Wang TL, Ma HP. Fatal outcome after Star fruit (Averrhoa carambola) ingestion in patients with chronic renal insufficiency. Ann Disaster Med 2004; 3:56-59. Chang CH, Yeh JH. Non-convulsive status epilepticus and consciousness disturbance after star fruit (Averrhoa carambola) ingestion in a dialysis patient. Nephrology 2004; 9:362-365 Tsai MH, Chang WN, Lui CC, Chung KJ, Hsu KT, Huang CR, Lu CH, Chuang YC. Status epilepticus induced by star fruit intoxication in patients with chronic renal disease. Seizure 2005; 14:521-525 Chen LL, Fang JT, Lin JL. Chronic renal disease patients with severe star fruit poisoning: hemoperfusion may be an effective alternative therapy. 2005; 43:181-183 Wu MY, Wu IW, WuSS, Lin JL. Hemoperfusion as an effective alternative therapy for star fruit intoxication: a report of 2 cases. Am J Kidney Dis 2007; 49:E1-E5 Wang YCL, Liu BM, Supernaw RB, Lu YH, Lee PY. Management of star fruit-induced neurotoxicity and seizures in a patient with chronic renal failure. Pharmacotherapy 2006; 26:143-146 Chien CL, Fang HC, Chou KJ, Wang JS, Chung HM. Acute oxalate nephropathy after ingestion of star fruit. Am J Kidney Dis 2001; 37:418-422 Hruska KA, Seltzer JR, Grieff M: Nephrolithiasis, in Schrier RW, Gottschalk CW(eds): Diseases of the Kidney(ed 6). New York, Little Brown, 1977, pp 739-764. Sanz P. Reig R: Clinical and pathological findings in fatal plant oxalosis: A review. Am J Forensic Med Pathol, 1992; 13: 342-345 Fang HC, Chen CL, Wang JS, Chou KJ, Chiou YS, Lee PT, Yeh JH, Yeh MY, Chung HM. Acute oxalate nephropathy induced by star fruit in rats. Am J Kidney Dis 2001; 38:876-880. Fang HC, Lee Po-Tsang, Lu PJ, Chen CL, Chang TY, Hsu CY, Chung HM, Chou KJ. Mechanisms of star fruit-induced acute renal failure. Food Chem Toxicol 2008; 46:1744-1752. Niticharoenpong K, Chalermsanyakorn P, Panvichian R, Kitiyara C. Acute deterioration of renal function induced by star fruit ingestion in a patient with chronic kidney disease. J Nephrol 2006; 19:682-686 Muir, CK, Lam, CK. Depressant action of Averrhoa carambola. Med J Malaysia. 1980; XXXIV: 279-80. Chen CL, Chou KJ, Wang JS, Yeh JH, Fang HC, Chung HM. Neurotoxic effects of carambola in rats: the role of oxalate. J Formos Med Assoc. 2002; 101(5): 337-41.

911

NETO, CAROLINO, LOPES & GARCIA-CAIRASCO

29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.

45. 46. 47. 48. 49. 50. 51.

52. 53.

