Pergamon

0892-0362(94)00067-O

Neurotoxicology and Teratology, Vol. 17, No. 3, pp. 281-287, 1995 Copyright o 1995 Elsevier Science Ltd

Printed in the USA. All rights reserved 0892-0362/95 $9.50 + .OO

Trimethyltin Disrupts Loudness Recruitment and Auditory Threshold Sensitivity

in Guinea Pigs

YE LIU AND LAURENCE D. FECHTER’

Toxicology Program, College of Pharmacy, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73190

Received 4 February 1994; Accepted 13 October 1994

LIU, Y. AND L. D. FECHTER. Trimethyltin disrupts loudness recruitment and auditory threshold sensitivity in guinea pigs. NEUROTOXICOL TERATOL 17(3) 281-287, 1995. -Trimethyltin (TMT) impairs auditory thresholds within minutes of systemic administration. However, there are no data which relate to the output of the auditory nerve at sound levels above threshold. In this experiment, we evaluated the functional effects of TMT on the auditory threshold by identifying the sound level which just produced a detectable compound action potential (CAP). We also assessed outer hair cell function by measuring the cochlear microphonic (CM), a nonpropagated ac potential which is phase-locked to the stimulus. Finally, we measured the growth of the N, amplitude as a function of stimulus intensity at levels above threshold and of the summating potential (SP), a dc potential which has multiple generators. To isolate cochlear from systemic effects of TMT, the agent was applied directly to the round window, a structure separating the middle and inner ear, of anaesthetized guinea pigs. We show that TMT applied to the round window membrane can disrupt the function of the cochlea. Measurements of auditory function at supra-threshold levels showed clearly that TMT reduced the amplitude of N, while having no measurable effect on the SP. These findings indicate that TMT bIocks the recruitment of neuronal elements by Ioud sound. This pattern of impairment differs from that observed with aminoglycoside antibiotics, hypothermia, and presbycusis in which loudness recruitment has been reported.

Trimethyltin Ototoxicity Compound action potential Loudness recruitment Round window membrane

TRIMETHYLTIN (TMT) has been reported to produce audi- tory dysfunction in rat and guinea pig (5,6,9,10,11,34). Elec- trophysiological studies show that TMT impairs the response sensitivity of the cochlea to sound within 30 mm of injection such that the sound level required to elicit a compound action potential (CAP) had to be significantly increased (5). The ini- tial locus of the impairment appears to be at the synapse be- tween the sensory receptor cell, the inner hair cell, and the initial sensory neuron, the type 1 spiral ganglion cell. Subse- quently, the outer hair cell appears to be affected as the ampli- tude of the cochlear microphonic (CM) is reduced at 6 h fol- lowing i.p. TMT. However, there are no data that relate to the output of the auditory nerve at sound levels above thresh- old. In many instances, ototoxic chemicals as well as other sources of hearing impairments yield a phenomenon known as

loudness recruitment in which the growth of some biological measure of loudness is normal at supra threshold levels even though detection thresholds are impaired. This phenomenon is dependent on the availability of recruitable units that can be stimulated at high intensity levels even though responses to low level stimuli are impaired. Davis et al. (8) suggested that two separate populations of cells respond to auditory signals. One group stimulated at low sound levels, but having poor frequency specificity, play an important role in stimulus detec- tion, while a second much larger population with greater fre- quency specificity becomes increasingly important at higher stimulus intensities. Selective impairment of the former cells may disrupt detection of a stimulus but does not alter the maximal response of the cochlea to high intensity stimuli. Age-related hearing dysfunction (presbycusis) and aminogly-

’ Requests for reprints should be addressed to Laurence D. Fechter, Toxicology Program, College of Pharmacy, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73190.

281

282 LIU AND FECHTER

coside antibiotic ototoxicity represent two common instances where such a phenomenon, referred to as loudness recruit- ment, occurs (1,7,14,15).

This experiment is designed to determine whether at high sound levels the TMT impaired auditory nerve can produce a normal output. The positive round window summating poten- tial (SP), a dc shift generated by several cell types, was also measured at a range of stimulus intensities above threshold. This was done both as a control to determine the specificity of the outcome with respect to the propagated N, potential and also because a large positive SP occurring just prior to the onset of N, might mask the latter negative potential.

