Viral Neuropathies in the Temporal Bone Karger .pdf

Nom original: Viral Neuropathies in the Temporal Bone_Karger.pdf

Ce document au format PDF 1.6 a été généré par Adobe Acrobat 7.05 / Adobe Acrobat 7.05 Image Conversion Plug-in, et a été envoyé sur le 14/08/2011 à 16:12, depuis l'adresse IP 41.107.x.x. La présente page de téléchargement du fichier a été vue 1233 fois.
Taille du document: 4.6 Mo (150 pages).
Confidentialité: fichier public

Télécharger le fichier (PDF)

Aperçu du document

Viral Neuropathies in the Temporal Bone

Advances in
Vol. 60

Series Editor

W. Arnold


Viral Neuropathies in
the Temporal Bone

Richard R. Gacek Mobile, Ala
Mark R. Gacek Mobile, Ala

100 figures, and 8 tables, 2002

Basel ⭈ Freiburg ⭈ Paris ⭈ London ⭈ New York ⭈
New Delhi ⭈ Bangkok ⭈ Singapore ⭈ Tokyo ⭈ Sydney

Prof. Richard R. Gacek,
Prof. Mark R. Gacek
Division of Otolaryngology, Head and Neck Surgery
College of Medicine, University of South Alabama
307 University Blvd., N
HSB Suite 1600
Mobile AL 36688– 0002 (USA)

Library of Congress Cataloging-in-Publication Data
Gacek, Richard R.
Viral neuropathies in the temporal bone / Gacek, Richard R., Gacek, Mark R.
p. ; cm. – (Advances in oto-rhino-laryngology, ISSN 0065-3071 ; v. 60)
Includes bibliographical references and index.
ISBN 3805572956
1. Temporal bone–Diseases. 2. Facial nerve–Diseases. 3. Vestibular
apparatus–Diseases. 4. Virology. I. Gacek, Mark R. II. Title. III. Series.
[DNLM: 1. Temporal Bone–innervation. 2. Temporal Bone–virology. 3. Facial Nerve
Diseases–virology. 4. Vestibulocochlear Nerve Diseases–virology. WV 201 G121v2002]
RF260 .G334 2002

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents® and
Index Medicus.
Drug Dosage. The authors and the publisher have exerted every effort 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, including photocopying, recording, microcopying,
or by any information storage and retrieval system, without permission in writing from the publisher.
© Copyright 2002 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland)
Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel
ISSN 0065–3071
ISBN 3–8055–7295–6


VII Preface
XI Acknowledgment
Chapter 1

3 The Biology of Neurotropic Viruses
Gacek, R.R. (Mobile)
Chapter 2

32 Neuroanatomy of the Nerves in the Temporal Bone
Gacek, R.R. (Mobile)
Chapter 3

32 Meatal Ganglionitis: A Pathologic Correlate in Idiopathic Facial
Gacek, R.R.; Gacek, M.R. (Mobile)
Chapter 4

54 Vestibular Neuronitis: A Viral Neuropathy
Gacek, R.R.; Gacek, M.R. (Mobile)

Chapter 5

67 Ménière’s Disease: A Form of Vestibular Ganglionitis
Gacek, R.R.; Gacek, M.R. (Mobile)
Chapter 6

80 The Pathology of Benign Paroxysmal Positional Vertigo
Gacek, R.R.; Gacek, M.R. (Mobile)
Chapter 7

89 A Classification of Recurrent Vestibulopathy
Gacek, R.R.; Gacek, M.R. (Mobile)
Chapter 8

305 Efferent System Degeneration in Vestibular Ganglionitis
Gacek, R.R. (Mobile)
Chapter 9

324 Antiviral Therapy of Vestibular Ganglionitis
Gacek, R.R.; Gacek, M.R. (Mobile)

327 Appendix
337 Subject Index




A number of otologic disorders have mystified clinicians over the years.
These have been referred to as ‘idiopathic’ indicating lack of a known cause.
Although animal models are useful in elucidating basic physiologic mechanisms, recurrent neuropathies (vestibular, facial) of the temporal bone (TB)
are unique to humans. Therefore, human TB specimens represent the best
source of information providing insight into the pathology of these neuropathic
For hundreds of years, Bell’s palsy (IFP) and Ménière’s disease (MD) have
been regarded as idiopathic. Although displaced otoconia have been implicated
in the mechanism of benign paroxysmal positional vertigo, the precise stimulus
for degenerated otoconia has also been unknown (idiopathic). Only vestibular
neuronitis was assumed to be an inflammatory disorder of the vestibular nerve
because of its clinical association with viral-type illnesses and supported by
serologic evidence of elevated viral antibodies.
The description of endolymphatic hydrops (EH) in TB from patients with
the clinical symptoms of MD [1, 2] provided the impetus for a long series of
investigations into the concept of obstruction in longitudinal flow of endolymph
to the endolymphatic sac. The theory received support from the experimental
demonstration of EH following obstruction of the endolymphatic duct in some
animals (guinea pig, gerbil, rabbit) [3, 4]. However, failure to produce EH in
nonhuman primates [5] and the absence of vertigo in the successful animal
models of EH detracted from the EH theory of MD and accounted for the equivocal results obtained by treatments designed to reduce endolymph.

In a similar way, the previous concept of IFP held that an ischemic event
leads to edema of the facial nerve and compression within the surrounding bony
canal. Surgical decompression to relieve intraneural pressure did not achieve
superior results compared to no treatment in a large number of consecutive
patients with IFP [6]. Molecular amplification of herpes simplex virus 1 by
PCR on vestibular nerves (ganglia) from patients with MD [7] and IFP [8] supports a viral role in these idiopathic disorders.
We have demonstrated in human TB specimens from patients with IFP,
MD, vestibular neuronitis and benign paroxysmal positional vertigo a pattern of
degenerative changes in the facial nerve (meatal ganglion) and vestibular nerve
(and ganglion) which is similar to morphologic changes in herpes zoster of the
trigeminal nerve. This evidence has been summarized in the series of reports
contained in this volume of Advances in Otorhinolaryngology.
Harold F. Schuknecht, MD, predicted a viral cause for MD in his discussion
of delayed EH, a form of MD. ‘Assuming that viral labyrinthitis can occur in
infants as a subclinical disease that results in delayed endolymphatic hydrops,
we may have an explanation for the cause of Ménière’s disease. Viewed in this
context the disease entity known as delayed endolymphatic hydrops becomes
the missing link in understanding the pathogenesis of Ménière’s disease’ [9]. We
dedicate this series of studies to the memory of H.F. Schuknecht whose life-long
professional passion was the TB.
Armed with this concept of pathogenesis for the recurrent vestibulopathies,
the variable features and unpredictable nature of the ‘three faces’ of vestibular
ganglionitis can be understood. An antiviral approach is warranted but will
require substantive changes in present-day antiviral pharmaceuticals.
R.R. Gacek
M.R. Gacek


Hallpike CS, Cairns H: Observations on the pathology of Ménière’s syndrome. J Laryngol Otol
Yamakawa K: Über die pathologische Veränderung bei einem Ménière-Kranken.
J Otorhinolaryngol Soc Jpn 1938;44:2310–2312.
Kimura RS, Schuknecht HF: Membranous hydrops in the inner ear of the guinea pig after obliteration of the endolymphatic sac. Pract Otorhinolaryngol 1965;27:343–354.
Kimura RS: Animal models of endolymphatic hydrops. Am J Otolaryngol 1982;3:447–451.
Swant J, Schuknecht HF: Long term effects of destruction of the endolymphatic sac in a primate
species. Laryngoscope 1988;98:1183–1189.
Peitersen E, Andersen P: Spontaneous course of 220 peripheral non-traumatic facial palsies. Acta
Otolaryngol Suppl (Stockh) 1967;224:296–300.






Pitovski DZ, Robinson AM, Garcia-Ibanez F, Wirt R: Presence of HSV-I gives products characteristic of active infection in the vestibular ganglia of patients diagnosed with acute Ménière’s disease
(abstract 457). 22nd Annu Midwinter Res Meet Assoc Res Otolaryngol, St Petersburg Beach,
February 1999.
Burgess RC, Michaels L, Bale JF, Smith RH: Polymerase chain reaction amplification of herpes
simplex viral DNA from the geniculate ganglion of a patient with Bell’s palsy. Ann Otol Rhinol
Laryngol 1994;103:775–779.
Schuknecht HF: Pathology of the Ear, ed 2. Philadelphia, Lea & Febiger, 1993, pp 235–244.




The authors are grateful to L. Nan Johnson for excellent secretarial assistance in manuscript preparation. The professional help in medical illustrations
by Lynda Touart and Frank Vogtner is much appreciated.
Financial support for this publication was provided by Glaxo Smithkline
Pharmaceutical Co. and the University of South Alabama School of Medicine.

