pathophysiology of ménière's syndrome: are symptoms caused by

Pathophysiology of Ménière’s Syndrome: AreSymptoms Caused by Endolymphatic Hydrops?

Saumil N. Merchant, Joe C. Adams, and Joseph B. Nadol, Jr.

Otopathology Laboratory, Department of Otolaryngology, Massachusetts Eye and Ear Infirmary,Department of Otology and Laryngology, Harvard Medical School, Boston, Massachusetts, U.S.A.

Background: The association of Ménière’s syndrome with en-dolymphatic hydrops has led to the formation of a central hy-pothesis: many possible etiologic factors lead to hydrops, andhydrops in turn generates the symptoms. However, this hypoth-esis of hydrops as being the final common pathway has notbeen proven conclusively.Specific Aim: To examine human temporal bones with respectto the role of hydrops in causing symptoms in Ménière’s syn-drome. If the central hypothesis were true, every case ofMénière’s syndrome should have hydrops and every case ofhydrops should show the typical symptoms.Methods: Review of archival temporal bone cases with a clini-cal diagnosis of Ménière’s syndrome (28 cases) or a histopath-ologic diagnosis of hydrops (79 cases).Results: All 28 cases with classical symptoms of Ménière’ssyndrome showed hydrops in at least one ear. However, thereverse was not true. There were 9 cases with idiopathic hy-

drops and 10 cases with secondary hydrops, but the patients didnot exhibit the classic symptoms of Ménière’s syndrome. Areview of the literature revealed cases with asymptomatic hy-drops (similar to the current study), as well as cases wheresymptoms of Ménière’s syndrome existed during life but nohydrops was observed on histology. We also review recentexperimental data where obstruction of the endolymphatic ductin guinea pigs resulted in cytochemical abnormalities withinfibrocytes of the spiral ligament before development of hy-drops. This result is consistent with the hypothesis that hydropsresulted from disordered fluid homeostasis caused by disrup-tion of regulatory elements within the spiral ligament.Conclusion: Endolymphatic hydrops should be considered as ahistologic marker for Ménière’s syndrome rather than beingdirectly responsible for its symptoms. Key Words: Endolym-phatic hydrops—Ménière’s syndrome—Temporal bone.Otol Neurotol 26:74–81, 2005.

In 1861, Prosper Ménière (1) described a syndromeconsisting of continuous or intermittent head noises ac-companied by diminution of hearing and intermittent at-tacks of vertigo; uncertain gait and falling; accompaniedby nausea, vomiting, and syncope. The essence ofMénière’s hypothesis was that the association of auditoryand vestibular symptoms suggested the ear as the ana-tomical location of the disorder, as opposed to the centralnervous system as previously thought. By extension ofFlourens’ (2) experiments on pigeons, he hypothesizedthe site of lesion to be the semicircular canals. In 1938,Yamakawa (3) in Japan and Hallpike and Cairns (4) inEngland independently reported the finding of endolym-phatic hydrops in temporal bones from patients withMénière’s syndrome. Hallpike and Cairns regarded thehydrops as an essential morbid anatomy of a specificdisease of the labyrinth. Other investigators described

similar temporal bone findings over the next decade (5–9). The hydrops seen in temporal bones was described as“idiopathic” in that there was no obvious abnormalityin the temporal bones that could be considered causal.With the identification of a pathologic lesion (i.e., hy-drops), the term “Ménière’s disease” came to be appliedto the syndrome.

A large number of factors have been proposed as lead-ing to the development of endolymphatic hydrops. Thelist includes excessive endolymph production, decreasedendolymph absorption by the endolymphatic sac, ionicimbalance, genetic abnormalities, viral infection, auto-nomic imbalance, dietary factors, autoimmune reactions,vascular irregularities, allergic responses, and others(10–13). In the majority of cases of Ménière’s disease,light microscopy of temporal bone sections has failed todemonstrate losses of sensory or neural structures withinthe inner ear such as hair cells, neuronal cells, stria vas-cularis, dark cells, and so forth, that can be correlatedwith the premortem loss of auditory and vestibular func-tion (3–9,11). Because hydrops is the only consistentpathologic abnormality observed at light microscopy, acentral hypothesis has been articulated regarding thepathophysiology of Ménière’s disease: although many

Presented at the Annual Meeting of the American Otological Soci-ety, May 1–2, 2004, Phoenix, Arizona, U.S.A.

Address correspondence and reprint requests to Saumil N. Merchant,M.D., Department of Otolaryngology Massachusetts Eye and Ear In-firmary, 243 Charles Street, Boston, MA 02114-3096; U.S.A.; Email:[email protected]

Otology & Neurotology26:74–81 © 2005, Otology & Neurotology, Inc.

