Advances in Oto-Rhino-Laryngology Vol. 55
Series Editor
KARGER
W Arnold, Munich
Basel· Freiburg· Paris· London· New York· New Delhi· Bangkok· Singapore· Tokyo· Sydney
Vestibular Dysfunction and Its Therapy
Volume Editor
U Battner, Munich
28 figures. 3 in color and 10 tables. 1999
KARGER
Basel· Freiburg· Paris· London· New York· New Delhi· Bangkok· Singapore· Tokyo· Sydney
Prof. Dr. med. U. Buttner Department of Neurology Klinikum GroBhadem Ludwig Maximilians University Marchioninistrasse 15 D-81366 Munich (Germany)
Library of Congress Cataloging-In-Publicatlon Data Vestibuiar dysfunction and its therapy / voiume editor, U. Bottner. (Advances in oto-rhlno-Iaryngology; vol. 55) Includes bibliographical references and indexes. I. Vestibular apparatus - Diseases - Treatment. 2. Labyrinth (Ear) - Diseases - Treatment. 3. Nystagmus - Treatment. 4. Meniere's disease - Treatment. 5. Vertigo - Treatment. I. BLittner, U. II. Series. [DNLM: J. Labyrinth Diseases - therapy 2. Vestibular Diseases - therapy.
3. Vestibular Diseases - physiopathology. 4. Vestibule - physiopathology. WI AD701 v. 55 1999/ WV 255 V5826 19991 RF16.A38 vol. 55 [RF2601 617.51 s-dc21 [617.8821 ISBN 3-8055-6702-2 (hardcover: alk. paper)
Bibliographic Indices. This publication is listed In bibliographic services, including Current Contents Index Medlcus.
and
Drug Dosage. The authors and the publisher have exerted every e ort to ensure that drug selection and dosage set forth In this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions This is particularly Important when the recommended agent Is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical. Lncluding photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher.
Copyright 1999 by S. Karger AG, P.O. Box, CH-4009 Basel (Switzerland) Printed In Switzerland on acid-free paper by Reinhardt Druck, Basel ISBN 3-8055-6702-2
Contents
~Preface
C!::: Brainstem and Cerebellar Structures for Eye Movement Generation Horn, A.K.E.; Buttner, u.; Buttner-Ennever, JA. (Munich) [26 Intrinsic Physiological and Pharmacological Properties of Central Vestibular Neurons Vidal, P-P; Vibert, N. (Paris); Serafin, M. (Geneva); Babalian, A. (Paris); MOhlethaler, M. (Geneva); de Waele, C. (Paris) 82] Vestibular Compensation Curthoys, 1.S. (Sydney); Halmagyi, G.M. (Camperdown)
~
Vestibular I\leuritis Strupp, M.; Brandt, T. (Munich)
[1371 Meniere's Disease Hamann, K.-F.; Arnold, W (Munich)
11691 Benign Paroxysmal Positioning Vertigo Brandt, T. (Munich) 1195' Drug Therapy of Nystagmus and Saccadic Intrusions Buttner, u.; Fuhry, L. (Munich)
[228
Nonpharmacological Treatment of Nystagmus Leigh, R.J (Cleveland, Ohio)
1241J Subject Index
Preface
Vertigo and dizziness are one of the most common complaints of patients consulting a doctor. These symptoms can be very disturbing to the patient, but a precise diagnosis is often di cult to make and, in many instances, satisfying therapy is lacking. The diagnostic approach has to be multidisciplinary including otolaryngology, ophthalmology and neurology. In November 1996 an international conference on 'Therapy of ocular motility and related visual disturbances' was held at Case Western Reserve University, Cleveland, Ohio, and was organized by H.]. Kaminiski and R.J Leigh [conference summary see, Neurology 1997;48:1178-1184]. At this conference it became quite clear that impressive progress has been made on the basic neurophysiological and neuropharmacological mechanisms of ocular motility over the last 10 years, and has resulted in a number of successful therapeutical studies. However, it was also obvious that more research and clinical studies are required. Particularly in the field of drug therapy, the number of patients investigated in double-blind controlled studies is still very small. Over the last years the basic mechanisms of benign paroxysmal positioning vertigo (BPPV), one of the most common causes of vertigo, have successfully been worked out. The correct application of these findings to physical therapy has led to impressive, often astonishing results. With a single maneuver lasting less than 5 min patients who had su ered from vertigo for many years can often be cured. The chapters in this book present the current state of research and clinical studies in this widely relevant field. They are aimed at the basic scientist wUI'king in the field uf neuruphysiulugy and neurupharmacolugy uf the vestilmlar and oculomotor system, who wishes to become more familiar with the clinical aspects and therapy. They are also aimed at clinicians interested in neuro-otology and neuro-ophthalmology, providing both information about the neuropharmacological and neurophysiological basis and, in addition, the
Contents
VII
clinical and therapeutic approach to patients with vestibular and ocular motility disorders. The book is divided into eight chapters. The first two chapters provide an overview of the structures in the brainstem and cerebellum involved in oculomotor and vestibular control with the main emphasis on neuropharmacological aspects. The chapter by Vidal et al. covers the vestibular nuclei, and the contribution of Horn et at. other brainstem and cerebellar structures. Peripheral vestibular disorders are treated in the fol1owing chapters by Curthoys and Halmagyi (Vestibular compensation), Strupp and Brandt (Vestibular neuritis), Brandt (Benign paroxysmal positioning vertigo) and Hamann and Arnold (Meniere's disease). The last two chapters address central eye movement disorders (nystagmus, saccadic intrusions) and their pharmacological (Buttner and Fuhry) and non-pharmacological treatment (Leigh). The aim of this book is to aid the diagnosis and treatment of patients with vestibular and oculomotor disorders, and it will also perhaps stimulate research for better therapy. Ulrich Battner Munich, August 1998
Preface
VIII
Btittner U. (ed): Vestibular Dysfunction and Its Therapy. Adv Otorhinolaryngol. Basel, Karger, 1999, vol 55, pp 1-25
Brainstem and Cerebellar Structures for Eye Movement Generation AKE. Horn a , U Battner b, JA Battner-Ennever a a b
Institute of Anatomy, Ludwig Maximilians University, Munich and Department of Neurology, Klinikum Grosshadern, Munich, Germany
Over the last years much progress has been made in the identification and characterization of functional cell groups of the premotor system for eye movements in the brainstem and cerebellum of the monkey. In parallel, attempts have been made to identify the homologous cell populations in humans, which now can be analysed at a cellular level with postmortem examinations uf brains with eye movement uisun.lers [Hum et aI., 1996], With the uevelupment of immunocytochemical techniques and the improvement of tract -tracing methods, the histochemical properties and the neurotransmitter profile of some populations have been studied in mammals. The knowledge about the chemical properties of the functional cell groups for eye movement generation is a prerequisite for the development of possible pharmaceutical therapies in patients. The current state of research excluding the vestibular nuclei [see Chapter by Vidal et al.] is reviewed in the present chapter.
Rostral Interstitial Nucleus of the Medial Longitudinal Fascicle
Structure and Function I The rostral interstitial nucleus of the medial longitudinal fascicle (riMLF)1 is essential for the generation of vertical and torsional saccades [Buttner et a1., 1977; Vilis et aI., 1989; Crawford and Vilis, 1992]. The riMLF lies in the mesencephalic reticular formation and forms the rostral medial part of thel fields of Forel. In transverse sections the riMLF appears as a wing-shaped
Supported by the Deutsche Forschungsgemeinschaft (SFB 462).
nucleus below the thalamus, dorsomedial to the red nucleus, dorsally bordered by the posterior branch of the thalamosubthalamic artery. From the caudaUy adjacent interstitial nucleus of Cajal (iC) the riMLF is roughly separated by the traversing fibres of the tractus retroflexus and the rostral end is formed by the traversing fibres of the tractus thalamosubthalamicus [Bottner-Ennever and Bottner, 1988J. The riMLF contains medium-lead burst neurons that discharge with highfrequency bursts 8-15 ms before and during vertical and torsional saccades. The right riMLF contains up- and down-burst neurons with a clockwise torsional component, and the left riMLF up- and down-burst neurons with a counterclockwise torsional component [Vilis et aI., 1989J. In primates, upand down-burst neurons are intermingled [Bottner et aI., 1977; Moschovakis et aI., 1991a, b; Horn and BClttner-Ennever, 1998], although in cats a tendency of upward neurons lying more caudally than downward neurons was observed [Wang and Spencer, 1996]. Within the small to medium-sized neurons of the riMLF, the saccadic premotor burst neurons belong to the medium-sized cell population (mean diameter 22 m) and contain the calcium-binding protein parvalbumin, which is present in neurons with high-firing activity [Bairnbridge et aI., 1992; Horn and Bottner-Ennever, 1998J.
A erent and E erent Connections The burst neurons project monosynaptically to the motoneurons of the vertical pulling extraocular eye muscles in the oculomotor and trochlear nuclei providing the premotor signal for the saccadic eye movement [Moschovakis et aI., 1991a, b; Horn and Bottner-Ennever, 1998]. There are additional connections of the riMLF to the contralateral riMLF via the ventral commissure, to the iC, the paramedian tract (PMT) neurons and sparsely to the spinal cord [Moschovakis et aI., 1991a, b; Wang and Spencer, 1996; Holstege and Cowie, 1989]. The burst neurons in the riMLF receive a strong input from the inhibitory saccadic omnipause neurons within the paramedian pontine reticular formation (PPRF) [Horn et aI., 1994J. In addition, the riMLF receives a erents from neurons in the iC that are not oculomotor-projecting premotor neurons [Moschovakis et aI., 1991b], but perhaps the saccade-related burst neurons [Helmchen et al., 1996], and the superior colliculus [Nakao et aI., 1990]. A minor projection from the medial vestibular nucleus targets mainly the mediocaudal part of the riMLF, and might derive from secondary vestibular neurons [Buttner-Ermever and Lang, 1981; Matsuu et aI., 1994]. Transmitters Anatomical studies in the riMLF of the cat revealed the presence of inhibitory premo tor burst neurons using GABA as transmitter. These
HornlBottner/Bottner-Ennever
GABA-immunoreactive neurons are concentrated in the dorsomedial part of riMLF [Spencer and Wang, 1996J. In contrast, the riMLF of the monkey contains only few small GABA-immunoreactive neurons, presumably local interneurons, which are not premotor burst neurons. So far there is no evidence for the presence of GABAergic premotor burst neurons in the monkey [Carpenter et a!., 1992; Horn, pel's. observationsJ. Microinjections of muscimol, a GABA agonist, into the riMLF result in similar deficits as those observed after kainic acid lesions: a loss of the torsional component to the ipsilateral side after unilateral injections. This indicates the presence of GABAA receptors on premotor burst neurons [Vilis et aI., 1989; Suzuki et aI., 1995; Crawford et a!., 1992]. Immunocytochemical studies showed that the premotor saccadic burst neurons in the riMLF receive a strong innervation of GABA- and also glycine-immunoreactive terminals [Horn et a!., 1994; Horn, pers. observations). Possible sources of the GABAergic a erents are the saccade-related burst neurons in the iC, which might provide an inhibitory feedback signal to the riMLF [Moschovakis et aI., 1996; Helmchen et aI., 1996], or collaterals of the vestibula-oculomotor connection, which is GABAergic for the vertical system [Spencer et aI., 1992J. Glycinergic a erents arise from the saccadic omnipause neurons in the PPRF [Horn et aI., 1994]. There are limitations using immunocytochemlcal techniques for the detection of amino acid transmitters in cell bodies. Whereas GABA and glycine are thought to occur in high concentrations exclusively in the neurons, which use them as transmitters, glutamate and aspartate are metabolic products as well. Almost all neuronal somata express more or less glutamate- and aspartate-immunoreactivity [Yingcharoen et a!., 1989J. Only careful studies applying the appropiate antibody dilutions with additional quantitative measurements of the staining intensity at the ultrastructural level may di erentiate between transmitter pools and metabolic pools. This distinction is easier in nerve terminals, of which for example the glutamatergic were shown to contain 2-3 times the average level of glutamate [for review, see Storm-Mathisen et aI., 1995J. In the vestibula-oculomotor system such quantitative studies were performed only in cats so far and revealed glutamate- and aspartate-positive neurons that project to the motoneurons in the oculomotor and trochlear nucleus, suggesting that glutamate and aspartate are the transmitters of excitatory premotor burst neurons in the riMLF in this species [Spencer and Wang, 1996J. This study did not show whether aspartate and glutamate are culucalized in the same neuruns, !.Jut !.Juth transmitter populations, e.g. aspartate and glutamate, had di erent ultrastructural and synaptic features (synaptic vesicle shape, degree of postsynaptic specializations) , which indicates that aspartate and glutamate are released from di erent terminals [Spencer and Wang, 1996].
Brainstem and Cerebellar Structures for Eye Movement Generation
3
Interstitial Nucleus of Cajal
Structure and Function As the riMLF, the iC is related to vertical and torsional eye movements, but more involved in the integration of eye-velocity into eye-position signals [Fukushima et aI., 1992] and eye-head coordination [Fukushima, 1987], rather than in the generation of vertical saccades [Helmchen et aI., 1998]. The iC lies within the medial longitudinal fascicle lateral to the rostral pole of the oculomotor nucleus. In humans its rostral border to riMLF is di cult to define, because both nuclei have a similar cellular appearance in Nissl-stained sections. The distinction between both is made easier with histochemical markers, such as the immunostaining pattern for the calciumbinding protein parvalbumin [Horn and Bottner-Ennever, 1998J. The iC is a rather compact nucleus consisting of small to medium-sized neurons with few large-sized polygonal neurons interspersed [Bianchi and Gioia, 1991]. Recording experiments in alert cats and monkeys revealed several functional cell groups within the iC: (1) Burst-tonic and tonic neurons encode the eye position and they are involved in the vertical integrator function [review in Fukushima et aI., 1992]. Accordingly, a lesion of the posterior commissure, which contains the crossing fibres of the burst-tonic and/or tonic neurons, results in the inability to hold eccentric gaze after vertical saccades [Partsalis et aI., 1994J. (2) Approximately one third of the eye movement-related neurons are saccade-related burst neurons with a similar firing pattern as premotor burst neurons in the riMLF [Helmchen et al., 1996]. Since they do not project to the eye muscle motoneurons but send collaterals back to the riMLF, they are thought to be part of an inhibitory feed back system carrying eye displacement information [Moschovakis et aI., 1996J. This hypothesis is supported by the observation that only the saccade amplitude was reduced after muscimol injections into the iC, but not the saccade velocity [Helmchen et aI., 1998J. (3) In the cat another group of neurons was identified, classified as bursterdriving neurons (BONs) or 'vestibular plus saccade' neurons that discharge a burst of spikes shortly before (but with longer lead time of 34 ms than medium-lead burst neurons in the riMLF) and during downward saccades. During the upward slow phase of vestibular stimulation the BONs discharged at gradually increasing firing rates [Fukushima et al., 1991, 1995]. Vertical BONs were also identified in and around the iC in monkeys [Kaneko and Fukushima, 1993]. A erent and E erent Connections There are three main e erent projection systems leaving the iC [Kokkoroyannis et aI., 1996]: the ascending system has strong projections to the ipsilat-
HornlBottner/Bottner-Ennever
eral mesencephalic reticular formation including riMLF and zona incerta, weaker projections to the ipsilateral centromedian and parafascicular thalamic nuclei and bilateral to the mediodorsal, central medial and lateral nuclei of the thalamus. The descending system projects through the medial longitudinal fascicle to innervate the ipsilateral oculomotor and trochlear nucleus, the ipsilateral PPRF, the rostral cap of the abducens nucleus (VI) as part of the PMT cell groups [Bottner-Ennever, 1992), the vestibular nuclei, the nucleus prepositus hypoglossi, the gigantocellular portion of the reticular formation, the inferior olive and the ventral horns of Cl up to C4. The commissural system projects via the posterior commissure to the nucleus of the posterior commissure bilaterally, the contralateral iC and the contralateral oculomotor and trochlear nuclei to innervate monosynaptically the motoneurons of vertical pulling extraocular eye muscles [Kokkoroyannis et aI., 1996J. This system arises from medium-sized, presumably burst-tonic and/or tonic neurons, which contain the calcium-binding protein parvalbumin [Horn and Bottner-Ennever, 1998J. Saccade-related burst neurons appear not to project to eye muscle motoneurons, but have recurrent collaterals to the riMLF [Moschovakis et aI., 1996J. The iC receive inputs from premotor neurons that encode eye- or head-velocity signals: via collaterals from secondary vestibulo-ocular neurons, excitatory signals from the contralateral side and inhibitory signals from the ipsilateral side [Iwamoto et aI., 1990] and from the y-group of the vestibular nuclei [Fukushima et aI., 1986J. Most probably the burst-tonic and tonic neurons receive a erents from the saccadic burst neurons in the riMLF [Moschovakis et al.. 1991 a, bJ. TransmHlers
Up to date there is no systematic study about the transmitter profile in the ie. There is some evidence that the iC contains small and medium-sized GABAergic neurons, but their connectivity has not been studied yet [Horn, pers. observations]. These neurons could comprise inhibitory premotor neurons projecting to the oculomotor and trochlear nucleus [Nakao et al.. 1990; Nakao and Shiraishi, 1985; Schwindt et aI., 1974], or they form part of the inhibitory feedback system to riMLF as suggested by Moschovakis et aI., [1996J (see above). The presence of symmetric synaptic profiles at the somata of all neuron types, but predominantly at large neurons, indicate monosynaptic inhibitory a erents to iC [Bianchi and Gioia, 1995J. The iC contains numerous GABA-irIlInunoreactive terminals, in part uutlining neuronal sumata [Hum, pers. observationsJ. The source of these inputs has not been studied yet, but part of them could arise from collaterals of inhibitory secondary vestibulooculomotor projections from the ipsilateral superior vestibular nuclei, which were shown to be GABAergic [Wentzel et aI., 1995; De la Cruz et aI., 1992].
Brainstem and Cerebellar Structures for Eye Movement Generation
In the monkey, unilateral microinjections of the GABAA-receptor agonist muscimol result in a torsional and vertical spontaneous nystagmus (torsional fast phases always toward the lesion), a severe gaze-holding deficit for vertical and torsional saccades, and a tonic torsional eye-position shift to the contralesional side [Crawford and Vilis, 1992; Helmchen et al., 1998]. So far nothing is known about the nature of excitatory neurons that drive oculomotor neurons. Weakly choline acetyltransferase (Chat)-positive neurons were found in the iC of humans [Juncos et aI., 1991], which appear not to be premotor neurons projecting to the oculomotor nucleus [Carpenter et aI., 1992]. Only a weak innervation by cholinergic a erents is reported in cats, whose origin is unknown [Kimura et aI., 1981].
Paramedian Pontine Reticular Formation
Structure and Function The PPRF was introduced as a functional term as the site where lesions produce a horizontal gaze palsy [Cohen and Komatsuzaki, 1972]. Anatomically the PPRF extends from rostral to caudal from the trochlear to the abducens nucleus and includes the nucleus reticularis pontis oralis (NRPO) , the nucleus reticularis pontis caudalis (NRPC) with corresponding midline structures, and the nucleus paragigantocellularis dorsalis (PGD) [Bottner-Ennever and Buttner, 1988; Hepp et aI., 1989]. There are several classes of eye movement-related neurons within the PPRF: Excitatory burst neurons (EBNs) lie as a compact group withln the dorsomedial part of the NRPC just rostral to the saccadic omnipause neurons (OPNs) [Strassman et aI., 1986a; Horn et aI., 1995J. Inhibitory burst neurons (lENs) are located in the PGD just beneath the rostral pole of the abducens nucleus [Strassman et aI., 1986b; Horn et aI., 1995]. Both, the EBNs and lENs form populations of medium-sized neurons within the NPRC and PGD, respectively, and contain the calcium-binding protein parvalbumin [Horn et aI., 1995J. The EBNs and lENs exhibit a high-frequency burst during horizontal saccades and are otherwise silent. OPNs lie within a distinct nucleus between the traversing fibres of the abducens nerve at the midline, which was named nucleus raphe interpositus [Bottner-Ennever et aI., 1988]. In monkeys the OPNs lie in two compact cell columns, which appear more scattered around the midline in humans [Hurn et aI., 1994]. The medium-sized OPNs are hurizontally oriented with their long dendrites reaching across the midline. They contain the calcium-binding protein parvalbumin and a high level of cytochrome oxidase activity, which is presumably related to the high-firing activity [Buttner-Ennever et aI., 1988; Horn et aI., 1994]. Saccade-related long-
HornlBottner/Bottner-Ennever
6
lead burst neurons (LLBNs) exhibit an additional irregular low frequency activity before the saccade-related burst and they are thought to activate premotor medium-lead burst neurons. LLBNs are found at several locations in the brainstem and can be divided into several groups on the basis of their location and their projection targets [Moschovakis et al., 1996]: pontine LLBNs lie witWn the NRPC, intermingled with EBNs, and more rostrally in the NRPO and nucleus reticularis tegmenti pontis (NRTP) [Kaneko et a!., 1981; Hepp and Henn, 1983; Scudder et a!., 1996J and in the PGD intermingled with IBNs [Scudder et al., 1988J. One class of LLBNs, reticulospinal neurons, were first described in the cat as neurons with long-lead burst activity related to eye-neck (head) movements, which lie either rostrally or ventrally to the abducens nucleus [Grantyn et a!., 1987J. During fixation and slow eye movements the OPNs exert monosynaptically a tonic inhibition onto premotor EBNs and IBNs in the riMLF and NRPC. During saccades the OPNs are inhibited by polysynaptic inputs possibly via LLBNs - from the superior colliculus. A erent and E erent Connections EBNs project directly to the motoneurons and internuclear neurons in the abducens nucleus thereby activating the ipsilateral lateral rectus muscle and the contralateral medial rectus muscle. In addition, the EBNs send an ipsilateral projection to the IBNs, thereby inhibiting the contralateral abducens nucleus and preventing the contralateral lateral rectus muscle from abduction during saccades [Strassmann et al., 1986a]. The OPNs send direct projections to the vertical premotor burst neurons in the riMLF, and the horizontal EBNs and IBNs in the NRPC and PGD [Ohgaki et a!., 1989; Strassman et aI., 1987J. Single-cell reconstructions of identified LLBNs in the NRPO of the monkey revealed projections to the dorsomedial part of the NRPC (EBN region), the PGD (IBN region), the NRTP and the nucleus reticularis gigantocellularis (NRG) [Scudder et aI., 1996J. In the cat, reticulospinal neurons were shown to project to the abducens nucleus, the facial nucleus, the medial and lateral vestibular nuclei and the nucleus prepositus hypoglossi [Grantyn et al., 1987]. Only recently, identified reticulospinal neurons in the monkey were shown to project to the spinal cord giving 0 collaterals to the PGD, the caudal half of the nucleus prepositus hypoglossi, but none to the abducens or facial nucleus, indicating that the contwl uf eye and head movements is mediated by mUI'e separated pathways in primates than in cats [Robinson et aI., 1994; Scudder et al., 1996J. Recent work in monkeys showed that the IBNs and EBNs receive via the predorsal bundle a strong a erent input from the intermediate layers of the superior colliculus motor map mediating large horizontal saccades, but
Brainstem and CerebelJar Structures for Eye Movement Generation
7
not the OPNs [Bottner-Ennever et aI., 19971. A strong a erent input to the somata of OPNs was observed from the superior colliculus motor map representing small horizontal saccades and fixation, whereas the EBN and IBN areas receive only a weak innervation [Bottner-Ennever et a!., 1997; Everling et al., 1998]. TransmHlers
The IBNs use glycine as transmitter [Spencer et a!., 1989], whereas the transmitter of the EBNs is not known yet. The OPNs are glycinergic as well, and they receive a similar strong input of glycine- and GABA-immunoreactive terminals on their somata and proximal dendrites [Horn et aI., 1994], which appear the most likely source for providing the brisk inhibition of the tonic activity during saccades. However, it is not clear why in the cat the iontophoretic application of glycine showed none and that of G ABA only a weak suppression of the firing activity of OPNs [Ashikawa et aI., 1991]. So far the origin of these inhibitory inputs is not known. Glycinergic inputs can theoretically derive from glycinergic neurons within the neighbouring formatio reticularis, which lie within reach of the descending projection fibres of the predorsal bundle from the 'small- and large-horizontal-saccade zone' of the superior colliculus motor map [Bottner-Ennever et aI., 1997]. The strong supply with glutamate-immunoreactive terminals on the proximal dendrites of the OPNs [Horn et a!., 1994] might derive from the 'rostral pole' of the superior colliculus, mediating fixation [Munoz and Wurtz, 1993; Pare and Guitton, 1994; BottnerEnnever et a!., 1997]. Cortical projections from the frontal and supplementary eye fields could be another source of glutamatergic a erents to OPNs [Shook et a!., 1988; Stanton et aI., 1988]. The systemic or local iontophoretic application of serotonin results in a complete suppression of the tonic firing rate of OPNs, which can be abolished by a prior intravenous injection of methysergide, a serotonin blocker, as shown in the cat [Furuya et a!., 1992]. So far, nothing is known about the nature of the transmitters used by LLBNs including reticulospinal neurons, and whether they are excitatory or inhibitory.
Paramedian Tract Neurons
Structure
i:lIJ(.l
FunctiuIJ
PMT neurons are defined as groups of cerebellar-projecting neurons that lie around the midline fibre bundles of the pons and medulla. The PMT cell groups have been brought to the attention of oculomotor neuroanatomists on account of their projection to the flocculus and ventral paraflocculus region
Horn/Buttner/Buttner-Ennever
8
[Sato et a!., 1983; Blanks et al. 1983; Langer et a!., 1985; Blanks, 1990]. The PMT cell groups could provide the flocculus and ventral paraflocculus of the cerebellum and other areas of the brain with a motor e erence copy ofthe oculomotor output signal. Damage could lead to a disturbance in gaze-holding [Bottner et al. 1995]. There are at least six relatively separate 'PMT cell groups' scattered in the medial longitudinal fascicle, rostral to, and even within, the abducens nucleus, and continuing back to the level of the hypoglossal nucleus. In the cat, rat and monkey they have been given di erent names by di erent investigators: we use the individual terms introduced by Langer and colleagues, 1985. The use of the nomenclature 'medial' and 'caudal interstitial nuclei of the MLF' for the PMT cell groups rostral or caudal to the abducens nucleus appears less satisfactory (as well as unwieldy), partly because they overlook fine di erences in the cell grouping at least in the primate, and also because often the prefix medjalor caudal is dropped and then there is total confusion with the vertical premotor neurons of the riMLF and iC (see above). A erent and E erent ConnecUons Aside from their projection to the flocculus and ventral paraflocculus, the PMT neurons receive an a erent input from all premotor brainstem nuclei known to project to oculomotor motoneurons [Buttner-Ennever and Buttner, 1988; Bottner-Ennever et al.. 1989; Bottner-Ennever, 1992]. TransmHters The transmitter content of the PMT cell groups is unknown: however, we have found that the cell somata are not serotoninergic, or catecholaminergic [pers. observation]. In addition, the PMT cell groups contain high levels of cytochrome oxidase and acetylcholinesterase. The neurons are chromophilic, medium-sized cells that lie immediately lateral to the smaller-celled raphe nuclei. The di erence between them is easily demonstrated by immunocytochemical stainings with serotonin antibodies: the raphe nuclei are labelled, but the PMT cells are not.
Abducens Nucleus
Structure am} FUIlcUun The abducens nucleus (VD contains at least three functional cell groups: (1) motoneurons innervating the lateral rectus muscle; (2) internuclear neurons, and (3) floccular-projecting neurons in the rostral cap, which belong to the PMT neurons (see above) [Buttner-Ennever, 1992]. Motoneurons and internuclear neu-
Brainstem and Cerebellar Structures for Eye Movement Generation
9
rons exhibit the same burst-tonic firing pattern during eye movements. However, only the motoneurons carry conjugate- and vergence-related signals, whereas internuclear neurons do not carry vergence-related signals [Delgado-Garciaet aI., 1986a, b; Zhou and King, 1998J. The motoneurons contain the calcium-binding protein parvalbumin, but in the cat at least 80% of the internuclear neurons contain a di erent calcium-binding protein, calretinin, which could serve as a histological marker for internuclear neurons [De la Cruz et aI., 1998].
A erent and E erent Connections The internuclear neurons project to the motaneurons of the medial rectus muscle in the contralateral oculomotor nucleus, thereby forming the anatomical basis for conjugate eye movements [Bottner-Ennever and Akert, 1981]. The motaneurons and internuclear neurons receive bilateral a erents from secondary vestibula-ocular neurons in the medial vestibular nuclei, the nucleus prepositus hypoglossi (PPH), the excitatory and inhibitory saccadic burst neurons in the NRPC and PGD and from internuclear neurons of the oculomotor nucleus [Evinger, 1988J. Transmitters In contrast to the cholinergic motoneurons, identified internuclear neurons are not cholinergic [Spencer et aI., 1986; Carpenter et aI., 1992], but appear to use glutamate and aspartate as transmitter [Nguyen and Spencer, 1996]. In cats, serotonin-immunoreactive synaptic contacts were disclosed at the dendrites of abducens neurons, but the serotoninergic dorsal raphe nucleus lying above the caudal oculomotor nucleus was shown not to be the source of these a erents [May et aI., 1987]. The abducens nucleus receives a strong supply of glycinergic inhibitory a erents, which originate from IBNs in the contralateral PGD, the PPH and the ipsilateral medial vestibular nucleus [Spencer et al., 1989J. Anatomical studies revealed a rather weak GABAergic input to the abducens nucleus with a slight tendency of motoneurons being more heavily contacted than internuclear neurons [De la Cruz et aI., 1992; Lahjouji et al., 1995J. Combined tracer and immunocytochemical studies revealed a GABAergic input from internuclear neurons in and above the oculomotor nucleus, which project to the abducens nucleus [De la Cruz et aI., 1992].
Oculomotor and Trochlear Nucleus
Structure and Function The oculomotor nucleus contains the motoneurons innervating the ipsilateral inferior rectus (IR), inferior oblique (10) and medial rectus (MR) muscle
Horn/Buttner/Buttner-Ennever
10
and those for the contralateral superior rectus (SR) muscle organized in a topographic map [Evinger, 1988]. In primates there are three clusters of MR motoneurons, ventral the A group, dorsolateral the B group and medially at the border of the oculomotor nucleus the C group, consisting of smaller motoneurons [Bottner-Ennever and Akert, 1981]. These MR-motoneuron subgroups presumably have di erent functions as indicated by a selective input from the pre tectum to only the C group [Bottner-Ennever et a!., 1996]. Several popu lations of internuclear neurons within and around the oculomotor nucleus were identified, which di er in their projection targets: the spinal cord, the cerebellum, the abducens nucleus [for review, see Evinger, 1988J. SO far little is known about the physiology and their function. The trochlear nucleus contains only motoneurons of the contralateral superior oblique muscle.
A erent and E erent Connections The MR subgroup in the oculomotor nucleus receives a erents via the medial longitudinal fascicle from the internuclear neurons of the contralateral abducens nucleus [Bottner-Ennever and Akert. 1981J and from the ipsilateral ventral part of the lateral vestibular nucleus via the ascending tract of Deiters, a pathway presumably involved in the control of vergence [Baker and Highstein, 1978]. Secondary vestibula-oculomotor projections target on the motoneurons of vertical pulling eye muscles, Le. IR, 10, MR and SR, via excitatory fibres from the superior and medial vestibular nuclei from the contralateral side, and via inhibitory fibres from the superior vestibular nucleus of the ipsilateral side [for review, see Bottner-Ennever, 1992J. The motoneurons of vertical pulling eye muscles in the oculomotor and trochlear nuclei receive bilateral projections from the iC (see above), and predominantly ipsilateral projections from the riMLF [Horn and Bottner-Ennever, 1998]. Transmitters The motoneurons in the oculomotor and trochlear nuclei are cholinergic [Spencer and Wang, 1996] and express parvalbumin-immunoreactivity [De la Cruz et a!., 1998]. In contrast to the abducens nucleus, the motaneurons of vertical pulling eye muscles in the oculomotor and trochlear nuclei receive a strong GABAergic, but a rather weak glycinergic input [De la Cruz et a!., 1992; Spencer et a!., 1992]. Anatomical studies in the rabbit showed that glycine-immunoreactive terminals were evenly distributed throughout the uculumutor nucleus, amI they were preduminantly fuund at proximal and distal dendrites of motoneurons, and only a few at the somata [Wentzel et aI., 1993J. The authors showed in neighbouring ultrathin sections that all glycineimmunoreactive terminals in the oculomotor nucleus contain GABA as well, whereas only a small fraction of GABA-positive terminals express glycine-
Brainstem and Cerebellar Structures for Eye Movement Generation
11
immunoreactivity [Wentzel et aI., 1993]. The function ofthis colocalization is not clear yet, and it cannot be excluded that glycine is only metabolic. Glycine is also known as coactivator of N-methyl-D-aspartate (NMDA) receptors to potentiate the response to glutamate [Johnson and Ascher, 1987]. GABAergic a erents to the oculomotor and trochlear nucleus originate from inhibitory secondary vestibula-ocular neurons in the ipsilateral superior vestibular nucleus [rabbit: Wentzel et aI., 1995; cat: De la Cruz et aI., 1992] and, at least in the cat, from the riMLF (see above) [Spencer and Wang, 1996]. Possible GABAergic projections from the iC have not been proven yet. There are conflicting reports about a strong GABAergic input to medial rectus motoneurons mediating horizontal eye movements: some authors did not see an obvious di erence of the supply with GABA-immunoreactive terminals at MR motoneurons compared to other motoneuron subgroups in rabbits and cats [De la Cruz et aI., 1992; Wentzel et aI., 1996], whereas a much weaker innervation by GABAergic terminals was observed in cats and monkeys [Spencer et aI., 1992; Horn, pers. observations]. A possible source for GABAergic a erents to MR motoneurons are small GABAergic interneurons scattered in and above the oculomotor nucleus [De la Cruz et aI., 1992]. The contralateral excitatory a erents from secondary vestibula-ocular neurons in the medial and superior vestibular nuclei most probably use glutamate as transmitter [De memes and Raymond, 1982]. Recent anatomical studies in the cat indicate that excitation to MR motoneurons from the internuclear neurons of the contralateral abducens nucleus is mediated by glutamate and aspartate, whereas the a erents from the ascending tract of Deiters use only glutamate as transmitter [Nguyen and Spencer, 1996].
Cerebellum
Structure and Function The cerebellum can be divided into ten lobules, according to Larsell; the median part of each lobule forms the vermis, and the lateral regions - the hemispheres [for reviews, see Voogd et aI., 1996; Berry et aI., 1995]. The cerebellum refines, modifies and coordinates all types of movement, including eye movements. The areas of the cerebellar cortex involved in eye movements are (1) the floccular region, (2) nodulus/uvula and (3) the dorsal vermis. Lesions in each structme lead tu specific eye movement deficits. The fluccular regiuIJ consists of the flocculus and parts of the ventral paraflocculus (part of the tonsilla in man) [Buttner and Buttner-Ennever, 1988]. Lesions here cause a smooth pursuit eye movement (SPEM) deficit, which is more pronounced to the ipsilateral side [Zee et aI., 1981; Leigh and Zee, 1991]. Since the same
HornlBottner/Bottner-Ennever
12
Purkinje cells in the floccular region are involved in SPEM and the visual suppression of the vestibula-ocular reflex (VOR-supp) [Bottner and Waespe, 1984], the SPEM deficit is always combined with impaired VOR-supp [Zee et aI., 1981; Leigh and Zee, 1991; Bottner and Grundei, 1995]. In addition, lesions of the floccular region lead to gaze-evoked nystagmus, whose intensity is highly correlated with the SPEM deficit [Bottner and Grundei, 1995]. Common features are also downbeat and rebound nystagmus, particularly with bilateral lesions [Zee et aI., 1981; Leigh and Zee, 1991]. Pulse-step mismatch dysmetria as seen with lesions of the floccular region causes a postsaccadic drift, since the neural command for the saccade (pulse) and the following new eye position (step) do not match [Leigh and Zee, 1991 J. A characteristic feature of a nodulus/ uvula lesion is periodic alternating nystagmus (PAN), which is known to cause disturbing oscillopsia [Halmagyi et al., 1980; also see Chapter by Bottner and Fuhry]. The nodulus has an inhibitory e ect on the 'velocity storage' mechanism in the vestibular nuclei. Consequently, lesions of the nodulus/uvula lead to prolonged time constants of the postrotatory vestibular nystagmus. They also a ect the 'dumping' of vestibular nystagmus by otolith stimulation and cause spontaneous nystagmus in the dark [Waespe et aI., 1985J. It has also been implicated in the generation of motion sickness [Money, 1970J. Lobulus VI and VII of the posterior vermis are considered as the oculomotor vermis [Yamada and Noda, 1987]. Lesions here lead to saccadic step-size error dysmetria and SPEM deficits [Vahedi et aI., 1995; Takagi et aI., 1996J. In contrast to lesions of the underlying fastigial nuclei, which lead to hypermetric saccades [Buttner and Straube, 1995], oculomotor vermis lesions cause hypometric saccades [Vahedi et al., 1995J. The cerebellar cortex has a homogenous histological architecture in terms of the arrangement of the input, output, and interneuronal elements (fig. 1).
A erent and E erent Connections The cerebellum has di erent types of input fibres: (a) The mossy fibres, which comprise all the cerebellar a erents from the brain except those from the inferior olivary nucleus. Mossy fibres are a heterogeneous group of fibres, that converge on granular cells in the granular layer of the cerebellum, forming glomerular synapses 'rosettes' with a variety of morphologies in association with Golgi cells IMugnaini et a1., 1974]. The information is transferred to Purkinje cells via the granular cell axon, the parallel fibres, synapsing on dendritic spines (fig. 1). (0) Climoing fiores, which arise exclusively frum the inferior olive and provide the Purkinje cells with an important excitatory input (fig. 1). They may control the access of mossy fibre inputs to the Purkinje cells. (c) Fine beaded fibres: Some a erent fibres, described in the reviews above, are not typical mossy fibres. They are fine and beaded (varicose), and
Brainstem and Cerebellar Structures for Eye Movement Generation
13
terminate in all layers, the molecular, the Purkinje cell and granular layers (fig. 1). They could be considered as a third category of cerebellar a erents [Hokfelt and Fuxe, 1969]. The Purkinje cells provide the only output from the cerebellar cortex: their axons enter the cerebellar white matter and terminate in the cerebellar or vestibular nuclei. Thus the cerebellar nuclei form the main output of the cerebellum: the lateral or dentate nucleus carrying information from the lateral cerebellum, the hemispheres: the nucleus emboliformis and globosus (nucleus interpositus) from the intermediate hemispheres, while the fastigial nucleus receives e erents from the vermis. The flocculus, ventral paraflocculus and nodulus project directly to the vestibular nuclei. A erents and e erents of the cerebellum travel in the three cerebellar peduncles: the superior cerebellar peduncle is almost exclusively an e erent pathway: only the ventral spinocerebellar tract enters the cerebellum via this route. The brachium conjunctivum is the main e erent tract within the peduncle. The superior cerebellar peduncle decussates in the caudal mesencephalon and provides cerebellar inputs to the regions around the oculomotor nucleus, the red nucleus and the ventrolateral and intra laminar thalamic nuclei. The middle cerebellar peduncle is exclusively a erent, carrying the axons from the pontine nuclei and nucleus reticularis tegmenti pontis (also called nucleus papillioformis) to the cerebellum. While the interior cerebellar peduncle (restiform body and juxtarestiform body) carries a erents and e erents from the spinal cord and brainstem structures [Voogd et a!., 1996], the flocculonodular lobe (lobule X) receives a major input from the vestibular nuclei and is often referred to as the vestibulocerebellum. Its organization has been recently reviewed by Voogd et a!. [1996]. Transmitters The chemical substances used in the cerebellum for fast neurotransmission or neuromodulation have been recently reviewed [Otterson, 1993; Tohyama and Takatsuji, 1998]. Purkinje cells have long been suspected of using GABA as their neurotransmitter. However, the large cell bodies do not stain with GABA antibodies, because the transmitter is quickly transported to its ter-
Fig 1. The main transmitters and input-output pathways of the cerebellum: mossy fibres, from heterogeneous sources with eli erent transmitters, drive glutaminergic granule cells (Gr). Climbing fibres utilize L-glutarnate: both pathways provide an excitatory input to GABAergic Purkinje cells (P). GABAergic interneurons, such as stellate (S), Golgi (Go) and basket cells (B), modulate the signals. The Purkinje axons form the only cerebellar cortex output, and they control the cerebellar and vestibular nuclei. ACH Acetylcholine, ASP aspartate, DA dopamine, ENK enkephalin, 5HT serotonin, GABA -aminobutyric acid, GLU L-glutamate, GLY glycine, NA noradrenaline.
HornlBottner/Bottner-Ennever
14
Molecular
layer
Purk;r.rje cell fayer
Granular"
fDyer
- -G"kBA,- - - -5"1,' laltJri II
DA ' N~
"
GILU
GLUI
ACH
White
(ASP)
matte".
GABA
:CH
110
t
Fl II rl;l 10
I
Inferior (nive I
Brainstem and Cerebellar Structures for Eye Movement Generation
15
minals, where it is easily demonstrated. Similar results were observed with antibodies against the synthesizing enzyme glutamate decarboxylase (GAD), and somatal staining could only be achieved by prior injections of colchicine, which interrupts the intra-axonal transport [Mugnaini and Oertel, 1985J. The messenger RNA for GAD is confined to the soma of GABAergic cells and was clearly detected in the Purkinje cells using in situ hybridization methods [Wuenschell et a!., 1986]. Taurine is also present in the Purkinje cell terminals, but its role is unclear [Otterson, 1993J. In addition, the Purkinje ce]]s contain the calcium-binding proteins parvalbumin and calbindin [Andressen et a!., 1993]. The interneurons of the cerebellum, stellate and basket cells in the molecular layer, and the Golgi cells in the granular layer, are also inhibitory and use GABA as their transmitter. The majority of Golgi cells colocalize GABA and glycine and might corelease both transmitters, as indicated by in vitro studies [see Ottersen, 1993J. The basket and stellate cells contain parvalbumin, whereas the Golgi cells lack parvalbumin [Andressen et a!., 1993J. Although Golgi cells are considered as inhibitory neurons, a recent report in humans demonstrates that most of the Golgi cells in the vermis, flocculus and tonsilia are Chat-immunopositive, a marker for acetylcholine, but the function of the transmitter here is not clear yet [De LacalJe et al. 1993J. Glutamate is the most likely neurotransmitter in climbingfibres, although earlier studies had suggested aspartate could playa role as well [see Ottersen, 1993]. The mossy fibre excitation of the granule cells has been considered to utilize glutamate in a large proportion ofthe fibres, including spinocerebe]]ar, pontinocerebellar and vestibular nerve a erents [for reviews, see Otterson and Laake, 1992; Ji et a!., 1991; Barmack et a!., 1992c; Otterson, 1993J. Two subpopulations of mossy fibres, one GABAergic and one glutamatatergic, originate from the cerebellar nuclei and ascend to the cerebellar cortex [Batini et a!., 1992J. In several species including monkeys, Barmack and colleagues found that all lobules of the cerebellum received a di use cholinergic a erent projection, but particularly dense projections were detected in the vestibulocerebellum in three areas: (a) the uvula-nodulus (lobules IX and X), (b) the flocculus and ventral paraflocculus, and (c) lobules I, II and III of the anterior vermis [compare, rat: Barmack et a!., 1992a, b; with cat: Ikeda etal., 1991J. Similar results were reported for the human [De Lacalle et al. 1993J. In the rat, the cholinergic projection to the uvula and nodulus originated mainly in the caudal medial vestibular nuclei and to a lesser extent in nucleus prepositus hypoglossi [Barmack et a!., 1992bJ. In the light of this stIUng chulinergic mussy fibre input tu the vestibulucerebellum, alung with its association with motion sickness, it is of interest that the most e ective antimotion sickness agents are antimuscarinic. Certain populations of mossy and climbing fibres also contain peptides, such as corticotrophin-releasing factor, in addition to the amino acid transmit-
HornlBottner/Bottner-Ennever
16
tel'. They are considered to facilitate the neuronal response to the transmitter [King et aI., 1992J. The simultaneous stimulation of climbing and parallel fibres leads to a depression in the synaptic e cacy of the parallel fibres for long periods. This phenomenon is called long-term depression, and is considered to reflect cerebellar regulatory mechanisms. Nitric oxide plays a neuromodulatory role in this process [Dawson et aI., 1992; Holschner 1997J. Adenosine modulates the parallel fibre-Purkinje cell transmission [Cuenod et aI., 1989]. The fine-beaded, or varicose, fibres form a thick plexus of fibres terminating in all layers of the cerebellar cortex (fig. 1). They were thought to constitute the monoaminergic input to the cerebellum carrying either serotonin, noradrenaline or possibly dopamine [Ito, 1984], but Barmack et a1. [1992aJ report the presence of cholinergic beaded a erents as well. Thin varicose fibres immunopositive for serotonin could be found in all parts of the cerebellum except the lobule X in cats [Kerr and Bishop 1991], but they are more evenly distributed in the rat [Tohyama and Takatsuji. 1998]. The serotoninergic a erents do not arise from the raphe nuclei, as usually supposed, but from more lateral parts in the reticular formation. and the lateral tegmental field [Kerr and Bishop, 1991J. The well-known raphe input to the cerebellum originates most probably from the PMT celJ groups, which utilize a transmitter other than serotonin (see above). The fine noradrenergic fibres to the cerebellum arise from locus coeruleus, and a distinct dopamlnergic input arises from the A8, A9 and AID cell groups of the mesencephalon [see Otterson, 1993]. The presence of scores of peptides and transmitter-binding sites have been documented within the lobules of the cerebellar cortex by Tohyama and Takatsuji [1998]. There are only two receptors not evenly distributed: one is the dopamine receptor which is concentrated in lobule X (the flocculus), the second are the nicotinic acetylcholine receptor-binding sites, which are concentrated in the ventral cerebellar lobules (the vestibulocerebellum including lobule I).