Gacek RR, Gacek MR: Viral Neuropathies in the Temporal Bone.
Adv Otorhinolaryngol. Basel, Karger 2002, vol 60, pp 1–11

Chapter 3

The Biology of Neurotropic Viruses
Richard R. Gacek

Neurotropic (NT) viruses are characterized by their affinity for neural
structures, specifically sensory neurons. One group of NT viruses commonly
associated with neuropathy is the ␣-herpes virinae subfamily [1]. This group
of viruses has a propensity for invading sensory neurons, the establishment of
latency within ganglion cells and a possibility of reactivation at some later date
by a stressful stimulus. The best known members of this group of viruses are
the herpes simplex (HSV) types 1 and 2, and the herpes zoster or varicella virus
[2]. These viruses are responsible for the clinical syndromes of HSV labialis
and herpes zoster [3]. Other members of this family of NT viruses are the cytomegalic inclusion virus, pseudorabies and the Epstein-Barr virus [1]. Within
these types are hundreds of strains representing mutant varieties of the virus type.
NT viruses are important clinically because of the high incidence of exposure to HSV in the population worldwide [4]. Exposure to the virus and the establishment of latency in sensory nerves may occur in individuals as early as the
first 10 years of life. The incidence of exposure and establishment of latency by
virus increases with age and lower socioeconomic status [5]. It is estimated that
by the age of 25 years 75% of the population has elevated antibodies to the HSV
group, and by the age of 60 years the exposure is over 90%. Therefore, the potential for latency in various sensory ganglia of the body is high. Exposure to viral
organisms in the human body is high in the soft palate, oropharynx, hypopharynx,
nose and nasopharynx where viral invasion of the mucous membrane epithelium
Virus presence in the epithelium of the oral cavity represents an opportunity
for invasion of a sensory neuron dependent on complementary surface structures
of the virus envelope and the sensory neuron. Virus invasion of the sensory neuron is mediated by glycoproteins in the virus envelope. These glycoproteins have
specific and sometimes overlapping functional roles. At least 10 glycoproteins
in the HSV virus envelope play a role in virus behavior. Only glycoprotein B (gB),
glycoprotein D (gD), glycoprotein H (gH) and glycoprotein L (gL) are vital to

the process of infection [6–11]. The remaining 6 glycoproteins contribute in
some way to virus invasion and infectivity in host cells. These glycoproteins are
a reflection of the genetic makeup of the viral organism and therefore confer a
particular level of infectivity for each virus strain.
Infection of a sensory neuron occurs first by virus attachment to the cell
plasma membrane, followed by penetration of the virus nucleocapsid into the
cell cytoplasm and nucleus [12]. This virus attachment involves different glycoproteins in the virus envelope and receptors on the sensory cell surface. gB and
gC are primarily responsible for the initial attachment phase which depends on
the combination of positively charged viral envelope glycoprotein moieties with
negatively charged heparan sulfate receptors on the cell surface [13, 14]. The
heparan sulfate proteoglycan receptors are also genetically programmed cell
features that permit successful attachment and infection of the neuron by a viral
organism [6]. The initial attachment process facilitates a second attachment
phase in which gD binds to a cellular receptor belonging to the tumor necrosis
factor and nerve growth factor family [15]. gD is essential for fusion of the viral
envelope to the cell membrane and finally penetration of the cell membrane by
the virus [11, 12]. It has been determined in genetic studies that gB, gD and gH
are necessary for this fusion-penetration process.
Virus-binding receptors are unevenly distributed over the plasma membrane
of the neuron. The adsorption of both HSV-1 and HSV-2 is efficient in mouse
synaptosomal and glial cell preparations but virtually absent on neuronal cell
bodies [16, 17]. Synaptosomes adsorb virus better than glial cells. Such a favorable uptake in synapses may account for the efficiency of virus uptake by nerve
terminals and transmission along multisynaptic neuronal linkages when used as
a neurobiologic tracer. The lack of receptors on neuronal perikarya might be
responsible for a reduction in HSV spread from cell to cell in peripheral ganglia.
However, cell-to-cell virus spread may occur as a result of virus movement across
cell junctions between adjacent neurons or by a viral precursor rather than the
fully formed virus [18].
The animal model of HSV infection in a murine sensory ganglion indicates
that after arrival of the virus in the ganglion cell, it accumulates within the nucleus
but after a productive infection may leave the nucleus and the cytoplasm acquiring
a double envelope from the nuclear and the plasma membranes [19]. Upon arrival
in the extracellular space, the virus particles are surrounded by an increased number of satellite cells (SC) which incorporate the virus within the SC nucleus and
cytoplasm (fig. 1). As virus particles are enveloped by the SC, they are replaced
by cytoplasmic extensions of the SC which proliferate membrane in layers. These
membrane layers fuse into thick membranous envelopes in a whorl-like pattern
(fig. 2). In this manner, the enveloping SC surround and replace the ganglion cell
body (fig. 3). The cytoplasm of the ganglion cell may become vacuolated by the



Stages of Herpetic Ganglionitis




Satellite Cell



Fig. 1. A schematic summary of the major morphologic features in NT viral ganglionitis.
Drawing A represents the acute inflammatory phase, while B and C illustrate degenerative
phases based on the histologic features shown in figures 2 and 3.

infection and replaced by SC which have increased in number (fig. 4). The SC
may replace the cell perikaryon leaving nuclear remnants at the epicenter.
After the NT virus has entered the peripheral process of a sensory neuron,
the virion is carried by axoplasmic transport to the ganglion cell body (fig. 1).
This process has been demonstrated to require 20–24 h. Once the virus has
successfully reached the cytoplasm of the ganglion cell, it may spread to adjacent ganglion cells by several mechanisms [19]. The virus may simply leave the
infected cell and attach to membrane receptors on nearby cells with subsequent
invasion. Alternatively, the virus may move across junctions where cells are closely
attached (tight junctions, desmosomes) and thereby elude antibodies circulating

The Biology of Neurotropic Viruses


Fig. 2. The vestibular ganglion in a 75-year-old female with recurrent vertigo (duration
20–40 min) and positional vertigo shows ganglion cells being replaced by SC (open arrows)
and a collagen-like substance replacing another (solid arrow).

Fig. 3. The meatal ganglion (MG) in a 44-year-old man with a history of episodic vertigo
without hearing loss for several years (case 5, table 1, see Appendix). Several ganglion cells
(arrowheads) have been replaced by a collagen-like material surrounding nuclear remnants.
FN ⫽ Facial nerve; VN ⫽ vestibular nerve.



Fig. 4. The trigeminal ganglion in case 19 (table 1, see Appendix) shows vacuolated
ganglion cells surrounded by SC (open arrows). The solid arrow points to lipofuchsin granules
in a ganglion cell.

within the extracellular space [20]. Finally SC may spread the virus to nearby
ganglion cells. The tendency of NT viruses to involve adjacent neurons in a ganglionic mass results in clusters of infected ganglion cells (fig. 5). Reactivation
of virus in a group of infected ganglion cells will then result in lesions (vesicles)
that are tightly grouped (i.e. herpes simplex labialis, herpes zoster). Degeneration of a cluster of ganglion cells produces a group of degenerated axons in the
nerve trunk (fig. 6).
If the virus does not leave the neuron completely after the initial infection,
it may assume a latent (subviral) state within the nucleus of the cell. All the factors necessary to develop latency are not known. However, a transformation in the
necessary RNA genome for establishing latency, i.e. latency-associated transcript,
is an essential event [21, 22]. Once latency has been established, reactivation of
the virus into an active or productive infection may occur following a stimulus
that is unusually stressful or traumatic, physically or chemically. The animal model
of latent HSV infection has shown adrenaline to be capable of reactivating latent
HSV [23]. An additional underlying factor is host resistance; in the immunocompromised host or in the host with a senescent immune response, the tendency
for reactivation of a latent virus form is greater than in the young uncompromised
host subject.

The Biology of Neurotropic Viruses


Fig. 5. The vestibular ganglion in case 3 (table 1, see Appendix), a 62-year-old male
with recurrent vertigo and no hearing loss, contained clusters of ganglion cells in various
phases of degeneration (arrows). Some cells are surrounded by dark SC and inflammatory
cells, while others have been replaced. VN ⫽ Vestibular nerve.

Fig. 6. Vestibular nerve from a 61-year-old female with otosclerosis who died from
ovarian cancer. Two fascicles of degenerated axons (arrows) are seen in the nerve trunk.
VG ⫽ Vestibular ganglion.