In order to eliminate potential systemic effects of TMT from direct cochlear effects the compound was applied on the round window, a structure separating the middle ear from the inner ear, and auditory function was monitored. Diffusion of various substances into the cochlea through the round window membrane has been documented including low molecular weight substances such as streptomycin (29), gentamicin (33), salicylate, and indomethacin (18). The sodium radionuclide (22Na) (3,30), and high molecular weight substances such as horseradish peroxidase (HRP) (28), albumin (12), exotoxin, and endotoxin (19,21,22) have also been shown to diffuse across this membrane.

METHOD

Subjects

: 12D h cycle at 22’%.

Procedure

Animals were anesthetized with urethane (1.5 g/kg) admin- istered IP for general anesthesia. The disappearance of the blink reflex and the pedal withdrawal reflex to digit compres- sion were the indicators for adequacy of anesthetic level. As necessary, this dosage was supplemented with urethane IP (0.15 g/kg). Lidocaine (2Yo) was injected at the incision for local analgesia. The surgery was conducted in a double-walled audiometric booth (IAC122A). The subjects were positioned dorsally on a dc heating pad which maintained the body tem- perature at approximately 39OC as measured by a rectal probe. A low voltage lamp was directed at the cochlea to prevent cooling. The head was held rigidly by a head holder, the subjects were tracheotomized to ensure unobstructed breathing, and the right pinna removed so that the tympanic membrane could be visualized. The bulla was exposed ven- trally by removing the skin and muscles overlying the bulla and opened very gently using a surgical blade to avoid damag- ing noise. A Teflon coated 40 G silver wire recording electrode was positioned on the surface of the round window membrane and was referenced to a Ag/AgCl electrode placed in neck musculature. A custom made Plexiglas speculum was carefully positioned into the external auditory meatus such that the tympanic membrane was visible through the speculum. Sound stimuli were delivered via a I%” condenser microphone which was placed into the speculum and was driven by a high voltage amplifier.

Eiectrophysioiogical Recording

CAP and CM measurement from the round window. Mea- surement of CAP thresholds and the 1 FV RMS CM isopoten-

tial curve were made from the round window for 11 pure tones at approximate % octave intervals between 2 and 40 kHz by methods fully described by Clerici et al. (5). The electrophysi- ological responses to pure tone pips of 10 ms duration with 1 ms onset-offset ramps were bandpass filtered between 300 and 1000 Hz and preamplified x 10 by a Grass dc preamplifier (model P16) and x 100 by a Grass dc preamplifier (model P15) for measurement of the CAP. Stimuli were presented at a rate of 9/s. CAP thresholds were determined by a just de- tectable N, wave observed on a digital oscilloscope triggered by the onset of the acoustic stimuli. The CM was elicited by constant pure tone stimuli. Tone intensity was increased stepwise under computer control until a 1 PV RMS signal was achieved. The CM was high pass filtered above 300 Hz, preamplified x 1000, as above, and filtered using a lock-in amplifier. The CAP threshold and CM isopotential curves were assessed prior to TMT administration and 30 min follow- ing TMT administration. Subsequent to the last threshold de- termination, the sound intensity was calibrated for each sub- ject using a %” Bruel and Kjaer 4134 microphone coupled to a probe tube inserted into the speculum to within approximately 1 mm of the tympanic membrane. The voltage output of the microphone was fed to a measuring amplifier circuit contained within our stimulus control equipment.

Input-output functions for N, and SP. The amplitude of the N, and of positive round window SP were determined at 8 kHz, 16 kHz, and 32 kHz using tone intensities varying in 10 dB steps between O-80 dB SPL (re : 20pPa). Testing auditory function at these three octave steps was designed to sample a broad portion of the guinea pigs’s auditory range. It is ex- pected that amplitudes of both the N, and the positive round window SP will increase as stimulus intensity is increased above threshold. Stimuli were comparable to those used for assessing CAP amplitude with a duration of 10 msec, a 1 msec rise/fall time and stimulus rate set at 9.7/s. However, responses were averaged over 256 stimulus presentations by the computer. The N, amplitude was measured from the base- line to the nadir of the negative N, and the positive round window SP amplitude was from the onset peak of SP to the baseline.

Trimethyltin administration. After the initial measure- ments of CAP and CM isopotential curves and of the input- output series, a 1.5 x 1.5 x 1.5 mm gelfoam soaked with 1.5 mM TMT-Cl in saline (ICN Biomedicals) for the treated group, or saline for the saline control group was placed gently in the round window niche. Extreme care was used to avoid touching the structures of the middle ear to avoid mechanical injury which would result in a threshold shift. The gelfoam remained in place on the round window for 15 min to let the TMT solution permeate through the round window mem- brane. The gelfoam was then removed carefully and the bulla dried by using absorbent points. The CAP sensitivity and CM isopotential curve, N, and positive round window SP intensity series were remeasured to evaluate the effect of the TMT.