Fastigial Nucleus
Structure and Function The fastigial nucleus (FN) is the most medial deep cerebellar nucleus. Through a erent and e erent connections it is intimately related to the vestibular nuclei [Noda et aI., 1990J. It also receives a major input from the Purkinje cells of the overlying cerebellar vermis. which is topographically mganized. The anterior vermis (lobules I-V) projects to the rostral FN and the posterior vermis to the caudal FN. Functionally the anterior vermis is part of the spinocerebellum and is involved in the control of posture, gait and neck movements. In contrast, lobules VI and VII of the posterior vermis receive a
Brainstem and CerebelJar Structures for Eye Movement Generation
17
major pontine input, they are classified as pontocerebellum and playa role in oculomotor functions [Yamada and Noda, 1987]. A similar division is also present in the FN. Neurons in the rostral FN are modulated during vestibular stimulation [Siebold et a1., 1997], but not with individual eye movements. Unilateral lesions cause a falling tendency to the ipsilateral side [Kurzan et a1., 1993; Thach et a1., 1992]. In contrast, neurons in the caudal FN are modulated with individual eye movements, either saccades [Helmchen et aI., 1994; Fuchs et a1., 1993] or smooth pursuit eye movements [Buttner et a1., 1991; Fuchs et a1., 1994J. Accordingly the caudal FN has been termed fastigial oculomotor region [Yamada and Noda, 1987]. Lesions in the caudal FN lead to step-size error dysmetria, with saccades to visual targets either too large (hypermetric) or too small (hypometric) [Buttner et a1., 1994; BOttner and Straube, 1995; Robinson et a1., 1993J. Smooth pursuit eye movements can have a reduced gain (cogwheel smooth pursuit) [Kurzan et aI., 1993; Robinson et aI., 1997]. In an earlier experimental study spontaneous nystagmus in the dark was observed after cooling of the FN [Vilis and Hore, 1981]. This, however, could not be confirmed in recent studies, and the involvement of adjacent structures (nodulus, uvula, pathways to the vestibular nuclei) must be considered as a cause. Lesions to the nodulus/uvula region are known to cause spontaneous nystagmus in the dark [Waespe et a1., 1995]. In clinical studies macrosaccadic oscillations have been related to disturbed visually guided saccades and deep cerebellar nuclei lesions [Selhorst et a1., 1976J. However, it is quite clear from clinical as well as experimental lesion studies in FN that severe saccadic dysmetria can occur without macrosaccadic oscillations [Buttner et a1., 1994; Robinson et a1., 1993J.
A erent and E erent ConnecUons The FN receives an ipsilateral input from the vermis. Lobules I-V project to the rostral FN and lobules VI-IX to the caudal FN [Noda et a1., 1990]. Also, the nodulus Oobule X) projects to the FN aside from its major projection to the vestibular nuclei [Wylie et aI., 1994]. Collaterals ofmossy fibers generally originate bilaterally from the brainstem. They derive from all vestibular nuclei (except the lateral vestibular nucleus), nucleus prepositus hypoglossi, dorsolateral and dorsomedial pontine nuclei and nucleus reticularis tegmenti pontis [Noda et a1., 1990J. A erents from climbing fibers terminate in the deep cerebellar nuclei [Van der Want et al., 1989J. Manye erent projections from the FN go to the same structures in the orainstem, fwm which the FN receive a erents, uften un the cuntralateral side (nucleus reticularis tegmenti pontis, dorsomedial and dorsolateral pontine nuclei and perihypoglossal nuclei). Projections to the vestibular nuclei are mainly contralateral, but also ipsilateral [Noda et a1., 1990J. Some deep cerebellar nuclei including FN neurons project to the cerebellar cortex [Batini et a1., 1989].
Horn/Buttner/Buttner-Ennever
18
Transmitters The FN, as all deep cerebellar nuclei, consists of heterogeneous groups of excitatory and inhibitory projection neurons and interneurons. The excitatory projection neurons are general1y larger and use glutamate as their transmitter, while the smal1er inhibitory neurons mostly use GABA as their transmitter. The GABAergic (inhibitory) neurons project mainly to the inferior olive [Fredette and Mugnaini, 1991], but some might also project to the cerebellar cortex [Batini et aI., 1989] and the pontine nuclei [Aas and Brodal, 1990J. Interneurons within the deep cerebellar nuclei probably use glycine as an inhibitory transmitter. It is well established that Purkinje cells projecting onto deep cerebellar nuclei neurons use GABA as a transmitter [Ito, 1984J. Individual Purkinje cells can innervate both inhibitory (GABA-ergic) and excitatory (non-GABAergic) neurons [De Zeeuw and Berrebi, 1995J. They also project to the (inhibitory) glycinergic interneurons [De Zeeuw and Berrebi, 1995]. The GABAinduced inhibition was thought to involve only GABA A and not GABA B receptors [Bmard et aI., 1993J. However, recently also the presence of GABA B receptors on terminals of Purkinje cells on deep cerebel1ar nuclei has been demonstrated in vitro [Mouginot and Gahwiler, 1996J. Other inputs to the deep cerebellar nuclei derive from collaterals of mossy fibres and climbing fibres. Both are excitatory and in most instances glutamate is used as a transmitter.
Outlook
The growing knowledge about the histochemistry and transmitters of the functional cell groups of the eye movement system leads not only to a better understanding of the function of the neuronal elements, but also opens the possibility to develop pharmacological treatments for eye movement disorders. There is a qUickly expanding literature on the chemical control of the cerebellum and brainstem, and it will take many more years of research to understand their interactions.
References Aas lE, Brodal P: GABA and glycine as putative transmitters in subcortical pathways to the pontine nuclei. A combined immunocytochemical and retrograde tracing study in the cat with some observations in the rat. Neuroscience 1990;34:149-162. Andressen C, BlUmcke I, Celio MR: Calcium-binding proteins: Selective markers of nerve cells. Cell Tissue Res 1993;271:181-208. Ashikawa H, Furuya N, Yabe T: E ecls of serotonin, GABA and glycine on the activity of pause neurons during vestibular nystagmus in the cat. Acta OtoLaryngol 1991; 111:999-1005.
Brainstem and Cerebellar Structures for Eye Movement Generation
19
Baimbridge KG, Celio MR. Rogers JH: Calcium-binding proteins in the nervous system. Trends Neurosci 1992; 15:303-308. Baker R, Highstein SM: Vestibular projections to medial rectus subdivision of oculomotor nucleus. J Neurophysiol 1978;41: 1629-1646. Barmack NH, Baughman R\N, Eckenstein FP: Cholinergic innervation of the cerebellum of rat, rabbit, cat, and monkey as revealed by choline acetyltransferase activity and immunohistochemistry. J Comp Neural 1992a;317:233-249. Barmack NH, Baughman RW, Eckenstein FP: Cholinergic innervation of the cerebellum of the rat by secondary vestibular a erents. Ann NY Acad Sci 1992b;656:566-579. Barmack NH, Baughman RW, Eckenstein FP, Shojaku H: Secondary vestibular cholinergic projection to the cerebellum of rabbit and rat as revealed by choline acetyltransferase immunohistochemistry, retrograde and orthograde tracers. J Comp Neurol 1992c;317:250-270. Batini C. Buisseret-Delmas C, Compoint C, Daniel H: The GABAergic neurones of the cerebellar nuclei in the rat: Projections to the cerebellar cortex. Neurosci Lett 1989;99:251-256. Batinl C, Compoint C, Buissert-Delmas C, Daniel H, Guegan M: Cerebellum nuclei and the nucleocortical projections in the rat: Retrograde tracing coupled to GABA and glutamate immunohistochemistry. J Comp Neurol 1992;315:74-84, Berry MM, Standring SM, Bannister LH: Cerebellum; in Gray's Anatomy (ed): Nervous System. London. Churchill Livingstone, 1995, pp 1027-1065. Bianchi R, Gioia M: Accessory oculomotor nuclei of man. 2. The interstitial nucleus of Cajal - A Nissl and Golgi study. Acta Anat 1991;142:357-365. Bianchi R. Gioia M: Fine structure of the interstitial nucleus of Cajal of the cat. J Anat 1995; 187: 141-150. Billard JM, Vigot R, Batini C: GABA, THIP and baclofen inhibition of Purkinje cells and cerebellar nuclei neurons. Neurosci Res 1993; 16:65-69. Blanks RHI: A erents to the cerebellar flocculus in cat with special reference to pathways conveying vestibular, visual (optokinetic) and oculomotor signals. J Neurocytol 1990;19:628--£42. Blanks RHI, Precht W, Torigoe Y: A erent projections to the cerebellar flocculus in the pigmented rat demonstrated by retrograde transport of horseradish peroxidase. Exp Brain Res 1983;52:293306. BUttner U, Buttner-Ennever JA: Present concepts of oculomotor organization, Rev Oculomot Res 1988; 2:3-32. BOttner U. BOttner-Ennever JA. Henn V: Vertical eye movement related activity in the rostral mesencephalic reticular rormation of the alert monkey. Brain Res 1977;130:239-252. BOttner U, Fuchs AF, Markert-Schwab G, Buckmaster P: Fastigial nucleus activity in the alert monkey during slow eye and head movements. J Neurophysiol 1991;65:1360-1371. Buttner U, Grundei T: Gaze-evoked nystagmus and smooth pursuit deficits: Their relationship studied in 52 patients. J Neurol 1995;242:384-389. BOttner U. Helmchen C, Buttner-Ennever JA: The localizing value of nystagmus in brainstem disorders, Neuroophthalmology 1995; 15:283-290. BOttner U, Straube A: The e ect of cerebellar midline lesions on eye movements. Neuraophthalmology 1995; 15:75-82. I3tittner U, Straube A, Spuler A: Saccadic dysmetria and 'intact' smooth pursuit eye movements after bilateral deep cerebellar nuclei lesions. J Neurol Neurosurg Psychiatry 1994;57:832-834. Btittner U, Waespe W: Purkinje cell activity in the primate flocculus during optokinetic stimulation, smooth pursuit eye movements and VOR-suppression. Exp Brain Res 1984;55:97-104. Btittner-Ennever JA: Patterns of connectivity in the vestibular nuclei. Ann NY Acad Sci 1992a;656: 363-378. Btittner-Ennever JA: Paramedian tract cell groups: A review of connectivity and oculomotor function; in Shimazu H, Shinoda Y (eds): Vestihular and Brain Stem Control of Eye, Head and Body Movements. Basel, Japan Scientific Societies Press/Karger. 1992b, pp 323-330. BOttner-Ennever JA, Akert K: Medial rectus subgroups of the oculomotor nucleus and their abducens internuclear input in the monkey. J Comp Neural 1981;197:17-27. BOttner-Ennever JA, Buttner U: Neuroanatomy of the oculomotor system. The reticular formation. Rev Oculomot Res 1988;2: 119-176.
Horn/Buttner/Buttner-Ennever
20
BUttner-Ennever lA, Cohen B, Horn AKE, Reisine H: Pretectal projections to the oculomotor complex of the monkey and their role in eye movements. 1 Comp Neurol 1996;366:348-359. BUttner-Ennever lA, Cohen B, Pause M, Fries W: Raphe nucleus of the pons containing omnipause neurons of the oculomotor system in the monkey, and its homologue in man. 1 Comp Neuro11988; 267:307-321. BUttner-Ennever lA, Horn AKE, Henn V: Di erential projections from rostval and caudal superior colliculus to the horizontal saccadic premotor and omnipause neurons in pons of the primate. Soc Neurosci Abstr 1997;23:1296. BUttner-Ennever JA, Horn AKE, Schmidtke K: Cell groups of the medial longitudinal fasciculus and paramedian tracts. Rev Neurol (Paris) 1989;145:533-539. BUttner-Ennever lA, Lang W: Vestibular projections to the monkey thalamus and rostal mesencephalon: An autoradiographic study; in Gualtierotti T (ed): The Vestibular System: Function and Morphology. Berlin, Springer, 1981, pp 130-143. Carpenter MB, Periera AB, Guha N: Immunocytochemistry of oculomotor a erents in the squirrel monkey (Saimiri-sciureus). 1 I-lirnforsch 1992;33:151-167. Cohen B, Komatsuzaki A: Eye movements induced by stimulation of the pontine reticular formation. Evidence for integration in oculomotor pathways. Exp Neurol 1972;36: 101-117. Crawford lD, Vilis T: Symmetry of oculomotor burst neuron coordinates about Listing's plane. 1 Neurophysiol 1992;68:432-448. Cuenod M, Do KQ, Grandes P, Morino P, Streit P: The puzzle of the transmitters in the climbing fibres. Exp Brain Res 1989; 17: 161-176. Dawson TM, Dawson VL, Synder SH: A novel neuronal messenger molecule in the brain: The free radical nitric oxide. Ann Neurol 1992;32:297-311. De la Cruz RR, Pastor AM, Martinez-Guijarro Fl, Lopez-Garcia C, Delgado-Garcia 1M: Role of GABA in the extraocular motor nuclei of the cat: A postembedding immunocytochemical study. Neuroscience 1992;51:911-929. De la Cruz RR, Pastor AM, Martinez Guijarro F1, Lopez-Garcia C, Delgado-Garcia 1M: Localization of parvalbumin, calretinin, and calbindin D-28k in identified extraocular motoneurons and internuclear neurons of the cat. 1 Comp Neurol 1998;390:377-391. De Lacalle S, Hersh LB, Saper CB: Cholinergic innervation of the human cerebellum. J Comp Neurol 1993;328:364-376. De Zeeuw CI, Berrebi AS: Postsynaptic targets of Purkinje cell terminals in the cerebellar and vestibular neuclei of the rat. Eur 1 Neurosci 1995;7:2322-2333. Delgado-Garcia 1M, Del Pozo F, Baker R: Behavior of neurons in the abducens nucleus of the alert cat. II. Internuclear neurons. Neuroscience 1986a; 17:953-973. Delgado-Garcia 1M, Del Pozo F, Baker R: Behavior of neurons in the abducens nucleus of the alert cat. I. Motoneurons. Neuroscience 1986b; 17:929-952. Dememes D, Raymond J: Radioautographic identification of glutamic acid labeled nerve endings in the cat oculomotor nucleus. Brain Res 1982:231:433-437. Everling S, Par~ M, Dorris MC, Munoz DP: Comparison of the discharge characteristics of brainstem omnipause neurons and superior colliculus fixation neurons in monkey: Implications for control of fixation and saccade behavior. J Neurophysiol 1998:79:511-528. Evinger C: Extraocular motor nuclei: Location, morphology and a erents. Rev Oculomot Res 1988;2:81-117. Fredette Bl, Mugnaini E: The GABAergic cerebello-olivary projection in the rat. Anat Embryol 1991; 184:225-243. Fuchs AF, Robinson FR, Straube A: Role of the caudal fastigial nucleus in saccade generation. 1. Neuronal discharge patterns. 1 Neurophysiol 1993;70: 1723-1740. Fuchs AF, Robinson FR, Straube A: Participation of the caudal fastigial nucleus in smooth-pursuit eye movements. 1. Neuronal activity. 1 Neurophysiol 19911;72:27111-2728. Fukushima K: The interstitial nucleus of Cajal and its role in the control of movements of head and eyes. Prog NeurobioI1987;29:107-192. Fukushima K, Fukushima 1, Ohashi T, Kase M: Possible downward burster-driving neurons related to the anterior semicircular canal in the region of the interstitial nucleus of Cajal in alert cats. Neurosci Res 1991;12:536-544.
Brainstem and Cerebellar Structures for Eye Movement Generation
21
Fukushima K, Kaneko CRS, Fuchs AF: The neuronal substrate of integration in the oculomotor system. Prog Neurobiol 1992;39:609-639. Fukushima K, Ohashi T, Fukushima J, Kaneko CRS: Discharge characteristics of vestibular and saccade neurons in the rostral midbrain of alert cats. J Neurophysiol 1995;73:2129-2143. Fukushima K, Terashima T, Kudo J, Inoue Y, Kato M: Projections of the group Y of the vestibular nuclei and the dentate and fastigial nuclei of the cerebellum to the interstitial nucleus of Cajal. Neurosci Res 1986;3:285-299. Furuya N, Yabe T, Ashikawa H, Koizumi T: Neurotransmitters regulating the activity of nystagmusrelated brain stem neurons in the cat; in Shimazu H, Shinoda Y (eds): Vestibular and Brain Stem Control of Eye, Head and Body Movements. Basel, Japan Scientific Societies Press/Karger, 1992, pp 167-176. Grantyn A, Ong Meang Jacques V, Berthoz A: Reticulo-spinal neurons participating in the control of synergic eye and head movements during orienting in the cat. II. Morphological properties as revealed by intra-axonal injections of horseradish peroxidase. Exp Brain Res 1987;66:355-377. Halmagyi GM, Rudge P, Gresty MA, Leigh Rj, Zee DS: Treatment of periodic alternating nystagmus. Ann Neurol 1980;8:609-611. Helmchen C, Rambold H, Buttner U: Saccade-related burst neurons with torsional and vertical ondirections in the interstitial nucleus of Cajal of the alert monkey. Exp Brain Res 1996;112:63-78. Helmchen C, Rambold H, Fuhry L, Buttner U: Deficits in vertical and torsional eye movements after uni- and bilateral muscimol inactivation of the interstitial nucleus of Cajal of the alert monkey. Exp Brain Res 1998;119:436-452. Helmchen, C, Straube A, Buttner U; Saccade-related activity in the fastigial oculomotor region of the macaque monkey during spontaneous eye movements in light and darkness. Exp Brain Res 1994; 98:474-482. Hepp K, Henn V: Spatio-temporal recording of rapid eye movement signals in the monkey paramedian pontine reticular formation (PPRF). Exp Brain Res 1983;52:105-120. Hepp K, Henn V, Vilis T, Cohen B: Brainstem regions related to saccade generation. Rev Oculomot Res 1989;3: 105-212. Holschner C: Nitric oxide, the enigmatic neuronal messenger: Its role in synaptic plasticity. Trends Neurosci 1997;20:298-303. Holstege G, Cowie RJ: Projections from the rostral mesencephalic reticular formation to the spinal cord. Exp Brain Res 1989;75:265-279. Horn AKE, Blittner-Ennever JA: Premotor neurons for vertical eye movements in the rostral mesencephalon of monkey and man: The histological identification by parvalbumin immunostaining. J Comp Neurol 1998;392:413-427. Horn AKE, Bottner-Ennever JA, BOttner U: Saccadic premotor neurons in the brainstem: Functional neuroanatomy and clinical implications. Neuroophthalmology 1996;16;229-240. Horn AKE, Buttner-Ennever JA, Suzuki Y, Henn V: Histological identification of premotor neurons for horizontal saccades in monkey and man by parvalbumin immunostaining. J Comp Neurol 1995; 359:350-363. Horn AKE, Blittner-Ennever JA, Wahle P, Reichenberger 1: Neurotransmitter profile of saccadic omnipause neurons in nucleus raphe interpositus. J Neurosci 1994; 14:2032-2046. Hokfelt T, Fuxe K: Cerebellar monoamine nerve terminals, a new type of a erent fibers to the cortex cerebelli. Exp Brain Res 1969;9:63-72. Ikeda M, Houtani T, Ueyama T, Sugimoto T: Choline acetyltransferase immunoreactivity in the cat cerebellLun. Neuroscience 1991 ;45:671-690. Ito M: The Cerebellum and Neural Control. New York, Raven Press, 1984. Iwamoto Y, Kitama T, Yoshida K: Vertical eye movement-related secondary vestibular neurons ascending in medial longitudinal fasciculus in cat. I. Firing properties and projection pathways. J Neurophysiol 1990;63:902-917. Ji Z, Aas JE, Laake J, Walberg F. Ottersen OP; An electron microscopic analysis of glutamate and glutamine in terminals of rat spinocerebellar fibers. J Comp NeuroI1991;307:1-15. Johnson JW, Ascher P: Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature 1987;325:529-531.
HornlBottner/Bottner-Ennever
22
Juncos JL, Hirsch EC, Malessa S, Duyckaerts C, Hersh LB, Agid Y: Mesencephalic cholinergic nuclei in progressive supranuclear palsy. Neurology 1991;41:25-30. Kaneko CRS, Evinger C, Fuchs AF: Role of cat pontine burst neurons in generation of saccadic eye movements. J Neurophysiol 1981;46:387-408. Kaneko CRS, Fukushima K: Burst driver neurons in the mesencephalon of the trained rhesus monkey. Soc Neurosci Abstr 1993; 19:859. Kerr CWH, Bishop GA: Topographical organization in the origin of serotoninergic projections to di erent regions of the cat cerebellar cortex. J Comp Neurol 1991 ;304:502-515. Kimura H, McGeer PL, Peng JH, McGeer EG: The central cholinergic system studied by choline acetyltransferase immunohistochemistry in the cat. J Comp Neurol 1981;200:151-201. King JS, Cummings SL, Cummings GA: Peptides in cerebellar circuits. Prog Neurobiol 1992;39:423-442. Kokkoroyannis T, Scudder CA, Balaban CD, Highstein SM, Moschovakis AK: Anatomy and physiology of the primate interstitial nucleus of Cajal. 1. E erent projections. J Neurophysiol 1996;75:725-739. Kurzan R, Straube A, Bmtner U: The e ect of muscimol microinjections into the fastigial nucleus on the optokinetic response and the vestibulo-ocular reflex in the alert monkey. Exp Brain Res 1993; 94:252-260. Lahjouji F, Bras H, Barbe A, Chazal G: GABAergic innervation of rat abducens motoneurons retrogradely labelled with HRP: Quantitative ultrastructural analysis of cell bodies and proximal dendrites. J Neurocytol 1995;24:29-44. Langer T, Fuchs AF, Scudder CA, Chubb MC: A erents to the flocculus of the cerebellum in the rhesus macaque as revealed by retrograde transport of horseradish peroxidase. J Comp Neurol 1985;235: 1-25. Leigh RJ, Zee OS: The Neurology of Eye Movements. Philadelphia, Davis, 1991. Li WB, Shiraishi Y, Nakao S: Topographical organization of cat mesodiencephalic areas for monosynaptic activation of vertical oculomotoneurons. Exp Brain Res 1993;94:43-52. Matsuo S, Hosogai P, Nakao S: Ascending projections of posterior canal-activated excitatory and inhibitory secondary vestibular neurons to the mesodiencephalon in cats. Exp Brain Res 1994;100:7-17. May PJ, Baker H, Vidal PP, Spencer RF, Baker R: Morphology and distribution of serotoninergic and oculomotor internuclear neurons in the cat midbrain. J Comp Neurol 1987;266:150-170. Money KE: Motion sickness. Physiol Rev 1970;50:1-39. Moschovakis AK, Scudder CA, Highstein SM: Structure of the primate oculomotor burst generator. I. Medium-lead burst neurons with upward on-directions. J Neurophysiol 1991a;65:203-217. Moschovakis AK, Scudder CA, Highstein SM: The microscopic anatomy and physiology of the mammalian saccadic system. Prog Neurobiol 1996;50:133. Moschovakis AK, Scudder CA, l-lighstein SM, Warren JD: Structure of the primate oculomotor burst generator. II. Medium-lead burst neurons with downward on-directions. J Neurophysiol 199Ib;65: 218-229. Mouginot 0, Gahwiler BH: Presynaptic GABA B-receptors modulate IPSPs evoked in neurons of deep cerebellar nuclei in vitro. J Neurophysiol 1996;75:894-901. Mugnaini E, Atluri RL, Houk JC: Fine structure of granular layer in turtle cerebellum with emphasis on large glomeruli. J Neurophysiol 1974;37: 1-29. Mugnaini E, Oertel W1 I: An atlas of the distribution of GAI3Aergic neurons and terminals in the rat CNS as revealed by GAD immunohistochemistry; in Bjorklund A, Hokfelt T (eds): Handbook of Chemical Neuroanatomy. Amsterdam, Elsevier, 1985, vol 4, pp 436-608. Munoz DP, Wurtz RH: Fixation cells in monkey superior colliculus. I. Characteristics of cell discharge. J Neurophysiol 1993;70:559-575. Nakao S, Shiraishi Y: Direct excitatory and inhibitory synaptic inputs from the mesodiencephalic jLUlction to motoneurons innervating extraocular oblique muscles in the cat. Exp Brain Res 1985;61:62-72. Nakao S, Shiraishi Y, Li WB, Oikawa T: Mono- and disynaptic excitatory inputs from the superiorcolliculus to vertical saccade-related neurons in the cat Fore!'s field H. Exp Brain Res 1990;82:222-226. Nguyen LT, Spencer RF: Neurotransmitters of abducens internuclear and ascending tract of Deiters inputs to the medial rectus motoneurons in the cat oculomotor nucleus. Soc Neurosci Abstr 1996; 22:263.8, p 665.
Brainstem and Cerebellar Structures for Eye Movement Generation
23
Noda H, Sugita S, Ikeda Y: A erent and e erent connections of the oculomotor region of the fastigial nucleus in the macaque monkey. 1 Comp Neurol 1990;302;330-348. Ohgaki T, Markham CH, Schneider 1S, Curthoys IS; Anatomical evidence of the projection of pontine omnipause neurons to midbrain regions controlling vertical eye movement. 1 Comp Neurol 1989; 289:610-625. Ottersen OP: Quantitative electron microscopic immunocytochemistry of neuroactive amino acids. Anat EmbryolI989;180:1-15. Ottersen OP: Neurotransmitters in the cerebellum (abstract). Rev Neurol 1993;149:629-636. Ottersen OP, Laake 1 H: Light and electron microscopic immunocytochemistry of putative neurotransmitter amino acids in the cerebellum with some observations on the distribution of glutamine; in Llinas R, Sotelo C (eds): The Cerebellum Revisited. Berlin, Springer, 1992, pp 116-134. Pare M, Guitton D: The fixation area of the cat superior colliculus: E ects of electrical stimulation and direct connection with brainstem omnipause neurons. Exp Brain Res 1994;101:109-122. Partsalis AM, Highstein SM, Moschovakls AK: Lesions of the posterior commissure disable the vertical neural integrator of the primate oculomotor system. 1 Neurophysio1 1994;71:2582-2585. Robinson FR, Phillips JO, Fuchs AF: Coordination of gaze shifts in primates: Bralnstem inputs to neck and extraocular motoneuron pools. J Comp Neurol 1994;346:43-62. Robinson FR, Straube A, Fuchs AF: Role of the caudal fastigial nucleus in saccade generation. 2. E ects of muscimol inactivation. 1 Neurophysiol 1993;70:1741-1758. Robinson FR, Straube A, Fuchs AF: Participation of caudal fastigial nucleus in smooth pursuit eye movements. II. E ects of muscimol inactivation. 1 Neurophysiol 1997;78:848-859. Sato Y, Kawasaki T, Ikarashi K: A erent projections from the brainstem to the three floccular zones in cats. II. Mossy fiber projections. Brain Res 1983;272:37-48. Schwindt PC, Precht W, Richter A: Monosynaptic excitatory and inhibitory pathways from medial midbrain nuclei to trochlear motoneurons. Exp Brain Res 1974;20:223-238. Scudder CA, Fuchs AF, Langer TP: Characteristics and fLUlctional identification of saccadic inhibitory burst neurons in the alert monkey. 1 Neurophysiol 1988;59:1430-1454. Scudder CA, Moschovakis AK, Karabelas AB, Highstein SM: Anatomy and physiology of saccadic longlead burst neurons recorded in the alert squirrel monkey. 11. Pontine neurons. 1 Neurophysiol 1996; 76:353-370. Selhorst JB, Stark L, Ochs AL, Hoyt WF: Disorders in cerebellar ocular motor control. II. Macrosaccaclic oscillation: An oculographic, control system and clinico-anatomical analysis. Brain 1976;99:509-522. Shook BL, Schlag-Rey M, Schlag 1: Direct projection from the supplementary eye field to the nucleus raphe interpositus. Exp Brain Res 1988;73:215-218. Siebold C, Glonti L, Glasauer S, Bottner U: Rostral fastigial nucleus activity in the alert monkey during three-dimensional passive head movements. 1 Neurophysio1 1997;77:1432-1446. Spencer RF, Baker R: Histochemical localization of acetylcholinesterase in relation to motor neurons and internuclear neurons of the cat abducens nucleus. J Neurocytol 1986; 15: 137-154. Spencer RF, Baker R: GABA and glycine as inhibitory neurotransmitters in the vestibuloocular reflex; in Cohen B, Tomko DL, Guedry F (eds): Sensing and Controlling Motion. New York, New York Academy of Science, 1992, pp 602-611. Spencer Rfr', Wang SP: Immunohistochemical localization of neurotransmitters utilized by neurons in the rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF) that project to the oculomotor and trochlear nuclei in the cat. 1 Comp Neurol 1996;366:134-148. Spencer RF, Wenthold RJ, Baker R: Evidence for glycine as an inhibitory neurotransmitter of vestibular, reticular, and prepositus hypoglossi neurons that project to the cat abducens nucleus. 1 Neurosci 1989;9:2718-2736. Stanton G B, Goldberg ME, Bruce C1: Frontal eye field e erents in the macaque monkey. II. Topography of terminal fields in midbrain and pons. 1 Comp Neurol 1988;271:~93-506. Storm-Mathisen JJ. Danbolt NC, Ottersen OP: Localization of glutamate and its membrane transport proteins; in Stone TW (ed): CNS Neurotransmitters and Neuromodulators. Glutamate. Boca Raton, CRC Press, 1995, pp 1-18. Strassman A, Evinger C, McCrea RA, Baker RG, Highstein SM: Anatomy and physiology ofintracellularly labelled omnipause neurons in the cat and squirrel monkey. Exp Brain Res 1987;67:436-440.
HornlBottner/Bottner-Ennever
24
Strassman A. Highstein SM, McCrea RA: Anatomy and physiology of saccadic burst neurons in the alert squirrel monkey. II. Inbibitory burst neurons. J Comp Neurol 1986a;249:358-380. Strassman A. Highstein SM, McCrea RA: Anatomy and physiology of saccadic burst neurons in the alert squirrel monkey. 1. Excitatory burst neurons. J Comp NeuroI1986b;249:337-357. Suzuki Y, BOttner-Eonever JA. Straumaon D, Hepp K. Hess BJM, Henn V: Deficits in torsional and vertical rapid eye movements and shift of Listing's plane after uni- and bilateral lesions of the rostral interstitial nucleus of the medial longitudinal fasciculus. Exp Brain Res 1995;106:215-232. Takagi M, Zee DS, Tamargo R: E ect of dorsal cerebellar verma 1 lesions on saccades and pursuit in monkey (abstract). Soc Neurosci Abstr 1996;22:1458. Thach WT, Goodkin HP, Keating JG: The cerebellum and the adaptive coordination of movement. Annu Rev Neurosci 1992; 15:403-442. Tohyama M, Takatsuji K: Atlas of Neuroactive Substances and Their Receptors in the Rat. Oxford, Oxford University Press, 1998. Vahedi K, Rivaud S, Amarenco P, Pierrotdeseilligny C: Horizontal eye movement disorders after posterior vermis infarctions. J Neurol Neurosurg Psychiatry 1995;58:91-94. Van der WantJJ, Wiklund L, Guegan M, Ruigrok T, Voogd J: Anterograde tracing of the rat olivocerebellar system with Phaseolus vulgaris leucoagglutinin (PHA-L). Demonstration of climbing fiber collateral innervation of the cerebellar nuclei. J Comp Neurol 1989;288:1-18. Vilis T, Hepp K, Schwarz U, Henn V: On the generation of vertical and torsional rapid eye movements in the monkey. Exp Brain Res 1989;77: 1-11. Vilis T, Hore J: Characteristics of saccadic dysmetria in monkeys during reversible lesions of medial cerebellar nuclei. J Neurophysiol 1981;46:828-838. Voogd J, Gerrits NM, Ruigrok TJH: Organization of the Vestibulocerebellum. New York, New York Academy of Science, 1996, pp 553-579. Waespe W, Cohen B, Raphan T: Dynamic modifiction of the vestibulo-ocular reflex by the nodulus and uvula. Science 1985;228: 199-202. Wang SF, Spencer RF: Spatial organization of premotor neurons related to vertical upward and downward saccadic eye movements in the rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF) in the cat. J Comp Neural 1996;366:163-180. Wentzel PR, Dezeeuw C[, Holstege JC, Gerrits NM: Colocalization of GABA and glycine in the rabbit oculomotor nucleus. Neurosci Lett 1993:164:25-29. Wentzel PR, Dezeeuw CI, Holstege JC, Gerrits NM: Inhibitory synaptic inputs to the oculomotor nucleus from vestibulo-ocular reflex-related nuclei in the rabbit. Neuroscience 1995;65:161-174. Wentzel PR, Gerrits NM, Dezeeuw CI: GABAergic and glycinergic inputs to the rabbit oculomotor nucleus with special emphasis on the medial rectus subdivision. Brain Res 1996;707:314-319. Wuenschell CW, Fischer RS, Kaufman DL, Tobin AJ: [n situ hybridisation to localize mRNA encoding the neurotransmitter synthezising enzyme glutamate decarboxylase in mouse cerebellum. Proc Nati Acad Sci USA 1986;83:6192-6197. Wylie DR, Dezeeuw CI, Digiorgi PL, Simpson JI: Projections of individual Purkinje ceUs of identified zones in the ventral nodulus to the vestibular and cerebeUar nuclei in the rabbit. J Comp Neurol [994;349:448-463. Yamada J, Noda I I: A erent and e erent connections of the oculomotor cerebellar vermis in the macaque monkey. J Comp Neurol 1987;265:224-241. Yingcharoen K, Rinvik E, Stormmathisen J, Ottersen OP: GABA, glycine, glutamate, aspartate and taurine in the peri hypoglossal nuclei: An immunocytochemical investigation in the cat with particular reference to the issue of amino acid colocalization. Exp Brain Res 1989;78:345-357. Zee DS, Yamasaki A, Butler P, GOcer G: E ects of ablation of flocculus and paraflocculus on eye movements in primates. J Neurophysiol 1981 ;46:878-899. Zhou \N, King vVM: Premotor commands encode monocular eye movements. Nature 1998;393:692-695.
AK.E. Horn, Institute of Anatomy, Ludwig Maximilians University, Pettenkoferstrasse 11, D-80336 Munich (Germany)
Brainstem and Cerebellar Structures for Eye Movement Generation
25
Blittner U. (ed): Vestibular Dysfunction and Its Therapy. Adv Otorhinolaryngol. Basel, Karger, 1999, vol 55, pp 26-81
Intrinsic Physiological and Pharmacological Properties of Central Vestibular Neurons Pjerre-Paul VMaJ", Njcolas Vjber(", Mauro Serafin b, Alexander BabaJjan a, Mjchel A1{jhlethalerb, Catherjne de Waele a a b
Neurobiologie des Reseaux Sensorimoteurs, CNRS UPRES-A 7060, Paris, France et Departement de Physiologie, Centre Medical Universitaire, Geneve, Suisse
Introduction
Oculomotor and postural stabilizing responses result from a complex multisensory integration, which can be defined as the process of matching multiple internal representations of an external event (head and/or body movement) obtained through di erent sensory modalities (visual, vestibular and proprioceptive) , into a unique, intrinsic frame of reference in which appropriate motor commands can be coded. It is therefore not surprising that such complex sensorimotor transformations are disturbed by aging, by pathological damage to the inner ear or eye muscles, by excessive natural stimulation and/or by exposure to conflicting sensory perceptions. Hence, to be accurate enough to provide a correct estimation of body position and self-motion during the whole life span, the stabilizing, oculomotor and postural responses must display a high degree of plasticity. We are therefore dealing here with a model involving the maintenance of a stable, internal representation of self-motion by the central nervous system (eNS) in a continually changing internal and external environment. From an experimental point of view, several characteristics of the motor synergies stabilizing gaze and posture make them suitable for neurophysiological studies: these reflexes have a well-defined goal, and require computation of the parameters underlying self-motion. Moreover, both the inputs (retinal slip of the visual world detected by the eyes, velocity of the head given by the vestibular system, etc....) and the static and dynamic motor responses to these inputs (eye and head movements, changes in skeletal geometry) are quantifiable with great precision IVibert et al., 1997J.
As a result, the neuronal operations underlying gaze and postural control have been intensely scrutinized for the past 30 years using electrophysiological recordings as well as morphological and electroanatomical methods [for reviews, see Baker et aI., 1981; Berthoz, 1989], which led to the elaboration of realistic models describing the underlying neuronal computations. Amongst other examples, it was shown that the central vestibular networks could integrate (in the mathematical sense of the term) a velocity signal into a position signal, and that they were segregated in frequency-tuned channels. In addition, the gains of vestibula-ocular and vestibulospinal pathways have been shown to change according to vigilance and following learning; the dea erented vestibular neurons are even able to recover a normal resting discharge in a few days following labyrinthine dea erentation [for reviews, see Berthoz, 1985; Smith and Curthoys, 1989; Cohen et al.. 1992; Kawato and Gomi, 1992; Barnes, 1993; Curthoys and Halmagyi, 1995; Dieringer, 1995; du Lac et a!., 1995J. All these remarkable features are extremelyinteresting in one vital respect:
these complex neuronal operations can be very precisely related to behavior. The neuronal computations underlying gaze and postural control will be, as all operations in the CNS, the by-product of both the emerging properties of the vestibular networks and the individual properties of their components, i.e. the neurons. This chapter deals with the individual properties of central vestibular neurons, which can be broadly segregated in two categories. First, in vitro studies have demonstrated that all vertebrates' neurons are endowed with specific nonlinear, intrinsic membrane properties, which confer to them complex integrative capabilities [Llinas, 1988]. Second, multiple pre- and postsynaptic receptors responding to various neurochemical compounds have been shown to influence the time course and the nature of the response of any given neuron to its main synaptic inputs. The membrane properties and neuropharmacological profile of neurons have been investigated both in vivo and in vitro using electrophysiological recordings. However, these methods have their limitations. Action potentials are all-or-none phenomena in a neuron, but they can be generated by a variety of di erent mechanisms. Furthermore, molecular biology has revealed the existence of various subclasses of the main neurotransmitter receptors expressed by vestibular neurons. These di erent subclasses can have markedly di erent distributions, and their activation often induces distinct electrophysiological or biochemical responses in di erent subsets of neurons. However, few specific ligands have yet been described for most of these receptors, which precludes the use of electrophysiological methods to study their involvement in vestibular-related computations. On the other hand, recent morphological methods such as in situ hybridization, which detects the expression of messenger RNAs leading to the synthesis of receptors, can be used in various behav-
Intrinsic Properties of Central Vestibular Neurons
27
ioral contexts. Therefore, to understand how the emerging properties of the vestibular network combine with the physiological and pharmacological properties of its constituent elements to stabilize gaze and posture requires the use of complementary neurobiological methods, in several types of in vivo and in vitro preparations. Needless to say, such investigations pave the way for new pharmacological treatments of vestibular syndromes, and could have even wider clinical applications since vestibular compensation has been shown to be a valuable model of postlesional plasticity in the CNS.
Physiological Properties of Central Vestibular Neurons
This part of the chapter presents our current knowledge of the individual membrane and discharge properties of vestibular nuclei neurons. Indeed, these properties deeply constrain how these neurons process the information they receive from their various a erences, which include visual, vestibular, and proprioceptive pathways, as well as cerebellar and cortical inputs [for reviews, see Highstein and McCrea, 1988; Berthoz, 1989; Schor et aI., 1992; Shinoda et aI., 1993J. In addition, these properties of central vestibular neurons can be strongly modulated by various neurotransmitters and neuromodulators, as described in the next part of the chapter.
Scientific Context
Most studies have focused on medial vestibular nucleus neurons (MVNn). As an example, we will mainly describe the results we obtained using intracellular recordings in guinea pig brainstem slices [Serfin et aI., 1991a, bJ and in the isolated whole brain [Babalian et aI., 1997J which can be compared with the results of the various studies made in the same species, in either acute or awake preparations. The functional anatomy [Curthoys et al.. 1975; Vidal et aI., 1986; de Waele et aI., 1989b; Graf et aI., 1995a, bJ and performance [Gresty, 1975; Pettorossi et aI., 1986; Mirenowicz and Hardy, 1992; Escudero et aI., 1993; Escudero and Vidal, 1996; Vi bert et aI., 1993] of the guinea pig's oculomotor system have been well described. Morphological and electrophysiological in vivo studies have unraveled the main structural and physiological characteristics of its central vestibula-ocular networks [Azzena et aI., 1976; Curthoys, 1982; Gstoettner and Burian, 1987; Smith and Curthoys, 1988; Yagi and Ueno, 1988; de Waele et aI., 1988; Marlinsky, 1992; Ris et aI., 1995; Murofushi et aI., 1996J.
VidaINibertiSerafin/Babaiian/Miihlethaler/de Waele
28
Finally, the intrinsic membrane properties of MVNs will be compared with those of two other classes of cells involved in postural and oculomotor control, the reticular neurons of the gigantocellular nucleus, and the prepositus hypoglossi nucleus neurons [Serafin et aI., 1996a, b].
MVN Neurons Physiological Properties: In vivo Recordings We tried to investigate the functional significance of the intrinsic membrane properties of vestibular neurons in two ways: by quantifying the discharge of these neurons in vivo, and by recording them in a preparation of isolated, in vitro whole brain to try to correlate the results of slice recordings with in vivo data. The activity of identified, second-order vestibular neurons was recorded at rest and during horizontal sinusoidal head rotations in the head-fixed, alert guinea pig. At rest, second-order MVNn formed a continuum between regularly and irregularly discharging cells [Serafin et aI., 1994: Ris et al., 1995], as already described in other species [Precht and Shimazu, 1965; Shimazu and Precht, 1965]. Their mean coe cient of variation (CV) amounted to 0.50, and their mean discharge rate to about 30 Hz. These two variables were highly correlated. During natural vestibular stimulation, second-order MVNn coded head velocity. Some of them were sensitive to the horizontal component of eye position, and half of these cells displayed some modification of their firing rate during the qUick phase of vestibular nystagmus. They decreased their discharges during quick phases oriented towards the side ipsilateral to their soma, and generated a burst of impulses during quick phases towards the contralateral side. Altogether, these firing characteristics are very similar to those of the horizontal, second-order vestibular neurons observed in cats and monkeys [Berthoz et aI., 1989; Scudder and Fuchs, 1992J. The regularity of the resting discharge of both first- and second-order vestibular neurons in vivo has been used as a convenient marker of their dynamic properties. Very approximately (since regular and irregular cells actually form a continuum), tonic, regular primary a erents tend to have a lower gain and a slower conduction velocity than the irregular, phasic cells [Goldberg and Fernandez, 1971; Yagi et al., 1977; Goldberg et al., 1984J. What are the functional implications of that segregation? One hypothesis states that the tonic first- and second-order vestibular neurons would mainly encode low-frequency head movements, and would be mostly involved in static postural and oculomotor control. The kinetic cells would in contrast encode high-frequency stimuli, and would playa major role in the stabilization of gaze and posture during fast, transient postural perturbations. According to
Intrinsic Properties of Central Vestibular Neurons
29
several authors, the whole vestibular network controlling gaze and posture would actually be organized in frequency-tuned channels [Baker et aI., 1981; Godaux et ai., 1983; Lisberger et ai., 1983]. Indeed, the inputs of regular and irregular, first -order vestibular neurons remain partly segregated at the level of second-order vestibular neurons [Goldberg et ai., 1987; Highstein et aI., 1987; Sato and Sasaki, 1993]. On the other hand, the oculomotor plant has widely di erent biomechanical properties from the other mobile segments of the body (neck, trunk, limbs). Therefore, it has also been proposed that the tonic MVNn would mostly control the oculomotor system, while the kinetic cells would be more involved in vestibulospinal synergies [Highstein et aI., 1987; Iwamoto et aI., 1990; Boyle et ai., 1992]. In alert guinea pigs, the irregular second-order MVNn that we have recorded did not have a higher sensitivity to vestibular stimulation than the regular ones. However, we have been unable to record these irregular cells at frequencies higher than 3 Hz, which is probably insu cient to reveal the di erent dynamic sensitivities of these two types of cells. An opposite hypothesis [Angelaki and Perachio, 1993] would be that the irregular a erents are mainly used to modulate VOR amplitude during constant velocity rotations. They would specifically input the velocity storage integrator included in the vestibula-ocular network, which improves the compensatory properties of the VOR at low frequency and in response to longduration, step changes in angular velocity.
MVN Neurons Intrinsic Membrane Properties: Slice Recordings Oi erent Types of MVN Neurons Intracellular recordings in guinea pig's brainstem slices [Serafin et ai., 1991a, b] led us to define two main classes of MVNn, according to their intrinsic membrane properties. Similar properties have been observed for rat and chick MVNn [Gallagher et aI., 1985; Outia et ai., 1995; du Lac and Lisberger, 1995]. It should be emphasized however that, as pointed out by du Lac and Lisberger [1995], this segregation in two subclasses is somewhat artificial, in the sense that the membrane properties of MVNn are actually distributed as a continuum in between the two stereotyped schemes corresponding to canonical A and B cells (see the intermediate types classified as type C neurons in Serafin et al. [1991a, b]). Type A MVNn (about 30% of recorded cells) are characterized by wide action potentials ( 1 ms at threshold) followed by a deep, single afterhyperpolarization (AHP). They also exhibit a transient rectification due to the activation
VidaINibertiSerafin/Babaiian/Miihlethaler/de Waele
30
of an lA-like, 4-aminopyridine-resistant conductance. Finally, type A MVNn display small, high-threshold calcium spikes potentiated by barium. Type B MVNn (about 50% of recorded cells) are in contrast characterized by thinner action potentials, and a double-component AHP including a first, fast and small component followed by a delayed and slower one. The overall amplitude of this AHP is lower than for type A MVNn. Type B MVNn also displays large, high-threshold calcium spikes and prolonged, calciumdependent plateau potentials, as well as a persistent, subthreshold sodium conductance. Finally, about one quarter of B MVNn, namely type B LTS MVNn, displayed low-threshold calcium spikes (LTS) that confer to them bursting properties. All types of MVNn display a spontaneous, regular discharge in slices (mean frequency of about 10 Hz) which persists when synaptic transmission is blocked, leading to the hypothesis that their resting activity may partly rely on pacemaker properties of the membrane of these cells [for review, see Darlington et aI., 1995].
Rhythmic Activity in Type B MVN Neurons In rare cases, type B MVNn neurons can spontaneously display an oscillatory discharge with regular bursts of two or more action potentials. On the other hand, this rhythmic activity could be elicited very easily by perfusing various pharmacological compounds in the bath. The frequency and duration of these oscillatory discharges were always voltage-dependent. The three following types of oscillations were observed: Addition of2.10 4 MN-methyl-D-aspartate (NMDA) in the bath induced a long-lasting oscillatory behavior in type B cells (but not in type B LTS MVNn) hyperpolarized (by 10-30 mY) from their resting membrane potentia] [Serafin et aI., 1992aJ. These membrane potential oscillations were tetrodotoxin (TTX)-resistant, and could be suppressed by APV, a specific NMDA antagonist, or by replacing sodium with choline. Perfusion ofthe slice with a low Ca 2 Ihigh Mg 2 -containing solution (which suppresses synaptic transmission) also triggered a rhythmic discharge [de Waele et aI., 1993] in slightly hyperpolarized type B MVNn (but not in the B LTS ones). This behavior was APV-resistant, but could be suppressed by TTX. M apamin, a selective blocker of one type of calcium-dependent potassium conductance (the SK channels), induced rhythmic burst firing on slightly hyperpolarized cells of the Band B LTS subtypes [de Waele et aI., 1993J. This oscillatory behavior was APV-resistant, but could be abolished by TTX or with ouabain, a specific antagonist of the active sodium pump.