74

possible etiologic factors can lead to endolymphatic hy-drops, it is the hydrops that generates the symptoms ofMénière’s disease (10) (Fig. 1). The central hypothesishas not been conclusively proven. If the hypothesis weretrue, every case of Ménière’s syndrome should have en-dolymphatic hydrops and every case of hydrops shouldshow the clinical symptoms, unless the chain of neuralevents is interrupted. A temporal bone study from ourlaboratory in 1989 to assess the correlation between casehistories and postmortem histopathology in Ménière’ssyndrome raised questions about the validity of this cen-tral hypothesis (14). Since this 1989 study, we haveadded over 250 new temporal bone specimens to ourcollection. One goal of this study was to reexamine therelationship between hydrops and symptoms ofMénière’s syndrome by including the new specimens.

In 1965, Kimura and Schuknecht (15) reported thatobliteration of the endolymphatic duct in the guinea pigconsistently produced endolymphatic hydrops. This wasthe first description of a reliable animal model of en-dolymphatic hydrops. Guinea pigs with hydrops did notdemonstrate signs of episodic vestibular disturbances,but they showed shifts in auditory thresholds, as deter-mined by electrophysiologic studies (16,17). Light mi-croscopic studies revealed partial degeneration of haircells, stria vascularis, and neurons, especially in the api-cal turn, but this was insufficient to account for the shiftsin auditory thresholds (16,17), a finding analogous toobservations in human temporal bones. New light wasshed on the histopathology of hydrops in the guinea pigby the findings of Ichimiya et al. (18) and Nadol et al.(19), who demonstrated cytochemical and ultrastructurallesions affecting the fibrocytes within the spiral liga-ment, hair cells, and cochlear neurons. However, it wasnot known which came first—the hydrops or the cyto-pathologic changes—raising the question as to whetherhydrops had induced the observed abnormalities or viceversa. Another unanswered question was the mechanismof development of hydrops after blockage of the en-

dolymphatic duct. It had been traditionally assumed thatthe hydrops was the result of blockage of longitudinalflow of endolymph from the cochlea to the endolym-phatic sac (11,15). However, measurements of en-dolymph flow in the guinea pig showed the rate of lon-gitudinal flow to be exceedingly small and thereforeinsufficient to account for the development of hydropson the basis of blockage of flow (20,21). Some answersto these questions have emerged by the recent work ofShinomori et al. (22), who demonstrated that obliterationof the endolymphatic duct resulted in disruption of en-zymes within the spiral ligament, indicating dysregula-tion of inner ear fluids. Remarkably, these changes oc-curred before the development of hydrops. A second goalof this article is to discuss the implications of these recentobservations with respect to experimental hydrops andMénière’s syndrome.

METHODS

The human temporal bone collection at the MassachusettsEye and Ear Infirmary currently contains 1,750 specimens from963 individuals. These temporal bones were prepared in thestandard manner for light microscopic study including fixationin 10% formalin, decalcification with ethylene diamine tetra-acetic acid, embedment in celloidin, serial sectioning at a thick-ness of 20 �m, and staining of every 10th section with hema-toxylin and eosin (11). A search of the database containing therecords of our temporal bones was conducted in two ways: 1)searching for cases with a clinical diagnosis of Ménière’s syn-drome (a clinical diagnosis of Ménière’s syndrome was madewhen the case history included both episodic vertigo and sen-sorineural hearing loss (23)) and 2) searching for cases with ahistopathologic description of endolymphatic hydrops in atleast one ear. Endolymphatic hydrops was defined for the pur-pose of this study as hydrops present in any part of the mem-branous labyrinth with the exception of apical cochlear hy-drops. This was based on the observations of Yamashita andSchuknecht (24) that apical endolymphatic hydrops occurs inapproximately 15% of temporal bones, that it cannot be corre-lated with any type of ear disease, and that it is probably of nopathologic or functional significance. The extent and severityof hydrops was characterized in each sense organ of the innerear (cochlea, saccule, utricle, and the three canals) by examin-ing serial sections using the same criteria as we have in the past(14). The hydrops was characterized as idiopathic when therewas no obvious abnormality in the temporal bone that could beconsidered causal or as “secondary,” based on the coexistence ofother pathologic changes in the inner ear such as surgicaltrauma, temporal bone fracture, labyrinthitis, syphilis, Cogan’ssyndrome, and a variety of neoplasms affecting the labyrinth(11). The case histories and histologic slides of specimens thatmet the above criteria were reviewed to examine the relation-ship between hydrops and symptoms.

RESULTS

Searching for a clinical diagnosis ofMénière’s syndrome

Twenty-eight cases had a clinical diagnosis ofMénière’s syndrome (Fig. 2). Every patient with thisdiagnosis had endolymphatic hydrops on histopathologic

FIG. 1. Central hypothesis for Ménière’s syndrome (modifiedafter Kiang (10)). Many possible etiologic factors can lead toendolymphatic hydrops, which in turn generates the clinicalsymptoms.

