Infectious Diseases of the Female Genital Tract, 4th edition: by Richard L., Md. Sweet, Ronald S., MD Gibbs By Lippincot...
74 downloads
3473 Views
17MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Infectious Diseases of the Female Genital Tract, 4th edition: by Richard L., Md. Sweet, Ronald S., MD Gibbs By Lippincott Williams & Wilkins Publishers (January 15, 2002)
By OkDoKeY
Infectious Diseases of the Female Genital Tract CONTENTS Editors Acknowledgment Dedication Preface
INTRODUCTION 1 Clinical Microbiology of the Female Genital Tract 2 Use of the Microbiology Laboratory in Infectious Diseases SPECIAL ORGANISMS 3 roup B Streptococci 4 Genital Mycoplasmas 5 Chlamydial Infections 6 Herpes Simplex Virus Infection 7 Sexually Transmitted Diseases 8 Mixed Anaerobic-Aerobic Pelvic Infection and Pelvic Abscess 9 Hepatitis Infection 10 HIV-AIDS GYNECOLOGIC AND OBSTETRIC INFECTIONS
11 Toxic Shock Syndrome 12 Infectious Vulvovaginitis 13 Postabortion Infection, Bacteremia, and Septic Shock 14 Pelvic Inflammatory Disease 15 Urinary Tract Infection 16 Perinatal Infections 17 Premature Rupture of the Membranes 18 Intraamniotic Infection 19 Subclinical Infection as a Cause of Premature Labor 20 Postpartum Infection 21 Wound and Episiotomy Infection 22 Parasitic Disease in Pregnancy ANTIBIOTICS/ANTIVIRALS 23 Antimicrobial Agents 24 Antibiotic Prophylaxis in Obstetrics and Gynecology IMMUNIZATION 25 Immunization
EDITORS RICHARD L. SWEET, M.D. The Milton Lawrence McCall Professor and Chair Department of Obstetrics, Gynecology, and Reproductive Sciences University of Pittsburgh School of Medicine Magee-Women's Hospital Pittsburgh, Pennsylvania RONALD S. GIBBS, M.D. Professor and Chair Department of Obstetrics and Gynecology University of Colorado Health Sciences Center Denver, Colorado
ACKNOWLEDGMENT We acknowledge the administrative and editorial assistance of Phyllis Catrell, Jane Cook, and Susan Kostilnik in the preparation of this edition.
DEDICATION This book is dedicated to our wives, Rhea and Jane, for their continued love, support and understanding. To our children, Jennifer, Suzanne, Andrew, Eric, and Stuart and our grandchildren Hanna, Dylan, Benjamin and Emily and future grandchildren who provide us with the ongoing spark and impetus for our continued commitment to this work.
PREFACE During the writing of this fourth edition of Infectious Diseases of the Female Genital Tract, infections seem to have played an ever-expanding role, not only in the practice of obstetrics and gynecology but in the practice of medicine in general. Regularly the news media produced stories about new infections and antibiotic resistance. These, indeed, continue to be challenging and exciting times in the world of infectious diseases. For this fourth edition, we have maintained the goal of providing physicians with up-to-date knowledge that—in a highly readable form—deals with infectious diseases in the female. We have substantially revised and updated every chapter and have added several eye-appealing features. We are especially pleased that accompanying this fourth edition will be an atlas to add to the usefulness of this text. We have been gratified by the place this text has earned in the offices and libraries of many obstetricians and gynecologists, and we thank the readers for their support.
1 CLINICAL MICROBIOLOGY OF THE FEMALE GENITAL TRACT Infectious Diseases of the Female Genital Tract
1 CLINICAL MICROBIOLOGY OF THE FEMALE GENITAL TRACT Virulence Aerobic Organisms Gram-Positive Cocci Gram-Positive Bacilli Gram-Negative Bacilli Gram-Negative Cocci Anaerobic Isolates Gram-Positive Anaerobes Gram-Negative Anaerobes Yeasts and Actinomyces Trichomonads Mycoplasmas Chlamydia Trachomatis Viruses Changes in Vaginal Microflora Age Sexual Activity Contraception Pregnancy and Delivery Surgery Homeostasis Chapter References
The microbiology of the female genital tract is indeed complex. In healthy women, the vagina contains 109 bacterial colony-forming units per gram of secretions. Isolates commonly found in the lower genital tract include a variety of aerobic and anaerobic bacteria, yeast, viruses, and parasites (Table 1.1). Influences upon these microbes include phase of the menstrual cycle, sexual activity, contraceptive use, childbirth, surgery, and antibiotic therapy. The upper genital tract usually is sterile, but bacteria from the lower genital tract may ascend into the uterine cavity, fallopian tubes, or pelvic peritoneum because of menstruation, instrumentation, foreign bodies, surgery, or other predisposing factors.
TABLE 1.1. OVERALL CLASSIFICATION OF MICROORGANISMS FOUND IN THE FEMALE GENITAL TRACT
This chapter presents a working knowledge of genital tract microbes for the clinician. More detailed descriptions of selected microbes are provided in other chapters.
VIRULENCE Distinguishing “virulent” or “pathogenic” isolates from “nonvirulent” or “nonpathogenic” ones often is difficult because the behavior of a given isolate is so dependent upon the numbers of isolates present, host factors, and local conditions (presence of necrosis and foreign body). For example, although group B streptococcus is a leading cause of maternal and neonatal septicemia, most women with genital colonization by group B streptococci (GBS) suffer no consequences. On the other hand, Staphylococcus epidermidis is generally considered a low-virulence organism and is commonly considered part of the normal skin and vaginal flora, but it also may cause disease when conditions allow. For example, S. epidermidis has been recognized as a cause of infective endocarditis of neurologic shunts. Despite such widely ranging behavior patterns, it still is practical to distinguish “high”from “low”-virulence genital isolates. For practicality, the bacteria are divided into aerobes and anaerobes. Each of these groups is subdivided further into Gram-positive and Gram-negative organisms.
AEROBIC ORGANISMS Gram-Positive Cocci The organisms in this group include the aerobic streptococci and staphylococci (Table 1.2).
TABLE 1.2. ANTIBIOTIC THERAPY FOR AEROBIC AND FACULTATIVE BACTERIA FOUND IN FEMALE GENITAL INFECTIONS
Streptococci Many varieties of streptococci have been found in the pelvis, and they have been classified by two independent schemes. In one scheme, the streptococci are distinguished by their hemolytic properties on blood agar plates. Aerobic streptococci showing partial (or green) hemolysis are termed alpha (a) streptococci, those showing complete (or clear) hemolysis are termed beta (b) streptococci, and those showing no hemolysis are termed gamma (g) streptococci. In the other scheme, the Lancefield system, many streptococci can be classified according to surface antigens, designated by the letters A through O. In humans, streptococci of groups A, B, and D are common pathogens. Group A and B streptococci nearly always produce b hemolysis. Group D streptococci usually are nonhemolytic (g streptococci), but on occasion they may be either a or b hemolytic as well. Group A Streptococci (Streptococcus pyogenes) Streptococcus pyogenes causes pharyngeal, cutaneous, puerperal, and postoperative infections and necrotizing fasciitis. Sequelae of group A streptococcal infections may include rheumatic fever and acute glomerulonephritis. It is not generally considered to be a member of the normal vaginal flora, as it is isolated in less than 1% of asymptomatic women. Pelvic infection caused by group A streptococci may produce a characteristic clinical picture, with a high, early initial fever, chills, prostration, and diffuse tenderness. Historically, the group A b-hemolytic streptococcus was the organism responsible for fatal puerperal sepsis. On Gram stains, one sees Gram-positive cocci in chains. The microorganism remains exquisitely sensitive to penicillin, and erythromycin or a cephalosporin usually can be substituted for treatment of the penicillin-allergic patient. Clindamycin is an alternative agent, as is vancomycin. Group A streptococcal pelvic infections occur in both sporadic and epidemic forms. Epidemics usually are exogenous in origin, commonly resulting from nasopharyngeal
carriage of the microorganism or from skin infections in hospital staff members. Occasionally, a mother may be the source. Epidemics can be prevented by placing patients with group A streptococcal infections in strict isolation and by early antibiotic treatment of hospital employees with group A streptococcal infections. Employees with positive cultures should be relieved of duty on obstetric, neonatal, and postoperative wards until their cultures become negative. When epidemics have occurred, isolation and antibiotic therapy have not always sufficed to effect control. In some recent outbreaks, additional measures, such as identifying and treating all streptococcal carriers, canceling elective surgery, and prophylactic treatment of all patients and personnel, have been necessary. Group B Streptococci (Streptococcus agalactiae) Before the 1960s, GBS were not recognized as frequent pathogens, but they have now become a major cause of sepsis among neonates and postpartum women (see Chapter 3) (1,2). Unlike group A streptococci, group B organisms are considered part of the normal vaginal flora and can be recovered in about 20% of normal pregnant women. Isolation rates are enhanced by use of selective broth containing nalidixic acid and gentamicin. The clinical picture of GBS infection in puerperal women closely resembles that of group A infection. Epidemic GBS disease has not been reported among mothers, however. The neonate with GBS sepsis usually acquired the microorganism from the maternal genital tract. Even with appropriate therapy, early-onset neonatal GBS infection has a high fatality rate. In 1996, national guidelines were established to prevent GBS perinatal sepsis (see Chapter 3) (1,2). Group B streptococci are susceptible to penicillin, ampicillin, and the cephalosporin group of antibiotics. Although erythromycin and clindamycin are considered alternatives to penicillin, recent reports have documented a rising resistance rate among GBS (3). Group D Streptococci This group is composed of two subgroups: “group D enterococci” and “group D not enterococci.” The former, which includes Streptococcus faecalis, Streptococcus faecium, and other less common species, occurs frequently. Although these organisms cause endocarditis and urinary tract infection, their virulence in genital infections has been debatable. They are considerably less virulent than group A or B streptococci, but on occasion they have caused serious genital and abdominal infections. Enterococci are important pathogens, particularly in situations where cephalosporin prophylaxis has been used (4). They are the only streptococci not sensitive to penicillin. They are resistant to cephalosporins, alone or in combination with aminoglycosides (streptomycin, kanamycin, or gentamicin), and to clindamycin, alone or in combination with aminoglycosides. Enterococci organisms are susceptible to ampicillin, to penicillin or ampicillin and aminoglycoside in combination, and to vancomycin. Of recent concern have been reports of enterococcal resistance to ampicillin-aminoglycoside combination and vancomycin, but these appear rarely in pelvic infections (5). Failure of a patient to respond to cephalosporin-aminoglycoside or clindamycin-aminoglycoside combinations
may be seen when the primary pathogen is an enterococcus (4). Nonenterococcal group D streptococci can be commonly isolated. They are susceptible to penicillin. Other Aerobic Streptococci Other aerobic streptococci include a variety of common bacteria that are susceptible to penicillin. Examples are a and g streptococci. The most important of these are the viridans group streptococci, which include Streptococcus intermedius (Streptococcus milleri group), Streptococcus mitis, Streptococcus mutans, Streptococcus oralis, Streptococcus parasanguis, Streptococcus salivarius, and Streptococcus sanguis. Streptococcus pneumoniae is an uncommon but potentially serious cause of genital tract infections. Staphylococci The aerobic staphylococci include S. epidermidis, Staphylococcus saprophyticus (Micrococcus), and Staphylococcus aureus. The first two do not produce a coagulase enzyme (the “coag-negative” staphylococci), whereas S. aureus does (“coag-positive” staphylococci). Staphylococcus aureus is isolated from 5% to 10% of genital tract cultures. This organism has been recognized as a cause of abdominal wound infections, breast abscesses, and nursery outbreaks of infection. It has been isolated from nearly all patients with toxic shock syndrome. Most species of S. aureus, whether isolated in the community or in the hospital, elaborate penicillinase and are resistant to penicillin and ampicillin. Agents of choice for treatment of S. aureus infections are the penicillinase-resistant penicillins, such as cloxacillin, dicloxacillin, methicillin, oxacillin, and nafcillin. Resistance to these antibiotics by methicillin-resistant S. aureus has become a major nosocomial infection problem (6). Antibiotics for use in S. aureus infections in the penicillin-allergic patient are the “first-generation” cephalosporins and clindamycin. Vancomycin is the drug of choice for methicillin-resistant S. aureus. Staphylococcus saprophyticus has been recognized recently as an important cause of urinary tract infection. This organism is susceptible to a wide range of antibiotics, including penicillins, cephalosporins, and trimethoprim-sulfamethoxazole (TMP-SMX). Staphylococcus epidermidis is commonly isolated from the vagina and skin, but rarely causes infection. Conditions in which S. epidermidis is recognized as a pathogen include osteomyelitis, possibly late-onset neonatal sepsis, and association with foreign bodies and invasive lines. Gram-Positive Bacilli The Gram-positive bacilli are common organisms in the normal vaginal flora (Table 1.2). Lactobacillus sp are the most frequent component of the normal vaginal flora in women of reproductive age. Although lactobacilli are generally nonvirulent organisms, the strains that produce hydrogen peroxide play a major role in controlling the vaginal flora.
In unusual circumstances, the usually avirulent lactobacilli may produce invasive disease, such as bacteremia, which occurs in patients with underlying conditions such as cancer, recent surgery, and diabetes mellitus. Many of these patients received prior antibiotic therapy (7). When lactobacilli appear in cultures of the urine, it almost certainly represents a contamination. Listeria monocytogenes is present rarely in the vaginas of healthy women. Although the predominant route for severe intrauterine infections due to this organism during pregnancy is transplacental secondary to bacteremia, on occasion L. monocytogenes ascends from the lower genital tract to cause intrauterine infection (see Chapter 16). Epidemics due to Listeria organisms have occurred from contaminated dairy products. Gram-Negative Bacilli The Gram-negative bacilli include a large number of microorganisms with highly variable patterns of antimicrobial susceptibility. Many species have been identified, but only a few are commonly isolated from patients with pelvic infections (Table 1.2). Escherichia Coli Escherichia coli is one of the most common members of this group isolated in genital tract and urine specimens. It is present in approximately 70% of urinary tract infections. Escherichia coli infections usually are mild, but occasionally they may be fulminant, as it is the microorganism most commonly identified in bacteremic obstetric and gynecologic patients. Escherichia coli frequently is recovered in mixed infections of the pelvis, such as amnionitis, endometritis, and posthysterectomy cellulitis. Its susceptibility to antibiotics varies from hospital to hospital and, probably, from service to service. Gentamicin, tobramycin, amikacin, and chloramphenicol usually are effective against more than 95% of E. coli isolates. Increasingly, E. coli resistance to ampicillin has emerged. In general, more than 40% of E. coli (including community-acquired strains) are resistant to ampicillin. The first-generation cephalosporin antibiotics have remained active against E. coli isolates in most hospitals, but the newer cephalosporin agents (second and third generation) and newer penicillins are more active. The new quinolone agents, such as ciprofloxacin and ofloxacin, are very active against E. coli, as is TMP-SMX. Gardnerella vaginalis Formerly known as Haemophilus vaginalis and Corynebacterium vaginale, Gardnerella vaginalis is found in vaginal cultures of nearly all women with bacterial vaginosis, but it also can be found in vaginal cultures of 40% to 60% of asymptomatic women when a selective medium is used. It has been reported to cause endometritis and bacteremia. In some institutions, G. vaginalis is the most common Gram-negative aerobe recovered from the endometrium and blood of patients with postpartum endometritis. Gardnerella vaginalis has been frequently recovered from patients with pelvic inflammatory disease. Rather than being a pathogen on its own, G. vaginalis probably is involved by association with other bacterial vaginosis organisms. In vitro testing shows this organism to be susceptible to ampicillin and tetracycline, but these agents are of limited
value in curing bacterial vaginosis (see Chapter 12). Klebsiella Species Klebsiella sp are found in less than 10% of genital tract infections, but they also cause urinary tract infections and hospital-acquired pneumonia. Klebsiella pneumoniae is the most common member of this group recovered from genital tract and urinary tract infections. Klebsiella oxytocia is much less common. All the cephalosporin antibiotics are highly effective against Klebsiella organisms, as are the aminoglycosides and chloramphenicol. Ampicillin has little activity, but some of the newer penicillins, such as piperacillin and mezlocillin, have improved activity. Quinoline agents are effective, as are the new enzyme-blocking drugs, such as amoxicillin (Augmentin), ticarcillin (Timentin), ampicillin (Unasyn), and piperacillin (Zosyn). Enterobacter Species Although closely related to Klebsiella species, Enterobacter species are encountered much less frequently (less than 5% of genital infections). They are more resistant to antibiotics than are Klebsiella species. Until recently, Enterobacter infections usually required therapy with aminoglycoside antibiotics, but some of the newer cephalosporins and newer penicillins show good activity. The most common of this group are Enterobacter aerogenes and Enterobacter cloacae. Proteus Species Proteus sp are isolated in 10% to 15% of genital tract infections and in a similar percentage of urinary tract infections. Proteus mirabilis, by far the most commonly isolated species in obstetric and gynecologic patients, is susceptible to ampicillin and the cephalosporins, as well as the aminoglycosides. Proteus vulgaris occurs much less commonly. Former Proteus species, Proteus morganii and Proteus rettgeri, are now classified as Morganella morganii and Providencia rettgeri, respectively. These species are resistant to ampicillin and the first-generation cephalosporins, but they are sensitive to the aminoglycosides and some of the newer penicillin and cephalosporin antibiotics. Pseudomonas Species Opportunistic pathogens in severe, usually hospital-acquired, infections, Pseudomonas sp are found infrequently in infections in obstetrics and gynecology, but Pseudomonas colonization is seen commonly in patients receiving antibiotic therapy. Antibiotic susceptibility is good to gentamicin and usually better to tobramycin and amikacin. Activity of the newer penicillins and some of the newer cephalosporins is good, and combinations of antibiotics may produce higher cure rates in serious infections. Other Gram-Negative Bacilli Other Gram-negative bacilli include microorganisms such as Serratia, Citrobacter, Acinetobacter, and Providencia species, all of which show resistance to commonly used
antibiotics. Fortunately, these species are found rarely among obstetric and gynecologic patients, except in those who are debilitated or who are receiving antibiotic, immunosuppressive, or cytotoxic therapy. Gram-Negative Cocci In pelvic infections, the only significant member of the Gram-negative cocci is Neisseria gonorrhoeae, which may produce an asymptomatic colonization of the cervix, cervicitis, or salpingitis (Table 1.2). Disseminated infection with septicemia, arthritis, and dermatitis occurs not infrequently. Neisseria gonorrhoeae is a common cause of neonatal conjunctivitis and has been reported recently as an unusual cause of amnionitis and fetal scalp abscess. Penicillinase-producing strains of N. gonorrhoeae have become a major problem in the United States. In addition, chromosomally mediated resistance and tetracycline resistance have emerged in N. gonorrhoeae. Penicillin is no longer a recommended antibiotic. Cephalosporins or quinolones are preferred, usually in single-dose regimens, for uncomplicated gonococcal infections (8).
ANAEROBIC ISOLATES Anaerobic bacteria are likely to produce infection in the presence of traumatized or devitalized tissue, and often they produce a feculent odor (9). Anaerobic bacteria are major pathogens in obstetric and gynecologic infections (Table 1.3).
TABLE 1.3. ANTIBIOTIC THERAPY FOR ANAEROBIC BACTERIA FOUND IN FEMALE GENITAL INFECTION
Gram-Positive Anaerobes Peptostreptococcus Species All species formerly classified as Peptococcus, except Peptococcus niger, have been
transferred into the genus Peptostreptococcus (9). These strictly anaerobic cocci are isolated very commonly in the vagina and in cultures from obstetric and gynecologic infections. The most common of these organisms include Peptostreptococcus anaerobius, Peptostreptococcus asaccharolyticus, Peptostreptococcus magnus, Peptostreptococcus prevotii, and Peptostreptococcus tetradius (Gaffkya anaerobia). Penicillin is the drug of choice for infections known to be caused by peptostreptococci, but clindamycin, metronidazole, cefoxitin, cefotetan, first- and third-generation cephalosporins, and chloramphenicol are highly effective for the penicillin-allergic patient or for broader anaerobic coverage. Clostridium Species Strictly anaerobic, plump, Gram-positive rods, clostridia may be isolated from vaginal secretions of 5% to 10% of asymptomatic women. Clostridia are anaerobic, spore-forming rods that can produce potent toxins and result in severe life-threatening infections. The most commonly isolated species is Clostridium perfringens (also known as Clostridium welchii). Other clinically important clostridia include Clostridium novyi, Clostridium ramosum, Clostridium septicum, and Clostridium sordelli. Although clostridial species may produce gas gangrene (with septicemia, circulatory collapse, hemolysis, and peritonitis), they are more commonly associated with a much less disseminated infection that responds promptly to appropriate antibiotic therapy. Simply isolating this microorganism from a pelvic site does not, therefore, indicate life-threatening infection or the need for a hysterectomy. Initial therapy for clostridial infections usually consists of intravenous administration of large doses of penicillin and close patient monitoring. If signs of extension of infection become apparent, debridement (often by transabdominal hysterectomy and bilateral salpingo-oophorectomy) is required. These signs include worsening clinical condition or failure to respond promptly; parametrial tenderness, crepitus, or myalgia; or hypotension, falling central pressure measurements, or oliguria despite adequate volume replacement. Radiographic findings of interstitial gas are uncommon and develop late in the clinical course, but an abdominal x-ray film should be obtained. Clostridium difficile is isolated infrequently in genital tract infections but is pathogenic in antibiotic-associated pseudomembranous colitis. Clostridium difficile is resistant to many antibiotics, but it can be treated with metronidazole or vancomycin (Table 1.3). Propionibacterium, Eubacterium, And Bifidobacterium Species These Gram-positive anaerobic bacilli usually are isolated in anaerobic specimens. They are organisms of low virulence and most often do not require specific antibiotic therapy. Gram-Negative Anaerobes Bacteroides In the early 1990s, the genus Bacteroides underwent major taxonomic revisions (9).
This revision was necessitated by the large heterogeneity in biochemical and chemical properties demonstrated in this group of microorganisms. The genus Bacteroides now only includes species that were formerly described as the “Bacteroides fragilis group.” Bacteroides now includes the species B. fragilis, Bacteroides caccae, Bacteroides distasonis, Bacteroids eggerthii, Bacteroides merdae, Bacteroides ovatus, Bacteroides stercoris, Bacteroides thetaiotaomicron, Bacteroides uniformis, and Bacteroides vulgatus (Table 1.3). New genera have been established that include many of the microorganisms previously classified as Bacteroides. Table 1.4 lists the changed designations of species associated with genital tract infections that were formerly classified as Bacteroides. The two major new genera are Prevotella and Porphyromonas. The most important of these microorganisms in the pathogenesis of obstetric and gynecologic infections are Prevotella bivia (formerly Bacteroides bivius), Prevotella disiens (formerly Bacteroides disiens), Prevotella melaninogenica (formerly Bacteroides melaninogenicus), and Porphyromonas asaccharolytica (formerly Bacteroides asaccharolyticus). The taxonomic position of several non-B. fragilis group species of Bacteroides remains to be determined. These currently are listed as Bacteroides species other and include Bacteroides capillosus, Bacteroides coagulans, Bacteroides splanchnicus, and Bacteroides ureolyticus (Table 1.3).
TABLE 1.4. NEW NOMENCLATURE OF MICROORGANISMS ASSOCIATED WITH GENITAL TRACT INFECTIONS THAT WERE FORMERLY MEMBERS OF THE GENUS BACTEROIDES
These are strictly anaerobic bacilli that produce infections that often are protracted. From studies of obstetric infection, the most commonly identified species is P. bivia, which is isolated in 20% to 40% of obstetric and gynecologic infections. Although the susceptibility of this species to penicillin has decreased, it is sensitive to clindamycin, chloramphenicol, and metronidazole, as well as many of the newer cephalosporins and newer penicillins. Prevotella disiens also is isolated regularly from pelvic infections and has a susceptibility pattern similar to that of P. bivia. Bacteroides (formerly Bacteroides
fragilis group) have been identified less commonly in studies but still play an important role. Because members of this group have been recognized by clinicians for their resistance to many antibiotics and their frequent involvement in pelvic and abdominal infections, it is necessary to maintain this connotation for all these species with a simple designation. Thus, in clinical literature, these species commonly are referred to as the B. fragilis group. Antibiotics of choice for therapy of B. fragilis group infections are metronidazole, imipenem, some penicillin b-lactamase inhibitors, and chloramphenicol. Of the newer cephalosporin-type agents, cefoxitin and cefotetan have the best in vitro activity against the B. fragilis group (10). Piperacillin, mezlocillin, and related drugs, given in large parenteral doses, also are active against the vast majority of B. fragilis species. Similarly, the b-lactamase enzyme blocker agents are active against these organisms. Previously, it was presumed that Bacteroides species other than B. fragilis (e.g., P. bivia, P. disiens, P. melaninogenica) were susceptible to penicillin. Recently, multiinstitutional studies of in vitro susceptibility have noted increasing resistance by the Prevotella organisms. Thus, these organisms require antimicrobial therapy similar to that used for Bacteroides (B. fragilis group), as shown in Table 1.3. Fusobacterium Fusobacteria are isolated less frequently than Bacteroides, Prevotella, and Porphyromonas and usually are involved in polymicrobial infections. They appear to be less virulent than Bacteroides and Prevotella and are generally susceptible to penicillin, clindamycin, chloramphenicol, metronidazole, and many of the newer penicillins and cephalosporins. Fusobacterium organisms have been implicated as important pathogens associated with amnionitis, especially prior to term. Mobiluncus The taxonomic position of Mobiluncus remains uncertain. The genus Mobiluncus is assigned to the family Bacteroidaceae, which contains phenotypically similar organisms that are obligate anaerobic Gram-negative rods (straight, curved, or helical). Mobiluncus are curved, motile rods that are Gram-variable or Gram-negative. These rods have tapered ends. These organisms (Mobiluncus curtisii and Mobiluncus mulieris) are associated with bacterial vaginosis and thus may be involved in the adverse complications attributable to bacterial vaginosis.
YEASTS AND ACTINOMYCES Yeasts are commonly isolated from the vagina. Candida albicans is clearly the most common yeast found in vaginal cultures, but other Candida species are seen in 10% to 15%. These include Candida kefyr (formerly Candida pseudotropicalis), Torulopsis (Candida) glabrata, and Candida tropicalis. In vaginitis, yeasts are commonly identified by direct microscopy of a potassium hydroxide preparation, but culture is helpful in problem cases (11,12). Vaginal candidiasis is treated by a number of topical agents or by oral agents. Although the initial cure rate for vaginal candidiasis is good (80% to 90%), recurrent infection is a problem. New strategies are discussed in Chapter 12. Systemic Candida infections in obstetric or gynecologic patients are unusual and limited to those patients who have received protracted antibiotic or immunosuppressive therapy
or who have debilitating diseases. The genus Actinomyces is classified between true bacteria and molds. Actinomyces israelii, an anaerobic or microaerophilic organism, is a rare cause of pelvic infection, but it has been isolated in pelvic abscesses associated with intrauterine devices. Actinomyces organisms are susceptible to penicillin, most cephalosporins, tetracycline, and rifampin.
TRICHOMONADS Trichomonas vaginalis, a flagellated parasite, is found in vaginal secretions of approximately 6% of women and is responsible for up to one fourth of cases of infectious vaginitis (Table 1.5). As discussed in Chapter 12 and Chapter 19, Trichomonas vaginalis has been associated recently with preterm premature rupture of membranes and preterm labor and delivery. Trichomonas vaginalis is commonly detected by direct microscopy of a wet mount of vaginal secretions. These flagellates are larger than white cells and have great motility in fresh preparations. Trichomonas vaginalis also may be detected on the Papanicolaou smear, but the reliability of this technique is uncertain. Culture for T. vaginalis is more sensitive than the wet mount. Trichomoniasis is treated with metronidazole (Flagyl) or related 5-nitroimidazole derivatives. Because this organism is transmitted sexually, partners also should be treated. Relative resistance to metronidazole has become common. Most cases respond to a higher dose of metronidazole (13).
TABLE 1.5. OTHER MICRORGANISMS IN INFECTIONS OF THE GENITAL TRACT OR PREGNANCY
MYCOPLASMAS These cell wall-deficient microorganisms are distinctly different, morphologically and biochemically, from bacteria and L forms (Table 1.5). Mycoplasma hominis and
Ureaplasma urealyticum are isolated with great frequency from the genital tract (13). Mycoplasma hominis has been found in blood cultures of women with postpartum fever. It has been isolated commonly in amniotic fluid from patients with intraamniotic infection or those with preterm labor. Ureaplasma urealyticum has been implicated in chorioamnionitis, recurrent abortion, infertility, prematurity, and low-birth-weight infants. The role of genital mycoplasmas in causing adverse pregnancy outcomes is controversial and is discussed in detail in Chapter 19. Culturing of mycoplasmas requires special techniques that are not available in most hospital or commercial laboratories. Mycoplasma hominis is susceptible to tetracycline, clindamycin, and lincomycin, whereas Ureaplasma organisms are sensitive to tetracycline and erythromycin.
CHLAMYDIA TRACHOMATIS Chlamydia are intracellular bacteria that are involved in a variety of genital and perinatal infections (14,15). From 2% to 25% of women have positive cervical cultures for Chlamydia trachomatis (Table 1.5). Isolation rates of approximately 25% are found in high-risk groups, such as women attending sexually transmitted disease clinics. Overall, the incidence of cervical infection in the United States probably averages 3% to 6%. In males, chlamydia are responsible for many instances of nongonococcal urethritis. In females, evidence of their virulence has been accumulating. Chlamydia trachomatis plays a major role in pelvic inflammatory disease, especially in subsequent tubal obstruction. Newborns may acquire this microorganism from the maternal genital tract. Approximately 40% of infants delivered from an infected mother develop conjunctivitis, and 10% develop late-onset pneumonia. Chlamydia infections are described in detail in Chapter 5. Azithromycin and doxycycline are the antibiotics of choice in treating chlamydial infection, with erythromycin and ofloxacin used alternatively (8). Other effective agents include sulfisoxazole, clindamycin, ofloxacin, and azithromycin. High doses of ampicillin or amoxicillin also appear to provide some cures. In pregnancy, erythromycin base and amoxicillin are recommended, with azithromycin and other erythromycin preparations as alternatives (8).
VIRUSES Several viruses are commonly found in the female genital tract (Table 1.5). Herpes simplex virus is a well-known cause of an ulcerative, usually self-limited, lower genital tract infection. Herpesvirus also can cause asymptomatic cervical infection. In surveys of adult females, this virus has been isolated from the genitalia in 0.02% to 1%. Approximately 85% of isolates of herpesvirus in the genital tract are of serologic type 2 (16) and 15% are serologic type 1, which more commonly causes oral lesions. Serologic surveys show that 25% of reproductive age females have serum antibody to herpesvirus type 2 (16). The most reliable method for detecting herpes infection has been culture, but this technique has a recognized false-negative rate. Polymerase chain reaction techniques have been introduced. Clinical diagnosis, Papanicolaou smears, enzyme-linked immunosorbent assay, and monoclonal antibody tests are less reliable. Several drugs are available for treatment or prevention of genital herpes infections (see
Chapter 6). Cytomegalovirus (CMV) is the most common virus found in the female genital tract. It is found in the cervix in 3% of 18% of pregnant women, more commonly in indigent, young, primiparous women. Serologic evidence of past infection is found in 20% to 70% of adults. Genital infections are asymptomatic and occur mainly in seropositive women. CMV may be transmitted vertically to the fetus by transplacental infection. It is estimated that 0.5% to 2.5% of neonates have this congenital infection, but most cases are asymptomatic (17,18). An additional 3% to 5% of newborns acquire CMV during delivery. If a mother has CMV in her genital tract, there is a 30% to 50% chance that her neonate will acquire the virus. CMV can be detected by culture, and several antibody tests are available. Ganciclovir can be used for treatment of CMV infection, but this drug may be toxic (19). Human papilloma virus (HPV) is the causative agent of genital warts, condylomata acuminata. The recent marked rise in interest in this virus stems from its dramatic increase in frequency and its relationship to genital tract malignancy. The common HPV types are 6, 11, 16, and 18. The first two are more common and usually are associated with benign genital warts; the latter two more likely are associated with higher grades of cervical intraepithelial neoplasia or invasive carcinoma. Treatment has generally consisted of surgical excision, cryotherapy, chemical burning, or laser vaporization. Recently, treatment with parenterally or locally administered interferon has been successful. Another immune modulator, imiquimod (Aldara), has been approved for treatment. Although the likelihood is small, maternal HPV infection may lead to juvenile-onset respiratory papillomatosis. However, cesarean delivery is not recommended solely on the basis of maternal HPV infection (20). Other viruses that cause nongenital infections may have the genital tract as a portal of entry. Among these are human immunodeficiency virus, the causative agent in acquired immunodeficiency syndrome (AIDS), and hepatitis B virus. Hepatitis and AIDS are discussed in detail in Chapter 9 and Chapter 10, respectively. Other viruses cause adult systemic infection that may have serious fetal sequelae if the infection occurs in pregnancy (see Chapter 16).
CHANGES IN VAGINAL MICROFLORA One should not conclude from this description of vaginal microflora that it is a static situation. Certainly, there are vast differences in flora between different groups of women and interesting shifts from time to time in the flora of one particular woman. Age At birth, the vagina is sterile. Secondary to maternal estrogen effect, lactobacilli growth is enhanced for a short time. The estrogen effect is gone within several weeks, and lactobacilli disappear until the onset of puberty, when, under the influence of endogenous estrogen, the vaginal flora becomes dominated by lactobacilli. It is suggested that postmenopausal women have a decrease in Lactobacillus colonization,
but that treatment with estrogens results in a higher rate of recovery of lactobacilli and probably of diphtheroids. Thus, there seems to be an important interaction between vaginal colonization and hormonal milieu. Changes associated with aging have been reported in other groups of bacteria, but the conclusions are less uniform. Sexual Activity Sexual intercourse leads to changes in lower genital tract microorganisms, mainly sexually transmitted ones. In addition to introducing major pathogens such as N. gonorrhoeae, C. trachomatis, and herpesvirus, intercourse leads to increases in genital mycoplasmas. Contraception Use of oral contraceptives appears to have minimal effect on the vaginal ecosystem. On the other hand, use of intrauterine contraceptive devices increases the number of anaerobic bacteria in the cervix and augments the risk for bacterial vaginosis, thus increasing the risk for pelvic inflammatory disease. Pregnancy And Delivery A number of studies have suggested that there is a progressive increase in colonization by Lactobacillus organisms during pregnancy, but changes in other bacterial groups are not well established. After delivery, dramatic changes in vaginal flora occur. There are marked increases in anaerobic species by the third postpartum day. Possible predisposing features to anaerobic vaginal colonization in postpartum women include trauma, presence of lochia and suture material, examinations during labor, and changes in hormonal levels. By the sixth week postpartum, the vaginal flora is restored to a normal distribution. Surgery Major procedures, such as hysterectomy, lead to wide changes in vaginal flora, including decreases in lactobacilli and diphtheroids and increases in aerobic and anaerobic Gram-negative rods (predominantly E. coli and various Bacteroides [Prevotella]). In addition, most investigators have noted a further shift when prophylactic or therapeutic antibiotics are used. As expected, use of antibiotics results in a decrease in susceptible flora and a corresponding increase in resistant organisms. Homeostasis Homeostatic mechanisms have been identified that function to maintain the stability of the normal vaginal flora (21). Production of hydrogen peroxide by certain Lactobacillus species appears to play a crucial role in maintaining the normal vaginal ecosystem. In addition, the low pH (acidity) of the normal vagina protects against exogenous organisms. Antimicrobial agents can disrupt the normal vaginal ecosystem, especially if
they eliminate the hydrogen peroxide-producing lactobacilli. CHAPTER REFERENCES 1. Centers for Disease Control and Prevention. Prevention of perinatal group B streptococcal disease: a public health perspective. MMWR 1996;45:1–24. 2. American College of Obstetricians and Gynecologists. Prevention of early-onset group B streptococcal disease in newborns. ACOG Committee Opinion 173. Washington, DC: American College of Obstetricians and Gynecologists, 1996. 3. Rouse DJ, Andrews WW, Lin FC, et al. Antibiotic susceptibility profile of group B streptococcus acquired vertically. Obstet Gynecol 1998;92:931–934. 4. Walmer D, Walmer KR, Gibbs RS. Enterococci in post-cesarean endometritis. Obstet Gynecol 1988;71:159–162. 5. Centers for Disease Control and Prevention. Nosocomial enterococcus resistant to vancomycin—United States 1989–1993. MMWR 1993;42:597–599. 6. Neu HC. The crisis in antibiotic resistance. Science 1992;257:1064–1073. 7. Husni RN, Gordon SM, Washington JA, et al. Lactobacillus bacteremia and endocarditis: review of 45 Cases. Clin Infect Dis 1997;25:1048–1055. 8. Centers for Disease Control and Prevention. Sexually transmitted disease treatment guidelines 1998. MMWR 198;47:1–116. 9. Summanen P. Recent taxonomic changes for anaerobic gram-positive and selected gram-negative organisms. Clin Infect Dis 1993;16[Suppl 4]:S168–S174. 10. Cuchural GJ, Snydman DR, McDermott L, et al. Antimicrobial susceptibility patterns of the Bacteroides fragilis group in the United States, 1989. Clin Ther 1992;14:122–136. 11. Sobel JD. Vaginitis. N Engl J Med 1997;337:1896–1903. 12. American College of Obstetricians and Gynecologists. Vaginitis. Technical Bulletin 221. Washington, DC: American College of Obstetricians and Gynecologists, 1996. 13. Lossick JG, Muller M, TE Gorrell. In vitro drug susceptibility and doses of metronidazole required for cure in cases of refractory vaginal trichomoniasis. J Infect Dis 1986;153:948–955. 14. Centers for Disease Control and Prevention. Chlamydia trachomatis genital infections—United States, 1995. MMWR 1997;46:193–198. 15. Stamm WE. Chlamydia trachomatis infections: progress and problems. J Infect Dis 1999;179[Suppl 2]:S380–S383. 16. Fleming DT, McQuillan FM, Johnson RE. Herpes simplex virus type 2 in the United States, 1976 to 1994. N Engl J Med 1997;337:1105–1111. 17. Istas AS, Demmler GJ, Dobbins JG, et al. Surveillance for congenital cytomegalovirus disease: a report from the National Congenital Cytomegalovirus Disease Registry. Clin Infect Dis 1995;20:665–670. 18. Lipitz S, Yagel S, Shalev E, et al. Prenatal diagnosis of fetal primary cytomegalovirus infection. Obstet Gynecol 1997;89:763–767. 19. Whitley RJ, Cloud G, Gruber W, et al. Ganciclovir treatment of symptomatic congenital cytomegalovirus infection: results of a phase II study. J Infect Dis 1997:175:1080–1086. 20. Shah K, Kashima H, Polk F, et al. Rarity of cesarean delivery in cases of juvenile-onset respiratory papillomatosis. Obstet Gynecol 1986;68:795–799. 21. Eschenbach DA, Davick PR, Williams BL, et al. Prevalence of hydrogen peroxide-producing Lactobacillus species in normal women and women with bacterial vaginosis. J Clin Microbiol 1989;27:251–256.
2 USE OF THE MICROBIOLOGY LABORATORY IN INFECTIOUS DISEASES Infectious Diseases of the Female Genital Tract
2 USE OF THE MICROBIOLOGY LABORATORY IN INFECTIOUS DISEASES Principles of Diagnostic Molecular Microbiology Detection and Identification Methods Antimicrobial Susceptibility Testing Serology Diagnostic Virology Infections of the Lower Female Genital Tract Infections of the Upper Genital Tract Urinary Tract Infections Bacteremia Chapter References
The role of the microbiology laboratory in the clinical care of obstetric and gynecologic patients with infection has been affected by a variety of factors (1). First, the introduction of potent antimicrobial agents with their dramatic favorable impact on mortality due to infection in obstetrics and gynecology has lulled clinicians into complacency. Second, the trend to centralize laboratory facilities has made it difficult for (a) clinicians to work with laboratory personnel; (b) the central microbiology laboratory to meet the needs for obtaining and transporting specimens; and (c) the laboratory to recover and identify the tremendous variety of microorganisms now recognized to be human pathogens. Third, as academic microbiology departments have focused on basic research involving molecular biology, medical students and house officers often complete their training with a minimal background in clinical microbiology and infectious disease. Fourth, formal organized instruction in clinical and laboratory diagnosis and management of sexually transmitted diseases (STDs) is an often-neglected area of the medical school curriculum and house officer training. Fifth, the virtual explosion in knowledge and technology related to the field of microbiology has widened the gap between the reality of microbiology laboratory capabilities and the clinician's understanding of these capabilities and their application to patient care. In particular, the introduction of more sensitive, rapid, and cost-efficient diagnostic tests, such as enzyme-linked immunosorbent assay (ELISA), DNA hybridization techniques, monoclonal antibody techniques, and nucleic acid amplification techniques, has resulted in a shift from the use of classic microbiologic isolation and identification of microorganisms to methods that rely on advances in molecular biology. Last, diagnostic virology has now entered the mainstream of clinical practice (2). This chapter will review the role played by microbiology and diagnostic molecular biology laboratories in clinical medicine. A description of the proper collection, handling, and transport of specimens is presented, with emphasis on the important role clinicians
must take in providing laboratories with appropriate specimens in a timely manner so that laboratories can provide quality testing. The clinical application of recently developed molecular-based tests for STDs and other pathogens will be addressed. Figure 2.1 provides an overview of the types of methods used for the diagnosis of infectious agents. The specific methods and relative usefulness of the tests are dependent on the microorganism(s) being sought (3).
FIGURE 2.1. Methodologies used for the diagnosis of infectious agents. (Reprinted with permission from ref. 3.)
The continuous and rapid increase in knowledge and technology of microorganisms, the immune response to these organisms, and their molecular biology has exploded during the last 2 decades. The introduction of more sensitive and specific diagnostic tests, such as ELISA, in situ nucleic acid hybridization, and molecular assays [e.g., probes and nucleic acid amplification such as polymerase chain reaction (PCR) and ligase chain reaction (LCR)], has resulted in the evolution of tests from the classic microbiological practices of isolation on bacteriologic agar or for viruses in tissue culture. The past decade has resulted in explosive growth in the application of molecular biology to the diagnosis of infection. New organisms that could never be grown are being identified and their role in human infection is being demonstrated (e.g., hepatitis C virus, Helicobacter pylori). The basic purpose of the infectious disease laboratory is to identify microorganisms that cause clinical disease. As noted by Gill et al. (3), a critical component in elucidating the etiologic agent(s) in infections is a partnership and active communication that involves the clinician and the microbiology laboratory. The laboratory must be capable of handling a wide variety of specimens that come from a multitude of clinical infection sites and thus require a diversity of collection and transport systems (4,5). Physicians refer to the laboratory to obtain assistance in the diagnosis and treatment of infectious diseases, but they also expect the laboratory to (a) produce reliable, rapid, and clinically useful facts; (b) establish and maintain a consistent reporting system; (c) provide
specific and concise guidelines for proper specimen collection, handling, and transport; (d) initiate interpretive reporting for unusual microorganisms, normal flora, and monitoring of the therapeutic response; and (e) provide susceptibility data that will guide therapeutic decisions. Clinicians should realize that laboratory results are only as good and as useful as the specimens received by the laboratory. The physician must provide the infectious disease laboratory with pertinent patient information, perform appropriate specimen collection, ensure proper and prompt specimen transport to the laboratory, and request isolation or identification of suspected pathogenic organisms. Each microorganism has its own requirements for collection, handling, transport, growth, isolation, and detection in the laboratory. Clinics and laboratories throughout the world have variable capabilities for isolating microorganisms, such as anaerobic bacteria, Chlamydia trachomatis, mycoplasmas, ureaplasmas, and viral pathogens, which require sophisticated technology and experienced laboratory personnel. For example, in a study of chlamydia culture from the cervix, Schachter et al. (6) found that sensitivity varied from 51.8% to 92.0% among a series of five laboratories throughout the United States. Clearly, the accuracy of a laboratory's results for specific organisms varies according to its focus, expertise, and personnel. General guidelines for collection of clinical specimens from infected patients has been summarized by Gill et al (3). Specimen selection must be based on the characteristics of suspected infectious agents. The clinical specimen should be obtained from the site of infection and be representative of the infectious process. An adequate quantity of material is required for the laboratory to perform a complete evaluation. The clinician must be scrupulously careful to avoid contamination of the specimen by microorganisms indigenous to skin and/or mucous membranes (i.e., oral cavity, vagina, or bowel). Once a suitable specimen is obtained, it should be forwarded promptly to the laboratory. Finally, it is crucial that, whenever possible, specimens for culture be obtained before institution of antimicrobial therapy. Table 2.1 summarizes the general guidelines for specimen collection and transport of common specimens from obstetric and gynecologic infections. Recent advances in the use of molecular diagnostic techniques for the diagnosis of infectious diseases have revolutionized the laboratory diagnosis of these infections and added another approach to the traditional use of stain, culture, and serology for diagnosing infectious agents (3).
TABLE 2.1. GUIDELINES FOR COLLECTION AND TRANSPORT OF COMMONLY OBTAINED SPECIMENS IN INFECTIONS OF THE FEMALE GENITAL TRACT
PRINCIPLES OF DIAGNOSTIC MOLECULAR MICROBIOLOGY In 1980, Moseley et al. (7) reported the first clinically useful application of recombinant DNA technology to the diagnosis of an infectious disease using DNA–DNA hybridization in stool samples spotted on filters to detect enterotoxigenic Escherichia coli. The following decade resulted in the availability of DNA probes for the identification of bacterial, viral, fungal, parasitic, and helmintic pathogens (8,9). Nucleic acid probes are segments of DNA or RNA that are labeled with enzymes, antigenic substrates, chemiluminescent moieties, or radioisotopes and can bind with high specificity to complementary nucleic acid sequences from the microorganism of interest (3,9). Oligonucleotide probes are probes with less than 50 base pairs. They can be synthesized and purified commercially and hybridize more rapidly than layer probes. Probe technology uses the presence of nucleotide sequences unique to a given organism or species as a molecular fingerprint for identification (10). Probes can be developed that identify organisms grown in culture (culture confirmation assays) or detect and identify organisms directly in clinical samples. Several nucleic acid amplification techniques have been developed. These include PCR, LCR, self-sustaining sequence replication (3SR), strand displacement amplification (SDA), and QB replicase. Nucleic acid amplification methods include three general categories: (i) target amplification (PCR, 3SR, or SDA); (ii) probe amplification systems (LCR or QB replicase); and (iii) signal amplification where the signal generated from each probe molecule is increased by using compound probes or branched-probe technology (3,11). PCR involves three steps: (i) DNA or target organism is denatured to a single strand; (ii) two primers (single-stranded DNA 20 to 30 bases in length) bind to regions of DNA flanking the target sequence to be amplified; and (iii) the DNA between the two primers is reproduced. With each round of PCR, the target DNA equence increases by a power of two. At the end of 30 replication cycles, over a million-fold amplification of the target DNA has taken place. Nested PCR is a modification of PCR in which a first round of amplification is performed with a single primer for 15 to 30 cycles, followed by a second round using a second primer specific for the internal sequence amplified by the first primer. This approach leads to very high sensitivity. Nucleic acid hybridization requires that two molecules of single-stranded DNA meet and form a stable double-stranded molecule (12). Both DNA probe and amplification techniques are based on this technology (9). Hybridization can be accomplished using solution phase hybridization or solid phase hybridization. Although solid phase hybridization is somewhat less sensitive, it greatly facilitates the handling of multiple specimens (9). An important type of solid phase hybridization is in situ hybridization, in which whole cells or tissue sections affixed to microscopic slides are assayed by
hybridization. A variety of commercially available probe kits have become available (Table 2.2). These include probes for (i) STD organisms such as C. trachomatis and Neisseria gonorrhoeae; (ii) respiratory pathogens such as Mycobacterium sp, including Mycobacterium tuberculosis, and Legionella sp; (iii) Gram-positive organisms such as Streptococcus pneumoniae, Staphylococcus aureus, group B streptococci, enterococci, and Listeria monocytogenes; and (iv) Gram-negative organisms such as Haemophilus influenzae, E. coli, and thermophilic campylobacters (10). The initial assays were based on hybridization with DNA probe technology to detect specific ribosomal RNA (Gen-Probe system). Currently this system uses acridinium ester-labeled, single-stranded DNA probes to detect complementary ribosomal RNA sequences of the target organism. This reaction is measured using chemiluminescence, which is read on a luminometer (3). Only a few organisms, primarily C. trachomatis and N. gonorrhoeae, are detected directly using probe technology. More commonly this method is used for culture confirmation of organisms such as fungi and mycobacterial species (3).
TABLE 2.2. EXAMPLES OF COMMERCIALLY AVAILABLE PROBES
Polymerase chain reaction technology has been applied rapidly to the detection of microorganisms directly from clinical samples (2,3,11,13,14 and 15). Among the agents for which PCR is used are the DNA viruses, including herpes simplex virus (HSV), cytomegalovirus (CMV), Epstein-Barr virus (EBV), varicella-zoster virus, parvovirus B19, human herpes virus type 6, human papillomavirus, hepatitis B virus, and polyomavirus (Table 2.3). Inclusion of an initial reverse transcriptase step that converts RNA into DNA permits PCR detection of RNA-containing viruses such as human immunodeficiency virus (HIV) types 1 and 2, human T-cell lymphotrophic virus type I, enterovirus, hepatitis C virus, and rotavirus. PCR assays for bacterial also have been developed (Table 2.4). Examples of bacterial organisms for which PCR technology has been introduced include C. trachomatis, M. tuberculosis, Borrelia burgdorferi, Legionella pneumophila, and Bacteroides fragilis (10,14,15).
TABLE 2.3. VIRAL PATHOGENS DETECTED BY NUCLEIC ACID AMPLIFICATION
TABLE 2.4. BACTERIAL PATHOGENS DETECTED BY NUCLEIC ACID AMPLIFICATION
DNA ligase-dependent amplification (LCR) is an alternative nucleic acid probe amplification technique that is based on target-dependent ligation of oligonucleotide probes (16). Rather than using polymerase to copy genetic information, LCR uses DNA ligase to join two pairs of complementary oligonucleotide probes after they have bound to a target sequence. After the two probes have been ligated, the product, which mimics one strand of the original target sequence, serves as a template for ligation of complementary oligonucleotides. The components then are heated to denature the template and allowed to anneal to new probes at a lower temperature (16). Originally, fresh DNA ligase had to be added at the end of each round, which made the process cumbersome. Recently, the use of thermophilic DNA ligase, which remains active during the thermal cycling steps, has greatly simplified the LCR test (3,11). Ligase chain reaction has been used to detect microorganisms including N. gonorrhoeae, C.
trachomatis, B. burgdorferi, and Mycobacterium sp (15). Several additional amplification methods have been developed (11). Self-sustaining sequence replication uses the collective activities of three enzymes to isothermally amplify an RNA target. Up to 10 million-fold amplification occurs within a 1- to 2-hour incubation (11). Strand displacement amplification is an isothermal DNA amplification method using specific primers, a DNA polymerase, and a restriction endonuclease to produce exponential amplification of the target (11). It can be applied to either single- or double-stranded DNA. Transcription-mediated amplification (TMA) is an isothermal technique that amplifies rRNA target. The QB replicase method is a probe amplification technique that uses incorporation of a single-stranded oligonucleotide probe into an RNA molecule that can be amplified exponentially after target hybridization by the enzyme QB replicase (3). Detection And Identification Methods Figure 2.1 outlines the various approaches available for the diagnosis of agents causing infectious diseases. Microorganisms can be identified in patient specimens by either direct detection or culture (3). Microscopy can be used to detect microorganisms that have been stained directly in patient specimens. The Gram stain is the most commonly used direct stain; it is simple, reliable, and inexpensive. It provides preliminary information as to Gram-positive versus Gram-negative bacteria and rod versus coccus. In the hands of an experienced technician, the Gram stain can serve as a quality control tool for the adequacy of the specimen and accuracy of cultures. However, the Gram stain is nonspecific, and its sensitivity is limited by the amount of material that can be reviewed on a slide (3). Direct fluorescent antibody (DFA) stains have been used for some organisms (C. trachomatis, Legionella, some parasites), but their use is limited by inadequate sensitivity when too few organisms are present and the need to be performed by experienced, trained personnel. Wet mounts traditionally are used to look for yeast [potassium hydroxide (KOH)] and Trichomonas vaginalis (saline preparation). Many laboratories are using a calcofluor white stain for fungi (3). Although more sensitive, it requires a fluorescence microscope. Wet mounts also are used to detect the presence of “clue” cells as an aid in diagnosing bacterial vaginosis. Direct antigen detection of patient specimens is now very common and available for a wide variety of microorganisms (3). Antigen detection tests fall into two major categories: latex agglutination and enzyme immunoassay (EIA) (Table 2.5). Antigen detection tests are simple and rapid to perform; those with good sensitivity and specificity, as listed in Table 2.5, have been widely used by clinical laboratories (3).
TABLE 2.5. DIRECT ANTIGEN DETECTION TESTS USED COMMONLY IN CLINICAL PRACTICE
Molecular-based assays were described previously in the section on Principles of Diagnostic Molecular Microbiology. Briefly, these include (i) DNA probe technology to detect specific ribosomal RNA (Gen-Probe), and (ii) amplification techniques such as PCR, LCR, and TMA. Generally, the sensitivity of DNA probes is equivalent to that of good EIA tests and amplification techniques are more sensitive, thus requiring the presence of a smaller number of microorganisms (3). The use of molecular diagnostics has expanded the ability of microbiology laboratories to detect organisms present in very low quantities, organisms that are difficult or slow to grow, and infectious agents as yet “undiscovered” (3). Traditionally, diagnostic bacteriology was based on the growth of microorganisms on appropriate culture media and identification based on biochemical characteristics (3). Enriched all-purpose media such as blood or chocolate agar, which grow most ordinary bacterial pathogens, are used for common specimen types. For specimens from lower and upper genital tract infections, which often contain mixed microbial flora, additional plates such as MacConkey or eosin-methylene blue agar for Gram-negative bacteria, phenylethyl alcohol agar for Gram-positive bacteria, kanamycin-vancomycin laked blood agar for Prevotella sp, and Bacteroides bile esculin for B. fragilis group are inoculated (3). Broth media such as thioglycolate, brain-heart infusion, or trypticase soy are used as a backup because a large aliquot of specimen can be inoculated, allowing detection of small numbers of organisms. Once colony growth is detected, Gram-stain morphology, colony size, color, shape, presence or absence of hemolysis, and biochemical reactions are used to identify bacteria (3). Antimicrobial Susceptibility Testing In general, susceptibility testing of pathogens is indicated when the response of the organism is not predictable (3). The increasing occurrence of resistance among bacteria has resulted in more species requiring susceptibility testing (3). The National Committee for Clinical Laboratory Standards (NCCLS) has published standards related to bacterial
susceptibility testing (17,18,19 and 20). Use of these NCCLS criteria permits testing and reporting of results for nearly all the more common and nonfastidious bacteria (3). As reviewed by Finegold et al. (21), susceptibility testing of anaerobic bacteria has been controversial but is warranted, because anaerobes are significant pathogens and their susceptibility patterns are not predictable. Most clinical microbiology laboratories do not perform routine susceptibility testing of anaerobic bacteria. The NCCLS recommends that susceptibility testing of anaerobic bacteria be performed in certain clinical settings, such as brain abscess, endocarditis, osteomyelitis, prosthetic device infection, and refractory or recurrent bacteremia (20,22). The recommended procedures for susceptibility testing of anaerobes are agar dilution testing, microbroth dilution, and macrobroth dilution (21). The broth disk elution method, which was the simplest and most commonly used method, is no longer recommended by the NCCLS. An interesting alternative, the epsilometer (E-test), has been shown to be a promising test for anaerobic bacteria susceptibility testing (23). On the other hand, susceptibility testing of aerobic bacteria is widely performed by clinical microbiology laboratories. The laboratory must test and report those antimicrobial agents that are most appropriate for the infection site and the organism(s) isolated (24). In addition, the agents tested should be those that are on the hospital formulary (24). The macrodilution or tube dilution method was one of the first antimicrobial susceptibility testing methods introduced. Miniaturization and mechanization of this method resulted in “microdilution” susceptibility testing, which is the most common method used by clinical laboratories for susceptibility testing (24). This method uses plastic disposable microtiter trays containing 96 wells, which provide for 12 antibiotics to be tested in a range of 8 twofold dilutions in a single tray. The second most commonly used method for testing susceptibility of aerobic bacteria is the disk diffusion or Kirby-Bauer test (17). Antibiotic-containing filter paper disks are placed on an inoculated agar plate and the zones of growth inhibition around the disks are measured. These results are compared to the zone size interpretive criteria published by the NCCLS (17) or the U.S. Food and Drug Administration (drug package insert data). Agar dilution testing is an additional but cumbersome method (24). Its major advantage is its ability to test fastidious organisms that do not grow well in broth (e.g., N. gonorrhoeae). In addition, it provides a quantitative result [minimum inhibitory concentration (MIC)]. The MIC is the lowest concentration of an antimicrobial agent that inhibits growth of the test microorganism. The minimal bactericidal concentration (MBC) is the lowest concentration of an antimicrobial agent that kills the test microorganism. The NCCLS has published definitions of susceptible, intermediate, and resistant (19). Susceptible implies that an infection due to the test organism may be treated appropriately with the recommended dosage of an antimicrobial agent for that type of infection and infecting bacteria. The intermediate category identifies isolates with MICs that approach usually attainable blood and tissue levels and for which response rates may be lower than for susceptible isolates. The intermediate category implies clinical usefulness in body sites where the drug is physiologically concentrated (e.g., quinolones and b-lactams in urine) or when a high dosage of antimicrobial agent can be used (e.g., b-lactams). Resistant strains are not inhibited by usually achievable systemic concentrations of the antimicrobial agent using normal dosage schedules and/or fall in the range in which
specific microbial resistance mechanisms are likely (e.g., b-lactamase enzymes) and clinical efficacy has not been reliable in treatment trials. Serology Serology is useful for diagnosing infectious diseases that are caused by microorganisms that are not easily or quickly recovered by culture. There are two basic approaches to serologic testing: antigen detection and antibody detection. Antigen detection is generally rapid and accurate. It is performed on a specimen of infected tissue (biopsy or body fluid). Antigen detection is useful for many pathogens that are difficult or impossible to culture in the laboratory. Examples include C. trachomatis, HSV, rotavirus, HIV, Legionnaires disease, and syphilis. Monoclonal antibody technology has stimulated a tremendous explosion of current investigative efforts, and several antigen detection systems have become available (e.g., C. trachomatis, HIV). Serologic tests based on antibody detection (Table 2.6) generally are slower and less precise than those based on antigen detection. There are frequent cross-reactions between antibodies of the various organisms. A single specimen that measures immunoglobulin M (IgM) antibody may be diagnostic. However, most available tests measure immunoglobulin G (IgG) and thus cannot differentiate past from present infection. A convalescent specimen is obtained 14 to 21 days after the initial specimen and must show a fourfold rise in antibody titer as an indicator of acute infection.
TABLE 2.6. INFECTIOUS DISEASES COMMONLY DIAGNOSED BY ANTIBODY RESPONSE
Use of serology in the diagnosis of syphilis is fully described in Chapter 7. Initial evaluation involves the use of nontreponemal tests [Venereal Disease Research Laboratories (VDRL), rapid plasma reagin (RPR)]. They are nonspecific and are associated with many false-positives. The major clinical use of the nontreponemal tests is screening, because the tests are quick and inexpensive. Quantitative titers can be used to evaluate therapy. Confirmation of the screening tests for syphilis requires use of
specific treponemal tests. The fluorescent treponemal antibody-absorption (FTA-ABS) test is the most commonly used confirmatory test for syphilis. However, it is expensive and technically difficult. The microhemagglutination–Treponema palladium (MHA-TP) assay is easier to perform and less expensive. It has replaced the FTA-ABS test in many centers. The use of serology for diagnosis of HIV infection is discussed fully in Chapter 10, for the genital mycoplasmas in Chapter 4, and for HSV in Chapter 6. Diagnostic Virology In 2000, Storch (2) reviewed the diagnostic approach to clinically important viral infections. He noted that diagnostic virology is rapidly entering into the mainstream of clinical medicine. Several recent advances and/or developments contributed to this emergence of diagnostic virology, including the following. (i) Dramatic progress in antiviral therapy has increased the need for specific viral diagnoses. (ii) Technologic developments, especially molecular diagnostics, have provided new tools for viral diagnosis. (iii) The HIV/acquired immunodeficiency syndrome epidemic has led to a greater number of patients at risk for opportunistic viral infections. (iv) Modern management of HIV infection and hepatitis C has provided a new paradigm for using molecular techniques in the management of chronic viral infections (2). Multiple methods are available for the laboratory diagnosis of viral infections, including viral culture, antigen detection, nucleic acid detection with or without amplification, and serology (Table 2.7) (2).
TABLE 2.7. TECHNIQUES USED IN DIAGNOSTIC VIROLOGY
Cell culture has been the traditional mainstay of diagnostic virology. With the introduction of new molecular technologies, the relative importance of viral isolation has diminished significantly (2). However, viral culture remains necessary to provide viable isolates for characterization, such as phenotypic antiviral susceptibility testing (2). Herpes simplex virus is the virus for which culture is most useful (2). Viral culture also is
used to detect CMV, varicella-zoster virus, adenovirus, respiratory syncytial virus, influenzae and parainfluenzae viruses, rhinovirus, and enterovirus. However, rapid antigen detection and nucleic acid amplification detection are replacing viral culture for these viruses (2). Antigen detection methods for viral agents include fluorescent antibody staining, immunoperoxidase staining, and EIA. For diagnostic virology, fluorescent antibody staining is the most commonly used antigen detection method (2). The most important viruses identified by antigen detection methods include (i) respiratory syncytial virus, influenzae and parainfluenza viruses, and adenoviruses in respiratory specimens; (ii) HSV and varicella-zoster virus in cutaneous specimens; (iii) rotavirus in stool specimens; and (iv) CMV and hepatitis B virus (hepatitis B surface antigen [HBsAg]) in blood (2). The advent of PCR analysis made possible the diagnosis of viral infection using sensitive detection of specific viral nucleic acids (2). Potentially any virus can be detected using this technology. PCR analysis has been adapted to detect viral RNA by including the use of the enzyme reverse transcriptase. Quantitative nucleic acid detection techniques are available to assess the viral load of HIV and hepatitis C virus. Multiplex PCR can be used to simultaneously detect multiple infectious agents in a single specimen, such as from a genital ulcer. Nucleic acid amplification assays include target amplification (e.g., PCR, LCR, and TMA) or signal amplification assays (e.g., branched-chain DNA assay and hybrid capture assay) (2). Detection of specific antiviral antibodies is a traditional method for diagnosis of viral infections. However, in most instances, the need for acute and convalescent antibody titers has limited the clinical usefulness of serologic diagnosis of viral infection (2). Detection of virus-specific IgM antibodies can diagnose with a single specimen many viral infections, including EBV, CMV, hepatitis A virus, hepatitis B virus (IgM antibody to hepatitis B core antigen), parvovirus B19, measles, rubella, mumps, and arbovirus (2). Serologic testing is useful in chronic infections, such as HIV, where antiviral antibodies always indicate current infection, and hepatitis C virus, where antibodies usually (85%) indicate current infection (2). Serology also is useful to determine immunity to viruses such as varicella-zoster virus, CMV, EBV, HSV, measles, rubella, parvovirus B19, hepatitis A, and hepatitis B (anti-HBsAg). Infections Of The Lower Female Genital Tract The clinically important bacterial infections of the lower genital tract in women fall into four major categories (Table 2.8).
TABLE 2.8. CAUSATIVE AGENTS IN LOWER GENITAL TRACT INFECTION IN WOMEN
Diagnosis of urethral infections caused by N. gonorrhoeae can be made on a slide obtained from a direct smear and stained with methylene blue or Gram stain. Methylene blue is quicker and as accurate as the Gram stain for identifying the typical intracellular diplococci of gonococcus (GC). To obtain an appropriate urethral specimen, a sterile calcium alginate swab (Calgiswab) is inserted 2 to 3 cm into the urethral orifice, gently twirled slowly, and allowed to remain for 10 to 15 seconds. Direct smears for GC from the urethra have a high degree of sensitivity and specificity (11,25). Direct smear with Giemsa straining for detection of C. trachomatis intracellular inclusions can be used with conjunctivitis, but this technique is not appropriate for genital tract specimens. Use of PCR and LCR on female urine specimens for GC has not been the object of any serious study. Studies using urine from symptomatic and asymptomatic chlamydia-infected women have demonstrated the effectiveness of PCR and LCR technology for detecting C. trachomatis (26,27,28,29,30,31,32,33,34 and 35). In females with symptoms and/or signs of urethritis, GC specimens should be obtained for culture from the urethra, and urine cultures should be obtained to rule out the presence of a urinary tract infection (UTI). Isolation of N. gonorrhoeae requires careful specimen collection and handling by the clinician using an enriched bacteriologic medium, high humidity, appropriate incubation temperature, and CO 2 tension for optimal growth. The most common selective media is modified Thayer-Martin medium, which contains vancomycin, colistin, nystatin, and trimethoprim to prevent growth of fungi and non-Neisseria bacteria. Alternative selective media are Martin-Lewis and New York City media (11). The urethral swab is streaked directly onto the selective media using a “Z” format. The swab is rolled over the agar to adequately distribute the microbe on the swab onto the media, then the culture is incubated immediately in a CO2 incubator or candle jar. Cultures should be held overnight at 35°C ± 1°C before transport to the laboratory to establish adequate Neisseria growth. Shipment of specimens before overnight incubation can lower the sensitivity of GC culture by 50%. The agar plates are examined daily for 3 days. Visible colonies are Gram stained to identify Gram-negative diplococci and then tested for oxidase production to confirm the presence of N.
gonorrhoeae. Other commercially available transport media and systems, such as Transgrow, Amies, Neigon, MICROCULT, or JEMBEC, can be used, but their sensitivity is less than that of Thayer-Martin medium unless they reach the laboratory rapidly (less than 24 hours) (25,36). Culture of chlamydia from the urethra of women can be achieved readily if the requisite specimen collection materials are available, along with an appropriate cold chain leading from the patient to the testing laboratory. Thin, 2- to 3-cm Dacron urethral swabs introduced 2 to 3 cm into the female urethra for 10 to 15 seconds must obtain cellular material, as C. trachomatis is an obligate intracellular organism. These swabs are added immediately to a suitable chlamydial transport medium (2SP) that is refrigerated at 4°C or kept on wet ice for transport to the laboratory. Chlamydia culture specimens must reach the infectious disease laboratory within 24 hours of collection, whereupon aliquots can be cultured at once or the entire specimen, containing the swab, can be frozen at –80°C for storage. Chlamydial culture has been replaced to a large extent by antigen detection or nucleic acid amplification methods (Table 2.9). Quinn and coworkers (37) demonstrated that direct immunofluorescence staining (MicroTrak) of urethral specimens had sensitivity of 100%, specificity of 99%, and positive predictive value of 82%. An alternate technique for detecting chlamydial antigens is the use of an immunoassay (Chlamydiazyme). This EIA-type test has been shown to have sensitivity of 81% to 89% and specificity of 90% to 98% (38,39). Polymerase chain reaction amplification of urethral specimens from women attending an STD clinic in Pittsburgh, Pennsylvania, revealed that 35 patients (13.7%) were positive and that, of these 35 women, only 21 (60%) were positive for chlamydia in the cervix (40). Of all the women with urogenital chlamydial infections, chlamydia was identified using PCR in both the cervix and urethra in 52.5% of patients, 35% were positive in the urethra only, and 12.5% were infected solely in the cervix. These data suggest that routine screening for chlamydia infections in women should include an urethral specimen, as well as an endocervical specimen.
TABLE 2.9. STUDIES COMPARING CHLAMYDIA CULTURE, POLYMERASE CHAIN REACTION, AND ENZYME IMMUNOASSAY IN SPECIMENS FROM THE ENDOCERVIX
Cervicitis now is recognized to be a distinct clinical entity, and mucopurulent cervicitis (MPC) has been described as analogous to male urethritis (41). The clinical criteria for MPC are presence of mucopurulent discharge from the endocervix, and erythema, edema, and friability of the cervix. The two major etiologic agents of MPC are N. gonorrhoeae and C. trachomatis. Unlike urethral specimens, direct smears for N. gonorrhoeae from the cervix lack specificity and sensitivity. Diagnosis of N. gonorrhoeae infection of the cervix from symptomatic or asymptomatic patients requires isolation and identification of GC by culture, followed by identification using Gram staining and oxidase production. A single endocervical swab will result in recovery of approximately 90% of N. gonorrhoeae-infected women; two consecutive endocervical swabs or an endocervical plus an anal culture swab will recover 99% of N. gonorrhoeae-infected women. Thus, we recommend the use of two swabs that can be plated onto the same Thayer-Martin culture. Culture remains the “gold standard” for diagnosis of cervical and urethral gonorrhea, but in clinical practice it is rapidly being replaced by molecular diagnostic methods (3,11), particularly DNA hybridization (probe) assays such as PACE 2 (Gen-Probe, San Diego, CA) (42). Establishing chlamydia infections in women has progressed historically from tissue culture isolation to antigen detection tests (DFA, EIA) to nucleic acid amplification (PCR, LCR, TMA). Isolation of C. trachomatis in tissue culture shell vials, microtiter plates, or 24-well tissue culture plates requires the use of a cell culture line (usually cycloheximide-treated McCoy cells), where monolayers of cells are inoculated (using centrifugation) with the chlamydia transport medium containing a swab from either the urethra and/or endocervix. Identification of chlamydia-infected cells is conducted after 48 to 72 hours using a fluorescein-labeled monoclonal antibody. Cell culture for chlamydia currently is performed by only a few laboratories (primarily research centers) (3). Although antigen detection methods for identification of C. trachomatis revolutionized the clinical availability of chlamydial diagnostics, their relative lack of sensitivity and specificity compared to tissue culture and especially nucleic acid amplification tests has limited their clinical usefulness (43,44) (see Chapter 5 for detailed description). Development of specific DNA probes for C. trachomatis (PACE 2) has led to their being the most commonly used diagnostic test for C. trachomatis, especially for screening purposes. Compared to culture, antigen detection, and PCR, the sensitivity and specificity of the PACE 2 probe is in the range of good EIAs but less than that of nucleic acid amplification tests (45,46 and 47). Numerous investigations using PCR for chlamydial infections have been reported. Table 2.9 summarizes several studies of the sensitivity of chlamydia culture, PCR, and EIA for endocervical specimens. At the University of Pittsburgh/Magee-Womens Hospital, we have performed studies where the enhanced sensitivity of PCR analysis was exploited by the collection of vaginal introitus specimens (48,49). Reasoning that minute amounts of DNA from urethral and cervical infections might deposit adequate amounts of chlamydial DNA in the vagina for application and detection by PCR, vaginal swabs were obtained from patients, along with culture specimens from the urethra and cervix (48). Eleven women
were found to be positive for chlamydia infection by culture, whereas 10 of these 11 women were positive by PCR for chlamydial DNA obtained from vaginal swabs. Of the 11 women found to be culture positive from the cervix, eight were culture positive from the urethra. This preliminary work represents an alternative procedure that may offer the patient the opportunity to collect her own vaginal specimen for subsequent drop-off or mailing of the specimen to a centralized STD-PCR laboratory. This self-collected chlamydia specimen may eliminate the need for pelvic examinations in certain clinical situations, enhance the screening of adolescents for chlamydia, and permit greater access to patients who are discouraged from seeking chlamydia testing due to limited laboratory, physician, or health care facilities. In fact, we recently demonstrated the effectiveness of this approach among young female adolescents. A major limitation to widespread screening of asymptomatic males for chlamydial genital infection has been the need to obtain urethral specimens using a swab (50). Screening urine samples with a leukocyte esterase (LE) dipstick has been demonstrated to be a useful tool to identify men with urethritis attributable to C. trachomatis or N. gonorrhoeae (51). However, this approach has a relatively low specificity (false-positives) and does not identify the causal agent (50). Subsequent studies demonstrated that testing urine specimens by EIA to detect chlamydial antigen was a sensitive and specific screening test for asymptomatic male carriers of C. trachomatis (52,53). Gene et al. (50) reported that EIA screening of LE-positive urine specimens was a cost-effective strategy for detecting asymptomatic chlamydial infections in males. These authors noted that positive EIA results in low-risk populations need to be confirmed. Shafer et al. (54) demonstrated that the optimum clinical and cost-effective approach was a combination of nonspecific screening of first-void urine for polymorphonuclear leukocytes or LE, followed by specific testing with EIA with DFA confirmation. Unfortunately, this approach has not been demonstrated to be effective for screening asymptomatic females for chlamydial infection. Thus, to date, screening of asymptomatic females for chlamydial infection has required a pelvic examination with a speculum examination to obtain specimens from the endocervical canal. As described earlier, we demonstrated that a vaginal swab PCR-based test for C. trachomatis could be a useful, noninvasive screening test for asymptomatic chlamydial genital infection in women (48,49). Both PCR and LCR technology have been applied to urine-based screening to detect C. trachomatis among asymptomatic females as well as males. Preliminary results in genitourinary infections have been very encouraging (27,28,29,30,31 and 32). The third pathogen capable of producing cervicitis is HSV. Unlike N. gonorrhoeae or C. trachomatis, HSV involves the ectocervix as well as the endocervix. To date, identification of HSV is routinely accomplished with culture isolation of the viral agent in cell culture. The urogenital swab of the newest lesion is placed in a viral transport media such as Hanks or Earle balanced salt solution supplemented with broad-spectrum antimicrobial agents that will inhibit and kill contaminating bacteria and fungi. For best results, these viral specimens should be delivered to the microbiology laboratory immediately after collection. Anticipated delays of more than 24 hours between collection and culture require viral specimens to be frozen at 70°C to 80°C. Refrigeration of viral specimens held for less than 24 hours is recommended for samples awaiting inoculation into cell culture. Cytopathic effect is observed in most HSV cultures in 18 to 24 hours; 99% of positive cultures are identified within 96 hours. Cytologic examination of specimens demonstrating multinucleated giant
cells using Wright-Giemsa or fluorescent antibody stains from the cervix is diagnostic of HIV infection, but negative smears do not rule out HSV infection. Immunoperoxidase staining of HSV is helpful if the result is positive, but a negative result may not rule out the presence of low copy numbers of HSV particles. Vaginitis is the most common infectious disease seen by obstetrician-gynecologists. The agent responsible for the signs and/or symptoms of vaginitis can be identified rapidly in the office or clinic using simple and inexpensive methodology. The diagnosis of trichomoniasis is generally made by demonstration of motile T. vaginalis organisms on a wet mount preparation; culture of the organism in Diamond TYM medium or Trichicult over a 7-day period is the most sensitive procedure. Direct fluorescent antibody staining can be conducted. Candida albicans infection usually is identified by the presence of pseudohyphae on a 10% KOH preparation or methylene blue stain. Although culture is available, it is generally unnecessary; Nickerson or Sabouraud media are very sensitive and are commonly used by microbiology laboratories. Diagnosis of bacterial vaginosis does not require culture methodology. Nearly 60% of normal healthy women have Gardnerella vaginalis as part of their normal vaginal flora. Clinical criteria for diagnosis of bacterial vaginosis include demonstration of (a) a homogeneous vaginal discharge; (b) vaginal pH more than 4.5; (c) presence of “clue” cells (squamous epithelial cells to whose border are adhered G. vaginalis); and (d) release of a “fishy” amine odor on addition of 10% KOH to vaginal secretions (55). Spiegal (56) and Nugent et al. (57) described the use of a vaginal Gram stain for diagnosis of bacterial vaginosis. They suggest that a normal Gram stain contains lactobacilli, whereas patients with bacterial vaginosis have large numbers of a variety of other organisms. Infections Of The Upper Genital Tract Infections involving the upper genital tract (i.e., uterus, fallopian tubes, ovaries, and pelvic peritoneal cavity) are typically of mixed anaerobic-aerobic bacterial etiology (58) and are discussed in depth in Chapter 8. Pelvic inflammatory disease (PID) is associated with sexually transmitted organisms such as N. gonorrhoeae andC. trachomatis (59). Table 2.10 lists the more common type of soft tissue pelvic infections encountered in clinical practice, the appropriate site from which to obtain specimen(s), and the microorganisms assumed to be pathogens at these anatomic sites. The specific microorganisms involved in these pelvic infections are discussed in Chapter 1.
TABLE 2.10. APPROPRIATE MICROBIOLOGIC SPECIMENS FROM SOFT TISSUE INFECTIONS OF THE FEMALE UPPER GENITAL TRACT
The pathogens involved in upper genital tract infections generally arise from the normal flora of the vagina and cervix (except N. gonorrhoeae, C. trachomatis, and group A b-hemolytic streptococcus); therefore, microbiologic isolation attempts must bypass the heavy bacterial colonization of the lower female genital tract (58). In addition, anaerobic bacteria are such prevalent pathogens in pelvic infections that microbiologic procedures should be used that are adequate and appropriate for the recovery and identification of anaerobic bacteria. In patients with endomyometritis following delivery (vaginal or cesarean) or abortion, the appropriate specimen for culture is an endometrial isolate. This specimen can be best obtained using a telescoping double plastic cannula with an inner wire brush (60), which nicely prevents contamination with the vaginal and cervical flora, or the Pipelle device (Unimar, Wilton, CT), which also has been demonstrated to provide excellent endometrial specimens for microbiologic analysis (61). The cervix is not an appropriate site from which to obtain specimens for anaerobic and facultative bacteria, because the cervix has a rich microflora of both anaerobes and aerobes and does not accurately reflect the pathogens present in the upper genital tract. Patients with acute PID should have their cervix cultured, sampled for antigen, or assayed by nucleic acid amplification tests for both N. gonorrhoeae and C. trachomatis as described in the section on Infections of the Lower Female Genital Tract. Specimens from the endometrial aspirate are submitted for isolation of anaerobic and facultative bacteria, C. trachomatis, and N. gonorrhoeae. Nucleic acid amplification technology can be used to detect C. trachomatis from endometrial specimens. In situ hybridization also has been used on endometrial biopsy specimens (62). The fallopian tube is the ideal site to obtain microbial specimens from patients with acute PID. Laparoscopy is required to obtain specimens and cultures from the tube but unfortunately is limited due to logistic and economic concerns. When laparoscopy is performed, cultures should be obtained from the fallopian tubes and cul-de-sac for anaerobic and facultative bacteria, N. gonorrhoeae and C. trachomatis. The usefulness
of direct smears of fallopian tube specimens with fluorescein-labeled monoclonal antibody against C. trachomatis was described by Kiviat and coworkers (62). Culdocentesis was widely used to obtain peritoneal fluid as a specimen, which is processed for anaerobic and facultative bacteria, N. gonorrhoeae, and C. trachomatis. Culdocentesis is superior to cervical cultures but is of limited value because of the high degree of vaginal contamination of these specimens (63). Endometrial aspirates have been used for the culture of anaerobic and facultative bacteria, N. gonorrhoeae, and C. trachomatis from the upper genital tract of patients with acute PID (64). The endometrial aspirate is a more accurate reflection of the microbiologic etiology of PID than is culdocentesis and is better tolerated by patients (59). Specimens from patients with posthysterectomy pelvic cellulitis should be obtained from the peritoneal cavity. This is best accomplished via culdocentesis and submitted for anaerobic and facultative cultures. In the presence of postoperative cuff abscess, the aspirate of the abscess is the ideal specimen and should be cultured for anaerobic and aerobic bacteria. Similarly, the appropriate specimen from patients with a pelvic abscess is purulent material aspirated from the abscess (obtained in the operating room or percutaneously) and evaluated for anaerobic and facultative bacteria and, in the case of tuboovarian abscess, N. gonorrhoeae. The ideal specimen from an abscess for culture, DNA amplification, or in situ hybridization is material or tissue obtained from the abscess wall at the time of surgery. Collection, handling, and transport of specimens obtained from the upper genital tract require special procedures, because anaerobic bacteria must be obtained using methods that ensure their survival. Aspiration of fluid or pus into a syringe is the preferred method for collection of these specimens. The syringe should not be used for transport; rather, the specimen should be injected into an anaerobic transport vial. Swabs are not recommended to obtain specimens for anaerobic cultures if fluid is available. Use of transport media is easy and convenient, but is associated with a false-negative rate of 40%. The preferred method is to use anaerobically sterilized, prereduced cultures, such as CO2-filled tubes or gas packs. Rubber-stoppered collection tubes that had O2 evacuated with CO2 are reliable tools for collection of aspirates (65). Anaerobic vials are commercially available; these include Port-A-Cul (BBL Microbiology System, Cockeysville, MD), Anaerobic Transport Medium (Anaerobe Systems, San Jose, CA), and Anaport and Anatube (Scott Laboratories, Fiskeyville, RI). A single specimen will suffice for anaerobic and facultative organisms; the latter grow under anaerobic as well as aerobic conditions. All anaerobic specimens require prompt transport to the infectious disease laboratory if they are to survive. The Gram stain is an important clinical tool in the management of upper genital tract infections. Gram staining allows for prompt identification of bacterial pathogens and provides a quality control mechanism on the culture techniques being used, especially as related to the recovery of anaerobes. The presence of bacteria on the Gram-stained smear in the face of negative culture results should raise concerns as to the adequacy of the anaerobic procedures being used for the collection, handling, or transport of specimens to the laboratory and/or the laboratory's capability to grow and identify anaerobic bacteria.
Urinary Tract Infections Infections involving the urinary tract are among the most common infections seen by obstetrician-gynecologists in their practice. Thus, it is important to understand the methods available for proper collection and handling of urine specimens for microbiologic evaluation (1). Although the absolute criteria for diagnosis of UTI in both symptomatic and asymptomatic patients rest on the presence of a positive culture from a properly collected and handled urine specimen that has not been contaminated by vulvar/vaginal bacteria, immediate microscopic examination of urine sediment often may be helpful (1). Identification of bacteria in urine allows for initiation of empiric therapy in the symptomatic patient while waiting for culture confirmation of UTI. A midstream clean-catch urine is the most widely obtained specimen for both culture and microscopic evaluation of urine. It is important that the patient be properly instructed on how to obtain a midstream, clean-catch urine specimen. A minimum of 4 hours should have elapsed since the last voiding. An alternative method, favored by some urologists but not commonly used by obstetrician-gynecologists, is suprapubic bladder aspiration. If this method is used, the specimen should be identified as such so that all growth is reported and identified (3). Kass (66) popularized the use of quantitative urine cultures where the criterion of ³10 5 bacteria per milliliter became the diagnosis of UTI. This standard is reliable for acute pyelonephritis and asymptomatic bacteriuria. However, Stamm and coworkers (67) demonstrated that a criterion of ³105 bacteria per milliliter is unreliable in association with acute bacterial cystitis. These investigators noted a 50% false-negative rate with a single specimen using this criterion. In patients with acute cystitis, a midstream clean-catch urine with ³102 bacteria per milliliter is sufficient to confirm the diagnosis of cystitis (67). Urine at room temperature is an excellent culture medium for bacteria. In such conditions, the number of bacteria in urine doubles every 45 minutes. Thus, a 6-hour delay at room temperature will result in 103 bacteria per milliliter increasing to ³105 bacteria per milliliter. It is imperative that urine specimens be transported promptly to the laboratory. Refrigerate urine specimens if a delay of more than 1 to 2 hours will occur. The most rapid method to detect ³10 5 bacteria per milliliter is microscopic examination of a urine specimen. A Gram stain of a well-mixed, unspun urine demonstrating 2 bacteria per high-power field has 90% correlation with results of quantitative bacteriologic cultures. In addition, this procedure allows for determination of the adequacy of the specimen by demonstrating the absence of epithelial cells. In contradistinction, demonstration of pyuria (more than five white blood cells per high-power field of a centrifuged urine specimen) has variable sensitivity (pyuria is present in 90% of symptomatic UTIs, but in only 50% of asymptomatic bacteriuria) and poor specificity (3,68). Several chemical methods have been developed to detect significant bacteriuria in urine
specimens (3). These include the Griess test, tetrazolium reduction, glucose oxidase, and catalase. The best of the chemical tests is the Griess test, which measures nitrite. However, it is valid only for Gram-negative bacteria, is useful only if the result is positive, and has a false-negative rate of nearly 50%. It is commercially available as N-Multistix (Ames), Microstix-3 (Ames), Chemstrip 8 (Bio Dynamics), or Bac-U-Dip (Warner/Chilcott). In office practice, especially for routine screening of prenatal patients, a multitude of office urine culture kits are available. Basically, there are three categories of culture kits: (i) dip strip with culture pads (Microstix-3); (ii) coated shell, coated tube, or agar cup (Testuria, Speri-Test, Bacturcult); and (iii) dip slide (Uricult; Medical Technology Corp.) and Dipchex (York Scientific). Of these, the dip slide is the most versatile and easiest to use. Most importantly, it is the most accurate, with a 99% correlation with standardized quantitative culture techniques. Thus, it permits inexpensive culture both before therapy and as a test of cure. Any organism isolated remains viable for several weeks and is available to be sent to the laboratory for identification and susceptibility testing, if necessary. Bacteremia It is estimated that 200,000 cases of septicemia occur each year in the United States, with a 20% to 50% mortality rate (69,70). Fortunately, bacteremia is not a common occurrence in patients with infection on obstetric and gynecologic services (71). However, bacteremia represents a serious and potentially life-threatening complication of pelvic infections. Thus, it is important to recognize the presence of bacteremia. To optimize the ability to diagnose bacteremia, clinicians must recognize the clinical determinants of positive blood cultures. Most bacteremias are intermittent and usually precede the onset of fever and chills. An exception is intravascular sources, such as endocarditis, in which bacteremias may be continuous (72). The sensitivity of most media used in blood culture attempts is two to three organisms per milliliter; however, the magnitude for bacteremia is low. Adults with endocarditis or Gram-negative bacteremia generally have ten to 100 colony-forming units per milliliter (73,74). Previous antibiotic therapy (even up to 2 weeks before) may completely or partially suppress growth of bacteria, giving false-negative results. The two major variables in detection of bacteremia are the timing of blood cultures and the volume of blood collected. In endocarditis, where bacteremia is continuous, two cultures (from separate venipunctures) at one time usually are adequate. Soft tissue pelvic infections are associated with intermittent bacteremias. In such bacteremias, Washington (74) demonstrated that the sensitivity of a single culture is 80%. Two separate cultures drawn at least 1 hour apart have 89% sensitivity, and three separate cultures drawn at least 1 hour apart have 99% sensitivity. The volume of blood is probably the single most important determinant of the appropriateness of a blood culture specimen (3,75). A 30-mL sample is associated with a 50% increase in positive results compared with a 10-mL sample, as is drawn by most laboratories. The current recommendation for adult patients is to draw at least two separate blood cultures totaling 30 to 40 mL (3). The optimal time for collection of blood specimens is just before
onset of a chill (76); however, such an event cannot be anticipated. It is crucial to differentiate the presence of true pathogens from contaminants when interpreting blood culture results (77). With pathogenic organisms, numerous cultures are positive, and less than 3 days are required for detection of organisms. Pathogenic organisms include Gram-negative bacilli, S. aureus, Enterococcus faecalis, viridans streptococci, anaerobes, fungi, and yeasts. Staphylococcus epidermidis has become a recognized and serious pathogen in certain clinical situations. This is especially true of nosocomial (hospital-acquired) infections in patients with central lines and invasive monitoring. In contrast, contaminants are characterized by a single positive culture, and more than 3 days are required for detection of organisms. Typical contaminants in blood cultures include Corynebacterium sp, Propionibacterium sp, and, in most circumstances, especially community-acquired infections, S. epidermidis. To prevent such contamination, the venipuncture site must be carefully disinfected before obtaining the blood culture. This is crucial because these commensals may be involved in infections of implanted prosthetic material. To help identify true catheter-related infections, semiquantitative catheter tip cultures per the guidelines of Maki et al. (78) are used. A significant advance has been the introduction of continuous-monitoring blood culture systems (3,77). With these systems, a growth reading is performed automatically every 10 to 20 minutes, which allows detection of positive cultures as quickly as possible. In addition, the bottles need not be opened to detect growth, which significantly reduces the risk of contamination (3). Although the incidence of bacteremia arising from soft tissue pelvic infections is relatively low, blood cultures are a valuable diagnostic tool in evaluating such infections. A positive culture demonstrating bacteremia identifies a subgroup of patients with severe pelvic infections and who may require more lengthy courses of therapy. CHAPTER REFERENCES 1. Ledger WJ. Infection in the female. Philadelphia, Lea & Febiger, 1977. 2. Storch GA. Diagnostic virology. Clin Infect Dis 2000;31:739–751. 3. Gill VJ, Pedorko DP, Witebsky FG. The clinician and the microbiology laboratory In: Mandel GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. Philadelphia: Churchill-Livingstone, 2000:184–221. 4. Health Care Financing Administration. Medicare, Medicaid and CLIA programs. Regulations implementing the Clinical Laboratory Improvement Amendments of 1988 (CLIA). Fed Register 1992;57:7002–7186. 5. Elder BL, Hansen SA, Kellogg JA, et al. Verification and validation of procedures in the clinical microbiology laboratory. Cumitech 31. Washington, DC: American Society for Microbiology, 1997. 6. Schachter J, Stamm WE, Quinn T, et al. Ligase chain reaction to detect Chlamydia trachomatis infection of the cervix. J Clin Microbiol 1994;32:2540–2543. 7. Moseley SL, Hug I, Alim ARMA, et al. Detection of enterotoxigenic Escherichia coli by DNA colony hybridization. J Infect Dis 1980;142:892–898. 8. Tenover FC. Diagnostic deoxyribonucleic acid probes for infectious diseases. Clin Microbiol Rev 1988;1:82–101. 9. Tenover FC, Unger ER. Nucleic acid probes for detection and identification of infectious agents In: Persing DH, Smith TF, Tenover FC, et al., eds. Diagnostic molecular microbiology: principles and applications. Washington, DC: American Society for Microbiology, 1993:3–25. 10. Tenover FC. DNA hybridization techniques and other application to the diagnosis of infectious diseases. Infect Dis Clin North Am 1993;7:171–181. 11. Chernesky MA. Laboratory services for sexually transmitted diseases: overview and recent
12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.
developments. In: Mandel GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. Philadelphia: Churchill-Livingstone 2000:1281–1294. Britten RJ, Davidson EH. Hybridization strategy. In: Hames BD, Higgins SJ, eds. Nucleic acid hybridization: a practical approach.. Oxford: JRL Press, 1985:3. Saiki RK, Gelfand DH, Stoffel S, et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988;239:487–491. Yamashita Y, Kohno S, Koga H, et al. Detection of Bacteroides fragilis in clinical specimens by PCR. J Clin Microbiol 1994;32:679–683. Persing DH. In vitro nucleic acid amplification techniques In: Persing DH, Smith TF, Tenover FC, et al., eds. Diagnostic molecular microbiology: principles and applications. Washington, DC: American Society of Microbiology, 1993:51–87. Wu DY, Wallace RB. The ligase-amplification reaction (LAR): amplification of specific DNA sequences using sequential rounds of template dependent ligation. Genomics 1989;4:560–569. National Committee for Clinical Laboratory Standards. Performance for antimicrobial disk susceptibility tests, 6th ed. Approved standard. NCCLS document MS-A6. Wayne, PA: National Committee for Clinical Laboratory Standards, 1997. National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 4th ed. Approved standard. NCCLS document M7-A4. Wayne, PA: National Committee for Clinical Laboratory Standards, 1997. National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing, 8th ed. Informational supplement. NCCLS document M100-S8. Wayne, PA: National Committee for Clinical Laboratory Standards, 1998. National Committee for Clinical Laboratory Standards. Methods for antimicrobial susceptibility testing of anaerobic bacteria, 4th ed. Approved standard. NCCLS document M11-A4. Wayne, PA: National Committee for Clinical Laboratory Standards, 1997. Finegold SM, Jousimies-Somer HR, Wexler HM. Current perspectives on anaerobic infections: diagnosis approaches. Infect Dis Clin North Am 1993;7:257–275. National Committee for Clinical Laboratory Standards. Methods for antimicrobial susceptibility testing of anaerobic bacteria, 2nd ed. Approved standard. NCCLS publication MII-A1. Villanova, PA: National Committee for Clinical Laboratory Standards, 1990. Citron DM, Ostavari MI, Karlson A, et al. Evaluation of the epsilometer (E-test) for susceptibility testing of anaerobic bacteria. J Clin Microbiol 1991;29:2197. Jorgensen JH. Antimicrobial susceptibility testing of bacteria that grow aerobically. Infect Dis Clin North Am 1993;7:393–409. Riccardi NB, Felman YM. Laboratory diagnosis in the problem of suspected gonococcal infection. JAMA 1979;242:2703–2705. Jaschek G, Gaydos CA, Welsh LE, et al. Direct detection of Chlamydia trachomatis in urine specimens from symptomatic and asymptomatic men by using a rapid polymerase chain reaction assay. J Clin Microbiol 1993;31:1209–1212. Schachter J, Moncada J, Whidden R, Shaw H, Bolan G, Burczak JD, Lee HH. Noninvasive tests for diagnosis of Chlamydia trachomatis infection: application of ligase chain reaction to first-catch urine specimens of women. J Infect Dis 1995;172(5):1411–1414. Bauwens JE, Clark AM, Loeffelholz MR, et al. Diagnosis of Chlamydia trachomatis urethritis in men by polymerase chain reaction assay of first-catch urine. J Clin Microbiol 1993;31:3013–3016. Wiesenfeld HC, Uhrin M, Dixon BW, et al. Diagnosis of male Chlamydia trachomatis urethritis by polymerase chain reaction. Sex Transm Dis 1994;21:268–271. Schachter J, Stamm WE, Quinn TC, et al. Ligase chain reaction to detect Chlamydia trachomatis infection of the cervix. J Clin Microbiol 1994;32:2540–2543. Lee HH, Chernesky MA, Schachter J, et al. Diagnosis of Chlamydia trachomatis genitourinary infection in women by ligase chain reaction assay of urine. Lancet 1995;245:213–216. van Doornum GJ, Buimer J, Prins M, et al. Detection of Chlamydia trachomatis infection in urine samples from men and women by ligase chain reaction. J Clin Microbiol 1995;33:2042–2047. Vogels WHM, Van Voorst Vader PC, Schroder FP. Chlamydial trachomatis infections in a high-risk population: comparison of polymerase chain reaction and cell culture for diagnosis and follow-up. J Clin Microbiol 1993;31:1103–1107. Bass CA, Jungkind DL, Silverman NS, et al. Clinical evaluation of a new polymerase chain reaction assay for detection of Chlamydia trachomatis in endocervical specimens. J Clin Microbiol 1993;31:2648–2653.
35. Loeffelholz MJ, Lewinski CA, Silver SR, et al. Detection of Chlamydia trachomatis in endocervical specimens by polymerase chain reaction. J Clin Microbiol 1992;30:2847–2851. 36. Spence MR, Guzick DS, Katta RA. The isolation of Neisseria gonorrhoeae: a comparison of three culture transport systems. Sex Transm Dis 1983;10;138–140. 37. Quinn TC, Warfield P, Kappue E, et al. Screening for Chlamydia trachomatis infection in an upper-city population: a comparison of diagnostic methods. J Infect Dis 1985;152:419–423. 38. Jones MF. Detection of Chlamydia trachomatis in genital specimens by the Chlamydiazyme test. J Clin Microbiol 1984;20:465–467. 39. Amortegui AJ. Enzyme immunoassay for detection of Chlamydia trachomatis from the cervix. Obstet Gynecol 1985;65:523–526. 40. Wiesenfeld HC, Uhrin M, Dixon BW, et al. Rapid polymerase chain reaction-based test for the detection of female urogenital chlamydial infections. Infect Dis Obstet Gynecol 1994;1:182–187. 41. Brunham RC, Paavonen J, Stevens CE, et al. Mucopurulent cervicitis: the ignored counterpart in women of urethritis in men. N Engl J Med 1984;311:1–6. 42. Hale YM, Melton ME, Lewis JS, et al. Evaluation of the PACE-2 Neisseria gonorrhoeae assay by three public health laboratories. J Clin Microbiol 1993;31:458–459. 43. Stamm WE, Harrison HR, Alexander ER, et al. Diagnosis of Chlamydia trachomatis infections by direct immunofluorescence staining of genital secretion: a multicenter study. Ann Intern Med 1984;101:638–641. 44. Stamm WE. Diagnosis of Chlamydia trachomatis genitourinary tract infection. Ann Intern Med 1988;108:710. 45. Lebar W, et al. Comparison of DNA probe, monoclonal antibody enzyme immunoassay and cell culture detection of Chlamydia trachomatis. J Clin Microbiol 1989;27:826–828. 46. Clarke LM, et al. Comparison of Syva Microtrak enzyme immunoassay and Gen-Probe PACE 2 with cell culture for diagnosis of cervical Chlamydia trachomatis infection in a high-prevalence female population. J Clin Microbiol 1993;31:968–971. 47. Altwegg M, et al. Comparison of Gen-Probe PACE 2, Amplicor Roche, and a conventional PCR for the detection of Chlamydia trachomatis in genital specimens. Med Microbiol Lett 1994;3:181–1887. 48. Weisenfeld HC, Heine RP, Rideout A, et al. The vaginal introitus: a novel site for Chlamydia trachomatis testing in women. Am J Obstet Gynecol 1996;174:1542–1546. 49. Wiesenfeld HC, Lowry DB, Heine RP, et al. Self-collection of vaginal swabs for the detection of chlamydia, gonorrhea and trichomoniasis: opportunity to encourage STD testing among adolescents. Sex Transm Dis 2001;28:321–325. 50. Gene M, Ruusuvaara L, Mardh P-A. An economic evaluation of screening for Chlamydia trachomatis in adolescent males. JAMA 1993;270:2057–2064. 51. Shafer M-A, Schachter J, Moscicki AB, et al. Urinary leukocyte esterase screening test for asymptomatic chlamydial and gonococcal infections in males. JAMA 1990;262:2562–2566. 52. Stacy A, Steyrer K, Heller-Vitouch C, et al. Screening for Chlamydia trachomatis in military personnel by urine testing. Infection 1991;19:205–207. 53. Leonardi GP, Seitz M, Edstrom R, et al. Evaluation of three enzyme immunoassays for detection of Chlamydia trachomatis in urine specimens from asymptomatic males. J Clin Microbiol 1992;30:2793–2796. 54. Shafer M-A, Schachter J, Moncada J, et al. Evaluation of urine-based screening strategies to detect Chlamydia trachomatis among sexually active asymptomatic males. JAMA 1993;270:2065–2070. 55. Amsel R, Totten PA, Spiegal CA, et al. Nonspecific vaginitis: diagnostic criteria and microbiologic and epidemiologic associations. Am J Med 1983;74:14–22. 56. Speigal CE. Diagnosis of bacterial vaginosis by direct gram stain of vaginal fluid. J Clin Microbiol 1983;18:170–177. 57. Nugent RP, Krohn MA, Hillier SL. The reliability of diagnosing vaginosis is improved by a standardized method of Gram stain interpretation. J Clin Microbiol 1991;29:297–301. 58. Sweet RL. Treatment of mixed aerobic-anaerobic infections of the female genital tract. J Antimicrob Chemother 1981;8[Suppl D];105–114. 59. Sweet RL, Schachter J, Robbie MO. Failure of beta-lactam antibiotics to eradicate Chlamydia trachomatis from the endometrium of patients with acute salpingitis despite apparent clinical cure. JAMA 1983;250:2641–2645. 60. Knuppel RA, Scerbo JC, Mitchell GW, et al. Quantitative transcervical uterine culture with a new device. Obstet Gynecol 1981;57:243–248.
61. Martens MG, Faro S, Hammill HA, et al. Transcervical uterine cultures with a new endometrial suction curette: a comparison of three sampling methods in postpartum endometritis. Obstet Gynecol 1989;74:273–276. 62. Kiviat NB, Wolner-Hanssen P, Peterson M, et al. Localization of Chlamydia trachomatis infection by direct immunofluorescence and culture in pelvic inflammatory disease. Am J Obstet Gynecol 1986;154:865–873. 63. Sweet RL, Draper DL, Schachter J, et al. Microbiology and pathogenesis of acute salpingitis as determined by laparoscopy: what's the appropriate site to sample? Am J Obstet Gynecol 1980;138:985–989. 64. Sweet RL, Landers DV, Schachter J. Chlamydial infections in obstetrics and gynecology. Clin Obstet Gynecol 1983;26:143–164. 65. Baron EJ, Peterson LR, Finegold SM. Bailer and Scott's diagnostic microbiology, 9th ed. St. Louis: Mosby, 1994:57–58. 66. Kass EH. Asymptomatic infections of the urinary tract. Trans Assoc Am Physicians 1956;69:56. 67. Stamm WE, Counts GW, Running KR, et al. Diagnosis of coliform infection in acutely dysuric women. N Engl J Med 1982;307:463–468. 68. Kunin CM. Detection, prevention and management of urinary tract infections, 3rd ed. Philadelphia: Lea & Febiger, 1979. 69. Washington J II, Ilstrup D. Blood cultures: issues and controversies. Rev Infect Dis 1986;8:792–802. 70. Leibovici L, Konisberger H, Pitlik S, et al. Bacteremia and fungemia of unknown origin in adults. Clin Infect Dis 1992;14:436–443. 71. Ledger WJ, Norman M, Gee C, et al. Bacteremia on an obstetric-gynecologic service. Am J Obstet Gynecol 1975;121:205–212. 72. Werner AS. Studies on the bacteremia of bacterial endocarditis. JAMA 1967;202:130–134. 73. Kreger BE. Gram negative bacteremia. III. Reassessment of etiology and ecology in 612 patients. Am J Med 1980;68:332–343. 74. Washington JA, ed. the Detection of septicemia. Boca Raton, FL: CRC Press, 1978. 75. Ilstrup DM, Washington JA. The importance of volume of blood cultured in the detection of bacteremia and fungemia. Diagn Microbiol Infect Dis 1983;1:107–110. 76. Smith-Elekes S, Weinstein MP. Blood cultures. Infect Dis Clin North Am 1993;7:221–234. 77. Weinstein MP. Current blood culture methods and systems: clinical concepts, technology, and interpretation of results. Clin Infect Dis 1996;23:40–46. 78. Maki DG, Weise CE, Sarafin HW. A semiquantitative culture method for identifying intravenous catheter related infection. N Engl J Med 1977;296:1305–1309.
3 GROUP B STREPTOCOCCI Infectious Diseases of the Female Genital Tract
3 GROUP B STREPTOCOCCI Organism Epidemiology Clinical Manifestations Neonatal Infection with Group B Streptococcus Maternal Infection with Group B Streptococci Diagnosis of Group B Streptococcal Infection Treatment of Group B Streptococcal Infection Prevention of Group B Streptococcal Infection Antepartum Antibiotics Intrapartum Antibiotics Neonatal Antibiotics Maternal Vaccination Aftermath of the 1996 Guidelines Chapter References
Since the last edition of this text, considerable progress has been made in the prevention of perinatal group B streptococcal (GBS) infections. In the spring of 1996, the Centers for Disease Control and Prevention (CDC) published recommendations for two preventative strategies. The American College of Obstetricians and Gynecologists (ACOG) endorsed these recommendations. In the ensuing years, there was growing evidence that adoption of these guidelines reduced the likelihood of GBS perinatal infection. However, we do not have a panacea, as both immediate and long-term questions remain. This heavily revised chapter will detail the national guidelines as well as provide insight into their limitations. In 1933, Lancefield (1) used serologic techniques to subdivide b-hemolytic streptococci into specific groups, which she named A, B, D, and E. The various groups of streptococci, their taxonomic designations, and hemolytic reactions on blood agar are listed in Table 3.1. On blood agar, the b reaction is clear or there is complete hemolysis around the bacterial colony; the b reaction is a greenish discoloration or there is partial hemolysis around the colony; and the b reaction is an absence of hemolysis around the colony. Most microbiology laboratories report groups C and G “as b-hemolytic streptococci, not groups A, B, or D.”
TABLE 3.1. AEROBIC STREPTOCOCCI: TAXONOMIC CLASSIFICATION AND REACTION ON BLOOD AGAR
Although GBS were recognized in veterinary medicine for many years as a cause of bovine mastitis, they were virtually ignored as human pathogens until 1964, when Eickhoff and associates (2) noted the role of GBS in perinatal infections. In the following years, GBS became the leading bacterial pathogens in perinatal infections, replacing Escherichia coli as the most frequent microorganisms associated with bacteremia or meningitis among infants during the first 2 months of life (3). In 1990, it was estimated that there were 7,600 cases of neonatal GBS disease, accounting for an incidence rate of 1.8 per 1,000 liveborn infants. These cases resulted in an estimated 310 deaths from GBS disease among infants younger than 90 days of age in the United States (4,5). Group B streptococcus has been recognized as one of the most important pathogens in obstetric patients as well. This organism causes urinary tract infections, amnionitis, postpartum endometritis, wound infection, and intrapartum and/or postpartum bacteremia (6,7). It may lead to premature rupture of membranes (PROM) and preterm delivery (4,5,8).
ORGANISM Group B streptococcus (Streptococcus agalactiae) is a facultative Gram-positive diplococcus. Colonies of GBS on sheep blood agar produce a characteristic appearance with narrow zones of b hemolysis surrounding the colonies, which are gray-white, flat, and mucoid. Approximately 1% of GBS isolates are nonhemolytic or b-hemolytic. Definitive microbiologic identification of GBS requires serologic techniques for detection of the group B antigen. Group B streptococci can be subdivided further into serotypes based on antigenic carbohydrates as Ia, Ib, II, III, IV, and V. In the past, the distribution of serotypes of GBS has been similar for mother-neonate pairs. Approximately one third of isolates are type Ia or Ib; one third are type II; and one third are type III (3). More recently, serotypes IV and V have been identified. In addition, a former subtype Ic has been categorized as identical to subtype Ia. In recent surveys, distribution of GBS serotypes is as follows: type Ia, 27% to 36%; type Ib, 2.4% to 12.5%; type II, 9% to
12%; type III, 14% to 43%; and type V, 10% to 29%. Type IV isolates are seen very infrequently, and approximately 0% to 3% cannot be typed (9).
EPIDEMIOLOGY Asymptomatic vaginal colonization with GBS occurs in approximately 20% (range 4.6% to 40.6%) of pregnant women (Table 3.2) (2,10,11,12 and 13). The gastrointestinal tract is the likely reservoir (4). The reported prevalence of vaginal colonization with GBS in gravid women varies according to geographic locale, age, gravidity, duration of gestation, and location and number of sites cultured. Obtaining specimens for culture from both the anorectum and the distal vagina increases the likelihood of obtaining GBS by a considerable percentage (5% to 25%) over vaginal culture alone (4). Within the genital tract, the highest isolation rates are reported from the introitus and the lowest from the cervix. Pregnancy does not influence colonization rates.
TABLE 3.2. EPIDEMIOLOGY OF GROUP B STREPTOCOCCUS PERINATAL INFECTION
The choice of culture medium is a crucial determinant of the prevalence of GBS. The highest yield of GBS occurs when a selective broth medium, such as Todd-Hewitt broth with sheep blood, nalidixic acid, and gentamicin, is used. When selective media are not used for cultures of the genital tract, up to 50% of colonized women will be missed (i.e., the sensitivity will be as low as 50%). The presence of GBS in the maternal genital tract is the major determinant of both infection in the neonate and colonization of the newborn (10,12,14,15 and 16). Vertical transmission from mother to fetus occurs either via an ascending route in utero, most commonly with ruptured membrane (but occasionally with intact membranes), or by acquisition during passage through the birth canal intrapartum. The risk of transmission has been shown to range from 42% to 72% among neonates born to colonized mothers; approximately only 8% of infants born to noncolonized mothers become colonized (10,11,12 and 13). Based on these prospective studies, nearly two thirds of infants born
to colonized mothers will be colonized with GBS. Despite high prevalence rates for vertical transmission, the incidence of GBS infection during the first 7 days of life ranges from 1.3 to 3 per 1,000 livebirths; after 7 days, the range is 1 to 1.7 per 1,000 livebirths (4,10). (These data are from an era before use of prophylaxis.) Only 1% to 2% of infants of colonized women develop early-onset GBS infection. In addition to maternal-infant transmission, nosocomial acquisition of GBS occurs (17). From 16% to 47% of nursery personnel are carriers of GBS and may be sources for neonatal transmission (3). Studies have demonstrated nosocomial transmission rates of GBS in neonates born to culture-negative mothers. In addition, cross-contamination of infants may arise via nursery personnel (3). Nosocomial transmission rates of GBS in neonates born to culture-negative mothers range from 13% to 43% (17). Whether late-onset GBS infection is largely due to nosocomial acquisition of GBS is unclear. Both nonmaternal and maternal sources have been implicated (3). Risk factors that predispose the neonate to clinical infection have been identified (Table 3.3).
TABLE 3.3. RISK FACTORS FOR EARLY-ONSET NEONATAL INFECTION WITH GROUP B STREPTOCOCCIa
Investigations have shown consistently that there is a significant 10- to 15-fold increase in the risk for GBS early-onset infection in preterm infants (10,14,15 and 16,18). Baker and Barrett (10) reported that 80% of infants with GBS disease weighed less than 2,500 g. Boyer and coworkers reported that as birthweight decreased, attack rates for GBS increased (Table 3.4) (14).
TABLE 3.4. ATTACK RATES OF NEONATAL GROUP B STREPTOCOCCAL EARLY-ONSET DISEASE ACCORDING TO PERINATAL CHARACTERISTICSa
Numerous studies found duration of rupture of the membranes (ROM) to be an important correlate of neonatal GBS infection (15). Boyer and colleagues (15) noted that as the duration of ROM increased, attack rates for GBS infection also increased, from 0.8 per 1,000 infants with ROM of 6 hours or less to 10.8 per 1,000 in those with durations greater than 48 hours. In this study, the relative risk of developing GBS early-onset disease for infants with ROM for more than 48 hours was 7.2-fold higher than that for infants with ROM for less than 18 hours (Table 3.4) (15). Further, Boyer and colleagues estimated that, in their population, the attack rate of GBS sepsis was 45 per 1,000 infants born to colonized women with perinatal risk factors (birthweight less than 2,500 g, ROM more than 18 hours, or intrapartum fever) (15). The relationship of GBS to premature birth and PROM is discussed in Chapter 17. Intrapartum fever, clinical chorioamnionitis, and clinical intraamniotic infection are perinatal risk factors associated with increased risk for early-onset GBS sepsis (Table 3.3 and Table 3.4) (10,15). Boyer and coworkers (15) noted that attack rates for GBS were 6.5 per 1,000 infants among mothers with intrapartum fever versus 1.5 per 1,000 infants of afebrile mothers. Neonates delivered through birth canals heavily colonized with GBS are significantly more likely to acquire GBS than those born to mothers with small numbers of GBS in their vaginas (11). Additional risk factors have been suggested. Baker and Kasper (19) proposed that the offspring of mothers who lack antibodies to GBS type III have a greater risk of acquiring type-specific disease. Baker and Edwards (3) suggested two mechanisms that result in low levels of neonatal antibody to GBS: it either reflects low maternal levels and lack of acquired immunity in the mother, or failure of adequate concentrations of these immunoglobulin G (IgG) antibodies to be transported transplacentally to the fetus. The very small preterm infant (younger than 30 weeks) may be significantly compromised by the second mechanism, because nearly two thirds of maternally derived IgG is actively
transported from mother to fetus in the last 10 weeks of pregnancy. In a CDC review of 278 cases of infant disease, there were 247 (80.7%) early-onset cases and 59 (19.3%) late-onset cases. The case fatality rate was 5.8% (16/278), with no difference in rates for early and late-onset disease (5.7% and 6.0%, respectively). Thirteen (81.3%) of the deaths occurred in infants of less than 34 weeks' gestation, and eight of the infants were black (20). In a population-based study, 71 cases of early-onset disease and 37 cases of late-onset disease were compared with 65,000 births in the same period (1982 to 1983). Infants with early-onset disease were more likely to be black (relative risk [RR] 2.4, confidence interval [CI] 1.4–3.9), of low birth weight (RR 2.03, CI 1.04–4.0), and born of teenage mothers (RR 2.2, CI 1.3–3.7). [Relative risk values shown are from a multivariate analysis for all pregnancies (21).] Combined data from earlier (1964 to 1983) series indicated that the greatest threat of GBS infection was to the premature infant. The likelihood of an infected premature baby surviving was 34% (38/110), whereas the likelihood of survival for term infants was 83% (65/78). Later series revealed improved survival rates among both preterm and term infants with GBS sepsis. In two large series from 1992, overall survival was reported at 86% and 94%, but was still lower (72%) among preterm infants (20,22). Because bacteremia is present at birth in a large proportion of neonates with early-onset GBS disease, most cases (67% to 88%) of GBS early-onset disease have an intrauterine pathogenesis (15,23). The epidemiology and pathogenesis of late-onset GBS infection (occurring more than 7 days after birth) are not as clear as for early-onset disease (3). Serotype III strains account for most late-onset GBS infection. The GBS organisms responsible for late-onset disease are acquired either by maternal-neonatal vertical transmission or from nosocomial sources. In a CDC survey of 306 cases of GBS sepsis among infants younger than 90 days of age, only 19% were late in onset, and case fatality rates were similar for early- and late-onset disease (5.7% and 6.0%, respectively) (20).