912

Watt JM, Breyer-Brandwijk MG. The medicinal and poisonous plants of Southern and Eastern Africa. 2nd edition, Edinburgh, Livingstone, 1962. Fang HC, Chen CL, Lee PT, Hsu CY, Tseng CJ, Lu PJ, Lai SL, Chung HM, Chou KJ. The role oif oxalate in star fruit neurotoxicity of Five-sixths nephrectomized rats. Food Chem Toxicol 2007; 45:1746-1749. Wilson CW, Shaw PE, Knight-Jr RJ. Analysis of oxalic acid in carambola (Averrhoa carambola L.) and spinach by High-performance liquid chromatography. J. Agric Food Chem 1982; 30(6): 1106-8. Oliveira RB, Godoy SAP, Costa, FB. Plantas tóxicas: conhecimento e prevenção de acidentes. Ribeirão Preto, Holos, 1. ed., 2003. p64 Wilson CW Shaw PE, Kinght RJ, Nagy S, Klim M. Volatile constituents of carambola (Averrhoa carambola L.). J. Agric. Food Chem.1985; 33:199-201. Artain N, Maárifa Y, Hanafi M. Isolation and identification of active antioxidant compound from star fruit (Averrhoa carambola) mistletoe (Dendrophthoe pentandra (L.) Miq.) ethanol extract J Appl Science 2006; 8: 1659-1663. Shui G, Leong, Lai P. Residue from star fruit as valuable source for functional food ingredients and antioxidant nutraceutical. Food Chemistry 2006; 97: 277-284 Gross JIR., Eckhardt G. Carotenoids of the fruit of Averrhoa carambola, Phytochemistry 1983; 22: 1479-1481 Gunasegaran R. Flavonoids and anthocyanins of three oxalidaceae, Fitoterapia, 1992, 63: 89-90. Jabbar A, Taleb M, Rashid MA, Hasan CM. 5-Hydroxymethyl-2-furfural from Averrhoa carambola Fitoterapia 1995; 66: 377 Shui G, Leong LP. Analysis of polyphenolic antioxidants in star fruit using liquid chromatography and mass spectrometry. J Chromatogr A 2004; 1022: 67-75 Guaratini T, Vessecchi RL, Lavarda FC, Maia Campos PM, Naal Z, Gates PJ, Lopes NP. New chemical evidence for the ability to generate radical molecular ions of polyenes from ESI and HR-MALDI mass spectrometry. Analyst 2004;129: 1223-1226 Guaratini T, Vessecchi R, Pinto E, Colepicolo P, Lopes NP. Balance of xanthophylls molecular and protonated molecular ions in electrospray ionization. J Mass Spectrom 2005; 40: 963-968. Guaratini T, Gates PJ, Cardozo KH, Campos PM, Colepicolo P, Lopes NP. Letter: Radical ion and protonated molecule formation with retinal in electrospray and nanospray. Eur J Mass Spectrom 2006; 12:71-74. Guaratini T, Lopes NP, Pinto E, Colepicolo P, Gates PJ. Mechanism for the elimination of aromatic molecules from polyenes in tandem mass spectrometry. Chem Commun 2006;39: 4110-4112. Carolino RO, Beleboni RO, Pizzo AB, Vecchio FD, Garcia-Cairasco N, Moyses-Neto M, Santos WF, Coutinho-Netto J. Convulsant activity and neurochemical alterations induced by a fraction obtained from fruit Averrhoa carambola (Oxalidaceae: Geraniales). Neurochem Int. 2005; 46: 523-31. Bradford HF. Glutamate, GABA and epilepsy. Prog Neurobiology 1995; 47:477-511. Engelborghs S, D’Hooge R, De Deyn PP. Pathophysiology of epilepsy. Acta Neurol Belgica 2000; 100: 201-13. Gray EG, Whittaker VP. The isolation of nerve ending brain: and electron-microscopic study of cell fragments derived by homogenization and centrifugation. J Anat 1962; 96:79-87 Coimbra TM, Cielisnki DA, Humes HD. Epidermal growth factor accelerates renal repair in mercuric chloride nephrotoxicity. Am.J.Physiol 1990; 259: F438-F443. Paxinos, G., Watson, C.H., 1998. The rat brain in stereotaxic coordinates, Academic Press, New York, 256p. Moraes MF, Galvis-Alonso OY, Garcia-Cairasco N. Audiogenic kindling in the Wistar rat: a potential model for recruitment of limbic structures. Epilepsy Research 2000; 39:251-259. Garcia-Cairasco N, Del Vecchio F, Oliveira JAC, Dantas M, Moyses-Neto M. A video-EEG study of the neurotoxic effects of star fruit (Averrhoa carambola) juice ingestion in animals with acute renal failure. Proceedings of the 39th European Renal Association and the European Dialysis and Transplantation Meetings, Copenhagen, Denmark. Nephrol Dial Transpl 2000; 17: 89 Leite JP, Garcia-Cairasco N, Cavalheiro EA. New insights from the use of pilocarpine and kainate models. Epilepsy Research 2002; 50: 93-103 Tiffany-Castiglioni E, Ehrich M, Dees L, Costa LG, Kodavanti PRS, Lasley SM, Oortgiesen M, Durham HD. Bridging the gap between in vitro and in vivo models for neurotoxicology. Toxicol Sci 1999; 51:178-183.