75PATHOPHYSIOLOGY OF MÉNIÈRE’S SYNDROME

Otology & Neurotology, Vol. 26, No. 1, 2005

examination in the affected ear. The hydrops was idio-pathic in 26 cases and considered secondary in 2 cases.All 26 cases with idiopathic hydrops showed dilatationof the cochlear duct and the saccule, and the majority (17cases [65%]) also had hydrops involving the utricleand/or the ampullae. The two cases with secondary hy-drops consisted of 1) a 55-year-old woman with episodicvertigo and sensorineural hearing loss after a right sta-pedectomy whose temporal bone showed cochlear hy-drops, collapse of the wall of the saccule, and disruptionwith atrophy of the saccular macula (a case of posttrau-matic hydrops); and 2) a 63-year-old man with episodicvertigo, bilateral progressive sensorineural hearing loss,and intractable urticaria whose temporal bones showedbilateral cochlear hydrops, collapse of the utricular andampullary walls, degeneration of the auditory and ves-tibular end organs, focal areas of fibrous proliferationand new bone formation within the labyrinth, and scat-tered aggregates of inflammatory cells within the en-dolymphatic and perilymphatic spaces (a case of pre-sumed autoimmune inner ear disease).

Searching for a histopathologic description ofendolymphatic hydrops

Seventy-nine cases were identified as having en-dolymphatic hydrops in at least one ear, of which 35cases were idiopathic and 44 were considered to be sec-ondary (Fig. 2). Of the 35 cases with idiopathic hydrops,26 cases had a clinical history of Ménière’s syndromeand 9 cases did not. Additional details of the nine caseswith idiopathic hydrops are shown in Table 1. None ofthe nine individuals had a history of episodic vertigo,despite hydrops of the cochlea and/or the vestibular ap-paratus. One case (Case 1) had normal hearing, and the

temporal bones showed saccular hydrops on both sides.The remaining eight cases had varying degrees of sen-sorineural hearing loss, which was fluctuating, sudden,or progressive in nature. The audiometric pattern of theloss was variable, and included low tone, flat, and down-sloping patterns. The hydrops was limited to the cochleain one case, whereas the other seven cases had hydropsaffecting the cochlea, saccule, and utricle (with or with-out the ampullae). The severity and extent of hydrops inthese 7 cases was similar to that observed in the 26 caseswith Ménière’s syndrome. Many specimens with idio-pathic hydrops also demonstrated premortem rupture ofthe membranous labyrinth. Some of the nine cases withidiopathic hydrops showed partial loss of hair cellsand/or afferent neurons within the cochlear and vestibu-lar end organs. However, none of the ears showed severeor total degeneration of the end organs (such degenera-tion would confound the interpretation of the role ofhydrops in causing symptoms). An analysis of the 44cases with secondary hydrops revealed the followingbreakdown:

1. Only two individuals had the Ménière’s symptomcomplex during life; brief descriptions of bothcases are given above.

2. There were 10 cases without a history of episodicvertigo. The hydrops involved only the cochlea intwo cases, whereas the remaining eight had co-chleosaccular hydrops (with or without hydrops af-fecting the utricle and ampullae). The extent andseverity of hydrops in these eight cases was similarto that observed in bones from individuals withMénière’s syndrome, including the presence ofruptures affecting the membranous labyrinth inseveral bones. All 10 cases exhibited varying de-

FIG. 2. Results of searches of database containing temporal bone records.

76 S. N. MERCHANT ET AL.


grees of sensorineural hearing loss in the hydropicear. However, none of the 10 cases had severe orcomplete atrophy of the auditory or vestibularend organs.

3. There were an additional 32 cases without an ac-companying history of symptoms of Ménière’ssyndrome, but no firm conclusions could be drawnabout the relationship between hydrops and the ab-sence of symptoms in these cases because of con-founding factors such as degeneration of the endorgans, inadequate medical records, or young ageof the subject (which might preclude reporting ofvertigo or hearing loss). Two representative casehistories of individuals with idiopathic hydrops butwithout the Ménière’s symptom complex are pre-sented below along with the histopathologic find-ings (Cases 5 and 6 from Table 1).

Case 5A 56-year-old woman underwent bilateral myringoto-

mies for middle ear effusions. An audiogram at age 58revealed good hearing for air conduction in both ears(Fig. 3A). She developed a bilateral progressive, down-sloping, sensorineural hearing loss that was worse in theleft ear. Audiometric evaluations were performed at ages73 and 81 (Fig. 3A). She wore a hearing aid with successin her right ear. She did not experience episodic vertigo.At the age of 81, she was given radioactive iodine forhyperthyroidism. She developed diabetes mellitus at age82 and malignant lymphoma at age 83, for which shereceived therapy with cyclophosphamide for 6 months.She fell at the age of 87, breaking her hip, which wascorrected by surgery. At age 87, she stated that the hear-ing aid was working well in her right ear and that she had“difficulties with balance.” She was using a cane andwalker. She died at the age of 87.