CLINICAL MANIFESTATIONS Neonatal Infection With Group B Streptococcus Two clinically distinct neonatal GBS infections have been identified. Early-onset GBS disease appears within the first week of life, usually within the initial 48 hours. In 60% of infants, early-onset GBS sepsis presents within 24 hours of birth. Early-onset GBS infection often is characterized by rapid clinical deterioration. It occurs most often in infants born to mothers with risk factors for GBS neonatal infection. The three major presentations of early-onset GBS infection are septicemia (bacteremia and clinical signs of sepsis), pneumonia, and meningitis. Respiratory symptoms and signs, such as grunting, tachypnea, apnea, and/or cyanosis, usually are the earliest clinical findings (3). Hypotension occurs in 25%. Additional symptoms and signs similar to those associated with any bacterial infection (lethargy, poor feeding, hypothermia or
fever, pallor, and jaundice) are present. Meningitis is present in approximately 30% of early-onset GBS infections. Pneumonia is present in 40% of infants with early-onset GBS infection, and almost all of these infants present with grunting, tachypnea, and apnea. Late-onset GBS disease occurs more insidiously. It usually occurs after the first week of life and up until 12 weeks of age. Most of these infants (85%) have meningitis as the prominent clinical manifestation. The presenting symptoms in newborns with meningitis include fever (nearly 100%), irritability and/or lethargy, and poor feeding. Antecedent upper respiratory tract symptoms were present in 20% to 30% of meningitis cases. Late-onset disease may result in localized infections involving the middle ears, sinuses, conjunctiva, breasts, lungs, bones, joints, and skin. Meningitis is related to the serotype of GBS. More than 80% of early-onset GBS infections with meningitis present are due to type II organisms. More than 90% of late-onset disease (in which meningitis usually is present) are due to type III GBS. These findings are in contrast to the distribution of GBS serotypes among asymptomatic maternal carriers, where one third have type Ia, Ib, or Ic organisms; one third have type II; and one third have type III. Whereas early-onset disease has been associated with transmission from the mother's genital tract either before labor or during parturition, such a route of transmission is thought to occur less frequently in late-onset disease. Nosocomial transmission of GBS can occur in the nursery from colonized nursing staff or by cross-colonization from other infants. Maternal Infection With Group B Streptococci Group B streptococcus now is recognized as an important cause of maternal infections as well as of neonatal sepsis. Among cases of GBS puerperal infection, bacteremia occurred in 31% to 35% (usual rate in obstetrics is 10%). Characteristically, women with GBS puerperal infection develop high spiking fever within 12 hours of delivery. Other clinical features include tachycardia, chills, and tender uterine fundus and parametrium. However, at the time of the first fever spike, few localizing signs or symptoms may be present (see Chapter 20). Group B streptococci also cause maternal urinary tract infection (both bacteriuria and symptomatic disease), amnionitis, and wound infection (see Chapter 18 and Chapter 21).
DIAGNOSIS OF GROUP B STREPTOCOCCAL INFECTION The “gold standard” for diagnosing maternal genitourinary colonization with GBS is the culture, when it is performed using selective medium. The optimal medium for cultivating GBS is a selective broth medium, such as Lim or selective broth medium (SBM) broth, which contains gentamicin and nalidixic acid. These antibiotics inhibit the growth of Gram-negative Enterobacteriaceae and other bacteria in the normal genital tract flora
that could interfere with the recovery and identification of GBS. Because the great majority of colonized neonates (~9%) are asymptomatic, detection of colonization requires culture identification of GBS. None of the clinical manifestations of neonatal disease is sufficient to confirm GBS disease in the absence of a positive culture of the blood or cerebrospinal fluid. A presumptive or “clinical” diagnosis sometimes is made when there is a suggestive clinical picture accompanied by positive cultures from peripheral sites such as the umbilicus, oropharynx, and rectum. Within the past few years, there has been great interest in tests for rapid identification of GBS from the maternal genital tract. These tests include Gram stain, immunofluorescent antibody test, colorimetric assay using starch serum media, and antigen detection by a variety of methods, such as coagglutination, latex agglutination, enzyme immunoassay, and DNA probes. These tests were reviewed in an excellent work by Yancey and colleagues (24) (Table 3.5) The Gram stain has many limitations, mainly low sensitivity and poor positive predictive value (25).
TABLE 3.5. RAPID IDENTIFICATION TESTS FOR GROUP B STREPTOCOCCUS
Methods using starch serum media have better sensitivity and predictive values (in the 90% range when performed by laboratory technicians), but the requirement for an incubation period of 12 hours and difficulties in interpretation by labor and delivery staff make this approach impractical (24). Numerous reports have assessed the rapid antigen detection tests. Overall, the sensitivity of these tests for detection of all colonized women varies from as low as 4% to as high as 88%. In 17 studies evaluated, the sensitivity was 60% or less in 13 studies. Positive predictive values varied from 15% to 100%. These tests perform better with heavy colonization, and the inability of current methods to detect women with light colonization is a serious drawback. A promising rapid technique currently available is the DNA probe. Results from a
preliminary report have been encouraging, with sensitivity of 71%, specificity of 90%, positive predictive value of 61%, and negative predictive value (NPV) of 94% after an incubation of 3.5 hours (26). No currently available rapid diagnostic test has sufficient sensitivity and positive predictive value to be cost effective and to replace the culture using selective media as the screening technique of choice (27).
TREATMENT OF GROUP B STREPTOCOCCAL INFECTION Penicillin remains the drug of choice for symptomatic GBS infection in mothers or neonates. However, in most instances, treatment must begin empirically before culture results are available. Maternal infections such as amnionitis or postpartum endomyometritis usually are polymicrobic; thus, treatment requires a broad-spectrum approach, as discussed in Chapter 6. In these instances, a broader-spectrum approach for empirically treating the mother is required. Ampicillin is a frequently used and very effective agent in such situations and provides adequate treatment for GBS. The new semisynthetic penicillins (piperacillin, mezlocillin, ticarcillin) and first- and second-generation cephalosporins have very good in vitro activity. Erythromycin and clindamycin have been thought to provide very good coverage, but resistance is now a concern, as will be discussed later. Although GBS are resistant to aminoglycosides, addition of gentamicin or tobramycin to one of the penicillins results in a synergistic action against GBS. Because ampicillin is capable of crossing into the cerebrospinal fluid, the most common pediatric recommendation for GBS neonatal infection is an ampicillin-aminoglycoside combination. Among the multitude of new third-generation cephalosporins, cefotaxime has activity against GBS comparable to that of penicillin G. Cefoperazone and ceftazidime have somewhat less activity.
PREVENTION OF GROUP B STREPTOCOCCAL INFECTION As a result of the severity of GBS neonatal infection and the recognition that the major method of pathogenesis is vertical transmission, major efforts have addressed prophylactic strategies. The various approaches are listed in Table 3.6.
TABLE 3.6. STRATEGIES FOR PREVENTION OF GROUP B STREPTOCOCCUS
INFECTION
Antepartum Antibiotics Attempts at reducing maternal carrier rates by antepartum prophylaxis generally have been unsuccessful (28). Hall and associates (28) noted that administration of ampicillin to gravid women with cervical colonization of GBS resulted in a significantly decreased colonization rate within 3 weeks of therapy, but the women treated often were recolonized by the time of parturition. Infants of the treated mothers were colonized at the same rate as the control infants. To circumvent the problem of a high percentage of reacquisition of GBS after attempted prophylaxis in the early third trimester, Merenstein and coworkers (29) evaluated the efficacy of an oral penicillin regimen at 38 weeks' gestation. They noted a significant reduction in maternal and infant colonization with GBS in the treatment group (mothers and sexual partners treated). However, this approach misses the group at greatest risk, the preterm pregnancies, in which neonatal mortality with GBS infection is much greater. The attempt at antepartum prophylaxis against GBS is hindered by several factors. Mainly, it is difficult to eradicate GBS from the rectum because of the b-lactamase enzymes (which inactivate penicillin and ampicillin) produced by the Enterobacteriaceae in this locale. Accordingly, using antepartum antibiotics in pregnant women to prevent neonatal sepsis generally has been abandoned. Furthermore, use of oral antibiotics during prenatal care in women colonized with GBS (for the purpose of decreasing preterm birth) has not been shown to be effective (30). (Treatment of GBS bacteriuria, of course, is appropriate.) Intrapartum Antibiotics Use of intrapartum antibiotics has been established as an effective method to decrease neonatal colonization and early-onset disease (Table 3.7 and Table 3.8). Yow and colleagues (31) showed that use of intravenous ampicillin (500 mg/dose) during the intrapartum period in GBS-positive mothers prevented neonatal GBS colonization and disease. Boyer et al. (16) examined the effect of intrapartum ampicillin treatment on vertical transmission of GBS. Ampicillin virtually eliminated vertical transmission of GBS in the treatment group without perinatal risk factors and in the treatment groups with premature labor and/or prolonged ruptured membranes. Group B streptococcus colonization occurred in the neonates born to women who had intrapartum fever or less than 1 hour of ampicillin treatment before delivery. Combining a single antepartum screening culture for GBS with intrapartum intravenous ampicillin treatment at 26 to 28 weeks for mothers with suspected amnionitis, preterm labor, and/or PROM who are GBS carriers resulted in a significant reduction in the vertical transmission of GBS from
mother to infant.
TABLE 3.7. NONRANDOMIZED CLINICAL TRIALS OF INTRAPARTUM PROPHYLAXIS AND NEONATAL OUTCOMES—SEPSIS
TABLE 3.8. RANDOMIZED CLINICAL TRIALS OF INTRAPARTUM PROPHYLAXIS AND NEONATAL OUTCOME—SEPSIS OR PNEUMONIA
In a subsequent study, Boyer and Gotoff (32) conducted a randomized, nonblinded, controlled trial of the effect of selective use of intrapartum ampicillin in women with these risk factors. Patients received either ampicillin 2 g intravenously initially, then 1 g intravenously every 4 hours until delivery or no ampicillin. For infants of mothers randomized to intrapartum ampicillin, therapy was continued with four doses of ampicillin (50 mg/kg/dose) intramuscularly every 12 hours until initial neonatal cultures were available. Infants of mothers randomized to no treatment were given antibiotics only if they became symptomatic. The ampicillin-treated group had significantly less neonatal colonization and significantly reduced neonatal bacteremia. In our experience
at the University of Colorado, we found that using the latter strategy was complex (33). Between 1986 and 1994, numerous other intrapartum approaches have been suggested and/or evaluated. In 1994, Rouse and colleagues (27) reviewed 19 different intrapartum strategies in a decision analysis. Of these, a strategy of universal prophylaxis, consisting of intrapartum administration of antibiotics to all women in labor, was believed to be a cost-effective strategy, based on decision analysis. In addition, a strategy of giving prophylaxis to all high-risk women in labor, following the recommendation of Minkoff and Mead (34) and of the ACOG (5), was believed to be a cost-effective strategy. Universal screening based on cultures at the end of the second trimester had considerable limitations, especially because of the modest predictive value of the 26-week culture for maternal culture status of women in labor. Strategies based on screening women in labor using a rapid test were judged to be not effective. To avoid systemic side effects in the mother, Burman and coworkers (41) proposed and tested a novel technique. They reported a randomized, blinded trial of chlorhexidine douches given to women in labor to prevent neonatal GBS sepsis. Compared with placebo, chlorhexidine douches reduced the admission rate of infants to the special care nursery from 5.4% to 2.8% for GBS-colonized mothers (p = NS) and from 2.9% to 2.0% overall (p = 0.04). However, this endpoint is soft, and there was no decrease in the culture-confirmed sepsis rate (2/1,000 in both groups) (41). Because many infants with GBS sepsis already are bacteremic at birth, vaginal douching may be too little prophylaxis, too late. In response to the need for national guidelines, the CDC developed prevention strategies through critical analysis of clinical trial data and a subsequent review by a multidisciplinary group of consultants. These guidelines were published in spring 1996 (4). Supporting documents then were issued by the ACOG (5) and the American Academy of Pediatrics (42). These guidelines recommended the following: Use of either a strategy based on late prenatal culture (at 35 to 37 weeks) as the primary risk determinant or a strategy based solely on clinical risk factors When the culture-based strategy is used, the offer of intrapartum antibiotic prophylaxis to all women who have a positive culture whether or not there are intrapartum risk factors Use of penicillin G as an alternative to ampicillin for intrapartum prophylaxis In the approach based on screening cultures at 35 to 37 weeks, prophylaxis is recommended if any of the following conditions exists: a previously infected infant with invasive disease, GBS bacteriuria in this pregnancy, or delivery at less than 37 weeks. Patients are screened at 35 to 37 weeks, and all patients are offered intrapartum prophylaxis. A proviso of this strategy is that if a culture result is unknown, prophylaxis is given if the patient has a temperature greater than 38°C (100.4°F), the patient is in labor, or ROM duration is 18 hours or more. Special conditions required for the screening-based approach include the use of an optimal culture technique. This involves use of swabs from both the distal vagina and the rectum and use of selective broth media in the culture process. It is important to note that treatment of genital antenatal GBS colonization is not recommended, whereas treatment of GBS bacteriuria is recommended when diagnosed. Figure 3.1 shows an algorithm summarizing the
approach based on prenatal screening cultures at 35 to 37 weeks. This algorithm is complex, and it is recommended that hospitals using this algorithm have copies of this figure readily available in the labor and delivery unit.
FIGURE 3.1. Prevention strategy using prenatal screening at 35 to 37 weeks' gestation.
Under the approach using risk factors, prophylaxis is given with one or more of the following risks: Previous infant with invasive GBS disease Group B streptococcus bacteriuria in this pregnancy Delivery at less than 37 weeks Duration of ROM 18 hours or more Temperature of 38°C (100.4°F) The algorithm for this approach is shown in Fig. 3.2. Even though this approach is a simpler one, it is recommended that copies of this algorithm be available in hospitals using it so that compliance can be optimized. It is estimated that the approach based on screening at 35 to 37 weeks prevents a greater percentage of cases of early-onset neonatal sepsis due to GBS than does the approach based on risk factors. However, this estimate presumes there is excellent compliance with both approaches and that proper culture technique is followed with the screening-based approach. It is acknowledged that a larger percentage of gravid women will be exposed to intrapartum antibiotics with the screening-based approach. Another concern of the screening-based approach is that recognition of a positive culture at 35 to 37 weeks will lead to antibiotic treatment during the antepartum period in an attempt to eradicate the positive culture. Comparison of these two regimens is shown in Table 3.9. Box 1 shows the recommended regimens for intrapartum antimicrobial prophylaxis for perinatal GBS disease.
FIGURE 3.2. Prevention strategy using risk factors.
TABLE 3.9. COMPARISON OF TWO CENTERS FOR DISEASE CONTROL AND PREVENTION/AMERICAN COLLEGE OF OBSTETRICIANS AND GYNECOLOGIST's strategies
Box 1 Recommended Regimens for Intrapartum Antimicrobial Prophylaxis for Perinatal Group B Streptococcal Disease Recommended Penicillin G, 5 mU i.v. load, then 2.5 mU i.v. every 4 hr until delivery Alternative Ampicillin, 2 g i.v. load, then 1 g i.v. every 4 hr until delivery If penicillin-allergic: Recommended Clindamycin, 900 mg i.v. every 8 hr Alternative Erythromycin, 500 mg i.v. every 6 hr until delivery
These guidelines have been widely distributed and fairly widely adopted. A later section in this chapter discusses the aftermath of the 1996 guidelines, including areas of potential noncompliance, persisting clinical questions, and policy questions. Note: If a patient is receiving treatment for amnionitis with an antimicrobial agent active against GBS (e.g., ampicillin, penicillin, clindamycin, or erythromycin), additional prophylactic antibiotics are not needed (4). Neonatal Antibiotics Another strategy is to give antibiotic prophylaxis to the neonate. Evidence to support this approach initially arose from large retrospective and prospective descriptive series (18). However, a major drawback to this approach is the recognition that the overwhelming majority of neonates with early-onset GBS sepsis are bacteremic at birth or within 1 hour of delivery (15). Further, in a well-designed trial of low birthweight infants, infants were randomized to receive either 50,000 U of penicillin G within 90 minutes of birth and every 12 hours for 72 hours or only routine newborn care. In the treatment group of 589 infants, ten had GBS bacteremia, and six of these infants died (23). There were 14 cases of GBS bacteremia, with eight deaths among the 598 control patients. Most importantly, bacteremia was present in 90% of treated infants and 86% of controls. This study clearly demonstrated that penicillin prophylaxis in the neonate is ineffective at preventing bacteremia or reducing the mortality rate. Maternal Vaccination The ultimate goal of a maternal vaccination strategy for prevention of perinatal GBS disease is to induce protective immunity, both systemically to prevent invasive disease and mucosally to prevent or limit maternal lower genital tract colonization. This strategy
has a great appeal from a theoretical perspective. A vaccine would eliminate the problems of screening and intrapartum prophylaxis. It is estimated that maternal vaccination would reduce the number of neonatal GBS sepsis cases by 90% (43). Considerable evidence demonstrates that a maternal vaccination strategy is feasible. Immunization of pregnant women at 31 weeks' gestation with a GBS polysaccharide vaccine resulted in successful transfer of specific functional antibody to the newborn, even though this vaccine preparation was suboptimally immunogenic (44). Even though this strategy hypothetically would prevent most cases of GBS sepsis, as noted earlier, it may not adequately protect prematurely born infants because of inefficient transport of IgG across the placenta (45). An alternative timing of the vaccination may be given to women of childbearing age to avoid hypothetical concerns about vaccination in pregnancy. An additional advantage of a vaccination strategy is that it may reduce maternal GBS infection and potentially GBS-associated premature birth. It is most likely that the first generation of GBS vaccines to be used in humans will consist of GBS polysaccharides coupled with a carrier such as tetanus toxoid. The seminal work of Lancefield et al. (46) described the capsular polysaccharide (CPS) antigen of GBS as an important target of protective immunity. It has been shown that the lack of circulating CPS-specific IgG correlated with GBS disease among infants born of GBS-colonized women (19). These observations provide the rationale for a GBS vaccine based on CPS, which is surface expressed. Although these native CPS from GBS are safe in adults, they are poorly immunogenic (47). Much recent work has sought to improve the immunogenicity of GBS CPS while maintaining proper antigenicity. Group B streptococcal CPS has been coupled to an immunogenic protein such as tetanus toxoid to form a conjugate. All conjugate vaccines have been superior to the uncoupled vaccines in eliciting a high-titer protective IgG response specific to the serotype in both mice and rabbits (48,49 and 50). Group B streptococcus type III polysaccharide tetanus toxoid vaccine has been shown to be immunogenic in nonhuman primates when administered with an alum adjuvant (51). Accordingly, successful preclinical studies have led to individual phase I and phase II safety and immunogenicity clinical trials with GBS conjugate vaccines for serotypes Ia, Ib, II, III, and V (52,53 and 54), which are the serotypes of major importance in the United States and western Europe. In each trial, the conjugate was shown not only to be safe and well tolerated, but it also elicited higher levels of IgG than did the control uncoupled GBS vaccine. Clearly, a multivalent GBS conjugate vaccine is desirable and offers promise of providing protection simultaneously to both the mother and her neonate against the most prevalent disease causing GBS serotypes. At the same time, there is considerable interest in GBS proteins as the basis for vaccines. Advantages of protein-based vaccines include the ability to simplify and extend the coverage of the polysaccharides. For example, it is anticipated that a single vaccine containing the type III polysaccharide linked to one of the GBS proteins (such as a C) theoretically could cover the majority of clinically important GBS isolates. A protein-based vaccine also might have reduced toxicity, avoiding overuse of a limited number of existing carrier proteins such as tetanus toxoid. An especially attractive feature of protein-based vaccines is their versatility for use in alternative delivery systems. Unlike polysaccharides, proteins can be incorporated readily into viral, bacterial, or naked DNA vaccines (55). Such developments in vaccines may be of particular importance in eliciting vigorous mucosal immunity. At present, however, there are still major obstacles to vaccine availability. First, the composition of the vaccine for trials has not been established. Second, the
maternal antibody response must be IgG (so that the antibodies will cross the placenta), but IgG does not cross well before 32 weeks. Thus, the infants at highest risk for infection would be left with little protection. Third, the timing of vaccine administration (except for the type III polysaccharide) has not been described. Aftermath Of The 1996 Guidelines Several years after publication of these guidelines, numerous questions remain. In this section, we will consider areas of potential noncompliance, persisting clinical questions, and persisting policy questions. The CDC guidelines recommend obtaining cultures at 35 to 37 weeks' gestation. If a culture was obtained between 1 and 5 weeks before delivery, sensitivity was 87% and specificity was 97% (56). If a culture was obtained more than 5 weeks before delivery, there was a much greater chance that the results would not accurately predict colonization status at delivery. Compliance with the guidelines requires collection of the culture from the distal vagina and anorectum and use of proper laboratory technique. When clinicians do not follow these guidelines, they miss at least one fourth of patients who are culture positive for GBS and who may benefit from antibiotic prophylaxis. The guidelines specifically recommend the use of selective broth medium, which is an enriched medium that enhances the growth of GBS better than agar media and is supplemented with antibiotics to inhibit the growth of organisms other than GBS (57). When selective medium is used, there is a 50% higher rate of GBS detection (58,59). It is consistent with the guidelines to use a nonselective transport medium (such as Amies medium) to transport rectovaginal swabs to the laboratory. It is estimated that in 1998 there was considerable noncompliance in both the site of culture and the use of media. The guidelines recommend penicillin G as the first choice, with ampicillin as the second choice. In penicillin-allergic patients, the recommendation is to use clindamycin or erythromycin parenterally. Group B streptococci are universally sensitive to the penicillins. Because of its narrower spectrum, aqueous penicillin G is the first line drug of choice. As the interval between the first dose of penicillin and birth increases, the proportion of GBS-positive infants delivered from GBS-colonized mothers decreases (Fig. 3.3). When the interval is greater than 2 to 4 hours, few infants are colonized with GBS. It is estimated that there is noncompliance with guidelines in 10% to 20% of cases, even when a policy is in place at an institution. This may be unavoidable, because patients may refuse antibiotics or deliver before antibiotics can be administered (60). In other cases, patients may have barely met the criteria (by having just a few minutes more than 18 hours of membrane rupture or being within a few days of 37 weeks' gestation). These protocols are complex. In one university hospital, the overall noncompliance was 19.7%, but nearly half of these protocol deviations were due to factors beyond the control of the physician or due to marginal situations (60). Other reports indicate similar rates of noncompliance (61,62).
FIGURE 3.3. Timing of intrapartum antibiotic and vertical group B streptococcus transmission. Adapted from ref. 71.
To optimize compliance with protocols, hospitals are encouraged to be proactive, for example, by placing special labels on the chart of GBS-positive patients or instituting a reminder to practitioners on the labor and delivery board as to when a patient becomes at risk. Is Chemoprophylaxis Necessary For Group B Streptococcus in Elective Cesarean Section? The current guidelines state that “intrapartum” antibiotics should be given; thus, the guidelines provide no specific recommendations for women admitted for elective cesarean section. The risk of sepsis to the newborn in this situation is low and recently observed to be 0% in an estimated 530 infants at risk (95% CI, 0.0%–0.7%) (63). The number of additional antibiotic exposures is estimated to be about 1% of the total population (assuming that 20% of the population is colonized and the total rate of cesarean section for patients with no labor and no ruptured membranes is approximately 5% of the total obstetric population [i.e., approximately 25% of the total cesarean sections]). Although expert opinion on this issue is divided, a position paper from the Infectious Disease Society for Obstetrics and Gynecology (IDSOG) states that giving GBS antibiotic prophylaxis in this circumstance is not warranted (64). What Prophylaxis Regimen Should Be Used In Preterm Premature Rupture Of The Membranes Without Labor? The 1996 guidelines recommend that a GBS culture be collected and then (a) antibiotics given until the culture returns with a negative result or (b) antibiotics given once a positive culture result is available. Several clinical scenarios are not addressed. If the initial culture for GBS is negative, should the culture be repeated? Because the likelihood of a negative antenatal culture becoming positive in the 5 weeks
after it was obtained is 5% (56), the IDSOG's position is that it is not necessary to repeat a negative culture for up to 5 weeks in this setting. If the initial culture is positive, how long should antibiotics be given? Because this is “antibiotic prophylaxis,” we recommend giving antibiotics intravenously for 48 hours and then obtaining another culture. If the culture remains positive, antibiotics should be given for an additional 5 to 7 days. If the culture at 48 hours is negative, the antibiotic should be discontinued. Patients with preterm PROM should receive intrapartum antibiotics according to the guidelines. There is a strong rationale to use broad-spectrum antibiotics to prolong latency with premature PROM from 24 to 32 weeks' gestation (see Chapter 17). What Prophylaxis Regimen Should Be Used For Preterm Labor That Has Been Arrested With Tocolytics? The 1996 guidelines recommend that the patient in preterm labor should receive intrapartum antibiotics for GBS prevention, but these guidelines are incomplete. Unless preterm delivery is imminent, we believe that a culture for GBS should be obtained and the mother managed clinically as she would be for preterm PROM. If labor ensues within 5 weeks, the original culture can be relied on; if labor occurs after 5 weeks, a repeat culture should be obtained. What Is The Threshold Colony Count For Treating Group B Streptococcal Bacteriuria? Infants born to mothers who are heavily colonized with GBS are more likely to become colonized than are infants whose mothers are lightly colonized (39,65,66). Maternal GBS bacteriuria is a marker for heavy maternal genital colonization. Babies born to mothers with GBS bacteriuria during pregnancy are more frequently and more heavily colonized with GBS (21). In addition, these infants are at increased risk for invasive GBS disease (67,68 and 69). Moller et al. (70) found GBS bacteriuria in 2.5% of pregnant women screened between 12 and 38 weeks' gestation. There were five cases (7.35%) of confirmed GBS sepsis among 68 infants born to women with GBS bacteriuria compared to none of 2,677 women without GBS bacteriuria (p 10 5 colony-forming units (CFU) per milliliter of urine. If the patient is asymptomatic, treat prenatally if the GBS colony count in the urine is >102 CFU per milliliter. If the patient is asymptomatic and the colony count is >102 but 10 pounds, diarrhea ³4 weeks) in 8 (7%). The remaining 26 symptomatic women had syndromes suggestive of HIV infection, including thrush, tuberculosis, hairy leukoplakia, herpes zoster, PCP, AIDS encephalopathy, and CMV retinitis. Esophageal candidiasis was the most common AIDS-defining manifestation in 34%; PCP was the AIDS-defining illness in only 20% of HIV-infected women (424). Creagh et al. (425) reported that, in French women, esophageal candidiasis was a more common AIDS-defining manifestation than PCP. Similarly, Sha and coworkers (418) documented that esophageal candidiasis (42%) was the most common AIDS-defining illness in HIV-infected women, followed by PCP (35%). Over time, esophageal candidiasis accounted for 54% of all AIDS diagnosis in this group of women (418). These findings are in distinct contradiction to national data in which PCP was the most common AIDS-defining illness in men. However, other investigations have failed to confirm this finding and have not noted any gender differences in the frequency of esophageal candidiasis or other AIDS-defining opportunistic infections (426,427,428 and 429). Other opportunistic infections commonly seen in HIV-infected women include CMV, recurrent mucocutaneous HSV, mucocutaneous candidiasis, toxoplasmosis, and tuberculosis (417). Early in the AIDS epidemic, the increased occurrence of certain malignancies associated with HIV infection and immunosuppression was apparent (296,298). It is estimated that 40% or more of HIV-infected persons have cancer, either as a cause of mortality or morbidity (430,431). The malignancies most commonly associated with this morbidity/mortality in HIV-infected women include (i) non-Hodgkin lymphoma (NHL); (ii) Epstein-Barr-virus–associated tumors; (iii) Kaposi sarcoma due to human herpesvirus 8; and (iv) cervical intraepithelial neoplasia (CIN) due to human papillomaviruses (HPVs) (296,298,430,431). Increases in risk for Kaposi sarcoma and NHL in HIV-infected women have been reported (432). The AIDS-related risk for NHL is approximately the same across all HIV risk groups, and no gender difference has been noted (298,433). In contrast, although Kaposi sarcoma is present frequently in MSM, it is found in less than 2% of women as an AIDS-defining diagnosis (296). However, studies from Africa, the Caribbean, and Europe demonstrated that Kaposi sarcoma, as a first AIDS-defining illness, is present as frequently in men and women IDUs and heterosexuals (434,435 and 436). There is not a lower incidence or severity of Kaposi sarcoma in pregnant women (437).
Human papillomaviruses, CIN, and associated cancers appear to behave differently in HIV-infected women, especially as cell-mediated immunity is impaired. Human papillomaviruses and cervical neoplasia are discussed in the “Gynecologic Aspects of HIV Infection” section. Transmission of HIV depends on both the infectiousness of the index case (438) and the susceptibility of the exposed host (327,439). The virologic factors required for transmission of HIV include (i) certain viral quasi-species appear to be favored (207); (ii) different viral subtypes (clades) found in various geographic areas may be transmitted with variable efficiency (440); and (iii) the viral envelope has an important role in determination of transmissibility (441). As reviewed by Cohen and coworkers (442), susceptibility of individuals to HIV infection depends on both lack of immunity in the exposed host and lack of genetic resistance. As discussed previously, persons with mutations in chemokine cell receptors demonstrate resistance to HIV infection (95,443). Cofactors that amplify HIV transmission secondary to increasing the infectiousness of index cases or susceptibility of the exposed host have been described (443). Cohen et al. described (442) cofactors that cause inflammation of genital tract mucosa as the most important for transmission of HIV. Thus, untreated STDs may increase both the infectiousness and susceptibility to HIV (372,444).
HIV Infection In Pregnancy As discussed previously, an increasing percentage of AIDS patients and HIV-infected patients in the United States are women. More than 85% of these women are of reproductive age. Nearly 90% of pediatric AIDS cases are presumed to have acquired HIV infection perinatally from their mothers. Nearly 100% of new pediatric AIDS cases are acquired perinatally. This increasing infection rate in women and the increasing risk for perinatal transmission demonstrate that we will continue to be faced with the challenge of providing health care to pregnant women infected with HIV. As documented by the CDC, the number of reported cases of perinatally acquired AIDS rose rapidly, in parallel with increases in the number of HIV-infected women, in the late 1980s and early 1990s, reaching a peak in 1992. Subsequently, a decline in perinatal AIDS cases has occurred (Fig. 10.10) (445). The decline in perinatally acquired AIDS in the United States has occurred despite a situation in which women have increasingly shouldered the burden of HIV disease. Women accounted for 18% of cumulative AIDS cases in the United States as of 1999 and accounted for 23% of new AIDS diagnoses and 32% of newly reported HIV diagnoses in the United States (446). An estimated 120,000 to 160,000 HIV-infected women are living in the United States (447). From 1989 to 1994, an estimated 6,000 to 7,000 infants were born to HIV-infected women each year. By 1995, more than 16,000 children infected with HIV via the perinatal route had been born (448,449). Pediatric AIDS cases are concentrated along the eastern seaboard of the United States, especially in the New York City metropolitan area, with New York, New Jersey, and Florida accounting for approximately 50% of cases (450). The District of Columbia (6.3/10,000 births), New Jersey (4.3/10,000
births), Connecticut (3.8/10,000 births), and Maryland (3/10,000 births) have the highest incidence rates (450). Lindegren et al. (451) reviewed the trends in perinatal transmission of HIV/AIDS in the United States. These authors reported an 80% decline in the number of perinatal AIDS cases from 1992 to 1997 and a 69% decrease in the rates of AIDS among infants (per 100,000 births) from 8.9 in 1992 to 2.8 in 1996, but only a 17% decline in births to HIV-infected women from 1992 (n = 6,990) to 1995 (n = 5,797) (451). In 1994, the Pediatric AIDS Clinical Trials Group Protocol 076 (PACTG 076) demonstrated that ZDV, when administered to selected HIV-infected (mild) pregnant women and their newborns, reduced perinatal transmission by two thirds (452,453). This dramatic reduction in transmission resulted in the USPHS recommendation in August 1994 for the use of ZDV to reduce perinatal transmission of HIV (55) and in July 1995 for routine HIV counseling and voluntary prenatal testing of pregnant women for HIV infection (56). The percentage of perinatally exposed children born from 1993 to 1997 whose mothers were tested for HIV before delivery increased from 70% to 94%, and the percentage receiving ZDV increased from 7% to 91% (451). The dramatic decline in AIDS incidence among infants was associated with an increased use of ZDV to prevent perinatal transmission of HIV and reflected the tremendous success with which the PHS guidelines were implemented by health care providers for pregnant women (451). Worldwide, a different picture exists. UNAIDS estimates that 1.2 million children were living with HIV-1 infection at the end of 1999, and an additional 3.6 million children had died (454). An estimated 2.4 million HIV-infected women give birth annually, and 1,600 infants acquire HIV-1 infection every day (455). As noted by Mofenson and McIntyre (455), two perinatal epidemics exist. In more developed countries, integration of prenatal HIV counseling and testing programs into an existing antenatal infrastructure, availability of effective antiretroviral prophylaxis, and access to infant formula have resulted in new perinatal HIV infections becoming rare. In contradistinction, in less developed countries where antenatal care is limited, testing programs are almost nonexistent, effective interventions are not implemented, and prevention of transmission via breast-feeding while maintaining adequate infant nutrition is a major dilemma, vertical transmission of HIV from mother to infant remains a crisis. Initial attempts to assess the seroprevalence of HIV infection in pregnant women yielded a wide range of seropositivity. In 1987, the CDC reported preliminary results from surveys and studies of HIV antibody prevalence in selected groups of pregnant women (456). They reported that HIV seroprevalence ranged from 0.0% to 1.7% in prenatal patients, 0.4% to 2.3% at the time of delivery, and 9.4% to 29.6% in pregnant IDUs. Several large-scale statewide screening programs using newborn heel-stick samples were performed in an attempt to determine the seroprevalence of HIV in the general population (457,458). Anonymous testing of newborn specimens in Massachusetts revealed an overall seropositivity of 0.21%. In New York, the overall HIV seroprevalence was 0.66%, ranging from 0.16% in upstate New York to 1.25% in New York City. To monitor the prevalence of HIV infection in women delivering infants in the United States, an anonymous national HIV serosurvey (Survey of Childbearing Women) was conducted from 1988 to 1994 using residual dried-blood spot specimens collected
routinely from newborns for metabolic screening (136). A weighted, nationwide estimate of HIV infection prevalence was 1.6 per 1,000 childbearing women (136). The estimates of HIV prevalence in childbearing women in the United States remained stable from 1988 to 1994 (136). HIV prevalence in childbearing women varies by geographic area, with the highest prevalences in 1994 reported (per 1,000 women) in the District of Columbia (6.9), New York (5.2), Puerto Rico (4.7), and Florida (4.6) (136). Several areas of concern have emerged regarding HIV infection in pregnancy. These concerns include the (I) effect of pregnancy on HIV infection; (ii) effect of HIV infection on pregnancy outcome; (iii) risk of vertical transmission from infected mother to her offspring; (iv) management of HIV-infected pregnant women; (v) prevention of pediatric AIDS; and (vi) early diagnosis of newborn HIV infection. Effect of Pregnancy on HIV Infection There has been concern that pregnancy may accelerate the progression of HIV infection from the asymptomatic stage to symptomatic HIV infection and AIDS (459). Three lines of evidence have been used to suggest that such an occurrence is biologically plausible. Pregnancy has been associated with decreased levels of T4 cells (460) and decreased lymphocytic responses (461,462,463,464 and 465). During pregnancy, the immune response to viruses, such as CMV and rubella, appears to be depressed (466,467). Finally, certain viral diseases (e.g., influenza, varicella, poliomyelitis, and coxsackievirus) have been associated with increased morbidity and mortality during pregnancy (468). To date, whether pregnancy enhances HIV disease progression remains unclear. Initial reports by Scott et al. (152) and Minkoff et al. (153) suggested that such enhanced progression does occur. These reports noted that 45% to 75% of asymptomatic pregnant women followed for 28 to 30 months after delivery developed AIDS-related complex or AIDS. These rates are higher than the rates of progression reported for asymptomatic HIV antibody-positive homosexual men, IDUs, or hemophiliacs, in whom only 13% to 34% of those followed for up to 6 years developed AIDS (469,470). However, these studies were not controlled and evaluated only pregnant women known to have transmitted HIV to their offspring (152,153). Thus, they may represent a group with a longer duration of HIV infection and/or altered immune function whose progression to symptomatic disease was coincidental to, but not the result of, pregnancy. More recent controlled studies involving pregnant and nonpregnant asymptomatic HIV-infected women failed to demonstrate that pregnancy accelerates the progressive course of HIV infection (471,472,473,474,475,476,477 and 478). Of some concern was the report by Biggar et al. (479), who assessed immunologic changes over the course of pregnancy and postpartum in 37 HIV-infected and 63 HIV-negative women. Among HIV-negative women, levels of CD4+ counts fell during pregnancy to reach a nadir 8 weeks before delivery, but rose again rapidly before delivery and remained normal postpartum. Among HIV-infected women, a decrease in the number of CD4 + cells was seen through pregnancy, and no postpartum recovery occurred. The CD8+ cell levels were consistently higher in HIV-infected compared to HIV-negative pregnant women. In addition, Biggar et al. (479) noted that the rate of loss of CD4 + cells was 2% per month in pregnant HIV-positive women compared to average rates of 0.5% to 1.0% per month
in HIV-infected homosexual men or hemophiliacs. Burns and coworkers (480) assessed T-lymphocyte populations in 192 HIV-infected women compared with 148 non–HIV-infected controls who were followed for 2 years after delivery. They noted that the percentage of CD4 T cells began to increase during pregnancy and returned to normal postpartum. In the HIV-negative control group, the percentage of CD4 cells increased between the third trimester and 12 months postpartum. In contradistinction, in HIV-infected women, CD4 T-cell levels declined steadily during pregnancy and postpartum. However, the clinical significance of these laboratory findings is unclear. Moreover, in the Swiss HIV Cohort Studies, Wiesser et al. (481) noted that when taking into account preconception CD4 cell counts, acceleration of HIV disease progression is consistent in HIV-infected women who became pregnant. The consensus is that pregnancy does not have a major adverse effect on HIV progression (482,483). HIV Infection and Pregnancy Outcome Whether HIV infection is associated with an increased risk for adverse pregnancy outcome has been the focus of considerable attention. Initial but retrospective and uncontrolled investigations of infants with AIDS suggested that HIV-infected newborns were at high risk for preterm birth and/or low birthweight (484,485,486 and 487). However, prospective and controlled studies (especially controlled for IDU) in industrialized nations have demonstrated that pregnancy outcome does not appear to be adversely affected by HIV infection (160,161,476,488,489). In general, these studies found no differences in preterm delivery, low birthweight, or small-for-gestational-age rates between seropositive and seronegative IDUs. Selwyn et al. (476) reported no difference in birth outcomes in HIV-seropositive and seronegative intravenous drug users (Table 10.18). In addition, Selwyn and coworkers (476) reported no differences in the rates of spontaneous or elective abortion, ectopic pregnancy, premature rupture of the membranes (PROM), preeclampsia, anemia, weight gain less than 10 pounds, oligohydramnios, chorioamnionitis, or intrapartum fetal distress. However, seropositive women were significantly more likely to be hospitalized for bacterial pneumonia during pregnancy and had an unexplained increased tendency for breech presentation. Similarly, Johnstone et al. (488) in Scotland reported no significant difference in frequency of preterm birth, low birthweight, PROM, or low Apgar scores for asymptomatic HIV-infected and uninfected women when race, socioeconomic status, and intravenous drug use were controlled for. Minkoff et al. (490) also studied pregnancy outcome in asymptomatic HIV-infected women. In this prospective analysis, 91 HIV-infected and 126 HIV-negative pregnant women were controlled for drug, alcohol, and tobacco use. The HIV-infected women had a significantly higher rate of STDs (17.6% vs. 7.1%) and medical complications of pregnancy (43% vs. 25%), but serostatus was not related to any adverse outcomes, including peripartum fever, preterm labor, PROM, placenta previa, or abruptio placenta (490). In newborns, serostatus was not related to any immediate adverse outcomes, such as gestational age, birthweight, length, head circumference, or Apgar scores (490).
TABLE 10.18. BIRTH OUTCOMES IN HUMAN IMMUNODEFICIENCY VIRUS SEROPOSITIVE AND SERONEGATIVE INTRANVEOUS DRUG USERS
In contrast to the studies from the United States and Europe, prospective studies from Africa comparing pregnancy outcome in HIV-infected and noninfected women suggest that HIV infection adversely affects pregnancy outcome (165,166,491,492,493 and 494). Ryder et al. (166) studied 606 HIV-negative and 406 HIV-infected pregnant women from Kinshasa, Zaire. Acquired immunodeficiency syndrome was present in 18% of the HIV-infected women. These authors reported that rates of preterm birth, low birthweight, low head circumference-to-height ratio, and chorioamnionitis all were higher among infants of symptomatic seropositive mothers with AIDS (p < 0.01) than among infants of asymptomatic HIV-infected mothers or seronegative mothers (166). In a study of 177 HIV-infected and 326 uninfected women in Nairobi, Kenya, Braddick and colleagues (491) demonstrated that women with symptomatic HIV infection were more likely than asymptomatic HIV-infected women or uninfected women to deliver infants of low birthweight (17% vs. 6% vs. 3%). In this study, 17% of patients had advanced HIV disease, 28% had lymphadenopathy, and 55% were asymptomatic (491). In a study comparing 109 HIV-infected women with 40 uninfected women, Hira et al. (165) reported that HIV-infected women were 2.9 times more likely to deliver low birthweight infants. In this study, 50% of the HIV-infected women had symptomatic disease. In a study from Nairobi, Kenya, Temmerman and colleagues (492) controlled for the presence of STDs and reported that HIV-1 infection was significantly and independently associated with prematurity, low birthweight, and stillbirth. In an update of their study, Temmerman et al. (493) reported on 406 HIV-1 seropositive and 407 HIV-1 seronegative age-matched and parity-matched pregnant women. Maternal HIV-1 infection was associated with significantly lower birthweight (2,913 vs. 3,072 g; p = 0.0003) and with prematurity (21.2% vs. 9.4%; p < 0.0001), but not with small for gestational age (SGA) (4.2% vs. 3.2%; p = 0.7). The stillbirth rate was increased but was not statistically significant (3.8% vs. 1.9%) Women whose CD4+ cell count was less than 30% had a higher risk of preterm delivery (26.3% vs. 10.1%; p < 0.001). In addition, postpartum endometritis was more common in HIV-1 infected women than in seronegative controls (10.3% vs. 4.2%; p = 0.01) and was inversely
correlated with CD4 percentage (493). In a prospective cohort of 318 HIV-infected and 309 seronegative women from Rwanda, Bulterys et al. (494) noted that birthweight was significantly reduced in infants born to HIV-infected mothers. Crown-to-heel length, head circumference, and placental weight also were reduced. These authors reported that maternal HIV infection was significantly associated with intrauterine growth retardation but not preterm birth (494). The association of HIV infection and low birthweight and preterm birth demonstrated in these African studies probably is related to the advanced maternal HIV infection status noted in these studies. In the United States, pregnancy outcome is similarly adversely affected in symptomatic HIV-infected women (495). Minkoff et al. (495) noted an increased incidence of serious infections in 9 (45%) of 20 HIV-infected pregnant women with CD4 counts less than 300/mm3, whereas seronegative or HIV-infected women with CD4 counts greater than 300/mm3 had no serious infections. The most common serious infection was PCP, which occurred in six cases. It appears that the more advanced the maternal HIV disease, the greater the likelihood of adverse pregnancy outcomes. Also of concern is the consistent finding that infectious complications, including development of opportunistic infections, are increased in women with more advanced HIV disease (496,497). Two studies from more developed countries demonstrated that HIV infection was associated with reduced birthweight (498,499). In an analysis of the New York State Medicaid Program, Markson et al. (500) noted that HIV-infected women delivered low birthweight infants more than three times as frequently (29%) as the general sample of Medicaid-enrolled women. However, the authors did not have data related to maternal CD4 T lymphocyte counts and, thus, could not assess the role of HIV disease severity on their findings. Johnstone et al. (499) reported that, after multivariate analysis of their data, HIV had a statistically significant but modest effect on fetal growth, but not on preterm birth. This effect seemed to be associated with placental size, but the small reduction in birthweight seen was less than that attributable to smoking, and the authors questioned its clinical significance (499). Risk of Perinatal Transmission of HIV Vertical transmission of virus from HIV-infected mothers to their infants is a well-established phenomenon that has been a major concern (152,153,160,161,272,404,405,406,407,408,409,410,411,412,413,414 and 415,459,484,486,489,500,501,502 and 503). However, the timing of perinatal transmission, the rate of such transmission, and the determinants of vertical transmission of HIV required further elucidation. HIV transmission from infected mother to infant can occur antepartum (in utero), intrapartum (during labor and delivery), or postpartum (breast-feeding) (272). Transplacental infection early in pregnancy initially was suggested by description of the AIDS embryopathy syndrome with characteristic facial malformations by Marion et al. (406). However, studies controlling for drug abuse have failed to identify or confirm that such an embryopathy syndrome exists. Subsequent investigations have demonstrated that in utero transmission does occur, although it accounts for less than 25% to 30% of
perinatal transmission (272). Several different approaches have produced data supporting intrauterine transmission of HIV from mother to fetus. Culture of HIV and identification of HIV via molecular techniques have been reported (404,410,411 and 412,504,505 and 506). In addition, HIV has been identified in amniotic fluid and cells (507). Third, the detection of HIV in blood from newborns at birth has confirmed the presence of in utero transmission by culture (414,500,501,504,508,509 and 510), PCR (413,504,508), or p24 antigen with immune-complex dissociation (511). Infection intrapartum during labor and delivery secondary to ingestion or exposure to blood and other body fluids infected with HIV (i.e., amniotic fluid, cervical secretions) has gained favor as the major route for vertical transmission of HIV. The virus may enter via fetal skin, mucous membranes, or gastrointestinal tract. A number of studies have supported the concept of intrapartum transmission from mother to infant. The data suggest that, in industrialized countries, 70% to 75% of vertical transmission occurs intrapartum (272). Goedert et al. (512) reported the findings of the International Registry of HIV-exposed twins and demonstrated that the first-born twin had a greater risk of HIV infection. Among discordant twins, 25% of first-born twins were HIV infected compared to 8% of second-born twins (512). However, this study did not assess the status of membranes or differentiate elective from emergency cesarean section. On the other hand, in the Italian national study, concordance was present in 18 to 19 twin pairs (513). Ehrnst and colleagues (514) reported that, in a group of HIV-infected women delivering at term, 23 (85%) of 27 women had a positive virus culture (PBMC or plasma). Whereas none of the 27 infants had positive cultures in the newborn period, 5 (26%) of 19 infants had positive cultures by 6 months of age. This study strongly suggests a major role for intrapartum transmission. Rogers et al. (515) noted that detection of HIV-1 by culture, PCR, or p24 antigen methods occurs in 30% to 50% of newborn peripheral blood from infants ultimately shown to be HIV-1 infected. The failure to identify virus in the remaining 50% to 70% supports the concept of intrapartum transmission. Further support is the description by Blanche et al. (516) of a bimodal distribution for development of HIV-1 symptomatic infection in children that may differentiate children infected in utero from those who acquire HIV intrapartum. These authors suggest that the children who, by 12 months of age, have advanced HIV-1 infection with a 26% incidence of AIDS and a 17% mortality were infected in utero (515,516). Bryson et al. (409) have proposed a classification system for in utero versus intrapartum transmission of HIV. Early (in utero) infection requires detection of HIV-1 genome by PCR or isolation of HIV-1 from the blood within 48 hours of birth. Late (intrapartum) infection occurs when HIV isolation, PCR, or serum p24 antigen are negative during the first week of life but become positive from 7 to 90 days after birth in an infant who was not breast-fed. Dunn et al. (517) performed a meta-analysis of studies that evaluated the use of HIV-1 DNA PCR for early diagnosis of HIV infection. Within 24 hours of birth, HIV infection was detected in 38% (90% CI, 29%–46%), whereas by 28 days HIV was detected in 100% of 271 HIV-infected children. Thus, about 60% of transmission was intrapartum (517). Using serologic testing, DeRossi and coworkers (518) reported that the timing and pattern of antibody to HIV-1–specific proteins (develop at mean 54 days in 70% infected infants) also was consistent with intrapartum transmission and the time to seroconversion following acute infection with HIV in adults. Rouzioux et al. (519) reported similar estimates using mathematical modeling to estimate the timing of vertical transmission in a non–breast-feeding group. These authors estimated that 65% (95%
CI, 22%–92%) of perinatal transmission occurred intrapartum. Moreover, the model suggested that 92% of all transmission occurred during the last 2 months before or during the intrapartum period (519). More recently, Bertolli and coworkers (186), using virologic detection in a breast-feeding population, reported that the proportion of transmission estimated to occur in utero was 26% (95% CI, 14%–35%), intrapartum/early postpartum was 65% (95% CI, 53%–76%), and late postpartum via breast-feeding was 12% (95% CI, 5%–20%). Similarly, Kalish and coworkers (520) assessed 140 infected infants in the Women and Infants Transmission Study (WITS) and reported that HIV-1 culture was positive in 27% at 48 hours (in utero transmission), with transmission usually occurring during the intrapartum period. Clearly, the twin study with the first-born twin having twice the risk for acquiring HIV infection suggests that exposure to infectious secretions of the cervix and/or vagina plays a critical role in the intrapartum transmission of HIV from mother to infant (512). Subsequently, Duliege et al. (521) estimated that greater than 50% of the risk of transmission in first-born twins was the result of vaginal delivery; this increased to 76% when vaginally delivered first-born twins were compared to second-born twins delivered via cesarean section. Additional evidence that exposure to infected genital tract secretions is a substantial risk for perinatal transmission of HIV is provided by studies demonstrating an increased risk of vertical HIV transmission associated with increased duration of rupture of the membranes (ROM) prior to delivery (522,523). The third possible route for perinatal transmission of HIV is postpartum via breast-feeding. Human immunodeficiency virus has been recovered from the breast milk of infected women (179,524) and is found in the highest concentration in colostrum (525). Transmission of HIV via breast-feeding has been demonstrated (407,524,526). The viral load of HIV is significantly higher from a blood transfusion than with chronic HIV infection (180). Thus, Lederman (180) proposed that women who contracted HIV from a contaminated transfusion while breast-feeding represent a group at greater risk for transmission than mothers who have chronic HIV infection. Similarly, Van de Perre et al. (181) demonstrated that transmission via breast-feeding was associated with acute HIV infection that is characterized by high viral titers. Workers in Australia also documented an association between primary maternal HIV infection and a high risk of transmission via breast milk (182). Thus, the risk of transmission via breast-feeding appears to be very high when maternal primary infection occurs within the first few months after delivery (182). Studies from Europe have provided epidemiologic data (marginally statistically significant) supporting an association between breast-feeding and increased risk of HIV transmission (171,527). In Africa, the higher vertical transmission rates reported to some degree may be attributable to a high rate of breast-feeding (165,166). However, in a prospective study from Zaire, Ryder et al. (528) could not demonstrate a dose-response effect between breast-feeding and perinatal HIV transmission. Moreover, they confirmed a protective effect of breast-feeding against other common causes of childhood morbidity and mortality. Datta et al. (529) noted that the risk of transmission via breast-feeding was 32% if an infected mother breast-fed her infant for more than 15 months. Dunn and coworkers (524) performed a meta-analysis and reported that the attributable risk of transmission via breast-feeding was an estimated 29% (95% CI, 16%–42%) if the mother became infected postpartum compared to 14% (95% CI, 7%–22%) if the mother was infected before pregnancy. In industrialized countries, it is recommended that HIV-infected women not breast-fed (530). On the other hand, in developing countries, the benefits of breast-feeding
definitely outweigh the risk of HIV transmission via breast-feeding. In July 1998, the World Health Organization recommended that HIV-infected women in developing countries be informed regarding the benefits and risks of breast-feeding and be given an opportunity to make an informed choice about breast-feeding (531). To maximize prevention of perinatal AIDS, HIV-infected women must be identified as early as possible in pregnancy (272). Similarly, postdelivery evaluation of the infant at risk for HIV infection immediately after birth is crucial for early diagnosis and optimal medical treatment (272). Serologic methods have limited usefulness for the early diagnosis of perinatal HIV infection due to transplacental passage of maternal immunoglobulin G antibodies that may last up to 18 months (272). Detection of HIV proviral genome using PCR (DNA PCR) or HIV RNA using PCR is a highly sensitive, specific, rapid, and cost-effective screening test for vertical infection (532,533). Using DNA PCR, 25% to 30% of HIV-infected infants may be identified at birth, with the remaining 70% to 75% of HIV-infected infants diagnosed by 1 month of age (272). Mother-infant (vertical) transmission of HIV has been both a concern and a focus of intense research efforts. In the United States, vertical transmission of HIV accounts for virtually all new HIV infections in children (4). Initial studies demonstrated vertical transmission rates as high as 65% (152,153 and 154). However, these initial estimates of the risk for perinatal transmission were based on studies assessing subsequent infants born to mothers who previously had given birth to an HIV-infected child or mothers with advanced HIV disease (152,153 and 154). In retrospect, this most likely represented a subgroup of women with advanced HIV disease, high viral loads, and low CD4 T-lymphocyte counts. Subsequent prospective studies that followed HIV-seropositive pregnant women and their infants revealed lower estimates of transmission, with interesting differences in vertical transmission rates noted worldwide (534). Prior to widespread use of antiretroviral therapy, vertical transmission rates in the United States and western Europe ranged from 14% to 33%; in Africa, rates ranged from 20% to 45%; and in India, a rate of 48% has been reported (Table 10.19). Mofenson (534) suggested that these observed geographic variations in transmission rates may be due to several factors, including (i) duration and severity of maternal HIV infection; (ii) genotypic and phenotypic viral variants present in various geographic areas; (iii) differences in the incidence of cofactors (e.g., STDs, chorioamnionitis); and (iv) differences in infant feeding practices.
TABLE 10.19. REPORTED RATES OF PERINATAL TRANSMISSION OF HUMAN IMMUNODEFICIENCY VIRUS
As noted by the Institute of Medicine, the variable transmission rates reported reflects the multiple factors that influence perinatal transmission (272). This is not surprising, given that studies prior to widespread availability of antiretroviral therapy demonstrate that mother-infant transmission can be influenced by a variety of maternal biologic factors. Most recent interest has focused on the role of maternal viral load in determining perinatal transmission of HIV. Studies have linked high maternal viral load to increased risk of HIV perinatal transmission (548,549,550,551,552,553,554,555,556,557,558,559,560 and 561). Contopoulos-Ioannidis and Ioannidis (558) performed a meta-analysis of the predictive value of maternal cell-free viral load in vertical transmission of HIV-1 in 696 untreated and 419 treated mother-infant pairs. The pooled rate of transmission in untreated women in the 95%) of menstruation-associated TSS cases occur in white women. Approximately 95% of TSS cases in women occurred in association with menstruation (1,2,4,5,7,11). Nearly 99% of women with menstruation-associated TSS wore tampons (5,8,45,46 and 47). At this peak time, the estimated incidence of TSS ranged from three to 15 cases per 100,000 menstruating women per year (5,7,8,12,48). The CDC established national surveillance for TSS (16). Since 1980, with increased public awareness in tampon habits and absorbency, the estimated incidence of TSS has declined dramatically to a level of two to four cases per 100,000 women per year (Fig. 11.1). In 1980, there were 890 cases of TSS reported, of which 812 (91%) were associated with menstruation. By 1989, only 61 cases of TSS were reported, of which 45 (74%) were menstrual (16). In 1980, the mortality rate for menstrual TSS (MTSS) was 5% (38/772); by 1988 and 1989, there were no deaths among women with MTSS (15). During this same period, reporting of nonmenstrual cases of TSS remained constant. A multistate active surveillance study in 1986 to 1987 confirmed the trends noted by the CDC's national passive surveillance (49). The study encompassed a population of 34 million persons and reported that the rate for MTSS was one per 100,000 women aged 15 to 44 years. This was a significant reduction from the rates reported in similar active surveillance reports in 1980, with rate of 6.2 per 100,000 in Wisconsin (7), nine per 100,000 in Minnesota (50), and 12.3 per 100,000 in Utah (51). Active surveillance also demonstrated that the proportion of TSS associated with menstruation had decreased substantially; by 1988, MTSS accounted for only 55% of cases of TSS (16,49). The sudden increase in TSS cases reported in 1980 may have been the result of several factors: (a) the introduction of new tampon products, (b) changes in the constituent materials of tampons, and (c) a change in S. aureus. A dramatic decrease in the number of menstruation-associated TSS cases reported to the CDC was noted following the removal of Rely® tampons from the market in September 1980 (Fig. 11.1) (52). However, using an active-passive surveillance system, Osterholm and Forfang (50) reported there was no decrease in the number of TSS cases in the state of Minnesota following the removal of Rely® tampons from the market. In addition, Davis and Vergeront (53) have suggested that TSS reporting may have been influenced by news media publicity related to TSS in the late summer and fall
of 1980. In a review of MTSS from 1980 to 1990, the CDC suggested that the principal reason for the decreased incidence of MTSS may be the decreased absorbency of tampons (16). They noted that in 1980, very high absorbency tampons (>15.4 g) were used by 42% of tampon users. In 1982, the United States Food and Drug Administration (FDA) required tampon package labeling advising women to use the lowest absorbency possible. As a result, by 1983 the proportion of very-high-absorbency tampons had decreased to 18% (54); by 1986, very-high-absorbency tampons were used by only 1% of women using tampons (16). In March 1990, the FDA commenced standardized absorbency labeling of tampons ranging from 6 to 15 g. Very-high-absorbency tampons are no longer marketed in the United States. Tampon composition also is an independent risk factor for TSS and has changed since 1980 (16). The Rely ® tampon consisted of polyester foam and cross-linked carboxymethylcellulose, a combination no longer used in tampons. The unique composition of Rely® tampons may have been responsible for the significant increased risk for MTSS associated with that product (51). Polyacrylate-containing tampons were withdrawn from the market in 1985 (Fig. 11.1), resulting in a further decrease in reported cases of MTSS. Current tampons are manufactured from cotton and/or rayon (16). Although reported cases of MTSS have declined substantially since the peak in 1980, reporting of NMTSS has remained constant during this time (16). Now nearly one half of TSS cases are nonmenstrual, up considerably from the 6% of TSS cases not associated with menstruation prior to 1981 (45). As suggested by Duff (55), this increase probably is due to two factors: (i) a change in the patterns of tampon use, and (ii) improved recognition of TSS in a variety of clinical settings. As reported by Reingold and coworkers (13), patients with TSS not associated with menstruation differ significantly in age and racial distribution from those with menstruation-associated TSS. First, it can occur in males, and TSS associated with infections not involving the vagina occurs equally in men and women. Of patients with non–menstruation-associated TSS, 11% were not white compared with only 2% of menstruation-associated TSS cases (13). The age range for non–menstruation-associated TSS ranged from the newborn to the elderly rather than the reproductive age span seen with menstruation-associated TSS. Kain et al. (18) contrasted and compared the clinical and laboratory features of patients with NMTSS and those with MTSS. Compared to patients with MTSS, patients with NMTSS were a more heterogeneous group with varying host factors and clinical presentations. The NMTSS group differed from the MTSS group in terms of (i) frequency of prior antimicrobial therapy (46% vs. 16%; p = 0.05); (ii) rate of nosocomial acquisition (65% vs. 0%; p = 0.0001); and (iii) time of onset of fever and rash in relation to the initial symptoms (p = 0.005 and 0.03, respectively, with earlier onset in the NMTSS group). Patients with NMTSS experienced more frequent renal and central nervous system (CNS) complications and less frequent musculoskeletal involvement (p = -0.07 in all three). In this study, stepwise discriminant analysis identified four variables that differentiated NMTSS and MTSS patients: (i) delayed onset of TSS symptoms after the precipitating injury or event; (ii) more frequent CNS manifestations; (iii) less frequent
musculoskeletal involvement; and (iv) higher degree of anemia. In addition, Kain et al. (18) reported that S. aureus associated with NMTSS and MTSS produced TSST-1 with comparable frequency (62% vs. 84%; p = 0.2). However, production of staphylococcal enterotoxin A was less common in NMTSS than MTSS (33% vs. 74%; p = 0.01), and MTSS strains more commonly coexpressed TSST-1 and staphylococcal enterotoxin A than did NMTSS isolates (68% vs. 28%; p = 0.01). Although mortality was higher in the NMTSS group (12.5% vs. 4.8%), this was not statistically significant. These authors concluded that NMTSS is clinically and microbiologically distinct from MTSS (18). Prior to 1981, TSS occurring in adults was believed to be primarily a menstruation-associated syndrome related to tampon use. It now is recognized that TSS is associated with a wide variety of conditions unrelated to menses and/or tampon use (Table 11.1). In 1982, the CDC reported 54 cases of NMTSS (13). The symptoms and signs in the cases were similar to those described in menstruation-associated TSS, and S. aureus was recovered from 88% of nonmenstrual cases. Nonmenstrual TSS has been reported in a variety of surgical procedures and soft tissue infection (Table 11.1). In a retrospective review of 390,000 surgical cases, Graham and coworkers (56) demonstrated that the incidence of postoperative TSS was 0.003% (12 cases). In addition, TSS has been reported in association with the use of contraceptive diaphragm (68,69 and 70) and vaginal contraceptive sponges (65). As noted by Friedell and Mercer (14), TSS can occur in almost any situation in which S. aureus infection can be harbored and an appropriate environment established to favor production of TSST-1. Disruption of normal skin or mucous membrane barrier that allows a localized S. aureus infection to transmit toxins systemically appears to be a common factor in nonmenstrual cases of TSS (14). MacDonald and coworkers (66) reported that TSS can occur as a complication of influenza and influenza-like illness. These authors identified nine cases compatible with TSS complicating influenza or influenza-like illness. Staphylococcus aureus was isolated from respiratory tract secretions in all eight cases cultured, and five of seven S. aureus isolates available for determination of exotoxin production produced TSST-1. Langmuir et al. (67) hypothesized that the plague of Athens, which ended the golden age of Athens, was TSS complicating an epidemic of influenza.
TABLE 11.1. CONDITIONS ASSOCIATED WITH NONMENSTRUAL TOXIC SHOCK SYNDROME
Based on observations made in investigations demonstrating TSS associated with surgical wound infections (57,58), the incubation period for TSS ranged from 1 to 4 days (median 2 days). These data clearly demonstrate the potential for rapid development of a fulminant disease in previously healthy individuals who have been exposed to the specific factor(s) responsible for TSS (13). Initial reports suggested that the case fatality rate for TSS was as high as 13% (4,9). The three major causes of death in TSS are adult respiratory distress syndrome (ARDS), intractable hypotension, and hemorrhage due to disseminated intravascular coagulation (DIC). Subsequent reports noted a dramatic decrease in the case fatality rate for TSS, with a rate of 3.1% in 1981 (45) and no deaths associated with MTSS in 1988 and 1989 (6). At the present time, mortality from staphylococcal TSS usually is less than 5% (2). This improvement in survival most likely is a reflection of increased awareness of TSS by clinicians, with resultant early diagnosis and institution of appropriate therapeutic measures. A recurrence rate of approximately 30% was noted by investigators at the CDC (4,8). Multiple recurrences have been reported in the same patient (6,8,63). Kain et al. (18) reported that the recurrence rate was similar for patients with MTSS and NMTSS. In both instances, recurrence was significantly more frequent in the absence of specific antistaphylococcal therapy. Streptococcal Toxic Shock Syndrome In the last half of the 1980s, there was a reemergence of rheumatic fever and severe group A streptococcal infections in developed countries (1). Fulminant group A streptococcal infections associated with shock, organ failure, necrotizing fasciitis, bacteremia, and death occurred and were termed streptococcal toxic shock syndrome (1,2). In 1987, Cone et al. (23) reported two cases of severe group A streptococcal infection that had clinical features similar to staphylococcal TSS. Bartter et al. (25) applied the name toxic streptococcal syndrome or streptococcal toxic shock-like syndrome to this entity. Stevens et al. (26) further characterized the syndrome in a series of 20 patients. Patients were generally healthy and less than 50 years of age. All had invasive group A streptococcal infection associated with shock, multiorgan system involvement, and rapidly progressive destructive soft tissue infection (e.g., necrotizing fasciitis). The case fatality rate was 30% despite appropriate treatment. Among the isolates available for testing, M types 1 and 3 were the most common serotypes, and 80% produced pyrogenic exotoxin A (26). By 1995, Stevens (71) summarized more than 100 additional published reports of cases of streptococcal TSS. Cases of streptococcal TSS tend to be sporadic, with an estimated prevalence of 10 to 20 cases per 100,000 population at the maximum (1). Eriksson et al. (35) assessed the epidemiologic and clinical aspects of streptococcal
TSS. In their retrospective review of group A streptococcus invasive infections in Stockholm from 1987 to 1995, they noted an incidence of 2.3 (annual range 1.3 to 3.7) cases per 100,000 residents. The incidence increased with age, being 6.1 per 100,000 individuals over 65 years of age. Streptococcal TSS developed in 19 (13%) of the 151 episodes of invasive group A streptococcal infection. The case fatality rate was 11% overall, but among the cases of streptococcal TSS the fatality rate was 47%. In a multivariate logistic regression model, streptococcal TSS was significantly associated with alcohol abuse (odds ratio [OR], 6.3) and infection with a M1T1 strain (OR, 6.7). Case fatality was associated with age (OR, 14.5), immunosuppression (OR, 4.7), and streptococcal TSS (OR, 21.5). Hypotension was significantly associated with mortality, whether or not TSS developed. Among obstetric and gynecologic patients, streptococcal TSS has occurred with cellulitis, wound infection, septic abortion, postpartum endomyometritis, and peritonitis (26,29,30,71). Although streptococcal TSS in obstetrics and gynecology is seen most often in pregnant patients, this entity has occurred with after vaginal and abdominal hysterectomy infections, ovarian abscesses, and pelvic inflammatory disease, with or without an intrauterine device (26,72,73 and 74). Clostridium Sordellii Toxic Shock Syndrome Since the 1980s, there have been six obstetric cases of C. sordellii TSS documented in the literature (40,41,42,43 and 44). These cases of necrotizing subcutaneous and muscle infections with C. sordellii occurred in obstetric patients following vaginal or cesarean deliveries (44). In contrast to other clostridial soft tissue infections characterized by widespread undermining of the subcutaneous tissue, myonecrosis, and rapid progression of tissue destruction away from the wound site, puerperal C. sordellii infections present with relatively “clean” incisions and localized tissue damage (44). Rather, C. sordellii is associated with a profound exotoxin-mediated systemic response characterized by anasarca, refractory hypotension, and marked leukocytosis (44). None of the obstetric patients with C. sordellii TSS survived (40,41,42,43 and 44).
ETIOLOGY Staphylococcal Toxic Shock Syndrome A significant association between TSS and tampon use was identified in a series of case-control epidemiologic studies (4,5,7,8,75,76 and 77). Nearly 70% of women in the United States used tampons during their menstrual period; a significantly greater proportion used tampons during the menstrual period in which TSS occurred than did matched controls. Initial studies revealed no association between TSS and any specific brand of tampons (4,5,7,8). However, subsequent investigations demonstrated a significant association between TSS and the use of Rely® tampons (46,48,75,76 and 77). In September 1980, the CDC reported that 26% of controls used Rely® tampons, but 71% of TSS cases were exclusive Rely® users (46). Schlech and coworkers (76) reported that the relative
risk for development of TSS among Rely® users compared with users of other tampon brands was 7.7 (99% confidence interval [CI], 2.1 to 27.9). These workers noted no significant influence of tampon absorbency that could be separated from the risk associated with the use of Rely®. Osterholm and associates (77) demonstrated that the odds ratio for developing menstruation-associated TSS with any use of tampons compared with no use of tampons was 18 (p < 0.001). When exclusive use of particular tampon brands was evaluated, Rely® was the only brand with a significantly increased odds ratio (2.49; p = 0.005). This increased relative risk with Rely® tampons was beyond that predicted by absorbency alone. In contradistinction to the CDC findings (76), Osterholm et al. (77) demonstrated that women who used high-absorbency tampons had a greater relative risk of developing TSS than women who used low-absorbency tampons. Clearly, tampon use has been established as a risk factor for TSS in menstruating women, with an odds ratio ranging from 11 to 18 in the initial studies conducted in 1980 (7,8,51,76,77). Based on cases in 1983 and 1984, Berkley et al. (54) reported an odds ratio of 33 for tampon use versus no tampon use. In the 1980 studies designed to assess the role of specific tampon brands, the odds ratio for the use of Rely® tampons compared with the use of other brands ranged from 2.5 to 7.7 (51,76,77 and 78). The results of the 1983–1984 study (54) and the data from the 1980 CDC II study are compared in Table 11.2. The risk for Rely® tampons was substantially higher than for other brands. The brands available during both intervals had similar odds ratios.
TABLE 11.2. RISK OF TOXIC SHOCK SYNDROME IN SINGLE-BRAND USERS COMPARED WITH THAT IN REFERENCE USERS OF TAMPAX TAMPONS
Several mechanisms by which tampons increase the risk of TSS have been suggested. In the tristate study in 1980, a significant increase in risk with tampons of higher absorbency was found, and the authors raised a question whether the chemical composition of tampons might be a factor in the development of TSS (77). To assess the interaction of absorbency and chemical composition, Berkley et al. (54) (CDC study III) compared tampon brands used by women with MTSS reported to the CDC passive
surveillance system in 1983 and 1984 with those used by women in a national marketing survey during the same time period. The results clearly demonstrated an increased risk with the use of tampons with increased absorbency (Fig. 11.2). For each 1-g increase in absorbency, the risk of TSS increased by 37% (OR, 1.37; 95% CI, 1.29–1.45). Although absorbency has been evaluated because it is readily quantifiable and standardized measurements are available, as noted by Broome (17), absorbency may be a surrogate for some other dose-response effect, such as increased oxygen content or increased ability to bind cations such as magnesium.
FIGURE 11.2. Odds ratios for toxic shock syndrome based on tampon absorbency and chemical composition. *Adjusted for age group and year of survey.
Toxic shock syndrome occurs in nonmenstruating females and in men. Thus, it is apparent that tampons by themselves are not the explanation for TSS. However, the composition of materials used in tampons may play an important role in the development of TSS (47). As noted by Shands et al. (8), prior to 1977, all tampon products were made of rayon or a rayon/cotton blend. After 1977, 44% of tampon products with 65% of the market contained more absorbent synthetic materials, such as polyacrylate fibers, carboxymethylcellulose, high-absorbency rayon-cellulose, and polyester foams. Tierno (79) and Hanna (80) have postulated that these synthetic components interact with the bacterial flora of the vagina to provide an environment that is appropriate for the growth of toxin-producing staphylococci. However, most observers were of the opinion that the increased risk associated with the higher-absorbency tampons appeared to be independent of tampon composition (77,81). Subsequently, Tierno and Hanna (82) reported that cotton tampons lead to little, if any, TSST-1 production by S. aureus and that cotton absorbed any toxin that was produced. These authors concluded that tampons made solely of cotton would be safer than the cotton/rayon tampons currently available in the United States. These data, which are inconsistent with the epidemiologic data demonstrating that the risk of TSS is mainly dependent on tampon absorbency rather than composition, have been questioned and refuted by Schlievert (83) and Parsonnet et al. (84). Schlievert (83) demonstrated that cotton tampons neither prevent TSST-1 production nor significantly absorb toxin onto the fibers. Parsonnet et al. (84) noted that neither cotton nor rayon tampons consistently
increased toxin production, nor were there significant differences between them or their effects on TSST-1. On the other hand, tampons composed of carboxymethylcellulose and polyester foam increased TSST-1 production (84). The association of TSS with S. aureus was identified in the original case-control studies (7,8). Davis et al. (7) reported that S. aureus was recovered in 17 of 23 cervicovaginal cultures obtained from TSS patients. Similarly, Shands and coworkers (8) recovered S. aureus from 62 of 64 vaginal cultures from women with menstruation-associated TSS. Among TSS cases reported to the CDC as of October 1982, S. aureus was recovered from 210 of 215 patients (45). Crass and Bergdoll (85) studied cultures of S. aureus from 434 individuals with TSS. Over 90% of these isolates produced TSST-1. Isolates producing both staphylococcal enterotoxin C and TSST-1 had a greater association with nonmenstrual and fatal cases. Staphylococcus aureus may be present as part of the normal bacterial flora of the vagina and cervix in healthy women. It has been recovered from 0% to 15% of vaginal specimens (6,86,87,88,89,90 and 91), with 5% to 10% the generally accepted prevalence. Interestingly, several studies have noted an increase in the incidence of S. aureus colonization of the female lower genital tract at the time of menses (92,93). Noble et al. (93) reported the recovery of S. aureus from 17% of their patients during menstruation compared with 5.8% at mid cycle in the same women. Saunders and associates (94) suggested that an imbalance among the normal microflora of the female genital tract may be crucial to the development of menstruation-associated TSS. They reported that lactobacilli, which play a major role in maintaining the normal vaginal environment, have an inverse relationship with staphylococci, and they postulated that lactobacilli served as a natural defense against S. aureus. Moreover, they suggested that an increase in S. aureus may occur because tampons (especially superabsorbable types) remove substrates necessary for lactobacilli to exert this inhibitory effect. The abrupt onset of the clinical presentation, the multisystem involvement, the rarity of bacteremia, and the similarity of TSS to other illnesses known to be caused by bacterial toxins led investigators to hypothesize that a toxin, either alone or synergistically with another factor(s), was responsible for TSS. With the demonstration that S. aureus was recovered from nearly 100% of TSS cases, speculation focused on a staphylococcal toxin as the culprit. Investigations quickly focused on identification of the presence of staphylococcal toxins associated with isolates of S. aureus recovered from patients with TSS (48,95,96). Schlievert (95) identified a protein that he named pyrogenic exotoxin C. In this investigation, pyrogenic exotoxin C was present in 28 of 28 S. aureus recovered from TSS cases but in only five (16%) of 32 S. aureus recovered from controls. Altemeier et al. (96) found pyrogenic exotoxin C was produced by 131 (91%) of 144 TSS isolates of S. aureus. Bergdoll and associates (97) simultaneously identified a toxin with similar physical characteristics, which they called enterotoxin F. They reported that 61 (94%) of 65 S. aureus strains from TSS patients produced enterotoxin F, and only four (4.6%) of 87 non-TSS S. aureus strains produced enterotoxin F. Furthermore, in a prospective blinded study, Bergdoll et al. (97) noted that all 34 S. aureus strains from TSS produced enterotoxin F compared with three (11%) of 26 control strains. In the early 1980s, a consensus developed that pyrogenic exotoxin C and staphylococcal enterotoxin F were
identical (98,99). Subsequent work has demonstrated that these two toxins are coded by the same gene in the staphylococcal chromosome and thus are identical (100). The current name for this toxin is toxic shock syndrome toxin 1 (TSST-1) (101). Studies in baboons have shown that TSST-1 produces many of the clinical and laboratory manifestations of TSS (102). As described by Crass and Bergdoll (85), this toxin has been demonstrated to be a potent inducer for production of interleukin-1 (IL-1) by macrophages, is a nonspecific T-cell mitogen, and induces suppression of some immune responses. Schlievert (103) has demonstrated that TSST-1 significantly enhanced susceptibility to lethal endotoxin shock by 50,000-fold. Larsen and Schlievert (99) suggested that this enhancement of endotoxin may explain the manifestations of Gram-negative septic shock associated with TSS, and that the combination of TSST-1 and small amounts of circulating endogenous endotoxin from the normal bowel flora provides the best explanation for the pathogenesis of TSS. The massive release of IL-1 has been implicated in mediating many of the symptoms of TSS (104). Toxic shock syndrome toxin 1 is a potent stimulator of IL-1 release (104,105). Beezhold et al. (106) reported that TSST-1 is a more potent inducer of IL-1 than endotoxin, and a synergistic induction of IL-1 was observed when macrophages were stimulated with both TSST-1 and endotoxin. Toxic shock syndrome toxin 1 has been shown to stimulate production of tumor necrosis factor alpha (TNF-a) (107). Similarly, other TSS-associated exotoxins and endotoxins induce TNF production (108). These findings are explained by the concept of superantigens, which are a class of bacterially and virally encoded proteins that stimulate a massive immune response by their ability to bypass processing by antigen-processing cells and to bind directly with Class II major histocompatibility molecules, thus stimulating massive numbers of T cells (110). Toxic shock syndrome toxin 1 is such a superantigen. Nearly all MTSS is caused by TSST-1–producing S. aureus (110). Toxic shock syndrome toxin 1 accounts for half of NMTSS cases; staphylococcal enterotoxins B and, to a lesser extent, C account for the remainder (110). Bergdoll and coworkers (97) demonstrated that immunologic status, vis-à-vis TSST-1, may be important in the pathogenesis of TSS. They found that TSS patients had a greater serosusceptibility to enterotoxin F (TSST-1) than did controls; anti-staphylococcal enterotoxin F antibody was present in titers ³1:00 in five (17.2%) of 29 TSS patients, compared with 44 (78.6%) of 56 controls. Crass and Bergdoll (85) tested the sera of 284 patients with TSS for antibodies to TSST-1. They reported that 82.4% of sera from TSS patients had no detectable level of antibody, whereas nearly 80% of healthy controls had antibody levels ³1:800. In this study, the sera from TSS patients also had lower levels of antibody to staphylococcal enterotoxins A, B, and C than did the controls. These authors noted that their findings indicate that TSS patients may have an immunodeficiency that inhibits production of and/or maintenance of antibodies to TSST-1 and staphylococcal enterotoxins. Several variables that influence toxin production have been described (22). These include increased oxygen, increased carbon dioxide, temperature elevation, and a high
pH. Wagner et al. (111) noted that insertion of tampons changes the internal vaginal milieu from an anaerobic to an aerobic state. This increase in oxygen tension enhances the growth of, and toxin production by, S. aureus. Mills et al. (112) suggested that magnesium ion controls the production of TSST-1. With lowered concentrations of Mg2+, toxin production was increased significantly. Conversely, with higher concentrations of Mg2+, the production of TSST-1 was suppressed. There is a direct correlation between absorbency and binding of magnesium, with the most absorptive fibers having the greatest capacity to bind Mg2+. Thus, high-absorbency tampons are associated with low levels of Mg2+ and consequent high production of TSST-1 (22). Kass and Parsonnet (22) claim that absorbability and magnesium-binding capacity can be separated and that the magnesium-combining capacity is critical to the production of TSST-1. Kiyota et al. (113) reported that magnesium-deficient strains of S. aureus secrete more exoprotein than under conditions when magnesium is not limiting. Multiple studies have demonstrated that the concentration of magnesium ion controls the production of TSST-1 and other exoproteins secreted by strains of S. aureus (22,112,113,114 and 115). Duff (55) noted that three conditions are required for development of TSS. First, the patient must be colonized or infected with S. aureus. Second, the staphylococci must be capable of producing the toxin(s) that is associated with the clinical manifestations of TSS. Third, there must be a portal for the toxin(s) to enter the systemic circulation. In NMTSS, S. aureus is present in a localized supportive infection; as toxin is produced, it is absorbed from the infected site into the bloodstream. Duff (55) suggested several mechanisms by which staphylococcal toxin gains access to the systemic circulation in menstruation-associated TSS. Toxin could be absorbed through microulcerations in the vaginal mucosa secondary to trauma from tampons and/or inserters. Superabsorbable tampons may obstruct menstrual flow, producing reflux of menstrual blood containing toxin through the fallopian tubes into the peritoneal cavity. This would result in absorption of toxin across the peritoneum. Alternately, toxin could be absorbed across the denuded endometrial surface. Streptococcal Toxic Shock Syndrome Following World War II, the rates of morbidity and mortality due to group A streptococcal infections declined steadily (36). Since the mid-1980s, a resurgence of severe, invasive disease has occurred (1,2,23,24,25,26,27,28,29,30,31,32 and 33,35,36,116). The most severe form of invasive group A streptococcal infection was recognized as streptococcal TSS (32). The increased incidence of invasive group A streptococcal infections has been associated with an increase in virulence and severity, with many cases of streptococcal TSS and necrotizing fasciitis (117,118 and 119). Whether this change in epidemiology and clinical manifestations is due to the acquisition of new virulence by the organism or compromised host susceptibility to virulence factors produced by the organism remains uncertain (36). Group A streptococci produce multiple virulence factors that contribute to the pathogenesis of invasive group A streptococcal disease (36). These include surface M protein, hyaluronic capsule, proteases, DNases, lipoteichoic acid, streptolysins O and S,
and pyrogenic exotoxins (36). Most importantly, the streptococcal pyrogenic exotoxins function as superantigens that result in activation of large numbers of immune cells. This, in turn, leads to the synthesis and release of very large amounts of inflammatory cytokines that mediate many of the systemic manifestations of sepsis, including hypotension and multiorgan failure (120,121). Group A streptococci produce three pyrogenic exotoxins: streptococcal pyrogenic exotoxins A, B, and C (1). These exotoxins induce fever and facilitate development of shock by lowering the threshold to endotoxin (1,110). Streptococcal pyrogenic exotoxins A and B induce mononuclear cells to produce inflammatory cytokines, including TNF-g, IL-1b, and interleukin-6 (122,123). These inflammatory cytokines mediate the fever, hypotension, organ failure, and tissue injury associated with streptococcal TSS (36). Two recently identified pyrogenic exotoxins, streptococcal superantigen and mitogenic factor, also can induce a cytokine synthesis, but their role in streptococcal TSS has not been elucidated (36). Lack of protective immunity against the virulence factors produced by group A streptococci may increase host susceptibility to infection (36). Initial studies suggested that low levels of antibodies against specific streptococcal pyrogenic exotoxins or the M protein increase host susceptibility to invasive group A streptococcal infection (124,125). As a result, it was hypothesized that the low levels of anti-M1 antibody in the general populations may have contributed to the reemergence of severe group A streptococcal infections in the United States, Canada, and Scandanavia (124,125 and 126). However, Basma et al. (36) failed to confirm this hypothesis. These investigators studied 33 patients with invasive infection caused by genotypically indistinguishable M1T1 strains of S. pyogenes who had different disease outcomes. Although levels of anti-M1 antibodies and antistreptococcal superantigen antibodies were significantly lower in severe (with hypotension) and nonsevere (without hypotension) invasive group A streptococcal infection than in age-matched and geographically matched healthy controls, the levels of these protective antibodies from severe and nonsevere cases were not different (36). Basma et al. (36) concluded that low levels of protective antibodies may contribute to host susceptibility to invasive streptococcal infection but do not modulate disease outcome. Clostridium Sordellii Toxic Shock Syndrome Clostridium sordellii is one of the histotoxic clostridium species that has been associated (usually with other organisms) with production of gas gangrene. There have been descriptions of a distinct TSS attributed to soft tissue infection with C. sordellii alone (37,38,39,40,41,42,43 and 44). Since 1929, it has been recognized that the virulence of C. sordellii was due to a toxin (beta toxin) that caused severe, gelatinous edema in animal models (127). Subsequently, beta toxin was divided into an edema-producing or lethal toxin and a hemorrhagic toxin (128). The lethal toxin is primarily responsible for the morbidity and mortality associated with C. sordellii infection (44). The natural reservoir for C. sordellii is the gastrointestinal and genital tract. Sosolik et al. (44) reviewed the pathogenesis of C. sordellii TSS. Following damage to the mucosa, skin, and other natural barriers associated with cesarean delivery or episiotomy with
vaginal delivery, an anaerobic environment is created secondary to decreased local blood flow and proliferation of indigenous flora. If a toxin-secreting strain of C. sordellii is present and proliferates, tissue necrosis and production of lethal toxin and hemorrhagic toxin occur. Rapid dissemination of the toxins, especially lethal toxin, occurs, leading to the morbidity and mortality associated with C. sordellii TSS.
CLINICAL PRESENTATION Staphylococcal Toxic Shock Syndrome Toxic shock syndrome is a multisystem disease with a wide range of symptoms, signs, and laboratory findings. Characteristically, TSS occurs abruptly with the sudden onset of high fever, chills, myalgias, vomiting, diarrhea, hypotension, and generalized “sunburn-like” rash. The case definition proposed by the CDC is given in Table 11.3.
TABLE 11.3. CASE DEFINITION OF TOXIC SHOCK SYNDROME
The most commonly observed clinical signs and symptoms of TSS are given in Table 11.4. Involvement of the skin and mucous membranes is one of the most characteristic findings. The rash, which occurs early in the disease process, presents as a “sunburn-like” macular erythema; it can evolve to where the patient appears like a “broiled lobster.” Five to 12 days after the onset of TSS in surviving patients, a fine desquamation occurs on the face, trunk, and extremities. This is followed by the nearly pathognomonic full-thickness, peeling-like desquamation of the palms and/or soles. The staphylococcal toxin(s) has a predilection for mucous membranes, and their involvement is often a prominent feature of TSS (47). This manifests clinically as sore throat, oropharyngeal hyperemia, strawberry tongue, nonpurulent conjunctivitis, and/or vaginal hyperemia.
TABLE 11.4. CLINICAL SIGNS AND SYMPTOMS COMMONLY SEEN IN PATIENTS WITH TOXIC SHOCK SYNDROME
Gastrointestinal symptoms and signs are frequent and prominent in TSS. Vomiting and watery diarrhea have been noted in approximately 90% of cases. These symptoms occur early in the disease, usually are severe, and are associated with marked generalized abdominal tenderness. Hepatomegaly and pancreatitis have been reported occasionally. Another early characteristic finding is the presence of myalgias, which have been reported in 88% to 100% of TSS cases (6,9,10 and 11,59). Exquisite muscle tenderness can be elicited by movement or touch. Additional musculoskeletal findings include arthralgias, synovitis of hand joints, and sterile joint effusions. Neurologic involvement may be difficult to assess in the presence of high fever and/or severe hypotension associated with TSS. In general, neurologic symptoms and signs have included headache, confusion, loss of consciousness, disorientation, agitation, photophobia, seizures, meningeal irritation, and evidence of psychomotor retardation. Tofte and Williams (9) and Chesney et al. (11) noted normal glucose and protein levels and very low counts of mononuclear or polymorphonuclear cells in the spinal fluid. Hypotension and shock are prominent features of TSS. They occur early in the disease process and constitute one of the major criteria for the diagnosis of TSS. The presence of orthostatic syncope now has been accepted as an alternative criterion for documented shock. Multiple cardiac abnormalities have been reported in TSS cases. These include sinus or supraventricular tachycardia, first-degree heart block, premature ventricular beats, nonspecific ST-T wave changes, and T-wave inversion in precardial leads (6,9,11,47,59). Shands et al. (8) reported cases with pericarditis and vasculitis. Chesney and coworkers (11) noted that low central venous pressure was present in their TSS cases. McKenna and associates (6), using Swan-Ganz measurements, noted that TSS patients had high cardiac output, low peripheral resistance, normal pulmonary wedge pressure, and normal pulmonary resistance.
Renal involvement has been demonstrated in more than 80% of patients with TSS (6,8,9,10 and 11,59). This renal dysfunction has been manifested by sterile pyuria, mild proteinuria, hematuria, azotemia, hyponatremia, increased urinary sodium, hyperkalemia, and decreased creatinine clearance. The prognosis for patients with TSS often is determined by the presence of pulmonary involvement, especially ARDS or “shock lung.” Fisher and coworkers (10) reported that tachypnea and hypoxemia (Po2 50% may occur), decreased renal function (mean serum creatinine >2.5 normal), hypoalbuminemia, and hypocalcemia (125,134). With necrotizing fasciitis, an elevated or rising creatine phosphokinase level is present (125,134). Adult respiratory distress syndrome has been reported in up to 55% of patients, usually manifesting after onset of hypotension (125,134). Bacteremia occurs in 60% of patients with streptococcal TSS (1). The major complications associated with group A streptococcal soft issue infection and their frequency are given in Table 11.11. Mortality in association with streptococcal TSS is high, ranging from 30% to 70% (124,125,134,135). Among the survivors, morbidity is similarly very high (1). An estimated 50% of patients with streptococcal TSS undergo major surgical procedures, often as a life-saving procedure to address infection sites, including surgical debridement, exploratory laparotomy, hysterectomy (with or without bilateral salpingo-oophorectomy), fasciotomy, and amputation (125,134).
TABLE 11.11. COMPLICATIONS OF GROUP A STREPTOCOCCAL SOFT TISSUE INFECTION
A comparison of the clinical features seen in staphylococcal and streptococcal TSS is given in Table 11.12.
TABLE 11.12. COMPARISON OF STAPHYLOCOCCAL AND STREPTOCOCCAL TOXIC SHOCK SYNDROME
Stevens (1) has suggested that exposure to a virulent strain of group A streptococci usually is not sufficient by itself to cause streptococcal TSS. Rather, he noted that other factors may be required for streptococcal TSS to occur (1). Examples of such factors include (i) varicella infection, which can disrupt anatomic barriers in the skin and mucosal surfaces; (ii) influenza infection affects respiratory epithelium providing a portal of entry; (iii) viral infections may predispose the immune system to streptococcal TSS; (iv) disruption of skin barriers due to lacerations, burns, surgical incisions, intravenous drug use, and bites; (v) compromise of the integrity of the uterine mucosal surface
during the birthing process, which allows entry of group A streptococci from vaginal colonization or from hospital environment or personnel; (vi) use of nonsteroidal antiinflammatory agents (NSAIDs), which inhibit neutrophil function (chemotaxis, phagocytosis, and bacterial killing), augment cytokine production, and attenuate the cardinal manifestations of inflammation (136); and (vii) genetic predisposition to severe streptococcal infection secondary to factors such as human leukocyte antigen class II antigen type (1). Clostridium Sordellii Toxic Shock Syndrome Toxic shock syndrome has been associated with C. sordellii necrotizing subcutaneous infections in previously healthy postpartum patients (40,41,42,43 and 44). Histotoxic clostridia, such as C. sordellii, can produce a broad spectrum of soft tissue infections ranging from cellulitis to necrotizing fasciitis or frank myonecrosis (44). Patients initially present with gastrointestinal distress and generalized weakness. The cardinal findings associated with C. sordellii TSS are systemic toxicity and profound, rapidly progressive, widespread edema, often progressing to anasarca (44). The edema initially is localized to the perineum or surgical site (40,41,42,43 and 44). With C. sordellii TSS, cardiovascular status deteriorates rapidly, and progressive, refractory shock occurs (44). With C. sordellii TSS, fever is absent or the patient presents with hypothermia. Minimal purulent discharge is present at the localized site of infection (44). Typical laboratory findings include marked leukocytosis and left shift, hemoconcentration, evidence of DIC, and isolation of C. sordellii as the principal anaerobic organism from infected tissue site (44). Blood cultures are negative for pathogens, including C. sordellii.
DIAGNOSIS As described earlier, TSS presents with a broad spectrum of clinical manifestations, and multiple organ systems are involved. No definitive diagnostic or confirmatory laboratory test has been developed for TSS, and none of the variety of clinical and laboratory findings are pathognomonic for the disease. Moreover, after suspecting TSS, it is necessary to differentiate among the three types of TSS: staphylococcal, streptococcal, and C. sordellii. It is necessary not only to differentiate TSS from a multitude of other diseases listed in Table 11.13, but also to identify the causative organism.
TABLE 11.13. DIFFERENTIAL DIAGNOSIS OF TOXIC SHOCK SYNDROME
Staphylococcal Toxic Shock Syndrome An empiric diagnosis must be based on the typical constellation of clinical and laboratory findings previously described (Table 11.3, Table 11.4 and Table 11.5). Clinicians should consider the possibility of staphylococcal TSS in any patient with an unexplained febrile, exanthematous, multisystem illness, especially if the illness is associated with menstruation or a documented or suspected S. aureus infection. A high index of suspicion and early diagnosis are crucial to instituting appropriate therapy and gaining a favorable prognosis. The majority of diseases on the differential diagnosis list are easily differentiated from TSS based on clinical grounds and laboratory evaluation. However, several of the conditions are very similar and may be difficult to differentiate. Streptococcal scarlet fever is unusual after 10 years of age and generally occurs secondary to an upper respiratory tract illness from which the group A b-hemolytic streptococcus can be recovered. If seen early in the disease, the rash of scarlet fever is described as a “sandpaper rash.” Late in the disease, the desquamation of scarlet fever and TSS are clinically similar, although microscopically they occur in different levels of the skin. Serum antibodies to streptococcal extracellular products, such as antistreptolysin O, are always present with scarlet fever. Since 1987, the streptococcal toxic shock-like syndrome has been recognized (23,24,25,26,27,28,29,30,31 and 32). This syndrome is caused by group A b-hemolytic streptococcus (S. pyogenes) and is characterized by marked systemic toxicity, rapidly progressive multisystem organ failure, and a high mortality rate (30%–60%) (30). A comparison of the common clinical findings in staphylococcal and streptococcal TSS is given in Table 11.12. Other than the association of tampon use and menstruation with staphylococcal TSS, the clinical findings most helpful in differentiating these entities are as follows: (i) rash is more common with staphylococcal TSS; (ii) tissue necrosis is more common with streptococcal TSS; and (iii) pain is common with streptococcal TSS but rare with staphylococcal TSS. Definitive differentiation relies on isolation of S. aureus versus S.
pyogenes. Mucocutaneous lymph node syndrome (Kawasaki disease) presents with a clinical picture very similar to that of staphylococcal TSS. Some investigators have suggested that this disease may be a variant form of TSS (6,7,9). However, mucocutaneous lymph node syndrome is primarily a pediatric disease, and the majority of cases have occurred in children less than 5 years old. In addition, as noted by McKenna et al. (6), myalgias, elevated creatine phosphokinase, abdominal pain, hypotension and shock, ARDS, thrombocytopenia, and renal failure are very rare or absent in mucocutaneous lymph node disease. Diseases such as leptospirosis, rubeola, and Rocky Mountain spotted fever are clinically strikingly similar to TSS. However, serologic testing of acute and convalescent serum can establish or rule out these diseases. In addition, Leptospira can be cultured from spinal fluid and/or blood. Streptococcal Toxic Shock Syndrome The Working Group on Severe Streptococcal Infections defined group A streptococcal TSS (32), and their proposed case definition for streptococcal TSS is given in Table 11.10. The first criterion is isolation of group A streptococci. An illness in a patient with a group A streptococcus isolated from a normally sterile site (IA) that meets the remainder of the definition can be designated as a definite case of streptococcal TSS (32). An illness in a patient with group A streptococci isolated from a nonsterile site only (IB) but that fulfills other criteria and has no other identified etiology can be designated as a probable case. The second requirement of the case definition is the presence of signs associated with severe infection, especially shock (IIA) and organ system involvement (IIB). Classification of group A streptococcal infections is given in Table 11.14. The features of streptococcal TSS can be readily recognized clinically or by routine laboratory tests (32).
TABLE 11.14. CLASSIFICATION OF GROUP A STREPTOCOCCAL INFECTIONa
Clostridium Sordellii Toxic Shock Syndrome Clostridium sordellii TSS should be suspected when a previously healthy woman with recent “clean” obstetric wounds or incisions presents with rapidly spreading edema that initially was localized to the perineum or surgical site and rapid cardiovascular decompensation with progressive, refractory hypotension (44). Unlike streptococcal TSS, tissue necrosis associated with C. sordellii TSS remains localized to the area surrounding the wound (44). Intraoperative frozen-section biopsies are useful for diagnosing myonecrosis associated with C. sordellii (44). The absence of muscle involvement on the biopsy of necrotizing subcutaneous infection strongly suggests a diagnosis of necrotizing fasciitis (44). Definitive diagnosis requires isolation of C. sordellii from infected tissue. On the other hand, blood cultures are negative.
TREATMENT AND PREVENTION Patients with suspected TSS require prompt and aggressive therapy (1,2,47,55). Initial evaluation of the patient commences simultaneously with aggressive fluid resuscitation aimed at maintaining adequate circulating volume, cardiac output, blood pressure, and perfusion of vital organs. Staphylococcal Toxic Shock Syndrome Initial patient evaluation requires a thorough physical examination, which includes a pelvic examination to look for and remove any tampon, diaphragm, cervical cap, or contraceptive sponge that may be present. If none of these items is present, a thorough search for a localized S. aureus infection must be undertaken. Cultures for S. aureus are obtained from the vagina, rectum, conjunctiva, oropharynx, anterior nares, and any localized sites of infection. Blood, urine, and spinal fluid also are obtained for culture. Serum is obtained for serologic studies to rule out leptospirosis, rubeola, and Rocky Mountain spotted fever. Additional laboratory studies are obtained for a complete blood count, with differential, platelet count, and coagulation screen; electrolytes; liver function tests; calcium, phosphorus, creatine phosphokinase, BUN, creatinine, and urinalysis; chest x-ray; and electrocardiogram. Similar to the management of septic shock, aggressive fluid replacement is the initial priority and is the cornerstone of therapy in patients with TSS. Massive amounts of fluid replacement generally are required, with 8 to 12 L per day not being uncommon (47). To facilitate and monitor administration of such massive volume replacement, placement of a Swan-Ganz catheter is necessary. This central line will allow monitoring of cardiovascular status. An arterial line is established to allow for continuous blood pressure and arterial oxygenation assessment. A urinary catheter is inserted to monitor urine output. When no blood loss has occurred, volume replacement is achieved with
isotonic crystalloid solution, such as normal saline or lactated Ringer's solution; colloids such as Plasmanate or salt-poor albumin; or combinations of crystalloid and colloid. In patients with blood loss secondary to DIC, packed red blood cells and coagulation factors should be included in the fluid replacement scheme. Fluid replacement is guided by measurement of pulmonary artery wedge pressure (PAWP), which should be maintained at 10 to 12 mm Hg (55). Shubin et al. (137) have proposed a “7-3” rule for fluid replacement in shock patients (137). A fluid bolus of 5 to 20 mL/min for 10 minutes is given. If PAWP increases more than 7 mm Hg above baseline, the infusion is temporarily discontinued. If PAWP does not rise more than 3 mm Hg, a second similar fluid challenge is given and the “7-3” rule reapplied. As fluid replacement is initiated, use of military antishock trousers (MAST unit) may be beneficial in producing immediate improvement in cardiac function. This unit mobilizes pooled blood from the lower extremities and lower abdomen/pelvis and returns it into the central circulation. McSwain (138) reported that the MAST unit results in an autotransfusion with 750 to 1,000 mL of autologous blood. The MAST unit is a temporizing measure. Once fluid replacement has reestablished normal perfusion blood pressure, the unit should be deflated gradually. If aggressive volume replacement does not promptly restore blood pressure and perfusion to vital organs, use of vasopressor therapy is indicated. The drug of choice is dopamine, which in low doses has a weak b-mimetic effect that results in increased myocardial contractility and heart rate while simultaneously stimulating dopaminergic receptors that cause vasodilation of the renal, mesenteric, coronary, and cerebral vasculature (1,2). In addition, dopamine produces vasoconstriction in skeletal muscle. This results in diversion of blood from peripheral beds to vital organs (i.e., heart, brain, kidney). Dopamine is administered by continuous intravenous infusion. An initial dose of 2 to 5 µg/kg/min is instituted, and the infusion is increased in small increments until an appropriate perfusion pressure is obtained. As in patients with septic shock, use of glucocorticoids in the management of TSS is no longer recommended. Their efficacy or necessity in TSS has not been demonstrated. Two additional therapeutic modalities have been suggested for management of TSS patients (as well as septic shock patients) with intractable hypotension. Several studies have demonstrated successful results with the use of naloxone (Narcan) (139,140). This narcotic antagonist could block the action of endorphins present in elevated levels in stressed patients, profoundly depressing the cardiovascular system. Although not of proven value in TSS, this approach deserves further investigation. Duff (55) has suggested that hemodialysis might remove the toxin responsible for the hypotension. Such an approach has been used with success in clostridial septicotoxemia secondary to the exotoxins of C. perfringens. This approach has not been evaluated adequately in TSS. Although b-lactamase–resistant, antistaphylococcal antibiotics have not been
demonstrated to shorten the acute episode of TSS or to be necessary for clinical response in the acute episode, they are recommended as part of the therapeutic approach. First, they are advisable because of the risk of bacteremia with S. aureus. More importantly, patients who received b-lactamase–resistant, antistaphylococcal antibiotics had a significantly lower rate of TSS recurrences (7,9,11). Available choices include methicillin, nafcillin, and oxacillin. For penicillin-allergic patients, vancomycin, clindamycin, and gentamicin are alternatives. A major objective in the management of TSS is preventing ARDS, which is one of the leading causes of death in TSS and is associated with very high mortality rates when it complicates septic shock. Oxygen is administered by mask or nasal cannula at the initiation of therapy. Arterial blood gases are monitored frequently to detect hypoxia and/or metabolic acidosis. Close monitoring of fluid replacement is mandatory to avoid excessive fluid replacement. There should be no hesitancy to intubate the patient and institute mechanical ventilation with the earliest signs of decreased pulmonary compliance. With such an aggressive approach, irreversible hypoxic damage to the pulmonary vasculature hopefully will be prevented. Additional therapeutic measures may be necessary on an individual basis, depending on the clinical presentation of TSS. Coagulation abnormalities are treated with fresh frozen plasma, cryoprecipitate, fresh whole blood, and/or platelets. Calcium supplementation may be necessary if hypocalcemia and/or cardiac dysfunction is present. Electrolyte abnormalities must be corrected by administration of electrolyte solutions. Dialysis may be necessary in cases of acute renal failure. Finally, in nonmenstrual cases of TSS, a thorough search for localized S. aureus infection(s) must be undertaken. Such infection sites must be managed appropriately, as indicated by the particular type of infection (i.e., incision and drainage of abscesses, debridement of necrotic tissue). Chesney et al. (11) have suggested that consideration be given to administering intravenous immunoglobulin (IVIG) to patients severely ill with TSS. High levels of antibody to TSST-1 have been demonstrated in IVIG (141). Recommendations have been made to reduce the risk of TSS (6). Women can significantly decrease their risk of developing TSS by not using tampons. The risk also may be reduced by avoiding continuous tampon use and alternating tampon use with napkins. Based on studies demonstrating a very significant increased risk to develop TSS among users of high-absorbency tampons (40,60), only low-absorbency tampons are available in the United States. Women, especially young adolescents, must be educated about the signs and symptoms of TSS. They should be warned that if such signs and symptoms occur, the tampon must be removed and medical care sought promptly. Women with a previous history of TSS should, ideally, not use tampons. In summary, good results in the management of TSS depend on a high index of suspicion; prompt and aggressive fluid replacement; close monitoring of cardiovascular status, respiratory status, and urine output; and prevention or early recognition, and treatment, of ARDS. Streptococcal Toxic Shock Syndrome
Unlike in staphylococcal TSS where antibiotic therapy has not been shown to affect the acute clinical response, antibiotic therapy is a critical component of treating streptococcal TSS. Group A streptococci are very sensitive to b-lactam antibiotics, and multiple studies have demonstrated the efficacy of penicillin in treating cutaneous infections and pharyngitis due to penicillin (1). However, most aggressive, invasive group A streptococcal infections (e.g., necrotizing fasciitis, subcutaneous gangrene, myositis) have been shown to respond less well to penicillin (1,2). The explanation for this clinical finding relates to growth rates of the organism and the quantity of penicillin-binding sites present on the group A streptococci (1,2). With mild infections or early in the infection, organisms are present in low concentrations and are rapidly growing (1,2). Subsequently, in the later stages of infection, growth slows and there is a loss of penicillin-binding proteins. This loss of binding sites decreases the effectiveness of b-lactam antibiotics. As a result, clindamycin in a dose of 900 mg intravenously every 8 hours has become the drug of choice for treating streptococcal TSS (1,2). Several factors have been linked to the greater efficacy of clindamycin (1,2): (i) inoculum size or stage of growth does not affect efficacy; (ii) clindamycin is a potent suppressor of bacterial toxin synthesis; (iii) clindamycin facilitates phagocytosis of S. pyogenes by inhibiting M protein synthesis; (iv) clindamycin suppresses synthesis of penicillin-binding proteins, which are enzymes involved in cell wall synthesis; (v) clindamycin has a longer postantibiotic effect than most b-lactam antibiotics; and (vi) clindamycin suppresses lipopolysaccharide-induced monocyte synthesis of TNF-a by 40% (1,2,142,143). Erythromycin (1 g intravenously every 6 hours) is less active than clindamycin but better than penicillin (142). Although single-agent antimicrobial therapy is effective against group A streptococci, combination broad-spectrum antimicrobial therapy generally is recommended due to the severe nature of these group A streptococcal infection (2). Although antibiotic treatment is extremely, important additional intervention including prompt and aggressive exploration and debridement of suspected deep-seated group A streptococcal infection are required (1). Early surgical intervention is often a lifesaving step. General supportive measures as described in the management of staphylococcal TSS are an important component of managing patients with suspected streptococcal TSS (1,2). Intravenous immunoglobulin has been used to neutralize toxin as well as cytokines (144). Pooled IVIG contains antibodies to streptococcal superantigens and decreases production of inflammatory cytokines (1,2). Weiss and Laverdiere (142) reported that, in patients with streptococcal TSS, IVIG significantly decreased the death rate (40% vs. 80%). Additional reports have shown dramatic improvement with IVIG (1,144). The recommended dose for IVIG is 2g/kg for 2 days. Clostridium Sordellii Toxic Shock Syndrome Management of C. sordellii TSS among pregnant women is similar to the treatment of C. perfringens myonecrosis and/or septicotoxemia. Treatment includes aggressive and early surgical debridement, broad-spectrum antimicrobial agents, high-dose penicillin,
hyperbaric oxygen therapy, and intensive supportive care. Despite such aggressive therapy, all reported maternal infections, to date, have been fatal. Cardiovascular collapse secondary to marked “third spacing” with sequestration of fluid has been the presumed cause of death (44). Once C. sordellii commences production of the exotoxins (especially lethal toxin), aggressive debridement and antibiotic therapy do not appear capable of reversing the pathologic chain of events initiated by the exotoxin (44). It appears that either neutralization or elimination of the exotoxins is required if survival is to occur (44). However, exogenous antisera are not available. Use of plasma exchange to diminish toxin levels might be of benefit. Use of this approach in the presence of the significant hypotension associated with C. sordellii TSS is problematic (44). CHAPTER REFERENCES 1. Stevens DL. The toxic shock syndromes. Infect Dis Clin North Am 1996;10:727–746. 2. Reiss MA. Toxic shock syndrome. Prim Care Update Obstet Gynecol 2000;7:85–90. 3. Todd J, Fishaut M, Kapral F, et al. Toxic-shock syndrome associated with phage-group-1 staphylococci. Lancet 1978;2:1116–1118. 4. Toxic-shock syndrome—United States. MMWR 1980;29:229. 5. Follow-up on toxic-shock syndrome—United States. MMWR 1980;29:297. 6. McKenna U, Meadows JA, Brewer NS, et al. Toxic-shock syndrome, a newly recognized disease entity: report of 11 cases. Mayo Clin Proc 1980;55:663–672. 7. Davis JP, Chesney PJ, Wand PJ, et al. Toxic shock syndrome: epidemiologic features, recurrence, risk factors, and prevention. N Engl J Med 1980;303:1429–1435. 8. Shands KN, Schmid GP, Dan BB, et al. Toxic-shock syndrome in menstruating women: association with tampon use and Staphylococcus aureus and clinical features in 52 cases. N Engl J Med 1980;303:1436–1442. 9. Tofte RW, Williams DN. Toxic-shock syndrome: clinical and laboratory features in 15 patients. Ann Intern Med 1981;94:149–155. 10. Fisher RF, Goodpasture HC, Peterie JD, et al. Toxic shock in menstruating women. Ann Intern Med 1981;94:156–163. 11. Chesney PJ, David JP, Purdy WK, et al. Clinical manifestations of toxic-shock syndrome. JAMA 1981;246:741–748. 12. Toxic-shock syndrome—United States, 1970–1980. MMWR 1981;30:25. 13. Reingold AL, Shands KN, Dans BB, et al. Toxic-shock syndrome not associated with menstruation. Lancet 1982;1:1–4. 14. Friedell S, Mercer LJ. Nonmenstrual toxic shock syndrome. Obstet Gynecol Surv 1986;41:336–341. 15. Centers for Disease Control. Summary of notifiable diseases—United States, 1986. MMWR 1987;35:1–57. 16. Centers for Disease Control. Reduced incidence of menstrual toxic-shock syndrome—United States, 1980–1990. MMWR 1990;39:421–23. 17. Broome CV. Epidemiology of toxic shock syndrome in the United States: overview. Rev Infect Dis 1989;11[Suppl 1]:S14–S21. 18. Kain KC, Schulzer M, Chow AW. Clinical spectrum of nonmenstrual toxic shock syndrome (TSS): comparison with menstrual TSS by multivariate discriminant analysis. Clin Infect Dis 1993;16:100–106. 19. Schlievert PM, Blomster DA, Kelly JA. Toxic shock syndrome Staphylococcus aureus: effect of tampons on toxic shock syndrome toxin-1 production. Obstet Gynecol 1984;64:666–671. 20. Tierno PM Jr, Hanna BA. In vitro amplification of toxic shock syndrome toxin-1 by intravaginal devices. Contraception 1985;31:185–194.