The mode of virus presence in the cell is also determined by the genotype
of host neurons. Margolis et al. [24] presented evidence that two types of
neurons in the mouse dorsal root ganglion allowed a different virus presence
following a single inoculation of virus. In one population of ganglion cells, viral
protein synthesis was high, but transcription of latency-associated transcripts
was minimal, while in a second type of neuron viral gene expression was restricted
but latency-associated transcript synthesis was abundant. These observations
suggest that following virus uptake in a nerve, latent infection will occur in one
type of neuron and active productive infection in another type of ganglion cell
within the same ganglion.
The SC has long been felt to be intimately related to its ganglion cell. SC
support the neuron metabolically during prolonged activity. This suggestion is supported by the decreased nucleic acid content in SC while neuronal nucleic acid is
increased in the superior cervical ganglion following prolonged (3-hour) stimulation [25, 26]. The role of SC in the NT viral infection of a ganglion may represent
a response to increased neural activity as well as the need to limit the spread of
virions released from ganglion cells. The increased density associated with membrane proliferation may be responsible for the collagen layers found in the onion
bulb pattern observed in some types of neuronal degeneration (fig. 2, 3) [27].
The number of SC associated with normal ganglion cells varies in different
cranial nerves. This variation may be dependant on the embryologic origin of
the ganglionic mass. The ganglion cells of the eighth cranial nerve (vestibular
and cochlear) are derived from the otic placode and typically have 1–2 SC per
ganglion cell [28]. However, the ganglion cells of the seventh cranial nerve
(geniculate and meatal) are derived from the epibranchial placode and the neural
crest epithelium; these ganglia are normally surrounded by many SC, the precise
number is not known.
Since the SC increase is part of the host response to NT virus infection in
a ganglion, a significant increase in SC in the vestibular ganglion can be recognized with confidence, but not in the ganglia of the seventh nerve where a large
number of SC is found normally. Therefore, evidence of ganglion cell degeneration is necessary to conclude that NT virus has infected the geniculate and meatal
ganglia. Since direct evidence of ganglion cell degeneration is uncommonly
seen in the vestibular ganglion, it is necessary to rely on indirect evidence in the
form of axonal degeneration to reflect NT damage of the vestibular ganglion.
Since NT virus will typically also infect adjacent ganglion cells (clusters), focal
axonal degeneration in the vestibular nerve trunk represents virus destruction of
ganglion cells. Focal axonal degeneration has been described in trigeminal
nerve zoster [29].
Although viruses are protected within the environment of the ganglion cell
and nucleus, and therefore shielded from antibodies or antiviral drugs, their

The Biology of Neurotropic Viruses


infectious effects may be manifested by the release of nucleic acids [20]. Nucleic
acids (DNA and RNA) have a low level of infectivity compared to the virus from
which they are derived. However, their release is capable of producing clinical
syndromes similar to that resulting from virus infection and yet be unaffected by
the antibody response of the host since nucleic acids are not a viral protein.
Nucleases released from blood components are capable of neutralizing nucleic
acids. White blood cells may release nuclease inhibitors and consequently disturb
the normal equilibrium between nucleic acid infectivity and nuclease control
during infection with fever (i.e. sinusitis). Fever as a precipitating effect in the
clinical manifestation of a latent virus neuropathy may be understood in the
context of this hypothesis.
Since exposure of the population to HSV is so high, reasons for the absence
of a similarly high incidence of cranial neuropathies should exist. Although this
evidence has not been reported, several possibilities exist.
(1) The makeup of the virus envelope as well as receptors on host neuronal
membrane represent a major determinant of virus invasion of a sensory ganglion.
The glycoprotein composition of the virus envelope and compatible proteoglycan receptors in the neuronal plasma membrane are genetically determined
features of the virus-host neuron complex. The absence of surface structures
essential for virus attachment and invasion would present a major deterrent to
NT invasion and establishment of latency in sensory nerves.
(2) The availability of sufficient ganglion cells to harbor a virus pathogen
may be important when the ganglion represents the initial repository for the
invading virus. The meatal ganglion of the facial nerve represents the location
for neuronal pathology associated with vestibular ganglion degeneration in recurrent vertigo possibly because it receives input from the soft palate and nasopharynx
where virus invasion occurs [27, 30]. Since the meatal ganglion is represented by
a very small ganglion cell population in most human temporal bones (TB), the
likelihood of a large virus load in this region of the facial nerve is correspondingly low.
(3) Host resistance reflects the genetic makeup of prospective neuronal
elements (i.e. No. 1), as well as the genotype of the immune system (i.e. lymphocytes) that are important for controlling virus invasion. The recrudescence
of virus from latency has frequently been noted with the immunocompromised state (chemotherapy, radiation therapy) as well as with senescent immune
NT viruses and their reactivation from latency assume a direction of flow
within the central or peripheral processes of a sensory neuron dependent on
virus strain [31, 32]. The flow from neuron to the brainstem is referred to as
anterograde flow, since it is in the direction of normal axoplasmic flow in the
neuron, whereas flow toward the periphery (over the dendrite of the sensory



neuron) is regarded as retrograde flow (fig. 1). The herpes family viruses
are characterized by their ability to flow bidirectionally between the neuron
and its peripheral or central terminus. Certain strains of the HSV preferentially travel in a retrograde direction (toward the periphery), and others flow
preferentially in an anterograde direction. The H 129 strain of HSV-1 flows in
an anterograde direction while the McIntyre B strain follows a retrograde direction of flow. This correlation is important, as to a large degree it may determine the clinical presentation of reactivated virus. This principle provides the
basis for the use of NT viruses as a neurobiologic tracing method, since the
anterograde virus strain will allow it to be transported centrally over several
synaptic connections to demonstrate the higher neuronal members of a sensory
Intracellular pathogens have long been known to produce plaques as a
result of their cytopathic effect [33]. Uncommonly associated with bacterial
pathogens such as Ehrlichia but commonly seen with viral agents such as vaccinia, psittacosis, western equine encephalomyelitis virus and HSV, plaques
have been used to detect and quantify virus presence in vitro because of the linear relationship of plaque number with the number of virus particles. Plaque
size may differ with virus type and strain. Plaque shape is roughly spherical
with a sharp border and represents necrosis of tissue.
Histological changes visible by light microscopy may reflect the accumulation of viral nucleic acids and antigen in cells infected with HSV. HeLa cells
infected with HSV in tissue culture show that virus DNA accumulates intracellularly before viral antigen can be detected [34]. Successive stages show that
more diffuse DNA accumulates as viral antigen is synthesized. These changes
are also associated with the formation of giant cells which may represent fusion
of infected individual cells. Since the nucleic acid content in cells is responsible
for nuclear staining, it is possible that nuclear stains can demonstrate high intracellular levels of DNA accumulation. One component of the hematoxylin and
eosin stain used in human TB histopathology is an excellent nuclear (nucleic
acid) stain. Hematoxylin (C16H14O6) is the compound which results after ether
extraction from the wood portion of Haematoxylon campechianum. Upon oxidation, hematoxylin is converted to hematein which stains certain structures
(i.e. nuclei) a deep blue.
Histopathologic TB studies are important to our understanding of disorders
caused by viral organisms. Understanding the events which accompany NT
viral infection and reactivation in sensory ganglia of the cranial nerves associated with the TB can guide the interpretation of morphologic changes caused
by these microorganisms. Complemented by direct immunofluorescence microscopy and molecular biology, an informative research approach can enhance the
value of human TB collections.