44

Drug dosage in renal failure Ali J. OLYAEI1 and William M. BENNETT2 1Oregon 2Legacy

Health Sciences University, Portland, Oregon, USA Good Samaritan Hospital, Portland, Oregon, USA

Introduction ___________________________________________________________ 914 Principles of pharmacokinetics in uremia ____________________________________ 914 Absorption Volume of distribution Protein binding Eliminations Dosing regimens

915 915 915 916 916

Therapeutic drug monitoring ______________________________________________ 916 Dialysis and drug clearance _______________________________________________ 916 References _____________________________________________________________ 918 Appendix: Drug dosing in renal failure _____________________________________ 919 Table 2. Antibacterial agents _________________________________________ 920 Table 3. Analgesic agents ____________________________________________ 926 Table 4. Antihypertensive and cardiovascular agents______________________ 927 Table 5. Endocrine and metabolic agents _______________________________ 932 Table 6. Gastrointestinal agents _______________________________________ 933 Table 7. Neurologic and anticonvulsant agents __________________________ 934 Table 8. Rheumatologic agents _______________________________________ 937 Table 9. Sedative agents _____________________________________________ 939 Table 10. Anti-Parkinson agents_______________________________________ 940 Table 11. Antipsychotic agents _______________________________________ 941 Table 12. Corticosteroid agents _______________________________________ 942 Table 13. Anticoagulant agents _______________________________________ 943

OLYAEI & BENNETT

Introduction

E

volution within the field of dialysis and advances in surgical procedures providing for superior access for shunt placement have made it possible to treat patients with end-stage renal disease with dialysis therapy for more than 50 years. Improved pharmacotherapy of pre- and post- dialysis has also contributed to these remarkable advancements. Drug therapy is also evolving. Health care provider for patients with renal diseases needs to understand the latest drug therapy and ensure appropriateness of therapy in each individual patient [1]. Renal insufficiency and dialysis alter the pharmacokinetics and pharmacodynamics of most commonly used drugs. Of note, an average of eight different classes of drugs per patient are prescribed in patients with renal failure. In comparison to the general population, patients with renal insufficiency experience significantly more adverse drug reactions. Therefore, clinicians caring for these patients must be familiar with the pharmacokinetic behavior of each agent and of the impact of renal failure on the drug elimination process. Understanding the time course of pharmacotherapy is based on knowledge of the relationship between drug concentration and effect. Drugs act by affecting biochemical and physiological processes in the body. Most drugs act at specific receptors but may produce multiple effects because of the location of the receptor in various organs. Knowledge of these properties helps to predict the behavior of a drug in the body and is an important guide in the selection of appropriate doses and dosage intervals. A particular area of concern is that many patients with renal insufficiency are elderly which in itself may effect drug disposition. Most therapeutic agents or their metabolites are completely or partially eliminated by the kidneys. In patients with renal insufficiency both metabolism and elimination is impaired, therefore, these patients are at a greater risk of adverse drug reactions or drug toxicity. Because of co-morbid conditions, most patients with advanced renal diseases require multiple medications for the treatment of hypertension, hyperlipidemia, hyperuricemia and congestive heart failure. Patients with chronic renal failure are at a greater risk of drug-drug interactions. Finally, depending on various factors such as the size of the drug molecule and degree of protein binding, a

914

significant amount of drug removal may occur during dialysis. Most drug dosages may be adjusted based on plasma therapeutic concentrations (Table 1). To prevent toxicity and optimize efficacy, it is critical that these factors be taken into account and appropriate dosage adjustments made when prescribing drugs for dialysis patients [2-12]. This chapter discusses the pharmacological principles for prescribing drugs in this population and provides specific dosage recommendations (Tables 2-13).

Principles of pharmacokinetics in uremia Pharmacokinetics is the study of drug behavior (absorption, distribution, metabolism and elimination) in the body. The ability of the body to remove a drug is called clearance. Clearance indicates the intrinsic ability of the body to decrease plasma drug concentration. The three major processes effecting drug clearance are metabolism by the liver, metabolism by the gastrointestinal tract (cytochrome P-450 and P-glycoproteins) or elimination and metabolism by the kidney. At steady state, the overall rate of clearance is equal to rate of drug absorption [1]. The important elements of a drug’s pharmacokinetics are shown in Figure 1. All important pharmacokinetic parameters, such as drug absorption, volume of distribution, protein-protein binding, and drug metabolism must be considered when dose modifications are made in uremic patients. For example, gastroparesis in diabetic patients, slow gastric emptying time and edema of gastrointestinal tract in patients with advanced renal failure may affect drug absorption. Iron preparations and phosphate binders may also alter drug absorption [2]. Tissue and receptor sites Bound

Absorption

Free

Free drug

Excretion

Systemic circulation

Biotransformation

Figure 1. Interrelationship of absorption, distribution, biotransformation and excretion.

