Triple Semicircular Canal Occlusion in Guinea Pigs with Endolymphatic Hydrops

Transcription

1 Otology & Neurotology 27:78 85 Ó 2005, Otology & Neurotology, Inc. Triple Semicircular Canal Occlusion in Guinea Pigs with Endolymphatic Hydrops *Shankai Yin, *Dongzhen Yu, *Manna Li, and Jian Wang *Department of Otolaryngology, Affiliated Sixth People s Hospital, Shanghai Jiao Tong University, Shanghai, China, and School of Human Communication Disorders, Dalhousie University, Halifax, Nova Scotia, Canada Hypothesis: Triple semicircular canal occlusion will eliminate rotatory stimulation to the vestibular peripheral system (as it blocks endolymphatic fluid movement) and therefore release rotatory vertigo attack. This surgery is safe in ears with endolymphatic hydrops. Background: Semicircular canal occlusion has been used as an alternative treatment of intractable benign paroxysmal positional vertigo with varied success. Triple semicircular canal occlusion in animal models blocks the responses of the semicircular canals to rotation and spares cochleae and the otolithic apparatus. This result suggests that triple semicircular canal occlusion is a prospective method in vertigo management for patients with Ménière s disease. However, the effectiveness and safety of triple semicircular canal occlusion has not been fully evaluated in ears with endolymphatic hydrops. Methods: Endolymphatic hydrops was established in 20 guinea pigs by endolymphatic sac obliteration. Triple semicircular canal occlusion was performed in 12 of them 120 days after endolymphatic hydrops surgery, whereas 8 others were killed for morphologic observation to confirm endolymphatic hydrops. Auditory and vestibular functions were monitored from the time before endolymphatic hydrops until 1 month after triple semicircular canal occlusion. Endolymphatic hydrops and canal occlusion were confirmed by morphologic observation. Results: Successful establishment of endolymphatic hydrops was indicated by mild elevation of the auditory brainstem response threshold and tentative asymmetry in nystagmus. Endolymphatic hydrops was confirmed by cochlear morphology in all eight animals that were killed 120 days after endolymphatic hydrops surgery. After triple semicircular canal occlusion, all 12 animals showed spontaneous nystagmus with a slow component toward the side that had been operated on, head tilt, rotated walking, and tentative asymmetry in rotatory nystagmus. The static symptoms disappeared within 1 month after triple semicircular canal occlusion. Caloric nystagmus was only slightly reduced after endolymphatic hydrops as compared with the contralateral ears but could not be elicited at all after triple semicircular canal occlusion. No significant elevation in auditory brainstem response threshold was found after triple semicircular canal occlusion. The canal occlusion and endolymphatic hydrops were confirmed in all surgical ears. Conclusion: Triple semicircular canal occlusion is effective for eliminating the response of semicircular canals to rotation and caloric stimulation and is safe in ears with endolymphatic hydrops. Also, the static compensation to the disequilibrium is quick and complete. These results suggest that triple semicircular canal occlusion should be an option for controlling rotatory vertigo in Ménière s disease. Key Words: Endolymphatic hydrops Guinea pig Ménière s disease Semicircular canal occlusion Vertigo. Otol Neurotol 27:78 85, Ménière s disease is an idiopathic inner ear disorder that has a host of symptoms including episodic vertigo, tinnitus, fluctuating hearing loss, and a sensation of pressure or pain in the affected ear(s). Although the same symptoms may result from different causes (e.g., circulation abnormalities, viral infections, allergies, autoimmune reactions, and genetic disorders), most researchers believe that the pathologic basis of this disorder is excess Address correspondence and reprint requests to: Jian Wang, Ph.D., 5599 Fenwick Street, Halifax, Nova Scotia B3H 1R2, Canada; jian.wang@dal.ca Supported by Science and Technology Committee of Shanghai grant 98QB endolymph in the inner ear, and a vertigo attack occurs when potassium-rich endolymph leaks into the sodiumrich perilymph (1). The role of endolymphatic hydrops (ELH) in Ménière s disease is supported by studies using animal models. Since the discovery of ELH in temporal bones from patients with Ménière s disease by Hallpike and Cairns in 1938 (2), animal models have been established via several different approaches (3 5). The auditory and vestibular dysfunctions observed in these animal models show remarkable similarity to cases of Ménière s disease. In animals with unilateral ELH, the number of instances of nystagmus induced by rotation toward the operated ear decreases, suggesting the impact of ELH on 78