The Biology of Neurotropic Viruses











Meier J, Straus S: Comparative biology of latent varicella-zoster virus and herpes simplex virus
infections. J Infect Dis 1992;166(suppl I):S13–S23.
Straus SE: Clinical and biological differences between recurrent herpes simplex virus and
varicella-zoster virus infections. JAMA 1989;262:3455–3458.
Baringer R, Swoveland M: Persistent herpes simplex virus infection in rabbit trigeminal ganglia.
Lab Invest 1974;30:230 –240.
Smith IW, Peutherer JF: The incidence of herpes virus hominis antibody in the population. J Hyg
(Camb) 1967;65:395– 408.
Whitley RJ: Herpes simplex viruses; in Fields BN, Knipe DM (eds): Virology, ed 2. New York,
Raven Press, 1990, pp 1843–1888.
Laquerre S, Argnani R, Anderson D, Zucchini S, Manservigi R, Glorioso J: Heparan sulfate
proteoglycan binding by herpes simplex virus type I glycoproteins B and C attachment, which differ
in their contributions to virus penetration, and cell-to-cell spread. J Virol 1998;72:6119– 6130.
Forrester AJ, Farrell G, Wilkinson G, Kaye J, Davis-Poynter N, Minson AC: Construction and
properties of a mutant herpes simplex type I deleted for glycoprotein H sequences. J Virol
Roop C, Hutchinson L, Johnson D: A mutant herpes simplex virus type I unable to express glycoprotein L cannot enter cells and its particles lack glycoprotein H. J Virol 1993;67:2285–2297.
Ligas MW, Johnson DC: A herpes simplex virus mutant in which glycoprotein D sequences are
replaced by B-galactosidase sequences binds to but is unable to penetrate into cells. J Virol
1998;62:1486 –1494.
Ca W, Gu B, Person S: Role of glycoprotein B of herpes simplex virus type I in viral entry and cell
fusion. J Virol 1988;62:2596 –2604.
Campadelli-Eiume G, Arsenakis M, Farabegali F, Roizman B: Entry of herpes simplex virus I in BJ
cells that constitutively express viral glycoprotein D is by endocytosis and results in degradation of
the virus. J Virol 1988;62:159–167.
Nicola AV, Peng C, Lou H, Cohen GH, Eisenberg RJ: Antigenic structure of soluble herpes simplex
virus (HSV) glycoprotein D correlates with inhibition of HSV infection. J Virol 1997;71:
2940 –2946.
Herold RC, Wu Dunn D, Sultys N, Spear PG: Glycoprotein C of herpes simplex virus type I plays
a principal role in the adsorption of virus to cells and infectivity. J Virol 1991;65:1090–1098.
Wu Dunn D, Spear PG: Initial interaction of herpes simplex virus with cells is binding to heparan
sulfate. J Virol 1989;63:52–58.
Montgomery RI, Warner MS, Lurn BJ, Spear PG: Herpes simples 1 entry into cells mediated by a
novel member of the TNF/NGF receptor family. Cell 1996;87:427–436.
Vahlne A, Svennerholm B, Sandberg M, Hamberger A, Lycke E: Differences in attachment
between herpes simplex type 1 and type 2 viruses to neurons and glial cells. Infect Immun 1980;
Haywood A: Mini review – Virus receptors: Binding, adhesion strengthening, and changes in viral
structure. J Virol 1994;68:1–5.
Dingwell KS, Brunetti CR, Hendricks RL, Tang Q, Tang M, Rainbow AJ, Johnson DC: Herpes
simplex virus glycoproteins E and I facilitate cell-to-cell spread in vitro and across junctions of
cultured cells. J Virol 1994;68:834–845.
Cook ML, Stevens JG: Pathogenesis of herpetic neuritis and ganglionitis in mice: Evidence for
intra-axonal transport of infection. Infect Immun 1973;7:272–288.
Herriott RM: Infectious nucleic acids, a new dimension in virology. Science 1961;134:256–260.
Croen KD, Ostrove JM, Dragovic LJ, Smialek JE, Straus SE: Latent herpes simplex virus in human
trigeminal ganglion: Detection of an immediate-early gene ‘antisense’ transcript. N Engl J Med
Dealty AH, Spivack JG, Lavi E, O’Boyle DR, Fraser NW: Latent herpes simplex virus type I
transcripts in peripheral and central nervous system tissues of mice map to similar regions of the
viral genome. J Virol 1988;62:749–756.




Schmidt J, Rasmussen AF: Activation of latent herpes simplex encephalitis by chemical means.
J Infect Dis 1960;106:154 –158.
Margolis TP, Sedarati F, Dobson AT, Feldman LT, Stevens JG: Pathways of viral gene expression
during acute neuronal infection with HSV-I. Virology 1992;189:150–160.
Pevzner LZ: Topochemical aspects of nucleic acid and protein metabolism within the neuronneuroglia unit of the superior cervical ganglion. J Neurochem 1965;12:993–1002.
Schwyn RC: An autoradiographic study of satellite cells in autonomic ganglion. Am J Anat 1967;
Gacek RR: The pathology of facial and vestibular neuronitis. Am J Otolaryngol 1999;20:202–210.
Ona AI: The mammalian vestibular ganglion cells and the myelin sheath surrounding them. Acta
Otolaryngol (Stockh) 1993;503:143–149.
Denny-Brown D, Adams RD, Fitzgerald PJ: Pathologic features of herpes zoster: A note on geniculate herpes. Arch Neurol Psychiatry 1949;51:216–231.
Gacek RR: On the duality of the facial nerve ganglion. Laryngoscope 1998;108:1077–1086.
Zemanick MC, Strick PL, Dix RD: Direction of transneural transport of herpes simplex virus I in
the primate motor system is strain-dependent. Proc Natl Acad Sci USA 1991;88:8048–8051.
Kuypers HG, Ugolini G: Viruses as transneuronal tracers. Trends Neurosci 1990;13:71–75.
Dulbecco R: Production of plaques in monolayer tissue cultures by single particles of an animal
virus. Proc Natl Acad Sci USA 1952;38:747–752.
Ross RW, Orlans E: The redistribution of nucleic acid and the appearance of specific antigen in
Hela cells infected with herpes virus. J Pathol Bacteriol 1958;76:393–402.

The Biology of Neurotropic Viruses


Gacek RR, Gacek MR: Viral Neuropathies in the Temporal Bone.
Adv Otorhinolaryngol. Basel, Karger 2002, vol 60, pp 12–31

Chapter 2

Neuroanatomy of the Nerves in the
Temporal Bone
Richard R. Gacek

The cranial nerves associated with the temporal bone (TB) have both sensory and motor components and may be invaded by neurotropic (NT) viruses.
These are the fifth, seventh, eighth and ninth cranial nerves. NT (herpetic)
involvement of the trigeminal nerve is a common viral neuropathy which has
been studied clinically by microbiologic and histopathologic methods [1, 2].
However, virus-mediated neuropathy of the seventh, eighth and ninth cranial
nerves has been a controversial subject. A viral etiology for idiopathic facial
palsy (Bell’s palsy) is now generally recognized [3–5]. Virus-mediated neuropathy of the eighth cranial nerve has only recently been supported by morphologic
evidence in human TB [6]. Evidence to support a similar pathology in the ninth
nerve is lacking thus far. Viral neuropathy of the tenth cranial nerve is also probable but is not included in this discussion since its anatomic course through the
TB is not included in routine specimens. The eleventh cranial nerve is not affected
by NT viruses because it does not have a sensory component. The relationship
of these cranial nerves to areas of the oral cavity, oropharynx, nasopharynx and
nose which are a habitat for NT viruses represents a basis for virus recrudescence
from latency later in life.

Trigeminal Nerve

The trigeminal (fifth cranial) nerve is the largest of the cranial nerves and
conveys common sensation from the superficial and deep regions of the face as
well as a smaller motor component to the muscles of mastication [7]. The fifth
nerve trunk is attached to the lateral part of the pons by a large sensory root and
a small motor root. The two nerve roots travel forward in the posterior cranial

Fig. 1. Low-power view of the trigeminal ganglion (T) in Meckel’s cave on the superior surface of the petrous apex (PA). Figure 4 in the previous chapter 1 was taken from this

fossa to enter the middle cranial fossa after passing between the attachment of
the tentorium cerebelli and the upper border of the petrous portion of the TB.
The sensory root overlies the motor root as they pass into the trigeminal ganglion
lying in Meckel’s cave on the superior surface of the petrous apex (fig. 1). The
ganglion gives rise to the three main divisions of the nerve: ophthalmic, maxillary
and mandibular.
Details of the diverse functional components in these three divisions may
be found in texts of anatomy [7]. An overview of the sensory input conveyed
by these divisions aids the understanding of the vulnerability of this cranial
nerve to virus invasion. Although each division conveys common sensation
from the upper, middle and lower thirds of the face as well as the anterior half
of the scalp, deeper structures lined with mucous membrane within bony cavities of the skull are also richly innervated by the corresponding division.
Accordingly, the ophthalmic division carries sensory input from the nasal cavity and ethmoid sinuses over the ethmoidal nerves, the maxillary division conveys
input from the alveolar ridge, maxillary and sphenoid sinuses (sphenopalatine
nerves), and the mandibular division supplies the floor of the mouth and alveolus.
NT viral invasion of the terminals (synaptosomes) and ganglion of the trigeminal nerve is made possible by nature of the epithelial surfaces in these areas.

Neuroanatomy of the Nerves in the Temporal Bone


Fig. 2. Photo of the FN (F) and vestibular nerve (V) in the internal auditory canal of a
dissected human TB. The myelinated nerves are stained with Sudan black. M ⫽ Location of the
meatal ganglion adjacent to the vestibular ganglion; G ⫽ geniculate ganglion; S ⫽ saccule;
U ⫽ utricular macula; LC, SC, PC ⫽ lateral, superior and posterior canal cristae.