44. Drug dosage in renal failure

Absorption Following oral drug administration, only a certain proportion of the drug is absorbed reaching systemic circulation (F or bioavailability). The percentage of a drug dose that appears in the systemic circulation following oral administration compared with the intravenous route for the same drug defines its bioavailability. In general, drugs given by the intravenous route reach the central compartment directly and usually have a more rapid onset of action. Drugs given by other routes must pass through a series of biologic membranes before entering the systemic circulation. For many drugs, only a fraction of the administered dose may reach the circulation to exert any pharmacodynamic effect [3]. Chronic renal failure may influence drug absorption and bioavailability. The dissolution rate, chemical forms, rout of administration, the gastrointestinal stability and dosage form may alter drug’s bioavailability. Bioavailability only indicates the extent of drug absorption not the rate of drug absorption. Drugs can be highly bound to plasma proteins (e.g. aspirin) or unbound (free active moiety). Only the free or unbound concentration of the drug interacts with specific receptors at the site of pharmacologic action. The liver can either metabolize drugs in the ‘first pass’ as the drug is absorbed into the portal circulation, or later when the drug is delivered to the liver via the systemic blood flow prior to reaching systemic circulation. First pass metabolism can significantly reduce the rate and extent of drug absorption. For renal failure patients, gastric pH is often high due to the use of antacids or anti-ulcer medications that may result in decreased absorption of medications requiring an acid milieu. Aluminum- or calcium-containing antacids may also form non-absorbable chelation products with certain drugs, such as digoxin or tetracycline and impair these agents’s absorption [4-6].

Volume of distribution Following drug absorption, individual drugs distribute throughout the body in a characteristic manner. The apparent volume of distribution (Vd) is the quantity of drug in the body (L/kg body weight) divided by the plasma concentration at steady state. Volume distribution also represents the amount of water that is needed for a drug to dissolve to reach an

observed plasma concentration. Therefore, lipophilic agents or drugs with high tissue binding capacity most commonly have a large volume of distribution. In contrast, drugs with high circulating protein binding or water-soluble drugs have a small volume of distribution. Drugs that are largely confined to the intravascular compartment usually have a volume of distribution less then 0.2 L/kg. Uremia, edema and renal failure may alter the volume of distribution of most commonly used agents in patients with renal insufficiency [7-9]. Changes in volume of distribution are usually not clinically significant except for those drugs which have a small volume of distribution under normal circumstances (i.e., >0.7 L/kg).

Protein binding Unbound or free drugs are pharmacologically active. Therefore, the degree of protein binding is an important issue in adjustment of drug dosing in renal failure. Low plasma albumin or increase in plasma albumin can potentially increase the pharmacodynamic effects of highly bound drugs. Organic acids usually have a single binding site on albumin whereas organic bases tend to have multiple binding sites and their behavior in the presence of increasing renal insufficiency is less predictable. In general, acidic drugs have reduced plasma protein binding in patients with renal failure; this reduction is attributable to a combination of decreased albumin concentration and a reduction in albumin affinity, which is, in turn, influenced by either structural changes in the albumin molecule or accumulation of competing endogenous inhibitors of protein binding. For some agents with high protein binding, the reduced sites or decreased plasma protein can cause potentially important pharmacologic consequences. For example in patients with renal failure, the free plasma concentration of phenytoin increases from 0.1 to 0.35. Therefore, the observed plasma concentration of 4 mg/L is comparable to 10-15 mg/L in patients with normal renal function. Finally only unbound or free drugs are available for drug metabolism or excretion. Uremia decreases binding capacity of most drugs and result in increased metabolism in patients with renal failure. For any given drug therapeutic concentration (bound plus unbound), the proportion of free or active drug is increased. It is more desirable to obtain free drug plasma concentrations in patients with renal 915

OLYAEI & BENNETT

failure [10-13].