2 TRIPLE SEMICIRCULAR CANAL OCCLUSION IN ELH 79 vestibular function (6,7). This is usually accompanied by characteristic changes in compound action potentials and summation potentials that are seen in patients with Ménière s disease. Also, audiograms in both animal models and human cases usually reveal fluctuated hearing loss, which is typically initiated at low frequency, and gradually extended to higher frequencies. Currently, there is no cure for Ménière s disease. In the early stage of Ménière s disease, the attack can be controlled via conservative medical management, including a low-salt diet and the use of diuretics, in the majority of patients. However, vertigo persists despite optimal conservative treatments in approximately 10% of cases; these cases are considered to have intractable Ménière s disease (8,9). So far, many different methods have been adopted in the management of these patients, including chemical labyrinth ablation, endolymphatic sac surgery, labyrinthectomy, and vestibular neurectomy. Aminoglycoside antibiotics have been used as a method of chemical labyrinth ablation for over 50 years in Ménière s disease. These antibiotics can be administrated systematically or locally. Each approach has its own advantages and disadvantages. A common problem is the difficulty in controlling vertigo effectively while preserving hearing. This problem is remarkable because ears with ELH are more vulnerable to the ototoxicity of the drugs (10). Some aminoglycosides, such as gentamicin, are identified as being relatively vestibulotoxic. Appropriate application of these antibodies can effectively reduce vertigo while preserving hearing function (11,12). Single intratympanic injection of gentamicin was reported to effectively control vertigo in most patients (12,13). The effect was attributed to the damage to the treated labyrinth caused by gentamicin as indicated by the reduced caloric response and the gain of angular vestibulo-ocular reflex. However, recurrent vertigo attacks were reported in some cases 1 or 2 years after treatment (14). In addition, symptoms caused by chemical labyrinthectomy required a much longer time to be compensated. For example, Magnusson et al. reported that both static and dynamic compensation were not completed in animals even 100 days after chemical labyrinthectomy (15). In contrast, static compensation is nearly completed within 1 month after surgical labyrinthectomy (16). Surgical procedures for the treatment of Ménière s disease can be classified into three categories. The first category involves selective vestibular neurectomy or labyrinthectomy. These procedures have been reported to achieve control of vertigo in 90% of cases of intractable Ménière s disease (8). The major problem for this approach is the relatively high chance of different complications including hearing loss, facial nerve paralysis, cerebrospinal fluid leak, and headache. The second category includes operations on endolymphatic sac that are aimed to release ELH by decompressing the sac or draining endolymph from it. Such procedures typically control vertigo in only 50 to 75% of cases (17 19). In addition, recurrence of vertigo is common on long-term follow-up (17). The third category is semicircular canal occlusion, which aims to ablate movement of endolymph in the canals and therefore to eliminate vertigo attacks aroused by rotation. In 1960, Kristensen reported that hearing was not affected after extramembranous labyrinth manipulation (20). On the basis of this finding, he proposed this operation as a potential method of treating intractable vertigo attacks. Selective occlusion of a single semicircular canal has been widely recognized as an effective treatment of intractable benign paroxysmal positional vertigo (21 29). In recent years, several studies have reported that triple semicircular canal occlusion (TSCO) effectively eliminated the function of the semicircular canals and spared hearing and otolithic function (16,30,31). In addition, postoperative disequilibrium was reported to be much less in cats treated with TSCO than in those receiving labyrinthectomy (30). This advantage was attributed to the preservation of otolithic function in cats treated with TSCO. Nevertheless, the development of TSCO opens a new opportunity for management of intractable Ménière s disease. However, TSCO has not been fully evaluated in ears with ELH, which is considered to be a common abnormality of Ménière s disease. In this study, we explored the effect of TSCO on cochlear and vestibular function in animals with hydrops. Our purpose was to identify whether such an operation could actually eliminate semicircular canal response to rotation and preserve hearing at the same time in ears in which ELH had been established. PATIENTS AND METHODS Twenty healthy, adult albino guinea pigs (male and female, weighing between 300 and 400 g) were used in this experiment. Normal hearing and vestibular functions were warranted by the tests of auditory brainstem response (ABR) and electronystagmography (ENG). ELH was established unilaterally by endolymphatic sac (ELS) obliteration. ABR and ENG evaluations were performed before and at days 15, 25, 30, 45, 60, 75, 90, 105, and 120 after ELH surgery. Then, TSCO was performed in 12 animals in the ear in which ELH had been established. Another eight animals were killed and the cochleae were harvested for morphologic observation to confirm the establishment of ELH. ABR was repeated at days 1, 3, 5, 10, 15, and 30 after TSCO surgery, whereas ENG was repeated at days 5, 10, 15, and 30 after TSCO. Caloric testing was carried out before and 30 days after TSCO. Finally, the occlusion and ELH were confirmed by morphologic observation in all surgical ears. Morphologic observations were also performed in the control ear. Establishment of ELH Guinea pigs were anesthetized with sodium pentobarbital (35 mg/kg, intraperitoneally) and lidocaine was infiltrated into retro-occipital skin as a local anesthetic. ELS obliteration was performed as described (32,33). Briefly, a paramedial incision was made on the retro-occipital skin, followed by the separation of the soft tissues over the right occiput bone. Part of the occiput bone was removed to expose sigmoid sinus, which was slightly pushed medially to expose the ELS crest