21. Bonventre PF, Weckbach L, Harth G, et al. Distribution and expression of toxic shock syndrome toxin 1 gene among Staphylococcus aureus isolates of toxic shock syndrome and non-toxic shock syndrome origin. Rev Infect Dis 1989;11:S90. 22. Kass EH, Parsonnet J. On the pathogenesis of toxic shock syndrome. Rev Infect Dis 1987;9:482–489. 23. Cone LA, Woodard DR, Schlievert PM, et al. Clinical and bacteriologic observations of a toxic-shock-like syndrome due to Streptococcus pyogenes. N Engl J Med 1987;317:146–149. 24. Hribalova V. Streptococcus pyogenes and the toxic-shock syndrome. Ann Intern Med 1988;108:772. 25. Bartter T, Dascal A, Carroll K, et al. Toxic strep syndrome. Arch Intern Med 1988;148:1421–1424. 26. Stevens DL, Tanner MH, Einship J, et al. Severe group A streptococcal infections associated with a toxic-shock-like syndrome and scarlet fever toxin A. N Engl J Med 1989;321:1–8. 27. Bergovac J, Marton E, Lisic M, et al. Group A beta-hemolytic streptococcal toxic shock-like syndrome. Pediatr Infect Dis J 1990;9:369–370. 28. Lersch C, Gain T, Siemens MV, et al. Toxic-shock-like syndrome due to severe hemolytic group A streptococcal infection. Klin Wochenschr 1990;68:523–525. 29. Whitted WR, Yeomans ER, Hankins GDV. Group A b-hemolytic streptococcus as a cause of toxic shock syndrome. A case report. J Reprod Med 1990;35:558–560. 30. Dotters DJ, Katz VL. Streptococcal toxic shock associated with septic abortion. Obstet Gynecol 1991;78:549–551. 31. Hoge CW, Schwartz B, Talkington DF, et al. The changing epidemiology of invasive group A streptococcal infections and the emergence of streptococcal toxic-shock-like syndrome. JAMA 1993;269:384–389. 32. The Working Group on Severe Streptococcal Infections. Defining the group A streptococcal toxic shock syndrome. JAMA 1993;269:390–391. 33. Forni AL, Kaplan EL, Schlievert PM, et al. Clinical and microbiological characteristics of severe Group A Streptococcus infections and streptococcal toxic shock syndrome. Clin Infect Dis 1995;21:333–340. 34. Stevens DL, Bryant AE, Hackett SP, et al. Group A Streptococcal bacteremia: the role of tumor necrosis factor in shock and organ failure. J Infect Dis 1996;173:619–626. 35. Eriksson BKG, Andersson J, Holm SE, et al. Epidemiological and clinical aspects of invasive Group A Streptococcal infections and the streptococcal toxic shock syndrome. Clin Infect Dis 1998;27;1428–1436. 36. Basma H, Norby-Teglund A, Guedez Y, et al. Risk factors in the pathogenesis of invasive Group A Streptococcal infections: role of protective humoral immunity. Infect Immun 1999;67:1871–1877. 37. Browdie DA, Davis JH, Koplewitz MJ, et al. Clostridium sordelli infection. J Trauma 1975;15:515–519. 38. Hatheway CL. Toxigenic clostridia. Clin Microbiol Rev 1990;1:66–98. 39. Grimwood K, Evans GA, Govender ST, et al. Clostridium sordelli infection and toxin neutralization. Pediatr Infect Dis 1990;9:582–588. 40. Golde S, Ledger WJ. Necrotizing fasciitis in postpartum patients: a report of four cases. Obstet Gynecol 1977;50:670–673. 41. Ewing TL, Smale LE, Elliott FA. Maternal deaths associated with postpartum vulvar edema. Am J Obstet Gynecol 1979;134:173–179. 42. Soper DE. Clostridial myonecrosis arising from an episiotomy. Obstet Gynecol 1986;68:S26–S28. 43. McGregor JA, Soper DE, Lovell G, et al. Maternal deaths associated with Clostridium sordellii infection. Am J Obstet Gynecol 1989;161:987–995. 44. Sosolik RC, Savage BA, Vaccarello L. Clostridium sordelli toxic shock syndrome: a case report and review of the literature. Infect Dis Obstet Gynecol 1996;4:31–35. 45. Toxic shock syndrome—United States, 1970—1982. MMWR 1982;31:201. 46. Follow-up on toxic-shock syndrome. MMWR 1980;29:441. 47. Wager GP. Toxic-shock syndrome: a review. Am J Obstet Gynecol 1983;146:93–102. 48. Toxic-shock syndrome—Utah. MMWR 1980;29:495. 49. Gaventa S, Reingold AL, Hightower AW et al. Active surveillance for toxic shock syndrome in the United States 1986. Rev Infect Dis 1989;12[Suppl S1]:S35–S42. 50. Osterholm MT, Forfang JC. Toxic-shock syndrome in Minnesota: results of an active-passive surveillance system. J Infect Dis 1982;145:458–464. 51. Latham RH, Kehrberg MW, Jacobson JA, et al. Toxic shock syndrome in Utah: a case-control and
surveillance study. Ann Intern Med 1982;96:906–908. 52. Reingold AL, Hargrett NT, Shands KN, et al. Toxic-shock surveillance in the United States, 1980–1981. Ann Intern Med 1982;96:875–880. 53. Davis JP, Vergeront JM. The effect of publicity on the reporting of toxic-shock syndrome in Wisconsin. J Infect Dis 1982;145:449–457. 54. Berkley SF, Hightower AW, Broome CV, et al. The relationship of tampon characteristics to menstrual toxic shock syndrome. JAMA 1987;258:917–920. 55. Duff P. Recognizing and treating toxic shock. Contemp Obstet Gynecol 1983;22:43–60. 56. Graham DR, O'Brien M, Hayes JM, et al. Postoperative toxic shock syndrome. Clin Infect Dis 1995;20:895–859. 57. Dornan KJ, Thompson DM, Conn AR, et al. Toxic shock syndrome in the postoperative patient. Surg Gynecol Obstet 1982;154:65–68. 58. Bartlett P, Reingold AL, Graham DR, et al. Toxic shock syndrome associated with surgical wound infections. JAMA 1982;247:1448–1450. 59. Fisher CJ, Horowitz BZ, Nolan SM. The clinical spectrum of toxic-shock syndrome. West J Med 1981;135:175–182. 60. Green SL, LaPeter KS. Evidence for postpartum toxic-shock syndrome in a mother-infant pair. Am J Med 1982;72:169–172. 61. Chow AW, Wittman BK, Bartlett BA, et al. Variant postpartum toxic shock syndrome with probable intrapartum transmission to the neonate. Am J Obstet Gynecol 1984;148:1074–1079. 62. Bracero L, Bowe E. Postpartum toxic-shock syndrome. Am J Obstet Gynecol 1982;143:478–479. 63. Tofte RW, Williams DN. Toxic-shock syndrome: evidence of a broad clinical spectrum. JAMA 1981;246:2163–2167. 64. Schlossberg D. A possible pathogenesis for recurrent toxic shock syndrome. Am J Obstet Gynecol 1981;41:348–349. 65. Faich G, Pearson K, Fleming D, et al. Toxic shock syndrome and the vaginal contraceptive sponge. JAMA 1986;255:216–218. 66. MacDonald KL, Osterholm MT, Hedberg CW, et al. Toxic shock syndrome. A newly recognized complication of influenza and influenzalike illness. JAMA 1987;257:1053–1058. 67. Langmuir AD, Worthen TD, Solomon J, et al. The Thucydides syndrome: a new hypothesis for the cause of the plague of Athens. N Engl J Med 1985;313:1027–1030. 68. Buehler EA, Dillon WP, Cumbo TJ, et al. Prolonged use of a diaphragm and TSS. Fertil Steril 1982;38:248–250. 69. De Young PD, Martyn J, Wass H, et al. TSS associated with a contraceptive diaphragm. Can Med Assoc J 1982;127:611. 70. Hyde L. TSS associated with diaphragm use. J Fam Pract 1983;16:616–618. 71. Stevens D. Streptococcal toxic shock syndrome: spectrum of disease, pathogenesis, and new concepts in treatment. Emerg Infect Dis 1995;1:69. 72. Ispahani P, Donald FE, Aveline AJD. Streptococcus pyogenes bacteremia: an old enemy subdued, but not defeated. J Infect 1988;16:37–46. 73. Jewett JF. Coagulopathy syndrome due to streptococci. N Engl J Med 1973;189:3–4. 74. Ledger WJ, Headington JT. Group A b-hemolytic streptococcus. An important cause of serious infection in obstetrics and gynecology. Obstet Gynecol 1972;39:374–382. 75. Kehrberg MW, Latham RH, Haslam BT, et al. Risk factors for staphylococcal toxic shock syndrome. Am J Epidemiol 1981;114:873–879. 76. Schlech WF, Shands KN, Reingold AL, et al. Risk factors for development of toxic shock syndrome: association with a tampon brand. JAMA 1982;248:835–839. 77. Osterholm MT, Davis JP, Gibson RW, et al. Tristate toxic-shock syndrome study. I. Epidemiologic findings. J Infect Dis 1982;145:431–440. 78. Helgerson SD, Foster LR. Toxic shock syndrome in Oregon. Ann Intern Med 1982;96:909–911. 79. Tierno PM. Cellulase activity of microorganisms on carboxymethylcellulose from tampons. Lancet 1981;2:746. 80. Hanna BA. Microbial degradation of carboxymethylcellulose from tampons. Lancet 1982;1:279. 81. Reingold AL, Broome CV, Gaventa S, et al. Toxic Shock Study Group. Risk factors for menstrual toxic shock syndrome: results of a multivariate case-control study. Rev Infect Dis 1989;11[Suppl 1]:S35–S42. 82. Tierno PM, Hanna BA. Propensity of tampons and barrier contraceptives to amplify Staphylococcus aureus toxic shock syndrome toxin-1. Infect Dis Obstet Gynecol 1994;2:140–145.
83. Schlievert PM. Comparison of cotton and cotton/rayon tampons for effect on production of toxic shock syndrome toxin. J Infect Dis 1995;172:1112–1114. 84. Parsonnet J, Modern PA, Giacobbe KD. Effect for tampon composition on production of toxic shock syndrome toxin-1 by Staphylococcus aureus in vitro. J Infect Dis 1996;173:98–103. 85. Crass BA, Bergdoll MS. Toxin involvement in toxic shock syndrome. J Infect Dis 1986;153:918–926. 86. Gorbach SL, Menda KB, Thadepalli H, et al. Anaerobic microflora of the cervix in healthy women. Am J Obstet Gynecol 1973;117:1053–1055. 87. Ohm MJ, Galask RP. Bacterial flora of the cervix from 100 prehysterectomy patients. Am J Obstet Gynecol 1975;122:683–687. 88. Grossman JH III, Adams RL. Vaginal flora in women undergoing hysterectomy with antibiotic prophylaxis. Obstet Gynecol 1979;53:23–26. 89. Bartlett JG, Onderdonk AB, Drude E, et al. Quantitative bacteriology of the vaginal flora. J Infect Dis 1977;136:271–277. 90. Tasjian JH, Coulan CB, Washington JA III. Vaginal flora in asymptomatic women. Mayo Clin Proc 1976;51:557–561. 91. Osborne NG, Wright RC. Effect of a preoperative scrub on the bacterial flora of the endocervix and vagina. Obstet Gynecol 1977;50:148–151. 92. Guinan ME, Dan BB, Guidott RJ, et al. Vaginal colonization with Staphylococcus aureus in healthy women. Ann Intern Med 1982;96:944–947. 93. Noble VS, Jacobson JA, Smith CB. The effect of menses and use of catamenial products on cervical carriage of Staphylococcus aureus. Am J Obstet Gynecol 1982;144:186–189. 94. Saunders CC, Saunders WE Jr, Fagnant JE. Toxic shock syndrome: an ecologic imbalance within the genital microflora of women? Am J Obstet Gynecol 1982;142:977–982. 95. Schlievert PM, Shands KN, Dan BB, et al. Identification and characterization of an exotoxin from S. aureus associated with toxic-shock syndrome. J Infect Dis 1981;143:509–516. 96. Altemeier WA, Lewis S, Schlievert PM, et al. Studies of the staphylococcal causation of toxic shock syndrome. Surg Gynecol Obstet 1981;153:481–485. 97. Bergdoll MS, Crass BA, Reiser RF, et al. A new staphylococcal enterotoxin, enterotoxin F, associated with toxic-shock syndrome, S. aureus isolates. Lancet 1981;1:1017–1021. 98. Bennett JV. Toxins and toxic-shock syndrome. J Infect Dis 1981;143:631–632. 99. Larsen B, Schlievert PM. TSS: the mystery unfolds. Contemp Obstet Gynecol 1981;18:219–229. 100. Kreiswirth BN, Lofdahl S, Betley MJ, et al. The toxic shock syndrome gene is not detectably transmitted by a prophage. Nature 1983;305:709–712. 101. Bergdoll MS, Schlievert PM. Toxic shock syndrome toxin. Lancet 1984;2:691. 102. Quimby FW, Nguyen HT. Animal studies of toxic shock syndrome. CRC Crit Rev Microbiol 1985;12:1–44. 103. Schlievert PM. Enhancement of host susceptibility to lethal endotoxin shock by staphylococcal pyrogenic exotoxin type C. Infect Immun 1982;36:123–128. 104. Ikejima T, Dinarello CA, Gill DM, et al. Induction of human interleukin-1 by a product of Staphylococcus aureus associated with toxic shock syndrome. J Clin Invest 1984;73:1312–1320. 105. Parsonnet J, Hickman RK, Eardley DD, et al. Induction of human interleukin-1 by toxic shock syndrome toxin-1. J Infect Dis 1985;151:514–522. 106. Beezhold DH, Best GK, Boventre PF, et al. Synergistic induction of interleukin-1 by endotoxin and toxic shock syndrome toxin-1 using rat macrophages. Infect Immun 1987;55:2865–2869. 107. Fast DJ, Schlievert PM, Nelson RD. Nonpurulent response to toxic shock syndrome toxin-1–producing Staphylococcus aureus: relationship to toxin-stimulated production of tumor necrosis factor. J Immunol 1988;140:949–953. 108. Fast DJ, Schlievert PM, Nelson RD. Toxic-shock syndrome-associated staphylococcal pyrogenic toxins are potent inducers of tumor necrosis factor production. Infect Immun 1989;57:291–294. 109. Buyalos RP, Rutanen E-M, Tsui E, et al. Release of tumor necrosis factor alpha by human peritoneal macrophages in response to toxic-shock syndrome toxin-1. Obstet Gynecol 1991;78:182–186. 110. Murray DL, Ohlendorf DH, Schlievert PM. Staphylococcal and streptococcal superantigens: their role in human diseases. ASM News 1995;61:229–235. 111. Wagner G, Bohr L, Wagner P, et al. Tampon induced changes in vaginal oxygen and carbon dioxide tensions. Am J Obstet Gynecol 1984;148:147–150. 112. Mills JT, Parsonnet J, Tsai Y-C, et al. Control of production of toxic-shock syndrome toxin-1
(TSST-1) by magnesium ion. J Infect Dis 1985;151:1158–1161. 113. Kiyota H, Kendrick MI, Kass EH. Influence of magnesium concentration on production of exoprotein and b-lactamase by Staphylococcus aureus and Staphylococcus hemolyticus. J Infect Dis 1989;160:1061–1063. 114. Mills JT, Dodel AW, Kass EH. Regulation of staphylococcal toxic shock syndrome toxin-1 and total exoprotein production by magnesium ion. Infect Immun 1986;53:663–670. 115. Kass EH. Magnesium and the pathogenesis of toxic shock syndrome. Rev Infect Dis 1989;11[Suppl 1]:S167–S175. 116. Low DE, Schwartz B, McGear A. The reemergence of severe group A streptococcal disease: an evolutionary perspective. In: Scheld WM, Armston D, Hughes JM, eds Emerging infections I. Washington, DC: American Society of Microbiology, 1997:93–123. 117. Chelsom JO, Halstensen A, Haga T, et al. Necrotizing fasciitis due to group A streptococci in western Norway: incidence and clinical features. Lancet 1994;344:1111–1115. 118. Bisno A, Stevens D. Streptococcal infections of skin and soft tissues. N Engl J Med 1996;334:240–245. 119. Kaul R, McGeer A, Low DE, et al. Population-based surveillance for group a streptococcal necrotizing fasciitis: clinical features, prognostic indicators, and microbiologic analysis of seventy-seven cases. Ontario Group A Streptococcal Study. Am J Med 1997;103:18–24. 120. Sullivan GW, Mandell GL. The role of cytokines in infection. Curr Opin Infect Dis 1991;41;344–349. 121. Koth M. Bacterial exotoxins as superantigens. Clin Microbiol Rev 1995;8:411–426. 122. Hackett SP, Stevens DL. Streptococcal toxic shock syndrome: synthesis of tumor necrosis factor and interleukin-1a by monocytes stimulated with pyrogenic exotoxin A and streptolysin O. J Infect Dis 1992:165:879–885. 123. Stevens DL, Bryant AE, Hackett SP, et al. Group A streptococcal bacteremia: the role of tumor necrosis factor in shock and organ failure. J Infect Dis 1996;173:619–626. 124. Holm SE, Norby A, Bergholm AM, et al. Aspects of pathogenesis of serious group A streptococcal infections in Sweden, 1988–89. J Infect Dis 1992;166:31–37. 125. Stevens D. Invasive group A streptococcus infections. Clin Infect Dis 1992;14:2–13. 126. Ferrieri P. Microbiological features of current virulent strains of Group A streptococci. Pediatr Infect Dis J 1991;10[Suppl]:S20–S24. 127. Hall IC. The occurrence of Bacillus sordelli in icterohemoglobinuria of cattle in Nevada. J Infect Dis 1929;45:156–162. 128. Arseculeratne SN, Panabokke RG, Wijesunder S. The toxins responsible for the lesions of Clostridium sordelli gas gangrene. J Med Microbiol 1969;2:37–53. 129. Sperber SJ, Blevins DD, Francis JB. Hypercalcitoninemia, hypocalcemia and toxic shock syndrome. Rev Infect Dis 1990;12:736–739. 130. Sahs AL, Helms CM, DuBois C. Complication of toxic shock syndrome. Arch Neurol 1983;40:414–415. 131. Rosene KA, Copass MK, Kastner LS, et al. Persistent neuropsychological sequelae of toxic shock syndrome. Ann Intern Med 1982;96:865–870. 132. Chesney PJ, Crass BA, Polyak MB, et al. Toxic-shock syndrome: management and long-term sequelae. Ann Intern Med 1982;96:847–851. 133. Efstratiou A, George RC, Holm SE et al. Report on informal consultation of directors of WHO collaborating centers on streptococci. WHO/CDS/BVI/95.6. Geneva: World Health Organization, 1996. 134. Stevens DL, Tanner MH, Winship J, et al. Reappearance of scarlet fever toxin A among streptococci in the Rocky Mountain West: severe group A streptococcal infection associated with a toxic-shock-like syndrome. N Engl J Med 1989;321:1. 135. Demers B, Simor AE, Vellend H, et al. Severe invasive group A streptococcal infection in Ontario, Canada: 1987–1991. Clin Infect Dis 1993;16:792. 136. Stevens DL. Could nonsteroidal antiinflammatory drugs (NSAIDs) enhance the progression of bacterial infections to toxic shock syndrome. Clin Infect Dis 1995;21:977–980. 137. Shubin H, Weil MH, Carson RW. Bacterial shock. Am Heart J 1977;94:112–114. 138. McSwain NE. Pneumonitic trousers and the management of shock. J Trauma 1977;17:719–724. 139. Tiengo M. Naloxone in irreversible shock. Lancet 1980;2:690. 140. Peters WP, Friedman PA, Johnson MW, et al. Pressor effect of naloxone in septic shock. Lancet 1981;1:529–532. 141. Melish ME, Frogner K, Hirata S, et al. Use of IVIG for therapy in the rabbit model of TSS. Clin Res
1987;35:220A(abst). 142. Weiss KA, Laverdiere M. Group A streptococcus invasive infections: a review. Can J Surg 1997;40;18–24. 143. Holm SE, Kohler W, Kaplan EL et al. Streptococcal toxic shock syndrome: an update. A round table presentation. Adv Exp Med Biol 1997;418:193–199. 144. Chiu CH, Lin TY, Gung C, et al. Successful treatment of severe/streptococcal toxic shock syndrome with a combination of intravenous immunoglobulins, dexamethasone, and antibiotics. Infection 1997;25:47–48.
12 INFECTIOUS VULVOVAGINITIS Infectious Diseases of the Female Genital Tract
12 INFECTIOUS VULVOVAGINITIS Prevalence and Distribution of Vaginitis General Diagnostic Approach to Infectious Vulvovaginitis Trichomoniasis Yeast Infection Bacterial Vaginosis Desquamative Inflammatory Vaginitis Chapter References
Symptoms of vulvovaginitis, which include discharge, odor, itching, and vaginal discomfort, account for a considerable proportion of outpatient gynecologic visits. Although diagnosis and therapy of vulvovaginitis is straightforward in most cases, many patients experience persistence or recurrence, and still others who have less distinctive types of vaginitis do not receive satisfactory therapy. Developments in the last 5 years have included improved knowledge of the complications associated with vaginitis (upper genital tract infection and adverse pregnancy complications), new diagnostic techniques, and refinements of therapy.
PREVALENCE AND DISTRIBUTION OF VAGINITIS In their classic paper, Gardner and Dukes (1) analyzed 1,181 private patients with symptoms of vaginitis or with gross evidence of vaginitis on inspection. They described normal vaginal secretions as usually white and curdy (i.e., not homogeneous), with a pH of 4.0 to 4.7. Their findings are summarized in Table 12.1. The most common type of vaginitis was what they called “nonspecific vaginitis” (now called bacterial vaginosis [BV]). Trichomoniasis and candidiasis each accounted for about one fourth of cases.
TABLE 12.1. ETIOLOGY OF INFECTIOUS VAGINITIDES IN 1,181 PATIENTS
Vaginal symptoms occur commonly. Trichomoniasis, moniliasis, and BV account for the vast majority of cases, but some cases (1%–10%) cannot be accounted for by these three diagnoses. Further, the prevalence of vulvovaginal organisms varies from population to population. Asymptomatic women commonly will harbor yeasts, Trichomonas vaginalis or Gardnerella vaginalis.
GENERAL DIAGNOSTIC APPROACH TO INFECTIOUS VULVOVAGINITIS Most cases of infectious vulvovaginitis can be diagnosed with a few inexpensive tests, provided the tests are used and interpreted properly. In all cases, the physical examination should note the following: vulvar erythema; edema; excoriation or lesions; and color, amount, and consistency of the vaginal discharge. Normal vaginal secretions are heterogenous. The cervix should be examined for eversion, friability, and color of mucus. For patients with vaginal discharge or irritation, the following tests are indicated: pH (using a paper that will distinguish pH values in the range from 4.0 to 5.0 (Fig. 12.1)
FIGURE 12.1. pH paper used in the differential diagnosis of vulvovaginitis. The pH range is from 4.0 to 7.0.
Amine (“fishy”) odor by adding a drop of discharge to a drop of 10% potassium hydroxide (KOH) solution
Saline preparation for true clue cells and trichomonads KOH preparation for hyphae and pseudomycelia of yeasts Culture for yeasts, if the other tests are nondiagnostic Saline preparation of an endocervical smear (to detect microscopic cervicitis) True “clue cells” (Fig. 12.2) are so heavily stippled with adherent bacteria as to obscure the cell margin. Normal epithelial cells commonly have some adherent bacteria (Fig. 12.3).
FIGURE 12.2. Vaginal Gram stain showing a true clue cell and abnormal bacteria characteristic of bacterial vaginosis.
FIGURE 12.3. Gram stain showing normal vaginal flora with a predominance of Lactobacillus morphotypes.
Culturing for G. vaginalis is not indicated. Although 95% of women with BV are positive
(when selective media are used), so are 40% of women with normal secretions. The Gram stain may be of help in diagnosing vaginitis, as the variety and numbers of bacteria can be readily identified, but this technique is infrequently used clinically (see section on bacterial vaginosis). Diagnostic criteria are given in Table 12.2. Approximately 8% to 10% of patients with abnormal vulvovaginal symptoms will not be diagnosed by the tests described. In these patients, cervicitis due to gonorrhea or chlamydia should be excluded. Cervicitis due to gonorrhea or chlamydia should be excluded by specific testing. In addition, a saline preparation of an endocervical smear should be examined microscopically to detect evidence of cervicitis, which often is nongonococcal and nonchlamydial in origin. Although the etiology of this cervicitis is not clearly elucidated, empiric treatment with a quinolone, erythromycin, doxycycline, or azithromycin is reasonable in patients with persistent symptoms. Because of the limited sensitivity of the wet mount for trichomonads (estimated at 50%–80%), additional diagnostic tests have been developed. Media for culturing trichomoniasis are available commercially, as Diamond or Trichosel media or as the InPouch TV system (see section on trichomoniasis). Some women presenting with symptoms of vulvar burning and irritation may be characterized as having vulvodynia or, alternatively, vulvar vestibulitis. This condition most likely is neuropathic rather than infectious in origin. This entity has a fairly characteristic presentation, marked by chronic vulvar (rather than vaginal) symptoms, point tenderness along the vaginal vestibule, and perhaps erythema in the vestibular glands at the five and seven o'clock positions (i.e., posteriorly at the introitus, just to the right and left of the midline).
TABLE 12.2. CHARACTERISTICS OF NORMAL SECRETIONS AND VAGINITIS
TRICHOMONIASIS Trichomoniasis is caused by T. vaginalis, a protozoan first described in 1836. The organism has been responsible for approximately one fourth of all cases of clinically
evident vaginal infections (1), but its prevalence appears to be decreasing lately (2). Previous estimates placed the prevalence at 3% to 5%, with a range of 2% to 3% in middle-class women to 56% in women attending sexually transmitted disease (STD) clinics (3). Approximately 50% of women with T. vaginalis are asymptomatic, but about 30% of these asymptomatic women develop symptoms when they are observed for 6 months (3). Among symptomatic women with trichomoniasis, the vaginal secretions usually are copious, homogeneous, and malodorous, with a pH above 4.5. A frothy, yellow-green discharge often is cited as the typical finding, but, in objective series, frothiness was detected in only 12% to 34% (1). Gardner and Dukes (1) described the color of the discharge as gray in 46% of cases, yellow-green in 36%, and yellow-green in 10%. Punctate mucosal hemorrhages of the cervix, the so-called strawberry or “flea-bitten” cervix, are seen infrequently with the naked eye but are seen frequently with colposcopy. Because of the variations in signs and symptoms, one cannot rely on these findings. Under most circumstances, the clinical diagnosis can be confirmed by microscopic examination of a wet mount, made by mixing a drop of secretions with saline on a slide. Because trichomoniasis may produce a heavy polymorphonuclear infiltrate, it is easy to miss the trichomonads. One must examine the preparation in an area with relatively few white cells. Usually, the trichomonads are evident because of the extreme activity of these flagellates in freshly made preparations. A helpful clue for microscopic diagnosis of trichomoniasis is using fresh saline (intended for intravenous use) with no preservative. Most reports have found the sensitivity of culture to be greater than that of wet mounts (4). The sensitivity of clinical microscopy varies with the experience and thoroughness of the examiner (3). Overall, sensitivity varied from as low as 42% to as high as 92%. When experienced technicians perform the test, the sensitivity runs at about 80%, compared with culture; when busy clinicians perform it, the sensitivity drops to about 50%. Culturing currently is considered the best method for making the diagnosis. Culture for T. vaginalis is not difficult, but special media, such as Diamond or Kupferberg medium, is needed. A drop or swabful of the vaginal secretions is placed in the broth, which is incubated. A drop of the media is examined daily under the microscope for 5 to 7 days for appearance of the motile trichomonads. Because there is limited demand for T. vaginalis cultures, few clinical laboratories have this available, even though the media are commercially available. In a comparison of a commercially available kit system (InPouch TV) to Diamond modified medium for culture of T. vaginalis from vaginal secretions, both culture techniques performed with similar high sensitivity (5). The diagnosis of trichomoniasis is possible on a Papanicolaou (Pap) smear, but the sensitivity of this method is modest (52%–67%). There is a low specificity, even when the smear is prepared with a vital stain such as acridine orange (6,7). New diagnostic methods using monoclonal antibodies, enzyme-linked immunosorbent assay, and latex agglutination have been introduced. Although initial reports show encouraging results, there has not been much use in the clinical setting to determine the
practical reliability of these (3,8). Several publications also have assessed a T. vaginalis polymerase chain reaction (PCR) test. Although these diagnostic tests each have used different PCR primers, each test has reported higher sensitivity with the PCR than with wet-mount microscopy, culture on Diamond or modified Kupferberg medium, or Pap smear. The PCR test also appears to be highly specific without interaction with other species of Trichomonas other than T. vaginalis and without cross-reactivity with other human parasites or other human STDs such as Chlamydia trachomatis or Neisseria gonorrhoeae. The improved sensitivity of PCR appears to apply both to women without and women with symptoms of vaginal discharge (9). One additional advantage of the PCR test is that satisfactory specimens can be obtained by self-collection from the distal vagina, which eliminates the requirement for a vaginal speculum examination (6,7,10,11 and 12) Urine sediment can be examined for trichomonads, and this is the preferred site for detection in males. The sediment can be cultured as well as inspected microscopically. The diagnosis of trichomoniasis is difficult, limited by the recognized modest sensitivity of the wet mount. For most practice situations, however, the wet mount continues to be the diagnostic technique of choice, even among experts (2). In special circumstances, culture using either Diamond medium or the InPouch Kit seems important, and PCR technology may become commercially available within the next few years. The only treatment for trichomoniasis available in the United States is metronidazole (Flagyl). Related 5-nitroimidazole derivatives (tinidazole, ornidazole) are available in other countries. In the early 1970s, evidence indicated that metronidazole was carcinogenic in certain rodent species but not in others. When tumors evolved, it was only after chronic, high-dose use, making the application of these data to humans difficult. The United States Food and Drug Administration (FDA) interpreted the data as showing a very low risk of human carcinogenesis and denied a petition to remove metronidazole from the market (13). Later, a study from the Mayo Clinic provided data on 771 women given metronidazole for trichomoniasis (14). With surveillance of 10 to 20 years, these investigators found no appreciable carcinogenicity attributable to metronidazole. Additional recent information has provided further reassurance. In a study of 1,387 women who filled prescriptions for metronidazole in pregnancy, there was no excess of overall birth defects (in their offspring), compared with 1,387 control women (risk ratio [RR], 1.2, 95% confidence interval [CI], 0.9–—1.6). The authors also were unable to find an excess risk in any category of birth defects (15). In 1976, in view of concerns about carcinogenicity, there was a reduction in standard dose of metronidazole (250 mg orally three times a day in women) from 10 to 7 days, and topical metronidazole preparations for treatment of trichomoniasis were removed from the market (13). A single 2-g dose was evaluated and found to be effective (7). Direct comparisons of the 7-day and single-dose regimens have found them to be equivalent (16). Cure rates with either regimen have varied from 86% to 97%, and side effects (mainly nausea) have not been significantly different between the two regimens. The CDC guidelines recommend single-dose treatment as the regimen of choice (17). Possible causes of failure of metronidazole therapy include (a) pharmacokinetic
problems in either absorption or delivery to the infected site, (b) inactivation of metronidazole by vaginal bacteria, (c) interference by other drugs, (d) resistance to metronidazole, (e) noncompliance or gastrointestinal intolerance, and (f) reinfection. In the past, resistance was discounted, but development of a suitable aerobic in vitro assay has permitted the demonstration of resistance in strains from Europe and the United States (18). Sexual partners should be treated. Although there are concerns raised by prescribing metronidazole to the male partner who does not come to the office or clinic, the CDC clearly recommends concurrent treatment of male partners of women with trichomoniasis (Table 12.3)(17).
TABLE 12.3. CENTERS FOR DISEASE CONTROL AND PREVENTION RECOMMENDED REGIMENS FOR TRICHOMONAS VAGINALIS VAGINITIS
Use of metronidazole occasionally is accompanied by headache, metallic or bitter aftertaste, or nausea and vomiting. Seizures and central nervous system toxicity can accompany metronidazole therapy, almost exclusively in patients on exceedingly high doses (>3 g/day). Blood dyscrasias are rare consequences, but, in the product information, total and differential leukocyte counts are recommended before and after treatment, especially when a second course is necessary. Alcohol should be avoided during treatment and for 3 days afterward. Metronidazole gel (0.75%) has become available and is indicated for treating BV. It has not been evaluated for trichomoniasis and should not be used for this infection in view of the poor efficacy of past local therapies. The FDA has approved a new preparation of metronidazole, marketed as Flagyl 375™ mg, twice daily for 7 days for treatment of trichomoniasis. This approval was based on the pharmacokinetic equivalency of this regimen with metronidazole 250 mg three times daily for 7 days. No clinical data are available to demonstrate clinical equivalency of the two regimens (17). Follow-up examination is not recommended for patients who become asymptomatic
after treatment. For treatment of trichomoniasis in pregnancy, metronidazole is recommended by the CDC in a single 2-g dose (17). Symptomatic nonpregnant women with documented trichomoniasis should be treated to prevent sexual transmission and future symptomatic infection. Symptomatic pregnant women with trichomoniasis should be treated with 2 g of metronidazole in a single dose. The CDC guidelines no longer restrict use of metronidazole to the second or third trimester. For pregnant women with asymptomatic trichomoniasis, the decision to treat is a dilemma. There is some evidence linking this infection to adverse pregnancy outcomes (see later), and sexual transmission remains a possibility. In selected circumstances, such as in patients undergoing cerclage or patients presenting in preterm labor, screening for and treatment of T. vaginalis may still be a very reasonable approach. However, a recent treatment trial conducted by the National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Unit Network concluded that metronidazole treatment increased the risk of preterm birth in asymptomatic women with trichomoniasis. In a multicenter, double-blind, placebo-controlled trial, women who were culture positive for trichomoniasis were randomized at 16 to 23 weeks of gestation to receive either metronidazole or placebo. The metronidazole regimen was 2 g given under observation and another dose 48 hours later. Sexual partners of women in both groups were given prescriptions for metronidazole. The regimen was repeated at 24 to 29 weeks of gestation. The two groups, with approximately 300 women in each group, did not differ significantly in terms of demographic characteristics. However, total preterm birth was significantly greater in the group receiving metronidazole (19% vs. 10.7%; RR, 1.8; 95% CI, 1.19–—2.66). There were significant increases in preterm births due to preterm labor (10% vs. 3.5%; RR, 2.87; 95 CI, 1.43–5.75). There were no significant differences in preterm births due to preterm premature rupture of the membranes (PROM), births at less than 32 weeks, birthweights less than 2,500 g, or birthweights less than 1,500 g. The risk of preterm birth in the metronidazole group was elevated among women with a prior preterm birth (RR, 1.93; 95 CI, 0.93–3.98; p > 0.05) (18). In view of the surprising results of this trial, treatment of asymptomatic women in pregnancy with trichomoniasis cannot be recommended. Treatment for patients with human immunodeficiency virus (HIV) infection and trichomoniasis currently is the same as the treatment for patients without HIV (18). When a patient has apparent persistence of T. vaginalis despite appropriate initial therapy, determine whether there has been compliance by both the patient and her sexual partner(s). Exclude interference by other medications such as phenytoin or phenobarbital, which induce microsomal liver enzymes and accelerate elimination of metronidazole. If these problems are excluded, presumptive treatment for a relatively resistant strain is appropriate. Relatively resistant strains have become increasingly common. Muller and colleagues (19) determined in vitro susceptibility to metronidazole in 146 isolates from patients who responded to a single 2-g oral dose. The mean aerobic minimum lethal
concentration (MLC) was 24.1 µg/mL. In comparison, the mean aerobic MLC was 195.5 µg/mL in 53 isolates from patients who had persistent infection. The corresponding mean anaerobic MLC values were 1.6 and 5.0 µg/mL (19). Lossick and colleagues (20,21) reported treatment in 53 women with persistent infection. Eighty percent of the women (27/31 from whom clinical follow-up data were available) were ultimately cured with a higher dose of metronidazole. Regimens ranged from 2 to 4 g/day for 5 to 14 days, often with vaginal insertion of a 500-mg oral tablet nightly. Doses greater than 3 g/day were accompanied by a high risk of serious side effects, including irreversible neurologic problems. Further intravenous administration appeared to have no advantage. The CDC recommendations for treatment failures are re-treatment with metronidazole 500 mg twice daily for 7 days and, if repeated failure occurs, a single 2-g dose once daily for 3 to 5 days. For patients who still have persistent infection, the CDC recommendation is to exclude reinfection, evaluate in vitro susceptibility of the isolate, and manage “in consultation with an expert” (17). Consultation is available through the CDC. We check a complete blood count before repeat high-dose treatment. Cost-effective alternatives to metronidazole are not available. Patients who have severe vaginal trichomoniasis and serious adverse reactions to metronidazole may be managed by desensitization. In a report of two patients, women were given incremental intravenous doses of metronidazole, beginning at 5 µg and building up to 125 mg stepwise at 15- to 20-minute intervals. Oral doses then were given up to 2 g orally. It is important to note that this desensitization approach was carried out after obtaining a documented history and a positive wheal test after application of metronidazole gel to the vaginal mucosa. The desensitization was performed in a monitored bed with additional precautions that included placement of two large intravenous lines and availability of a cardiopulmonary resuscitation team. In both cases, the desensitization was carried out without complications, and both patients were cured (22). One report described resolution of resistant vaginal trichomoniasis when the patient used intravaginal nonoxynol-9 (23). Prevention of trichomoniasis is achieved by using safe sexual practices, which include limiting numbers of sexual partners, using condoms, and treating sexual partners. Trichomonas vaginalis has not been linked, by itself, to genital tract cancer, infertility, abortion, or endometritis. However, as an STD, it should prompt detection of other STDs. Trichomoniasis has been linked to BV, which has been associated with several obstetric and gynecologic complications (see later). Trichomoniasis has been associated with adverse pregnancy outcomes, such as PROM and premature birth (24). One treatment trial performed to determine whether treatment of trichomoniasis in pregnancy reduces such adverse outcomes showed an increase in adverse outcomes in the treated group (18). This subject is discussed in detail in Chapter 19.
YEAST INFECTION Yeasts are commonly isolated in the female lower genital tract, with rates of 22% among asymptomatic college women, 26% among patients in an STD clinic, and 39% among women with vulvovaginal symptoms (25). It is estimated that 75% of women will have at least one episode of yeast vulvovaginitis, with 4% to 45% having two or more episodes (17). In all studies, Candida albicans was the predominant yeast isolate (in approximately 90% of cases), with Candida glabrata and other Candida sp making up the remainder. Among patients with a positive culture for C. albicans, Oriel and coworkers (25) reported vaginal pruritus with or without discharge in 50% and discharge alone in 30%. Commonly noted predisposing features to growth of yeast in the vagina are glycosuria, diabetes mellitus, pregnancy, obesity, and recent use of antibiotics, steroids, or immunosuppressants. Pregnancy is associated with an increased vaginal carriage rate, increased susceptibility to infection, and lower cure rates. The exact mechanism is uncertain, but pregnancy may act to increase glycogen in the vaginal epithelium, or pregnancy may provide a positive hormonal effect for yeasts. Antibiotics that most commonly result in yeast colonization are ampicillin, tetracycline, and cephalosporins. After 2 weeks of tetracycline use, vaginal carriage increased from 10% to 30%. Presumably, antibiotics suppress normal bacterial population and allow opportunistic colonization by yeasts. Prophylactic local antifungal agents are an appropriate measure in patients susceptible to yeast infections when they require antibiotics for other infections. In a population of women attending an urban STD clinic, risk factors for a positive genital C. albicans culture included condom use, presentation in the second half of the cycle, sexual intercourse more than four times per month, recent antibiotic use, young age, past gonococcal, infection and absence of current gonorrhea or BV (26). The role of oral contraceptives (OCs) is controversial (25). Oriel et al. (25) found that OC users were more likely to have yeast isolates in vaginal cultures (32% of 241 women taking OCs vs. 18% of women not taking OCs), but symptoms or signs of vaginal yeast infection were not increased (23% OC vs. 26% no OC). Other investigators also reported increased vaginal carriage among OC users. However, recent studies of low-dose OCs have consistently reported no increase in Candida isolation among OC users. Proposed mechanisms include increased adherence or receptivity, increased vaginal glycogen, and increased yeast virulence. It is commonly suggested that wearing tight-fitting undergarments predisposes to yeast infection by increasing local humidity and temperature. Yeasts usually are not acquired through sexual intercourse (17). Evidence supporting sexual transmission in yeast vulvovaginitis includes (a) a fourfold increase in yeast colonization in male partners of infected women and (b) isolation of the same strains in infected couples. Further, in a recent case-control study of 85 students with vulvovaginal candidiasis, Foxman (27) reported that frequent intercourse (defined as seven or more times per week) was the strongest risk factor. The odds ratio was 4.3 (95% CI,
1.4–12.9) compared with student health service controls with no intercourse (27). However, epidemiologic evidence against sexual transmission includes (a) no direct association between yeast infection and other STDs, (b) no difference in yeast isolation rates in STD versus non-STD populations, and (c) absence of any study showing that treatment of the male partner benefits the female. Routine treatment of sexual partners is not recommended, but may be considered for women with recurrent infection (17). These recommendations suggest treatment of male partners who have balanitis, as evidenced by erythematous areas of the glans with local pruritus or irritation. Topical treatment to relieve symptoms is recommended such cases. Despite much anecdotal information, purported risk factors, including wiping from back to front after using the toilet, use of feminine hygiene products, type of clothing, diet, and stress, showed no association with vulvovaginal candidiasis among college women (27). Kalo-Klein and Witkin (28) pointed out that hormonal status may influence the pathogenicity of Candida by modulation of immune system activity. They examined the influence of the phase of the menstrual cycle on the cellular immune response to C. albicans, the efficiency of C. albicans germination in sera, and the ability of products from activated lymphoid cells to inhibit germination. In the luteal phase, host responses to C. albicans were decreased. The characteristic findings of vaginal candidiasis are reddened vulval or vaginal areas with vulval scaling, edema, or excoriation, and raised, white, or yellow adherent vaginal plaques. However, Oriel et al. (25) noted such findings in only 38% of women with positive cultures for C. albicans. The diagnosis can be confirmed by observing mycelia and/or pseudohyphae upon direct microscopy in a 10% KOH preparation, but this is a test with limited sensitivity. In the study by Oriel et al. (25), only half of the patients with positive cultures showed organisms on direct microscopy, and the women with the most florid findings often gave positive results only by culture. On the other hand, Van Slyke et al. (29) found that, among symptomatic patients, the KOH wet mount was “usually sufficient for office use.” Compared with cultures, direct microscopy gave 2.3% false-positive and 6.2% false-negative results. McCormack et al. (30) helped place the role of vaginal cultures in perspective. Among 144 women, 42 were culture positive for yeasts (30). Of these patients, 25 (60%) had vulvovaginal itching or irritation. Of the 42 culture-positive women, eight also had positive KOH preparations, and 7 (88%) of these 8 women were symptomatic. In comparison, the 34 remaining culture-positive women had negative KOH preparations, and 18 (53%) of these 34 women were symptomatic. Thus, only seven of 25 culture-positive and symptomatic women had positive KOH preparations. Interestingly, of 102 culture-negative (and KOH-negative) women, 16 (16%) had itching or irritation. McCormack et al. recommended obtaining a fungal culture for patients who have symptoms and a negative KOH preparation, before excluding yeast vulvovaginitis. A vaginal yeast culture may reveal that the infecting species is a non–C. albicans variety. In the Wayne State University vaginitis clinic in Detroit, approximately 10% of all
yeasts isolated are Torulopsis glabrata (31). In Connecticut, Horowitz et al. (32) found that Candida tropicalis accounted for fully 23% of all genital fungal infections, second only to C. albicans, which accounted for 65% of isolates. In comparison to the Detroit reports, this group found Candida (Torulopsis) glabrata in less than 1% of cases. Eight other species of yeast were isolated, all infrequently, by the Connecticut group. Because these reporting groups saw referral populations, it is possible that their respective rates of non–C. albicans species were higher than those seen in most practices. Although these reports are preliminary, they suggest that infection with non–C. albicans species may be more difficult to treat and more likely to recur. The vaginal pH of women with yeast vulvovaginitis is normal (100 colonies per milliliter was sufficient for diagnosing acute cystitis. Multiple dipstick culture or nitrite test techniques have been used to simplify and decrease the cost of screening urine for bacteria, and although appropriate for diagnosing acute cystitis in nonpregnant patients, in pregnancy the gold standard remains culture (93). The reported incidence of acute cystitis in pregnancy ranges from 0.3% to 1.3% (235,236). Harris and Gilstrap (235) reported that although the increased diagnosis and treatment of ASB resulted in a decreasing incidence of pyelonephritis at their institution, the incidence of acute cystitis remained constant. On initial screening of urine cultures during prenatal care, 64% of patients who developed cystitis had negative cultures. This is in contrast to the patients with acute pyelonephritis, in which only a minority had negative initial screening cultures. These authors noted that the recurrence pattern of patients with acute cystitis differed from that of patients with either bacteriuria or acute pyelonephritis. Whereas 75% of patients with acute pyelonephritis developed a recurrence if not given suppressive antimicrobial therapy, and one third of patients with ASB subsequently developed positive urine cultures as evidence of recurrence, only 17% of patients with acute cystitis developed subsequent positive cultures (235). They reported that only 6% of patients with acute cystitis had a positive fluorescent antibody test result, suggesting upper urinary tract involvement; this is in comparison to the nearly 50% of patients with ASB and the two thirds of the patients with acute pyelonephritis who had positive fluorescent antibody test results and, presumably, renal bacteriuria. This low incidence of renal involvement may well explain the decreased likelihood for recurrence among patients with acute cystitis. The cases of acute cystitis tended to occur in the second trimester (235). This is in contrast to reports that put most cases of acute pyelonephritis in the first and third trimester and almost all cases of ASB in the first trimester. The only morbidity associated with acute cystitis in pregnancy is the discomfort present with symptomatic UTIs (93). No studies, to date, have demonstrated that cystitis increases the risk for preterm birth, low birthweight, or acute pyelonephritis (93). In fact, Harris and Gilstrap (235) did not have a single case of acute pyelonephritis after acute cystitis in their series.
Pregnant patients with acute cystitis should begin immediate therapy. The antimicrobial agents recommended for treatment of acute cystitis are similar to those for ASB. The recommended dosage schedules for these agents are listed in Table 15.24. The new fluoroquinolone agents are very effective agents for UTIs but are not approved for use in pregnancy. The most commonly isolated microorganisms from the urine in patients with acute cystitis have been E. coli and other Gram-negative facultative organisms, such as Klebsiella, Proteus, or Enterobacter; less commonly, group B streptococcus and S. saprophyticus. Unlike in nonpregnant women with acute cystitis, in pregnancy, a culture should be obtained before instituting empiric therapy. Either a catheterized specimen or a clean-catch midstream urine should be obtained, before the institution of antibiotics, for urinalysis and urinary culture and sensitivities. The duration of therapy of cystitis is 3 days. Because of the symptomatology associated with acute cystitis, it is not possible to wait for the results of the culture; therefore, the constellation of symptoms and a urinalysis (or dipstick test, leukocyte esterase and nitrite) revealing white blood cells and bacteria should be sufficient to initiate therapy. The physician must obtain follow-up culture for examination of the urine to determine the effectiveness of therapy. Acute Pyelonephritis In Pregnancy Epidemiology Acute pyelonephritis is one of the most frequent medical complications of pregnancy. The overall incidence is reported to be between 1% to 2.5% of all obstetric patients, with an estimated recurrence rate of 10% to 18% during the same gestation (8,93). The incidence varies depending on the population, prevalence of ASB, and whether screening for and treatment of ASB in pregnant patients occurs. Predisposing factors for acute pyelonephritis during pregnancy include obstructive and neurologic diseases of the urinary tract, ureteral or renal calculi, and previous episodes of pyelonephritis and ASB of pregnancy, which is the major predisposing factor. As screening for and treatment of ASB increases, the incidence of pyelonephritis decreases (94,213,214). Diagnosis Diagnosis of pyelonephritis is based on a history of shaking chills, fever, and flank pain, symptoms and signs of nausea and vomiting, frequency, urgency, and dysuria, physical findings showing costovertebral angle tenderness and fever, and laboratory reports reveling pyuria and bacteriuria. White blood cell casts on urinalysis are strong support for the diagnosis. Leukocytosis is often present. The diagnosis is confirmed by a positive urine culture. Pregnant women with pyelonephritis generally present with clear-cut signs and symptoms that allow one to easily make the diagnosis (219). Most women will have chills and documented fever and will complain of back pain (85%). A significant number (40%) of patients have symptoms of lower UTI, such as dysuria and frequency. Almost one of four has nausea and vomiting. Fever is universal, and the diagnosis should be suspect if this is not present. Although blood cultures are positive in 10% to 15% of pregnant women with acute pyelonephritis, their usefulness in
uncomplicated acute pyelonephritis has been questioned (93,237). MacMillan and Grimes (237) noted that blood cultures are expensive, the organism isolated is invariably the same as that recovered from urine, and change of antibiotics is usually based on lack of clinical response, not culture results. Complications Pyelonephritis poses a serious threat to maternal and fetal well-being. Maternal complications are primarily the result of bacterial endotoxin-induced tissue damage (93). Cox and Cunningham (238) noted that up to one fourth of pregnant women with severe pyelonephritis have evidence of multisystem derangement. Hemodynamic changes due to endotoxemia and hypovolemia secondary to dehydration result in hypotension. Although bacteremia occurs in approximately 10% to 15% of pregnant women with severe pyelonephritis, full-blown septic shock is rare. However, hypotension and decreased perfusion secondary to endotoxemia may not respond to fluid resuscitation alone, as well as hypotension secondary to dehydration from fever and vomiting. Such patients will present with the full-blown septic shock syndrome (238). Moderate to severe anemia can occur as the result of hemolysis initiated by endotoxemia (239,240). Anemia has been noted in 25% to 66% of pregnant patients with pyelonephritis (206,241). Mild evidence of disseminated intravascular coagulation (thrombocytopenia and elevated fibrin split products) may be present. Approximately 25% of pregnant women with pyelonephritis have renal dysfunction as documented by decreased creatinine clearance (242). The renal dysfunction associated with acute pyelonephritis is transient and clears over a few days (93). If the initial creatinine clearance is elevated, fluid resuscitation should be initiated and serial serum creatinine clearance tested. In addition, antibiotics with potential nephrotoxicity should be withheld or administered in reduced dosages and those excreted via the kidney administered in reduced dosages (93). Approximately 1% to 2% of pregnant women with severe pyelonephritis develop respiratory insufficiency (238,243). The pulmonary injury is analogous to adult respiratory distress syndrome (ARDS) (shock lung) and is the result of endotoxin producing altered alveolocapillary membrane permeability, leading to pulmonary edema. Thus, pregnant women with acute pyelonephritis should be monitored closely for dyspnea and tachypnea. In most cases, these pulmonary effects are transient (238). Cunningham et al. (244) reported that evidence of pulmonary injury and respiratory insufficiency occurred in 2% of women with severe antepartum pyelonephritis. Towers et al. (245) noted an incidence of 8% if tocolysis was given to suppress uterine contractions. ARDS is life threatening, so close attention must be paid to the respiratory rate and other evidence of respiratory compromise or failure. The presence of tachypnea or dyspnea requires prompt chest x-ray and arterial blood gas analysis. Prompt recognition and appropriate intervention, including intubation and mechanical ventilation, are necessary to prevent severe hypoxemia resulting in fetal death or preterm delivery. Towers et al. (245) compared 11 patients with pyelonephritis and pulmonary injury with 119 patients with pyelonephritis only. The presence of maternal tachycardia (more than 110 beats per minute) and a fever of 103°F 12 to 24 hours
before the occurrence of respiratory symptoms in a gestation of more than 20 weeks was highly predictive of pulmonary injury (245). In this study, the most significant predictive factors associated with pulmonary injury were aspects of treatment including fluid overload, use of tocolytic agents, and choice of antibiotic (e.g., ampicillin). As discussed already, the major fetal complications associated with acute antepartum pyelonephritis are preterm birth and low birthweight (3,13,18,21,22,162,199,200,201,202,203,204,205,206 and 207). Aggressive antimicrobial therapy and judicious tocolytic therapy may decrease this risk. However, the optimum approach is prevention of acute pyelonephritis by screening for and treating ASB in pregnant women (213,214 and 215). Treatment Obstetric patients with acute pyelonephritis require hospitalization for vigorous treatment with intravenous fluids and parenteral antimicrobial agents and close monitoring of renal function and respiratory status (Table 15.25). Nausea, vomiting, anorexia, and pyrexia frequently result in severe dehydration. An indwelling Foley catheter should be placed to obtain urine for microscopic examination and culture, as well as to closely monitor urine output. (Once the patient is stable and has adequate urine output, the Foley catheter can be discontinued.) Intravenous fluids are instituted promptly, and fluid replacement with isotonic crystalloid solutions should be given until adequate urine flow is established. Either Ringer's lactate or normal saline is used, and 2,000 mL often must be administered before adequate hydration is evident (i.e., improved urine output). If septicemia with shock is present, the amount of fluids required to restore isovolume may be considerably larger.
TABLE 15.25. MANAGEMENT OF PREGNANT WOMEN WITH ACUTE PYELONEPHRITIS
As noted already, approximately 25% of pregnant women with pyelonephritis have transient renal dysfunction, as documented by decreased creatinine clearance (242).
This factor is particularly important when the antimicrobial agents used are nephrotoxic, eliminated by the kidney, or both. The initial choice of antimicrobial agent for the treatment of acute pyelonephritis is empiric (Table 15.26). This choice should be based on knowledge of the common etiologic agents and recognition of the need for bactericidal agents. The most common isolates recovered from pregnant women with pyelonephritis are E. coli (80% to 85%), rather than organisms of the Klebsiella-Enterobacter group and Proteus sp (8,93). Both the enterococci and the group B streptococci are increasingly recognized as pathogens in acute pyelonephritis of pregnancy. In the past, empiric therapy with ampicillin (1 to 2 g intravenously every 6 hours) or cephalothin (in similar doses) was sufficient. Because of increasing resistance of E. coli to ampicillin, this agent is no longer recommended as single-agent therapy of acute pyelonephritis in pregnancy. In many geographic areas, high-level resistance to first-generation cephalosporins has emerged. As an example, Millar et al. (247) noted that 12% of E. coli isolates from patients with pyelonephritis were resistant to cefazolin. However, Dunlow and Duff (246) reported that cefazolin was very effective for treating acute pyelonephritis. More recently, we have recommended that empiric therapy for acute pyelonephritis during pregnancy be instituted using ceftriaxone (1 to 2 g intravenously as a single daily dose). This third-generation cephalosporin provides coverage against the major uropathogens (except Enterococcus) and can be administered once daily, which facilitates home parenteral therapy and lowers the cost. In a prospective randomized trial, Sanchez-Ramos et al. (248) demonstrated that daily single-dose intravenous ceftriaxone was as effective as multiple-dose cefazolin in the treatment of acute pyelonephritis in pregnancy.
TABLE 15.26. TREATMENT OF ACUTE PYELONEPHRITIS IN PREGNANT WOMEN
A combination of either ampicillin or cefazolin with gentamicin is an alternative choice. These combinations are appropriate for the following reasons: (a) There has been a steadily increasing pattern of resistance by E. coli to ampicillin and first-generation cephalosporins; (b) women with recurrent infections (or occasionally initial infections) are more likely to have microorganisms resistant to ampicillin or first-generation
cephalosporins (e.g., Enterobacter or Klebsiella); and (c) the high probability for bacteremia or endotoxin sepsis argues for inclusion of a bactericidal agent that is effective against a wide range of Gram-negative facultative bacteria including resistant organisms such as Klebsiella, Enterobacter, Pseudomonas, or Serratia. Because either group B streptococci or enterococci are responsible for a small percentage of acute pyelonephritis in pregnancy, we prefer the ampicillin-gentamicin combination, which covers these two pathogens as well. As in nonpregnant patients, trimethoprim-sulfamethoxazole (160/800 mg intravenous every 12 hours) is another effective antimicrobial regimen for treating pyelonephritis in pregnancy. Recently, Wing et al. (249) conducted a randomized prospective trial of the treatment of pyelonephritis in pregnancy comparing three antibiotic regimens: (a) intravenous ampicillin and gentamicin; (b) intravenous cefazolin; and (c) intramuscular ceftriaxone. There were no statistically significant differences in clinical response or birth outcomes among the three regimens for treating acute pyelonephritis in pregnancy before 24 weeks of gestation. Intravenous antibiotics are continued until the patient is afebrile for 24 to 48 hours and can tolerate oral medications. With prompt treatment, a rapid clinical response occurs and most patients are afebrile and asymptomatic within 48 hours. Monitoring of serum levels of gentamicin is crucial in pregnancy, in which the associated increase in vascular volume and glomerular filtration rate results in a high prevalence of subtherapeutic serum levels with aminoglycosides (250). Peak levels of gentamicin should be less than 10 µg/mL and the trough level should be less than 2 µg/mL. We recommend obtaining peak and trough serum levels approximately 24 hours after initiation of therapy, and if these levels, together with the serum creatinine clearance, are within the reference range, one should repeat these evaluations every 3 days. After intravenous antimicrobial therapy, 85% of patients become afebrile within 48 hours, and nearly 100% of patients do so within 4 days (219). If this does not occur and the patient continues to be febrile, it must be assumed that resistant organisms are present or obstructive uropathy with or without calculi exists. A change or an addition of antimicrobial agents may be necessary, based on the sensitivities of the microorganisms recovered on urine culture. Investigation should be undertaken to rule out obstructive lesions of the urinary tract in patients not responding within 48 to 72 hours. A “one-shot” intravenous pyelogram or real-time ultrasonography should be used. Few patients will develop a perinephric abscess, which requires surgical drainage. Angel et al. (251) proposed that pregnant women with acute pyelonephritis could be treated solely with oral antimicrobial agents. In a randomized trial of oral cephalexin versus intravenous cephalothin, Angel et al. (251) reported equal efficacy rates (91% vs. 93%). However, 13 (14.4%) of the patients were bacteremic and thus were excluded. When the bacteremic patients were included, oral therapy was only effective in 32 (71%) of 45 patients, compared with 39 (87%) of 45 patients in the intravenous group. Subsequently, Millar et al. (247) performed a randomized controlled trial comparing inpatient and outpatient treatment of pregnant women with acute pyelonephritis before 24 weeks of gestation. The patients were initially observed in the emergency department to verify that they were stable and able to tolerate oral intake. The patient group (n = 60) received intravenous cefazolin and the outpatients (n = 60) received ceftriaxone (1 g intramuscularly in the effective dose and 1 g intramuscularly at home
within 18 to 36 hours after discharge), followed by oral cephalexin (500 mg four times a day to complete a 10-day course). Home health care nurses monitored the patients' progress for 48 to 72 hours after initial treatment. Patients with evidence of severe pyelonephritis were excluded from the study. Rates of persistent or recurrent bacteriuria and recurrent pyelonephritis were the same in both groups (247). Six (10%) of the inpatients were switched to gentamicin because of failure to respond to cefazolin, whereas none of the outpatients required a change in antibiotic (p < 0.03). Although on the surface, this study suggests that outpatient antibiotic therapy may be an appropriate option for selected pregnant women with pyelonephritis, the inpatient arm was inferior, as 12% of E. coli isolates were resistant to cefazolin (247), thus biasing the results in favor of the ambulatory regimen. Moreover, this study incorporated close monitoring, stabilization for up to 24 hours in the effective dose, and home visits. Thus, extrapolation to general clinical practice is difficult. More recently, Wing et al. (46) compared outpatient management with inpatient management of pyelonephritis in pregnancy beyond 24 weeks of gestation (46). All patients received two 1-g doses of ceftriaxone intramuscularly at 24-hour intervals while hospitalized. The outpatients were discharged after 24 hours of hospital observation if they were stable and prescribed cephalexin (500 mg orally four times a day for 10 days). Inpatients received oral cephalexin until they were afebrile for 48 hours and then were discharged to complete a 10-day course of oral cephalexin. There were no significant differences in clinical responses or birth outcomes of inpatients or outpatients if they completed their assigned protocol (46). However, 30% of outpatients were unable to complete their protocol. Moreover, most women with acute pyelonephritis were not candidates for outpatient therapy because of suspected sepsis (blood pressure of less than 90/50 mm Hg, temperature of more than 39.8°C, or sustained tachycardia of 110 beats per minute), signs of ARDS, serious underlying medical disorders (such as diabetes and systemic lupus), renal or urologic disorders, or intolerance of oral intake. Because of concerns with the previously discussed studies, our policy is that pregnant women with pyelonephritis at any gestational age should be hospitalized for initiation of antimicrobial therapy, rehydration, close monitoring for complications, and monitoring for preterm labor. Once patients have stabilized and are afebrile for 24 to 48 hours, they are discharged on oral therapy to complete a 10-day course. Recurrence for acute pyelonephritis during the same gestation has been reported in 10% to 18% of all patients with pyelonephritis. In a retrospective analysis, Harris and Gilstrap (252) reported that patients with acute pyelonephritis who did not receive suppressive antimicrobial therapy for the remainder of the gestation had a 60% incidence of recurrence, requiring rehospitalization. This was in contrast to the low incidence (2.7%) of recurrence and rehospitalization in patients who were maintained on suppressive antimicrobial therapy for the duration of the gestation. Cunningham et al. (198) noted a similar high incidence of recurrence among patients not receiving suppressive therapy. Lenke et al. (253) performed a prospective randomized trial assessing antibiotic prophylaxis after an episode of pyelonephritis during pregnancy versus use of sequential urine cultures to identify patients for re-treatment. These authors noted that recurrent or persistent bacteriuria was more common in the group followed with sequential cultures. However, the rates of recurrent pyelonephritis in the two groups were similar, 7% and 8%. Suppression is accomplished with nitrofurantoin,
100 mg each, or trimethoprim-sulfamethoxazole, 80/160 mg every night before bed (Table 15.24). An acceptable alternative to suppressive therapy is continued examination every 2 weeks of urine cultures for the detection and prompt treatment of recurrent bacteriuria. We feel that the risk of recurrence of pyelonephritis is too great not to suppress the UTI with antimicrobial therapy after an acute episode of pyelonephritis. Nosocomial Urinary Tract Infection Epidemiology Hospital-acquired UTIs are among the most frequent of nosocomial infections. An exact prevalence for nosocomial UTIs is difficult to come by because of various surveillance systems and different definitions (i.e., clinically manifest infection versus laboratory evidence of significant bacteriuria). Based on data obtained from surveillance studies of nosocomial infections, approximately one third of these hospital-acquired infections involve the urinary tract (254,255,256,257 and 258). Sixty percent of the nosocomial infections occurring in gynecologic patients involve the urinary tract (259). Current estimates are that approximately 500,000 patients per year acquire UTIs in acute care hospitals in the United States (260,261). Most of these UTIs are associated with usage of indwelling urethral catheterization or other types of urinary instrumentation (262,263). It has been estimated that 1% of nosocomial UTIs (i.e., 5,000 cases) are associated with bacteremia and potentially life-threatening illness (265). Platt et al. (266) report that the acquisition of UTI during indwelling bladder catheterization is associated with nearly a threefold increase in mortality rate among hospitalized patients. These investigators concluded that 14% of the deaths among the catheterized patients represented the excess mortality that was associated with acquisition of infection. Based on estimates that nearly 400,000 deaths per year occur in the approximately 7.5 million persons catheterized annually in the United States, they estimated that the excess mortality associated with catheter-related infection is 56,000 patients per year. Etiology And Pathogenesis As in other UTIs, E. coli is the most common etiologic agent in nosocomial infections involving the urinary tract; it accounts for approximately 50% of nosocomial bacteriuria cases (267). The Klebsiella-Enterobacter group of organisms are recovered in 13% to 15%, Proteus sp in 3% to 13%, and Pseudomonas aeruginosa, Serratia, enterococci, staphylococci, and yeast make up the remainder of etiologic agents (265,267,268,269 and 270). E. coli and the Klebsiella-Enterobacter group tend to occur among patients who have not received antimicrobial therapy. Among patients who are debilitated, immunosuppressed, or on antibiotic therapy, organisms such as Proteus, Pseudomonas, Serratia, Enterococcus, and yeasts are recovered with greater frequency. Stamm et al. (258) reported that Serratia and Klebsiella are associated with an inordinate frequency of bacteremia. It has been estimated that 75% to 80% of nosocomial bacteriuria cases follow urinary tract catheterization (258,263,264). Certain risk factors have been identified, which are unavoidable; these include increasing age, debilitating illness, and female sex. Turck
and Stamm (265) also described additional factors involved in the risk of acquiring bacteriuria, which are alterable. These factors relate to both the method and the duration of catheterization. Thus, a single in-and-out catheterization is associated with an incidence for infection of less than 1%, whereas 100% of patients with indwelling urethral catheters draining into an open system for more than 4 days develop bacteriuria (1,270). Garibaldi et al. (267) noted that even with use of a closed drainage system, the risk of infection increased 5% to 10% per day of catheterization. Bacteria that cause catheter-associated UTI gain access to the urinary tract by three major routes. The first is introduction of microorganisms from the external genitalia or distal urethra into the bladder at the time of catheterization. In general, bacteria introduced in this way are well tolerated and controlled by voiding and the antibacterial defense mechanism of the bladder (265). A second mechanism by which bacteria gain access to the bladder is via a thin film of urethral fluid on the outside of the catheter (271). Catheters have been shown to contain biofilm on their surface (272,273). Bacteria contained within such biofilms are protected from antibiotics and the natural defense provided by flow of urine (274). Third, once the drainage system has been contaminated, bacteria may migrate up inside the catheter lumen (267,268). Turck and Stamm (265) believe that the intraluminal ascending route accounted for most nosocomial UTIs. Such contamination may be due to failure to use sterile technique in disconnecting the catheter and drainage tube to obtain a specimen or to irrigate the catheter (259). Inadvertent disconnection occurs and frequently results in contamination. An additional important factor in the pathogenesis of catheter-associated UTI is cross-contamination of catheters by transmission of bacteria from patient to patient on the hands of hospital personnel (1,259). Treatment And Prevention In general, we do not use antimicrobial agents to treat those patients in whom bacteriuria occurs during catheterization, but who remain asymptomatic. This concern relates to the risk of persistent colonization or emergence of more resistant nosocomial organisms. Fortunately, in many such situations, removal of the catheter results in eradication of the bacteriuria. If signs and symptoms of cystitis or acute pyelonephritis occur with the presence of an indwelling catheter, systemic antimicrobial therapy should be administered for 10 to 14 days. In addition, if bacteriuria without symptoms persists once the catheter is removed, treatment should be initiated. Regimens similar to those used in treating acute cystitis are appropriate for this situation (Table 15.24). The most important and effective method for prevention of nosocomial UTIs is to limit the use of indwelling catheters only to instances in which they are necessary. Additional preventive measures have been recommended (259): (a) Urinary catheters should be removed as soon as possible; (b) only adequately trained hospital personnel should insert urinary catheters; (c) use of aseptic technique in inserting catheters to avoid introducing bacteria; (d) cleansing the metal catheter junction with soap and water once or twice a day to reduce periurethral bacterial contamination; (e) maintain unobstructed “downhill” flow in a closed drainage system; (f) disconnect the drainage system only to irrigate an obstructed catheter, not to obtain specimens or to routinely irrigate; and (g) whenever feasible, separate catheterized patients from each other to prevent
cross-contamination. Proposed measures to aid in prevention but for which supporting data are lacking to demonstrate an advantage over a closed drainage system include the following: (a) a flutter valve to prevent reflux of urine from the collection bag to the drainage tube; (b) continuous bladder irrigation via a triple-lumen catheter with either acetic acid or a neomycin-polymyxin solution; (c) use of suprapubic catheterization; and (d) prophylactic systemic antibiotics. The final aspect of catheter care is to obtain a follow-up culture after the catheter is removed and to institute appropriate antimicrobial therapy if significant bacteriuria persists after removal of the catheter. CHAPTER REFERENCES 1. Stamm WE, Hooton TM. Management of urinary tract infections in adults. N Engl J Med 1993;329:1328–1334. 2. National Institutes of Health, The National Kidney and Urologic Diseases Advisory Board. 1990 long-range plan-window of the 21st century. Bethesda, MD: National Institutes of Health, 1990. NIH publication no 90-583. 3. Johnson JR, Stamm WE. Urinary tract infections in women: diagnosis and treatment. Ann Intern Med 1989;111:906–917. 4. Patton JP, Nash DB, Abrutyn E. Urinary tract infection: economic considerations. Med Clin North Am 1991;75:495–513. 5. Hooton TM, Scholes D, Hughes JP, et al. A prospective study of risk factors for symptomatic urinary tract infection in young women. N Engl J Med 1996;335:468–474. 6. US Bureau of the Census. Statistical abstracts of the United States: 1995, 115th ed. Washington: US Government Printing Office; 1995:82. 7. Hooton TM, Stamm WE. Diagnosis and treatment of uncomplicated urinary tract infection. Infect Dis Clin North Am 1997;11:551–581. 8. Sweet RL. Bacteriuria and pyelonephritis during pregnancy. Semin Perinatol 1977;1:25–40. 9. Stamey TA, Goven DE, Palmer JM. The localization and treatment of urinary tract infections: the role of bactericidal urine levels as opposed to serum levels. Medicine (Baltimore) 1965;44:1–36. 10. Stamey TA, Timothy M, Millar M, et al. Recurrent urinary infections in adult women. The role of introital enterobacteria. Calif Med 1971;115:1–19. 11. Johnson JR, Stamm WE. Diagnosis and treatment of acute urinary tract infections. Infect Dis Clin North Am 1987;1:773–791. 12. Kass EH, Finland M. Asymptomatic infections of the urinary tract. Trans Assoc Am Phys 1956;69:56–64. 13. Kass EH. The role of asymptomatic bacteriuria in the pathogenesis of pyelonephritis. In: Quinn EL, Kass EH, eds. Biology of pyelonephritis. Boston: Little, Brown and Company, 1960:399–412. 14. Paterson L, Miller A, Henderson A. Supra-pubic aspiration of urine in diagnosis of urinary tract infection during pregnancy. Lancet 1970;2:1195–1196. 15. Stamm WE, Counts GW, Running KR, et al. Diagnosis of coliform infection in acutely dysuric women. N Engl J Med 1982;307:463–468. 16. Latham RH, Wong ES, Larson A, et al. Laboratory diagnosis of urinary tract infection in ambulatory women. JAMA 1985;254:3333–3336. 17. Sabath LD, Charles D. Urinary tract infections in the female. Obstet Gynecol 1980;55[Suppl]:162–169. 18. Kass EH, Zinner SH. Bacteriuria and pyelonephritis in pregnancy. In: Charles D, Finland M, eds. Obstetric and perinatal infections. Philadelphia: Lea & Febiger, 1973:407–446. 19. Nicolle LE, Harding GKM, Preiksaitis J, et al. The association of urinary tract infection with sexual intercourse. J Infect Dis 1982;146:579–583. 20. Ronald A. Sex and urinary tract infection. N Engl J Med 1996;335:511–512. 21. Turck M, Goffe BS, Petersdorf RG. Bacteriuria of pregnancy: relationship to socioeconomic status.
N Engl J Med 1962;266:857–860. 22. Henderson M, Entwiste G, Tayback M. Bacteriuria and pregnancy outcome: preliminary findings. Am J Public Health 1962;52:1887–1893. 23. Williams JD, Reeves DS, Condie AP, et al. The treatment of bacteriuria in pregnancy. In: O'Grady F, Brumfitt W, eds. Urinary tract infection. London: Oxford University Press, 1968:160–169. 24. McNealey SG. Treatment of urinary tract infection in pregnancy. Clin Obstet Gynecol 1988;31:480–487. 25. Strom BL, Collins M, West SL, et al. Sexual activity, contraceptive use, and other risk factors for symptomatic and asymptomatic bacteriuria: a case-control study. Ann Intern Med 1987;107:816–823. 26. Remis RS, Gurwith MJ, Gurwith D, et al. Risk factors for urinary tract infection. Am J Epidemiol 1987;126:685–694. 27. Foxman B, Frerichs RR. Epidemiology of urinary tract infection. I. Diaphragm use and sexual intercourse. Am J Public Health 1985;75:1308–1313. 28. Foxman B, Chi J-W. Health behavior and urinary tract infection in college-aged women. J Clin Epidemiol 1990;43:329–337. 29. Fihn SD, Latham RH, Roberts P, et al. Association between diaphragm use and urinary tract infection. JAMA 1985;254:240–245. 30. Foxman B, Geiger AM, Palin K, et al. First time urinary tract infection and sexual behavior. Epidemiology 1995;6:162–168. 31. Adatto K, Doebele KG, Galland L, et al. Behavioral factors and urinary tract infection. JAMA 1979;241:2525–2526. 32. Foxman B, Frerichs RR. Epidemiology of urinary tract infection. II. Diet, clothing and urination habits. Am J Public Health 1985;75:1314–1317. 33. Sheinfeld J, Schaeffer AJ, Cordon-Cardo C, et al. Association of the Lewis blood-group phenotype with recurrent urinary tract infection in women. N Engl J Med 1989;320:773–777. 34. Kinane DF, Blackwell CC, Brettle RP, et al. ABO blood group, secretor state, and susceptibility to recurrent urinary tract infection in women. BMJ 1982;285:7–9. 35. Hooton TM, Roberts PL, Stamm WE. Effects of recent sexual activity and use of a diaphragm on the vaginal microflora. Clin Infect Dis 1994;19:247–248. 36. Hooton TM, Winter C, Tiu F, et al. Association of acute cystitis with the stage of the menstrual cycle in young women. Clin Infect Dis 1996;23:635–636. 37. Foxman B, Zhang L, Tallman P, et al. Transmission of uropathogens between sex partners. J Infect Dis 1997;175:989–992. 38. Smith HS, Hughes JP, Hooton TM, et al. Antecedent antimicrobial use increases the risk of uncomplicated cystitis in young women. Clin Infect Dis 1997;25:63–68. 39. Schwartz MA, Wang CC, Eckert LO, et al. Risk factors for urinary tract infection in the postpartum period. Am J Obstet Gynecol 1999;181:547–553. 40. Hopkins WJ, Heisey DM, Lorentzen DF, et al. A comparative study of major histocompatibility complex and red blood cell antigen phenotypes as risk factors for recurrent urinary tract infections in women. J Infect Dis 1998;177:1296–1301. 41. Stapelton A, Hooton TM, Fennell C, et al. Effect of secretor status on vaginal and rectal colonization with fimbriated Escherichia coli in women with and without recurrent urinary tract infection. J Infect Dis 1995;171:717–720. 42. Stamm WE, Raz R. Factors contributing to susceptibility of postmenopausal women to recurrent urinary tract infections. Clin Infect Dis 1999;28:723–725. 43. Raz R, Stamm WE. A controlled trial of intravaginal estradiol in post menopausal women with recurrent urinary tract infections. N Engl J Med 1993;329:753–756. 44. Jordan PA, Iravani A, Richard GA, et al. Urinary tract infection caused by Staphylococcus saprophyticus. J Infect Dis 1980;142:510–515. 45. Latham RH, Runing K, Stamm WE. Urinary tract infections in young adult women caused by Staphylococcus saprophyticus. JAMA 1983;250:3063–3066. 46. Wing DA, Hendershott CM, Debuque L, et al. Outpatient treatment of acute pyelonephritis in pregnancy after 24 weeks. Obstet Gynecol 1999;94:683–688. 47. Hooton TM, Winter C, Tiu F, et al. Randomized comparative trial and cost analysis of 3-day antimicrobial regimens for treatment of acute cystitis in women. JAMA 1995;273:41–45. 48. Cox CE, Lacey SS, Hinman F Jr. The urethra and its relationship to urinary tract infection. The urethral flora of the female with recurrent urinary tract infection. J Urol 1968;99:632–638.
49. Brumfitt W, Hamilton-Miller JMT, Bakhtiar M, et al. New technique for investigating bacterial flora of female periurethral area. Br Med J 1976;2:1471–1472. 50. Lidin-Janson G, Lindberg U. Asymptomatic bacteriuria in school girls. VI. The correlation between urinary and faecal Escherichia coli. Correlation of the relation of bacteriuria and the sampling technique. Acta Paediatr Scand 1977;66:349–354. 51. Stamey TA, Sexton CC. The role of vaginal colonization with Enterobacteriaceae in recurrent urinary infections. J Urol 1975;113:214–217. 52. Fowler JE, Stamey TA. Studies of introital colonization in women with recurrent urinary infection. VII. The role of bacterial adherence. J Urol 1977;117:472–473. 53. Stamey JA, Wehner N, Mihara G, et al. The immunologic basis of recurrent bacteriuria; role of cervicovaginal antibody in enterobacterial colonization of the introital mucosa. Medicine 1978;57:47–56. 54. Hooten TM, Hillier S, Johnson C, et al. Escherichia coli bacteriuria and contraceptive method. JAMA 1991;265:64–69. 55. Stamm WE, McKevitt M, Roberts PL, et al. Natural history of recurrent urinary tract infections in women. Rev Infect Dis 1991;13:77. 56. Hooten TM, Fennell CL, Clark AM, et al. Nonoxynol-9: differential antibacterial activity and enhancement of bacterial adherence to vaginal epithelial cells. J Infect Dis 1991;164:1216–1219. 57. Nicolle LE, Ronald AR. Recurrent urinary tract infection in adult women: diagnosis and treatment. Infect Dis Clin North Am 1987;1:793–806. 58. Nabor KJ. Use of quinolones in urinary tract infections. Rev Infect Dis 1989;11[Suppl 5]:1321. 59. Sandberg T, Kaijser B, Lidin-Janson G, et al. Virulence of Escherichia coli in relation to host factors in women with symptomatic urinary tract infection. J Clin Microbiol 1988;26:1471–1476. 60. Johnson JR. Virulence factors in Escherichia coli urinary tract infection. Clin Microbiol Rev 1991;4:80–128. 61. Stapelton A, Mosley S, Stamm WE. Urovirulence determinants in Escherichia coli isolates causing first-episode and recurrent cystitis in women. J Infect Dis 1991;163:773–779. 62. Svanberg C, Godaly G. Bacterial virulence in urinary tract infection. Infect Dis Clin North Am 1997;11:513–529. 63. Lindberg F, Lund B, Johanssen L, et al. Localization of the receptor-binding protein adhesin at the tip of the bacterial pilus. Nature 1987;328:84–87. 64. Svanborg C, Agace W, Hedges S, et al. Bacterial adherence and mucosal cytokine production. Ann N Y Acad Sci 1994;162–181. 65. Leffler H, Svanborg-Eden C. Glycolipids as receptors for Escherichia coli lectins or adhesins. In: Mirelman D, ed. Microbial lectins and agglutinins: properties and biological activity. New York: John Wiley and Sons, 1986:83–111. 66. Mabeck CE, Orskov F, Orskov I. Escherichia coli serotypes and renal involvement in urinary tract infection. Lancet 1971;1:1312–1314. 67. Orskov I, Orskov F, Birch-Andersen A, et al. O,K,H and fimbrial antigens in Escherichia coli serotypes associated with pyelonephritis and cystitis. Scand J Infect Dis Suppl 1982;33:18–25. 68. Vaisanen-Rhen V, Elo J, Vaisanan A, et al. P-fimbriated clones among uropathogenic Escherichia coli strains. Infect Immun 1984;43:149–155. 69. Lidin-janson G, Hanson LA, Kaijser B, et al. Comparison of Escherichia coli from bacteriuric patients with those from feces of healthy schoolchildren. J Infect Dis 1977;136:346–353. 70. Agace W, Hedges S, Anderson U, et al. Selective cytokine production by epithelial cells following exposure to Escherichia coli. Infect Immun 1993;61:602–609. 71. Hedges S, Agace WM, Svanborg C. Epithelial cytokine responses and mucosal cytokine networks. Trends Microbiol 1995;3:266–270. 72. Connell H, Agace W, Klemm P, et al. Type 1 fimbrial adhesion enhances Escherichia coli virulence for the urinary tract. Proc Natl Acad Sci U S A 1996;93:9827–9832. 73. Orskov I, Svanborg Eden C, Orskov F. Aerobactin production of serotyped Escherichia coli from urinary tract infections. Med Microbiol Immunol (Berl) 1988;177:9–14. 74. Jacobson SH, Hammarlind M, Lidefeldt KJ, et al. Incidence of aerobactin-positive Escherichia coli strains in patients with symptomatic urinary tract infection. Eur J Clin Microbiol Infect Dis 1988;7:630–634. 75. Cavalieri SJ, Bohach GA, Snyder IS. Escherichia coli x-hemolysin: characteristics and probable role in pathogenicity. Microbiol Rev 1984;48:326–343. 76. Stapelton A, Nudelman E, Clausen H, et al. Binding of uropathogenic Escherichia coli R45 to
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.
glycolipids extracted from vaginal epithelial cells is dependent on histo-blood group secretor status. J Clin Invest 1992;90:965–972. Sobel JD. Pathogenesis of urinary tract infection: role of host defenses. Infect Dis Clin North Am 1997;11:531–549. Cox CE, Hinman F Jr. Experiments with induced bacteriuria, vesicle emptying and bacterial growth on the mechanism of bladder defenses to infection. J Urol 1961;83:739. Vivaldi E, Munoz J, Colran R. Factors affecting the clearance of bacteria within the urinary tract. In: Kass EH, ed. Progress in pyelonephritis. Philadelphia: FA Davis Co, 1965. Schulte-Wissermann H, Mannhardt W, Schwarz J, et al. Comparison of the antibacterial effect of uroepithelial cells from healthy donors and children with asymptomatic bacteriuria. Eur J Pediatr 1985;144:230. Svanborg-Eden C, Anderson B, Hagberg L, et al. Receptor analogues and antipili antibodies as inhibitors of bacterial attachment in vivo and in vitro. Ann N Y Acad Sci 1983;409:580. Orskov I. Ferencz A, Orskov F. Tamm-Horsfall protein or uromucoid is the normal urinary slime that traps type 1 fimbriated Escherichia coli. Lancet 1980;1:887. Parsons CL, Greenspan C, Mulholland SG. The primary antibacterial defense mechanism of the bladder. Invest Urol 1975;13:72. Jarvinen A, Sandholm M. Urinary oligosaccharides inhibit adhesion of E. coli onto canine urinary tract epithelium. Invest Urol 1980;17:443. Agace WW, Patarrayo M, Svensson M, et al. Escherichia coli induces trans-epithelial neutrophil migration by an ICAM-1 dependent mechanism. Infect Immun 1995;63:4054. Rene P, Silverblatt FJ. Serological response to Escherichia coli pili in pyelonephritis. Infect Immun 1982;37:749. Rene P, Dinolfo M, Silverblatt FJ. Serum and urogenital antibody response to Escherichia coli pili in cystitis. Infect Immun 1982;38:542. Remis RS, Gurwith MJ, Gurwith D, et al. Risk factors for urinary tract infection. Am J Epidemiol 1987;126:685. McNeeley SG, Kmak DC. Lower urinary tract infections in pregnancy. Contemp Obstet Gynecol 2000(January):15–19. Rubin UH, Shapiro ED, Andriole VT, et al. Evaluation of new anti-infective drugs for the treatment of urinary tract infection. Clin Infect Dis 1992;15:S216. Hindman R, Tronic B, Bartlett R. Effect of delay on culture of urine. J Clin Microbiol 1976;4:102–103. Campbell-Brown M, McFadyen IR, Scal DW, et al. Is screening for bacteriuria in pregnancy worthwhile? Br Med J 1987;294:1579. Millar LK, Cox SM. Urinary tract infections complicating pregnancy. Infect Dis Clin North Am 1997;11:13–26. Wadland WC, Plante DA. Screening for asymptomatic bacteriuria in pregnancy. A decision and cost analysis. J Fam Pract 1989;29:372–376. Rouse DJ, Andrews WW, Goldenberg RL, et al. Screening and treatment of asymptomatic bacteriuria of pregnancy to prevent pyelonephritis: a cost-effectiveness and cost-benefit analysis. Obstet Gynecol 1995;86:119–123. Plauche WC, Janney FA, Curole DN. Screening for asymptomatic bacteriuria in pregnant patients: Three of five screening systems versus quantitative culture. South Med J 1981;74:1227. McNeeley SG, Baselski VS, Ryan GM. An evaluation of two rapid bacteriuria screening procedures. Obstet Gynecol 1987;69:550. Hagay Z, Levy R, Miskin A, et al. Uriscreen, a rapid enzymatic urine screening test: useful predictor of significant bacteriuria in pregnancy. Obstet Gynecol 1996;87:410–413. Boutros P, Mourtada H, Ronald AR. Urinary infection localization. Am J Obstet Gynecol 1972;112:379–381. Kaitz AL. Urinary concentrating ability in pregnant women with asymptomatic bacteriuria. J Clin Invest 1961;40:1331–1338. Clark H, Ronald AR, Cutler RE, et al. The correlation between site of infection and maximal concentrating ability in bacteriuria. J Infect Dis 1969;120:47–51. Percival A, Brumfitt W, Delouvois J. Serum antibody level as an indication of clinically inapparent pyelonephritis. Lancet 1964;2:1027–1033. Brumfit W, Reeves DS. Recent developments in the treatment of urinary tract infection. J Infect Dis 1969;120:61–81.
104. Vinacur JC, Casellas JM, Rubi RA, et al. Serum anti–Escherichia coli antibodies and urinary b-glucuronidase for the diagnosis and control of evolution of urinary infection during pregnancy. Am J Obstet Gynecol 1974;120:812–816. 105. Heineman HS, Lee JH. Bacteriuria in pregnancy. A heterogeneous entity. Obstet Gynecol 1973;41:22–26. 106. Thomas V, Shelokov A, Farland M. Antibody coated bacteria in the urine and the site of urinary tract infection. N Engl J Med 1974;290:588–590. 107. Jones SR, Smith JW, Sanford JP. Localization of urinary tract infections by detection of antibody-coated bacteria in urine sediment. N Engl J Med 1974;290:591–593. 108. Carlson KJ, Mulley AG. Management of acute dysuria: a decision-analysis model of alternative strategies. Ann Intern Med 1985;102:244–249. 109. Pappas PG. Laboratory in the diagnosis and management of urinary tract infections. Med Clin North Am 1991;75:313–325. 110. Fihn SD, Johnson C, Roberts PL, et al. Trimethoprim-sulfamethoxazole for acute dysuria in women: a single-dose or 10 day course: a double-blind randomized trial. Ann Intern Med 1988;108:350–357. 111. Inter-Nordic Urinary Tract Infection Study Group. Double-blind comparison of 3-day versus 7-day treatment with norfloxacin in symptomatic urinary tract infections. Scand J Infect Dis 1988;20:619–624. 112. Hooten TM, Stamm WE. Management of acute uncomplicated urinary tract infection in adults. Med Clin North Am 1991;75:339–357. 113. Norby JR. Short-term treatment of uncomplicated lower urinary tract infections in women. Rev Infect Dis 1990;12:458–467. 114. Hooton TM, Johnson C, Winter C, et al. Single-dose and three-day regimens of ofloxacin versus trimethoprim-sulfamethoxazole for acute cystitis in women. Antimicrob Agents Chemother 1991;35:1479–1483. 115. Ronald AR, Nicolle LE, Harding GK. Standards of therapy for urinary tract infections in adults. Infection 1992;20[Suppl 3]:S164–S170. 116. Hooton TM, Winter C, Tiu F, et al. Randomized comparative trial and cost analysis of 3-day antimicrobial regimens for treatment of acute cystitis in women. JAMA 1995;273:41–45. 117. Warren JW, Abrutyn E, Hebel JR, et al. Guidelines for antimicrobial treatment of uncomplicated acute bacterial cystitis and acute pyelonephritis in women. Clin Infect Dis 1999;29:745–748. 118. Ronald AR, Boutros P, Mourtado H. Bacteriuria localization and response to single-dose therapy in women. JAMA 1976;35:1854–1856. 119. Gupta K, Scholes D, Stamm WE. Increasing prevalence of antimicrobial resistance among uropathogens causing uncomplicated acute cystitis. JAMA 1999;281:736–738. 120. Talan DA, Stamm WE, Hooten TM, et al. Comparison of ciprofloxacin (7 days) and trimethoprim-sulfamethoxazole (14 days) for acute uncomplicated pyelonephritis in women. JAMA 2000;283:1583–1590. 121. Gruneberg RN. Changes in urinary pathogens and their antibiotic sensitivities, 1971–1992. J Antimicrob Chemother 1994;33[Suppl A]:1. 122. Pfau A, Sacks T, Engelstein D. Recurrent urinary tract infections in premenopausal women: prophylaxis based on an understanding of the pathogenesis. J Urol 1983;129:1153–1157. 123. Fowler JE, Pulask ET. Excretory urography, cystography, and cystoscopy in the evaluation of women with urinary tract infection: a prospective study. N Engl J Med 1981;304:462–465. 124. Stapelton A, Latham RH, Johnson C, et al. Postcoital antimicrobial prophylaxis for recurrent urinary tract infection: a randomized, double-blind, placebo-controlled trial. JAMA 1990;264:703–706. 125. Harding GKM, Ronald AR, Nicoelle LE, et al. Long-term antimicrobial prophylaxis for recurrent urinary tract infection in women. Rev Infect Dis 1982;4:438–443. 126. Nicolle LE, Harding GKM, Thompson M, et al. Efficacy of five years of continuous low-dose trimethoprim-sulfamethoxazole prophylaxis for urinary tract infection. J Infect Dis 1988;157:1239–1242. 127. Wong ES, McKevitt M, Running K, et al. Management of recurrent urinary tract infections with patient administered single-dose therapy. Ann Intern Med 1985;102:302–309. 128. Svanborg-Eden C, Hausson S, Jodal U, et al. Host-parasite interaction in the urinary tract. J Infect Dis 1988;157:421–426. 129. Duff P. Urinary tract infection. Primary Care Update Obstet Gynecol 1994;1:12–16. 130. Rubin RH, Shapiro ED, Andriole VT, et al. Evaluation of new anti-infective drugs for the treatment
of urinary tract infection. Clin Infect Dis 1992;15[Suppl 1]:S216–S227. 131. Johnson JR, Lyons MF II, Pearce W, et al. Therapy for women hospitalized with acute pyelonephritis: a randomized trial of ampicillin versus trimethoprim-sulfamethoxazole for 14 days. J Infect Dis 1991;163:325–330. 132. MacMillan MC, Grimes DA. The limited usefulness of urine and blood cultures in treating pyelonephritis in pregnancy. Obstet Gynecol 1991;78:745–748. 133. Stamm WE, McKevitt M, Counts GW. Acute renal infection in women: treatment with trimethoprim-sulfamethoxazole or ampicillin for two or six weeks: a randomized trial. Ann Intern Med 1987;106:341–345. 134. Safrin S, Siegel D, Black D. Pyelonephritis in adult women: inpatient versus outpatient therapy. Am J Med 1988;85:793–798. 135. Pinson AG, Philbrick JT, Lindbeck GH, et al. ED management of acute pyelonephritis in women: a cohort study. Am J Emerg Med 1994;12:271–278. 136. Bailery RR, Peddie BA. Treatment of acute urinary tract infection in women. Ann Intern Med 1987;107;430. 137. Bialey RR. Duration of antimicrobial treatment and the use of drug combinations for the treatment of uncomplicated acute pyelonephritis. Infection 1994;22[Suppl 1]:S80–S82. 138. Bach D, van den Berg-Segers A, Hubner A, et al. Rulfloxacin once daily versus ciprofloxacin twice daily in the treatment of patients with acute uncomplicated pyelonephritis. J Urol 1995;154:19–24. 139. Bergeron MG. Treatment of pyelonephritis in adults. Med Clin North Am 1995;79:619–649. 140. Ronald AR, Harding GKM. Complicated urinary tract infections. Infect Dis Clin North Am 1997;11:583–592. 141. Zhanel GG, Harding GKM, Guay DRP. Asymptomatic bacteriuria: which patients should be treated? Arch Intern Med 1990;150:1389–1396. 142. Hooton TM, Scholes D, Stampleton AE, et al. A prospective study of asymptomatic bacteriuria in sexually active young women. N Engl J Med 2000;343:992–997. 143. Nicolle LE. Asymptomatic bacteriuria—important or not? N Engl J Med 2000;343:1037–1039. 144. Millar LK, Cox SM. Urinary tract infections complicating pregnancy. Infect Dis Clin North Am 1997;11:13–26. 145. Meis PJ, Michielutte R, Peters TJ, et al. Factors associated with preterm birth in Cardiff, Wales. Am J Obstet Gynecol 1995;173:597–602. 146. Romero R, Oyarzun E, Mazor M, et al. Meta-analysis of the relationship between asymptomatic bacteriuria and preterm delivery/low birthweight. Obstet Gynecol 1989;73:576–582. 147. Schieve LA, Handler A, Hershow R, et al. Urinary tract infection during pregnancy: its associations with maternal morbidity and perinatal outcome. Am J Public Health 1994;84:405–410. 148. Kass EH. Bacteriuria and pyelonephritis of pregnancy. Arch Intern Med 1960;205:194–198. 149. Elder HA, Kass EH. Renal function in bacteriuria of pregnancy: its relationship to prematurity, acute pyelonephritis and excessive weight gain. In: Kass EH, ed. Progress in pyelonephritis. Philadelphia: FA Davis Co, 1965:81–86. 150. Elder HA, Santamarina BAG, Smith S, et al. The natural history of asymptomatic bacteriuria during pregnancy: the effect of tetracycline on the clinical course and the outcome of pregnancy. Am J Obstet Gynecol 1971;111:441–462. 151. Andriole V, Cohn GL. The effect of diethylstilbestrol on the susceptibility of rats to hematogenous pyelonephritis. J Clin Invest 1964;43:1136–1145. 152. Harle EMJ, Bullen JJ, Thompson DA. Influence of estrogen on experimental pyelonephritis caused by Escherichia coli. Lancet 1975;2:283–286. 153. Braude AI. Current concepts of pyelonephritis. Medicine (Baltimore) 1973;52:257–264. 154. Fass RJ, Klainer AS, Perkins RL. Urinary tract infection. Practical aspects of diagnosis and treatment. JAMA 1973;225:1509–1513. 155. Whalley P. Bacteriuria of pregnancy. Am J Obstet Gynecol 1967;97:723–738. 156. Andrews W, Cox S, Gilstrap LC. Urinary tract infections in pregnancy. Int Urogynecol J 1990;1:155–163. 157. Bryant RE, Windom RE, Vineyard JP, et al. Asymptomatic bacteriuria in pregnancy and its association with prematurity. J Lab Clin Med 1964;63:224–231. 158. Condie AP, Williams JD, Reeves DS, et al. Complications of bacteriuria in pregnancy. In: O'Grady F, Brumfitt W, eds. Urinary tract infection. London: Oxford University Press, 1968:148–159. 159. Gruneberg RN, Leigh DA, Brumfitt W. Relationship of bacteriuria in pregnancy to acute pyelonephritis, prematurity and fetal mortality. Lancet 1969;2:1–3.
160. Little PJ. The incidence of urinary infection in 5000 pregnant women. Lancet 1966;4:925–928. 161. McFadyen IR, Eykyn SJ, Gardner NHN, et al. Bacteriuria in pregnancy. J Obstet Gynaecol (Br Commonw) 1973;80:385–405. 162. Norden CW, Kilpatrick WH. Bacteriuria of pregnancy. In: Kass EH, ed. Progress in pyelonephritis. Philadelphia: FA Davis Co, 1965:64–72. 163. Stuart KL, Cummins GT, Chin WA. Bacteriuria, prematurity and the hypertensive disorders of pregnancy. Br Med J 1965;1:554–556. 164. Williams GL, Campbell HM, Davies KJ. The influence of age, parity and social class on the incidence of asymptomatic bacteriuria in pregnancy. J Obstet Gynaecol (Br Commonw) 1969;76:229–239. 165. Blunt A, Williams RH. Asymptomatic bacteriuria in pregnancy. Aust N Z J Obstet Gynecol 1969;9:196–198. 166. Carrol R, MacDonald D, Stanley JC. Bacteriuria in pregnancy. Obstet Gynecol 1968;32:525–527. 167. Gower PE, Haswell B, Sidaway ME, et al. Follow-up of patient with bacteriuria of pregnancy. Lancet 1968;2:990–994. 168. LeBlanc AL, McGanity WJ. The impact of bacteriuria in pregnancy—a survey of 1300 pregnant patients. Tex Rep Biol Med 1964;22:336–347. 169. Pathak UN, Tang K, Williams LL, et al. Bacteriuria of pregnancy: results of treatment. J Infect Dis 1969;120:91–95. 170. Robertson JG, Livingstone JRB, Isdale MH. The management and complications of asymptomatic bacteriuria in pregnancy. J Obstet Gynaecol (Br Commonw) 1968;75:59–65. 171. Savage WE, Hajj SN, Kass EH. Demographic and prognostic characteristics of bacteriuria in pregnancy. Medicine (Baltimore) 1967;46:385–407. 172. Turner GC. Bacilluria in pregnancy. Lancet 1967;2:1062–1064. 173. Eykin SJ, McFadyen IR. Suprapubic aspiration of urine in pregnancy. In: O'Grady F, Brumfitt W, eds. Urinary tract infection. London: Oxford University Press, 1968:141–147. 174. Kaitz AL, Holder EW. Bacteriuria and pyelonephritis of pregnancy. N Engl J Med 1961;265:667–672. 175. Kincaid-Smith P. Bacteriuria in pregnancy. In: Kass EH, ed. Progress in pyelonephritis. Philadelphia: FA Davis Co, 1965:11–26. 176. Sleigh JD, Robertson JG, Isdale MH. Asymptomatic bacteriuria in pregnancy. J Obstet Gynecol (Br Commonw) 1964;71:74–81. 177. Layton R. Infection of the urinary tract in pregnancy: an investigation of a new routine in antenatal care. J Obstet Gynecol (Br Commonw) 1964;71:927–933. 178. Monson OT, Armstrong D, Pion RJ, et al. Bacteriuria during pregnancy. Am J Obstet Gynecol 1963;85:511–518. 179. Patrick MJ. Renal infection in pregnancy. The natural development of bacteriuria in pregnancy. J Obstet Gynecol (Br Commonw) 1966;73: 793–796. 180. Wren BG. Subclinical urinary infection in pregnancy. Med J Aust 1969;2:1220–1226. 181. Wilson MG, Hewitt WI, Monzon OT. Effect of bacteriuria on the fetus. N Engl J Med 1966;274:1115–1118. 182. Dixon HG, Brandt HA. The significance of bacteriuria in pregnancy. Lancet 1967;1:19–20. 183. Harris RE, Thomas VL, Shelokov A. Asymptomatic bacteriuria in pregnancy: antibody-coated bacteria, renal function and intrauterine growth retardation. Am J Obstet Gynecol 1976;126:20–25. 184. Kincaid-Smith P, Bullen M. Bacteriuria in pregnancy. Lancet 1965;1:395–399. 185. Kunin CM. Asymptomatic. Annu Rev Med 1966;17:383–406. 186. Norden CW, Kass EH. Bacteriuria of pregnancy—a critical appraisal. Annu Rev Med 1968;19:431–470. 187. Brumfitt W. The effects of bacteriuria in pregnancy on maternal and fetal health. Kidney Int 1975;8[Suppl]:113–119. 188. Naeye RL. Urinary tract infections and the outcome of pregnancy. Adv Nephrol 1986;15:95–102. 189. Patrick MJ. Influence of maternal renal infection on the fetus and infant. Arch Dis Child 1967;42:208–213. 190. Reddy J, Campbell A. Bacteriuria in pregnancy. Aust N Z J Obstet Gynecol 1985;25:176–178. 191. Pritchard JA, Scott DE, Whalley PJ, et al. The effects of maternal sickle hemoglobinopathies and sickle cell trait on reproductive performance. Am J Obstet Gynecol 1973;117:662. 192. Lye WC, Chan RK, Lee EJ, et al. Urinary tract infections in patients with diabetes mellitus. J Infect 1992;24:169.
193. Stamm WE, McKevitt M, Roberts RL, et al. Natural history of recurrent urinary tract infections in women. Rev Infect Dis 1991;13:77. 194. Committee on Technical Bulletins, American College of Obstetricians and Gynecologists. Antimicrobial therapy for obstetric patients. Am Coll Obstet Gynecol Bull 1988;117:1–5. 195. US Preventive Services Task Force. Guide to clinical preventative services: an assessment of the effectiveness of 169 interventions. Baltimore: Williams & Wilkins, 1989:155–162. 196. Canadian Task Force on the Periodic Health Examination. The periodic health examination. Can Med Assoc 1979;121:1193–1254. 197. Wood EG, Dillon HC Jr. A Prospective study of group B streptococcal bacteriuria in pregnancy. Am J Obstet Gynecol 1981;140:515–520. 198. Cunningham FG, Morris GB, Mickal A. Acute pyelonephritis of pregnancy: a clinical review. Obstet Gynecol 1973;42:112–117. 199. Baird D. The upper urinary tract in pregnancy and the puerperium with special reference to pyelitis of pregnancy. J Obstet Gynaecol Br Emp 1936;43:1–59. 200. Dodds GH. Bacteriuria in pregnancy, labor and puerperium. J Obstet Gynaecol Br Emp 1931;38:773–787. 201. Crabtree E. Urological diseases in pregnancy. Boston: Little, Brown and Company, 1942. 202. McLane CM. Pyelitis of pregnancy. Am J Obstet Gynecol 1939;38:117–123. 203. Hibbard L, Thrupp L, Summeril S, et al. Treatment of pyelonephritis in pregnancy. Am J Obstet Gynecol 1967;98:609–615. 204. Kass EH. Pregnancy, pyelonephritis and prematurity. Clin Obstet Gynecol 1973;13:239–254. 205. Wren BG. Subclinical renal infection and prematurity. Med J Aust 1969;2:956–960. 206. Gilstrap LC, Hankins GDV, Snyder RR et al. Acute pyelonephritis in pregnancy. Compr Ther 1986;12:38–42. 207. Graham JM, Oshiro BT, Blanco JD, et al. Uterine contractions after antibiotic therapy for pyelonephritis in pregnancy. Am J Obstet Gynecol 1993;168:577–580. 208. Calderyo-Barcia R. In: Kowlessor M, ed. Physiology of prematurity, 5th conference. Josiah Macy Jr Foundation, 1960:105. 209. Wiederman J, Stone ML, Pataki R. Urinary tract infections and uterine activity. Effect of Escherichia coli endotoxins on uterine motility in vitro. 210. Cox SM. Infection-induced preterm labor In: Gilstrap LC, Faro S, eds. Infectious in pregnancy. New York: Alan R. Liss, 1990:247–253. 211. Romero R, Mazor M. Infection and preterm labor. Clin Obstet Gynecol 1988;31:553–584. 212. Whalley PJ. Bacteriuria in pregnancy. In: Kass EH, ed. Progress in pyelonephritis. Philadelphia: FA Davis Co, 1965:50–57. 213. Harris RE. The significance of eradication of bacteriuria during pregnancy. Obstet Gynecol 1979;53:71–73. 214. Gratacos E, Torres P-J, Vila J, et al. Screening and treatment of asymptomatic bacteriuria in pregnancy prevents pyelonephritis. J Infect Dis 1994;169:1390–1392. 215. Smaill F. Antibiotic versus no treatment for asymptomatic bacteriuria. In: Enkin MW, Keirse MJNC, Renfrew MJ, et al, eds. Pregnancy and childbirth module. Cochrane database of systematic reviews [Review no 03170, 22 April 1993]. Oxford: Update Software; 1993. Disk issue no 2. 216. Norden CW. Significance and management of bacteriuria of pregnancy. In: Kaye D, ed. Urinary tract infection and its management. St. Louis: Mosby, 1972:171–187. 217. Fairley KF, Whitworth JA, Radford HJ, et al. Pregnancy bacteriuria. The significance of site of infection. Med J Aust 1973;2:424–427. 218. Giles C, Brown JAH. Urinary infection and anemia in pregnancy. Br Med J 1962;2:10–13. 219. Gilstrap LC, Cunningham FG, Whalley PJ. Acute pyelonephritis in pregnancy: an anterospective study. Obstet Gynecol 1981;57:409–413. 220. Zinner SH, Kass EH. Long-term (10–12 years) follow up of bacteriuria of pregnancy. N Engl J Med 1971;285:820–824. 221. Leigh DA, Gruneberg RN, Brumfitt W. Long term follow up of bacteriuria in pregnancy. Lancet 1968;2:603–605. 222. Cobbs CG, Stickler JC, McGovern JH, et al. The postpartum renal status of women with untreated asymptomatic bacteriuria during pregnancy. Am J Obstet Gynecol 1967;99:221–227. 223. Leveno KJ, Harris RE, Gilstrap LC, et al. Bladder versus renal bacteriuria during pregnancy: recurrence after treatment. Am J Obstet Gynecol 1981;139:403–406. 224. Norden CW, Levy PS, Kass EH. Predictive effect of urinary concentrating ability and
225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252.
hemagglutinating antibody titer upon response to antimicrobial therapy in bacteriuria of pregnancy. J Infect Dis 1970;121:588–596. Kass EH. Bacteriuria and the diagnosis of infection of the urinary tract. Arch Intern Med 1957;100:709–714. Kass EH. Pyelonephritis and bacteriuria. A major problem in preventive medicine. Ann Intern Med 1962;56:46–53. Bachman, JW, Heise RH, Nassens JM, et al. A study of various tests to detect asymptomatic urinary tract infections in an obstetric population. JAMA 1993;270:1971–1974. Lenke RR, Van Dorsten JPV. The efficacy of the nitrite test and microscopic urinalysis in predicting urine culture results. Am J Obstet Gynecol 1981;140:427. Boggess KA, Benedetti TJ, Rahgu G. Nitrofurantoin-induced pulmonary toxicity during pregnancy: a report of a case and review of the literature. Obstet Gynecol Survey 1996;51:367–370. Jick SS, Jick H, Walker AM, et al. Hospitalizations for pulmonary reactions following nitrofurantoin use. Chest 1989;96:512–515. Harris RE, Gilstrap LC, Pretty A. Single dose antimicrobial therapy for asymptomatic bacteriuria during pregnancy. Obstet Gynecol 1982;59:546–549. Campbell-Brown M. Bacteriuria in pregnancy treated with a single dose of cephalexin. Br J Obstet Gynecol 1983;90:1054. Bailey RR. Single dose antibacterial treatment for bacteriuria in pregnancy. Drugs 1984;27:183. Gallagher DJA, Montgomerie JZ, North JDK. Acute infections of the urinary tract and the urethral syndrome in general practice. Br Med J 1965;1:622–626. Harris RE, Gilstrap LC. Cystitis during pregnancy: a distinct clinical entity. Obstet Gynecol 1981;57:578–580. Stenquist K, Sanberg G, Lidin-Janson F, et al. Virulence factors of E. coli in urinary isolates from pregnant women. J Infect Dis 1987;156:870–877. MacMillan MC, Grimes DA. The limited usefulness of urine and blood cultures in treating pyelonephritis in pregnancy. Obstet Gynecol 1991;78:745–748. Cox SM, Cunningham FG. Urinary tract infection. In: Charles D, Glover R, eds. Current therapy in obstetrics. Toronto: BC Decker Inc, 1993, 209–215. Gilstrap LC, Leveno KJ, Cunningham FG, et al. Renal infection and pregnancy outcome. Am J Obstet Gynecol 1981;141:709–716. Cox SM, Shelburne P, Mason R, et al. Mechanisms of hemolysis and anemia associated with acute antepartum pyelonephritis. Am J Obstet Gynecol 1991;164:587–590. Cox SM, Cunningham FG. Ureido penicillin therapy for acute antepartum pyelonephritis. Curr Ther Res 1988;44:1029–1034. Whalley PJ, Cunningham FG, Martin FG. Transient renal dysfunction associated with acute pyelonephritis of pregnancy. Obstet Gynecol 1975;46:174–177. Cunningham FG, Leveno KJ, Hankins GDV, et al. Respiratory insufficiency associated with pyelonephritis during pregnancy. Obstet Gynecol 1984;63:121–125. Cunningham FG, Lucas MF, Hankins GDV. Pulmonary injury complicating antepartum pyelonephritis. Am J Obstet Gynecol 1987;156:797. Towers CV, Kaminskas CM, Garite TJ, et al. Pulmonary injury associated with antepartum pyelonephritis: can patients at risk be identified. Am J Obstet Gynecol 1991;164:974–980. Dunlow S, Duff P. Prevalence of antibiotic-resistant uropathogens in obstetric patients with acute pyelonephritis. Obstet Gynecol 1990;76:241–244. Millar LK, Wing DA, Paul RH, et al. Outpatient treatment of pyelonephritis in pregnancy: a randomized controlled trial. Obstet Gynecol 1995;86:560–564. Sanchez-Ramos L, McAlpine KJ, Adair CD, et al. Pyelonephritis in pregnancy: once a day ceftriaxone versus multiple doses of cefazolin. A randomized double-blind trial. Am J Obstet Gynecol 1995;172:129–133. Wing DA, Hendershott CM, Debuque L, et al. A randomized trial of three antibiotic regimens for the treatment of pyelonephritis in pregnancy. Obstet Gynecol 1998;92:249–253. Zaske DE, Cipolle RJ, Strate RG, et al. Rapid gentamicin elimination in obstetric patients. Obstet Gynecol 1980;56:559. Angel JL, O'Brien WF, Finan MA, et al. Acute pyelonephritis in pregnancy: a prospective study of oral versus intravenous antibiotic therapy. Obstet Gynecol 1990;76:28–32. Harris RE, Gilstrap LC. Prevention of recurrent pyelonephritis during pregnancy. Obstet Gynecol 1974;44:637–641.
253. Lenke RR, VanDorsten JP, Schifrin BS. Pyelonephritis in pregnancy: a prospective randomized trial to prevent recurrent disease evaluating suppressive therapy with nitrofurantoin and close surveillance. Am J Obstet Gynecol 1983;146:953–957. 254. Kislak JW, Eickhoff TC, Finland M. Hospital-acquired infection and antibiotic usage in the Boston City Hospital. N Engl J Med 1964;271:834–835. 255. Barrett FF, Casey JI, Finland M. Infections and antibiotic use among patients at Boston City Hospital. N Engl J Med 1968;278:5–9. 256. Moody ML, Burke JP. Infections and antibiotic use in a large private hospital. Arch Intern Med 1972;130:261–266. 257. McNamara MJ, Hill MC, Balows A, et al. A study of bacteriologic patterns of hospital infections. Ann Intern Med 1967;66:486–488. 258. Stamm WE, Martin SM, Bennett JV. Epidemiology of nosocomial infections due to Gram-negative bacilli. Aspects relevant to development and use of vaccines. J Infect Dis 1977;136[Suppl]:151–160. 259. Prevention of hospital acquired urinary tract infections in gynecologic patients. American College of Obstetricians and Gynecologists; May 1977. Technical bulletin no 46. 260. Centers for Disease Control and Prevention. National Nosocomial Infections Study Report: annual summary 1979. Atlanta, GA: Centers for Disease Control and Prevention; March 1982. 261. American Hospital Association. Guide to the health care field. Chicago: American Hospital Association, 1980. 262. McGowan JE Jr, Finland M. Infection and antibiotic usage at Boston City Hospital: changes in prevalence during the decade 1964–1973. J Infect Dis 1974;129:421–428. 263. Martin CM, Bookrajian EN. Bacteriuria prevention after indwelling urinary catheterization: a controlled study. Arch Intern Med 1962;110:703–711. 264. Warren JW. Catheter-associated urinary tract infections. Infect Dis Clin North Am 1997;11:609–622. 265. Turck M, Stamm W. Nosocomial infection of the urinary tract. Am J Med 1981;70:651–654. 266. Platt R, Polk BF, Murdock B, et al. Mortality associated with nosocomial urinary-tract infection. N Engl J Med 1982;307:637–642. 267. Garibaldi RA, Burke JP, Lickman ML, et al. Factors predisposing to bacteriuria during indwelling urethral catheterization. N Engl J Med 1974;291:214–219. 268. Kunin CM, McCormack RC. Prevention of catheter-induced urinary tract infections by sterile closed drainage. N Engl J Med 1966;274:1155–1161. 269. Buddington WT, Graves RC. Management of catheter drainage. J Urol 1949;62:387–393. 270. Turck M, Goffe B, Petersdorf RG. The urethral catheter and urinary tract infection. J Urol 1962;88:834–837. 271. Kass EH, Schneiderman LJ. Entry of bacteria into the urinary tract of patients with inlying catheters. N Engl J Med 1970;282:33–35. 272. Cox AJ, Hukins DWL, Sutton TM. Infection of catheterized patients: bacterial colonization of encrusted Foley catheters shown by scanning electron microscopy. Urol Res 1989;17:349–352. 273. Nickel JC, Gristina P, Costerton JW. Electron microscopic study of an infected Foley catheter. Can J Surg 1985;28:50–52. 274. Ladd TI, Schmiel D, Nickel JC, et al. The use of a radiorespirometric assay for testing the antibiotic sensitivity of catheter-associated bacteria. J Urol 1987;138:1451–1456.
16 PERINATAL INFECTIONS Infectious Diseases of the Female Genital Tract
16 PERINATAL INFECTIONS Viruses Rubella Cytomegalovirus Hepatitis Varicella-Zoster Infection Measles (Rubeola) Mumps Influenza Enteroviruses Condylomata Acuminata and Human Papillomavirus Infection Human Parvovirus Bacteria Syphilis Listeriosis Tuberculosis Protozoa Toxoplasmosis Lyme Disease Chapter References
VIRUSES Rubella Rubella (also known as German measles) is usually a mild viral illness with fever, postauricular or suboccipital lymphadenopathy, arthralgia, and a transient erythematous rash. It has been gratifying that since 1966, the incidence rate of reported rubella cases has fallen dramatically from 28 cases per 100,000 population to approximately 1 case per 100,000 in 1982 (1). Yet, there was a resurgence of rubella and congenital rubella syndrome (CRS) from 1989 to 1991, with outbreaks continuing to occur. By 1992 and 1993, however, the reported number of cases of rubella was the lowest ever (2). Further, in 1996, there were only 238 cases of rubella reported in the United States and only 4 cases of CRS (3). Since Gregg, an Australian ophthalmologist, observed in 1941 that rubella in early pregnancy was teratogenic, this disease has been of special concern to the obstetrician. Epidemiology Wild rubella virus is spread by droplets or direct contact with infected persons or articles
contaminated with nasopharyngeal secretions. Although it is considered primarily a disease of childhood, in the past few years, more than half the cases have occurred in patients older than 9 years. From 1992 through 1997, 65% of reported cases of rubella occurred in individuals 20 years or older (5). With so large a percentage in adults of reproductive age, those providing care must have a heightened awareness about rubella (2). By reproductive age, about 75% to 85% of the population has had rubella, and about half of the cases are subclinical infections. Once wild virus infection occurs (even if subclinical), immunity is lifelong. Before vaccination became available in 1969, rubella occurred in 6- to 9-year cycles, but the last major epidemic was in 1964. Since 1982, the rate in the United States has been less than 1 case per 100,000 population; but outbreaks occur, particularly among members of religious communities that traditionally refuse vaccination (Fig. 16.1). In 1991, these groups in Michigan, New York, Ohio, Pennsylvania, and Tennessee accounted for approximately 90% of the 1,007 cases in the United States. Since 1992, outbreaks have occurred among young adults in specific racial/ethnic groups (for example, Hispanics and Asian/Pacific Islanders) who frequently have not been vaccinated and in persons who have immigrated from countries, such as Brazil and Mexico, where vaccination is not routine.