Histologic findingsThe left inner ear shows endolymphatic hydrops af-

fecting all turns of the cochlea (Fig. 3B). There is severedilatation of the vestibular cecum that has expanded tofill the bony vestibule. There is a fistula between thehydropic vestibular cecum and the saccule (Fig. 3B).There is hydrops involving the saccule, the utricle, andthe ampulla of the superior semicircular canal. The organof Corti is intact in all turns. The basilar membrane isthickened in the basal 5 to 6 mm of the organ of Corti.There is moderate to severe atrophy of the stria vascu-laris in the middle and apical turns. The cochlear neuro-nal population appears normal for age. All vestibularsense organs show good populations of hair cells, sup-porting cells, and nerve fibers. There is a granular baso-philic deposit on the cupula of the posterior canal. In theright inner ear, the cochlea is similar to the opposite sidebut there is no endolymphatic hydrops. The sacculeshows degenerative changes involving the otolithicmembrane associated with cytoarchitectural irregulari-ties of the sensory epithelium. The utricular macula andall three cristae appear normal. There is a clump of amor-phous basophilic granular material in the nonampullatedend of the duct of the lateral canal.

CommentThis patient had bilateral, progressive, down-sloping

sensorineural hearing loss, worse in the left ear, withoutepisodic vertigo. Although there is extensive endolym-phatic hydrops of the cochlear and vestibular apparatusin the left ear, she did not experience episodic vertigoor fluctuating hearing loss. The down-sloping sensori-neural hearing loss cannot be fully explained on the basisof the observed otopathologic findings and is consistentwith a diagnosis of “cochlear conductive presbycusis”(11). The “difficulties with balance” at the end of her lifemay have been multifactorial in nature, associated with

TABLE 1. Cases with idiopathic hydrops without the classic Ménière’s symptom complex

CaseSex/age

(yr) SideEpisodicvertigo

Sensorineuralhearing loss

Audiometric patternand severity

Hydrops Ruptures ofmembranous

labyrinthCo S U Sup Lat Post

1 M/6 Right No No Normal hearing +Left No No Normal hearing +

2 M/34 Right No Yes, fluctuating 35-dB low-tone loss + + + + Between sacculeand utricle

3 F/42 Left No Yes, fluctuating 50 to 70-dB flat loss + + + Reissner’s saccule4 M/62 Left No Yes, fluctuating 30 to 50-dB

down-sloping loss+

5 F/87 Left No Yes, progressive 75 to 105-dBdown-sloping loss

+ + + + Vestibular cecum

6 M/78 Left No Yes, progressive 70 to 90-dB loss,greater in low tones

+ + + + + Lat ampulla, saccule,common crus

7 F/86 Right No Yes, progressive 90 to 100-dB flat loss + + + + + + Saccule8 M/77 Right No Yes, progressive Profound loss + + + + + + Reissner’s saccule9 M/91 Right No Yes, initially sudden,

then progressive80 to 95-dB flat loss + + + + + + Lat, post ampullae

Left No Yes, severe losssince childhood

70 to 95-dB flat loss + + + + + + All three ampullae,utricle

M, male; F, female; +, presence of hydrops; Co, cochlea; S, saccule; U, utricle; Sup, superior canal; Lat, lateral canal; Post, posterior canal.



hip replacement surgery, degenerative vestibulopathy in-volving the saccule in the right ear and cupulolithiasis ofthe posterior canal in the left ear, and possible diabeticperipheral neuropathy.

Case 6A 78-year-old man was hospitalized for squamous cell

carcinoma of the right maxillary sinus. He reported hav-ing progressive hearing loss in both ears for several yearsand that it was worse on the left. He had occasionaltinnitus. He denied vertigo. He had a history of exposureto noise as a factory worker. An audiogram 6 days beforedeath showed a bilateral, moderate to profound, 70- to

90-dB sensorineural hearing loss that was worse on theleft. Speech discrimination was 92% on the right and60% on the left.

Histologic findingsThe left inner ear shows moderate to severe endolym-

phatic hydrops in all cochlear turns along with severehydrops of the saccule (Fig. 4A). Although there is com-pression artifact, the hair cells and stria vascularis appearto be intact. There is a partial loss of cochlear neurons inthe basal turn. The macula of the saccule appears normal.The utricle and its macula appear normal. There is hy-drops of the posterior ampulla and hydrops with a pre-mortem rupture of the lateral ampulla (Fig. 4B). Thereare ruptures involving the saccule and common crus. Allthree cristae, the vestibular nerves, and the bony laby-rinth appear normal. The right inner ear shows a normalorgan of Corti and stria vascularis with a partial loss of

FIG. 3. Case 5 from Table 1, an 87-year-old woman. (A) Au-diograms at ages 58, 73, and 81 showing a bilateral, progressive,down-sloping sensorineural hearing loss, which was worse onthe left. (B) Photomicrograph of the left inner ear. There is mod-erate to severe endolymphatic hydrops affecting the middle andapical turns of the cochlea. There is severe dilatation of the ves-tibular cecum that has expanded to fill the bony vestibule. Thereis also a fistula between the hydropic vestibular cecum and thesaccule. IAC, internal auditory canal.