Intrinsic Properties of Central Vestibular Neurons
31
Neuronal oscillations are by no mean restricted to vestibular neurons. NMDA-induced oscillations were observed in abducens motoneurons [Durand, 1993], in the spinal cord, the caudate nucleus, the cortex [for review, see Serafin et al., 1992a], and in cholinergic nucleus basalis neurons of the basal forebrain [Khateb et aI., 1995]. In vitro recorded thalamocortical cells can display either a tonic firing mode or an oscillatory discharge, and can switch from one to the other in response to various modulatory transmitters [for review, see McCormick, 1992].
MVN Neurons Intrinsic Membrane Properties: Recordings in the in vitro Whole Brain The overall viability of the vestibula-ocular network in the isolated whole brain preparation (IWB) has been precisely assessed in a recent study [Babalian et al., 1997]. The advantage of the IWB preparation is that, since the connectivity of the brain is preserved, stable intracellular recordings revealing the intrinsic membrane properties of neurons can be made in functionally identified groups of cells. Field potential recordings can be used in association with stereotaxic atlases [Gstoettner and Burian, 1987] to localize unambiguously small brain regions like the abducens nucleus ( 1 mm~. The abducens nucleus was then used as landmark to localize other neighboring structures such as the medial vestibular nucleus, the prepositus hypoglossi nucleus and to characterize the recordings obtained in the reticular formation. The four known types of medial vestibular nucleus neurons (types A, B, B LTS and C) were recorded in the isolated brain, with similar membrane properties than on slices (fig. 1). This means that this classification was not an artifact of the slice preparation. 80-85% of the MVNn recorded in the IWB could be identified as second-order cells, in which stimulation of the stump of the ipsilateral vestibular nerve evoked monosynaptic, excitatory postsynaptic potentials (EPSP). This proportion did not vary Significantly among the MVNn cell types. Stimulation of the contralateral vestibular nerve evoked di- or trisynaptic inhibitory postsynaptic potentials (IPSP) in about 75% of these second-order neurons, in accordance with previous in vivo studies [Shimazu and Precht, 1966; Precht et al., 1973]. The mean resting discharge of second-order MVNn amounted to about 10 Hz, similar to that found in slices or in anesthetized guinea pigs [Smith and Curthoys, 1988] but lower than the average 36 Hz obtained in the alert guinea pig [Ris et aI., 1995]. A crucial di erence was that second-order MVNn, which all discharged regularly on slice, displayed highly variable spontaneous activities in the IWB, like in vivo (fig. 2). Whereas type A cells had regular
Vidal/Vibert/SerafinlBabalian/Muhlethaler/de Waele
32
3
2
A
B
4
-l,- >~_) tlLUJ 16
-L,- 14- illlli ~ >
'[T~~ l
lliL1l1~, Iwm' lOOms
Fig J. Identification of three distinct cell types in the region of the MVN of isolated whole brains [reprinted with perrrtission from Babalian et aI., 19971. Intracellular recordings from individual type A (first row), type B (second row) and type B LTS (third row) vestibular neurons at resting potentials of 66, 64 and 58 mY, respectively. Column I: Individual spikes shOWing the single component AHP (arrow) of type A neurons and the double component AHP (single and double arrows) of type Band B LTS neurons. Column 2: Response of neurons to current pulses passed through the intracellular electrode. Arrowheads indicate the level of the resting membrane potential. The dot shows the slowing down of the repolarization of a type A neuron after a hyperpolarizing pulse, possibly due to an IA like rectifying current. Depolarizing pulses applied from a hyperpolarized level of membrane potential produce an LTS-like response with superimposed spikes (asterisk) in B LTS neurons, but ordinary spikes in B neurons. Column 3: Spontaneous discharge of the three neurons. Column 4: Monosynaptic activation of the neurons by stimulation of the ipsilateral vestibular nerve.
firing rates, type Band B LTS cells could present very irregular patterns of spontaneous activity. More precisely, while type B MVNn tended to be more irregular than type A neurons, they actually displayed a wide range of CVs (from 0.09 to 0.77), with the more regular type B MVNn being as regular as type A cells. Basically, the regularity of type B MVNn seems to depend on the amount of spontaneous synaptic activity reaching each neuron, whereas type A MVNn would be less sensitive to synaptic inputs. The type A and type B (including B LTS) MVNn identified in vitro in the IWB appeared
Intrinsic Properties of Central Vestibular Neurons
33
~ type A VNn
1O
•
CIl
c:
2:::l
8
'"c:
6
"-'
type B VNn
0
....
4
E :::l
2
15 c:
0
a
N.
"l. v.
0
0
0
"'0
"'. "". 0
0
<Xl. 0
"'.
A
-.
q
0
N.
"l. v.
CV
c: 0
.
•
0,8
c
":;l
'" "@ ....>0 i:
0,2
'"0 U
0,0
i:B
•
0,4
'"
"C
•
0,6
type A type B type B+LTS
B 15
1O
5
20
25
AHP (mV)
•
15 CIl
c:
2:::l
....'"c:
extracell ular
10
0
....
'"
.0
E :::l
c:
0
N
C')
V
Ll)
a a a a a
"'. "". 0
0
a:>.
0
"'. 0
e o.
~.
"!. "l. v.
CV
Fig 2. Characteristics of the spontaneous discharge of second-order VNn in the isolated whole brain [reprinted with permission from Babalian et aI., 19971. A Histogram showing the distribution of cae cients of variation for spontaneous discharges of type A (gray columns) and type B (black columns) cells" B Diagram displaying the relationship between AHP amplitudes (abscissa) and CVs (ordinate) for type A (open squares), type B (dots), ilnd type B LTS (fiJled squilres) neurons. The slope of the regression line is 0.018. e Distribution of CVs for spontaneous discharges of extracellularly-recorded units.
Vidal/Vibert/SerafinlBabalian/Muhlethaler/de Waele
34
therefore to correspond to the regular and irregular second-order vestibular neurons described in vivo. Note that in both cases, we are dealing with a continuum. In addition, stimulation of the ipsilateral eighth nerve could often trigger, in type B MVNn, long subthreshold plateau potentials on top of the monosynaptic evoked EPSP, sometimes following evoked action potentials. Similarly, the eighth nerve shock could induce low-threshold spikes with superimposed bursts in second-order, type B LTS neurons maintained slightly hyperpolarized. These nonlinear responses were obtained without any pharmacological manipulation, which supports the hypothesis that the membrane properties of MVNn described on slices could have a functional significance in the behaving animal.
Oscjllatory Behavior of MVN Neurons: Functional Speculations Up to now, there is no direct experimental evidence of a functional relevance of the MVNn oscillations. The simplest hypothesis would be that these oscillations are just a side e ect of the artificial MVNn exposure to exogenous, biologically-active compounds. For instance, apamine perfusion would be e ective because it completely blocks part of the Ca 2 -dependent K channels, which are indeed the target of various endogenous neuromodulators [for reviews, see Nicoll et ai., 1990; McCormick, 1992J. The neuronal oscillations observed in slices, the head oscillations induced by unilateral infusion of apamin into the vestibular complex of alert guinea pigs [de Waele et aI., 1993J would be the consequence of a nonphysiological blockade of these Ca 2 dependent conductances. The same reasoning could apply to NMDA-triggered oscillations, which could be a side-e ect of a nonphysiological, massive activation of all the NMDA receptors present on the MVNn membranes. Another pOSSibility would be that the MVNn oscillations result from an inherent instability of these cells, a side e ect due to the particular set of membrane conductances they have to express in order to properly encode head velocity in space. In that context, the periodic discharges recorded in vitro would be functionally irrelevant. Nevertheless, it could reveal a tendency of MVNn to oscillate in some unusual contexts, leading in vivo to pathological syndromes. Very prolonged exposures to conflicting sensory information [Collewijn, 1979; de Waele et aI., 1989a] led for instance to long-lasting oscillations of the rabbit and guinea pig oculomotor system, in the absence of any sensory activation. Similarly, the mal de debarquement syndrome, or some of the vertigoes of central origin, could be related to such putative, pathological oscillations of the MVNn in vivo.
Intrinsic Properties of Central Vestibular Neurons
35
During locomotion, vestibular neurons discharge rhythmically in phase with extensor limb activity in cat [Orlovsky, 1972J and guinea pig [Marlinsky, 1992J. Moreover, NMDA-induced oscillations in spinal motoneurons induce fictive swimming in the lamprey spinal cord preparation [Headley and Grillner, 1990; Hochman et al., 1994]. Therefore, the activation of the MVNn NMDA receptors during locomotion could help these neurons to sustain an oscillatory mode of activity. Finally, both NMDA receptors and neuronal oscillations such as the rhythm [Larson et aI., 1986; Steriade et aI., 1993] have been involved in CNS plasticity. Given the extremely developed adaptive properties of vestibuloocular and vestibulocollic synergies, MVNn oscillations might contribute to plastic modifications of the vestibular network. Some of the MVNn, playing a key role in these plastic processes, the so-called floccular target neurons, receive a powerful inhibitory drive from the cerebellum which could initiate the oscillatory behavior. More generally, several authors suggested that oscillations are used as a tool by the CNS to synchronize populations of neurons [for review, see Steriade et aI., 1993J. In invertebrates [for review, see Meyrand et aI., 1994], switching neurons to an oscillatory mode of firing allows to reconfigure complex neuronal networks according to the behavioral context.
MVN Neurons Intrinsic Membrane Properties: Functional Speculations We and others [Gallagher et aI., 1985; Dutia et aI., 1995; du Lac and Lisberger, 1995] have compared the physiological characteristics of MVNn in vivo and in vitro to search whether the heterogeneous, highly nonlinear membrane properties of these neurons may have a functional relevance. It must be stressed first that these membrane properties are not artifacts ensuing from the slicing procedure. Indeed, the same nonlinear responses could be recorded both in the IWB and on slices. This is quite reassuring in the sense that in the IWB, MVNn keep their normal connections; their axonal and dendritic trees are preserved. Moreover, LTS and plateau potentials were obtained without any pharmacological manipulation. Not surprisingly, our results confirm that the regularity ofMVNn is linked with their intrinsic membrane properties. In vivo, the regularity of the resting discharge has been used as a convenient marker of the dynamic properties of first- and second-order vestibular neurons: regular and irregular MVNn could correspond to tonic and phasic neurons, respectively. Therefore, the regular, tonic MVNn in vivo would correspond, in vitro, to the regular, type A MVNn. The irregular phasic MVNn in vivo would correspond in vitro to the irregular, type B MVNn. The membrane properties would then contribute to determine
Vidal/Vibert/SerafinlBabalian/Muhlethaler/de Waele
36
the dynamic properties ofMVNn. We have seen that the inputs of the regular and irregular, first -order vestibular neurons remained partly segregated at the level of second-order vestibular neurons, which would lead to the segregation of the vestibular networks in frequency-tuned channels. This segregation might therefore rely on the distinct membrane properties of the neurons involved in these di erent channels. In these schemes, MVNn having the same connectivity should altogether display a whole range of widely di erent membrane properties. In line with that hypothesis, the study of du Lac and Lisberger [1995J demonstrates that the cellular properties of MVNn indeed contribute to the processing of temporal information in VOR pathways. Our own model of intracellularly-recorded MVNn also indicates that their membrane properties are strong determinants of their dynamic properties [Av-Ron and Vidal, 1997J. On the other hand, our results in the guinea pig do not fit with the hypothesis that the regular and irregular vestibular neurons would correspond to the vestibula-ocular and vestibulospinal neurons, respectively [Highstein et aI., 1987; Iwamoto et a!., 1990; Boyle et aI., 1992J. In the IWB, a similar proportion of regular and irregular neurons were identified as vestibulospinal neurons by their antidromic activation from the cervical spinal cord [Babalian et aI., unpubl. observationsJ. The subthreshold plateau potentials evoked in the second-order (irregular) type B MVNn by orthodromic stimulation of the eighth nerve in the IWB raise the question of the capability of these neurons to 'store' information transmitted by the sensory vestibular a erents. For example, temporal summation of these long-lasting plateau potentials could be one of the neuronal mechanisms underlying the velocity storage system included in vestibuloocular networks [Raphan et a!., 1979J. This would fit well with the hypothesis of Angelaki and Perachio [1993] stating that irregular neurons would specifically input the velocity storage integrator. The plateau potentials of type B MVNn could also playa role in the transformation of the head velocity signal into a position signal by the so-called neuronal integrator [for reviews, see Fukushima et aI., 1992; Anastasio, 1994] localized in the prepositus hypoglossi nucleus (PHN) and the medial vestibular nucleus [Baker et a!., 1981; Fuchs, 1981; Godaux et aI., 1993J. In this respect. it is interesting to note that the type B neurons recorded in the PHN exhibit the same plateau potentials as type B MVNn [Serafin et aI., 1996b]. Finally, we would like to ask what could be the functional correlate(s) of the low-threshold spikes recorded in B LTS MVNn? We have proposed that these LTS could participate in vivo to the burst of discharge recorded in second-order MVNn during the fast phases of nystagmus [Serafin et aI., 1990]. However, irregular B LTS MVNn represented only 10-15% of all MVNn recorded in the IWB, whereas studies in the alert guinea pig have shown that
Intrinsic Properties of Central Vestibular Neurons
37
at least 50% of MVNn bursted during fast phases, whatever the regularity of their resting discharge. Hence, LTS was clearly not a prerequisite for this phasic activity to occur. The functional relevance of LTS remains therefore to be determined.
Membrane Properties of Other Types of Neurons Participating in Gaze Control. Towards a Unifying Scheme?
Assuming that the intrinsic membrane properties of the MVNn could help to specify the various dynamic properties of these neurons, and assuming that the existence of frequency-tuned channels are a valid hypothesis, then other types of neurons which participate in gaze and postural control should display the same type of various membrane properties as the MVNn. We have tried to answer that question by exploring the membrane properties of the gigantocellular reticular nucleus neurons (GCRNn) and of the prepositus hypoglossi nucleus neurons (PHN n). GCRNn have been mainly involved in locomotion, respiration and stabilization of gaze and posture [for reviews, see Vertes, 1979; Peterson, 1984; Grantyn and Berthoz, 1987]. They are input by visual, vestibular and somatosensory a erents, and are under both tectal and cortical control. GCRNn are organized in a somatotopic way, and project onto motoneurons at every level of the spinal cord [for reviews, see Peterson, 1979, 1984]. In guinea pig's brainstem slices [Serafin et aI., 1992b, 1996a], we have shown that they were endowed with intrinsic membrane properties which were strikingly similar to those of MVNn. GCRNn could indeed be subdivided in the same type A and type B cells, even if no B LTS neuron could be found in this nucleus. It is noteworthy, however, that pontine reticular neurons probably involved in gaze control [Vidal et aI., 1983; Grantyn and Berthoz, 1987] displayed LTS in vitro [Greene et aI., 1986]. Both regular and irregular GCRNn were found in vivo [Siegel, 1979; Vertes, 1979; Steriade et aI., 1984] and in the isolated whole brain [Serafin et al., 1992b]. As a first approximation, the regular neurons recorded in the IWB tend to be mostly type A cells, whereas the majority of type B neurons seem to have an irregular resting discharge. As mentioned above, the PHNn playa major role in oculomotor control: in close association with MVNn, they transform a velocity signal arising from the MVN into a position signal necessary to control the eye movements. We are currently investigating the intrinsic membrane properties of PHNn on guinea pig's brainstem slices [Serafin et aI., 1996b]. Again, these neurons are endowed with intrinsic membrane properties strikingly similar to those of MVNn and GCRNn. Type A, type B and type B LTS neurons could be
VidaINibertiSerafin/Babaiian/Miihlethaler/de Waele
38
found in the PHN. A fourth, distinct type of neurons displayed subthreshold oscillations and spontaneous clusters of spikes in addition to a strong, lA-like current. This last cell type is very similar to cells recorded within the basal forebrain, a structure of the diencephalon known to be strongly involved in cortical activation across the sleep-waking cycle [Khateb et at., 1995). Given the tight dependency of oculomotor behavior on the vigilance level [MelvillJones and Sugie, 19721, these cells might be involved in some of the interactions between the neuronal structures underlying gaze control and those setting the level of vigilance. Finally, both in vivo and in vitro intracellular recordings revealed that the membrane properties of abducens motoneurons, which display highly phasic firing patterns in vivo, were rather similar to those of type B MVNn [Durand, 1989; Babalian et al., 1997J. In summary, the description of intrinsic membrane properties of other types of brainstem neurons also involved in gaze control tend to support the hypothesis that: (a) these membrane properties contribute to shape the dynamics of responses of any given class of neurons; (b) they could participate in the segregation of neuronal networks controlling gaze and posture in frequency-tuned channels. Such concepts could also apply to the networks controlling respiration, locomotion, etc. However, a large part of these nonlinear membrane properties could also contribute to other aspects of neuronal processing, such as integration. Finally, the functional relevance of some of the characteristics of the recorded brainstem neurons still remain unknown.
Neurotransmitters and Neuromodulators of the Vestibular Neurons
We will subdivide the neurotransmitters acting on central vestibular neurons in three main groups. The excitatory and inhibitory amino acids, which include aspartate, glutamate, GABA and glycine, mediate fast synaptic events by acting mainly on postsynaptic, ionotropic receptors. The five monoamines (histamine, dopamine, serotonin, noradrenaline, and adrenaline) constitute a second category, together with acetylcholine. They have more di use and moderate e ects on vestibular neurons. Most of them activate only metabotropic receptors acting through second messenger systems, and have therefore much slower actions on the neuronal activity [Hille, 1992). Finally, several neuroactive peptides have been shown to be e cient on vestibular nuclei neurons. A brief technical comment: both in case of in vivo microiontophoretic applications, or of in vitro bath application on slice, the e ects of a tested
Intrinsic Properties of Central Vestibular Neurons
39
compound on any given neuron normally represent the summation of the direct e ect of the drug on the recorded cell combined with the action of the drug on the inhibitory and excitatory (inter)neurons contacting the neuron under scrutiny. On slice and in the IWB, synaptic transmission can be blocked by perfusing a high Mg 2 low Ca 2 solution, or adding TTX to the bath. It is therefore possible to isolate the e ect of the drug on the recorded neuron, and to record it independently of the rest of the network.
The Excitatory and Inhibitory Amino Acids in Vestibular Networks The excitatory and inhibitory amino acids (EAA and IAA) include aspartate and glutamate on one side, and GABA and glycine on the other side. Glutamatergic receptors can be subdivided into ionotropic and metabotropic receptors, named after their main specific agonists [for review, see Nakanishi, 1992]. The ionotropic receptors include the AMPA/kainate and NMOA receptors, while the eight distinct metabotropic receptors sensitive to trans-ACPO can be divided into three main groups [Pin and Ouvoisin, 1995).
Excitatory Amino Acid (EAA) Receptors of the Vestibular Nuclei Neurons Anatomical studies have revealed the presence of all types of EAA receptors in the vestibular nuclei [for reviews, see Raymond et aJ., 1988; de Waele et aI., 1995; Vidal et aI., 1996). including metabotropic receptors of the mGluRl, mGluR2, mGluR5 and mGluR7 subtypes [Shigemoto et aI., 1992; Ohishi et aI., 1995; Neki et aI., 1996]. In situ hybridization techniques have also revealed some of the subunits which compose ionotropic EAA receptors in the vestibular nuclei: high densities of the four subunits of the AMPA receptors (GluRl, GluR2/3, GluR4), and of the Rl and R2C subunits of the NMOA receptors were detected, whereas the R2B and R20 subunits were expressed at lower levels [Petralia and Wenthold, 1992; de Waele et aI., 1994; Watanabe et aI., 1994]. These anatomical data fit with numerous in vitro, electrophysiological studies which demonstrated that vestibular neurons are responsive to both agonists and antagonists of the AMPAIkainate, NMOA and trans-ACPO receptors [for reviews, see Gallagher et aI., 1992; Smith et aI., 1992; de Waele et aI., 1995; Vidal et aI., 1996). Moreover, most MVNn are depolarized by AMPA, kainate, NMOA and trans-ACPO [Vibert et aI., 1992, 1994), whatever their intrinsic membrane properties (table 1). Given the persistence of these
VidaINibertiSerafin/Babaiian/Miihlethaler/de Waele
40
Table 1. E ects of excitatory amino acids on MVNn recorded in slices: the nature and number of e ects obtained with six agonists of the excitatory amino acid receptors on the various parameters characterizing intracellularly-recorded MVNn are given for type A. type B and type B LTS neurons
Experimen tal conditions
Control
potential and discharge
Type A neurons
20
Type B neurons
20
Type B LTS neurons
TTX
NMOA
AMPA
121
(95%)
120
resistance
13
114
potential and discharge 22
124
(93%)
(92%)
8
31
18
135
Control
17
120
180
(88%)
(85%)
(56%)
14
26
8
117
126
(82%)
(100%)
(25%)
10
4
13
4 15
8
6
(80%)
(100%)
110
(100%)
12
14
(100%)
112
114
(93%)
31
(100%)
20
131
13
(100%) /14
120
14
18
(19%)
113
(100%)
250/31
150/18
(81%)
(83%)
16
/]4
116
(100%)
(100%)
(100%)
Kainate
Quisqualate
Glutamate
potential and discharge 6
Type B neurons
8
Increase;
18
resistance
(89%)
Type A neurons
Type B LTS neurons
7
potential and discharge
(100%)
14
Experimen tal conditions
resistance
(100%)
(100%)
Synaptic uncoupling
Trans-ACPO
resistance
16
6
(100%)
19
(89%) 3
resistance
16
14 115 (93%)
13
5
15
(100%) 3
13
(100%)
potential and discharge
12
113
(92%)
5
16
4 18 (50%)
(83%)
I
II
(100%)
no e ect.
Intrinsic Properties of Central Vestibular Neurons
resistance
4 15 (80%)
(100%)
(100%)
decrease; 0
potential and discharge
41
responses both during bath application of TTX and in a high Mg 2 flow Ca 2 -containing solution, all these responses are at least partly mediated by postsynaptic receptors [see also Darlington and Smith, 1995J. As mentioned above, long-term NMDA perfusion can generally induce a long-lasting oscillatory behavior in type B MVNn [Serafin et aI., 1992a]. There is now a general agreement that an excitatory amino acid like glutamate and/or aspartate mediates synaptic transmission between first- and second-order vestibular neurons [for reviews, see Raymond et aI., 1988; Gallagher et aI., 1992; de Waele et aI., 1995; Yamanaka et aI., 1997J. Also, at least in the frog, several groups of a erents to the vestibular nuclei such as the proprioceptive fibers originating in the spinal cord and the excitatory commissural pathways linking together the two vestibular complexes use glutamate and/or aspartate as transmitter [Cochran et aI., 1987; Dieringer, 1995J. Finally, the excitatory second-order vestibular neurons which input the contralateral abducens motoneurons and some of the spinal motoneurons involved in postural stabilization are also believed to release excitatory amino acids [for reviews, see Raymond et aI., 1988; de Waele et aI., 1995; Dieringer, 1995J.
Pharmacological Analysis of EAA-Mediated Synaptic Transmission in the MVN The contribution of NMDA receptors to the synaptic transmission between first- and second-order vestibular neurons [for reviews, see de Waele et aI., 1995; Dieringer, 1995; Vidal et aI., 1996J has been a matter of debate. It was first believed that the monosynaptic input of vestibular a erents was only mediated through non-NMDA receptors [Cochran et aI., 1987; Lewis et aI., 1989; Doi et aI., 1990J. Subsequent slice studies indicated that it might not be the case [Kinney et aI., 1994; Takahashi et aI., 1994; Straka et aI., 1995bJ. In slices, however, isolated electrical stimulation of the first-order vestibular neurons has been di cult to obtain. We therefore recorded the field potentials evoked in the MVN, and the monosynaptic EPSPs evoked in second-order MVNn following single-shock stimulation of the vestibular nerve in the IWB of guinea pig [Babalian et aI., 1997J. Our data confirmed that this transmission was mediated by EAA. Perfusion of CNQX (an antagonist of AMPA/kainate receptors) suppressed a major part of the field potentials and EPSPs evoked following the stimulation of the stump of the eighth nerve. In about 50% of the cases, perfusion of APV (an antagonist of NMDA receptors) subsequently abolished a small and variable part of field potentials or EPSPs which persisted following CNQX perfusion. The results obtained with APV demonstrated that NMDA receptors
Vidal/Vibert/SerafinlBabalian/Muhlethaler/de Waele
42
contributed to the transmission between first- and second-order MVNn in mammals. However, this was the case in only 50% of the recorded neurons, and this NMDA contribution was highly variable. Such conclusions were in agreement with a previous study by Straka et a!. [1995bJ in the isolated frog's brainstem, which demonstrated in addition that only the thickest vestibular a erents (presumably the kinetic neurons) would activate NMDA receptors on second-order vestibular neurons. In both studies, it was clear that NMDA receptors were involved in the direct, monosynaptic transmission linking the sensory a erents to the second-order vestibular neurons. The low amplitude of the CNQX-insensitive, APV-sensitive component of the EPSP evoked by stimulation of the eighth nerve does not mean that NMDA-mediated transmission is of minor importance in vivo. NMDA receptors are subjected to a voltage-dependent block by extracellular Mg 2 [Ascher and Nowak, 1988], and a quantitative assessment of the NMDA-mediated transmission in a totally dea erented brain was therefore irrelevant. It should also be noted that we found a CNQX- and APV-resistant component in the transmission between sensory vestibular a erents and second-order MVNn. This component could result from the activation of the postsynaptic, metabotropic glutamate receptors present on most MVNn, or from the involvement of another transmitter like acetylcholine. Finally, it is noteworthy that in the frog, disynaptic IPSPs seem to be often superimposed upon the monosynaptic EPSPs elicited in vestibular neurons by stimulation of the ipsilateral vestibular nerve [Straka and Dieringer, 1996J. Presynaptic EAA receptors could also regulate glutamate release by the glutamatergic a erents reaching vestibular neurons. Presynaptic NMDA and trans-ACPD receptors are present on the terminal arborization ofaxons in the vestibular nucleus [Gallagher et a!., 1992; Kinney et a!., 1993J. In particular, they may be localized on the synaptic endings of the first-order neurons since several NMDA and trans-ACPD receptor subunits are expressed by vestibular ganglion cells [Doi et a!., 1995; Safieddine and Wenthold, 1997].
Functional Roles of NMDA Receptors in the Vestibular Nuclei: Correlation with in vivo Data Chronic, unilateral perfusion of APV in the vestibular complex of aler guinea pigs induced a massive postural and oculomotor syndrome, similar to! the one observed following ipsilateral, acute vestibular dea erentation [de Waele et a!., 1990J. Perfusion ofCNQX, in contrast, failed to induce any static postural syndrome or eye deviation. This experiment demonstrated that (a) the cation channels associated with NMDA receptors were open in the alert animal;
Intrinsic Properties of Central Vestibular Neurons
43
(b) they were absolutely essential to maintain a normal resting discharge ofneurons in the vestibular nuclei, which was not the case for AMPA/kainate receptors. This result was quite interesting, because it linked in an intimate way the static control of posture to the NMDA subtype ofEAA receptor. In addition, the long duration of the NMDA-mediated EPSPs might facilitate the summation of synaptic potentials [Mettens et aI., 1994b], and NMDA receptors could therefore be involved in the integration of the velocity signal encoded by the first-order vestibular neurons into a position signal necessary to stabilize the eyes: in the alert cat indeed, APV perfusion in the central part of the medial vestibular nucleus induced important gaze-holding failures [Mettens et aI., 1994a]. The absence of the voltage-dependent block of the NMDA receptors of the central vestibular neurons by Mg 2 in alert animals is explicable. First, many of their a erent fibers have high, spontaneous firing rates. For instance, first-order neurons have a mean resting discharge of about 30-40 spikes/so This presumably maintains the membrane potential of vestibular neurons at a su ciently depolarized level to prevent the Mg 2 block. Second, glYCine, an inhibitory transmitter acting on strychnine-sensitive receptors [for review, see Betz et aI., 1994], is also a coagonist of glutamate at a strychnine-insensitive site of the NMDA receptors [for review, see Wood, 1995]. It turns out that in frog and rat, glycine is colocalized with glutamate in the largest, sensory vestibular neurons [Reichenberger and Dieringer, 1994], precisely those which trigger in vitro NMDA-mediated responses in vestibular neurons [Straka et aI., 1995a]. Hence, corelease of glutamate and glycine by these fibers could potentiate postsynaptic I\lMDA receptors, and contribute to decrease the Mg 2 block. The functional relevance of the glycinergic modulation of NMDA receptors is unclear. It was even stated that it might not operate in normal conditions, since it has been suggested that the strychnine-insensitive site may be saturated in vivo [Wood, 1995]. In order to investigate that problem, we have checked on slices that MVNn NMDA receptors were sensitive to bath application of a specific agonist (D-serine) and antagonist (7 -chlorokynurenate) of the strychnine-insensitive binding site of NMDA receptors [Lapeyre and Benazet, unpub!. results]. Both compounds were then chronically perfused unilaterally in the vestibular complex of alert, unrestrained guinea pigs [Benazet et aI., 1993]. D-Serine induced an asymmetry of the HVOR, and a reversible postural syndrome consistent with a hyperactivity of vestibular neurons on the perfused side. In contrast, 7-chlorokynurenate induced the same syndromes in the opposite direction, revealing a hypoactivity of vestibular neurons. These results demonstrated that, in vivo, the vestibular NMDA receptors could be modulated through their glycinergic site. The extent to which this may be of functional importance remains to be clarified.
Vidal/Vibert/SerafinlBabalian/Muhlethaler/de Waele
44
EAA Receptors and Vestibular Plasticity NMDA receptors have been shown to play a key role in di erent types of synaptic plasticity [for review, see Nakanishi, 1992], which raised the question of their involvement in vestibular plasticity. Postlesional Plasticity Vestibular compensation after unilateral labyrinthectomy is considered as an excellent model of plasticity of the adult CNS [Llinas and Walton, 1979J. Indeed, the postural and oculomotor syndromes observed at the acute stage largely disappear over time in all species of vertebrates studied [for review, see Schaefer and Meyer, 1974). The static syndromes observed at the acute stage result from an asymmetry of the resting discharges between the intact and dea erented vestibular nuclei: immediately following the lesion, the dea erented MVNn are silent, whereas the spontaneous activity of the contralateral medial vestibular neurons is increased [for review, see Smith and Curthoys, 1989J. At the compensated stage, the dea erented vestibular neurons recover a quasinormal resting activity [for review, see Ris et a!., 1995). As a result a new, symmetrical pattern of resting activity is restored between the intact and the dea erented vestibular complexes; the emergence of this new balance plays an essential role in the compensation process. Since we have shown that NMDA receptors were essential for the maintenance of the resting discharge of central vestibular neurons, we suggested that they could be strongly involved in the recovery of resting discharge after lesion. This was tested by comparing the e ects of APV perfusion in the vestibular complex of normal and compensated, unilaterally labyrinthectomized guinea pigs [de Waele et a!., 1990J. APV induced similar postural and oculomotor syndromes in intact and lesioned animals, which indicated that, indeed, a denervation supersensitivity of the vestibular NMDA receptors present on the dea erented vestibular neurons could contribute to vestibular compensation. To check that hypothesis, the distribution of mRNAs coding for the NMDA R1 subunit in the two vestibular nuclei were investigated in both intact and compensated rats [de Waele et a!., 1994). On both sides of the brain, the mean density of mRNA coding for the NMDA R1 subunit in the MVN decreased by 20% just after the lesion. Three days later, the intensity of labeling was back to normal. Hence, unilateral labyrinthectomy induced a transient decrease of the NMDA receptor synthesis in both vestibular nuclei, which disappeared during compensation. This result is compatible with the postulated denervation supersensitivity of NMDA receptors. Other studies support that hypothesis [for reviews, see Smith et a!., 1992; de Waele et al., 1995; Vidal et al.,
Intrinsic Properties of Central Vestibular Neurons
45
1996]. Systemic injections of MK-801, another NMDA antagonist, impaired compensation in the guinea pig [Smith and Darlington, 1988; Pettorossi et aI., 1992; Kitahara et aI., 1995]. However, studies in the frog did not find any evidence for an NMDA supersensitivity [Knopfel and Dieringer, 1988; Dieringer, 1995]. More studies are needed to ascertain whether modifications of the NMDA receptors occurring during vestibular compensation are causally related to the behavioral recovery.
Functional Plasticity Habituation and adaptation of the vestibulo-ocular and vestibulospinal reflexes are well-known illustrations of the functional plasticity of the central vestibular system. They are believed to rely on long-term modifications of synaptic strengths occurring at di erent levels of vestibulo-ocular and/or vestibulospinal pathways [for reviews, see Kawato and Gomi, 1992; Cohen et aI., 1992; du Lac et aI., 1995]. Based on extracellular field recordings, the occurrence of such long-term synaptic modulations, and the concomitant involvement of the NMDA receptors appeared likely at the level of the vestibular nuclei [Racine et aI., 1986; Capocchi et aI., 1992; Grassi et aI., 1995]. However, our intracellular recordings of MVNn, either in vivo or in the IWB, have failed up to now to detect long-term potentiation and/or long-term depression phenomena at the level of vestibular neurons.
Inhibitory Amino Acid (IAA) Receptors of the Vestibular Nuclei Neurons
Anatomical Studies GABA and glycine are the main inhibitory transmitters in the CNS [Sivilotti and Nistri, 1991; Sato et aI., 1991]. Vertebrates' GABA receptors can be of two types: The ionotropic GABA A receptors include chloride ion channels [for review, see Kaila, 1994]. The GABA s metabotropic receptors activate second messenger systems [for review, see Misgeld et aI., 1995]. Glycine receptors are ionotropic receptors quite similar to the GABA A ones [for review, see Betz et aI., 1994]. Anatomical studies have revealed a dense innervation of all vestibular nuclei by GABAergic and glycinergic a erent fibers [for reviews, see Raymond et aI., 1988; de Waele et al.. 1995; Rampon et aI., 1996; Reichenberger et aI., 1997]. In situ hybridization and immunocytochemical techniques have demonstrated that vestibular neurons were endowed with GABAA , both pre- and postsynaptic GABA s receptors [Holstein et aI., 1992], and glycinergic receptors. 30% of MVNn were shown to be GABAergic neurons expressing glutamate decarboxylase (GAD), the specific enzyme for GABA synthesis [de Waele
Vidal/Vibert/SerafinlBabalian/Muhlethaler/de Waele
46
et a1., 1994; Holstein et a1., 1996]. These cells would correspond to the inhibitory intemeurons previously described in the MVN [Shimazu and Precht, 1966; Nakao et a1., 1982] and to the inhibitory, second-order MVNn projecting to extraocular andfor spinal motoneurons [for reviews, see de Waele et aI., 1995; Graf et aI., 1997]. Electrophysiological Studies In vitro studies confirmed the importance of inhibition in the processing of egomotion information in the vestibular nuclei. In slices, extracellularlyrecorded MVNn were inhibited by GABA through both GABA A and GABA B receptors [Smith et a1., 1991; Dutia et aI., 1992]. Intracellular recordings [Vibert et aI., 1995a, c] in a high Mg 2 flow Ca 2 solution, or in presence of TTX, confirmed that all types of MVNn were directly hyperpolarized and inhibited (table 2) by GABA, muscimol (a specific GABA A agonist), and baclofen (a specific GABA B agonist). However, in normal medium, while many MVNn were still hyperpolarized by bath application of GABA and muscimol, others could be depolarized. MVNn which were depolarized in control conditions were always hyperpolarized by muscimol when TTX was added in the bath. These results indicated that (a) local inhibitory intemeurons were spontaneously active in the slice and often exerted a tonic inhibition on the recorded neurons; (b) bath application of GABA and muscimol inhibited these interneurons, interrupting the tonic inhibition they exerted on the recorded cell and often leading to its disinhibition in normal medium; (c) this disinhibition could supersede the direct inhibition induced by bath application of GABA and muscimol on the recorded MVNn, which provoked the apparent depolarizing e ects observed with these compounds. Ultimately, this suggests that both the recorded MVNn and the local inhibitory interneurons present in the vestibular nuclei are endowed with postsynaptic, GABA A receptors. This hypothesis is strengthened by a recent morphological study [de Zeeuw and Berrebi, 1996], which demonstrated that indeed, individual Purkinje cell axons (which use GABA as their main transmitter) terminate on both inhibitory and excitatory neurons in the vestibular nuclei. Bath application of glycine produced a dose-dependent decrease in the MVNn resting discharge [Lapeyre and de Waele, 1995]. This inhibitory action was suppressed by strychnine, and persisted in a high Mg 2 flow Ca 2 -containing solution, which indicated that MVNn display postsynaptic, strychninesensitive glycinergic receptors. Therefore, glycine can modulate the MVNn at two levels: it might potentiate the depolarizing action of glutamate through the glycinergic, strychnine-insensitive modulatory site of NMDA receptors, but can also have a hyperpolarizing e ect by acting on their strychnine-sensitive receptors.
Intrinsic Properties of Central Vestibular Neurons
47
Table 2. E ects of inhibitory amino acids on MVNn recorded in slices: the nature and number of e ects obtained with three agonists and one antagonist of the GABA receptors on the various parameters characterizing intracellularly-recorded MVNn are given for type A, type B and type B LTS neurons (bicuculline is a specific antagonist of GABA A receptors)
-.
'"~
(/)
35 30
•
control
o
vestibular exercises
'i-I--"r,"
25 20 15
10
-£-----------,.-~'-"-'-2n~·20
5 0 0
10
15
20
25
30
time (day after symptom onset) JOC Fig JOe. Time course of the changes in total sway path values (SP) for the control and physiotherapy groups: vestibular exercises improved central vestibulospinal compensation. For postural control [39J we measured the SP values (mlmin, mean SD) of patients with eyes closed and standing on a compliant foam-padded posturography platform. The total SP is the length of the path described by the center of force during a given time (20 s), which is generated by the inherent instability of a subject standing on a recording platform. SP is approximated by the sum of the distances between two consecutive sampling points in the anteroposterior (sagittal x) plane, i.e., sagittal sway (calculated as x ), mediolateral (frontal y) plane, i.e., frontal sway (calculated as ( y ), or for two dimensions as - the y2 )). There was a significant di erence (ANOVA, total SP - (calculated as ( ( x 2 p 0.00l) between the two groups at the statistical end point (day 30 after symptom onset). The dotted line indicates the normal range. a During the first days after symptom onset not all patients could stand long enough ( 20 s) on the platform to permit accurate quantitative measurement of the SP values [from 78[.
Animal experiments have shown that visual and physical exercises promote central compensation of spontaneous nystagmus [14] as well as postural reflexes in locomotion [42, 49J. A recent prospective clinical study has also shown that vestibular exercises (table 1) improve central vestibulospinal compensation in humans (fig. 10). Central compensation of unilateral peripheral vestibular lesions involves multiple processes occurring in distributed networks at ui erent lucatiuns (spinal coru, vestibular nuclei, commissural brainstem connections between vestibular nuclei) and with di erent time courses. For example, after hemilabyrinthectomy in the frog, 50% of the postural recovery is accomplished within the first 2 weeks. At that time the commissural vestibular changes have reached only 5% of maximum, which
Vestibular Neuritis
131
Table 1. Physical therapy for acute, unilateral labyrinthine lesions Clinical stage
Physical exercise
Strategy
No exercise; bed rest Head immobilization
Prevent falls Avoid active head accelerations leading to 'cross-coupled' e ects Avoid visual-vestibular mismatch
1. Approximately days 1-3
Nausea Spontaneous nystagmus with fixation
Eyes closed II. Approximately days 3-5
No spontaneous nausea Incomplete suppression of spontaneous nystagmus by fixation straight ahead
III. Approximately days 5-7 Suppression of spontaneous nystagmus with fixation gaze nystagmus in the straight ahead, but continued direction of fast phase, and spontaneous nystagmus with Frenzel's glasses
IV Approximately weeks 2-3 No spontaneous vertigo Weak spontaneous nystagmus with Frenzel's glasses
Strupp/Brandt
Exercise in bed (supine and sitting) with rapid mobilization I) Fixation straight ahead; voluntary saccades and eccentric gaze-holding (l0, 20, and 40° horizontaVvertical) ; reading exercise Smooth pursuit (finger movements or pendulum 20-40°; 20-60 0 /s) Active head oscillations with fixation of a stationary target at 1 m distance (0.5-2 Hz; 20-30°; yaw pitch roll) 2) First balance exercise-free sitting and stance and gUided gait (eyes open, eyes closed)
Visual control of stabilization of gaze in space by suppressing spontaneous nystagmus through voluntary fixation impulse (retinal slip) Visually guided control of target fixation Provoke vestibular stimuli for recalibration of VOR under visual control of retinal slip of the viewed target Circulatory training, prophylaxis of thrombosis
1) Static stabilization: Four point stance. Stance on one knee and one foot Upright stance (eyes open/eyes closed; head upright/head extension) 2) Dynamic stabilization: Smooth pursuit and head oscillation exercises during free stance as described in preceding section. Exercises with rope, ball, and club with fixation (eye and head) of the object (sitting/standing/ walking)
Recalibrate visuovestibulospinal reflexes for postural control and eye-head coordination during free body movements
Complex balance exercise, successive increase in di culty, above the demands for postural control under everyday conditions
Expose the subjectively 'recovered patient' increasingly to unstable body positions in order to facilitate rearrangement and recruitment of control capacities
132
is reached for postural recovery as well as commissural changes about 60 days after hemilabyrinthectomy [48 ,77J. Pharmacological and metabolic studies in animals suggest that the state of central compensation for peripheral vestibular lesions is both dynamic and fragile [86]. Alcohol, phenobarbital, chlorpromazine, diazepam, Ca 2 channel antagonists [15], and ACTH antagonists [25J may retard compensation; ca eine, amphetamines, and ACTH accelerate compensation; cholinomimetics, cholinesterase inhibitors, adrenergic agents, GAB A agonists, and alcohol can (re)produce decompensation. It still remains to be proven if the use of drugs accelerates compensation in patients [76]. As mentioned above, many patients develop phobic postural vertigo [7], secondary VN, even if peripheral vestibular function has completely recovered. Our therapeutic regimen for phobic postural vertigo consists mainly of: (a) relieving the patients of their fear of a severe organic disease; (b) providing them with a detailed explanation of the causative mechanism and the factors that provoke phobic postural vertigo attacks; (c) initiating self-controlled 'desensitization' - within the context of behavioral therapy - by repeated exposure to situations that evoke the vertigo, and (d) advocating regular but not overly strenuous physical activity to improve the sense of diminished fitness. In summary: (1) Antiemetlcs and sedatives should be given only during the period of 1-3 days after symptom onset, because they prolong the time required to achieve central compensation; (2) on the basis of the probable viral etiology of vestibular neuritis, steroids in combination with antiviral substances (acyclovir) may improve the outcome, although this has not yet been systematically studied, and (3) vestibular exercises improve vestibulospinal compensation and should be begun as early as possible after symptom onset.
References Adour KK, Ruboyianes JM. Von Doersten PC, Byl FM. Trent CS, Quesenberry CP Jr. Hitchcock T: Bell's palsy treatment with acyclovir and prednisone compared with prednisone alone: A doubleblind, randomized, controlled trial. Ann Otol Rhinol Laryngol 1996;105:371-378.
2
Arbusow V, Dieterich M, Strupp M. Dreher V, Jager L, Brandt T: Herpes zoster neuritis involving superior and inferior parts of the vestibular nerve cause ocular tilt reaction. Neuro-ophthalmology
3
Ariyasu L, Byl FM, Sprague MS, Adour KK: The beneficial e ect of methylprednisolone in acute vestibular vertigo. Arch Otolaryngol Head Neck Surg 1990; 116:700-703. Bergstrom B: Morphology of the vestibular nerve. II. The number of myelinated vestibular nerve fibers in man at various ages. Acta Otolaryngol (Stockh) 1973;76:173-179. Bohmer A, Rickenmann]: The subjective visual vertical as a clinical parameter ofvestibular function in peripheral vestibular diseases. J Vestib Res 1995;5:35-45. Brandt T: Vertigo: Its Sensorimotor Syndromes, ed 2. London, Springer. 1999.
1998;19:17-22.
4 5 6
Vestibular Neuritis
133
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
27 28 29 30
31 32 33 34
Brandt T: Phobic postural vertigo. Neurology 1996;46: 1515-1519. Brandt T. Allum JH, Dichgans J: Computer analysis of optokinetic nystagmus in patients with spontaneous nystagmus of peripheral vestibular origin. Acta Otolaryngol (Stockh) 1978;86:115-122. Brandt T. Daro RB: The multisensory physiological and pathological vertigo syndromes. Ann Neurol 1980;7: 195-203. Brandt T, Krafczyk S, Malsbenden I: Postural imbalance with head extension: Improvement by training as a model for ataxia therapy. Ann J\.ry Acad Sci 1981;374:636-649. Brandt T, Strupp M, Arbusow V, Dieringer N: Plasticity of the vestibular system: Central compensation and sensory substitution for vestibular deficits. Adv Neurol 1997;73:297-309. Buchele W, Brandt T: Vestibular neuritis - A horizontal semicircular canal paresis? Adv Otorhinolaryngol 1988;42:157-161. Cawthorne T: The physiological basis for head exercises. J Chart Soc Physiother 1944;106-107. Courjon JH, Jeannerod M, Ossuzio I, Schmid R: The role of vision in compensation of vestibuloocular reflex after hemilabyrinthectomy in the cat. Exp Brain Res 1977;28:235-248. Darlington CL, Smith PF: Pre-treatment with a Ca' channel antagonist facilitates vestibular compensation. Neuroreport 1992;3: 143-145. Davis LE: Viruses and vestibular neuritis: Review of human and animal studies. Acta Otolaryngol Suppl (Stockh) 1993;503:70-73. Davis LE, Johnson RT: Experimental viral infections of the inner ear. 1. Acute infections of the newborn hamster labyrinth. Lab Invest 1976;34:349-356. Depondt M: Vestibular neuronitis. Vestibular paralysis with special characteristics. Acta Otorhinolaryngol Belg 1973;27:323-359. Dieterich M. Buchele W: MR1 findings in lesions at the entry zone of the eighth nerve. Acta Otolaryngol Suppl (Stockh) 1989:468:385-389. Dix MR, Hallpike CS: The pathology. symptomatology, and diagnosis of certain common disorders of the vestibular system. Ann Otol 1952;61:987-991. Ewald R. Physiologische Untersuchungen nber das Endorgan des Nervus octavus. Wiesbaden, Bergmann. 1892. Fetter M, Dichgans J: Vestibular neuritis spares the inferior division of the vestibular nerve. Brain 1996;119:755-763. Francis DA. Bronstein AM. Rudge p. Du Bou]ay EP: The site of brainstern lesions causing semicircular canal paresis: An MRI study. J Neurol Neurosurg Psychiatry 1992;55:446-449. Furuta Y. Takasu T, Fukuda S, Inuyama Y, Sato KC. Nagashima K. Latent herpes simplex virus type 1 in human vestibular ganglia. Acta Otolaryngo] Supp] (Stockh) 1993;503:85-89. Gilchrist DP, Smith PF, Darlington CL: ACTH(4-1O) accelerates ocular motor recovery in the guinea pig following vestibular dea erentation. Neurosci Lett 1990;118:14-16. Goldberg JM. Fernandez C: PhysIology of peripheral neurons innervating semicircular canals of the squirrel monkey.!. Resting discharge and response to constant angular accelerations. J Neurophysial 1971;34:635-660. Hain TC, Fetter M, Zee DS: Head-shaking nystagmus in patients with unilateral peripheral vestibular lesions. Am J Otolaryngo] 1987;8:36-47. Ilallpike CS: The pathology and di erential diagnosis of aural vertigo. Proc 4th Int Congress Oto]aryngo], London. Br Med Assoc 1949;2:514 Ha]magyi GM, Curthoys IS: A clinical sign of canal paresis. Arch Neuro] 1988;45:737-739. Ha]magyi GM, Curthoys IS, Cremer PD, Henderson Cj, Todd Ivl], Staples MJ, D'Cruz DM: The human horizontal vestibula-ocular reflex in response to high-acceleration stimulation before and after unilateral vestibular neurectomy. Exp Brain Res 1990;81:479-490. Hasuike K, Sekitani T, Imate Y: Enhanced MRI in patients with vestibular neuronitis. Acta Otolaryngo] Suppl (Stock h) 1995;519:272-274. I-lelmchen C, Jager L, Buttner U, Reiser M, Brandt T: Cogan's syndrome: High-resolution MRI as an Indicator of activity. J Vestib Res 1998;8:155-167. Herdman SJ: Vestibular Rehabilitation. Philadelphia, Davis. 1994. HUding DA, Kanda T, House WF: Vestibular neuronitis and small acoustic neuroma: Electron microscopic observations. Oto] Clin North Am 1968;112:305-318.