FIGURE 16.1. Incidence rates of rubella and congenital rubella syndrome (CRS) in the United States, 1966 to 1993. Cases of rubella were reported to the National Notifiable Disease Surveillance System per 100,000 population. Cases of CRS were reported to the National CRS Registry per 100,000 livebirths. (From Centers for Disease Control and Prevention. Rubella and congenital rubella syndrome—United States, January 1, 1991–May 7, 1994. MMWR Morb Mortal Wkly Rep 1994;43:397, with permission.)
The main problem is infection during early pregnancy, when primary maternal rubella may lead to involvement of the embryo or fetus. As shown in Table 16.1, the risk of CRS after maternal infection in the first trimester is as high as 85% when infants have been followed for up to 40 years. The risk of any defect decreases to approximately 50% if infection occurs during the 9th through 12th week of gestation, and the risk of CRS is
low when infection occurs after the 20th week of gestation (5). In 1996, the Centers for Disease Control and Prevention (CDC) (3) revised the case definitions for rubella and CRS (Box 16.1 and Box 16.2).
TABLE 16.1. RISK OF CONGENITAL RUBELLA AFTER MATERNAL INFECTION BY GESTATIONAL AGE AT INFECTION
Cataracts, patent ductus arteriosus, and deafness are the most common abnormalities. When these children have been followed for a few years, additional disorders such as diabetes have occurred more frequently than expected. CRS has not been eliminated from the United States, but progress is being made. In 1969, when the vaccine was licensed, there were 62 cases of CRS in the United States. The number of cases decreased generally over the next two decades, so by the late 1980s, only one to three cases were reported per year. Then paralleling the rise in rubella overall, there was a dramatic surge in CRS cases between 1990 and 1991 to 25 and 31 cases, respectively. In 1991, two thirds of the cases occurred in infants born to Amish mothers in Pennsylvania. By 1992, there were only five cases in the United States, and in 1993—for the first time ever—there were no cases of CRS among infants born in the United States. (Not included in these reported cases are several imported cases, for example, five in 1992 and one in 1993.) Diagnosis With rubella infection, the virus can be isolated from the bloodstream and throat 7 to 10 days after exposure. Shedding of virus from the throat continues for about a week. The rash, which typically starts in the face, commonly develops 16 to 18 days after exposure. The diagnosis of rubella should never be based solely on clinical criteria. To confirm rubella infection, various antibody tests are available. Hemagglutination inhibition (HI) antibody, an IgG class antibody, has been most commonly used for screening, but a number of newer antibody tests have replaced the HI because of lower cost and/or greater sensitivity. The most commonly used tests are enzyme immunoassay (EIA) tests, but others include latex agglutination, fluorescence
immunoassay, passive hemagglutination, hemolysis in gel, and virus neutralization (6). A number of kits have been approved by the Food and Drug Administration (FDA), but in comparative testing, not all kits are equivalent. The most important aspect of a diagnostic kit for rubella screening tests is a high specificity (i.e., a low false-positive rate), as false-positive test results could result in failure to identify susceptible women. Further, many clinical laboratories now report rubella status as simply present or not present (corresponding to immune or nonimmune). Any antibody level is considered evidence of immunity (5). Thus, when acute rubella is suspected, the laboratory must be consulted and paired serum obtained so that quantitative antibody levels can be measured. Also an IgG antibody, complement-fixing (CF) antibody appears later than HI antibody and may be useful in some diagnostic situations, such as in patients with high HI levels or patients first seen 1 to 5 weeks after exposure. It may thus be possible to demonstrate a significant rise in CF antibody when it is too late to demonstrate this change in HI antibody. The CF test has to be requested, because it is not performed routinely. Rubella-specific IgM antibodies appear early and last for only a few weeks. This test may be helpful in some diagnostic situations but is available in few laboratories. Although a positive rubella IgM titer indicates recent primary rubella infection, its absence does not necessarily exclude it, as in some patients, IgM may disappear in less than 4 to 5 weeks. When a patient is first seen with an elevated titer, it is not always possible to determine whether there has been acute primary rubella infection, even though these other tests are properly used (7). According to the CDC (5), persons generally can be presumed to be immune to rubella if they have documented vaccination with at least one dose of measles, mumps, and rubella (MMR) vaccine or other live rubella–containing vaccines, when administered on or after the first birthday, or if they have laboratory evidence of rubella immunity, or if they were born before 1957 (except women who could become pregnant). Birth before 1957 is not acceptable evidence for rubella immunity for women who could become pregnant because it provides only presumptive evidence of rubella immunity and does not guarantee that a person is immune. If a person has an “equivocal” serologic test result, that individual should be considered susceptible to rubella unless they have evidence of adequate vaccination or a subsequent serologic test result indicating rubella immunity. Postinfection immunity to rubella appears to be long lasting and is probably lifelong.
Box 16.1 Case definition for rubella (revised, 9/96) Clinical case definition: An illness that has all of the following characteristics: Acute onset of generalized maculopapular rash Temperature of more than 99.0°F (more than 37.2°C), if measured Arthralgia/arthritis, lymphadenopathy, or conjunctivitis Laboratory criteria for diagnosis include the following: Isolation of rubella virus, or Significant rise between acute and convalescent phase titers in serum rubella immunoglobulin G (IgG) antibody level by any standard serologic assay, or Positive serologic test result for rubella immunoglobulin M (IgM) antibody Case classification: Suspected: any generalized rash illness of acute onset Probable: a case that meets the clinical case definition, has no or noncontributory serologic or virologic testing, and is not epidemiologically linked to a laboratory-confirmed case Confirmed: a case that is laboratory confirmed or that meets the clinical case definition and is epidemiologically linked to a laboratory-confirmed case Comments Serum rubella IgM test results that are false-positive have been reported in persons with other viral infections (e.g., acute infection with Epstein-Barr virus [infectious mononucleosis], recent cytomegalovirus [CMV] infection, and parvovirus infection) or in the presence of rheumatoid factor. Patients who have laboratory evidence of recent measles infection are excluded.
Box 16.2 Case definition for Congenital Rubella Syndrome (CRS) (revised, 9/96) Clinical description: An illness usually manifesting in infancy resulting from rubella infection in utero and characterized by signs or symptoms from the following categories: Cataracts or congenital glaucoma, congenital heart disease (most commonly patent ductus arteriosus or peripheral pulmonary artery stenosis), loss of hearing, pigmentary retinopathy Purpura, splenomegaly, jaundice, microcephaly, mental retardation, meningoencephalitis, radiolucent bone disease Clinical case definition: Presence of any defects or laboratory data consistent with congenital rubella infection Laboratory criteria for diagnosis include the following: Isolation of rubella virus, or Demonstration of rubella-specific IgM antibody, or Infant rubella antibody level that persists at a higher level and for a longer period than expected from passive transfer or maternal antibody (i.e., rubella titer that does not drop at the expected rate of a twofold dilution per month) Case classification: Suspected: a case with some compatible clinical findings but not meeting the criteria for a probable case Probable: a case that is not laboratory confirmed and that has any two complications listed in the first clinical description paragraph or one complication from each paragraph and lacks evidence of any other etiology Confirmed: a clinically compatible case that is laboratory confirmed Infection only: a case that demonstrates laboratory evidence of infection, but without any clinical symptoms or signs Comment In probable cases, either or both of the eye-related findings (i.e., cataracts and congenital glaucoma) are interpreted as a single complication. In cases classified as infection only, if any compatible signs or symptoms (e.g., hearing loss) are identified later, the case is reclassified as confirmed.
Comment In probable cases, either or both of the eye-related findings (i.e., cataracts and congenital glaucoma) are interpreted as a single complication. In cases classified as infection only, if any compatible signs or symptoms (e.g., hearing loss) are identified later, the case is reclassified as confirmed.
Counseling And Management Pregnant women with confirmed rubella infection must have proper counseling about the risks and types of congenital anomalies. Culturing amniotic fluid for rubella virus is not advised, because this does not reliably distinguish the infected fetus from the uninfected one among pregnancies at risk. However, the Advisory Committee to the CDC notes that immunoglobulin (Ig) given after exposure prevents neither infection nor viremia, although it may alter symptoms (5). Further, infants with congenital rubella have been born to women who received Ig shortly after exposure. Thus, the committee does not recommend Ig for routine use for postexposure prophylaxis. Yet, it suggests that Ig might be useful in susceptible, exposed pregnant women for whom pregnancy termination would not be acceptable. In this case, Ig might offer some protection against fetal infection (5). Prevention When the rubella virus vaccine became available in 1969, the strategy in the United States was to control rubella in preschool and young school-aged children who were known reservoirs. It was believed that this approach would prevent exposure of susceptible pregnant women to the wild virus. By 1977, however, 10% to 20% of women of childbearing age still were susceptible—a proportion similar to that in the prevaccine years. To meet the national goal of eliminating CRS by 1996, improved control strategies were necessary, particularly targeting young adults. These measures included increasing vaccination in children, implementing laws requiring vaccination among students, encouraging all providers to vaccinate young adults and adolescents, encouraging vaccination among religious groups that do not seek traditional health care, and targeting young adults who are likely to be unvaccinated (2). In January 1979, the rubella virus vaccine RA27/3 replaced the HPV77 vaccine in the United States. All rubella virus vaccines contain live attenuated virus. RA27/3 is administered subcutaneously. After vaccination, approximately 95% of susceptible individuals develop HI antibodies, which provide long-term (probably lifelong) protection. Among adults who did not show a positive HI titer after vaccination, nearly all show detectable antibody when a more sensitive test is used. According to the CDC, any detectable rubella antibody or a history of rubella vaccination is presumptive evidence of immunity. Those vaccinated may shed the attenuated virus from the nasopharynx for a few weeks, but there is no evidence that the vaccine virus can be transmitted. Thus, there appears to be no risk to susceptible pregnant women who contact recently vaccinated children or adults. Rubella-susceptible women of reproductive age should be considered candidates for
immunization, but it is recommended that pregnancy be avoided for 90 days. The immediate postpartum period is often suggested as an excellent time for immunization. Vaccinated women may breast-feed without fear of adverse effects to the newborn, and Rh-negative women may receive Rho(D) immune globulin, if indicated, as well as the rubella virus vaccine. It has been suggested that to eradicate CRS, one must make an increased effort to ensure that patients are either immune to rubella or vaccinated as part of routine medical and gynecologic care. For reproductive-age women, recommendations have included determining rubella immunity or vaccinating women in family planning clinics, during any hospitalization, at the time of premarital serology testing, or at college entry examinations. Rubella outbreaks in hospitals have led to the recommendation to screen for rubella immunity and vaccinate susceptible hospital employees who may have contact with pregnant women (8). Again, it is recommended that the rubella virus vaccine not be given to pregnant women and that pregnancy be avoided for 90 days. Nevertheless, many rubella-susceptible pregnant women have received the RA27/3 rubella virus vaccine within 3 months after the time of conception. As of late 1986, 172 of these women delivered at term. After vaccination with the HPV77 preparation, the virus has been isolated from the products of conception in about 20% of the cases. With the RA27/3 vaccine, cases are fewer, but this virus appears to be isolated less frequently (approximately 3%). Even though the virus may be isolated in the products of conception, none of these cases of maternal RA27/3 vaccination resulted in any anomalies consistent with CRS. The risk of teratogenicity from the RA27/3 vaccine virus is considerably lower than that with the wild virus and has been estimated by the CDC to be 1.2% (5). The risk estimate was derived by the binomial distribution. Side effects of the vaccine include arthralgias, but true arthritis occurred in less than 1%. In susceptible adult women, joint symptoms are more frequent and tend to be more severe than in children, but adults have not usually had to disrupt work. Other complaints such as pain or paresthesias have been rare. Besides pregnancy, contraindications to vaccination are febrile illnesses and immunosuppression. Precautions are necessary in few individuals with neomycin allergy. Breast-feeding is not a contraindication to vaccination. Even though vaccine virus can be transmitted via breast milk to the infant, the infant remains asymptomatic (5). The eradication of CRS and rubella is feasible worldwide because of the high efficacy of the RA27/3 rubella vaccine and the low cost of vaccine outside of the United States, making it affordable except in the poorest countries. However, universal vaccination of infants without associated vaccination of adults is not likely to be successful. Efforts need to be made to strengthen postpartum vaccination of susceptible women and women at all encounters in the health care system (9). Prenatal Diagnosis Of Congenital Rubella To diagnose fetal infection after primary rubella infection in early pregnancy, Daffos et al. (10) obtained fetal blood under ultrasound guidance at 20 to 26 weeks of gestation in
18 patients. Rubella-specific IgM was present in 12. Of these 12 pregnancies, terminations were performed in 6 of 6 pregnancies with rubella before 12 weeks of gestation and in 2 of 6 with rubella after 12 weeks, by parental decision. Among the six fetuses with blood-negative rubella-specific IgM, one was infected at birth. The missed diagnosis was believed to be due to sampling too early in gestation (10). Cytomegalovirus CMV, a DNA virus of the herpesvirus group, causes cytomegalic inclusion disease. Characteristic large cells with prominent intranuclear inclusion bodies have been identified with this disease since the early 20th century, but the virus was not isolated until 1956 (Fig. 16.2). Although this “owl's eye” appearance is pathognomonic for CMV infection, these cells are relatively rare in infected individuals. Initially, CMV was considered to be rare because only the classic clinically severe form of the disease was appreciated. Subsequent studies using viral cultures have identified the frequent presence of “silent” CMV in which no clinical manifestations are present. Weller (1) initially described the scope and impact of congenital CMV. It has been estimated that there are approximately 40,000 infants born annually in the United States with congenital infection (2,3 and 4). CMV is now recognized as the most common cause of intrauterine infection, and congenital CMV infection has been reported to occur in 0.5% to 2.5% of all births in the United States (Table 16.2). An additional 3% to 5% of newborns acquire CMV during the perinatal period.
FIGURE 16.2. Intranuclear (arrow) and cytoplasmic inclusion (circle) in a pathognomonic “owl's eye” cell of cytomegalic inclusion disease.
TABLE 16.2. RISK OF CONGENITAL CYTOMEGALOVIRUS INFECTION
Absence of detectable infection at birth may not be innocuous. The persistent and progressive nature of these inapparent congenital infections may result in central nervous system (CNS) pathology and neurologic sequelae, which represent the major impact of CMV infection (5). The teratogenic potential of CMV is unsettled. Although the virus can cause a reduction in the absolute number of cells in various organs, whether this is due to a direct effect on the cells or whether it is secondary to endothelial and vascular damage is not known. Malformations such as cataracts or congenital heart lesions are seldom seen with CMV. Thus, congenital CMV bears a closer resemblance to congenital toxoplasmosis, in which defects are secondary to destruction of tissue than to congenital rubella. Epidemiology Approximately 40% of women in the United States and Europe are susceptible to CMV by the time they reach reproductive age, and the highest rate of seroconversion occurs between the ages of 15 and 35. Socioeconomic status is a major determinant of susceptibility; only 15% of lower income women are susceptible, compared with 45% of higher income women (3). The rate of seroconversion in women in the reproductive age range is approximately 2% annually in higher socioeconomic groups, compared with 6% among lower socioeconomic groups (6). Chandler et al. (7) demonstrated that seropositivity to CMV correlated with lower socioeconomic status, multigravidity, older age, first pregnancy when younger than 15 years, and a greater number of sex partners. Absence of these risk factors identifies those women who are most susceptible to primary CMV infection during pregnancy. In a study from Alabama, Hunter et al. (8) found that 30% to 40% of pregnant women were susceptible to CMV. Primary CMV infection, as evidenced by seroconversion, occurred in 1.6% of 8,000 pregnancies (8). CMV infection is more likely in lower socioeconomic groups. In a longitudinal study of CMV in 4,578 pregnant women in Houston, Yow et al. (9) reported that 52% had CMV antibody, and that among the
susceptible women, 2.2% developed primary CMV during pregnancy. Although CMV infection is widespread, it produces serious illness only in fetuses, immunodeficient individuals, and patients receiving immunosuppressive therapy. Most adult infections are subclinical, and the remainder (approximately 10%) cause a mononucleosis-like illness. CMV is a ubiquitous organism that has developed a remarkably successful form of parasitizing humans. It is persistently excreted and is communicable for long periods. Infants infected congenitally excrete CMV on an average of 4 years. Those acquiring CMV at the time of birth excrete the virus for 2 years. Many seropositive young adults shed CMV intermittently. Recurrent excretion of CMV in asymptomatic persons may be due to several possible mechanisms. After primary infection, a low-grade chronic infection might be established in which viral excretion periodically reaches detectable levels. Reinfection could occur in immune persons because of antigenic and genetic disparity among CMV strains. Also, like herpes simplex virus (HSV), CMV may become latent during the primary infection and in later life be reactivated by various stimuli or with suppression of the cell-mediated immune system (e.g., human immunodeficiency virus [HIV] infection, transplant recipients on immunosuppressive medication, and pregnancy). Asymptomatic CMV infection and excretion is common during pregnancy. The cervix is involved in 3% to 18% of cases, urinary tract in 3% to 9%, breast milk in up to 27%, and the pharynx in 1% to 2%. Overall, CMV can be cultured (cervix and urine) in 2% to 28% of pregnant women. The incidence of CMV infection is highest in low-income, young, primiparous, lower educational status, and unmarried women. Longitudinal studies have demonstrated that the average incidence of cervical excretion of CMV in pregnancy increases from 2.6% (range, 0 to 7.1) in the first trimester to 7.6% (range, 2% to 28%) near term. These asymptomatic infections occur mainly in seropositive women whose antibody status does not change in spite of viral shedding. Pregnancy itself may either increase a woman's susceptibility to CMV infection or reactivate latent infection. Congenital CMV infection is acquired in utero, usually as the result of transplacental transmission of CMV. However, ascending infection from the cervix may also occur. Neonates with congenital CMV are culture positive for the virus at birth. Most commonly, CMV is excreted in urine. On average, 1% (range, 0.5% to 2.5%) of all newborns excrete CMV at birth and are congenitally infected. In addition to those fetuses, an additional 3% to 5% of liveborn infants acquire CMV peripartum, presumably as a result of exposure to infected cervical secretions, ingestion of infected milk, or exposure to infected transfused blood. If a maternal genital CMV infection is present at the time of birth, 30% to 50% of the neonates will acquire the virus. Perinatal CMV is the term applied to infants whose initial urine culture at birth is negative but have subsequent positive cultures several days to months after delivery. The infections transmitted in utero are the major concern, particularly because they relate to infant development. Perinatally acquired CMV infection does not result in serious complications or sequelae except in very-low-birthweight neonates (less than 1,200 g). Congenital infections may occur after either primary or recurrent maternal infection. A demonstrably high rate of congenital infection in infants born to previously
immune mothers suggests that recurrent maternal CMV is an important cause of intrauterine transmission of CMV. Indeed, as pointed out by Stagno and Whitley (3), in lower income women (approximately 85% of these women are immune), most intrauterine infections occur in immune, rather than in susceptible, women (3). In fact, the birth of an infant with congenital CMV infection does not ensure that a subsequent fetus will not become infected in utero. Congenital CMV has been reported in consecutive pregnancies up to 3 years apart. In a prospective study of seroimmune women, Stagno et al. (10) found that the rate of congenital CMV infection was 1.9% among 541 infants born to seropositive women (Table 16.3). Similarly, Schopfer et al. (11) reported a 1.4% prevalence of congenital CMV in a population from the Ivory Coast that was virtually 100% seropositive because of previous CMV infection.
TABLE 16.3. INCIDENCE OF CONGENITAL CMV INFECTION IN A LOW-INCOME POPULATION
However, primary CMV infection acquired during pregnancy poses a significantly more severe risk to the infant than does recurrent infection (6,13). Stagno et al. (6) reported that symptomatic congenital CMV infection occurred only with primary maternal infection and that sequelae or complications were significantly more frequent in the primary group (35% vs. 7%) (Table 16.4). In a more recent update from the Alabama group, Fowler et al. (13) again demonstrated that only infants born to mothers with primary CMV infection had symptomatic CMV infection at birth (18% vs. 0%). Sequelae were noted in 25% of the primary group, compared with 8% in the recurrent CMV infection group (13). Mental impairment (Ig level of less than 70) was noted in 13% of infants exposed to primary CMV, versus 0% in the recurrent group (13). Sensorineural hearing loss was present in 15% of infants born to mothers with primary CMV infection during pregnancy, compared with 5% of infants born to mothers with recurrent infection. Most importantly, only children born to mothers with primary CMV infection developed bilateral hearing loss (8%).
TABLE 16.4. SEQUELAE IN CHILDREN WITH CONGENITAL CYTOMEGALOVIRUS INFECTION ACCORDING TO TYPE OF MATERNAL INFECTIONa
The exact means by which pregnant women acquire primary CMV is not known. Because CMV is not highly contagious, infection with the virus requires close contact with infected bodily secretions. CMV is sexually transmitted, and pregnant women may also be infected by household spread of CMV from young children, who often have a high prevalence of the virus (14). As noted by Yow (4), children attending day care centers have become a major source for parental transmission of CMV. From 25% to 80% of children attending day care centers acquire CMV, which they shed in their urine and saliva for up to 2 years (14,15 and 16). Clearly, these toddlers transmit CMV to their mothers and to other family members (14,15 and 16). Fifty percent of susceptible household members will seroconvert when CMV is introduced into the household (17). With increasing use of day care centers, CMV transmitted in this setting has become a major source of primary maternal CMV infection during pregnancy. Stagno et al. (6) reported that risk factors for primary CMV infection in pregnancy include (a) presence of young child in the home; (b) white race; (c) younger age; and (d) middle to upper income. These characteristics fit the women most likely to use day care centers. Adler (18) suggested that approximately 25% of serious congenital CMV infections (primary CMV) are attributable to the day care setting. There are multiple potential sources of perinatal infection with CMV. Transplacental vertical transmission from mother to fetus has been confirmed. In addition, in utero infection may possibly be due to ascending infection across intact membranes from an infected cervix. The frequent presence of CMV in the cervix and birth canal is an obvious source for acquired neonatal CMV infection, similar to that seen with HSV. CMV has been isolated in breast milk. Stagno et al. (19) demonstrated that 70% of children of seropositive mothers excreting CMV in breast milk will acquire CMV (i.e., positive cultures) within 3 months of commencing breast-feeding). Another potential source of infection is sibling contact. However, as noted already, perinatal CMV is only a cause of neurologic sequelae in very-low-birthweight infants. Presumably this is due to immune system incompetence.
In general, primary maternal infection is viewed as potentially more dangerous for the fetus, but not all fetuses of mothers with primary CMV infection become infected (20,21). Hunter et al. (8) reported that 46% (17 of 37) of infants born to women with primary CMV infection were infected in utero. Only 2 (11%) of these 17 congenital infections were clinically detected in the nursery (8). Despite the high prevalence of CMV infection in pregnant women (documented by cervical excretion or viruria), infection in the mother is three to four times greater than that in neonates. Of more significance is the timing of infection in gestation. Manif et al. (21) showed that the more severely affected infants are those who acquire CMV infection in the first or second trimester of pregnancy. Those born to mothers with maternal infection after the third trimester were normal at birth but had positive cord serum for CMV IgM antibodies, suggesting “silent” congenital CMV infection. Stern and Tucker (20) showed that 50% of infants whose mothers developed a primary infection during pregnancy were excreting virus after delivery. Of the eight women who had reactivation of CMV infection during pregnancy, none of their infants were excreting virus after delivery. Stagno et al. (6) reported that primary CMV infection during pregnancy was associated with a 30% to 40% risk of intrauterine transmission. Adverse outcomes are more likely when infection occurs within the first 20 weeks of gestation (Table 16.5) (6). Recently Boppana et al. (22) demonstrated that the important determinants of CMV vertical transmission in utero were (a) higher levels of maternal anti-GB antibody at delivery; (b) longer duration from seroconversion to peak levels anti-GB antibody; and (c) higher levels of anti-CMV IgM antibody at the first prenatal visit and at delivery. Thus, it appears that primary CMV infection during the first two trimesters presents a greater risk for fetal infection than does infection occurring in the third trimester.
TABLE 16.5. PRIMARY CYTOMEGALOVIRUS INFECTION: EFFECT OF GESTATIONAL AGE ON TRANSMISSION IN UTERO AND DISEASE IN OFFSPRING
The public health impact of congenital CMV infection in the United States is immense (Table 16.6 and Fig. 16.3) (23). With an estimated 4 million births annually and an average rate of congenital CMV of 1%, there are more than 7,000 infants born each
year who either die or develop significant neurologic sequelae due to congenital CMV.
TABLE 16.6. PUBLIC HEALTH IMPACT OF CONGENITAL CYTOMEGALOVIRUS INFECTION IN THE UNITED STATES
FIGURE 16.3. Public health impact of congenital cytomegalovirus infection in the United States.
The prognosis is poor for babies with clinically apparent disease at birth. The mortality in these infants is estimated to be as high as 20% to 30%, and 90% will have late complications (3). CNS and perceptual disabilities usually result in severe mental retardation. The major site for this chronic morbidity is the brain and perceptual organs, with resultant seizures, spastic diplegia, optic atrophy, blindness, and sensorineural deafness. With recent emphasis shifting from the obviously diseased infant, the major focus of interest is on the prognosis of the 90% of congenitally infected neonates who appear normal at birth. These children may not develop normally, as significant neurologic sequelae may become apparent with time. Long-term longitudinal follow-up
studies have documented progressive sensorineural hearing loss and apparently subtle brain damage resulting in lowered intelligence quotient (IQ) and school-associated behavioral problems, which develop over several years after delivery of infants with subclinical CMV. The prevalence of CMV infection suggests that this agent may be a leading cause of deafness, a major contributor to school-related learning disabilities, and a significant public health problem (24). Diagnosis Clinical Manifestations Maternal Infection. More than 90% of maternal infections with CMV, primary or recurrent, are asymptomatic. Occasionally, CMV infection presents as a heterophil-negative mononucleosis syndrome with leukocytosis, with relative and absolute lymphocytosis, abnormal liver function test results, abrupt onset of spiking temperature, and constitutional symptoms such as malaise, myalgias, and chills. The mildness of the pharyngitis, minimal lymphadenopathy, and absence of hepatosplenomegaly and jaundice help to differentiate CMV from the infectious mononucleosis syndrome. Neonatal Infection. The spectrum of disease caused by CMV in the fetus and neonate is very broad. Of the congenitally infected infants, 90% are completely asymptomatic at birth. Clinically apparent disease occurs in 10% of infants with congenital CMV and ranges from isolated organ involvement to the classic multiorgan system disease. Approximately 50% of the symptomatic infants have the typical generalized pattern of the characteristic cytomegalic inclusion disease (3). In severely infected neonates, the clinical features include hepatosplenomegaly, jaundice, thrombocytopenia, purpura, microcephaly, deafness, chorioretinitis, optic atrophy, and cerebral calcifications (Table 16.7). The cerebral calcifications of CMV are characteristically periventricular in the subependymal region. A characteristic tetrad of findings has been described in infants who have survived fulminant clinically apparent CMV infection. These are (a) mental retardation, (b) chorioretinitis, (c) cerebral calcification, and (d) microcephaly or hydrocephaly. Such severe symptomatic CMV disease occurs in an estimated 1 in 10,000 to 1 in 20,000 newborns.
TABLE 16.7. CLINICAL FINDINGS IN 34 NEWBORN INFANTS WITH SYMPTOMATIC CONGENITAL CYTOMEGALOVIRUS INFECTION
The long-term prognosis for symptomatic congenital CMV infection is poor (25). The mortality rate is 20% to 30%, and more than 90% of the survivors develop significant neurologic sequelae (Table 16.8) (25). These sequelae include microcephaly, psychomotor retardation, neuromuscular disorder, unilateral and bilateral hearing loss, chorioretinitis, optic nerve atrophy, and dental defects.
TABLE 16.8. COMPLICATIONS IN PATIENTS WITH SYMPTOMATIC OR ASYMPTOMATIC CONGENITAL CYTOMEGALOVIRUS INFECTION
Ninety percent of newborns with congenital (intrauterine acquired) CMV infection are asymptomatic. However, 5% to 15% of newborns with asymptomatic CMV infection at birth will go on to develop late sequelae (2,26,27) such as sensorineural hearing loss, subnormal intelligence, and behavioral problems.
Hanshaw et al. (26) screened 3,300 cord bloods for CMV-specific IgM and followed the positive infants for 4 to 6 years (Table 16.9). These authors demonstrated that CMV IgM–positive newborns were at significantly increased risk to have school failure, an IQ of less than 90, microcephaly, and sensorineural hearing loss (26). Similarly, the Alabama group reported on their 4-year follow-up of neonates with asymptomatic congenital infection and identified complications associated with asymptomatic congenital CMV infection (27). Approximately 2% developed microcephaly with various degrees of mental retardation and neuromuscular defects by age 2. An additional 1% and 7% developed chorioretinitis and sensorineural hearing loss, respectively.
TABLE 16.9. FOLLOW-UP STUDIES OF CMV-SPECIFIC IGM-POSITIVE NEONATES
Laboratory Maternal Infection. Because maternal infection is almost always asymptomatic, the diagnosis is rarely suspected or confirmed in pregnancy. Even when clinical disease occurs, it is generally mild, and CMV is usually overlooked as a diagnostic possibility. Several reliable IgG antibody tests are available for CMV. These include indirect hemagglutination, enzyme-linked immunosorbent assay (ELISA), indirect fluorescent antibody, and neutralization tests. Because approximately 40% of adults have antibody, a single positive result does not necessarily indicate recent or current infection. Demonstration of seroconversion is the best documentation of primary infection. If infection has occurred within the previous 4 to 8 months, IgM-specific antibody can be detected in the serum. However, currently available IgM assays have not been evaluated on a widespread clinical basis (14). Moreover, IgM may remain positive for up
to 18 months and in 10% of women with recurrent CMV, IgM can be detected. Another way to establish the presence of CMV infection is by isolating the virus. Virus isolation does not differentiate primary and recurrent infections. On the other hand, diagnosis of asymptomatic recurrent CMV depends on viral isolation from urine or cervix, because no change in antibody level occurs in normal hosts with recurrent infection. CMV from swabs or urine may require 2 to 6 weeks before cytopathic effects of the virus are seen in tissue culture. Polymerase chain reaction (PCR) detection of CMV is available and provides a sensitive and specific diagnostic test for the presence of CMV. Infections In Neonates. A small percentage of clinically apparent infections in neonates are similar in presentation to other congenital infections, such as toxoplasmosis, rubella, syphilis, and herpes. The characteristic periventricular calcifications may be helpful in clinically differentiating congenital CMV infection from these other infections, but laboratory confirmation is necessary to document CMV infection. Although serology can be used as an aid in the diagnosis of congenital CMV infection, virus isolation is more sensitive and direct. As in adults, newer methods such as an indirect hemagglutination test, ELISA, and a fluorescent antibody test have replaced the CF test. Most neonates with congenital CMV have antibody to CMV when tested with the newer methods. Approximately 80% of congenitally infected infants have IgM-specific antibody in the serum during the first few months of life. This test is rather sophisticated and available from only a few medical research laboratories. Virus isolation is the best method available for documenting newborn CMV infection. Specimens can be taken from the urine, nasopharynx, conjunctiva, and spinal fluid. PCR is also available. Characteristic cytomegalic inclusions may be seen in tissues collected at autopsy or by biopsy, for example, of the kidney. The laboratory abnormalities found in neonates with symptomatic congenital CMV infection are listed in Table 16.10 (25). Similar abnormalities can be detected in the in utero infected fetus and are included as part of the assessment for prenatal diagnosis of congenital CMV.
TABLE 16.10. LABORATORY ABNORMALITIES IN SYMPTOMATIC CONGENITAL CMV INFECTION
Prenatal Diagnosis Recently, prenatal diagnosis of in utero acquired (congenital) CMV infection of the fetus has become available using ultrasound, amniocentesis, and cordocentesis (percutaneous umbilical cord sampling). On ultrasound, the most common findings include microcephaly; hydrocephalus; necrotic, cystic, or calcified lesions in the periventricular region of the brain, liver, or placenta; intrauterine growth retardation; oligohydramnios; ascites; pericardial or pleural effusion; hypoechogenic bowel; and hydrops. Direct fetal sampling via fetoscopy (28) and cordocentesis (29) has detected elevated levels of anti-CMV IgM. Subsequent studies by Weiner and Grose (30), Hohlfeld et al. (31), and Lamey et al. (32) demonstrated that amniocentesis with culturing of amniotic fluid is an excellent method for detection of in utero CMV infection. With amniotic fluid culture, these authors correctly diagnosed all cases of congenital CMV infection. The accuracy of amniotic fluid cultures for CMV is not unexpected, because the fetal kidney is one of the major sites for CMV involvement. Weiner and Grose (30) recommended amniotic fluid culture for CMV in pregnant women with documented primary CMV infection or when ultrasonography suggests intrauterine growth retardation, hydrops, ascites, and CNS abnormalities. Hohlfeld et al. (31) suggested that fetal blood sampling could provide additional information about the fetal condition. Grossly abnormal laboratory findings such as severe thrombocytopenia, anemia, or signs of hepatic involvement were associated with a rapidly fatal outcome after birth (31). With cordocentesis for fetal blood sampling used to complement ultrasound and amniocentesis, assessment of in utero CMV infection includes specific and nonspecific tests for fetal CMV infection (Table 16.11). Recently, Donner et al. (33) reported that the combination of these tests resulted in antenatal diagnosis of CMV in 13 of 16 infected fetuses (sensitivity, 81%). Amniocentesis diagnosed 12 of 13 antenatally identified cases of CMV infection. Of these 12 cases, 4
had a negative result on the first amniocentesis before 20 weeks; 4 to 8 weeks later, the results of a second amniocentesis were positive. Thus, with a strong suspicion (i.e., documented maternal primary CMV) if the initial assessment is negative, the testing should be repeated 4 to 8 weeks later. Detection of CMV IgM antibody in fetal blood had a sensitivity of 69% (9 of 13 cases). Antsaklis et al. analyzed 42 pregnant women with primary CMV infection and reported that in 14 (33.3%) of pregnancies vertical transmission of CMV occurred (34). Positive amniotic fluid culture and positive PCR were present in 9 (64.3%) and 12 (78.6%) infected fetuses. Combining both tests diagnosed 12 of 14 infected fetuses (sensitivity 85.7%). Guerra et al. assessed the relationship of CMV viral load in amniotic fluid with fetal or neonatal outcome using PCR and quantitative PCR (35). CMV infection was noted in 16 (23%) fetuses and neonates, 5 of whom were symptomatic. Using quantitative PCR, these authors demonstrated that the presence of ³ 105 genome equivalents predicted development of symptomatic infection. Most recently, Azam and co-workers assessed amnioncentesis, fetal blood sampling and serial ultrasounds in the prenatal diagnosis of CMV (36). nearly 23% (26 of 114) of fetuses were infected with prenatal diagnosis identifying 20 of these cases, resulting in a sensitivity of 77% and specificity of 100%. In this study, amniocentesis best diagnosed fetal infection while ultrasound examination and fetal blood sampling identified fetuses at risk for severe sequelae (36).
TABLE 16.11. DIAGNOSIS OF FETAL CMV INFECTION
Treatment There is no specific treatment of CMV infection. In mothers with the infectious mononucleosis-like illness, treatment is symptomatic. No satisfactory therapy is available for congenital CMV infection. Attempts have been made to use antiviral agents such as adenosine arabinoside and cytosine arabinoside for neonates with severe clinical infection, but these drugs are quite toxic. Further, although they temporarily suppress the excretion of CMV, virus shedding resumes when the drugs are stopped. Because of their toxicity, these antiviral agents are not used in asymptomatic CMV infection. Acyclovir is not effective against CMV, which unlike HSV, does not include its
own thymidine kinase. Ganciclovir has recently been demonstrated to be effective in the treatment of CMV retinitis in HIV-infected patients (37). Another agent, foscarnet, is also approved for treatment of CMV retinitis. However, no published experience with these drugs in pregnancy or neonates is available to date. The development of a CMV vaccine has been suggested as a means of preventing congenital CMV infection with its associated morbidity and mortality. Although CMV persists in the host even in the presence of high levels of specific antibody and the existing maternal antibody does not invariably protect against congenital infection, previous infection significantly reduces the risk of severe infection in the infant, particularly protecting against mortality, mental retardation, and other significant neurologic handicaps (13). These factors suggest that such an approach is likely to be successful. A live attenuated CMV vaccine (Towne strain) has been tested in volunteers and prospective renal transplant recipients (4). This vaccine elicits antibodies and cellular immune responses. Clearly, with molecular biology techniques, a recombinant vaccine can be developed for CMV. Use of such a vaccine could prevent nearly all of the neonatal deaths (n = 1,000) and approximately 90% of the severe neurologic sequelae associated with congenital CMV secondary to primary maternal infection that occur each year in the United States (based on data from Fowler et al. [13] comparing primary with recurrent maternal infection). Screening And Counseling Although reliable tests are available for detecting IgG class antibody to CMV in maternal serum, mass screening is not recommended (3,14). Such a screening program would be expensive, because detection of primary infection requires serial testing of women who are seronegative initially or testing for IgM. (As noted, tests for CMV-specific IgM have not been evaluated as a screening tool.) Then, even if primary infection is documented, present information does not allow straightforward decisions. Because most maternal infections are asymptomatic, the time in gestation of the infection is nearly always unknown. Prospective studies have shown that infections early in pregnancy are more likely to cause serious fetal injury than those late in pregnancy. Once maternal primary CMV infection has been documented, a combination of ultrasound, amniocentesis, and fetal blood sampling is necessary to identify the 30% to 40% of fetuses infected with CMV. A major limitation of a screening program is that many cases of congenital infection occur in immune women. Because there is no marker to detect these at-risk fetuses, all of these cases would be missed by a screening program. Stagno and Whitley (3) conclude that “there is inadequate information to serve as a basis for recommendations regarding termination of pregnancy after a primary CMV infection. Similarly, there is no information regarding how long conception should be delayed after primary infection.” Pass et al. (14) note that “since the most likely outcome of pregnancy complicated by maternal CMV infection is a normal infant, the well-informed patient—in our experience—will usually not choose to terminate the pregnancy.” Because large quantities of CMV are excreted by infected neonates, it has been
suggested that pregnant pediatric health care workers might be at high risk of developing CMV infection. However, Dworsky et al. (38) recently reported that the annual primary CMV infection rate (as determined by seroconversion) was no higher among pediatric house staff (2.7%) and pediatric nursing staff (3.3%) than it was among young women in the community (2.5% during pregnancy and 5.5% between pregnancies). The higher than expected rate of conversion between pregnancies suggests that exposure to young infants by family or other social exposure may be the most important factor in horizontal transmission. Pass et al. (14) showed that in several families, toddlers were the most likely source of CMV for both the mother and the fetus or infant. The toddler acquired the CMV in a day care center. As a result, the CDC currently recommends that day care workers be counseled regarding risk and instructed in careful, frequent hand washing. Use of gloves to handle material contaminated with body fluids is also useful. Seronegative workers may be offered the option of working with older children. A metaanalysis of primary CMV infection in pediatric nurses has been performed (39). These authors identified six controlled studies performed to evaluate the risk of CMV infection among pediatric nurses (37,38,40,41,42,43 and 44). The pooled risk ratio for CMV infection (using cumulative incidence data) was statistically significant (risk ratio, 2.7; 95% confidence interval [CI], 1.33–5.52). However, person-year analysis only demonstrated a trend, which did not reach statistical significance (risk ratio, 1.8; 95% CI, 0.88–3.55). They concluded that in studies published before widespread adoption of universal precautions, pediatric nurses may have been at increased risk for CMV infection because of occupational exposure. However, inadequate design and sample size prevented a definite conclusion (39). More recently, Balcarek et al. (44) evaluated CMV infection risk in employees of a children's hospital. Of the 300 workers who initially were seronegative and followed, 13 seroconverted over a mean follow-up interval of 1.96 years, 2.2% per year. There was no statistically significant difference in the incidence of CMV infection by job type, number of hours per week of patient contact, or nursing unit. This incidence of CMV was similar to that expected in the general population (45). Standard infection control measures for handling potentially contaminated material in pediatric units would seem to provide sufficient protection from CMV infection. Further, other isolation measures in infants known to have CMV infection would be “useless or even dangerous” (3,38). Hepatitis Acute viral hepatitis is a systemic infection predominately affecting the liver. Distinct hepatotropic viral agents cause hepatitis types A, B, C (which was formerly known as non-A, non-B hepatitis), D, and E. Other transmissible agents causing secondary hepatitis include CMV, Epstein-Barr virus, varicella-zoster (VZ) virus, coxsackievirus B, HSV, and rubella virus. The major impact of hepatitis on the fetus and neonate relates to hepatitis B and hepatitis C. These viruses are discussed fully in Chapter 9 (Hepatitis). Varicella-Zoster Infection VZ virus is a member of the herpesvirus group. Similar to other members of the herpesvirus group, VZ is a DNA virus and exhibits viral latency (1). Initial (primary)
infection with VZ virus results in varicella (chickenpox). This common childhood disease is usually marked by typical skin lesions, which progress from macules and papules to vesicles that occur in successive crops and evolve into pustules that form crusts and scabs (2). A highly contagious disorder, it is acquired by most persons in the United States before reproductive age and is generally self-limited. Among adults who contract the disease, constitutional and pulmonary symptoms may be more severe. Zoster (shingles) is the result of reactivation of the latent virus. It generally occurs in the older adult population or in immunocompromised patients. Characteristically, zoster presents as painful vesicular lesions that occur in a pattern of distribution that follows segmental dermatomes. Availability of the varicella vaccine in the United States has led to a dramatic decrease in new cases of varicella (3). This discussion focuses on the effects of VZ virus in pregnancy, the management and prevention of VZ infections during the perinatal period, and the possibly increased severity of the infection in pregnancy. Two major problems exist when VZ infection occurs in pregnancy. First, the infection itself poses a risk for significant morbidity and mortality for mother and neonate. Second, VZ virus has a teratogenic effect, and infection early in pregnancy may result in congenital anomalies. Epidemiology In 1998, more than 82,000 cases of chickenpox were reported in the United States (3), and this disease was third after chlamydia and gonorrhea as the most frequently reported infectious disease in the United States (3). Because of extensive underreporting, it has been estimated that 2 to 3 million cases of chickenpox occur annually in the United States (4,5 and 6). VZ is endemic in the United States and is extremely contagious; more than 90% of the population has been infected before adulthood (4). Consequently, chickenpox is uncommon among women of childbearing age and thus is uncommon in pregnancy. In the Collaborative Perinatal Research Study's prospective analysis of 30,059 pregnancies, 20 VZ infections were diagnosed, of which 14 were confirmed; thus, at a minimum, there were 5 VZ cases out of every 10,000 pregnancies (7). Balducci et al. (8) recently estimated the incidence of varicella in pregnancy to be 0.7 per 1,000 patients. Clinical Presentation VZ virus is highly contagious. It is spread by respiratory droplets and close personal contact. The incubation period of varicella ranges from 10 to 20 days but usually is 13 to 17 days. In children, fever and rash occur simultaneously, whereas in adults, fever and generalized malaise precede the rash by several days. Characteristically, the rash begins on the face and scalp and spreads to the trunk. The extremities tend to be minimally involved. Skin lesions begin as macules, then progress to a vesicular stage followed by pustules, crusts, and scabs. Itching is a common and prominent feature of the disease. For 2 to 5 days, new crops of lesions occur, and the various stages (i.e., vesicles, pustules, and scabs) are present simultaneously. Patients are contagious from 1 to 2 days before the onset of rash until the lesions have dried and become crusted. The most common complication of chickenpox is secondary bacterial infection of the
skin lesions (2). Rarely, encephalitis, meningitis, myocarditis, glomerulonephritis, and arthritis occur. The most serious complication of chickenpox is varicella pneumonia. It occurs in about 5% to 10% of adults with chickenpox (and in a much smaller percentage of children) and is associated with a significant mortality risk. Varicella pneumonia develops several days after onset of the rash. The severity of varicella pneumonia ranges from a mild illness to severe life-threatening disease characterized by cough, tachypnea, dyspnea, hemoptysis, chest pain, and cyanosis. On radiographic examination, the chest x-ray demonstrates diffuse, nodular peribronchial infiltrates, which in severe cases can progress to adult respiratory distress syndrome (Fig. 16.4).
FIGURE 16.4. Chest x-ray of patient with varicella pneumonia during pregnancy.
Varicella In The Mother Varicella is an unusual infection among adults and probably occurs with no greater frequency among pregnant women. In the past, pregnant women were believed to be at greater risk of developing severe or fatal varicella than were nonpregnant adults. However, the current consensus holds that although adults are at increased risk of developing varicella pneumonia, this risk is no greater among pregnant women (4,9,10). Recently, in a review of the literature on varicella pneumonia in pregnancy, Gershan (2) reported that among 198 cases of chickenpox in pregnant women, 57 (28%) developed varicella pneumonia. All 16 deaths occurred in the group with pneumonia, for a pneumonia mortality rate of 28% (overall mortality rate for varicella was 10%) (Table 16.12). This results in 14 (31%) deaths among 45 cases of pneumonia. Thus, it seems clear that uncomplicated chickenpox poses no severe risk to the pregnant woman. Although pneumonia (in the past) was associated with a significant mortality rate, pneumonia is an uncommon complication, and our current ability to manage severe respiratory distress and failure is much enhanced. Moreover, acyclovir in high doses is effective against VZ (6). While the mortality rate for varicella pneumonia during pregnancy has improved since the introduction of high dose acyclovir (1985), mortality
remains a concern.
TABLE 16.12. MATERNAL MORTALITY ASSOCIATED WITH VARICELLA DURING PREGNANCY
The clinician must maintain a high index of suspicion that pneumonia may complicate varicella in pregnant women. Pulmonary symptoms begin on the second to sixth day after appearance of the rash and usually consist of a mild nonproductive cough. If the disease is more severe, there may be pleuritic chest pain, hemoptysis, dyspnea, and frank cyanosis. Physical examination reveals fever, rales, and wheezes. The chest x-ray characteristically reveals a diffuse nodular or miliary pattern, particularly in the perihilar regions. With varicella in pregnancy, the patient should be warned to contact the physician immediately if even mild pulmonary symptoms develop. Hospitalization with full respiratory support, if necessary, should then be made available. Smego and Asperilla (6) described their experience with the use of acyclovir in the management of severe varicella (i.e., pneumonia) during pregnancy. They reported on 21 cases, of which 12 required intubation and mechanical ventilation for severe pneumonia. The mortality rate was 14% (three deaths), with all deaths occurring in the third trimester. The recommended dosage of acyclovir for treatment of severe varicella is 10 to 15 mg per kilogram of body weight intravenously three times daily for 7 days. The equivalent oral dosage is 800 mg five times per day. These authors noted no adverse effects due to acyclovir (6). Paryani and Arvin (18) reported the consequences of maternal varicella in 43 pregnancies. Nine women (21%) developed associated morbidity. Four women (9%) developed varicella pneumonia, and one of these women died (at 6 months into the pregnancy). Another required ventilator support. Premature labor developed in 4 (10%) of 42 and premature delivery in 2 (5%) of 42. On the other hand, morbidity was rare when herpes zoster occurred (18). In 13 (93%) of 14 cases, no complications occurred. A single patient developed cutaneous disseminated disease (18).
Because varicella infection in pregnancy is associated with significant morbidity and mortality, prevention of varicella among susceptible exposed patients is paramount. For a pregnant woman with exposure to varicella, infection is very likely if she is not immune. As recently pointed out by McGregor et al. (25), most pregnant women (12 of 17; 71%) with a negative history of chickenpox had detectable antibody. Further, of those with indeterminate histories, an even greater percentage of pregnant women were immune (18 of 20; 90%). Thus, it is appropriate and cost-effective in the face of maternal exposure to test for maternal antibody by any of the following methods: fluorescent antibody to membrane antigen, ELISA, immune adherence hemagglutination, and the enhanced neutralization test (20) (Fig. 16.5).
FIGURE 16.5. Protocol for management of pregnant women exposed to varicella.
Then, in susceptible women, varicella-zoster immune globulin (VZIG) may be given (18,19). When given intramuscularly within 96 hours of exposure, it is likely that VZIG ameliorates the course of the maternal disease, as it does in children. There is no certainty, however, that passive immunization prevents fetal infection. The recommended dose of VZIG is 125 units per 10 kg of body weight up to a maximum of 625 units (five vials) intramuscularly. The average pregnant woman requires the maximum dose, at a cost of approximately $625. Effects Of Varicella In Early Pregnancy Maternal VZ infection has been reported to result in spontaneous abortion, stillbirth, and congenital anomalies (19,26). Congenital anomalies due to varicella in early pregnancy were not recognized until LaForet and Lynch (27) (in 1947) reported the birth of an infant, exposed to VZ virus during the first trimester, who had limb hypoplasia, cicatricial skin lesions, atrophic digits, bilateral cortical atrophy, severe psychomotor retardation, and growth retardation. It is now appreciated that maternal varicella infection in the first trimester of pregnancy can produce a congenital varicella syndrome, which consists of cutaneous scars, limb hypoplasia, rudimentary digits, ocular abnormalities (e.g., optic
atrophy, microphthalmia, and cataracts), cerebral cortical atrophy, mental retardation, and growth retardation. Because both gestational varicella and the reported cases of varicella congenital syndrome are uncommon, the risk of a fetus developing congenital anomalies if the mother acquired chickenpox during the first trimester has been quantitated only recently. The risk of developing the congenital varicella syndrome after exposure in the first trimester is reviewed in Table 16.13. Overall, only 1% of infants exposed to VZ virus during the first trimester developed stigmata of congenital varicella, with a range of 0% to 9%. Recently Pastuszak et al. (25) studied 120 women with varicella in pregnancy; 106 had varicella during the first 20 weeks of gestation. When the authors controlled for elective terminations and spontaneous abortions, congenital defects occurred in four infants (1.2%) in the varicella group (95% CI, 0–2.4%). Enders et al. (26) recently reported a large prospective study in Germany and the United Kingdom of 1,373 women who had varicella before 36 weeks of gestation. Nine cases of congenital varicella syndrome were identified, for a risk rate of 1%. The highest risk (2.0%) was noted between 13 and 20 weeks of gestation (7 of 351 pregnancies; 95% CI, 0.8–4.1%). Before 13 weeks, only 2 (0.4%) of 472 pregnancies (95% CI, 0.05–1.5%) were identified.
TABLE 16.13. RISK OF CONGENITAL VARICELLA-ZOSTER SYNDROME WITH ACUTE MATERNAL VARICELLA-ZOSTER INFECTION DURING THE FIRST TRIMESTER OF PREGNANCY
Higa et al. (33) suggested that many of the congenital malformations related to VZ virus may not be caused by acute varicella in the fetus but are the result of sequelae of recurrent herpes zoster in the fetus. In particular, recurrent herpes zoster infection would explain the cutaneous and limb abnormalities that follow nerve distributions. On the other hand, acute varicella is responsible for the CNS and neurologic lesions present in the congenital varicella syndrome. The frequency of clinical findings in infants with congenital varicella syndrome is presented in Table 16.14. Cicatricial skin lesions, eye abnormalities, and hypoplastic limb are the most common findings. Nearly one fourth of the affected infants died. Although the risk of congenital varicella syndrome is low, administration of VZIG to pregnant women without evidence of previous varicella as
soon as possible (but within 96 hours of exposure) is recommended, because this may protect the fetus during the viremia (see the previous section). Once the mother has the rash, there would be no reason to administer VZIG, as viremia will have already occurred. VZIG is prepared from donors with high antibody titers to VZ virus.
TABLE 16.14. CLINICAL DATA IN INFANTS WITH DEVELOPMENTAL DEFECTS BORN TO 37 WOMEN WITH VARICELLA-ZOSTER INFECTIONS DURING EARLY PREGNANCY
As with other perinatal infections, ultrasonography, amniocentesis, chorionic villus biopsy, and cordocentesis have been proposed as techniques for diagnosing in utero varicella infection (32,34,35 and 36). Although virus-specific IgM antibody can be detected in cord blood by 19 to 22 weeks of gestation and VZ virus can be recovered from amniotic fluid and identified by in situ hybridization in placental tissue, detection of either antibody or virus does not provide information about the severity of fetal infection. Ultrasound assessment appears to be the best method available for assessing the severity of fetal involvement with varicella infection. Pretorius et al. (36) reported ultrasound findings in 37 cases with maternal varicella infection in early pregnancy and that all five fetuses with sonographic abnormalities on sonography had varicella embryopathy at postdelivery examination or autopsy. All sonographic abnormalities were observed before 20 weeks. These abnormalities included polyhydramnios, hydrops fetalis, and multiple hypoechogenic foci within the liver (36). Ultrasound has also been successful in identifying abnormal limb development. These authors noted that 14 of 17 fetuses with hypoplastic limbs also either had brain damage or died in early infancy. More recently, Mouley et al. utilized PCR of amniotic fluid for pernatal diagnosis of V-1 infection in 107 pregnant women with varicella prior to 24 weeks gestation (32). Nine (8.4%) of specimens were positive by PCR versus 2 (1.8%) by culture. The reported incidence of congenital varicella-zoster was 3 (2.8%); all were PCR positive and 2 had abnormal sonographic findings at 21–22 weeks (32). The third infant had bilateral microopthalmia in the face of a normal ultrasound at 24 weeks (32).
Varicella In The Newborn P>Acquisition of maternal antibody usually protects the fetus. However, if an infant is born after the maternal viremia but before the mother has developed an antibody response, the fetus is at high risk for life-threatening neonatal varicella infection. Infants at risk are those whose mothers contract varicella within 2 days of birth or within the first 5 days after delivery (37,38). Congenital varicella infection has been reported in 10% to 20% of full-term infants born to mothers with varicella within 4 to 5 days of delivery, and the case fatality rate was 20% to 30% (4,37). Infants born 5 or more days after the onset of maternal illness developed either mild varicella or no infection at all. Management of varicella in the newborn should focus on prevention. Ideally, delivery should be delayed until 5 to 7 days after the onset of maternal varicella. This will allow transfer of protective IgG antibody from the mother to the fetus. However, if delay is not possible, then the neonate should receive VZIG passive immunization as soon as possible after delivery. VZIG modifies or prevents varicella in normal children, and Brunell (10,39) recommended its use in preventing severe neonatal varicella. Thus, infants at risk (those born to mothers who develop varicella between 5 days before and 2 days after delivery) should receive VZIG as passive immunization. The dose is 125 units. Recently, Miller et al. (38) reported on the outcome of 281 newborns whose mothers had chickenpox during the perinatal period. All infants had received VZIG shortly after birth. However, 169 of the children (60%) were noted to be infected—134 (48%) with chickenpox and 35 (13%) without clinical features. Although VZIG did not prevent neonatal varicella, it did prevent fatal outcome. The authors concluded that VZIG is still indicated for newborn infants whose mothers have chickenpox within 7 days before or after delivery. This differs from the recommendation of 5 days before delivery in the United States. The CDC has published guidelines for the prevention of varicella (40). A live attenuated virus vaccine (Varivax) has been approved in the United States since 1995. See Chapter 25 for a detailed discussion of varicella vaccine. Because Varivax is a live attenuated virus it is recommended that pregnant women nor women attempting to conceive not receive varicella vaccine (40). A Pregnancy Registry for Varivas has been established and from May 1995 through December 1998 there were 371 women vaccinated within 3 months before or during pregnancy prospectively reported. There were zero cases of congenital varicella reported (95% C.I. 0.0–0.01). This finding plus the 1% rate of congenital varicella resulting from mild varicella virus infection resulted in the recommendation that exposure to varicells is not an indication for pregnancy termination (40). Also, because the vaccine virus is not transmissible, presence of a pregnant woman in a household is not a contraindication for vaccination of household members (40). Effect Of Zoster On Pregnancy Herpes zoster is caused by the same virus that causes varicella. It occurs very rarely in
pregnancy. Because it is a reactivation of latent VZ virus and maternal antibodies are present in normal healthy women, zoster poses no threat to the fetus or neonate (18). Measles (Rubeola) Rubeola is an acute illness that most commonly occurs in childhood. It is the most communicable of the childhood exanthems. Rubeola is characterized by fever, coryza, conjunctivitis, cough, and a generalized maculopapular rash that usually appears 1 to 2 days after the pathognomonic Koplik spots in the oral cavity. The rubeola virus is a paramyxovirus that contains RNA as its nuclear protein. Epidemiology The virus is spread chiefly by droplets expectorated by an infected person and gains access to susceptible people via the nose, oropharynx, and conjunctival mucosa. The incubation time is between 10 and 14 days. Measles is most communicable during the prodrome and catarrhal stages of the infection. Approximately three fourths of exposed susceptible contacts acquire rubeola. Before the availability of live measles vaccines, epidemics of measles occurred at intervals of 2 to 3 years in the United States. The use of attenuated measles vaccine since 1963 has had a major impact in decreasing the number of measles cases in the United States (1). Measles occurs less frequently during pregnancy than chickenpox or mumps. Before the introduction of the measles vaccine, there were 0.4 to 0.6 cases of measles per 10,000 pregnancies. This figure is probably even lower since the measles vaccine was introduced. Clinical Manifestations The prodrome of fever and malaise begins 10 to 11 days postexposure and is followed within 24 hours by coryza, sneezing, conjunctivitis, and cough. This catarrhal phase is exacerbated over the next several days, and a marked conjunctivitis and photophobia occur. The pathognomonic Koplik spots appear at the end of the prodrome. These are tiny, granular, slightly raised white lesions surrounded by a halo of erythema, which are located on the lateral buccal mucosa. The rash appears 12 to 14 days after exposure. It begins on the head and neck, particularly postauricularly, and subsequently the maculopapular rash spreads to the trunk, upper extremities, and finally the lower extremities. The respiratory tract is the most frequent site for complications of measles. Otitis media and croup are frequent occurrences, but bacterial pneumonia is the complication most frequently associated with mortality. The most common bacterial organisms involved in rubeola pneumonia are Streptococcus pneumoniae, Haemophilus influenzae, Staphylococcus aureus, and Streptococcus pyogenes. Encephalitis, a less common but serious complication, is estimated to occur with a frequency of 1 per 1,000 cases of
measles. Other complications of measles include thrombocytopenic purpura, myocarditis, and subacute sclerosing panencephalitis, which is a progressive neurologic disease associated with chronic rubeola infection of the CNS. Maternal Effects Of Measles It is unclear whether pregnant women with measles are at greater risk for serious complications and death than nonpregnant adults. More recent studies in the United States and Australia have noted that measles in pregnant women is only rarely associated with pneumonia or other complications (2). Fetal Effects Of Measles Sever et al. (2) summarized the results of reports in the literature concerning measles in pregnancy, suggesting that there is an increased rate of prematurity in pregnancies complicated by measles, particularly when the disease occurs late in gestation. However, no clear evidence suggests that maternal measles is associated with an increased risk of spontaneous abortion. Because of the rarity of measles in pregnancy, no statement can be made regarding the teratogenic potential of gestational measles for the fetus. No particular constellation of abnormalities has been found among the sporadic instances of congenital defects reported in association with maternal measles (3). In general, if there is any increased risk of malformations after measles, the risk is small. Perinatal Measles Measles that becomes clinically apparent in the first 10 days of life is considered transplacental in origin (i.e., congenital), whereas those cases occurring at 14 days or later are acquired postnatally. Postnatally acquired measles is usually associated with a mild course. Congenital measles includes cases in which the disease is present at birth or infection acquired in utero appears during the first 10 days of life. The spectrum of illness in congenital measles varies from a mild illness to a rapidly fatal disease. However, maternal measles immediately before delivery does not involve the fetus and neonate commonly. Diagnosis In general, the diagnosis of measles relies on a history of recent exposure and the typical clinical presentation of the disease. However, the diagnosis is more difficult during the prodrome (when the illness is most communicable) or when illness and the exanthema are attenuated by passively acquired measles antibodies. Included in the differential diagnosis are (a) drug eruptions and allergies, (b) rubella, (c) scarlet fever, (d) meningococcemia, (e) roseola, (f) Rocky Mountain spotted fever, (g) toxoplasmosis, (h) enterovirus, and (i) infectious mononucleosis.
Treatment And Prevention The treatment of uncomplicated measles is symptomatic. When otitis media or pneumonia develops, appropriate antibiotic therapy should be instituted on the basis of a Gram stain and culture. Passive immunization is recommended for the prevention of measles in susceptible exposed pregnant women, neonates, and their contacts in the delivery room or nursery. Immune serum globulin (ISG) in a dose of 0.25 mL per kilogram of body weight administered within 6 days after exposure may prevent or at least modify the infection (4,5). Children born to women who have measles in the last week of pregnancy or the first week postpartum should be given ISG as soon as possible in a dose of 0.25 mL/kg (4). Measles, rubella, and mumps vaccines are available as monovalent measles vaccine, monovalent rubella, monovalent mumps, or in various combinations. Each dose of combined or monovalent vaccines contains human albumin, neomycin, sorbitol, and hydrolyzed gelatin. Live measles vaccine and live mumps vaccines are produced in chick embryo cell culture. Live measles vaccine, as a component of measles and rubella or MMR vaccines, should not be given in pregnancy or within 3 months of conception. Women who are given monovalent measles vaccine should not become pregnant within 30 days. Although no evidence exists to substantiate a risk of birth defects, these precautions are based on the theoretical risk of giving a live vaccine in early pregnancy (4). Mumps Mumps is an acute generalized infection with a predilection for the parotid and salivary glands, but that also may affect the brain, pancreas, and gonads. There is no associated rash. The mumps virus is a member of the paramyxovirus family and is thus an RNA virus. Epidemiology Mumps virus is transmitted by saliva and droplet contamination. The virus has been recovered from saliva and respiratory secretions from 7 days before the onset of parotitis until 9 days afterward. The usual incubation period is 14 to 18 days. Mumps is primarily a disease of childhood, and only 10% of cases occur after the age of 15. Many adults are immune as a result of clinical or subclinical infection (one third of cases). However, mumps is much less contagious than measles or chickenpox, and even among susceptible subjects exposed to household members, the attack rate is low. Mumps occurs more frequently in pregnant women than does measles or chickenpox. The incidence in prospective studies has been variously reported as 0.8 to 10 cases per 10,000 pregnancies (1).
Clinical Manifestations The prodrome of mumps consists of fever, malaise, myalgia, and anorexia. Parotitis occurs within 24 hours and is characterized by a swollen and tender parotid gland. The orifice of the Stensen duct is usually red and swollen. In most cases, parotitis is bilateral. The submaxillary glands are involved less often and almost never without parotid gland involvement. The sublingual glands are rarely affected. Mumps is generally a self-limited and complication-free disease. However, it can be a significant cause of morbidity. Orchitis occurs in about 20% of postpubertal men and is the most common manifestation other than parotitis in this group. Oophoritis is far less common. The most common neurologic complication of mumps is aseptic meningitis. However, the course of mumps-associated aseptic meningitis is almost always benign and self-limited. In addition, mumps may cause pancreatitis, mastitis, thyroiditis, myocarditis, nephritis, or arthritis. Maternal Effects Of Mumps Mumps in pregnancy is generally benign and no more severe than in nonpregnant patients. Aseptic meningitis in pregnant patients is neither more frequent nor more severe. Mortality in association with mumps is extremely rare in both pregnant and nonpregnant women. Fetal Effects Of Mumps Retrospective studies have suggested that mumps during the first trimester of pregnancy is associated with a twofold increase in the incidence of spontaneous abortion (2). No significant association between maternal mumps infection and prematurity, intrauterine growth retardation, or perinatal mortality has been demonstrated. Mumps infection in pregnancy is not associated with congenital abnormalities (3). Diagnosis The diagnosis of mumps is usually made on clinical grounds. When there is acute bilateral painful parotitis with a history of recent exposure, the diagnosis is straightforward. It may be more difficult when disease is unilateral or confined to organs other than the parotid gland. In these cases, diagnosis depends on virus isolation or more usually by demonstration of a rising CF, HI, or neutralizing antibody titer in paired acute and convalescent serum. Treatment The treatment of mumps is symptomatic in both pregnant and nonpregnant patients.
Analgesics, bed rest, and application of cold or hot compresses to the parotids are useful. Maternal mumps is not an indication for termination of pregnancy. Prevention The live attenuated mumps virus vaccine has been effective in preventing primary mumps. In susceptible subjects, 95% develop antibodies without clinically adverse reactions. The duration of protection afforded by immunization is not known. Immunization with the live mumps virus vaccine in pregnancy is contraindicated on the theoretical grounds that the developing fetus might be harmed. Although the risk to the fetus seems negligible, the innocuous nature of mumps in pregnancy suggests that any risk from vaccination in pregnancy is unwarranted. Influenza Influenza is an epidemic disease that has been known since antiquity. The influenza viruses are myxoviruses. Three antigenically different influenza viruses have been identified (1). Type A influenza is responsible for most epidemics and is associated with severe cases. Less frequently, type B is involved in epidemics, but it tends to cause milder clinical disease. The third, type C, is the least frequent. Epidemiology Both the frequency and severity of influenza epidemics have been related to antigenic changes in the virus (1). The major antigenic changes that occur at 10- to 30-year intervals are associated with severe infection; the minor antigenic changes that occur annually are not. Two major pandemics with influenza have occurred in this century. The pandemic of 1918 was responsible for 20 million deaths worldwide (2). More recently, the Asian influenza pandemic of 1957 to 1958 caused considerable morbidity and mortality. Epidemics of influenza occur nearly every year during the winter months and are responsible for substantial morbidity and mortality in the United States, averaging approximately 114,000 hospitalizations and 20,000 deaths annually (3,4). Clinical Presentation The incubation type for influenza is 1 to 4 days. Influenza presents clinically with an abrupt onset of a respiratory infection associated with fever, malaise, myalgias, and headache. The severity of the disease varies from mild to severe with pneumonia present. The major portion of the clinical disease lasts, on average, 3 days. Definitive diagnosis is made by virus isolation from throat washings during the acute illness or by serologic confirmation of a fourfold rise in antibody with paired acute convalescent serum. Either CF tests or HI tests may be done.
Maternal Effects Of Influenza The major concern during pregnancy is the increased likelihood of life-threatening pneumonia as a complication of influenza among pregnant women. Reports from the epidemics of 1918, as well as from 1957, all indicate that pregnant women were disproportionately represented among individuals dying of influenza. The maternal mortality rate associated with influenza during the 1918 pandemic was approximately 30% (5,6). In 1919, Harris (6) reported that although the overall maternal mortality rate was 27%, in cases complicated by pneumonia, the mortality rate rose to 50%. Finland (7) noted that during the 1918 pandemic, the worldwide mortality rate was 10% in the general population, but that in some areas, the mortality rate in pregnant women was 80%. During the 1957 pandemic, in Minnesota, 50% of all deaths from Asian influenza among women occurred during pregnancy (8). It is not clear that pregnant women are more likely to develop influenza or that they are more likely to develop influenza pneumonia. However, if influenza pneumonia develops in pregnancy, it appears to be more severe. Deaths among pregnant women with influenza may result not only from secondary bacterial infection (such as with S. aureus, S. pneumoniae, or Klebsiella sp), but also from primary influenza pneumonia without bacterial superinfection. Fetal Effects Of Influenza The effect of influenza on rates of abortion, prematurity, and congenital anomalies is difficult to determine because the evidence is contradictory. In part, confusion may arise from variations of the virus itself from epidemic to epidemic and from lack of well-controlled studies. Studies that did not include serologic confirmation of influenza infection have noted an increased risk for developmental anomalies in pregnancies with a history of influenza. In 1955, Coffey and Jessup (9) (in Ireland) noted that women who gave birth to infants with malformations (neural tube defects primarily) were more likely to have had a history of influenza during pregnancy (18.4%) than mothers delivering healthy babies (3.6%). In a subsequent prospective study, these investigators reported that the malformation rate was more than doubled in pregnancies with a history of influenza at the time of the 1957 Asian flu pandemic (10). Similarly, Doll et al. (11) (in Scotland) noted that congenital anomalies occurred at a higher rate among infants born to women with histories of influenza infection during pregnancy. Hakosalo and Saxen (12) confirmed this association among Finnish women. However, studies using serologic confirmation of influenza infection were not used. Hardy et al. (13) noted a 5.3% incidence of congenital anomalies among women infected during the first trimester; cardiac anomalies were the most common. Griffiths et al. (14) also noted an increase in anomalies, but they all occurred in pregnancies with influenza infection in the second and third trimesters. Wilson et al. (15) noted no increase in anomalies associated with influenza infection in early pregnancy. Similarly, neither the Collaborative Perinatal Research Study (16) nor the study by Brown (17) revealed any association between maternal influenza infection and congenital anomalies among offspring. In summary, most women who have influenza in pregnancy have normal outcomes, and there seems to be little influence on congenital anomalies, intrauterine growth, prematurity, or stillbirth.
Management And Prevention Management of the uncomplicated pregnant woman with influenza consists of symptomatic relief (1), with bed rest, analgesia, liberal fluid intake, and fever control with acetaminophen. The physician must be alerted to the development of pneumonia. If pneumonia occurs in pregnant women with influenza, prompt hospitalization is indicated and broad-spectrum antibiotic coverage (i.e., ceftriakone) for bacterial superinfection pneumonia is required. Respirator support may be needed if there are problems with adequate oxygenation, with retention of carbon dioxide, or with excessively labored breathing. Use of amantadine, which blocks the replication of influenza A virus, has been efficacious in nonpregnant patients, to prevent symptoms, shorten the clinical course, and improve pulmonary function (18). However, the drug has been associated with teratogenic effects in animals and is not recommended for use in pregnancy (19). In years of epidemics, it is generally considered advisable to vaccinate pregnant women. During the 1977 Swine flu vaccination program, however, pregnancy was not considered among the high-risk conditions, such as rheumatic heart and chronic lung disease. Flu vaccines are as immunogenic in pregnant women as in other adults, and no unusual complications have been encountered in pregnant women. The vaccines are killed virus preparations and, thus, safe for use during pregnancy. Prevention of influenza in pregnant women can be achieved with vaccination. See Chapter 25 for detailed discussion of influenza vaccine. The Advisory Committee on Immunization Practices (ACIP) recommends that all pregnant women who will be in second or third trimester during flu season receive influenza vaccine (4). Enteroviruses The enteroviruses consist of three major groups: the polioviruses, coxsackieviruses, and echoviruses. These viruses are a subgroup of the picornaviruses. Enteroviruses are small viruses (18 to 30 nm) with an RNA core. They occur worldwide, both in sporadic and epidemic form, and cause various illnesses (2,3,4,5 and 6). Congenital and neonatal infections have been associated with polioviruses, echoviruses, and coxsackieviruses (1,2,3,4,5,6,7 and 8). Cherry (2) noted that enterovirus infections of the fetus and newborn are more severe than similar infections in older age-groups. He felt that the relatively immature neonatal immune system might explain this phenomenon. Polioviruses Since the introduction of polio vaccines, poliomyelitis is an uncommon disease in Western industrialized nations. Only a brief review of the effects of polioviruses on pregnancy and the neonate is presented. Investigations in the prevaccine era clearly demonstrated that poliovirus infections during pregnancy could result in spontaneous abortion, stillbirth, low-birthweight infants, and neonatal poliomyelitis (1,9,10 and 11).
Although poliovirus can be transmitted across the placenta to the fetus, most pregnant women (nearly two thirds) with clinically apparent poliomyelitis delivered healthy full-term babies (1). There is no evidence that polioviruses are teratogens, and no increase in congenital malformations has been noted (9,10 and 11). Echoviruses A total of 33 echoviruses have been identified (several have been reassigned to other groups of viruses). The echoviruses are responsible for various illnesses in adults and children, including respiratory disease, rashes, gastroenteritis, conjunctivitis, aseptic meningitis, and pericarditis (12). Echovirus infection in pregnancy has not been associated with spontaneous abortions, premature delivery, stillbirths, or congenital malformations (1,2,12,13,14 and 15). There have been reports of neonatally acquired infections caused by many of the echoviruses. The clinical findings associated with neonatal echovirus infection include fever with splenomegaly and lymphadenopathy, macular rashes, diarrhea and vomiting, pneumonitis, otitis media, jaundice, coryza with cough, and septic meningitis (1,2,12). However, congenital echovirus infection can produce severe disease and damage to the neonate. Echovirus 14 was reported to be the cause of a febrile illness that developed at 3 days of life and progressed to cyanotic episodes, hypothermia, hepatomegaly, bradycardia, and purpura, with resultant death at 7 days of life (15). Echovirus 19 was reported to have caused hepatic necrosis and massive hemorrhage in three infants (17). No specific treatment or vaccines are available for echovirus infections. Coxsackieviruses The coxsackieviruses are divided into two major groups. Group A coxsackievirus contains 23 types, and group B includes 6 types. Group A coxsackieviruses do not cause significant perinatal illness, except in rare cases. Group B coxsackieviruses can cause pleurodynia, meningoencephalitis, and myocarditis. Hepatitis, the hemolytic uremic syndrome, and pneumonia are infrequent but severe manifestations of group B coxsackievirus infection. Transplacental transmission of group B coxsackievirus has been demonstrated (18,19,20 and 21). However, the magnitude of the risk to the fetus has not been defined. Most maternal group B coxsackievirus infections result in no demonstrable adverse effects on the fetus. No evidence demonstrates a role for coxsackieviruses in spontaneous abortion (15) or preterm labor and delivery. In a study involving nearly 23,000 pregnancies, Brown and Karunas (13) reported that coxsackieviruses B2, B3, B4, and A9 had a positive correlation between maternal infection and neonatal anomalies (13). Coxsackievirus B4 infection in the first trimester has been associated with urogenital malformations, such as hypospadias, epispadias, and cryptorchidism. Coxsackievirus A9 maternal infection was associated with digestive tract anomalies, and types B3 and B4 were associated with cardiovascular defects. There was an association between the B1 through B5 coxsackievirus group and congenital heart disease. In addition, Brown and Karunas (13) found no correlation
between reported maternal illness and serologic evidence of infection in the offspring. Thus, even asymptomatic maternal coxsackievirus infection may result in fetal maldevelopment. Congenital coxsackievirus infection within 48 hours of birth is a rare occurrence (1). Neonatal infection with coxsackievirus can be acquired from mother, nursing personnel, or infected babies in the nursery. Many studies have documented the variety and severity of neonatal coxsackievirus infection, particularly types B1 through B5. Most have focused on myocarditis and CNS infection (1,2). Myocarditis seems to be a particularly prominent manifestation of group B coxsackievirus infection in the neonate. Diagnosis of coxsackievirus infection is based on virus isolation from throat or rectum and serologic evidence of increasing antibody titer during the convalescent period. HI or CF tests may be performed. Condylomata Acuminata And Human Papillomavirus Infection The condylomata acuminata anogenital warts are caused by infection due to human papillomavirus (HPV) (1). Today, they are of great interest because of a dramatic increase in frequency and their implication in genital tract malignancy and juvenile and respiratory papillomatosis. HPV has not yet been cultured but is known to be a DNA virus. There are more than 60 types of HPV, which have been classified by risk of neoplasia. Low-risk types are 6, 11, 42, 43, 44, 53, 54, and 55; high-risk types are 16, 18, 45, and 56; and intermediate-risk types are 30, 31, 33, 35, 39, 51, 52, 58, and 66 (2). Epidemiology From 1970 to 1985, there was a fourfold increase in this infection. The prevalence of HPV in the genital tract varies widely, depending on the detection technique used and the population tested. In patients with normal cervical cytology and examination, the rate is as low as 6%, but in women with cervical neoplasia, it is detected in more than 60% (2). It is widely believed that condylomata may proliferate and become friable in pregnancy (3), perhaps because of the “immunosuppressive” state of pregnancy. Lesions may cause soft tissue dystocia, extreme discomfort, difficulty in urinating or defecating, and hemorrhage with vaginal delivery (4). HPV infections, most likely, are spread by direct skin-to-skin contact. The major mode of spread is sexual activity. However, HPV DNA has been detected on underwear and surgical instruments, as well as on swabs in 20% of virginal college-age women when sensitive polymerase chain reaction (PCR) techniques were used (2). Thus, other modes of transmission—other than sexual—are possible. It is estimated that contagion is relatively high. Condylomata acuminata have been reported on an infant at birth. Respiratory papillomatosis results in warts of the larynx and trachea. In about 60% of juveniles with this condition, genital condylomata were present in the mother at delivery. HPV infection has a relatively long, though variable, incubation period, with a range of 3
to 8 months. As with other sexually transmitted diseases (STDs), warts may occur in concert with gonorrhea or other STDs. Most cases occur between ages 16 and 25. Diagnosis The classic wart is a soft excrescence, several millimeters in diameter, often taller than it is wide. They may occur singularly or in clusters. Occasionally, there are “giant” condylomata up to 3 cm or more in diameter. These are soft and round, with a pebbled surface. Warts are commonly located in moist areas and are often found in the vaginal introitus, vagina, and vestibule. Minute flat warts, often visible only with the colposcope, are common on the cervix. Often, the diagnosis is made clinically. Condylomata lata should be ruled out, by use of a syphilis serologic test. In suspect or atypical lesions, biopsy may be necessary to rule out malignancy. Juvenile Respiratory Papillomatosis Respiratory papillomatosis may be of either juvenile or adult onset. About one third to one half of cases are evident by age 5. With the vocal cords, the most common site, hoarseness is the most frequent symptom. Occasionally, papillomas may produce respiratory obstruction. Because of the historic similarity between respiratory papillomas and genital warts, an HPV etiology had been suspected in the former. Recent studies have established that warts of these two locations are virologically indistinguishable, with 90% from both locations being HPV-6 or HPV-11 (5). Perinatal transmission has been the subject of recent study. In approximately 60% of mothers with infants afflicted with respiratory papillomatosis, there is detection of HPV DNA (2). Although it has been estimated that 2% to 5% of all births in the United States are at risk for perinatal HPV exposure, juvenile respiratory papillomatosis remains are (approximately 1,550 cases per year; a rate of 0.04% of all vaginal deliveries) (2). It is speculated that its perinatal transmission may occur by ascension into the uterus or by direct contact during birth. To estimate the rate of transmission during vaginal delivery, Shah et al. (5) examined the frequency of cesarean delivery in 109 cases of respiratory papillomas (onset before the age of 14). It was presumed that cesarean delivery, before rupture of membranes, would prevent intrapartum transmission but not prevent other modes. Because of the rarity of the condition, cases were collected from three sources. All cases were biopsy proven. Only 1 (1%) of the 109 cases had a history of cesarean delivery. On the basis of national cesarean section rates in relevant years, the authors estimated that there would have been ten cesarean births in these 109 patients. The child developing respiratory papillomatosis after cesarean delivery was born by section before membrane rupture. The authors concluded that in juvenile-onset respiratory papillomatosis, mother-to-infant transmission occurs most often during vaginal delivery but that in utero infection is also
possible. Further, in view of the frequency of genital warts and the rarity of juvenile-onset respiratory papillomatosis, the authors estimated the risk of developing disease for a child born to an infected mother. On the basis of crude annual rates of births and new cases of respiratory papillomatosis, the estimate was one in several hundred (range, 1 : 80 to 1 : 1,500). Perinatal transmission of HPV DNA has been evaluated in several recent works. Using a PCR assay, Puranen et al. (6) assessed 106 infants born by vaginal delivery or by cesarean section and their 105 mothers. Both mothers' and infants' samples were positive for the same type of HPV in 29 mother-infant pairs. It was noted that five infants born by cesarean delivery were found to have the same type as the mother. Overall concurrence between HPV types and her newborn was 69% (29 of 42). HPV DNA was found in 39 (37%) of 106 infants. It should be noted that infant specimens were taken immediately after birth from a nasopharyngeal aspirate (6). In a study of similar design in the Republic of China, specimens from the neonate were collected by swab from the buccal mucosa and from the genitalia, on day 3 or 4 of life. The investigators found that the overall frequency of HPV transmission from HPV-16–positive or HPV-18–positive mothers was 39% (27 of 68). A significantly higher rate of transmission was found when infants were delivered vaginally than by cesarean delivery (51.4% vs. 27.3%; p = 0.042) (7). Both of these studies assessed newborns very early in the newborn period. On the other hand, Watts et al. (8) found much lower transmission rates when infants were followed for longer intervals. During pregnancy, 112 (74%) of 151 women had evidence, by history, clinical, or DNA testing, of genital HPV infection. At nearly 500 infant visits, HPV DNA was detected with low prevalence, from only 5 of 335 genital specimens (1.5%), from 4 of 324 anal specimens (1.2%), and from none of 372 oral or nasopharyngeal specimens. The rate of isolation did not vary over the interval of follow-up from 6 weeks to 36 months. Although the authors were not able to rule out perinatal transmission, they observed that the upper 95% CI for detection of perinatal transmission from women with evidence of human genital HPV infection was only 2.8% (8). Based on the use of cesarean section to prevent neonatal herpes when the mother is infected with herpes at labor, delivering babies of mothers with genital warts by cesarean section might be suggested. However, authors including the American College of Obstetricians and Gynecologists and CDC do not support that position, and we know of no current data to recommend that strategy (2,3). First, the risk of transmission of the disease is low (probably ten times lower than with herpes). Second, cesarean delivery does not offer complete protection, and we do not know exactly what protection is offered. Third, it is likely that the risks of cesarean section for all women with any genital warts would outweigh the potential benefits (2,5). Treatment Of Condylomata Acuminata In Pregnancy General predisposing features, such as other vaginitis, should be corrected, and the genital area should be kept clean with avoidance of excess moisture. In addition to the concern for transmission of HPV to the neonate, warts in pregnancy may lead to tearing or bleeding during delivery, particularly when they are large or
extensive. Thus, genital warts in pregnancy should be treated when recognized in pregnancy. Many have noted that condylomata acuminata may thrive in pregnancy, perhaps due to a relative maternal immunosuppression. Specific treatment considerations of genital warts should be modified in pregnancy. Podophyllum is best avoided in pregnancy, as it may be absorbed and can be toxic to the fetus. Large doses of podophyllum (7.5 mL of 25% solution, containing 1.88 g of podophyllum) used to treat florid vulvar warts were associated with fetal death 2 days later (9). Maternal neurologic and respiratory symptoms were also reported in this overdose. In animals, podophyllum is both teratogenic and neurotoxic (9). Congenital anomalies were reported in the infant of a woman who ingested podophyllum tablets. A purified preparation, 0.5% podofilox, is available for treatment of external warts, but this preparation should not be used on the vagina or cervix or in pregnancy (2). Topical 5-fluorouracil (5-FU) is contraindicated in pregnancy (2). Several years ago, there were individual case reports of multiple congenital anomalies associated with use in the first trimester and of “neonatal intoxication” from 5-FU used at midpregnancy. More recently, there have been case reports of periconceptional 5-FU exposure without adverse effects, but these few cases do not exclude a high potential risk (10,11). Interferons have been used to treat persistent condylomata, but with varying effectiveness. There has been little use of these agents in pregnancy, and they are considered contraindicated in pregnancy (2). Topical trichloracetic acid (85%) or bichloracetic acid may be used in pregnancy. They work best on moist mucosal warts. Cure rates from a single application are limited (estimated at 20% to 30%, except for very small lesions), thus necessitating re-treatment every 7 to 10 days (4). Trichloracetic acid has also been combined with laser treatment for condylomata in pregnancy (4). Condylomata in pregnancy may also be treated with cryosurgery when they occur in the vagina or on the vulva. In 1987, Berman et al. (12) reported a series of 28 pregnant women treated with cervical cryotherapy. They reported favorable responses with no adverse effects on the pregnancies (12). For isolated large condylomata, surgical excision would be appropriate in a pregnant woman. Treatment may be carried out with electrocoagulation with curettage (13). A recent addition to the armamentarium for treating HPV is imiquimod (Aldara Cream), an immune response modifier. It is available as a 5% cream with applications to be made by the patient three times per week for up to 16 weeks until the warts are gone. The exact mechanism of action is unknown, although it does not have direct antiviral activity in cell culture. It is presumed that imiquimod induces cytokines such as interferon alpha, interleukins, and tumor necrosis factor. After application of imiquimod cream, there is minimal systemic absorption, and the cream is generally well tolerated. Less than 2% of patients discontinued therapy because of adverse effects, but the most common application site reactions were itching (32%), burning (26%), pain (8%), and soreness (1%). Imiquimod 5% cream is rated as pregnancy category B by the FDA; however, there are no adequate controlled studies of
use of this preparation in pregnancy (14). Human Parvovirus The human parvoviruses are a family of DNA viruses, of which human parvovirus B19 is the only known human pathogen (1,2). Parvovirus B19 is a small (12-nm) single-stranded DNA virus. Human parvovirus was discovered in 1975 and in 1981 was shown to be an etiologic agent of transient aplastic anemia (3,4). The virus was identified as the etiologic agent of erythema infectiosum (fifth disease), a common childhood viral exanthem in 1983 (5,6). The virus has a predilection for the hematopoietic system and is cytotoxic for erythroid progenitor cells. The most common manifestation of human parvovirus B19 is erythema infectiosum (fifth disease). Parvovirus has also been implicated in aplastic crises in patients with chronic hemolytic anemia. In 1984, after epidemics of erythema infectiosum, parvovirus B19 was shown to cause in utero infection and nonimmune hydrops of the fetus (7,8). Human parvovirus B19 is worldwide in distribution and is most common in young children, ages 5 to 14 years. Parvovirus B19 is a very common infection in which seroprevalence of antibody to parvovirus B19 IgG is age dependent. As reviewed by Anderson (1), the reported prevalence ranges, by age, are as follows: (a) 1 to 5 years, 2% to 15%; (b) 5 to 19 years, 15% to 60%; and (c) adults, 30% to 60%. It is estimated that 50% to 75% of reproductive-age women are immune to parvovirus B19, based on serologic evidence of previous infection. Clinical disease may be epidemic or sporadic. The most common manifestation in children is erythema infectiosum, which has a winter and spring seasonality. In children, disease tends to be clinically mild and presents with a low-grade prodromal fever that precedes to the development of a very characteristic rash; the rash is an erythematous, warm “slapped cheek”–like facial rash. This is followed by a morbilliform rash of the extremities. In adults, the disease usually is asymptomatic. However, in clinical cases, it presents with fever, adenopathy, arthralgias, and mild arthritis, particularly of the hands, wrists, and knees. This polyarthropathy occurs in 50–80% of infected adult women (9). Rash is usually not present in adults. In the United States, 50% to 75% of women in the reproductive age-group are immune and demonstrate antibodies against human parvovirus B19. The organism is a highly infectious agent and 60% to 80% of susceptible household contacts will become infected when exposed to childhood disease. Among seronegative schoolteachers and day care workers who are at risk during epidemics, 20% to 30% will develop the disease (10). Among susceptible teachers, two factors were shown to be associated with an increased risk of parvovirus B19 infection (10): (a) teachers of young children and (b) exposure in the classroom to many children with rash. Bell et al. (11) reported that 36% and 38% of susceptible health care workers became infected after exposure to children with aplastic anemia. Recently, Adler et al. (12) identified risk factors for parvovirus B19 infection for hospital and school employees during nonepidemic periods. In the absence of an epidemic of parvovirus B19 infection, the annual seroconversion rate for hospital workers was 0.42% and that for school employees was 2.93%. This compares with
secondary attack rates of 20% to 40% during epidemic periods. Their data demonstrated that contact with elementary school-aged children at home or at work was the most important risk factor for parvovirus B19 acquisition (12). Individuals had approximately a twofold greater risk of acquiring parvovirus B19 from children at work than at home (12). However, they showed a low seroconversion rate among hospital employees without known contact with children. This is in contrast to exposure of hospital workers to patients with aplastic anemia (11). Thus, Adler et al. (12) recommended that pregnant women with unknown or susceptible serologic status should not care for patients with hemolytic anemia and fever or aplastic crisis. Recently, Valeur-Jensen et al. screened nearly 31,000 pregnant women in Denmark and noted that 65% had evidence of past Parvovirus B19 infection (13). Annual seroconversion rates among susceptible women during endemic and epidemic periods were 1.5% (95% CI 0.2–1.9) and 13% (95%) CI 8.7–23.1) respectively (13). Baseline seropositivity was significantly associated with increasing number of siblings, having a sibling of the same age, number of own children, and occupational parvovirus infection increased in parallel with the number of children in the household. Compared with other pregnant women, nursery school teachers had a 3-fold increased risk of acute infection (OR 3.09%; 95% CI 1.62–5.89) (13). These authors etimated that the population-attributable risk of seroconversion was 55.4% for number of own children and 6% for occupational exposure (13). Thus, the risk of infection is high for susceptible pregnant women during epidemics and is associated with the level of contact with children (13). Nursery school teachers had the highest occupational risk, but most infections appear to result from exposure to the woman's own children (13). The diagnosis is generally made on clinical grounds, particularly in children. Antibody tests for parvovirus B19 have become available. An ELISA method is currently recommended for IgG- and IgM-specific antibodies to human parvovirus B19 (9). Acute parvovirus infection can be documented by a combination of clinical symptoms and signs of parvovirus B19 infection and the presence of parvovirus B19 IgM antibodies, the virus (e.g., by culture, PCR, in situ hybridization), or typical histologic changes in red blood cell precursors (14,17). Parvovirus B19 IgM antibody appears within a few days of onset of illness (13). The gold standard test for detection of anti-B19 IgM antibody is the IgM antibody capture radioimmunoassay (MACRIA) (18). The most sensitive tests for detection of virus include PCR for parvovirus B19 DNA and nucleic acid hybridization for parvovirus B19 DNA (13). The fetus has also been shown to be at risk for parvovirus B19 infection. The risk to the fetus has received tremendous attention over the last decade since initial reports in the late 1980s suggested that during pregnancy, human parvovirus B19 may result in fetal hydrops and stillbirth (19,20,21 and 22). The pathogenesis of this disease revolves around the predilection of human parvovirus B19 for human erythroid progenitor cells. With intrauterine transmission to the fetus, the fetus is at risk for developing aplastic anemia secondary to this predilection. The rapidly expanding red blood cell volume, a shorter half-life of red blood cells in the fetus, and an immature immune system make the fetus particularly susceptible to the effects of parvovirus B19. In general, fetal hydrops occurs 4 to 6 weeks after maternal infection, with a range of 1 to 12 weeks. Parvovirus B19 causes erythroid hypoplasia, shortened red blood cell survival, and hemolysis. The resultant anemia leads to high-output cardiac failure, which results in
fetal hydrops. Although early reports suggested a high rate of transmission to the fetus (38%) and a significant morbidity to the fetus and mortality with human parvovirus infection, more recent work has demonstrated that maternal infection produces no adverse effects on the fetus in most cases. Early small cases studies suggest a high rate of transmission to the fetus. Woernle et al. (20) reported that 4 of 12 pregnant women at risk for parvovirus B19 were IgM positive. In this group, one of the four fetuses was a stillbirth due to hydrops. Previous reports had described 22 pregnant women with serologic evidence of parvovirus B19 infection, in whom nine fetal deaths occurred. In 1988, Rodis et al. (21) reviewed 37 reported cases of women exposed and infected during pregnancy. In this retrospective study, they reported that 14 of the pregnancies (38%) had adverse outcomes, including spontaneous abortions, intrauterine fetal death, and congenital anomalies. Eleven of the 37 fetuses had intrauterine nonimmune fetal hydrops. Similarly, in another retrospective study, Schwartz et al. (23) reported that 10 (26%) of 39 susceptible pregnancies developed fetal hydrops; 7 of these infants died in utero, and 3 who had received intrauterine transfusions survived. More recent prospective studies have cast doubt on this high rate of fetal transmission. In England, the Public Health Laboratory Service Working Party on Fifth Disease reported a fetal death rate of 16% in a large series of 190 pregnancies; however, only 6 of the 14 fetuses tested were DNA positive for human parvovirus B19 (24). These authors reported a fetal loss rate of 14% (1 of 14) after 20 weeks of gestation and a 17% loss before 20 weeks, which is similar to that in the general population (24). Moreover, if the documented rate of parvovirus B19 infection (6 of 14) is applied to all 30 fetal deaths, an upper limit estimate for parvovirus B19–related fetal death among infected women would be 9.7% (24). In a prospective study undertaken during a 1988 epidemic of erythema infectiosum in Connecticut, Rodis et al. (25) reported that among 39 pregnant women with evidence of recent infection during pregnancy, 37 (95%) delivered healthy newborns and only 2 (5%) had spontaneous abortions, of which only 1 was related to human parvovirus B19. No infants developed fetal hydrops in this series. Similarly, in the United States in a large prospective study that screened 3,526 pregnant women, the CDC (21) concluded that although the risk for fetal death was increased, particularly in the first 20 weeks of gestation, this did not reach statistical significance (Table 16.15). There was a 5.9% fetal death rate among infected women, compared with 3.4% among controls (26). The CDC noted no increase in preterm delivery; birthweights were similar in those infected with parvovirus B19 and in controls; and no increase in birth defects occurred among the infants infected with parvovirus B19. Table 16.15 summarizes the findings of prospective studies assessing the effect of maternal parvovirus B19 infection on pregnancy outcome. Overall the risk for fetal death was 7% and the risk for non-immune hydrops 1% (24,25,26,27,28,29 and 30).
TABLE 16.15. PREGNANCY OUTCOMES AMONG WOMEN SCREENED FOR HUMAN PARVOVIRUS B19 INFECTION
Women exposed to erythema infectiosum during pregnancy should be screened serologically for IgG and IgM antibodies against parvovirus. If IgG antibody is present, this denotes immunity against the disease, and these patients can be reassured. Those who are IgG negative and IgM negative are susceptible to infection and should be cautioned to reduce their risk of exposure, particularly if they are schoolteachers or work in day care centers during epidemics of fifth disease. The women who are IgG negative and IgM positive and in whom acute parvovirus infection is thus confirmed should be closely evaluated and monitored for the development of intrauterine fetal hydrops (Fig. 16.6). For women with a documented infection, maternal serum a fetoprotein (AFP) determinations (31,32) and diagnostic ultrasound examinations are used to detect the development of hydrops. In general, maternal serum AFP level will be elevated before the onset of hydrops, as demonstrated by sonography. Serial maternal serum AFP determinations are obtained. If the maternal serum AFP level becomes elevated, serial ultrasound examinations are obtained. More commonly, only serial ultrasound screening for detection of hydrops is utilized. Ultrasounds should be repeated weekly for up to 12 weeks. Hydrops usually occurs 4 to 6 weeks postinfection.
FIGURE 16.6. Suggested protocol for management of pregnant women exposed to parvovirus B19.
Once intrauterine evidence for hydrops develops, two alternatives are available. Until recently, intrauterine blood transfusion, usually via percutaneous umbilical blood sampling, has been proposed for the treatment of the hydropic fetus with parvovirus B19–induced anemia and has been very successful (23,32,33,34,35 and 36). However, several recent studies have noted the occurrence of fetal hydrops in association with acute parvovirus infection that resolved spontaneously over 4 to 6 weeks without intrauterine transfusion (37,38). Thus, the role of intrauterine blood transfusion in preventing intrauterine stillbirth due to hydrops secondary to aplastic anemia remains unclear. Possibly, the younger fetus, with less than 20 to 22 weeks of gestation, is at risk and requires intrauterine transfusion, whereas the older fetus, in the latter part of the second trimester, with a more functional immune system, may be better able to tolerate the insult from the parvovirus infection. Fairley et al. (39) demonstrated the benefit of intrauterine transfusion in a series of 66 cases of fetal hydrops in England and Wales. Among the 38 fetuses alive at the time of initial ultrasound, 3 of 12 transfused and 13 of 26 not transfused died in utero. After adjustment for severity of hydrops (ultrasound) and gestational age, the risk of death among those transfused was significantly reduced (O.R. 0.14; 95% C.I. 0.02–0.96) (39). No vaccine or treatment is available for human parvovirus infection. Thus, prevention by avoiding exposure to erythema infectiosum is the only method available. As discussed in previous sections, parvovirus B19 is highly contagious. High-risk groups for exposure and acquisition of infection are day care workers, elementary schoolteachers, hospital employees caring for patients with aplastic anemia, and household members with an infected child. As reviewed by Adler (12), secondary attack rates for high-risk susceptible women are 20% to 40% during epidemic periods and 2% to 5% during endemic periods. Although these are worrisome rates, available data now show that the risk to the fetus after maternal infection is low (12,19,20 and 21). Adler (12) calculated (assuming that on average 50% of pregnant women are susceptible, during an endemic period 1% to 4% of susceptible women acquire infection, and the rate of fetal death after maternal infection is 5% to 10%) that the occupational risk of fetal death due to parvovirus B19 infection for a pregnant woman with unknown serologic status is between 1 of 500 and 1 of 4,000. Thus, during endemic periods, these rates are so low that intervention such as serologic screening of pregnant women or furloughing or temporarily transferring pregnant seronegative employees to non–child care areas is not justified (12). On the other hand, during epidemic periods, when infection rates may be 5- to 20-fold higher, serologic testing or temporary transfer of pregnant employees may be appropriate (12). However, for most women, the source of fetal parvovirus B19 infection is exposure to school-aged children at home. No good method of prevention is applicable to this
situation. Thus, to a large extent, prevention of maternal parvovirus must await development of an effective vaccine that can be administered to infants.
BACTERIA Syphilis The epidemiology, clinical presentation, diagnosis, and treatment of syphilis is discussed in detail in Chapter 7 (Sexually Transmitted Diseases). Treatment Penicillin is the drug of choice for the treatment of syphilis in nonpregnant patients and pregnant women. The recommended treatment schedule suggested by the CDC is presented in Table 16.16.
TABLE 16.16. CENTERS FOR DISEASE CONTROL AND PREVENTION–RECOMMENDED TREATMENT OF SYPHILIS IN PREGNANCY (2001)
Current CDC recommendations suggest that patients with a history of allergy to penicillin should be tested to confirm the allergy and then desensitized to penicillin and treated with penicillin in the dosage schedule appropriate for the duration of syphilis (Chapter 7, Sexually Transmitted Diseases). Listeriosis Listeriosis is an infection caused by Listeria monocytogenes, a facultative, motile non–spore-forming Gram-positive rod. Although seven species of Listeria have been identified, L. monocytogenes is the principal pathogen in humans (1). L. monocytogenes
is indistinguishable from diphtheroids on morphologic grounds and thus is often discarded as a contaminant by clinical laboratories (1). Sixteen serotypes (serovars) of L. monocytogenes have been identified, of which three (i.e., 46, 1/2b, 1/2a) are responsible for more than 90% of clinical infections (1). Although the organism is easily isolated from normally sterile sites (e.g., placenta, amniotic fluid, and blood) on routine culture media, it can be difficult to recover from mixed cultures such as vaginal or cervical specimens (1). Use of selective media and enrichment techniques will improve the yield from such sites. Listeriosis causes an estimated 2500 serious illnesses and 500 deaths annually in the United States. Patients who are immunocompromised and pregnant women and their newborns are particularly susceptible to infection with L. monocytogenes. Of concern to the obstetrician is the association between maternal listerial infection and preterm labor and fetal infection. High perinatal morbidity and mortality rates have been reported for listerial infection in pregnancy (2,3,4,5 and 6). As with group B streptococcal infection, neonatal listeriosis has been divided into two serologically and clinically distinct areas. Serotypes Ia and IVb have been associated with early-onset listeriosis. Early-onset disease takes the form of a diffuse sepsis with multiorgan involvement including pulmonary, hepatic, and the CNS. Early-onset listeriosis is associated with high stillbirth and neonatal mortality rates. Early-onset listerial infection appears to occur more frequently in low-birthweight infants. Like group B streptococcus infection, late-onset listeriosis presents as meningitis. These infants are usually full-term and born to mothers with uneventful perinatal courses. Neurologic sequelae such as hydrocephalus or mental retardation are common with late-onset disease. In addition, a mortality rate approaching 40% is reported. Although Charles (2) suggested that an ascending route of infection from cervical colonization with L. monocytogenes (even across intact membranes) plays a role in the pathogenesis of neonatal infection, the more important and common route of infection is secondary to maternal infection, leading to placental infection that then leads to fetal septicemia and multiorgan involvement in the fetus. It is felt that amniotic fluid becomes infected with Listeria secondary to excretion of the organism via fetal urination. Aspiration and swallowing of infected amniotic fluid may then result in respiratory tract involvement of the infection in the fetus. Human listeriosis presents in both epidemic and sporadic forms. The epidemic orm has clearly been associated with contamination of food and food products. Since 1981, four major outbreaks of listeriosis (three in North America and one in Switzerland) have occurred, which clearly document food-borne transmission in epidemic listeriosis (7,8,9 and 10). The first was a 1981 outbreak of listeriosis in Nova Scotia (7). In this outbreak, 34 perinatal cases and 7 adult cases were identified. The case fatality rate for infants born alive was 27%, and there were 19 intrauterine deaths (7). In the Massachusetts outbreak of 1983, the epidemiology was different, with most cases (42 of 49) occurring in immunocompromised adults and only seven cases of perinatal listeriosis were reported (8). In both adult and neonatal cases, the case fatality rate was 29% (8). In Los Angeles, from January 1, 1985, through August 15, 1985, 142 cases of human listeriosis were reported in association with ingestion of cheese contaminated by L. monocytogenes. Two thirds of these infections occurred in pregnant women or their
offspring, and 30 of the 48 deaths in this epidemic occurred in fetuses or neonates (9). The case fatality rate for the 49 nonpregnant adults was 33% (9). In 1988, an epidemic of listeriosis was reported in Switzerland, involving 122 cases (64 perinatal and 58 nonpregnant adults); the case fatality rate was 28% (10). In a recent report from Finland of 74 cases of systemic listeriosis from 1971 to 1989, a mortality rate of 30% was noted among the neonates infected with L. monocytogenes (11). Among the six pregnant women with listerial infection, five had fetal complications: three spontaneous abortions and two preterm deliveries. In the presence of underlying disease, the mortality rate in adults was 32% (11). All healthy nonimmunocompromised nonneonatal patients survived the listerial infection. Cherubin et al. (12) reviewed more than 120 cases of listeriosis from four medical centers in three geographically separated sites (Los Angeles, Nashville, and Chicago). Although in Los Angeles, more than two thirds of cases occurred during the perinatal period, at Vanderbilt University Hospital in Nashville, most cases occurred in older adults who had received organ transplants (Table 16.17) (12). In Chicago, an intermediate pattern with a more equal mix of perinatal and adult cases of meningitis, bacteremic, or neurologic infections was seen (12). These authors identified risk factors for listeriosis (Table 16.18). Among the most important are pregnancy, neonatal status, organ transplantation, malignancy, renal failure, systemic lupus erythematosus, steroid therapy, and HIV infection (12). These authors stressed that mortality associated with listeriosis occurred predominantly among premature infants and stillbirths delivered by infected pregnant women (12). The earlier the stage of gestation when infection occurred, the higher the risk for fetal death, with a mean gestational age of 25.9 weeks for fetuses or neonates who died, compared with 33.7 weeks for surviving infected infants (12). Mortality rates were markedly decreased among neonates (particularly term) and adults (12). These studies demonstrate that Listeria is prone to adversely affect immunocompromised adults and fetuses or neonates with immature immune systems.
TABLE 16.17. SUMMARY OF CASES AND MORTALITY FOR LISTERIOSIS BY GEOGRAPHIC SITE IN THE UNITED STATES
TABLE 16.18. RISK FACTORS FOR LISTERIOSIS
More commonly, listeriosis occurs sporadically (sporadic or endemic listeria) (1). Two studies by the CDC have established a baseline incidence for sporadic listeriosis in the United States (13,14). Using hospital discharge data from the Commission on Professional and Hospital Activity, Ciesielski et al. (13) reported that from 1980 to 1982, there were an estimated 800 cases of listeriosis annually, for an incidence of 3.6 cases per 1 million population and a minimum of 150 deaths, for a case fatality rate of 19%. Attack rates were noted to be highest among newborns (4.7 per 100,000 livebirths per year) and those 70 years or older (11 per 1 million population). Using a population-based active-surveillance study, the CDC estimated that approximately 1,700 cases of listeriosis occurred in the United States in 1986 (14). This computes to an overall annual incidence rate of 7.1 cases per 1 million population. Among perinatal cases, the attack rate was 12.4 cases per 100,000 livebirths, and for nonpregnant adults, 5.4 cases per 1 million population (14). Among persons older than 70 years, the incidence was 21 cases per 1 million population (14). In nonperinatal cases, the overall fatality rate was 35%, ranging from 11% in persons younger than 40 years to 63% in persons older than 60 (14). In summary, listeriosis is estimated to cause at least 1,700 serious infections and contribute to approximately 450 deaths and 100 stillbirths annually in the United States (1). It is unclear what role food-borne transmission plays in sporadic listeriosis. Schwartz et al. (15) hypothesized that such is the case. Individuals infected with Listeria from contaminated food may be asymptomatic and become chronic carriers of the organism in their gastrointestinal tract (16). Subsequently, such female organisms could colonize the vagina and cervix with L. monocytogenes. Unfortunately, many pregnant women with listerial infection remain asymptomatic. When symptomatic, a flulike syndrome is characterized by fever, chills, headache, malaise, myalgia, back pain, and upper respiratory complaints. Associated gastrointestinal symptoms (e.g., diarrhea and abdominal cramping) are less common. These symptoms, or prodrome, occur in about two thirds of cases (17). This prodrome represents the bacteremic stage of listeriosis and is probably the time during which the
uterine contents are seeded (1). Within 3 to 7 days, this prodrome can progress to amnionitis with resultant preterm labor and delivery, septic abortion, in utero fetal infection, stillbirth, or neonatal infection. Maternal infection tends to be mild and not associated with significant maternal morbidity. On occasion, diffuse sepsis may occur. Listeriosis occurs most frequently during the third trimester (1). Unfortunately, no specific clinical manifestations help distinguish listeriosis from other infections that may occur during pregnancy. Thus, pregnant women in the late second or early third trimester with the above-described prodrome need to be assessed for possible listeriosis. Unlike maternal infection, neonatal listeriosis is commonly associated with morbidity and mortality, with fatality rates ranging from 3 to 50% in liveborn infants (14,17). As in group B streptococcus infection, two distant neonatal patterns of listerial infection occur—early and late onset. Early-onset neonatal listeriosis occurs in infants infected in utero before the onset of labor. Usually the mothers of these infants experienced the flulike prodrome (1). However, ascending infection from a colonized lower genital tract has also been documented (2,18). Early-onset infection manifests within the first few hours to few days of life. Widely disseminated granulomas are typical in severe newborn disease acquired in utero. The granulomas are found most commonly in the placenta and liver. Such granulomas may be clues to the diagnosis when neonatal sepsis presents within the first few days of life. Late-onset neonatal infection due to L. monocytogenes usually occurs in full-term infants from uncomplicated pregnancies (1). Such infants appear healthy at birth and do not manifest their infection until several days to several weeks after delivery. With late-onset listerial infection, meningitis is more common than sepsis. It is assumed that postpartum acquisition is the route for late-onset infection with L. monocytogenes. Possibly, the organism is transmitted during passage of the fetus through the birth canal. Nosocomial transmission of L. monocytogenes has also been documented as a cause of late-onset neonatal listeriosis (1). Because of the high mortality rate associated with both early- and late-onset neonatal listerial infection, it is crucial that the obstetrician maintain a high index of suspicion that any febrile illness in pregnancy may be due to L. monocytogenes. In such patients, cervical and blood cultures should be obtained for L. monocytogenes as soon as possible. Because colonies of L. monocytogenes may be mistaken on Gram stain for diphtheroids and thus ignored, it is important to inform the microbiologist that Listeria is a concern. In febrile pregnant women, a Gram stain revealing Gram-positive pleomorphic rods with rounded ends is very suggestive of L. monocytogenes. Penicillin G and ampicillin are effective in vivo against L. monocytogenes. Current opinion holds that optimum therapy includes a combination of ampicillin and an aminoglycoside. Maternal treatment consists of ampicillin (1 to 2 g intravenously every 4 to 6 hours) and gentamicin (2 mg/kg intravenously every 8 hours). For the newborn, the ampicillin dosage is 200 to 300 mg/kg per day administered in four to six divided doses. Treatment is generally provided for 1 week. A recent report suggested that with
documentation via amniocentesis of intrauterine listerial infection, antibiotic treatment without immediate delivery may be successful and result in a normal healthy fetus (19). Guidelines for prevention of listeriosis are similar to those for preventing other food bourne illnesses (20). The general recommendations are: 1) cook thoroughly raw food from animal sources; 2) wash raw vegetables thoroughly before eating; 3) keep uncooked meats separate from vegetables and from cooked foods and ready-to-eat foods; 4) avoid raw (unpasteurized) milk or foods made from raw milk; and 5) wash hands, knives, and cutting boards after each handling of uncooked foods (20). Tuberculosis In just the last few years, there has been renewed interest in tuberculosis because of an increasing number of cases in women of reproductive age, reports of new cases of congenital tuberculosis, and development of multidrug resistance. In June 2000, the CDC revised its recommendations for tuberculin testing and treatment of latent tuberculosis infection (1). Epidemiology Between 1985 and 1992, the number of cases in reproductive-age women increased 41% (2). In two hospitals in New York City, there was a dramatic increase among pregnant women—from 12.4 cases per 100,000 deliveries (five cases in 6 years) in 1985 to 1990 to 94.8 cases per 100,000 deliveries (11 cases in 2 years) in 1991 to 1992 (3). This increase parallels a national increase in tuberculosis cases (4). From 1985 through 1991, the number of reported tuberculosis cases increased 18%, largely due to the HIV epidemic, deterioration of the health care infrastructure, and increases of cases among foreign-born persons. The recently reported case series from New York illustrates key points (3). Of the 16 patients in the study, only 5 had prenatal care before the diagnosis was made. The remaining 11 patients had their diagnoses confirmed when they sought care in the emergency department because of symptoms related to the infection. Ten of the 16 cases were proven active pulmonary tuberculosis, 2 were tuberculous meningitis, and 1 each were mediastinal, renal, gastrointestinal, and pleural tuberculosis. There was a high rate of anergy; only 6 of 15 patients tested had positive skin test results for tuberculosis. Of 11 tested for HIV, 7 were positive. There was a high rate of complications: preterm labor in five, fetal growth restriction in five, and oligohydramnios in one. One infant with growth restriction also had fetal ascites and was delivered at 33 weeks of gestation. One HIV-infected mother with severe pulmonary tuberculosis died of respiratory failure after a cesarean delivery for fetal distress. It is unclear whether the high rate of complications was due to tuberculosis alone, to the combination of tuberculosis and HIV, or to other factors. All women were treated with multiple-drug regimens (14 with isoniazid, ethambutol, and rifampin; 2 also with pyrazinamide, for CNS infection and for persistent positive sputum cultures on the three drugs). Because all the women had negative cultures by delivery, infants did not receive prophylaxis except for one whose mother had tuberculous meningitis. There were no cases of true
congenital tuberculosis reported. Diagnosis As noted in the series discussed already, diagnosis of tuberculosis in pregnancy may be difficult, because a high percentage of patients are anergic and seek care late (3). Demonstration of acid-fast bacilli in sputum smears should raise suspicion, but culture confirmation is required for diagnosis, because sputum may contain nontuberculous mycobacteria. Diagnosis of extrapulmonary tuberculosis may be more difficult because specimens of urine and cerebrospinal fluid (CSF) are less frequently positive for acid-fast bacilli on smears (3). Demonstration of Mycobacterium tuberculosis in culture is thus mandatory for the definitive diagnosis, but treatment should be initiated with first-line drugs when there is an abnormal chest x-ray and suggestive clinical findings, particularly in high-risk groups. Pregnancy does not appear to increase acquisition or activation of tuberculosis in HIV-infected pregnant women (3). No special recommendations have been made for tuberculosis screening in pregnancy, but recent CDC recommendations for screening high-risk populations include some pregnant women (6). These recommendations are summarized in Table 16.19.
TABLE 16.19. CDC RECOMMENDATIONS FOR TUBERCULOSIS SCREENING IN HIGH-RISK POPULATIONS
The CDC has provided revised guidelines for interpretation of tuberculin-skin tests (11) (Table 16.20).