3 80 S. YIN ET AL. and vestibular aqueduct niche. The endolymphatic sac and duct were occluded with bone wax under direct visualization. TSCO Surgery One hundred twenty days after endolymphatic sac occlusion, TSCO was performed with the method previously reported (30,31,34). Under the same anesthetic procedure as that used for ELH surgery, an incision was made at the right postauricular skin and the bulla was opened between the facial nerve canal and the posterior ridge of the occiput bone to expose the posterior semicircular canal. A fenestration was made at a place on the canal as far as possible away from the ampulla. Then, a small piece of soft tissue was filled into the fenestration to tightly block the canal over a short segment. Thereafter, the lateral attic wall was widely opened to expose the superior and horizontal semicircular canals, which were blocked with the same method. Evaluation of Auditory and Vestibular Functions For ABR recording, guinea pigs were anesthetized with sodium pentobarbital (35 mg/kg, intraperitoneally). The response was picked up by three subcutaneous needle electrodes. The active electrode was placed at vertex, and the reference electrode under one ear and grounding electrode at the nose. Clicks were presented at a rate of 11.1 per second through a TDH 39 earphone (Telephonics, Chester, U.K.), which was placed firmly against the external canal of the ear tested, whereas the contralateral ear was plugged with wax to prevent crossover. The evoked potential was filtered with a bandpass between 100 and 3,000 Hz, and averaged for 128 times (Nicolet Spirit evoked potential system, [Madison, Wisconsin, U.S.A.]). The ABR threshold was approached in a downward sequence of 5-dB steps and was determined as the lowest sound pressure level to which a visible, repeatable Wave III could be identified. To evaluate the functional status of the vestibular system, nystagmus was examined via rotation and caloric tests using ENG. Two disk electrodes were attached to the outer and inner canthus of one eye of the animal, and a grounding electrode was attached to the vertex of the animal head. To test rotatory nystagmus, the guinea pig was placed in a custom-made box fixed on a Fod-03 rotary chair (Dallchi Medical Corporation, Japan) with the head pitched downward 48 degrees so the horizontal semicircular canal was parallel to the rotation plane. Sinusoidal rotations of different magnitudes ( degrees in 30-degree step) were performed at the same speed (0.1 Hz, 10 seconds per circle). The nystagmus was recorded on paper using RM-6100 mini polygraph (Nikon Kohden, Japan). The number of instances of nystagmus in half a cycle (5 seconds) of rotation was counted as the indicator of response strength. To measure caloric nystagmus, the animal heads were pitched upward 42 degrees so that the horizontal canal was vertical to the horizontal plane. One milliliter of ice water was applied to the posterior and superior walls of the external ear canal for 40 seconds and the number of instances of nystagmus was counted during this period. Morphologic Observation To confirm ELH and TSCO, the guinea pigs were decapitated under deep anesthesia and the temporal bones were fixed in 10% formalin, decalcified in 10% ethylenediamine tetra-acetic acid, and embedded in paraffin. The samples were sectioned and stained with hematoxylin and eosin and observed under light microscopy. The protocol for the care and use of animals involved in this study was approved by the animal ethical committee of Shanghai Second Medical University. RESULTS Changes in Vestibular Function After ELH surgery, no apparent vestibular dysfunction was identified from the daily behavior of the guinea pigs, although subtle changes were evident in laboratory evaluations. Before the operation, rotatory nystagmus in all 20 guinea pigs was symmetric. However, directional differences were found after surgery. Figure 1 illustrated the difference as a function of postoperation days. The positive points (above zero line) indicated the prevalence of ipsilateral rotation (toward the side that had been operated on), whereas the negative points indicated contralateral prevalence. Ipsilateral prevalence was evident at postsurgery Day 25 and was significant for rotations of 60 and 120 degrees (paired t test, p, 0.05). This prevalence was soon replaced by the contralateral one that was generally not significant except at days 45 and 75 for rotations of 90 and 120 degrees. Successful TSCO was first indicated by behavior changes and spontaneous nystagmus. Immediately after TSCO, all animals underwent characteristic head tilt and postural disturbances, including deviation of the head and unstable and rotated walking toward the surgical side, which resulted in difficulty in eating and drinking. These symptoms persisted for only 6 to 8 hours and then quickly ceased. Immediately after TSCO, all animals demonstrated spontaneous nystagmus with slow phase toward the surgical side. The spontaneous nystagmus existed FIG. 1. The difference in nystagmus induced by ipsilateral versus contralateral rotation after ELH. The nystagmus was counted in a half cycle in response to the rotation of 60, 90, 120, 150, and 180 degrees. Positive values indicate that more nystagmus is elicited by contralateral rotation.