Strupp/Brandt
134
35 36 37 38 39 40 41 42
43 44 45 46 47 48 49 50 51 52 53 54 55
56 57 58 59 60 61
62
Hirata T, Sekitani T, Okinaka Y, Matsuda Y: Serovirological study of vestibular neuronitis. Acta Otolaryngol Suppl (Stock h) 1989;468:371-373. Hirata Y, Gyo K, Yanagihara N: Herpetic vestibular neuritis: An experimental study. Acta Otolaryngal Suppl (Stock h) 1995;519:93-96. Honrubia V: Quantitative vestibular function tests and the clinical examination; in Herdman SJ (ed): Vestibular Rehabilitation. Philadelphia, Davis, 1994, pp 113-164. Hopf HC: Vertigo and masseter paresis. A new local brainstem syndrome probably of vascular origin. J Neurol 1987;235:42-45. Hufschmidt A, Dichgans J, Mauritz KH, Hufschmidt M: Some methods and parameters of body sway quantification and their neurological applications. Arch Psychiatr Nervenkr 1980;228; 135-150. Huppert D, Brandt T, Dieterich M, Strupp M: Phobic vertigo. The second most common diagnosis in specialized ambulatory care for vertigo. Nervenarzt 1994;65:421-423. Hyden D, Odkvist LM, Kylen P: Vestibular symptoms in mumps deafness. Acta Otolaryngol Suppl (Stockh) 1979;360: 182-183. Igarashi M, Levy JK, 0 Uchi T, Reschke MF: Further study of physical exercise and locomotor balance compensation after unilateral labyrinthectomy in squirrel monkeys. Acta Otolaryngol (Stockh) 1981;92;101-105. Ishikawa K, Edo M, Togawa K: Clinical observation of 32 cases of vestibular neuronitis. Acta Otolaryngol Suppl (Stockh) 1993;503:13-15. Ishizaki H, Pyykko I, Nozue M: Neuroborreliosis in the etiology of vestibular neuronitis. Acta Otolaryngol Suppl (Stock h) 1993;503:67-69. Jager L, Strupp M, Brandt T, Reiser M: Imaging of the labyrinth and vestibular nerve: Clinical relevance for di erential diagnosis of vestibular disorders. Nervenarzt 1997;68:443-458. Kamei T: The two-phase occurrenceofhead-shaking nystagmus. Arch OtorhinolaryngoI1975;209:59--£7. Katsarkas A, Galiana HL: Bechterew's phenomenon in humans. A new explanation. Acta Otolaryngal Suppl (Stock h) 1984;406:95-100. Kunkel AW, Dieringer N: Morphological and electrophysiological consequences of unilateral preversus postganglionic vestibular lesions in the frog. J Camp Physiol A 1994; 174 :621-632. Lacour M, Roll JP, Appaix M: Modifications and development of spinal reflexes in the alert baboon (Papio papio) following a unilateral vestibular neurectomy. Brain Res 1976; 133;255-269. Lindsay JR, Hemenway WG: Postural vertigo due to unilateral sudden partial loss of vestibular function. Arch Otolaryngol 1956;65:692-706. Langridge NS: Recurrent vestibulopathy: Support for a viral etiology. J Otolaryngol 1989;18:99-100. Lorente de No R: Vestibula-ocular reflex arc. Arch Neurol Psychiatry 1933;30:245-291. Matsuo T: Vestibular neuronitis - Serum and CSF virus antibody titer. Auris Nasus Larynx 1986; 13:11-34. Matsuo T, Sekitani T, Honjo S, Imate Y, Inokuma T: Vestibular neuronitis. Pathogenesis in the view of virological study of CSF. Acta Otolaryngol Suppl (Stockh) 1989;468:365-369. Matsuzaki M, Kamei T: Stage assessment of the progress of continuous vertigo of peripheral origin by means of spontaneous and head-shaking nystagmus findings. Acta Otolaryngol Suppl (Stockh) 1995;519: 188-190. Meran A, Pfaltz CR: The acute vestibular paralysis. Arch Otorhinolaryngol 1975;209:229-244. Nadal JB Jr: Vestibular neuritis. Otolaryngol Head Neck Surg 1995;112:162-172. Nahmias N, Roizman B: Infection with herpes simplex viruses I and 2. II. N Engl J Med 1973; 289:719-725. Nylen CO: Some cases of ocular nystagmus due to certain positions of the head. Acta Otolaryngol (Stockh) 1924;6:106-137. Ogata Y, Sekitani T, Shimogori H, Ikeda T: Bilateral vestibular neuronitis. Acta Otolaryngol Suppl (Stock h) 1993;503:57-60. Ohbayashi S, Oda M, Yamamoto M, Urano M, Harada K, (-lorikoshi I-I, Orihara H, Kitsuda C: Recovery of the vestibular function after vestibular neuronItis. Acta Otolaryngol Suppl (Stockh) 1993;503:31-34. Okinaka Y, Sekitani T, Okazaki H, Miura M, Tahara T: Progress of caloric response of vestibular neuronItis. Acta Otolaryngol Suppl (Stockh) 1993;503:18-22.
Vestibular Neuritis
135
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
79
80 81 82
83 84 85 86 87
Proctor L. Perlman H, Lindsay J, Matz C: Acute vestibular paralysis in herpes zoster oticus. Ann Otol Rhinal Laryngol 1979;88:303-310. Ruttin B: Zur Di erentialdiagnose der Labyrinth- und Hornerverkrankungen. Z Ohrenheilkunde 1909;57:327-333. SafranAB, Vibert 0, Issoua 0, Hausler R: Skew deviation after vestibular neuritis. Am J Ophthalmol 1994; 118:238-245. Sando I, Black FO, Hemenway WC: Spatial distribution of vestibular nerve in internal auditory canal. Ann Otol 1972;81:305-319. Schuknecht HF: Pathology of the Ear, ed 2. Philadelphia, Lea & Febinger, 1993. Schuknecht HF, Donovan ED: The pathology of idiopathic sudden sensorineural hearing loss, Arch Otorhinolaryngol 1986;243; 1-15. Schuknecht HF, Kitamura K: Vestibular neuritis. Ann Otol 1981;90(suppI78);1-19. Schuknecht HF, Witt RL: Acute bilateral sequential vestibular neuritis. Am J Otolaryngol 1985;6: 255-257. Sekitani T, Imate Y. Noguchi T, Inokuma T: Vestibular neuronitis: Epidemiological survey by questionnaire in Japan. Acta Otolaryngol Suppl (Stockh) 1993;503:9-12. Shimizu T, Sekitani T, Hirata T, Hara H: Serum viral antibody titer in vestibular neuronitis. Acta Otolaryngol Suppl (Stock h) 1993;503:74-78. Shirabe S: Vestibular neuronitis in childhood. Acta Otolaryngol Suppl (Stockh) 1988;458:120-122. Silvoniemi P: Vestibular neuronitis. An otoneurological evaluation. Acta Otolaryngol Suppl (Stockh) 1988;453:1-72. Sloane PO, Baloh RW. Honrubia V: The vestibular system in the elderly: Clinical implications, Am J Otolaryngol 1989; 10:422-429, Smith PF, Darlington CL: Can vestibular compensation be enhanced by drug treatment? J Vestib Res 1994;4:169-179. Straka H, Dieringer N: Spinal plasticity after hemilabyrinthectomy and its relation to postural recovery in the frog. J Neurophysiol 1995;73:1617-1631. Strupp M, Arbusow V. Brandt T: Vestibular exercises improve central vestibula-spinal compensation after an acute unilateral peripheral vestibular lesion: A prospective clinical study. Neurology 1998. in press, Strupp M, Arbusow V, Dieterich M, Sautier W. Brandt T: Perceptual and oculomotor e ects of neck muscle vibration in vestibular neuritis; Ipsilateral somatosensory substitution of vestibular function. Brain 1998; 121:677-685. Strupp M, Brandt T: Vestibulo-okularer Reflex; in Huber A, Kampf 0 (eds): Klinische Neuroophthalmologie. Stuttgart, Thieme, 1997, pp 78-85. Strupp M, Brandt T, Steddin S: Horizontal canal benign paroxysmal positioning vertigo: Reversible ipsilateral caloric hypoexcitability caused by canalolithiasis? Neurology 1995;45:2072-2076, Strupp M, Jager L. Moller-Lisse U. Arbusow V, Reiser M, Brandt T; High resolution MRI in 60 patients with vestibular neuritis: No contrast enhancement of the labyrinth or vestibular nerve. J Vestib Res 1998;8:1-7. Tahara T, Sekitani T. Imate Y, Kanesada K. Okami M: Vestibular neuronitis in children. Acta Otoiaryngol Suppl (Stock h) 1993;503:49-52. Tran Ba Huy P: Physiopathology of peripheral non-Meniere's vestibular disorders. Acta Otolaryngol Suppl (Stockh) 1994;513:5-10. Vibert 0, Hausler R, Safran AB, Koerner F: Diplopia from skew deviation in LUlilateral peripheral vestibular lesions. Acta Otolaryngol (Stockh) 1996; 116: 170-176. Zee OS: Perspectives on the pharmacotherapy of vertigo. Arch Otolaryngol 1985;111:609-612. Zee OS, Preziosi TJ, Proctor LR: Bechterew's phenomenon in a human patient (letter). Ann Neurol 1982; 12;~95-~96.
Dr. Michael Strupp, MD, Department of Neurology, University of Munich, KJinikum Crosshadern, Marchioninistrasse 15, 0-81377 Munich (Cermany) Tel. 49 89 7095 2571, Fax 49 89 7095 8883, E-Mail
[email protected] Strupp/Brandt
136
Btittner U. (ed): Vestibular Dysfunction and Its Therapy. Adv Otorhinolaryngol. Basel, Karger, 1999, vol 55, pp 137-168
Meniere's Disease K. -F Hamann, W Arnold Department of Otorhinolaryngology, Head and Neck Surgery, Klinikum rechts der Isar, Technical University of Munich, Germany
Definition
Meniere's disease can be defined as an inner ear disorder of unknown cause [92J. Clinically it is characterized by fluctuating hearing loss, vertigo attacks, tinnitus and a sensation of aural fullness in which its pathological anatomical correlate is endolymphatic hydrops (fig. 1). This definition makes it clear that the term Meniere's disease does not fulfil the criteria of a nosologic entity since the etiology is not fully known. However, it is more than just a syndrome, it is a state of illness. Although there are restrictions to the term Meniere's disease, it will be used in the following chapter as it is used worldwide. In addition, other diseases of the cochlea-vestibular system or central nervous system (e.g, acoustic neuroma) tend to mimic the disease by presenting with similar clinical symptoms. Therefore, di erential diagnosis of Meniere's disease is challenging. Even for an experienced clinician it is not simple to recognize and diagnose Meniere's disease until the complete picture of the classical triad exists, and in addition a retrocochlear cause has been excluded. Even though the International Meniere Society has attempted to classify an 'inner ear profile' for the symptoms of the definite diagnosis, the AAOHNS (American Academy of Otolaryngology, Head and Neck Surgery) has published specific guidelines to define Meniere's disease and the development of endolymphatic hydrops [23J. Classification of Meniere's disease depends on the severity uf tile symptums, vertigu, hearing luss and tinnitus (table 1). These recommendations should serve the purpose of comparing diagnosis and clinical studies with di erent neum-otological centers and clinics. Similar gUidelines have been presented by the AAO-HNS [1) in order to compare the di erent types of treatment (see below).
Table 1. Diagnostic scale for Meniere's disease of the AAO-HNS 1 [1]
Certain Meniere's disease Definite Meniere's disease, plus histopathological confirmation Definite Meniere's disease Two or more episodes of vertigo of at least 20 min Audiometrically documented hearing loss on at least one occasion Tinnitus or aural fullness Probable Meniere's disease One definite episode of vertigo Audiometrically documented hearing loss on at least one occasion Tinnitus or aural fuJJness Possible Meniere's disease Episodic vertigo without documented hearing loss Sensorineural hearing loss, fluctuating or fixed, with dysequilibrium, but without definite episodes I In all scales, other causes must be excluded using any technical method (e.g. imaging, laboratory, etc,),
The Development of Research in Meniere's Disease
Following Hard's [45] first description of the combination of symptoms of vertigo, hearing loss and tinnitus in 1821, it was the merit of Prosper Meniere [62], who in 1861 presented in Paris the clinical a ictions of the inner ear disorder thereby crediting him with the recognition of the syndrome as more than a sign of 'apoplectic cerebral congestion'. P. Meniere was influenced by the findings of the physiology of the inner ear by Flourens [31J well known to him. Thereby he correctly classified the symptoms to the inner ear. A few years later the syndrome of fluctuating hearing loss accompanied by vertigo and tinnitus was named by Duplay [27J as Meniere's disease. In 1921, G. Portmann [78] performed an investigation on Selacian fish and determined endolymphatic hydrops as a possible cause for Meniere's disease, the proof however was not given until 1938 by Hallpike and Cairns [39] in London and at the same time independently, by Yamakawa [117] in Osaka. This finding not only made for a better understanding of Meniere's disease, but was alsu an impurtant landmark in experimental research. Schuknecht and Igarashi [91J demonstrated in a series of temporal bone examinations of Meniere patients that endolymphatic hydrops always involved the sacculus and utriculus although hydrops in the endolymphatic compartments of the cochlea developed in completely di erent regions (fig. 1).
Hamann!Arnold
138
2
Fig. 1. Left cochlea from a 67-year-old patient suffering from Menie'res disease on both sides. The endolymphatic space (El is enlarged and Reissners membrane (RMl is not in its normal anatomic position. The folding of the lateral part of RM indicates a collapse of this area. PPerilymphatic space: S = spiral ganglion. Fig. 2. Guinea pig cochlea 6 weeks following destruction ofthe endolymphatic sac. The endolymphatic spaces (El of the scala media are enlarged indicating endolymphatic hydrops.
Based on the experimental results in animals from Guild [36] in which he demonstrated that by infusing dye into the cochlear endolymph it reappears in concentrated form in the endolymphatic sac, in 1959, Naito [69] was able to produce endolymphatic hydrops in the guinea pig by surgically destroying the endolymphatic duct and sac thereby producing hydrops. This method was then improved by Kimura and Schuknecht [48J in 1965 and is currently the most common model for experimental studies on endolymphatic hydrops. It was also proven that the endolymphatic sac is a decisive factor for the homeostasis of the endolymph (fig. 2).
Merrie'res Disease
139
In order to better understand the relationships between endolymphatic hydrops and the appearance of characteristic symptoms, Tasaki and Fernandez [100] in 1952 made the important finding that a reversible reduction of cochlear microphonic potentials and action potentials takes place - equivalent to impaired hearing - when the perilymphatic space of the cochlea is injected with Ringer solutions containing 0.25% potassium chloride. Two years later, Tasaki et al. [101] could show that this phenomenon only appears when the electrode into the scala tympani is inserted, however not when it is located in the scala vestibuli. In the mid-1950s it was proven that the endolymph only contains a limited amount of sodium concentration (3-5 mval), however an abundant amount of potassium (130-145 mval); the opposite is true for the perilymph [94J. Finally, Lawrence and McCabe [56J postulated that ruptures of the endolymph membranes extended by the hydrops, followed by a mixing of potassium-rich endolymph and sodium-rich perilymph, are the cause of the symptoms of the attacks. This hypothesis was confirmed when they succeeded in simulating Meniere attacks in animals by injecting potassium-rich solution into the perilymphatic space until the concentration was comparable to that of endolymph [25J. As a result, the concept for potassium intoxication of the demyelinated nerve fibers within the sensory epithelia of the cochlear vestibular organ was accepted as the basis of the fundamental mechanism for attacks and the cause of a lesion nystagmus in monkeys [25]. Current research supports the hypothesis that Meniere's disease is caused by a malabsorption of the endolymph in the endolymphatic sac. Therefore, special interest is focused on this structure and its pathological changes. Schuknecht [92] demonstrated on temporal bones of Meniere patients that not only are endolymphatic hydrops present in this area but also a fibrosis or an osseous obliteration of the endolymphatic duct and sac. Immunological investigations on the human endolymphatic sac have proven that this organ has additional immunological functions [3, 7, 8] which will be discussed below.
Clinical Course and Epidemiology
Meniere's disease is characterized by a clinical triad which consists of attacks of vertigo, fluctuating hearing loss, tinnitus and a sensation of fullness uf the ear. Huwever, the disease dues nut always lJegin with a complete picture of the above-mentioned symptoms and often develops monosymptomatically, yet acute vertigo or an acute loss of hearing has been observed in the same frequency. In 90% of the patients the complete symptom picture does not become clear until 1 year after the first symptoms have been experienced [77J.
Hamann!Arnold
140
The course of the disease is also not homogeneous. Frequency and length of vertigo attacks, the most disturbing symptom, are variable so that the e ect of di erent therapies is di cult to evaluate. The spontaneous remission rate is remarkable, but di cult to determine since intervals between attacks can last days, months or sometimes even years. Nevertheless, many patients' symptoms normally subside within 8-10 years because the disease is 'burnt out'. Loss of hearing may worsen during the course of the disease which can result in deafness. Tinnitus may remain or disappear, vertigo attacks may yield slightly to a constant unsteadiness [74]. Wide variation exists in the published incidence of Meniere's disease. Incidence varies from 4/100,000 persons in Japan [109] to 160/100,000 persons in Great Britain [20] and 15/100,000 in the USA [116J. It is unclear whether these di erences can be deduced back to genetic or environmental factors or whether it is more the di erences of diagnostic criteria. Contrary to a genetic disposition is a study by Kitahara et al. [51] which demonstrated no di erences in occurrence of Meniere's disease in di erent ethnic groups in America. A sexual preponderance of the disease is not clearly defined although some reports show a slight predominance in women. The disease can appear at any age, however a peak has been noted between the 3rd and 4th decade [74]. Remarkable is the frequent appearance of occurrence of bilateral Meniere's disease which continually increases within the period of observation and according to Paparella's data can increase up to 50%. These figures allow for the assumption that Meniere's disease has a systematic cause, e.g. an underlying immunological mechanism a ecting both sides of the inner ear.
Etiology
Today, more than a hundred years after Meniere's first publication, the disease is still classified as 'a disorder of unknown or multiple causes' [92J. Many di erent etiologies have been discussed in the last decades, each supported by numerous arguments although convincing evidence for one of these hypotheses is still outstanding. In the following the most important hypotheses of the past will be presented, due to its actuality and high plausibility the immunological hypothesis will be discussed in more detail. Since it is generally accepted that the pathological anatomical correlate of Meniere's disease is endolymphatic hydrops (see below), the etiological question is, which factors lead to the development of endolymphatic hydrops. It must certainly be taken into consideration that endolymphatic hydrops is not only present in Meniere's disease, but has also been found in inner ears
Meniere's Disease
141
of patients not su ering from Meniere's disease, in the form of delayed endolymphatic hydrops [68J. Fundamental far the possible development of the mechanism in Meniere's disease is the assumption that endolymphatic hydrops is caused by an imbalance of endolymph production in the stria vascularis and endolymph reabsorption in the endolymphatic sac, in other words either by overproduction or a reduced absorption of the endolymph. The di erent hypotheses must be measured with respect to these basic assumptions.
Genetic Hypothesis Several authors have inferred genetic connections although they have not been confirmed by others [15]. The e arts of Bernstein et al. [14], in which the role of the MHC gene (major histocompatibility complex) with particular antigens is connected to Meniere patients, have not been convincing. Psychogenk Factors The hypothesis that psychological factors play an important role in Meniere's disease has been suggested numerous times. Fowler and Zeckel [32] favored this opinion since they relied on the observation of their own patients. Other authors dismiss this point of view and suggest a more somatopsychological e ect [113J. The main argument against this hypothesis is however the fact that there are not direct connections between inner ear structures and higher cerebral structures such as in the limbic system; a vegetative innervation is also not known. Vascular Hypothesis A more historical aspect is the hypothesis [88] that a circulatory insu ciency may cause Meniere's disease. This hypothesis must however be dismissed in that there is no valid evidence. The fluctuating characteristics of Meniere's disease led to the assumption that a vascular compression can cause the symptoms. This presumption leaves out the fact that the development of endolymphatic hydrops is the fundamental morphological basis in Meniere's disease. Supporters of the vascular compression hypothesis consider hydrops only as an epiphenomenon. They support their viewpoint with electron microscopic investigations [22] in which lesions to the central parts of the vestibular nerve were found in Meniere patients. Therefore, they cuncluded the cause uf these changes stems frum a vascular compression. Moreover, other authors argue that lesions in the central parts of the vestibular nerve must be interpreted as a degenerative change following years of Meniere attacks [95].
Hamann!Arnold
142
Viral Hypothesis In analogy to the modern view of the cause of facial palsy, a viral etiology of Meniere's disease [115] is postulated. Bergstrom et a1. [13] found viral antibodies in the serum of 20 of 21 Meniere patients against the herpes simplex virus in comparison to 18 of 21 controls. However, the increased antibody reactivity indicates individual HSV1 proteins in Meniere's patients in which a viral reactivation took place. In 1996, Kumagami demonstrated in healthy human endolymphatic sacs a homing of herpes simplex type I micro-organisms. Perilymph samples from Meniere patients demonstrated in contrast to those of otosclerosis patients or cochlear implant patients a significant increase of specific anti-HSV IgG titer, which is also remarkably higher than in the sera of the a ected patients [9]. These results can also be interpreted as a sign of viral reactivation of the endolymphatic sac virus which finally induced the development of endolymphatic hydrops and results as an immunological (inflammatory) reaction to the endolymphatic sac (see below). Immunological Hypothesis In the 1920s, Duke [26J first suspected a relationship between Meniere's disease and allergies. He observed in several patients who su ered from this disease, an increase in frequency of vertigo attacks related to food intolerance. This paint of view was rejected for decades since the inner ear was thought of as an organ which was privileged against immunological reactions. Based on experimental results of the last 20 years, a fundamental re-examination of these viewpoints is necessary. In 1979, McCabe [59] postulated that specific inner ear disorders can be caused by autoimmune phenomena. He reached this conclusion due to several etiologically unclear inner ear diseases which reacted positively to immunosuppressives. In his first reports he explicitly distinguished Meniere's disease to that of an autoimmune disease. Although these viewpoints were later revised, this working hypothesis triggered systematic examinations. Special interest was focused on the endolymphatic sac where endolymph reabsorption occurs and a disorder in this area could explain the development of endolymph hydrops. The first indications for the evidence that the endolymph sac presents as a site of immunological defense was made by RaskAnderson and Stahle [83]. They found in the guinea pig in the epithelia of the endolymphatic sac lymphocytes and in the lumen macrophages. Arnold et a1. [7] succeeded in 1984 to show irIlInunoglolJin G and A as well as secretory components in the human endolymphatic sac. These findings proved that the immunocompetence of the cochlea to the endolymphatic sac exists. Yet the question arises which immunological mechanisms can lead to these conceivable morphological changes found in Meniere patients. A first step in clarifying
Meniere's Disease
143
this was given by the animal experiments of Harris [42] in which he could demonstrate that local immune reactions in the inner ear could be produced by applying exogen antigen stimulants thereby causing specific antibodies to appear. Gloddek and Harris [34] showed that a transfer of antigens injected in the cochlea to the endolymphatic sac exists. For the local immunological response the endolymphatic sac is responsible, with a high degree of probability [7, 83]. These observations were supported by findings made with the same experimental approach, where destruction of the endolymphatic sac leads to a reduction in the immune reaction [67]. However, lymphocytes can also enter the cochlea through the bloodstream. This takes place through the modiolar vessel and the posterior modiolar spiral veins [43]. Therefore, the inner ear is also connected to the control system of the systematic immune apparatus. The endolymphatic sac can also react to di erent antigens coming either through the bloodstream to the inner ear or through the membranes of the cochlear windows of the middle ear, in other words the endolymph sac responds immunologically by inflammatory reactions. This determines the decisive defense mechanism in the inner ear. Although these seem to be biologically reasonable processes, they can however have negative consequences if the reocurrence of the inflammations of the endolymphatic sac leads to morphological persisting changes of the subepithelial inner surface of the saccus. A disorder of the reabsorption function of the endolymphatic sac is therefore assumed which leads to a predominant change in the homeostasis between continuing endolymph secretion and endolymph reabsorption in the endolymph's favor. As a consequence, endolymphatic hydrops develops, as has been determined histologically in patients with Meniere's disease [8]. An important step in investigation of the causes of Meniere's disease was the development of a method in which endolymphatic hydrops could be produced experimentally. Kimura and Schuknecht [48] perfected this method by producing endolymphatic hydrops in the guinea pig by obliterating the endolymphatic sac. It must however be mentioned here that although this model leads to endolymphatic hydrops in the guinea pig, this could not be reproduced in higher species such as monkeys. The question therefore remains as to whether the results of animal experiments along with their consequences can be transferred directly to humans if the recurrent inflammations of the endolymphatic sac lead to morphological persisting changes. In spite of these restrictiuns, the must convincing puint uf view in tile develupment ufMeniere's disease remains the current immunological hypothesis gained by clinical observations and animal models. A very important support for this hypothesis came from Rask-Andersen et al. [84], who removed the endolymphatic sac of a patient su ering from
Hamann!Arnold
144
Exogenous agent (e.g. virus)
~
Via coch ear windows or bloodstream
Cochlear immmunoreaction
~
Via endolymphatic flow
Inflammation of the saccus endolymphatlcus (SE)
~
Chroniflcatlon
Rbrous tissue and ossification In the (SE)
~ DiSlur
nce
:~::~::~:ti:n K:f the endolymph
Endolymphatic hypdrops
Fjg 3. Etiologic cascade of the development of endolymphatic hydrops.
active Meniere's disease for therapeutical reasons. The electron microscopic examination showed granular lymphocytes adhering to the disturbed and degenerating epithelial cells in combination with further signs of immunologic - finally cytotoxic - interactions on epithelial cells. These changes were interpreted as immunological, autoaggressive, and tissue-damaging reaction. Therefore, they named it 'saccitis' which could explain the disturbance of endolymph reabsorption followed by the development of an endolymph hydrops leading to the symptoms of Meniere's disease (fig. 3).
Pathophysiology of Meniere Attacks
The knowledge that the pathological anatomical equivalent to Meniere's disease consists of endolymphatic hydrops, provoked by an imbalance of endolymph production and endolymph reabsorption, led to the development of a pathophysiological model in explaining Meniere attacks. The basis of this model was developed by Lawrence and McCabe [56], elaborated by Schuknecht and Igarashi [91J and modified by Jahnke [46]. It is highly probable that the disturbed reabsUI'ption of the endolymph in the endolymphatic sac results in an increase in potassium while steady endolymph secretion continues. This gives rise to an increase of the osmotic pressure in the endolymphatic area whereby water from the surrounding areas, especially from the scala vestibuli, pours in. By this mechanism the semiperme-
Meniere's Disease
145
sv
RM
....
Fig. 4. Chronic Menie'res disease. There is a large connection between the scala media (SM) and scala vestibuli (SV) caused by rupture of Reissners membrane (RM).
able Reissner membrane distends and then bulges out. The Reissner membrane or the sacculus and utricular membranes, the locus minoris resistentiae in this system, are initially distended until finally they rupture (fig. 4, 5). Because the natural diaphragm between endo- and perilymph is lost. a mixture of both lymphs which originally each have different ion concentrations results [100]. The inflow of higher potassium levels from the endolymph causes at the level of the sensory cells, which are normally surrounded by the perilymph, such an intense depolarization that abnormal stimulation of the affected sensory cells results. This stage is comparable to that of an attack and clinically implies hearing loss and vertigo seizures. As a result of this decompression, leaks which developed during the rupture can close again. Since new ion gradients are reproduced, the acute symptoms of the attack disappear. Simultaneously however, conditions are created for hydrops to form again, leading to yet another rupture thereby causing a new attack. The appearance of irreversible cochlear and vestibular losses in the course of Menie'res disease can be explained by the repeated bulging of the endolymph space causing persisting damage [92]. This model concept explains Menie're attacks with all symptoms (fig. 4). It is however unclear why some patients experience attacks of shorter intervals yet may recur within days, whereas others can remain symptom-free for months or even years.
Hamann!Arnold
146
1. Normal
2. Endolymphatic hydrops
~/
Perilymph
Membranous labyrinth adherent _.------------to bony wall
Membranous labyrinth~~ • \
~ Endolymph
Endolymphatic duct
'I?
p-
. } 00 0
~~~~=c,.,..JjU '
S ense organ
I'
Internal auditory canal Cerebrospinal fluid (subarachnoid space)
3. Rupture of membranous labyrinth
"8~~'"
~
Decreased resorptive function of the endolymphatic sac
4. Healed membranous labyrinth and recurring hydrops
Partial collapse of membranous Iabyrlnth Rupture
Potassium K" .... contamination of perylymph with paralysis of sensory and neural tissues
Endolymph reaccumulating
Healed rupture Permanent alterations in sense organs
Fjg 5. Schematic drawing of di erent steps of development of seizures in Meniere's disease; from Schuknecht [92J.
Symptomatology - Disorders of the Vestibular System
Subjective Complaints The most common complaints of Meniere's disease are the disturbing attacks of vertigo in which the patient finally seeks the advice of a doctor. Hence, the necessity of a well-taken history, mainly the typical pattern of vertigo complaints, plays a vital role for the diagnosis of Meniere's disease. The most frequent type of vertigo in Meniere's disease is the sudden attack of torsional vertigo. Sensation of linear movements or 'drop attacks', which are usually typical for Tumarkin's otolithic crisis, are rare. The frequency and intensity of the attacks are irregular and unpredictable. Some patients report uf singular attacks during the course uf years whereas uther su er more intense vertigo attacks on several days in the course of a week. Especially characteristic, almost pathognomonic, for Meniere's disease is the time course of a vertigo seizure, the duration as well as its presentation of the attack. The duration of a typical Meniere vertigo seizure lasts minutes
Meniere's Disease
147
to hours, however never seconds or several days. An exception is only the first seizure which can last longer than 24 h and then often conceal the cochlear symptoms (see below). In advanced Meniere's disease acute torsional vertigo attacks change to a constant feeling of unsteadiness.
Oculomotor Symptoms Examination of oculomotor functions must take into account that Meniere's disease is characterized by an exchange of acute exacerbation and more or less silent periods so that objective signs of a vestibular disorder are to vary considerably. During the acute phase of a seizure, spontaneous nystagmus has been noted to beat towards the a ected ear (so-called potassium intoxication nystagmus or paralytic nystagmus) [19, 60J; however, shortly after the acute phase, spontaneous nystagmus beats toward the intact ear. This phenomenon reflects the change between the irritative stage and the destructive stage during the acute seizure. The same can be demonstrated with caloric testing in that hyperexcitability is present during the very acute stage of the a ected side, however in the symptom-free interval hypoexcitability or balanced excitability is present. Therefore vestibular testing as a diagnostic criteria for Meniere's disease is not suitable and does not lead to a final diagnosis. In the early stages of the disease no pathological signs of vestibular disorders in between attacks can be noted, neither spontaneous nor provocation nystagmus; caloric testing shows no side di erence in excitability. Only in the later stages of the disease is a persisting hypofunction of the a ected ear seen [110J. Rotatory testing shows corresponding findings, a directional preponderance of the nystagmus of the nona ected side. In the symptom-free (or symptom-poor) interval, depending on the activity of the disease, a balanced reaction on rotational stimuli or a directional preponderance corresponding to the side of hypofunction is determined. Oculomotor testing such as the 'eye tracking test' or triggering of an optokinetic nystagmus have not shown any specific changes in Meniere's disease. Test results are generally normal. That the otolith system is also a ected by the course of the disease can be deduced from the barbecue rotations in patients with Meniere's disease [49J. Torsional eye movements were seen in some patients as asymmetric, regardless if caloric testing showed side di erences in excitability or not. It should be mentioned again that even if nystagmus reaction was observed in the acute stage uf Menihe's disease, it dues nut pwvide adequate evidence of the a ected side although this is more so the case in the later stages of the disease.
Hamann!Arnold
148
Vestibular-Spinal Signs Reports on posture disorders due to Meniere's disease are rare. GeneraUy a patient's posture, with the exception during acute attacks, remains stable in the Romberg test if visual or proprioceptive information is available. Under sharpened conditions such as visual stabilization ('visual conflict stimulation') or destabilization by suddenly tilting the platform, pathological signs appear corresponding to the marked momentary vestibular lesion [71].
Symptomatology - Disorders of the Cochlear System
The fluctuating hearing loss of the a ected ear which increases during an attack in Meniere's disease is one of the most common symptoms of the disease. In the early stages it may be the only symptom resulting in an initial false diagnosis of sudden hearing loss. Often diplacusis is reported in addition to hearing loss, giving the patients the impression that their hearing is distorted. Although it is certain to say that hearing loss in Meniere's disease develops at the level of Corti's organ, it is di cult to find a plausible explanation. It can only be partly explained by the hydrostatical tension of the basilar membrane, because the outer hair cells should function normally, which is rarely the case [76J, as it is proven by recording of otoacoustic emlssions. Whether this is caused as a result of hydrostatical pressure of the tectorial membrane on the stereocilia of the outer hair cells, is only speculation. Hearing loss is generally recovered after the first attacks. Also after further attacks it may appear a 'restitutio ad integrum'. A remaining loss of cochlear function is only seen in later stages of the disease. Studies of large numbers of patients have shown that low-frequency hearing loss is most frequently seen in the early stages of the disease, however other forms have also been reported. During the course of the disease, high-frequency hearing is also incorporated and finally almost aJl hearing frequencies are a ected (fig. 6). The above cited findings describe why there is no uniform audiometric as well as reliable model for tone decay in the hydrops ear. Speech audiometry has not always shown, especially in long-lasting cases of Meniere's disease, typical findings of cochlear damage, more often a higher degree of speech discrimination loss is found than seen in pure tone audiometry. The reason can be found in the increasing deterioration of the spiral ganglia cell as a result of a retrograde nerve degeneration, which was proven by histological examinations [95J. Measurement of the stapedius reflex confirms the diagnosis of a cochlear disorder identified by the Metz recruitment. For many years evidence of recruitment' which was proven in the Fowler test (positive in 73-100% according to
Meniere's Disease
149
125 til
250
500
1,000
2,000
4,000
8,000 Hz
-10
E-O
0
z
10
A
20
30 B
40 50 60
C
70 80 90 100 110 120 1,500
A
3,000
6,000
12,000
Hg 6. Typical form of pure tone audiograms in patients with Menihe's disease. Early stage; B progressive stage; C end stage.
Hallpike) as well as in the Luscher or SISI test was accepted as a decisive clinical finding in Meniere's disease. These tests however lost their importance due to their limited validity. In recent years, objective methods for measuring the function of the outer hair cells such as the transitorial otoacoustic emissions (TEOAE) and the distortion products otoacoustic emissions (DPOAE) were developed in order to topographically classify hearing loss. In Meniere's disease they have not yet been useful in determining the final diagnosis. Only in rare cases can OAE be recorded as proof of intact outer hair cells, otherwise they cannot be recorded or only in a reduced form [76]. The main interest of audiological tests was directed to prove the evidence of endolymphatic hydrops. Two tests were favored: the glycerol test [52J and electrucuchleugraphy [29, 30]. In the Klockho test [52], adapted from the glaucoma test in ophthalmology, the intake of glycerol, a powerful osmotic agent, should temporarily induce dehydration of the endolymphatic hydrops. If this succeeds, a temporary increase of function of the hair cells as well as the hearing threshold curve is
Hamann!Arnold
150
2 ms
f----------1 Fjg 7. Electrocochleographic potential of a patient with endolymphatic hydrops: pathologic relationship between summating potential (SP) and NAP (nerve action potential) 35%.
to be expected. If this e ect occurs, it is considered proof for endolymphatic hydrops. However, a negative result does not indicate that Meniere's disease does not exist. On the other hand, a positive result of this test. in which hearing loss improves within a short period of time, indicates that the cause is related to the mechanical processes of the inner ear thereby leading to a functional disorder. Electrocochleography is among the 'evoked response audiometry' methods which comprises very early potentials (up to 2 ms). The source is usually seen in the cochlea itself and at the beginning of the n. cochlearis. The electrocochleographical potential consists of three components in which microphonic potentials are eliminated by the change in stimulation polarity, in which the summating potential (SP) represents a DC component, which reflects the distension of the basilar membrane and in which the action potential of the nerve has its source at the beginning of the n. cochlearis [29, 30]. The SP is significantly higher in Meniere's patients than in normals [3D]. The augmented SP is considered as the correlate of a basilar membrane distension to the scala tympani, which is caused by endolymphatic hydrops. Since potential amplitudes vary between individuals, Eggermont [29] introduced a practical method for evaluating electrocochleographic potentials by determining the ratiu uf the amplitudes uf the SP tu the actiun putential thereby identifying the existence of endolymphatic hydrops [30] (fig. 7). If the SP/AP ratio is 0.35, endolymphatic hydrops can be assumed. At the actual stage of knowledge, electrocochleography is currently the best objective method in determining endolymphatic hydrops. However. the rule also applies here, that
Meniere's Disease
151
only a positive test result indicates the existence of endolymphatic hydrops. A negative test result however, does not indicate that Meniere's disease does not exist since drainage of the hydrops could have occurred shortly prior to testing.
Tinnitus
Tinnitus a ects up to 90% of Meniere patients [74J and thereby ranks to the classical Meniere triad. Tinnitus is otherwise only present in approximately 5% of the population. Tinnitus may remain as an uncomfortable symptom long after vertigo attacks and hearing loss have stabilized. In Meniere's disease, tinnitus presents as a constant low-pitched rumbling or roaring with fluctuating intensity. Many times this tinnitus is masked by noises from the surrounding environment. Tinnitus in Meniere's disease can be compressed with a Politzer balloon, it disappears as long as the air column is compressed in the auditory canal. This phenomenon points to a peripheral, therefore cochlear origin of tinnitus [IOJ. The pathophysiological substrate for tinnitus is not known. The Tonndorf model [103J can be used to determine its origin by uncoupling the tectorial membrane from the stereocilia of the outer hair cells. This could also be an explanation for the positive result of the compression test. Interestingly, tinnitus as well as vertigo often disappear when a small hole in the stapedial foot plate is made and a small amount of perilymph is drained. Since the exact etiology of Meniere's disease is not known, the management of tinnitus can only be symptomatic. Although there is no valid general concept for treatment, an overview of the di erent types of therapies was given at the last tinnitus seminars [80, 81J.
Sensation of Pressure and Fullness of the Ear (Aural Fullness) Pressure or fullness of the ear do not belong to the classical Meniere triad, although many patients describe these sensations in the sense of aura before the onset of an attack (74%) [74]. This sensation of pressure can last in a milder form during as well as after the attack. Because pressure receptors have not yet been proven, an anatomical basis for inner ear complaints as an explanation does not exist [47]. Another explanation focuses on the sensory innervation of the dura mater which surrounds the emlulymphatic sac. This hyputhesis has a certain amuunt uf probability since it is known in the course of the disease that inflammatory reactions occur within the endolymphatic sac leading to morphologically persisting changes [83J. However, a satisfying general explanation for this frequently reported symptom is still unknown.
Hamann!Arnold
152
Special Forms
Lermoyez Syndrome In 1919, M. Lermoyez [57] described a recidiving of vertigo attacks accompanied by improved hearing of the a ected ear: 'Ie vertige, qui fait entendre' (Lermoyez syndrome). Because all signs of Meniere's disease were apparent, these complaints were seen as a special form. It is however rarely seen and its frequency is estimated as only 1% in Meniere patients [92]. Lermoyez himself postulated the cause as a spasm of the labyrinth vessels, resulting in long-lasting hearing loss and tinnitus. The sudden release of the spasm during the attack should lead to both a reduction in hearing loss and vertigo attacks. A clear pathophysiological explanation for Lermoyez syndrome actually does not exist. It is possible that in this unusual form of endolymphatic hydrops, ruptures solely of the sacculus or utricular area are present so that during a vertigo attack a decompression without ruptures in the cochlear areas results in a functional recovery of cochlear hair cells. Results of pathological-anatomical examinations from patients who su ered from Lermoyez syndrome are currently not known. Tumarkin Otolithic Crisis In 1936 the British otologist A. Tumarkin [105] reported on 3 patients who su ered from so-called 'drop attacks'. These are sudden falls in which the patient collapses. They are accompanied by a strong sense of pseudomovement in a vertical direction. 'Drop attacks' occur without warning, without loss of consciousness or relation to movement, yet often lead to injury. Tumarkin crises are typicalJy short, lasting only up to a minute. Complaints of other existing types of vertigo are rare, these appear only towards the final stages of this disease. The other symptoms fulfil the criteria for Meniere's disease. The exact incidence of Tumarkin crisis is not known; Schuknecht [92J estimated it at 10%. Tumarkin [105J himself postulated a vestibular origin for 'drop attacks'. Today it is assumed that due to its typical symptom characteristics, endolymphatic hydrops are localized in the otolith apparatus and not in the semicircular canals. Therefore it is called an otolithic crisis. In a case study of 11 patients with Tumarkin crisis, Owen Black et a1. [72] demonstrated that only with conservative treatment complete recovery was not pOSSible. All patients were finally surgically treated either with labyrinthectomy or neurectomy.
Meniere's Disease
153
Imaging Techniques
The value of imaging techniques as a diagnostic tool in Meniere's disease was, until recently, underestimated. Due to its small size, structural changes in the temporal bone could not be detected with conventional methods so that reliable evidence of the disease could not be produced. Only in the past few years have there been attempts by using new high-resolution imaging techniques in detecting the structural changes in the area of the endolymphatic sac typically seen in Meniere's disease. During the 1970s and 1980s a perisaccular reduction of pneumatization was described [114J. With the aid ofa lateral polytomography, a computer-assisted radiography, a hypoplastic endolymphatic duct could be proven in Meniere's patients. The 3-DFT-CISS-MR technique allows the presentation of soft tissue of the endolymphatic sacs in the human temporal bone. The results of a study on 20 patients, in which 23 ears were a ected by Meniere's disease, were compared to 50 healthy normals [2). In 74% of the diseased ears an altered or not visualized endolymphatic duct and sac could be determined. Interestingly, the same pathological findings existed in 79% of nondiseased ears [2J. It is di cult to decide, based on these findings, whether these positive cases concern ears which may later be a ected by Meniere's disease. At the moment, imaging techniques as a diagnostic tool in Meniere's disease are of minor contribution, unless it is used to exclude other diseases (for example acoustic neuroma).
Differential Diagnosis
Vestibular Neuritis Vestibular neuritis presents as an acute reduction in unilateral function of the peripheral vestibular apparatus or of the vestibular nerve, clinically seen as severe systematic vertigo. The etiology is unclear although it has been suggested that an acute circulation disorder at the level of the vestibular apparatus may be a cause. However, since 1981 [89J a viral infection of the vestibular nerve was made probable by pathological examinations. Gradual recovery of vertigo takes place within days as a result of the spontaneous course of compensation mechanisms, however a full recovery is not always seen. Long-term prognosis is good in that the compensation mechanisms are supported by active rehabilitation management [40J. Meniere's disease is usually easy to define from that of vestibular neuritis in that in the latter cochlear symptoms are not present and in the former vertigo complaints have a longer duration.