TABLE 16.20. CRITERIA FOR TUBERCULIN POSITIVITY, BY RISK GROUP
Congenital Tuberculosis Congenital tuberculosis is rare, with less than 300 cases reported. Before two cases reported in 1994, the last reported case in the United States was in 1982. The frequency has ranged from 3 cases among 100 infants of mothers with active tuberculosis to no cases in two series of 260 and 1,369 infants (2). In 29 cases of congenital tuberculosis reported since 1980, the median age of presentation was 24 days (range, 1 to 84), and common presenting findings were hepatosplenomegaly in 76%, respiratory distress in 72%, fever in 48%, lymphadenopathy in 38%, abdominal distention in 24%, lethargy or irritability in 21%, ear discharge in 17%, and papular skin lesions in 14%. Chest x-rays were abnormal in 23 infants, and skin test results at the time of diagnosis were negative in all 9 tested. Only 12 mothers had active tuberculosis; 15 were asymptomatic and were diagnosed only after their infants were diagnosed. Mortality from congenital tuberculosis was striking: 38% overall and 22% (5 of 23) of infants treated with chemotherapy. The observation that congenital tuberculosis commonly occurs in infants of asymptomatically infected mothers reinforces the need for careful evaluation of mothers of infants with suspected congenital infection. On the basis of this review, Cantwell et al. (2) proposed revised diagnostic criteria for the diagnosis of congenital tuberculosis. The infant must have proven tuberculous lesions and at least one of the following: (a) lesions in the first week of life; (b) a primary hepatic complex or caseating hepatic granulomas; (c) tuberculous infection of the placenta or the maternal genital tract; or (d) exclusion of the possibility of postnatal transmission by a thorough investigation of contacts. They proposed that different restriction-fragment-length polymorphisms in the isolates from mother and infant would exclude congenital infection, but that identical patterns could also come from either congenital or postnatal infection. Treatment Of Latent Tuberculosis Infection The mainstay of treatment of latent tuberculosis infection in the United States over the
past three decades has been isoniazid for 6 to 12 months (1). However, because of poor adherence and concerns about toxicity, there has been interest in the development of shorter rifampin-based regimens as alternatives to isoniazid for treatment of latent tuberculosis. In June 2000, the CDC provided revised recommendations for the treatment of latent tuberculosis. Four regimens are recommended for the treatment of nonpregnant adults with latent tuberculosis. These four regimens, together with their rating and evidence, are summarized in Table 16.21. Details of the medications used to treat latent tuberculosis, including doses, toxicities, and monitoring requirements, are shown in Table 16.22 and Table 16.23. Highlights of the recommended drug regimens for treatment of latent tuberculosis in pregnancy are summarized in Box 16.3 (1).
TABLE 16.21. REGIMENS FOR TREATMENT OF ADULTS WITH LATENT TUBERCULOSIS INFECTION
TABLE 16.22. MEDICATIONS TO TREAT LATENT TUBERCULOSIS INFECTION: DOSES, TOXICITIES, AND MONITORING REQUIREMENTS
TABLE 16.23. RECOMMENDED DRUG REGIMENS FOR TREATMENT OF LATENT TUBERCULOSIS (TB) INFECTION IN ADULTS
Thus, for pregnant, HIV-negative women, isoniazid given daily or twice weekly for 9 or 6 months is recommended. The CDC recommends that for women at risk for progression of latent tuberculosis to active disease, particularly those who are infected with HIV or who have likely been infected recently, initiation of therapy should not be delayed on the basis of pregnancy alone, even during the first trimester. For women whose risk of active tuberculosis is lower, some experts recommend waiting until after delivery to start treatment. Pregnancy has minimal influence on the pathophysiology of tuberculosis or the likelihood of its progression to active disease (1). The CDC notes that the current classification scheme for determining the tuberculin skin test is likely valid in pregnancy. The CDC further notes that there is no evidence that the tuberculin skin test has adverse effects on the pregnant mother or the fetus. Pregnant women should be targeted for tuberculin skin testing only if there has been a specific risk factor for latent tuberculosis or for progression of latent tuberculosis to disease. Whereas the need for treatment of active tuberculosis during pregnancy is unquestioned (see later discussion), the treatment of latent tuberculosis in pregnancy has remained somewhat controversial. Thus, some experts recommend delaying treatment until after delivery because pregnancy itself does not increase the risk of progression and because women in pregnancy and the early puerperium may be more susceptible to isoniazid-induced hepatotoxicity. On the other hand, in view of the concern regarding progression of latent tuberculosis to disease and its effect on the mother and fetus, many experts agree that pregnant women with latent tuberculosis should be treated during pregnancy and have careful clinical and laboratory monitoring. It is felt that the risk of isoniazid-induced hepatotoxicity must be weighed against the risk of developing active tuberculosis and the consequences for the mother and fetus. As noted already, the preferred treatment regimen for latent tuberculosis in pregnancy is isoniazid administered either daily or twice weekly. Whereas rifampin is probably safe,
there are no efficacy data to support its use in pregnancy. The CDC notes clearly that for women at high risk for progression of latent tuberculosis to active disease, particularly those who are infected with HIV or who have recently been infected, initiation of therapy should not be delayed on the basis of pregnancy alone, even during the first trimester. For these women, however, careful clinical and laboratory monitoring for hepatitis should be undertaken (Table 16.22). Pregnant women taking isoniazid should receive pyridoxine supplementation at 50 mg per day. There have been no reported adverse effects of antituberculosis medications to infants who have been breast-feeding while mothers have taken them. Thus, breast-feeding is not contraindicated when the mother is being treated for latent tuberculosis. Yet infants whose breast-feeding mothers are taking isoniazid should receive supplemental pyridoxine. The amount of isoniazid provided by breast milk is inadequate for the treatment of the infant. The June 2000 guidelines recommended for pregnant women, as for other adults, that there should be targeted tuberculin skin testing (1). The target in tuberculin testing for latent tuberculosis infection is a strategic component of tuberculosis control that identifies persons at high risk for developing tuberculosis who would benefit by treatment of latent tuberculosis if detected. New recommendations indicate that persons with increased risk for developing tuberculosis include those who have had recent infection with M. tuberculosis and those who have clinical conditions that are associated with an increased risk of progression from latent to active tuberculosis. These conditions are as follows: recent tuberculosis, particularly when it occurs within the past year, HIV infection, injection drug abuse, silicosis, radiographic findings consistent with prior tuberculosis, being underweight, diabetes mellitus, chronic renal failure or hemodialysis, gastrectomy, jejunoileal bypass, solid organ transplant (particularly renal or cardiac), and carcinoma of the head or neck. Combined with this recommendation, targeted tuberculin testing should be conducted only among groups at high risk and discouraged in low risk. Infected persons who are considered to be at high risk for developing active tuberculosis should be offered treatment of latent tuberculosis irrespective of age. Cutpoints for defining a positive tuberculin skin reaction are shown in Table 16.20. In summary, for persons who are at the highest risk of developing active tuberculosis if they are infected with M. tuberculosis, more than or equal to 5 mm of induration is considered positive. For other persons with an increased probability of recent infection or with clinical conditions that increase the risk of progression to tuberculosis, more than or equal to 10 mm of induration is considered positive. For persons at low risk for tuberculosis for whom tuberculin testing is no longer generally indicated, more than 15 mm of induration is considered positive. Because of the complexity of this decision making, the reader is referred to the full CDC recommendations (1). The changes from prior recommendations on tuberculin testing and treatment of latent tuberculosis are summarized in Table 16.24.
Box 16.3 Recommended drug regimens for latent tuberculosis infection in pregnancy are as follows: In pregnancy, the preferred regimens include Isoniazid daily for 9 months or Isoniazid twice weekly for 9 months or Isoniazid daily for 6 months or Isoniazid twice weekly for 6 months as shown in Table 16.23. Note that some experts would use rifampin and pyrazinamide for 2 months as an alternative regimen for HIV-infected pregnant women, although pyrazinamide should be avoided during the first trimester.
TABLE 16.24. CHANGES FROM PRIOR RECOMMENDATIONS ON TUBERCULIN TESTING AND TREATMENT OF LATENT TUBERCULOSIS INFECTION (LTBI)
Treatment Of Active Tuberculosis Emergence of tuberculous isolates with multidrug resistance has prompted revised treatment recommendations from the CDC (4). In New York City, 33% of cases showed resistance to at least one drug, and 19% were resistant to both isoniazid and rifampin. Treatment regimens for children and nonpregnant adults are shown in Table 16.25 and Table 16.26. The initial regimens include four drugs for the first 2 months, followed by an altered regimen based on drug susceptibility testing.
TABLE 16.25. REGIMEN OPTIONS FOR INITIAL TREATMENT OF TB IN CHILDREN AND NONPREGNANT ADULTS
TABLE 16.26. DOSAGE RECOMMENDATION FOR THE INITIAL TREATMENT OF TB AMONG CHILDRENa AND ADULTS
Special consideration is given to the treatment of tuberculosis in pregnancy. Effective therapy is essential, but streptomycin may cause congenital deafness. Routine use of pyrazinamide is not recommended during pregnancy because the risk of teratogenicity has not been determined. A minimum of 9 months of therapy is recommended because the 6-month regimens cannot be used. Thus, in pregnancy, the initial regimen is isoniazid plus rifampin, and ethambutol, if potential isoniazid resistance is suspected. If sensitivities prove all drugs to be effective, treatment may be given twice weekly after the first 2 months. Breast-feeding should not be discouraged for women taking these drugs (Box 16.4) (5).
Box 16.4 Recommended initial treatment regimen for active tuberculosis in pregnancy is as follows: Isoniazid, 300 mg daily, plus rifampin, 600 mg daily. Add ethambutol, 2.5 g daily, if there is potential isoniazid resistance All drugs for a minimum of 9 months. Give pyridoxine, 50 mg daily with isoniazid Note that pyrazinamide is not recommended during pregnancy because teratogenicity has not been assessed. Streptomycin may cause congenital deafness.
In an extensive review, Snider et al. (8) provided data on ethambutol, isoniazid, streptomycin, and rifampin. Ethionamide is the only antituberculous agent considered clearly teratogenic, but it is rarely indicated. Ethambutol has replaced p-aminosalicylic acid as a first-line drug. It is easier to tolerate, and there is no evidence of adverse fetal effects (8). On the basis of more than 600 cases, Snider et al. (8) considered ethambutol safe for use in pregnancy. Isoniazid also has no special adverse effects in the pregnant woman or the fetus. Of 1,480 pregnancies in which isoniazid was used, there were only 16 abnormal fetuses—a rate lower than that in the general population (8). Streptomycin, when used for periods of several weeks in the treatment of maternal tuberculosis, commonly causes eighth nerve damage in exposed fetuses (9). Snider et al. (8) reported eighth nerve damage in 15% of infants exposed in utero to streptomycin in the treatment of maternal tuberculosis. The overall risk of congenital anomalies does not seem to be increased in fetuses with in utero exposure to rifampin. Fetal malformations have been seen in 14 of 442 women (8). The anomalies covered a wide spectrum, suggesting no cause-effect relationship.
PROTOZOA Toxoplasmosis Toxoplasmosis is a widely distributed illness, caused by Toxoplasma gondii, an intracellular parasite. In the last few years, there has been renewed interest in this infection because of four developments: growing numbers of patients with acquired immunodeficiency syndrome (AIDS) who are particularly susceptible to toxoplasmosis, particularly severe manifestations; an FDA initiative in food safety; concerns about
current diagnostic techniques; and availability of prenatal diagnosis. Epidemiology Although T. gondii is found in many mammalian species, the cat is the only definitive host. This parasite may exist in three forms (trophozoite, cyst, or oocyst). Trophozoites are the invasive forms, and the cysts are the latent forms. The oocysts are found only in cats. Human infection may be acquired by consuming raw or undercooked meat of infected animals (particularly mutton and lamb) or by contact with oocysts from the feces of an infected cat. The oocysts may be spread to humans or to food by hand or by insects. Cats acquire toxoplasmosis by eating infected mice or other animals. Oocysts in cat feces do not become infective for 4 to 5 days. Once infected by oocysts from cat feces or by cysts from infected meat, persons may experience a parasitemia, during which fetal involvement may occur. Later, T. gondii cysts appear in tissues, particularly striated muscle and brain, where they persist indefinitely. In the general population, the seroprevalence of Toxoplasma varies widely, from approximately 55% in France to 26% in Denmark to 10% in the United States among military recruits (1). Among pregnant women, recent data from the United States (1988 to 1994) show that the current seroprevalence is only 14% (Table 16.27). Older data in the United States showed a seroprevalence of 20% to 40% (2), whereas in Paris, the prevalence was up to 84% (3). Thus, compared with western Europeans, Americans have lower seroprevalence. In addition, seroprevalence in the United States appears to have decreased in the general population in the last 30 years, as judged by seroprevalence in military recruits.
TABLE 16.27. EPIDEMIOLOGY OF TOXOPLASMOSIS IN THE UNITED STATES
Domestic cats acquire the infection as early as 6 to 10 weeks of age. Once infected, they excrete the oocysts for a short period of time (approximately 1 to 2 weeks) and usually are seronegative while shedding occurs. The duration of immunity in cats is
uncertain but is estimated to be from approximately 1/2 a year to up to 6 years. Consumption of infected food may be another source of toxoplasmosis. In the United States, the greatest source of food-related transmission is meat from infected pigs. However, the seroprevalence in pigs is decreasing and is now only approximately 2.5% in pigs sold for market, in comparison to seroprevalence rates of 24% approximately two decades ago (1). Overall risk factors for seroprevalence for toxoplasmosis in the United States include increasing age (older than 30 to 40 years), less than a high school education, rural or small town residence, and foreign birth. Rates of seroconversion in pregnancy also vary widely on a geographic basis. In France, the rate of infection in pregnancy is 1.5%, whereas in the United States, it is estimated that the rate is 0.2% to 1%. In 1983, the seroconversion rate in urban Alabama was only 0.06% (2 of 5,142) (1,4,5). Rates of congenital infection with toxoplasmosis have been reported for western Europe and the United States. In France, the rate is approximately 0.2%. In New England, a recent study showed the rate was 0.01% (52 cases of 600,000 births), whereas estimates for the overall United States placed the rate at 0.01% to 0.1%. This estimate would project to approximately 400 to 4,000 cases per year in the United States (1,5,6). Recent data have clarified the risk of fetal infection when there is primary toxoplasmosis in pregnancy. Clearly, the risk of fetal infection increases with gestational age; that is, the later in pregnancy toxoplasmosis infection occurs, the more frequently the fetus is infected. As shown in Fig. 16.7, when there is no maternal treatment, the rate of fetal infection rises from approximately 10% in the first trimester to 30% in the second trimester to 60% in the third trimester. When there are seroscreening and treatment programs, the rate of fetal infection is reduced by more than 50%. Figure 16.7 displays the rate of congenital infection by gestational age of maternal infection in more than 2,000 pregnancies. When maternal infection occurred in the first 2 weeks of pregnancy, the fetal infection rate was 0. Thereafter, the rate of fetal infection continued to climb as gestation advanced (Fig. 16.8). Further, these are rates that occur when there are seroscreening and maternal treatment programs. However, when fetal infection occurs in the first trimester, it is likely to be severe or to lead to stillbirth, whereas with fetal infection later in gestation, the consequences are less severe. Classic data from France are shown in Table 16.28 (7). Figure 16.9 shows severe hydrocephalus in a stillborn infant with congenital toxoplasmosis.
FIGURE 16.7. Incidence of fetal infection by trimester of maternal infection. Note that mothers were treated with spiramycin (3 g per day).
FIGURE 16.8. Congenital toxoplasmosis by time of maternal infection in 2,281 pregnancies.
TABLE 16.28. FETAL INFECTION AFTER PRIMARY MATERNAL TOXOPLASMOSIS
FIGURE 16.9. Severe hydrocephalus in a stillborn infant with congenital toxoplasmosis.
Congenital infection does not affect more than one infant in a particular mother (8). The role of toxoplasmosis in chronic abortion remains unresolved after 20 years of study. Diagnosis Subclinical disease is the rule in both adults and newborns with toxoplasmosis. When it is apparent clinically in a normal host, the most common manifestation is lymphadenopathy (most commonly cervical). Fever, fatigue, sore throat, maculopapular rash, and occasionally hepatosplenomegaly may also be noted. Examination of the peripheral blood shows lymphocytosis or an occasional atypical lymphocyte. Accordingly, this disease is often thought to be the “flu” or infectious mononucleosis. An occasional adult may have mainly ocular symptoms including haziness of vision, pain, and photophobia. In these cases, ophthalmologic examination shows clusters of yellow-white patches in the optic fundus, representative of a focal necrotizing retinochoroiditis. In healthy adults, clinical toxoplasmosis is mild and self-limited; in immunosuppressed individuals such as those with AIDS, it leads to serious pulmonary or CNS involvement. As noted, most infants with congenital toxoplasmosis have only serologic abnormalities. Of those with clinical disease, few have the commonly suggested triad of intracerebral calcifications, chorioretinitis, and hydrocephaly in the past. Findings from a recent French series are shown in Table 16.29, where we see that of 210 infants with congenital infection diagnosed during a study of infection in pregnancy, most (55%) had subclinical infection. Thus, there is a wide spectrum of disease. Acute primary
toxoplasmosis in pregnancy has also been associated with abortion, prematurity, and growth retardation.
TABLE 16.29. SELECTED CLINICAL FINDINGS IN CONGENITAL TOXOPLASMOSIS (N = 210)a
Numerous diagnostic approaches are used for toxoplasmosis, as shown in Table 16.30. Toxoplasmic cysts may be identified histologically in infected tissue such as lymph nodes, but this technique is cumbersome and probably of limited sensitivity. In selected laboratories, T. gondii may be cultured either using tissue culture (such as in mouse fibroblasts) or by peritoneal inoculation of mice. Recent modifications of the mouse inoculation method include assay of mouse serum 6 weeks after inoculation for Toxoplasma-specific antibody by EIA. Tissue culture is faster, but mouse inoculation is more sensitive. Toxoplasma antigens may also be demonstrated by immunofluorescent techniques, but these also probably have limited sensitivity.
TABLE 16.30. DIAGNOSIS OF TOXOPLASMOSIS
Serologic techniques are generally used to detect toxoplasmosis: at least eight different techniques including EIA, indirect fluorescent antibody, complement fixation, and the “classic” Sabin-Feldman dye test (8). IgG antibodies peak within 1 to 2 months of the onset of infection, and low titers persist for years. IgM antibodies appear within a week and usually last for a few months. Commercially available kits for diagnosis of Toxoplasma-specific IgG are generally reliable, but the FDA has issued a Public Health Advisory regarding the limitations of Toxoplasma-specific IgM in commercial kits (10). Specifically, physicians were advised to interpret results “with caution.” Further, physicians were advised that they should not rely on any single test in diagnosing recent infection and in deciding medical action. The concern was with false-positive Toxoplasma IgM leading to false-positive diagnosis of recent toxoplasmosis and hence the likelihood of fetal infection. Thus, it is complicated to make the diagnosis of recent toxoplasmic infection. Often, multiple samples are needed. It is possible to make the diagnosis, but there are still some false-positives even in expert laboratories. Physicians are advised to consult an expert in making the diagnosis (1). Advice on the diagnosis may be obtained from the FDA or from the laboratory at Stanford University (laboratory of Dr. Jack Remington). Thus, serologic results for Toxoplasma antibodies should not be accepted at face value from all laboratories because accuracy varies among laboratories. Some commercial facilities do not store serum for more than a few days. This practice is most unfortunate, because repeated testing and running the acute and convalescent serum in parallel are often needed to make the diagnosis. The clinician is advised to use a reliable laboratory and to know how to interpret its results. Otherwise, incorrect information and incorrect counseling may result (11). New diagnostic techniques employing DNA testing have been developed (such as PCR detection in amniotic fluid). These diagnostic approaches have shown rapid and sensitive identification of the infected fetus. In 1988, Daffos et al. (12) first reported prenatal diagnosis of 746 pregnancies at risk for congenital toxoplasmosis based on culture of fetal blood and amniotic fluid for T. gondii, testing of fetal blood for specific IgM and for nonspecific measures for infection (including platelet count, serum glutamic-oxaloacetic transaminase, serum glutamic-pyruvic transaminase, lactate dehydrogenase, white blood cell count, eosinophilia, and total IgM), and fetal ultrasound examination. Overall, infection was diagnosed in 39 (93%) of 42 fetuses. Subsequent series have confirmed the validity of diagnosis in utero (13). In 89 cases of proven fetal toxoplasmosis, ultrasound abnormalities have been cataloged. Of these cases, 32 (36%) showed abnormal findings: ventricular dilation, 25; intracranial densities, 6; placental thickness (higher than the 95th percentile), 11; placental hyperdensity, 2; intrahepatic densities, 4; hepatomegaly, 2; ascites, 5; pericardial effusion, 2; and pleural effusion, 1 (14).
With the advent of PCR, rapid diagnosis of congenital toxoplasmosis has been reported. In 1990, Grover et al. (15) described a study of 43 documented cases of acute maternal infection in which PCR was compared with standard methods. PCR correctly identified five of five samples of amniotic fluid from four proven cases of congenital infection. There were no false-positive diagnoses reported by any of the methods. Thus, detection of T. gondii in amniotic fluid by PCR appeared to be very promising, particularly because it is rapid and highly sensitive. In a larger series from Paris, Hohlfeld et al. (16) compared PCR on amniotic fluid with their “classic” methods of ultrasonography, amniocentesis, and fetal blood sampling. In 2,632 cases of maternal infection, the overall risk of congenital infection was 7.3%, varying from 2% in weeks 3 to 10, to 22% in weeks 27 to 30, to 67% in weeks 31 to 34. There was no observed risk of congenital infection among 100 women with infection during the 2 weeks after the last menstrual period. In 339 cases, PCR on amniotic fluid was used. Overall, the classic diagnostic methods had a sensitivity of 92%, specificity and positive predictive value of 100%, and a negative predictive value of 99.4%. With PCR, the sensitivity, specificity, and predictive values were all 100%. The authors concluded that in utero diagnosis of congenital toxoplasmosis may be simplified because fetal blood sampling has been replaced by PCR of amniotic fluid and inoculation of amniotic fluid into mice. This new approach also offers safer and faster results. Further data on the reliability of diagnosis of congenital toxoplasmosis have recently been reported from a retrospective study from six reference centers in Europe, where amniocentesis and cordocentesis were used in 122 patients between 1986 and 1996. The diagnosis of congenital toxoplasmosis was based on the persistence of specific IgG at age 1 year or reappearance of specific IgG in the child after cessation of therapy. Overall, the most sensitive diagnostic combination was amniotic fluid PCR plus amniotic fluid mouse inoculation. Fetal blood tests were much less sensitive (Table 16.31). Thus, amniotic fluid PCR has been confirmed as the single most reliable test, but the combination of amniotic fluid PCR and amniotic fluid mouse inoculation achieves a sensitivity of more than 90%.
TABLE 16.31. PRENATAL DIAGNOSIS OF CONGENITAL TOXOPLASMOSIS
An alternative to in utero diagnosis is early neonatal diagnosis and prompt treatment (6). In Massachusetts and New Hampshire, newborns were screened for intrauterine infection with T. gondii by an immunoassay performed on routinely collected blood. If the screening test result was positive, confirmatory testing was performed for specific IgG and IgM antibodies in maternal and neonatal serum. Infants with serologic evidence of infection had 1 year of treatment and detailed clinical follow-up. Treatment consisted of combinations of pyrimethamine, sulfadiazine, and folinic acid (leucovorin), but five different regimens were used at the discretion of attending physicians. Spiramycin was also given to two infants. Among 635,000 screened infants, 100 had positive screening test results, and 52 (1 of 12,211) had confirmed infection. Fifty infants were identified only through neonatal screening (not by initial clinical examination). More detailed clinical evaluation, performed after serologic results were available, showed abnormalities of the eyes or CNS in 19 (40%) of 48 infants. Yet, after treatment, only 1 of 46 children had any neurologic deficit. Of 39 treated children with follow-up eye examinations, 4 (10%) had lesions that may have developed postnatally. Treatment in the children was well tolerated in 32 of 47 for whom toxicity information was available. The remaining infants had mild anemia, leukopenia, or rash. All infants had normal growth. The estimated cost per case identified was less than $30,000, an amount considered reasonable in view of the costs of rearing an afflicted child. Thus, routine neonatal screening for toxoplasmosis, coupled with early treatment, may reduce long-term sequelae, and this approach may be more practical than in utero diagnosis and treatment in the United States, where the incidence is low. However, a limitation of the approach of screening newborns is that this misses the diagnosis in severely affected fetuses who wind up with either abortion or stillbirth. Prevention To prevent toxoplasmosis in pregnancy, most authorities advise (a) avoiding undercooked meat; (b) hand washing after handling a cat, particularly before eating; (c) having someone else change the litter box daily; (d) not permitting indoor cats to go outside, where they may attack an infected mouse; (e) not allowing stray cats in the house; and (f) not feeding raw meat to cats. Based on their data in Alabama, Hunter et al. (4) did not favor routine serologic screening programs in view of the expense of screening many women who are susceptible and the low attack rate. Alternatively, patient education is recommended by Frenkel (18). Wilson and Remington (19), however, note that regional programs are needed in the United States to provide data pertinent in this country. Recently, Foulon et al. (20) reported that the impact of primary prevention of congenital toxoplasmosis by hygienic measures was limited. They noted that such a program in Brussels, Belgium, only reduced the seroconversion rate for acquisition of toxoplasmosis in pregnancy from 1.43% to 0.95% (not a statistically significant
decrease). Whether such an effort would significantly reduce the seroconversion rate in the United States is not clear. Treatment For women with toxoplasmosis confirmed in the first trimester, the physician should offer counseling regarding the risk of serious congenital infection, the availability of in utero diagnosis, and pregnancy termination. As noted, widespread screening and treatment programs are practiced in many western European countries. One treatment protocol is shown in Table 16.32. In brief, spiramycin treatment is begun when maternal diagnosis is made, and diagnostic evaluation of the fetus is begun after 18 weeks. If fetal diagnosis is confirmed, then treatment is begun with pyrimethamine plus sulfadiazine or termination is offered. When there is late infection (after 26 to 34 weeks), there is presumptive treatment with pyrimethamine and sulfadiazine without diagnostic workup. In late maternal infection, although there is a high risk of fetal infection, the likelihood of severe infection is small (1). Specific regimens available in the United States are shown in Box 16.5 (21). Note that treatment is available with pyrimethamine and sulfadiazine, but these should be accompanied by administration of folinic acid to pregnant women. In addition, spiramycin is available but only through the FDA at the telephone number shown.
Box 16.5 Maternal treatment regimens for toxoplasmosis include the following (from Sever JL, 1998): 1. Pyrimethamine 25 mg orally daily (after first trimester) plus sulfadiazine 1 g orally four times a day for 28 days, followed by half dose for 28 days more 2. Also give 6 mg folinic acid intramuscularly or orally three times a week 3. Spiramycin may be used initially (available in the United States through the FDA, at 301-827-2335)
TABLE 16.32. TREATMENT OF TOXOPLASMOSIS: FRENCH PROTOCOL
All authorities agree that symptomatic infants with congenital toxoplasmosis should be treated with sulfadiazine, pyrimethamine, and folinic acid supplementation. Details of the treatment regimens are available (8). In the infant with asymptomatic toxoplasmosis at birth, late CNS sequelae are possible. Based on the recent data from New England, early treatment of asymptomatically infected infants appears to reduce long-term sequelae (6).
LYME DISEASE Since Lyme disease was first described in 1977, it has rapidly emerged as an important infection. In 1999 there were 16,273 Lyme disease cases reported to the CDC for an overall incidence of 6.0 per 100,000 population (1). This represents over a 30-fold increase in the less than 500 cases reported in 1982. Most cases were reported from the Northeast, mid-Atlantic, and north central regions, as well as California (Fig. 16.10). During 1999, the highest reported rates in the Northeast and north central areas were seen in Connecticut (98 per 100,000), (Rhode Island 55.1; New York 24.2; Pennsylvania 23.2; Delaware 22.2; New Jersey 21.2; Maryland 17.4; Massachusetts 12.7; Wisconsin 9.3 [1]). The CDC suggested that the tremendous increase in reported Lyme disease cases since 1982 was due to four factors: (a) heightened awareness of Lyme disease by patients and physicians; (b) increased use of laboratory testing for the diagnosis of Lyme disease; (c) increased surveillance and health department requirements for reporting; and (d) a true increase in the number of cases (1). The latter reflects increasing penetration of suburban development into rural forest areas where deer are present in large numbers. Lyme disease is currently the most commonly reported vector-borne disease in the United States and now accounts for 90% of vector-borne infections reported to the CDC (1).
FIGURE 16.10. Reported cases of Lyme disease—United States, 1992. (From Centers for Disease Control. Lyme disease—United States, 1991–1992. MMWR Morb Mortal
Wkly Rep 1993;42:345–348, with permission.)
Lyme disease is a tick-borne infection that is caused by the spirochete Borrelia burgdorferi. Lyme disease is a multisystem illness that is characterized by a distinct lesion, erythema chronicum migrans, which often is followed by neurologic, cardiac, and arthritic manifestations (2,3 and 4). The Lyme disease spirochete is transmitted by Ixodes ticks. The most common vector is the deer tick, Ixodes dammini, whose distribution coincides with endemic areas of the disease in the northeast United States. In the western United States, Ixodes pacificus, the black-legged deer tick, is responsible for transmission of disease. The white-footed mouse is host for the larval and nymph stages of the disease. White-tailed deer are the preferred hosts for the adult-stage I. dammini. Most human infections with B. burgdorferi occur from May through August, a time when both nymphal-stage activity and human outdoor activity are at their highest (4). As noted by Steere (4), Lyme disease generally occurs in stages characterized by differing clinical manifestations. The initial manifestation is erythema migrans, which is followed several weeks or months later by meningitis or Bell palsy and subsequently followed months or years later by arthritis. Asbrink and Hovmark (5) suggested a classification based on the system used in syphilis, in which Lyme disease is divided into early and late infection. Early infection consists of stage I (localized erythema migrans), followed within days or weeks by stage II (disseminated infection), and within weeks or months by intermittent symptoms. Late infection, or stage III (persistent infection), begins generally a year or more after the onset of disease. Shapiro et al. (6) recently assessed the risk of infection with B. burgdorferi and the efficacy of prophylactic antimicrobial treatment after a deer tick bite. Of the 344 deer ticks studied, only 15% were infected with B. burgdorferi, as analyzed by PCR technology (6). Erythema migrans was noted in only two patients, both of whom were in the placebo group (6). The risk of infection in the placebo-treated patients was 1.2% (95% CI, 0.1–4.1%). After transmission of B. burgdorferi, approximately 60% to 80% of individuals develop erythema migrans. This lesion begins as a small erythematous papular macule, which is then followed by a gradual centrifugal expansion over 3 to 4 weeks. In general, the lesions clear centrally and take on an annular configuration that averages 15 cm in diameter. The initial skin lesion is often accompanied by systemic symptoms including fever, flulike symptoms with migratory arthralgias, myalgias, headaches, and regional lymphadenopathy (4). Even without treatment, erythema migrans usually fades within 3 to 4 weeks. Stage II early infection, or disseminated infection, occurs within days or weeks after transmission. Disseminated infection is often associated with characteristic symptoms involving skin, nervous symptoms, or the musculoskeletal system. Nearly half the patients will develop secondary annular skin lesions that resemble the primary erythema
migrans lesions. These are usually smaller and migrate less. Patients commonly develop severe headaches and mild stiffness of the neck, which occur in short attacks. Interestingly, a recent report using PCR assay for Borrelia-specific DNA has demonstrated the presence of B. burgdorferi in the CNS during acute disseminated infection (7). The musculoskeletal pain associated with Lyme disease is migratory, lasting only hours or days in any given location; it may involve the joints, bursa, tendons, muscle, and bone. During the disseminated stage, patients frequently have severe malaise and fatigue. As the infection begins to localize, approximately 15% to 20% of patients develop frank neurologic involvement. The classic triad of neurologic Lyme disease includes meningitis, cranial nerve palsies, and peripheral radiculopathies. The predominant symptoms of Lyme meningitis are severe headaches and mild neck stiffness, which fluctuate for several weeks. Cerebral spinal fluid demonstrates a leukopoietic pleocytosis, slightly elevated protein levels, and normal glucose levels. Nearly one half of the cases of meningitis have an associated mild encephalitis that leads to loss of concentration, emotional lability, lethargy, sleep disturbances, or focal cerebral dysfunction. The most commonly affected cranial nerve is the seventh, leading to Bell palsy. One third of the patients with Bell palsy have bilateral involvement. Peripheral radiculopathies are characterized by severe neuritic pain, dysesthesia, focal weakness, and areflexia. Within several weeks of the onset of disease, 5% to 10% of patients have cardiac involvement. Fluctuating degrees of atrioventricular block are the most common abnormalities; this ranges from first-degree to complete heart block. Occasionally patients have acute myopericarditis, mild left ventricular dysfunction, or (rarely) cardiomegaly or fatal pancarditis. Cardiac abnormalities usually last from 3 days to 6 weeks. Approximately 6 months after the onset of disease, about 60% of the patients in the United States begin to have brief attacks of asymmetric oligoarticular arthritis, primarily in the large joints, particularly the knee. Stage III, or late infection, is characterized by episodes of arthritis during the second and third year of the illness, which often become chronic arthritis. Nocton et al. (8) recently used PCR to detect B. burgdorferi DNA in the synovial fluid of patients with Lyme arthritis. Nearly all joint fluid samples from untreated patients with Lyme arthritis contained detectable B. burgdorferi DNA, whereas most posttreatment samples and all control samples were negative (8). In addition, several late syndromes of the CNS have now been described. These include progressive encephalomyelitis (9), subacute encephalitis, dementia, and a syndrome suggesting demyelination. Experience with Lyme disease during pregnancy is rather limited (10). Concern has arisen because other spirochetes such as T. pallidum cross the placenta and produce adverse effects on the fetus or neonate. In 1985, the first case of transplacental transmission of B. burgdorferi was documented by identification of the spirochete in multiple organs of an infant who died of congenital heart disease shortly after birth (11). Subsequently, Weber et al. (12) isolated the organism from the brain and liver of an infant who died within the first 24 hours of life. Both of these infants were born to mothers who had erythema migrans in the first trimester. Three stillbirths with recovery of B. burgdorferi from multiple organs subsequently were reported (13,14 and 15). The CDC evaluated the effect of Lyme disease during pregnancy on two occasions. In a retrospective study, the CDC identified five adverse outcomes in fetuses born to 19 women with Lyme disease (15). Among the adverse outcomes noted were prematurity, cortical blindness, fetal demise, syndactyle, and a rash in the neonate. However, they
could not prove a teratogenetic pattern or show a reduction in fetal morbidity if the mothers had been appropriately treated for Lyme disease. In the second prospective study, the CDC evaluated 17 women with documented Lyme infection in the first trimester (16). Only two of these pregnancies were abnormal, with one resulting in a spontaneous abortion at 13 weeks. In the second, the infant had syndactyle. In a survey of core blood serum from 421 infants born in an endemic area for B. burgdorferi, Williams et al. (17) found no relationship between congenital malformation and the presence of antibody to Lyme disease. In a recent European study, 0.85% of core blood obtained from more than 1,400 pregnancies demonstrated elevated titers to B. burgdorferi (18). Of these, only one patient had clinical disease during her pregnancy, and her infant had a ventricular septal defect without other anomalies. In the remaining 11 infants with elevated IgG cord titers, 6 had abnormal neonatal courses; 2 with hyperbilirubinemia, 1 macrocephaly, 1 intrauterine growth retardation, and 1 supraventricular extrasystoles. However, at an average follow-up of 9 months, all of these latter six infants were normal. Thus, any relationship between positive IgG titers and abnormalities in the fetus or neonate is inconclusive. Recently, Strobino et al. (19) conducted a prospective study to determine whether prenatal exposure to Lyme disease was associated with an increased risk of adverse pregnancy outcome. Nearly 2,000 women were enrolled after completing a questionnaire and having serum tested for antibody to B. burgdorferi at their first prenatal visit and at delivery (19). These authors reported that neither maternal Lyme disease nor an increased risk of exposure to Lyme disease was associated with fetal death, decreased birthweight, or length of gestation at delivery (19). Tick bites or Lyme disease around the time of conception was not associated with congenital malformations (19). The diagnosis of Lyme disease is hindered by the various manifestations of the disease. Culture or direct visualization of B. burgdorferi from patient specimens is difficult, so serology is currently the only practical laboratory diagnosis available. Indirect fluorescent antibody and ELISA are the most commonly used tests for the diagnosis of Lyme disease. In general, ELISA is preferred because of its increased sensitivity and specificity. It is important to recognize that false-positive results can be seen with other spirochetal diseases, infectious mononucleosis, autoimmune disorders, and Rocky Mountain spotted fever. Thus, an indirect screening test for syphilis should be done on all patients with positive results to rule out syphilis. False-negative results may be seen early in the disease, that is, during the first 2 weeks when infection is localized to the skin, or if antibiotics have been given before an immune response can be mounted. Thus, most laboratories use a two-test approach for the serologic diagnosis of Lyme disease. Specimens are initially screened with the more sensitive ELISA or IFA. Positive (or equivocal) specimens are confirmed by the more specific IgG and IgM Western blot. The sensitivity and specificity of the ELISA and Western blot vary, depending on the time of specimen acquisition and, thus, clinical and exposure histories are important components of interpreting the serologic results. Because laboratory diagnosis is uncertain, a case definition has been developed to aid in recognition of Lyme disease: (a) the occurrence of erythema migrans no more than 30 days after exposure in an endemic area (i.e., where vector ticks are known to exist) or (b) the involvement of at least one of the three commonly affected organ systems producing neurologic, cardiovascular, or arthritic systems; and either (c) a positive serology, (d) isolation of B. burgdorferi, or (e) erythema migrans or positive serology
without a history of exposure. As noted already, the use of PCR technology to detect B. burgdorferi DNA in spinal fluid and synovial fluid may be useful in determining the natural history of Lyme disease and improving the reliability of the diagnosis. The diagnosis of Lyme disease has primarily been based on the presence of a characteristic clinical picture, exposure in an endemic area, and an elevated antibody response to B. burgdorferi (18). Unfortunately, serologic testing for Lyme disease is not standardized and both false-negative and more commonly false-positive results occur (20,21). Recently, Steere et al. analyzed the diagnoses, serologic test results, and treatment results of patients evaluated in a Lyme disease clinic (22). Among the 788 patients studied, 180 (23%) had active Lyme disease, usually arthritis, encephalopathy, or polyneuropathy (22). An additional 156 (20%) had evidence of previous Lyme disease and another current illness, most commonly chronic fatigue syndrome (22). The remaining 452 patients (57%) did not have Lyme disease. Of the patients without Lyme disease, 45% had positive serologic test results for Lyme disease in other laboratories (22). The overwhelming reason for failure to respond to antibiotic therapy was incorrect diagnosis in 322 (79%) of 409 patients who had been treated before referral (22). Thus, misdiagnosis, particularly overdiagnosis, and unreliable serologic testing are major problems that clinicians and patients face in dealing with Lyme disease. Treatment of Lyme disease is most successful when given early. Updated guidelines for treatment of Lyme disease have been recently published in 2000 (23). Treatment regimens are listed in Table 16.33. These guidelines from the Infectious Diseases Society of America suggest that for early Lyme disease, administration of doxycycline (100 mg twice daily) or amoxicillin (500 mg 3 times daily) for 14–21 days is the recommended treatment of early localized or early disseminated disease associated with erythema migrans, in the absence of neurological involvement or third-degree atrioventricular heartblock (23). Cefuroxime axetil (500 mg orally twice daily) should be reserved (because of higher cost) as an alternative agent for persons unable to take doxycycline or amoxicillin. Ceftriaxone (2g iv daily) is effective but is not superior to oral agents for the treatment of Lyme disease in the absence of neurologic involvement or third-degree atrioventricular heart block; thus it is not recommended as a first line agent in this situation (23).
TABLE 16.33. RECOMMENDED ANTIMICROBIAL REGIMENS FOR TREATMENT
OF PATIENTS WITH LYME DISEASEa
For early disease with acute neurological disease (e.g. meningitis or radiculopathy) ceftriaxone (2 g once daily) iv for 14–28 days) is recommended (23). Intravenous pencillin G 18–24 million units daily, divided into doses given every 4 hours is an acceptable alternative (23). In patients unable to tolerate penicillins and cephalosporins, doxycycline (200–400 mg/d) in 2 divided doses orally for 14–28 days is suggested; iv therapy with doxycycline is indicated in patients unable to take oral medication (23). Patients with third-degree atrioventricular heart block are treated similarly to those with neurologic manifestations except that hospitalization is recommended and a temporary pacemaker may be required (23). Other than doxycycline, which is contraindicated in pregnancy, the treatment of pregnant women is identical to that for non-pregnant patients. Lyme arthritis can usually be treated successfully with antimicrobial agents administered orally or intravenously (Table 16.33); doxycycline (100 mg twice daily orally) or amoxicillin (500 mg 3 times daily) for 28 days is recommended in patients without clinical manifestations of neurologic disease (23). Patients with arthritis and objective evidence of neurologic disease, the recommendation is iv ceftriaxone (2g once daily for 14–28 days) (23). Alternatives include iv cefotaxime (2g every 8 hours) or iv penicillin (18–24 million units daily) (23). The role of primary antimicrobial prophylaxis to prevent Lyme disease after tick bites is controversial. Kaslow (24) suggested that the geographic variability in the proportion of ticks infected, the relatively long duration of tick attachment required for transmission, the low likelihood of significant symptomatic infection, and the availability of highly effective therapy for erythema migrans and other early stages of Lyme disease make routine prophylaxis for a tick bite unnecessary. Shapiro et al. (6) recently provided data to support this approach, in a prospective, randomized, placebo-controlled study. They found that the risk of Lyme disease in the placebo-treated patients was 1.2% (95% CI, 0.1–4.1%), which was not statistically different from the risk in the amoxicillin-treated patients (0%; 95% CI, 0–1.5%). They concluded that even in an area in which Lyme disease is endemic, the risk of infection with B. burgdorferi after a recognized deer tick bite is so low that prophylactic antimicrobial treatment is not routinely indicated (6). On the other hand, Magid et al. (25), using a decision analysis approach, concluded that empiric treatment of patients with tick bites is indicated when the probability of B. burgdorferi infection after a bite is 0.036 or higher and may be preferred when the probability of infection ranges from 0.01 to 0.035. With a risk of infection less than 0.01, empiric therapy is not indicated. The IDSA guidelines suggest that routine prophylaxis is not indicated (23). Presently, the major measures available for prevention of Lyme disease include wearing protective clothing in tick-infested areas during summer months, showering after exposure, and checking for attached ticks (10). Any ticks that are found must be removed. This is best accomplished by grasping the tick with forceps near the point of
attachment and applying gentle traction. With the availability of a Lyme disease vaccine (LYMErix) another method of prevention is now available (26). this vaccine was approved by the FDA in 1998. Reported trials have demonstrated the efficacy of Lyme disease vaccine (27,28). The CDC estimates that the Lyme disease vcaccine is 76 % effective in preventing Lyme disease after three doses (26). See Chapter 25 for a detailed discussion of Lyme disease vaccine. CHAPTER REFERENCES Viruses Rubella 1. Centers for Disease Control and Prevention. Rubella and congenital rubella—United States, 1980–1983. MMWR Morb Mortal Wkly Rep 1983;32:505. 2. Centers for Disease Control and Prevention. Rubella and congenital rubella syndrome—United States, January 1, 1991–May 7, 1994. MMWR Morb Mortal Wkly Rep 1994;43:391–401. 3. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States 1996. MMWR Morb Mortal Wkly Rep 1996;45:3. 4. Centers for Disease Control and Prevention. Case definitions for infectious conditions under public health surveillance. MMWR Morb Mortal Wkly Rep 1997;46:29–30. 5. Centers for Disease Control and Prevention. Measles, mumps, and rubella-vaccine use and strategies for elimination of measles, rubella, and congenital rubella syndrome and control of mumps: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1998;47:1–57. 6. Centers for Disease Control and Prevention. Recommendations of the Immunization Practices Advisory Committee (ACIP): rubella prevention. MMWR Morb Mortal Wkly Rep 1984;33:301. 7. Mann JM, Preblud SR, Hoffman RE, et al. Assessing risks of rubella infection during pregnancy: a standardized approach. JAMA 1981;245:1647. 8. Centers for Disease Control and Prevention. Rubella in hospitals—California. MMWR Morb Mortal Wkly Rep 1983;32:37. 9. Plotkin SA, Katz M, Cordero JF. The eradication of rubella. JAMA 1999;281:561–562. 10. Daffos F, Grangeot-Keros L, Lebon P, et al. Prenatal diagnosis of congenital rubella. Lancet 1984;7:1.
Cytomegalovirus 1. Welier TH. The cytomegaloviruses; ubiquitous agents with protean clinical manifestations. N Engl J Med 1971;285:203–214,267–278. 2. Hanshaw JB. Congenital cytomegalovirus infection; a 15 year prospective study. J Infect Dis 1971;123:555–561. 3. Stagno S, Whitley RJ. Herpesvirus infections of pregnancy. Part I. Cytomegalovirus and Epstein-Barr virus infections. N Engl J Med 1985;313:1270. 4. Yow MD. Congenital cytomegalovirus disease: a NOW problem. J Infect Dis 1989;159:163–167. 5. Hanshaw JB, Scheiner AP, Maxley AW, et al. School failure and deafness after “silent” congenital cytomegalovirus. N Engl J Med 1976;295:468–470. 6. Stagno S, Pass RF, Cloud G. Primary cytomegalovirus infection in pregnancy: incidence, transmission to fetus and clinical outcome. JAMA 1986;256:1904–1908. 7. Chandler SH, Holmes KK, Wentworth BB, et al. The epidemiology of cytomegaloviral infection in women attending a sexually transmitted disease clinic. J Infect Dis 1985;152:597–604. 8. Hunter K, Stagno S, Capps E, et al. Prenatal screening of pregnant women for infections caused by cytomegalovirus, Epstein-Barr virus, herpesvirus, rubella, and Toxoplasma gondii. Am J Obstet Gynecol 1983;145:269. 9. Yow MO, Williamson DW, Leeds L, et al. Epidemiologic characteristics of cytomegalovirus
infection in mothers and infants. Am J Obstet Gynecol 1988;158:1189–1195. 10. Stagno S, Reynolds DW, Huang ES, et al. Congenital cytomegalovirus infection: occurrence in an immune population. N Engl J Med 1977;296:1254–1258. 11. Schopfer K, Louber E, Krech U. Congenital cytomegalovirus infection in newborn infants of mothers infected before pregnancy. Arch Dis Child 1978;53:536–539. 12. Stagno S. Maternal cytomegalovirus infection and perinatal transmission. Clin Obstet Gynecol 1982;25:563–576. 13. Fowler KB, Stagno S, Pass RF, et al. The outcome of genital cytomegalovirus infection in relation to maternal antibody status. N Engl J Med 1992;326:663–667. 14. Pass RF, Little EA, Stagno S, et al. Young children as a probable source of maternal and congenital cytomegalovirus infection. N Engl J Med 1987;316:1366. 15. Pass RF, Hutlo C. Group day care and cytomegaloviral infections of mothers and children. Rev Infect Dis 1986;8:599–605. 16. Adler SP. Molecular epidemiology of cytomegalovirus: viral transmission among children attending a day care center, their parents and care takers. J Pediatr 1988;112:366–372. 17. Taber LH, Frank AL, Yow MD, et al. Acquisition of cytomegaloviral infection in families with young children: a serological study. J Infect Dis 1985;151:948–952. 18. Adler SP. Cytomegalovirus and pregnancy. Curr Opin Obstet Gynecol 1992;4:670–675. 19. Stagno S, Reynolds DW, Pass RF, et al. Breast milk and the risk of cytomegalovirus infection. N Engl J Med 1980;302:1073–1076. 20. Stern J, Tucker SM. Prospective study of cytomegalovirus infection in pregnancy. Br Med J 1973;2:268. 21. Manif GRG, Egan EA, Held B, et al. The correlation of maternal cytomegalovirus infection during varying stages in gestation and neonatal involvement. J Pediatr 1972;80:17. 22. Boppana SB, Pass RF, Britt WJ. Virus-specific antibody responses in mothers and their newborn infants with asymptomatic congenital cytomegalovirus infections. J Infect Dis 1993;167:72–77. 23. Stagno S. Cytomegalovirus. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn. Philadelphia: WB Saunders, 1990:241–281. 24. Reynolds DW, Stagno S, Stubbs KG, et al. Inapparent congenital cytomegalovirus infection with elevated cord IgM levels: causal relation with auditory and mental deficiency. N Engl J Med 1974;290:291. 25. Pass RF, Stagno S, Meyers GJ, et al. Outcome of symptomatic congenital CMV infection: results of long-term longitudinal follow-up. Pediatrics 1980;66:758–762. 26. Hanshaw JB, Scheiner AP, Maxley AW, et al. School failure and deafness after “silent” congenital cytomegalovirus. N Engl J Med 1976;295:468–470. 27. Stagno S, Pass RF, Dworsky ME, et al. Congenital and perinatal cytomegalovirus infection. Semin Perinatol 1983;7:30–42. 28. Lange J, Rodeck CM, Morgan-Capner P. Prenatal serological diagnosis of intrauterine cytomegalovirus infection. Br Med J 1982;284:1673–1674. 29. Hogge WH, Thiagarajah S, Brerbridge AN, et al. Fetal evaluation by percutaneous blood sampling. Am J Obstet Gynecol 1988;158:132–136. 30. Weiner CP, Grose C. Prenatal diagnosis of congenital cytomegalovirus infection by virus isolation from amniotic fluid. Am J Obstet Gynecol 1990;163:1253–1255. 31. Hohlfeld P, Vial Y, Maillard-Brignon C, et al. Cytomegalovirus fetal infection: prenatal diagnosis. Obstet Gynecol 1991;78:615–618. 32. Lamey ME, Mulongo KN, Gadisseux J-F, et al. Prenatal diagnosis of fetal cytomegalovirus infection. Am J Obstet Gynecol 1992;166:91–94. 33. Donner C, Liesnard C, Content J, et al. Prenatal diagnosis of 52 pregnancies at risk for congenital cytomegalovirus infection. Obstet Gynecol 1993;82:481–486. 34. Antsaklis AJ, Daskalakis GJ, Mesogitis SA, et al. Prenatal disagnosis of fetal primary cytomegalovirus infection. Brit J Obstet Gynecol 2000;107:84–88. 35. Guerra B, Lazzarotto T, Quarta S, et al. Prenatal diagnosis of symptomatic congenital cytomegalovirus. Am J Obstet Gynecol 2000;183:476–482. 36. Azam AZ, Vialy, Fawer CL, Zufferey J, Hohlfeld P. Prenatal diagnosis of congenital cytomegalovirus infection. Obstet Gynecol 2001;97(3):443–448. 37. Collaborative DHPG Treatment Study Group. Treatment of serious cytomegalovirus infections with 9 (1,3-dihydroxy-2-proxymethyl) guanine in patients with AIDS and other immunodeficiencies. N Engl J Med 1986;314:801–805.
38. Dworsky ME, Welch K, Cassady G, et al. Occupational risk for primary cytomegalovirus infection among pediatric health care workers. N Engl J Med 1983;309:950–953. 39. Flowers RH, Torner JC, Farr BM. Primary cytomegalovirus infection in pediatric nurses: a meta-analysis. Infect Control Hosp Epidemiol 1988;9:491–496. 40. Yeager AS. Longitudinal, serological study of cytomegalovirus infection in nurses and in personnel without patient contact. J Clin Microbial 1975;2:448–452. 41. Ahlfors K, Ivarsson S-A, Johnsson T, et al. Risk of cytomegalovirus infection in nurses and congenital infection in their offspring. Acta Pediatr Scand 1981;70:819–823. 42. Friedman HM, Lewis MR. Nemerofsky DM, et al. Acquisition of cytomegalovirus infection among female employees at a pediatric hospital. Pediatr Infect Dis 1984;3:233–235. 43. Balfour CL, Balfour HH Jr. Cytomegalovirus is not an occupational risk for nurses in renal transplant and neonatal units. Results of a prospective surveillance study. JAMA 1986;256:1909–1914. 44. Hatherley LI. Is primary cytomegalovirus infection an occupational hazard for obstetric nurses? A serologic study. Infect Control 1986;7:452–455. 45. Balcarek KB, Bagley R, Cloud GA, et al. Cytomegalovirus infection among employees of a children's hospital: no evidence for increased risk with patient care. JAMA 1990;263:840–844.
Varicella-Zoster Virus 1. Charles D. Infections. In: Charles D. Infections in obstetrics and gynecology. Philadelphia: WB Saunders, 1980:162–166. 2. Gershan AA. Chicken pox, measles, and mumps. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn. Philadelphia: WB Saunders, 2001:683–732. 3. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States 1998. MMWR Morb Mortal Wkly Rep 1999;47:3. 4. Hermmann KL. Congenital and perinatal varicella. Clin Obstet Gynecol 1982;25:605–609. 5. Preblud SR, D'Angelo LJ. Chicken pox in the United States, 1972–1977. J Infect Dis 1979;140:257–260. 6. Smego RA, Asperilla MO. Use of acyclovir for varicella pneumonia during pregnancy. Obstet Gynecol 1991;78:1112–1116. 7. Sever J, White LR. Intrauterine viral infections. Annu Rev Med 1968;19:471. 8. Balducci J, Rodis JF, Rosengren S, et al. Pregnancy outcome following first trimester varicella infection. Obstet Gynecol 1992;79:5–6. 9. Pearson HE. Parturition varicella-zoster. Obstet Gynecol 1964;32:21–27. 10. Brunell PA. Varicella-zoster infections in pregnancy. JAMA 1967;199:315. 11. Garcia AGP. Fetal infection in chickenpox and alastrim, with histologic study of the placenta. Pediatrics 1963;32:895. 12. Abler C. Neonatal varicella. Am J Dis Child 1964;107:492. 13. Newman CGH. Perinatal varicella. Lancet 1965;2:1159. 14. Harris RE, Rhoades ER. Varicella pneumonia complicating pregnancy: report of a case and review of literature. Obstet Gynecol 1965;23:734. 15. Pickard RE. Varicella pneumonia in pregnancy. Am J Obstet Gynecol 1968;101:504. 16. Mendelow DA, Lewis GC. Varicella pneumonia during pregnancy. Obstet Gynecol 1969;33:98. 17. Geeves RB, Lindsay DA, Robertson TI. Varicella pneumonia in pregnancy with varicella neonatorum: report. Aust N Z J Med 1971;1:63. 18. Paryani SG, Arvin AM. Intrauterine infection with varicella-zoster virus after maternal varicella. N Engl J Med 1986;314:1542. 19. Siegel M, Fuerst HT. Low birth weight and maternal virus diseases. JAMA 1966;197:88. 20. Esmonde TF, Herdman G, anderson G. Chickenpox pneumonia: an association with pregnancy. Thorax 1989;44;812. 21. Cox SM, Cunningham FG, Luby J. Management of varicella pneumonia complicating pregnancy. Am J Perinatol 1990;7:300. 22. Broussard OF, Payne DK, George RB. Treatment with acyclovir of varicella pneumonia in pregnancy. Chest 1991;99:1045. 23. Baren J, Henneman P, Lewis R. Primary varicella in adults: pneumonia, pregnancy and hospital admission. Ann Emerg Med 1996;28:165.
24. Figueroa-Damian R, Arrendondo-Gareia JL. Parinatal outcome of pregnancies complicated with varicella infection during the first 20 weeks of gestation. Am J Perinatol 1997;14:411. 25. McGregor JA, Mark S, Crawford GP, et al. Varicella zoster antibody testing in the care of pregnant women exposed to varicella. Am J Obstet Gynecol 1987;157:281. 26. Brunell PA. Fetal and neonatal varicella-zoster infections. Semin Perinatol 1983;7:47. 27. LaForet E, Lynch CL. Multiple congenital defects following maternal varicella. N Engl J Med 1947;236:534–537. 28. Siegel M. Congenital malformations following chicken pox, measles, mumps and hepatitis: results of a cohort study. JAMA 1973;226:1521–1524. 29. Enders G. Varicella-zoster virus infection in pregnancy. Prog Med Virol 1984;29:166–196. 30. Pastuszak AL, Levy M, Schick B, et al. Outcome after maternal varicella infection in the first 20 weeks of pregnancy. N Engl J Med 1994;330:901–905. 31. Enders G, Miller E, Crodock-Watson J, et al. Consequences of varicella and herpes zoster in pregnancy: prospective study of 1739 cases. Lancet 1994;343:1547–1550. 32. Mouly F, Mirlesse V, Merited JF, et al. Prenatal diagnosis of fetal varicella-zoster virus infection with polymerase chain reaction of amniotic fluid in 107 cases. Am J Obstet Gynecol 1997;177:894–898. 33. Higa K, Dan K, Manabe H. Varicella-zoster infection during pregnancy: hypothesis concerning the mechanisms of congenital malformations. Obstet Gynecol 1987;69:214–222. 34. Scharf A, Scherr O, Enders, et al. Virus detection in the fetal tissue of a premature delivery with congenital varicella syndrome. J Perinatal Med 1990;18:317–322. 35. Isada NB, Paar DP, Johnson MP, et al. In utero diagnosis of congenital zoster virus infection by chorionic villus sampling and polymerase chain reaction. Am J Obstet Gynecol 1991;165:1727–1730. 36. Pretorius DH, Hayward I, Jaones KL, et al. Sonographic evaluation of pregnancies with maternal varicella infection. J Ultrasound Med 1992;11:459–463. 37. Meyers JD. Congenital varicella in term infants: risk considered. J Infect Dis 1974;129:215–217. 38. Miller E, Craddock-Watson JE, Ridehilgh MKS. Outcomes in newborn babies given anti-varicella-zoster immunoglobulin after perinatal maternal infection with varicella-zoster virus. Lancet 1989;2:371–373. 39. Brunell PA. Fetal and neonatal varicella-zoster infections. In: Amsley MS, ed. Virus infections in pregnancy. Orlando: Grune & Stratton, 1984:131. 40. Centers for Disease Control and Prevention. Prevention of Varicella. Recommendations of ACIP. MMWR 1996;45(RR-11): 1–36.
Rubeola (Measles) 1. Centers for Disease Control and Prevention. Annual summary 1982: reported morbidity and mortality in the United States. MMWR Morb Mortal Wkly Rep 1983;31:48–51. 2. Sever JL, Larsen JW Jr, Grossman JH III. Handbook of perinatal infections. Boston: Little, Brown and Company, 1979:63–64. 3. Siegel M. Congenital malformations following chicken pox, measles, mumps, and hepatitis. Results of a cohort study. JAMA 1973;226:1521–1524. 4. Measles prevention. Recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR Morb Mortal Wkly Rep 1989;38:9–12. 5. Centers for Disease Control and Prevention. Measles, mumps, and rubella-vaccine use and strategies for elimination of measles, rubella, and congenital rubella syndrome and control of mumps: recommendations of the advisory committee on immunization practices (ACIP). MMWR Morb Mortal Wkly Rep 1998;47:1–57.
Mumps 1. Siegel M, Fuerst HT. Low birth weight and maternal virus diseases. A prospective study of rubella, measles, mumps, chicken pox, and hepatitis. JAMA 1966;197:88. 2. Siegal M, Fuerst HT, Peress NS. Comparative fetal mortality in maternal virus disease. A prospective study on rubella, measles, mumps, chicken pox, and hepatitis. N Engl J Med
1966;274:768. 3. Centers for Disease Control and Prevention. Measles, mumps, and rubella-vaccine use and strategies for elimination of measles, rubella, and congenital rubella syndrome and control of mumps: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1998;47:1–57.
Influenza 1. Sever JL, Larsen JW Jr, Grossman JH III. Handbook of perinatal infections. Boston: Little, Brown and Company, 1980:45–50. 2. Weinstein L. Influenza-1918: a revisit? N Engl J Med 1976;294:1058–1060. 3. Centers for disease control and prevention. Surveillance for influenza—United States, 1994–95, 1995–96, and 1996–97 seasons. MMWR 2000;49:13–28. 4. Centers for disease control and prevention. Prevention and control of influena: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 2000;49(RR-3). 5. Bland PB. Influenza in relation to pregnancy and labor. Am J Obstet Dis Woman 1919;79:184–197. 6. Harris JW. Influenza occurring in pregnant women. A statistical study of thirteen hundred and fifty cases. JAMA 1919;72:978–980. 7. Finland M. Influenza complicating pregnancy. In: Charles D, Finland M, eds. Obstetrics and perinatal infections. Philadelphia: Lea & Febiger, 1973:355–398. 8. Freeman DW, Barno A. Deaths from Asian influenza associated with pregnancy. Am J Obstet Gynecol 1959;78:1172–1175. 9. Coffey VP, Jessup WJE. Congenital abnormalities. Ir J Med Sci 1955;349:30. 10. Coffey VP, Jessup WJE. Maternal influenza and congenital deformities: a prospective study. Lancet 1959;2:935–938. 11. Doll RA, Hill AB, Sakula J. Asian influenza in pregnancy and congenital defects. Br J Prev Soc Med 1960;14:167–172. 12. Hakosalo J, Saxen L. Influenza epidemic and congenital defects. Lancet 1971;2:1346–1347. 13. Hardy JMB, Ararowicz EN, Mannini A, et al. The effect of Asian influenza on the outcome of pregnancy, Baltimore, 1957–58. Am J Public Health 1961;51:1182–1188. 14. Griffiths PD, Ronalds CJ, Heath RB. A prospective study of influenza infections during pregnancy. J Epidemiol Community Health 1980;34:1224. 15. Wilson MG, Heins HL, Imagawa DT, et al. Teratogenic effects of Asian influenza. JAMA 1959;171:638–641. 16. Elizan TS, Ajero-Froehlich L, Labiyi A, et al. Viral infection in pregnancy and congenital CNS malformations in man. Arch Neurol 1969;20:115. 17. Brown GC. Maternal virus infection and congenital anomalies. A prospective study. Arch Environ Health 1970;21:362. 18. Amantadine for high risk influenza. Med Lett Drugs Ther 1978;20:25. 19. Larsen JW Jr. Influenza and pregnancy. Clin Obstet Gynecol 1982;25:599–603.
Enteroviruses 1. Hanshaw JB, Dudgeon JA. Viral diseases of the fetus and newborn. Philadelphia: WB Saunders, 1978:182–191. 2. Cherry JD. Enteroviruses. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn infant. Philadelphia: WB Saunders, 1983:290–334. 3. Cherry JD, Nelson DB. Enterovirus infections: their epidemiology and pathogenesis. Clin Pediatr 1966;5:659. 4. Kibrick S. Current status of coxsackie and ECHO viruses in human disease. Prog Med Virol 1964;6:27. 5. Grist NR, Bell EJ, Assad F. Enteroviruses in human diseases. Prog Med Virol 1978;24:114. 6. Cherry JD. Nonpolio enteroviruses: coxsackie viruses, echoviruses and enteroviruses. In: Feign RD, Cherry JD, eds. The textbook of pediatric infectious diseases. Philadelphia: WB Saunders, 1981.
7. Hardy JB. Viral infections in pregnancy. A review. Am J Obstet Gynecol 1965;93:1052. 8. Horstmann DM. Viral infections in pregnancy. Yale J Biol Med 1969;42:99. 9. Horn P. Poliomyelitis in pregnancy. A twenty year report from Los Angeles County, California. Obstet Gynecol 1955;6:121. 10. Bates T. Poliomyelitis in pregnancy, fetus and newborn. Am J Dis Child 1955;90:189. 11. Siegel M, Greenberg M. Poliomyelitis in pregnancy: effect on fetus and newborn infant. J Pediatr 1956;49:280. 12. Sever JL, Larsen JW Jr, Grossman JH III. Handbook of perinatal infections. Boston: Little, Brown and Company, 1979:64–66. 13. Brown GC, Karunas RS. Relationship of congenital anomalies and maternal infection with selected enteroviruses. Am J Epidemiol 1972;95:207–217. 14. Kleinman H, Prince JT, Mathey WE, et al. ECHO 9 virus infection and congenital abnormalities: a negative report. Pediatrics 1962;29:261–269. 15. Landsman JB, Grist NR, Ross CAC. ECHO 9 virus infection and congenital malformations. Br J Prev Soc Med 1964;18:152–156. 16. Hughes J, Wilfert CM, Moore M, et al. ECHO virus 14 infection associated with fatal neonatal hepatic necrosis. Am J Dis Child 1972;123:61–67. 17. Philip AGS, Larsen EJ. Overwhelming neonatal infection with ECHO 19 virus. J Pediatr 1973;82:391–397. 18. Kibrick S, Benirschke K. Acute septic myocarditis and meningoencephalitis in the newborn child infected with coxsackie virus group B, type 3. N Engl J Med 1956;255:883–884. 19. Brightman VJ, Scott TFM, Westphal M, et al. An outbreak of coxsackie B-5 virus infection in the newborn nursery. J Pediatr 1966;69:179–192. 20. Benirschke K, Pendleton ME. Coxsackie virus infection. An important complication of pregnancy. Obstet Gynecol 1958;12:305–309. 21. McLean DM, Donohue WL, Snelling CE, et al. Coxsackie B-5 virus as a cause of neonatal encephalitis and myocarditis. Can Med Assoc J 1961;85:1046.
Condylomata Acuminata 1. Lynch PJ. Condylomata acuminata (anogenital warts). Clin Obstet Gynecol 1985;28:142. 2. Genital human papillomavirus infections. Am Coll Obstet Gynecol Technical Bulletin no 193; June 1994. 3. Centers for Disease Control and Prevention. 1993 sexually transmitted disease guidelines. MMWR Morb Mortal Wkly Rep 1993;42[RR-14]:87. 4. Schwartz DB, Greenberg MD, Daoud Y, et al. Genital condylomas in pregnancy: use of trichloroacetic acid and laser therapy. Am J Obstet Gynecol 1988;158:1407–1416. 5. Shah K, Kashima H, Polk DF, et al. Rarity of cesarean delivery in cases of juvenile-onset respiratory papillomatosis. Obstet Gynecol 1986;68:795. 6. Puranen MH, Yliskoski MH, Saarikoski V, et al. Exposure of an infant to cervical human papillomavirus infection of the mother is common. Am J Obstet Gynecol 1997;176:1039–1045. 7. Tseng CJ, Liang CC, Soong YK, et al. Perinatal transmission of human papillomavirus in infants: relationship between infection rate and mode of delivery. Obstet Gynecol 1998;91:92–96. 8. Watts DH, Koutsky LA, Holmes KK, et al. Low risk of perinatal transmission of human papillomavirus: results from a prospective cohort study. Am J Obstet Gynecol 1998;178:365–373. 9. Chamberlain MJ, Reynolds AL, Yeoman WB. Toxic effect of podophyllum application in pregnancy. Br Med J 1972;3:391–392. 10. Odom LD, Plouffe L, Butler WJ. 5-Fluorouracil exposure during the period of conception: report on two cases. Am J Obstet Gynecol 1990;163:76–77. 11. Otaño L, Amestay G, Paz J, et al. Periconceptional exposure to topical 5-fluorouracil [Letter]. Am J Obstet Gynecol 1992;166(1):263–264. 12. Berman A, Matsunaga J. Bhatia NN. Cervical cryotherapy for condylomata acuminata during pregnancy. Obstet Gynecol 1987;69:47–50. 13. Young RL, Acosta AA, Kaulman RH. The treatment of large condylomata acuminata complicating pregnancy. Obstet Gynecol 1973;41:65–73. 14. Buck HW. Imiquimod (Aldara Cream). Infect Dis Obstet Gynecol 1998;6:49–51.
Human Parvovirus 1. Anderson IJ. Role of parvovirus B19 in human disease. Pediatr Infect Dis 1987;6:911. 2. Thurn J. Human parvovirus B19: historical and clinical review. Rev Infect Dis 1988;10:1005–1011. 3. Sergeant GR, Topley JM, Mason K, et al. Outbreak of aplastic crisis in sickle cell anemia associated with parvovirus-like agent. Lancet 1981;2:595–597. 4. Pattison JR, Jones SE, Hodgan J, et al. Parvovirus infection and hypoplastic crisis in sickle cell anemia. Lancet 1981;1:664–665. 5. Anderson MJ, Pattison JR. The human parvovirus: brief review. Arch Virol 1984;82:137–148. 6. Anderson MJ, Jones SE, Fisher-Hoch SP, et al. Human parvovirus, the cause of erythema infectiosum (fifth disease)? Lancet 1983;1:1378. 7. Knott PD, Welpy GAC, Anderson MJ. Serologically proved intrauterine infection with parvovirus. Br Med J 1984;289:1660. 8. Brown T, Anand A, Ritchie LD, et al. Intrauterine parvovirus infection associated with hydrops fetalis. Lancet 1984;2:1033–1034. 9. Woolf AD, Campion GV, Chishick A, et al. Clinical manifestations of human parvovirus B19 in adults. Arch Int Med 1989;149:1153. 10. Gillespie SM, Cartler ML, Asch S, et al. Occupational risk of human parvovirus B19 infection for school and daycare personnel during an outbreak of erythema infectiosum. JAMA 1990;263:2061. 11. Bell LJ, Naides SJ, Staffman P, et al. Human parvovirus B19 infection among hospital staff members after contact with infected patients. N Engl J Med 1989;321:485–491. 12. Adler SP, Manganello AMA, Koch WC, et al. Risk of human parvovirus B19 infections among school and hospital employees during endemic periods. J Infect Dis 1993;168:361–368. 13. Valeur-Jensen AK, Pedersen CB, Westergaard T, et al. Risk factors for Parvovirus B19 infection in pregnancy. JAMA 1999;281:1099–1105. 14. Anderson LJ. Human parvovirus. J Infect Dis 1990;161:603–608. 15. Levy R, Weissman A, Blomberg G, Hagay ZJ. Infection by Parvovirus B19 during pregnancy: a review. Obstet Gynecol Survey 1997;52:254–59. 16. Markenson GR, Yancey MK. Parvovirus B19 infections in pregnancy. Seminars Perinatol 1998;22:309–17. 17. Rodis JF. Parvovirus infection. Clinical Obstet Gynecol 1999;42:107–20. 18. Tolfvenstam T, Ruden U, Broliden K. Evaluation of serological assays for identification of parvovirus B19 immunoglobulin M. Clin Diag Lab Immunol 1996;3:147–50. 19. Anand A, Gray ES, Brown T, et al. Human parvovirus infection in pregnancy and hydrops fetalis. N Engl J Med 1987;156:17. 20. Woernle CH, Anderson LJ, Tatersall P, et al. Human parvovirus B19 infection during pregnancy. J Infect Dis 1987;156:17. 21. Rodis JF, havick TJ, Quinn DL, et al. Human parvovirus infection in pregnancy. Obstet Gynecol 1988;72:733. 22. Centers for Disease Control. risks associated with human parvovirus B19 infection. MMWR 1989;38:81. 23. Schwartz TF, Roggendor M, Hattentrager B, et al. Human parvovirus B19 infection in pregnancy. Lancet 1988;2:566–567. 24. Public Health Laboratory Service Working Party on Fifth Disease. Prospective study of human parvovirus (B19) infection during pregnancy. Br Med J 1990;300:1166–1170. 25. Rodis JF, Quiinn DL, Gary W, et al. Management and outcomes of pregnancies complicated by human B19 parvovirus infection. A prospective study. Am J Obstet Gynecol 1990;163:1168–1171. 26. Torok TJ, Anderson LJ, Gary GW, Yoder CGT. Reproductive outcomes following humman parvovirus B19 infection in pregnancy (abstract 374). Program and abstracts of the 31st Interscience Conference on Antimicrobial Agents and Chemotherapy for Microbiology 1991;328. 27. Guidozzi F, Ballot D, Rothberg A. Human B19 parvovirus infection in a obstetric population: a prospective sutdy determining fetal outcome. J Reprod Med 1994;39:36–38. 28. Gratacos E. Torres P, Vidal J, et al. The incidence of human parvovirus B19 infection during pregnancy and its impact on perinatal outcome. J Infect Dis 1995;171:1360–1363. 29. Harger JH, Adler SP, Koch WC, et al. Prospective evaluation of 618 pregnant women exposed to parvovirus B19: risks and symptoms. Obstet Gynecol 1998;91:413–20.
30. Miller E, Fairley CK, Cohen BJ, et al. Immediate and long-term outcome of human parvovirus B19 infection in pregnancy. Brit J Obstet Gynecol 1998;105:174–78. 31. Bernstein IM, Capeless EL. Elevated maternal serum alphafetoprotein and hydrops fetalis in association with fetal parvovirus B19 infection. Obstet Gynecol 1989;74:456–457. 32. Carrington D, Gilmore DH, Whittle MJ, et al. Maternal serum alphafetoprotein. A marker of fetal aplastic crisis during intrauterine human parvovirus infection. Lancet 1987;1:433–435. 33. Naicles SJ, Weiner CP. Antenatal diagnosis and palliative treatment of nonimmune hydrops fetalis secondary to fetal parvovirus B19 infection. Prenat Diagn 1989;9:105–114. 34. Peters MT, Nikofoides KH. Cordocentesis for the diagnosis and treatment of human fetal parvovirus infection. Obstet Gynecol 1990;75:501–504. 35. Sahakian V, Weiner CP, Naides SJ, et al. Intrauterine transfusion treatment of nonimmune hydrops fetalis secondary to human parvovirus B19 infection. Am J Obstet Gynecol 1991;164:1090–1091. 36. Soothill P. Intrauterine blood transfusion for nonimmune hydrops fetalis dur to parvovirus B19 infection. Lancet 1988;2:566–567. 37. Humphrey W, Magoon M, O'Shoughnessy R. Severe nonimmune hydrops secondary to parvovirus B19 infection. Spontaneous reversal in utero and survival of a term infant. Obstet Gynecol 1991;78:900–902. 38. Torok TJ, Wang AY, Gary GW, et al. Prenatal diagnosis of intrauterine infection with parvovirus B19 by the polymerase chain reaction technique. Clin Infect Dis 1992;14:149–155. 39. Fairles CK, Smoleniec JS, Caul OE, et al. Observational study of the effect of intrauterine transfusions on outcome of fetal hydrops after parvovirus B19 infection. Lancet 1995;346:1335–37.
Bacteria Listeriosis 1. Gellin BG, Broome CV. Listeriosis. JAMA 1989;261:1313–1320. 2. Charles D. Infections in obstetrics and gynecology. Philadelphia: WB Saunders, 1980:192–197. 3. Sever JC, Larsen JW Jr, Grossman JH III. Listeriosis. In: Sever JL, Larsen JW Jr, Grossman JH III, eds. Handbook of perinatal infections. Boston: Little, Brown and Company, 1980:141–144. 4. Ahlfors C, Goetzman BW, Holsted CC, et al. Neonatal listeriosis. Am J Dis Child 1977;131:405–408. 5. Anderson G. Listeria monocytogenes septicemia in pregnancy. Obstet Gynecol 1975;46:102–104. 6. Seeliger HPR, Finger H. Listeriosis. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn infant. Philadelphia: WB Saunders, 1983:264–289. 7. Schlech WF, Lavigne PM, Bortolusse R, et al. Epidemic listeriosis: evidence for transmission by food. N Engl J Med 1983;308:203–206. 8. Fleming D, Cochi S, MacDonald K, et al. Pasteurized milk as a vehicle of infection in an outbreak of listeriosis. N Engl J Med 1985;312:404–407. 9. Linnan MJ, Mascola L, Dong Lou X, et al. Epidemic listeriosis associated with Mexican-style cheese. N Engl J Med 1988;319:823–828. 10. Bula C, Bille J, Mean F, et al. Epidemic food-borne listeriosis in western Switzerland. I. Description of 58 adult cases. In: Program and abstracts of the 28th Interscience Conference of Antimicrobial Agents and Chemotherapy; October 26, 1988; Los Angeles, CA. 11. Skogberg K, Syrjanen J, Johkola M, et al. Clinical presentation and outcome of listeriosis in patients with and without immunosuppressive therapy. Clin Infect Dis 1992;14:815–821. 12. Cherubin CE, Appleman MD, Heseltine PN, et al. Epidemiological spectrum and current treatment of listeriosis. Rev Infect Dis 1991;13:1108–1114. 13. Ciesielski CA, Hightower AW, Parsons SK. Listeriosis in the United States 1980–1982. Arch Intern Med 1988;148:1416–1419. 14. Gellin BG, Broome CV, Hightower AW. Geographic differences in listeriosis in the United States. In: Programs and abstracts of 27th Interscience Conference on Antimicrobial Agents and Chemotherapy. October 5, 1987; New York, NY. 15. Schwartz B, Ciesielski CA, Gaventa S, et al. Association of sporadic listeriosis with consumption of uncooked hot dogs and undercooked chicken. Lancet 1988;2:779–782. 16. Kampelmacher EH, van Noorle Jansen LM. Isolation of Listeria monocytogenes from feces of
clinically healthy humans and animals. Zentralbl Bakterial Mikrobiol Hyg A 1969;211:353–359. 17. Boucher M, Yonekura ML. Perinatal listeriosis (early onset): correlation of antenatal manifestations and neonatal outcome. Obstet Gynecol 1986;68:593–597. 18. Filice GA, Contrell HF, Smith AB, et al. Listeria monocytogenes infection in neonates. Investigation of an epidemic. J Infect Dis 1978;138:17–23. 19. Kalstone C. Successful antepartum treatment of listeriosis. Am J Obstet Gynecol 1991;164:57–58. 20. Centers for disease control and prevention. Multistate outbreak of Listeriosis—United States, 2000. MMWR 2000;49:1129–1130.
Tuberculosis 1. Centers for Disease Control and Prevention. Targeted tuberculin testing and treatment of latent tuberculosis infection. MMWR Morb Mortal Wkly Rep 2000;49[RR-6]:1–51. 2. Cantwell MF, Shehab ZM, Castello AM, et al. N Engl J Med 1994;330:1051–1054. 3. Margono F, Mroueh J, Garely A. Resurgence of active tuberculosis among pregnant women. Obstet Gynecol 1994;83:911–914. 4. Centers for Disease Control and Prevention. Initial therapy for tuberculosis in the era of multidrug resistance. MMWR Morb Mortal Wkly Rep 1993;42[RR-7]:1–8. 5. Sanders CV and Hill MK. Tuberculosis in pregnancy. Cont Obstet Gynecol 1999;Mar:15–20. 6. Centers for Disease Control and Prevention. Screening for tuberculosis and tuberculous infection in high-risk populations and the use of preventive therapy for tuberculous infection in the United States. MMWR Morb Mortal Wkly Rep 1990;39[RR-8]:1–12. 7. Centers for Disease Control and Prevention. Essential components of a tuberculosis prevention and control program. Screening for tuberculosis and tuberculosis infection in high-risk populations: recommendations of the advisory council for the elimination of tuberculosis. MMWR Morb Mortal Wkly Rep 1995;44:1–34. 8. Snider DE, Layde PM, Johnson MW, et al. Treatment of tuberculosis during pregnancy. Am Rev Respir Dis 1980;122:65. 9. Conway N, Birt BD. Streptomycin in pregnancy: effect on the fetal ear. Br Med J 1965;2:260.
PROTOZOA Toxoplasmosis 1. Lopez A, Dietz VJ, Wilson M, et al. Prevailing congenital toxoplasmosis. MMWR Morb Mortal Wkly Rep 2000;49:59–75. 2. Sever JL, Ellenberg JH, Ley AC, et al. Toxoplasmosis: maternal and pediatric findings in 23,000 pregnancies. Pediatrics 1988;82:181–192. 3. Fuchs F, Kimball AC, Kean BH. The management of toxoplasmosis in pregnancy. Clin Perinatol 1974;1:407. 4. Hunter K, Slagno S, Capps E, et al. Prenatal screening of pregnant women for infections caused by cytomegalovirus. Epstein-Barr virus, herpesvirus, rubella, and Toxoplasma gondii. Am J Obstet Gynecol 1983;145:269–273. 5. Wong SY, Remington JS. Toxoplasmosis in pregnancy. Clin Infect Dis 1994;18:853–861. 6. Guerina NG, Hsu HW, Meissner HC, et al. Neonatal serologic screening and early treatment for congenital Toxoplasma gondii infection. N Engl J Med 1994;330:1858–1863. 7. Desmont G, Couvreur J. Congenital toxoplasmosis. A prospective study of 378 pregnancies. N Engl J Med 1974;290:1110. 8. Remington JS, Desmonts G. Toxoplasmosis. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn infants, 4th ed. Philadelphia: WB Saunders, 1995. 9. Couvreur J, Thulliez P, Daffos F. Toxoplasmosis. In: Charles D. Koss handbook of infectious disease obstetric and perinatal infections. St. Louis: Mosby–Year Book, 1993. 10. Food and Drug Administration. FDA public health advisory: limitations of Toxoplasma IgM commercial test kits; July 25, 1997. 11. Sever JL. Torch tests and what they mean. Am J Obstet Gynecol 1985;152:495–498.
12. Daffos F, Forestier F, Capella-Pavlovsky M, et al. Prenatal management of 746 pregnancies at risk for congenital toxoplasmosis. N Engl J Med 1988;318:271. 13. Foulon W, Naessens A, Mahler T, et al. Prenatal diagnosis of congenital toxoplasmosis. Obstet Gynecol 1990;76:769–772. 14. Hohlfeld P, MacAleese J, Capella-Pavlovski M, et al. Fetal toxoplasmosis: ultrasonographic signs. Ultrasound Obstet Gynecol 1991;1:241–244. 15. Grover CM, Thulliez P, Remington JS, et al. Rapid prenatal diagnosis of congenital toxoplasma infection by using polymerase chain reaction and amniotic fluid. J Clin Microbiol 1990;28:2297–2301. 16. Hohlfeld P, Daffos F, Costa JM, et al. Prenatal diagnosis of congenital toxoplasmosis with a polymerase-chain-reaction test on amniotic fluid. N Engl J Med 1994;331:695–699. 17. Foulon W, Pinon JM. Stray-Pederson B. Prenatal diagnosis of congenital toxoplasmosis: a multicenter evaluation of different diagnostic parameters. Am J Obstet Gynecol 1999;181:843–847. 18. Frenkel JK. Congenital toxoplasmosis: prevention or palliation? Am J Obstet Gynecol 1981;141:359. 19. Wilson CB, Remington JS. What can be done to prevent congenital toxoplasmosis? Am J Obstet Gynecol 1980;138:357. 20. Foulon W, Naessans A, Louwers S, et al. Impact of primary prevention on the incidence of toxoplasmosis during pregnancy. Obstet Gynecol 1988;72:363–366. 21. Sever JL. Toxoplasmosis in pregnancy. Cont Obstet Gynecol 1998;21–24.
Lyme Disease 1. Centers for Disease Control and Prevention. Lyme disease—United States, 1999. MMWR Morb Mortal Wkly Rep 2001;50:181–185. 2. Steere AC, Malawista SE, Snydman DR, et al. Lyme arthritis in children and adults in three Connecticut communities. Arthritis Rheum 1977;20:7–17. 3. Steere AC, Malawista SE, Hardin JA, et al. Erythema chronicum migrans and Lyme arthritis: the enlarging clinical spectrum. Ann Intern Med 1977;86:685–698. 4. Steere AC. Lyme disease. N Engl J Med 1989;321:586–596. 5. Asbrink E, Hovmark A. Early and late cutaneous manifestations of Ixodes-borne borreliosis (erythema migrans borreliosis, Lyme borreliosis). Ann N Y Acad Sci 1988;539:4–15. 6. Shaprio ED, Gerber MA, Holabird NB, et al. A controlled trial of antimicrobial prophylaxis for Lyme disease after deer-tick bites. N Engl J Med 1992;237:1769–1773. 7. Luft BJ, Steinman CR, Neimark HC, et al. Invasion of the central nervous system by Borrelia burgdorferi in acute disseminated infection. JAMA 1992;267:1364–1367. 8. Nocton JJ, Dressler F, Rutledge BJ, et al. Detection of Borrelia burgdorferi: DNA by polymerase chain reaction in synovial fluid from patients with Lyme arthritis. N Engl J Med 1994;330:229–234. 9. Ackerman R, Rehse-Klupper B, Gallmer E, et al. Chronic neurologic manifestations of erythema migrans borreliosis. Ann N Y Acad Sci 1988;539:16–23. 10. Smith LG Jr, Pearlman M, Smith LG, et al. Lyme disease: a review with emphasis on the pregnant woman. Obstet Gynecol Survey 1991;46:125–130. 11. Schlesinger PA, Duray PH, Burke BA, et al. Maternal-fetal transmission of the Lyme disease spirochete, Borrelia burgdorferi. Ann Intern Med 1985;103:67–69. 12. Weber K, Bratzke HJ, Neubert U, et al. Borrelia burgdorferi in a newborn despite oral penicillin for Lyme borreliosis during pregnancy. Pediatr Infect Dis J 1988;7:286–289. 13. MacDonald A. Human fetal borreliosis, toxemia of pregnancy, and fetal death. Zentralbl Bakteriol Hyg A 1986;263:189. 14. MacDonald AB, Benach JL, Burgdorfer W. Stillbirth following Lyme disease. N Y J Med 1987;8:615. 15. Markowitz LE, Steere AC, Benach JL, et al. Lyme disease during pregnancy. JAMA 1986;255:3394–3396. 16. Ciesielski CA, Russell H, Johnson S, et al. Prospective study of pregnancy outcome in women with Lyme disease. In: Program and abstracts of the 27th Interscience Conference Antimicrobial Agents Chemotherapy, 1987. 17. Williams CL, Benach JL, Curran AS, et al. Lyme disease during pregnancy: a cord blood serosurvey. Ann N Y Acad Sci 1988;539:504–506.
18. Nadal D, Hunziker VA, Bucher HV, et al. Infants born to mother with antibodies against Borrelia burgdorferi at delivery. Eur J Pediatr 1989;148:426. 19. Strobino BA, Williams CL, Abid S, et al. Lyme disease and pregnancy outcome. A prospective study of two thousand prenatal patients. Am J Obstet Gynecol 1993;169:367–374. 20. Hedberg CW, Osterhold MT, MacDonald KL, et al. An interlaboratory study of antibody to Borrelia burgdorferi. J Infect Dis 1987;155:1325–1327. 21. Bakken LL, Case KL, Callster SM, et al. Performance of 45 laboratories participating in a proficiency testing program for Lyme disease serology. JAMA 1992;268:891–895. 22. Steere AC, Taylor E, McHugh GI, et al. The over diagnosis of Lyme disease. JAMA 1993;269:1812–1816. 23. Kaslow RA. Current perspective on Lyme borreliosis. JAMA 1992;267:1381–1383. 24. Magid D, Schwartz B, Craft J, et al. Prevention of Lyme disease after tick bites. A cost-effectiveness analysis. N Engl J Med 1992;327:534–541.
17 PREMATURE RUPTURE OF THE MEMBRANES Infectious Diseases of the Female Genital Tract
17 PREMATURE RUPTURE OF THE MEMBRANES Definitions Incidence Etiology Diagnosis Consequences of Premature Rupture of the Membranes Onset of Labor Effect of Tocolytic Drugs Complications Treatment Considerations Preterm Premature Rupture of the Membranes Term Premature Rupture of the Membranes Maternal Mortality Management Options Premature Rupture of the Membranes at Less than 25 Weeks Premature Rupture of the Membranes at 25 to 32 Weeks Premature Rupture of the Membranes at 33 to 36 Weeks Premature Rupture of the Membranes at or Near Term (35 Weeks or More) Chapter References
Premature rupture of the fetal membranes (PROM) is one of the most common problems in obstetrics, complicating approximately 5% to 10% of term pregnancies and up to 30% of preterm deliveries. Within the last 5 years, considerable progress has been made in aspects of our understanding and treatment of PROM. Although the etiology of PROM usually is not clinically evident in any given case, there is a considerable degree of consensus regarding management options. The problem is intricate; major variables influencing the outcome of a study are gestational age, date of the study, and population features. Variables affecting outcome include use of corticosteroids, tocolytics, and more potent antibiotics, and innovative approaches using various tests (such as amniocentesis, ultrasound, biophysical testing, and C-reactive maternal serum protein). Of major importance is the marked improvement in survival of low-birthweight infants.
DEFINITIONS Nearly all recent publications are in agreement by defining PROM as rupture at any time before the onset of contractions. Unfortunately, “premature” also carries the connotation of preterm pregnancy. To avoid confusion, we use the word “preterm” to refer to gestational age less than 37 weeks. The latent period is defined as the time from membrane rupture to onset of contractions. It is to be distinguished from a similar term, “latent phase,” which designates the early
phase of labor before the active phase. Respiratory distress syndrome (RDS) requires careful definition. Commonly used criteria include early onset after birth; tachypnea, expiratory grunting, and retractions; cyanosis and hypoxia; and a chest radiograph showing a reticulogranular pattern with air bronchograms. Most studies use these criteria, but some use the term without defining RDS. Various terms have been used recently to describe presumed maternal or perinatal infections related to PROM. During labor, designations have included “fever in labor,” “intrapartum fever,” “chorioamnionitis,” “amnionitis,” and “intrauterine infection.” The degree of temperature used to define “fever” is selected arbitrarily. The latter three terms usually are presumptive, based on combinations of maternal fever, uterine irritability or tenderness, leukocytosis, or purulent cervical discharge. After delivery, maternal infection is referred to as “endometritis” or “postpartum infection.” These diagnoses usually are based on fever, uterine tenderness, and exclusion of other sources of fever. In a few recent studies, presumed maternal infections were confirmed by reports of blood or genital tract cultures. In neonates, the most common term used to report infection was neonatal sepsis, but this may mean strictly a positive blood culture or simply clinical signs or symptoms of sepsis. Neonatal meningitis and pneumonia also were noted in a few studies. Prophylactic administration of antibiotics to the mother, as in many reports, likely would influence detection of bacteremia in the newborn.
INCIDENCE The incidence of PROM ranges from 5% to 10% of all deliveries, and preterm PROM occurs in approximately 1% of all pregnancies (1). Approximately 70% of cases of PROM occur in pregnancies at term, but more than 50% of cases in referral centers may occur in preterm pregnancies (2). Premature rupture of the membranes is the clinically recognized precipitating cause of about one third of all preterm births. Despite some progress in prolonging the latent period after preterm PROM and preventing recurrence (in women with bacterial vaginosis), preterm PROM remains a leading contributor to the overall problem of premature birth.
ETIOLOGY In the vast majority of cases, the etiology is not clinically evident. Earlier studies had identified selected clinical conditions, such as cervical incompetence and polyhydramnios, as risk factors evident in some cases of PROM. In a large case-control study (341 cases of preterm PROM and 253 gestational age-matched controls), three factors were associated with preterm PROM in a multifactorial analysis (3). These three factors were previous preterm delivery (odds ratio [OR], 2.5; 95% confidence interval [CI], 1.4–2.5), cigarette smoking (stopped during pregnancy: OR, 1.6; 95% CI, 0.8–3.3; continued during pregnancy: OR, 2.1; 95% CI, 1.4–3.1), and bleeding (first trimester: OR, 2.4; 95% CI, 1.5–3.9; third trimester: OR, 6.5; 95% CI, 1.9–23; more than one
trimester: OR, 7.4; 95% CI, 2.2–26). This study enrolled controls at the same gestational age as cases (thus correcting for the decreasing frequency of coitus closer to term) and found no association between coitus and PROM. In a scholarly review of the etiology of preterm PROM, Parry and Strauss (4) identified numerous potential causes in any given case. These causes included a generalized decrease in tensile strength of membranes, local defects in the membranes, decreased amniotic fluid (AF) collagen, a change in collagen structure, uterine irritability, apoptosis, collagen degradation, and membrane stretch. There is substantial evidence that subclinical infection may be a cause of PROM and not merely its result. In data from the Collaborative Perinatal Project (1959–1966), acute inflammation of the placental membranes was twice as common when membranes ruptured within 4 hours before labor than when the membranes ruptured after the onset of labor (5). Further evidence is provided by older bacteriologic studies showing associations between anaerobic isolates in endocervical cultures and PROM. More support for a role of infection is provided by studies showing an association between clinically diagnosed bacterial vaginosis (a condition with a shift in flora from lactobacilli to anaerobes, genital mycoplasmas, and Gardnerella vaginalis) and preterm birth/preterm PROM (see Chapter 19 for details.) Investigations into placental histology have provided correlates with clinical outcomes in cases of preterm PROM. Overall, in 235 cases, acute inflammation was seen in 43.4%, vascular lesions in 20.4%, inflammatory plus vascular lesions in 20.4%, normal findings in 13.2%, and “other” findings in 2.5% (Fig. 17.1).
FIGURE 17.1. Placental histology in 235 cases of preterm premature rupture of the membranes. From ref. 6.
When acute inflammation was seen in the placenta (either by itself or mixed with vascular lesions), birth at less than 26 weeks was more common. Delivery for suspected or proved clinical infection also was more common (Fig. 17.2 and Fig. 17.3) (6). McGregor et al. (7) demonstrated that some genital bacteria elaborate enzymes such as proteases and collagenases that may act to weaken the membranes. Additional discussion of the relationship between infections and PROM is presented in Chapter 19.
FIGURE 17.2. Outcome of premature rupture of the membranes by placental histology: percentage of births at less than 26 weeks. From ref. 6.
FIGURE 17.3. Outcome of premature rupture of the membranes by placental histology: delivery for infection. *Suspected or proved. From ref. 6.
DIAGNOSIS In most instances, PROM is readily diagnosed by history, physical findings, and simple laboratory tests. Standard tests are based on determination of pH (Nitrazine test) or detection of a “ferning” pattern or fetal cells. Although these tests are accurate in approximately 95% of cases, each has well-known false-positive and false-negative results, especially in patients with small amounts of AF in the vagina. Other approaches to diagnosing PROM in such equivocal cases include biochemical and histochemical tests and intraamniotic injection of various dyes; however, these methods have not been widely used in practice.
Ultrasound examination has been used widely, because oligohydramnios suggests PROM, but there have been no evaluations of its sensitivity and specificity. In 1991, Lockwood and colleagues (8) observed that the presence of fetal fibronectin in the cervicovaginal secretions of second-trimester and third-trimester pregnant women identified a group at high risk for preterm delivery. The fibronectins are a family of proteins found in plasma and extracellular matrix. Fetal fibronectin normally is found in AF and placental tissue (8). Eriksen et al. (9) reported a multicenter clinical trial comparing fetal fibronectin detection with standard tests for detection of ROM at term. There were 339 women with a clinical history of ROM and 67 controls. Fetal fibronectin showed excellent sensitivity (98.2%) but low specificity. The authors speculated that fetal fibronectin in cervicovaginal secretions may be a marker for impending labor, even without frank ROM.
CONSEQUENCES OF PREMATURE RUPTURE OF THE MEMBRANES Onset Of Labor At term, the onset of labor occurs within 24 hours after membrane rupture in 80% to 90% of patients (1). Among patients with PROM prior to term, longer latent periods occur. Latent periods of more than 24 hours occur in 57% to 83%, more than 72 hours in 15% to 26% (10), and 7 days or more in 19% to 41%. Johnson et al. (10) showed an inverse relationship between gestational age and the proportion of patients with latent periods longer than 3 days. For pregnancies between 25 and 32 weeks, 33% had latent periods longer than 3 days; for pregnancies of 33 to 34 weeks and 35 to 36 weeks, the corresponding values were 16% and 4.5%, respectively. Additional data on the natural history of preterm PROM (before 34 weeks) were provided by a population-based study at a hospital serving mainly indigent patients. (Data from referral institutions are skewed toward patients with longer latent periods; thus, a population-based study provides additional insight.) In 1988, Cox and colleagues (11) from Parkland Memorial Hospital in Dallas noted that 204 (76%) of 267 patients were already in labor at the time of admission. An additional 13 (5%) had an indicated delivery within a short time after admission, leaving only 50 patients (19%) as candidates for expectant management. Of these 50 patients, 30 (60%) went into spontaneous labor within less than 48 hours (11). Effect Of Tocolytic Drugs Two controlled trials have evaluated the use of tocolytic drugs in patients with PROM. Garite et al. (12) randomly assigned patients with PROM at 25 to 30 weeks' gestation to either expectant management or ritodrine tocolysis when labor developed. There were 39 patients in the tocolysis group and 40 in the expectant group. Corticosteroids were not used. Results are summarized in Table 17.1. Use of tocolytics showed no statistically significant or clinically meaningful benefit in increasing birthweight, gestational age, or time to delivery or in decreasing RDS. Of interest, there was a twofold increase in chorioamnionitis in the ritodrine group (p = 0.12), implicating labor as a potential sign of subclinical infection.
TABLE 17.1. RITODRINE VERSUS EXPECTANCY IN PREMATURE RUPTURE OF THE MEMBRANE AT 25 TO 30 WEEKS' GESTATION
In 1988, Weiner and coworkers (13) reported a similarly designed study. Patients with PROM at 34 weeks' gestation were randomized to liberal use of tocolytics (usually ritodrine, when there were three or more contractions in an hour) or to expectancy. As in the study by Garite et al. (12), no corticosteroids were used. Magnesium sulfate was added to ritodrine as needed in the tocolytic group in an attempt to ablate contractions. Outcomes of this well-designed study are given in Table 17.2. No significant differences in overall outcome measures were seen. The authors observed prolongation of intrauterine time after the onset of contractions in the tocolytic group (105 ± 157 hours vs. 62 ± 72 hours; p = 0.06) and a significant increase in intrauterine time after onset of contractions for pregnancies of less than 28 weeks (232 ± 312 hours vs. 53 ± 87 hours; p = 0.05). However, there was no identifiable perinatal benefit accompanying these increases in intrauterine time, and there were no cost savings.
TABLE 17.2. NEONATAL OUTCOME IN A TRIAL OF TOCOLYTICS VERSUS EXPECTANCY WITH PREMATURE RUPTURE OF THE MEMBRANES £34 WEEKS
Earlier studies had shown a significant reduction in delivery at less than 24 hours when tocolytics were used, but these studies were small and had substantial design problems (14). In 1995, a comparison of short-term versus long-term tocolysis in preterm PROM at 26 to 35 weeks showed an adverse effect of “long-term tocolysis” (15). Patients with preterm PROM at 26 to 35 weeks were randomized to receive an intravenous b-mimetic drug either for less than 48 hours (n = 105) or until delivery (n = 136). All patients received corticosteroids, and group B streptococci (GBS) and gonococci were treated. There was no significant difference in the latent period or in neonatal infection, but there was a significant increase in both chorioamnionitis and endometritis with long-term tocolysis. Accordingly, use of tocolytics in patients with preterm PROM remains controversial, but the bulk of the evidence shows no benefit. If tocolytics are used, such as during transfer to a tertiary care center or possibly to obtain benefits of corticosteroids, we strongly believe that the course of tocolytics should be short term, i.e., limited to less than 48 hours.
COMPLICATIONS The risks of PROM generally have been viewed as those of infection versus those of prematurity. Despite the limitations of using data from descriptive studies over a long period, it is clear that the most common complication among pregnancies with PROM before 37 weeks is RDS, which is found in 10% to 40% of neonates. Bona fide neonatal sepsis was documented in less than 10%, and amnionitis (based always on clinical criteria) occurred in approximately 3% to 31%. Endometritis developed in 0% to 29% in most groups, but it is not clear whether patients with amnionitis also were included in the endometritis category. Abruption after PROM has been reported in 4.0% to 6.3% of cases, which is higher than the usually quoted rate of 0.5% to 1.0% (16). Pulmonary hypoplasia is a serious fetal complication occurring in preterm PROM. Pulmonary hypoplasia is more common when there is very early preterm PROM, especially when it occurs in the presence of prolonged PROM with severe oligohydramnios. Vergani et al. (17) estimated a nearly 100% probability of lethal pulmonary hypoplasia when PROM occurred before 23 weeks and when there was severe oligohydramnios. With later gestational age at the onset of preterm PROM, the likelihood of pulmonary hypoplasia decreased. Notably, pulmonary hypoplasia was rare with preterm PROM more than 28 to 29 weeks, even with oligohydramnios. Kilbride and colleagues (18) provided a formula for predicting the probability of lethal pulmonary hypoplasia after mid-trimester PROM. When preterm PROM occurred at less than 25 weeks with severe oligohydramnios lasting more than14 days, the likelihood of lethal pulmonary hypoplasia was estimated to be 80%. At the other extreme, when preterm PROM occurred at more than 25 weeks and when there was either no severe oligohydramnios or severe oligohydramnios for less than 5 days, then the predicted probability of lethal pulmonary hypoplasia was only 2%. These data provide important information for counseling patients with mid-trimester PROM. In 1991, the group at Irvine provided unique data on the recurrence risk for PROM. In a 5-year period, the recurrence rate was 32% (39/121) in patients who had PROM in an
index pregnancy. Based on these data, the risk of recurrence is considerable, prompting patient education and close follow-up in subsequent pregnancies (19).
TREATMENT CONSIDERATIONS The overall approach to management of PROM takes into consideration neonatal survival at the gestational age when rupture occurs. As shown in Figure 17.4, management usually is divided into four different phases of pregnancy. During the second trimester, neonatal survival is nil, leading numerous investigators to adopt a policy of expectant management. Early in the third trimester, neonatal survival rises markedly, but there is still considerable morbidity associated with delivery at this gestational age. In the mid third trimester, neonatal survival is high, but there is still attendant morbidity; in the late third trimester (at or near term), neonatal mortality and morbidity are low. In the sections that follow, neonatal outcome is one of the driving features in determining clinical management.
FIGURE 17.4. Management of premature rupture of the membranes by gestational age.
Preterm Premature Rupture of the Membranes Infection Diagnosis of Infection After Premature Rupture of the Membranes Because of the great variation in risk of infection, attention has been directed at determining the infection risk of a given individual. Garite et al. (20) found that observing bacteria on a smear of AF (obtained by amniocentesis) was 78% specific and 81% sensitive for prediction of infection. However, AF was available in only half the patients, and amniocentesis may be accompanied by trauma, bleeding, initiation of labor, or introduction of infection. Romero and colleagues (21) compiled data indicating that, in the presence of preterm PROM, positive cultures of AF are obtained in 25% to 35% of
patients. Others have looked for less invasive predictors of infection (Table 17.3). The biophysical profile when performed every 48 hours has been reported to discriminate between cases at high and low risk for “infection” (22). Vintzeleos and colleagues (22) reported that of 52 cases with a good score (8 or above), “infection” developed in only 6%. In comparison, of 16 cases with a low score (i.e., 7), “infection” was diagnosed in 15 (94%; p < 0.01). Ohlsson and Wang (23) reviewed the tests purporting to predict clinical maternal infection, histologic chorioamnionitis, positive cultures of AF, or neonatal sepsis after preterm PROM. As summarized in Table 17.3, these tests have widely varying predictive values. Because the prevalence of infection in either mother or neonate is never high, a high negative predictive value for a test is neither impressive nor especially useful clinically. Because none of these tests is ideal, they should be performed selectively and interpreted in the context of clinical findings.
TABLE 17.3. PREDICTION OF INFECTION IN PREMATURE RUPTURE OF THE MEMBRANES
Effects of Steroids on Infection There still is not complete agreement on the risks and benefits of steroids in preterm PROM. In the first metaanalysis in 1989, Ohlsson (24) included five randomized trials and concluded that the use of steroids “increases the incidence of endometritis and may increase neonatal infections.” In a 1990 metaanalysis, Crowley and coauthors (25) included seven trials in the assessment of maternal infection and five trials for neonatal infection. The odds ratios for maternal infection of 1.26 (95% CI, 0.6–2.4) and for neonatal infection of 1.6 (95% CI, 0.9–3.0) suggest that corticosteroid use is more likely to increase than to decrease infection after PROM, but the results are not statistically significant. The 1994 National Institutes of Health (NIH) Consensus Conference concluded that the risk of maternal and infant infection may be increased with corticosteroid use after PROM but that the magnitude of this risk was small. The
recommendations are summarized in Box 1 (26).
Box 1 Corticosteroid Use in Preterm Premature Rupture of the Membranes: Recommendations from the 1994 NIH Consensus Conference Antenatal steroids in preterm PROM reduced the risk of RDS in randomized clinical trials, but the effect was less than with intact membranes. Strong evidence suggests reduced neonatal mortality and intraventricular hemorrhage with use in preterm PROM. “In fetuses 2,500 g), Maberry and colleagues (62) found no differences in mean umbilical artery pH (7.28 in both groups) or rate of acidemia (umbilical artery pH Spectrum of Activity Methicillin was the first antistaphylococcal penicillin introduced into clinical medicine. Although it is active against pneumococci and streptococci, its efficacy against these organisms is many-fold less than that of penicillin G. Methicillin is active against most strains of S. aureus, but increasingly, methicillin-resistant S. aureus (MRSA) has been reported. MRSA is resistant by virtue of altered penicillin-binding proteins. MSRA is also resistant to penicillin G, the cephalosporins, and other penicillinase-resistant penicillins such as nafcillin, oxacillin, cloxacillin, and dicloxacillin. MRSA has been associated predominantly with hospital-acquired infections. In addition, the incidence of methicillin-resistant Staphylococcus epidermidis is higher than that for MRSA. The isoxazolyl penicillins—oxacillin, cloxacillin, and dicloxacillin—have a spectrum of activity similar to that of methicillin. Nafcillin also has a similar antibacterial spectrum. Its
stability in the presence of staphylococcal penicillinase is comparable to that of methicillin but greater than that of the isoxazolyl penicillins. Dosage and Route Methicillin is administered only by the intravenous route. The dosage can be varied depending on the site and severity of the infection. For infections of moderate severity, 1 g every 4 hours is generally used, and for serious infections such as endocarditis, a dose of 2 g every 2 to 3 hours may be given. After an intravenous injection of 1 g of methicillin, a peak serum level of 60 µg/mL is obtained. There is a rapid fall in serum levels, to 7 µg/mL by 1 hour and less than 1 µg/mL by 2 to 3 hours. The peak level can be doubled by doubling the dose. Although oral oxacillin is available, it is absorbed erratically. Oxacillin may be given in a dosage of 250 to 500 mg every 4 to 6 hours, intravenously or intramuscularly, for mild to moderate infections. The dosage for severe infections is 1 g intravenously every 4 to 6 hours. Peak serum levels after a single intramuscular injection of oxacillin (500 mg) reach 14 to 16 µg/mL. Cloxacillin and dicloxacillin are analogs of oxacillin. They are acid stable and thus well absorbed after oral ingestion. After an oral dose of 500 mg of cloxacillin, a peak serum level of 8 µg/mL is reached in 30 to 60 minutes. Dicloxacillin produces serum levels that are twice as high as those of cloxacillin. Serum levels of flucloxacillin are similar to those seen with dicloxacillin. The usual oral adult dosage of these penicillins is 500 mg every 6 hours. For dicloxacillin, a dosage of 125 to 250 mg every 6 hours may suffice. Oxacillin, cloxacillin, and flucloxacillin can be given intramuscularly or intravenously, although dicloxacillin can be given intramuscularly. For mild infections, 250 mg every 6 hours is recommended. For moderate infections, the usual dosage is 1 g every 4 hours, and in severe infections, this is increased to 2 g every 4 hours. Nafcillin is poorly absorbed and is unreliable when used orally. It can be given either intramuscularly or intravenously. The usual adult dosage is 1 g every 4 hours but can be increased to 2 g every 4 hours for severe infections. After a 1-g intramuscular injection, a peak serum level of 8 µg/mL of nafcillin is reached in about 1 hour. By 6 hours, the level is 0.5 µg/mL. Side Effects Methicillin, isoxazolyl penicillins, and nafcillin are contraindicated in penicillin-allergic patients because they can produce any of the hypersensitivity reactions described for penicillin G. Methicillin has been associated with drug fever and leukopenia. In addition, interstitial nephritis may result from the use of large doses of intravenous methicillin. Most patients recover from the methicillin-associated interstitial nephritis after stopping the agent. The isoxazolyl penicillins may produce nausea and diarrhea with oral administration.
Oxacillin may cause fever, nausea, and vomiting in association with abnormal liver function test results. Neutropenia has been noted as well. With very large doses, neurotoxicity occurs, particularly in patients with impaired renal function. Nafcillin has been reported to cause nephropathy and neutropenia on rare occasion. Cost The penicillinase-resistant group of penicillins are more-expensive drugs than penicillin G. They should be used only in the treatment of staphylococcal strains resistant to penicillin G. Use in Pregnancy All the penicillinase-resistant penicillins are capable of crossing the placenta. No untoward effects on the fetus have been described. Metabolism Methicillin is excreted in urine, both by glomerular filtration and by tubular secretion. Up to 80% of an injected dose can be recovered from urine. A small percentage (2% to 3%) of the administered drug is excreted in bile. The unexcreted methicillin is inactivated in the body by the liver. Methicillin is widely distributed throughout the body. Isoxazolyl penicillins are mainly excreted in the urine, with dicloxacillin and flucloxacillin present in larger amounts than cloxacillin, of which 30% of an oral dose is excreted in the urine. Oxacillin is excreted in smaller amounts than cloxacillin. Excretion occurs by glomerular filtration and tubular secretion. These penicillins are also excreted in the bile to a small extent. Inactivation of the isoxazolyl penicillins probably occurs in the liver. After administration of nafcillin, 30% of the dose can be recovered from the urine. Between 5% and 10% of a dose is excreted in bile. The remainder of the nafcillin is inactivated in the liver. Nafcillin reaches high concentrations in the CSF of patients with normal meninges and in those with inflamed meninges. Indications in Obstetrics and Gynecology The penicillinase-resistant penicillins are indicated solely for the treatment of infections due to S. aureus. Examples include wound infection, mastitis, endocarditis, osteomyelitis, and toxic shock syndrome. Aminopenicillins The aminopenicillins are semisynthetic derivatives of 6-APS and include ampicillin, amoxicillin, and antibiotics structurally related to ampicillin-amoxicillin, such as epicillin,
cyclacillin, hetacillin, pivampicillin, talampicillin, bacampicillin, and metampicillin. Spectrum of Activity Ampicillin and amoxicillin have identical spectra. They are active against most of the bacteria that are sensitive to penicillin G. However, they are also active against some of the Gram-negative bacteria that are resistant to penicillin G. Like penicillin G, they are bactericidal antibiotics that inhibit cell wall synthesis. Against aerobic Gram-positive cocci, ampicillin is generally as effective as penicillin G. Thus, group A and group B b-hemolytic streptococci, most S. pneumoniae, and the b-hemolytic streptococci of the viridans group are susceptible. Nearly all S. aureus organisms are resistant to ampicillin. Unlike penicillin G, ampicillin is effective against enterococcal strains of the group D streptococci. The anaerobic Gram-positive cocci such as Peptostreptococcus species are almost always sensitive to ampicillin. Among Gram-positive aerobic bacilli, C. diphtheriae, B. anthracis, and L. monocytogenes are generally susceptible to ampicillin. Ampicillin is active against anaerobic Gram-positive spore-forming bacilli such as clostridia and anaerobic Gram-positive non–spore-forming bacilli such as Actinomyces, Eubacterium, Bifidobacterium, Propionibacterium, and Lactobacillus species. Unlike penicillin G, ampicillin is effective against some of the Enterobacteriaceae. Although E. coli is often sensitive, increasing resistance by this organism has made ampicillin an unreliable choice for therapy. Unless it is a b-lactamase–producing strain, Proteus mirabilis is susceptible to ampicillin. Enterobacter, Klebsiella, Serratia, Citrobacter, and indole-positive Proteus strains are generally resistant to ampicillin. Salmonellae are usually sensitive, but resistant strains have occurred. On the other hand, Shigella species tend to be resistant. Among the other Gram-negative aerobic bacteria, P. aeruginosa is always resistant to ampicillin. Haemophilus influenzae type b was generally sensitive to ampicillin, but plasmid-mediated ampicillin resistance has occurred. N. gonorrhoeae strains are now less susceptible to ampicillin. Penicillinase-producing N. gonorrhoeae (PPNG) strains are completely ampicillin resistant. N. meningitides organisms remain sensitive. Similar to the pattern with penicillin G, many of the Bacteroides and Fusobacterium species are ampicillin sensitive. The notable exceptions are B. fragilis group, P. bivia, and Prevotella disiens. Mycoplasmas and rickettsiae are ampicillin resistant, chlamydiae are relatively resistant, but Chlamydia trachomatis cervicitis may be treated with amoxicillin, which is an alternative treatment in pregnancy (2). Dosage and Route The usual oral dosage of ampicillin is 250 to 500 mg every 6 hours. After oral administration of 500 mg, peak serum levels of 3 µg/mL are obtained at approximately 2 hours, and the drug is still detectable at 6 hours. Doubling the dose results in a doubling of the serum level. Oral absorption of ampicillin is significantly lowered if it is taken with
meals. For amoxicillin, the usual dosage is 250 mg three times a day. An intramuscular preparation of ampicillin is available but rarely necessary. The usual dosage is 0.5 to 1.0 g every 6 hours. After a 500-mg injection, a peak serum level of 10 µg/mL is achieved in 1 hour, and levels persist for 4 hours. The usual intravenous dosage is 1 to 2 g every 4 to 6 hours by intermittent infusion, depending on the severity of the clinical infection. Amoxicillin is significantly better absorbed than ampicillin when given orally, and peak serum levels are approximately 2 to 2.5 times those achieved with a similar dose of ampicillin. The usual adult dosage of amoxicillin is 250 to 500 mg every 6 to 8 hours. Bacampicillin, when administered orally, is totally hydrolyzed to free ampicillin. It is absorbed better than ampicillin in the presence of food. Peak serum levels of 1.5 to 2 times greater are achieved with bacampicillin, compared with a similar dose of ampicillin. The usual adult dosage is 200, 400, or 800 mg every 8 to 12 hours. Side Effects Ampicillin and the other aminopenicillins may cross-react with other penicillins and should, therefore, not be used in patients with a history of penicillin allergy. Any of the hypersensitivity reactions described for penicillin G may occur with ampicillin. The incidence of rash in ampicillin users is much greater than that reported with use of penicillin G. Five percent to 7% of patients treated with ampicillin develop a diffuse macular rash. However, these rashes are specific for ampicillin and do not represent a true penicillin allergy. Gastrointestinal side effects such as nausea and diarrhea occur commonly but are generally not serious. Because of their enhanced gastrointestinal absorption, the aminopenicillins, except ampicillin, are associated with fewer gastrointestinal side effects. More rarely, pseudomembranous colitis has been reported with ampicillin use. Unusual side effects noted in ampicillin treatment include nephropathy, agranulocytosis, and encephalopathy. Ampicillin may impair the absorption of oral contraceptives and thus can be associated with breakthrough bleeding. Monilia vaginitis may develop secondary to ampicillin suppression of the normal vaginal microflora. Cost Ampicillin is a relatively inexpensive antibiotic, in both oral and parenteral forms. For oral amoxicillin, the cost is similar. The other aminopenicillins are more costly than ampicillin, and the additional cost may limit their use.