Eliminations The presence of progressive renal insufficiency affects most body biochemical processes including drug biotransformation. In addition, some drugs have pharmacologically active metabolites, which, although unimportant in patients with normal renal function, may accumulate in patients with renal insufficiency causing adverse drug reactions [13-17]. Some of these pharmacologically active metabolites may account for the high incidence of adverse drug reactions in patients with renal failure. Some of the best-known examples of this phenomenon are the accumulation of pharmacologically active metabolites of meperidine causing seizures, nitrofurantoin causing peripheral neuropathy and morphine sulfate causing excess respiratory depression. The metabolic biotransformation of drugs to another more water-soluble chemical moiety also may be altered in uremia. In patients with renal failure, chemical reduction, acetylating, ester or peptide hydrolysis may be delayed, whereas metabolism by hepatic microsomal oxidation is usually normal. Drug elimination rate is usually expressed as elimination half-life (t2). Drug half-life is the time required for the plasma concentration to decrease by 50%. The half-life is dependent upon Vd and clearance (renal, hepatic, or other) as expressed by the formula: t2 = 0.693 x Vd/clearance For drugs eliminated primarily intact through the kidneys, as the renal clearance decreases, t2 will increase (assuming that Vd is unchanged). It should be noted that active drug metabolites may also be excreted by the kidney and therefore have a prolonged half-life in renal failure.

Dosing regimens Most drugs or their metabolites that are normally excreted unchanged by the kidney will require dosage modification in advanced renal failure. The loading dose of a drug will stay the same unless the Vd is significantly altered. The maintenance regimen may be modified by the interval extension method or dosage reduction. The interval extension method utilizes the same dose at greater intervals and is useful for drugs 916

with long half-lives. The dosage reduction method reduces the dosage and leaves the interval between doses unchanged. This method generally leads to more constant serum levels. Therapeutic drug monitoring is a useful method in guiding drug therapy and preventing toxicity. Interpretation of drug levels must be made in light of the amount of drug given; the time elapsed since the last dose, and the route of administration and clinical scenario of the patients [14-18].

Therapeutic drug monitoring Dosage and interval modifications do not necessarily protect against drug toxicity. Therefore, monitoring drug levels in some specific agents with a narrow therapeutic window is essential in the patient with renal impairment. In order to correctly interpret therapeutic drug monitoring it is important to know the exact time when a dose given, the route of administration, time since the last dose and the particular drug’s half-life. Peak drug levels represent the highest drug concentration achieved after initial rapid distribution and in most drugs predict overall drug efficacy. Trough drug levels are obtained immediately before the next dose, represent the lowest serum concentration and predict drug toxicity and accumulations. Drug level monitoring can be expensive and is not always available. Drug level monitoring does not always reduce the incidence of toxicity. Aminoglycoside antibiotics, for instance, can concentrate in tissues such as the inner ear and renal cortex and toxicity is not always correlated with high blood levels. Ongoing clinical assessment is important even when drug levels are within the established therapeutic range. In the presence of metabolic acidosis or hypokalemia, digoxin toxicity may occur despite acceptable therapeutic levels. Most assays do not distinguish between free and protein-bound drug in the plasma. An increase in unbound drug is common in patients with renal failure. Table 1 summarized the therapeutic drug monitoring in renal insufficiency for drugs which monitoring of drug levels is routinely recommended.

Dialysis and drug clearance Dialysis and renal replacement therapy (RRTx) are common treatments option for the treatment of acute renal failure in the hospital setting. Many

44. Drug dosage in renal failure

drugs are substantially cleared by dialysis. Therefore, scheduling of drug administration and the possibility of dosage supplementation should be considered in patients receiving dialysis. Scheduled doses should be given upon completion of dialysis therapy. If this is not possible and dialytic treatment increases total body clearance of a given drug by greater than 30%, dosage supplementation may be necessary. Dialyzability is primarily determined by molecular weight (< 500 daltons), water solubility of the drug and extent of protein binding (unbound drugs are more readily

cleared). Other factors of a drug which determine dialyzability include Vd, non-renal excretion, ionic charge and erythrocyte partitioning. Some properties of the dialysate and the dialyzer membrane also affect drug clearance and include flow rate, temperature, pH, solute composition, volume (peritoneal dialysis), pore size and surface area. Creatinine clearance rates of up to 30 to 50 ml/min are currently being achieved with continuous renal replacement therapies (CRRTs) such as continuous venovenous hemofiltration and continuous arteriov-