Facial Nerve

After emerging from the brainstem, the facial nerve (FN) travels together
with the vestibular division of the eighth cranial nerve the length of the internal
auditory canal (fig. 2). The FN then enters the labyrinthine segment of the fallopian canal which conveys it throughout a tortuous course through the TB. The
FN is derived from the second branchial arch and innervates structures that are
derived from Reichert’s cartilage. Four groups of functional neurons constitute
the FN complex [8].
(1) The special efferent FN axons supply the striated muscles of facial
expression, as well as the stapedius muscle, the stylohyoid muscle and the posterior belly of the digastric muscle.
(2) General visceral efferent fibers represent the preganglionic portion of the
autonomic pathway to glandular and vascular structures (fig. 3). The main glandular structures are the lacrimal gland and the seromucinous glands in the nasal




Superior Salivary
Motor Nucleus

Autonomic Pathways

Fig. 3. Diagram of the pre- and postganglionic parasympathetic motor pathways of
the FN. VI ⫽ Abducens nucleus; N VII ⫽ seventh cranial nerve.

cavity. These fibers travel in the greater superficial petrosal nerve (GSPN) to
synapse in the sphenopalatine ganglion, which contains the postganglionic neurons providing secretomotor function. Secretory fibers are carried by the chorda
tympani nerve and synapse with postganglionic neurons in the submandibular
ganglion innervating the submandibular and sublingual salivary glands.
(3) Special sensory fibers (taste) (fig. 4) are carried over two pathways. The
majority of the taste receptors inputting to the FN are located in the anterior two
thirds of the tongue. Peripheral dendrites supplying these sensory receptors in
the chorda tympani nerve join their cell bodies in the geniculate ganglion (GG).
A second group of taste receptors are located in the soft palate and nasopharyngeal mucosa and are innervated by fibers in the GSPN which belong to ganglion
cells (meatal ganglion, MG) located in the meatal segment of the FN.
(4) Somatic sensory neurons supply the skin of the external auditory canal
and the concha.
The brainstem nuclei which give rise to FN axons are:
(1) the motor nucleus of the FN, which is located in the caudal brainstem
adjacent to the superior olivary nucleus of the auditory system; just caudal to the
facial nucleus is the rostral limit of the nucleus ambiguus which provides motor
innervation to the intrinsic laryngeal musculature; the number of facial motor
neurons has been estimated at approximately 10,000–20,000; the motor neurons
for various facial muscle groups are topographically arranged in subnuclei

Neuroanatomy of the Nerves in the Temporal Bone


Trigeminal G.

Nasal &


Meatal G.
Nucleus &

Geniculate G.


Motor Nucleus

Lingual Nerve
Submandibular G.


Sublingual &

Special Sensory Pathways

Fig. 4. Diagram of the special sensory pathways of the FN. G ⫽ Ganglion;
VI ⫽ abducens nucleus; N VII ⫽ seventh cranial nerve.

within the facial nucleus [9]; however, the axons from these subnuclei intermix
as they leave the facial nucleus in a dorsal direction to loop around the abducens
nucleus near the floor of the fourth ventricle [10]; the axons converge at this
point and then bend in a ventrolateral direction just medial to the vestibular
nerve (VN) root before exiting the brainstem;
(2) the location of motor neurons for the stapedius muscle and the posterior
belly of the digastric muscle are separately clustered in the brainstem; stapedius
motor neurons are located in the interface between the facial nucleus and the
superior olivary nucleus, where they are strategically located to receive stimuli
from the afferent auditory pathway and carry out reflex contraction of the
stapedius muscle (stapedius reflex); the motor neurons for the posterior belly of
the digastric muscle are located along the course of the emerging FN root in the
lateral brainstem region;
(3) the superior salivary nucleus is responsible for secretomotor (autonomic) neurons in the FN system; this nucleus is located dorsally to the motor
facial nucleus and gives rise to the preganglionic parasympathetic secretomotor
neurons entering the submandibular and the sphenopalatine ganglia;
(4) the nucleus of the solitary tract, also located in the medulla, receives
taste input over sensory fibers of the FN.
The major portion of the FN is comprised of motor axons to the facial musculature. Although arising from regional groups of motor neurons in the facial



Fig. 5. Low-power view of the tympanic (T), geniculate (G), petrosal (P) and
meatal (M) segments of the FN. V ⫽ Vestibular ganglion; C ⫽ basal turn of the cochlea.

nucleus, these fibers intermix throughout the course of the FN in its intracranial
and intratemporal segments [10]. After exiting the stylomastoid foramen, the
motor axons gather together in functional groups before forming the 4–5
branches which supply the regional facial muscle groups. For purposes of this
discussion, the important divisions of the FN trunk are the meatal segment, the
labyrinthine (petrosal) portion, the geniculate portion and the tympanic part
(fig. 5). Except for the meatal portion which lies free in the internal auditory
canal, the remaining segments of the FN are contained within a bony canal (fallopian). Accompanying the FN trunk is the nervus intermedius which carries
secretomotor axons of the preganglionic neurons in the superior salivary nucleus,
as well as proximal axons of sensory neurons in the FN ganglia (geniculate and
meatal), traveling to the nucleus solitarius in the brainstem.

Neuroanatomy of the Nerves in the Temporal Bone


Fig. 6. a Photograph of the GG (G) at the junction of the tympanic (T) and petrosal (P)
segments of the FN. b The MG of the FN (F) is located adjacent to the vestibular ganglion (V).


Composition (%)











Cases (n)

Fig. 7. Graphic ordering of the percentage composition of the FN ganglia (geniculate
and meatal) in 100 human TB. ⫽ Meatal; ⫽ geniculate.

The sensory ganglia of the FN (geniculate and meatal) are important to the
subject of virus-mediated neuropathy (fig. 6). A quantitative study of 100 TB
described these ganglionic masses quantitatively (fig. 7) [11]. These two ganglia
are derived from different embryologic anlagen, the GG from the epibranchial
placode (second branchial arch), while the MG develops from the neural crest
primordium. In most TB (88%), the GG contains most of the sensory neurons in
the FN while the MG is very small. In approximately 12% of FN, the MG may



Fig. 8. a In this case where the GG is absent (*), the greater superficial petrosal nerve
(GSP) travels toward the labyrinthine segment (L) of the FN. T ⫽ Tympanic FN. b A large MG
represents the sensory ganglion of the FN (F) when the GG is missing. V ⫽ Vestibular ganglion.

equal or exceed the number of ganglion cells in the GG. The study found that
the number of neurons in the GG ranged from 66 to 4,017 (mean 1,713) while
the MG contained from 0 to 2,764 cells (mean 448). Fourteen percent of the GG
contained less than 1,000 cells, while 88% of the MG contained under 1,000
cells. Sixty-four percent of the MG held fewer than 500 cells, and 34% had less
than 200. In approximately 2% of TB, the MG represents the entire ganglion
associated with the FN.
In instances where the GG is absent and the MG represents the only sensory ganglion of the FN, TB specimens indicate that the GSPN inputs to the MG
(fig. 8). This observation supports a conclusion that the afferent input from taste
receptors in the soft palate and nasopharynx is carried over the GSPN to the
MG, while the GG contains sensory neurons for taste receptors in the anterior
two thirds of the tongue [11]. Furthermore, the MG location in the inner auditory canal portion of the FN is juxtaposed to the vestibular ganglion (Scarpa’s
ganglion; fig. 6b). Although these two ganglionic masses are derived from two
separate embryologic sources, their intimate anatomic association permits a
common involvement in inflammatory processes [6].

Eighth Cranial Nerve

The eighth cranial nerve is made up of two portions, the vestibular and the
cochlear, supplying the balance and the auditory portions of the labyrinth,
respectively. Both of these nerve divisions are primarily afferent in function and

Neuroanatomy of the Nerves in the Temporal Bone






Supporting cell

Nerve chalice
nerve ending

Synaptic bar
nerve ending

nerve ending

Type 2

Type 1
Hair cells

Fig. 9. Drawing of the afferent and efferent innervation of type 1 and 2 vestibular hair
cells. KC ⫽ Kinocilium.

composed of bipolar ganglion cells derived from the auditory vesicle [8]. They
are of placodal origin. The human VN is comprised of approximately 18,000
bipolar neurons, of which a third are classified as large afferents and two thirds
are small afferents. The large and small afferent neurons supply hair cells in all
five vestibular sense organs. The large afferents supply the type 1 hair cells with
a calyx-like ending, in a 1 : 1 or 1 : 2 ratio (fig. 9). The small afferents supply
type 2 hair cells in the vestibular sense organs with small bouton-type endings.
Each small afferent fiber branches generously to contact type 2 hair cells over a
wide area in the sensory epithelium. There is an orderly distribution of type 1
and type 2 hair cells in the sense organs [12]. In the crista of the three semicircular
canals, type 1 hair cells are located primarily at the crest of the crista, whereas
type 2 hair cells predominate along the slopes. In the maculae of the utricle and
the saccule, type 1 hair cells predominate near the striola line of the macula,
whereas the type 2 hair cells are denser over the peripheral regions. The two






Saccular nerve
Utricular nerve

Fig. 10. Drawing of the input from vestibular sense organs in the ear. The dark area from
the superior (SCA) and lateral (HCA) canals is closest to the FN. PCA ⫽ Posterior canal
crista; SG ⫽ Scarpa’s ganglion; OCB ⫽ olivocochlear bundle; PCN ⫽ posterior canal nerve.

types of afferent neurons in the VN possess different neurophysiologic properties. The large afferents supplying type 1 hair cells display an irregular spontaneous discharge pattern, while the small afferents to type 2 hair cells exhibit a
regular spontaneous discharge pattern [13, 14]. Each ganglion cell in the human
VN is normally surrounded by 1–2 satellite cells which have an important and
close metabolic relationship to its ganglion cell [15]. The distribution and course
of the bipolar afferent neurons in the VN are organized as to their projection pattern from sense organs (fig. 10) [16]. The lateral and superior canal cristae of
the superior vestibular division input to the brainstem over large afferent ganglion cells in the most anterior portion of the vestibular trunk, which are closest
to the MG in the FN. The small afferents supplying type 2 hair cells in the lateral
and superior canal cristae are located in a more caudal portion of the superior
division of the VN, while the ganglion cells supplying the utricular macula lie in

Neuroanatomy of the Nerves in the Temporal Bone













Coch. eff.