Use in Pregnancy Ampicillin has been used extensively in pregnancy. No adverse fetal effects have been associated with ampicillin, so the drug is considered safe for use in pregnancy. In pregnant woman, serum levels of ampicillin are approximately 50% of those obtained in nonpregnant women. This is the result of the significant increases in plasma volume and renal clearance of ampicillin that occur in pregnancy. Ampicillin rapidly crosses the placenta, and after a single maternal dose, peak cord blood levels are 40% of the peak maternal levels. In turn, amniotic fluid levels are lower than those in the umbilical cord. Experience with the other aminopenicillins in pregnancy is limited to date, and they are not generally recommended for use in pregnant women. Metabolism Like penicillin G, ampicillin is largely excreted in urine as the result of glomerular filtration and tubular secretion. After oral administration, 75% of a dose is excreted in the urine. The remainder of an ampicillin dose is either excreted in the bile or inactivated in the liver. Ampicillin is evenly distributed throughout body tissues, with levels in the kidneys and liver being significantly greater than serum levels. Although only very low levels can be detected in normal CSF, high levels of ampicillin are achieved in the CSF with inflamed meninges. Indications in Obstetrics and Gynecology Ampicillin is widely used in the therapy for asymptomatic bacteriuria, acute cystitis, and acute pyelonephritis in pregnant and nonpregnant women. Because of increasing resistance of E. coli, ampicillin should not be used by itself in the empiric treatment of urinary tract infection (UTI), particularly pyelonephritis. Although in the past ampicillin alone or in combination with an aminoglycoside was commonly used for the treatment of mixed aerobic-anaerobic soft tissue pelvic infections such as endomyometritis, pelvic cellulitis, or pelvic inflammatory disease (PID), the recognition of the need to provide therapy against Gram-negative anaerobes in such infections has resulted in a significant decrease in the use of ampicillin on obstetric and gynecologic services. Ampicillin and penicillin are the drugs of choice for group B streptococci and are included in the treatment of intraamniotic infection (chorioamnionitis) in combination with agents effective against anaerobes and Gram-negative aerobes. Ampicillin is also the optimal drug for the treatment of L. monocytogenes infections. Ampicillin is an alternative to penicillin G in intrapartum prophylaxis of perinatal group B streptococcal
infections (4). Finally, ampicillin has been used effectively as a prophylactic antibiotic with cesarean sections and vaginal hysterectomy. Because the other aminopenicillins have an antibacterial spectrum similar to that of ampicillin, the indications for their use are similar to those for ampicillin. Because of enhanced oral absorption, amoxicillin has replaced oral ampicillin in the treatment of UTIs. Antipseudomonal Penicillins The antipseudomonal group of penicillins include the carboxypenicillins—carbenicillin and ticarcillin—and the ureidopenicillin azlocillin. The major importance of these antimicrobial agents is their activity against Pseudomonas. However, they are all susceptible to b-lactamase enzymes produced by Gram-negative and Gram-positive organisms. Spectrum of Activity The spectrum of activity of carbenicillin is similar to that of ampicillin. However, it does possess activity against P. aeruginosa and certain indole-positive Proteus species. Carbenicillin is less active than ampicillin against the facultative streptococci, although its activity against N. gonorrhoeae, N. meningitides, and Haemophilus is similar to that of ampicillin. Klebsiella pneumoniae is resistant. At high concentrations, carbenicillin is also active against anaerobic bacteria including B. fragilis, although not to the extent of clindamycin, chloramphenicol, metronidazole, or cefoxitin. Carbenicillin acts synergistically with aminoglycosides to inhibit P. aeruginosa. This is clinically important because resistance to carbenicillin often develops among P. aeruginosa during therapy. The antibacterial spectrum of ticarcillin is similar to that of carbenicillin with the exception that it is two to four times more active against P. aeruginosa. However, strains of P. aeruginosa that are highly resistant to carbenicillin are also resistant to ticarcillin. Azlocillin is a ureidopenicillin whose main advantage is its significantly enhanced activity against P. aeruginosa, compared with that of carbenicillin or ticarcillin. All three of these agents are ineffective against b-lactamase–producing S. aureus. Dosage and Route Carbenicillin is not absorbed after oral administration and is available in intramuscular or intravenous forms. An oral form, carbenicillin indanyl sodium, is available but is useful only for UTIs because adequate serum concentrations are not achieved. For the treatment of systemic Pseudomonas infections or anaerobic infection, intravenous intermittent doses of carbenicillin (5 g every 3 to 4 hours) are required (total 30 to 40 g daily). Parenteral carbenicillin for Pseudomonas UTIs requires 1 to 2 g every 4 to 6 hours. The usual adult dosage for oral carbenicillin (provided as carbenicillin indanyl sodium) is 0.5 to 1.0 g every 6 hours. After an intramuscular dose of 1 g, peak carbenicillin serum levels of 20 µg/mL are achieved in 1 hour, and no drug is detectable by 4 hours. Although these serum levels
are inadequate for Pseudomonas or anaerobic infections in soft tissue, the urine levels reach 2,000 to 4,000 µg/mL, which are sufficient for the treatment of Pseudomonas UTI. With an intravenous infusion of 70 to 100 mg of carbenicillin per kilogram of body weight, serum levels of 150 to 200 µg/mL can be obtained; these levels are sufficient, in many instances, for the treatment of soft tissue Pseudomonas or anaerobic infections. With ticarcillin, an adult dosage of 18 to 24 g per 24 hours is recommended for systemic Pseudomonas infections or anaerobic infections; this is given as a 3-g dose every 3 to 4 hours. For treatment of UTIs, a dosage of 1 g every 6 hours is recommended. After a 1-g dose of ticarcillin intramuscularly, a mean peak serum level of 35 µg/mL is achieved in 1 hour, and by 6 hours, it is 6 µg/mL. After a 3-g intravenous infusion, mean serum levels postinfusion are 239 µg/mL, and at 4 hours, they are 94 µg/mL. Azlocillin is administered intravenously in a dosage of 12 to 16 g per day divided into four doses. Administration of a 2-g dose of azlocillin results in peak serum levels of 60 µg/mL at 1 hour. Side Effects Carbenicillin, ticarcillin, and azlocillin may provoke any of the hypersensitivity reactions that occur with penicillin G. High doses of carbenicillin may result in neurotoxicity. Each gram of carbenicillin contains 4.7 mEq of sodium. With the use of the required large dosage of 30 to 40 g per day, the potential for sodium overload may occur in patients with cardiac or renal disease. In addition, hypokalemia may occur as a result of this sodium overload. Finally, carbenicillin rarely is associated with bleeding as the result of diminished platelet adhesiveness. In high doses, ticarcillin also may result in neurotoxicity, electrolyte disturbances similar to those described with carbenicillin, and altered platelet function. The lower required dose of ticarcillin (compared with that of carbenicillin) is associated with a lower sodium load and less platelet dysfunction. Cost The antipseudomonal penicillins are costly antimicrobial agents because of the large doses required. Use in Pregnancy Carbenicillin, ticarcillin, and azlocillin have not been used extensively during pregnancy. Like other penicillins, they are most likely safe to use during pregnancy. However, they should be used in pregnant women only when no safe alternative is available. Metabolism Carbenicillin is excreted via the kidney, with about 95% of a parenteral dose excreted into the urine during the first 6 hours after administration. It is excreted via glomerular
filtration and tubular secretion. A small amount (less than 1%) of carbenicillin is excreted in the bile or inactivated in the liver. Similarly, ticarcillin and azlocillin are also primarily excreted via the kidneys. Indications in Obstetrics and Gynecology The clinical role for carbenicillin and ticarcillin in obstetrics and gynecology is now very limited. Although clinical studies have demonstrated relatively good efficacy with these agents (as single agents or in combination with aminoglycosides) in the treatment of mixed aerobic-anaerobic soft tissue infections of the pelvis (5,6), the introduction of the ureidopenicillins, mezlocillin and piperacillin, into clinical practice has significantly reduced their usefulness. Although Pseudomonas infections of soft tissue are rare on obstetric and gynecologic services, when these infections do occur, azlocillin is an appropriate therapeutic agent. Its major advantage over aminoglycosides is a decreased potential for nephrotoxicity or ototoxicity. For serious Pseudomonas infection, azlocillin should be combined with an aminoglycoside. Extended-Spectrum Penicillins The new ureidopenicillins, such as mezlocillin (Mezlin) and piperacillin (Pipracil), have replaced the carboxypenicillins (carbenicillin and ticarcillin). Compared with carbenicillin and ticarcillin, the ureidopenicillins have extended coverage against the Enterobacteriaceae, a diminished sodium load, and increased bioavailability in CSF, amniotic fluid, and cord blood. In addition, they provide increased activity against many aerobic and anaerobic Gram-positive cocci, including enterococci. Spectrum of Activity Mezlocillin is fairly similar to carbenicillin and ticarcillin in its antibacterial spectrum, with several important differences. It is more active against Proteus and Enterobacter strains. Mezlocillin is significantly more active than carbenicillin or ticarcillin against K. pneumoniae, with nearly 75% of Klebsiella species being susceptible. Mezlocillin is more active against enterococci and somewhat more active against B. fragilis; however, most S. aureus strains are resistant. Piperacillin demonstrates a wide spectrum of antibacterial activity (8). It is active against most clinically important Gram-negative facultative bacteria, particularly indole-positive Proteus species, Klebsiella, many Serratia marcescens, and P. aeruginosa; E. coli, P. mirabilis, Enterobacter, Citrobacter, Salmonella, and Shigella species are sensitive to low concentrations of piperacillin. Piperacillin demonstrates activity against Gram-positive aerobic bacteria, including the enterococcus. In addition, virtually all clinically important anaerobes, including B. fragilis and P. bivia, are sensitive to piperacillin. The major defect in the spectrum of these agents is lack of coverage for S. aureus, 10%
of E. coli, and some Klebsiella species (particularly with piperacillin, which covers only half of Klebsiella strains). Dosage and Route Mezlocillin (Mezlin) may be administered intramuscularly or intravenously (preferred). The recommended adult dosage is 200 to 300 mg/kg per day, usually given as 3 g intravenously every 4 hours (18 g per day). Up to 24 g per day may be given for severe infections. A 30-minute infusion of 3 g of mezlocillin results in peak serum concentrations of about 260 µg/mL, and by 1 hour and 4 hours, the serum levels are 57 µg/mL and 4.4 µg/mL, respectively. Piperacillin (Pipracil) is available for intramuscular or intravenous (preferred) administration. The usual adult dosage is 3 to 4 g every 4 to 6 hours as an intravenous infusion (total, 12 to 18 g per day). After a 30-minute intravenous infusion of 4 g of piperacillin, peak serum levels of 215 µg/mL are reached; by 1 hour, the level is 105 µg/mL, and it falls to 15 µg/mL by 4 hours. Side Effects As with other penicillins, mezlocillin and piperacillin have been associated with hypersensitivity reactions, nausea, diarrhea, CNS irritability with large doses, and local reactions such as phlebitis. A major advantage of mezlocillin and piperacillin over carbenicillin and ticarcillin is a significant decrease in sodium load. The ureidopenicillins contain 1.85 mEq/g of sodium, compared with 5.23 mEq/g for carbenicillin and 5.20 mEq/g for ticarcillin. This dramatically reduces the risk of congestive heart failure in patients with cardiovascular compromise. Cost Both mezlocillin and piperacillin are expensive antimicrobial agents because of the need for large doses, frequent intravenous administration, and a moderately expensive per-gram cost. Use in Pregnancy Because of altered protein binding, the ureidopenicillins can enter cord blood and amniotic fluid in therapeutic concentrations. Experience with these agents in pregnancy is very limited. Being penicillins, they are probably safe for use during pregnancy, but long-term follow-up studies are unavailable. Metabolism Like other penicillins, both mezlocillin and piperacillin are primarily excreted by the
kidneys. Approximately 55% of a mezlocillin dose appears in urine within 6 hours, and 26% is excreted in the bile. For piperacillin, nearly 80% of a dose is excreted in urine within 24 hours. Excretion also occurs via the biliary route for piperacillin. Both mezlocillin and piperacillin are widely distributed throughout body tissues. Although little of either drug penetrates the CSF in normal patients, mezlocillin and piperacillin both achieve satisfactory levels in the CSF with inflamed meninges. Indications in Obstetrics and Gynecology The in vitro studies reviewed previously suggested that mezlocillin or piperacillin might be effective single agents for the treatment of mixed aerobic-anaerobic soft tissue infection in the female pelvis. Clinical investigations have confirmed the efficacy of these agents in the treatment of such infections with clinical cure rates usually, but not always, similar to those achieved using standard regimens such as clindamycin plus gentamicin. These results suggest mezlocillin and piperacillin can be used as reasonable single-agent therapy for many polymicrobial infections in obstetrics and gynecology. Yet, they have not been evaluated as thoroughly as other regimens (e.g., clindamycin-gentamicin, and cefoxitin, or cefotetan). Both these agents have been used as prophylactic antibiotics in patients undergoing cesarean section or hysterectomy. The coverage of enterococci and P. aeruginosa is a theoretic advantage for mezlocillin or piperacillin over commonly used cephalosporins or ampicillin in prophylactic attempts. However, they are significantly more costly than older drugs, and no studies have demonstrated that these agents are better than ampicillin or first-generation cephalosporins for prophylaxis. Amidino Penicillins Mecillinam, although derived from the penicillin nucleus, 6-APS, is actually in a new class of penicillins. Pivmecillinam, an ester of mecillinam, is the preparation for oral administration. Spectrum of Activity Mecillinam differs from most penicillins in its spectrum of activity. Unlike other penicillins, mecillinam has poor activity against Gram-positive bacteria and is much more active against Gram-negative organisms. Thus, it is highly active against most Enterobacteriaceae including E. coli, Klebsiella, Enterobacter, P. mirabilis, Proteus vulgaris, and Citrobacter. Most S. marcescens are resistant, as are almost all P. aeruginosa and B. fragilis. Both N. gonorrhoeae and H. influenzae are significantly less susceptible to mecillinam than to ampicillin. Interestingly, mecillinam acts synergistically with other b-lactam antibiotics—penicillins and cephalosporins—against most Enterobacteriaceae. Dosage and Route Mecillinam is administered intramuscularly or intravenously. The dosage is 200 to 400
mg every 6 hours but can be increased to 600 mg. The peak serum level after a 200-mg intravenous dose is 6.5 µg/mL, which falls to 2 µg/mL by 1 hour. Pivmecillinam is usually administered in 400-mg doses every 6 hours, orally. Mean peak serum levels of 2.5 µg/mL are achieved 1.5 hours after administration of a 400-mg dose. Side Effects Mecillinam use is associated with few toxic effects. Gastrointestinal side effects are uncommon, as are the penicillin-type hypersensitivity reactions. Use in Pregnancy The use of mecillinam in pregnancy is not clear. Animal studies suggest that it achieves low levels in the fetus. Metabolism Mecillinam is excreted in the urine; 60% of parenterally administered and 40% of orally administered drug appear in the urine within 24 hours. The drug is also excreted in bile, where levels higher than those in serum are obtained. Indications in Obstetrics and Gynecology Little available information on the use of mecillinam in obstetrics and gynecology exists. Its major appeal would be as a safe b-lactam agent in the treatment of infections due to Enterobacteriaceae (e.g., pyelonephritis) as a single agent or in combination with drugs effective against Gram-positive aerobes and anaerobic bacteria in mixed infections. The final answer must await such clinical trials. Penicillin-b-Lactamase-Inhibitor Combinations A new approach to developing new antimicrobial agents that are resistant to degradation by b-lactamase enzymes involves developing compounds that inhibit these enzymes and combining them with older b-lactam antibiotics. This results in an enhanced spectrum of activity against a wide variety of aerobic and anaerobic bacteria. Three enzyme inhibitors, clavulanic acid, sulbactam, and tazobactam, have been developed, and their combination with penicillins has led to the introduction of clinically available agents: amoxicillin–clavulanic acid (Augmentin), ticarcillin–clavulanic acid (Timentin), sulbactam-ampicillin (Unasyn), and piperacillin-tazobactam (Zosyn). Amoxicillin–Clavulanic Acid (Augmentin) The addition of clavulanic acid, a suicide inhibitor of b-lactamases, to amoxicillin has resulted in an antimicrobial agent (Augmentin) that has a markedly increased spectrum
of activity against a wide variety of aerobic and anaerobic bacteria. Augmentin has demonstrated success in the treatment of skin infections due to b-lactamase–producing S. aureus, otitis media due to H. influenzae, UTIs due to b-lactamase–producing E. coli and Klebsiella, and PPNG. However, it is available only as an oral agent. Theoretically, Augmentin plus doxycycline is an excellent ambulatory regimen for PID. However, attempts to use Augmentin plus doxycycline for the treatment of PID have been limited by a high incidence of nausea and vomiting (9). Thus, Augmentin has a limited role in the treatment of pelvic infections. Ticarcillin–Clavulanic Acid (Timentin) Timentin is the combination of ticarcillin (3g) and clavulanic acid (100 mg). The addition of clavulanic acid, a b-lactamase inhibitor, to ticarcillin resulted in a combination compound that has markedly enhanced activity. Compared with ticarcillin alone, Timentin has increased activity against S. aureus, E. coli, Klebsiella, B. fragilis, other b-lactamase–producing Bacteroides, and N. gonorrhoeae, including PPNG (10). Timentin has demonstrated excellent in vitro activity against the B. fragilis group of organisms, which was equivalent to that of imipenem (11). However, enterococci and MRSA remain relatively resistant. The pharmacology of ticarcillin is not significantly altered by the addition of clavulanic acid and thus is similar to that discussed already. Both drugs have a serum half-life of about 1 hour after intravenous infusion. Sixty percent to 90% of ticarcillin and 35% to 45% of clavulanic acid are excreted unchanged in the urine. Renal insufficiency leads to accumulation, so dosage adjustment is necessary for patients with renal impairment. After intravenous infusion, both ticarcillin and clavulanic acid are widely distributed, with peak serum levels of 400 µg/mL for ticarcillin and 11 µg/mL for clavulanic acid. The dosage of Timentin (3 g of ticarcillin and 100 mg of clavulanic acid) is every 4 to 6 hours, depending on the severity of infection. This combination can inactivate aminoglycosides and should not be infused at the same time as an aminoglycoside. Side effects are rare and usually mild. Among the most common are phlebitis, rash, diarrhea, mild elevations in liver function test results, and rarely, hematologic abnormalities such as eosinophilia, leukopenia, thrombocytopenia, or anemia. The cost per day for Timentin (3.1 g every 6 hours) is high. In comparative studies (12,13), Timentin was found to be as effective as clindamycin-gentamicin. These results make Timentin a reasonable choice for many polymicrobial infections, but there has been more variability in cure rate than is seen with comparable regimens. Ampicillin-Sulbactam (Unasyn) The addition of sulbactam to ampicillin significantly extended the antibacterial spectrum of ampicillin to include b-lactamase–producing strains of H. influenzae, N. gonorrhoeae, many anaerobes (including B. fragilis, B. bivius, and B. disiens), E. coli, Klebsiella,
Enterobacter aerogenes, S. aureus, and S. epidermidis. However, ampicillin-sulbactam is not active against P. aeruginosa, Enterobacter cloacae, or Serratia species. Administration of ampicillin and sulbactam together has no effect on the pharmacokinetics of these agents. The serum half-life for both drugs is approximately 1 hour. After intravenous infusion, ampicillin and sulbactam are widely distributed, and after 500 mg infusion of each agent, serum levels rapidly peak at about 30 µg/mL for ampicillin and 43 µg/mL for sulbactam. Eighty-five percent of both drugs are excreted in urine within 24 hours. Small amounts of both drugs appear in breast milk (10). Ampicillin-sulbactam is available as 1 g of sulbactam to every 2 g of ampicillin. The usual dosage is 0.5 to 1 g of sulbactam plus 1 to 2 g of ampicillin every 6 hours. As with ampicillin itself, the kinetics of ampicillin-sulbactam is altered in pregnancy. There is more rapid elimination and a shorter serum half-life (14). The most common side effects have generally been nausea, vomiting, diarrhea, or allergic reactions. Minor elevations in liver function test results have been infrequently reported. In clinical studies, ampicillin-sulbactam has been demonstrated to be effective in the treatment of soft tissue pelvic infections including endometritis and PID. However, ampicillin-sulbactam is not recommended as a treatment of PID by itself (2,17). In a large study, equivalent response rates were also found by McGregor et al. (17). For endometritis, the good response rate was 89% (141 of 159) for ampicillin-sulbactam versus 91% (139 of 153; p values were not significant [NS]) for clindamycin-gentamicin; for PID, the good response rate was 86% (47 of 55) in the ampicillin-sulbactam group versus 90% (43 of 48; p = NS) in the cefoxitin-doxycycline group (17). The cost per day for 3 g every 6 hours is moderately high but may be less than that of other new antibiotics. Based on the efficacy in clinical trials and the moderate cost, ampicillin-sulbactam is also a reasonable choice for treatment of polymicrobial infections. Among the penicillin b-lactamase combinations, it is the most widely studied. Piperacillin-Tazobactam (Zosyn) Another combination is piperacillin-tazobactam (Zosyn). Tazobactam is also an irreversible b-lactamase inhibitor. In vivo, this combination also shows enhanced activity compared with the penicillin alone. In a large multicenter comparative trial, Sweet et al. (18) compared Zosyn (3 g of piperacillin and 375 mg of tazobactam) with clindamycin-gentamicin in various pelvic infections (50% endometritis, 39% PID, 11% other gynecologic infections) and reported the following favorable response rates: 85% (166 of 196) for Zosyn and 87% (90 of 103) for clindamycin-gentamicin; p = NS). In this trial, diarrhea occurred more commonly in the piperacillin group (9.7% vs. 2.9%; p = 0.04), although most cases were mild to moderate and generally did not require therapy. In noncomparative trials, this antibiotic also led to high response rates (19).
CEPHALOSPORINS Many of the new antibiotics introduced over the past 15 to 20 years have been cephalosporins. Not surprisingly, these agents are very commonly prescribed antibiotics. Cephalosporins are classified according to their chemical structure, differences in pharmacology, b-lactamase resistance, and spectrum of activity. The most commonly used classification system is by generations according to antimicrobial activity (Table 23.5). Since the introduction of the first cephalosporin in 1964, there has been a virtual explosion of the cephalosporins and 21 are currently available in the United States (1). Three generations of cephalosporins have been marketed. In addition, cefepime has been classified as a fourth-generation cephalosporin by some authorities because of its unique extended spectrum of activity (1).
TABLE 23.5. CLASSIFICATION OF CEPHALOSPORINS
First-Generation Cephalosporins The cephalosporin antibiotics make up an extensive group, all of which are comprised of a 7-aminocephalosporanic acid with a d-aminodipic acid side chain. Changes in the side chain of this molecule have led to the development of cephalothin (Keflin), cefazolin (Kefzol, Ancef), cephapirin (Cefadyl), cephalexin (Keflex), and other so-called first-generation cephalosporins. Because of similar properties, these antibiotics are discussed as a group (2,3 and 4). Currently, only cefazolin, cephalexin, and cefadroxil are marketed in the United States. Spectrum of Activity The first-generation cephalosporins are primarily active against many aerobic Gram-positive cocci (including S. aureus, but excluding the enterococcus). They possess moderate activity against N. gonorrhoeae (exclusive of PPNG strains) and
some aerobic Gram-negative rods, including community-acquired E. coli, P. mirabilis (indole negative), K. pneumoniae, and Moraxella catarrhalis. Resistant aerobic Gram-negative organisms include Enterobacter, Pseudomonas, and Serratia species. First-generation cephalosporins have poor activity against H. influenzae, methicillin-resistant staphylococci (despite in vitro sensitivity), and penicillin-resistant pneumococci. Although some anaerobes are susceptible to cephalosporins, other antibiotics are preferable because of their broader activity, particularly against the Bacteroides and Prevotella group of bacteria, which produce b-lactamase enzymes. C. trachomatisand the genital mycoplasmas are resistant. The mechanism of action of cephalosporin antibiotics is quantitatively and qualitatively similar to that of penicillin. Cephalosporins interfere with cell wall synthesis. In particular, cephalosporins interfere with the biosynthesis of peptidoglycan in the cell wall by binding to and inactivating penicillin-binding proteins (1). As reviewed by Asbel and Levison (1), there are three mechanisms of microbial resistance to the cephalosporins: (a) alteration and decreased affinity of penicillin-binding proteins; (b) production of b-lactamases that inactivate the agent; and (c) changes in the bacterial outer cell wall (loss of porin channels) that limit the ability of the drug to reach the penicillin-binding proteins. Dosage and Route Two first-generation cephalosporin antibiotics, cephalexin (Keflex) and cefadroxil (Duricef) are absorbed orally. Cephalexin reaches maximum serum concentrations of 15 to 18 µg/mL within 1 hour after a 500-mg oral dose and has a half-life of 0.9 hours. Cefadroxil has a half-life of 1.2 hours and reaches a peak serum concentration of 16 µg/mL. For adults, the usual dosage is 250 mg every 6 hours for cephalexin and 1 g twice daily for cefadroxil. Oral cephalosporins are very well absorbed from the gastrointestinal tract. When parenteral cephalosporins are administered by either the intramuscular or the intravenous route, they are readily absorbed and widely distributed. Cephalosporins are detectable in ascitic, synovial, and pericardial fluids. Cefazolin (Ancef, Kefzol) may be administered by the intramuscular or intravenous route. After a 1-g intravenous infusion, cefazolin reaches a peak serum concentration of 80 µg/mL. The half-life of cefazolin is 1.8 hours, so adult dosing ranges from 0.5 to 1.5 g every 6 to 8 hours (the usual adult dosage is 1 g every 8 hours). Because excretion of most cephalosporins is delayed in patients with diminished renal function, the dose should be decreased modestly for these persons. The first-generation cephalosporins have similar antibacterial spectra. Although slight differences may be observed in the activity of these antibiotics against particular strains, these differences appear not to be clinically significant. Although cefazolin achieves serum concentrations that are significantly higher than those of the other cephalosporins, all these compounds
achieve concentrations adequate to inhibit most cephalosporin-susceptible bacteria. Side Effects The cephalosporins are generally very well tolerated. Fever, eosinophilia, serum sickness, and rashes occur rarely, and transient neutropenia has been observed. Although Coombs positivity is commonly observed in patients receiving cephalosporins, particularly cephalothin, hemolytic anemia is extremely uncommon. Pain on intramuscular injection is uncommon with cefazolin. Phlebitis has been observed with intravenous administration of all the cephalosporins. Nephrotoxicity has been very rarely observed after administration of current cephalosporins. Cross-reactivity may exist between penicillins and cephalosporins (5). Although precise incidence data are not available, persons who have had an immediate hypersensitivity reaction to penicillin, manifested by urticaria, wheezing, or anaphylaxis, are at greater risk for the development of similar reactions after cephalosporin administration. Cephalosporins are contraindicated in such individuals. Cost Costs of the first-generation cephalosporins have come down in the last few years. For oral preparations, the cost per day for cephalexin (500 mg orally four times a day) is approximately $1, but other preparations may be more expensive. For injectable preparations, the cost per day for cefazolin (1 g every 8 hours) is approximately $4. Use in Pregnancy Cephalosporins cross the placenta rapidly after intravenous injection to the mother and achieve adequate levels in cord blood and amniotic fluids. To date, no adverse fetal effects have been reported for this group of antibiotics. Metabolism Cefazolin is excreted by the renal route. The primary mode of excretion is tubular excretion, and excretion is delayed in the presence of diminished renal function. Approximately 80% of cefazolin is bound to protein. Indications in Obstetrics and Gynecology The cephalosporins have gained an important role in the practice of medicine; they are among the most commonly used group of antibiotics in hospitals in this country. In other specialties, they have been frequently used as prophylactic agents in cardiac, orthopedic, vascular, and gastrointestinal surgery. They are also active and effective agents in the treatment of staphylococcal and a limited number of Gram-negative
infections. Yet, there are no situations in which they are clearly the drugs of choice for therapy. In most situations, other antibiotics are clearly preferable. Thus, for a penicillin-sensitive organism, penicillin should be used, and for treating infections caused by S. aureus, a penicillinase-resistant penicillin should be used. In a number of situations, however, the first-generation cephalosporin antibiotics are valuable. The first of these is in the patient with delayed-type penicillin hypersensitivity. A cephalosporin may be used safely in most of these patients in place of penicillin, ampicillin, or methicillin, oxacillin, cloxacillin, and nafcillin. The second situation for using a cephalosporin antibiotic may be in treating some UTIs. Depending on the susceptibility pattern of the causative microorganism, a first-generation cephalosporin may be a good choice for the empiric treatment of pyelonephritis in pregnancy. With rising resistance of bacteria to ampicillin, cephalosporins may offer a high degree of both efficacy and safety. In patients with severe pyelonephritis and shock or with multiple previous infections, a broader spectrum of therapy is indicated. Finally, a cephalosporin may be used in the initial treatment (with or without an aminoglycoside) of postoperative (or other hospital-acquired) pneumonia, because it may be caused by either Gram-positive cocci or Gram-negative aerobes, particularly Klebsiella species. First-generation cephalosporins have been studied widely for prophylaxis and are now considered the drugs of choice by many experts for prophylaxis in obstetric and gynecologic procedures (Chapter 24, Antibiotic Prophylaxis in Obstetrics and Gynecology). Second-Generation Cephalosporins As noted by Karchmer (6), the second-generation cephalosporins actually contain two distinct groups: the true cephalosporins and the cephamycins. The former group includes cefamandole (Mandol), cefaclor (Ceclor), cefonicid (Monocid), cefprozil (Cefzil), cefuroxime (Zinacef), cefuroxime axetil (Ceftin), and loracarbef (Lorabid), whereas cefoxitin (Mefoxin) and cefotetan (Cefotan) are in the latter group (Table 23.5). Cefoxitin (Mefoxin) and cefamandole (Mandol) became available in late 1978. Both have an extended spectrum of activity compared with previous cephalosporins. As a result of this expanded spectrum of activity, particularly against anaerobes by cefoxitin, effective single-agent antimicrobial therapy for polymicrobic intraabdominal and pelvic infections became available and revolutionized the approach to treatment of these mixed aerobic-anaerobic infections. Most of these second-generation cephalosporins are not commonly used in pelvic infections. More recently, additional cephalosporins that belong to this group have been introduced. Thus, among the second-generation drugs, only cefoxitin and cefotetan are discussed in detail. Spectrum of Activity (7,8) Like the penicillins and other cephalosporins, the second-generation cephalosporins are bactericidal, inhibiting cell wall synthesis. The true cephalosporins among the second-generation agents have significantly improved activity against H. influenzae, M.
catarrhalis, N. meningitidis, and N. gonorrhoeae compared with first-generation drugs (6). They provide comparable to improved activity against staphylococci and nonenterococcal streptococci. In addition, they have enhanced activity against some Enterobacteriaceae (6). The cephamycins provide significantly enhanced activity against anaerobic bacteria, particularly the B. fragilis group, Bacteroides species, Prevotella species, and Porphymonas species, compared with first-generation cephalosporins and the true cephalosporins among the second-generation cephalosporins. Cefotetan, in particular, has enhanced activity against many Enterobacteriaceae: Cefoxitin and cefotetan have good activity against Neisseria species but decreased activity against staphylococci, compared with first-generation cephalosporins or the true cephalosporins among the second-generation cephalosporins (6). In addition, Pseudomonas species are highly resistant to all of these new agents. N. gonorrhoeae, including penicillinase-producing strains, is susceptible to cefoxitin and cefotetan. Gram-positive anaerobic organisms are susceptible to these agents and to penicillin and older cephalosporin antibiotics. Against Gram-negative anaerobes, cefoxitin and cefotetan clearly have better activity than older cephalosporins. Cuchural et al. (8) reported that surveillance of roughly 1,200 isolates of the B. fragilis group showed only 5% resistance to cefoxitin during 1984 and 1985. More recently, Wexler et al. (10) have shown that cefoxitin has maintained good activity against B. fragilis species (90% susceptible at 32 µg/mL, 99% susceptible at 64 µg/mL) and excellent activity against most other anaerobes (90% susceptible at 16 µg/mL or less). Cefamandole is not as active against the B. fragilis group as cefoxitin. In vitro susceptibility testing has demonstrated that the MIC50 (minimum inhibitory concentration) for P. bivia and P. disiens is twofold to fourfold higher for cefotetan than cefoxitin. However, clinical studies have demonstrated equal efficacy in the treatment of obstetric and gynecologic infections (see the “Indications in Obstetrics and Gynecology” section). Susceptible organisms have MIC values of less than 16 µg/mL; intermediate organisms have MIC values of 32 µg/mL. The antimicrobial spectra of cefonicid, ceforanide, and cefuroxime are very similar to that of cefamandole (11). These agents are not active against the B. fragilis group of organisms or other b-lactamase–producing Bacteroides such as B. bivius, B. disiens, or the pigmented Bacteroides. Thus, their use in obstetrics and gynecology is limited. Cefaclor is an oral agent that is more active against Gram-negative bacilli than the oral first-generation cephalosporin cephalexin. As with older cephalosporins, C. trachomatis and the genital mycoplasmas are resistant to the second-generation cephalosporins. Dosage and Route Cefoxitin, cefotetan, and cefamandole are administered parenterally. For initial treatment of most infections with cefoxitin, the dosage is 1 to 2 g every 6 hours intravenously. With cefotetan, the dosage is 1 to 2 g every 8 to 12 hours. For cefamandole, the comparable dosage is 2 to 3 g every 6 hours. At the lower range of these dosages, these agents can be given intramuscularly. Because these agents are excreted mainly in the urine, the dose should be reduced somewhat in patients with renal impairment. For example, the dosage for a patient with
moderate renal failure (those with a creatinine clearance of 10 to 50 mL per minute) should be reduced to 1 g every 8 to 12 hours, and for a patient with severe renal failure (creatinine clearance of less than 10 mL per minute, 1 g every 24 hours. After a 1-g intravenous injection of either drug, peak levels average approximately 100 µg/mL in 0.5 hours, and concentrations fall to 1 µg/mL in about 3 to 4 hours. Cefotetan has a longer half-life. Cefuroxime has a longer serum half-life than cefamandole (1.5 vs. 0.5 hours) and can be dosed every 8 hours. The serum half-life of ceforanide is 3 hours, and it is administered every 12 hours. Cefonicid has a long serum half-life of 4.5 hours, and one dose a day has been effective for infections due to susceptible organisms (12). Side Effects These drugs, like other cephalosporins, are generally well tolerated. (Please see previous discussion for comments on hypersensitivity, renal, liver, and hematologic effects of the cephalosporins.) One of the problems with some antibiotics in this class is local irritation on intramuscular or intravenous injection. In clinical trials, pain on intramuscular injection and thrombophlebitis on intravenous injection were reported infrequently (approximately 5% to 6% of cases) with cefamandole. Pain on intramuscular injection is more common with cefoxitin, although it can be decreased by diluting the drug with 0.5% lidocaine solution. Cefamandole and cefotetan contain a methyltetrazolethiol (MTT) side chain that inhibits the conversion of inactive to active prothrombin. Thus, these drugs have been associated, on occasion, with hypoprothrombinemia and bleeding (13). Cefoxitin has been associated infrequently with pseudomembranous colitis. Cost These agents are relatively expensive and are considerably more costly per gram than the older first-generation cephalosporins. Because of competition between cefoxitin and cefotetan, costs of these drugs have been lowered. In most settings, the cost per day is lower for cefotetan because of its twice-daily dosing schedule, versus four times daily for cefoxitin. Use in Pregnancy and Placental Transfer Giamarellou et al. (14) studied the pharmacokinetics of cefoxitin in patients undergoing mid-trimester abortion. They noted that 1 hour after a 2-g intravenous infusion, mean maternal serum levels averaged 30 µg/mL, and by 6 hours after the infusion, the levels had fallen to 0.6 µg/mL. These levels were similar to those noted in nonpregnant women. Amniotic fluid levels of cefoxitin were 2 to 6 µg/mL between 1 and 5 hours after infusion. Cefoxitin, cefotetan, and cefamandole have been widely used in pregnancy and the puerperium and are considered safe for use in pregnancy.
Metabolism As with other cephalosporins, these antibiotics are excreted mainly by the kidney. Approximately 75% of an intravenous dose can be recovered in the urine within 8 hours. Probenecid inhibits tubular secretion and results in higher blood levels. Indications in Obstetrics and Gynecology The extended activity of cefoxitin to include 95% of B. fragilis group isolates has made it an attractive single-agent therapy for obstetrics and gynecologic infections. In the late 1980s, cefoxitin emerged as the most commonly used single-agent antimicrobial used for the treatment of pelvic infections. In 1979, Sweet and Ledger (15) evaluated use of 2 g every 8 hours in 109 patients with genital infections. The overall cure rate was 92% (100 of 109). Included were seven patients with pelvic abscess, only three of whom were cured. For salpingitis, endomyometritis, and pelvic abscess, cure rates were 97%, 92%, and 89%, respectively. In a similar report, Ledger and Smith (16) noted a 94% cure rate in 178 patients treated with cefoxitin for various genital tract infections. However, cefoxitin has not led to high cure rates in other studies. Duff and Keiser (17) compared cefoxitin (2 g intravenously every 8 hours) with penicillin-gentamicin. For cefoxitin, the cure rate was only 61% (19 of 31), and for penicillin-gentamicin, it was 63% (27 of 43). In this study, the authors used a lower dose of cefoxitin, encountered a large number of abscesses, and decided on success or failure of therapy at 48 hours. An intermediate cure rate was reported by Hager and McDaniel (18), who treated 25 cases of pelvic infection with cefoxitin (2 g intravenously every 6 hours). The response rate was 84% (21 of 25). Of the four failures, three required surgical drainage and the fourth showed a response to clindamycin-gentamicin. Larsen et al. (19) reported that cefoxitin was as effective as the clindamycin-gentamicin combination in the treatment of pelvic infections; 37 (90%) of 41 patients treated with cefoxitin and 42 (84%) of 50 patients treated with clindamycin-gentamicin were considered clinical cures. In several studies, cefotetan has been demonstrated to have clinical efficacy equivalent to that of cefoxitin for the treatment of obstetric and gynecologic infections (20,21,22 and 23). In an open study of cefotetan (2 to 4 g per day) for the treatment of various pelvic infections (predominantly postpartum endomyometritis and salpingitis), Poindexter et al. (20) reported a clinical cure rate of 99% in 118 patients. Subsequently, Poindexter et al. (21) reported that 93% of 133 patients with obstetric and gynecologic infections responded to cefotetan (2 to 4 g per day). In randomized, comparative clinical trials versus cefoxitin, Hemsell et al. (22) and Sweet et al. (23) reported cure rates with cefotetan (2 to 4 g per day) of 90% and 94%, respectively. The study by Sweet et al. (23) compared cefotetan-doxycycline with cefoxitin-doxycycline for the treatment of acute PID, demonstrating equivalent cure rates (94% and 92%, respectively). Cefotetan has a serum half-life of 3.5 hours and is administered in a dosage of 2 g every 12 hours. It is excreted primarily by a glomerular filtration and thus dosing must be
adjusted in patients with renal failure. Most hospital formularies consider these two drugs equivalent. Thus, the choice is usually based on cost per day on an institution-by-institution basis. Because of activity against N. gonorrhoeae, including penicillinase-producing species, and anaerobic bacteria, including the Bacteroides groups, cefoxitin and cefotetan have been recommended by the Centers for Disease Control and Prevention (CDC) as part of combination therapy for PID and cefoxitin is an alternative choice for uncomplicated N. gonorrhoeae (24). Although cefoxitin has been shown to be effective for prophylaxis, it has no advantage over older less-expensive agents such as cefazolin (25). Because cefoxitin may be used for therapy and is more expensive than older agents, its use for prophylaxis has been questioned. (The use of prophylactic antibiotics is discussed more fully in Chapter 24, Antibiotic Prophylaxis in Obstetrics and Gynecology.) Because of its activity against S. pneumoniae, H. influenzae, and M. catarrhalis, cefuroxime is commonly used in the treatment of community-acquired bacterial pneumonia (6). Oral second-generation cephalosporins (cefuroxime axetil, cefaclor, cefprozil, and loracarbef) are used to treat various mild to moderate community-acquired infections, including skin and soft tissue infection and UTI (6). However, more cost-effective or narrow-spectrum treatment is available for these indications. Thus, the primary use of oral second-generation cephalosporins is for the treatment of respiratory tract infections. In summary, among the second-generation cephalosporins, cefoxitin and cefotetan have the most appropriate antimicrobial spectra for treating pelvic infections. They are well tolerated and achieve high concentrations in serum. In most clinical trials, efficacy has been very good. Cefoxitin or cefotetan is a reasonable initial choice for pelvic infections such as endomyometritis, salpingitis, and cuff cellulitis after hysterectomy. Third-Generation Cephalosporins This group of antibiotics is made up of a number of parenteral agents with an even more extended spectrum of activity (26) (Table 23.6). Modification of the side chains provides these new antibiotics with changes in spectrum and alterations in metabolism. In general, the third-generation cephalosporins are less active than the first-generation agents against Gram-positive cocci but are significantly more active against the Enterobacteriaceae. The third-generation cephalosporins are not as active against the B. fragilis group of bacteria, as are cefoxitin and cefotetan. On the other hand, they are more active against a broader spectrum of Gram-negative aerobic and facultative bacteria. In addition, as noted in Table 23.5 and Table 23.6, several of these agents have antipseudomonal activity.
TABLE 23.6. ANTIBIOTIC SENSITIVITIES OF COMMON PATHOGENS RECOVERED FROM OBSTETRIC AND GYNECOLOGIC INFECTIONS
Spectrum of Activity Although all of these new compounds possess very broad activities, they should not be considered to be interchangeable. Particularly in regard to the array of organisms involved in pelvic infections, there are important in vitro differences in activity. For aerobic streptococci, these agents have activity that is inferior to that of penicillin G or ampicillin and the first-generation cephalosporins. Within the group of third-generation cephalosporins, cefotaxime, ceftizoxime, and ceftriaxone have greater intrinsic activity than cefoperazone and ceftazidime. Yet, with usual doses, all of these agents provide sufficient coverage for all aerobic streptococci except enterococci. For S. aureus, an antistaphylococcal penicillin (such as nafcillin) or a first-generation cephalosporin (cephalothin or cefazolin) would be preferable, because the newer agents have 10- to 80-fold less activity. The MIC 90 values for these new agents range from 2 to 16 µg/mL, compared with 0.2 µg/mL for cefazolin, but satisfactory coverage is still supplied by usual doses of the new antibiotics. However, with the exception of ceftazidime, these agents are effective against methicillin-sensitive S. aureus. Methicillin-resistant staphylococci are not susceptible to the new agents. Activity against N. gonorrhoeae is excellent, with MIC90 values of less than 0.1 µg/mL for cefotaxime, ceftazidime, cefoperazone, and moxalactam. All of these inhibit penicillinase-producing species. Although not quite as active, the MIC90 for cefotetan against N. gonorrhoeae is 0.5 µg/mL. Among the third-generation cephalosporins, ceftriaxone has the best activity against N. gonorrhoeae, including PPNG strains (24,27). Activity of these agents against aerobic Gram-negative rods is impressive. MIC90 values of most of these agents for E. coli, K. pneumoniae, and P. mirabilis are less than 0.5
µg/mL, but for cefoperazone, the MIC90 is 16 µg/mL for the first two species. Enterobacter species have higher MIC90 values, but these are generally within the susceptible range. The exception is that for E. cloacae, the MIC90 of cefoperazone is 32 µg/mL. Many aminoglycoside-resistant strains are susceptible to low concentrations of the third-generation cephalosporins. P. aeruginosa occurs infrequently in pelvic infections, and activity of these new agents varies widely. Ceftazidime is the most active (MIC90 = 8 µg/mL), with cefoperazone and moxalactam having moderate activity (MIC90 = 32 µg/mL). For cefotaxime, the MIC 90 is 64 µg/mL. Other Pseudomonas species are likely to be resistant. Gram-positive anaerobic cocci are inhibited by the third-generation cephalosporins. MIC90 values for Peptococcus species and Peptostreptococcus species were less than 1 µg/mL in the early 1980s (28,29). In 1991, susceptibility of Peptostreptococcus species remained good to representative third-generation cephalosporin drugs. For cefonicid and ceftizoxime, 90% of isolates were susceptible to 16 and 8 µg/mL, respectively (8). C. perfringens usually has low MIC values, whereas Clostridium difficile is resistant. Among the anaerobic Gram-negative species, P. bivia and P. disiens are inhibited by 4 to 8 µg/mL of cefoperazone and cefotaxime, but the B. fragilis group has a more variable response. Cuchural et al. (9) found that only 59% of strains were inhibited by 16 µg of cefotaxime and only 46% by 16 µg of cefoperazone. For moxalactam, 85% of strains were inhibited by 16 µg/mL. With ceftizoxime, 63% of B. fragilis group strains were sensitive to 16 µg. For cefotetan, only 74% of strains were sensitive at 16 µg and 86% at 32 µg. For comparison, metronidazole and chloramphenicol inhibited 100% of B. fragilis group strains at 8 µg and 8 µg/mL, respectively. There was 5% resistance to clindamycin at 4 or 8 µg in this study (9). Cefoxitin inhibited 95% of these strains at 16 µg/mL (9). In general, ceftazidime and ceftriaxone have poor activity against the B. fragilis group. This limits their usefulness as single-agent therapy for polymicrobic pelvic infections. Considerable controversy exists over the activity of ceftizoxime against the B. fragilis group. The differences were related to variability in the susceptibility test conditions. As noted already, Cuchural et al. (9) reported that one third of B. fragilis group bacteria were resistant to ceftizoxime. More recently, Wexler et al. (10) reported modest activity of cefonicid and ceftizoxime against B. fragilis species (only 8% and 37% susceptible, respectively, at 32 µg/mL). For other Bacteroides species, activity is greater for cefonicid and ceftizoxime (57% and 93% susceptible, respectively, at 32 µg/mL) (10). In this report, metronidazole and chloramphenicol maintain excellent activity (MIC90 = 4 and 8 µg/mL, respectively, for B. fragilis, and 16 and 8 µg/mL, respectively, for other Bacteroides species). None of these agents are active against either C. trachomatis or the genital mycoplasmas.
Dosage, Route, and Metabolism Until recently third-generation cephalosporins were available only for parenteral administration. More recently, oral forms have become available. They possess different routes of metabolism and different half-lives. Cefotaxime has a relatively short half-life of 1 hour and should be administered every 6 to 8 hours. After a 2-g infusion over 30 minutes, peak levels of 80 to 90 µg/mL are attained. By 6 hours, the level is approximately 5 µg/mL. It is excreted mainly in the kidney, by tubular excretion and glomerular filtration. In renal insufficiency, cefotaxime does not accumulate, but the desacetyl derivative, which is not active biologically, has a longer half-life. Cefoperazone has a longer half-life of 1.6 to 2.4 hours. After a 2-g intravenous dose, serum levels equal 250 µg/mL. At 12 hours, levels are 1 to 2 µg/mL. Accordingly, cefoperazone has been used with dosing every 8 or 12 hours. In contrast to most cephalosporins, this agent is cleared mainly by biliary excretion and does not accumulate in patients with renal failure. The serum half-life of ceftazidime is nearly 2 hours. It is 10% protein bound, and 80% to 90% of the drug is excreted unchanged by the kidneys in the first 24 hours. The usual dosage is 1 to 2 g every 8 to 12 hours, and the dosage must be adjusted for patients with renal failure. For ceftizoxime, the serum half-life is approximately 1.7 hours, and the drug is excreted unchanged in the urine. Ceftriaxone is dosed at 2 g every 8 hours. Ceftriaxone is a long-acting drug with a serum half-life of 6 to 9 hours, which is the longest of any cephalosporins. The usual dosage for moderate to severe infections is 1 to 2 g intravenously every 12 to 24 hours. Newly introduced oral third-generation cephalosporins include cefixime (Suprax), cefpodoxime proxetil (Vantin), and ceftibuten (Cedax). Investigational oral third-generation agents include cefdinir (Omnicef). With a single oral dose of 400 mg, the peak serum concentration of cefixime is 3.9 µg/mL. The half-life is 3.7 hours. Thus, cefixime is provided as a single daily dose of 400 mg. Cefpodoxime proxetil achieves a maximum serum concentration of 4 µg/mL after a 400-mg oral dose. However, its half-life is 2.2 hours and cefpodoxime proxetil is dosed at 400 mg every 12 hours. With ceftibuten, peak serum levels of 15 to 17 µg/mL are achieved after a dose of 400 mg orally. Its half-life is 2.5 hours. The dose for adult infections is 400 mg as a single daily dose. Cost All of the third-generation cephalosporins have a high direct cost per gram, with a cost per day of about $25 to $35 or more. However, because of less frequent doses and more safety, they probably have lower indirect costs (such as pharmacy and nursing time and tests to monitor safety). Use in Pregnancy and Placental Transfer
Data on the use of antimicrobial agents during pregnancy, particularly their pharmacokinetics and placental transfer, are limited. Thus, it is not surprising that only a few studies have assessed the use of third-generation cephalosporins during pregnancy. Evaluating placental transfer of cefotaxime, Kafetzis et al. (30) reported serum levels in women undergoing mid-trimester abortion of 8.2 µg/mL at 1 hour after 1-g intravenous infusion and 0.9 µg/mL at 3 hours after infusion. The corresponding mean cord level at 1 hour was 1.9 µg/mL (23% of maternal levels). The serum levels were lower than expected for a nonpregnant woman. Cefotaxime was also found in breast milk in low concentrations (mean, 0.35 µg/mL) (30). In patients in their mid trimester, Giamarellou et al. (14) studied the kinetics of moxalactam and ceftazidime (1-g infusion). For ceftazidime, peak values occurred 2 hours after the infusion and were 13.2 µg/mL. At 6 hours after infusion, the serum levels averaged 1.9 µg/mL. In amniotic fluid, ceftazidime levels ranged from 1 to 4 µg/mL from 2 to 6 hours after infusion. Thus, for ceftazidime, serum levels in pregnant women are nearly 50% of those in nonpregnant women. Cefoperazone differs from the other cephalosporins in that it has a dual excretory pattern, primarily via the biliary tract and secondarily via the kidney. Although significant alterations occur in renal function during pregnancy, only minimal changes occur in biliary function. Gonik et al. (31) reported that the serum half-life (152 minutes) and the total clearance (1.1 ± 0.4 mL/kg per minute) for cefoperazone in pregnant women were similar to those seen in nonpregnant women. In addition, the trough level of cefoperazone was similar to that seen in nonpregnant women. This is in contrast to the findings for other cephalosporins. Approximately 45% of an intravenous dose of cefoperazone is transferred across the placenta (32). Only minimal transfer of cefoperazone occurs into breast milk, with a mean concentration of 0.3 µg/mL after a 1-g intramuscular dose (30). Kafetzis et al. (33) evaluated the pharmacokinetics of ceftriaxone in pregnant women. Despite its high protein binding (85% to 95%), ceftriaxone rapidly reached umbilical cord, amniotic fluid, and placenta concentrations in the clinically relevant range. The serum half-life for ceftriaxone was lower in pregnant women than in nonpregnant women. The elimination rates of ceftriaxone from fetal tissues were almost identical to those from maternal serum. This finding differs from those seen with other antibiotics, which tend to accumulate in the fetus. These authors reported that low concentrations that were only 3% to 4% of maternal serum levels of ceftriaxone occurred in breast milk (33). Side Effects In general, adverse effects with these new agents are similar to those seen with older cephalosporins, but there have been several special problems (26). As with the firstand second-generation cephalosporins, hypersensitivity reactions are the most common systemic adverse reactions seen with the third-generation drugs. Allergic reactions, rashes, local pain, and phlebitis have been reported infrequently. They probably occur in a similar frequency as with first-generation cephalosporins, but comparative studies
have not been performed. Diarrhea has been observed in 1% to 7% of patients receiving these compounds. Because cefoperazone is mainly excreted into the bile, it might be anticipated that it would cause diarrhea more often, but this has not been the case. Two special adverse effects have been seen with some of these new antibiotics: bleeding and a disulfiram reaction. Hypoprothrombinemia and bleeding have been associated with the use of moxalactam, cefoperazone, cefotetan, and cefamandole (34,35,36,37,38,39,40,41 and 42). These drugs have in common the presence of a MTT side chain on their molecule (38,39 and 40). The MTT side chain has been demonstrated to inhibit gamma carboxylation of glutamic acid, which is the vitamin K–dependent step in the synthesis of prothrombin (39,40). To date, this adverse effect has been reported to occur most frequently with the use of moxalactam (37), a drug no longer available. Most likely, the ability of moxalactam to also interfere with adenosine diphosphate–induced platelet aggregation is responsible for this greater risk for clinical bleeding. Bleeding due to moxalactam generally developed in patients with the following risk factors: renal or liver disease, poor nutrition, use of anticoagulants or aspirin, or thrombocytopenia. Obstetric-gynecologic patients would rarely have any of these risk factors. Kline et al. (41) reported that cefotetan also resulted in a statistically significant increase in prothrombin time in healthy volunteers. Hypoprothrombinemia and bleeding diathesis have been associated with cefotetan therapy (42). Also seen with moxalactam, cefoperazone, and cefotetan has been a disulfiram reaction (38,41), that is, acute alcohol intolerance due to inhibition of acetaldehyde dehydrogenase. This reaction has been seen in patients taking the antibiotics before alcohol, but not in intoxicated individuals who are treated with these antibiotics. The cephalosporins associated with disulfiram-type reactions also all contain the MTT side chain. Other adverse effects including mild elevation of liver enzymes and neutropenia have occurred infrequently and have been mild and transient. Because of their very broad spectra, superinfection may result. Enterococcal and candidal infections have been reported, but the incidence is low. In obstetric-gynecologic patients, particularly those who develop infection after prophylaxis with a first-generation cephalosporin, enterococci are isolated frequently, usually in mixed culture. Nevertheless, most of these patients respond promptly to these new antibiotics. Development of resistance during therapy has been seen with P. aeruginosa and E. cloacae. In debilitated patients, this has led to failure of therapy. In obstetric-gynecologic patients, colonization with resistant organisms has been noted with other antibiotics, but in healthy patients receiving short courses of therapy, poor clinical response has not been attributed to the development of resistance. When response has not been satisfactory and when enterococci are present, it is appropriate to use antibiotic therapy active against this organism. Indications in Obstetrics and Gynecology In view of their broad in vitro activity, these agents have been used in a number of trials as single-agent therapy in pelvic infections. Of these compounds, the one most
thoroughly evaluated in obstetric-gynecologic infections has been moxalactam. As shown in Table 23.7, reported experience includes more than 200 cases in which a consistent cure rate of approximately 90% has been achieved (43,44,45 and 46). In a double-blind comparison of moxalactam (6 g per day) and clindamycin-gentamicin for the treatment of post–cesarean section endomyometritis, Gibbs et al. (45) found the regimens to be equivalent. Similarly, Sweet et al. (46) noted excellent results (cure rates of higher than 95%) in a prospective controlled study of moxalactam versus clindamycin-tobramycin in the treatment of PID and tuboovarian abscess. Despite these excellent results, the risk of hypoprothrombinemia and bleeding diathesis has limited the use of moxalactam in obstetrics and gynecology and it is no longer marketed in the United States.
TABLE 23.7. CLINICAL TRIALS WITH THIRD-GENERATION CEPHALOSPORIN ANTIBIOTICS IN PELVIC INFECTIONS
Cefotaxime has been evaluated by Hemsell et al. in several reports (47,48 and 49). As with moxalactam, a higher cure rate has been seen with 6 g per day than with 3 g per day (97% vs. 84%; p < 0.05). Of the total of 143 women treated with the larger doses, there were 5 failures. Of these, two responded to the addition of other antibiotics and three required surgical drainage. In one phase of these studies, cefotaxime was compared with clindamycin-gentamicin. No difference in rates was found. These same investigators have evaluated cefotaxime in 53 cases of PID and 21 posthysterectomy infections. The cure rate was 96% for PID and 100% for the 21 other infections (48). More recently, Hemsell et al. (49) reported that cefotaxime (2 g every 8 hours) resulted in clinical cure in 39 (95%) of 41 patients with pelvic abscesses. Despite these excellent results, there is a high rate of resistance in the B. fragilis group to cefotaxime (9). This drug, though, is a reasonable single agent in most polymicrobial pelvic infections. In an open noncomparative study, Strausbaugh and Llorens (50) reported that cefoperazone achieved a clinical cure in 97 (91%) of 107 patients with various pelvic
infections. Gilstrap et al. (51) compared cefoperazone with a combination of clindamycin-gentamicin in the treatment of obstetric and gynecologic infections. They demonstrated clinical cure in 47 (92%) of 51 patients treated with cefoperazone and 48 (94%) of 51 patients in the clindamycin-gentamicin group. The high rate of resistance to cefoperazone among the B. fragilis group limits the usefulness of cefoperazone as a single agent for the treatment of severe pelvic infections. Studies of the effectiveness of ceftizoxime in the management of pelvic infections are limited. Both Harding et al. (53) and Lou et al. (54) demonstrated good clinical results in patients with pelvic and intraabdominal infections treated with ceftizoxime. However, Apuzzio et al. (55) reported clinical cure in only 49 (72%) of 68 patients with post–cesarean section endomyometritis treated with ceftizoxime (2 g intravenously every 12 hours) and 15 (72%) of 21 patients treated with ceftizoxime (3 g intravenously every 8 hours), compared with 28 (88%) of 32 patients receiving clindamycin-gentamicin and 15 (62%) of 24 patients receiving cefoxitin. These authors advised caution with the use of single-agent cephalosporins in the treatment of post–cesarean section endomyometritis. However, they used a 48-hour treatment time, rather than 72 hours to determine failure. Clinical studies of ceftriaxone as single-agent therapy in obstetric and gynecologic infections have not been reported. This is not surprising, because its lack of in vitro activity against B. fragilis, B. bivius, and B. disiens limits its use as a single agent in mixed anaerobic-aerobic infections of the pelvis. On the other hand, ceftriaxone (125 mg intramuscularly) is the drug of choice for uncomplicated anogenital gonorrhea. Ceftriaxone results in cure rates of 98% to 100% for uncomplicated anogenital and pharyngeal gonorrhea, both penicillin-sensitive and PPNG strains. In addition, ceftriaxone-doxycycline has been the primary outpatient regimen for acute PID during the past 15 years (24). In general, none of these agents have an extensive enough spectrum of activity to warrant its use in septic shock or similar serious infections. When the organism is known, other antibiotics are often preferable, as in staphylococcal or streptococcal wound infections. Although these agents may be effective for prophylaxis, there is no evidence that they are more effective than older less-expensive antibiotics. Indeed, it seems unwise to use them for prophylaxis. Accordingly, some of these agents would seem to be reasonable choices in the moderate polymicrobial infections commonly seen on obstetric-gynecologic services. These infections include endometritis after cesarean section and pelvic cellulitis after hysterectomy. For PID, all of these agents yield a high immediate cure rate, particularly in patients without a tuboovarian abscess. Yet, because none of these antibiotics are active against C. trachomatis, they must be combined with an antimicrobial agent with activity against C. trachomatis (i.e., tetracycline, azithromycin, erythromycin). Although it is not possible to discern a differential effect from clinical studies, the third-generation cephalosporins with somewhat better spectra are moxalactam and possibly ceftizoxime in preference to cefotaxime and cefoperazone. Side effects in obstetric and gynecologic patients are
seen infrequently and are mild, by and large. The cost of these agents is considerable. Decisions regarding the use of these versus a combination with an excellent efficacy (such as clindamycin-gentamicin) must take this aspect into consideration. Fourth Generation Cephalosporins Cefepime (Maxipime) is the first of the fourth-generation cephalosporins to become Food and Drug Administration (FDA) approved in the United States. The fourth-generation cephalosporins differ from third-generation agents because of a positively charged quaternary ammonium side chain on the cephem nucleus. This combines with the negative charge on the cephem nucleus to produce a zwitterion effect, which allows for increased penetration via porin channels through the outer membrane of Gram-negative bacteria (1,6). As a result, these agents have broad antimicrobial activity, including against P. aeruginosa and organisms producing many b-lactamases (6). Cefepime is active against a wide range of Gram-positive and Gram-negative bacteria (56). Its antipseudomonal coverage is similar to that seen with ceftazidime (56). Cefepime has excellent activity against non–cephalosporinase-producing Enterobacteriaceae such as E. coli, Klebsiella and Proteus species, N. gonorrhoeae and H. influenzae (1). Although the MIC values of cefepime against cephalosporinase-producing Enterobacteriaceae such as Enterobacter, Serratia, and Pseudomonas are higher than those seen with non–cephalosporinase-producing Enterobacteriaceae, it is superior to most third-generation cephalosporins against such bacteria (1). In addition, cefepime demonstrates excellent activity against streptococci, but only moderate activity against methicillin-sensitive S. aureus. Cefepime does not have significant activity against the B. fragilis group or MRSA. Cefepime is approved for treatment of uncomplicated and complicated UTIs, skin and soft tissue infections, and pneumonia. It appears to be equivalent to third-generation cephalosporins as therapy for community-acquired pneumonia and UTI (1). However, older less-expensive antibiotics are available for such infections. Cefepime has demonstrated clinical efficacy in the treatment of nosocomial infections (57,58). In particular, cefepime may be superior to third-generation cephalosporins, to which many nosocomial bacteria (e.g., P. aeruginosa and Enterobacteriaceae) have become resistant while remaining sensitive to cefepime (1). A combination of cefepime plus metronidazole has been demonstrated to be efficacious in the treatment of severe intraabdominal infections (59). New Oral Cephalosporins The first useful oral cephalosporin introduced was cephalexin (Keflex). This was followed by the development of cephradine (Anspor, Velosef, which is no longer available) and cefadroxil (Duricef). All three of these oral cephalosporins have virtually the same antimicrobial activity and are very active against staphylococci, Streptococcus pyogenes, S. pneumoniae, and Enterobacteriaceae (some E. coli, Klebsiella species, and P. mirabilis) (1). Cephalexin and cephradine also have similar pharmacokinetic
properties, whereas cefadroxil has a longer half-life, allowing twice-daily dosing. These early oral cephalosporins had poor activity against H. influenzae and b-lactamase–producing Enterobacteriaceae. In general, cephalexin and cefadroxil are useful agents for the treatment of UTIs due to susceptible Gram-negative bacteria and minor skin and soft tissue infections due to S. aureus and streptococci (2). However, these drugs (particularly cefadroxil) are expensive and are not recommended for routine treatment of UTIs. Cefaclor was introduced in the late 1970s and had the advantage of activity against H. influenzae, including some b-lactamase producers (1). However, 10% to 15% of ampicillin-resistant strains of H. influenzae are resistant to cefaclor. The synthesis of cefuroxime introduced an oral cephalosporin that was resistant to b-lactamases and inhibited many pathogens responsible for respiratory, skin, and UTIs (3). Conversion of cefuroxime to an axetil ester made it orally absorbable. Cefuroxime axetil (Ceftin) is an oral agent with activity against S. aureus, groups A and B streptococci, viridans streptococci, and many Gram-negative aerobes including E. coli, K. pneumoniae, H. influenzae, and N. gonorrhoeae (including b-lactamase–positive strains). Cefuroxime axetil is effective for treating UTIs, skin and soft tissue infection, and upper respiratory infections due to susceptible organisms (2). However, it is also expensive. Recently, several new oral cephalosporins have become clinically available in the United States. Cefprozil (Cefzil) is an oral second-generation cephalosporin approved for use in the treatment of pharyngitis, bronchitis, otitis media, and skin and soft tissue infections (4,5). Its activity is similar to that of cefaclor and cefuroxime axetil (4), with good activity against methicillin-susceptible S. aureus, groups A and B streptococci, S. pneumoniae, L. monocytogenes, H. influenzae, and community-acquired Enterobacteriaceae (E. coli, Klebsiella species, and P. mirabilis). Cefprozil has a half-life of 1.3 hours, which allows once-daily or twice-daily dosing. Like other oral cephalosporins, cefprozil is expensive. Whether it has any unique advantage over other available agents is unclear. Loracarbef (Lorabid) is an oral b-lactam antibiotic of the carbacephem class. Its in vitro activity is similar to that of cefaclor (2,6). Loracarbef is approved for mild to moderate acute bronchitis; pneumonia caused by S. pneumoniae or non–b-lactamase-producing H. influenzae; otitis media caused by S. pneumoniae, M. catarrhalis, S. pyogenes, and H. influenzae; pharyngitis or tonsillitis caused by S. pyogenes; and acute maxillary sinusitis (2). However, other drugs that are cheaper have similar activities, and the role of this agent awaits further study. Cefixime (Suprax) is a third-generation cephalosporin that is active against S. pneumoniae, H. influenzae, Neisseria. catarrhalis, N. gonorrhoeae, and many of the Enterobacteriaceae (1). S. aureus is resistant to cefixime, as is P. aeruginosa. Cefixime is more active than other oral cephalosporins against E. coli, Klebsiella species, P. mirabilis, and S. marcescens. This agent has a long half-life of 4 hours and thus can be administered once or twice daily. However, cefixime is expensive and for most indications is not any more effective than less-expensive agents (2). Thus, for otitis media, sinusitis, pharyngitis or tonsillitis, bronchitis, or UTIs, cefixime is an alternative agent but not the first-line choice. Exceptions to this are otitis media due to
b-lactamase–producing H. influenzae or M. catarrhalis and uncomplicated anogenital gonorrhea. For the latter, cefixime has been shown to be as effective in a single dose of 400 mg orally as intramuscular ceftriaxone (7,8). Cefpodoxime proxetil (Vantin) is another cephalosporin available as an oral agent. It is active in vitro against groups A and B streptococci, S. pneumoniae, H. influenzae (b-lactamase positive and negative), and N. gonorrhoeae. Against S. aureus, cefpodoxime proxetil has only modest activity. It is also active against many Enterobacteriaceae but not Enterobacter, Serratia, or Morganella species (2). Although cefpodoxime has been shown to be effective in the treatment of bronchopneumonia, acute otitis media, pharyngitis, UTI, skin and soft tissue infections, acute and chronic bronchitis, and uncomplicated gonorrhea (2,9), it is expensive and does not offer any advantage over previously available drugs for the treatment of any infection (10). Ceftibuten (Cedax) is the most recently marketed oral cephalosporin (third generation) in the United States. Its antimicrobial spectrum of activity is limited to streptococci, H. influenzae and most Enterobacteriaceae such as E. coli, P. mirabilis, and K. pneumoniae (1,6). Ceftibuten has poor activity against anaerobic bacteria. Although ceftibuten is clinically effective in the treatment of acute bronchitis, pneumonia (except penicillin-resistant S. pneumoniae), acute otitis media and pharyngitis, it provides no significant advantage over the older less-expensive cephalosporins. Table 23.8 summarizes the indications, dosing, and cost of the oral cephalosporins.
TABLE 23.8. ORAL CEPHALOSPORINS: INDICATIONS, DOSING, AND COST
MONOBACTAMS Aztreonam is the first monobactam to be clinically available. It has an unusual spectrum in that it is mainly active against the aerobic Gram-negative rods (1). However, it is inactive against Gram-positive organisms or anaerobic bacteria. Yet, like the cephalosporins and penicillins, it has a wide margin of safety and readily achieves therapeutic levels. Aztreonam inhibits most Enterobacteriaceae at levels of less than 0.5 µg/mL. Most P. aeruginosa are inhibited by less than 16 µg/mL, but some are resistant.
Aztreonam has the advantage of remaining active in an anaerobic environment (e.g., abscess), whereas the aminoglycosides are inactive in anaerobic situations. In addition, aztreonam is not associated with the nephrotoxicity that occurs with aminoglycosides. Hence, this is the potential advantage of these antibiotics over aminoglycosides as drugs of choice in treating Gram-negative infections. In the empiric treatment of pelvic infections, however, aztreonam must be combined with an agent such as clindamycin to provide anaerobic and Gram-positive activity. Aztreonam is available only in parenteral form. Intravenous boluses of 500, 1,000, and 2,000 mg produced peak serum levels of aztreonam of 58, 125, and 242 µg/mL, respectively (2). Aztreonam is 50% to 70% protein bound. The serum half-life of aztreonam is 1.7 hours, and 60% of the drug is excreted in urine. In the presence of renal failure, aztreonam accumulates, so the dose must be altered accordingly. Aztreonam crosses the placenta and is found in low concentrations in fetal serum and breast milk. Similar to b-lactam agents, aztreonam is a safe antimicrobial for which no major adverse reactions have been reported. The most common adverse reactions have been rashes and increases in serum transaminase levels. Aztreonam, in a 1-g intramuscular dose, was noted to be 100% effective for uncomplicated anogenital gonorrhea. However, aztreonam has no activity against C. trachomatis, and an agent active against Chlamydia must be given concurrently when treating gonorrhea. In treating endometritis, Gibbs et al. (3) compared aztreonam (2 g intravenously every 8 hours) with gentamicin (1.5 mg/kg intravenously every 8 hours), each given with clindamycin. Both regimens were highly efficacious and well tolerated. The aztreonam-clindamycin combination resulted in an 88% cure rate, versus a 91% cure rate with gentamicin-clindamycin. Dodson et al. (4) reported that aztreonam-clindamycin produced clinical cure in 98% of patients with acute PID. Although the toxicity of aztreonam is clearly less than that of aminoglycosides, obstetric and gynecologic patients are usually at low risk for aminoglycoside toxicity, and aztreonam is considerably more expensive than the aminoglycosides. Potential benefits of the monobactams have to be weighed against their high cost. Thus, aztreonam should be reserved for special situations such as in patients at high risk for aminoglycoside toxicity (e.g., renal disease).
CARBAPENEMS Carbapenems are derivatives of thienamycin, and currently two of these agents are marketed in the United States: imipenem-cilastatin (Primaxin) and meropenem (Merrem). Imipenem (Primaxin) was the first clinically available member of a new class of antimicrobial agents, the carbapenems. These agents differ structurally from penicillins and cephalosporins. The formulation clinically available is a combination of imipenem and cilastatin, a dipeptidase inhibitor that prevents the nephrotoxicity associated with the use of imipenem alone. Imipenem has the broadest spectrum of in vitro activity of any b-lactam agent (1,2 and
3). This drug has excellent activity against aerobic Gram-positive bacteria, including hemolytic streptococci groups A and B, S. pneumoniae, Enterococcus, non-MRSA, and S. epidermidis. Listeria is also susceptible. Among the Enterobacteriaceae (E. coli, Klebsiella, Proteus, Enterobacter), most are sensitive to imipenem levels of 1 µg/mL or less. Although P. aeruginosa is sensitive in vitro, resistance to imipenem has emerged during treatment. Imipenem has excellent activity against anaerobic bacteria, including B. fragilis, B. bivius, and B. disiens. Methicillin-resistant staphylococci, mycoplasmas, and C. trachomatis are resistant to imipenem. Imipenem is 20% protein bound; cilastatin is 40% protein bound. After a 1-g dose of imipenem, peak levels of 70 µg/mL are obtained in serum. The serum half-life of imipenem is approximately 1 hour, and when combined with cilastatin, 70% of the imipenem is excreted unchanged in urine. No studies are available on transplacental transfer or breast milk levels for imipenem. In an open noncomparative study, Berkeley et al. (4) reported that imipenem resulted in clinical cure in 43 (88%) of 49 obstetric and gynecologic patients. Three of the failures had pelvic abscesses that required surgical intervention. Of concern was the occurrence of pseudomembranous colitis in two patients. Similarly, Sweet (5) demonstrated the excellent clinical efficacy of imipenem for the treatment of obstetric and gynecologic infections. In this study, clinical response occurred in 70 (97%) of 72 patients treated with imipenem, including all 7 patients with tuboovarian abscesses (documented with ultrasound). In a small comparative study, Berkeley et al. (6) reported a 100% cure rate with imipenem versus a 62% cure rate with moxalactam in the treatment of serious obstetric and gynecologic infections. As with other b-lactam antibiotics, the most common adverse reactions with imipenem are of the allergy variety in 2% to 3% of patients. Patients who are allergic to penicillin should not be given imipenem. Other reported adverse effects include nausea (1% to 2%), phlebitis and pain at infusion site (2% to 4%), and pseudomembranous colitis (less than 1%). Rarely, seizures have occurred, primarily in elderly patients with underlying CNS disease. The cost of imipenem remains substantial: approximately $60 per day at a dosage of 500 mg every 6 hours. In view of its unusual spectrum and high cost, it seems best to reserve this drug for special situations including infections with known or suspected resistant pathogens. Meropenem is the second carbapenem group of antimicrobial agents to become clinically available in the United States. Unlike imipenem, it is stable to dehydropeptidase activity in the kidney (7). Meropenem is slightly more active against Gram-negative bacteria than imipenem. It is active against P. aeruginosa resistant to imipenem. Otherwise, its spectrum of activity is very similar to that of imipenem with excellent activity against Gram-positive organisms, Gram-negative aerobes, and anaerobic bacteria. The mean peak serum concentration after a single 500-mg intravenous dose is 30 µg/mL with a half-life of 1 hour. The side effects seen with meropenem are similar to those seen with imipenem, except that meropenem is less
likely to produce seizure activity (7). Clinically, meropenem appears to be equivalent to imipenem and has been used with success to treat pneumonia, meningitis, bacteremia, UTI, intraabdominal sepsis, and febrile episodes in neutropenic patients (7). The recommended adult dosage is 500 mg to 1 g intravenously every 6 to 8 hours.
TETRACYCLINES All tetracyclines have a similar molecular structure consisting of four benzene rings and generally the same spectrum of activity. The tetracyclines are primarily bacteriostatic. There are two major groups of tetracyclines, which are based on pharmacokinetic properties: short-acting compounds of which tetracycline hydrochloride is the prototype and long-acting compounds, including doxycycline. Spectrum of Activity Tetracyclines had a broad spectrum of activity that includes Gram-positive, Gram-negative, aerobic, and anaerobic bacteria, spirochetes, chlamydiae, mycoplasmas, rickettsiae, and even some protozoa. However, because many organisms have acquired resistance, the role of tetracyclines in obstetric-gynecologic infections is limited, mainly to treating sexually transmitted diseases (STDs) (1,2). The tetracyclines are not drugs of choice for infection caused by Gram-positive aerobes (3). N. gonorrhoeae and N. meningitides had been generally susceptible to tetracyclines. However, increasingly, strains of N. gonorrhoeae that are resistant to tetracycline have been reported (4,5). Gonococci resistant to penicillin are likely to be resistant to tetracyclines; approximately 50% of PPNG are resistant to tetracycline as well (5). Not only are PPNG strains resistant to tetracycline, but chromosomally mediated and plasmid-mediated high-level tetracycline-resistant N. gonorrhoeae are also a major problem in the United States. Thus, the CDC no longer recommends tetracycline or doxycycline for the treatment of N. gonorrhoeae (except as concomitant therapy for Chlamydia). Tetracyclines are active against some Enterobacteriaceae such asE. coli (particularly community-acquired strains), Enterobacter, and Klebsiella. However, many Enterobacteriaceae and Pseudomonas species are resistant to tetracyclines. Although the tetracyclines inhibit the growth of many anaerobic bacteria, their activity is not comparable to that of agents such as clindamycin, chloramphenicol, metronidazole, or the newer cephalosporins and extended-spectrum penicillins. Tetracyclines (tetracycline hydrochloride, doxycycline, and minocycline) are the drugs of choice against C. trachomatis, mycoplasmas (Mycoplasma hominis, Ureaplasma urealyticum, and Mycoplasma pneumoniae), and rickettsiae. Tetracyclines are bacteriostatic. They enter the bacterial organism, where they bind reversibly to the 30S ribosomal unit of susceptible organisms. As a result, polypeptide synthesis is inhibited.
Dosage and Route Tetracycline hydrochloride and doxycycline are available for administration by the oral, intramuscular, and intravenous routes. The tetracyclines are most commonly administered orally. In Table 23.9, we show the dosing schedule for oral administration of tetracyclines.
TABLE 23.9. TETRACYCLINES: PHARMACOKINETICS AND DOSING
Of the short-acting forms, tetracycline is the one that is best absorbed, obtains the highest serum levels, and is the least costly. The usual adult oral dosage is 500 mg every 6 hours. Absorption takes place in the proximal small bowel. Peak serum levels of 4 µg/mL are reached 1 to 3 hours after oral administration of a 500-mg dose of tetracycline. After a 500-mg oral dose, oxytetracycline reaches a peak serum level of 2 to 3 µg/mL, and chlortetracycline, 1 to 2 µg/mL. For doxycycline, the usual adult oral dosage is 100 mg every 12 hours. Doxycycline is almost completely absorbed in the proximal bowel after oral administration. The serum half-life is 18 to 22 hours. After an oral dose of 200 mg, peak serum levels of 2 to 4 µg/mL are attained. Although available, intramuscular forms of the tetracyclines are irritating and rarely used. After intravenous administration of 500 mg of tetracycline, serum levels reach about 8 µg/mL at 30 minutes and decrease to 2 to 3 µg/mL by 5 hours. Intravenous doxycycline or minocycline in a dose of 200 mg produces a serum level of 4 µg/mL at 30 minutes. Food of any type markedly reduces the absorption of tetracycline, oxytetracycline, or chlortetracycline. These short-acting agents also have a marked affinity for divalent cations (calcium, magnesium, aluminum), to which they bind and chelate, and thus are
excreted in the feces. Tegretol, phenytoin, and barbiturates decrease the normal half-life of doxycycline by almost 50% (3). Side Effects The tetracyclines may be associated with local and systemic untoward effects. Allergic reactions including rash, urticaria, and anaphylaxis occur but are uncommon. Photosensitivity reactions associated with onycholysis occur in areas exposed to sunlight. These drugs are irritating, and frequently oral administration produces gastrointestinal symptoms such as nausea, vomiting, and epigastric distress. Diarrhea (including pseudomembranous colitis) may develop, particularly with the forms that are poorly absorbed (i.e., chlortetracycline and oxytetracycline). Secondary infection with Candida albicans is a frequent complication of tetracycline treatment. Tetracycline is deposited in the deciduous teeth of children early in life or if they were exposed as a fetus. The period of mineralization of the decidual teeth commences during the mid trimester of pregnancy and ends 2 to 3 months after birth. Exposure to tetracycline during this time results in yellow discoloration of the “baby” teeth. If tetracyclines are administered to children younger than 6 years, a lifelong discoloration of the permanent teeth may occur. The deposition of tetracycline in the bones of infants causes temporary inhibition of bone growth. A recent case-control analysis did not show any teratogenic potential of doxycycline (6). Nevertheless, tetracyclines should be avoided during pregnancy or while breast-feeding. A dose-related hepatotoxicity occurs, particularly during the third trimester of pregnancy, with doses of more than 2 g intravenously of tetracycline. It appears pathologically as “fatty liver” of pregnancy and is associated with a high mortality rate (7). Tetracycline therapy may result in nephrotoxicity as well, either as a result of “acute fatty liver” or by a direct toxic renal effect that aggravates preexisting renal failure. The latter phenomenon results from tetracycline's ability to inhibit protein synthesis, which increases the azotemia from amino acid metabolism. Minocycline, but not the other tetracyclines, can cause vestibular disturbances with dizziness, ataxia, vertigo, and tinnitus. It is reversible when the drug is stopped. This side effect is much more common in women than in men, occurring in 50% to 70% of women receiving minocycline. As a result of this frequent and bothersome side effect, minocycline is rarely used in female patients. Cost Tetracycline is a relatively inexpensive antibiotic. Chlortetracycline, oxytetracycline, doxycycline, and minocycline are more costly for a 7- to 10-day oral course of treatment. However, the cost of doxycycline has recently been lowered. Use in Pregnancy and Placental Transfer Tetracyclines cross the placenta and are deposited in decidual teeth and growth centers
of long bones, with resultant discoloration of decidual teeth and inhibition of bone growth. Thus, tetracyclines are best not used in pregnancy and while breast-feeding unless alternative drugs are not available. Metabolism All the tetracyclines are excreted by the kidneys via glomerular filtration. Nearly 20% of an orally administered dose of tetracycline is excreted in the urine, whereas about 50% of a parenterally administered tetracycline dose is excreted in the urine within 24 hours. Tetracyclines are also excreted in the bile, but a large proportion of this is reabsorbed from the intestine. Except for doxycycline and minocycline, tetracyclines are incompletely absorbed from the gastrointestinal tract, and this unabsorbed drug is excreted in the feces. Both tetracycline and doxycycline should be taken either 3 hours before or 2 hours after taking iron or antacids. The absorption of tetracycline is impaired by food. Thus, tetracycline should not be taken with meals. Doxycycline's absorption is not inhibited by food in the upper gastrointestinal tract (1). Indications in Obstetrics and Gynecology Tetracyclines should not be administered to pregnant or lactating women. The major indication for tetracycline in gynecology practice is the treatment of C. trachomatis infections, either singly or in combination with other agents depending on the clinical setting (Chapter 3 [Group B Streptococci], Chapter 4 [Genital Mycoplasmas], and Chapter 8 [Mixed Anaerobic-Aerobic Infection and Pelvic Abscess]). Doxycycline is used in the treatment of PID. The tetracyclines are commonly used as prophylactic agents in women undergoing pregnancy termination. In nonpregnant patients allergic to penicillin, it is the alternative drug for the treatment of syphilis. Tetracyclines are the drug of choice against M. hominis and Ureaplasma. Tetracyclines are also used to treat granuloma inguinale, lymphopathia venereum, actinomyces, and nongonococcal urethritis. In primary care practice, tetracyclines are used to treat acne and Lyme disease (1).
CLINDAMYCIN Clindamycin (Cleocin), which is derived from lincomycin, is a macrolide antibiotic. Because of its spectrum of activity, particularly against anaerobes and Gram-positive aerobes, clindamycin is important in treating pelvic infections (1). Spectrum of Activity Clindamycin attaches to the 50S ribosome and inhibits bacterial protein synthesis. It is bacteriostatic. Against Gram-positive aerobic organisms, clindamycin has very good activity, including nafcillin-sensitive S. aureus and nonenterococcal streptococci. Clindamycin maintains excellent activity against group A streptococci. Approximately 5% of group B streptococci show in vitro resistance. Among hospital isolates, 5% to 20% of S. aureus isolates are resistant to clindamycin, and MRSA strains are usually resistant. N. gonorrhoeae is variably susceptible to clindamycin in vitro. However, in
combination with aminoglycosides, clindamycin has effectively eradicated N. gonorrhoeae. The entire group of aerobic Gram-negative bacilli are highly resistant. Its activity against the anaerobic organisms provides clindamycin with its main value. With the exception of very few strains of Clostridium species and a few strains of Bacteroides species, anaerobic organisms are susceptible to this antibiotic. Cuchural et al. (2) reported that 5% of B. fragilis group organisms were resistant to clindamycin during 1984 and 1985. However, more recent reviews have demonstrated a prevalence of resistance among B. fragilis group bacteria, as high as 20% to 25% in some geographic areas (3,4). Clindamycin is active against M. hominis but not U. urealyticum. Clindamycin has very good in vitro activity against C. trachomatis (5). Clinical studies have demonstrated that in large doses, clindamycin has excellent clinical efficacy against C. trachomatis (6). The MIC for susceptible organisms varies from 0.01 to 4 µg/mL. Dosage And Route Clindamycin may be administered intravenously, intramuscularly, orally, or intravaginally as a cream. For intravenous or intramuscular administration, clindamycin phosphate is used. For initial treatment of serious infections, the dosage for intravenous injection is usually 600 mg every 6 hours or 900 mg every 8 hours by “piggyback” injection with an infusion over 30 minutes. After a 300-mg infusion over 30 minutes, serum levels peak at 12 to 15 µg/mL and fall to 5 µg/mL in 2 to 3 hours. For intramuscular injection, the maximum dosage is 300 to 600 mg every 8 hours. For oral administration, clindamycin hydrochloride is used, in a dosage of 150 to 300 mg every 6 hours. For treatment of C. trachomatis in patients with acute PID, an oral dosage of 450 mg every 6 hours is recommended. Gastrointestinal absorption is impaired with food in the stomach. No reduction is necessary for patients in renal failure, unless it is severe. For patients with liver disease, the dose should be decreased. Side Effects Gastrointestinal Symptoms Nausea, vomiting, and particularly diarrhea occur fairly commonly during clindamycin therapy. The incidence of diarrhea varies widely in large series, but in obstetric-gynecologic patients, the incidence has been from 2% to 6%. Pseudomembranous Colitis (Clostridium difficile–Associated Diarrhea) Pseudomembranous colitis was first observed in patients taking lincomycin. Severe diarrhea and pseudomembranous colitis were subsequently reported in the 1970s in patients taking either oral or parenteral clindamycin. The pathogenesis of antibiotic-associated diarrhea was linked primarily to the anaerobic organism C. difficile. The diagnosis of C. difficile–associated diarrhea (CDAD) should be suspected whenever
a patient who has received antibiotic begins to have diarrhea, particularly if accompanied by abdominal cramps and fever. It should be recognized that the onset of CDAD may be delayed for months (7). Pseudomembranous colitis has been reported as a complication with various broad-spectrum antibiotics, including tetracycline, ampicillin, and the cephalosporin antibiotics, in addition to clindamycin. C. difficile is resistant to clindamycin and many other antibiotics. Thus, various antibiotics may eliminate susceptible intestinal organisms and allow C. difficile, if present, to proliferate and elaborate its toxin. The organism is susceptible to vancomycin and metronidazole. Practice guidelines for the diagnosis of CDAD are shown in Table 23.10A. Both the American College of Gastroenterology and the Society for Health Care Epidemiology of America recommend that testing be done on a stool specimen for the presence of C. difficile or its toxins. The Society for Health Care Epidemiology of America indicates that stool culture is the most sensitive test for CDAD, whereas the C. difficile toxin is the most specific and further suggest that for maximal diagnostic sensitivity and specificity, performance of both tests is recommended. The American College of Gastroenterology further recommends that if the results of these tests are negative but diarrhea persists, one or two additional stool samples can be sent for testing with these tests. Proctoscopic examination is now reserved for special situations, for example, when a rapid test is needed, when other test results are delayed, or when the test is not highly sensitive. Although CDAD will resolve in 15% to 25% of patients, antibiotic therapy is indicated. Metronidazole, vancomycin, and other antibiotics have all been shown to be effective in CDAD, but metronidazole is now considered the initial drug of choice. It has comparable efficacy to vancomycin and it has a lower cost. In addition, there is a hypothetical concern about the use of vancomycin primarily in a situation leading to further resistance with organisms such as enterococci. The recommended therapeutic regimen of first choice is metronidazole (250 mg four times daily or 500 mg three times daily for 10 days). (Please see Table 23.10B.) No diagnostic testing is indicated at the end of therapy, but repeated testing is recommended if diarrhea recurs. Although most patients respond to a specific antibiotic therapy, up to 30% of patients will have recurrence of CDAD. This usually occurs within 1 to 2 weeks.
TABLE 23.10. IN VITRO ACTIVITY OF ERYTHROMYCIN, AZITHROMYCIN, AND
CLARITHROMYCIN
The practice guidelines for the prevention and control of C. difficile infection by the American College of Gastroenterology recommend the following: limiting the use of antimicrobial drugs, washing hands between contact with all patients, using stool isolation precautions for patients with CDAD, wearing gloves when in contact with patients who have CDAD or within their environment, disinfecting objects contaminated with C. difficile with appropriate agents, and educating the medical, nursing, and other appropriate staff about the disease and its epidemiology. For patients who develop diarrhea while receiving antibiotic therapy, we stop suppositories and laxatives and next discontinue clindamycin. If treatment of the B. fragilis group is still needed, we will empirically replace it with an alternative agent such as metronidazole. Lomotil and related compounds should be avoided. Continued use of antibiotics has been shown to prolong the course of the colitis and to worsen the prognosis. Vigorous supportive therapy is essential, but use of Lomotil is dangerous. Other Side Effects For the most part, other adverse reactions to clindamycin are rare. Minor elevations in liver function test results (particularly serum glutamic-oxaloacetic transaminase) occur in a small percentage of patients. Allergic reactions such as rash are rare. Occasionally, neutropenia and thrombocytopenia have occurred. Cost Both the oral and parenteral preparations of clindamycin remain expensive. Use in Pregnancy Although clindamycin carries the standard FDA warning that safety of the use of this drug in pregnancy has not been established, there is indirect evidence that it is probably safe. In a long-term prospective evaluation of lincomycin in pregnancy, no adverse fetal effects were noted. In view of the close structural similarity of lincomycin and clindamycin, the latter is probably also a relatively safe agent. Thus, pregnancy should not be considered a contraindication to the appropriate use of clindamycin. Placental Transfer After intravenous administration to the mother at term, clindamycin rapidly appears in the cord blood. Peak levels are achieved within 20 minutes and are approximately 40%
of peak maternal levels. Clindamycin did not appear in the amniotic fluid in detectable concentrations during the first hour after maternal injection. Metabolism Clindamycin is removed from the body largely by direct excretion in the bile. A smaller amount is excreted by the kidney, and some is also inactivated in the body, probably in the liver. Clindamycin is excreted in breast milk in small concentrations (1). Indications in Obstetrics and Gynecology Clindamycin is a potent and valuable antibiotic, but its use has been accompanied on rare occasion by serious adverse effects. Parenteral clindamycin is indicated in the empiric treatment of pelvic infections such as endometritis, pelvic cellulitis after hysterectomy, PID, tuboovarian abscess, and pelvic abscess. Because these infections are polymicrobial, clindamycin should be used initially in combination with other agents such as aminoglycosides that have activity against Gram-negative aerobic bacteria. Because of widespread clinical use and consistently high cure rates in many studies, clindamycin plus an aminoglycoside has been considered the standard, against which other antibiotics are measured for the treatment of mixed anaerobic-aerobic pelvic infections (8). Excluding side-effect failures, cure rates with this combination are generally from 90% to 97%. Oral clindamycin may be used to complete a course of therapy in a patient who has shown a response to parenteral clindamycin, but for most pelvic infections, it is unnecessary to follow parenteral therapy with oral therapy in patients who have responded. Exceptions to this rule are treatment of C. trachomatis or abscesses. Although S. aureus is susceptible to clindamycin, it is not the drug of choice for infections caused by these organisms. Penicillinase-resistant penicillins (such as methicillin and others) are preferred, and the cephalosporin antibiotics would be the alternative in most patients with a penicillin allergy. In the rare patient with anaphylaxis to penicillin or a cephalosporin, clindamycin may be used to treat infections with S. aureus. A more recent use of clindamycin is for the treatment of bacterial vaginosis. Both oral clindamycin (300 mg twice daily for 7 days) (9) and clindamycin cream (2% intravaginally twice daily for 5 days) (10,11) are very effective (equal efficacy to oral metronidazole) for the treatment of bacterial vaginosis. Clindamycin is currently listed as an alternative for intrapartum prophylaxis for the prevention of perinatal group B streptococcal infections. Several reports have raised the concern of resistance of group B streptococci to clindamycin. These reports generally indicate that up to 4% or 5 % may be resistant, but resistance rates are considerably higher with erythromycin (12,13).
ERYTHROMYCIN, AZITHROMYCIN, AND CLARITHROMYCIN
Erythromycin Erythromycin is a macrolide antibiotic composed of a many-membered lactose ring with one or more deoxy sugars attached. Erythromycin is most active in an alkaline medium and is primarily bacteriostatic. Spectrum of Activity The antibacterial spectrum of erythromycin is limited. It is active against various aerobic Gram-positive cocci including groups A and B streptococci and S. pneumoniae. S. aureus is generally susceptible to erythromycin, but emergence of resistance does occur during treatment (1,2 and 3). Erythromycin is the drug of choice for C. diphtheriae. Similarly, it is the drug of choice for Legionella pneumophila and Haemophilus ducreyi. Erythromycin is also active against N. gonorrhoeae, Borrelia burgdorferi, L. monocytogenes, T. pallidum, and M. pneumoniae. Many Gram-positive and Gram-negative anaerobic bacteria are inhibited, but by relatively high concentrations. Erythromycin also is active against C. trachomatis, C. pneumoniae, M. hominis, and U. urealyticum. Table 23.10 compares the in vitro activity of erythromycin with that of azithromycin and clarithromycin. Erythromycin is bound to the 50S ribosomal unit of susceptible microorganisms and inhibits polypeptide synthesis in ribosomal complexes. The drug accumulates 100 times more in Gram-positive than in Gram-negative bacteria. Dosage and Route Erythromycin preparations are available for administration by the oral, intramuscular, and intravenous routes. Oral preparations of erythromycin include erythromycin base (E-Mycin), erythromycin stearate (Bristamycin, Ethril, Pfizer, SK-Erythromycin, Erypar, Erythrocin Stearate), erythromycin estolate (Ilosone), and erythromycin ethylsuccinate (Pediamycin, Wyamycin) (1,2,3 and 4). Erythromycin base is a very bitter, weak base (pK = 8.8). To make it tolerable, enteric coatings have been applied. In some preparations of erythromycin base, gastrointestinal absorption has been erratic, but E-Mycin (Upjohn) is well absorbed in both the fasting and nonfasting state. Mean peak serum levels of E-Mycin are 0.73 µg/mL 5 hours after a single 250-mg dose. Similar levels are achieved in nonfasting subjects. After multiple dosages of 250 mg four times daily, peak levels of 1.5 µg/mL are achieved. Erythromycin stearate hydrolyzes to erythromycin base in the duodenum, where absorption takes place. After a 250-mg dose in a fasting subject, levels of 0.82 µg/mL are achieved 2 hours later. Multiple doses every 6 hours lead to levels of 1.0 to 1.5 µg/mL in the fasting state. This preparation should be given 1 hour before or 2 hours after meals. When given with meals, erythromycin is destroyed considerably because of longer periods in the stomach and greater exposure to acid. Acceptable bioavailability of
this preparation also depends on administration with adequate amounts of water. Erythromycin estolate is acid stable, and its absorption is unaffected by food. Although this preparation achieves higher levels than erythromycin base or stearate esters, erythromycin estolate is apparently absorbed primarily as the propionate ester, which is not active until hydrolyzed to the base. Erythromycin ethylsuccinate is tasteless and is reported to be stable in acid. The usual adult dosage is 250 to 500 mg orally every 6 hours. Because intramuscular administration of erythromycin is associated with extreme pain, the antibiotic should not be administered by this route. Intravenous administration of erythromycin is associated with a significant incidence of thrombophlebitis at the intravenous site. The dosage for this route is usually 500 mg every 6 hours. Side Effects The erythromycins have a good safety record but may be associated with both local and systemic untoward effects. Epigastric distress and gastrointestinal upset occur often after oral administration, particularly in pregnant women. Individuals experiencing these symptoms at 500 mg every 6 hours are often able to tolerate 250 mg every 6 hours. Pseudomembranous enterocolitis has rarely been observed. Pain may develop after intramuscular administration. Hypersensitive reactions, manifested by fever, eosinophilia, and skin eruption, may occur. The only major adverse effect is hepatotoxicity, which has been seen with the estolate and ethylsuccinate esters. This syndrome is not dose related and usually does not occur after the first exposure to the drug. After 1 to 3 weeks of therapy, patients may develop abdominal cramps, nausea, vomiting, jaundice, fever, leukocytosis, and abnormal liver function. All manifestations usually resolve when administration of the drug has been discontinued. Cost As an older preparation, erythromycin is a relatively inexpensive antibiotic. Use in Pregnancy and Placental Transfer Erythromycin has not been associated with adverse effects on the fetus, but the estolate salt may cause mild hepatotoxicity more commonly in pregnant women than in other adults (5). Approximately 10% of 161 women treated with the estolate ester in the second trimester had elevated serum glutamic-oxaloacetic transaminase levels. These returned to within the reference range after the antibiotic was discontinued. Erythromycin crosses the placenta but achieves cord blood concentrations of only 6% to 20% of maternal levels (6). Like many other antibiotics, erythromycin may lower urinary estriol levels (7).
Metabolism When erythromycin is administered orally, 2% to 5% of the antibiotic is excreted in active form in the urine. After intravenous administration, 12% to 15% is excreted in the urine. Erythromycin is concentrated in the liver and is excreted primarily in the active form in the bile. Indications in Obstetrics and Gynecology Erythromycin has been indicated for the treatment of staphylococcal and streptococcal infections in penicillin-allergic patients. Recently, recognition of the importance of infections due to C. trachomatis broadened the clinical indications for the use of this antibiotic, particularly in pregnancy (when tetracycline is contraindicated). Schachter et al. (8) demonstrated that the treatment of pregnant women with cervical chlamydial infection with erythromycin ethylsuccinate (800 mg three times a day) eradicated C. trachomatis from mothers and prevented vertical transmission to their newborns in more than 90% of cases. Erythromycin is also an alternative to tetracycline for the treatment of Ureaplasma, but clinical indications for treating this organism are not clear. M. pneumoniae is among the microorganisms most frequently responsible for the production of lower respiratory infection (“walking pneumonia”) in people 20 to 40 years of age. L. pneumophila has been recognized as an important cause of sporadic pneumonia (Legionnaires disease). Erythromycin is the first choice for the treatment of these two organisms. Erythromycin is no longer recommended for treatment of gonorrhea or syphilis. Azithromycin Azithromycin (Zithromax) is the first new azalide antibiotic to be clinically available. Inserting a nitrogen atom into the lactone ring of erythromycin resulted in an agent with unique properties (9). Thus, azithromycin has an expanded spectrum of activity, high and sustained tissue levels of antibiotic (higher than serum levels), and a prolonged tissue half-life (2 to 4 days), which permits fewer doses per course of therapy and a shorter duration of therapy. Spectrum of Activity Azithromycin, like erythromycin, is a broad-spectrum antimicrobial agent that is active against Gram-positive and some Gram-negative bacteria, chlamydiae, mycoplasmas, and some spirochetes (1,2 and 3,10,11). Azithromycin is active against strains of S. aureus, groups A and B streptococci, S. pneumoniae, and coagulase-negative staphylococci that are susceptible to erythromycin. If resistant to erythromycin, strains of these Gram-positive bacteria are also resistant to azithromycin. In general, azithromycin is twofold to fourfold less active against staphylococci and streptococci.
Although azithromycin is more active than erythromycin in vitro against H. influenzae and Neisseria species, Enterobacteriaceae (E. coli, Klebsiella species, Proteus, Enterobacter, and Serratia) and Pseudomonas species are resistant (10). Azithromycin has greater activity than erythromycin against many anaerobes (10). C. trachomatis, N. gonorrhoeae, M. pneumoniae, and B. burgdorferi are all susceptible to azithromycin. T. pallidum and Toxoplasmosis gondii also appear to be susceptible (1,2 and 3,10,11). Azithromycin has excellent activity against H. ducreyi (2,3). Dosage and Route Azithromycin is available in an oral form as a 250-mg capsule. It is recommended that azithromycin be taken in the fasting state (i.e., 1 hour before or 2 hours after a meal). As noted already, azithromycin has the unique property of producing tissue levels that are 10- to 100-fold higher than serum levels (12,13). It is more stable than erythromycin at acid pH and thus is better absorbed from the gastrointestinal tract (37% vs. 25%). The half-life in tissue is 2 to 4 days, so 5 days of therapy (daily dosages) results in therapeutic tissue concentrations of azithromycin for 5 or more additional days after completion of therapy (12,13). In addition, azithromycin is highly concentrated in polymorphonuclear (PMN) leukocytes, and migration of PMN leukocytes to sites of infection may facilitate transport of this agent to the site of infection (12). Depending on the type of infection, azithromycin is given as single-dose therapy or as a 5-day regimen. For nongonococcal urethritis or chlamydial cervicitis, a single 1,000-mg dose is indicated. A single 2-g oral dose of azithromycin has been shown to be effective for uncomplicated gonorrhea (14). The 5-day regimen is 500 mg as a single dose on day 1, followed by 250 mg once daily on days 2 to 5 for a total dose of 1.5 g. This regimen is recommended for upper respiratory infections (e.g., pharyngitis, otitis media, and sinusitis), lower respiratory tract infections, susceptible skin and soft tissue infections (e.g., bronchitis and community-acquired pneumonia) (81–91), and pharyngitis or tonsillitis (second-line therapy) (1,2 and 3). Azithromycin is an effective alternative choice for the treatment of Lyme disease (3). Currently, azithromycin is not recommended for use in pregnant or lactating women. However, it should be safe in these circumstances, and use during pregnancy only awaits studies confirming safety in infants and fetuses. Side Effects Azithromycin appears to be well tolerated (15). It has fewer gastrointestinal side effects than erythromycin, but diarrhea, nausea, abdominal pain, and vomiting can occur. With the 1-g and particularly with the 2-g single dose, gastrointestinal side effects are more common. Mild headache or dizziness occurs in approximately 1% of patients. Allergic reactions are rare, as is superinfection with vaginitis. Liver function test results may be minimally elevated on occasion (1). Unlike erythromycin, azithromycin does not interact with theophylline or warfarin. However, it may interact in a life-threatening manner (like erythromycin) with terfenadine (Seldane) and astemizole (Hismanal), and these agents should not be used together
until clinical studies demonstrate lack of such an interaction (1). Cost Azithromycin is expensive (11). The average wholesale cost to the pharmacy for a 250-mg tablet is $8.13, with the single 1-g dose costing $35.52 and the 5-day 1.5-g total dose cost being $48.78 (1). Because only a few tablets are required, azithromycin remains relatively competitive with agents requiring a 10-day course. Use in Pregnancy and Placental Transfer Limited data currently exist as to the safety of azithromycin in pregnancy or during breast-feeding. Theoretically, this agent as a macrolide should be safe in these circumstances. The extent to which azithromycin crosses the placenta and the concentration of azithromycin in the fetus and amniotic fluid are unknown. Preliminary studies have confirmed the safety of azithromycin in pregnancy (16,17 and 18). Metabolism Most azithromycin remains unmetabolized in the body (12). Approximately 20% of the drug is excreted unchanged in the urine. Indications in Obstetrics and Gynecology Azithromycin is indicated for the treatment of chlamydial cervicitis and urethritis as a single 1-g dose (19). This single dose ensures compliance but is expensive. To date, studies have not been published assessing azithromycin in the treatment of acute PID. Unfortunately, the single 1-g dose of azithromycin is not sufficiently effective to recommend it for the treatment of uncomplicated gonorrhea (11,14). Treatment of mild community-acquired pneumonia and bronchitis in nonpregnant women with the 5-day regimen (total, 1.5 g) of azithromycin is better tolerated than erythromycin and is very effective. Use of azithromycin in obstetric patients awaits clinical studies confirming its safety in infants and fetuses. Several small studies suggest that azithromycin is an effective and safe treatment agent for chlamydial infection in pregnancy (16,17 and 18). Azithromycin is an alternative agent for Mycobacterium avium and toxoplasmosis in patients with acquired immunodeficiency syndrome (AIDS) (Chapter 10, Acquired Immunodeficiency Syndrome [AIDS]). A single 1-g dose of azithromycin is one of the recommended treatments for chancroid (19). Clarithromycin Clarithromycin (Biaxin) contains an O-methyl substitution at position 6 of the macrolide ring. Except for enhanced activity against H. influenzae, its spectrum of activity is similar
to that of erythromycin. However, it has better pharmacokinetic properties, resulting in a twice-daily dose regimen (1,2 and 3,20,21). Spectrum of Activity Like erythromycin and azithromycin, clarithromycin is a broad-spectrum antimicrobial that is active against Gram-positive and some Gram-negative bacteria, chlamydiae, mycoplasmas, and some mycobacteria (Table 23.10). Clarithromycin is inactive against MRSA. Staphylococci and streptococci resistant to erythromycin are also resistant to clarithromycin. It has only modest activity against anaerobes, and Enterobacteriaceae are resistant. Clarithromycin is more active than azithromycin or erythromycin against L. pneumophila and C. pneumoniae (2,3). Clarithromycin has excellent activity against Helicobacter pylori, the etiologic agent of peptic ulcer disease (2,3). Dosage and Route Clarithromycin is available only as an oral agent. Dosing depends on the pathogen and type of infection. For pharyngitis or tonsillitis, the dosage is 250 mg every 12 hours for 10 days, and for sinusitis, it is 500 mg every 12 hours for 14 days. Acute exacerbation of chronic bronchitis or pneumonia due to S. pneumoniae is treated with 250 mg every 12 hours for 7 to 14 days. Bronchitis due to H. influenzae requires 500 mg every 12 hours for 7 to 14 days. M. pneumoniae pneumonia is treated with 250 mg every 12 hours for 7 to 14 days. Side Effects Clarithromycin appears to be well tolerated and safe (1). In preclinical trials, only 1% of patients had severe side effects, most of which were gastrointestinal (17). No significant hematologic, hepatic, or renal toxicity was reported (17). The most common side effect is gastrointestinal symptoms, but these are much less frequent than with erythromycin. Like azithromycin and erythromycin, clarithromycin should not be given concomitantly with terfenadine (Seldane) or astemizole (Hismanal). Other medications such as theophylline, carbamazepine, warfarin, triazolam, ergots, and cyclosporine should be used cautiously when given concomitantly with clarithromycin (2). Cost Clarithromycin is more expensive than erythromycin. A 10-day course of 250 mg orally twice daily or 500 mg orally twice daily costs about $50 wholesale (1). Use in Pregnancy and Placental Transfer There are no adequate and well-controlled studies of clarithromycin in pregnant women. Thus, it should be used in pregnancy only if the potential benefit justifies the potential risk to the fetus. Clarithromycin, in high doses, during pregnancy has caused cardiovascular anomalies in rats, cleft palates in mice, and fetal growth retardation in
monkeys (11). The most promising use for clarithromycin appears to be in the therapy for M. avium complex in patients with AIDS (2,3,11). Clarithromycin (500 mg orally twice daily), in addition to ethambutol, is the drug of choice for this entity. In patients with AIDS with CD4 counts of less than 100 cells/mm3, prophylaxis of disseminated M. avium complex infection with clarithromycin (500 mg once or twice daily) is effective (22). Metabolism Clarithromycin is extensively metabolized in the liver. Indications In Obstetrics and Gynecology The role of clarithromycin in obstetrics and gynecology awaits future studies. For most infections, it is an expensive alternative, and whether its convenient twice-daily dosing offsets this cost is unclear. In pregnant and nonpregnant patients with AIDS with M. avium complex disease, clarithromycin may be an appropriate choice of therapy.
AMINOGLYCOSIDES The aminoglycoside-aminocyclitol antibiotics include streptomycin, kanamycin (Kantrex), gentamicin (Garamycin), tobramycin (Nebcin), amikacin (Amikin), and netilmicin. Streptomycin now has little use, aside from the treatment of tuberculosis, and kanamycin has largely been replaced by newer members of this group. The remaining agents are the most effective against Gram-negative aerobic organisms, but they have a relatively narrow margin of safety (1). Spectrum of Activity The aminoglycoside antibiotics are bactericidal agents that induce defective protein molecules by inhibiting the 30S subunit of the ribosome. Resistance to aminoglycosides may develop because (a) penetration of the antibiotic into the bacterial cell is prevented, (b) the bacterial ribosomes are modified to prevent binding of the aminoglycosides, and (c) bacterial enzymes destroy the antibiotics. Gentamicin is the most widely used member of this group in obstetric-gynecologic practice. It is active against nearly all strains of aerobic Gram-negative bacteria, including E. coli, Klebsiella species, indole-positive and indole-negative Proteus species, Enterobacter species, and P. aeruginosa. Widespread use has led to the development of strains of P. aeruginosa that are totally resistant to gentamicin. In view of the rarity of infections due to P. aeruginosa and other organisms uniquely susceptible to gentamicin (particularly in obstetric and gynecologic patients), less toxic antibiotics can often be used. Moreover, the introduction of new broad-spectrum agents that are
safer, such as third-generation cephalosporins and b-lactam agents plus enzyme blockers, has reduced the need for and use of aminoglycosides. Gentamicin acts synergistically with the penicillins against enterococci. The antibacterial spectrum of tobramycin is similar to that of gentamicin, with two exceptions. Many strains of P. aeruginosa that are resistant to gentamicin are susceptible to tobramycin. On the other hand, S. marcescens is significantly more susceptible to gentamicin than tobramycin. Susceptible organisms have MIC values of less than 4 µg/mL of gentamicin or tobramycin. Amikacin is a more recently introduced aminoglycoside antibiotic with activity against a wide range of Gram-negative bacteria, including P. aeruginosa. Its use is restricted to the treatment of infections produced by organisms resistant to gentamicin and tobramycin. Spectinomycin is an aminocyclitol antibiotic and not a true aminoglycoside. The primary clinical indication for spectinomycin has been in the treatment of N. gonorrhoeae, particularly in pregnant, penicillin-allergic patients, and treatment of uncomplicated anogenital gonorrhea due to PPNG strains. However, the emergence of PPNG strains resistant to spectinomycin and the availability of ceftriaxone, which is cheaper and also effective against pharyngeal gonorrhea, have limited its use. When used for the treatment of uncomplicated gonorrhea, 2 g intramuscularly as a single dose is recommended. Dosage and Route Because their principal toxic effects are dose related, dosages of aminoglycoside antibiotics should be based on the patient's weight. For gentamicin and tobramycin, the usual dosage has been 1.0 to 1.5 mg/kg of body weight every 8 hours, and for amikacin, either 7.5 mg/kg every 12 hours or 5.0 mg/kg every 8 hours. These preparations may be administered intravenously or intramuscularly in the same dose. These are approximations only, and blood levels should be determined, particularly in patients with courses longer than 10 days, with renal disease, with use of other nephrotoxic agents, with marked obesity, or with poor clinical response. In addition, many authors recommend that peak and trough levels be obtained after 24 hours of therapy. If these values are in the recommended range, then levels should be repeated every 3 to 4 days. Table 23.11 lists the expected peaks and troughs of the aminoglycosides. In the last 5 years, interest has focused on once-daily dosing with aminoglycosides. The rationale for this approach is (a) efficacy is comparable to every-8-hour dosing regimens, (b) nephrotoxicity and ototoxicity are less than with once-daily dosing, (c) aminoglycosides have a postantibiotic effect on aerobic Gram-negative bacteria, and (d) cost of administration is reduced.