Table 1. Therapeutic drug monitoring. Drug name Aminoglycosides (Conventional dosing) Gentamicin, Tobramycin, Amikacin

When to draw sample Trough: Immediately prior to dose Peak: 30 min after a 30-45 min infusion

How often to draw levels Check peak and trough with 3rd dose For therapy less than 72 h, levels not necessary. Repeat drug levels weekly or if renal function changes

Aminoglycosides (24-h dosing) Gentamicin, Tobramycin, Amikacin Carbamazepine

Therapeutic range Gentamicin and Tobramycin: Trough: 0.5–2 mg/L Peak: 5–8 mg/L Amikacin: Peak: 20–30 mg/L Trough: < 10 mg/L 0.5–3 mg/L

Obtain random drug level 12 h after dose

After initial dose. Repeat drug level in 1 week or if renal function changes

4–12 mcg/mL

Trough: Immediately prior to dosing

Cyclosporin Digoxin

150–400 ng/mL 0.8–2.0 ng/mL

Trough: Immediately prior to dosing 12 h after maintenance dose

Check 2–4 days after first dose or change in dose Daily for first week, then weekly. 5–7 days after first dose for patients with normal renal and hepatic function; 15–20 days in anephric patients

Lidocaine Lithium

1–5 mcg/mL Acute: 0.8–1.2 mmol/L Chronic: 0.6–0.8 mmol/L 15–40 mcg/mL 10–20 mcg/mL 1–2 mcg/mL 4–10 mcg/mL Trough: 4 mcg/mL Peak: 8 mcg/mL 10–30 mcg/mL

8 h after i.v. infusion started or changed Trough: Before a.m. dose at least 12 h since last dose Trough: Immediately prior to dosing Check 2 weeks after first dose or change in dose. Follow-up level in 1–2 months. Trough: Immediately prior to dosing 5–7 day after first dose or after change in dose Trough: Immediately prior to next dose or 12–18 h after starting or changing an infusion Draw with procainamide sample

1–5 mcg/mL 10–20 ng/dL 10–15 ng/mL 15–20 mcg/mL

Trough: Immediately prior to next dose Trough: Immediately prior to next dose Trough: Immediately prior to next dose Trough: Immediately prior to next dose

40–100 mcg/mL

Trough: Immediately prior to next dose

Trough: 5–15 mg/L Peak: 25–40 mg/L

Trough: Immediately prior to dose Peak: 60 min after a 60 min infusion

Phenobarbital Phenytoin Free Phenytoin Procainamide NAPA (n-acetyl procainamide) a procainamide metabolite Quinidine Sirolimus Tacrolimus(FK-506) Theophylline p.o. or Aminophylline i.v. Valproic acid (divalproex sodium) Vancomycin

Daily for first week, then weekly

Check 2–4 d after first dose or change in dose With 3rd dose (when initially starting therapy, or after each dosage adjustment). For therapy less than 72 h, levels not necessary. Repeat drug levels if renal function changes

917

OLYAEI & BENNETT

enous hemofiltration. CRRTs are enjoying widespread application in both medical and surgical intensive care units. Limited data is available regarding drug removal during CRRT. Dosage adjustments can be determined through close monitoring of drug levels and clinical status of the patient. During CRRT solutes and drugs are removed by convective transport. Drugs also may

be substantially removed by membrane-drug binding. Drugs and solutes not bound to plasma proteins and dissolved in the plasma cross the dialysis membrane through plasma water ultrafiltration. The ulltrafiltrate drug concentration is equal to the plasma concentration multiplied by the percentage of unbound drug.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