FIG. 4. Case 6 from Table 1, a 78-year-old man. (A) Photomi-crograph of left inner ear showing severe endolymphatic hydropsaffecting the cochlea and saccule. There is an artifactual tear ofthe basilar membrane in the lower middle turn (*). IAC, internalauditory canal. (B) Photomicrograph of ampulla of the lateralsemicircular canal of the left ear, showing hydrops, areas of thin-ning of the wall of the membranous labyrinth, and an area of out-pouching/rupture. The smooth outline of the area of the ruptureand the lack of torn membranes suggest that the rupture waspremortem in nature. The sensory epithelium of the crista andnerve fibers appears normal.



cochlear neurons in the basal turn. There is no endolym-phatic hydrops. The utricular and saccular maculae andthe ampullae of all three canals including hair cells, sup-porting cells, and vestibular nerve fibers appear normal.

CommentThis patient did not experience vestibular complaints

or fluctuating hearing loss, despite having moderatelysevere endolymphatic hydrops in the left ear involvingthe auditory and vestibular end organs.

DISCUSSION

The finding that endolymphatic hydrops was presentin every individual with a clinical diagnosis of Ménière’ssyndrome establishes the correlation between Ménière’ssyndrome and hydrops. Such a correlation, however,does not imply a cause-and-effect relationship. If thecentral hypothesis (Fig. 1) were true, every case ofMénière’s syndrome should have endolymphatic hy-drops and every case of hydrops should show the clinicalsymptoms. We identified nine cases of idiopathic hy-drops from patients who did not have the classic symp-toms of Ménière’s syndrome during life. One of the ninecases who had saccular hydrops had normal hearing andno vertigo (Case 1), whereas the remaining eight caseshad sensorineural hearing loss without vertigo. The hear-ing loss was fluctuant in only three of the eight cases.The audiometric pattern varied and included low tone,and flat and down-sloping losses. With the exception ofCase 1, the extent and severity of the hydrops in thesecases were similar to that observed in the temporal bonesof individuals with the Ménière’s symptom complex.Furthermore, we found a number of cases of secondaryhydrops who also did not suffer from the classic symp-tom complex of Ménière’s syndrome.

Our results indicate that 1) endolymphatic hydrops ofthe cochlea is invariably associated with a sensorineuralhearing loss, but not necessarily a fluctuating type or alow-tone pattern; and 2) hydrops of the cochlea and/orvestibular system is not necessarily associated with ahistory of episodic vertigo. Therefore, these results arenot consistent with the central hypothesis of hydrops asthe final common pathway for production of symptomsin Ménière’s syndrome. The results are more consistentwith hydrops being a marker for disordered homeostasisof the labyrinth, in which some factor (as yet unknown)produces both the clinical symptoms of Ménière’s syn-drome and endolymphatic hydrops.

The results of the current study are similar to those ofthe previous temporal bone study from our laboratory in1989 regarding the relationship between hydrops andsymptoms (14). The current study included the additionaltemporal bone cases accrued over the past 15 years aswell as cases with secondary hydrops. A review of theliterature reveals reports of cases with asymptomatic hy-drops (25–28) as well as cases of Ménière’s syndromediagnosed during life without demonstrable hydrops athistologic examination (27,29–33). The central role ofhydrops in mediating the pathophysiology of the

Ménière’s symptom complex has also been questionedby other authors (34,35).

We point out that temporal bone studies have someinherent limitations that prevent an unequivocal rejectionof the central hypothesis. For example, the lack of ves-tibular symptoms in patients with hydrops can be ex-plained on the basis that symptoms were present but notdocumented in the medical history. Although we cannotdiscount such an explanation, they appear unlikely in atleast some of our cases (e.g., Cases 5, 7, and 9 in Table1) who had multiple otolaryngologic evaluations duringlife. It has also been suggested that the lack of vertigo incases with hydrops might be explained on the basis thatthese bones did not have ruptures of the membranouslabyrinth or that the hydrops was nonprogressive (11,25).Many of the cases in the current study showed clearexamples of membrane ruptures, but there was no historyof vertigo, which brings into question the rupture theoryas an explanation for the episodic vertigo in Ménière’ssyndrome. We have no way of establishing whether thehydrops in asymptomatic cases was progressive or not.However, the hydrops was quite severe in many of ourasymptomatic cases, with the dilated membranous laby-rinth having reached the limits of expansion within thebony confines of the cochlea or the vestibule. Thus, thehydrops must have been progressive at some point intime. A more direct way to confirm or refute the centralhypothesis is to visualize the endolymphatic space inliving subjects by imaging techniques (10). Indeed, mag-netic resonance imaging has been recently used to diag-nose hydrops in vivo in the guinea pig (36,37). Thesestudies demonstrate the potential of in vivo imaging topermit a systematic assessment of the dynamic eventsover time in patients with and without Ménière’s syndrome.

Much research has focused on the pathology andpathophysiology of hydrops in the guinea pig model inan attempt to better understand Ménière’s syndrome. Hy-drops starts to develop within a few days after oblitera-tion of the endolymphatic duct in the guinea pig and isconsistently observed by 1 week postoperatively (15,16,37). Although it had been traditionally assumed that thehydrops was the result of blockage of longitudinal flowof endolymph from the cochlea to the endolymphatic sac(11,15), measurements of the rate of endolymph flow inthe guinea pig showed the rate of longitudinal flow to beexceedingly small and inadequate to account for the de-velopment of hydrops (20,21). Thus, the mechanism ofhow hydrops occurred after blockage of the endolym-phatic duct remained unknown. The cause of auditorythreshold shifts in hydropic animals also remained ob-scure and was attributed to being the result of the hy-drops (11,16).