Statistics. The CAP detection threshold and 1 FV RMS CM amplitude isopotential values were analyzed using sepa- rate analysis of variance (ANOVA) tests with repeated mea- sures (SuperANOVA) to compare the different in CAP sensi- tivity or in the CM isopotential curve between groups. The two within variables were pre- versus post-treatment and fre- quency. The between subjects variable was Treatment group (TMT vs. Saline group). The N, and the positive round win- dow SP input-output curves were also evaluated by ANOVA with repeated measures. Treatment group (TMT vs. saline control) was analyzed as a between-subject factor and pre-

TMT DISRUPTS LOUDNESS RECRUITMENT 283

versus post-treatment, sound intensity, and frequency were analyzed as within subject factors (1 between, 3 within). Two- tailed analyses were used in all instances and a p value < 0.05 was used as the criterion for statistical significance.

CM 120

=i & %

100

P

2 v *O

A RESULTS

TMT placed on the round window disrupted the sensitivity of the CAP at the middle and high frequencies after 15 min of exposure (see Fig. 1A). The range of threshold impairment ____~._. Pre TMT

Post TMT

70

1

70 -

60 -

CAP

A

____o___ Pre TMT

- Post TMT

(n=S) Frequency (kHz)

10

Frequency (kHz)

B

- - a- - Pre Saline

- Post Saline

$ (r&=5)

(n=S)

0 i

1

1

10

Frequency (kHz)

I

100

FIG. 1. Compound action potential thresholds 15 min after TMT (A) and saline (B) administration through the round window mem- brane(Mean + SE).

B

- - Q- - Pre Saline

--t- Post Saline

1 10

Frequency (kHz)

100

FIG. 2. Effect of TMT (A) and saline (B) administration through round window membrane on the 1 pV RMS cochlear microphonic (Mean f SE).

was from 20 to 40 dB. This is consistent with previous reports using systemic administration of TMT given at a dose of 2 mg/kg IP (5). There was no change in the sensitivity of CAP in the saline control group (see Fig. 1B). Statistical analysis confirmed a significant difference between the groups as a function of test time (pre- vs. post-treatment) F(1, 8) = 181.20, p < 0.001. On the other hand, the CM data showed about a lo-25 dB increase in the sound intensity necessary to elicit a 1 PV signal after TMT administration especially at frequencies higher than 20 kHz (see Fig. 2A). A statistically

284 LIU AND FECHTER

significant interaction was obtained between treatment group (TMT and saline treated) and test (pre- vs. post-test) F(1, 8) = 76.04, p < 0.001, see Fig. 2A and B. The intensity func- tion for the N, potential amplitude is presented in Fig. 3 A-C for the TMT group and Fig. 3 D-F for the control subjects. The CAP thresholds prior to and following TMT or saline are indicated for reference based on the CAP threshold value reported in Fig. 1 and Fig. 2. The data show that the N, amplitude decreased dramatically at all sound intensity levels in those subjects which received TMT. It is apparent at 8 kHz and 16 kHz that there is some slow growth in the N, amplitude when the TMT subjects receive supra threshold sound stimuli (see Fig. 3 A and B), but it is very shallow in comparison to that observed for the same subjects prior to TMT or for the control subjects (see Fig. 3 D and E). At the highest test fre- quency, there is little evidence for growth in the N, as a func- tion of stimulus intensity after TMT treatment (see Fig. 3C). Thus, even at sound levels of 80 dB SPL, the N, amplitude is profoundly reduced compared to either the subjects’ pretreat- ment levels or the response observed in saline treated subjects (see Fig. 3 D, E, F). Statistical analysis showed a significant difference between the TMT and saline-treated subjects in the amplitude of N, between the pretest and post-test, F(1, 8) = 12.084, p < 0.01, and in the growth of N, over stimulus intensity, F(8, 64) = 2.684, p < 0.05. A significant interac- tion was also observed between treatment and stimulus inten- sity and frequency, F(16, 128) = 3.359, p < 0.0001. Finally, a significant four-way interaction between treatment groups x time (pre- vs. post-test), by frequency by intensity was found, F(16, 128) = 9.128, p < O.OOOl.The growth in ampli- tude of the positive round window SP as a function of stimu- lus intensity is shown in Fig. 4 A-F. There is some reduction in the amplitude of the positive round window SP at the high- est test frequency (32 kHz) among TMT-treated subjects (see Fig. 4C). However, the shift is small particularly when com- pared to the pronounced effect which this toxicant had on N, amplitude at sound levels above threshold. There was no evidence for a shift in the positive round window SP input- output functions at 8 or 16 kHz. Whereas the ANOVA showed significant main effects due to stimulus intensity, F(8, 64) = 74.552, p < 0.0001, and frequency, F(2, 16) = 8.656, p < 0.01, the ANOVA did not show any significant interac- tions which included treatment group as a factor.