4 TRIPLE SEMICIRCULAR CANAL OCCLUSION IN ELH 81 after postsurgery Day 1 in some animals but disappeared in all animals within 3 days. The imbalanced semicircular canal function after TSCO and dynamic process of compensation were also indicated by the changes in directional difference of rotation-induced nystagmus (Fig. 2). At Day 1 after TSCO, the difference, which appeared as the contralateral prevalence (or ipsilateral weakness), reached maximal values because the nystagmus could not be induced by ipsilateral rotation. After Day 1, this imbalance was gradually reduced. It had totally disappeared at Day 30 after surgery. Finally, successful TSCO was further supported by the negative caloric nystagmus in operated ears. It was found that ELH surgery induced a slight but significant reduction in response to caloric stimulation. As evaluated 120 days after ELH, caloric nystagmus was per second (all data presented in this report are mean 6 standard deviation) for contralateral irrigation of ice water and / per second for ipsilateral irrigation, with a significant binaural difference (paired t test, p, 0.05). Thirty days after TSCO, the averaged caloric response from contralateral ears was per second, not significantly changed from the control value. However, caloric nystagmus could not be elicited in TSCO ears. It is interesting to note that, at this time, rotatory nystagmus had totally recovered and balanced. Changes in Auditory Functions Hearing function was evaluated with click-evoked ABR. Because the ABR thresholds of contralateral ears FIG. 2. The difference in nystagmus elicited by rotation to different directions (ipsilateral versus contralateral rotation) after TSCO. The nystagmus was tested in response to the rotation of different degrees from 60 to 180 degrees and was counted in half a cycle of rotation (5 seconds). The evaluation was repeated at different times after TSCO. A tentative drop in nystagmus in response to ipsilateral rotation was maximal 1 day after TSCO. It had totally recovered 30 days after surgery. FIG. 3. The dynamic change in binaural difference of ABR thresholds after ELH. Day 15 indicated control recording before ELH surgery. The bars represent standard deviation. did not change after two operations and there was no significant binaural difference in ABR thresholds before ELH surgery, the ABR threshold of contralateral ears served as an excellent internal control. Therefore, the binaural threshold difference obtained after unilateral surgery was a valid indication of the threshold shift induced by the operation. Figure 3 shows the averaged binaural difference and standard deviation in ABR threshold as a function of postsurgery time. Starting at 15 days after ELH surgery, the ABR threshold of the operated ears was gradually elevated and became significantly higher than the control ears (paired t test, p, 0.05). The threshold shift reached plateau at 90 days after surgery; the maximal shift was db (n = 20). Figure 4 shows the further ABR threshold shift induced after TSCO. The ABR threshold was slightly elevated after TSCO starting at Day 5 after surgery and quickly stabilized. The maximal shift was only db, which was not significant (p = 0.069, paired t test) when compared with the threshold measured 120 days after ELH surgery from 12 guinea pigs that had undergone TSCO. Morphologic Evaluation of ELH and TSCO ELH appeared in the cochlea as a great expansion of scala media, resulting in a remarkable invasion of the Reissner s membrane into the scala vestibuli and a slight bulge of the basilar membrane toward the scala tympani. Such an appearance was especially severe in the third and fourth turns of the cochlea and was evident in every surgical ear. Figure 5 shows a typical change of a surgical ear in a cross-sectional view of the cochlear canal, which demonstrates the invasion of the Reissner s membrane toward the scala vestibuli and a slight dislocation of the basilar membrane toward the scala tympani as a result of hydrops in the scala media. No significant damage could be found in the organ of Corti under light microscopy.