New light was shed on the histopathology of hydropsin the guinea pig by the findings of Ichimiya et al. (18)and Nadol et al. (19), who demonstrated cytochemicaland ultrastructural lesions at the level of hair cells, co-chlear neurons, and the spiral ligament. The essence oftheir investigations was that the Type I fibrocytes of thespiral ligament were the most severely affected cochlear



cells, although Type II fibrocytes were also clearly af-fected in animals with long-term survival. However, itwas not known which came first—the hydrops or thecytopathologic changes—raising the question as towhether hydrops had induced the observed abnormalitiesor vice versa. More recently, Shinomori et al. (22) ex-tended these findings to show that the cytochemistry offibrocytes and other nonsensory cells changed before theinduction of hydrops. Shinomori et al. (22) blocked theendolymphatic duct to induce hydrops in 22 guinea pigsand studied the cytochemistry of the inner ear at variouspostoperative survival times ranging from 1 day to 3months. A striking finding was that there were changesin the cytochemistry of Type I and Type II fibrocytes andof nonsensory epithelial cells 1 day after surgery, beforethe development of hydrops. The Type I fibrocytesshowed increased immunostaining for the NaK2Cl co-transporter (NKCC1), along with decreased immuno-staining for taurine and C-Jun-N-terminal kinase (JNK).

These results are interesting because the fibrocytes ofthe spiral ligament, which contain a variety of gap junc-tions, enzymes, and proteins, are known to play crucialroles in maintaining fluid homeostasis within the cochlea(38–41). For example, the Type I and Type II fibrocytesare involved in the transport of K+ ions from the rootcells to the stria vascularis, thus enabling the recycling ofK+ ions within the scala media (38,40). NKCC1 is in-volved in regulation of cellular volume in response toosmotic stresses (42). Taurine is an amino acid thatwidely serves as an osmolyte (43). JNK is known to bea critical player in many forms of cellular stress re-sponses (44). The increased immunostaining for NKCC1and decreased staining for taurine in the Shinomori et al.(22) study suggest that the Type I fibrocytes were os-motically stressed and that the cytochemical changes re-flected compensatory mechanisms for regulation of cellvolume. The decrease in staining for JNK indicates thatthe fibrocytes were under stress, probably because of achange in perilymphatic conditions induced by the sur-gical procedure. The connective tissue surrounding theendolymphatic duct is in communication with the peri-lymphatic fluid spaces of the vestibular and cochlear endorgans (45). We hypothesize that surgical obstruction ofthe endolymphatic duct altered the cytochemistry of theperilymph (in some unknown manner), resulting in cel-lular stress and dysfunction of fibrocytes of the spiralligament. In turn, this may have interfered with the re-cycling of K+ ions, resulting in an osmotic imbalance andexpansion of the endolymphatic compartment (i.e., en-dolymphatic hydrops). In other words, the findings sug-gest that hydrops is the result (rather than the cause) ofdisordered cochlear homeostasis. There is another line ofevidence to support the hypothesis that dysfunction ofthe spiral ligament results in endolymphatic hydrops.Transgenic mice with deletion of the Brn-4 transcriptionfactor, which is expressed in fibrocytes of the spiral liga-ment, develop endolymphatic hydrops (46,47). Theseanimals also suffer from a hearing loss and show de-creased immunoreactivity for NaK-ATPase, NKCC1,

and connexin-31 within the Type II fibrocytes of theligament (47).

It has long been a practice to consider endolymphatichydrops as a sign of excessive endolymph productionwith respect to its absorption (11). This conceptualframework has contributed to misconceptions concern-ing the cause of hydrops. It should be remembered thatthe fluid volume within the bony labyrinth remains con-stant (21). Changes in the volumes of the endolymphaticand perilymphatic compartments are responses to os-motic gradients between the compartments. In the co-chlea, the major potential for creation of abnormal os-motic gradients is associated with the flux of K+ ions intothe endolymphatic compartment from the stria vascu-laris, a flux that reflects the current flow underlying theendolymphatic potential (21). This current flow is con-tinuous and is largely independent of levels of sound-induced activity of the sensory cells (48). For osmoticpressure to remain constant in the presence of this con-stant inward flux of potassium ions, there must be anequal efflux from the scala media via a return pathway.Zidanic and Brownell (48) measured current flow in thecochlea and concluded that return paths of K+ ions in-cluded sensory and nonsensory cells on the basilar mem-brane, as well as Reissner’s membrane. Chiba and Mar-cus (49) reported a potential K+ return pathway throughcells in the region below the spiral prominence. Irrespec-tive of the exact return paths, delineation of cochlear gapjunctional pathways together with considerations of tightjunction barriers and specializations for ionic uptake ledKikuchi et al. (38) to conclude that Type II fibrocytes arethe major site of uptake of K+ ions from the perilym-phatic space for their return to the stria vascularis. TypeI fibrocytes are part of that return pathway. The results ofIchimiya et al. (18) and Nadol et al. (19) indicated thatspiral ligament fibrocytes and other cells are profoundlyaffected in the hydropic guinea pig. The results of Shi-nomori et al. (22) show that Type I fibrocytes and othercells are affected before the onset of hydrops, whichindicates that spiral ligament abnormality is not abyproduct of the processes that induce the hydrops.Clearly, much remains to be clarified regarding the un-derlying cytologic changes that lead to the induction ofhydrops. The key point of the present publication is thatthere is a pressing need for a search for the cellular andmolecular bases of the various symptoms of Ménière’ssyndrome, because the evidence indicates that hydropsper se is not the cause. Therefore, therapeutic strategieswhose main goal is the reduction of hydrops (e.g., en-dolymphatic sac shunting, sacculotomy) are unlikely tocontrol the disorder. A better understanding of the syn-drome at a cellular level may uncover targets for noveltherapeutic interventions.