918

Winters ME, Basic clinical pharmacokinetics. Winter ME, Applied Therapeutic Inc. Vancouver, WA 1994. Zhang Y. Benet LZ. The gut as a barrier to drug absorption: combined role of cytochrome P450 3A and P-glycoprotein. Clinical Pharmacokinetics. 40(3):159-68, 2001 Olyaei AJ, Bennett WM. “The Effect of Renal Failure on Drug Handling”. Oxford Textbook of Critical Care, 1996. Benedetti P. de Lalla F. Antibiotic therapy in acute renal failure. Contributions to Nephrology. (132):136-45, 2001. Olyaei AJ, deMattos AM, Bennett WM. “Prescribing Drugs in Renal Failure”. Brenner and Rector’s, The Kidney, 2000. Olyaei AJ, deMattos AM, Bennett WM. “Drug-drug Interaction and most Commonly Used Drug in Transplant Recipients” . Primer on Transplantation, American Society of Transplantation. 2000 Olyaei AJ “ Drug Dosing in Renal Failure ” in Treatment Strategies In Nephrology and Hypertension. By Bennett 1st edition. PocketMedicine.com Inc 2001 Olyaei AJ, deMattos AM, Bennett WM “Principle of Drug Usage in Dialysis Patients” Dialysis Therapy, Hanley & Belfus Inc. Medical Publishers. May 2001 Olyaei AJ, deMattos, Bennett WM, Drug usage in dialysis patients. In: Nissenson AR, Fine RN, Gentile DE, eds. Clinical dialysis. Norwalk, NJ: Appleton and Lange; 2001. Bennett WM. Guide to drug dosage in renal failure. In: Speight TM, Holford N, eds. Avery’s drug treatment, 4th edn. Auckland: ADIS International; 1997:1725–92. Beauchamp D, Labrecque G. Aminoglycoside nephrotoxicity: do time and frequency of administration matter?. Current Opinion in Critical Care 2001:7: 401-408. Livornese LL, Jr., Slavin D, Benz RL, Ingerman MJ, Santoro J. Use of antibacterial agents in renal failure. Infectious Disease Clinics of North America 2001:15: 983-1002. Chertow GM, Lee J, Kuperman GJ et al. Guided medication dosing for inpatients with renal insufficiency. JAMA 2001:286: 28392844. Bugge JF. Pharmacokinetics and drug dosing adjustments during continuous venovenous hemofiltration or hemodiafiltration in critically ill patients. Acta Anaesthesiologica Scandinavica 2001:45: 929-934. Izzedine H, Launay-Vacher V, Baumelou A, Deray G. An appraisal of antiretroviral drugs in hemodialysis. [Review] [47 refs]. Kidney International 2001:60: 821-830. Subach RA, Marx MA. Drug dosing in acute renal failure: the role of renal replacement therapy in altering drug pharmacokinetics. Advances in Renal Replacement Therapy 1998:5: 141-147. Joy MS, Matzke GR, Armstrong DK, Marx MA, Zarowitz BJ. A primer on continuous renal replacement therapy for critically ill patients. Annals of Pharmacotherapy 1998:32: 362-375. Joos B, Schmidli M, Keusch G. Pharmacokinetics of antimicrobial agents in anuric patients during continuous venovenous haemofiltration. Nephrology Dialysis Transplantation 1996:11: 1582-1585.

44. Drug dosage in renal failure

Appendix: Drug dosing in renal failure Abbreviations used: ACE: angiotensin-converting enzyme AV: atrioventricular BUN: blood urea nitrogen CCr: Creatinine clearance CAPD: continuous ambulatory peritoneal dialysis CHF: congestive heart failure CMV: cytomegalovirus CNS: central nervous system CRRT: continuous renal replacement therapy CSA/FK: cyclosporine A and tacrolimus CVD: cardiovascular disease CVVH: Continuous venovenous hemofiltration DVT: deep vein thrombosis ESRD: end-stage renal disease GI: gastrointestinal GFR: glomerular filtration rate HBV: hepatitis B virus HD: hemodialysis HDL: high-density lipoprotein HIT: heparin-induced thrombocytopenia HSV: herpes simplex virus INR: international normalized ratio IV: intravenous MI: myocardial infarction MMF: mycophenolate mofetil NA: not applicable NC: No data: no change required NSAIDs: nonsteroidal anti-inflammatory drugs TB: tuberculosis TDM: therapeutic drug monitoring VD: volume of distribution VZV: varicella zoster virus.

919

920

Normal dosage

% of drug Dosage adjustment for renal failure with GFR (ml/min): excreted >50 10−50