5 82 S. YIN ET AL. FIG. 4. The dynamic change in binaural threshold difference of ABR after TSCO. Slight but significant elevation was evident 5 days after TSCO. The difference became stable after that. No such dislocations of Reissner s membrane or the basilar membrane were found in any of the control ears. Successful TSCO was confirmed in every surgical ear. Figure 6 shows a typical occlusion of a semicircular canal after TSCO. The canal was totally blocked for a short segment, and the blocking point was far away from the ampulla. DISCUSSION The main purpose of this study was to evaluate whether TSCO can be safely performed in hydropic ears. In this study, the nonhydropic ear was used as an internal control to verify the hearing threshold shifting, and therefore TSCO was not performed in nonhydropic ears. However, the safety of this surgery on nonhydropic ears had been evaluated in our previous report (16). FIG. 5. ELH morphology (hematoxylin and eosin staining). The expansion of the scala media caused by ELH is evident as the displacement of Reissner s membrane into scala vestibuli. The organ of Corti is generally intact. FIG. 6. Occlusion of semicircular canals (hematoxylin and eosin staining). Arrows point to the segment of occlusion in a posterior semicircular canal. The pathogenesis of Ménière s disease remains to be explored (1), although it is generally agreed that the vertiginous symptoms result from the abnormality in the vestibular apparatus of the inner ear. ELH has been considered to be the pathologic basis of this mysterious disorder (1,2,35,36). Endolymph is either overproduced or not promptly absorbed, resulting in a pressure increase and expansion of the endolymphatic space, which is most consistently found in the cochlea and saccule. Although experimental animals with ELH do not typically display characteristic signs of Ménière s disease, namely, rotatory vertigo, functional disturbances in the vestibular system have been well documented in these models (37,38). On the basis of the differential response of semicircular canals to angular acceleration and the otolithic organs to linear acceleration, it is hypothesized that rotatory vertigo is associated with dysfunctions in the semicircular canals, whereas gait disturbances and drop attacks are associated with dysfunctions in otolithic organs. In most cases, patients with Ménière s disease are disabled by rotatory vertigo (1). Under the condition of hydrops, the increased endolymphatic fluid is assumed to increase the pressure in the vestibular apparatus and causes more stimulation to the ampullary crista (39). As a consequence, the vestibular apparatus in the involved ear increases excitability and therefore generates imbalanced inputs to the central vestibular system. Such a hypothesis is supported by the vertigo attacks in patients with Ménière s disease (37,38) and the asymmetric rotatory nystagmus in animals with unilateral ELH (40 42). Consistent with the previous findings, ipsilateral prevalence was evident shortly after ELH in the present experiment. Therefore, it is hypothesized that the vertigo attack resulted from ELH can be released if the fluid movement induced by rotation can be eliminated by occluding the semicircular canals in the involved ear. The TSCO surgery itself will reduce the vestibular input from the surgical side and therefore will produce

6 TRIPLE SEMICIRCULAR CANAL OCCLUSION IN ELH 83 imbalanced inputs. However, such disequilibrium can be quickly compensated for by central plasticity as suggested by the outcome of TSCO on normal animals (16,30). One of the complications of endolymphatic sac obliteration is fenestration of the posterior and/or superior canals, which will cause spontaneous nystagmus. In the current study, however, we did not find spontaneous nystagmus after ELH surgery. This result suggested that this complication did not occur in ELH ears. Various methods of semicircular canal ablation have been evaluated in the treatment of rotatory vertigo, including ultrasound therapy (43), labyrinthine cryosurgery (44), and ablation using carbon dioxide laser (45). Clinical use of those methods has been limited by surgical difficulty, trauma, and/or complications including hearing loss and lesions of facial nerve. Adamczyk and Antonelli recently reported a method of TSCO using potassium-titanyl-phosphate laser in guinea pigs with and without ELH (46). The auditory function was evaluated with electrocochleography in this experiment, which showed no significant elevation of auditory threshold after ELH. The lack of hearing loss after ELH was different from the outcome of most ELH experiments by others. Moreover, the vestibular function was not evaluated in this experiment. The only indication of ELH establishment was the morphologic observation that, unfortunately, was performed in only three ears. In the present experiment, TSCO was performed in ears in which ELH had been established and was confirmed later by morphologic observation. A tentatively enhanced excitability in the vestibular system was evident at Day 25 after ELS obliteration as the nystagmus prevalence in response to ipsilateral rotation. The enhancement was soon replaced by a decline in ipsilateral excitability resulting in an ipsilateral weakness. An auditory threshold shift of the surgical ears reached plateau 90 days after ELH surgery. The chronologic change in vestibular and auditory functions was generally consistent with what had been reported