CONCLUSION

Endolymphatic hydrops should be considered as a his-tologic marker for Ménière’s syndrome rather than beingdirectly responsible for its symptoms.



Acknowledgments: The authors thank Mr. Axel Eliasen forsupport of this work and the NIDCD National Temporal Bone,Hearing and Balance Pathology Resource Registry for identi-fying temporal bone cases with Ménière’s syndrome or en-dolymphatic hydrops.

REFERENCES

1. Ménière P. Sur une forme de surdité grave dépendant d’une lésionde l’oreille interne. Gaz Méd de Paris 1861;16:29.

2. Flourens MJP. Expériences sur les canaux semi circulaires del’oreille. Mém Acad de Sci 1830;9:455–77.

3. Yamakawa K. Uber die pathologisch Veranderung bei einemMénière-Kranken. J Otorhinolaryngol Soc Jpn 1938;4:2310–2.

4. Hallpike CS, Cairns H. Observations on the pathology ofMénière’s syndrome. J Laryngol Otol 1938;53:625–55.

5. Lindsay JR. Labyrinthine dropsy and Ménière’s disease. Arch Oto-laryngol 1942;35:853–67.

6. Altmann F, Fowler EP Jr. Histological findings in Ménière’s symp-tom complex. Ann Otol Rhinol Laryngol 1943;52:52–80.

7. Cawthorne T. Ménière’s disease. Ann Otol Rhinol Laryngol 1947;56:18–38.

8. Day KM, Lindsay JR. Hydrops of the labyrinth: case report—coagulation operation, clinical course and histopathology. Laryn-goscope 1949;59:213–27.

9. Nager FR. Zur Histopathologie des Ohrschwindels. Pract Otorhi-nolaryngol 1949;11:360–77.

10. Kiang NYS. An auditory physiologist’s view of Ménière’s syn-drome. In Nadol JB Jr, ed. Second International Symposium onMénière’s disease. Amsterdam: Kugler & Ghedini, 1989:13–24.

11. Schuknecht HF. Pathology of the Ear. 2nd ed. Philadelphia: Lea &Febiger, 1993.

12. Merchant SN, Rauch SD, Nadol JB. Meniere’s disease. Eur ArchOtorhinolaryngol 1995;252:63–75.

13. Nadol JB Jr. Pathogenesis of Meniere’s syndrome. In Harris JP, ed.Ménière’s Disease. The Hague, The Netherlands: Kugler Publica-tions, 1999:73–9.

14. Rauch SD, Merchant SN, Thedinger BA. Ménière’s syndrome andendolymphatic hydrops: a double-blind temporal bone study. AnnOtol Rhinol Laryngol 1989;98:873–83.

15. Kimura RS, Schuknecht HF. Membranous hydrops in the inner earof the guinea pig after obliteration of the endolymphatic sac. PractOtorhinolaryngol 1965;27:343–54.

16. Kimura RS. Animal models of endolymphatic hydrops. Am J Oto-laryngol 1982;3:447–51.

17. Horner KC. Functional changes associated with experimentallyinduced endolymphatic hydrops. Hear Res 1993;68:1–18.

18. Ichimiya I, Adams JC, Kimura RS. Changes in immunostaining ofcochleas with experimentally induced endolymphatic hydrops. AnnOtol Rhinol Laryngol 1994;103:457–68.

19. Nadol JB Jr, Adams JC, Kim JR. Degenerative changes in theorgan of Corti and lateral cochlear wall in experimental endolym-phatic hydrops and human Ménière’s disease. Acta Otolaryngol1995;519(Suppl):47–59.

20. Salt AN, Thalmann R, Marcus DC, et al. Direct measurement oflongitudinal endolymph flow rate in the guinea pig cochlea. HearRes 1986;23:141–51.

21. Salt AN. Regulation of endolymphatic fluid volume. Ann N Y AcadSci 2001;942:306–12.

22. Shinomori Y, Kimura RS, Adams JC. Changes in immunostainingfor Na+, K+, 2Cl-cotransporter 1, taurine and c-Jun N-terminalkinase in experimentally induced endolymphatic hydrops. AROAbstr 2001;24:134.