Fig. 11. Drawing of the origin and course of the efferent vestibular pathway. The stippled
area denotes the efferent cochlear pathway. DCN ⫽ Dorsal cochlear nucleus; VCN ⫽ Ventral
cochlear nucleus; LVN ⫽ Lateral vestibular nucleus; MVN ⫽ Medial vestibular nucleus;
V ⫽ Descending trigeminal nucleus; VII ⫽ Facial nerve genu; ASO ⫽ Accessory superior
olivary nucleus; VI ⫽ Abdueens nucleus; LSO ⫽ Lateral superior olivary nucleus; Coch.
eff. ⫽ Cochlear efferent.

the inferior portion of the superior vestibular division. The ganglion cells for
the posterior canal crista are located most caudally in the inferior vestibular
ganglion and project their axons rostrally to join those of the superior division
cristae before entering the brainstem. The saccular ganglionic input is located
in the most caudal portion of the VN trunk. The distal process (dendrite) of
vestibular ganglion cells is approximately half the diameter of the proximal
axon and is intermixed in the nerve branches before terminating in the sense
organ neuroepithelium. On the other hand, the proximal axons of ganglion cells
project in a straightforward fashion from Scarpa’s ganglion cells to the brainstem. Therefore, degeneration of the thicker proximal axons in the nerve trunk
is more easily detected by light microscopy than degeneration of the thinner
distal process (dendrite) in VN branches.
The efferent pathway to the vestibular labyrinth arises from small neurons
located bilaterally near the medial vestibular nucleus and the abducens nucleus
close to the floor of the fourth ventricle (fig. 11) [17]. The number of efferent
neurons supplying the cat labyrinth is approximately 200–300 [18]. However,
because of a profuse branching pattern, the number of efferent terminals almost



Fig. 12. Cross-section of the seventh and eighth nerves of the cat stained for acetylcholinesterase. EF ⫽ Efferent fibers (cochlear and vestibular) bundle in the VN (V);
C ⫽ cochlear nerve; F ⫽ FN; NI ⫽ nervus intermedius.

equals the number of afferent terminals provided by 12,000 afferent neurons [19].
Fine efferent axons collect as they travel laterally before entering the vestibular
root in the brainstem, join the efferent cochlear fibers (olivocochlear bundle)
and travel together as a compact group of small axons in the VN (fig. 12). These
fine axons emerge from the brainstem between the superior and inferior VN divisions [19]. At the saccular portion of Scarpa’s ganglion, the efferent axons pass
through the ganglionic mass and then diverge toward the sense organs (fig. 13).
Vestibular efferents are dorsally located in the parent efferent bundle before
dispersing into the superior and inferior vestibular divisions, first in fascicles
and then as individual fibers which branch as they travel peripherally (fig. 14).
After penetrating the basement membrane of the sense organs, they ramify further before forming many vesiculated small bouton terminals contacting type 2
hair cells predominantly [20, 21]. Efferent termination also occurs on the large
calyx-like endings which engulf type 1 hair cells. The density of efferent terminals is greatest on the type 2 hair cells along the slopes of the cristae and in
peripheral regions of the maculae [21]. Efferent fibers are cholinergic and the
distribution of efferent fibers can be selectively demonstrated by using a histochemical method to localize acetylcholinesterase activity [19].

Neuroanatomy of the Nerves in the Temporal Bone


Fig. 13. A more distal section of the same specimen as in figure 12 demonstrates the
olivocochlear bundle (OCB) as it leaves the saccular nerve (S) to join the cochlear nerve (C).
Vestibular efferent fibers (VE) travel as bundles in the superior division and as scattered
fibers in the posterior canal nerve (PC) and saccular nerve. F ⫽ FN.

The human cochlear nerve is composed of approximately 30,000 bipolar
ganglion cells, of which 95% are type 1 with myelinated axons, and 5% are type 2
ganglion cells with unmyelinated cell processes [22]. The type 1 spiral ganglion
cells project in a straightforward manner to the inner hair cells (IHC) of the
organ of Corti where they terminate directly on the IHC (fig. 10). Since approximately 10–20 type 1 dendrites terminate on each IHC, the innervation pattern
is very dense at the base of the IHC (fig. 15). The spontaneous discharge pattern
of these type 1 ganglion cells is irregular, somewhat similar to the large afferents in the vestibular ganglion which terminate on type 1 vestibular hair cells.
The small type 2 spiral ganglion cells project fine unmyelinated dendrites along
the floor of the tunnel space in the organ of Corti, enveloped by tunnel cell
processes to form spiral fiber bundles between Deiters’ cells and terminate on
outer hair cells (OHC) in a diffuse pattern (fig. 16). They travel basally in a longitudinal direction before terminating on OHC [22]. The central termination of
type 2 ganglion cells is unknown, although it has been suggested that they terminate in the dorsal cochlear nucleus. The function of type 2 spiral ganglion
cells is unknown at this time.



Fig. 14. A high-power view of the posterior canal nerve shows the individual efferent
nerve fibers (arrows) scattered throughout the nerve. Ganglion cells are in the upper left corner.

The efferent cochlear system (olivocochlear bundle) has been described for
over 50 years as having a bilateral origin with the major portion of the efferent
axons to one cochlea arising from periolivary neurons near the contralateral
accessory superior olivary nucleus in the brainstem (fig. 17) [24–26]. These
myelinated axons pass in a dorsal direction before decussating under the floor of
the fourth ventricle with the contralateral olivocochlear bundle and interdigitate
with FN fibers before merging with the vestibular efferent bundle in the VN
root. They are then joined by the ipsilateral limb of the olivocochlear bundle
which is given off by small neurons surrounding the lateral superior olivary
nucleus. The efferent neuronal supply to the cat cochlea numbers 1,500–2,000
compared to 50,000 afferent ganglion cells in the cat [26]. As with vestibular
innervation, extensive branching in the efferent system accounts for near equality in the number of afferent and efferent terminals within the organ of Corti.
Numerically, the ipsilateral limb of the olivocochlear bundle is about 25% of the
size of the contralateral limb. Furthermore, the axons in the ipsilateral olivocochlear bundle are unmyelinated or thinly myelinated while those in the contralateral limb are well myelinated. These cochlear efferent axons, together with
vestibular efferent fibers, travel in the VN through the saccular portion of the

Neuroanatomy of the Nerves in the Temporal Bone


Fig. 15. A transmission electron micrograph of the base of IHC shows bundles
of nerve fibers (NF) and endings (NE) tightly surrounded by supporting cells (S) after
penetrating the basilar membrane (BM). Efferent fibers in a spiral bundle (E) pass near the
hair cell.

Fig. 16. A phase contrast micrograph of the organ of Corti in the cat illustrates the
relationship of outer spiral bundles (open arrows) at the base of OHC. D ⫽ Deiters’ cells;
H ⫽ Hensen’s cells; P ⫽ pillars; S ⫽ supporting cells with IHC; T ⫽ tectorial membrane.



Fig. 17. Drawing of the origin and course of the olivocochlear efferent bundle by
Rasmussen [24].

vestibular ganglion before the efferent cochlear fibers emerge as the vestibulocochlear anastomosis (Oort’s), enter Rosenthal’s canal and then travel apically
in a spiral direction as the intraganglionic spiral bundle. As the bundle travels
apically in the cochlea, it gives off individual fibers which mix with afferent
dendrites within the osseous spiral lamina before exiting through the habenula
perforata to enter the organ of Corti.
Since cochlear efferent axons are also cholinergic, the acetylcholinesterase
technique has been used to demonstrate their course and termination. The fibers
of the contralateral olivocochlear bundle cross high in the tunnel space and give
rise to large vesiculated terminals which contact the base of OHC (fig. 18, 19).
The density of the efferent innervation of OHC is greatest in the upper basal
turn, with decreasing innervation density in both apical and basal directions
[27]. This decrease is seen first in the outermost row of OHC, then the middle
and innermost OHC in the apical direction. The smaller fibers of the ipsilateral
efferent system form a dense inner spiral bundle of fibers under the IHC which
provides contact by small bouton-shaped endings on afferent axons near their
termination on IHC.

Glossopharyngeal Nerve (Ninth Cranial Nerve)

Although the ninth cranial nerve has a motor component to the stylopharyngeus muscle, it is largely a sensory nerve which innervates the carotid body,

Neuroanatomy of the Nerves in the Temporal Bone


Fig. 18. An acetylcholinesterase preparation of the guinea pig organ of Corti demonstrates the course and termination of efferent fibers (EF). Large terminal swellings are located
at the base of OHC, while the large accumulation under an IHC represents inner spiral fibers
as well as terminals on afferent endings.