previously (40,41,47 49). The success in establishment of ELH was further found in morphologic observation in all surgical ears. All of the above suggested that ELH was successfully established in this experiment, and 4 months of waiting was appropriate for stabilizing ELH. The dynamic change of nystagmus after TSCO in this experiment was similar to our previous report in which TSCO was performed in normal guinea pigs (16). The optimal stimulus to semicircular canal is angular acceleration, which results in ampullary crista displacement caused by endolymphatic flow. Shortly after TSCO, ipsilateral rotation could not induce nystagmus (Day 1 after TSCO), resulting in a large directional difference (Fig. 2). This early imbalance or directional difference in nystagmus suggested that the semicircular canals were disabled in response to rotation. However, the nystagmus gradually recovered, which was seen as the reduced directional difference at Days 3, 5, and 10 after operation, and the directional difference totally disappeared at Day 30. Because the nystagmus induced by contralateral rotation did not change after TSCO, the reduction in nystagmus difference was attributed to the recovery of nystagmus induced by ipsilateral rotation. At this time, however, the operated ears failed to respond to caloric stimulation, suggesting that fluid movement induced by the caloric stimulation was still blocked in TSCO ears. These results favor the idea that the recovery of nystagmus to ipsilateral rotation results from a process of central compensation rather than a regained peripheral input from the surgical ear. Recovery of balanced rotatory nystagmus was quick and complete in our guinea pigs after TSCO. Similar to what has been reported before, TSCO showed a remarkable advantage over labyrinthectomy in terms of the speed and completeness of compensation (15,30,50). At this moment, however, we do not have clear reasons for the slow and incomplete recovery after labyrinthectomy over TSCO. One of possibilities is that unilateral labyrinthectomy totally destroys the vestibular peripheral system, including the sensory organ, which may be more difficult to compensate for than the dysfunctioning by blocking the fluid movement in the canals. The mechanisms for regaining balanced nystagmus after unilateral labyrinthectomy have been widely explored in terms of central vestibular plasticity, particularly in the modulation of neural activities in vestibular nuclei (see review by Kitahara et al. (51)). In guinea pigs after unilateral labyrinthectomy, the neurons in ipsilateral vestibular nuclei cease firing; however, the neural activity recovers to normal levels in 1 week. Two synergistic mechanisms are considered to be involved in this plastic process: a significant increase in the intrinsic excitability of neurons and a remarkable decrease in the functional efficacy of GABA(A) and GABA(B) receptors in the inhibitory circuits (51 55). Until now, there has been no report regarding the central mechanisms of vestibular compensation after semicircular canal occlusion. Further studies are obviously needed to address this issue, although it is very possible that similar mechanisms should apply. A major concern in selecting a surgical treatment of Ménière s disease is the preservation of cochlear function. The results of this experiment and our previous report (16) demonstrate that surgical occlusion of the semicircular canal is safe and cochlear function can be well preserved if the surgery is performed carefully. In one of our previous study, we reported that vertigo was well controlled in patients with positional vertigo after posterior canal occlusion and hearing was not changed after operation as indicated by ABR and distortion product otoacoustic emission testing (56). Also, there were no significant or permanent changes in auditory functions evaluated with ABR and distortion product otoacoustic emission testing in guinea pigs after TSCO (16). In this study, a similar result was presented, suggesting that hearing was well preserved after TSCO in the ears that had become vulnerable after establishment of ELH. At present, it is not entirely clear what the critical factors are for preservation of the cochlear function

7 84 S. YIN ET AL. during TSCO. However, some clues can be proposed on the basis of previous reports and our experiences. First, there exists membrane limitans in the perilymphatic space, which separates the cochlea and saccule from the utricle and canal system and limits the fluid exchange between the vestibular apparatus and the cochlea (57). Second, the endolymphatic valve in the utricle exerts a protection role, which may prevent excessive loss of perilymph during surgery (58). It is recommended, however, that surgery be performed as quickly as possible because excessive loss of perilymph may cause permanent damage to the cochlea. Third, transient endolymphatic fistula during TSCO surgery may induce endolymph into the perilymphatic space, resulting in alleviation of endolymphatic hydrops. However, excessive loss of endolymph in the cochlea is prevented by the existence of the endolymphatic valve in the utricle and because the cochlear duct is far away from the surgical points. Fourth, to avoid the loss of endolymph and damage to the otolith organs, the site of fenestration should be as far away from the ampullae as possible; care should be taken to avoid damage to the membranous labyrinth. In summary, the results of this experiment