TABLE 23.11. EXPECTED PEAKS AND TROUGHS OF THE AMINOGLYCOSIDES AFTER A TYPICAL 8-HOUR DOSE
After a 30-minute intravenous infusion of a typical 8-hourly dose, a concentration of gentamicin or tobramycin of 6 to 10 µg/mL is achieved rapidly, and concentrations fall to 1 to 2 µg/mL within 6 to 8 hours. For once-daily gentamicin therapy dosing, the usual dosage is 4.5 to 7 mg/kg every 24 hours by the intravenous route. It may also be given intramuscularly in selected circumstances. The dosing interval is increased if creatinine clearance initially is less than 60 mL per minute. For monitoring, a single serum level is obtained between 6 and 14 hours after the first dose, and the level is checked on a nomogram. The level is checked on the nomogram and adjustments in the dosing interval are made if necessary. If therapy is continued for more than 4 days, a second sample is obtained. Pregnancy is considered an exclusion for 24-hour dosing because of the rapid glomerular filtration rate. In the puerperium, however, 24-hour dosing appears to be as safe and effective as 8-hour dosing (2,3). Aminoglycosides are not absorbed after oral administration; oral preparations are intended only for bowel sterilization. Determining the proper dose for very obese patients has been a problem (4,5,6,7 and 8). Until recently, it has been thought that aminoglycosides were distributed only in extracellular fluid and that dosages should, therefore, be calculated on lean body weight. Accordingly, when obese patients were given an aminoglycoside dose based on total body weight, serum levels were higher than expected. Very recently, however, it has been recognized that obese patients, given a dose based on lean body weight, may have seriously low antibiotic levels, because there is partial distribution into adipose tissue. Based on recent pharmacokinetic study, it seems best to determine aminoglycoside doses in obese nonpregnant patients as follows: 1. For mildly obese patients (i.e., those with less than 30% excess weight), use total body weight to calculate dose.
2. For moderately obese patients (i.e., those with more than 30% excess weight), use lean body weight plus 40% of weight of adipose tissue. 3. For severely obese patients, do not exceed 150 mg every 8 hours for initial doses. 4. For moderately and severely obese patients, obtain antibiotic levels and adjust dose accordingly. Pregnancy and the immediate puerperium represent additional situations when it is difficult to estimate the correct dose. Duff et al. (4) and Blanco et al. (5) found “subtherapeutic” peak levels (i.e., less than 5 µg/mL) in approximately one third of postpartum women given 3.0 mg of gentamicin per kilogram of body weight per day and 4.5 mg of tobramycin per kilogram of body weight per day (4,5). Zaske et al. (6) found it necessary to use 3.0 to 11.6 mg/kg per day (in three divided doses) in pregnant women to achieve proper levels in 77 puerperal patients (6). These investigators emphasized the wide interpatient variation and the need to measure serum levels before using these increased doses. A poor correlation was found between elimination rate of gentamicin and creatinine clearance, but a high correlation was noted between distribution volume and gentamicin elimination. The latter point suggested that changes in fluid volumes in pregnancy affected gentamicin kinetics. Zaske et al. (7) also found widely ranging needs for gentamicin in gynecologic patients. In 249 patients with normal renal function (serum creatinine, less than 1.5 mg/dL), the daily dose ranged from 1.9 to 14.0 mg/kg (in divided doses) (7). Aminoglycosides are excreted by the kidney and must be used with caution in patients with renal disease. In such persons, adjustments in the dose of this antibiotic must be made based on serum concentrations. Side Effects The principal toxic effects of all aminoglycoside agents are ototoxicity and nephrotoxicity. A relationship between gentamicin dose and the development of ototoxicity has been noted. Significant factors correlating with ototoxicity are length of therapy, bacteremia, and higher temperature. Ototoxicity related to aminoglycoside administration may involve either the auditory or the vestibular portion of the eighth nerve. The ototoxicity may be unilateral or bilateral and is nearly always irreversible. It also tends to occur late in the course of therapy and may even develop or progress after the antibiotic has been stopped. Patients at high risk for ototoxicity are those who have received a high cumulative dose or a protracted course of aminoglycosides. Ototoxicity defined as greater than 15-dB loss of hearing has been reported in about 20% of patients and is detected clinically in about 2%. Nephrotoxicity develops in approximately 2% of the patients who receive gentamicin. It ranges from minimal abnormalities of the urine sediment to frank renal failure. In most patients who develop gentamicin-induced nephrotoxicity, renal function gradually returns to normal after the drug has been discontinued. Neuromuscular blockade is a rare complication of all aminoglycosides. It is dose related and is reversed by anticholinesterases and calcium salts.
Tobramycin possesses the potential to produce both ototoxicity and nephrotoxicity, but a number of recent studies have suggested that the incidence of such untoward effects may be lower than with gentamicin. However, most obstetric patients are at low risk for aminoglycoside toxicity. Accordingly, potentially less nephrotoxicity is not a compelling reason to abandon gentamicin. Cost The price of gentamicin has fallen dramatically. Amikacin is the most expensive aminoglycoside. Tobramycin is intermediate in cost. Use in Pregnancy More than 35 years ago, hearing loss was reported in children born to women who received streptomycin in pregnancy for the treatment of tuberculosis. Data are not available for other aminoglycosides, but the potential exists for fetal eighth nerve or renal toxicity. Nevertheless, pregnancy should not be considered a contraindication to the use of aminoglycosides. Concentrations of aminoglycosides in pregnancy are about 25% lower than expected, probably because of more rapid renal clearance (8). Care should be taken to avoid overdosing. If a pregnant patient does not respond to an aminoglycoside despite in vitro susceptibility, the antibiotic concentration should be determined. Metabolism As noted already, aminoglycoside antibiotics are excreted mainly by the kidneys, with a small amount being excreted in the bile. Indications in Obstetrics and Gynecology Because of their activity against aerobic Gram-negative organisms, aminoglycoside antibiotics are widely used in genitourinary infections, usually in combination with other antibiotics. Soft Tissue Pelvic Infections In the treatment of serious soft tissue pelvic infections, the combination of an aminoglycoside with other agents such as clindamycin (or less often, metronidazole) is usually considered the standard. Yet, aerobic Gram-negative organisms are isolated in perhaps only 25% of genital infections, and many of these are susceptible to less toxic antibiotics. Thus, even in serious infections, therapy may be modified on the basis of culture results and clinical response. Because of the potential toxicity, the difficulty in achieving the proper level of aminoglycosides, and the expense and toxicity of combination regimens, there has been a major stimulus to the search for newer, safe
single agents for therapy. Urinary Tract Infection Although nearly all organisms causing UTI (except enterococci) are susceptible to gentamicin, other agents are usually indicated. Aminoglycosides should be used to treat infections caused by organisms with in vitro resistance to less toxic agents. Aminoglycoside antibiotics may also be used in the initial treatment (before cultures are available) of recurrent infection when highly resistant organisms are likely. Similarly, an aminoglycoside can be added to ampicillin or cefazolin for initial empiric therapy of acute pyelonephritis.
METRONIDAZOLE Metronidazole is a nitroimidazole drug that was initially introduced for the treatment of trichomoniasis. Subsequently, it was used extensively for infections due to Giardia lamblia and Entamoeba histolytica. Metronidazole is widely recognized as an effective antimicrobial for the treatment of anaerobic infections and bacterial vaginosis (1,2). Spectrum of Activity The mode of action of metronidazole is believed that the biologic activity of the drug is related to reduction of the nitro group at the S position on the imidazole ring. This mode of action of metronidazole on anaerobic microorganisms requires four steps: (a) entry into the microorganisms, (b) reductive activation of the agent, (c) toxic effect of the reduced products or microorganisms, and (d) release of inactivated end products. The mechanism of the toxic action exerted by the reduced derivatives of metronidazole is assumed to be due to unstable intermediate products. N-(-2-hydroxyethyl)-oxalic acid and acetamide form when metronidazole is reduced. It is possible that these partially reduced intermediates are also responsible for the mutagenicity and carcinogenicity associated with metronidazole. The intracellular target of the toxic action is unclear but believed to be an interaction with DNA or a degradation of DNA. This causes extensive damage to nucleic acids and destroys the organism. Metronidazole is active against the anaerobic protozoa Trichomonas vaginalis, E. histolytica, and G. lamblia. Metronidazole resistance in strains of T. vaginalis has been rarely reported (3). Usually, this is relative resistance requiring higher doses of metronidazole. Metronidazole is active against only those bacteria with primarily anaerobic metabolism (obligate anaerobes). Of the Gram-negative anaerobic bacteria, Fusobacterium and Bacteroides species, particularly B. fragilis and Bacteroides melaninogenicus, are the most susceptible. Among the Gram-positive anaerobes, the Clostridium species are the most sensitive; Peptostreptococcus and Eubacterium species are also frequently sensitive. However, Actinomyces and Bifidobacterium species are less commonly sensitive. Metronidazole has no significant activity against aerobes or facultative anaerobes. In addition, it has less activity against microaerophilic streptococci. Other microaerophilic bacteria such as Campylobacter fetus and
Gardnerella vaginalis are susceptible to metronidazole. Dosage and Route Metronidazole may be administered intravenously, orally, rectally, or vaginally. With oral administration, metronidazole should be taken just after or during a meal. Oral doses range from 2 to 4 g as a single dose or 250 to 750 mg three to four times a day. For treatment of trichomoniasis, the usual recommended dosage is either a single 2-g regimen or 250 mg three times a day for 7 to 10 days. For giardiasis, the dose is 250 mg three times a day for 7 days; for nondysenteric amebiasis, 500 mg three times a day for 10 days; and for dysenteric amebiasis or amebic liver abscess, 750 mg three times a day for 10 days. If oral dosing is used for the treatment of anaerobic soft tissue infections, 500 mg three to four times a day is recommended. In bacterial vaginosis, the recommended dosage is 500 mg twice a day for 7 days. Intravenously administered metronidazole is available as the hydrochloride salt in 100-mL bottles and buffered with sodium bicarbonate. It can be infused over 20 to 30 minutes. The dosage varies from 250 to 750 mg three to four times a day. In general, intravenous metronidazole is reserved for the treatment of moderate to severe anaerobic infections, and the dose is determined by the severity of the clinical infection. Intravenous administration of 2 to 4 g of metronidazole has been used on occasion to treat resistant T. vaginalis. In Europe, rectal administration of a 1-g metronidazole suppository has been used in a dosage regimen of 1 g every 8 hours. Metronidazole vaginal gel (0.75%) is available for the treatment of bacterial vaginosis. Side Effects When metronidazole has been used in relatively low doses for short periods, very few side effects occur. However, with administration of higher doses over a more prolonged period, side effects occur more frequently. The most common side effects are gastrointestinal disturbances such as nausea, an unpleasant metallic taste, a furred tongue, and abdominal cramps (1,2). Metronidazole has an Antabuse-like effect, so alcoholic beverages should be avoided while taking metronidazole. CNS symptoms with lower doses include headache, ataxia, vertigo, sleepiness, and depression. With very large doses, reversible peripheral neuropathies have occurred, as well as myalgia, transient encephalopathies, and persistent paresthesia. Other side effects noted even with low-dose therapy are vaginal burning, a disulfiram-like intolerance to alcohol, and the presence of dark red-brown urine. Transitory rashes and leukopenia have also occurred. Concern had been voiced about the safety of metronidazole because of reports suggesting that it may have mutagenic, carcinogenic, or teratogenic properties. Several reviews have given evidence putting aside concerns of metronidazole induced
teratogenesis (4,5). Recently, attention has been focused on the interaction of metronidazole with other drugs. The concomitant administration of metronidazole augments the hypoprothrombinemic effect of warfarin (Coumadin). On the other hand, diphenylhydantoin (Dilantin) and phenobarbital increase the metabolism of metronidazole. Cost Oral metronidazole is not under patent protection any longer and is thus relatively inexpensive. Intravenous metronidazole is a costly preparation. Use in Pregnancy and Placental Transfer Metronidazole passes across the placental barrier and can be found in fetal tissue, cord blood, and amniotic fluid in high concentrations. As described already, metronidazole has been used extensively during pregnancy without apparent ill effects (5). In previous editions of the text, we had voiced the precaution that metronidazole should be avoided in the first trimester of pregnancy and that alternative agents should be preferred in the second and third trimesters of pregnancy. Some texts (6) still note that manufacturers do not recommend that metronidazole be used in the first trimester of pregnancy or during lactation and that if the drug is used in the second or third trimester, a large single-dose treatment should be avoided. However, in view of recent reassuring reviews, as noted already, metronidazole may be used without regard to adverse effects to the pregnancy in the second or third trimester. It is also noted that in the 1998 STD treatment guidelines, metronidazole use in pregnant patients as a 2-gm and a single dose is recommended for the treatment of trichomoniasis, without any restriction about use in the first trimester (7). Metronidazole is excreted in breast milk and is present in breast milk in levels comparable to those in serum (8). Thus, it has been suggested that either lactating women should not receive metronidazole or that breast-feeding should be temporarily discontinued for 24 hours after a single oral dose of metronidazole. Metabolism Metronidazole is almost completely absorbed after oral administration and diffuses well into nearly all tissues, resulting in wide distribution throughout the body. After single oral doses of 250 and 500 mg, peak serum levels of 6 and 12 µg, respectively, have been reported. Oral 500-µg doses of metronidazole administered four times daily resulted in peak serum levels of 20 to 50 µg/mL. The peak blood levels achieved with intravenous metronidazole approximate those of the oral route. After inserting a 1-g suppository, a mean serum level of 2.3 µg/mL was detected at 1 hour and 10.5 µg/mL at 4 hours. The serum half-life of metronidazole is about 8 hours. Metronidazole binds to plasma protein in the range of 20%. In humans, it is primarily
eliminated through metabolism, which occurs mainly in the liver via oxidation, hydroxylation, or conjugation of side chains on the imidazole rings. Metronidazole is excreted primarily via the kidneys but is found in feces as well. Indications in Obstetrics and Gynecology Metronidazole is a potent antimicrobial agent against anaerobic protozoa and bacteria. In the United States, metronidazole is the only effective drug available for the treatment of T. vaginalis. Outside the United States, other nitroimidazoles are available. A 2-g single dose and 250 mg three times a day for 7 days are very effective. The FDA has approved a preparation called Flagyl 375 (375 mg twice a day for 7 days) for the treatment of trichomoniasis. Approval is on the basis of pharmacokinetic equivalency of this regimen with metronidazole (250 mg three times a day for 7 days). No clinical data are available to demonstrate the clinical equivalency of these regimens (7). Metronidazole is used for the treatment of amebiasis and giardiasis. It is the drug of choice for amebiasis, and an alternative therapy in giardiasis, where quinacrine is the drug of choice. Bacterial vaginosis is a synergistic polymicrobial infection associated with G. vaginalis and anaerobic bacteria, particularly Peptococcus and Bacteroides species. Metronidazole (500 mg orally twice daily for 5 days) or metronidazole vaginal gel 0.75% (one application once or twice daily) is effective therapy for bacterial vaginosis. The important role of anaerobic bacteria in soft tissue infections of the upper genital tract of women is well recognized, with anaerobes recovered from approximately two thirds of pelvic infections. With the recognition of the excellent bactericidal activity of metronidazole against anaerobic bacteria, it has been used widely as a therapeutic agent in the treatment of infections associated with anaerobic bacteria. For this type of infection, 250 to 750 mg three to four times a day is the recommended dosage. Because these infections are generally mixed infections involving facultative bacteria and anaerobes, metronidazole must be used in combination with an agent effective against Gram-negative facultative bacteria, such as an aminoglycoside. However, this combination does not provide activity against facultative streptococci such as group B streptococci, which are common pathogens in the obstetrics and gynecology. Thus, an agent such as ampicillin or penicillin must often be added. Alternative regimens with metronidazole include combining it with second- or third-generation cephalosporins or monobactam antimicrobial agents. However, this results in a rather expensive combination. In the treatment of intraabdominal and pelvic infections, therapeutic trials comparing a metronidazole combination with a clindamycin combination have failed to demonstrate an enhanced clinical efficacy for metronidazole (9,10). Because metronidazole has not resulted in an enhanced clinical efficacy over current standard therapies and because concern exists about its carcinogenic potential, metronidazole should not be considered a first-line antimicrobial agent for treating mixed aerobic-anaerobic pelvic infections. Rather, it should be used as a backup agent when other treatment regimens have failed or when anaerobes resistant to clindamycin, chloramphenicol, or cefoxitin are present. However, many authorities disagree and suggest that metronidazole is the drug of
choice for the treatment of anaerobic infections. A multitude of inexpensive antimicrobial agents are available for use as prophylactic antibiotics in surgical procedures. These agents are not required for treating severe infection and make ideal prophylaxis choices. It seems inappropriate to use metronidazole as a prophylactic antibiotic at this time (1).
FLUOROQUINOLONES During the past decade, the new quinolone antimicrobial agents have aroused great interest. The quinolone agents act on bacterial DNA gyrase, an enzyme that helps maintain the structure of DNA and thus inhibit DNA synthesis. Nalidixic acid (NegGram) was the first quinolone introduced into clinical practice. Although it possesses in vitro activity against many Enterobacteriaceae, nalidixic acid is weakly active against Pseudomonas species and Gram-positive organisms. In addition, nalidixic acid achieved low serum levels after oral administration, had poor tissue penetration, led to rapid development of resistance during therapy, and led frequently to adverse CNS effects (1,2,3,4 and 5). Thus, its use has been limited to treatment of UTIs (1,2). In an attempt to expand the spectrum of activity of nalidixic acid and the clinical usefulness of quinolones, several derivatives of nalidixic acid were developed. However, cinoxacin (Cinobac) and oxolinic acid (Utibid) did not provide any real clinical advantages over nalidixic acid and were also classified as urinary tract antiseptics (6). Over the past 15 years, various fluorinated derivatives of nalidixic acid have been developed and investigated. Addition of a fluorine molecule at position 6 of the basic quinolone nucleus enhanced the spectrum of quinolone activity 1,000-fold and increased penetration into bacterial cells (3,7). Five of these fluoroquinolones—norfloxacin (Noroxin), ciprofloxacin (Cipro), ofloxacin (Floxin), lomefloxacin (Maxaquin), and enoxacin (Penetrex)—became clinically available by the early 1990s. As the new millennium commenced, six new fluoroquinolones—levofloxacin, sparfloxacin, grepafloxacin, trovafloxacin, moxifloxacin, and gatifloxacin—have been approved for clinical use in the United States (3,4). Additional new fluoroquinolone agents such as clinafloxacin, temafloxacin, and tosufloxacin should soon be available. These compounds have a markedly expanded spectrum of activity compared with nalidixic acid. In addition, their pharmacokinetic characteristics allow their use in various systemic infections. Spectrum of Activity The fluoroquinolones, like nalidixic acid, kill susceptible bacteria by inhibiting DNA synthesis secondary to a direct effect on DNA gyrase (1). The spectrum of activity of the fluoroquinolones is wide. As shown in Table 23.12, these new fluoroquinolones are much more active than nalidixic acid.
TABLE 23.12. IN VITRO ACTIVITY OF THE QUINOLONE ANTIBIOTICS: MIC90 (µg/mL)
Enterobacteriaceae are highly susceptible to the fluoroquinolones with MIC values of 0.5 µg/mL or less. In addition, these compounds are active against P. aeruginosa. The fluoroquinolones also have excellent activity against other Gram-negative bacteria such as H. influenzae and N. gonorrhoeae including PPNG strains. Gastrointestinal pathogens such as Shigella species, Salmonella species, Campylobacter jejuni, Vibrio species, and Yersinia enterocolitica are highly susceptible to these new fluoroquinolones (1,6,9,10 and 11). The older fluoroquinolones—norfloxacin, ciprofloxacin, and ofloxacin—are active primarily against a wide variety of aerobic Gram-negative bacteria, including Enterobacteriaceae, H. influenzae, Neisseria species, Campylobacter species, and M. catarrhalis (3). These older agents also have activity against P. aeruginosa, with ciprofloxacin being the most active (8). However, an increasing number of studies have demonstrated higher MIC levels at or above susceptibility breakpoints of the older fluoroquinolones of Klebsiella species, Enterobacter species, S. marcescens, and E. coli (8). Although the newer fluoroquinolones with enhanced activity against Gram-positive bacteria have maintained their excellent Gram-negative aerobic coverage, they have lost activity against P. aeruginosa (3). The MIC values against Gram-positive bacteria are generally higher than those against Gram-negative organisms. The older fluoroquinolones have relatively high MIC values against streptococci (1,6,9). However, streptococci and enterococci are susceptible at levels achievable in the urine. Although the older fluoroquinolones have only modest activity against Gram-positive bacteria, the newer fluoroquinolone agents have enhanced activity against these agents (3,4). Thus, newer fluoroquinolones such as grepafloxacin, sparfloxacin, trovafloxacin, gatifloxacin, and moxifloxacin have MIC90 values ranging from 0.25 to 0.5 µg/mL against methicillin-susceptible S. aureus (8). Levofloxacin is somewhat less active. Gatifloxacin, sparfloxacin, and trovafloxacin also have some activity against MRSA (Table 23.12). Other staphylococci species (e.g., Staphylococcus saprophyticus and S. epidermidis) are also susceptible to the newer
agents (8). The newer fluoroquinolones all are active against S. pneumoniae (8). Most importantly, the MIC values of the new fluoroquinolones are similar for both penicillin-susceptible and penicillin-resistant strains (12,13). As a result, the newer fluoroquinolones have become popular agents for the treatment of community-acquired lower respiratory tract infections (bronchitis and pneumonia) (3). However, Chen et al. (14) recently reported the occurrence of decreased susceptibility of pneumococci to fluoroquinolones between 1988 and 1998 in association with increased prescribing of these agents. Close monitoring of this trend will be required. The newer fluoroquinolones are also active against S. pyogenes (group A b-hemolytic streptococci) and Streptococcus agalactiae (group B b-hemolytic streptococci). On the other hand, activity against enterococci remains variable even among the newer fluoroquinolones, with sparfloxacin and gatifloxacin being most active (3). In general, the MIC values against anaerobic bacteria for the older fluoroquinolones are even higher than those against Gram-positive bacteria (6,9,10 and 11). Thus, the older quinolones have limited anaerobic activity. Among the newer fluoroquinolones, only trovafloxacin has been approved by the FDA for the treatment of anaerobic infections. Although gatifloxacin, moxifloxacin, sparfloxacin, and clinafloxacin have demonstrated activity against anaerobic bacteria in vitro, few clinical treatment trials have been reported (3,4,15). The older fluoroquinolones have demonstrated variable in vitro activity against C. trachomatis (7,8 and 9). Only ofloxacin is recommended for the treatment of C. trachomatis infections among the older quinolones (8). In addition, these fluoroquinolones have activity against other intracellular organisms such as L. pneumophila, U. urealyticum, M. hominis, M. pneumoniae, M. avium, and Mycobacterium intracellulare. All of the newer fluoroquinolones have activity against C. trachomatis, C. pneumoniae, U. urealyticum, M. hominis, and M. pneumoniae (3,4,8). All of the newer agents also have excellent activity against L. pneumophila (19). Unfortunately, the initial view that bacterial resistance to the fluoroquinolones would not be a significant problem has proven to be unfounded. Clinically important resistance to the fluoroquinolones is an increasing problem with S. aureus, P. aeruginosa, and S. marcescens (10,11,20,21). Nearly 80% of MRSA strains reported to the CDC were resistant to ciprofloxacin, with prior exposure to ciprofloxacin being a major risk factor (22). Similarly, increasing resistance (10% to 30%) is being reported with methicillin-susceptible S. aureus (10,11). The CDC has also reported the occurrence of decreased susceptibility of N. gonorrhoeae to the fluoroquinolones. Dosage and Route of Administration In addition to the increased bacterial spectrum provided by the fluoroquinolones, the pharmacokinetic characteristics of these new compounds have significantly enhanced their clinical usefulness. The pharmacokinetic characteristics of these agents are compared in Table 23.13. Norfloxacin has poor oral bioavailability, achieves lower peak concentrations than other quinolones, and is not useful for systemic infections. Other fluoroquinolones are well absorbed orally with a bioavailability of more than 50% for all agents and approaching 100% for several (Table 23.13) (4,25,26). Although norfloxacin
and ciprofloxacin undergo hepatic metabolism and are eliminated by both hepatic and renal mechanisms (27), ofloxacin and lomefloxacin are metabolized to a lesser degree and are primarily excreted by the kidneys (26,28). Thus, ofloxacin and lomefloxacin achieve higher peak concentrations and larger areas under the time-concentration curve.
TABLE 23.13. COMPARISON OF PHARMACOKINETIC CHARACTERISTICS OF THE FLUOROQUINOLONES
Among the newer fluoroquinolones, levofloxacin and sparfloxacin are cleared by the kidneys and thus require dosing medication with impaired renal function (29). Hepatic metabolism and biliary excretion are the primary routes of elimination for sparfloxacin, grepafloxacin, and trovafloxacin (3,4). The pharmacokinetic characteristics of the newer fluoroquinolones are superior to those of the older agents (3). In general, their serum half-lives are longer, which permits once-daily dosing for most newer fluoroquinolones. As a consequence of once-daily dosing, higher peak levels are achieved, which results in maximal bacterial killing (3). The newer agents have an increased volume of distribution, which results in excellent tissue penetration (8). All the fluoroquinolones are bactericidal antimicrobial agents with a concentration-dependent killing effect (3). Thus, the high peak concentrations achieved by fluoroquinolones result in more rapid and complete killing of susceptible bacteria (3,8). All the fluoroquinolones demonstrate a postantibiotic effect, which is concentration dependent (8). Among the newer fluoroquinolones, this postantibiotic effect lasts for 1 to 6 hours, further enabling these agents to be dosed once daily (3). One of the major advantages of the fluoroquinolones is that they can be administered orally and are generally well absorbed after oral administration, reaching peak serum concentrations 1 to 2 hours after oral dosing. Maximum serum concentrations achieved after oral administration range from 1.1 to 6.4 µg/mL. These levels are higher than the MIC values of most organisms. Moreover, urinary levels of the fluoroquinolones are very high, up to 100 times the serum concentration. These new fluoroquinolones have low
protein binding (15% to 40% protein bound), have large volumes of distribution, and penetrate well into various body fluids and tissues (2,6). Concomitant administration of magnesium- or aluminum-containing antacids reduces the absorption of fluoroquinolones. Antacids should not be taken until at least 2 hours after fluoroquinolone administration (6). Sucralfate reduces the bioavailability of the fluoroquinolones by 85% to 90%; administration more than 6 hours before fluoroquinolones is necessary (6). Ciprofloxacin and enoxacin interfere with excretion of theophylline. Norfloxacin, ciprofloxacin, ofloxacin, enoxacin, lomefloxacin, pefloxacin, sparfloxacin, levofloxacin, grepafloxacin, moxifloxacin, and trovafloxacin are available in an oral form. Ciprofloxacin, ofloxacin, pefloxacin, levofloxacin, and trovafloxacin also are available as intravenous preparations. In general, these agents provide 1 µg/mL per 100 mg of antibiotic infused intravenously (11). Although the fluoroquinolones are potent broad-spectrum antibiotics, the Medical Letter only lists them as the drugs of choice for UTIs due to P. aeruginosa, Shigella infections, and C. jejuni infections (30). All of the fluoroquinolones are very effective against most bacteria causing UTIs (Table 23.12). However, it is recommended that the fluoroquinolones not be used for uncomplicated UTI (e.g., cystitis), an infection for which other less-expensive and effective agents are available (Chapter 15, Urinary Tract Infection). For complicated (pyelonephritis) or recurrent UTI, these agents are best reserved for the treatment of more resistant pathogens such as P. aeruginosa or other hospital-acquired Gram-negative aerobes. Among the newer fluoroquinolones, only levofloxacin is approved for the treatment of cystitis and pyelonephritis. The fluoroquinolones provide good to excellent activity against many STDs (Table 23.12) (3,4,31,32). All available fluoroquinolones have excellent activity against N. gonorrhoeae. A single oral dose of ciprofloxacin (500 mg) or ofloxacin (400 mg) is effective for the treatment of uncomplicated gonococcal cervicitis, urethritis, or rectal infection. Among the newer fluoroquinolones, trovafloxacin, and grepafloxacin are FDA approved for the treatment of gonococcal infection. However, trovafloxacin is no longer recommended due to its liver toxicity (3). None of the fluoroquinolones is effective in a single-dose regimen against C. trachomatis. Of the older fluoroquinolones, only ofloxacin (300 mg orally twice daily for 7 days) has proven consistent efficacy against C. trachomatis. All the newer fluoroquinolones have activity against C. trachomatis that is equivalent or superior to ofloxacin (3). The most recent CDC guidelines recommend that for outpatient treatment of PID, ofloxacin (400 mg twice daily orally for 14 days) in combination with clindamycin or metronidazole is an alternative for the ceftriaxone or cefoxitin-doxycycline regimen (31). Against H. ducreyi (chancroid), ciprofloxacin (500 mg orally twice daily for 3 days) is effective (32). All the new fluoroquinolones have excellent activity against H. ducreyi (3). Fluoroquinolones are useful agents for the treatment of pneumonia and bronchitis due to susceptible Gram-negative pathogens. However, the older agents should not be used for empiric therapy of common community-acquired pneumonias in which S. pneumoniae is the most common pathogen. As discussed already, the newer fluoroquinolones are commonly used agents for treating community-acquired bronchitis
and pneumonia. All the fluoroquinolones are excellent agents for the treatment of bacterial gastroenteritis (10,11). These agents are the drugs of choice for Shigella infections as a single oral dose or 3 days of therapy (30). These agents are all effective for the treatment of traveler's diarrhea as a 3- to 5-day course. The recommended doses are ciprofloxacin (500 mg every 12 hours), ofloxacin (300 mg every 12 hours), and norfloxacin (400 mg every 12 hours). The fluoroquinolones are useful agents in treating mixed soft tissue infections. They are effective against the Gram-negative aerobes involved in these mixed infections (e.g., cellulitis, wound infection, and ischemic ulcers), but an agent effective against anaerobes must be added with the older agents. Table 23.14 and Table 23.15 summarize the dose schedules of the fluoroquinolones. Intravenous therapy is reserved for seriously ill patients who are unable to take oral therapy.
TABLE 23.14. DOSING OF CIPROFLOXACIN AND OFLOXACINa
TABLE 23.15. DOSING OF NEWER FLUOROQUINOLONES
Side Effects In general, the fluoroquinolones are well tolerated and considered relatively safe compared with other commonly prescribed antibiotics (5,7). Overall adverse reactions among the older fluoroquinolones have been reported in 2% to 8% of patients (10,11,33). The adverse reactions are similar for the various older fluoroquinolones. The most common adverse effects associated with use of quinolones are gastrointestinal (nausea, vomiting, diarrhea, and anorexia) and CNS related (dizziness, headache, and insomnia). In addition, rash and transient elevation in liver enzymes may occur. Quinolones inhibit theophylline metabolism, and their concomitant administration leads to elevated serum theophylline levels. Adverse events ascribed to the use of fluoroquinolones can be either associated with the chemical and structural modifications made to improve these agents or unrelated to these changes (3). Crystalluria, phototoxicity, genetic toxicity, and drug interaction are toxicities related to chemical modifications, whereas gastrointestinal symptoms and chondrotoxicity/arthropathy are not (3,34). According to O'Donnell and Gelone (3), the most commonly reported adverse events with the use of fluoroquinolones are listed by organ system and their manifestations in Table 23.16. The relative frequencies among the fluoroquinolones for some of the clinically significant adverse events are shown in Table 23.17 (3).
TABLE 23.16. COMMON ADVERSE EVENTS ASSOCIATED WITH FLUOROQUINOLONES
TABLE 23.17. SUMMARIES OF ADVERSE EVENT: POTENTIAL ASSOCIATED WITH FLUOROQUINOLONES
Gastrointestinal side effects are the most frequent reported adverse effects associated with fluoroquinolone use (3,4 and 5,7) and have been reported in 0.8% to 11% of patients (5). Nausea is the most commonly reported gastrointestinal side effect (5). An unpleasant taste has been reported in 9% to 17% of patients receiving grepafloxacin (35). This side effect appears to be dose related. Neurotoxicity is the second most frequent category of adverse effects related to fluoroquinolones (3,4 and 5,7). Overall, CNS reactions have occurred in 0.9% to 11% of patients (3,4 and 5). Mild reactions such as headache, dizziness, tiredness, and sleeplessness are most commonly noted (3,4 and 5). These CNS side effects appear to be more commonly associated with ofloxacin and lomefloxacin than with other fluoroquinolones (5). Severe neurotoxicity is rare (less than 0.5%) and includes psychosis, hallucinations, depression, and seizures (5). These reactions usually commence a few days after beginning treatment with a fluoroquinolone and resolve when the agent is stopped (5). Neurotoxicity is more likely to occur in elderly patients, particularly those with significant arteriosclerosis, and in individuals with CNS impairments (5). It is recommended that fluoroquinolones should not be given to patients with a history of seizures (5). Allergic and skin reactions occurred in 0.4% to 2.2% of patients (3). Phototoxicity is a potentially significant side effect that must be considered with the clinical use of fluoroquinolones (3,4 and 5,7). Quinolone phototoxicity appears to be related to fluoridation at the 8 position and thus is more common and possibly more severe with lomefloxacin, fleroxacin, and sparfloxacin (5,36). Moderate to severe phototoxicity manifests as an exaggerated sunburn reaction with reddening, blistering, and subsequent peeling of skin (5). Most likely, phototoxicity is an idiosyncratic reaction, although the risk appears to increase with higher doses and longer courses of fluoroquinolone treatment (5,37). Thus, it is appropriate to warn all patients taking fluoroquinolones of the phototoxicity side effect. It can occur with both direct and indirect
sunlight. Both older and newer quinolones induce arthropathy with cartilage erosions and noninflammatory effusions in the weight-bearing joints of juvenile animals (3,4 and 5). It has been suggested that the pathogenic mechanism responsible for this arthropathy is chelation of magnesium by fluoroquinolones (38). As a consequence of these findings in juvenile animals, concern has been raised over potential cartilage toxicity in children and quinolones have not been recommended for routine pediatric use or for pregnant and lactating women (3,4 and 5). However, there is little evidence of quinolone-induced arthropathy in humans (3,4 and 5,36,39,40). More than 10,000 pediatric cases (many with cystic fibrosis) treated with fluoroquinolones have been reported (40). Concurrent joint disease has been explained in most instances by hyperimmune mechanisms of the so-called cystic fibrosis arthropathy or hypertrophic pulmonary osteoarthropathy. Among the pediatric cases treated with fluoroquinolones, the incidence of arthralgia was no greater than that expected as a result of cystic fibrosis (3). Moreover, none of the fluoroquinolones evaluated had negative effects on linear growth of children (41). Recently, Church et al. (42) assessed sequential ciprofloxacin therapy in 1,795 cases of pediatric cystic fibrosis and noted that there were no unequivocal cases of quinolone arthropathy. Burkhardt et al (43) recently reviewed the clinical treatment of more than 7,000 children and adolescents with fluoroquinolones and did not demonstrate any association with the development of arthropathy. This absence of human arthropathy with increasing exposure of pediatric patients to fluoroquinolones has led some authorities to suggest that in some children, particularly those with cystic fibrosis, the benefits of fluoroquinolones outweigh the small risk of joint toxicity (43,44). Although expanded pediatric use of fluoroquinolones is being considered (43,44), the quinolones are not currently recommended for routine use in children, have not been approved for pediatric use in the United States, and should not be given to nursing mothers (5). In addition, their safety in pregnancy remains to be established (7,45). Quinolones have also been reported to cause tendonitis and rupture of the Achilles tendon, with more than 200 cases reported in the literature (40). These complications have occurred even with short-term use. Most cases of tendonitis have been associated with pefloxacin with much lower incidences occurring with ciprofloxacin, ofloxacin, norfloxacin, and enoxacin (3). Post–marketing surveillance of trovafloxacin disclosed an unexpected rare occurrence of adverse hepatic reactions (46). From February 1998 through early May 1999, 2.5 million prescriptions for trovafloxacin were written in the United States and 140 patients were reported to have experienced a hepatic adverse event (incidence rate, 0.0056%). In 14 of the cases, the FDA determined that acute liver injury was strongly associated with concomitant administration of trovafloxacin (5,46). Of these cases, four patients required liver transplantation (one subsequently died) and five patients died. Although the hepatic reactions occurred between 1 and 60 days after commencing treatment, the risk of serious hepatic injury increases with more than 14 days of trovafloxacin therapy (5). As a result of these adverse events, it has been recommended that trovafloxacin use be limited to serious infections in hospitalized patients with concurrent monitoring of hepatic enzymes (46). This has limited the usefulness of trovafloxacin in obstetric and gynecologic patients, particularly for STDs and PID in which the drug held great
promise. Prolongation of the QT interval has been reported in patients receiving sparfloxacin and to a lesser degree grepafloxacin (47). Other fluoroquinolones have not been associated with this side effect. Hemolytic anemia associated with renal failure, coagulation disorders, or both have been rarely reported except with temafloxacin, which was removed from the market when post–marketing surveillance identified an incidence of 1 in 5,000 prescriptions (4). Ocular side effects of fluoroquinolones include blurred vision, diplopia, photophobia, abnormal accommodation, and changes in color perception (5). These symptoms resolve once therapy is halted (7). The safety of quinolones in pregnancy has not been established. However, preliminary reports of babies born to pregnant women exposed to norfloxacin or ciprofloxacin during the first trimester have not demonstrated any increased teratogenic risk (48,49). Teratogenicity studies have not identified gross structural defects such as limb reductions in association with fluoroquinolone use (35). Prenatally formed cartilage appears to be much less sensitive to fluoroquinolones than joint cartilage (5). Moreover, reproductive toxicity studies with ciprofloxacin in animals did not demonstrate any effect on fertility or on prenatal or postnatal development (36). Similarly, no embryonic or teratogenic effects were noted in animal studies (36). Certain precautions are advised with the use of fluoroquinolones (10,11,33,50). They should not be used in patients allergic to nalidixic acid. Because these agents produce cartilage erosions in young animals, the fluoroquinolones should not be used in children and are not recommended for use in pregnant or lactating women. The exact age above which quinolones are safe is debatable. The CDC recommended quinolone use in adolescents 17 years of age or older. One exception is cystic fibrosis in children, where the benefit of fluoroquinolones outweighs the potential risks. Significant interactions between the fluoroquinolones and other drugs have been reported (Table 23.18). Antacids with magnesium or aluminum markedly decrease oral absorption. Theophylline levels must be monitored, because its metabolism is diminished by fluoroquinolones. Ciprofloxacin and enoxacin also interfere with caffeine metabolism.
TABLE 23.18. FLUOROQUINOLONES AND INTERACTIONS WITH OTHER DRUGS
AND FOOD
Coadministration of drugs containing aluminum, magnesium, iron, and calcium may result in significant decreased bioavailability of fluoroquinolones (3). Thus, products containing any of these cations should be staggered, so the fluoroquinolone agent is administered 2 hours before or after the offending agent (3). Clearance of theophylline is reduced by enoxacin, ciprofloxacin, clinafloxacin (investigational), and grepafloxacin, resulting in significantly increased levels of theophylline (enoxacin, 84%; ciprofloxacin, 30%) (3). Acute theophylline toxicity may occur and is characterized by nausea, vomiting, and rarely seizures (3). These are expensive antibiotic agents and should be used only when clearly indicated. Moreover, inappropriate or excessive use of the fluoroquinolones is an emerging concern (10,11,33). In addition to the cost issue, their overuse is leading to resistance, particularly among S. aureus and P. aeruginosa. Cost The fluoroquinolones are significantly more expensive than trimethoprim-sulfamethoxazole, sulfisoxazole, or ampicillin for the treatment of acute UTIs. However, these agents are very cost-effective when used to treat UTIs due to resistant organisms such as P. aeruginosa that would otherwise require parenteral therapy. Similarly, despite its expense, ciprofloxacin is very cost-effective as an oral agent for the treatment of osteomyelitis due to susceptible organisms. As noted already, overuse of these agents must be avoided. Use in Pregnancy Animal studies have demonstrated that the quinolones are deposited in cartilage and result in irreversible arthropathy (6). Moreover, these agents work by inhibiting DNA synthesis. Thus, these agents are not recommended for use in pregnant or lactating women. However, clinical studies have not confirmed such findings (48,49). Indications in Obstetrics and Gynecology To date, the fluoroquinolones have found their greatest use in treating uncomplicated gonorrhea, UTIs (particularly those due to multidrug-resistant Gram-negative organisms), and diarrheal diseases due to Shigella species, Salmonella species, or C. jejuni (3,4 and 5,51,52,53 and 54). Ofloxacin is effective for the treatment of chlamydial cervicitis and urethritis. The fluoroquinolones have undergone clinical investigation in the treatment of acute PID. Crombleholme et al. (55) reported that ciprofloxacin (300 mg intravenously every 8 hours) for the treatment of acute PID resulted in clinical cure in 31 (94%) of 33 patients with acute PID, compared with 34 (95%) of 35 patients treated with clindamycin-gentamicin. However, as a single agent, ciprofloxacin was much less
effective than the clindamycin regimen in eradicating anaerobic bacteria from the endometrial cavities of these patients. The clinical significance of this finding is not clear, but such a finding suggests that ciprofloxacin may not be useful as a single agent for the treatment of acute PID. Ofloxacin as a single agent for the treatment of PID has been clinically effective, particularly when N. gonorrhoeae is the predominant organism (56,57). The lack of anaerobic coverage is problematic though. As noted already, the fluoroquinolones are not recommended for use in pregnant or lactating women.
VANCOMYCIN Vancomycin (Vancocin) is a glycopolypeptide that is unrelated to any of the other antimicrobial agents. It has a narrow spectrum of activity and is bactericidal. Vancomycin inhibits synthesis and assembly of cell wall peptidoglycan polymers. Until recently, vancomycin was limited to use as an alternative agent against enterococci and penicillin-resistant staphylococci in penicillin-allergic patients. However, the emergence of methicillin-resistant staphylococci and antibiotic-associated colitis due to C. difficile has led to a dramatically increased clinical role for vancomycin, which is the drug of choice for the former and an alternative for the latter condition (1,2). Both S. aureus and S. epidermidis are susceptible to vancomycin, with MIC values ranging from 1 to 5 µg/mL. Similarly, group A and group B b-hemolytic streptococci and S. pneumoniae are highly susceptible. Generally, the enterococcus is inhibited by concentrations of vancomycin that are obtainable in the serum with parenteral therapy. Since the mid-1980s, however, acquired vancomycin resistance in enterococci has begun to appear (1). According to data from the CDC, vancomycin resistance has increased more than 20-fold, from less than 0.5% in 1989 to more than 10% in 1995. Fortunately, from the viewpoint of obstetrics and gynecology, nearly all of the vancomycin-resistant enterococci are species other than Streptococcus faecalis. In other words, they are enterococcal species that rarely occur in obstetric and gynecologic patients. Nevertheless, the concern of vancomycin-resistant enterococci heightens overall concern regarding the development of antibiotic-resistant bacteria. C. difficile and C. perfringens are usually susceptible. All strains of methicillin-resistant staphylococci are susceptible to low concentrations of vancomycin, although some strains have demonstrated tolerance to its bactericidal action (1). Vancomycin is given intravenously in 100 to 250 mL of 5% dextrose in water or 0.9% saline over 30 to 60 minutes. The usual intravenous dose of vancomycin is 1 g every 12 hours or 500 mg every 6 hours. The dose must be reduced in the presence of renal disease. For the treatment of C. difficile enterocolitis, an oral dose of 125 to 500 mg every 6 hours is recommended. Rapid or bolus administration is dangerous and contraindicated; it can result in extreme flushing and anaphylactic reactions. Vancomycin is poorly absorbed orally and is used by the oral route only for the treatment of enterocolitis. After intravenous administration of a 500-mg dose, peak serum levels of 50 µg/mL are obtained, and serum levels of 6 to 10 µg/mL occur at 1 to 2 hours. Vancomycin is excreted almost entirely via the kidney, and 80% to 90% of the drug appears in the urine within 24 hours. From 10% to 55% of the drug is protein bound. The serum half-life is about 6 hours. With renal failure, the dosage must be
reduced. Most serious toxicity effects such as nephrotoxicity associated with vancomycin in the past were due to impurities present with the active drug. With the currently available purified preparations, adverse reactions are significantly less frequent. The most commonly reported side effects associated with vancomycin are fever, chills, and phlebitis at the infusion site. With too rapid infusion of the drug, tingling and flushing may occur. Most importantly, shock has been reported to occur with rapid intravenous infusion of vancomycin. Allergic reactions occur in 4% to 5% of patients. The most important adverse reaction to vancomycin is neurotoxicity, which presents with auditory nerve damage and hearing loss. However, with serum concentrations less than 30 µg/mL, this complication is infrequent. Although vancomycin no longer carries a significant risk for nephrotoxicity, high doses of vancomycin should be avoided, and when used in conjunction with an aminoglycoside, the dose of vancomycin should not exceed 0.5 g every 8 hours. Serum levels of vancomycin can be monitored. With a 1-g dose, peak levels of 20 to 50 µg/mL and troughs of 5 to 10 µg/mL are expected. The major uses of vancomycin include treatment of (a) methicillin-resistant staphylococcal infections, (b) enterococcal endocarditis in penicillin-allergic patients (in combination with an aminoglycoside), and (c) antibiotic-associated C. difficile colitis. In addition, vancomycin is recommended by the American Heart Association for use as a prophylactic agent to prevent bacterial endocarditis in penicillin-allergic patients undergoing certain dental or surgical procedures. Vancomycin is not contraindicated during pregnancy or the puerperium. During pregnancy, the indications and cautions for its use are similar to those in nonpregnant patients.
SULFONAMIDES AND TRIMETHOPRIM-SULFAMETHOXAZOLE Sulfonamides were the first antimicrobial agents introduced into clinical practice in the 1930s. Shortly after Prontosil became available, it was recognized that its antibacterial activity was due to sulfanilamide. Subsequently, hundreds of sulfanilamide derivatives, the sulfonamides, were synthesized. More recently, trimethoprim was developed and found to be a potentiator of sulfonamide activity. The sulfonamides that are currently available are divided into short-acting, medium-acting, and long-acting sulfonamides (Table 23.19). In addition, there are sulfonamides limited to the gastrointestinal tract (poorly absorbed) and topical sulfonamides. These drugs are bacteriostatic and inhibit bacterial growth by interfering with bacterial synthesis of folic acid. This action occurs because sulfonamides competitively inhibit the incorporation of p-aminobenzoic acid into tetrohydroopteroic acid. Trimethoprim inhibits bacterial dihydrofolate reductase, the enzyme step in the synthesis of folic acid immediately after the step blocked by sulfonamides.
TABLE 23.19. CLASSIFICATION OF THE AVAILABLE SULFONAMIDES
Initially, the sulfonamides exhibited a broad spectrum of activity against Gram-positive and Gram-negative organisms. However, widespread resistance has developed to these agents. Thus, the primary use of these agents is in the treatment of UTIs due to susceptible bacteria. Most first-episode UTIs are due to susceptible Enterobacteriaceae. For recurrent or chronic UTIs, therapy must be based on in vitro susceptibility tests. The most common sulfonamide used for the treatment of UTIs is sulfisoxazole (Gantrisin) in a dose of 1 g orally every 6 hours. Short-acting sulfonamides are safe to use during pregnancy. However, because of concern over competition with bilirubin for binding on albumin, some authorities do not use sulfonamides during the third trimester of pregnancy. Trimethoprim is active in vitro against most Gram-positive cocci and Gram-negative rods, except for P. aeruginosa. Most anaerobes are resistant. The action of trimethoprim is potentiated in combination with sulfamethoxazole. The combination of trimethoprim-sulfamethoxazole (TMP-SMX, Bactrim, Septra) is active against 90% to 95% of S. aureus, S. pneumoniae, group A streptococci, E. coli, P. mirabilis, Shigella species, Salmonella species, and N. gonorrhoeae (1). Almost all P. aeruginosa strains are resistant to TMP-SMX. The major indication for TMX-SMX is the treatment of UTIs. Most Enterobacteriaceae are sensitive to TMX-SMX. The usual adult dosage is two tablets (trimethoprim [80 mg per tablet] and sulfamethoxazole [400 mg per tablet]) every 12 hours or a single double-strength tablet (TMX, 160 mg; SMX, 800 mg) every 12 hours. TMX-SMX is an effective agent for the treatment and prevention of gastroenteritis caused by enteropathogenic E. coli and traveler's diarrhea (2,3). Although TMX-SMX can treat chlamydial urethritis and cervicitis, its activity is due to the sulfonamide (4). TMX-SMX has been demonstrated to be effective in the treatment of Pneumocystis carinii infections in patients with AIDS. It is the first-line drug for the treatment of and primary and secondary prophylaxis against Pneumocystis carinii pneumonia (PCP) in human immunodeficiency virus–infected patients (5,6). Sulfonamides have been used to treat toxoplasmosis in patients with or without AIDS and chloroquine sensitive of
chloroquine-resistant Plasmodium falciparum malaria (with pyrimethamine) (7). TMX-SMX use in pregnancy is appropriate when indicated and other agents are not available. Side effects associated with sulfonamides include nausea, vomiting, diarrhea, rash, fever, headache, depression, jaundice, hepatic necrosis, drug-induced lupus, and a serum sickness–like syndrome (7). Serious adverse effects caused by sulfonamides include acute hemolytic anemia related to erythrocyte glucose-6-phosphate dehydrogenase deficiency, aplastic anemia, agranulocytosis, thrombocytopenia, and leukopenia (7). In addition, significant hypersensitivity reactions occur with administration of sulfonamides; examples include erythema nodosum, erythema multiforme (including Stevens-Johnson syndrome), vasculitis, and anaphylaxis. CHAPTER REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9.
Gold HS, Mollering RC. Antimicrobial-drug resistance. N Engl J Med 1996;335:1445–1453. Polk R. Optimal use of modern antibiotics: emerging trends. Clin Infect Dis 1999;29:264–274. Swartz MN. Use of antimicrobial agents and drug resistance. N Engl J Med 1997;337:491. Jacoby GA, Archer GL. New mechanisms of bacterial resistance to antimicrobial agents. N Engl J Med 1991;324:601–612. Antimicrobial therapy for obstetric patients. American College of Obstetricians and Gynecologists Educational Bulletin 1998;245. Kucers A, Crowe SM, Grayson ML, et al. The use of antibiotics: a clinical review of antibacterial, antifungal and antiviral drugs, 5th ed. Oxford, UK: Butterworth-Heineman, 1997. Philipson A. Pharmacokinetics of ampicillin in pregnancy. J Infect Dis 1977;136:370. Chow AW, Jewesson PJ. Pharmacokinetics and safety of antimicrobial agents during pregnancy. Rev Infect Dis 1985;7:287–313. Charles D. Placental transmission of antibiotics. Obstet Gynecol Annu 1979;8:19–86.
Penicillins 1. Gold HS, Mollering RC. Antimicrobial-drug resistance. N Engl J Med 1996;335:1445–1453. 2. Centers for Disease Control and Prevention. 1998 guidelines for treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1997;47[RR-1]:1–118. 3. Heikkila AM, Erkkola RU. The need for adjustment of dosage regimen of penicillin V during pregnancy. Obstet Gynecol 1993;81:919–921. 4. Centers for Disease Control and Prevention. Prevention of perinatal group B streptococcal disease: a public health perspective. MMWR Morb Mortal Wkly Rep 1996;45[RR-7]:1–24. 5. Harding GKM, Buckwold FJ, Ronald AR, et al. Prospective randomized comparative study of clindamycin, chloramphenicol and ticarcillin each in combination with gentamicin in therapy for intra-abdominal and female genital tract sepsis. J Infect Dis 1980;142:384. 6. Faro S, Sanders CV. Aldridge KE. Use of single-agent antimicrobial therapy in the treatment of polymicrobial female pelvic infections. Obstet Gynecol 1982;60:232–236. 7. Winston DJ, Wang D, Young LS, et al. In vitro studies of piperacillin, a new semisynthetic penicillin. Antimicrob Agents Chemother 1978;13:944. 8. Sweet RL, Robbie MO, Ohm-Smith M, et al. Comparative study of piperacillin versus cefoxitin in the treatment of obstetric and gynecologic infections. Am J Obstet Gynecol 1983;145:342–349. 9. Wolner-Hanssen P, Paavanen J, Kiviat N, et al. Treatment of pelvic inflammatory disease with Augmentin with and without doxycycline in ambulatory patients. Am J Obstet Gynecol 1988;158:577–579. 10. Dinsmoor MJ, Gibbs RS. The role of the newer antimicrobial agents in obstetrics and gynecology. Clin Obstet Gynecol 1988;31:423–434. 11. Cuchural GJ, Tally FP, Jacobus N, et al. Susceptibility of the Bacteroides fragilis group in the United States. Analysis by site of isolation. Antimicrob Agents Chemother 1988;32:717–722.
12. Apuzzio JJ, Ganesh V, Kaminski Z, et al. Comparative clinical evaluation of ticarcillin plus clavulanic acid versus clindamycin-gentamicin in treatment of post-cesarean endometritis. Am J Med 1985;79[Suppl 58]:164. 13. Faro S, Martens M, Phillips LE. Ticarcillin disodium/clavulanate potassium versus clindamycin/gentamicin in the treatment of postpartum endometritis. J Reprod Med 1988;33[Suppl 6]:603–606. 14. Chamberlain A, White S, Bawdon R, et al. Pharmacokinetics of ampicillin and sulbactam in pregnancy. Am J Obstet Gynecol 1993;168:667–673. 15. Crombleholme WR, Ohm-Smith M, Robbie MO, et al. Ampicillin/sulbactam versus metronidazole-gentamicin in the treatment of soft tissue pelvic infections. Am J Obstet Gynecol 1987;156:507–512. 16. Stovall TG, Thorpe EM, Ling FW. Treatment of post-cesarean section endometritis with ampicillin and sulbactam or clindamycin and gentamicin. J Reprod Med 1993;38:843–848. 17. McGregor JA, Crombleholme WR, Newton E. Randomized comparison of ampicillin-sulbactam to cefoxitin and doxycycline or clindamycin and gentamicin in the treatment of pelvic inflammatory disease or endometritis. Obstet Gynecol 1994;83:998–1004. 18. Sweet RL, Roy S, Faro S, et al. Piperacillin and tazobactam versus clindamycin and gentamicin in the treatment of hospitalized women with pelvic infection. Obstet Gynecol 1994;83:280–286. 19. Culver SM, Martens MG. Piperacillin/Tazobactam (Zosyn). Infect Dis Obstet Gynecol 1997;4:258:62.
Cephalosporins 1. Asbel LE, Levison ME. Cephalosporins, carbapenems, and monobactams. Infect Dis Clin North Am 2000;14:435–447. 2. Moellering RC, Swartz MH. The newer cephalosporins. N Engl J Med 1976;294:24. 3. Nightingale CH, Greene DS, Quintiliani R. Pharmacokinetics and clinical use of cephalosporin antibiotics. J Pharmacol Sci 1975;64:1899. 4. Garrod LP. Choice among penicillins and cephalosporins. Br Med J 1974;3:96. 5. Grieco MH. Cross-allergenicity of the penicillins and the cephalosporins. Arch Intern Med 1967;199:141. 6. Karchmer AW. Cephalosporins. In: Mandel GL, Bennett JE, Dolin R, eds. Philadelphia: Churchill Livingstone, 2000:274–290. 7. Sutter VL, Kirby B, Finegold SM. In vitro activity of cefoxitin and parenterally administered cephalosporins against anaerobic bacteria. Rev Infect Dis 1979;1(1):218. 8. Cefamandole and cefoxitin. Med Lett 1979;12(3):13. 9. Cuchural GJ, Tally FP, Jacobus NV, et al. Susceptibility of the Bacteroides fragilis group in the United States: analysis by site of isolation. Antimicrob Agents Chemother 1988;32:717–722. 10. Wexler HM, Molitoris E, Finegold SM. Effect of b-lactamase inhibitors on the activities of various b-lactam agents against anaerobic bacteria. Antimicrob Agents Chemother 1991;35:1219–1224. 11. Smith BR, LeFrock JL. Cefuroxime: antimicrobial activity, pharmacology and clinical efficacy. Ther Drug Monit 1983;5:149. 12. Gremillion DH, Winn RE, Vandenbout E. Clinical trial of cefonicid for treatment of skin infections. Antimicrob Agents Chemother 1983;23:944. 13. Bruch K. Hypoprothrombinemia and cephalosporins. Lancet 1983;1:535–536. 14. Giamarellou H, Gazis J, Petrikkos G, et al. A study of cefoxitin moxalactam, and ceftazidime kinetics in pregnancy. Am J Obstet Gynecol 1983;147:914. 15. Sweet RL, Ledger WJ. Cefoxitin: single-agent treatment of mixed aerobic-anaerobic pelvic infections. Obstet Gynecol 1979;54:193. 16. Ledger WJ, Smith D. Cefoxitin in obstetric and gynecologic infections. Rev Infect Dis 1979;1:199. 17. Duff P, Keiser JF. A comparative study of two antibiotic regimens for the treatment of operative site infections. Am J Obstet Gynecol 1982;142:996. 18. Hager WD, McDaniel PS. Treatment of serious obstetric and gynecologic infections with cefoxitin. J Reprod Med 1983;28:337. 19. Larsen JW, Voise CT, Grossman JH III. Comparison of cefoxitin and clindamycin-gentamicin for pelvic infections. Clin Ther 1986;9:77–83. 20. Poindexter AN, Sweet R, Ritter M. Cefotetan in the treatment of obstetric and gynecologic
infections. Am J Obstet Gynecol 1986;154:946–950. 21. Poindexter AN, Ritter M, Faro S, et al. Results of noncomparative studies of cefotetan in the treatment of obstetric and gynecologic infections. Am J Obstet Gynecol 1988;158:717–721. 22. Hemsell DL, Wendel GD, Goll SA, et al. Multicenter comparison of cefotetan and cefoxitin in the treatment of acute obstetric and gynecologic infections. Am J Obstet Gynecol 1988;158:722–727. 23. Sweet RL, Schachter J, Landers DV, et al. Treatment of hospitalized patients with acute pelvic inflammatory disease: comparison of cefotetan plus doxycycline and cefoxitin plus doxycycline. Am J Obstet Gynecol 1988;158:736–743. 24. Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted diseases. MMWR 1998;47(RR-1):1–116. 25. Stiver HG, Forward KR, Livingstone RA, et al. Multicenter comparison for cefoxitin versus cefazolin for prevention of infectious morbidity after nonelective cesarean section. Am J Obstet Gynecol 1983;145:158. 26. Neu HC. The new beta-lactamase-stable cephalosporins. Ann Intern Med 1982;97:408. 27. Handsfield HH, Murphy VL. Comparative study of ceftriaxone and spectinomycin for treatment of uncomplicated gonorrhea in men. Lancet 1983;2:67. 28. Ohm-Smith MJ, Hadley WK, Sweet RL. In vitro activity of new b-lactam antibiotics and other antimicrobial drugs against anaerobic isolates from obstetric and gynecological infections. Antimicrob Agents Chemother 1982;17:711. 29. Jorgensen JH, Crawford SA, Alexander GA. Comparison of moxalactam (LY127935) and cefotaxime against anaerobic bacteria. Antimicrob Agents Chemother 1980;17:901–904. 30. Kafetzis DA, Lazarides CV, Sifas CA, et al. Transfer of cefotaxime in human milk from mother to fetus. J Antimicrob Chemother 1980;6[Suppl A]:135. 31. Gonik B, Feldman S, Pickering LK, et al. Pharmacokinetics of cefoperazone in the parturient. Antimicrob Agents Chemother 1986;30:874–876. 32. Dinsmoor MJ, Gibbs RS. The role of the newer antimicrobial agents in obstetrics and gynecology. Clin Obstet Gynecol 1988;31:423–434. 33. Kafetzis DA, Broler DC, Fanourgakis JE, et al. Ceftriaxone distribution between maternal blood and fetal blood and tissues at parturition and between blood and milk postpartum. Antimicrob Agents Chemother 1983;23:870–873. 34. Antimicrobials and haemostasis [Editorial]. Lancet 1983;1:510–511. 35. Bruch K. Hypoprothrombinemia and cephalosporins. Lancet 1983;1:535–536. 36. Quintiliani R. Bleeding disorders associated with newer cephalosporins. Clin Pharm 1983;2:360–361. 37. Weitekamp MR, Aber RC. Prolonged bleeding times and bleeding diathesis associated with moxalactam administration. JAMA 1983;249:69–71. 38. Kammer RB. Moxalactam: clinical summary of efficacy and safety. Rev Infect Dis 1982;4[Suppl]:712. 39. Lipsky JJ. N-methyl-thiotetrazole inhibition of gamma carboxylation of glutamic acid: possible mechanism for antibiotic associated hypoprothrombinemia. Lancet 1983;1:192–193. 40. Lipsky JJ. Mechanism of the inhibition of the gamma-carboxylation of glutamic acid by N-methyl-thiotetrazole-containing antibiotics. Proc Natl Acad Sci U S A 1984;81:2893–2897. 41. Kline SS, Mauro VF, Forney RB, et al. Cefotetan-induced disulfiram-type reactions and hypoprothrombinemia. Antimicrob Agents Chemother 1987;31:1328–1331. 42. Halt J. Hypoprothrombinemia and bleeding diathesis associated with cefotetan therapy in surgical patients. Arch Surg 1988;123:523. 43. Gibbs RS, Blanco JD, Castaneda YS, et al. Therapy of obstetrical infections with moxalactam. Antimicrob Chemother 1980;17:1004. 44. Cunningham FG, Hemsell DL, DePalma RT, et al. Moxalactam for obstetric and gynecologic infections in vitro and dose-finding studies. Am J Obstet Gynecol 1981;139:915. 45. Gibbs RS, Blanco JD, Duff P. A double-blind, randomized comparison of moxalactam versus clindamycin-gentamicin in treatment of endomyometritis after cesarean section delivery. Am J Obstet Gynecol 1983;146:769. 46. Sweet RL, Ohm-Smith M, Landers DV, et al. Moxalactam versus clindamycin plus tobramycin in the treatment of obstetric and gynecologic infections. Am J Obstet Gynecol 1985;152:808–817. 47. Hemsell DL, Cunningham FG, DePalma RT, et al. Cefotaxime sodium therapy for endomyometritis following cesarean section: dose-finding and comparative studies. Obstet Gynecol 1983;62:489. 48. Hemsell DL, Cunningham FG. Combination antimicrobial therapy for serious gynecological and
obstetrical infections—obsolete? Clin Ther 1981;4:81. 49. Hemsell DA, Rigoberto SR, Cunningham FG, et al. Cefotaxime treatment for women with community acquired pelvic abscess. Am J Obstet Gynecol 1985;151:771–777. 50. Strausbaugh IJ, Llorens AS. Cefoperazone therapy for obstetric and gynecologic infections. Rev Infect Dis 1983;5:S154. 51. Gilstrap LC III, St. Clair PJ, Gibbs RS, et al. Cefoperazone versus clindamycin plus gentamicin for obstetric and gynecologic infections. Antimicrob Agents Chemother 1986;30:808–809. 52. Blanco JD, Gibbs RS, Duff P, et al. Randomized comparison of ceftazidime versus clindamycin-tobramycin in the treatment of obstetrical and gynecological infections. Antimicrob Agents Chemother 1983;24:500. 53. Harding G, Vincelette J, Rochlis A, et al. A preliminary report on the use of ceftizoxime vs. clindamycin/tobramycin for the therapy of intra-abdominal and pelvic infection. Antimicrob Agents Chemother 1982;10[Suppl]:191–192. 54. Lou MA, Chen DF, Bansal M, et al. Evaluation of ceftizoxime in acute peritonitis. J Antimicrob Chemother 1982;10[Suppl]:183–189. 55. Apuzzio JJ, Ganesh V, Pelosi M, et al. Comparative clinical evaluation of ceftizoxime with clindamycin and gentamicin and cefoxitin in the treatment of post-cesarean endomyometritis. Surg Gynecol Obstet 1985;161:518–522. 56. Cunha BA, Gill MW. Cefepime. Med Clin North Am 1995;79:721–732. 57. Oster S, Edelstein H. Cassano K, et al. Open trial of cefepime (BMY 28142) for infections in hospitalized patients. Antimicrol Agents Chemother 1990;34:954 58. Sharifi R, Gecker R, Childs S. Treatment of urinary tract infections: selecting an appropriate broad-spectrum antibiotic for nosocomial infections. Am J Med 1996;100:76–82. 59. Baire PS, Vogel SB, Dellinger EP, et al. A randomized double-blind clinical trial comparing cefepime plus metronidazole with imipenem-cilastatin in the treatment of complicated intra-abdominal infections. Cefepime intra-abdominal infection study group. Arch Surg 1997;132:1294–1302.
New Oral Cephalosporins 1. Neu HC. Oral b-lactam antibiotics from 1960–1993. Infect Dis Clin Pract 1993;2:394–404. 2. Reese RE, Betts RF. Handbook of antibiotics. Boston: Little, Brown and Company, 1993. 3. Neu HC, Fu KP. Cefuroxime, a b-lactamase-resistant cephalosporin with a broad spectrum of Gram-positive and negative activity. Antimicrob Agents Chemother 1978;13:657–664. 4. Cefprozil [Letter]. Med Lett Drugs Ther 1992;36:63. 5. Klein JO. Evaluation of new oral antimicrobial agents and the experience with cefprozil—a broad spectrum oral cephalosporin. Clin Infect Dis 1992;14[Suppl 2]:183. 6. Doern G. In vitro activity of loracarbef and effects of susceptibility test methods. Am J Med 1992;92[Suppl 6A]:7S. 7. Handsfield HH. A comparison of single-dose cefixime with ceftriaxone as treatment for uncomplicated gonorrhea. N Engl J Med 1991;325:1337. 8. Plourde PJ. Single dose cefixime versus single dose ceftriaxone in the treatment of antimicrobial-resistant Neisseria gonorrhoeae infection. Infect Dis 1992;166:919. 9. Dobernat H. In vitro activity of cefpodoxime against pathogens responsible for community-acquired respiratory tract infection. J Antimicrob Chemother 1990;26[Suppl E]:T. 10. Cefpodoxime proxetil—a new oral cephalosporin [Letter]. Med Lett Drugs Ther 1992;34:107.
Monobactams 1. Davis JD. Aztreonam. PrimCare Update Obstet Gynecol 1997;4:61–64. 2. Swabb EA, Sugarimon A, Platt TB, et al. Single-dose pharmacokinetics of the monobactam aztreonam (SQ26,776) in healthy subjects. Antimicrob Agents Chemother 1982;21:944. 3. Gibbs RS, Blanco JD, Lipscomb KA, et al. Aztreonam versus gentamicin, each with clindamycin, in the treatment of endometritis. Obstet Gynecol 1985;65:825. 4. Dodson MG, Faro S, Gentry LO. Treatment of acute pelvic inflammatory disease with aztreonam, a
new monocyclic b-lactam antibiotic, and clindamycin. Obstet Gynecol 1986;67:657.
Carbapenems 1. Kesada T, Hashizume T, Asahi Y. Antibacterial activities of a new stabilized thienamycin, N-formimidayl thienamycin, in comparison with other antibiotics. Antimicrob Agents Chemother 1980;17:912. 2. Kropp H, Sundelof JG, Kahan JS, et al. MK 0787 (N-formimidayl thienamycin): evaluation of in vitro and in vivo activities. Antimicrob Agents Chemother 1983;17:993. 3. Kropp H, Gerckens L, Sundelof JG, et al. Antibacterial activity of imipenem: the first thienamycin antibiotic. Rev Infect Dis 1985;7[Suppl 2]:389–410. 4. Berkeley AS, Freedman K, Hirsch J, et al. Imipenem/cilastatin in the treatment of obstetric and gynecologic infections. Am J Med 1985;78[Suppl]:79–84. 5. Sweet RL. Imipenem/cilastatin in the treatment of obstetric and gynecologic infections: a review of worldwide experience. Rev Infect Dis 1985;7[Suppl]:522–527. 6. Berkeley AS, Freedman KS, Hirsch JC, et al. Randomized, comparative trial of imipenem/cilastatin and moxalactam in the treatment of serious obstetric and gynecologic infections. Surg Gynecol Obstet 1986;162:204. 7. Chambers HF. Other b-lactam antibiotics. In: Mandel EL, Bennett JE, Dolin R, eds. Principles and practice of infectious disease. Philadelphia: Churchill Livingstone, 2000:291–298.
Tetracyclines 1. Lynn SF. Tetracycline and doxycycline applications. Primary Care Update Obstet Gynecol 1996;3:224–232. 2. The choice of antibacterial drugs [Letter]. Med Lett Drugs Ther 1992;34:49. 3. Reese RE, Betts RF. Handbook of antibiotics. Boston: Little, Brown and Company, 1993. 4. Centers for Disease Control and Prevention. Penicillinase (beta-lactamase) producing Neisseria gonorrhoeae—worldwide. MMWR Morb Mortal Wkly Rep 1978;27:10. 5. Centers for Disease Control and Prevention. Antibiotic-resistant strains of Neisseria gonorrhoeae. Policy guidelines for management, detection, and control. MMWR Morb Mortal Wkly Rep 1987;36[Suppl]:1–18. 6. Czeizel AE, Rockenbauer M. Teratogenic study of doxycycline. Obstet Gynecol 1997;89:524–528. 7. Whalley PJ, Adams RH, Combes B. Tetracycline toxicity in pregnancy. JAMA 1964;189:357.
Clindamycin 1. Stevens HA. Primary Care Update Obstet Gynecol 1997;4:251–253. 2. Cuchural GJ, Tally FP, Jacobus N, et al. Susceptibility of the Bacteroides fragilis group in the United States: analysis by site of isolation. Antimicrob Agents Chemother 1988;32:717–722. 3. Cuchural GJ, Tally FP, Jacobus NV, et al. Comparative activities of newer beta-lactam agents against members of the Bacteroides fragilis group. Antimicrob Agents Chemother 1990;34:479–480. 4. Turgeon P, Turgeon V, Gardeau M, et al. Longitudinal study of susceptibilities of species of the Bacteroides fragilis group to five antimicrobial agents in three medical centers. Antimicrob Agents Chemother 1994;38:2276–2279. 5. Bowie WR. In vitro activity of clindamycin against Chlamydia trachomatis. Sex Transm Dis 1987;8:220–221. 6. Wasserheit JN, Bell JA, Kiviat NB, et al. Microbial causes of proven pelvic inflammatory disease and efficacy of clindamycin and tobramycin. Ann Intern Med 1986;104:187. 7. Johnson S, Gerding DN. Clostridium difficile–associated diarrhea. Clin Infect Dis 1998;26:1027–1036. 8. Gibbs RS, Blanca JD, Castaneda YS, et al. A double-blind randomized comparison of clindamycin-gentamicin versus cefamandole for treatment of post-cesarean section endomyometritis. Am J Obstet Gynecol 1982;144:261.
9. Greaves WL, Chungafung J, Morris B, et al. Clindamycin versus metronidazole in the treatment of bacterial vaginosis. Obstet Gynecol 1988;72:799–802. 10. Hillier S, Krohn MA, Watts H, et al. Microbiologic efficacy of intravaginal clindamycin cream for the treatment of bacterial vaginosis. Obstet Gynecol 1990;76:407–413. 11. Livengood CH, Thomason JL, Hill GB. Bacterial vaginosis: treatment with topical intravaginal clindamycin phosphate. Obstet Gynecol 1990;76:118–123. 12. Pearlman MD, Pierson CL, Faix RG. Frequent resistance of clinical group B streptococci isolates to clindamycin and erythromycin. Obstet Gynecol 1998;92:258–261. 13. Rouse DJ, Andrews WW, Feng-Ying CL, et al. Antibiotic susceptibility profile of group B streptococcus acquired vertically. Obstet Gynecol 1998;92:931–934.
Erythromycin, Azithromycin, And Clarithromycin 1. Reese RE, Betts RF. Handbook of antibiotics. Boston: Little, Brown and Company, 1993. 2. Steigbigel NH. Macrolides and Clindamycin. In: Mandrel GL, Bennett JE, Dolin R, eds. Philadelphia: Churchill Livingstone, 2000:366–381. 3. Zuckerman JM. The newer macrolides: azithromycin and clarithromycin. Infect Dis Clin North Am 2000;14:449–462. 4. Frazer DG. Selection of an oral erythromycin product. Am J Hosp Pharmacol 1980;37:1199. 5. McCormack WM, George H, Donner A, et al. Hepatotoxicity of erythromycin estolate during pregnancy. Antimicrob Agents Chemother 1977;12:630. 6. Philipson A, Sabath LD, Charles D. Transplacental passage of erythromycin and clindamycin. N Engl J Med 1973;288:1219. 7. Gallagher JD, Ismall MA, Aladjem S. Reduced urinary estriol levels with erythromycin therapy. Obstet Gynecol 1980;56:381. 8. Schachter J, Sweet RL, Grossman M, et al. Experience with the routine use of erythromycin for chlamydial infections in pregnancy. N Engl J Med 1986;314:276–279. 9. Moellering RC Jr. Introduction: revolutionary changes in the macrolide and azalide antibiotics. Am J Med 1991;91[Suppl 3A]:1S. 10. Nev HC. Clinical microbiology of azithromycin. Am J Med 1991;9[Suppl 3A]:12S. 11. Clarithromycin and azithromycin [Letter]. Med Lett Drug Ther 1992;34:45. 12. Schenlag JJ, Ballow CH. Tissue directed pharmacokinetics. Am J Med 1991;91[Suppl 3A]:5S. 13. Piscitelli SC, Danziger LH, Radvold KA. Clarithromycin and azithromycin: new macrolide antibiotics. Clin Pharmacol 1992;11:137. 14. Handsfield HH, Dalv ZZ, Martin DH, et al. Multicenter trial of single-dose azithromycin vs. ceftriaxone in the treatment of uncomplicated gonorrhea. Sex Transm Dis 1994;21:107–111. 15. Hopkins S. Clinical toleration and safety of azithromycin. Am J Med 1991;92[Suppl 3A]:405. 16. Bush MR, Rosa C. Azithromycin and erythromycin in the treatment of cervical chlamydial infection during pregnancy. Obstet Gynecol 1994;84:61–63. 17. Edwards MS, Newman RB, Carter SG, et al. Randomized clinical trial of azithromycin vs. erythromycin for the treatment of chlamydial cervicitis in pregnancy. Infect Dis Obstet Gynecol 1996;41:333–337. 18. Adair CD, Gunter M, Stovall TG, et al. Chlamydia in pregnancy: a randomized trial of azithromycin and erythromycin. Obstet Gynecol 1998;91:165–168. 19. Centers for Disease Control and Prevention. 1993 sexually transmitted diseases treatment guidelines. MMWR Morb Mortal Wkly Rep 1993;42:1–102. 20. Olsson-Liljequist B, Hoffman BM. In vitro activity of clarithromycin combined with its 14-hydroxy metabolite A-62671 against Haemophilus influenzae. J Antimicrob Chemother 1991;27[Suppl A]:11. 21. Nev HC. The development of macrolides. Clarithromycin in perspective. J Antimicrob Chemother 1991;27[Suppl A]:1. 22. Pierce M, Crampton S, Henry D, et al. A randomized trial of clarithromycin or prophylaxis against disseminated Mycobacterium avium complex infection in patients with advanced AIDS. N Engl J Med 1996;335:384–391.
Aminoglycosides 1. Bennett CC. The aminoglycosides. Primary Care Update Obstet Gynecol 1996;3:186–191. 2. Del Priore G, Jackson-Stone M, Shim EK, et al. A comparison of once-daily and 8-hour gentamicin dosing in the treatment of postpartum endometritis. Obstet Gynecol 1996;87:994–1000. 3. Perry KG, Theilman KA, Coker JN. A prospective randomized trial comparing gentamicin administration every 8 hours versus gentamicin administration every 24 hours for the treatment of postcesarean section endometritis: proceedings of the Society of Perinatal Obstetricians, 1997. Abstract 177. 4. Duff P, Jorgensen JH, Alexander G, et al. Serum gentamicin levels in patients with post-cesarean endomyometritis. Obstet Gynecol 1983;61:723. 5. Blanco JD, Gibbs RS, Duff P, et al. Serum tobramycin levels in puerperal women. Am J Obstet Gynecol 1983;147:466. 6. Zaske DE, Cipolle RJ, Strate RG, et al. Rapid gentamicin elimination in obstetric patients. Obstet Gynecol 1980;56:559. 7. Zaske DE, Cipolle RJ, Strate RG, et al. Increased gentamicin dosage requirements: rapid elimination in 249 gynecology patients. Am J Obstet Gynecol 1981;139:896. 8. Weinstein AJ, Gibbs RS, Gallagher M. Placental transfer of clindamycin and gentamicin in term pregnancy. Am J Obstet Gynecol 1976;124:688.
Metronidazole 1. Robbie MO, Sweet RL. Metronidazole use in obstetrics and gynecology: a review. Am J Obstet Gynecol 1983;145:865–881. 2. Simms-Cendan JS. Metronidazole. Primare Care Update Obstet Gynecol 1996;3:153–156. 3. Muller M, Meingassner JG, Miller WA, et al. Three resistant metronidazole-resistant strains of Trichomonas vaginalis from the United States. Am J Obstet Gynecol 1980;138:808–812. 4. Beard CM, Noller KL, O'Fallan M, et al. Lack of evidence for cancer due to use of metronidazole. N Engl J Med 1979;301:519–522. 5. Burtin P, Taddio A, Ariburnu O. Safety of metronidazole in pregnancy: a meta-analysis. Am J Obstet Gynecol 1995;172:525–529. 6. Kucers A, Crowe SM, Grayson ML, et al. The use of antibiotics. A clinical review of antibacterial, antifungal and antiviral drugs, 5th ed. Oxford, UK: Butterworth-Heineman, 1997. 7. Centers for Disease Control and Prevention. 1998 guidelines for treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1997;47[RR-1]:1–118. 8. Erickson SH, Oppenheim GL, Smith GH. Metronidazole in breast milk. Obstet Gynecol 1981;57:48. 9. Collier J, Calhoun EN, Hill PL. A multicenter comparison of clindamycin and metronidazole in the treatment of anaerobic infections. Scand J Infect Dis 1981;26[Suppl]:96. 10. Gall SA, Kohan AP, Ayers OM. Intravenous metronidazole or clindamycin with tobramycin for therapy of pelvic infection. Obstet Gynecol 1981;57:51.
Fluoroquinolones 1. Wolfson JS, Hooper DC. The fluoroquinolones: structures, mechanisms of action and resistance, and spectra of activity in vitro. Antimicrob Agents Chemother 1985;28:581–586. 2. Dinsmoor MJ, Gibbs RS. The role of the newer antimicrobial agents in obstetrics and gynecology. Clin Obstet Gynecol 1988;31:423–434. 3. O'Donnell JA, Gelone SP. Fluoroquinolones. Infect Dis Clin North Am 2000;14:489–513 4. Hooper DC. Quinolones. In: Mandel GL, Bennett J, Dolin R, eds. Principles and practice of infectious disease. Philadelphia: Churchill Livingstone, 2000:4004–4023. 5. Dembry LM, Farrington JM, Andriole VT. Fluoroquinolone antibiotics: adverse effects and safety profiles. Infect Dis Clin Pract 1999;8:421–428. 6. Rodandi LC. Quinolones: an update. Pharmacy and therapeutics forum. San Francisco: University of California, 1987;35(July/Aug):4.
7. Andriole VT. Quinolones. In: Gorbach SL, Bartlett JG, Blacklow NR, eds. Infectious diseases, 2nd ed. Philadelphia: WB Saunders, 1998:275–289. 8. Schentag JJ, Scully BE. Quinolones. In: Yu VL, Merigan TC Jr, Barriere S, eds. Antimicrobial therapy and vaccines. Baltimore: Williams & Wilkins, 1999:875. 9. Eggleston M, Park S-Y. Review of the 4-quinolones. Infect Control 1987;8:119–125. 10. Hooper DC, Wolfson JS. Fluoroquinolone antimicrobial agents. N Engl J Med 1991;324:384. 11. Neu HC. Quinolone antimicrobial agents. Am Rev Med 1992;43:463. 12. Sprangler SR, Jacobs MR, Applebaum PC. Susceptibilities of penicillin-susceptible and resistant strains of S. pneumoniae to RP59500, vancomycin, erythromycin, PD-131628, sparfloxacin, temafloxacin, WIN 57273, ofloxacin and ciprofloxacin. Antimicrob Agents Chemother 1992;36:856. 13. Visalli MA, Jacobs MR, Applebaum PC. Activity of CP99,219 (trovafloxacin) compared with ciprofloxacin, sparfloxacin, clinafloxacin, lomefloxacin and cefixime against ten penicillin-susceptible and penicillin-resistant pneumococci by time-kill methodology. J Antimicrob Chemother 1996;37:77. 14. Chen DK, McGeer A, deAzavedo JC, et al. Decreased susceptibility of Streptococcus pneumoniae to fluoroquinolones in Canada. N Engl J Med 1999;341:233. 15. Goldstein EJC. Possible role of fluoroquinolones (levofloxacin, grepafloxacin, trovafloxacin, sparfloxacin, DU-6859a) in the treatment of anaerobic infections: review of current information and efficacy. Clin Infect Dis 1996;23[Suppl 1]:S25. 16. Liebowitz LD, Saunders J, Fehler G, et al. In vitro activity of A-56619 (difloxacin), A-56620, and other new quinolone antimicrobial agents against genital pathogens. Antimicrob Agents Chemother 1986;30:948–950. 17. Bailey JMG, Heppleston C, Richmond SJ. Comparison of the in vitro activities of ofloxacin and tetracycline against Chlamydia trachomatis as assessed by indirect immunofluorescence. Antimicrob Agents Chemother 1984;26:13–16. 18. Heppleston C, Richmond S. Antichlamydial activity of quinolone carboxylic acids. J Antimicrob Chemother 1985;15:645–646. 19. Edelstien PH. Antimicrobial chemotherapy for Legionnaires disease: time for a change. Ann Intern Med 1998;15:328. 20. Wolfson JS, Hooper DC. Bacterial resistance to quinolones: mechanisms and clinical importance. Rev Infect Dis 1989;11[Suppl 5]:S960. 21. Trucksis M, Hooper DC, Wolfson JS. Emerging resistance to fluoroquinolones in staphylococci: an alert. Ann Intern Med 1991;114:424. 22. Raviglone MC. Ciprofloxacin-resistant Staphylococcus aureus in an acute care hospital. Antimicrob Agents Chemother 1990;34:2050.23. 23. Nightingale CH. Overview of the pharmacokinetics of fleroxacin. Am J Med 1993–94;[Suppl 3A]:38S–42S. 24. Neuman M. Clinical pharmacokinetics of the new antibacterial 4-quinolones. Clin Pharmacokinet 1988;14:96–121. 25. Drusano GL, Standiford HC, Plaisance K, et al. Absolute oral bioavailability of ciprofloxacin. Antimicrob Agents Chemother 1986;30:444–446. 26. Yuk JH, Nightingale CH, Quintilani R, et al. Bioavailability and pharmacokinetics of ofloxacin in healthy volunteers. Antimicrob Agents Chemother 1991;35:384–386. 27. Stratton C. Fluoroquinolone antibiotics: properties of the class and individual agents. Clin Ther 1992;14:348–375. 28. Freeman CD, Nicolau DP, Belliveau PP, et al. Lomefloxacin clinical pharmacokinetics. Clin Pharmacokinet 1993;25:6–19. 29. Stein GE. Pharmacokinetics and pharmacodynamics of newer fluoroquinolones. Clin Infect Dis 1996;23[Suppl 1]:519 30. The choice of antibacterial drugs [Letter]. Med Lett Drugs Ther 1992;34:49. 31. Centers for Disease Control and Prevention. 1998 sexually transmitted diseases treatment guidelines. MMWR Morb Mortal Wkly Rep 1998;47(RR-1):1–116. 32. Drugs for sexually transmitted diseases [Letter]. Med Lett Drugs Ther 1991;33;119. 33. Walker RC, Wright AJ. The fluoroquinolones. Mayo Clin Proc 1992;66:1249. 34. Lipsky BA, Baker CA. Fluoroquinolone toxicity profiles: a review focusing on newer agents. Clin Infect Dis 1999;28:352. 35. Stahlmann R, Schwabe R. Safety profile of grepafloxacin compared with other fluoroquinolones. J Antimicrob Chemother 1997;40:83–92.
36. Stahlmann R. Lode H. Safety overview: toxicity, adverse effects, and drug interactions. In: Andriole VT, ed. The quinolones, 2nd ed. London: Academic Press, 1998:370–417. 37. Garyson ML. Ciprofloxacin. In: Kucers A, Crowe S, Grayson ML, et al, eds. The use of antibiotics, 5th ed. Oxford, UK: Butterworth-Heineman, 1997:981–1060. 38. Stahlmann R, Forster C, Shakibaei M, et al. Magnesium deficiency induces joint cartilage lesions in juvenile rats which are identical to quinolone-induced arthropathy. Antimicrob Agents Chemother 1995;39:2013. 39. Ball P, Tillotson G. Tolerability of fluoroquinolone antibiotics. Drug Safety 1995;13:343–358. 40. Christ W, Esch B. Adverse reactions to fluoroquinolones in adults and children. Infect Dis Clin Pract 1994;3[Suppl 3]:S168. 41. Bethell DB, Hien TT, Phi LT, et al. Effects on growth of single short courses of fluoroquinolones. Arch Dis Child 1996;74:44. 42. Church DA, Kanga JF, Kuhn RJ, et al. Sequential ciprofloxacin therapy in pediatric cystic fibrosis: comparative study versus ceftazidime/tobramycin in the treatment of acute pulmonary exacerbations. Pediatr Infect Dis J 1997;16:97. 43. Burkhardt JE, Walterspiel JN, Schaad UB. Quinolone arthropathy in animals versus children. Clin Infect Dis 1997;25:1196–1204. 44. Schaad UB, Abdus SM, Aujard Y, et al. Use of fluoroquinolone in pediatrics: consensus report of an International Society of Chemotherapy commission. Pediatr Infect Dis J 1995;14:1–9. 45. Giamarellou H, Kilokuthas E, Petrikkos G, et al. Pharmacokinetics of three newer quinolone in pregnant and lactating women. Am J Med 1989;87:[Suppl 5A]:49–51. 46. Lumpkin MM. Public health advisory [Letter correspondence]. Food and Drug Administration: Trovan (trovafloxacin/alatrofloxacin); June 10, 1999. 47. Jaillon P, Morganroth J, Brumpt I, et al. Overview of electrocardiographic and cardiovascular safety data for sparfloxacin. J Antimicrob Chemother 1996;37:161–167. 48. Berkovitch M, Pastuszack A, Gazarian M, et al. Safety of the new quinolones in pregnancy. Obstet Gynecol 1994;84:535–538. 49. Loebstein R, Addis A, Ho E, et al. Pregnancy outcome following gestational exposure to fluoroquinolones: a multicenter prospective controlled study. Antimicrob Agents Chemother 1998;42:1336–1339. 50. Reese RE, Betts RF. Handbook of antibiotics. Boston: Little, Brown and Company, 1993. 51. Sabbaj J, Hoagland VL, Shih WJ. Multiclinic comparative study of norfloxacin and trimethoprim-sulfamethoxazole for treatment of urinary tract infections. Antimicrob Agents Chemother 1985;27:297–301. 52. Romanowski B, Wood H, Draker J, et al. Norfloxacin in the therapy of uncomplicated gonorrhea. Antimicrob Agents Chemother 1986;30:514–515. 53. Raddy RE, Handsfield HH, Hook EW III. Comparative trial of single-dose ciprofloxacin and ampicillin plus probenecid for treatment of gonococcal urethritis in men. Antimicrob Agents Chemother 1986;30:267–269. 54. Fass RJ. Efficacy and safety of oral ciprofloxacin for treatment of serious urinary tract infection. Antimicrob Agents Chemother 1987;31:148–150. 55. Crombleholme WR, Schachter J, Ohm-Smith M, et al. Efficacy of single agent therapy for the treatment of acute pelvic inflammatory disease with ciprofloxacin. Am J Med 1989;87[Suppl 5A]:S142. 56. Soper DE, Brockwell NJ, Dalton HP. Microbial etiology of urban emergency department acute salpingitis: treatment with ofloxacin. Am J Obstet Gynecol 1992;167:653–660. 57. Wendel GD, Cox SM, Bawdon RE, et al. A randomized trial of ofloxacin versus cefoxitin and doxycycline in the treatment of acute salpingitis. Am J Obstet Gynecol 1991;164:1390–1396.
Vancomycin 1. Gold HS, Moellering RC. Antimicrobial drug resistance. N Engl J Med 1996;355:1445–1454. 2. Johnson S, Gerding DN. Clostridium difficile–associated diarrhea. Clin Infect Dis 1998;26:1027–1036.
Sulfonamides And TMP-SMX 1. Bushby SRM. Sensitivity patterns and use of a combined disc of trimethoprim-sulfamethoxazole. In: Bernstein LS, Salter AJ, eds. Trimethoprim/sulfamethoxazole in bacterial infections. London: Churchill Livingstone, 1973:31. 2. Tharen A, Walde-Mariam I, Stinlzing G, et al. Antibiotics in the treatment of gastroenteritis caused by enteropathogenic Escherichia coli. J Infect Dis 1980;141:27. 3. DuPont HL, Evans DG, Rios N, et al. Prevention of traveler's diarrhea with trimethoprim-sulfamethoxazole. Rev Infect Dis 1982;4:533. 4. Hammerschlag MR. Activity of trimethoprim-sulfamethoxazole against Chlamydia trachomatis in vitro. Rev Infect Dis 1982;4:500. 5. Sattler RF, Remington JS. Intravenous trimethoprim-sulfamethoxazole therapy for Pneumocystis carinii pneumonia. Am J Med 1981;70:1215. 6. Young LS. Trimethoprim-sulfamethoxazole in the treatment of adults with pneumonia due to Pneumocystis carinii. Rev Infect Dis 1982;4:608. 7. Zinner SH, Mayer KH. Sulfonamides and trimethoprim. In: Mandel GL, Bennett JE, Dolin R, eds. Principles and practice of infectious disease. Philadelphia: Churchill Livingstone, 2000:394–403.
24 ANTIBIOTIC PROPHYLAXIS IN OBSTETRICS AND GYNECOLOGY Infectious Diseases of the Female Genital Tract
24 ANTIBIOTIC PROPHYLAXIS IN OBSTETRICS AND GYNECOLOGY Uses in Obstetrics Cesarean Delivery Other Obstetric Considerations Uses in Gynecology Vaginal Hysterectomy Abdominal Hysterectomy Therapeutic Abortion Gynecologic Oncology Other Gynecologic Procedures Prophylaxis of Bacterial Endocarditis Recommendations Regimen Chapter References
Because infections have been nearly an every day encounter in a busy obstetric-gynecologic practice, there has been great interest in antibiotic prophylaxis for operative procedures in the specialty. Over the last three decades, a large body of data has accumulated to allow for the establishment of practical recommendations for use in many procedures. Quality standards have very recently been published for antimicrobial prophylaxis in all surgical procedures (1). In previous editions of this book, this chapter included detailed tables listing numerous publications and giving specific results of these individual tables. Because these studies are now aging and because antibiotic prophylaxis is so well established for many procedures, we have decided to eliminate these detailed and cumbersome tables and replace them with practice recommendations. There are special conditions regarding the use of antibiotic prophylaxis in an obstetric-gynecologic population. First, nearly all obstetric and most gynecologic patients are healthy and free of serious underlying disorders. Second, because the lower genital tract is a contaminated field, operation through or adjacent to this field leads to a moderate to high incidence of infection (in the absence of antibiotic prophylaxis), but serious infection measured by abscess or death is unusual. Third, use of certain antimicrobials for prophylaxis in pregnancy is often contraindicated because of the potential for adverse effects on the fetus, newborn, or mother. In this chapter, antibiotic prophylaxis is defined as the use of antibiotics for the prevention of infection in the absence of current signs or symptoms of infection.
USES IN OBSTETRICS In obstetric patients, most attention has been directed at prophylaxis of the patient undergoing cesarean delivery. Antibiotic prophylaxis for prevention of perinatal neonatal group B streptococcal infection and for use in patients having premature or prolonged rupture of the fetal membranes is discussed in Chapter 3 (Group B Streptococci) and Chapter 19 (Subclinical Infection as a Cause of Premature Labor), respectively. Cesarean Delivery As noted in Chapter 20 (Postpartum Infection), cesarean delivery is accompanied by more frequent and more serious puerperal infections. Indeed, cesarean delivery is the single most important risk factor for maternal postpartum infection. Patients undergoing cesarean delivery have a 5-fold to 20-fold greater risk for puerperal infection than patients having vaginal delivery. Patients having nonelective cesarean delivery (i.e., in labor with or without membrane rupture) are at more risk than patients having electively scheduled procedures. Particularly among indigent populations, the risks in this subgroup have been reported as 45% to 85%. Labor, membrane rupture, and vaginal examination increase postpartum infection, probably because they allow ascent of bacteria into the amniotic cavity before surgery. Since the early 1970s, more than 30 randomized placebo-controlled trials have been reported on the use of prophylactic antibiotics representative in cesarean section (2,3). In one analysis, prophylactic antibiotics were found to decrease the overall rate of “serious infection” in patients undergoing cesarean section (emergency and elective combined) dramatically and significantly (typical odds ratio [OR], 0.24; 95% confidence interval [CI], 0.18–32) (2). Prophylactic antibiotics have been shown to have profound and consistent effects in reducing endometritis (typical OR, 0.25; 95% CI, 0.22–0.29) and wound infection (typical OR, 0.35; 95% CI, 0.28–0.44). Compared with placebo, both broad-spectrum penicillins and cephalosporins are highly effective (typical ORs, 0.33 and 0.31, respectively), whereas metronidazole (in five studies) showed no significant decrease (typical OR 0.72; 95% CI, 0.48–1.08). Several studies have compared one antibiotic with another. There were no statistically significant differences in infection rates, but there were relatively few patients in these studies, and the absolute differences in infection rates were small (3% to 8%). Canadian investigators compared cefoxitin with cefazolin in nonelective cesarean sections and found no significant differences in rate of genital tract infection or in hospital stay (4). When the efficacy of broad-spectrum penicillins has been compared directly with that of cephalosporins, there has been no significant difference (typical OR, 0.89; 95% CI, 0.60–1.32) (2). Studies comparing administration of the prophylaxis before versus after cord clamping
have shown similar effectiveness (5,6). Single-dose prophylaxis has been compared with multiple doses. Several individual studies have found equivalent efficacy between one dose and two or three doses of antibiotic for prophylaxis. In an analysis of nine studies comparing single and multiple doses, postoperative febrile morbidity was more common with single-dose prophylaxis (typical OR, 1.36; 95% CI, 0.95–1.95) (2), but the CI includes 1.0. Hemsell (3) noted that cesarean section after prolonged rupture of the membranes, presumptively associated with a large intrauterine inoculum, “may be the one procedure for which single-dose prophylaxis is not so effective as two doses administered at an interval shorter than the 4 to 6 hours.” The recent educational bulletin from the American College of Obstetricians and Gynecologists concluded that single-dose prophylaxis usually is sufficient. We support this position and would reserve two- or three-dose courses of prophylaxis for selected cases such as those with ruptured membranes for more than 12 hours, as noted by Hemsell (6). As an alternative to parenteral administration of prophylactic antibiotics, there was a flurry of interest in intraoperative irrigation of the uterus and peritoneal cavity with antibiotic-containing solution. Publications 15 to 20 years ago established that this route is often equivalent to, but in some cases was less effective than, intravenous prophylaxis. Over the last 15 years, interest in this approach has waned (7,8,9 and 10). Use of prophylaxis routinely (including in “low-risk” patients) remains controversial. The Oxford group concluded that giving antibiotics routinely at cesarean section would reduce costs by between £1,300 and £3,900 per 100 cesarean deliveries (at 1988 British prices), based on a literature review (11). In the United States, Ehrenkranz et al. (12) performed prospective and retrospective studies of women at low risk (defined as those having a scheduled procedure without an urgent indication, with any duration of ruptured membranes being 12 hours or less, and among patients in community hospitals). Absence of antibiotic prophylaxis was associated with endometritis (p < 0.013) or endometritis with wound infection (p < 0.01). Without prophylaxis, such infections occurred in 3.7% (37 of 957) versus 0.9% (8 of 906) with prophylaxis. It was estimated that routine antibiotic prophylaxis in low-risk cesarean sections “could lead to an annual national savings of approximately $9 million” (12). On the other hand, Howie and Davey (13), in editorial for the Oxford group's report, noted that prophylactic antibiotics were “important but not always necessary.” The editorialists noted that the Oxford group's claim must be tested in prospective trials and that there have been both direct adverse effects of prophylactic antibiotics and development of resistant bacteria. It is one thing to draw conclusions in a metaanalysis, but quite another to decide on an individual mother. Similarly, Hemsell (3) concluded that a woman with intact membranes, not in labor, undergoing electively scheduled cesarean section “probably does not require prophylaxis.” The recent position of the American College of Obstetricians and Gynecologists (6) is that “prophylaxis is not recommended routinely in low-risk patients because of concerns about adverse side effects, bacteriologic shifts toward resistant organisms, and relaxation of standard infection control measures and proper operative technique. We agree with these latter philosophies. Untoward effects such as changes in flora and direct toxic reactions may accompany
the use of antibiotics for prophylaxis (14). Systematic investigations have detected significant changes in antimicrobial flora after prophylactic antibiotics (14). Overall, there were decreases in highly susceptible organisms and increases in enterococci and Enterobacteriaceae organisms. Although many of the Enterobacteriaceae organisms were susceptible to the prophylactic agent, some isolates of Pseudomonas sp also appeared. In these cases, the bacterial changes were without consequence. Recently Newton and Wallace (15) reported the results of prophylactic antibiotics on endometrial flora in women with postcesarean endometritis. Patients who received cefazolin prophylaxis had a significant increase in enterococci (p < 0.015) and a significant decrease in Proteus sp (p < 0.05) when samples were taken from the endometrium by a sheathed aspiration technique at the time of endometritis. In contrast, patients who received ampicillin prophylaxis had a significant increase in Mycoplasma species (p < 0.5), Klebsiella pneumoniae (p < 0.0001), Escherichia coli (p = 0.04), and any aerobic Gram-negative rod (p = 0.003). Ampicillin prophylaxis was associated with a decrease in Prevotella bivia (formerly Bacteroides bivius) (p < 0.05) and any anaerobic isolates (p < 0.01). Patients who received cephalosporin prophylaxis, followed by a cephalosporin for treatment, had significantly more wound infections than those who received other prophylaxis treatment combinations (19% vs. 16%; p < 0.01). Considerable evidence shows that antibiotic use for prophylaxis alters endometrial flora and has the potential to influence subsequent response rates to therapy (15). In patients who develop infection after prophylaxis, it is essential to obtain appropriate cultures to guide antibiotic therapy. Further, the antibiotic use for prophylaxis should not be used to treat a subsequent infection when one develops. In addition, extended-spectrum antibiotics should not be used for prophylaxis but should be reserved for treatment (6). Knowledge of the shifts in genital tract flora precipitated by antibiotic prophylaxis should considered when selecting antibiotics to be used for therapy of postprophylaxis endometritis. Direct toxic effects are unlikely. No serious allergic or toxic reactions were noted among 1,443 patients receiving prophylaxis in 26 studies (5). Less serious reactions such as rash were reported on occasion. However, several cases of fatal anaphylactic reaction to prophylactic antibiotics were reported in orthopedic patients (16), and two cases of fatal pseudomembranous enterocolitis have been attributed to a combination of prophylactic therapeutic antibiotics (17). More recently, pseudomembranous colitis has been reported in 15 women having short-course prophylaxis (18,19 and 20). One case followed cefazolin prophylaxis; the other 14 followed cefoxitin prophylaxis. Block et al. (18) found Clostridium difficile–associated colitis to be significantly more common after cefoxitin (9 of 162) than after other antibiotics (0 of 8; p = 0.02). For patients who have documented immediate hypersensitivity reactions to penicillin, antibiotic choices for prophylaxis are limited. In such cases, cephalosporins are contraindicated, as are all penicillin-type antibiotics. One recommendation is for the use of single-dose clindamycin (900 mg), perhaps with a single dose of gentamicin (1.5 to 2.0 mg/kg) (6). Vancomycin may be used for prophylaxis of cesarean section endometritis in patients with an immediate hypersensitivity reaction to penicillin (Box 24.1).
Box 24.1. Recommendations for Use Of Prophylactic Antibiotics in Cesarean Section 1. Antibiotic prophylaxis is recommended for women undergoing nonelective cesarean section because such antibiotic use reduces the risk of endometritis and wound infections. Use of prophylaxis in this setting appears to be cost-effective. 2. Prophylaxis is not recommended routinely in low-risk patients such as those with documented low infection rates after undergoing an electively scheduled cesarean section. 3. Preferred antibiotics include a first-generation cephalosporin such as cefazolin (1 g intravenously) or ampicillin (1 to 2 g intravenously). Newer extended-spectrum cephalosporins or penicillins are no more effective and add cost to the prophylaxis. 4. Single-dose prophylaxis is usually sufficient, but in selected circumstances such as cesarean section after prolonged rupture of the membranes (e.g., 12 hours), a regimen of two- or three-dose prophylaxis is supported. Current information suggests that additional intraoperative doses of an antibiotic for prophylaxis should be given at intervals of one or two times the half-life of the antibiotic to maintain adequate levels of the antibiotic throughout the surgical procedure. Half-lives of representative antibiotics (in patients with normal renal function) are as follows: for cefazolin, 1.8 hours; cefoxitin, 60 minutes; cefotetan, 4 hours; and clindamycin, 3 hours (1). 5. The antibiotic used for prophylaxis should be initiated immediately after cord clamping (unless the rationale is to prevent group B streptococcal perinatal infection). This timing of antibiotic prophylaxis is equally effective as regimens beginning prophylaxis before cord clamping, and this regimen avoids direct and indirect adverse effects on the newborn. 6. Antibiotics used for prophylaxis dramatically shift upper genital tract flora. Extended-spectrum antibiotics should not be used for prophylaxis. For patients who develop genital tract infection after prophylaxis, a culture for aerobes should be obtained to direct subsequent antibiotic therapy and to direct antibiotic policies in a given hospital. Antibiotics used for prophylaxis should not be used for therapy in the same patient. If a cephalosporin was used for prophylaxis, a treatment regimen active against enterococci should be considered. If ampicillin is used for prophylaxis, a treatment regimen active against Klebsiella organisms should be considered. 7. For patients who have immediate hypersensitivity reactions to penicillin, antibiotics for prophylaxis are limited. One recommendation is to use a single dose of clindamycin (900 mg), either with or without gentamicin in a dose of 1.5 to 2.0 mg/kg. If there is also a contraindication to clindamycin, vancomycin may be used. The Infectious Diseases Society of America recommends vancomycin for prophylaxis instead of cefazolin or alternative cephalosporins in patients who are allergic to cephalosporins. Further, the recommendation is that because vancomycin provides no activity against Gram-negative bacilli, another antibiotic with Gram-negative activity should be added to this regimen.
clindamycin, vancomycin may be used. The Infectious Diseases Society of America recommends vancomycin for prophylaxis instead of cefazolin or alternative cephalosporins in patients who are allergic to cephalosporins. Further, the recommendation is that because vancomycin provides no activity against Gram-negative bacilli, another antibiotic with Gram-negative activity should be added to this regimen. Such an alternative includes aztreonam or aminoglycoside (1). 8. Direct adverse effects of antibiotics used for prophylaxis have been reported, including life-threatening complications such as anaphylaxis and pseudomembranous colitis. 9. Use of prophylactic antibiotics must not result in relaxation of standard infection control measures. 10. Antibiotics administered by irrigation are not more effective than those given by intravenous injection.
A common concern about widespread use of prophylaxis is a dangerous relaxation of standard infection control measures. Clearly, hand-washing, appropriate isolation techniques, proper disposal of infected materials and dressings, and changing of soiled scrub suits remain important elements in the control of infections and cannot be replaced capriciously by antibiotic prophylaxis. Recommendations for use of antibiotics in patients undergoing cesarean section are provided in Table 24.1 and Box 24.1.
TABLE 24.1. SUMMARY OF EFFICACY OF ANTIBIOTIC PROPHYLAXIS IN OBSTETRIC AND GYNECOLOGIC PROCEDURES TO PREVENT PELVIC INFECTION
Other Obstetric Considerations For a full discussion of the use of prophylactic antibiotics for prevention of group B streptococcal perinatal infection and in premature rupture of the membranes, see Chapter 3 (Group B Streptococci) and Chapter 19 (Subclinical Infection as a Cause of Premature Labor), respectively.
USES IN GYNECOLOGY Studies in gynecologic patients have focused mainly on vaginal and abdominal
hysterectomy, but there have been additional studies of other procedures. Table 24.1 summarizes the efficacy of antibiotic prophylaxis. Vaginal Hysterectomy The risk of postoperative infection in patients undergoing vaginal hysterectomy varies widely. In the Professional Activities Study, 38% of 3,500 patients having vaginal hysterectomy had fever of more than 101°F (21). Premenopausal women are at high risk for infection after vaginal hysterectomy. Women with a hysterectomy within 24 to 72 hours after cervical conization also encounter a higher rate of postoperative infection (22). Among double-blind placebo-controlled studies, there is a consistent finding of decreased postoperative infection (most studies showing statistically significant differences) with use of antibiotic prophylaxis. Polk (23) calculated an overall preventive fraction of 82% in 11 studies of short-course prophylaxis. Numerous studies have compared different antibiotics for prophylaxis in patients with vaginal or abdominal hysterectomy. These studies have compared short courses (single dose to 12 hours vs. long courses [48 to 72 hours]), different antibiotics in the same or different dosing regimens, the same antibiotic in differing regimens, and antibiotics versus suction drainage. Few of these studies showed any significant differences in rates of postoperative infection. The findings of these well-designed studies are remarkably consistent in showing statistically significant and clinically impressive decreases in postoperative infection. Abdominal Hysterectomy Infection after abdominal hysterectomy appears to be less frequent than infection after vaginal hysterectomy, possibly due to less contamination from the vagina (21). In a metaanalysis, “serious infections” (including abdominal wound infection, pelvic cellulitis, pelvic abscess, vaginal cuff abscess, and septicemia) developed in 21% (373 of 1,768) of patients undergoing abdominal hysterectomy without prophylaxis (24). Of numerous double-blind placebo-controlled studies of antibiotic prophylaxis in patients who underwent abdominal hysterectomy, few individual studies show a significant decrease in postoperative infection, but nearly all show an absolute decrease. Polk (23) calculated that in studies using short-course prophylaxis, 49% of postoperative infections are prevented. A recent metaanalysis of rigorously conducted trials found a significant reduction in “serious postoperative infection” (as defined previously) among patients who received antibiotic prophylaxis compared with those who did not (9.0% [166 of 1,836] vs. 21.1% [373 of 1,768]; p = 0.00001). Individual antibiotics with demonstrated significant decreases were cefazolin (p = 0.0002), metronidazole (p = 0.015), and tinidazole (a drug similar to metronidazole, p = 0.034) (24). As for vaginal hysterectomy, many studies have compared different regimens, with little overall difference among regimens. Thus, broader spectrum, more expensive antibiotics for prophylaxis are no more effective than older antibiotics such as cefazolin. Further, because single-dose cefazolin prophylaxis appears to be equivalent to three-dose regimens, we do not recommend use of newer longer acting agents for prophylaxis. As pointed out by Shapiro et al. (25), the benefit-cost ratio is decreased considerably by the use of the more expensive antibiotics for prophylaxis. Under special conditions, such as
long cases or those with high blood loss, second doses of the prophylactic antibiotic may be beneficial. Adverse Effects of Prophylaxis in Hysterectomy Among patients receiving cephalosporin prophylaxis for abdominal or vaginal hysterectomy, there was a shift toward more resistant isolates. Changes in flora have also been noted in patients who received short-course prophylaxis for hysterectomy, but these changes were similar to those in patients who received placebo. Other adverse effects, including rashes and abnormalities in chemistry or hematology studies, have been reported infrequently. Alternatives to Prophylaxis in Hysterectomy Significant decreases in postoperative infections had been demonstrated by the use of suction drainage through a T-tube catheter (26). Later studies, however, showed prophylactic antibiotics to be superior to T-tube suction drains (27). Patients with an abnormal vaginal flora are more likely to have postoperative infection. Thus, if a condition such as bacterial vaginosis is detected preoperatively, it should be treated either topically or orally before the patient is admitted to the hospital, whether the condition is symptomatic or asymptomatic (28). Recommendations for Hysterectomy Please see Box 24.2. Therapeutic Abortion The risk of upper genital tract infection after therapeutic abortion ranges from 5% to 20% (29). Retrospective reports from the United States had suggested that the incidence might be less than 1%. It is recognized that there were long-term sequelae of postabortal infection including chronic pelvic pain and infertility. It may be argued that because most women undergoing suction abortion in the United States are young, unmarried, and nulliparous, there is an urgency in preventing upper genital tract infections, which can adversely effect fertility. Problems with earlier studies of prophylactic antibiotics for therapeutic abortions included the rather vague definition of infection. It was noteworthy that nearly all of the infections were minor or minimal (30). Additional problems were that side effects from the antibiotics were common, particularly when doxycycline was used for prophylaxis. For example, vomiting developed in 18% of patients (31). More recent views support universal prophylaxis for therapeutic abortion (28,29). Organisms most likely to cause postabortal endometritis include the sexually transmitted organisms Neisseria gonorrhoeae and Chlamydia trachomatis, as well as the array of aerobic and anaerobic organisms found in the pelvis (28). Accordingly, regimens to be used for prophylaxis include doxycycline, ofloxacin, and ceftriaxone (28). The course of prophylaxis should be short and may be given orally, such as 200 mg of doxycycline before and 12 hours after the procedure.
Erythromycin orally and metronidazole orally have also been used (29). A recent metaanalysis by Sawaya et al. (29) assessed 12 studies and reported an overall summary relative risk for developing postabortal upper genital tract infection in women receiving antibiotic prophylaxis of 0.58 (95% CI, 0.47–0.72) compared with women receiving placebo. In all subgroups of women, there was a significant protective effect. In high-risk women, such as those with a history of pelvic inflammatory disease, there was a summary relative risk of 0.56 (95% CI, 0.37–0.84). Women who had positive culture results for chlamydiae at the time of abortion had a summary relative risk of 0.38 (95% CI, 0.15–0.92). Even in women with no risk factors, the relative risk was protective at 0.65 (95% CI, 0.47–0.90). Thus, it is recommended that routine perioperative antibiotics be used in patients having abortion in the United States to prevent approximately half of the cases of postprocedural infection.
Box 2. Recommendations for Antibiotic Prophylaxis with Hysterectomy 1. Short-course prophylaxis significantly reduces the risk of pelvic infection in patients undergoing vaginal or abdominal hysterectomy. Antibiotic prophylaxis is likely to be more effective in lowering infection after vaginal hysterectomy than after abdominal hysterectomy. 2. Antibiotic selection for prophylaxis for abdominal hysterectomy includes primarily first-generation cephalosporins such as cefazolin (1). An alternative antibiotic to be used is ampicillin (28). Some experts also recommend second- and third-generation cephalosporins such as cefoxitin, cefotetan, or ceftizoxime and broader spectrum penicillin such as mezlocillin and piperacillin (28). 3. As with cesarean section, prophylactic antibiotics should be given as a single dose for most procedures. If the operation lasts more than approximately 3 hours, or if the blood loss exceeds 1,500 mL, a second dose may be beneficial (28). 4. Shifts in bacteria undoubtedly occur after antibiotic prophylaxis, as in cesarean section. Accordingly, recommendations for performing cultures and for therapeutic strategies should be the same as recommended under cesarean sections. Specifically, cephalosporins are likely to lead to emergence of resistant enterococci, resistant E. coli, and Bacteroides fragilis; penicillins tend to select for resistant Gram-negative bacteria. 5. It is important to monitor infection rates within each hospital to assess changes in effectiveness of an antibiotic regimen (28). 6. In patients who are allergic to penicillins or cephalosporins, alternative antibiotics for prophylaxis include clindamycin, doxycycline, metronidazole (28), or in special cases, vancomycin (1). 7. Again, as with cesarean section, antibiotic prophylaxis may cause direct adverse effects. 8. As with cesarean section, good intraoperative technique remains essential.
Gynecologic Oncology Antibiotic prophylaxis is widely used in major surgery for gynecologic cancer, but randomized trials have less extensive data than in hysterectomy for benign indications (32,33,34,35 and 36). Current data suggest that prophylaxis is effective for radical surgery but that individual determinations of the need for prophylaxis should be made. When used in such cases, recommendations are the same as those for other hysterectomy procedures. Overall, it is likely that infections are decreased, because there is a substantial likelihood of a beta error in individual studies because of the small sample size.