23. Committee on Hearing and Equilibrium guidelines for the diagno-sis and evaluation of therapy in Meniere’s disease: AmericanAcademy of Otolaryngology-Head and Neck Foundation, Inc. Oto-laryngol Head Neck Surg 1995;113:181–5.

24. Yamashita T, Schuknecht HF. Apical endolymphatic hydrops.Arch Otolaryngol 1982;108:463–6.

25. Schuknecht HF, Gulya AJ. Endolymphatic hydrops: an overviewand classification. Ann Otol Rhinol Laryngol Suppl 1983;106:1–20.

26. Sperling NM, Paparella MM, Yoon TH, Zelterman D. Symptom-atic versus asymptomatic endolymphatic hydrops: a histopatholog-ic comparison. Laryngoscope 1993;103:277–85.

27. Linthicum FH. Histopathology of Meniere’s-like conditions. InHarris JP, ed. Ménière’s Disease. The Hague, The Netherlands:Kugler Publications, 1999:53–66.

28. Vasama JP, Linthicum FH Jr. Meniere’s disease and endolym-phatic hydrops without Meniere’s symptoms: temporal bone his-topathology. Acta Otolaryngol 1999;119:297–301.

29. Arnvig J. Histological findings in a case of Ménière’s disease withremarks on the pathologic-anatomic basis of this lesion. Acta Oto-laryngol 1947;35:453–66.

30. Brunner H. Ménière’s disease. J Laryngol Otol 1948;62:627–38.31. Berggren S. Histological investigation of three cases with

Ménière’s syndrome. Acta Otolaryngol 1949;37:30–5.32. Belal A Jr, Ylikoski J. Pathologic significance of Ménière’s symp-

tom complex: a histopathologic and electron microscopic study.Am J Otolaryngol 1980;1:275–84.

33. Fraysse BG, Alonso A, House WF. Ménière’s disease and en-dolymphatic hydrops: clinical-histopathologic correlations. AnnOtol Rhinol Laryngol Suppl 1980;89:2–22.

34. Rickenstein MJ, Harrison RV. Cochlear pathophysiology in Me-niere’s disease: a critical appraisal. In Harris JP, ed. Ménière’sDisease. The Hague, The Netherlands: Kugler Publications,1999:195–202.

35. Gacek RR, Gacek MR. Meniere’s disease as a manifestation ofvestibular ganglionitis. Am J Otolaryngol 2001;22:241–50.

36. Niyazov DM, Andrews JC, Strelioff D, Sinha S, Lufkn R. Diag-nosis of endolymphatic hydrops in vivo with magnetic resonanceimaging. Otol Neurol 2001;22:813–7.

37. Zou J, Pyykko I, Bretlau P, Klason T, Bjelke B. In vivo visual-ization of endolymphatic hydrops in guinea pigs: magnetic reso-nance imaging evaluation at 4.7 tesla. Ann Otol Rhinol Laryngol2003;112:1059–65.

38. Kikuchi T, Kimura RS, Paul DL, Adams JC. Gap junctions in therat cochlea: immunohistochemical and ultrastructural analysis.Anat Embryol 1995;191:101–18.

39. Spicer SS, Schulte BA. The fine structure of spiral ligament cellsrelates to ion return to the stria and varies with place-frequency.Hear Res 1996;100:80–100.

40. Steel KP. The benefits of recycling. Science 1999;285:1363–4.41. Wangemann P. K(+) cycling and the endocochlear potential. Hear

Res 2002;165:1–9.42. Flatman PW. Regulation of Na-K-2C1 cotransport by phosphory-

lation and protein-protein interactions. Biochem Biophys Acta2002;1566:140–51.

43. Lambert IH. Regulation of the cellular content of the organic os-molyte taurine in mammalian cells. Neurochem Res 2004;29:27–63.

44. Barr RK, Bogoyevitch MA. The c-Jun N-terminal protein kinasefamily of mitogen-activated protein kinases (JNK MAPKs). Int JBiochem Cell Biol 2001;33:1047–63.

45. Yeo SW, Gottsschlich S, Harris JP, Keithley EM. Antigen diffu-sion from the perilymphatic space of the cochlea. Laryngoscope1995;105:623–8.

46. Minowa O, Ikeda K, Sugitani Y, et al. Altered cochlear fibrocytesin a mouse model of DFN3 nonsyndromic deafness. Science 1999;285:1408–11.

47. Xia A-P, Kikuchi T, Minowa O, et al. Late-onset hearing loss in amouse model of DFN3 non-syndromic deafness: morphologic andimmunohistochemical analyses. Hear Res 2002;166:150–8.

48. Zidanic M, Brownell WE. Fine structure of the intracochlear po-tential field: I—the silent current. Biophys J 1990;57:1253–68.

49. Chiba T, Marcus DC. Nonselective cation and BK channels inapical membrane of outer sulcus epithelial cells. J Membr Biol2000;174:167–79.



pathophysiology of ménière's syndrome: are symptoms caused by

Documents