Hamann!Arnold
154
Acute Hearing Loss Acute hearing loss is defined, in the closer sense, as a unilateral sudden deafness of the inner ear of unknown cause. The clinical picture presents as a cochlear counterpart to vestibular neuritis. Spontaneous recovery of sudden deafness is shown in 70% of cases [Ill]. Since it cannot be determined at the onset of the disease whether patients will experience a spontaneous remission or not, it must be treated as an urgent case. Treatment is usually polypragmatical among which corticosteriods are used due to the assumed inflammatory pathogenesis of sudden hearing loss and circulatory-promoting agents providing oxygen for the recovery of cochlear hair cells. The acute hearing loss is generally simple to diagnose in that the typical vertiginous symptoms of Meniere's disease are not apparent. Cases which are di cult to determine are those presenting as the onset of Meniere's disease with an acute hearing loss, but can only retrospectively be diagnosed. Loss of Labyrinthine Function (Apoplexia labyrinthi) The acute loss of labyrinthine function, cochlear and vestibular, presenting with vertigo, tinnitus and hearing loss can mimic Meniere's disease initially. The final diagnosis is determined by the varying time course of the symptoms. In contrast to an acute Meniere attack lasting minutes to hours, loss of labyrinthine function can last up to days. The etiology in this syndrome is also unknown and therapy is polypragmatical and symptomatic (see above). Suggested treatment corresponds to that of vestibular neuritis and of acute hearing loss. Benign Paroxysmal Positioning Vertigo (BPPY) BPPV is characterized with brief (seconds) attacks of vertigo without cochlear symptoms, typically evoked by head movements [24). In recent years, findings have suggested that BPPV can be attributed to a canalolithiasis [18, 112). Its pathogenesis, namely the pathological stimulation of sensory cells in the semicircular canal by free-floating otolith particles, explains the course of vertigo. Although BPPV is rotational as in Meniere's disease, its typical varying time characteristics prevent it from being confused with Meniere's disease particularly since cochlear symptoms are never present. Acuustic NeuIUIIla
Acoustic neuromas have the tendency to mimic Meniere's disease due to the various constellation of symptoms. Acoustic neuroma is defined as a schwannoma stemming from the vestibular portion of the eighth cranial nerve in the internal auditory canal. The
Meniere's Disease
155
primary damage is therefore done to the vestibular nerve yet rarely leads to vertigo, since the slow progressing lesion is almost synchronously compensated in the vestibular nuclei. Progressive hearing loss or occasional sudden deafness may be a result of an acoustic neuroma in which a lesion of the tumor to the cochlear nerve or compression to the labyrinth artery has been produced. Hearing loss in these cases can recover so that their seizure-like attacks lead to diagnosis of Meniere's disease. A careful neuro-otological examination including caloric test, brainstem evoked response audiometry (BERA) and, if necessary, MRI can determine if an acoustic neuroma exists. Perilymph Fistulas Perilymph fistulas are caused by overpressure to the inner ear compartments, e.g. that experienced when diving or strenuous physical exertion or without any recognizable cause [35, 98]. Apart from the acute sensorineural hearing loss, vertigo and tinnitus may also present making a di erential diagnosis of Meniere's disease often di cult. The temporary characteristics also point to correct diagnosis. Although a pronounced reduction of symptomatology in the initial stages of Meniere's disease can be seen, this is not the case with perilymph fistulas. Final diagnosis can be made with a tympanotomy and control of the windows to the inner ear. In these cases the fistula can be surgically closed. VertebrobasiJar lnsu ciency (VBI) VBI, with short ischemias of the vertebral and basilar arteries, is seen in a sense as a TIA (transient ischemic attacks) or a PRIND (prolonged reversible ischemic neurologic deficit). Vertigo is indeed one of the most frequent symptoms of VBI, however it is never the sale symptom [16J. Based on the vascular supply of the brainstem, in case of an impairment of circulation, symptoms apart from the cochlea-vestibular symptoms must appear. Reduced visual acuity or oscillopsia are the typical ophthalmological symptoms. The risk of mistaking VBI for Meniere's disease is due to its having the same time characteristics. Di erential diagnosis to Meniere's disease can only be made by a thorough neurological examination since there is no current imaging technique available to determine VBI except for complete infarction. Pharmacological Side E ects Numerous medications have side e eets which can either damage the peripheral cochlea-vestibular system or its central part. Of relevance are the ototoxic antibiotics which cause damage to the peripheral sensory epithelium [l1J. Drugs which contain typical agents with central e ects to the vestibular system are nonsteroidal anti-inflammatories as well as anti-epileptics and sed-
Hamann!Arnold
156
Table 2. Treatment of Menieres disease [data from 1]
Drug therapy Conservative (e.g. betahistine) Destructive (e.g. gentamicin) Surgical therapy Nondestructive Ventilation tube Cochleosacculotomy Saccotomy Destructive Labyrinthectomy Neurectomy of vestibular nerve
atives [17]. In these cases a close chronological relationship exists between intake and onset of symptoms. On the other hand, these unwanted side e ects often last longer than the typical symptoms of a Meniere attack. A complete drug anamnesis would bring prompt clarification.
Treatment
General Remarks Treating Meniere's disease has per se two special problems: on the one hand, an etiologic therapy is excluded due to unknown cause, and on the other, because of its cycliC course with frequent spontaneous remissions, it is di cult to evaluate the e ect of each treatment. Until recently, most of the treatment regimens aimed at influencing endolymphatic hydrops, since it is the only pathological substrate presenting in Meniere's disease, or they were directed at relieving the symptoms, especially vertigo. Recently published studies [e.g. 108] have also brought about new therapeutic perspectives, which now attempt to focus on the suppression of the immunological reactions in the endolymphatic sac. Management can be divided into two large groups: medical and surgical. most being destructive techniques (table 2). Because evaluation of the therapy's e ect is often di cult, guidelines have been developed by the AAO-HNS [1], in order to compare di erent types of therapy for Meniere's disease. They emphasize that a lung uuservatiun periud is needed iII order tu determine the e ects of therapy, since a long symptom - free interval can either be due to the success of treatment or a spontaneous remission (table 3). A general rule for treating Meniere's disease is that prior to a destructive and irreversible surgical procedure, other medical regimens should be tried.
Meniere's Disease
157
Table 3. AAO-HNS guidelines for the evaluation of therapy in Meniere's disease
Frequency of vertigo
Functional level scale
Treatment e cacy is assessed using the following formula: XlY numerical value (rounded to the nearest whole number) where X is the average number of definitive spells per month for the 6-month period 18-24 months after therapy and Y is the average number of definitive spells per month for the 6 months before surgery. The resulting value is used to classifY the patient as follows:
1. Dizziness has no e ect on activities at all. 2. Dizziness does not necessitate changes in plans or activities. 3. Dizziness necessitates some changes in plans. 4. Patient is able to engage in essential activities but constant adjustments are required. 5. Patient is unable to work, drive, take care of a family member, or do most active things. Even essential activities are limited. 6. Patient has been disabled for 1 year or longer and receives compensation.
Class Numerical value A B C D E F
0 1-40 41-80 81-120 120
Secondary treatment initiated because of disability from vertigo
The advantage is a gain of time which brings the patient closer to spontaneous remission. Drug Therapy
Numerous studies on medical therapy for Meniere's disease have been published, however none has been able to describe an e ective treatment regimen unanimously accepted so that the discussion can be closed. It has been noticed in meta-analyses that e cacy for vertigo is almost always between 60 and 80% [21, 104J. For this reason a placebo e ect cannot be ruled out. In the acute stage normally lasting no longer than 3 days, however accompanied by severe attacks of vertigo, agents with sedative and antiemetic properties have been empluyed tu diminish the sensatiun uf vertigu, Amung these are antihistamines such as dimenhydrinate, meclizine and diphenhydramine (table 4). Neuroleptics have been successfully used as an acute therapy, especially in cases of severe attacks because they can be applied in parenteral form. Diazepam, a minor tranquilizer, can also be taken in the active stage of a
Hamann!Arnold
158
Table 4. Drug management of Acute stage Symptomatic
Meni~re's
disease
Dimenhydrinate Dimenhydrinate
150 mg supp. or 62 mg Lv.
'Etiologic' 1st day 2nd day 3rd day
1.000 mg prednisolone 1,000 mg prednisolone 1,000 mg prednisolone
Lv. Lv.
or 100 mg prednisolone 80 mg prednisolone 60 mg prednisolone 40 mg prednisolone 20 mg prednisolone IO mg prednisolone 5 mg prednisolone 2.5 mg prednisolone
orally orally orally orally orally orally orally orally
Chronic stage betahistine
3
12 mg/day
Lv.
2 2 2 150 150 150 150 150 150 150 150
150 mg ranltidine 150 mg ranitidine 150 mg ranitidine mg mg mg mg mg mg mg mg
ranitidine ranitidine ranitidine ranitidine ranitidine ranitidine ranitidine ranitidine
for for for for for for for for
2 2 2 2 2 2 2 2
days days days days days days days days
for months
Meniere attack. These agents however are not suitable for long-term treatment in that they are counteractive to the vestibular compensation processes [41, 85J, and can lead to side e ects in the extrapyramidal motor system. Placebo-controlled double-blind studies only for betahistine and diuretics have been proven e ective long-term treatments [21, 86J. However, these studies - explainable at the time of their publication - have not yet met the criteria of the AAO-HNS gUidelines (table 3). The relatively short observation period of these studies allows for a more critical judgement of the results. Studies have shown the higher e cacy of betahistine to placebo or other substances [63J. However it must be critically noted that the observation period lasted no longer than 3 months and e cacy did not exceed 80% [86] (table 4). Diuretics have been proven to have a dehydrating e ect on endolymphatic hydrops, which is diagnostically applied in the Klockho test [52]. Although placebo-controlled double-blind studies have shown diuretics as an e ective lung-term treatment [53], its use has been limited because uf the accumpanying e ects of changes in electrolyte levels which must be corrected. Convincing evidence for the frequently recommended 'vasoactive agents' does not exist. In addition, it has not been proven that the development of endolymphatic hydrops is of vascular origin.
Meniere's Disease
159
In recent years, glucocorticoids have been frequently used in the treatment of Meniere's disease as a direct consequence of the possible immunopathogenesis of endolymphatic hydrops [108] (table 4). However, controlled longterm studies are needed to prove the clinical e ects of this well-established concept. Drug therapy used in treating Meniere's disease has been described as being relatively e ective in reducing the symptoms of vertigo, however hearing loss improvement has only been reported in the early stages. Only for corticosteroids a direct influence on the course of disease can be expected (table 4), although it has not yet been proven. Currently there is no evidence for all other drugs used. Table 4 shows the management actually used in our hospital. Aminoglycoside Therapy Based on the fact that aminoglycosides are ototoxic, Schuknecht [87] introduced this pharmacological principle for therapeutic purposes in the form of a parenteral application. The intention was to destroy the vestibular hair cells so that they could not be irritated by endolymph flow disorders. When it was made clear that the sensory cells of the a ected ear reacted more sensitively to aminoglycoside than those of the healthy cells, a systematic application through intramuscular injection was attempted. This method was ultimately not accepted since the unwanted bilateral e ects as well as damage to the cochlear hair cells could not be avoided. Lange [12, 55] introduced locally applied aminoglycoside treatment. Initially, gentamicin was applied via a small plastic tube by tympanotomy, later via an ear tube which was placed into the middle ear as close as possible to the windows. In individually varying periods, gentamicin di used through the cochlear windows into the inner ear leading to a local toxic e ect. Although di erent types of dosage regimens exists, a common factor in all is that the prescribed dosage is generally dependent on and determined by objective symptoms of functional loss. Prior to each new application, audiograms are performed for early detection of cochlear damage thereby discontinuing therapy and for spontaneous nystagmus indicating functional loss of the end organ. In 90%, satisfactory reduction in vertigo has been published by several authors, cochlear complications have been shown in 15% [65]. In order to classify local toxic e ects of arninoglycoside, histological investigations have shown that vestibular hair cells are not essentially destroyed as previously assumed. Instead the so-called 'dark cells' presenting the secretory epithielium [75] are damaged. Consequently, the result of this therapy is a decrease in endolymph production thereby positively influencing endolymphatic hydrops. In this manner improved hearing can be explained,
Hamann!Arnold
160
in some cases, observed as a result of an earlier intratympanical application of streptomycin. A reduction of endolymph production leads to a decompression of endolymphatic hydrops resulting in a recovery of cochlear hair cells. The question as to whether streptomycin or gentamicin is more destructive to vestibular hair cells has undoubtedly not been answered. Although streptomycin has been shown in animal experiments to be more destructive to vestibular than cochlear hair cells, the reverse seems to be true in favor of gentamicin in humans [11, 70J. In order to instill ototoxic agents more e ectively into the inner ear, the 'chemical labyrinthectomy' by means of alcohol injection was modified [64], in which streptomycin was locally injected into the perilymphatic space of the lateral semicircular canal [28J. However, a relative high amount of hearing loss is noticeable. Therefore it is doubtful if this method can be expanded. The e cacy of locally applied gentamicin is so high that it reaches the figures of successful surgical destruction [12J. However, since cochlear damage cannot be excluded with certainty, intratympanical application of gentamicin should be performed primarily on patients experiencing severe hearing loss [28J.
Nondestructive Surgery
Transtympanic Ventilation Tubes Since some patients with Meniere's disease reported the disappearance of vertigo with changes in atmospheric pressure, Tumarkin [106] in 1966 attempted to equalize middle ear pressure by inserting a transtympanal tube. His encouraging results were confirmed by a group of British otologists [54] in which they proved in 77% a reduction or temporary relief of vertigo. Contradictory evidence was presented by Hall and Brackmann [38]. Findings in a long-term study by Montandon et al. [66] showed an improvement in vertigo in 82% (23/28), however no e ect on hearing or tinnitus. Independent of this, this observation explains the reduction of vertigo following a simple mastoidectomy [102J since pressure equalization is also gained between middle ear and mastoid. Cochleosacculotomy In 1982, Schuknecht [901 detected in a histopathological study in humans and animals in the later stage of Meniere's disease, following the disappearance of vertigo, persisting fistulas between the endolymphatical space and the peri-
Meniere's Disease
161
lymphatical space. Based on these findings he developed the 'cochleosacculotomy', a surgical procedure in which a permanent fistula between the round window and the sacculus is produced. The advantage of this relatively simple procedure is a remarkable reduction in vertigo, the disadvantage is the increased rate of hearing deterioration. Therefore this technique is mainly recommended for older patients in which the more extensive destructive procedure is not an option and in patients with a pre-existing advanced hearing loss [50J. Saccotomy In 1921, G. Portmann [78J performed experiments which led him to the conviction that overpressure in the endolymphatic sac is involved in the development of Meniere's disease. Consequently, in 1927 [79J he performed the first decompression operation of the endolymphatic sac, the saccotomy. Many studies have reported on good results [5, 82], the e cacy being at 80% in reducing vertigo. Nevertheless, this method is not without risk to hearing, the complication rate for hearing loss is assessed at 15%. For many years, surgery of the endolymphatic sac was the method of choice when all other conservative techniques failed in treating the intolerable symptoms of vertigo in Meniere's disease. The success rate however was seen in another light when Thomsen and Bretlau [102J published a 'double-blind study', a controlled study in which they compared the results of a saccotomy with the results of a single mastoidectomy. Based on these data they could not find a significant therapeutic di erence between either surgical techniques so that they classified the saccotomy as a 'sham operation'. This critical opinion led to lively discussions. Arenberg et a1. [5J dismissed the interpretation of a mastoidectomy not having a specific e ect. By recording early evoked potentials during a mastoidectomy which precedes a saccotomy, changes are registered in the electrocochleographic potentials so that a mastoidectomy was seen as active surgery and not as an unspecific procedure. That is not to say that the observations by Bretlau and co-workers are refuted. The positive e ect, which can be achieved only with a mastoidectomy, can be explained as such, that during a mastoidectomy the middle ear and the mastoid are decompressed, leading to the same phenomenon when a transtympanical ventilation tube is used (see above). These simple sac decompression techniques were refined by inserting a small silicun slice keeping the sac upenlunger, and in order tu prevent adhesiuns by valves which only open with higher pressure [4]. Despite the studies by Thomsen and Bretlau, saccotomies continue to be performed and their positive results to be published [82J. It must be critically noted that most of these studies do not have control groups so that the
Hamann!Arnold
162
benefit of endolymphatic sac surgery cannot be viewed as an approved method. Opposing opinions for a saccotomy in patients with Meniere's disease have been given by honest surgeons who state that locating the sac can be di cult due to a heavy fibrosis to the area (see above). Even if the endolymphatic sac is identified, the lumen is frequently not found and surgery is not applicable.
Destructive Procedures
To free patients from the intolerable symptoms of vertigo, a unilateral dea erentation of the vestibular apparatus either with a labyrinthectomy or a vestibular neurectomy is e ective as a last resort. Labyrinthectomy The purpose of labyrinthectomy is to achieve a unilateral dea erentation of the peripheral receptors by destroying the vestibular sensory apparatus allowing for a central compensation. Labyrinthectomy is primarily performed on patients with existing deafness or severe cochlea dysfunction because of the inevitable loss of cochlear function. The procedure is usually performed by a transtympal approach, is of little discomfort to the patient, requires only a short stationary stay and is cost-e cient [33]. Almost a 100% resolution of vertigo has been reported in the literature [see outlines in 44, 61]. Despite these excellent results, the procedure can be applied only to a limited patient group, due to its related consequence of deafness which must be considered especially in bilateral Meniere's disease. Neurectomy of the Vestibular Nerve The goal of a vestibular neurectomy is also to dea erentate the a ected peripheral vestibular apparatus. This is gained by a selective vestibular nerve section in the inner auditory canal. Both of these mainly used approaches, the suboccipital ( retrosigmoidal [58]) and the transtemporal [37]. have the advantage of maintaining not only the cochlea nerve, but cochlear function as well. The cost and risk of complications is higher, stationary stay longer than with a labyrinthectomy [33], yet all of these can be justified if hearing is preserved. Approximately 90% resolution of vertigo has been reported [see outlines iII 37,58,61,93]. Complications such as cerebral spinal fluid fistulas, facial palsy and meningitis are rarely seen. On the other hand its e cacy is not substantially higher than that of transtympanlcal gentamicin treatment. Therefore neurectomy of the vestibular nerve should only be used when all other therapies have proven
Meniere's Disease
163
unsuccessful. A surgeon's ambition should not be a reason for not choosing a so-called simpler method with similar e cacy such as the application of a drainage tube or locally applied gentamicin treatment. References
2
3
6
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
23
AAO-HNS Guidelines for the evaluation of therapy in Meniere's disease. Otolaryngol Head Neck Surg 1995;113:181-185. Albers FWJ, Casselman JW: 3-DFT-magnetic resonance imaging of the inner ear in Meniere's disease; in Filipo R, Barbara M (eds): Proc 3rd Int Symp Men Dis. Amsterdam. Kugler. 1994, pp 43-46. Altermatt HJ, Gebbers JO, Muller C, Arnold W, Laissue JA: Human endolymphatic sac: Evidence fur a role in inner ear immune defence. ORL 1990;82:143-148. Arenberg IK, Stahle J: Unidirectional ear valve implants: A nondestructive alternative to labyrinthectomy in Meniere's disease. Am J Otol 1981;3:9-10. Arenberg IK, Serkowsky M, Mihalco L, Wu CM, Moorhead JF: Danish sham surgery revisited: Intraoperative electrocochleographic evidence that mastoidectomy/drilling actively changes hydrops; in Filipo R, Barbara M (eds): Proc 3rd Int Symp Men Dis. Amsterdam, Kugler, 1994, pp 533-536. Arenberg IK Lemke C, Shambough GE Jr: Viral theory for Meniere's disease and endolymphatic hydrops: Overview and new therapeutic option for viral labyrinthitis. Ann NY Acad Sci 1997;830: 306-312. Arnold W, Altermatt HJ, GebbersJO: Qualitativer Nachweis von Immunglobulinen immenschlichen Saccus endolymphaticus, Laryng Rhinol Otol 1984:63:464-467, Arnold W, Altermatt HJ: The significance of the endolymphatic sac and its possible role in Meniere's disease. 1\ct" Otol"ryngol (Stockh) 1995(suppl 519):36-42. Arnold W, Niedermeyer HP: Herpes simplex virus antibodies in the perilymph of patients with Meniere's disease. Arch Otolarnygol Head Neck Surg 1997;123:53-56. Arnold W: Personal communication. Bagger-Sjoback 0: E ect of streptomycin and gentamicin on the inner ear. Ann NY Acad Sci 1997;830:120-129. Beck C, Schmidt CL: Ten years of experience with intratympanically applied streptomycin (gentamicin) in the therapy of morbus Meniere. Arch OtolaryngoI1978;21:149-152. Bergstrom T, Edstrom S, Tjellstrom A, Vahlne A: Meniere's disease and antibody reactivity to herpes simplex virus type 1 polypeptides. Am J Otolaryngol 1992;13:259-300. Bernstein JM, Shanahasu TC, Scha er FM: The genetics of hearing loss in Meniere's disease and otosclerosis. Acta Otolaryngol (Stockh) 1996;116:666-671. Birgerson L, Gustavson KH, Stahle J: Familiar Meniere's disease: A genetic investigation. Am J Otolaryngol 1987;8:323-326. Brandt T: Vascular vertigo: in Brandt T (ed): Vertigo. Berlin, Springer, 1991. pp 171-188. Brandt T: Drug and vertigo; in Brandt T (ed): Vertigo. Berlin, Springer, 1991, pp 213-228. Brandt T, Stedclin S, Daro RB: Therapy for benign paroxysmal positioning vertigo revisited. Neurology 1994;44:796-800. Brown DH, McClure JA, Down-Zapolski Z: The membrane rupture theory of Meniere's disease: Is it valid? Laryngoscope 1988;98:599-60 I. Cawthorne T, Hewlett AB: Meniere's disease. Proc R Soc Med 1954;47:663-670. Claes J, van de Heyning PH: Medical treatment of Meniere's disease: A review of literature. Acta Otolaryngol (Stockh) 1997(suppI526):37-42. Coletti V, Carner M, Fiorino FG, Sharbati A: Electron microscopic evaluation of the vestibular nerve in patients su ering from Meniere's disease and vascular cross-compression syndrome; in Filipo R, Barbara M (eds): Proc 3rd Int Symp Men Dis. Amsterdam, Kugler, 1994, pp 221-225. Committee on Hearing and Equilibrium: Guidelines for the diagnosis and evaluation of therapy in Meniere's disease. Otolaryngol Head Neck Surg 1995;113:181-185.
Hamann!Arnold
164
24 25 26 27 28 29
30 31 32 33 34 35 36 37
38 39 40 41
42 43 44
45 46 47 48 49 50 51 52
Dix MR, Hallpike CS: The pathology symptomatology and diagnosis of certain common disorders of the vestibular system. Ann Otol (St Louis) 1952;61:987. Dohlmann GF, Johnson VVH: Experiments on the mechanism of Meni~re's attacks. Proc Can Otol Soc 1965; 19:73. Duke "'iW: Meni~re's syndrome caused by allergy. JAMA 1923;81:2179-2181. Duplay S: Des maladies de l'oreille interne. Arch Gen Med 1872;1:711-722. Ecke U, Begall K. Amedee RG, Norris CH, Mann Wj: Selective chemical vestibulectomy. ORL 1997;59:209-214. Eggermont J]: Summating potentials in electrocochleography: Relation to hearing disorders; in Ruben RJ, Elberling C, Salomon G (eds): Electrocochleography, Baltimore, University Press, 1967, pp 67-88. Eggermont JJ. Odenthal OW: Methods in electrocochleography. Acta Otolaryngol (Stockh) 1974(suppI316):17-24. Flourens P: Recherches experimentales sur les proprietes du systeme nerveux, ed. 2, Paris 1842. Fowler EP, Zeckel A: Psychosomatic aspects of Meniere's disease. JAMA 1952;148:1265-1268. Gacek RR, Gacek MR: Comparison of labyrinthectomy and vestibular neurectomy in the control of vertigo, Laryngoscope 1996; 106:225-230, Gloddek B, Harris JP: Role of lymphokines in the immune response of the inner ear. Acta Otolaryngal (Stockh) 1989;108:68-75. GoodhiJI V: Leaking labyrinth lesions, deafness, tinnitus and dizziness. Ann Otol 1981;90:99-105. Guild S: The circulation of the endolymph. Am J Anat 1927;39:57-81. Haid CT, WolfSR, Gjuvic M, Wigand ME: Surgical treatment of Meniere's disease via the enlarged middle cranial fossa approach: Long-term follow-up; in Filipa R. Barbara M (eds): Proc 3rd lnt Symp Men Dis. Amsterdam, Kugler, 1994, pp 547-556. Hall C, Brackman DE: Eustachian tube blockage and Meniere's disease. Arch Otolaryng 1997; 103: 355-360. Hallpike CS, Cairns H: Observations on the pathology of Meniere's syndrome. J Laryngol Otol 1938;53:625-655. Hamann KF: Training gegen Schwindel. Berlin, Springer, 1987, Hamann KF, Lannou j: Dynamic characteristics of vestibular nuclear neurons responses to vestibular and optokinetic stimulation during vestibular compensation in the rat. Acta Otolaryngol (Stockh) 1988(suppl 445):1-18. Harris JP: Immunology of the inner ear: Evidence of local antibody production. Ann Otol Rhinal Laryngol 1984 ;93: 152-162. Harris JP, Fukuda S, Keithley EM: Spiral modiolar veins: Its importance in inner ear inflammation. Acta Otolaryngol (Stockh) 1990;110:356-357. Hausler R, Vibert 0: Treatment of unilateral delayed endolymphatic hydrops by transmeatal mechanical or argon laser labyrinthectomy: in Filipo R, Barbara M (eds): Proc 3rd Int Symp Men Dis, Amsterdam, Kugler, 1994, pp 519-523. ltard JMG: TraM des maladies de l'oreille et de l'audition. Paris 1821. Jahnke K: Zur Pathogenese der akuten Symptome des Morbus Meniere. Laryngol Rhinal Otol 1977;56: 402-406. Kiang NYS: An auditory physiologist's view of Meniere's syndrome; in Nadal JB (ed): 2nd Int Symp Men Dis. Amsterdam, Kugler, 1989, pp 13-24. Kimura RS, Schuknecht HF: Membranous hydrops in the inner ear of the guinea pig after ablation of the endolymphatic sac. Pract Otorhinolaryngol 1965;27:343-354. Kingma H, Wuyts FL, Bonmans L: Clinical testing of the statolith system in patients with Meniere's disease. Acta Otolaryngol (Stockh) 1997 (suppl 526) :24-26. Kinney WC, Nahpa N, Hughes GB, Kinney SE: Cochleosacculotomy for the treatment of Meniere's disease in the elderly patient. Laryngoscope 1995;105:934-937. Kitahara M, Futaki T, Nakano K: Ethnic aspects of Meniere's disease, Equilib Res 1971 (suppl!): 104-110. KJockho 1. Lindblom U: Endolymphatic hydrops revealed by glycerol test: Preliminary report. Acta Otolaryngol (Stockh) 1966;61:459-462.
Meni~re 's
Disease
165
53 54 55 56 57
Klockho I. Lindblom U, Stahle J: Diuretic treatment of Meniere's disease: Long-term results with chlorthalidone. Arch Otolaryngol 1974; 100:262-265. Lall M: Meniere's disease and the grommet (a survey of its therapeutic e ects). J Laryngol Otol 1969;83:787-791. Lange G: Die intratympanale Behandlung des Morbus Meniere mit ototoxischen Antibiotika. Laryngol Rhinol Otol 1977;56:409-414. Lawrence M, McCabe BF: Inner ear mechanics and deafness, special considerations of Meniere's syndrome. JAMA 1959;171:1927-1932. Lermoyez M: Le vertige, qui fait entendre (angiospasme labyrinthique). Presse Med 1919;27:13,
58
59 60 61 62 63 64 65 66 67 68 69 70
71 72 73 74 75 76
77 78
Magnan JP, Chays AA. Locatelli PO, Ulmer EG, Scharf B: Hearing results in Meniere's disease following retrosigmoid vestibular neurotomy. Proc 3rd Int Symp Men Dis. Amsterdam, Kugler, 1994, pp 565-569. McCabe BF: Autoimmune sensorineural hearing loss. Ann Otol Rhinol LaryngoI1979;88:585-589. McCiure JA, Copp JC, Down-Zapolski Z: Recovery nystagmus in Meniere's disease. Laryngoscope 1981 ;91: 1727-1737. McKee GJ, Kerr AG, Toner JG, Smyth GDL: Surgical control of vertigo in Meniere's clisease, Clin OtolaryngoI1991;16:216-227. Meniere P: Memoire sur des lesions de I'oreille interne donnant lieu a des symtOmes de congestion cerebrale apoplectiforme. Gaz Med Paris 1861 ;38:597-601. Meyer ED: Zur Behandlung des Morbus Meniere mit Betahistidinmesilat (Aequamen). Doppelblindstudie gegen Plazebo (crossover). Laryngol Rhinol Otol 1985;64:269-272. Mollison WV: Surgical treatment of vertigo by opening the external semicircular canal and injecting alcohol. Acta Otolaryngol (Stockh) 1939;27:222-229. Monsell EM, Cass SP, Rybak LP: Therapeutic use of aminoglycosides in Meniere's disease. Otolaryngol Clin North Am 1993;26:734-746. Montadon P, Guillemin P, Hausler R: Prevention of vertigo in Meniere's syndrome by means of transtympanic ventilation tubes. ORL 1988;50:377-381. Morgenstern C: Pathogenesis of experimental endolymphatic hydrops. Acta Otolaryngol 1983; (suppl 406) ;56-58, Nadol JB, Weiss AD, Parker SW: Vertigo of delayed onset after sudden deafness, Ann Otol Rhinol Laryngol 1975;84:841-846. Naito T: Clinical and pathological studies on Meniere's disease. Proc 60th Annu Meet Jpn ORL Soc, 1959, pp 1-65. Norris CH, Aubert A, Shea JJ: Comparison of cochleotoxicity and vestibulotoxicity of streptomycin and gentimicin; in Filipo R. Barbara M (eds): Proc 3rd lnt Symp Men Dis. Amsterdam, Kugier, 1994, pp 485-490, Owen Black F: Vestibular function assessment in patients with Meniere's disease: The vestibulospinal system. Laryngoscope 1982;92:1419-1436. Owen Black F, E ron MZ, Burns OS: Diagnosis and management of drop attacks of vestibular origin: Tumarkin's otolithic crisis. Otolaryngol Head Neck Surg 1982;90:256-262. Paparella MM: Pathogenesis of Menieres disease and Meniere's syndrome. Acta Otolaryngol 1984 (suppl 406):10-25. Paparella MM: The natural course of Meniere's disease; in Filipo R. Barbara M (eds): Proc 3rd lnt Symp Men Dis. Amsterdam, Kugler, 1994, pp 9-20. Pender OJ: Gentamicin tympanoclisis. e ects on the vestibular secretory cells. Am J Otol 1985;6: 358-367. Perez N, Garcia-Tapia R: Distortion product otoacoustic emission testing in patients with Meniere's disease; in Filipo R. Barbara M (eds): Proc 3rd Int Symp Men Dis. Amsterdam, Kugler, 1991, pp 323-327. Pfaltz CR. Matefi L: Meniere's disease - or sydrome? A critical review of diagnose criteria; in Vosteen KH (ed): Meniere's Disease. Stuttgart, Thieme, 1981, pp 1-10. Portmann, G: Recherches sur la physiologie du sac et du canal endolymphatiques: valeur fonctionelle de I'organe endolymphatique des Selaciens. C R Soc Bioi 1921;85:1070-1077.
Hamann!Arnold
166
79 80 81 82 83 84
85
86 87 88 89 90 91 92 93
94 95 96 97 98 99 100 101 102 103 104 105 106 107
Portmann G: The saccus endolymphaticus and an operation for draining the same for the relief of vertigo. J LaryngoI1927;42:809-817. Feldmann H (ed): Proc III Int Tinnitus Semin. Karlsruhe, Harsch, 1987. Reich GE, Vernon JA (eds): Proc V Tinnitus Semin. Portland, Am Tinnitus Assoc 1996. Quaranta A, Marini F, Sallustio V: Long-term outcome of Meniere's disease: Endolymphatic shunt versus natural history. Audiol Neurootol 1998;3:54-60. Rask-Andersen H, Stahle J: Immunodefence of the inner ear. Acta Otolaryngol (Stockh) 1980;89: 283-294. Rask-Andersen H, Danckwardt-LiIJiestrom N, Friberg U, Kinnefors A: Lymphocyte-induced autoaggressive cytotoxicity in the endolymphatic sac in the active Meniere's disease: A TEM histopathological investigation; in Magi G, Veldman JE, Kawauchi H (eds): Immunobiology in Otorhinolaryngology. Amsterdam, Kugler, 1994, pp 343-350. Schaefer KP, Wehner H: Zur pharmakologischen Beeinflussung zentralnervoser Kompensationsvorgange nach einseitiger Labyrinthaussschaltung durch Krampfgifte und andere erregende Substanzen. Naunyn Schmiedebergs Arch Pharmacal 1966;254: 1-17. Schmidt JT, Huizing EH: The clinical drug trial in Meniere's disease with emphasis on the e ect of beta-histine. Acta Otolaryngol (Stockh) 1992(suppl 497):1-187. Schuknecht HF: Ablation therapy in the management of Meniere's disease Acta Otolaryngol (Stockh) 1957(suppl 132):1-42. Schuknecht HF: Pathology of the Ear. Cambridge, Harvard Univ Press, 1974. Schuknecht HF, Kitamura K: Vestibular neuritis. Ann Otol Rhinol Laryngol 1981(suppl 78):1-19. Schuknecht HF:Cochleosacculotomy for Meniere's disease: Theory, technique and results. Laryngoscope 1982;92:853-858. Schuknecht HF, Igarashi M: Pathophysiology of Meniere's disease; in Pfaltz CR (ed): Controversial Aspects of Meniere's Disease. New York, Thieme, 1986, pp 46-54. Schuknecht HF: Pathology of the Ear, ed 2. Philadelphia, Lea & Febiger, 1993. Silverstein H, Ho er ME, Rosenberg S[, Nichols ML, Willcox TO, Rowan PT: Posterior fossa vestibular neurectomy: 14 years' experience; in Filipo R, Barbara M (eds): Proc 3rd Int Symp Men Dis. Amsterdam, Kugler, 1994, pp 537-542. Smith CA, Lowry OH, Wu ML: The electrolytes of the labyrinthine fluids. Laryngoscope 1954;64: 141-153. Spoendlin H, Balk V, Bock G: Multicenter evaluation of the temporal bones obtained from a patient with suspected Meniere's disease. Acta Otolaryngol (Stockh) 1992(suppl 499):1-21. Stahle J, Stahle C, Arenberg K: Incidence of Meniere's disease. Arch Otolaryngol 1978;104:99102. Stahle J, Arenberg [K, Goldman G: Staging Meniere's disease: Description of a vertigo disability profile. Am J Otol [981;2:357-364. Strohm M: Trauma of the middle ear: Clinical findings. postmortem observations and results of experimental studies. Adv Otol Rhinol LaryngoI1986;35:1-254. Swart J, Schuknecht HF: Long-term e ects of destruction of the endolymphatic sac in a primate species. Laryngoscope 1988;98:1183-1189. Tasaki I, f7ernandez C: Modification of the cochlear microphonics and action potentials by KCl solution and by direct current. J Neurophysiol 1952;15:497-512. Tasaki I, Davis H, Eldredge DH: Exploration of cochlear potentials in guinea pig with a microelectrode. J Acoust Soc Am 1954 ;26:765-773. Thomsen J, Bretlau P, Tos M: Placebo e ect in surgery for Meniere's disease: A double-blind study on endolymphatic sac surgery. Arch Otolaryngol 1981;107:271-278. Tonndorf J: Stereociliary dysfunction, a case of sensory hearing loss, recruitment, poor speech discrimination and tinnitus. Acta Otolaryngol (Stockh) 1981;91:169-179. Torok N: Old and new in Meniere's disease. Laryngoscope 1977;87:1870-1877. Tumarkin A: Otolithic catastrophe, a new syndrome. Br Med J 1936;ii:175-177. Tumarkin A: Thoughts on the treatment of labyrinthopathy. J Laryagol Otol 1966;80:1041-1053. van Deelen GW, Huizing EH: Use of a diuretic (Dyzide) in the treatment of Meniere's disease. A double-blind cross-over placebo-controlled study. ORL 1986;48:287-292.
Meniere's Disease
167
108 109 110 III 112 113 114 115 116 117
Vischer M, Arnold W: Kortisonsensible Innenohrschwerhorigkeit. ORL Nova 1991;1:75-79. Watanabe J: Meniere's disease with special emphasis on epidemiology, diagnosis and prognosis. ORL 1980;42:20--45. Watanabe I: Vestibular aspects of Meniere's disease; in Filipa R, Barbara M (eds): Proc 3rd Tnt Symp Men Dis. Amsterdam, Kugler, 1994, PP 337-345. Weinaug P: Die Spontanrernission beirn HOrsturz. HNO 1984;32:346-351. Welling DB, Parnes LS, O'Brian B, Bahaletz LO, Brackmann DE, Hinojosa R: Particulate matter in the posterior semicircular canal. Laryngoscope 1997;107:90-94. Wexler M, Crary WG: Meniere's disease: The psychosomatic hypothesis. Am j Otol 1986;7:93-96. Wilbrand HF, Stahle j, Rask-Andersen H: Tomography in Meniere's disease: Why and how? Adv Otorhinolaryngol 1978;24:71-93. Williams LL, Lowery NW, Shannon BT: Evidence of persistent viral infection in Meniere's disease. Arch Otolaryngol Head Neck Surg 1987;113:397-400. Wladislavosky-Waserman P, Facer GW, Moleri B, Kurland LT: Meniere's disease: A 36-year epidemiologic and clinical study in Rochester. Laryngoscope 1984;94:1098-1102. Yamakawa K: Ober die pathologische Veranderung bei einem Meniere-Kranken. j Otolaryngol Soc jpn 1938;4:2310-2312.
K.-F. Hamann, Department of Otorhinolaryngology, Head and Neck Surgery, Technical University of Munich, Ismaningerstrasse 22, D-S 1675 MLmich (Germany) Tel. 49 89 4140-2370, Fax 49 89 4140-4853
Hamann!Arnold
168
Buttner U. (cd): Vestibular Dysfunction and Its Therapy. Adv Otorhinolaryngol. Basel, Karger. 1999, vol 55, pp 169-194
Benign Paroxysmal Positioning Vertigo Thomas Brandt Department of Neurology, Ludwig-Maximilians University, Munich, Germany
Benign paroxysmal positioning vertigo (BPPV; also known as positional vertigo) was initially defined by Barany in 1921. The term itself was coined by Dix and Hallpike [1952]. Lanska and Remler [1997] describe in detail the history of BPPV, its original description, the proper eponymic designation for the provocative positioning test, and the steps leading to our current understanding of its pathophysiology. BPPV is the most common cause of vertigo, particularly in the elderly. By age 70, about 30% of all elderly subjects have experienced BPPV at least once. This condition is characterized by brief attacks of rotatory vertigo and concomitant positioning rotatory-linear nystagmus which are elicited by rapid changes in head position relative to gravity. BPPV is a mechanical disorder of the inner ear in which the precipitating positioning of the head causes an abnormal stimulation, usually of the posterior semicircular canal (p-BPPV) of the undermost ear, less frequently of the horizontal (h-BPPV) or the anterior semicircular canal (a-BPPV). Schuknecht [1969] and Schuknecht and Ruby [1973J hypothesized that heavy debris settle on the cupula (cupulolithiasis) of the canal, transforming it from a transducer of angular acceleration into a transducer of linear acceleration. It is now generally accepted, however, that the debris float freely within the endolymph of the canal ('canalolithiasis') [Parnes and McClure, 1991; Epley, 1992; Brandt and Steddin, 1992]. The debris - possibly particles detached from the otoliths - congeal to form a free-floating clot (plug). Since the clot is heavier than the endolymph, it will always gravitate to the most dependent part of the canal during changes in head position which alter the angle of the cupular plane relative to gravity. Analogous to a plunger, the clot induces bidirectional (push or pull) forces on the cupula, thereby triggering the BPPV attack. Canalolithlasis explains all the features of BPPV: latency, short duration, fatigability (diminution with repeated positioning), changes in direction of nystagmus with changes
in head position, and the e cacy of physical therapy [Brandt and Steddin, 1993; Baloh et al., 1993; Brandt et a\., 1994]. In 1980, Brandt and Daro proposed the first e ective physical therapy (positioning exercises) for BPPV Based on the assumption that cupulolithiasis was the underlying mechanism, the exercises were a sequence of rapid lateral head/trunk tilts, repeated serially to promote loosening and, ultimately, dispersion of the debris toward the utricular cavity. In 1988, Semont et a\. introduced a single liberatory maneuver, and Epley promoted a variation in 1992, which Herdman et al. [1993] later modified. If performed properly, all three forms of therapy (Brandt-Daro exercises and Semont and Epley's liberatory maneuvers) are e ective in BPPV patients [Herdman, 1990; Herdman et aI., 1993]. The e cacy of physical therapy makes selective surgical destructions such as transection of the posterior nerve [Gacek, 1978] or nonampullary plugging of the posterior semicircular canal [Pace-Balzan and Rutka, 19911 largely unnecessary. About 5-10% of BPPV patients su er from horizontal canalolithiasis (h-BPPV) [McClure, 1985]. h-BPPV is elicited when the head of the supine patient is turned from side to side, around the longitudinal z-axis. Combinations are possible, and transitions from p-BPPV to h-BPPV occur, if the clot moves from one to the other semicircular canal. Transitions from canalolithiasis to cupulolithiasis in h-BPPV patients have been described [Steddin and Brandt, 19961. Most of the cases appear to be idiopathic (degenerative?), their incidence increasing with advancing age. Prolonged bedrest also facilitates their occurrence. Other cases arise due to trauma, vestibular neuritis, or inner ear infections. The diagnosis of typical BPPV is simple and safe: the patient must have the usual history and exhibit positioning nystagmus toward the causative, undermost ear. Diagnosis is less easy in rare cases, for example, in patients with horizontal semicircular canal cupulolithiasis (p. 188) who exhibit positional nystagmus beating toward the uppermost ear for several minutes. Di erential diagnosis includes di erent forms of central vestibular vertigo or nystagmus, vestibular paroxysmia, perilymph fistula, drug or alcohol intoxication, vertebrobasilar ischemia, Meniere's disease, and psychogenic vertigo. The following is largely adopted from a more detailed presentation in the book 'Vertigo: Its Multisensory Syndromes' by Th. Brandt (2nd ed.).
The Clinical Syndrome
Patients with typical BPPV report attacks of rotatory vertigo, postural imbalance, and sometimes nausea preCipitated by the following maneuvers:
l3randt
170
(a) sitting up from a supine position (particularly after awaking in the morning); (b) when first lying down in bed; (c) turning over in bed from one side to the other; (d) extending the neck (head) to look up or get something from above, and (e) flexing the neck (head) when bending over. The occurrence of BPPV in the supine position is very disturbing and makes patients afraid of falling backward, an almost unique complaint. In the upright position, vertigo attacks produced by changes in head position are incapacitating and can be dangerous, for example when a su erer is looking up at the ceiling while standing on a ladder. In such a situation, BPPV can cause a catastrophic fall. Sometimes the 'probable' diagnosis of BPPV is simply based on the typical patient history, because the condition has spontaneously resolved by the time of examination in accordance with its usually benign course. However, there is no absolute reliability of a diagnosis based on history, and some patients may describe their vertigo in a rather atypical way [None, 1995]. Most patients are aware that tilting or rotating of the head toward the right, the left, or in both directions will induce the attacks. Since as a rule the first positioning maneuver triggers the strongest attack of vertigo and the most obvious positioning nystagmus, the first examination of a patient with a history of BPPV should always be a positioning maneuver toward the side of the posterior semicircular canal thought to be a ected. This initial step is all the more important as repeated positioning maneuvers cause fatigue of the induced vertigo and nystagmus, thus making diagnostic evaluations more di cult and uncertain. The following rules have proven sound for examining patients with BPPY: (1) Positioning testing should be done first during the physical examination, and the initial positioning maneuver should be directed toward the ear assumed to be a ected, because vertigo and nystagmus will fatigue during repeated maneuvers. (2) Frenzel's lenses should be used whenever possible, to avoid partial suppression of the positioning nystagmus by fixation. (3) Patients should be instructed before the positioning maneuver to ensure better cooperation. They should also be asked not to shut their eyes when vertigo begins. (4) Vertigo and nystagmus are maximal if the patient is positioned rapidly and the head is abruptly halted at its final position. We usually perform the lateral head-trunk tilt with the patient in a sitting position on a couch (fig. 1). Others prefer the so-called Dix-Hallpike maneuver [Dix-Hallpike, 1952], during which the patient is tilted backward until the head is both turned and hanging (fig. 1). To confirm a suspected canalolithiasis of the posterior semicircular canal of the right labyrinth, the head of the sitting patient is rotated 45° to the left. Then the observer tilts the passive patient quickly to the right side. After a
Benign Paroxysmal Positioning Vertigo
171
Fig 1. Precipitating positioning maneuvers for typical posterior semicircular canal BPPV: lateral head-trunk tilt toward the a ected ear.
Brandt
172
latency of seconds, rotatory vertigo and positioning nystagmus occur in a crescendo/descrescendo mode. The horizontal rotatory nystagmus beats toward the undermost ear. After vertigo and nystagmus cease, the patient is again quickly returned to the initial upright position. In most cases this will result in a less violent and shorter vertigo and nystagmus toward the opposite direction. To confirm a suspected canalolithiasis of the posterior semicircular canal of the left labyrinth, the head of the sitting patient is turned 45° to the right and the tilting maneuver is performed to the left for the first time. Thereafter, one should always check for canalolithiasis of the horizontal semicircular canal (p. 187), since combinations of both conditions may manifest simultaneously in the same patient. Finally, one should check for rare anterior canal BPPV. Observation of the positioning nystagmus provides the definite diagnostic criteria for typical p-BPPV. They include: (1) Latency: vertigo and nystagmus begin 1 s or more after the head is tilted toward the a ected ear and increase in severity to a maximum. (2) Duration less than 40 s: nystagmus gradually reduces after 10-40 s and ultimately abates even when the precipitating head position is maintained. (3) Linear-rotatory nystagmus: the nystagmus is best seen with Frenzel's glasses (for example, lenses 16 dpt), which prevents suppression by fixation. The nystagmus is linear-rotatory, with the fast phase beating toward the undermost ear or upward when gaze is directed to the uppermost ear. (4) Reversal: when the patient returns to the seated position, the vertigo and nystagmus may recur less violently in the opposite direction. (5) FatiguabjJjty: constant repetition of this maneuver will result in everlessening symptoms. These five criteria are crucial for further discussion of the confusing literature on the mechanism of BPPV. They provide the major arguments to prove or disprove any hypothetical explanation of cupulolithiasis or canalolithiasis as the causative factor.
Natural Course The natural history of BPPV is considered benign because it resolves spontaneously within weeks or months in most patients. However, in about 20-30% of the patients the condition persists when untreated, and it recurs in another 30% after variable periods for years. Di erential Diagnosis The most important di erential diagnosis is that of central vestibular positional nystagmus. It should be suspected in all cases of positional nystagmus without concomitant subjective vertigo, although some lesions around the fourth ventricle may also induce very violent vertigo and nausea. Central
Benign Paroxysmal Positioning Vertigo
173
positional nystagmus does not subside with maintenance of the head in the precipitating position, it may not exhibit fatigability with repetitive stimulation, it may change its direction when di erent head positions are assumed, or it may occur as downbeat nystagmus only in the head-hanging position. Frequently, additional neurological (in particular oculomotor) signs point toward a central cause of the symptoms. BPPV must also be di erentiated from positional nystagmus in Meniere's disease, perilymph fistulas, drug andJor alcohol intoxication, neurovascular cross-compression (vestibular paroxysmia), and rare conditions such as WaldenstrOm's macroglobulinemia.
Pathomechanism
Typical posterior canal BPPV is caused by canalolithiasis, a free-floating clot within the endolymph of the posterior canal. There was, however, a longcontinuing controversy in the quite extensive literature, which arose over the questions: Is cupulolithiasis compatible with the clinical features of BPPV? Is canalolithiasis compatible with the clinical features of BPPV?
The Traditional View of Cupulolithiasis The early assumption by Barany [Barany, 1921; Dix and Hallpike, 19521 that the underlying lesion must be situated in the vestibular end-organ and must involve the otolith was later supported by Schuknecht and Ruby [Schuknecht, 1962, 1969; Schuknecht and Ruby, 1973], who postulated a mechanical pathogenesis, which they termed ·cupulolithiasis'. They found basophilic deposits on the cupula of the causative posterior semicircular canal in individual patients who manifested unilateral BPPV prior to death from unrelated disease. These deposits exceeded the size of those found in more than 30% of temporal bones in a control population. They argued that inorganic particles, detached from the otoconiallayer by spontaneous degeneration or head trauma, gravitate to and settle on the cupula of the posterior semicircular canal, which is situated directly inferior to the utricle when the head is upright. The posterior semicircular canal thus serves as a receptacle for the detached sediment. In fact, otoconia are easily dislodged by linear accelerations or centrifuging in animals [Hasegawa, 1933; Igarashi and Nagaba, 1968J. Otoconial debris become displaced in old age [Johnson and Hawkins, 1972] and lodge either in the posterior semicircular canals [Vyslonzil, 1963] or in the cochlea in cochleosaccuJar degeneration [Gussen, 1980J. Moriarty et al. [1992] identified basophilic granular deposits on 22% of the cupulae of 1,038 canals of 566 examined individual postmortem temporal bones. They determined a higher
Brandt
174
B
Fjg 2. A, B Cupulolithiasis: schematic drawing of the cupula in the ampulla of the posterior semicircular canal carrying otoconia! debris. If this were the pathology causing BPPV, then the static lateral head position (B) rather than the positioning maneuver should be the precipitating factor, irrespective of which side the debris lodge on. Clinically, however, typical BPPV is a positioning vertigo, and cupulolithiasis of this type cannot explain aU clinical features. [From Brandt and Steddin, 1993.J
incidence for the posterior than for the horizontal or anterior semicircular canals. Naganuma et al. [1996J found an even higher percentage of basophilic deposits (horizontal canal: 41 %, posterior canal: 37%, anterior canal: 26%), which increased with increasing age. The cupula normally has the same specific gravity as the endolymph and is a transducer of angular acceleration only. When heavily loaded, it should theoretically become sensitive to changes in head position relative to the gravitational vector (buoyancy hypothesis). The buoyancy mechanism is used to explain some forms of positional vertigo/nystagmus which arise after the ingestion of compounds with di ering specific gravities, such as alcohol [Money et a\., 1974], glycerol [Rietz et a\., 1987], or 'heavy water' [Money and Myles, 1974]. The common view was that BPPV simply reflects transformation of the a ected cupula from a transducer of angular acceleration to one of linear acceleration and (abnormal) angular acceleration, secondary to the acquired specific gravity di erential between the cupula and endolymph [Gacek, 1984; Schuknecht, 1969; Rietz et aI., 1987J. Therefore, in cupulolithiasis it should be irrelevant on which side of the cupula the heavy debris become attached (fig. 2). The traditional view of cupulolithiasis and the buoyancy mechanism must be incorrect for several reasons. BPPV is a positioning rather than a positional
Benign Paroxysmal Positioning Vertigo
175
vertigo/nystagmus, because it is induced only by rapid changes in head position, with the paroxysmal nystagmus being compatible with the cupulogram of an ampullofugal stimulation of the posterior semicircular canal. If cupulolithlasis were valid, then an ongoing positional vertigo/nystagmus would be expected, as occurs in positional alcohol nystagmus [Money et aI., 1974]. The mechanism of cupulolithiasis may account for the subtle static (persistent) positional nystagmus, which Baloh et al. [1987] observed in 41 % of their patients with BPPV There was, however, no consistent relation between the side of the lesion and the static positional nystagmus. This lack of directional correspondence makes interpretation di cult. Arguments for CanaJolithiasis There is histological proof that inorganic 'heavy particles' detached from the otoconiallayer (by degeneration or head trauma) gravitate into the posterior semicircular canal. This is moreover supported by large series of patients in whom the common clinical finding indicates that the following conditions figure in the etiology of BPPV: head (labyrinthine) trauma, viral neural labyrinthitis, vertebrobasilar ischemia, postsurgery (ear and general), prolonged bedrest due to unrelated diseases [Gyo, 1988], and aging [Baloh et aI., 1987; Katsarkas and Kirkham, 1978]. A clot formed by specific heavy material floating in the ampullofugal branch of the posterior canal would be compatible with all five clinical criteria mentioned above. The diameter of the semicircular canal (0.32 mm in humans; Curthoys and Oman [1987]) is about one-fifth to one-seventh of the ampulla. A clot approximately this size in diameter would act as a plunger on the endolymph and the cupula, if the slightly turned head is positioned from an upright to a precipitating lateral position (fig. 3) as soon as the angle between the canal's plane and the gravity vector altered. The clot produces pressure or suction in the canal, thereby deflecting the cupula and eliciting a BPPV attack. This mechanism would be compatible [Brandt and Steddin, 1993] with (1) a latency of a few seconds (time needed for the clot-induced flow mechanism to develop by gravitational force); (2) the ine ectiveness of a very slow positioning maneuver (then the clot would slowly gravitate along the undermost wall of the canal without a ecting the cupula); (3) the limited duration of the positioning vertigo/nystagmus (cupula deflection due to elastic restoring force ends when the heavy clot reaches its lowest position in the canal with respect to the earth surface); (4) the fatigability with repetitive provocation (explained by dispersion of single particles from the clot, which decreases the plunger e ect); (5) the reactivation of the vertigo after prolonged bedrest (the result of a new clot being formed by the particles, and (6) the direction of the nystagmus during the positioning maneuvers as explained below.
Brandt
176
A
B
C
Fig 3. 'Canalolithiasis'; schematic representation of a free-floating' heavy clot' of otoconial debris acting like a plunger on endolymph and cupula of the posterior semicircular canal, the proposed mechanism for BPPV. In the normal upright position (A) the debris rest at the base of the cupula without any noticeable e ect. If the head is turned and rapidly positioned to the side in the plane of the posterior semicircular canal. the clot. thanks to its greater specific weight, gravitates downward (B) and, together with endolymph flow, deflects the cupula in an ampullofugal direction. When the clot has gravitated to the lowest curvature of the posterior canal. vertigo and nystagmus subside because the cupula assumes its normal resting position (C). If the patient returns to the seated position, a similar e ect could explain the reversed direction of nystagmus and vertigo (for explanation of fatiguability due to repeated provocation and physical therapy, p. 176). [From Brandt and Steddin 1993.]
Attempts were made earlier to attribute BPPV to floating particles in the semicircular canal [Hall et aI., 1979; Epley, 1980; Pagnini et al., 1989; McClure, 1988]. We believe there are now convincing arguments that canalolithiasis and a clot-induced endolymph flow mechanism are the significant causative factors [Brandt and Steddin, 1993]. Despite ongoing discussions about possible vestibular habituation e ects [Steenerson and Cronin, 1996; Smouha, 1997], canalolithiasis is the only mechanism that explains the success of the highly e ective physical therapy by positioning maneuvers, as proposed by Brandt and Daro [1980], Semont et a1. [1988], Epley [1992], or Lempert et a1. [1996], who demonstrated the e cacy of canal-clearing maneuvers by performing backward rotation of the posterior canal during the use of a flight simulator. As figure 4 shows, we believe that the floating clot of particles can be sluiced down by both of the described positioning maneuvers via the upper ampullofugal branch of the posterior canal into other labyrinthine recesses where they are no longer causative. The mechanism demonstrated in figure 4 coincides with the hitherto unexplained observation [Pace-Balzan and Rutka, 1991; Semont et al., 1988; Hausler and Pampurik, 1989J that after a positioning
Benign Paroxysmal Positioning Vertigo
177
c Fjg 4. Schematic drawing of the clot of otoconial debris dispersed and sluiced by positioning maneuvers from the normal (A) to the challenging (B) position and to the opposite side (C). Depending on the change of the head position relative to the gravitational vector, the clot settles to the lowermost part of the canal, and after transition from position B to C, leaves the canal in order to enter other labyrinthine recesses where it no longer causes vertigo attacks. This scheme demonstrates why physical therapy with positioning maneuvers is e ective when the clot or the debris leave the a ected semicircular canal. The direction of induced nystagmus in C indirectly proves that canalolithiasis rather than cupulolithiasis is the significant mechanism. Only with a free-floating clot is the direction of positioning nystagmus in C the same as in B. [From Brandt and Steddin, 1993.1
maneuver from the position B (with the a ected ear undermost) to C (with the a ected ear uppermost), the induced nystagmus still rotates toward the uppermost ear. Cupulolithiasis predicts an ampullopetal deflection of the cupula, which results in nystagmus toward the undermost ear. The unexpected direction, however, can be easily explained by the clot-induced endolymph flow mechanism, which acts in the same direction in the two di erent positioning maneuvers in figures 4B and C. We interpret the direction of the nystagmus thus induced by the positioning maneuver as indirect proof that canalolithiasis is valid. Free-floating endolymphatic particles were first found intraoperatively during posterior semicircular canal occlusion by Parnes and McClure [1992J.
l3randt
178
In a prospective study on 26 patients undergoing the posterior canal occlusion procedure and 73 patients undergoing labyrinthine surgery for vestibular Schwannoma or labyrinthectomy, particulate matter was observed in 8 of 26 patients with BPPV; no particles were observed in any of the 73 patients [Welling et al., 1997J. The possibility of a combination of canalolithiasis with cupulolithiasis has been demonstrated for h-BPPV [Steddin and Brandt, 1996]. There are positions of the head in which the free-floating clot should settle on the cupula, and subsequently after initial canalolithiasis, cupulolithiasis should occur, according to the buoyancy mechanism. Do some of the patients with intractable BPPV have precipitate settled on the cupula? If this is the case, the nystagmus pattern should be di erent from that in treatable cases. On the basis of our observations and the assumption of a free-floating heavy clot (leading to canalolithiasis and cupulolithiasis), head-positioning maneuvers were described in which unilateral BPPV mimics bilateral BPPV [Steddin and Brandt, 1994J. Suzuki et a!. [1996J described a functional animal model of BPPV using an isolated frog semicircular canal. When otoconia were placed on the cupular surface to mimic cupulolithiasis, ampullary nerve action potential instantaneously responded; when otoconia were dropped into the canal to mimic canalolithiasis, action potentials responded with the otoconial flow after a latent period. Etiology Particles have been frequently found in the membranous labyrinth of symptomatic and asymptomatic patients. These particles seem to be identical to otoconia or otoconial (calcite) fragments [Cussen, 1974; Kveton and Kashgarian, 1994J. The particular matter from within the membranous posterior semicircular canal from a patient at the time of canal occlusion for intractable BPPV was examined by scanning electron microscopy. It appeared to be morphologically consistent with degenerated otoconia [Welling et a!., 1997]. However, there is still some uncertainty about the origin of these deposits, and it seems likely that more than one possible explanation is needed to account for their existence [Moriarty et a!., 1992J. Large series of patients [Katsarkas and Kirkham, 1978; Baloh et a!., 1987] provided support for the common clinical finding that the following playa role in the etiology of BPPY: head (labyrinthine) trauma, vestibular neuritis, vertebrobasilar ischemia, postsurgery (ear and general), prolonged bedrest due to unrelated diseases, and most often 'idiopathic' conditions (e.g., aging). Single case descriptions include postoperative bedrest [Cyo, 1988] or neurosurgical removal of an osteoma using hammer and chisel [Andaz et aI., 1993J. In the
Benign Paroxysmal Positioning Vertigo
179
early stages, BPPV is usually experienced on awaking in the morning rather than on first lying down. In the series of 240 patients described by Baloh et al. [1987J, the origin was idiopathic in about half of the cases. In the remainder the most commonly identified causes were head trauma (17%) and vestibular neuritis (15%). When patients present with post-traumatic BPPV [Gordon, 1954; Barber, 1964]. it is sometimes di cult to determine retrospectively whether the trauma caused the vertigo or vice versa. In less than 10% of patients, BPPV is bilateral (mostly asymmetrical); this is particularly more frequent in posttraumatic cases [Longridge and Barber, 1978J. We found that 12% of a total of 104 patients with unilateral BPPV had su ered from vestibular neuritis days or years previously [Bochele and Brandt, 1988]. The relatively frequent occurrence of BPPV following vestibular neuritis was first attributed to ischemia of the anterior vestibular artery [Lindsay and Hemenway, 1956]. but it is more likely to be due to viral inflammation at the site of the vestibular nerve (see Vestibular Neuritis, p. 111). Vestibular neuritis is a partial rather than complete unilateral vestibular paresis [Bochele and Brandt, 1988]. since the resulting BPPV requires that the function of the posterior canal be preserved. In a retrospective population-based study in Minnesota, the age- and sexadjusted incidence of BPPV was 64/100,000 population per year [Froehling et aI., 1991J. Incidence increased by 38% with each decade of life. From epidemiological surveys, the incidence of BPPV in Japan was estimated to be between 11 and 17/100,000 population [Mizukoshi et aI., 19881. In a retrospective study of 806 patients 70 years of age or older who complained of dizziness, 41 % had a history strongly suggestive of BPPV [Bloom and Katsarkas, 1989J. The age of onset of BPPV ranges from adolescence to old age, and in the idiopathic group exhibits a peak incidence in the sixth and seventh decades, but onset tends to be earlier on the average in symptomatic forms of BPPV. In the idiopathic group, females exceed males by 2: 1 [Katsarkas and Kirkham, 1978; Baloh et aI., 1987]. whereas the sexes are about equally distributed in the post-traumatic and postvestibular neuritis forms.
Management Positional Exercises and Liberatory Maneuvers The positional exercises proposed by Brandt and Daro in 1980 were the first e ective physical therapy (fig. 5). The exercises were a sequence of rapid lateral head/trunk tilts, repeated serially to promote dispersion of the debris toward the utricular cavity. We instructed the patients to sit; to then move rapidly into the challenging position to induce the correct plane-specific stimulation of the posterior semicircular canal; to remain in the position until the
l3randt
180
Fig 5. Positional exercises for e ective physical therapy for BPPV as proposed by Brandt and Daro in 1980. Patients are instructed to sit and then to move rapidly into the challenging position, to remain in the position for at least 30 s, and then to sit up for 30 s before assuming the opposite head-down position for 30 s. These exercises are repeated serially 5-10 times a day.
evoked vertigo subsided, or for at least 30 s, and then to sit up for 30 s before assuming the opposite head-down position for an additional 30 s. Troost and Patton [1992] reviewed and diagrammed this exercise protocol. The Semont and Epley liberatory maneuvers require only a single sequence, making them preferable to the multiple repetitions over many days required by the Brandt-Daro exercises. With canalolithiasis as the established mechanism of BPPV, we can now explain the e cacy of the therapies according to anatomic and physical principles. Figure 6 illustrates the Semont maneuver in a patient with typical (posterior canal) left-sided BPPV The clot causes no deflection of the cupula in the upright position. When the patient is quickly tilted toward the a ected left
Benign Paroxysmal Positioning Vertigo
181
Fig 6. Schematic drawing of the Semont liberatory maneuver in a patient with typical BPPV of the left ear. Boxes from left to right: position of body and head, position of labyrinth in space, position and movement of the clot in the posterior canal and resul ting cupula deflection, and direction of the rotatory nystagmus. The clot is depicted as an open circle within the canal; a black circle represents the final resting position of the clot. 1 In the sitting position, the head is turned horizontally 45° to the una ected ear. The clot, which is heavier than endolymph, settles at the base of the left posterior semicircular canal. 2The patient is tilted approximately 105 0 toward the left (a ected) ear. The change in head position, relative to gravity, causes the clot to gravitate to the lowermost part of the canal and the cupula to deflect downward, inducing BPPV with rotatory nystagmus beating toward the undermost ear. The patient maintains this position for 3 min. 3The patient is turned approximately 195 0 with the nose down, causing the clot to move toward the exit of the canal. The endolymphatic flow again deflects the cupula such that the nystagmus beats toward the left ear, now uppermost. The patient remains in this position for 3 min. 4 The patient is slowly moved to the sitting position; this causes the clot to enter the utricular cavity. A, P, and H Anterior, posterior, horizontal semicircular canals; UT utricular cavity; RE right eye, and LE left eye. [From Brandt et aI., 1994.1
Brandt
182
ear with a 45° head rotation to the right (moving the left posterior canal to a plane corresponding to the plane of the head tilt), the clot gravitates toward the lower part of the canal, causing the cupula to deflect downward (ampullofugal), and triggering a typical BPPV attack. These events explain the latency of a few seconds (the time needed for the clot-induced endolymph flow to develop by gravitational force), the ine ectiveness of a very slow positioning movement (the clot would then slowly gravitate along the undermost wall of the canal without plugging the canal and deflecting the cupula) , and the short duration of the positional vertigo/nystagmus (the cupula deflection ends when the clot reaches its lowest position in the canal) [Brandt and Steddin, 1993]. If the patient is swung toward the opposite right side with the nose down, the clot will gravitate downward, causing stimulation of the posterior canal of the a ected left ear (now uppermost). If no vertigo and nystagmus are elicited, we gently shake the patient's head in this position; this sometimes seems to facilitate settlement of the clot. The patient is then slowly moved to the upright position; the clot will gravitate downward through the common crus of the posterior and anterior canals and enter the utricular cavity, where it becomes harmless. We share the experience of others [Serafini et a!., 1996J that complete recovery after a single maneuver is achieved in about 50-70% of cases. Semont et a!. [1988] recommended having the patient maintain the upright position for 48 h following the liberation, but we have not found this to be necessary. Figure 7 illustrates the Epley maneuver [Epley, 1992J as modified by Herdman et a!. [1993J and Harvey et a!. [1994J in a patient with typical (posterior canal) left-sided BPPV The clot causes no deflection of the cupula in the upright position with the head turned horizontally 45° to the a ected ear. When the patient is qUickly tilted backward into a slight head-hanging position, the clot gravitates downward in the posterior canal, deflecting the cupula downward and inducing a BPPV attack. Rotation of the head and trunk toward the una ected right ear causes further movement of the clot downward (ampullofugal) toward the exit of the canal, resulting in positioning vertigo and nystagmus toward the a ected (now uppermost) ear. The final uprighting of the patient causes the clot to enter the utricular cavity, and it becomes harmless. Li [1995J found that the success rate improved if the procedure is combined with mastoid vibration; this corresponds to our attempts to promote 'canal-clearing' by additional head shaking. In our opinion, the frequently used term 'canalith repositioning maneuver' [Epley, 1992] is incorrect, since it is unlikely that the clot is 'repositioned' to its original location. Following e ective physical liberation, approximately 50% of patients [BaJoh et a!., 1987J will experience a recurrence of attacks; 10-20% occur in the first 2 weeks [Herdman et a!., 1993J. The recurrences may be due to re-
Benign Paroxysmal Positioning Vertigo
183
5 RE
LE
FIg 7. Schematic drawing of modified Epley liberatory maneuver. Patient characteristics and abbreviations as in figure 6 (Cup cupula). I In the sitting position, the head is turned horizontaUy 45° to the a ected (left) ear. 2The patient is tilted approximately 105° backward into a slight head-hanging position, causing the clot to move in the canal, deflecting the cupula downward, and inducing the BPPV attack. The patient remains in this position for 3 min. 3a The head is turned 90° to the una ected ear, now undermost, and 3b the head and trunk continue turning another 90° to the right, causing the clot to move toward the exit of the canal. The patient remains in this position for 3 min. The positioning nystagmus beating toward the a ected (uppermost) ear in positions 3a and 3b indicates e ective therapy. 5 The patient is moved to the sitting position. [From Brandt et aI., 1994.1 Brandt
184
entry of the debris into the posterior canal from the utricular cavity and should be treated with the same maneuver that induced resolution of the initial episode. The process illustrated in figures 6 and 7 explains the seemingly paradoxic observation [Brandt and Steddin, 1993; Semont et a\., 1988; Pace-Balzan and Rutka, 1991; Hausler and Pampurik 1989J that the finalliberatory positioning with the a ected ear uppermost (fig. 6, panel 3; fig. 7, panel 3b) induces nystagmus that beats toward that ear. As described above, cupulolithiasis predicts an ampullopetal deflection of the cupula that would cause nystagmus to beat toward the undermost ear, whereas in canalolithiasis, the clot-induced endolymphatic flow causes ampullafugal deflection of the cupula and nystagmus beating to the uppermost ear. Moreover, the upward direction of the nystagmus induced by the final positionings is a clinically relevant observation in that it provides reasonable certainty that the clot has exited the canal (or will so exit in the modified Epley maneuver) and the patient will be free of symptoms ('liberated') . If the nystagmus does not beat upward toward the a ected ear, the clot is probably still inside the canal; if the nystagmus beats downward toward the una ected ear, the clot must have moved toward the cupula, causing an ampullopetal deflection. In either situation, the procedure should be repeated. If the nystagmus fails to beat upward following the second procedure and the BPPV persists, we schedule a return visit for the same maneuver. If the second session fails, we try a di erent Iiberatory maneuver (Le., modified Epley, if we first used Semont, or vice versa). If both Iiberatory maneuvers fail, we prescribe BrandtDaro exercises. A possible complication of liberatory maneuvers is that the clot leaves the posterior canal but instead of staying in the utricular cavity enters the anterior (via common crus) or the horizontal canal. Thus, p-BPPV may convert to h- or a-BPPV This occurred in 5 of 85 patients originally with typical pBPPV (horizontal canal: 3, anterior canal: 2) after they had undergone liberatory maneuvers [Herdman and Tusa, 1996]. 'Canalithjam' is another speculative description of hitherto unexplained transient phenomena that rarely occur during physical treatment [Epley, 1995]: 'An interesting phenomenon that I have occasionally observed while undertaking the canalith repositioning procedure is a sudden conversion of transient nystagmus to a rapid form that persists irrespective of head position. Simultaneously the patient usually complains of intense vertigo. I believe the mechanism to be a jamming of the canaliths when migrating from a wider to a narrower segment (Le., from ampulla to canal or at the bifurcation of the common crus). For treatment the crus is repositioned (inverted) and vibration is applied. Gravity backs dense debris out of the jam.'
Benign Paroxysmal Positioning Vertigo
185
In conclusion, there are three highly e ective therapies for the common posterior canal BPPV The Semont and modified Epley liberatory maneuvers are easily performed by a physician or trained therapist, and the instructions for home use of the Brandt-Daro exercises are simple. With both liberatory maneuvers, nystagmus beating toward the uppermost a ected ear confidently predicts therapeutic success. Surgjcal Procedures In those rare patients who do not respond even to appropriate and prolonged physical therapy (1) surgical plugging of the posterior semicircular canal via a transmastoid approach or (2) surgical transection of the posterior ampullary nerve via a middle ear approach can be considered. In our experience with more than 1,000 patients with typical BPPV, however, only a few individual patients did not respond to physical therapy, and they ultimately required selective surgical transection or canal plugging. We believe that an indication for surgical intervention is still too frequently complied with before the possibilities of physical therapy are completely exhausted. This view is shared by Epley [1995], who has invented his own e ective liberatory procedure. He believes that the disability ensuing from multiple, unpredictable recurrences is over the long term a more common indication for surgery.
Singular Neurectomy Transection of the posterior ampullary nerve provides relief of vertigo; it is, however, not easy to locate surgically the relevant semicircular canal [Leuwer and Westhoven, 1996], and sensorineural hearing loss is a possible complication among several others [Gacek, 1978, 1984; Epley, 1980]. In his series of 137 patients, Gacek [1995J found complete relief of vertigo in 94% with sensorineural hearing loss in 3%.
Plugging of the Posterior Semicircular Canal Parnes and McClure [1990, 1991] first described transmastoid posterior semicircular canal occlusion with complete relief ofBPPV and preserved lateral semicircular canal function as a simpler and safer alternative to singular neurectomy. Others have confirmed the success of fenestration and occlusion of the posterior canal [Pace-Balzan and Rutka, 1991; Hawthorne and ElNaggar, 1994J. Possible complications are surgical labyrinthitis with reversible
Brandt
18G
ataxia and sensorineural hearing loss, which also mostly recovers, or inadvertent plugging of neighboring (e.g., horizontal) semicircular canals. Arai et al. [1989J observed direction-changing positional nystagmus that continued for several months after vertical canal plugging in monkeys. Modifications of the procedure using laser techniques have also been reported [Anthony, 1991, 1993; Kartusch and Sargent, 1995; Nomura et aI., 1995], but convincing proof of their superiority is lacking. It should be noted here that in 1950, Vogel was the first not only to speculate that endolymph flow could be the cause of BPPV attacks, but also to propose the plugging of semicircular canals as a possible treatment.
Horizontal Semicircular Canal BPPV (h-BPPV)
The first cases of h-BPPV were described by McClure [1985J. Later, Pagnini et ai. [1989) and Baloh et al. [1993J presented more detailed clinical descriptions of this condition. Now h-BPPV accounts for about 10-20% of all patients presenting with BPPV It may be combined with p-BPPV of the same or the contralateral ear. Transitions between h-BPPV and p-BPPV are also possible, particularly as a result of therapeutic positioning maneuvers. Whereas typical h-BPPV is caused by canalolithiasis, atypical h-BPPV may occur with ageotropic positioning nystagmus caused by cupulolithiasis. The etiological factors are the same as in p-BPPV Patients do not report experiencing episodic vertigo when getting in or out of bed, but when rolling the head from side to side while supine. Consequently, positioning testing of BPPV patients should include the sitting-to-lateral head positioning maneuver for p-BPPV and the supine-to-lateral head rotation maneuver for h-BPPV (fig. 8). The most e ective physical therapy seems to be a forced prolonged bedrest, during which the a ected ear remains uppermost. Relapses of h-BPPV seem to occur more frequently than relapses of p-BPPV The Clinical Syndrome The clinical characteristics of the horizontal semicircular canal variant ofBPPV are easy to distinguish from those of posterior canal BPPV [McClure, 1985; Pagnini et aI., 1989; Baloh et aI., 1993). The diagnosis of h-BPPV is based on the following features [Strupp et aI., 1995J: (1) The patient has a history of brief episodes of vertigo, usually induced by rolling the head from side to side while supine. (2) Positioning testing reveals a linear horizontal nystagmus toward the undermost ear (geotropic) when the head of the supine patient is rapidly turned from side to side around the longitudinal z-axis (barrel roll). (3) Horizontal positioning nystagmus with the head turned to either side
Benign Paroxysmal Positioning Vertigo
187
Fjg 8. Precipitating positioning maneuvers for horizontal sernicircular canal SPPV: the head of the supine patient is rapidly turned from side to side around the longitudinal z-axis (barrel roll).
beats stronger toward the a ected ear (fig. 9); when tills position is maintained, nystagmus often reverses its direction. (4) Positioning nystagmus in h-BPPV exhibits short latencies ( 5 s) and lasts longer (20-60 s) than in p-BPPY. (5) h-BPPV rarely fatigues with repetitive positioning maneuvers. (6) About one-third of the patients show moderate, horizontal semicircular paresis during caloric irrigation of the a ected ear. (7) Positioning vertigo attacks are often more severe than in p-BPPV and more frequently associated with nausea. As distinct from p-BPPV, attacks are not elicited by the patient getting in or out of bed. bending over, or extending the neck. However, some patients report brief episodes of vertigo when turning their head while erect [Baloh et aI., 1993J.
Atypical h-BPPV with Apogeotropic Positional Nystagmus Atypical cases of h-BPPV have been described in which the positional rather than positioning nystagmus beats toward the uppermost ear [Agus et aI., 1995; Baloh et al., 1995; Nuti et aI., 1996J. They have been identified as horizontal canal 'cupulolithiasis' as distinct from typical 'canalolithiasis' [Baloh et a1.. 1995; Steddin and Brandt. 1996]. In these exceptional patients. episodic vertigo and nystagmus can be termed positional, since they depend on the head position relative to gravity rather than on the positioning maneuver.
l3randt
188
AY~~~~........J"".I""""""'-'---: I~
B C
~
lW
40"
}to"
,,
.~~·I4oP L I 1-
5S
.
Fig 9. h-BPPV of the right ear. Schematic drawing of the right horizontal canal membranous duct illustrating the mechanism of canalolithiasis in di erent head positions while supine. Rotation of the head of the patient around the longitudinal z-axis from (A) the supine (nose up) to the right lateral, (B) right lateral to left lateraL and (C) left lateral to right lateral positions while recumbent. Lower part shows the induced horizontal eye movements, which were most intense with ampullopetal stimuli (B versus C) and with maximal rotation angles of the head (A versus C). The maximum slow-phase velocity (mean SO of n number of positioning maneuvers) was 54.6 13.6°/s (n 3) in A, 17.4 1O.4°/s (n 7) in B, and 176.2 13.00 /s (n 9) in C. [From Strupp et aI., 1995.[
Vertigo and nystagmus may be more or less severe than in canalolithiasis. The following are the typical di erential diagnostic criteria [Baloh et aI., 1995; Steddin and Brandt, 1996J: (1) the direction of positional nystagmus is toward the uppermost ear (apogeotropic); (2) nystagmus may last for minutes when the precipitating position is maintained, and (3) nystagmus depends only on the assumed head position rather than the net angle of head rotation. Thus, positional nystagmus beating toward the uppermost ear is not a pathognomonic sign of central vestibular disturbance. However, it can indicate occasional cupulolithiasis of the horizontal semicircular canal. EUoJogy and Pathomechanism
The etiology of h-BPPV is the same as that of p-BPPV [Baloh et aI., 1993; De la Meilleure et al., 1996J. ltis 'idiopathic' in most cases, posttraumatic
Benign Paroxysmal Positioning Vertigo
189
in about 20%, or the result of physical liberatory maneuvers for treating pBPPV (transition from p-BPPV to h-BPPV), general or temporal bone surgery, or prolonged bedrest. In an attempt to explain the clinical features of h-BPPV, McClure [1985] proposed canalolithiasis as the underlying mechanism, a theory that was backed by Baloh et ai. [1993J. This concept is strongly supported by the intensity of the positioning nystagmus: it is maximal when a patient with hBPPV of the right horizontal semicircular canal turns his head about the longitudinal z-axis from the left lateral to the right lateral position while supine, and it is much slower when he turns his head to the right lateral position while in a supine position (nose up) (fig. 9) [Strupp et aI., 1995J. The fact that this di erence in intensity of nystagmus depends on the initial head position and direction of head rotation indirectly proves that the clot moves freely (to and fro) within the segment of the horizontal semicircular canal diametrically opposite the ampulla. Canalolithiasis is compatible with all clinical features of typical h-BPPY.
Management Several features of h-BPPV remain unclear and are still a subject of speculation. For instance, why does canalolithiasis of the horizontal semicircular canal occur despite the fact that debris leave the canal on a simple head or body tilt from one side to the other, as is often performed while lying in bed? We suspect that the following two conditions must be fulfilled for the debris to remain in the canal [Steddin and Brandt. 1996]: (1) The diameter of the congealed debris must be greater than that of the bottleneck-like narrowing of the distal branch of the canal [Curthoys and Oman, 1987J. (2) The configuration of the debris, which congeal in the canal, must be so stable that the clot does not break into pieces small enough to pass the bottleneck. If the fatigability of symptoms on repetitive testing is explained by transient dissolving of the debris, then nonfatigability, which is frequently seen in h-BPPV, may prove the above assumption. Consequently, as long as there is no fatigue of vertigo and nystagmus in h-BPPV, maneuvers intended to sluice the debris out of the canal should have minor success. In fact, the liberatory maneuvers that Lempert [Lempert. 1994; Lempert and Tiel-Wilck, 1996] and Baloh [1994] proposed were each based on only 2 patients. Lempert [1994] described a single 270 0 'barbecue rotation' toward the una ected side, which was performed in rapid steps of 90 at 30-s intervals. Baloh [1994J suggested a 360 0 rotation around the yaw axis in four quick steps of 90 0 with the initial motion toward the healthy side; each position was held for about 1 min. Vannucchi et ai. [1997] compared the therapeutic results obtained by maintaining a prolonged position on the healthy side (35 patients) with repeti0
l3randt
190
tive head shaking in a supine position (24 patients) and no therapy (15 patients). More than 90% of the patients treated with prolonged position recovered within 3 days, although 6 of 35 patients subsequently developed p-BPPV (which responded successfully to repositioning maneuvers). The rationale was obviously that the 'heavy particles' in the nonampullary arm of the horizontal canal gradually moved out when the patient maintained a prolonged position on the side of the nona ected ear. This maneuver was not e ective if performed only for a duration of 10-20 min; it was intended to be maintained for up to 12 h. In those who failed to respond, the Brandt-Daro exercises [Brandt and Daro ,1980) were performed and within a matter of days led to full recovery. This agrees with the report of Baloh et a1. [1993) and our own experience with such patients. Thus, for the time being, we propose that prolonged bedrest with the head turned toward the una ected ear [Vannucchi et aI., 1997] be maintained for up to 12 h. If this is still unsuccessful after 2 days, we advise the patients to perform the Brandt-Daro exercises (fig. 5). Both physical therapies can be performed at home and do not require the presence of a physical therapist.
References Agus G. Puxeddu R, Demontis GP, Puxeddu P: Atypical 'reversed' paroxysmal positioning nystagmus in benign paroxysmal positional vertigo. Acta Otolaryngol Suppl (Stockh) 1995;520: 143-147. Andaz C, Whittet HB, Ludman H: An unusual cause of benign paroxysmal positional vertigo. J Laryngol Otol 1993;107:1153-1154. Anthony PF: Partitioning of the labyrinth: Application in benign paroxysmal positional vertigo. Am J Otol 1991; 12:388-393. Anthony PF: Partitioning the labyrinth for benign paroxysmal positional vertigo: Clinical and histologic findings. Am J Otol 1993; 14 :334-342. Arai Y, Henn V, Boehmer A, Suzuki 1: How could canal-pluggings result in intensive direction changing type of positional nystagmus? Acta Otolaryngol Suppl (Stockh) 1989;468:159-164. Baloh RW: Reply to the letter by Lempert: Horizontal benign positional vertigo. Neurology 1994;44: 2214. Baloh RW, Hornrubia V, Jacobson K: Benign positional vertigo. Neurology 1987;37:371-378. Baloh RW, Jacobson K. Honrubia V: Horizontal semicircular canal variant of benign positional vertigo. Neurology 1993:43:2542-2549. Baloh RW. Yue Q. Jacobson KM, Honrubia V: Persistent direction-changing positional nystagmus: Another variant of benign positional vertigo? Neurology 1995;45:1297-1301. Barany R: Diagnose von Krankheitserscheinungen im Bereiche des Otolithenapparates. Acta Otolaryngol (Stockh) 1921 ;2:334-437. Barber HO: Positional nystagmus especially after head injury. Laryngoscope 1964;73:391-944. Bloom J. Katsarkas A: Paroxysmal positional vertigo in the elderly. J Otolaryngol 1989; 18:96-98. Brandt Th: Positional and positioning vertigo and nystagmus (review). J Neurol Sci 1991:95:2-28. Brandt Th: Vertigo: Its Multisensory Syndromes. 2nd Ed. London, Springer, 1999. Brandt Th, Daro RB: Physical therapy for benign paroxysmal positional vertigo. Arch Otolaryngol 1980; 106:484-485. Brandt Th, Dieterich M: Vestibular falls. J Vestib Res 1993;3;3-14.
Benign Paroxysmal Positioning Vertigo
191
Brandt Th, Steddin S: Reply to the letter by Gordon: 'Benign paroxysmal positional vertigo (BPPV) or bubble provoked positional vertigo?' J Neurol Sci 1992;111:231-233. Brandt Th, Steddin S: Current view of the mechanism of benign paroxysmal positioning vertigo: Cupulolithiasis or canalolithiasis? J Vestib Res 1993;3:373-382. Brandt Th, Steddin S, Daro RB: Therapy for benign paroxysmal positioning vertigo, revisited. Neurology 1994;44;796-800, Biichele W, BrandtTh; Vestibular neuritis - A horizontal semicircular canal paresis? Adv Otorhinolaryngol. Basel, Karger, 1988, vol 42, pp 157-161. Curthoys JS, Oman CM: Dimensions of the horizontal semicircular duct ampulla and utricle in humans. Acta Otolaryngol (Stockh) 1987;103:254-261. De la Meilleure G, Dehaene T, Depondt M, Damman W, Crevits L, Vanhooren G: Benign paroxysmal positional vertigo of the horizontal canal. J Neurol Neurosurg Psychiatry 1996;60:68-71. Dix R. Hallpike CS; The pathology, symptomatology and dIagnosis of certain common disorders of the vestibular system, Ann Otol Rhinol Laryngol 1952;6:987-1016, Epley JM: New dimensions of benign paroxysmal positional vertigo. Otolaryngol Head Neck Surg 1980; 88:599-605. Epley JM: The canalith repositioning procedure for treatment of benign paroxysmal positional vertigo. Otolaryngol Head Neck Surg 1992; 107:399-404. Epley JM; Positional vertigo related to semicircular canalithiasis, Otolaryngol Head Neck Surg 1995; 112; 154-161. Froehling DA, Silverstein MD, Mohr ON, Beatty CW, 0 ord KP, Ballard OJ: Benign positional vertigo: Incidence and prognosis in a population-based study in Olmsted County, Minnesota. Mayo Clin Proc 1991 ;66:596-601. Gacek RR: Further observations on posterior ampulla!)' nerve transection for positional vertigo, Ann Otol Rhinol Laryngol 1978;87:300-306, Gacek RR: Cupulolithiasis and posterior ampullary nerve transection. Ann Otol Rhinol Laryngol 1984; 93(suppl 112):25-29. Gacek RR: Technique and results of singular neurectomy for the management of benign paroxsymal positional vertigo. Acta Otolaryngol (Stockh) 1995;115:154-157. Gordon N: Post-traLuuatic vertigo, with special reference to positional nystagmus. Lancet 1954;i: 1216-1218. Gussen R: Saccule otoconia displacement into cochlea in cochleosaccular degeneration. Arch Otolaryngol 1974; 106: 161-166, Gyo K: Benign paroxysmal positioning vertigo as a complication of postoperative bedrest. Laryngoscope 1988;98:332-333. Hall SF, Ruby RR McClure JA: The mechanics of benign paroxysmal vertigo. J Otolaryngol 1979;8: 151-158, Harvey SA, Hain TC, Adamiec LC: Modified liberatory maneuver: E ective treatment for benign paroxysmal positional vertigo. Laryngoscope 1994;104:1206-1212. Hasegawa T: Die Veranderung der labyrintharen Refiexe bei zentrifugierten Meerschweinchen. PfiOgers Arch 1933;232:454-465. Hausler R Pampurik JC: Die chirurgische und die physiotherapeutische Behandlungdes benignen paroxysmalen Lagerungsschwindels. Laryngorhinootologie 1989;68:342-346. Hawthorne M, EI-Naggar M: Fenestration and occlusion of posterior semicircular canal for patients with intractable benign paroxysmal positional vertigo. J Laryngol Otol 1994;108:935-939. Herdman Sj: Treatment of benign paroxysmal positional vertigo. Phys Ther 1990;70:381-387. Herdman SJ, Tusa R): Complications of the canalith repositioning procedure. Arch Otolaryngol Head Neck Surg 1996;122:281-286. Herdman SJ, Tusa R), Zee OS, Proctor LR Mattox DE: Single treatment approaches to benign paroxysmal positional vertigo. Arch Otolaryngol Head Neck Surg 1993; 119:450-454. Igarashi M, Nagaba M: Vestibular end-organ damage in squirrel monkeys after exposure to intensive linear acceleration; in NASA SP: Third Symposium on the Role of the Vestibular Organs in space Exploration, no 152, pp 63-71. l Johnson L-G, Hawkins JE: Sensory and neural degeneration with ageing, as seen in microdissections ofl the human inner ear. Ann Otol 1972;81:179-193.
I3randt
192
Kartusch JM, Sargent EvV: Posterior semicircular canal occulusion for benign paroxysmal positional vertigo-C0 2 laser-assisted technique: Preliminary results. Laryngoscope 1995; 105:268-273. Katsarkas A, Kirkham Th: Paroxysmal positional vertigo as a complication of postoperative bedrest. Laryngoscope 1978;98:332-333. Kveton JF, Kashgarian M: Particulate matter within the membranous labyrinth: Pathologic or normal? Am J OtoI1994;15:173-176. Lanska OJ. Remler B: Benign paroxysmal positioning vertigo: Classic descriptions, origins of the provocative positioning technique, and conceptual developments. Neurology 1997;48: 1167-1177. Lempert T: Horizontal benign positional vertigo (letter). Neurology 1994;44:2213-2214. Lempert T, Tiel-Wilck K: A positional maneuver for treatment of horizontal-canal benign positional vertigo. Laryngoscope 1996; 106:476-478. Lempert T, Wolsley C, Davies R, Gresty MA, Bronstein AM: Curing benign positional vertigo in a 3D night simulator. Lancet 1996;347:1192. Leuwer RM, Westhofen M: Surgical anatomy of the singular nerve, Acta Otolaryngol (Stockh) 1996; 116:576-580. Li JC: Mastoid oscillation: A critical factor for success in the canalith repositioning procedure. Otolaryngol Head Neck Surg 1995; 112:670-675. Lindsay JR, Hemenway vVG: Postural vertigo due to unilateral sudden partial loss of vestibular function. Ann Otol Rhinal Laryngol 1956;65:692-708. Langridge NS, Barber HO: Bilateral paroxysmal positioning nystagmus. Can J Otol 1978;7:395-400. McClure JA: Horizontal canal BPPY. J Otolaryngol 1985;14:30-35. McClure JA: Functional basis fOr horizontal canal BPPV; in Barber HO, Sharpe JA (eds): Vestibular Disorders. Chicago, Year Book Medical. 1988, pp 233-238. tVlizukoshi K, Watanabe Y, Shojaku H, Okubo J, Watanabe l: Epidemiological studies on benign paroxysmal positional vertigo in Japan. Acta Otolaryngol Suppl (Stockh) 1988;447:67-72. Money KE, Myles WS: Heavy water nystagmus and e ects of alcohol. Nature 1974;247:404-405. Money KE, Myles WS, Ho ert BM: The mechanism of positional alcohol nystagmus. Can J Otolaryngol 1974;3:302-313. Moriarty B, Rutka J, Hawke M: The incidence and distribution of cupular deposits in the labyrinth. Laryngoscope 1992;102:56-59. Naganuma H, Kohut RI, Tokumasu K, Okamoto M, Fujino A, Arai M: Basophilic deposits on the cupula: Preliminary findings describing the problems involved in studies regarding the incidence of basophilic deposits in the cupula, Acta Otolaryngol Suppl (Stockh) 1996;524:9-15, Nomura Y, Ooki S, Kukita N, Young YH: Laser labyrinthectomy, Acta Otolaryngol (Stockh) 1995;115: 158-161. Norn~ ME: Reliability of examination data in the diagnosis of benign paroxysmal positional vertigo. Am J OtoI1995;16:806-81O. Nuti 0, Vannucchi P, Pagnini P: Benign paroxysmal positional vertigo of the horizontal canal: A form of canalolithiasis with variable clinical features. J Vestib Res 1996;6:173-184. Pace-Balzan A, Rutka JA: Non-ampullary plugging of the posterior semicircular canal for benign paroxysmal positional vertigo, J Laryngol Olol 1991; 105:901-906, Pagnini P, Nuti D, Vannucchi P: Benign paroxysmal vertigo of the horizontal canal. ORL 1989;51: 161-170. Parnes LS, McClure JA: Posterior semicircular canal occlusion for intractable benign paroxysmal positional vertigo. Ann Otol Rhinal Laryngol 1990;99:330-334. Parnes LS, McClure JA: Posterior semicircular canal occlusion in the normal hearing ear. Otolaryngol Head Neck Surg 1991;104:52-57. Parnes LS, McClure JA: Free-floating endolymph particles. Laryngoscope 1992;102:988-992. Rietz R, Troia BW, Yonkers AJ, Norris TW: Glycerol-induced positional nystagmus in human beings. Otolaryngol Head Neck Surg 1987;97:282-287. Schuknecht HF: Positional vertigo. Clinical and experimental observations. Trans Am Acad Ophthal Otolaryngol 1962;66:319-331. Schuknecht J-lF: Cupulolithiasis. Arch Otolaryngol 1969;90:765-778.
Benign Paroxysmal Positioning Vertigo
193
Schuknecht HF, Ruby RRF: Cupulolithiasis Adv OtorhinolaryngoL Basel. Karger, 1973, vol 22, pp 434443.
Semont A Freyss C, Vitte E: Curing the BPPV with a liberatory maneuver. Adv OtorhinolaryngoL Basel, Karger, 1988, vol 42, pp 290-293. Serafini C, Palmierei AMR, SimoncelJi C: Benign paroxysmal positional vertigo of posterior semicircular canal: Results in 160 cases treated with Semont's maneuver. Ann Otol Rhinal Laryngol 1996;105: 770-775.
Smouha EE: Time course of recovery after Epley maneuvers for benign paroxysmal positional vertigo. Laryngoscope 1997;107:187-191. Steddin S, Brandt Th: Unilateral mimicking biJateral benign paroxysmal positioning vertigo. Arch Otolaryngol Head Neck Surg 1994; 120: 1339-1341. Steddin S, Brandt T: Horizontal canal benign paroxysmal positioning vertigo (h-BPPV): Transition of canalolithiasis to cupulolithiasis, Ann Neurol 1996;40:918-922, Steenerson RL, Cronin CW: Comparison of the canalith repositioning procedure and vestibular habituation training in forty patients with benign paroxysmal positional vertigo. Otolaryngol Head Neck Surg 1996;114:61-64. Strupp M, Brandt Th, Steddin S: Horizontal canal benign paroxysmal positioning vertigo: Reversible ipsilateral caloric hypoexcitability caused by canalolithiasis? Neurology 1995;45:2072-2076. Suzuki M, Kadir A Hayashi N, Takamoto M: Functional model of benign paroxysmal positional vertigo using an isolated frog semicircular canaL J Vest Res 1996;6: 121-125. Troost BT, Patton JM: Exercise therapy for positional vertigo. Neurology 1992;42:1441-1444. Vannucchi P, ciaonnoni B, Pagnini P: Treatment of horizontal semicircular canal benign paroxysmal positional vertigo. J Vestib Res 1997;7:1-6. Vogel K: Zur Entstehung des peripheren Lagenystagmus. Arch OhrenheiJkd 1950; 157:89-98. VyslonziJ E: Ober eine umschriebene Ansammlung von Otokonien in hinteren hautigen Bogengangen, Monatschr Ohrenheilkd 1963;97:63. Welling DB, Parnes LS, O'Brien B, Bakaletz LO, Brackman DE, Hinojosa R: Particulate matter in the posterior semicircular canaL Laryngoscope 1997;107:90-94.
Prof Dr. med. Thomas Brandt, Direktor der Neurologischen Universitats-Klinik, Klinikum crosshadern, Marchioninistrasse 15, 0-81377 Munchen (Germany) Tel. 49 89 70 95 25 70/1, Fax 49 89 70 95 88 83
13randt
194
Btittner U. (ed): Vestibular Dysfunction and Its Therapy. Adv Otorhinolaryngol. Basel, Karger, 1999, vol 55, pp 195-227
Drug Therapy of Nystagmus and Saccadic Intrusions U Battner, L. Fuhry Neurologische Klinik, Klinikum Grosshadern der Ludwig-Maximilians-Universitat, MOnchen, Deutschland
Nystagmus, consisting of a slow compensatory and a fast resetting phase, is the normal oculomotor response to extended head movements during the vestibulo-ocular reflex (VOR) and to large moving visual fields during optokinetic nystagmus (OKN) , This nystagmus helps to stabilize the visual environment, However, in pathological nystagmus with both the head and visual surround stationary, the eye movements cause movements to the visual surround on the retina, called oscillopsia, which are the most frequent of complaints from the a ected patient. In addition, these movements also decrease visual acuity, which is the other reason for treating these pathological eye movements [95J, Fortunately, the movements of the visual surround on the retina during pathological nystagmus are not as strongly perceived as predicted by the nystagmus velocity. Motion perception during long-lasting visual stimuli or continuous nystagmus is considerably lower than the actual velocity of the visual stimuli or the nystagmus, which can both be demonstrated during physiological and pathological conditions [19J, With congenital [52J or downbeat nystagmus the threshold for visual motion detection is Significantly higher than in normal subjects, Also patients with high velocity congenital nystagmus only seldom complain about oscillopsia [50]. This reduction of motion perception has been attributed to a central adaptation mechanism [19], aimed at suppressing unwanted visual motion. Thus, the central nervous system can partly suppress the e ects of self-generated visual motion. However, this e ect is only su cient up to a certain level. If nystagmus velocity can be reduced below SOls, then oscillopsia is eliminated and vision is improved [98J. With higher velocities the e ects can still be very disturbing.
There are several disorders which can lead to oscillopsia. They usually occur with a peripheral vestibular disorder, where nystagmus and oscillopsia are combined with vertigo. In contrast to oscillopsia, vertigo also persists with the eyes closed. [Peripheral Vestibular Disorders and Their Management, see chapter by Strupp and Brandt]. Central causes of oscillopsia are jerk nystagmus as in downbeat, upbeat, torsional, horizontal and periodic alternating nystagmus. Other forms of nystagmus, such as congenital and acquired pendular, also lead to oscillopsia. In general, not only nystagmus, but all pathological, frequently occurring eye movements lead to oscillopsia. This includes ocular flutter and opsoclonus, ocular myoclonus and superior oblique myokymia. The latter eye movements have no relation to a vestibular disorder, i.e. opsoclonus and ocular flutter are considered as a saccadic disorder [26]. Thus, the symptoms (oscillopsia, reduced visual acuity) to be cured are very similar in all conditions, although the causes are very di erent. This does not only apply to the di erences between a saccadic disorder and jerk nystagmus. Even jerk nystagmus can have di erent causes, which often are not related to a central vestibular disorder. For spontaneous nystagmus, which is nystagmus while looking straight ahead, several di erent pathomechanisms can be distinguished [24]. Usually spontaneous nystagmus is attributed to a peripheral or central vestibular imbalance. With head movements under normal conditions the vestibular nerve activity increases on one side and decreases on the other side (vestibular imbalance). Accordingly, lesions of peripheral or central vestibular structures also lead to nystagmus. The main feature of this type of nystagmus is a constant velocity slow phase [32, 139]. In contrast, nystagmus due to a neural integrator deficit is characterized by an exponential decay ofthe slow phase velocity After each saccade, neuronal (and muscle) activity is required in order to maintain the eye in its new position. Since premotor structures for eye movements including those for saccades encode eye velocity, this signal has to be integrated (in the mathematical sense) to obtain eye position. Such a neural integrator for horizontal eye movements has been localized in the region of the medial vestibular nucleus/nucleus praepositus hypoglossi complex in the lower pons [32] and for the vertical and torsional system in the interstitial nucleus of Cajal (iC) in the mesencephalon [73]. The cerebellum, particularly the floccular region, supports the neural integrator, i.e. gaze-hulding. With a neural integrator deficit the eye drifts back after each saccade to a null-position and a new saccade has to be made in an attempt to maintain the desired eye position. This manifests as gazeevoked nystagmus and characteristically the slow phase decays exponentially. The null-position does not have to coincide with the midposition of the eye,
BiittnerlFuhry
196
i.e. gaze straight ahead. It has indeed been shown experimentally that the nulland midposition can di er up to 35° [139J. In these instances a spontaneous nystagmus is evident, which characteristically has an exponentially decayjng slow phase. There are also reports which consider some instances of spontaneous nystagmus as a consequence of a smooth pursuH jmbalance. This has been attributed mainly to vertical disorders, particularly downbeat nystagmus, but also to horizontal nystagmus. Finally, it has recently been shown that a saccade generator deficHcan lead to spontaneous nystagmus. The saccade generator for the horizontal system is located in the paramedian pontine reticular formation (PPRF) and for torsional and vertical saccades in the rostral interstitial nucleus of the medial longitudinal fasciculus (rostral iMLF). Lesions of the PPRF lead to a horizontal ipsilateral gaze palsy and lesions of the rostral iMLF to a loss of torsional saccades to the ipsilateral side [142]. In addition to the loss of saccades in a specific direction after a unilateral saccade generator lesion, spontaneous nystagmus to the contralateral side with a PPRF lesion [75] and contratorsional nystagmus with a rostral iMLF lesion have been found [73J. Thus, a similar phenomenon (spontaneous nystagmus) can have completely di erent causes, which of course has to be considered when a drug therapy based on neurophysiological and neuropharmacological mechanisms is conceived. Furthermore, lesions in di erent locations in the bralnstem and/ or the cerebellum can lead to nystagmus of the same appearance [24]. Since the vestibular nuclei are a major site of central vestibular control, it should be stressed that not only are vestibular functions relayed in these nuclei, but that they also play (among others) an important role for smooth pursuit generation and the oculomotor neural integration. There are several means to reduce oscillopsia caused by involuntary eye movements. Often the patients develop strategies themselves. It is a common experience that patients with congenital nystagmus turn their head in order to bring their gaze direction close to the null-position, i.e. the eye position with minimal or no nystagmus velocity. As mentioned above, this null-position often does not coincide with the midposition of the eye, i.e. gaze straight ahead. Individual patients with periodic alternating nystagmus (PAN) can also reduce nystagmus by appropriate VOR interaction. First they can figure out in which direction their nystagmus is currently beating by looking to the side. If they look to the right and the nystagmus intensity (velocity) increases, they have right nystagmus (Alexander's law). With this infUI'matiun and their experience that their vestibular responses are functioning [60], they can induce postrotatory nystagmus by a sudden stop after rotating several times around their body axis. Tills postrotatory nystagmus counteracts the PAN for a willIe and reduces the disturbing oscillopsia. Unfortunately, this method is rather
Drug Therapy of Nystagmus and Saccadic Intrusions
197
elaborate and mistakes in the turning direction lead to a worsening of the symptoms. For therapy several approaches are used [95; also see chapter by Leigh]. They consist of surgery, injection of botulinum toxin into selected extraocular muscles, optical treatments and drug therapy. Eye muscle surgery is most often applied in patients with congenital nystagmus. The aim here is to move the null-position to the midposition of the eye (Anderson-Kestenbaum procedure). Injection of botulinum toxin into selected extraocular muscles or into the retrobulbar space reduces muscle strength and can thereby reduce unwanted eye movements. For optical treatments, prisms and lenses are used. Depending on the method used they move the null-position, alter the vergence angle (which a ects some forms of acquired nystagmus) or stabilize images on the retina [see chapter by LeighJ. In the following only drug therapy will be considered. Often di erent approaches are used to treat the same disorder. Rarely are di erent approaches combined to treat one disorder. Hardly any treatment is successful in all instances.
Principles of Drug Therapy
Drugs to reduce involuntary eye movements aim at neurotransmitter receptor sites in the central nervous system. In the following a few points important for the understanding of drug therapy will be reviewed [46; also see chapter by Vidal et a1.J. Neurotransmitters can be divided into fast- and slow-acting substances. With regard to the system under consideration, fast-acting neurotransmitters are excitatory and inhibitory amino acids, which include glutamate, -aminobutyric acid (GABA) and glycine. About 30-40% of synaptic connections in the central nervous system use GABA as their transmitter, and another 15-20% use glutamate. They act mainly on postsynaptic, ionotropic receptors. In contrast. the slower acting substances mostly activate metabotropic (second messenger-linked) receptors. The monoaminergic substances noradrenaline, dopamine, serotonin and histamine belong to this group, together with acetylcholine (muscarinic receptors). All these substances are universal in the central nervous system or at least have been found in many areas, so they are by no means specific to vestibular structures. In experimental animal studies these substances ur related drugs can be specifically delivered by microinjections into circumscribed central nervous structures and the resulting oculomotor changes can be observed [46, 139]. This, however, is not yet possible for treating patients. Drugs have to be given systemically, which often cause intolerable side e ects, since the drug is
BllttnerlFuhry
198
not only acting on central vestibular structures. Furthermore, with systemic application the blood-brain barrier has to be taken into account, which often decreases the e ciency of the proposed drug on the site of action [11]. Several drugs have been shown to be e ective by intravenous application [11]. which however rarely provides a long-lasting e ect. It has also been shown that the chronic use of anticholinergic drugs (muscarine antagonists) increase the receptor density in many brain areas, which is not seen after single applications [12]. and reduces the e ect of long-term therapy. Glutamate is an excitatory amino acid. In central structures at least two di erent classes of glutamatergic receptors have been distinguished: Ionotropic receptors including NMDA (N-methyl-D-aspartate) and AMPA ( -amino3-hydroxy-5-methyl-4-isoxalone-propionic acid) subtypes and metabotropic receptors. In neurons, which act on NMDA receptors, glutamate can be colocalized with glycine [119]. The co-release of glutamate and glycine thus then potentiates the postsynaptic excitatory e ect on NMDA receptors. Otherwise glycine is an inhibitory transmitter acting on strychnine-sensitive ionotropic receptors (see below). Apart from the vestibular nerve, a erents from other structures to the vestibular nuclei might also use glutamate as a transmitter [46]. Glutamate is also the excitatory transmitter for some of the mossy fibers [130] and the granule cells in the cerebellum. It also acts as the excitatory transmltter for vestibula-ocular pathways on AMPA receptors on abducens motaneurons [I36J. Memantine, a glutamate antagonist, has been successfully applied in the treatment of patients with acqUired pendular nystagmus (APN) [I34J. This drug is considered to block the NMDA receptor channel and to modulate the AMPA receptor. In addition it also has some dopaminergic action. The site of action for memantine within the brain is not yet clear, except for APN the vestibular nuclei seem to be an unlikely location. One possible site would be the excitatory input to the motoneurons, particularly since in APN both eyes can be a ected di erentially. GABA is the main inhibitory substance in the central nervous system. An ionotropic GABA A and a metabotropic GABA B receptor can be distinguished. All vestibular nuclei contain dense innervation by GABA-ergic a erent fibers. They include Purkinje cell axons, which use GABA as their main transmitter. A GABA-ergic projection originating from the contralateral inferior olive has also ueen suggested. Vestiuular nuclei neurons contain GABA A and pre- and postsynaptic GABA B receptors. A high percentage of vestibular nuclei neurons also use GABA as their transmitter. These neurons would correspond to inhibitory interneurons within the vestibular nuclei and to inhibitory neurons projecting to motoneurons mediating the verticalVOR. In contrast, the inhibi-
Drug Therapy of Nystagmus and Saccadic [ntrusions
199
tory neurons for motoneurons of the horizontal VOR utilize glycine as a transmitter [131]. GABA is also a putative neurotransmitter for mossy fibers in the cerebellum [72]. Behavioral e ects of GAB A have been shown in a number of experimental studies. In the monkey unilateral microinjections of the GABA A agonist muscimol in the vestibular nuclei lead to failure of the common neural integrator for horizontal eye movements [139]. For torsional/vertical eye movements muscimol yields such an e ect after injection in the iC [73]. A failure of the neural integrator leads to a severe gaze-holding deficit with gaze-evoked nystagmus. In contrast to muscimol, injections of the GABA A receptor-antagonist bicuculline in the vestibular nuclei lead to a vestibular imbalance with constant velocity slow phases of nystagmus [139J. Systemic application of baclofen (GABA B agonist) strongly impairs the velocity storage mechanisms in normal and nodulus/uvula lesioned monkeys [38, 155]. The inhibitory action of systemically applied baclofen is probably not due to a direct e ect on GABA-ergic synapses, since it has been shown that in usual oral doses baclofen is not a GABA-mimetic agent [57J and that one e ect is the presynaptic inhibition of glutamate release [106, 115]. In addition, baclofen also increases the glycine turnover (at least in the spinal cord) suggesting that bac10fen might also a ect inhibitory glycinergic interneurons [115]. Benzodiazepines (clonazepam, diazepam) are thought to modify GABA actions acutely and chronically [144] by acting on GABA A receptors. Barbiturates are also intimately related to the ionotropic GABA receptor complex [113J. Gabapentin, which has recently been shown to be successful in the treatment of APN in a double-blind controlled study [7], also e ects GABA metabolism and release [81J. However, although structurally similar to GABA it shows no direct e ect on GABA A or GABA B receptors [148J. Besides GABA, glycine is the main inhibitory transmitter in the central nervous system. Glycine receptors are ionotropic. Vestibular nuclei neurons express postsynaptic, strychnine-sensitive glycinergic receptors, which proves the hyperpolarizing e ect of these receptors. Glycine is used (beside GABA) as an inhibitory transmitter for commissural interneurons in the vestibular nuclei and for the inhibition of motoneurons during the horizontalVOR [131J. It is also the transmitter of the pause neurons, which inhibit saccadic burst neurons in the paramedian pontine reticular formation [82]. Cholinergic substances act on nicotinic (ionotropic) and muscarinic (metabutrupic) receptors. Buth classes uf receptors have beell detected ill all vestibular nuclei [46J. The origin for many of the neurons acting on these vestibular nuclei neurons has yet to be determined. They could be localized within the vestibular nuclei (intrinsic neurons) or originate in the contralateral inferior olive. Cholinergic neurons from the vestibular nuclei project to the flocculus,
BllttnerlFuhry
200
nodulus [10], dorsal cap of the inferior olive and the spinal cord. The visual climbing fiber input to the cerebellum also has a high cholinergic activity [123]. At least 5 subtypes of muscarinic receptors (M I-M5) with di erent distribution in the central nervous system can be distinguished. Whereas granule cells in the cerebellum primarly express M2 and M3 receptors [1], neurons in the pons and medulla almost exclusively express M2 [116]. Cholinergic substances have also been shown to be e ective on a behavioral level. In intact animals unilateral microinjections into the vestibular nuclei lead to a postural deficit. Injections of carbachol (cholinergic agonist) or betanechol (muscarinic agonist) into the rabbit flocculus increase the gain of sinusoidal OKN [145-147]. Interestingly, in patients, physostigmine (acetylcholine-esterase inhibitor) might be beneficial in some central vestibular disorders (hyperactive horizontal VOR, disturbed fixation suppression of caloric nystagmus) [149. 151], but deteriorating in others (downbeat nystagmus) [51].
Downbeat Nystagmus
Clinical Aspects Downbeat nystagmus is usually present during gaze straight ahead and increases on lateral gaze, downward gaze and during convergence. According to Alexander's law. downbeat nystagmus decreases on upward gaze. Fixation usually does not eliminate the nystagmus. There are also many exceptions to these rules [24J. In the head-hanging position. slow phase velocity of the downbeat nystagmus is often enhanced [100J. Downbeat nystagmus is probably the most common form of acquired jerk nystagmus of central origin. There are a variety of clinical conditions resulting in downbeat nystagmus. In most cases it is seen with cerebellar lesions (Arnold-Chiari malformation. cerebellar degeneration). It has rarely been clinically related to defined brainstem lesions [15J. Other common causes are drugs and nutritional deficiencies. In about 40% of cases with downbeat nystagmus, the cause remains unclear [42]. Although downbeat nystagmus can resolve, particularly when due to drugs or nutritional deficiencies, it can often last for many years. Pathuphysiulugy and Expeljmental Studies Downbeat nystagmus is always of central origin. As one possible cause an imbalance of the central vertical vestibular tone has been assumed. The vertical VOR is driven by signals deriving from the posterior and anterior semicircular canals. Posterior semicircular canal a erences mediate downward,
Drug Therapy of Nystagmus and Saccadic [ntrusions
201
signals from the anterior semicircular canals mediate upward eye movements. The vertical VOR is di erentially controlled for up and down. Imbalance can therefore be either achieved by a relative decrease of the signals originating in the posterior semicircular canals or a relative increase of the anterior semicircular canal a erences. Lesions to the posterior cerebellar midline structures can evoke downbeat nystagmus as well. Anatomically, these structures only have an inhibitory influence on the VOR elicited by the anterior semicircular canals. In contrast, a erent information from the posterior semicircular canals is not a ected by means of this pathway [851. Due to this asymmetry, loss of Purkinje cells in the midline structures of the posterior cerebellum could lead to a disinhibition of anterior canal input resulting in both upward drift ofthe eyes and compensatory downward beating quick eye movements [1 OOJ. Accordingly, experimental ablation of the flocculus and paraflocculus invariably produces downbeat nystagmus [165]. Thus, downbeat nystagmus is one of the few disorders which can be elicited experimentally. However, it has not yet been pharmacologically attempted to treat this type of nystagmus in experimental animals. Brainstem lesions causing downbeat nystagmus are rare. In the monkey a midsagittal section in the dorsal tegmentum at the pontomedullary border causes downbeat nystagmus [44J. This lesion site. however. has not yet been established in patients [24J.
Patient Studies As a result of the anatomical. physiological and neuropharmacological data. most studies investigating the e ects of drug therapy on downbeat nystagmus concentrated on agents interfering with the GABA-ergic or cholinergic system (table 1). In an open study, baclofen reduced the slow phase velocity resulting in a long-lasting improvement [51J. However. in a recent double-blind controlled study ofbaclofen vs. gabapentin. neither ofthese two drugs showed a significant improvement of downbeat nystagmus in 6 patients when administered for 2 weeks [7J. There was also no response to baclofen seen in another patient [33]. Downbeat nystagmus improved with clonazepam in all 8 patients in an open study [42]. With single doses of 1-2 mg. the sedative side e ects were tolerable. In addition, there are casuistic reports of the beneficial e ect of clonazepam [34, 105]. but unsuccessful treatments are also reported [33, 162]. The usefulness uf drugs acting un the chulinergic transmissiun was tested in several studies. While cholinergic agents had a deteriorating e ect on downbeat nystagmus. the influence of anticholinergic drugs seems to be potentially beneficial. In a double-blind. randomized study [IIJ the centrally acting scopolamine and benztropine were tested in comparison to the peripheral acting
BiittnerlFuhry
202
Table 1. Drugs used for the treatment of nystagmus and saccadic intrusions: after each reference the total number of investigated patients (second value) and the responding patients (first value) is shown; studies in which no patient responded are not listed Baclofen (15-60 mg/day) Downbeat: Dieterich et aI., 1991 (2/3) Upbeat: Dieterich et aI., 1991 (2/2) PAN: Halmagyi et aI., 1980 (2/3); Carlow, 1986 (2/2); Isago et aI., 1985 SSN: Carlow, 1986 (1/1) CN: Yee et aI., 1982 (417)
I
(l/q
Clonazepam (1-6 mg/day) Downbeat: Currie and Matsuo, 1986 (8/8); Chambers et aI., 1983 (111); McConnell et aI., 1990 (1/1); Yee, 1989 (2/4) SSN: Cochin et a!., 1995 (111); Currie and Matsuo. 1986 (111) APN: Currie and Matsuo. 1986 (2/2) Opsoclonus: Leigh and Zee. 1983 (111); Carlow. 1986 (1/1); Anderson et a!., 1988 (2/4) MSWJ; Fukazawa et a!.. 1986 (Ill) Gabapentin (900 mg/day) APN: Averbuch-Heller et a!.. 1997 (10/15 including 2 OM and 3 SSN); Stahl et a!.. 1996 (2/2) OM: Stahl et a!., 1996 (Ill). Averbuch-Heller et a!., 1997 (2/2) Isoniazid (400-600 mg/day) APN: Traccis et al.. 1990 (2/3) Phenobarbital CN: MSWJ;
Gay et aI., 1969 (oral 60 mg/day) (111) Fukazawa et al.. 1986 (single dose 150 mg Lm.) (Ill)
Amobarbital (single dose 50-300 mg Lv.) CN: Bender. 1946 (7) APN: Nathanson et al.. 1953 (717) OM: Bender et aI., 1952 (3/3); Lawrence and Lightfoote. 1975 (111) Carbamazepine (200-1,200 mg/day) SOM: Susac et a!.. 1973 (4/5); Herzau et al.. 1978 (2/4): Morrow et a!.. 1980 (112): Rosenberg and Glaser. 1983 (617): Brazis et al.. 1994 (117) Diphenylhydantoin (200-300 mg/day) PAN: Davis and Smith, 1971 (111) Valproic acid (750-2.000 mg/day) OM: Lefkowitz and Harpold. 1985 (111)
Drug Therapy of Nystagmus and Saccadic [ntrusions
203
Table 1 (continued)
Scopolamine (single dose 0.4 mg Lv.) Downbeat: Barton et al.. 1994 (2/2) APN: Barton et al.. 1994 (SiS); Starck et aI.. 1997 (TIS: 0.5 mg172 h) (2/8); Gresty et aI.. 1982 (SIS) Benztropine (single dose 2 mg Lv.) Downbeat: Barton et aI.. 1994 (2/2) APN: Barton et al., 1994 (SiS) Trihexyphenidyl (15-40 mg/day) Downbeat: Leigh et aI.. 1991 (111) APN: Herishanu and Louzoun. 1986 (III); Jabbari et al.. 1987 (414) OM: Herishanu and Zigoulinski. 1991 (111) Tridihexethyl (100 mg/day) APN: Leigh et al., 1991 (2/4) SSN: Leigh et al.. 1991 (Ill) Glycopyrrolate (single dose 0.2 mg Lv.) APN: Barton et at.. 1994 (3/5) Memantine (15-60 mg/day) APN: Starck et al., 1997 (11111 Propranolol Opsoclonus: SOM:
Fowler. 1976 (2 mg/kg/24 h) (2/2 infants) Tyler and Ruiz. 1990 (10 mg/day) (Ill); Brazis et al..1994 (dosage?) (1/3)
Methylphenidate (single dose 20 mg oral) SWJ: Currie et al.. 1986 (?) Acetazolamide (60-750 mg/day) Episodic ataxia: Wolf. 1980 (3/3); Zasorin et al.. 1983 (III); Griggs et al.. 1978 (3/3); Brunt and van Weerden, 1990 (8/12) Sulthiame (50-300 mg/day) Episodic ataxia: Brunt and van Weerden, 1990 (9/12); Wolf, 1980 (3/3) ACTH. corticosteroids Opsoclonus: PranzatelJi, 1992 Immunoglobulin Opsoclonus:
Petruzzi and de Alarcon. 1995 (1 child); Sugie et at.. 1992 (1 child)
Plasma exchange Opsoclonus:
Cher et at.. 1995 (3/6)
APN Acquired pendular nystagmus; CN congenital nystagmus; MSWJ macrosquare wave jerks; OM ocular myoclonus; PAN periodic alternating nystagmus; SSN seesaw nystagmus; SWJ square wave jerks.
BiittnerlFuhry
204
glycopyrrolate. All results were derived from single dose intravenous injections. Scopolamine led to a marked improvement in both patients tested, in one the nystagmus was eliminated completely. Benztropine was less e ective and glycopyrrolate deteriorated the nystagmus [IIJ. Thus, the authors demonstrated that only central acting anticholinergic drugs may be beneficial in the treatment of downbeat nystagmus. The di erential e ect of scopolamine and benztropine might be caused by the diverse profile of action on di erent muscarinic receptor subtypes (see above). Another double-blind study showed slight improvement with the centrally acting trihexyphenidyl[96]. In this investigation, medication was administered orally over a period of 1 month. Thus, at present, GABA-ergic drugs (baclofen, clonazepam) are the drugs of choice for the treatment of downbeat nystagmus with tolerable side e ects. However, it is quite clear that the drugs are not e ective in all instances and that an e ect could not be demonstrated in the only available controlled double-blind study [7J. Anticholinergic drugs might help, but so far an e ect has only been shown with intravenous administration which limits the use of long-term therapy. Side e ects are also more disturbing. For both GABAergic and anticholinergic drugs, long-term e ects are not known. Casuistic reports described single patients with improvement of their downbeat nystagmus after treatment with magnesium. This was particularly the case when magnesium depletion was diagnosed as the reason for the nystagmus [125]. Substitution of thiamine may also occasionally improve downbeat nystagmus [125J. None ective drugs: Gabapentin (up to 900 mg/day) had no e ect in 6 patients in a double-blind controlled study [7J. Drugs causing downbeat nystagmus: Cholinergic agents like physostigmine had a deteriorating e ect on the downbeat nystagmus in all 5 patients after administration of 1 mg intravenously [51J. Lithium intoxication can cause downbeat nystagmus [40, 160J. Anticonvulsant medication with phenytoin [2J or carbamazepine [158] is a common cause of downbeat nystagmus.
Upbeat Nystagmus
Clinical Aspects Like downbeat nystagmus, upbeat nystagmus usually does not disappear with fixatiun. Fulluwing Alexander's law, sluw phase velucity is greatest in upward gaze, though upbeat nystagmus can be transformed to downbeat nystagmus on upward gaze in some patients [1001. Oi erent from downbeat nystagmus, lateral gaze often does not enhance slow phase velocity. Convergence can increase and decrease upbeat nystagmus or convert it to downbeat
Drug Therapy of Nystagmus and Saccadic [ntrusions
205
nystagmus. The head-hanging position also increases upbeat nystagmus in some cases [100J. In contrast to downbeat nystagmus, which often lasts many years, upbeat nystagmus frequently subsides within weeks, requiring no specific treatment [138J. Upbeat nystagmus is less common than downbeat nystagmus. Lesions of the anterior cerebellar vermis and metabolic or toxic changes (drugs, tobacco, see below) as a cause have been described [100]. In contrast to downbeat nystagmus, lesions of the brainstem are frequent in patients with upbeat nystagmus. A critical structure for upbeat nystagmus is located at the midline in the lower medulla, but more rostal lesions have also been encountered [24J.
Pathophysiology and Experimental Studies Several clinical reports on upbeat nystagmus showing lesions of the medulla prompted to hypothesize similar to downbeat nystagmus an imbalance of central vertical vestibular tone to be the cause of upbeat nystagmus [20, 100]. As has been described for downbeat nystagmus, a erences from the posterior semicircular canals conveying vestibular signals which mediate downward eye movements follow pathways di erent from those deriving from the anterior semicircular canals mediating upward movements. In principle, tonic imbalance leading to upbeat nystagmus can be due to a relative decrease of anterior semicircular canal a erences or due to a relative increase of the signals from the posterior semicircular canals. Besides vertical vestibular tone imbalance a gaze-holding deficit with a shift of the null-position has also been considered as a cause [24]. Excitatory a erences originating in the anterior semicircular canals project via di erent pathways: after being relayed in the vestibular nuclei, signals are either transferred to the contralateral oculomotor complex through the brachium conjunctivum, the medial longitudinal fasciculus (MLF) or the ventral tegmentum. Lesions in both the brachium conjunctivum and the ventral tegmentum have been found in patients with upbeat nystagmus. A critical structure for upbeat nystagmus may be the paramedian tract in the lower medulla [24], which is located at the midline ventrally to the nucleus prepositus hypoglossi and possibly involved in vertical gaze-holding [29J. Bilateral muscimol injections into the iC can occasionally lead to upbeat nystagmus in the monkey which often has a torsional component [74J. Ketanest (a narcotic glutamate antagonist) elicited transient upbeat nystagmus in the monkey [51]. Patient Studies Dieterich et a1. [51J showed in an open study a long-lasting improvement with a reduction of the slow phase velocity after administration of baclofen for 2-4 weeks.
BiittnerlFuhry
206
Besides intramuscular or retrobulbar injection of botulinum toxin A [see chapter by LeighJ there are no other systematic studies showing an improvement of upbeat nystagmus. It is remarkable that thus far there is only one study on drug therapy for upbeat nystagmus. One reason may be that upbeat nystagmus in contrast to downbeat nystagmus usually reverses within a few weeks by itself. None ective drugs: Although tobacco. which has a cholinergic e ect. is known to produce upbeat nystagmus in healthy subjects, no therapeutical implications were derived using both cholinergic or anticholinergic acting drugs [138]. Drugs causing upbeat nystagmus: Upbeat nystagmus with tobacco consumption (cholinergic e ect) lasts for 10-20 min [128J.
Episodic Ataxia
Clinical Aspects Two groups of episodic ataxia can be distinguished. In type I episodic ataxia. paroxysms of ataxia are short (seconds to minutes) and deficits between paroxysms consist of general motor activity with myokymia and neuromyotonia. In type II the paroxysms of ataxia last longer (hours to days) usually combined with vertigo and between paroxysms pathological nystagmus is present. The nystagmus for type II consists of upbeat, downbeat and horizontal nystagmus during gaze straight ahead. Type I usually starts between the ages of 5-7. For type II the onset varies from early childhood to late in adult life. Spontaneous improvement or occasional remission but also progressive ataxia have been described for type II [21J. Pathophysiology and Experimental Studies Both type I and II episodic ataxia are autosomal-dominant disorders. Both types are ion channel disorders. type I a potassium channelopathy and type II a calcium channelopathy. Type II has been localized on chromosome 19p, I.e. the same region to which familial hemiplegic migraine and CADASIL (cerebral autosomal-dominant arteriopathy with subcortical infarcts and leukencephalopathy) is linked [68]. Interestingly, also familial hemiplegic migraine is often combined with nystagmus and ataxia. There is evidence that attacks in episodic ataxia are linked with abnormally high intracellular cerebellar pH values (which reduces putassium cumJuctance uf the cell membrane). Patient Studies Acetazolamide (a carbonic anhydrase inhibitor) reduces the pH and is e ective in more than 75% of the patients. who tolerate the side e ects [23. 67,
Drug Therapy of Nystagmus and Saccadic Intrusions
207
161, 164]. Prolonged treatment can lead to long-lasting improvement. Side e ects such as hyperhydrosis, paraesthesia, muscle sti ening and easy fatigability can be reduced by potassium chloride supplement [23]. Sulthiame, another carbonic anhydrase inhibitor, has also been a successful treatment. With less side e ects than acetazolamide and equal e ectiveness it is often the drug of choice [23J. A child was also successfully treated with the calcium entry blocker flunarizine (10 mg/day) [18]. None ectivedrugs: Phenobarbital, phenytoin, carbamazepine, betahistine, cinnarizine, clomipramine, clonazepam [23J. Worsening ofsymptoms has been reported under phenytoin and phenobarbital for a type II patient [67J.
Periodic Alternating Nystagmus
Clinical Aspects PAN is a well-defined and uniform oculomotor disorder. It consists of horizontal nystagmus with alternating directions while looking straight ahead. Reversal of nystagmus occurs every 70-150 s, Le. within a period of 2-3 min. One distinguishes between an acquired and a congenital form. The acquired form is mostly due to a lesion of the nodulus/uvula region of the cerebellar vermis as seen in posterior-fossa malformation. Loss of vision due to a vitreous hemorrhage or cataract, can also lead to PAN. The congenital form is considered as a special form of congenital nystagmus [100]. Pathophysiology and Experimental Studies PAN is closely associated with the 'velocity storage' mechanism of the vestibular system. 'Velocity storage' normally extends the postrotatory vestibular time constant and is under inhibitory control of the nodulus/uvula region. Lesions here have shown in experimental animals to induce slow nystagmus oscillations in the dark with reversing directions [155J. Vision normally has an inhibitory e ect on velocity storage, which would explain the occurrence of PAN in patients with eye disease (cataract, etc.) [97]. The inhibitory e ect of the nodulus/uvula region is directly transmitted to the vestibular nuclei via Purkinje cells (Pcs) , which use GABA as an inhibitory transmitter. In experimental studies orally administered baclofen [38J and picmtuxin [61J had an inhibitury e ect un the velucity sturage mechanism in normal monkeys. Diazepam mildly enhanced velocity storage [16J. This is remarkable, considering that baclofen (GABA B) and diazepam (GABA A ) are agonists and picrotoxin is a GABA A receptor antagonist. Baclofen eliminated the alternating nystagmus in monkeys with nodulus/uvula lesions [155].
BllttnerlFuhry
208
Patient Studies In 2 patients the acquired PAN was eliminated with baclofen in midposition and patients were relieved of oscillopsia [70). Carlow [33) and Nuti et al. [112J (1 patient with aperiodic alternating nystagmus) also reported improvement after baclofen. An e ect of baclofen can be seen within a few days [33). However, Furman et a1. [60) reported on 2 patients treated with 30 mg baclofen/ day, who did not improve. One patient had impaired vision and cerebellar vermis atrophy and the other a cerebellar atrophy. Congenital PAN, which might have a di erent cause, has been reported to improve [33, 84J or remain unchanged [65, 70) after baclofen therapy. Hydantoin can decrease the duration of the nystagmus cycle with a slight slowing of the oscillations [43J. Intravenous chlorpromazine or barbiturates only transiently eliminated the nystagmus [17J. There are no systematic studies on long-term e ects, which however have been reported in individual cases [33). One patient took baclofen successfully for 3 months, before the drug was discontinued due to indigestion [70). None ective drugs: Trials of meclizine or carbamazepine showed no improvement [43J. Atropine had no significant e ect in a patient with congenital PAN [9J. Drugs causing PAN: PAN has been observed after phenytoin [31) and after primidone/phenobarbital intoxication [127) (l patient each), which disappeared after toxic levels were discontinued.
Seesaw Nystagmus
Clinical Aspects Seesaw nystagmus (SSN) is a combination of a conjugate torsional and dissociated vertical nystagmus. While one eye is elevated and intorted, the other eye is synchronously depressed and extorted. In the second half of a full cycle the vertical and torsional movements reverse. SSN often has a pendular waveform and has been identified in mesodiencephalic lesions due to parasellar masses, head trauma, brainstem stroke, Arnold-Chiari malformation or septo-optic dysplasia [69J. SSN due to parasellar masses or head trauma is often associated with bitemporal hemianopia. It can also be congenital [118J. In unilateral mesencephalic lesions, jerk-waveform SSN has been observed [69J. Pathophysiology and Experimental Studies Olle critical structure ill patiellts withjerk-wavefurm SSN might ue the iC [69J. Electrical unilateral stimulation of the iC and its vicinity in the monkey resulted in a sustained ocular tilt reaction including a vertical divergent eye position and conjugate torsion [157J. However, unilateral muscimol injection into the iC in the monkey only led to downbeat and upbeat nystagmus in combination
Drug Therapy of Nystagmus and Saccadic [ntrusions
209
with ipsilesionally beating torsional nystagmus, but not to SSN [74]. Thus, the precise role of the mesencephalon for SSN has yet to be determined. Alternatively, lesions to the central vertical-torsional VOR and the otolithocular VOR have been proposed to cause jerk-waveform SSN by tonic imbalance. Lesions to the MLF can disrupt the connections between the vertical semicircular canals and the vertical-torsional eye muscles. The combination ofjerkwaveform SSN and internuclear ophthalmoplegia seen in patients, supports this hypothesis [54]. SSN in animal experiments has not yet been elicited.
Patient Studies SSN is a rare syndrome. Therefore most therapeutical data evolved from casuistic reports, which concentrated on GABA-ergic drugs. Carlow [33] described a patient with marked improvement after administration of baclofen. Other studies, however, could not demonstrate the beneficial e ect of baclofen (1 patient [37], 3 out of 3 patients [7]). Cochin et al. [37] demonstrated a benefit in 1 patient with clonazepam. SSN disappeared and did not return after withdrawal. A beneficial e ect of clonazepam was also reported by Currie and Matsuo [42] and Carlow [33]. Gabapentin (900 mg/day) also reduced median eye speed by more than 25% in 2 of the 3 patients [7]. There are two case reports on e cacy of ethanol in SSN [58, 10 1]. The benefit, however, only lasted for 1 h and the patient was noticeably inebriated at an ethanol intake of 1.2 g/kg body weight [58]. The dampening e ect of diazepam on SSN for a few minutes after intravenous administration [126] also supports the hypothesis that GABA-ergic transmission is important in SSN. There is a report on anticholinergic drugs in 1 patient with SSN [96], which showed marked decrement of nystagmus after administration of the peripherally acting tridihexethyl. The centrally acting trihexyphenidyl decreased slow phase velocity at gaze straight ahead but increased it in other directions. The results are limited due to the concomitant medication in this patient (carbamazepine, baclofen and protryptiline) which potentially could interact with the nystagmus. None ective drugs: Not known. Drugs causing SSN: Not known.
Congenital Nystagmus
Clinical Aspects Congenital nystagmus (CN) usually manifests itself in the first few months of life and only seldom becomes evident during adult life. The nystagmus has variable waveforms ranging from jerk to pendular nystagmus.
BllttnerlFuhry
210
It is almost always conjugate and horizontal. Vertical or torsional forms are rare exceptions. Usually there is an eye position in the orbit where the eye is quieter, i.e. the null-position, which often does not coincide with the midposition of the eye in the orbit. Another characteristic feature of CN is inverted OKN or inverted smooth pursuit eye movements. These findings supposedly result from interaction of CN and a dynamic shift of the null-position. with the smooth pursuit mechanisms being basically intact [90J. Although retinal image velocities in CN can exceed 100 0 /s. patients seldom complain about oscillopsia [50]. It is thought that in CN visual perception takes place mainly during short foveation periods (lasting about 50 ms). i.e. episodes during each nystagmus cycle in which the eye is not moving. To improve vision it is common that patients turn their head to bring the null-position of the eyes and gaze direction together. Patients might also benefit from convergence. Some patients show intermittent head-shaking, when attending to visual tasks. This stimulation of the VOR might improve vision in a few cases [95].
Pathophysiology and Experimental Studies The cause of CN is unknown. It may be familial. The well-known association with albinism is of interest. since these patients present evidence for wrongly directed visual pathways. This has also been proven in albino rabbits. which can exhibit inverted optokinetic nystagmus and nystagmus slow phases with increasing velocity as well. Visual depriviation in monkeys can cause ocular oscillations which resemble CN. Miswiring of the neural integrator leading to unstable gaze-holding has also been suggested as a cause of CN. Patient Studies Most patients do not require a specific therapy. If required it mostly consists of nondrug approaches such as surgery or biofeedback [see chapter by Leigh]. Bender [13J reported in 1946 that small doses of barbiturates given intravenously improved vision and decreased nystagmus in patients with CN. This was also seen in 1 patient with CN receiving phenobarbital [62J. The approach has not been routinely pursued due to the side e ects of barbiturates. Other drug trials include 5-hydroxytryptophan (antidepressant drug) [91J and uaclofen. which moderately improved vision and decreased nystagmus amplitude [163]. None ective drugs: Not known. Drugs causing eN: Not known.
Drug Therapy of Nystagmus and Saccadic [ntrusions
211
Acquired Pendular Nystagmus
Clinical Aspects APN is a spontaneous sinusoidal nystagmus of 3-7 Hz (amplitude 1-4°) which cannot be divided into fast and slow phases and is typically enhanced during fixation. The direction of APN is not restricted to the horizontal plane like in most patients with congenital forms of pendular nystagmus. Depending on the phase/amplitude relationship between di erent components, the resulting eye movements in APN are horizontal, vertical, oblique or circular/ elliptical. APN never occurs without additional brainstem or cerebellar signs. In contrast to CN, APN does not show inversion of the OKN. There is often a dissociation between the nystagmus found in both eyes and it can also be monocular. Di erent than CN, patients with APN almost always su er from osci1lopsia. APN is most commonly caused by demyelination, in adults as a consequence of multiple sclerosis or toluene abuse, in children by leukodystrophies. Pathophysiology and Experimental Studies Three lesion sites have been suggested leading to APN. Cerebellar midline structures are frequently lesioned in patients su ering from APN caused by multiple sclerosis. This is supported by the fact that electrical stimulation of cerebellar nuclei in man can evoke pendular nystagmus. On the other hand, disconjugacy of both eyes in APN has led to the hypothesis that structures in the vicinity of the oculomotor nuclei must be damaged in this disease. A third possible cause could be the dysfunction of the reciprocal feedback circuits between cerebellum and brainstem nuclei [8]. To date, there is no animal model simulating APN. Therefore, from both clinical and experimental data, it is still unclear what the critical structure for the occurrence of APN is. Patient Studies GABA-ergic, anticholinergic or antiglutaminergic medication has been tested. The e ect of the GABA-ergic system was tested in a double-blind controlled study (n 15) of baclofen vs. gabapentin [7]. Gabapentin significantly reduced median eye speed and enhanced visual acuity, whereas baclofen did nut shuw a beneficial e ect [7]. In anuther upen study, gabapentin was also e ective [133J. Clonazepamimproved APN in an open study [42], although no benefit was observed in 2 patients [66, 133J. Treatment with barbiturate [110J or ethanol consumption [108J can also diminish pendular nystagmus. Isoniazid eliminated APN in 2 out of 3 patients [152].
BiittnerlFuhry
212
Di erent anticholinergic agents were tested in six studies. In 1 patient with APN, after massive brainstem hemorrhage due to eclampsia, chronic trihexyphenidyl treatment markedly improved the pendular nystagmus [77]. The same therapeutic agent with higher dosages (up to 40 mg/day) also showed significant beneficial e ects in all 4 patients [86]. In a double-blind study, however, trihexyphenidyl failed to demonstrate an improvement in any of the 4 patients [96]. Tridihexethyl chloride, which does not cross the blood-brain barrier, improved visual acuity and reduced eye speed moderately in 2 out of 4 patients. It was proposed that anticholinergic agents suppress nystagmus by peripheral rather than central mechanisms [96J. In contrast, single-dose application of the centrally acting scopolamine (hyoscine) eliminated APN in all 5 patients in a double-blind controlled study [11]. This is in accordance with an earlier open study [66]. In an open study of scopolamine (TIS applied as a plaster for 3 days) APN was only reduced in 2 out of 8 patients [134J. Response failures to scopolamine have also been reported [133]. Benztropine, which crosses the blood-brain barrier and glycopyrrolate, which acts peripherally, had only a mild e ect [IIJ. As has been discussed above, the diverse profile of action on di erent muscarinic receptor subtypes might be responsible for the di erential e ect of scopolamine and benztropine [11]. In conclusion, there are controversial results for the e ect of anticholinergic agents on APN. One reason for the di erence between the results from the studies mentioned above might be the type of administration of the therapeutical agents. While single-dose intravenous application eliminates APN, long-term usage of related drugs is less beneficial. This might be due to a dose-dependent increase of receptor density after chronic exposure of muscarinic antagonists [12, 53]. Further double-blind studies have to be conducted in order to solve this discrepancy. In a recent open study the glutamate antagonist memantine eliminated APN in 11 patients after oral administration for 1 week [134J. A positive e ect of memantine was followed up for 3 years in 6 of the patients. None ective drugs: Two patients responding to gabapentin did not improve with valproic acid and baclofen [133J. The ine ectiveness of baclofen also agrees with other studies (14 patients) [7, 66J. The Na-channel blocker mexiletine did not show any significant e ect on APN in 4 patients [134J. Without supplying any detailed information, Gresty et al. [66] stated that there is no improvement of APN from L-DOPA, prochlorperazine, carbamazepine and tetraoenazine. Drugs causing acquired pendular nystagmus: Toluene abuse in sni ers can cause APN by di usely damaging the white matter [102, 122].
Drug Therapy of Nystagmus and Saccadic [ntrusions
213
Ocular Myoclonus
Clinical Aspects Ocular myoclonus (OM) consists of binocular, usually regular pendular eye movements of 1-3 Hz. It is always combined with rhythmic movements of nonocular muscles. Most often the soft palate is involved. In these instances, OM is also called oculopalatal myoclonus. The nystagmus frequently has a vertical direction and is not a ected by fixation. The movements of the nonocular muscles are often sychronized with the nystagmus. OM is caused by lesions ofthe inferior olivary nucleus (io) orits connections. Most frequently these lesions are due to bleedings and tumors as weU as due to strokes, trauma or inflammation. Characteristically, OM develops within months after the primary damage. OM seldom disappears spontaneously. Pathophysiology and Experimental Studies As described above, OM (like pure palatal myoclonus) can be found after destruction of the inferior olivary nucleus or its connections. Within months, hypertrophy of io can be found on MRI scans [132]. Vacuolated neurons causing the pseudohypertrophy are seen with postmortem histological workup. Increased levels of acetylcholinesterase reaction products have been found in these neurons [88J. Such neurons show cholinergic hyperexcitability and rhythmic spontaneous activity which is caused by damage to inhibitory dentatoolivary tracts [104J. 10 neurons are known to project to the cerebel1ar flocculus mediating adaptive properties of the VOR. Ocular oscillations in OM may be caused by an instability of this adaptive mechanism [109J. Electrical stimulation of io and its vicinity in the monkey led to palatal myoclonus, sometimes involving the eyeballs [156]. Patient Studies There are a variety of studies describing the e ect of medication on palatal myoclonus. As there is a di erential benefit on palatal myoclonus and concurrent OM (see below), results from studies on palatal myoclonus cannot simply be used to estimate e cacy of these drugs in OM [14, 92]. Cholinergic hyperexcitability of io has been shown to be a main factor in the evolution of OM. Therefore the use of anticholinergic agents should be the logical consequence. However, only one study in OM reported the successful administration of the anticholinergic drug trihexyphenidyl [78]. Treatment had to be discontinued due to psychomotor irritability. Amobarbital ceased nystagmus, but nonocular movements remained unchanged [14, 92J. Valproic acid was reported to strikingly decrease OM in 1 patient [94]. In a patient with palatal myoclonus it had no e ect [55]. Gaba-
BiittnerlFuhry
214
pentin decreased the mean slow phase velocity of OM by more than 25% in 2 patients in a double-blind, controlled study [7J. Another patient also improved after single oral doses of 600 mg gabapentin [133]. Unfortunately, all patients reported side e ects (feeling drunk, worsening of ataxia). There are a number of reports concerning treatment in palatal myoclonus without ocular oscillations. It has been supposed that both disorders, palatal and ocular myoclonus, have the same etiology. Although it has been shown that both deficits can respond di erently, some of the substances used for palatal myoclonus, which have not been tried for OM, will be mentioned. Combination of 5-hydroxytryptophan and carbidopa resulted in cessation of myoclonus in 2 patients [103, 159]. Carbamazepine had a beneficial e ect in 1 patient with palatal myoclonus [124], but worsened the clinical symptoms in another [55]. None ective drugs: Trifluoperazine (1 patient [36]), clonazepam (3 patients [33]), baclofen (3 patients [33], 2 patients [7]), diazepam (1 patient [124]) and diphenylhydantoin (1 patient [92]). Drugs causjng ocular myoclonus: Not known.