OTHER GYNECOLOGIC PROCEDURES After hysterosalpingography, the risk of pelvic inflammatory disease is 0.3% to 3.4% (3). Tests for sexually transmitted diseases should be performed before the procedure, and appropriate treatment given. There are no randomized blind trials, and indeed they would be difficult to complete given the low overall infection rate. Nevertheless, prophylaxis has been advised for decades presumably based on the observation that bacteria are commonly carried with the dye into the endometrial cavity, through the fallopian tubes and into the peritoneum (28). Before intrauterine device (IUD) insertion, tests for sexually transmitted diseases should be performed and appropriate treatment given. In a randomized trial of doxycycline versus placebo, there was no significant reduction in postinsertion infection (1.9% vs. 1.3%; p = 0.17), but those with prophylaxis had fewer unscheduled postinsertion visits (3). An additional randomized controlled trial was recently reported among nearly 2,000 patients taking either azithromycin (500 mg) or placebo 1 hour before insertion of a copper T-shaped IUD. There was no significant difference in the rate of IUD removal for any reason (other than partial expulsion) (3.8% in the antibiotic group and 3.4% in the placebo group). The two groups had a similar number of subsequent visits, and in the 90 days after IUD insertion, only one woman from each group had a diagnosis of salpingitis. Thus, the risk of upper genital tract infection after IUD insertion was negligible with or without the administration of prophylactic antibiotics (37). One report indicated that changing the time of insertion from during menses to within 2 days of predicted ovulation eliminated “inflammatory-type reactions.” Overall, prophylactic antibiotics are not recommended for routine use with IUD insertion. For infertility surgery, prophylactic antibiotics are commonly used, but there are no prospective randomized trials. It is reasonable to use a single dose of cefazolin for tubal reconstructive surgery and a single oral dose of doxycycline for laparoscopy with hydrotubation (3). For clean procedures such as myomectomy, ovarian cystectomy, and resection of endometriosis, prophylaxis is not recommended (3). Bacteriuria was not decreased after combined urodynamics and cystourethroscopy among women taking a 1-day course of nitrofurantoin versus placebo (38). Similarly, among women having curettage for incomplete abortion, prophylactic doxycycline did not decrease the rate of
postoperative febrile morbidity (39).
PROPHYLAXIS OF BACTERIAL ENDOCARDITIS Bacteremia develops in perhaps 1% to 5% of women during uncomplicated delivery and in an undetermined percentage of women undergoing pelvic surgery. Bacteremia has been found in 3% to 20% of obstetric patients with infection. In 1997, the American Heart Association issued revised guidelines for antibiotic prophylaxis to prevent endocarditis (40). These recommendations included a number of changes. These include stratification of cardiac conditions into high-, moderate-, and negligible-risk groups based on the potential outcome if endocarditis develops; identification of procedures that may cause bacteremia and for which prophylaxis is recommended; development of an algorithm to more clearly define when prophylaxis is recommended in patients with mitral valve prolapse; and simplification of a prophylactic regimen for patients with gastrointestinal and genitourinary procedures. Cardiac conditions for which endocarditis prophylaxis is either recommended or not recommended appear in Table 24.2. A clinical approach to determine the need for prophylaxis in patients with suspected mitral valve prolapse was provided. Overall, if there is a murmur of mitral regurgitation, then prophylaxis is indicated. If the presence or absence of mitral regurgitation is either not determined or not known, and if confirmation is not available, prophylaxis is recommended if there is an immediate need for the procedure. Otherwise, the patient should be referred for evaluation and a determination of the need for prophylaxis should be based on detection of the murmur by a cardiologist, by echocardiography or by Doppler flow studies.
TABLE 24.2. CARDIAC CONDITIONS ASSOCIATED WITH ENDOCARDITIS
Recommendations The revised recommendations show that endocarditis prophylaxis is recommended with
cystoscopy and with urethral dilation. However, endocarditis prophylaxis is not recommended routinely with vaginal hysterectomy, vaginal delivery, cesarean section, or with uninfected tissues with dilation and curettage, therapeutic abortion, sterilization procedures, or insertion or removal of IUDs. However, the committee continues to note that prophylaxis is optional for high-risk patients (Table 24.2) undergoing vaginal hysterectomy or vaginal delivery. Antibiotic therapy is, of course, indicated in the treatment of patients with infected products of conception undergoing dilation and curettage or for the removal of an IUD in the presence of pelvic inflammatory disease (Table 24.3).
TABLE 24.3. OBSTETRIC GYNECOLOGIC PROCEDURES AND ENDOCARDITIS PROPHYLAXIS
Regimen Endocarditis prophylaxis during pelvic procedures is directed primarily against enterococci (e.g., Streptococcus faecalis). Although Gram-negative organisms are among the most common microbes causing bacteremia in this population, these organisms rarely cause endocarditis. Table 24.4 lists the recommended regimens.
TABLE 24.4. PROPHYLACTIC REGIMEN FOR GENITOURINARY PROCEDURES
In patients with significantly compromised renal function, it may be necessary to modify the dose of antibiotics used. CHAPTER REFERENCES 1. Gross PA, Barrett TL, Dellinger EP, et al. Infectious Diseases Society of America quality standards for infectious diseases: purpose of quality standards for infectious diseases. Clin Infect Dis 1994;18:421–426. 2. Enkin M, Enkin E, Chalmers I, et al. Prophylactic antibiotics in association with caesarean section. In: Chalmers I, Enkin M, Kearse MJNC, eds. Effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1989:1246–1269. 3. Hemsell DL. Prophylactic antibiotics in gynecologic and obstetric surgery. Rev Infect Dis 1991;13[Suppl 10]:S821–S841. 4. Silver HG, Forward KR, Livingstone RA, et al. Multicenter comparison of cefoxitin versus cefazolin for prevention of infectious morbidity after nonelective cesarean section. Am J Obstet Gynecol 1983;145:158. 5. Swartz WH, Grolle K. The use of prophylactic antibiotics in cesarean section, a review of the literature. J Reprod Med 1981;26:595. 6. Antimicrobial therapy for obstetric patients. American College of Obstetricians and Gynecologies Educational Bulletin 1998;245. 7. Long WH, Rudd EG, Dillon MB. Intrauterine irrigation with cefamandole nafate solution of cesarean section: a preliminary report. Am J Obstet Gynecol 1980;138:755. 8. Rudd EG, Cobey EA, Long WH, et al. Prevention of endomyometritis following cesarean section. Paper presented at: the Armed Forces district meeting of the American College of Obstetricians and Gynecologists; October 1981; Phoenix, AZ. 9. Elliott JP, Flaherty JF. Comparison of lavage or intravenous antibiotics at cesarean section. Obstet Gynecol 1986;67:29. 10. Gonen R, Samberg I, Levinski R, et al. Effect of irrigation of intravenous antibiotic prophylaxis on infectious morbidity at cesarean section. Obstet Gynecol 1986;67:545. 11. Mugford M, Kingston J, Chalmers I. Reducing the incidence of infection after caesarean section: implications of prophylaxis with antibiotics for hospital resources. Br Med J 1989;299:1003–1006. 12. Ehrenkranz NJ, Blackwelder WC, Pfaff SJ, et al. Infections complicating low-risk cesarean sections in community hospitals: efficacy of antimicrobial prophylaxis. Am J Obstet Gynecol 1990;162:337–343. 13. Howie PW, Davey PG. Prophylactic antibiotics and caesarean section. Br Med J 1990;300:2–3. 14. Gibbs RS, Weinstein AJ. Bacteriologic effects of prophylactic antibiotics in cesarean section. Am J Obstet Gynecol 1976;130:226. 15. Newton ER, Wallace PA. Effects of prophylactic antibiotics on endometrial flora in women with postcesarean endometritis. Obstet Gynecol 1998;92:262–268. 16. Spruill FG, Minette LJ, Sturner WO. Two surgical deaths associated with cephalothin. JAMA 1974;229:440. 17. Ledger WJ, Puttler OL. Death from pseudomembranous enterocolitis. Obstet Gynecol 1975;45:609. 18. Block BS, Mercer LJ, Ismail MA, et al. Clostridium difficile–associated diarrhea follows perioperative prophylaxis with cefoxitin. Am J Obstet Gynecol 1985;153:835. 19. McNeeley SG, Anderson GD, Sibai BM. Clostridium difficile colitis associated with single-dose cefazolin prophylaxis. Obstet Gynecol 1985;66:737. 20. Arsura EL, Fazio RA, Wickremesinghe PC. Pseudomembranous colitis following prophylactic antibiotic use in primary cesarean section. Am J Obstet Gynecol 1985;151:87.
21. Ledger WJ, Child MA. The hospital care of patients undergoing hysterectomy: an analysis of 12,026 patients from the Professional Activities Study. Am J Obstet Gynecol 1973;117:423. 22. Forney JP, Morrow CP, Townsend DE, et al. Impact of cephalosporin prophylaxis in conization-vaginal hysterectomy morbidity. Am J Obstet Gynecol 1976;125:100. 23. Polk BF. Antimicrobial prophylaxis to prevent mixed bacterial infections. J Antimicrob Chemother 1981;8[Suppl]:115. 24. Mittendorf R, Aronson MP, Berry RE, et al. Avoiding serious infections associated with abdominal hysterectomy: a meta-analysis of antibiotic prophylaxis. Am J Obstet Gynecol 1993;169:1119–1124. 25. Shapiro M, Schoenbaum SC, Tager IB, et al. Benefit-cost analysis of antimicrobial prophylaxis in abdominal and vaginal hysterectomy. JAMA 1983;249:1290. 26. Swartz WH, Tanaree P. Suction drainage as an alternative to prophylactic antibiotics for hysterectomy. Obstet Gynecol 1975;45:305. 27. Poulsen HK, Vorel J, Olsen H. Prophylactic metronidazole or suction drainage in abdominal hysterectomy. Obstet Gynecol 1984;63:291. 28. Antibiotics and gynecologic infections. American College of Obstetricians and Gynecologies Educational Bulletin 1997;237. 29. Sawaya GF, Grady D, Kerlikowski K, et al. Antibiotics at the time of induced abortion: the case for universal prophylaxis based on a meta-analysis. Obstet Gynecol 1996;87:884–890. 30. Hodgson JE, Major B, Portman K, et al. Prophylactic use of tetracycline for first trimester abortions. Obstet Gynecol 1975;45:574. 31. Darl E, Stralin EB, Nilsson S. The prophylactic effect of doxycycline on postoperative infection rate after first-trimester abortion. Obstet Gynecol 1987;70:755. 32. Sevin BU, Ramos R, Lichtinger M, et al. Antibiotic prevention of infections complicating radical abdominal hysterectomy. Obstet Gynecol 1984;64:539. 33. Marsden DE, Cavanagh D, Wisniewski BJ, et al. Factors affecting the incidence of infectious morbidity after radical hysterectomy. Am J Obstet Gynecol 1985;152:817. 34. Micha JP, Kucera PR, Birkett JP, et al. Prophylactic mezlocillin in radical hysterectomy. Obstet Gynecol 1987;69:251. 35. Sevin BU, Ramos R, Gerhardt RT. Comparative efficacy of short-term versus long-term cefoxitin prophylaxis against postoperative infection after radical hysterectomy: a prospective study. Obstet Gynecol 1991;77:729–734. 36. Orr JW, Sisson PF, Patsner B. Single-dose antibiotic prophylaxis for patients undergoing extended pelvic surgery for gynecologic malignancy. Am J Obstet Gynecol 1990;162:718–721. 37. Walsh T, Grimes D, Grezieres R, et al. Randomised controlled trial of prophylactic antibiotics before insertion of intrauterine devices. Lancet 1998;351:1005–1008. 38. Cundiff GW, McLennan MT, Bent AE. Randomized trial of antibiotic prophylaxis for combined urodynamics and cystourethroscopy. Obstet Gynecol 1999;93:749–752. 39. Prieto JA, Eriksen JL, Blanco JD. A randomized trial of prophylactic doxycycline for curettage in incomplete abortion. Obstet Gynecol 1995;85:692–696. 40. Dajani AS, Bisno AL, Chung KJ. Prevention of bacterial endocarditis. Recommendations by the American Heart Association. JAMA 1997;277:1794–1801.
25 IMMUNIZATION Infectious Diseases of the Female Genital Tract
25 IMMUNIZATION General Principles of Immunization Historical Perspective Safety and Efficacy of Vaccines Immunization Programs for Adults General Contraindications to Vaccines/Immune Therapy Diphtheria and Tetanus Toxoid and Pertussis Background Indications Specific Contraindications and Complications Dosage Use During Pregnancy or Lactation Measles, Mumps, and Rubella Background Indication Specific Contraindications/Complications Dosage Use During Pregnancy or Lactation Hepatitis B Vaccine Background Indications Specific Contraindications/Complications Dosage Use During Pregnancy or Lactation Hepatitis A Vaccine Background Indication Contraindications/Complications Dosage Use During Pregnancy or Lactation Varicella Vaccine Background Indications Specific Contraindications and Complications Dosage Use During Pregnancy and Lactation Pneumococcal Vaccine Indications Specific Contraindications and Complications Dosage Use During Pregnancy or Lactation Influenza Vaccine Background
Indications Specific Contraindications and Complications Dosage Use During Pregnancy and Lactation Lyme Disease Vaccine Indications Specific Contraindications and Complications Dosage Use During Pregnancy and Lactation Immune Therapy Immune Globulin Hepatitis B Immune Globulin Varicella-Zoster Immune Globulin Use of Vaccines: Special Circumstances Pregnancy International Travel Occupational Exposure Chapter References
Immunization of adults has not received the same high priority as immunization of children (1,2,3 and 4). Paradoxically, in the United States, deaths from vaccine-preventable diseases occur predominantly in adults, with an estimated 50,000 to 70,000 adults dying each year from such diseases (1,2,3,4 and 5). Predominantly, mortality is due to pneumonococcal infection, influenza, and hepatitis B (Table 25.1). In fact, vaccine-preventable deaths exceed those resulting from automobile accidents or acquired immunodeficiency syndrome (AIDS) (2,3 and 4). Factors that contribute to the poor record of adult immunization include (a) concerns about safety and efficacy of vaccines among the general public and health care workers; (b) uncertainty about specific recommendations; (c) liability concerns; (d) inadequate reimbursement for immunizations; (e) a poorly informed public; and (f) an independently developed system for immunization of adults (2,4,5).
TABLE 25.1. ESTIMATES OF THE EFFECT OF FULL USE OF THE VACCINES ADVOCATED FOR ADULTS
Of the more than 50 biologic products available in the United States for immunization, seven major vaccines are designed for routine use in adults (Table 25.2) (1,2,3,4 and 5). It is important that obstetrician-gynecologists be involved in providing immunization of adults. Unfortunately, vaccination of adult patients has been inadequately addressed by obstetrician-gynecologists and other primary care providers (6,7). For example, the American College of Physicians' Task Force on Immunizations has recommended that age 50 be established as the time to review preventive health measures including an emphasis on risk factors that identify patients in need of pneumococcal and hepatitis B vaccines and initiation of annual influenza immunization (8). This chapter discusses the general principles of immunization, specific details for use of these seven vaccines, and special circumstances related to immunizations such as pregnancy, occupational exposure, travel, postexposure immunizations, and immunocompromised states.
TABLE 25.2. MAJOR VACCINES RECOMMENDED FOR ROUTINE USE IN ADULTS
General Principles Of Immunization Immunization is the process by which either immunity is artificially induced or protection from disease is provided (9). Immunization can be active or passive. Active immunization induces the body to develop defenses against infection; it is accomplished with vaccines or toxoids that stimulate the immune system to produce antibodies, cell-mediated immunity, or both, which in turn protects against an infectious disease (9). Passive immunization is a process that provides temporary protection against an infectious agent by administration of exogenously produced antibody (9). The most common methods of passive immunization are transplacental passage of maternal antibodies to the fetus and the use of immunoglobulin (Ig) to prevent specific infectious diseases (9). Immunizing agents include the following: (a) vaccine, a suspension of attenuated live or killed microorganisms or fractions thereof administered to induce immunity; (b) toxoid, a modified bacterial toxin that has been rendered nontoxic but is still capable of producing antitoxin; (c) Ig, a solution containing antibody from human
blood that can provide passive immunization against certain infections (e.g., measles and hepatitis A) or routine protection of immunodeficient patients; and (d) specific Ig, special preparations obtained from donor pools with a high antibody content against a specific disease (e.g., hepatitis B immune globulin [HBIG] and varicella-zoster immune globulin [VZIG]) (9). Historical Perspective Jenner (10) undertook the first scientific attempt to control an infectious disease through deliberate inoculation (11). In 1796, he demonstrated that milkmaids who had contacted cowpox (vaccinia) were immune to smallpox. Subsequently, Jenner inoculated susceptible persons with vesicular fluid from cowpox lesions and induced protection against smallpox (9,10 and 11). This commenced the era of immunization, whose impact on reducing mortality and augmenting population growth is surpassed only by improved sanitation and safe water (11). In the 200 years since Jenner, the introduction of vaccines has controlled the following ten major infectious diseases in at least some areas of the world: smallpox, diphtheria, tetanus, pertussis, yellow fever, poliomyelitis, measles, mumps, rubella, and Haemophilus influenzae type B disease. Vaccination has resulted in the global eradication of smallpox (12) and the elimination of poliomyelitis caused by wild virus in the United States (9,13,14). Great progress has been made worldwide toward the goal of polio eradication (14,15). Recently, the Centers for Disease Control and Prevention (CDC) reported that the western Pacific region of the World Health Organization (WHO) (e.g., Australia, China, Japan, Korea, Philippines, and New Zealand) as of October 2000 has been certified as free of indigenous wild poliovirus transmission (16). No cases of polio due to wild virus have been reported in the Western Hemisphere since 1991 (14) and no new indigenous cases have been detected in the European region of the WHO since 1998 (16,17). The WHO recently suggested that global circulation of poliovirus may be interrupted by 2002 (17). In addition to the eradication of smallpox and poliomyelitis in the United States, a dramatic reduction in the incidence rates of other diseases has occurred with the widespread use of vaccines (Table 25.3) (9). There has been a more than 90% reduction in measles, diphtheria, and rubella in developed countries (18). Although H. influenzae type B (Hib) vaccine has only recently become available on a widespread basis, it has markedly reduced the incidence of this disease in many developed countries (9,19). With more widespread availability, hepatitis B vaccine use in infants should have an impact on disease reduction comparable to that of the other childhood vaccines (9). Similarly, major advances have been made by vaccines against influenza and pneumococcal infection (11).
TABLE 25.3. BASELINE 20th CENTURY ANNUAL MORBIDITY AND 1998 MORBIDITY FROM NINE DISEASES WITH VACCINES RECOMMENDED BEFORE 1990 FOR UNIVERSAL USE IN CHILDREN—UNITED STATES
Safety and Efficacy of Vaccines Although modern vaccines are very effective and safe, all vaccines are associated with some adverse effects and all vaccines are not 100% effective (8). Despite the demonstrated high efficacy of vaccines, controversy has arisen over adverse effects attributed to vaccine use (8). In response to these concerns, the Institute of Medicine (IOM) undertook an assessment of the available data for 9 of the 11 vaccines universally recommended for children and the serious adverse events reported to be associated with these vaccines (20,21 and 22). In most cases, there was insufficient evidence to establish causation (9). Table 25.4 summarizes the IOM findings for adverse events in which data were available to reach a conclusion. Subsequently, the IOM findings were reviewed by the Advisory Committee on Immunization Practices (ACIP), which assessed newer data related to Guillain-Barré syndrome. Based on these newer data, the ACIP did not confirm a causal relationship between Guillain-Barré syndrome and either oral poliovirus (OPV) vaccine, diphtheria toxoid, tetanus toxoid, and pertussis (DTP) vaccine, or tetanus toxoid (T) vaccine (23,24). In addition, recent studies failed to substantiate an increased risk for chronic arthritis among women inoculated with RA27/3 (rubella virus) vaccine (25,26).
TABLE 25.4. SUMMARY OF INSTITUTE OF MEDICINE FINDINGS ON THE RELATIONSHIP OF ADVERSE EVENTS TO INDIVIDUAL VACCINES
In the case of pregnancy, there is no direct evidence of risk to the fetus when pregnant women are inoculated with a particular vaccine (9). However, most live virus vaccines result in viremia, which can lead to infection of the fetus (8). Thus, live virus vaccines are not administered to pregnant women except in special circumstances (9,27). Immunization Programs for Adults In the United States, recommendations for vaccine use are developed by three bodies: (a) The ACIP of the CDC develops recommendations oriented toward the public health arena (28,29); (b) the Committee on Infectious Diseases of the American Academy of Pediatrics (AAP) (Red Book) develops recommendations for use in private pediatric practice (30); and (c) the Task Force on Adult Immunization of the American College of Physicians and the Infectious Diseases Society of America develop recommendations for adults in the private sector (8). Currently, the ACIP, AAP, and the American Academy of Family Physicians collaborate and issue a consensus schedule for childhood immunizations that is updated annually (28). The immunization schedule recommended for adults is divided into three age-groups (Table 25.5) (8,29). In the 18- to 24-year-old age-group, it is recommended that (a) diphtheria and tetanus booster be given at age 16; (b) Measles, mumps, and rubella (MMR) vaccination with a second dose from kindergarten to college entry; (c) influenza virus vaccine given to high-risk groups and considered for all patients; and (d) varicella virus vaccine given to patients with no history of clinical varicella. Individuals in the 25to 64-year-old age-group should receive (a) tetanus and diphtheria toxoids (Td) booster every 10 years or a single booster at age 50; (b) MMR vaccine for persons born during or after 1957; (c) influenza virus vaccine for all persons in this age-group is strongly recommended, particularly health care providers and others at increased risk for exposure or transmission; (d) pneumococcal vaccine for persons with risk factors for acquiring invasive pneumococcal disease; and (e) varicella virus vaccine in patients with no clinical history of varicella. For those 65 years of age or older, it is recommended that
they receive influenza virus vaccine annually and pneumococcal vaccine every 6 years.
TABLE 25.5. OVERVIEW OF RECOMMENDED ROUTINE VACCINATIONS FOR ADULTS
Currently, five major types of vaccines are available: (a) live attenuated viruses or bacteria (e.g., cowpox, OPV, typhoid, cholera); (b) killed whole viruses (e.g., Salk poliovirus vaccine and influenza virus vaccine); (c) killed bacteria (e.g., pertussis, typhoid, and cholera); (d) purified component vaccines such as polysaccharide vaccines (e.g., pneumococcal and Hib vaccines) or toxins (e.g., tetanus toxoid and diphtheria toxoid); and (e) genetically engineered proteins produced by recombinant technology (e.g., hepatitis B vaccine). In the future, a sixth type, DNA vaccines, will be available that will provide direct inoculation with purified DNA. The advantages of DNA vaccines are that they stimulate both the hormonal and the cell-mediated arms of the immune system, they do not require new technology, they are more stable, and they are less expensive. This technology will allow for rapid creation of new vaccines. General Contraindications to Vaccines/Immune Therapy As noted by Orenstein et al. (9), the decision to use a vaccine must take into account the risks of disease, the benefits of vaccination, and the risks associated with vaccination. There are several contraindications to vaccination that can be generalized to all vaccines: (a) previous anaphylactic reaction to the same vaccine; (b) previous anaphylactic reaction to a vaccine constituent; and (c) presence of moderate or severe illness, with or without a fever. Although concern has been raised regarding the use of vaccines in pregnant and lactating women, the risks in pregnancy and lactation are primarily theoretical. In pregnancy, the benefit is greater than the risk when (a) the risk for disease exposure is high (e.g., travel to endemic areas); (b) infection would pose a special risk to pregnant women (e.g., influenza); and (c) vaccine is unlikely to cause harm. Breast-feeding is generally not a contraindication to vaccines or immune therapy. Killed and attenuated
vaccines do not replicate. Although live virus vaccines (e.g., rubella virus vaccine) may replicate and be excreted in breast milk, the neonate is usually not infected and any infection is well tolerated. From 1966 until recently, vaccination in pregnant women was discouraged and not recommended. The rationale for such an approach was not based on scientific data but occurred as a “knee-jerk” reaction to protect pregnant women and their fetuses. Vaccination in pregnant women to protect both mother and offspring was practiced extensively in the United States from 1957 to 1966 at a time when immunization with influenza virus and poliovirus vaccines was recommended during pregnancy. The safety of vaccination during pregnancy was demonstrated by the Collaborative Perinatal Project (31), which enrolled more than 50,000 pregnant women between 1959 and 1965 and evaluated more than 9,000 vaccine doses administered during the first 4 months of pregnancy. The vaccines most commonly used were inactivated poliovirus vaccines (18,342 women), live attenuated OPV vaccine (3,056 women), and influenza virus vaccine (2,291 women). These immunizing agents were not associated with the principal outcomes of the study. Based on extrapolation from the immunization rates in the project, an estimated 2 million doses of vaccine were given to pregnant women each year from 1959 to 1965. Moreover, the vaccines available currently are less “reactogenic,” more immunogenic, and better standardized. A contraindication means that a vaccine should not be administered (32,33). On the other hand, a precaution identifies a situation in which a vaccine may be appropriate if after careful assessment of the benefit of vaccination is judged to outweigh the risk (32,33). As reviewed by Watson and Peter (32), the contraindications and precautions can be generic and applicable to all vaccines or they can be specific to a particular vaccine. The specific contraindications and precautions as recommended by the ACIP and AAP are discussed in the sections for each vaccine. Immunosuppression, as a result of either underlying disease or therapy, is a contraindication to use of most live virus or bacterial vaccines (32). Measles vaccination of persons infected with the human immunodeficiency virus (HIV) (not severely immunocompromised) is an exception (34). In a similar vein, live virus vaccines should not be administered to individuals who have received high doses of systemic corticosteroids for 14 days or more until 1 month or more after steroid therapy was discontinued (35,36). Although most live virus vaccines, based on theoretical grounds, are contraindicated for pregnant women, the small theoretical risk associated with live virus vaccines in pregnant women may be far outweighed by the benefit of not contracting a disease with serious consequences for the mother and fetus (32).
Diphtheria And Tetanus Toxoid And Pertussis Background In the past, diphtheria was a major cause of morbidity and mortality, with death rates in the late 19th century averaging nearly 200 per 100,000 population annually and the proportion of total deaths attributable to diphtheria annually ranged from 3% to 10% (37). In that era, there were 50% more deaths from diphtheria than cancer. With the introduction of diptheria toxoid, the incidence of diphtheria fell dramatically, so from
1980 to 1995, only 41 cases of respiratory diphtheria were reported to the CDC (38). In 1998, only one case of diphtheria was reported in the United States (39). Tetanus is caused by the bacterium Clostridium tetani and is unique among the vaccine-preventable diseases because it is not a communicable disease (40). Tetanus, despite the presence of a highly effective toxoid, remains a major public health problem worldwide, with an estimated 1 million deaths per year due to neonatal tetanus and an estimated 310,000 to 700,000 nonneonatal cases, resulting in 122,000 to 300,000 deaths annually in developing countries (40). On the other hand, there are approximately 2,000 cases and 1,000 deaths annuawith improved hygiene and childbirth practices and improvement in wound care. Pertussis is caused by Bordetella pertussis and at one time was a major cause of morbidity and mortality in infants and children in the United States (39). From the time pertussis became a reportable disease (1920s) through the early 1940s, 115,000 to 270,000 cases of pertussis were reported each year in the United States, with 5,000 to 10,000 deaths annually (40). The availability and widespread use of pertussis vaccine in children led to a dramatic decrease in the incidence of pertussis, with 7,405 cases reported to the CDC in 1998 (39). Indications Diptheria and tetanus toxoids combined with acellular pertussis (DTaP) vaccine is the preferred vaccine and is recommended for all infants. The Td vaccine is recommended as a booster for all children aged 10 to 11 years with subsequent boosters recommended every 10 years (28). Use of whole-cell DTP remains an acceptable alternative. Specific Contraindications and Complications Severe reactions to the DT, tetanus and diptheria toxoids, or tetanus toxoid alone are unusual. As noted in Table 25.4, anaphylaxis has been established to result (rarely) from diphtheria and tetanus toxoids. The data favor a causative role of diphtheria and tetanus toxoids for brachial neuritis. Although initial data favored causation of Guillain-Barré syndrome (21), more recent assessments do not. The IOM found no evidence of causation for DT, tetanus and diphtheria, or tetanus toxoid alone for encephalopathy or infantile spasms. The pertussis vaccine given in combination with diphtheria and tetanus toxoids (i.e., the DTP vaccine) is felt to be causative for anaphylaxis (rare) and protracted inconsolable crying in infants (Table 25.4) (20,22). The IOM report favored a causative role of pertussis vaccine for acute encephalopathy and shock, as well as an unusual shocklike state characterized by hypotonia and unresponsiveness (20,22). No evidence implicates pertussis vaccine in infantile spasms, hypsarrhythmia, Reye syndrome, or sudden infant death syndrome (20,22). Whole-cell pertussis vaccines have been recognized for a long time as the vaccine associated with the greatest risk for adverse reactions—mostly minor but on occasion serious (41). Concern about the use of whole-cell pertussis vaccine led to the production of an effective, less reactogenic pertussis vaccine
culminating in the development of purified component (acellular) pertussis vaccines (41). Incorporation of DTaP has become the preferred vaccine for childhood vaccination (28). The only contraindication specific to the use of DTaP/DTP is encephalopathy within 7 days of administration of a previous dose. The contraindications general for all vaccines also apply (see previous discussion). Precautions for the use of DTaP/DTP include (a) fever of 40.5°C or higher within 48 hours after vaccination with prior dose and not attributable to another identifiable cause; (b) collapse or shocklike state within 48 hours of receiving prior dose; (c) convulsions within 3 days of receiving prior dose; (d) persistent inconsolable crying lasting 3 hours or more, within 48 hours of receiving prior dose; and (e) Guillain-Barré syndrome within 6 weeks of prior dose (32). Dosage DTaP and DTP are combinations used in infants and children younger than 7 years. Universal use of DTaP or DTP in infancy and childhood is recommended unless contraindications exist (28). Tetanus and diphtheria toxoid (i.e., Td vaccine) absorbed (for adult use) is used in persons 7 years of age or older. Single-antigen tetanus toxoids are available for use in persons 7 years of age or older. Td vaccine is the recommended formulation for routine booster inoculation in older children, adolescents, and adults. The usual dose for these vaccines is 0.5 mL intramuscular. The recommended schedule is to receive DPaT or DPT vaccine at 2 months, 4 months, 6 months, and 15 to 18 months of age. Td vaccine is recommended at age 10 to 11 years, followed by a booster every 10 years. Use During Pregnancy or Lactation No increased risk to mother or fetus from vaccination with Td or T vaccine during pregnancy, including in the first trimester, has been demonstrated (8,33,36). Widespread vaccination with tetanus toxoid of pregnant women in developing countries with high rates of neonatal tetanus has demonstrated the safety and the efficacy of this approach (32). In the United States, administration of Td vaccine is recommended for pregnant women who have not completed a primary series of vaccination or who are due for a booster (8,33).
Measles, Mumps, And Rubella Background Widespread vaccination of children against MMR has dramatically reduced the incidences of these infections and their associated complications worldwide (9,11,42,43 and 44). Measles is a ubiquitous, highly contagious infection that before vaccine availability, affected nearly 100% of the population by adolescence (42). In the United States during the pre–measles virus vaccine era, approximately 500,000 cases of measles were reported annually, with an estimated actual 4 million cases annually (45).
The morbidity and mortality associated with measles was also extensive, with an estimated annual occurrence of 150,000 cases of respiratory complications (pneumonia); 100,000 cases of otitis media; 48,000 hospitalizations; 7,000 seizure episodes; 4,000 cases of encephalitis, of which 25% had permanent neurologic sequelae or deafness; and 500 deaths (45). After measles virus vaccine licensure in the United States, the incidence of measles was dramatically reduced, from a reported 450,000 cases in 1965 to 100 cases in 1998 (less than 1 case per 1 million population), of which 71% were associated with international importation, suggesting measles is no longer an indigenous disease in the United States (39). On a global basis, dramatic control of measles has been accomplished in many areas (42). However, measles still remains the leading cause of vaccine-preventable deaths in children (42). In the Global Burden of Disease Study, measles ranked eighth overall as a cause of mortality, with 1 million deaths worldwide in 1996 (46). Before the licensing of live attenuated mumps virus vaccine in 1967, mumps was a common communicable disease of childhood. As noted by Plotkin and Wharton (43), the morbidity of mumps is best appreciated when the disease is perceived as a respiratory infection that is often accompanied by viremia resulting in multiorgan involvement, predominantly salivary glands. Complications of mumps are more common in adults and include orchitis in postpubertal men, pancreatitis, central nervous system involvement (meningoencephalitis), mastitis, nephritis, arthropathy, and myocarditis (rare) (43). In addition, mumps is a major cause of sensorineural deafness. Postlicensure of mumps virus vaccine, the reported cases of mumps in the United States declined from 152,209 in 1968 to 666 cases in 1998 (39,43). Similar dramatic decreases have accompanied the introduction of mumps virus vaccine globally. Rubella was also a common infection of children and young adults before vaccine availability. In the United States, rubella was both endemic and epidemic, with a cycle of major epidemics at 7-year intervals (44). Rubella was generally perceived to be a benign disease until Gregg (47), an Australian ophthalmologist, suggested that rubella infection of pregnant women was associated with congenital cataracts. Subsequently, his findings were expanded to include congenital heart disease and deafness as the classic triad of the congenital rubella syndrome (44). The last major epidemic of rubella in the United States occurred from 1964 to 1965 and clearly demonstrated the significant adverse effects of rubella in pregnancy and congenital rubella (48). During the epidemic, an estimated 12.5 million cases of rubella occurred, with approximately 2,000 encephalitis cases reported. In pregnant women, there were 5,000 elective abortions, 6,250 spontaneous abortions, and 2,100 neonatal deaths or infants who were stillborn. Congenital rubella syndrome was noted in 20,000 surviving infants, of whom 11,600 were deaf, 3,580 blind, and approximately 1,800 were mentally retarded. In addition, arthralgias and arthritis are common complications of rubella in adults; thrombocytopenia and encephalitis are less common (44). A syndrome of progressive rubella panencephalitis rarely occurs (49) and Guillain-Barré syndrome after rubella has been reported (50). Since licensing of rubella vaccine in 1969 in the United States, no major epidemics of rubella have occurred and the incidence of disease (rubella and congenital rubella syndrome) has progressively declined (39,51). In 1998, only 364 cases of rubella were reported to the CDC and 7 cases of congenital syphilis (39). Of these cases, 71% were
in persons 20 years of age or older, many of whom were immigrants who had not received vaccinations. Indication MMR vaccines are all live attenuated vaccines. These three vaccines are combined into a combination vaccine, which has been licensed in the United States since 1971. (42,43 and 44). In the United States, the ACIP has recommended two doses of MMR for all children and certain high-risk adolescents and adults (Fig. 25.1) (29). All children should receive the first dose of MMR vaccine at 12 to 15 months of age. The second dose should be administered at age 4 to 6 years; if the dose at age 4 to 6 years is missed, it is recommended that MMR vaccine be given at age 11 to 12 years. Although the second dose is primarily given to provide adequate protection against measles, it is recommended that both doses be given as MMR by both the Committees on Infectious Diseases of the AAP and the ACIP (42,43 and 44).
FIGURE 25.1. Recommended childhood immunization schedule—United States, January to December 2001. DTaP, diphtheria toxoid, tetanus toxoid, and acellular pertussis; DTP, diphtheria toxoid, tetanus toxoid, and pertussis; HepB, hepatitis B; Hib, Haemophilus influenzae type b; Td, tetanus and diphtheria toxoids; MMR, measles, mumps, and rubella; Var, varicella.
In addition to the recommended routine schedule of immunization with MMR vaccine, vaccination is recommended for all susceptible persons unless there is a contraindication. Persons born before 1957 are likely to have been naturally infected with MMR and thus can be considered immune (9). However, in the case of rubella, it is critical to ensure that women of childbearing age even those born before 1957, are immune to rubella. Thus, adult women before pregnancy and adult seronegative pregnant women in the postpartum period have been targeted for rubella vaccine. In the United States, persons are considered susceptible to MMR unless they (a) have documentation of adequate vaccination (two doses after age 12 months); (b) have laboratory proof of immunity; or (c) have evidence of physician-diagnosed measles or
rubella (9). Among adults, particular groups have been the focus for vaccination of susceptible individuals. These include (a) attendees of college or other higher educational institutions; (b) health care workers; (c) military recruits; and (d) international travelers (9). Susceptible adults should receive two doses of MMR separated by at least 28 days. In addition, anyone inoculated between 1963 and 1967 with killed vaccine (measles) or a vaccine of unknown type should be reinoculated (9). Specific Contraindications/Complications In general, MMR vaccine is contraindicated in (a) pregnant women (theoretical risk with live attenuated virus); (b) patients with moderate to severe illness or a high fever; (c) persons with a history of anaphylactic reaction to neomycin (measles virus vaccine), gelatin (mumps virus vaccine), or egg (mumps virus vaccine); (d) administration of Ig within 3 months; or (e) immunosuppression (42,43 and 44). History of allergy to eggs (without) anaphylaxis is generally not believed to be a contraindication to MMR vaccination (42,43 and 44). Side effects associated with measles virus vaccines are generally mild and limited to susceptible vaccinees (Table 25.4) (5,8,9,21,42,52). Fever 39.4°C (103°F) or higher occurs in approximately 5% to 15% of vaccine recipients; it is usually not bothersome but rarely may induce febrile seizures (42). Rash occurs in about 5% of recipients (42). These side effects are less frequent with the second dose of vaccine because most individuals are already immune. The IOM found evidence to establish a casual relationship between measles virus vaccine and anaphylaxis, thrombocytopenia, and death from vaccine strain viral infection in severely immunocompromised persons (21). A relationship between measles virus vaccine and central nervous system effects such as encephalitis or encephalopathy has been debated; the IOM report concluded that inadequate data were available to accept or reject such a casual relationship (21). The most common side effects associated with mumps virus vaccine are parotitis and low-grade fever (43). Rash, pruritus, and purpura, although reported, are uncommon, mild, and transient (43). The rate of aseptic meningitis after vaccination with the Jeryl Lynn strain used in the United States is very low at 1 of 800,000 (53). Moreover, sequelae to postvaccinal meningitis have been rare or absent (43). Adverse events after rubella vaccination have also been a topic of controversy (44). Low-grade fever and rash occur in 5% to 10% of rubella virus vaccine recipients (9). Approximately 25% of susceptible adult women have transient arthralgias after rubella vaccination, and acute arthritis occurs in approximately 10% of susceptible women (54,55). The IOM review concluded that the rubella virus vaccine is an established cause of acute arthritis (21). However, the finding that the data also favored a casual role for the rubella virus vaccine in chronic arthritis (21) has been refuted by more recent studies (25,26). Furthermore, the IOM found insufficient data demonstrating a causal relationship between rubella and radiculoneuritis, other neuropathies, and thrombocytopenia (21). In contradistinction, Plotkin (44) concluded that thrombocytopenia is causally associated with RA27/3 vaccination.
Dosage Use of the combined MMR vaccine is encouraged for most indications. However, monovalent forms of the MMR vaccine are also available for use in specific circumstances (e.g., rubella virus vaccine postpartum for susceptible pregnant women). In the United States, and increasingly so in other areas of the world, most childhood vaccination is accomplished with a triple vaccine containing the Moraten attenuated measles virus (1,000 TCID50), the Jeryl Lynn strain of mumps virus (5,000 TCID 50), and the RA27/3 rubella virus (1,000 TICD50). TCID50 is the median “tissue culture infective dose” of a virus. Measles virus vaccine is also combined only with the rubella vaccine as an alternative approach in adults. Either as the combined MMR or as monovalent vaccines, the dosage is 0.5 mL given subcutaneously. Two doses are recommended for routine childhood immunization, as discussed previously. For adults, two does are required for measles, but single doses are sufficient for rubella and mumps when used in the monovalent form. Either as combined MMR or as monovalent vaccines in the recommended schedule, MMR vaccines are very effective and probably provide lifelong immunity (42,43 and 44). Use During Pregnancy or Lactation On theoretical grounds, MMR vaccines (combined or monovalent) should not be administered to pregnant women (9,42,43 and 44). The rationale for such a recommendation is that these are live attenuated vaccines that may produce a viremia with the potential for transplacental passage to the fetal compartment. Thus, it is recommended that women of childbearing age should avoid pregnancy for 3 months after vaccination with these live attenuated vaccines. Measles virus vaccine has not been demonstrated to cross the placenta and infect the fetus (42). There is no evidence that measles or mumps virus vaccine can cause congenital malformations in humans (9,42,43). Although transplacental passage of virus has rarely been documented, evidence of fetal damage from the rubella virus vaccine does not exist (56,57,58,59 and 60). Specifically, there have been no documented cases of congenital rubella syndrome in the offspring of 226 susceptible women who received the RA27/3 rubella virus vaccine (strain currently used) within 3 months of conception and who carried their pregnancies to term (60). As a result, the ACIP recommends that rubella vaccination during pregnancy should not by itself be a reason to undertake an elective abortion (35). However, although no observable risk is associated with the rubella virus vaccine given during pregnancy, the rubella virus vaccine should not knowingly be given to a pregnant women (9). The MMR vaccine or monovalent forms of the MMW vaccine are safe for use in breast-feeding women (9). Although live viruses may replicate and be excreted in the breast milk, the neonate is usually not infected and any infection is well tolerated (32,36). The widespread use of the rubella virus vaccine in seronegative pregnant women clearly demonstrates the safety of the use of these vaccines in breast-feeding
women.
Hepatitis B Vaccine Background Hepatitis B virus (HBV) is transmitted by percutaneous or permucosal contact with HBV-containing body fluids, sexually by exposure to an infected partner, or perinatally from an infected mother to her infant (61). Based on seroprevalence data from the National Health and Nutrition Examination Survey, an estimated 5% of persons (12.5 million persons) living in the United States have been infected with HBV—approximately 300,000 persons each year during the 20 years preceding the survey (62). In the United States, it is estimated that there are more than 1 million chronic carriers of HBV (i.e., individuals with chronic HBV infection) and that currently post–hepatitis B vaccine availability, 100,000 to 150,000 people are still infected each year (61,62,63 and 64). Furthermore, it is estimated that 5,000 people die each year in the United States as the result of HBV-related liver disease. Groups at high risk for HBV infection in the United States include (a) injection drug abusers; (b) men having sex with men; (c) persons having heterosexual contact with multiple partners; (d) household contacts of person with chronic HBV infection; (e) hemophiliacs; (f) hemodialysis patients and staff; (g) inmates of long-term correctional facilities; (h) persons with occupational exposure to blood and infected body fluids; (i) institutionalized persons with developmental disabilities; and (j) immigrants from geographic areas where HBV infection is endemic (63,64). The public health impact of HBV infection is even more dramatic worldwide (61). An estimated 5% of the world's population (300 million people) have chronic HBV infection, which is a leading cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma (65). Beasley (66) demonstrated that persons with chronic HBV infection have more than a 100-fold increased risk to develop hepatocellular carcinoma. It is estimated that 500,000 to 1,000,000 people die each year worldwide as the result of HBV-related liver disease (61,65). Both in the United States and in other areas of the world, the availability of the hepatitis B vaccine has led to significant declines in the incidence of HBV disease (39,61,67,68). In Taiwan, the incidence of liver cancer in children was reduced after the introduction of newborn universal vaccination (68). Indications The ACIP has developed recommendations for the prevention of HBV transmission in the United States (64). These recommendations address both preexposure vaccination and postexposure vaccination. Preexposure vaccination to prevent HBV infection addresses three groups: (a) routine vaccination of infants; (b) catch-up vaccination of children and adolescents; and (c)
vaccination of adults in high-risk groups (61,64). The hepatitis B vaccine is recommended for all infants; infants born to mothers who are hepatitis B surface antigen (HBsAg) positive should receive the first dose of vaccine within 12 hours of birth concomitantly with HBIG, whereas those born to HBsAg-negative mothers should receive the first dose within 2 months of birth. All children and adolescents not previously inoculated with hepatitis B vaccine should be inoculated with the age-appropriate dose of vaccine; it is recommended that all 11 to 12 year olds be offered hepatitis B vaccine to ensure comprehensive coverage of all adolescents not inoculated at birth. Lastly, postexposure vaccination is recommended for adults in high-risk groups, as listed in Table 25.6, who have not previously been inoculated.
TABLE 25.6. GROUPS AT HIGH RISK FOR HEPATITIS B INFECTION
Postexposure prophylaxis with hepatitis B vaccine to prevent HBV infection is recommended for (a) the prevention of perinatal HBV infection; (b) persons with accidental percutaneous or permucosal exposure to blood; (c) sexual partners of people with acute HBV infection; and (d) household contacts of persons with acute HBV infection. All pregnant women should be screened for HBsAg early in prenatal care. HBsAg-negative women at high risk of HBV infection (Table 25.6) should be retested during the early third trimester. Infants born to HBsAg-positive women should receive the first vaccine dose plus HBIG within 12 hours of birth. Women admitted for delivery who were not screened prenatally should have blood drawn for HBsAg screening, but the infant should receive the hepatitis B vaccine within 12 hours of birth while the result is pending. Specific Contraindications/Complications The initial hepatitis B vaccines consisted of purified inactivated HBsAg particles obtained from the plasma of chronic HBV carriers. In the United States, the plasma-derived vaccines have been replaced by recombinant DNA vaccines derived from yeast.
Adverse effects after hepatitis B vaccine are uncommon and consist mainly of local reactions or low-grade fever. Anaphylaxis occurs rarely, but other serious adverse events have not been shown to be caused by the hepatitis B vaccine. The only specific contraindication to the hepatitis B vaccine is a history of allergy to yeast. Dosage The hepatitis B vaccine should be administered according to the recommended doses and schedule listed in Table 25.7. Administration of the vaccine should be in the deltoid muscle of children, adolescents, and adults or in the anterolateral thigh of neonates and infants (61). Comvax combines PedvaxHIB (H. influenzae type b conjugate) and Recombivax HB vaccines for administration at 2 months, 4 months, and 12 to 15 months of age in infants of HBsAg-negative mothers.
TABLE 25.7. RECOMMENDED DOSES AND SCHEDULES OF HEPATITIS B VACCINE
After three doses of hepatitis B vaccine, 95% to 99% of persons have protective antibody titers. There is a steady decline in protective antibody titers over time (61). However, studies have failed to identify acute cases of HBV among vaccine responders. Asymptomatic infections have been detected in 2.6% of persons followed for up to 11 years, but none had evidence of chronic HBV infection (61). These asymptomatic infections do not produce the sequelae associated with chronic HBV infection (61). To date, it has been demonstrated that the hepatitis B vaccine provides protection against serious HBV infection for at least 12 years (61). At this time, in the United States, routine booster doses of the hepatitis B vaccine are not recommended for persons who responded to vaccination (61). Use During Pregnancy or Lactation
The hepatitis B vaccine is safe to use in pregnant and breast-feeding mothers. Recombinant vaccines contain noninfectious DNA particles, so there is no risk to the fetus or breast-feeding infant. Thus, neither pregnancy nor lactation is a contraindication to the hepatitis B vaccine.
Hepatitis A Vaccine Background Hepatitis A virus (HAV) infection is usually a mild self-limited disease without any chronic sequelae (69). In 1998, there were 23,229 cases of HAV reported in the United States (39). The CDC estimates that there were 90,000 cases of symptomatic HAV infection and overall 180,000 persons had HAV infection (39). The mortality rate with HAV infection is low, with an estimated 100 persons dying each year in the United States as the result of acute liver failure due to fulminant HAV (69). No chronic carrier state exists for HAV infection. In developing countries of the world where HAV infection is endemic, HAV infection is nearly universal in childhood (69). However, HAV disease in childhood is usually subclinical (69). In the United States, risk factors associated with HAV infection have been identified (70). These include (a) personal contact with infected person; (b) staff and children in day care centers; (c) international travel to geographic areas in which hepatitis A is endemic; (d) homosexual men; (e) injection drug abusers; and (f) staff and patients of institutions for developmentally challenged persons. Approximately 45% of persons with sporadic community-acquired hepatitis do not have an identified risk factor (70). Indication The hepatitis A vaccine is recommended for persons 2 years of age or older who are at increased risk of HAV infection (9,69). Preexposure vaccination is indicated for (a) international travelers to countries where hepatitis A is endemic; (b) military personnel; (c) high-risk ethnic or geographic populations (e.g., Native Americans and Alaskan natives); (d) homosexual or bisexual men; (e) intravenous drug abusers; (f) regular recipients of blood or plasma-derived products (e.g., factor VIII); (g) persons engaged in high-risk employment (e.g., primate handlers, employees of institutions for developmentally challenged, and staff of day care centers); and (h) persons chronically infected with hepatitis C (40% chance of developing fulminant disease if infected with HAV) (71). Vaccination should also be considered for food handlers and children 2 years of age or older in group day care centers (69). The ACIP recommends that children living in states, counties, or communities where the reported annual rates of hepatitis A were more than or equal to 20 per 100,000 between 1987 and 1997 should be routinely inoculated with the hepatitis A vaccine at 2 years of age or older (72). In addition, the hepatitis A vaccine should be considered for all children living in states, counties, or
communities with reported annual rates of HAV between 10 to 20 per 100,000. Contraindications/Complications Two inactivated hepatitis A vaccines are available in the United States, Havrix (SmithKline Beecham Biologicals) and Vaqta (Merck and Company). No serious side effects have been associated with either of these vaccines (9). Local reactions at the injection site are common but mild (9). The only contraindication for hepatitis A vaccine is a history of allergy to any vaccine components (73). In children younger than 2 years, currently available hepatitis A vaccines are not approved for use. Dosage Havrix is available in two concentrations (69). The 720 enzyme-linked immunosorbent assay units (EU) per milliliter formulation is designed as a pediatric vaccine in persons aged 2 through 18 years and is administered in three 0.5 mL (360 EU) intramuscular injections at 0, 1, and 6 to 12 months. The formulation with 1,440 EU/mL can be used for either pediatrics or adults; the pediatric dose is 0.5 mL (720 EU) and the adult dose at 1 mL (1,440 EU) administered as an intramuscular injection, followed by a booster 6 to 12 months after the primary dose for individuals with repeated or long-term exposure. Vaqta is available as 50 U/mL. For children and adolescents (2 to 18 years of age), a single dose of 0.5 mL (25 units) is recommended. A single 1-mL dose (50 units) is recommended for adults. At 6 to 12 months, a booster dose is recommended to provide long-term protection (9,69). These inactivated hepatitis A vaccines are highly immunogenic and provide protection against infection for at least 10 years in persons receiving the primary vaccine and the booster (69). In Europe a third hepatitis A vaccine, Avaxim, is licensed for use in persons older than 15 years (69). It is administered intramuscularly as a 0.5-mL injection, followed by a booster dose 6 to 12 months later. Use During Pregnancy Or Lactation Hepatitis A vaccines are inactivated viruses. Thus, hepatitis A vaccine is not specifically contraindicated in pregnancy. However, it should be given only if clearly needed (e.g., with exposure to infected contact or travel to endemic area). Similarly, hepatitis A vaccine is safe in breast-feeding mothers.
Varicella Vaccine
Background Varicella-zoster (VZ) virus is a very contagious virus that is spread by respiratory droplets or airborne transmission (74,75). Acute varicella (chickenpox) most commonly occurs in childhood, but susceptible adults may also acquire VZ (74,75). Chickenpox is one of the most communicable human infections (75). As a result, more than 90% of persons in temperate climates have been infected by the age of 20 (74). As noted by Gershon (75), until vaccination against varicella becomes universally used, chickenpox will continue to be endemic in the United States. An estimated 4 million cases of varicella occur annually in the United States (74,75). In addition, it has been estimated that 5.2 million cases of herpes zoster (recurrent form of VZ virus infection) occur each year in the United States (76). Worldwide, an estimated 60 million cases of varicella occur annually (77). Balducci et al. (78) calculated the incidence of varicella in pregnancy to be 7 cases per 10,000 pregnancies. Thus, an estimated 2,800 pregnant women each year develop acute varicella. Healthy children produce few extracutaneous manifestations of varicella (74,75). These are uncommon and include pneumonitis, encephalitis, cerebellar ataxia, arthritis, hepatitis, glomerulonephritis, pericarditis, and orchitis (74,75). In contradistinction, adults with varicella have more morbidity associated with primary VZ virus infection (74,79,80). In adults, fever is higher and of longer duration and the severity of constitutional symptoms such as malaise, myalgias, and dehydration is greater (74). Complications of varicella such an encephalitis, pneumonia, and hepatitis are more common and more severe in adults. For instance, encephalitis occurs seven times more often in adults than in healthy children (74). Adults with varicella complications are hospitalized nine times more often than children and have a case fatality rate that is estimated to be 25 times more than that associated with children (81,82). The CDC estimates that 60 to 100 previously healthy persons die of varicella complications annually in the United States (83). The morbidity of acute varicella is increased in varicella-susceptible pregnant women and their fetuses (75). Gershon (75) reviewed reported cases of varicella in pregnancy and noted that among the 198 cases, 57 (29%) women developed pneumonia and the case fatality rate for pneumonia was 28% (16 of 57). As for the fetus, Patuszak et al. (84) in their metaanalysis estimated that the risk of congenital varicella syndrome is 2% if acute varicella occurred during the first 20 weeks of gestation (84). Indications The Oka varicella vaccine was licensed in Japan in the late 1980s and in the United States in 1995 (74). The ACIP recommended universal use of varicella vaccine for all healthy, varicella-susceptible children 12 months to 12 years of age (83). As the primary target, it is recommended that all children should be routinely inoculated at 12 to 18 months of age (83,85). To maximize vaccine compliance, the ACIP recommends review of vaccination status at 11 to 12 years of age and administration of varicella vaccine to
those not previously inoculated or infected with chickenpox (83). Persons 13 years of age or older should be assessed for varicella immune status, and those who are susceptible should be inoculated (83). Those with a reliable history of varicella are considered immune (83). In the absence of such a history, patients are considered susceptible and can be either tested to determine their immune status or inoculated without testing (83). The ACIP recommends that priority should be given to vaccination of susceptible adolescents and adults who are at high risk for exposure and for transmitting varicella (83). Vaccination is recommended for susceptible persons who have close contact with persons at high risk for serious complications (e.g., health care workers, family contacts of immunocompromised persons). Vaccination should be considered for susceptible persons in the following groups who are at high risk for exposure: (a) persons living or working in environments in which transmission of VZ virus is likely (e.g., teachers of young children, day care employees, and residents and staff in institutional settings); (b) persons living or working in environments in which varicella transmission can occur (e.g., college students, inmates and staff of correctional institutions, and miliary personnel); (c) nonpregnant women of childbearing age; and (d) international travelers. Women in the reproductive age-group should be asked if they are pregnant and advised to avoid pregnancy for 1 month after vaccination (83). Vaccination of other susceptible adolescents and adults is desirable and may be offered during routine health care visits (83). Specific Contraindications and Complications Varicella vaccine, which is a live attenuated vaccine, is contraindicated for persons who have a history of anaphylactic reaction to any component of the vaccine, including neomycin or gelatin (83). Varicella vaccine is not recommended for persons with active, untreated tuberculosis. It is contraindicated for use in persons with malignant conditions affecting the bone marrow or lymphatic system (83). However, varicella vaccine is available on a compassionate basis for use in patients with acute lymphoblastic leukemia who (a) have disease that has been in remission for 12 continuous months; (b) have a negative history of varicella; (c) have a peripheral lymphocyte count of more than 700 cells/mm2; and (d) have a platelet count of more than 100,000 cells/mm2 within 24 hours of vaccination (83). Varicella vaccine should not be administered to persons with primary or acquired immunodeficiency, including HIV infection or AIDS, cellular immunodeficiencies, hypogammaglobulinemia and dysgammaglobulinemia; varicella vaccine is also contraindicated in patients on immunosuppressive therapy (e.g., high-dose steroids and cancer chemotherapy) (83). Varicella vaccine in pregnancy is contraindicated (see later discussion) (83). The most common side effects seen with varicella vaccine are mild tenderness and redness at the infection site (15% to 20%), fever (10% to 15%), and mild rash (1% to 5%) (74,83). Transmission of vaccine virus to healthy susceptible persons has occurred rarely, but the disease is invariably mild or subclinical (74). This risk is higher if a rash develops and may be higher with immunocompromised vaccines (74,83).
Dosage Varicella vaccine is a live attenuated VZ virus known as the Oka strain (83). The vaccine contains more than 1,350 plaque-forming units in each 0.5-mL dose. Children 12 years of age or younger should be administered one 0.5-mL dose of vaccine subcutaneously (83). Preferably, varicella virus vaccine should be administered routinely to children at the same time as MMR vaccine (12 to 18 months of age) (83). Persons 13 years of age or older should receive two 0.5-mL doses of vaccine subcutaneously 4 to 8 weeks apart (83). The seroconversion rate after one dose of vaccine among susceptible children 12 months to 12 years of age is 97% (83). Among adolescents and adults 13 years of age or older, 78% of vaccines seroconverted after the first dose, and 99% seroconverted after a second dose (83). In clinical trials, varicella vaccine has proven to be effective for more than 10 years in preventing varicella (9,83). Most vaccinees who acquire varicella tend to have mild illness. Overall, varicella vaccine provides 70% to 90% protection against acute varicella infection and 95% protection against serious disease for more than 10 years (9,86). Use During Pregnancy and Lactation Varicella vaccine is a live attenuated virus and thus because of a theoretical risk of viremia and transplacental passage of vaccine virus to the fetus, pregnant women should not be inoculated with varicella vaccine (74,83). In addition, it is recommended that vaccine recipients should not become pregnant within 1 (ACIP) or 3 (Merck) months of vaccination (74,83). However, no infants with the congenital varicella syndrome secondary to vaccine-type virus have been reported (74,87). The manufacturer (Merck) in collaboration with the CDC has established the Varivax Pregnancy Registry to monitor maternal-fetal outcomes of pregnant women inadvertently administered varicella virus vaccine 3 months before or during pregnancy (telephone, 800-986-8999) (87). Initial results demonstrated no evidence of congenital varicella syndrome among 257 livebirths followed prospectively (95% confidence interval, 0–0.1%). Moreover, the risk of congenital varicella syndrome with wild virus is low (83), with an estimated 0.4% with maternal infection from conception to 12 weeks and 2% when infected between 13 to 20 weeks (78,84,88). The virulence of the attenuated vaccine virus is less than that of the wild type virus, so the risk to the fetus, if any, should be even lower (83). Thus, although a pregnant woman who inadvertently was inoculated within 1 month of conception or during pregnancy should be counseled about potential effects on the fetus, a decision to terminate a pregnancy should not be based on whether vaccine was administered just before or during pregnancy (83). According to the CDC, varicella virus vaccine may be administered to nursing mothers (83). It is not known whether attenuated vaccine virus is excreted in human milk and whether infants can be infected (83). However, most live vaccines (rubella is exception) have not been demonstrated to be secreted in breast milk (83).
Pneumococcal Vaccine Pneumococcal infections caused by Streptococcus pneumoniae remain an important cause of morbidity and mortality (89,90). Major clinical syndromes associated with S. pneumoniae include otitis media, sinusitis, bronchitis, and pneumonia, which occur secondary to direct spread of the organism from the nasopharynx (89,90). Hematogenous spread of S. pneumoniae can result in meningitis, endocarditis, arthritis, or peritonitis (89,90). Population-based studies have demonstrated that the overall rate of invasive pneumococcal disease (i.e., isolation of S. pneumoniae from a normally sterile site such as blood, pleural fluid or cerebrospinal fluid) is about 15 per 100,000 persons per year (89). Certain populations such as Native Americans have an incidence rate up to tenfold higher (90). Invasive pneumococcal infection is common in newborns and infants 2 years of age or younger and among adults 65 years of age or more (90). Among persons 65 years of age or older, the incidence of mortality was 42 to 57 cases per 100,000 population (89). Approximately 200,000 cases of invasive pneumococcal disease occur annually in the United States, with nearly 40,000 deaths per year (91). In the United States, pneumococcal infection is the leading cause of community-acquired pneumonia requiring hospitalization, with 30% to 50% of such cases caused by S. pneumoniae (89,90). The mortality rate for community-acquired pneumonia is 5% to 10% overall, increasing to 10% to 30% in people 65 years of age or older (91). Most fatal cases of pneumococcal pneumonia are associated with bacteremia (89). Conversely, 70% to 90% of pneumococcal bacteremia is caused by pneumococcal pneumonia (89). The mortality rate associated with pneumococcal bacteremia range from 16% to 36% among adults and 28% to 51% among persons 65 years of age or older (89,90). Risk factors that predispose to pneumococcal infection have been identified (90). These include (a) defective antibody formation (congenital or acquired); (b) patients with multiple myeloma, lymphoma, or chronic lymphocytic leukemia; (c) HIV infection; (d) neutropenia from any cause; (e) persons 65 years of age or older; (f) chronic diseases; and (g) prior respiratory viral infection, particularly caused by influenza virus (90). Over the past two decades, S. pneumoniae has increasingly become more resistant to penicillin and other antibiotics (92,93 and 94). This clearly has major implications for the treatment and complications of pneumococcal disease. Indications Recommendations for the use of pneumococcal vaccine have been promulgated by the ACIP, most recently in 1997 (95). The ACIP strongly recommends vaccination with pneumococcal vaccine for all persons 65 years of age or older and for those persons between 2 and 65 years of age who are at increased risk for serious pneumococcal infection. Those at increased risk include (a) persons with functional or anatomic asplenia (e.g., sickle cell disease, splenectomy); (b) persons with chronic cardiovascular or pulmonary disease; (c) those with diabetes mellitus; (d) alcoholics and those with chronic liver disease; (e) persons with cerebrospinal fluid leak; and (f) persons
inhabiting special environments (e.g., nursing homes, chronic care facilities, homeless shelters) or social settings (e.g., Native Alaskan, Native American) (95). Disappointingly, large national surveys have demonstrated that only 30% to 35.6% of persons 65 years of age or older in the United States have received pneumococcal vaccine (96,97). Although pneumococcal vaccine is less effective in immunocompromised persons, it is still recommended (95). The rationale is both the increased risk of pneumococcal disease in these groups and benefits, safety, and low cost of vaccination with pneumococcal vaccine. Thus, patients in the following groups should be inoculated with pneumococcal vaccine: (a) HIV infection; (b) congenital immunodeficiency; (c) leukemia, lymphoma, Hodgkin disease, and multiple myeloma; (d) generalized malignancy; (e) chronic renal failure; (f) nephrotic syndrome; and (g) organ or hematopoietic cell transplantation (95). Routine revaccination with pneumococcal vaccine is not recommended in the United States (95). Revaccination once is recommended for persons 65 years of age or older who received vaccine more than 5 years before the age of 65. Revaccination once after 5 years is also recommended for persons with asplenia or who are immunocompromised (95). Specific Contraindications and Complications Other than a severe reaction to a previous dose of vaccine, there are no contraindications to pneumococcal vaccine. Local side effects are common, with 30% to 50% of recipients of pneumococcal vaccine experiencing erythema, induration, and pain (90,95,98). These reactions are generally mild and of short duration (1 to 3 days). Severe systemic reactions are uncommon and severe febrile reactions (more than 103°F) are extremely rare (89). Dosage Pneumococcal vaccine is a polysaccharide vaccine that contains the capsular polysaccharides of 23 serotypes of S. pneumoniae. These serotypes are responsible for approximately 90% of invasive pneumococcal disease in developed countries (89). In the United States, two pneumococcal vaccines are marketed: Pneumovax 23 (Merck and Company) and Pnu-Imune 23 (Lederle Laboratories). Pneumococcal vaccine is administered as a single 0.5-mL dose intramuscularly or subcutaneously. The intramuscular route is preferred (95). The duration of immunity is unclear and revaccination should be considered after 5 years in patients at high risk of decline in antibody levels (e.g., those with chronic renal failure, nephrotic syndrome, or organ transplants) or of fatal infection (asplenia) (95). Although not as effective as most monovalent vaccines (90% to 95% or more), the efficacy of pneumococcal vaccine is still good (89). Overall effectiveness rate approaches 60%, and in patients 65 years of age or older, the effectiveness rate is 75% (89). To overcome the limitations of the 23-valent pneumococcal polysaccharide
vaccine, newer pneumococcal conjugate and protein vaccines have been developed (89). Initial results with pneumococcal conjugate vaccines have been very encouraging (99,100 and 101). However, conjugate vaccines are only effective against the serotypes included in the vaccines. Thus, research currently is attempting to identify other protective antigens—primarily proteins essential to pneumococcal virulence for candidate pneumococcal protein vaccines (89). Use During Pregnancy Or Lactation The safety of pneumococcal vaccine for use in pregnant women has not been determined (89). However, there is no reason to suspect that vaccination with pneumococcal vaccine, even in the first trimester, would have an adverse effect on the fetus or mother (89). Thus, pneumococcal vaccine use is not contraindicated in pregnancy. However, its use should be limited to those pregnant women at high risk (see “Indications”) not previously inoculated. Ideally, such high-risk women should be inoculated before rather than during pregnancy (89). Pneumococcal vaccine is safe to use in breast-feeding women. In fact, it has been suggested that maternal vaccination may stimulate protection in newborns via breast milk (89).
Influenza Vaccine Background Influenza is an acute febrile, prostrating infection of sudden onset that primarily involves the respiratory tract (e.g., nonproductive cough, nasal discharge, and substernal burning) but is associated with systemic symptoms of myalgia and headache that are out of proportion to the severity of respiratory symptoms (102). In adults, gastrointestinal symptoms are uncommon. Primary influenza virus pneumonia is rare except in persons with chronic cardiopulmonary disease; if it occurs, it is usually fatal (102). Secondary bacterial pneumonia is responsible for most fatal cases of influenza (102). Human influenza viruses are classified into three types: A, B, and C (102). Influenza A infection is the most frequent, is associated with the most morbidity and mortality, and is responsible for worldwide pandemics (104). Influenza B causes regional epidemics that are less severe than influenza A epidemics, but influenza B does not cause pandemics (104). Influenza C rarely causes epidemics (104). Because influenza A and B cause epidemic human disease, influenza vaccines contain two strains of A and one strain of B (102,103). Influenza A is divided into two subtypes based on surface antigens, hemagglutinin and neuraminidase (103). Influenza B is not categorized into subtypes. Both influenza A and B are further separated into groups based on antigenic characteristics (103). Frequent antigenic charges (antigenic drift) due to point mutations that occur during viral replication result in new influenza variants (103). This antigenic drift is the virologic basis for seasonal epidemics and why each year's influenza vaccine contains one or more new strains of influenza virus (103). Major antigenic differences (antigenic shift) occur at irregular intervals of 10 to 40 years (102). As a result of
antigenic shift, viruses with major antigenic differences from prevalent subtypes circulate in the community. Because these subtypes are antigenically unique, they are rapidly dispersed, resulting in widespread (pandemic) disease, affecting all age-groups (102). In the past century, such pandemics occurred in 1918, 1957, and 1968 (102). Influenza epidemics occur nearly every year during the winter months and the CDC estimates that influenza is responsible for approximately 20,000 deaths per year in the United States (103). Rates of infection are highest in children, but rates of serious illness and mortality are highest in person 65 years of age or older or in persons of any age with existing medical conditions associated with a high risk for complications (103). During the pandemic of 1957 to 1958 and the subsequent epidemic during 1960, pregnant women, particularly during the third trimester, were also noted to be at higher risk for influenza-associated disease (105,106 and 107). The devastating impact of influenza was best demonstrated during the pandemic of 1918 to 1919, when there were 20 million deaths worldwide and 500,000 deaths in the United States (102). The risks for complications, hospitalization, and deaths from influenza are highest in persons 65 years of age or older, very young children, and persons of any age with certain underlying medical conditions (103). During the past two decades, the estimated number of influenza-associated hospitalizations in the United States ranged from approximately 20,000 to more than 300,000 per epidemic, with an average of approximately 114,000 excess hospitalizations per year related to influenza (103). It has been estimated that the annual economic cost of influenza in the United States is $3 to $5 billion annually (108). According to the CDC, vaccination levels in persons 65 years of age or older increased from 33% in 1989 to 63% in 1997 (103). This rate exceeded the Healthy People 2000 goal of 60% (109). However, the vaccination for persons younger than 65 years with high-risk conditions was less than 30%, well short of the Healthy People 2000 goal of 60%. Moreover, only 34% of health care workers reported that they received influenza vaccine (110). Thus, there clearly exists the need to increase influenza vaccination utilization, which will in turn lead to a significant improvement in public health (102). As reviewed by the CDC, influenza vaccine is effective in preventing secondary complications and reducing the risk for influenza-related hospitalizations and deaths among persons 65 years of age or older and those with chronic diseases that place them at high risk for such complications (103,111). Among the older adult population not residing in nursing homes or chronic care facilities, influenza vaccine is 30% to 70% effective in preventing hospitalizations for influenza and pneumonia (103,112,113). Although vaccination prevents only 30% to 40% of influenza infection in elderly persons residing in nursing homes, influenza vaccine is effective in this group at preventing severe illness, secondary complications, and deaths; vaccine prevents 50% to 60% of hospitalizations or pneumonia and 80% of the mortality (103). Nichol et al. (114) demonstrated that influenza vaccination in healthy young adults significantly reduced the frequency of upper respiratory tract infection by 25%, work absenteeism by 36% to 43%, and physician office visits for upper respiratory tract infection by 44% (114).
Indications The ACIP strongly recommends that influenza vaccine be given to any persons 6 months of age or older who because of age or underlying medical condition are at increased risk for complications of influenza (103). In addition, it is recommended that health care workers and other individuals (including household members) in close contact with persons in high-risk groups be inoculated (103). Vaccination with influenza vaccine is recommended for the following groups of persons who are at increased risk for complications from influenza, who have a higher risk for complications from influenza, or who have a higher prevalence of chronic medical conditions that place them at risk for influenza-related complications: (a) persons aged 50 years or older; (b) residents of nursing homes and other chronic care facilities that house persons of any age who have chronic medical conditions; (c) adults and children who have chronic pulmonary or cardiovascular disease, including asthma; (d) adults and children with chronic metabolic diseases (e.g., diabetes mellitus), renal dysfunction, hemoglobinopathies, or immunosuppression; (e) children and adolescents (aged 6 months to 18 years) who are receiving long-term aspirin therapy and might be at risk for developing Reye syndrome after influenza infection; and (f) women who will be in the second or third trimester of pregnancy during the influenza season (103). In addition, the ACIP recommends that the following persons who can transmit influenza be inoculated: (a) physicians, nurses, and other personnel in both hospital and adjacent outpatient care settings; (b) employees of nursing homes and chronic care facilities who have contact with patients or residents; (c) persons who provide home care to persons in high-risk groups, and (d) household members (including children of persons in high-risk groups) (103). Additional groups in which influenza vaccine is recommended include (a) pregnant women who will be beyond the first trimester of pregnancy during the influenza season; (b) HIV-infected patients; and (c) any persons who wish to reduce the likelihood of acquiring influenza (103). Specific Contraindications and Complications Inactivated influenza vaccine should not be administered to persons with a history of anaphylactic hypersensitivity to eggs or other components of influenza vaccine (102,103). Some authorities believe that most persons with a history of egg allergy can be inoculated, although it should be approached with caution (102). The CDC suggests that persons who have developed hives, have had swelling of lips or tongue, or have experienced acute respiratory distress or collapse after eating eggs should consult a physician for appropriate evaluation to determine whether vaccine should be administered (103). Protocols have been developed for safely administering influenza vaccine to persons with egg allergies (115,116). Influenza vaccines are contraindicated in infants younger than 6 months and only split-virus vaccines should be used in children from 6 months to 12 years of age (102). In addition, adults with acute febrile illnesses should not be inoculated until their
symptoms have abated. However, minor illnesses with or without fever are not contraindications for use of influenza vaccine (103). The CDC suggests that when educating patients about potential side effects, it should be emphasized that (a) inactivated influenza vaccine contains noninfectious killed viruses and cannot cause influenza and (b) coincidental respiratory disease unrelated to influenza vaccination can occur postvaccination (103). Local reactions (e.g., erythema, pain, tenderness, and itching) are the most frequent side effects of influenza vaccination, effecting 10% to 64% of patients and lasting up to 2 days (102,103). These local reactions are generally mild and rarely interfere with daily activities (103). Systemic reactions include fever, malaise, myalgias, headache, and arthralgia. These are uncommon (e.g., 1% of adults develop moderate fever) and most often affect individuals who have had no exposure to the influenza virus antigens in the vaccine (e.g., young children) (103). These systemic reactions begin 6 to 12 hours postvaccination and may persist for 24 to 48 hours (103). The high incidence of febrile reactions to whole-virus vaccine in infants and children (8% to 50%) has led to the use of a two-dose immunization schedule and administration of split-virus vaccine to children younger than 12 years (102). Immediate allergic reactions including hives, angioedema, allergic asthma, and systemic anaphylaxis rarely occur after influenza vaccination (115). Most reactions are probably induced by residual egg protein in the vaccine (see previous discussion for details) (103). Several neurologic syndromes have been temporally associated with influenza vaccination (102). These include rare instances of optic neuritis, brachial neuritis, and cranial palsies; however, no statistically significant association with influenza vaccine has been established (102). However, a statistically significant association for Guillain-Barré syndrome with influenza vaccine was established only with the swine influenza vaccine of 1976 (102,103). The CDC emphasizes that no large increases in Guillain-Barré syndrome associated with influenza vaccine (other than swine influenza vaccine in 1976) have been noted, and that if influenza vaccine does pose a risk, it is probably quite small, slightly more than one case per million people inoculated (103). During epidemics from 1972 to 1973 through 1994 to 1995, the estimated rates of influenza-associated deaths ranged from approximately 300 to more than 1,500 per million persons no younger than 65 years of age (90% of influenza mortality occurs in this group) (103). Thus, the potential benefit of influenza vaccine in preventing serious illness, hospitalization, and death greatly outweighs the possible risk (if any) for developing influenza vaccine-associated Guillain-Barré syndrome. Dosage Influenza vaccine is an inactive virus vaccine composed of three influenza virus strains; usually two strains of influenza A and one strain of influenza B (102,103). The WHO makes recommendations as to the antigenic properties of influenza virus strains for use in influenza vaccines each year (102). It is recommended that influenza vaccine should be administered before influenza outbreaks occur, which is from December through March (103). The optimal time for vaccination against influenza is from the beginning of
October through mid-November (103). Dosage recommendations for influenza vaccine vary according to age-group (103). Among adults, a single 0.5-mL intramuscular injection is recommended (103). Adults should be inoculated in the deltoid muscle. Children younger than 9 years should receive two doses administered at least 1 month apart. Children younger than 12 years should be inoculated with split-virus vaccine, whereas adults generally receive whole-virus vaccine (103). In the United States, three influenza virus vaccines are licensed and available (102). These are Fluzone (Connaught), Fluvirin (Evans), and Flu-Shield (Wyeth-Ayerst). Use During Pregnancy and Lactation Currently available influenza vaccines are inactivated viral vaccines and thus are considered safe to use during any stage of pregnancy (103). Heinoven et al. (117) demonstrated no adverse fetal effects associated with influenza vaccine in a study of more than 2,000 pregnant women inoculated with influenza vaccine (117). Several other studies also evaluated antepartum influenza vaccination and documented no adverse effects on mother or infant (118,119 and 120). Some experts prefer to wait until after the first trimester of pregnancy is completed to administer influenza vaccine (as well as all vaccines) (103). Excess deaths due to influenza were documented among pregnant women during the pandemics of 1918 to 1919 and 1957 to 1958 (106,121,122 and 123). Additional studies have suggested that pregnancy can increase the risk for serious medical complications of influenza as a result of the many physiologic changes associated with pregnancy, including increases in heart rate, stroke volume, and oxygen consumption; decreases in lung capacity; and changes in immunologic function (e.g., decreased cell-mediated immunity) (107,124,125). euzil et al. (126) reviewed the impact of influenza during 17 interpandemic influenza seasons and demonstrated that the relative risk for hospitalization for selected cardiorespiratory conditions among pregnant women increased from 1.4 during weeks 14 to 20 of gestation to 4.7 during weeks 37 to 42, compared with women who were 1 to 6 months postpartum. In this study, women in the third trimester were hospitalized at a rate of 250 per 100,000 pregnant women, a rate comparable to that of nonpregnant women who had high-risk medical conditions for serious influenza infection (103). Based on the data in the study by Neuzil et al. (126), an average of one to two hospitalizations could be prevented for every 1,000 pregnant women inoculated (103,126). Thus, the benefit of influenza vaccination during pregnancy outweighs the potential risks. Women who will be beyond the first trimester of pregnancy (no less than 14 weeks of gestation) during the influenza season should be inoculated (103). Some experts suggest that all women at any stage of pregnancy should be inoculated against influenza before the influenza season. Pregnant women who have medical conditions that increase their risk for complications from influenza should be inoculated before the influenza season, regardless of the stage of pregnancy (103).
Influenza vaccine does not affect the safety of mothers who are breast-feeding or their infants (103). Furthermore breast-feeding does not adversely affect the mother's immune response (103). Thus, breast-feeding is not a contraindication for vaccination with influenza vaccine.
Lyme Disease Vaccine Lyme disease is a multisystem illness that primarily affects the skin, joints, heart, and nervous system (127,128). It is a tick-borne disease caused by infection with the spirochete Borrelia burgdorferi (128). Lyme disease is a multistage disease (127,128). Early localized disease manifests with skin lesions, erythema migrans, and within days or weeks, some infected patients undergo lymphatic or hematogenous dissemination of the organism to multiple organs, resulting in multiple skin lesions, hepatitis, cranial and peripheral neuropathies, meningitis, encephalitis, heart block and myocarditis, and arthritis (127). In the untreated or inadequately treated patient, B. burgdorferi infection can progress to late disseminated disease (127,128). The most common manifestation of late Lyme disease is intermittent swelling and pain of one or more joints, usually large weight-bearing joints such as the knee (128). Some cases manifest with chronic axonal polyneuropathy or encephalopathy, which presents with cognitive disorders, sleep disturbance, fatigue, and personality changes (128). Since the CDC initiated surveillance for Lyme disease in 1982, the number of annually reported cases in the United States has increased approximately 25-fold (128). In 1998, there were 16,801 cases of Lyme disease reported, the highest number ever reported (39). In the United States, the disease is primarily localized to states in the northeastern, mid-Atlantic, and upper north central regions and to several areas in northwestern California (129). The following nine states had incidence rates higher than the annual national average of 6.39 cases per 100,000 population and accounted for 93% of reported cases: Connecticut (105 of 100,000), Rhode Island (79.6), New York (25.5), New Jersey (24.0), Pennsylvania (22.9), Maryland (13.1), Massachusetts (11.5), Wisconsin (12.8), and Delaware (10.7) (39). Two Lyme disease vaccines, which use recombinant B. burgdorferi lipidated outer surface protein A (OspA) as immunogen, have been developed (128): LYMErix (SmithKline Beecham Biologicals) and ImuLyme (Pasteur Merieux Connaught). Only LYMErix is licensed by the Food and Drug Administration and available for use in the United States. Indications Recently the CDC issued recommendations for use of Lyme disease vaccine (128). Lyme disease vaccine is recommended for persons who reside, work, or recreate in areas of high or moderate risk for acquiring Lyme disease; Lyme disease vaccination should be considered for individuals aged 15 to 70 years who engage in activities that result in frequent or prolonged exposure to tick-infected habitat. Lyme disease vaccination may be considered for persons age 15 to 70 years with exposure to tick-infected habitats that is neither frequent nor prolonged. Vaccination should be considered for travelers to areas of high incidence if frequent or prolonged exposure to
tick-infected habitats is anticipated. Vaccination should also be considered for persons with a history of previous uncomplicated Lyme disease who are at continued high risk. Specific Contraindications and Complications Until the safety and immunogenicity of recombinant OspA vaccines in children have been established, Lyme disease vaccine is not recommended for children younger than 15 years (128). Similarly, the safety and efficacy of Lyme disease vaccine have not been established for persons older than 70 years (128). Because the safety of recombinant OspA vaccines administered during pregnancy has not been established, vaccination of pregnant women is not recommended (128). Because persons with HIV, joint swelling, or diffuse musculoskeletal pain were excluded from the phase III Safety and Efficacy Trial of Lyme Disease Vaccine, no data exist regarding vaccine use in these groups (128). In the phase III clinical trails of LYMErix, nearly 5,000 vaccine and 5,000 placebo recipients were compared (130). Soreness at the injection site was the most frequent side effect, occurring in 24.1% of vaccine recipients, versus 7.6% of placebo recipients. Redness and swelling at the injection site occurred in less than 2%; myalgia, influenza-like illness, fever, and chills were noted in up to 3.2%. Arthralgias were more common in vaccine recipients. No episodes of immediate hypersensitivity reactions among vaccine recipients were noted. Dosage LYMErix is administered by intramuscular injection, 0.5 mL (30 µg), into the deltoid muscle (128). Three doses are required for optimal protection (128,130). The first dose is followed by a second dose 1 month later and a third dose administered 12 months after the first dose. Vaccine administration should be timed so that the second dose (year 1) and the third dose (year 2) are given several weeks before the beginning of the B. burgdorferi transmission season, which usually begins in April (128). Use During Pregnancy and Lactation The safety of recombinant OspA vaccines administered during pregnancy has not been studied. Thus, vaccination of women known to be pregnant is not recommended (128). No evidence has demonstrated that pregnancy increases the risk to acquire Lyme disease or its severity (128). Moreover, acute Lyme disease during pregnancy responds well to antibiotic therapy and adverse fetal or maternal outcomes have not been reported in pregnant women receiving standard courses of treatment (128). SmithKline Beecham Biologicals (manufacturer of LYMErix) has established a Lyme disease vaccine pregnancy registry for pregnant women inadvertently inoculated. Health care providers are encouraged to register such vaccination by calling the company (800-366-8900, ext. 5231).
Immune Therapy Passive immunization to provide temporary protection against specific infections is accomplished by injection of Ig containing exogenously produced antibody (9). This can be pooled human Ig or hyperimmunoglobulin obtained from persons with known high titers of antibody to a specific infectious agent. Immune Globulin Ig is a preparation made up of pooled human Igs containing antibodies against infectious agents, particularly hepatitis A and measles (9). Ig is indicated for postexposure or preexposure prophylaxis depending on the infectious agent of concern. In the case of exposure to hepatitis A, Ig is effective in preventing disease when administered within 14 days of exposure. The postexposure dose of immunoglobulin G (IgG) to prevent HAV is 0.02 mL/kg intramuscularly (9,131). Ig is an alternative to hepatitis A vaccine as prophylaxis for short-term exposure (e.g., single international trip of less than 2 to 3 months' duration). For preexposure prophylaxis (i.e., international travel), the dose is 0.06 mL/kg intramuscularly (9,131). The efficacy of Ig for the prevention of HAV is 80% to 90% if given early in the incubation period. The other major indication for Ig is postexposure prophylaxis for measles (9,131). Ig may prevent or modify measles if administered within 6 days of exposure (9). The dose of Ig to prevent measles is 0.25 mL/kg intramuscularly for healthy persons; maximum total dose is 15 mL (9,131). Other than a history of anaphylactic reaction to Ig, there is no contraindication to the use of Ig. The most common side effect is local tenderness; rarely anaphylaxis has occurred (9). Ig inhibits the immune response to live virus vaccines (e.g., measles and rubella), for 3 months with hepatitis prevention dose and 5 months with measles prevention dose (9,132). Ig is safe for use in pregnant and lactating women. Hepatitis B Immune Globulin HBIG is hyperimmunoglobulin prepared from plasma that has been preselected to contain high titers of antibody against HBsAg (9). HBIG is recommended for postexposure prophylaxis in susceptible persons who have been exposed to blood or body fluids containing HBsAg by percutaneous or mucous membrane routes or by HBV-infected sexual partners (9,64,131). HBIG is also recommended for infants born to HBsAg-positive women (64). In adults, for both sexual contacts and percutaneous exposure, the dose is 0.6 mL/kg intramuscularly given immediately. For newborn passive immunization, a dose of 0.5 mL should be given within 12 hours of delivery (64). The hepatitis B vaccine series should be started in conjunction with HBIG in adults and newborns (64). For percutaneous and sexual exposures, postexposure prophylaxis with HBIG is 75% effective. Used in
combination with hepatitis B vaccine, HBIG is more than 90% effective in neonates (61,64,68). There are no known contraindications or precautions with HBIG (9). The use of HBIG in pregnant and breast-feeding women appears to be safe. Varicella-Zoster Immune Globulin VZIG is a hyperimmunoglobulin prepared by obtaining serum from individuals with high titers of VZ antibodies. VZIG is recommended for postexposure prophylaxis against varicella in susceptible immunocompromised persons with significant exposure to varicella such as household contact, close contact indoors of more than 1 hour, sharing same hospital room, or prolonged direct face-to-face contact (83). In addition, VZIG is recommended for susceptible pregnant women exposed to varicella and newborns delivered to mothers who developed clinical varicella within 5 days before to 48 hours after delivery (83). In adults, the dose of VZIG is 125 units per 10 kg of body weight (maximum of 625 units) intramuscularly. VZIG should be administered within 96 hours of exposure (83). It has been suggested that VZIG may also be useful in ameliorating the clinical presentation of varicella in susceptible adults, particularly pregnant women who may be at increased risk for complications of varicella (9,83). Local reactions (e.g., pain, redness, and swelling) are uncommon (less than 10%) and severe side effects are rare (83). There are no known contraindications to VZIG (9,83). VZIG is safe to use in pregnant and breast-feeding women (83). Its primary benefit is to decrease the incidence and severity of maternal varicella (83). However, its effect on vertical transmission of varicella is unknown.
Use Of Vaccines: Special Circumstances Table 25.2 and Table 25.5 summarize those immunizations that are routinely recommended in adults. However, there are special circumstances in which various additional vaccines are recommended (Table 25.8).
TABLE 25.8. SPECIAL CIRCUMSTANCES FOR IMMUNIZATIONS IN ADULTS
Pregnancy Immunization of pregnant women is generally not recommended because of theoretical risks to the fetus (3,9,27,32). However, in certain circumstances, selected vaccines may be indicated when the benefits of vaccinating a pregnant woman outweigh any potential risks (3,9,27,32,33,133). Such instances include (a) if the risk of infection is high; (b) if the infection can have serious consequences for the pregnant woman or her fetus; and (c) if the vaccine is unlikely to be associated with an increased incidence of adverse effects (33,133). A summary of vaccines that are recommended for use in pregnant women either on a routine basis or in selected high-risk situations is provided in Table 25.9.
TABLE 25.9. SUMMARY OF VACCINES FOR PREGNANT WOMEN
In general, live organism vaccines are contraindicated in pregnancy because they
contain attenuated viruses or bacteria that multiply in the vaccine recipient (3,9,32). In particular, the live virus vaccines, some of which prevent infections such as rubella and varicella that are known to be teratogenic, are usually contraindicated in pregnancy (32,33,36,133). However, it is recommended that both OPV vaccine and yellow fever vaccine can be administered to nonimmune pregnant women who are at substantial risk of imminent exposure to infection (e.g., international travelers to endemic areas) (23,33,134). Unfortunately, despite warnings, some pregnant women are inadvertently inoculated with live virus vaccines. Studies, to date, have demonstrated the safety of such inadvertent vaccination in pregnancy (35,83). Moreover, the risk to the fetus is largely theoretical (9,32). Thus, administration of a live virus vaccine during pregnancy, including rubella or varicella, is not an indication for performing an abortion to terminate the pregnancy (35,83). As discussed in earlier sections, the Rubella Vaccination in Pregnancy Registry documented the apparent safety of inadvertent rubella vaccination in pregnancy (35) and currently there exists a Varicella Vaccination in Pregnancy Registry to monitor prospectively maternal and fetal outcomes in pregnant women inadvertently injected with varicella vaccine (telephone, 800-986-8999) (36,83,87). Killed and inactivated vaccines do not contain viruses or bacteria that can multiply within the body and thus do not pose a risk during pregnancy. In the past, concern arose over the use of inactivated poliovirus (IPV) vaccine and suspected association of IVP vaccine use in pregnancy with neural malignant neoplasms in offspring (117). However, other studies have not confirmed this association and IPV vaccine can be provided to pregnant women requiring immediate protection against poliomyelitis (23,33,36). No adverse effects on pregnant women or their fetuses have been demonstrated with the use of other killed inactivated vaccines and toxoids (9,32,33,36). In fact, these vaccines and toxoids are indicated in pregnancy to prevent infections associated with serious adverse events in the mother or fetus. Thus, influenza vaccine is recommended on a routine basis for pregnant women who will be in the second or third trimester during the influenza season because influenza has been associated with increased morbidity in pregnant women (103). An example of protection for the fetus or neonate is vaccination of pregnant women with tetanus toxoid in developing areas of the world where neonatal tetanus is a common and devastating disease (32). Maternal immunization provides protection to the newborn as the result of transplacental passage of maternal IgG antibodies to the fetus. In the United States, administration of combined tetanus-diphtheria toxoid is recommended for any pregnant women who have not completed their primary vaccination series or who need a booster dose (33,133). Hepatitis B vaccine is produced with recombinant DNA technology and contains no infectious agent. Thus, it also is safe to use when indicated in pregnant women at high risk for acquiring HBV (64). Because of theoretical concerns, some physicians prefer to avoid vaccination during the first trimester of pregnancy and wait until after 14 weeks of gestation to administer vaccines or toxoids to pregnant women (133). In reality, there is no evidence demonstrating that there is an increased risk to the mother or fetus with vaccination in the first trimester (32). Furthermore, in some cases, vaccination will be required before 14 weeks of gestation (32). Thus, vaccines such as influenza vaccine, hepatitis B vaccine, and tetanus-diphtheria toxoid may be given in the first trimester
(33,64,103,133). Neither live nor killed vaccines are contraindicated in breast-feeding mothers (30,33). No adverse events in mother or infant have been associated with vaccination of breast-feeding mothers (30,33). The killed or inactivated virus vaccines do not multiply within the body and thus carry no risk for lactating mothers or their infants (30). Although live virus vaccines contain attenuated live viruses or bacteria that replicate in vaccine recipients, most live viral or bacterial vaccines have not been demonstrated to be secreted in breast milk and thus are not a risk to lactating women or their infants (32). Thus, it is recommended that lactating mothers may also receive live virus vaccines such as MMR, varicella, rubella, OPV, and yellow fever safely (33,35,83,133). Attenuated rubella virus vaccine is an exception and has been detected in breast milk and from the nasopharynx and throat of some breast-fed infants (135). However, no adverse effects have been noted and rubella vaccination postpartum is commonly performed. International Travel Various vaccines are required for admission into some countries or are recommended for travelers to countries where vaccine-preventable diseases are endemic (136). Information can be obtained by calling the CDC Travel Information System at 404-332-4559. Examples of vaccines commonly considered for travelers include measles, polio, and tetanus-diphtheria boosters. Travelers to endemic areas should consider hepatitis B, hepatitis A, typhoid, yellow fever, rabies, plague, and Japanese encephalitis prophylaxis (32). Occupational Exposure Vaccine recommendations have not been promulgated for most occupational categories (32). However, specific recommendations have been established for health care workers (137). All health care workers and public safety workers who may be exposed to blood or blood-derived body fluids should receive hepatitis B vaccine. Any health care worker who may transmit rubella to pregnant patients should be immune to rubella vaccine. In addition, health care workers are at greater risk of measles than the general population and should receive measles virus vaccine (two doses) if they are susceptible. To prevent transmission of influenza to patients at high risk for complications of influenza, all health care workers should receive influenza vaccine before the influenza season annually (103). Varicella vaccination is also recommended for health care workers not immune to varicella (83). There are other occupations at need for selected vaccination. These include plague vaccine for laboratory workers with potential exposure to plague and anthrax vaccine for animal workers. CHAPTER REFERENCES 1. Centers for Disease Control and Prevention. Vaccine-preventable diseases: improving vaccination coverage in children, adolescents, and adults. MMWR Morb Mortal Wkly Rep 1999;48[RR-8]:1–15.
2. Gardner P, Schaffner W. Immunization of adults. N Engl J Med 1993;328:1252–1258. 3. Glezen WP, Alpers M. Maternal immunization. Clin Infect Dis 1999;28:219–224. 4. Reilly JG, Haque R, Pavlides A, et al. Adult vaccinations: an underused method of disease prevention. Primary Care Update Obstet Gynecol 1996;3:220–223. 5. National Coalition for Adult Immunization. Resource guide for adult and adolescent immunization, 3rd ed. Bethesda, MD: National Coalition for Adult Immunization, 1998. 6. Bartman BA, Weiss KB. Women's primary care in the United States: a study of practice variation among physician specialties. J Womens Health 1993;2:261–266. 7. Gonik B, Jones T, Contreras D, et al. The obstetrician-gynecologist's role in vaccine-preventable diseases and immunizations. Obstet Gynecol 2000;96:81–84. 8. American College of Physicians Task Force on Adult Immunization and Infectious Diseases Society of America. Guide for adult immunization, 3rd ed. Philadelphia: American College of Physicians, 1994. 9. Orenstein WA, Whartom M, Bart KJ, et al. Immunization. In: Mandel GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. Philadelphia: Churchill Livingstone, 2000:3207–3234. 10. Jenner E. An inquiry into the causes and effects of the variola vaccine. London: Low, 1798. 11. Plotkin SL, Plotkin SA. A short history of vaccination. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:1–12. 12. Global Commission for the Certification of Smallpox Eradication. The achievement of the global eradication of small pox. Geneva, Switzerland: World Health Organization, 1979. 13. Centers for Disease Control and Prevention. Ten great public health achievements—United States, 1900–1999. MMWR Morb Mortal Wkly Rep 1999;48(12):245. 14. Centers for Disease Control and Prevention. Certification of poliomyelitis eradication—the Americas, 1994. MMWR Morb Mortal Wkly Rep 1994;43:720–722. 15. Centers for Disease Control and Prevention. Progress toward global poliomyelitis eradication 1997–1998. MMWR Morb Mortal Wkly Rep 1999;48:416–421. 16. Centers for Disease Control and Prevention. Certification of poliomyelitis eradication—Western Pacific Region, October 2000. MMWR Morb Mortal Wkly Rep 2001;50:1–3. 17. World Health Organization. Global polio eradication initiative. Strategic plan 2001–2005 (WHO/Polio/00.05). Geneva, Switzerland: World Health Organization, 2000. 18. Centers for Disease Control and Prevention. Recommendations of the International Task Force for Disease Eradication. MMWR Morb Mortal Wkly Rep 1993;42:1–38. 19. Centers for Disease Control and Prevention. Progress towards eliminating Haemophilus influenzae type B disease among infants and children—United States, 1987–1997. MMWR Morb Mortal Wkly Rep 1998;47:993–998. 20. Howson CP, Howe CJ, Fineberg HV, eds. Institute of Medicine: adverse effects of pertussis and rubella vaccines. Washington: National Academy Press, 1991. 21. Stratton KR, Howe CJ, Johnston RB Jr, eds. Institute of Medicine: adverse events associated with childhood vaccines. Evidence bearing on causation. Washington: National Academy Press, 1994. 22. Howson CP, Fineberg HV. Adverse effects following pertussis and rubella vaccines: summary of a report by the Institute of Medicine. JAMA 1992;267:392–396. 23. Centers for Disease Control and Prevention. Poliomyelitis prevention in the United States: introduction of a sequential vaccination schedule of inactivated polio virus vaccine followed by oral polio virus vaccine. MMWR Morb Mortal Wkly Rep 1997;46:1–25. 24. Rantala H, Cherry JD, Shields WD, et al. Epidemiology of Guillain-Barré syndrome in children: relationship of oral polio vaccine administration to occurrence. J Pediatr 1994;124:220–223. 25. Slater PE, Ben-Zvi T, Fogel A, et al. Absence of an association between rubella vaccination and arthritis in underimmune postpartum women. Vaccine 1995;13:1529–1532. 26. Ray R, Black S, Shinefield H, et al. Risk of chronic arthropathy among women after rubella vaccination. JAMA 1991;278:551–556. 27. Amstey MS. Immunization in pregnancy. Contemp Obstet Gynecol 2000;October:65–72. 28. Centers for Disease Control and Prevention. Recommended childhood immunization schedule—United States, 2000. MMWR Morb Mortal Wkly Rep 2001;50:7–10. 29. Centers for Disease Control and Prevention. Update on adult immunization. Recommendation of the Immunization Advisory Committee (ACIP). MMWR Morb Mortal Wkly Rep 1991;40[RR-12]:56. 30. Committee on Infectious Diseases, American Academy of Pediatrics. Report of the Committee on Infectious Diseases, 24th ed. Elk Grove Village, IL: American Academy of Pediatrics, 1997.
31. The Collaborative Perinatal Study of the National Institute of Neurological Diseases and Stroke. The women and their pregnancies. Philadelphia: WB Saunders. 32. Watson JC, Peter G. General immunization practices. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:47–73. 33. Centers for Disease Control and Prevention. General recommendations on immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1994;43[RR-1]:1–38. 34. Centers for Disease Control and Prevention. Measles pneumonitis following measles-mumps-rubella vaccination of a patient with HIV infection, 1993. MMWR Morb Mortal Wkly Rep 1996;45:603–606. 35. Centers for Disease Control and Prevention. Measles, mumps, and rubella-vaccine use and strategies for measles, rubella, congenital rubella syndrome elimination and mumps control. Recommendations of the Advisory Committee on Immunizations. MMWR Morb Mortal Wkly Rep 1998;47[RR-8]:1–57. 36. American Academy of Pediatrics. Immunization in special clinical circumstances. In: Peter G, ed. 1997 Red book: report of the Committee on Infectious Diseases, 24th ed. Elk Grove Village, IL: American Academy of Pediatrics, 1997:48–71. 37. Mortimer EA Jr, Wharton M. Diphtheria toxoid. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:140–157. 38. Bisgard KM, Hardy IRB, Popovic T, et al. Respiratory diphtheria in the United States, 1980–1995. Am J Public Health 1998;88:787–791. 39. Centers for Disease Control and Prevention. Summary of notifiable diseases, United States 1998–1999. MMWR Morb Mortal Wkly Rep 2000;47(53):1–92. 40. Wassilak SG, Orenstein WA, Sutler RW. Tetanus toxoid. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:441–474. 41. Edwards KM, Decker MD, Mortimer EA Jr. Pertussis vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:293–344. 42. Redd SC, Markowitz LE, Katz S. Measles vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:222–266. 43. Plotkin SA, Wharton M. Mumps vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:267–285. 44. Plotkin SA. Rubella vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:409–439. 45. Bloch AB, Orenstein WA, Stetler HC, et al. Health impact of measles vaccination in the United States. Pediatrics 1985;76:524–532. 46. Murray CJL, Lopez AD. Mortality by cause for eight regions of the world: global burden of disease study. Lancet 1997;349:1269–1276. 47. Gregg NM. Congenital cataract following German measles in the mother. Trans Opthalmol Soc Aust 1941;3:35–46. 48. Rubella Surveillance. National Communicable Disease Center. US Dept of Health, Education and Welfare; June 1969; No 1. 49. Townsend JJ, Baringer JR, Wolinsky JS, et al. Progressive rubella panencephalitis: late onset after congenital rubella. N Engl J Med 1975;292:990–993. 50. Yaginuma Y, Kawamura M, Ishikawa M. Landry-Guillain-Barré-Strohl syndrome in pregnancy. J Obstet Gynecol 1996;22:47–49. 51. Centers for Disease Control and Prevention. Rubella and congenital rubella syndrome—United States, 1994–1997. MMWR Morb Mortal Wkly Rep 1997;46:350–354. 52. Peltola H, Heinonen OP. Frequency of true adverse reactions to measles-mumps-rubella vaccine. A double-blind placebo controlled trial in twins. Lacent 1986;1;939–942. 53. Nalin D. Mumps vaccine complications: which strain [Letter]? Lancet 1989;2:1396. 54. Freestone DS, Prydie J, Smith SG, et al. Vaccination of adults with Wistar RA27/3 rubella vaccine. J Hygiene 1971;69:471–477. 55. Polk BF, Modlin JF, White JA, et al. A controlled comparison of joint reactions among women receiving one of two rubella vaccines. Am J Epidemiol 1982;115:19–25. 56. Fleet WF, Benz EW, Karzon DT, et al. Fetal consequences of maternal rubella vaccination. JAMA 1974;227:621–626. 57. Centers for Disease Control and Prevention. Rubella vaccination during pregnancy—United States, 1973–1983. MMWR Morb Mortal Wkly Rep 1984;33:365–373.
58. Bart SW, Stetler HC, Preblud SR, et al. Fetal risk associated with rubella vaccine: an update. Rev Infect Dis 1985;7[Suppl]:S95–S102. 59. Enders G. Rubella antibody titers in vaccinated and nonvaccinated women and results of vaccination during pregnancy. Revs Infect Dis 1985;7[Suppl]:S103–S107. 60. Centers for Disease Control and Prevention. Rubella vaccination during pregnancy—United States, 1971–1988. MMWR Morb Mortal Wkly Rep 1989;38:289–293. 61. Mahoney FJ, Kane M. Hepatitis B vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:158–182. 62. Coleman PJ, McQuillan GM, Moyer LA, et al. Incidence of hepatitis B virus infection in the United States, 1976–1994: estimates from the National Health Nutrition Examination Surveys. J Infect Dis 1998;178:954–959. 63. Margolis HS, Alter MJ, Hadler SC. Hepatitis B: evolving epidemiology and implications for control. Semin Liver Dis 1991;11:84–92. 64. Centers for Disease Control and Prevention. Hepatitis B virus: a comprehensive strategy for eliminating transmission in the United States through universal childhood vaccination: recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR Morb Mortal Wkly Rep 1991;40[RR-13]. 65. Maynard JE, Kane MA, Alter MJ, et al. Control of hepatitis B by immunization: global perspectives. In: Vyas GN, Dienstag JL, Hoofnagle JH, eds. Viral hepatitis and liver disease. New York: Grune & Stratton, 1988:967–969. 66. Beasley RP. Hepatitis B virus: the major etiology of hepatocellular carcinoma. Cancer 1988;61:1942–1956. 67. Mahoney F, Smith N, Alter MJ, et al. Progress towards the elimination of hepatitis B virus transmission in the United States. Viral Hepatitis Rev 1997;3:105–119. 68. Chang MH, Chen CJ, Lai MS, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. N Engl J Med 1997;336:1855–1859. 69. Feinstone SM, Gust ID. Hepatitis A virus. In: Mandell Gl, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. New York: Churchill Livingstone, 2000:1920–1940. 70. Shapiro CN, Coleman PJ, McQuillan GM, et al. Epidemiology of hepatitis A: seroepidemiology and risk groups in the United States. Vaccine 1992;10[Suppl 1]:S59–S62. 71. Vento S, Garofano T, Renzini C, et al. Fulminant hepatitis associated with hepatitis A virus superinfection in patients with chronic hepatitis C. N Engl J Med 1998;338:286–290. 72. Centers for Disease Control and Prevention. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 1999;48[RR-12]. 73. Centers for Disease Control and Prevention. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 1996;45:1–30. 74. Gershon AA, Takahashi M, White CJ. Varicella vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines Philadelphia: WB Saunders, 1999:475–507. 75. Gershon AA. Chickenpox, measles, and mumps. In: Remington JS, Kelin JO, eds. Infectious diseases of the fetus and newborn infant. Philadelphia: WB Saunders, 2001:683–732. 76. Ragozzino ME, Melton LF, Kurkland, et al. Population-based study of herpes zoster and its sequelae. Medicine (Baltimore) 1982;61:310–316. 77. Plotkin SA, Arbeter AA, Starr SE. The future of varicella vaccine. Postgrad Med 1985;61:155–162. 78. Balducci J, Rodis JF, Rosengren S, et al. Pregnancy outcome following first-trimester varicella infection. Obstet Gynecol 1991;79:5–6. 79. Choo PW, Donahue JG, Manson JE, et al. The epidemiology of varicella and its complications. J Infect Dis 1995;172:706–712. 80. Gogos CA, Bassaris HP, Vagenakis AG. Varicella pneumonia in adults. A review of pulmonary manifestations, risk factors and treatment. Respiration 1992;59:339–343. 81. Preblud SR. Varicella: complications and costs. Pediatrics 1986;76[Suppl]:728–735. 82. Preblud SR, Orenstein WA, Bart K. Varicella: clinical manifestations, epidemiology, and health impact on children. Pediatr Infect Dis 1984;3:505–509. 83. Centers for Disease Control and Prevention. Prevention of varicella: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1996;45:1–36. 84. Pastuszak AL, Levy M, Schick B, et al. Varicella infection in pregnancy. N Engl J Med
1994;330:901–905. 85. Committee on Infectious Diseases. Live attenuated varicella vaccine. Pediatrics 1995;95:791–796. 86. Centers for Disease Control and Prevention. Impact of vaccines universally recommended for children—United States, 1990–1998. MMWR Morb Mortal Wkly Rep 1999;48:243–248. 87. Centers for Disease Control and Prevention. Establishment of Varivax Pregnancy Registry. MMWR Morb Mortal Wkly Rep 1996;45:239. 88. Enders G, Miller E, Craddock-Watson J, et al. Consequences of varicella and herpes zoster in pregnancy: prospective study of 1739 cases. Lancet 1994;343:1548–1550. 89. Fedson DS, Musher DM, Eskola J. Pneumococcal vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:553–607. 90. Musher DM. Streptococcus pneumoniae. In: Mandel GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. New York: Churchill Livingstone, 2000:2128–2147. 91. Markowitz JS, Pashko S, Gutterman, et al. Death rates among patients hospitalized with community-acquired pneumonia: a examination with data from three states. Am J Public Health 1996;86:1152–1154. 92. Applebaum PC. Antimicrobial resistance in Streptococcus pneumoniae. An overview. Clin Infect Dis 1992;15:77–83. 93. Spika JS, Facklam RR, Plikaytis BD, et al. Antimicrobial resistance of Streptococcus pneumoniae in the United States, 1979–1987. J Infect Dis 1991;163:1273–1278. 94. Doern GV, Pfaller MA, Kugler K, et al. Prevalence of antimicrobial resistance among respiratory tract isolates of Streptococcus pneumoniae in North America: 1997 results from the SENTRY antimicrobial surveillance program. Clin Infect Dis 1998;27:764–770. 95. Centers for Disease Control and Prevention. Prevention of pneumococcal disease. Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1997;46[RR-8]:1–24. 96. Centers for Disease Control and Prevention. Pneumococcal and influenza vaccination levels among adults age ³65 years—United States, 1993. MMWR Morb Mortal Wkly Rep 1996;45:853–859. 97. Centers for Disease Control and Prevention. Pneumococcal and influenza vaccination levels among adults aged 65 years—United States 1995. MMWR Morb Mortal Wkly Rep 1997;46:913–919. 98. Nichol KL, MacDonald RM, Hague M. Side effects associated with pneumococcal vaccination. Am J Infect Control 1997;25:223–228. 99. Robbins JB, Schneerson R. Polysaccharide-protein conjugates: a new generation of vaccines. J Infect Dis 1990;161:821–832. 100. Siber GR. Pneumococcal disease. Prospects for a new generation of vaccines. Science 1994;265:1385–1387. 101. Kayhty H, Eskola J. New vaccines for the prevention of pneumococcal infections. Emerg Infect Dis 1996;2:289–298. 102. Arden NH, Kilbourne ED. Inactivated influenza vaccines. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:531–551. 103. Centers for Disease Control and Prevention. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 2000;49[RR-3]:1–38. 104. Maassab HF, Herlocher ML, Bryant ML. Live influenza virus vaccine In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:909–927. 105. Eickhoff TC, Sherman IL, Serfling RE. Observations on excess mortality associated with epidemic influenza. JAMA 1961;176:776–782. 106. Freeman DW, Barno A. Deaths from Asian influenza associated with pregnancy. Am J Obstet Gynecol 1959;78:1172–1175. 107. Schoenbaum SC, Weinstein L. Respiratory infection in pregnancy. Clin Obstet Gynecol 1979;22:293–300. 108. Schoenbaum SC. Economic impact of influenza: the individual's perspective. Am J Med 1987;82[Suppl 6A]:26–30. 109. Healthy people 2000: national health promotion and disease prevention objectives. Washington: US Dept of Health and Human Services, Public Health Service; 1991; DHHS publication no (PHS)91-50212. 110. Walker FJ, Singleton JA, Lu PJ, et al. Influenza vaccination of health-care workers in the United
States, 1989–1997. Infect Control Hosp Epidemiol 2000;21:113. 111. Gross PA, Hermogenes AW, Sacks HS, et al. The efficacy of influenza vaccine in elderly persons: a meta-analysis and review of the literature. Ann Intern Med 1995;123:518–527. 112. Nichol KL, Margolis KL, Wuorenma J, et al. The efficacy and cost-effectiveness of vaccination against influenza among elderly persons living in the community. N Engl J Med 1994;331:778–784. 113. Mullooly JP, Bennett MD, Hornbrook MC, et al. Influenza vaccination programs for elderly persons: cost-effectiveness in a health maintenance organization. Ann Intern Med 1994;121:947–952. 114. Nichol K, Lind A, Markgolis K, et al. The effectiveness of vaccination against influenza in healthy, working adults. N Engl J Med 1995;333:889–903. 115. Bierman CW, Shapiro GG, Pierson WE, et al. Safety of influenza vaccination in allergic children. J Infect Dis 1997;136:S652–S655. 116. James JM, Zeiger RS, Lester MR, et al. Safe administration of influenza vaccine to patients with egg allergy. J Pediatr 1998;133:624–628. 117. Heinoven OP, Shapiro S, Monson RR, et al. Immunization during pregnancy against poliomyelitis and influenza in relation to childhood malignancy. Int J Epidemiol 1973;2:229–235. 118. Murray Dl, Imagawa DT, Okada DM, et al. Antibody response to monovalent A/New Jersey/8/76 influenza vaccine in pregnant women. J Clin Microbiol 1979;10:184–187. 119. England JA, Mbawinke IN, Hammil H, et al. Maternal immunization with influenza or tetanus toxoid vaccine for passive antibody protection in young adults. J Infect Dis 1993;168:647–656. 120. Stanwell-Smith R, Parker AM, Chakraverty P, et al. Possible association of influenza A with fetal loss: investigation of a cluster of spontaneous abortions and stillbirths. Commun Dis Rep 1994;4:28–32. 121. Harris JW. Influenza occurring in pregnant women: a statistical study of thirteen hundred and fifty cases. JAMA 1919;72:978–980. 122. Noble GR. Epidemiological and clinical aspects of influenza. In: Beare AS, ed. Basic and applied influenza research. Boca Raton, FL: CRC Press, 1982:41–42. 123. Widelock D, Csizmas L, Klein S. Influenza, pregnancy and fetal outcome. Public Health Rep 1963;78:1–11. 124. Schhab GZ, Glezen WP. Influenza virus. In: Gonik B, ed. Viral diseases in pregnancy. New York: Springer-Verlag New York, 1994:215–223. 125. Kort BA, Cefalo RC, Baker VV. Fatal influenza A pneumonia in pregnancy. Am J Perinatol 1986;3:179–182. 126. Neuzil KM, Reed GW, Mitchell EF, et al. Impact of influenza on acute cardiopulmonary hospitalizations in pregnant women. Am J Epidemiol 1998;148:1094–1102. 127. Evans J, Fikrig E. Lyme disease vaccine. In: Plotkin SA, Orenstein WA, eds. Vaccines. Philadelphia: WB Saunders, 1999:968–982. 128. Centers for Disease Control and Prevention. Recommendations for the use of Lyme disease vaccine. Recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep 1999;48[RR-7]:1–17. 129. Dennis DT. Epidemiology, ecology and prevention of Lyme disease. In: Rahn DW, Evans J, eds. Lyme disease. Philadelphia: American College of Physicians, 1998:7–34. 130. Steere AC, Sikand VK, Meurice F, et al. Vaccination against Lyme disease with recombinant Borrelia burgdorferi outer-surface lipoprotein A with adjuvant. N Engl J Med 1998;339:216–222. 131. Jenson HB. Pocket guide to vaccination and prophylaxis. Philadelphia: WB Saunders, 1999;144–196. 132. Siber GR, Werner BC, Hasley NA. Interference of immune globulin with measles and rubella immunization. J Pediatr 1993;122:204–211. 133. American College of Physicians, Task Force on Adults Immunizations and Infectious Diseases Society of America. Immunizations for special groups of patients. In: Guide for adult immunization, 3rd ed. Philadelphia: American College of Physicians, 1994:25–41. 134. Centers for Disease Control and Prevention. Yellow fever vaccine: recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR Morb Mortal Wkly Rep 1990;39[RR-6]:1–6. 135. Losonsky GA, Fishaut JM, Strussenberg J, et al. Effect of immunization against rubella on lactation products II. Maternal-neonatal interactions. J Infect Dis 1982;145:661–666. 136. Centers for Disease Control and Prevention. Health information for international travel 1996–97. Washington: US Government Printing Office, 1996. 137. Centers for Disease Control and Prevention. Immunization of health-care workers. MMWR Morb
Mortal Wkly Rep 1997;46:1–42.