Fig. 19. An electron micrograph at the base of an OHC in the guinea pig demonstrates
large efferent terminals (E) filled with vesicles and an afferent ending (A) from type 2 spiral
ganglion cells.



Middle ear
Inferior salivatory


Otic ganglion

Motor nucleus





Soft p




carotid artery


carotid artery

Carotid body

Fig. 20. Drawing summarizing the afferent and efferent projections of the glossopharyngeal (ninth) nerve. Note that the tympanic nerve contains both afferent and efferent
nerve fibers.

the pharyngeal tonsil, the base of the tongue and the lingual surface of the
epiglottis. Sensory taste receptors located in the posterior third of the tongue,
the adjacent epiglottis and the soft palate project over the glossopharyngeal
nerve and ganglion to the nucleus solitarius in the brainstem (fig. 20). Ganglion
cells responsible for these sensory inputs are located in the inferior ganglion
within the jugular foramen. A smaller superior ganglion is variably present and
may contain sensory neurons of the tympanic branch.
The tympanic branch of the ninth nerve is important clinically because it
carries preganglionic efferent parasympathetic axons as well as afferents from
middle ear mucosa through the middle ear space as Jacobson’s nerve which continues as the lesser superficial petrosal nerve before synapsing in the otic ganglion. Postganglionic neurons in the otic ganglion complete the efferent link to
the parotid salivary gland.
The presence of sensory ganglion cells carrying input from taste receptors
in the oral cavity over the seventh and ninth cranial nerves represents a common
pathway for entrance of NT viruses into these cranial nerves. NT viral ganglionitis as a cause of recurrent ear pain requires morphologic evidence in
human TB.

Neuroanatomy of the Nerves in the Temporal Bone









Baringer JR, Griffith JF: Experimental herpes simplex encephalitis: Early neuropathologic
changes. J Neuropathol Exp Neurol 1970;29:89–104.
Baringer RJ, Swoveland MA: Persistent herpes simplex virus infection in rabbit trigeminal ganglia.
Lab Invest 1974;80:230–240.
Adour K, Bell DN, Hilsinger R: Herpes simplex virus in idiopathic facial paralysis (Bell’s palsy).
JAMA 1975;233:527–530.
Burgess RC, Michaels L, Bole JF, Smith RH: Polymerase chain reaction amplification of herpes
simplex viral DNA from the geniculate ganglion of a patient with Bell’s palsy. Ann Otol Rhinol
Laryngol 1994;103:775–779.
Gacek R, Gacek M: Meatal ganglionitis: Clinical pathologic correlation in idiopathic facial paralysis (Bell’s palsy). Otorhinolaryngol Nova 1999;9:229–238.
Gacek R: The pathology of facial and vestibular neuronitis. Am J Otolaryngol 1999;20:202–210.
Dunward A: Peripheral nervous system – Trigeminal nerve; in Brash JC (ed): Cunningham’s Text
Book of Anatomy. Oxford, Oxford University Press, 1951, pp 1018–1028.
Gacek R, Gacek M: Anatomy of the auditory and vestibular systems; in Ballenger J, Snow J (eds):
Ballenger’s Otorhinolaryngology, Head and Neck Surgery, ed 16. San Diego, Singular Publications,
San Diego, in press.
Radpour S, Gacek RR: Further observations on the organization of the facial nucleus. Laryngoscope
Gacek RR, Radpour S: Fiber orientation of the facial nerve: An experimental study in the cat.
Laryngoscope 1982;92:547–556.
Gacek RR: On the duality of the facial nerve ganglion. Laryngoscope 1998;108:1077–1086.
Wersall J, Flock A, Lundquist RG: Structural basis for directional sensitivity in cochlear and
vestibular sensory receptors. Cold Spring Harbor Symp Quant Biol 1965;30:115–132.
Walsh BT, Miller JB, Gacek RR, Kiang NYS: Spontaneous activity in the eighth cranial nerve of
the cat. Int J Neurosci 1972;3:221–236.
Goldberg JM, Fernandez C: Physiology of peripheral neurons innervating semi-circular canals
of the squirrel monkey. I. Resting discharge and response to constant angular accelerations.
J Neurophysiol 1971;34:635–660.
Ona A: The mammalian vestibular ganglion cells and the myelin sheath surrounding them. Acta
Otolaryngol (Stockh) 1993;suppl 503:143–149.
Gacek RR: The course and central termination of first order neurons supplying vestibular end
organs in the cat. Acta Otolaryngol (Stockh) 1969;suppl 254:1– 66.
Gacek RR, Lyon M: The localization of vestibular efferent neurons in the kitten with horseradish
peroxidase. Acta Otolaryngol (Stockh) 1974;suppl 77:92–101.
Gacek RR: Efferent component of the vestibular nerve; in Rasmussen GL, Windle WF (eds):
Neural Mechanisms of the Auditory and Vestibular Systems. Springfield, Thomas, 1960,
pp 276 –284.
Gacek RR, Nomura Y, Balogh K: Acetyl cholinesterase activity in the efferent fibers of the statoacoustic nerve. Acta Otolaryngol 1965;59:541–533.
Wersall J: Studies on the structure and innervation of the sensory epithelium of the cristae
ampullares in the guinea pig: A light and electron microscope investigation. Acta Otolaryngol
(Stockh) 1956;suppl 126:1–85.
Smith CA, Rasmussen GL: Nerve endings in the maculae and cristae of the chinchilla vestibule,
with a special reference to the efferents. 3rd Symp Role Vestib Organs Space Exploration, NASA
Status Post-152, Washington, USGPO, pp 183–201.
Spoendlin H, Schrott A: The spiral ganglion and the innervation of the human organ of Corti. Acta
Otolaryngol (Stockh) 1990;105:403–410.
Spoendlin HH, Gacek RR: Electron microscopic study of the efferent and afferent innervation of
the organ of Corti in the cat. Ann Otol Rhinol Laryngol 1963;72:660–687.
Rasmussen GL: The olivary peduncle and other fiber projections of the superior olivary complex.
J Comp Neurol 1946;84:141–220.





Rasmussen GL: Further observations of the efferent cochlear bundle. J Comp Neurol 1953;
Warr B, Guinan J, White JS: Organization of the efferent fibers: The lateral and medial olivocochlear systems; in Altschuler RA, Hoffman GW, Bobbin RP (eds): Neurobiology of Hearing:
The Cochlea. New York, Raven Press, 1986, pp 333–348.
Ishii D, Balogh K: Distribution of efferent nerve endings in the organ of Corti: Their graphic
reconstruction in cochleae by localization of acetylcholinesterase activity. Acta Otolaryngol
(Stockh) 1968;66:282–288.

Neuroanatomy of the Nerves in the Temporal Bone


Gacek RR, Gacek MR: Viral Neuropathies in the Temporal Bone.
Adv Otorhinolaryngol. Basel, Karger 2002, vol 60, pp 32–53

Chapter 3

Meatal Ganglionitis: A Pathologic
Correlate in Idiopathic Facial Paralysis
Richard R. Gacek, Mark R. Gacek

Evidence from many sources has accumulated to support the concept that
idiopathic facial paralysis (IFP) is an inflammatory neuropathy caused by a neurotropic (NT) virus of the herpes simplex or zoster family [1–10]. Because of the
high exposure to these viruses in the general population, numerous nerves in the
body are exposed to NT viruses which then have the tendency to acquire a latent
form within sensory ganglion cells of peripheral nerves. The previous chapters
have described the biology of herpes simplex ganglionitis and the propensity of
the virus to remain in a latent form within the ganglion from which it can be
reactivated at a later time [11]. Although the sensory ganglion cell is the locus
of inflammation, the motor portion of the facial nerve (FN) may be affected
because of a demyelinating autoimmune response to the viral agent in the ganglion cells [12, 13]. It is also probable that various virus types and strains, as
well as host resistance, play a role in the clinical manifestation of IFP.
Although it has generally been assumed that the geniculate ganglion (GG)
is the site of virus accumulation in IFP [14], ganglion cell degeneration within
the GG has never been described. On the other hand, recent attention has been
called to the meatal ganglion (MG) which is located in the meatal segment of
the FN [15, 16]. While the MG is present in all human temporal bones (TB), it
has a relatively minor presence compared to the GG in most TB. However, in
12% of TB, the MG may be as large or even exceed the GG in ganglion cell
number. Clinical observations made by Fisch and Esslen [17] indicated that the
most prominent location of FN swelling and edema is in the meatal segment
(that portion of the nerve that is proximal to the meatal foramen). The postulated
dural constriction at the entrance to the labyrinthine fallopian canal was felt to
be responsible for obstruction of axoplasmic flow which then causes a physiologic decrease in nerve conduction. Surgical decompression of this portion of
the FN canal was felt to be important in the treatment of IFP.