Opsoclonus and Ocular Flutter
Clinjcal Aspects Ocular flutter is defined as a series of involuntary back-to-back saccades in the horizontal plane without intersaccadic interval. For opsoclonus these pathological, involuntary eye movements do not only occur in the horizontal, but also in the vertical plane [26, 1OOJ. Originally the term opsoclonus referred to irregular, chaotic, conjugate and partly continuous eye movements. A bout of ocular flutter consists of 3-5 saccadic eye movements and lasts usually less than 1 s. Rarely more than 4 salves occur within a 10-s interval. As a rule, ocular flutter is only seen in adults. It usuaJly subsides spontaneously (without therapy) either within weeks or up to 5 months. In general. it presents as a uniform oculomotor pattern. In contrast, the eye movement patterns of opsoclonus are much more variable. In many instances the clinical picture (as well as the etiology, see below) is very similar to that of ocular flutter, except that the eye movements also have a vertical and/or an oblique component. In these instances, eye muvements between buuts are normal. This is in cuntrast tu patients with continuous opsoclonus. If cooperative, these patients can alter their gaze direction, otherwise there is little evidence for normal eye movements. Often patients with continuous opsoclonus are in severe clinical conditions. However, also alert and cooperative patients are not uncommon. Opsoclonus is seen more
Drug Therapy of Nystagmus and Saccadic [ntrusions
215
often in children than in adults. The course of opsoclonus is related to its etiology. For children as well as adults, paraneoplastic as well as viral infections are the most common cause for opsoclonus (and ocular flutter). In adults a toxic-metabolic disorder is also often present [26J. In parainfection cases, opsoclonus generally subsides within a week to a few months. Opsoclonus will disappear when the toxic substances are removed.
Pathophysiology and Experimental Studies Opsoclonus as well as ocular flutter are not elicited by a circumscribed brain lesion. Based on the frequent association with peripheral tumors and viral diseases, an autoimmune mediated oculomotor dyskinesia is discussed [117]. It is not known which neurons and transmitters in the brainstem and/ or cerebellum are involved. Earlier it was assumed that the lack of saccadic omnipause neurons in the brainstem would lead to saccadic oscillations. However, a more recent model supported by experimental evidence [87], showed that lesions to the omnipause area in the brainstem [30J actually only led to a slowing of saccades. Furthermore, with light microscopy the omnipause region appeared normal in the neuropathological examination of patients with opsoclonus [120J. Patient Studies In children, a neuroblastoma is the most common tumor associated with opsoc1onus. The detection of the neuroblastoma is often di cult and delays for years have been reported. When detected, removal is the therapy of choice and leads in approximately 50% of all cases to a complete remission of opsoclonus. Two-thirds of the children also respond to corticosteroids. Either ACTH or prednisone are used with doses similar to immunosuppressive therapy [117]. Hammer et al. [71J reported an improvement in children with ACTH but not with prednisone. Even if the eye movements improve under prednisone or ACTH, relapses during withdrawl or after discontinuation are common, prompting an increase or restart of therapy. Even with therapy, symptoms can last for several years. Parainfectious or idiopathic opsoclonus in children may also respond to ACTH [117J. In a few children treated so far. symptoms also improved after the application of immunoglobulin [114, 140J. In 2 infants unresponsive to corticosteroid treatment, propranolol (2 mg/kg/24 h) markedly improved all abnormal movements [56]. In genera!' ucular flutter is self-limiting and dues nut require a specific treatment. This also applies to opsoclonus of parainfectious origin. Adult patients with paraneoplastic opsoclonus usually die within months with the opsoclonus being present all the time. Also spontaneous remissions and recurrences have been reported for paraneoplastic opsoclonus.
BiittnerlFuhry
216
There is no established drug therapy for adult patients with paraneoplastic opsoclonus. In 3 patients an immunoadsorption therapy (plasma exchange) has been recently shown to be e ective in treating opsoclonus [35]. In analogy to children, corticosteriods have been applied successfully [76J. Drug trials were performed with clonazepam [3, 33, 99J. Clonazepam might help in cases which did not respond to corticosteroids or propranolol [33J. However, it might also be ine ective (1 patient [143]). Propranolol has shown to be e ective [56J as well as ine ective in 2 patients [33,143]. One patient with paraneoplastic opsoclonus improved with thiamine [111], which often has no e ect [3J. None ective drugs: Carbamazepine, primidone, baclofen, valproate, lisuride, methysergide, bromocriptine, cinnarizine and neuroleptic drugs [3J. Drugs causing ocular flutter or opsoc1onus: An overdose of amitriptyline [4], the combination of diazepam/phenytoin [48J and lithium/haloperidol [39] have been reported to cause opsoclonus. Gizzi et al. [64] reported ocular flutter after vidarabine.
Square Wave Jerks, Macro-Square Wave Jerks and Macrosaccadic Oscillations
Clinical Aspects All three disorders consist of saccades occurring at a high rate, which can lead to blurred vision. Patients do not complain of oscillopsia. Especially patients with square wave jerks (SWj) are often unaware of their involuntary eye movements. SWj consist of small saccades (0.5-5°) bringing the eyes away from fixation and after an intersaccadic interval of normal duration (about 200 ms) back to the starting point. SWj are found in normal subjects, slightly more common in the elderly. Frequencies over 25/min during fixation are considered pathological, which can be as high as 240/min. The amplitude should almost be the same for each saccade within a burst. For macro-square wave jerks (MSWj) the saccade amplitudes are larger (20-50°) and the intersaccadic intervals are shorter than 100 ms. MSWj occur in bursts with the amplitUde not varying within a single burst. They can also be continual with frequencies up to 120-180/min. MSWj are not seen in healthy subjects. Macro-saccadic oscillations (MSO) also have large amplitudes (20-60°) seen in bursts lasting several secunds. Huwever, in cuntrast tu MSWJ, the saccadic interval is longer (150-200 ms), the amplitude for MSO within a burst first increases and then decreases and the MSO take place around the point of fixation, whereas MSWj are to-and-fro saccades with regard to the point of fixation. All three of these forms of saccadic intrusions are mostly
Drug Therapy of Nystagmus and Saccadic Intrusions
217
restricted to the horizontal plane and are rarely seen with vertical or torsional directions [lOOJ.
Pathophysiology and Experimental Studies SWJ are seen in 20-40% of normal people. They have also been associated with a number of diseases, including progressive supranuclear palsy, multisystem degeneration, Parkinson's disease, Huntington's chorea, spinocerebellar degeneration or cortical cerebel1ar atrophia, schizophrenia and in connection with cerebral lesions [100]. Originally SWJ were considered as a cerebellar sign. However, since it has been shown that SWJ also occur with lesions outside the cerebellum and that cerebellar oculomotor deficits show no correlation with SWJ, this concept is no longer valid. In contrast to this, MSO and MSWJ can usually be related to disorders of the brainstem and cerebellum. MSO has been related to lesions of the midline cerebellum and the underlying deep cerebellar nuclei. They have been considered as a consequence of an increased gain in relation to visually guided saccades. But clearly severe saccadic hypermetria after bilateral deep cerebellar nuclei lesions can also occur without MSO [25, 27J. MSO has also been seen after a pontine lesion [5J. Recent experimental evidence has provided a model for SWJ related to the basal ganglia and the superior colliculus. The caudate nucleus projects to the nondopaminergic portion of the substantia nigra pars reticulata, which in turn has a tonic GABA-ergic inhibitory influence on the superior colliculus. Accordingly, injection of bicuculline into the superior colliculus leads to an increase of saccadic intrusions [80]. Patient Studies People with SWJ are usually not aware of their saccadic intrusions. If the frequency is not too high, leading to impaired vision, no specific treament is required. Dyslexic patients with SWJ have been successfully treated with methylphenidate [41]. There are no systematic studies about the long-term course of SWJ. In a patient who had both MSO and MSWJ, these eye movements were almost completely eliminated by clonazepam and phenobarbital [59]. In 1 patient with MSO the amplitude was reduced after a single dose of 600 mg gabapentin as well as the frequency by a single dose of 250 mg cycloserine, an antiuiotic that is a partial glycine agonist [6J. It has to ue considered however in these studies, that MSO and MSWJ are the consequence of an acute disease, mostly of the cerebellum, which usually disappears spontaneously. None ective drugs: Haloperidol and hydroxyzine pamoate had no e ect on MSO and MSWJ [59J.
BllttnerlFuhry
218
Drugs causing saccadic intrusions: L- Tyrosine, the precursor for dopamine biosynthesis, increased the frequency of saccadic intrusions in all 8 schizophrenic patients tested [49]. Depletion of catecholamines by metyrosine increased amplitude and frequency of saccadic intrusions in 3 normal subjects [153]. Tabacco smoking leads to saccadic intrusions during smooth pursuit eye movements [129].
Superior Oblique Myokymia
Clinical Aspects Superior oblique myokymia (SOM) is a recurrent episodical condition with short phasic contractions of the superior oblique muscle in one eye. This leads to small- and high-frequency monocular eye movements and as a consequence to vertical and torsional diplopia and oscillopsia. SOM is not a saccadic eye movement since there is no fixed relationship between amplitude and peak velocity of the movement (main sequence). SOM is most commonly found in otherwise normal patients. Occasionally SOM has been described in patients after su ering from superior oblique palsy. Lesions to the trochlear nucleus also seem to be a possible cause ofSOM in patients with posterior fossa tumors and subsequent tectal compression [47]. Pathophysiology and Experimental Studies Electromyographical examination in SOM showed polyphasic action potentials of high amplitude suggesting neurogenic disease. Neuronal activity in the antagonistic inferior oblique muscle was not reduced during SOM [89J. Therefore, SOM is unlikely to be caused by supranuclear lesions. As SOM followed su perior oblique palsy at intervals of months or years in some patients, misdirection-regeneration might be the cause of SOM in these patients [93]. As stated above, lesions to the trochlear nucleus itself can also lead to SOM [47]. Another cause could be a pathological contact between the trochlear nerve and a vessel, similar to hemifacial spasm [137]. Patient Studies The natural history of SOM typically shows a relapsing and remitting course for months to years. As a consequence, beneficial e ects of any treatment in SOM have tu be carefully interpreted. Anticonvulsant medication has been used most frequently in the medical treatment of SOM. There are a number of reports from patients responding to carbamazepine [22, 79, 107, 121, 141]. Unfortunately, if followed up for more than a year, most patients initially responding to carbamazepine only
Drug Therapy of Nystagmus and Saccadic [ntrusions
219
showed a temporarily beneficial e ect. In 6 out of 7 patients treated with carbamazepine, a prompt cessation in ocular symptoms was noted [121]. During the follow-up period of an average 31/ 2 years however, all patients su ered from at least one subsequent relapse [121J. Brazis et al. [22] reported that only 1 patient experienced a lasting benefit (more than 4 years) with carbamazepine. No improvement with carbamazepine was reported in all 3 patients of de Sa et al. [45] as well as in single patients [63, 135, 150, 154]. Propranolol (10 mg/ day) stopped SOM in a patient after carbamazepine did not show any benefit [154J. Transient improvement was also observed by Brazis et al. [22]. In conclusion, carbamazepine is the drug most often used. It is e ective for many patients, however relapses within 1-3 years are common. None ective drugs: Phenobarbital and chlordiazepoxide were orally administered in 1 patient without noticeable e ect [83]. The same patient also did not improve after intravenous edrophonium injection. Baclofen administered in a patient by Staudenmaier [135] showed no influence on SOM. Drugs causjng SOM: Not known.
Comment
Despite our detailed knowledge about the neurophysiology, neuroanatomy, neurochemistry and clinical aspects of central vestibular and oculomotor disorders [28, 46, 100], the drug therapy for nystagmus of central origin and saccadic intrusions still is only emerging and certainly many more studies are needed. Most reports are uncontrolled and based on single cases. Only a few double-blind controlled studies are available [7, 11, 96]. Since many of the described disorders are not very common, future studies have to be performed in international multicenter studies. There is certainly also the need for new and more e cient drugs. Single dose injections of anticholinergic drugs are e ective for APN [11], however comparable anticholinergic drugs for oral use needed for long-term therapy are still lacking [96]. Duwnbeat nystagmus has been shuwn tu respunu tu baclufen [51], yet many patients do not [7, 33]. When all studies are taken together a response has only been observed in 20% (2 out of 10) of the patients. That new drugs can provide promising results has recently been shown by using gabapentin [7] and the glutamate antagonist memantine [134] for treating APN.
Acknowledgments We would like to thank B. Pfreundner and 1. Wend I for preparing the manuscript and M. Seiche for editing it.
BllttnerlFuhry
220
References
I
2 3
4
5 6 7
8 9
10
11
12 13 14 15 16 17 18 19
20 21
22 23 24 25
Alonso R. Didier M, Soubrie P: [3H]N-methylscopolamine binding studies reveal M2 and MJ muscarinic receptor subtypes on cerebellar granule cells in primary culture. J Neurochem 1990;55;1 334-337. Alpert IN; Downbeat nystagmus due to anticonvulsant toxicity. Ann Neurol 1978;4:471-473. Anderson NE, Budde-Ste en C, Rosenblum MK, Graus F, Ford D, Synek BJ, Posner JB: Opsoclonus, myoclonus, ataxia, and encephalopathy in adults with cancer: A distinct paraneoplastic syndrome. Medicine (Baltimore) 1988;67: 100-109. Au WJ, Keltner JL; Opsoclonus with amitriptyline overdose. Ann Neurol 1979;6:87. Averbuch-Heller L, Leigh RJ: Eye movements. Curr Opin Neurol 1996;9:26-31. Averbuch-Heller L, Leigh RJ: Medical treatments for abnormal eye movements: Pharmacological, optical and immunological strategies. Aust NZ J Ophthalmol 1997;25:7-13. Averbuch-Heller L, Tusa RJ, Fuhry L, Rottach KC, Canser CL, Heide W, Buttner U, Leigh RJ; A double-blind controlled study of gabapentin and baclofen as treatment for acquired nystagmus. Ann Neurol 1997;41;818-825. Averbuch-Heller L, Zivotofsky AZ, Das YE, DiScenna AO, Leigh RJ; Investigations of the pathogenesis of acquired pendular nystagmus. Brain 1995;118:369-378. Baloh R\N, Honrubia Y, Konrad HR: Periodic alternating nystagmus. Brain 1976;99: 11-26. Barmack NH, Baughman R\N, Eckenstein FP, Shojaku H: Secondary vestibular cholinergic projection to the cerebellum of rabbit and rat as revealed by choline acetyltransferase immunohistochemistry, retrograde and orthograde tracers. J Comp Neurol 1992;317:250-270. Barton JJ, Huaman AC, Sharpe JA: Muscarinic antagonists in the treatment of acquired pendular and downbeat nystagmus: A double-blind, randomized trial of three intravenous drugs. Ann Neurol 1994;35:319-325. Ben-Barak J, Dudai Y: Scopolamine induces an increase in muscarinic receptor level in rat hippocamp"s. Rr-
>Il
-0.5 -1.0 0.0
0.2
0.4
0.6
Time (sec)
Fig 2. Suppression of nystagmus by changing the angle of vergence. The patient was a 41-year-old woman with multiple sclerosis. Representative records of the horizontal component of her acquired pendular nystagmus are shown as she viewed a target at far (top) or near (bottom). Note that her high-frequency nystagmus increased in amplitude at near. leading to complaints of blurred vision, especially during reading. Her near vision was improved by applying a 5-dpt base-in paste-on prism over her right spectacle lens (to induce divergence). Positive values correspond to rightward eye rotations.
high-plus spectacle lens wum in cumuinatiun with a high-minus cuntact lens. The system rests on the principle that stabilization of images on the retina could be achieved if the power of the spectacle lens focused the primary image close to the center of rotation of the eye. However, such images are defocussed, and a contact lens is required to extend back the focus onto the retina. Since
Nonpharmacological Treatment of Nystagmus
231
the contact lens moves with the eye, it does not negate the e ect of retinal image stabilization produced by the spectacle lens. With such a system it is possible to achieve up to about 90% stabilization of images upon the retina. There are several limitations to the system, however. One is that it disables all eye movements (including the vestibulo-ocular reflex and vergence), so that it is only useful while the patient is stationary and views monocularly. Another is that with the highest power components (contact lens of 58.00 dpt and spectacle lens of 32 dpt), the field of view is limited. Some patients with ataxia or tremor (such as those with multiple sclerosis) have di culty inserting the contact lens. However, initial problems posed by rigid polymethylmethacrylate contact lenses [Leigh et aI., 1988J have been overcome by development of gas- permeable [Yaniglos and Leigh, 1992J or even soft contact lenses. Most patients do not need the highest power components for oscillopsia to be abolished and vision to be useful. We have found that, in selected patients, the device may prove useful for limited periods of time, for example, if the patient wishes to watch a television program. Another technique that has been used experimentally to negate the visual consequences of nystagmus is to record the ocular oscillation and use this signal to move the visual stimulus in synchrony with the eyes. Thus, we have measured nystagmus with the magnetic search coil technique and used the electrical signal from the system's amplifiers to drive a mirror galvanometer that controlled the position of a visual stimulus (an optotype) projected onto a tangent screen [Leigh et a1., 1988J. We found that when we carried out this technique in the plane of maximum nystagmus oscillation, visual acuity improved - presumably because image slip was substantially reduced. However, to be of therapeutic value, a contactless, reliable method for recording eye movements would need to be employed and visual images also controlled in both horizontal and vertical directions. New infraredreflectance and video-based eye movement recording devices allow reliable measurement of both horizontal and vertical eye movements [DiScenna et aI., 1995J and might eventually be used to control the position of a visual signal on a video monitor. Such a system would have to allow for head movements (or the patient's head would have to be fixed), and there would also be need to be able to easily null any 0 set of the image from the fovea (which otherwise leads to a series of saccades as the eyes 'chase' the stabilized image). The issue of fading of images is, in practice, not a practical problem because electronically based systems are nut precise enuugh tu produce the high degrees of stabilization reqUired. Thus, as these technologies advance, a practical system for canceling out the visual e ects of nystagmus may become possible. However, such a system would also negate normal eye movements - like the optical device described above - unless a filter could
Leigh
232
be developed to di erentiate them from the abnormal oscillations. This ability also seems likely to become possible in the near future, perhaps using a 'neural network' approach.
Procedures for Weakening the Extraocular Muscles
One method to treat nystagmus that has gained popularity, has been injection of botulinum toxin either into the extraocular muscles or retrobulbar space [Crone et aI., 1984; Helveston and Pogrebniak, 1988]. Using both of these techniques, Ruben et a1. [1994] reported improvement of vision in most of their 12 patients with a variety of diagnoses; the major side e ect was ptosis. However, eye movements were not systematically measured and compared before and after injection. Repka et a1. [1994J also described improvement of vision following retrobulbar injection of botulinum toxin in 6 patients, and documented the e ects on eye movements. Their main reservation was the temporary nature of the treatment. We measured binocular eye rotations in three planes prior to and following monocular injection of botulinum toxin into the horizontal recti in 2 patients with acquired pendular nystagmus due to multiple sclerosis [Leigh et aI., 1992]. In both patients, the amplitude of the horizontal components was abolished (fig. 3), and visual acuity was slightly improved. However, the persisting vertical oscillations were more annoying. Furthermore, diplopia and ptosis were more annoying to the patients than visual symptoms due to the nystagmus. Finally, 1 patient reported a complication that illustrates the limitations of methods that aim to mechanically reduce ocular oscillations: the nystagmus got worse in the non-injected eye (compare fig. 3A and C). Since her better vision was in the injected eye, she preferred to use this to view her environment. The botulinum toxin had weakened all horizontal eye movements - not just those due to her nystagmus. This caused plastic-adaptive changes to take place, and the increased innervation caused saccades made by her left eye to be hypermetric (fig. 4A) and the vestibulo-ocular reflex, evaluated during rotation in darkness, to have increased gain for the left eye (fig. 4C). One other aspect of this result deserves further study: How were these plastic-adaptive changes related to an increase in the amplitude of the horizontal component of her nystagmus in the noninjected eye? This finding suggests that her acquired pendular nystagmus emanated eitller fwm une uf the eye movement systems that had undergone plastic-adaptive changes or, that the oscillations somehow arose from the adaptive mechanism itself. We also treated acquired nystagmus in 3 patients by injecting botulinum toxin into the retrobulbar space [Tomsak et aI., 1995]. Nystagmus was abol-
Nonpharmacological Treatment of Nystagmus
233
LEFT EYE
~
RIGHT EYE
0.5
00 .,
~
~
13 c..
0
...J
« u
vi
c.. ...J
0.0
« u
i=
'"
0.0
i=
'"
LLl
>
0.5
LLl
>
-0.5
..(l.S
B
A ..(l.S
0.0
0.5
..(l.s
HORIZONTAL POS. (deg)
00
0.5
13 c.. ...J
..:
0.5
HORIZONTAL POS. (deg)
00
~
0.0
0.5
~
vi
0
c.. ...J
0.0
«
u
u
'" >
'">
i=
0.0
J
i=
LLl
LLl
..(l.s
..(l.s
D
C ..(l.s
0.0
0.5
HORIZONTAL POS. (deg)
-0.5
0.0
0.5
HORIZONTAL POS. (deg)
Fig 3. E ects of botulinum toxin on acquired pendular nystagmus in a 27-year old woman with multiple sclerosis. A Oeft eye) and B (right eye) display representative I-second records of her nystagmus as 'scan paths' prior to injection with botulinum toxin. C and D display characteristics of her nystagmus, I week after injection of the right medial rectus and 2 weeks after injection of the right lateral rectus muscle. The horizontal component of nystagmus in the right eye was almost abolished, and visual acuity increased from 20/40 ' to 20/25 3 in this eye. The amplitude of the horizontal component of nystagmus in the left, noninjected eye has increased, however, and visual acuity declined from 20/70 to 20/100. Positive values correspond to rightward or upward eye rotations. [See Leigh et aI., 1992, for details.]
ished ur reduced ill the treated eye fur a!.Juut 2-3 mUIlths, !.Jut ptusis alld diplopia were even more troublesome than when botulinum was injected into the extraocular muscles. Furthermore, 1 patient developed filamentary keratitis, perhaps due to denervation of the ciliary ganglion, that has persisted for several years. No patient that we treated with retrobulbar botulinum toxin
Leigh
234
LEFT EYE
40
00 20
00 ... 20
Z
Z
~
9
E-
Vi
0
E
O
0
'"2
0..
UJ
A
TARGET
~
0
>UJ
RIGHT EYE
40
UJ
>UJ
-20
B
-40 0
2
4
-20
-40
6
6
0
TIME (sec)
TIME (se ~
o·
z o t=
iii oc. w
10 -1 VIBRATION ON NECK
-2
-3 '--_--'-_----'--_ _'--_--'-_-----'-_ _'--_---'-_-----'-_ _-'----------J 7 8 9 10 4 5 6 o 2 3 TIME (sec)
Fig 5. E ects of applying a vibratory stimulus to the neck of a patient with congenital nystagmus. The peak-to-peak amplitude of the nystagmus was reduced by over 60%, and gaze was stable enough for the image of the object of regard to remain within a 'foveation window' of 0.5°. [See Sheth et a!., 1995, for details.]
Application of Somatosensory or AUditory Stimuli to Suppress Nystagmus
Contact lenses alone sometimes suppress congenital nystagmus; this e ect is not due to the mass of the lenses but is probably mediated via trigeminal a erents [Dell 'Ossu et aI., 1988). A variety uf utiler treatments have been reported, principally for congenital nystagmus. Electrical stimulation or vibration over the forehead or neck may suppress congenital nystagmus [Sheth et aI., 1995); an example is shown in fig. 5. It is postulated that this e ect, as well as wearing contact lenses [Dell'Osso et al., 1988), may be exerted
Nonpharmacological Treatment of Nystagmus
237
via the trigeminal system, which receive extraocular proprioception. Not all patients gain benefit from this procedure and, in some, nystagmus is made worse. Acupuncture administered to the neck muscles may suppress congenital nystagmus in some patients, perhaps due to a similar mechanism [Ishikawa et aI., 1987]. Biofeedback has also been reported to help some patients with congenital nystagmus [Abadi et aI., 1980; Ciu reda et aI., 1982]. However, the role of any of these treatments outside the laboratory - during natural activities - have yet to be demonstrated.
Conclusions
Although the strategy of weakening the extraocular muscles is an inherently appealing approach for treating nystagmus, it often induces diplopia and provokes adaptive changes that may actually counteract the treatment. Optical devices that partially stabilize images on the retina despite eye movements provide a monocular, limited view of the world. Both of these approaches abolish the capacity to make eye movements that compensate for head movements and so make vision wor.se while patients are in motion. If nystagmus can be suppressed with convergence, then this may provide substantial benefit so that, for example, driving is possible. Methods that use somatosensory or auditory stimuli to suppress nystagmus are unreliable, especially during natural activities. Thus, all these methods are potentially inferior to pharmacological suppression of the ocular oscillations. Because e ective pharmacological treatment for some forms of nystagmus is not yet available, these alternative methods still require consideration and, with careful evaluations, may be found to appreciably help selected patients.
Acknowledgments Supported by USPHS grant EY067J7, the Department of Veterans A airs, and the Evenor Armington Fund.
References Abadi RV, Carden D, Simpson J: A new treatment for congenital nystagmus. Br J Ophthalmol 1980;64: 2-6. Anderson JR: Causes and treatment of congenital eccentric nystagmus. Br J OphthalmoI1953;37:267-281. Bender MB: Oscillopsia. Arch Neurol 1965; 13:204-213. Buchele W, Brandt T, Degner D: Ataxia and oscUlopsia in downbeat-nystagmus vertigo syndrome; in Pfaltz CR (ed): Neurophysiological and Clinical Aspects of Vestibular Disorders. Adv Otorhinolaryngo!. Basel, Karger, 1983, vol 30, pp 291-297.
Leigh
238
Burr DC. Morrone MC, Ross]: Selective suppression of the magnocellular visual pathway during saccadic eye movements. Nature 1994;371:511-513. Burr DC, Ross]: Contrast sensitivity at high velocities. Vision Res 1982;22:479-484. Carpenter RHS. The visual origins of ocular motility; in Cronly-Dillon JR (ed): Vision and Visual Function, vol 8: Eye Movements. London, Macmillan Press. 1991, pp 1-10. Campbell FW, Wurtz RH: Saccadic omission: \Nhy we do not see a grey out during a saccadic eye movement. Vision Res 1978;18:1297-1303. Ciu reda KJ, Goldrich SG, Neary C: Use of eye movement auditory biofeedback in the control of nystagmus. Am J Optom Physiol Optics 1982;59:396-409. Crone RA, de Jong PTVM, Notermans G: Behandlung des Nystagmus durch Injektion von Botulinustoxin in die Augenmuskeln. Klin Monatsbl Augenheilkd 1984;184:216-217. COppers C: Probleme der operativen Therapie des okularen Nystagmus. Klin Monatsbl Augenheilkd 1971; 159: 145-157. Dell'Osso LF: Improving visual acuity in congenital nystagmus; in Smith JL, Glaser JS (ed): Neuroophthalmology. St Louis, Mosby, 1973, vol 7, pp 98-106. Dell'Osso LF: Extraocular muscle tenotomy, dissection, and suture: A hypothetical therapy for congenital nystagmus, J Pediatr Ophthalmol Strabismus 1998. in press. Dell'Osso LF. Flynn JT: Congenital nystagmus surgery. A quantitative evaluation of the e ects. Arch Ophthalmol 1979;97:462-469. Dell'Osso LF, Leigh RJ; Ocular motor stability of foveation periods. Required conditions for suppression of oscillopsia. Neuroophthalmology 1992; 12:303-326. Dell'Osso LF, Traccis S, Abel LA, Erzurum SI: Contact lenses and congenital nystagmus. Clin Vis Sci 1988;3:229-232. D'Esposito M, Reccia R. Roberti G, Russo P: Amount of surgery in congenital nystagmus. Ophthalmologica 1989;198:145-151. Dickinson CM, Abadi RV: The influence of nystagmoid oscillation on contrast sensitivity in normal observors. Vision Res 1985;25:1080-1096. DiScenna AO, Das VE, Zivotofsky AZ, Seidman SH, Leigh RJ; Evaluation of a video tracking device for measurement of horizontal and vertical eye rotations during locomotion. J Neurosci Methods 1995;58;89-94, Flynn JT, Dell'Osso LF: Surgery of congenital nystagmus, Trans Ophthalmol Soc UK 1981;101:431-433, Helveston EM. Ellis FD. Plager DA: Large recession of the horizontal recti for treatment of nystagmus. Ophthalmology 1991;98:1302-1305. Helveston EM, Pogrebniak AE: Treatment of acquired nystagmus with botulinum A toxin. Am J Ophthalmol 1988; 106:584-586. Hoyt WF. Keane JR: Superior oblique myokymia: Report and discussion on five cases of benign intermittent uniocular microtremor, Arch OphthalmoI1970;84;461-467. JC: LiVing without a balancing mechanism, N Engl J Med 1952;246:458-460, Ishikawa S, Ozawa H, Fujiyama Y: Treatment of nystagmus by acupunture; in Highlights in Neuroophthalmology. Proceedings of the Sixth Meeting of the International Neuro-Ophthalmology Society (INOS). Amsterdam, Aeolus Press, 1987, pp 227-232. Kaufmann I I, Kolling G: Operative Therapie bei Nystagmuspatienten mit I3inokularfLmktionen mit und ohne Kopfzwangshaltung. Ber Dtsch Ophthalmol Ges 1981;78:815-819. Kestenbaum A: Nouvelle operation de nystagmus. Bull Soc Ophthalmol Fr 1953;6:599-602. Kommerell G, Schaubele G: Superior oblique myokymia: An electro-myographical analysis. Trans Ophthalmol Soc UK 1980; 100:504-506. Kosmorsky GS, Ellis BD, Fogt N, Leigh RJ: The treatment of superior oblique myokymia utilizing the Harada-Ito procedure. J Neuroophthalmol 1995; 15: 142-146. Lavin PJM, Traccis S, Dell'Osso LF, Abel LA, Ellenberger C Jr: Downbeat nystagmus with a pseudocycloid waveform: Improvement with base-out prisms. Ann Neurol 1983;13:621-624. Leigh RJ, Brandt Th; A re-evaluation of the vestibulo-ocular reflex; New ideas of its purpose, properties. neural substrate. and disorders. Neurology 1993;43:1288-1295. Leigh RJ. Rushton ON, Thurston SE, Hertie RW. Yaniglos SY: E ects of retinal image stabilization on acquired nystagmus due to neurological disease. Neurology 1988;38:122-127.
Nonpharmacological Treatment of Nystagmus
239
Leigh Rj, Tomsak RL, Grant MP, Remler BF, Yaniglos SS, Lystad L, DelrOsso LF: E ectiveness of botulinum toxin administered to abolish acquired nystagmus. Ann Neurol 1992;32:633-642. Leigh Rj, Tomsak RL, Seidman SH, DelrOsso LF: Superior oblique myokymia. Quantitative characteristics of the eye movements in three patients. Arch Ophthalmol 1991;109:1710-1713. Ott 0, Seidman SH, Leigh Rj: The stability of human eye orientation during visual fixation. Neurosci Lett 1992; 142: 183-186. Pedersen RA, Troost BT, Abel LA, Zorub, 0: Intermittent downbeat nystagmus and oscillopsia reversed by suboccipital craniectomy. Neurology 1980;30:1239-1242. Repka MX, Savino PJ, Reinecke RD: Treatment of acquired nystagmus with botulinum neurotoxin A. Arch Ophthalmol 1994;112:1320-1324. Ruben ST, Lee JP, O'Neil 0, Dunlop I, Elston JS: The use of botulinum toxin for treatment of acquired nystagmus and oscillopsia. Ophthalmology 1994;101:783-787. Rushton 0, Cox N: A new optical treatment for oscillopsia. J Neurol Neurosurg Psychiatry 1987;50: 411-415. Sendler S, Shallo-Ho mann J, MOhlendyck H: Die Artifizielle-Divergenz-Operation beim kongenitalen Nystagmus. Fortschr Ophthalmol 1990;87:85-89. Sheth NV, Delrosso LF, Leigh Rj, Van Doren CL, Peckham HP: The e ects of a erent stimulation on congenital nystagmus foveation periods. Vision Res 1995;35:2371-2382. Spooner JW, Baloh RVV: Arnold-Chiari malformation. Improvement in eye movements after surgical treatment. Brain 1981; 104 :51-60. Susac JO, Smith JL, Schatz NJ; Superior oblique myokymia. Arch Neurol 1973;29:432-434. Tomsak RL, Remler BF, Averbuch-Heller L, Chandran M, Leigh Rj: Unsatisfactory treatment of acquired nystagmus with retrobulbar botulinum toxin. Am J Ophthalmol 1995;119:489-496. von Noorden GK, Sprunger DT: Large rectus muscle recession for the treatment of congenital nystagmus. Arch OphthalmoI1991;109:221-224. Yaniglos SS, Leigh Rj: Refinement of an optical device that stabilizes vision in patients with nystagmus. Optom Vis Sci 1992;69:447-450. Zubkov AA, Stark N, Weber A, Wizov SS, Reinecke RD: Improvement of visual acuity after surgery for nystagmus. Opht ha lmology 1993;100: 14 88-1497.
R. John Leigh, MD, Department of Neurology, University Hospitals, 11100 Euclid Avenue, Cleveland, 01-1, 44106-5000 (USA) Tel. 1 2168443190, Fax 1 2168443160
Leigh
240
Subject Index
Abducens nucleus (VI) a erent and e erent connections 10 structure and function 9, 10 transmitters 10 Acetazolamide, episodic ataxia treatment 204, 207, 208 Acoustic neuroma, di erential diagnosis ISS, 156 Acquired pendular nystagmus clinical aspects 212 drugs causes of nystagmus 213 therapy 212, 213 pathophysiology and experimental studies 221 Acupuncture, nystagmus suppression 238 Acute hearing loss, di erential diagnosis 155 Acyclovir, vestibular neuritis management 129, 133 Adenosine triphosphate, sensitivity of central vestibular neurons 70 Adrenaline, modulation of central vestibular neurons 52 Adrenocorticotropin, modulation of central vestibular neurons 69 Aminoglycosides, Meniere's disease management 160, 161 Amobarbital, nystagmus treatment 203,214 AMPA receptors medial vestibular nucleus neurons 40 principles of drug therapy 199 Apoplexia labyrinthi, di erential diagnosis 155
Baclofen, nystagmus treatment 202, 203, 205, 206, 209, 212, 220 Benign paroxysmal positioning vertigo canalolithiasis in etiology 168, 169, 176-178 clinical presentation 170, 171 cupulolithiasis in etiology 174-176 diagnosis 171, 173 di erential diagnosis ISS, 173, 174 horizontal benign paroxysmal positioning vertigo atypical disease with ageotropic positional nystagmus 188, 189 clinical presentation 187, 188 conversion from posterior disease 187 etiology and pathomechanism 189, 190 incidence 170, 187 management 190, 191 incidence and age dependence 169, 180 natural course 173 nystagmus 173 physical therapy 170 Brandt-Daro exercise 180, 181, 185, 186 Epley maneuver 183, 186 recurrence following therapy 183, 185 Semont maneuver 181, 183, 186 surgical management guidelines 186 neurectomy, posterior ampuJlary nerve 186
241
Benign paroxysmal positioning vertigo, surgical management (continued) plugging of posterior semicircular canal 186, 187 unilateral versus bilateral disease 180 Benztropine, nystagmus treatment 204, 205 Botulinum toxin, nystagmus treatment 207, 233-235 Brandt-Daro exercise, benign paroxysmal positioning vertigo milnilgement 180,181, 185, 186 Canalolithiasis, see Benign paroxysmal positioning vertigo Carbamazepine, superior oblique myokymia treatment 203, 219, 220 Cerebellum a erent and e erent connections 13, 14 structure and function 12, 13 transmitters climbing fibers 16 in terneu rons 16 mossy fibers 16 Purkinje cells 14, 16 varicose fibers 17 Clonazepam, nystagmus treatment 202,203, 205, 212, 218 Cochleosacculotomy, Meniere's disease management 161, 162 Congenital nystagmus clinical aspects 210,211 drug therapy 211 pathophysiology and experimental studies 211 Contact lenses, nystagmus suppression 237, 238 Corticosteroids opsoclonus management 216,217 vestibular neuritis management 129 Cupulolithiasis, see Benign paroxysmal positioning vertigo Dimenhydrinate, Meniere's disease management 158, 159 Diphenylhydantoin, nystagmus treatment 203, 209
Subject Index
Dopamine, modulation of central vestibular neurons agonists in therapy 60 central pathways 52, 57, 58 electro physiological studies 58, 60 receptors 58, 60 Downbeat nystagmus clinical aspects 201 drugs causes of nystagmus 205 thempy 202, 205 pathophysiology and lesion studies 201, 202 Dramamine, vestibular neuritis management 129 Electrocochleography, Meniere's disease 151, 152 Endolymphatic hydrops, see Meniere's disease Episodic ataxia clinical aspects 207 drug therapy 207, 208 pathophysiology and lesion studies 207 Epley maneuver benign paroxysmal positioning vertigo management 183, 186 Eye movement generation abducens nucleus 9, 10 cerebellum 12-14, 16, 17 fastigial nucleus 17-19 interstitial nucleus of Cajal 4-6 oculomotor nucleus 10-12 paramedian pontine reticular formation 6-8 paramedian tract neurons 8, 9 rostral interstitial nucleus of the medial longitudinal fascicle 1-3 Fastigial nucleus a erent and e erent connections 18 lesion studies of eye movement 12, 13 structure and function 17, 18 transmitters 19 Gabapentin, nystagmus treatment 203, 210, 215,220
242
GABA receptors medial vestibular nucleus neurons anatomical studies 46, 47 electrophysiological studies 47 functional roles 49 principles of drug therapy 198-200 Gigantocellular reticular nucleus neurons, membrane properties and gaze control 38, 39 Glycine receptors mediul vestibulur nucleus neurons anatomical studies 46, 47 electrophysiological studies 47 functional roles 49 principles of drug therapy 198, 200 Glycopyrrolate, nystagmus treatment 204 Herpesviruses Meniere's disease 143 vestibular neuritis 126, 127 Histamine, modulation of central vestibular neurons binding sites 53 central pathways 52, 53 electrophysiological studies 53, 54 receptor antagonists 54, 55 -Histamine, Meniere's disease management 159 Horizontal benign paroxysmal positioning vertigo, see Benign paroxysmal positioning vertigo Interstitial nucleus of Cajal a erent and e erent connections 4, 5 structure and f,mction 4 transmitters 5, 6 Isoniazid, nystagmus treatment 203 Labyrinthectomy, Meniere's disease management 163 Lermoyez syndrome, see Meniere's disease Long-term depression, cerebeUar regulation 17 Macrosaccadic osciUations clinical aspects 217, 218 drugs causes of oscillations 219
Subject Index
therapy 218 pathophysiology and experimental studies 218 Macro-square wave jerks clinical aspects 217, 218 drugs causes of jerks 219 therapy 218 pathophysiology and experimental studies 218 Mugnetic resomllce imuging Meniere's disease diagnosis 154 vestibular neuritis cliagnosis 117, 120, 121 Medial vestibular nucleus neurons intrinsic membrane properties and functional speculations 36-38 in vivo recorclings 29, 30 modulation of central vestibular neurons adrenaline 52 dopamine agonisL~ in therapy 60 central pathways 52, 57, 58 electrophysiological stuclies 58, 60 receptors 58, 60 histamine binding sites 53 central pathways 52, 53 electrophysiological stuclies 53, 54 receptor antagonists in therapy 54, 55 noradrenaline central pathways 52, 61 electrophysiological stuclies 62, 63 receptors 61, 62 serotonin central pathways 55 electrophysiological studies 55-57 receptors 55 neuropeptides in central vestibular network regulation adrenocorticotropin 69 nerve growth factor 69 neurokinins 66, 69 opioid peptides 65, 66 somatostatin 64, 65 substance P 66, 67, 69
243
Medial vestibular nucleus neurons (continued) neurotransmitters AMPA receptors 40 classification 39 GABA and glycine receptors anatomical studies 46, 47 electro physiological studies 47 functional roles 49 isolation of electrophysiological e ects 39,40 metabotropic glutamate receptors 40 muscarinic and nicotinic receptors anatomical studies 50 behavioral studies 51, 52 electrophysiological studies 50, 51 NMDA receptors functional plasticity role 46 functional roles and correlation with in vivo data 43 postlesional plasticity role 45, 46 subunits 40 pharmamlogical analysis of excitatory amino acid transmission 42, 43 specificity for neuron types 64 operation relationship to behavior 27, 28 oscillatory behavior and functional speculations 35, 36 plasticity 26 purine receptors and ATP sensitivity 70 response to horizontal angular accelerations 98-100 slice recordings classification of neuron types 30, 31 rhythmic activity in type B neurons 31,32 whole-brain recordings, in vitro 32, 33, 35 Memantine, nystagmus treatment 204, 213, 220 Meniere's disease clinical course 140, 141 definition 137 diagnosis audiological testing 149, 150 diagnostic scale 137, 138 di erential diagnosis acoustic neuroma 155, 156 acute hearing loss 155
Subject Index
apoplexia labyrinthi 155 benign paroxysmal positioning vertigo 155 overview 137 perilymph fistula 156 pharmacological side e ects 156, 157 vertebrobasilar insu ciency 156 vestibular neuritis 154 magnetic resonance imaging 154 endolymphatic hydrops pathophysiological model 145, 146 epidemiology 141 etiology of endolymphatic hydrops autoimmunity 143-145 genetics 142 overview 141, 142 psychological factors 142 vascular compression 142 viral factors 143 history of research 138-140 Lermoyez syndrome 153 signs and symptoms aural filllness 152 cochlear function 149-152 oculomotor symptoms 148 subjective complaints 147, 148 tinnitus 152 vestibular-spinal signs 149 treatment aminoglycoside therapy 160, 161 decompression surgery cochleosacculotomy 161, 162 saccotomy 162, 163 transtympanic ventilation tubes 161 drug therapy 158-160 guidelines 158 labyrinthectomy 163 neurectomy of vestibular nerve 163,164 symptom relief 157 Tumarkin otolith crisis 153 Metabotropic glutamate receptors medial vestibular nucleus neurons 40 principles of drug therapy 199 Methylphenidate, nystagmus treatment 204 Muscarinic receptors medial vestibular nucleus neurons anatomical studies 50
244
behavioral studies 51, 52 electrophysiological studies 50, 51 principles of drug therapy 200, 201 Nerve growth factor, modulation of central vestibu lar neurons 69 Neurectomy posterior ampullary nerve 186 vestibular nerve in Meniere's disease management 163, 164 Neurokinins, moduliltion of centml vestibular neurons 66, 69 Nicotinic receptors medial vestibular nucleus neurons anatomical studies 50 behavioral studies 51, 52 electrophysiological studies 50, 51 principles of drug therapy 200, 201 NMDA receptors, medial vestibular nucleus neurons functional plasticity role 46 functional roles and correlation with in vivo data 43 postlesional plasticity role 45, 46 subunits 40 Noradrenaline, modulation of central vestibular neurons central pathways 52, 61 electrophysiological studies 62, 63 receptors 61, 62 Nystagmus see also Acquired pendular nystagmus, Congenital nystagmus, Downbeat nystagmus, Episodic ataxia, Periodic alternating nystagmus, Seesaw nystagmus, Spontaneous nystagmus, Upbeat nystagmus benign paroxysmal positioning vertigo 173, 188, 189 decaying of slow phase 196, 197 phases 195-197 treatment acupuncture 238 base-in prisms 230 botulinum toxin 207, 233-235 contact lenses 237, 238 drugs, see specific diseases and drugs
Subject Index
optical methods for negating visual consequences 230-233 surgery 236 vibratory stimuli 237 velocity reduction and oscillopsia elimination 195, 228, 229 visual consequences 228-230 Ocular flutter clinical aspects 215, 216 drugs causes of flutter 217 therapy 216,217 pathophysiology and experimental studies 216 Ocular myoclonus clinical aspects 214 drugs causes of myoclonus 215 therapy 214, 215 pathophysiology and experimental studies 214 Ocu lomotor nucleus a erent and e erent connections 11 structure and function 10, 11 transmitters 11, 12 Opioid peptides, modulation of central vestibular neurons 65, 66 Opsoclonus clinical aspects 215, 216 drugs causes of opsoclonus 217 therapy 216,217 pathophysiology and experimental studies 216 Oscillopsia causes 196 elimination by nystagmus velocity reduction 195, 228, 229 treatment 197, 198 Paramedian pontine reticular formation a erent and e erent connections 7,8 saccade generation 197 structure and function 6, 7 transmitters 8
245
'-"'-1 ............ '-' ........ '-' JV'
185. 186
Epley maneuver 183. 186 recurrence after therapy 183. 185 Semont maneuver 181, 183, 186 vestibular neuritis management 129, 130, 132
Positional vertigo. see Benign paroxysmal positioning vertigo Prednisolone. Meniere's disease management 159. 160 Prepositus hypoglossi nucleus neurons. membrane properties and gaze control 38. 39
.~
.....
clinical aspects 217. 218 drugs causes of jerks 219 therapy 218 pathophysiology and experimental studies 218 Substance P. modulation of central vestibular neurons 66. 67. 69 Superior oblique myokymia clinical aspects 219 drug therapy 219. 220 pathophysiology and experimental studies 219
Propranolol. nystagmus treatment 204 Rostral interstitial nucleus of the medial longitudinal fascicle a erent and e erent connections 2 saccade generation 197 structure and flJnction 1. 2 transmitters 2. 3 Saccotomy. Meniere's disease management 162. 163
Scopolamine nystagmus treatment 204 vestibular neuritis management 129 Seesaw nystagmus clinical aspects 209 drug therapy 210
Subject Index
Transtympanic ventilation tubes, Meniere's disease management 161 Tridihexethyl, nystagmus treatment 204,210,213
Tumarkin otolith crisis. see Meniere's disease Unilateral vestibular loss see also Vestibular neuritis clinical features and mechanism asymmetrical horizontal vestibuloocular response 85 maintained rolled ocular torsional position 85 spontaneous nystagmus 85 consequences. direct versus indirect 84-86
246