provide evidence for the safety of TSCO by mechanical surgery in ELH ears and the effectiveness of eliminating the response of the vestibular peripheral to rotatory stimulation. Therefore, this surgery may be an alternative approach with which to relieve patients with Ménière s disease from rotatory vertigo attack. REFERENCES 1. Minor LB, Schessel DA, Carey JP. Meniere s disease. Curr Opin Neurol 2004;17: Hallpike CS, Cairns H. Observation on the pathology of Meniere s syndrome. J Laryngol Otol 1938;53: Kimura RS. Experimental blockage of the endolymphatic duct and sac and its effect on the inner ear of the guinea pig: a study on endolymphatic hydrops. Ann Otol Rhinol Laryngol 1967;76: Andrews JC, Li J, Koyama S, et al. Vestibular and optokinetic function in the normal guinea pig. Ann Otol Rhinol Laryngol 1997; 106: Dunnebier EA, Segenhout JM, Wit HP, et al. Two-phase endolymphatic hydrops: a new dynamic guinea pig model. Acta Otolaryngol 1997;117: Horner KC, Aurousseau C, Erre JP, et al. Long-term treatment with chlorthalidone reduces experimental hydrops but does not prevent the hearing loss. Acta Otolaryngol 1989;108: Horner KC, Erre JP, Cazals Y. Asymmetry of evoked rotatory nystagmus in the guinea pig after experimental induction of endolymphatic hydrops. Acta Otolaryngol Suppl 1989;468: Glasscock ME III, Gulya AJ, Pensak ML, et al. Medical and surgical management of Meniere s disease. Am J Otol 1984;5: Brown JS. A ten year statistical follow-up of 245 consecutive cases of endolymphatic shunt and decompression with 328 consecutive cases of labyrinthectomy. Laryngoscope 1983;93: Kimura RS, Lee KS, Nye CL, et al. Effects of systemic and lateral semicircular canal administration of aminoglycosides on normal and hydropic inner ears. Acta Otolaryngol 1991;111: Wu IC, Minor LB. Long-term hearing outcome in patients receiving intratympanic gentamicin for Meniere s disease. Laryngoscope 2003;113: Harner SG, Driscoll CL, Facer GW, et al. Long-term follow-up of transtympanic gentamicin for Meniere s syndrome. Otol Neurotol 2001;22: Carey JP, Hirvonen T, Peng GC, et al. Changes in the angular vestibulo-ocular reflex after a single dose of intratympanic gentamicin for Meniere s disease. Ann N Y Acad Sci 2002;956: De Waele C, Meguenni R, Freyss G, et al. Intratympanic gentamicin injections for Meniere disease: vestibular hair cell impairment and regeneration. Neurology 2002;59: Magnusson AK, Lindstrom S, Tham R. GABA(B) receptors contribute to vestibular compensation after unilateral labyrinthectomy in pigmented rats. Exp Brain Res 2000;134: Zhang Y, Yin S, Gu J, et al. Establishment of the animal model of triple semicircular canal occlusion [in Chinese]. Zhonghua Er Bi Yan Hou Ke Za Zhi 2000;35: Monsell EM, Wiet RJ. Endolymphatic sac surgery: methods of study and results. Am J Otol 1988;9: Brackmann DE, Nissen RL. Meniere s disease: results of treatment with the endolymphatic subarachnoid shunt compared with the endolymphatic mastoid shunt. Am J Otol 1987;8: Moffat DA. Endolymphatic sac surgery: analysis of 100 operations. Clin Otolaryngol 1994;19: Kristensen HK. Acoustic-vestibular and histologic examinations in guinea-pigs after interruption of membranous labyrinth in semicircular canals or cochlea. Acta Otolaryngol 1960;51: Brantberg K, Bergenius J. Treatment of anterior benign paroxysmal positional vertigo by canal plugging: a case report. Acta Otolaryngol 2002;122: Agrawal SK, Parnes LS. Human experience with canal plugging. Ann N Y Acad Sci 2001;942: Kartush JM, Sargent EW. Posterior semicircular canal occlusion for benign paroxysmal positional vertigo CO2 laser-assisted technique: preliminary results. Laryngoscope 1995;105: Oki S, Nomura Y, Sugio Y, et al. Occlusion of the semicircular canal using argon laser. J Clin Laser Med Surg 1996;14: Parnes LS, McClure JA. Posterior semicircular canal occlusion for intractable benign paroxysmal positional vertigo. Ann Otol Rhinol Laryngol 1990;99: Parnes LS, McClure JA. Posterior semicircular canal occlusion in the normal hearing ear. Otolaryngol Head Neck Surg 1991;104: Parnes LS. Update on posterior canal occlusion for benign paroxysmal positional vertigo. Otolaryngol Clin North Am 1996; 29: Parnes LS, Agrawal SK, Atlas J. Diagnosis and management of benign paroxysmal positional vertigo (BPPV). CMAJ 2003;169: Crossland G, De R, Axon P. Far advanced otosclerosis and intractable benign paroxysmal positional vertigo treated with combined cochlear implantation and posterior semicircular canal occlusion. J Laryngol Otol 2004;118: Gianoli GJ, Duff B, Kartush JM, et al. Triple semicircular canal occlusion versus labyrinthectomy in the cat. Am J Otol 1997;18: Antonelli PJ, Bouchard KR, Kartush JM, et al. Triple semicircular canal occlusion in the guinea pig. Otolaryngol Head Neck Surg 1997;117: Lee JR, Wright CG, Meyerhoff WL. Modified occipital approach to the endolymphatic sac and cochlear aqueduct of the guinea pig. Am J Otol 1993;14: Andrews JC, Bohmer A. The surgical approach to the endolymphatic sac and the cochlear aqueduct in the guinea pig. Am J Otolaryngol 1989;10: Zhang Y, Yin S, Gu J. Establishment of the animal model of triple semicircular canal occlusion [in Chinese]. Zhonghua Er Bi Yan Hou Ke Za Zhi 2000;35: Anatoli-Candela F. The histopathology of Meniere s disease. Acta Otolaryngol Suppl 1976;340: Schuknecht HF, Igarashi M. Pathophysiology of Meniere s disease. In Pfaltz CR, ed. Controversial Aspects of Meniere s Disease. New York: George Thieme, 1986:46 54.