In 1999, we reported TB findings in a patient with IFP 6 years before death,
with subsequent complete recovery of the facial paralysis [18]. This patient had
undergone radiation therapy to the spleen for chronic lymphocytic leukemia. No
degenerated ganglion cells were found in the GG of the TB, but there were several degenerated ganglion cells in the MG. The adjacent vestibular ganglion to
the MG carrying innervation to the lateral and superior canal cristae was completely degenerated, and focal axonal degeneration was also seen in the vestibular nerve trunk. The concept is formed that IFP results from meatal ganglionitis
rather than geniculate ganglionitis. It is possible that the GG is involved in the
progression of IFP, since the two ganglia are connected by the nervus intermedius. The present report describes 6 TB from 4 patients with IFP. The major
finding was a confirmation of degenerated ganglion cells in the MG of the FN
and not the GG.

Materials and Methods
(1) A case report describes the MRI findings in a patient with IFP that was monitored
at 1, 8 and 15 weeks after the onset of paralysis. Spontaneous and complete recovery of facial
function occurred within 2 months in this patient. Six additional patients with IFP were followed with MRI.
(2) Eleven published studies [19–29] describing the use of MRI in IFP were reviewed
comparing the location of enhancement in the FN during the disorder. These studies included
patients who were monitored within 7 days as well as several weeks to months following the
onset of paralysis.
(3) Six horizontally sectioned TB from 4 patients with a history of IFP were examined
for morphologic changes in the FN as well as vestibular and cochlear ganglia which correlate
with the FN paralysis. These TB were formalin fixed, decalcified, embedded in celloidin and
sectioned at 20 m thickness. Every tenth section was stained with hematoxylin and eosin,
cover-slipped and examined in a light microscope.


Case Report
A 51-year-old female with a 7-day history of complete left facial paralysis
(grade VI/VI House-Brackman), otalgia and vertigo had an otherwise normal
head and neck examination. The remaining cranial nerve function including
hearing was normal. The patient had been treated with oral prednisone (40 mg
daily) since the onset of facial weakness. Famvir (500 mg t.i.d.) was added to the
steroid management. An enhanced MRI at this time revealed localized enhancement in the meatal segment of the left FN (fig. 1).

Meatal Ganglionitis: A Pathologic Correlate in Idiopathic Facial Paralysis


Fig. 1. Coronal (a) and axial (b) Gd-DTPA MRI demonstrates localized enhancement
in the internal auditory canal of a left TB (arrow) 7 days after onset of left IFP.

Medical treatment was discontinued after 1 additional week because the
patient demonstrated spontaneous partial recovery of function (grade III/VI).
Six weeks later, when the left facial weakness had improved significantly (grade
I/VI), an MRI with gadolinium showed enhancement of the geniculate, tympanic
and mastoid FN segments in addition to the meatal FN enhancement (fig. 2).
MRI 15 weeks after onset of IFP demonstrated enhancement of the GG and
greater superficial petrosal nerve but none in the meatal segment (fig. 3).
We have performed MRI on 6 additional patients with IFP. All demonstrated enhancement in the meatal segment of the FN during the first 2 weeks
after onset of paralysis.

MRI Studies
Table 1 lists 11 MRI studies of the FN in IFP reported from the years 1989–
1997 [19–29]. These studies are representative of the FN imaging studies dealing
with IFP likely caused by herpes simplex virus type 1 (HSV-1) and varicella-zoster



Fig. 2. Coronal (a) and axial (b) enhanced MRI 6 weeks after IFP onset demonstrates enhancement in the geniculate (G), proximal tympanic and mastoid (MA) segments of
the FN.

Fig. 3. Axial enhanced MRI at 15 weeks after onset of IFP demonstrates enhancement
in the GG (G) and greater superficial petrosal nerve (GSP) but minimal change in the
labyrinthine segment (L) of the FN.

Meatal Ganglionitis: A Pathologic Correlate in Idiopathic Facial Paralysis


Table 1. MRI of FN in IFP


Daniels et al. [19]
Schwaber et al. [20]
Tien et al. [21]
Doringer et al. [24]
Matsumoto et al. [23]
Murphy and Teller [25]
Yanagida et al. [26]
Korzec et al. [22]
Engstrom et al. [27]
Kohsyu et al. [28]
Engstrom et al. [29]





No enhancement
















– Not given; ME meatal; L labyrinthine; G geniculate; T tympanic; MA mastoid.

virus. Careful evaluation of enhancement in the FN in an inflammatory disorder
such as IFP requires consideration of enhancement in the geniculate, tympanic
and mastoid FN segments caused by pooling of gadolinium in the vascular network of the sheath surrounding the nerve in these portions of the fallopian
canal. Enhancement of the FN in these regions is frequently seen with increased
time after onset of IFP. This is due to increased intraneural edema and inflammatory dilatation of the perineural vessels in these segments [30]. Therefore,
enhancement in the FN most reliably reflects an inflammation in the meatal and
labyrinthine segments. In 7 of the studies [19, 20, 22, 24, 27–29], enhancement
occurred in the meatal FN segment in the majority of the patients with IFP. All
11 series reported enhancement in the meatal FN. Seven studies recorded
patients where no enhancement was found in the ipsilateral FN. Of the remaining 4 reports, 2 series [21, 28] recorded no FN without enhancement while 2
other reports [23, 26] gave no information.

TB Reports
Table 2 summarizes morphologic changes in the seventh and eighth cranial
nerves in the TB of the 4 patients with IFP. The ages of the 4 patients with IFP
ranged from 56 to 74 years; there were 2 males and 2 females. One patient



Table 2. Pathology in IFP (n 6 TB)



Cause of death





vestibular spiral
ganglion ganglion





chronic lymphocytic




and LC)





deg. coch.
and vest.

oat cell carcinoma of
the lung/terminal




total loss

total loss

W.R. (L) 74



myocardial infarction




50% loss

W.R. (R) 74



myocardial infarction





60% loss

E.J. (R)



herpes zoster





FN comp.


90% loss

E.J. (L)









90% loss

FP Facial paralysis; R resolved; P partial; T total; FS facial nerve swelling; SOM serous otitis media;
deg. degeneration; coch. cochlear; vest. vestibular; comp. complete; SNHL sensorineural hearing loss;
SC superior semicircular canal; LC lateral semicircular canal.

(A.B.) had a history of a partial facial paralysis with complete recovery 6 years
before her death from chronic lymphocytic leukemia. This TB has previously
been reported [18]. The remaining 3 patients had facial paralysis (partial 1, total 2)
at the time of death. Patient E.J. had right total facial paralysis from herpes
zoster oticus. Degeneration of ganglion cells or the FN in its meatal segment
was seen in all but 1 TB. No degenerated cells were observed in the GG of any
of the 6 TB. The vestibular ganglion was partially or totally degenerated in 4 TB,
and cochlear neurons were significantly degenerated in all 6 TB. In 1 case with
bilateral facial paralysis, there was significant swelling of the FN proximal to
the meatal foramen on both sides.
Case 1
At the age of 78 years, this patient experienced sudden onset of partial right
FN paralysis which recovered completely in 10 days. She did not complain of
hearing loss or vertigo. Three years before she had been diagnosed as having
chronic lymphocytic leukemia which was treated by radiation therapy to the
spleen. During the remaining 6 years of life she had recurrent episodes of

Meatal Ganglionitis: A Pathologic Correlate in Idiopathic Facial Paralysis


Fig. 4. The GG in an 81-year-old female (case 1) with recovered IFP contained many
satellite cells (arrows) but no degenerated ganglion cells.

septicemia treated with antibiotics. Her death was caused by overwhelming
Histopathology of the Right TB: Postmortem Time 13 h. The middle ear
mucosa was hypertrophic and contained numerous foci and diffuse infiltration
of lymphocytes. There were numerous fascicles of regenerating myelinated
nerve fibers passing around the GG. Numerous mononuclear cells resembling
satellite cells filled the space between GG neurons. No degenerated ganglion
cells were seen in the GG (fig. 4).
Ganglion cells in the MG and scattered between sensory fibers of the FN
were surrounded by an increased number of satellite cells. There were several
degenerated ganglion cells in the MG (fig. 5), and degenerated axons were
found in the nervus intermedius of the FN.
There was a loss of dendrites to the cristae ampullares of all three semicircular canals (fig. 6) while the utricular (fig. 7) and saccular nerve branches
were normal. Scarpa’s ganglion contained approximately 30% loss of ganglion
cells and the remaining cells were surrounded by an increase in satellite cells.
Focal axonal degeneration was present in the vestibular nerve trunk (fig. 8). There
was a patchy loss of the organ of Corti throughout the basal turn of the cochlea.
Atrophy of the stria vascularis was present in the middle and upper basal turns.



Documents similaires

viral neuropathies in the temporal bone karger
herpes virus
herpes cutaneomuqueux
ultrasound does not shorten the duration
maladies virales ppt

Sur le même sujet..