DISCUSSION

The principal new finding from this experiment is that TMT disrupts auditory function at sound levels above thresh- old such that the N, amplitude remains significantly depressed at sound levels that are above detection limits. This finding is not characteristic of ototoxic agents in general but relates to the target cells for TMT toxicity. For example, the aminogly- coside antibiotics (1,7), age-related hearing loss or presbycusis (14,15), hypothermia (4), nicotine (2), venous obstruction (8), and acoustic overstimulation (23) all produce a phenomenon known as loudness recruitment in which the N, amplitude or other measure of loudness shows normal growth as sound levels increase above threshold even though detection thresh- olds are impaired.The explanation commonly advanced to ex- plain loudness recruitment is that it reflects disruption of a population of cells which is important for sound detection but which plays relatively little role in encoding loudness. Conse- quently, once sound levels exceed a critical value, the unaf- fected cell population generates a rapid increase in amplitude of the propagated signal from the cochlea yielding a normal

cochlear output (26,27).Davis et al. (8) suggested that the two populations of cells contributing to the input-output function consisted of the inner and outer hair cells and the associated Type 1 and Type 2 spiral ganglion cells. The small population of neuronal cells innervating the outer hair cell is responsible for sound detection while the high intensity portion of the curve is subserved by a large population of cells connected to the inner hair cells. Pickles (26,27) suggested, based on single cell studies, that the low intensity part of the N, intensity function curve results from the tuned response of units so that they have a low threshold at their characteristic or best frequency and are less sensitive to stimuli presented at fre- quencies other than that to which the unit is optimally tuned. The current data demonstrate that TMT disrupts both stimu- lus detection sensitivity and loudness recruitment. TMT im- pairs the sensitive “tip” of the tuning curve so that a sensory unit is less responsive at its characteristic frequency. It also elevates the “tail” of the tuning curve so the unit cannot be recruited easily even when the sound stimulus is above the auditory response threshold. The data indicate that total co- chlear nerve output is reduced by TMT regardless of sound intensity. As in previous reports using systemic administra- tions, TMT delivered via the round window directly into the cochlea produced a rapid impairment of auditory thresholds especially for higher frequency tones. The size of the shift was comparable to that observed 30 min following an IP injection of 2 mg/kg TMT (5), but the extent of the effect across test frequencies appears to differ somewhat. When TMT was in- jected IP, threshold sensitivity was significantly impaired at frequencies as low as 4 kHz (5,20) whereas application of TMT to the round window showed no effect at frequencies below 8 kHz. One explanation for this difference is that sys- temic administration of TMT can be predicted to result in distribution along the entire length of the cochlea based on the vascular bed which subserves the cochlea. Because the round window is located adjacent to the basal turn of the cochlea, a larger concentration gradient may exist with higher concentra- tions of TMT found in the basal or high frequency turn than in the apex (16). Because flow rates for perilymph fluid within the inner ear are negligible (24), it would be predicted that an appreciable period of time would be required for TMT to diffuse through the cochlear fluids. The round window appli- cation also produced some slight loss of cochlear microphonic amplitude at high frequencies while systemic administration of TMT does not disrupt the cochlear microphonic typically until several hours following administration (10). The finding of reduced cochlear microphonic output in this study suggests that TMT may reach the site of action (i.e., basal turn of cochlea) faster and/or in higher concentration with a round window application than a systemic injection (16). The conse- quence of this is disruption of the outer hair cells. Consistent with this interpretation is the reduction in the positive round window SP observed at the highest test frequency (see Fig. 4C). The permeability of the round window membrane has gained much interest among investigators. The round window membrane is thought to be a semi-permeable membrane (25) which may normally function to allow diffusion of gas across the middle ear-inner ear partition. The function and the ultra- structure have been fully described by many investigators (12,13,32). Low molecular weight substances apparently have no difficulty entering the labyrinth while high molecular weight substances can pass through the round window mem- brane into the perilymph only with difficulty (17). The appli- cation of toxicants directly to the round window membrane may be effective in evaluating inner ear toxicity separately

TMT DISRUPTS LOUDNESS RECRUITMENT 285

8kHz 1

40 60 80 100

Intensity (dB SPL)

*” 4 0 60 SO 100

Intensity (dB SPL)

32 kHz

80

,^ 2 60

so

60

s 2 -z ‘3 ‘lo

9 s a

z 2”

0

---CT-. PIPTMT 4 : - Pos,ThlT : a’

E

Intensity (dB SPL)

I 0 2”

IAknsity TiB SPL)s” I””

32 kHz

0 2”

Inte%ty (di”SPL) 80 100 0 2” 4” 60 8” 1””

Intensity (dB SPL)

FIG. 3. Effects of TMT on the Nr intensity input-output function curve (Mean f SE) at 8 kHz (A); 16 kHz (B); 32 kHz (C) in subjects 15 min after receiving TMT and in saline controls subjects at the same frequencies (D,E,F). *Pre- TMT or Saline CAP Threshold; “Post- TMT or saline CAP Threshold according to Figs. 1 and 2.

286 LIU AND FECHTER

XI,

1 A 8 kHz T 8 kHz

,’

-1 ! ,’ ,’

,’ 7,’ I

C 32 kHz 32 kHz

: : P

0 tll 40 60 4”

Intensity (dB SPL)

FIG. 4. Effect of TMT on positive round window summating potential intensity input-output function curve (Mean k SE) at 8 kHz (A); 16 kHz (B); 32 kHz (C) in subjects 15 min after receiving TMT and in saline controls group at the same frequencies (D,E,F). *Pre- TMT or saline CAP threshold; ‘Post- TMT or saline CAP threshold according to Figs. 1 and 2.

TMT DISRUPTS LOUDNESS RECRUITMENT 287

from systemic effects. Whereas round window application of the case with vascular distribution following systemic adminis- ototoxicants permits the exclusion of nonauditory factors tration. which might indirectly impair auditory function (e.g., alter- ations in blood flow, etc.), the current data also suggest that the distribution of chemical within the cochlea following

ACKNOWLEDGEMENTS

round window application may be less homogenous than is This study was supported by a grant from NIH-NIEHS ES02852.

REFERENCES

1. Aran, J. M.; Darrouzet, J. Observation of click-evoked com- pound VIII nerve responses before, during and over seven months after Kanamycin treatment in the guinea pig. Acta. Otolaryngol. 79:24-32; 1975.

2. Bobbin, R. P.; Gondra, M. I. Effect of nicotine on cochlear function and noise-induced hair cell loss. Ann. Otol. Rhino]. Lar- yngol. 85:247-254; 1976.

3. Brady, D. R.; Pearce, J. P.; Juhn, S. K. Permeability of round window membrane to 22Na or RISA. Arch. Otorhino-Laryngol. 214:183-184; 1976.

4. Butler, R. A.; Konishi, T.; Fernandez, C. Temperature coeffi- cients of cochlear potentials. Am. J. Physiol. 199:688-692; 1960.

5. Clerici, W. J.; Ross, J. B.; Fechter, L. D. Acute ototoxicity of trialkyltin in the guinea pig. Toxicol. Appl. Pharmacol. 109:547- 556; 1991.

6. Crofton, K. M.; Dean, K. F.; Menache, M. G.; Janssen, R. Tri- methyltin effects on auditory function and cochlear morphology. Toxicol. Appl. Pharmacol. 105:123-132; 1990.

7. Dallos, P.; Wang, C. Y. Bioelectric correlates of Kanamycin in- toxication. Audiology 13:277-289; 1974.

8. Davis, H.; Deatherage, B. H.; Rosenblut, B.; Fernandez, C.; Kimura, R.; Smith, C. A. Modification of cochlear potentials produced by streptomycin poisoning and by extensive venous ob- struction. Laryngoscope 68:596-627; 1958.

9. Fechter, D. L.; Carlisle, L. Auditory dysfunction and cochlear vascular injury follow trimethyltin exposure in the guinea pig. Toxicol. Appl. Pharmacol. 105:133-143; 1990.

10. Fechter, L. D.; Clerici, W. J.; Yao, L.; Hoeffding, V. Rapid disruption of cochlear function and structure by trimethyltin in the guinea pig. Hear. Res. 58:166-174; 1992.

11. Fechter, L. D.; Young, J. S.; Nuttall, A. L. Trimethyltin ototoxi- city: Evidence for a cochlear site of injury. Hear. Res. 23:275- 282; 1986.

12. Goycoolea, M. V.; Paparella, M. M.; Goldberg, B.; Carpenter, A. M. Permeability of the round window membrane during otitis media. Arch. Otolaryngol. 106:430-433; 1980.

13. Hellstrom, S.; Johansson, U.; Anniko, M. Structure of the round window membrane. Acta. Otolaryngol. Suppl. 457:33-42; 1988.

14. Henry, K. R.; Lepkowski, C. M. Evoked potential correlates of genetic progressive hearing loss. Acta. Otolaryngol. 86:366-374; 1978.

15. Hunter, K. P.; Willott, J. F. Aging and the auditory brainstem response in mice with severe or minimal presbycusis. Hear. Res. 30:207-218; 1987.

16. Ito, Z.; Kusakari, J.; Salt, A. N. Solute movement across the round window membrane in comparison with that across the blood-labyrinth barrier. Nippon Jibiinkoka Gakkai Kaiho. 95: 1959-1967; 1992.

17. Juhn, S. K.; Hamaguchi, Y.; Goycoolea, M. Review of round window membrane permeability. Acta. Otolaryngol. Suppl. 457: 43-48: 1988.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Jung, T. T. K.; Miller, S. K.; Rozehnal, S.; Woo, H. Y .; Park, Y. M.; Baer, W. Effect of round window membrane application of salicylate and indomethacin on hearing and levels of arachi- donic acid metabolites in perilymph. Acta. Otolaryngol Suppl. 493:81-87; 1992. Kawauchi, H.; DeMaria, T. F.; Lim, D. J. Endotoxin permeabil- ity through the round window. Acta. Otolaryngol. Suppl. 457: 100-115; 1989. Liu, Y.; Fechter, L. D. Trimethyltin disrupt N, sensitivity, but has limited effects on the summating potential and cochlear microphonic. Assoc. Res. Otolaryngol. St. Petersburg Feb. 6-13; 1994. Lundman, L.; Santi, P.; Morizono, T.; Juhn, S. K.; Bagger- Sjoback, D.; Harada, T. Inner ear damage and passage through the round window membrane of pseudomonas aeruginosa exo- toxin: A in a chinchilla model. Ann. Otol. Rhinol. Laryngol. 101: 437-444; 1992. Lundman, L.; Juhn, S. K.; Bagger-Sjoback, D.; Svanborg, C. Permeability of the normal round window membrane to haemo- philus influenzae type b endotoxin. Acta. Otolaryngol. 112:524- 529; 1992. Mitchell, C.; Brummett, R.; Vernon, J. Frequency effects of tem- porary N, depression following acoustic overload. Arch. Otola- ryngol. 103:177-123; 1977. Ohyama, K.; Salt, A. N.; Thalmann, R. Volume flow rate of perilymph in the guinea-pig cochlea. Hear. Res. 35:119-130; 1988. Okuno, T.; Nomura, Y. Permeability of the round window mem- brane. Arch. OtoRhinolaringol. 240:103-106; 1984. Pickles, J. 0. An introduction to the physiology of hearing. San Diego, CA: Academic Press; 1988:297-320. Pickles, J. 0. An introduction to the physiology of hearing. San Diego, CA: Academic Press; 1988:26-76. Saijo, S.; Kimura, R. S. Distribution of HRP in the inner ear after injection into middle ear cavity. Acta. Otolaryngol. 97:593- 610; 1984. Schucknecht, H. F. Ablation therapy in the management of Me- niere’s disease. Acta. Otolaryngol. 132:1-42; 1957. Stecker, R. H.; Cody, D. T. Inontophoresis of “Na and “‘I into the inner ear. Arch. Otolaryngol. 83:213-217; 1966. SuperANOVA. SuperANOVA Users Guide. Abacus Concepts. Inc. 1993:141-216. Takahashi, M.; Kanai, N. T. Cell subsets in round window mem- brane after middle ear immunostimulation. Acta. Otolaryngol. Suppl. 493:147-153; 1992. Wersall, J.; Lundquist, P. G.; Bjorkroth, D. B. Ototoxicity of gentamicin. J. Infect. Dis. 119:410-416: 1969. Young, J. S.; Fechter, L. D. Trimethyltin exposure produces an unusual form of toxic auditory damage in rats. Toxicol. Appl. Pharmacol. 82:87-93; 1986.