8 TRIPLE SEMICIRCULAR CANAL OCCLUSION IN ELH Parnes LS, Riddell D. Irritative spontaneous nystagmus following intratympanic gentamicin for Meniere s disease. Laryngoscope 1993;103: Tsunoda R, Fukaya T, Komatsuzaki A. The furosemide test and vestibular status in Meniere s disease. Acta Otolaryngol 1998;118: Andrews JC. Intralabyrinthine fluid dynamics: Meniere disease. Curr Opin Otolaryngol Head Neck Surg 2004;12: Andrews JC, Honrubia V. Vestibular function in experimental endolymphatic hydrops. Laryngoscope 1988;98: Aran JM, Rarey KE, Hawkins JE Jr. Functional and morphological changes in experimental endolymphatic hydrops. Acta Otolaryngol 1984;97: Bohmer A, Dillier N. Experimental endolymphatic hydrops: are cochlear and vestibular symptoms caused by increased endolymphatic pressure? Ann Otol Rhinol Laryngol 1990;99: Peron DL, Kitamura K, Carniol PJ, et al. Clinical and experimental results with focused ultrasound. Laryngoscope 1983;93: Wolfson RJ. Labyrinthine cryosurgery for Meniere s disease present status. Otolaryngol Head Neck Surg 1984;92: Antonelli PJ, Lundy LB, Kartush JM, et al. Mechanical versus CO2 laser occlusion of the posterior semicircular canal in humans. Am J Otol 1996;17: Adamczyk M, Antonelli PJ. Selective vestibular ablation by KTP laser in endolymphatic hydrops. Laryngoscope 2001;111: Bohmer A. Hydrostatic pressure in the inner ear fluid compartments and its effects on inner ear function. Acta Otolaryngol Suppl 1993;507: He D, Zhou D, He L. Changes in vestibular function and furosemide test after experimental endolymphatic hydrops in guinea pigs [in Chinese]. Zhonghua Er Bi Yan Hou Ke Za Zhi 1994;29: Horner KC. Functional changes associated with experimentally induced endolymphatic hydrops. Hear Res 1993;68: Yin SK, Li MN, Zhang Y. Vestibular compensation after triple semicircular canal occlusion and labyrinthectomy. Chinese Arch Otolaryngol Head Neck Surg 2001;8: Kitahara T, Takeda N, Kiyama H, et al. Molecular mechanisms of vestibular compensation in the central vestibular system. Acta Otolaryngol Suppl 1998;539: Johnston AR, Him A, Dutia MB. Differential regulation of GABA(A) and GABA(B) receptors during vestibular compensation. Neuroreport 2001;12: Him A, Dutia MB. Intrinsic excitability changes in vestibular nucleus neurons after unilateral deafferentation. Brain Res 2001; 908: Kitahara T, Takeda N, Uno A, et al. Unilateral labyrinthectomy downregulates glutamate receptor delta-2 expression in the rat vestibulocerebellum. Brain Res Mol Brain Res 1998;61: Beraneck M, Idoux E, Uno A, et al. Unilateral labyrinthectomy modifies the membrane properties of contralesional vestibular neurons. J Neurophysiol 2004;92: Chen B, Shen P, Yin S, et al. Transantral orbit decompression to malignant ophthalmopathy [in Chinese]. Lin Chuang Er Bi Yan Hou Ke Za Zhi 1999;13: Hara M, Kimura RS. Morphology of the membrana limitans. Ann Otol Rhinol Laryngol 1993;102: Schuknecht HF, Belal AA. The utriculo-endolymphatic valve: its functional significance. J Laryngol Otol 1975;89: