0196/0202/90/ 1 101-0016$02.00/0 EAR A N D HEARING Copyright 0 I990 by The Williams & Wilkins Co

Vol. I I , No. I Printid i n I' S A

ELECTROPHYSIOLOGIC TECHNIQUES IN AUDIOLOGY AND OTOLOGY

The Effect of Tinnitus on ABR Latencies* Cynthia L. lkner and Arnold H. Hassen

West Virginia School of Osteopathic Medicine, Lewisburg, West Virginia

in tinnitus latencies were greater than the variations associated with gender, age, or hearing loss. Isolating a tinnitus effect on the ABR is more difficult if hearing loss is present (Stockard, Stockard, & Sharbrough, 1978; Vernon, 1983). Authors disagree about both the extent of the hearing loss required before increases are seen in ABR latencies, as well as the specific wave or interwave measures that are affected (Hyde. 1985: Tomasulo & Peele, 1988). Some authors reported that increased latencies occurred with hearing loss (wave I: Squires, 0110, & Jordan, 1986; Weber, 1985: wave V: Weber, 1985), yet others reported decreased interpeak latencies with hearing loss (1-111, 111-V: Coats & Martin, 1977; Weber, 1985; 1-111, IV: Tomasulo & Peele. 1988: Weber, 1985). In addition, there is no agreed upon parameter of peripheral hearing that correlates best with equivalent ABRs. The most commonly used parameters reported are acous0 (Squires et al, tic reflex thresholds at 4000 Hz of ~ 1 0 dB 1986), the 1000 to 4000 Hz thresholds (Hyde, 1985), and the 4000 Hz threshold (Bauch & Olsen, 1988; Hyde, 1985: Selters & Brackmann, 1977). In order to isolate a tinnitus effect, comparisons of nontinnitus and tinnitus groups must be made using one or more index of peripheral hearing status (Ikner, Hassen, & Shaver, 1986). The present study was designed to determine if tinnitus patients showed prolonged ABR latencies when nontinnitus and tinnitus samples were matched as closely as possible for age, gender, and peripheral hearing indices.

ABSTRACT

Comparisons were made of the ABR latencies of tinnitus (T) and nontinnitus(NT) patient groups balanced for age and gender and matched for acoustic reflex threshold (ART)s, 1000 to 4000 Hz and 4000 Hz auditory thresholds, and normal hearing. In the ART match, prolongations of wave I [r(94) = 4.42, p < O.OOl], wave 111 [f(94) = 2.72,p < 0.011,and wave V [t(94) = 3.32,p < 0.011and the Ill-V interval [t(94) = 2.48,p < 0.02)were seen in T subjects. Wave I in 1 to 4 kHz matched [2(62)= 3.13,p < 0.0051and normal-hearing subjects [r(30)= 2.58,p < 0.01]was prolonged in T females. The utility of using wave I as a diagnostic indicator for tinnitus in females is discussed.

Auditory evoked potentials (AEPs) are used to examine the synchronous discharge of fibers in the auditory pathway and identify the presence of abnormal neuronal activity. The waveforms that occur in the first 10 msec of an AEP are termed the ABR (Dobie, 1980; Hausler & Levine, 1980; Weber, 1985). The ABR is the test of choice when patients present with symptoms suggesting a cochlear or retrocochlear site of lesion (Antonelli, Bellotto, & Grandon, 1987; Musiek & Gollegly, 1985; Musiek, Josey, & Glasscock, 1986). However, the reliability of the ABR for differential diagnosis may be questioned if the effects of tinnitus, hearing loss, or the combined effects of both are not considered (Antonelli et al, 1987). Systematic analyses of the cumulative effects of age, hearing loss, gender, and tinnitus on ABRs are not prevalent in clinical literature (Borg & Lofqvist, 1982; Brackman & Sheehy, 1980; Brix & Ehrenberger, 1987; Chu, 1985). In studies of tinnitus patients, for example, it was reported that the ABRs showed quantitative increases in latency and poor reproducibility (Rubin, 1984; Shea & Harell, 1978; Shulman & Seitz, 1981; Shulman & Goldstein, 1984; Shulman, 1984; Shulman, 1987). However, these reports did not address whether or not the increases *This research was supported by American Osteopathic Association Grant No. 85-12-1 86.

METHOD A core group of subjects reporting tinnitus (T = 19 males, 16 females; mean age, 40) and denying tinnitus (NT = 18 males, 17 females; mean age, 36) was assembled. Comparisons were made between the groups for four audiometric conditions: ( 1 ) crossed acoustic reflex thresholds at 1000, 2000, and 4000 Hz: (2) audiometric 1000, 2000, 4000 Hz averages; (3) hearing at 4000 Hz; and (4) absence of hearing loss (thresholds < 20 dB at 1000, 2000, 3000, and 4000 Hz). Patient matching was done for age (by decade) and gender, using thresholds +5 dB for individual ears. For example, in the 1 to 4 kHz group, a 3 1 year old female with tinnitus could be matched with a 35 year old female. If the

16

Tinnitus on ABR

3 1 year old had mean auditory thresholds of left ear 40 dB, right ear 20 dB, then the second female would have scores of 40 k 5 dB for one ear and 20 5 5 dB for the other ear. No attempt was made to separate the tinnitus patients with regard to type of tinnitus, e.g. intermittent, constant, pulsatile, nonpulsatile, unilateral, or bilateral. Each subject met inclusion criteria for normal middle ear function (Gelford, 1984; Northern, 1984; Silman & Gelford, 1981). Pure-tone air conduction thresholds 1000 to 2000 Hz averages were s 2 5 dB to ensure that the 75 dB HL presentation level was adequate (Jerger, Oliver, & Stach, 1985). All subjects had identifiable wave forms at the 75 dB stimulus level. ABR data were collected with either a Nicolet CA 1000 or a Nicolet Compact IV, using TDH-39 (Audiocup) earphones in a sound isolated room. The subject reclined on a treatment table and Grass gold electrodes were applied and impedances were measured using standard protocol (Seabea, 1985). Each ear was stimulated monaurally with two successive runs of a 100 psec square wave at 75 dB nHL using a condensation click (21.9 clicks/sec). The ipsilateral tracings were added, identified, and labeled according to prescribed criteria for this laboratory. A preliminary analysis of paired comparisons of left and right ears showed no differences (Table 1); therefore, data from both ears were combined for analyses. Student’s unpaired t-test was used to compare nontinnitus and tinnitus subjects in each of the four groups of auditory matching.

old [NT = 12 + 1.7 dB; T = 20

Table 1. ABR latencies in mean msec

+ 2.0 dB, t(94) = 2.64, p

< 0.021 in the tinnitus population was observed. However, comparison of the NT and T populations was considered valid because the mean 4000 Hz threshold was commensurate with that required for a 75 dB presentation level (Jerger et al, 1985), and there were no differences in the mean acoustic reflex values at 4000 Hz (NT = 90 -+ 2.0; T = 92 k 3.2 dB). In auditory reflex matched subjects (Table 3), the T population exhibited prolonged latencies for waves I [t(94) = 4.42, p < 0.001], 111 [t(94) = 2.72, p < 0.011 and V [(94) = 3.32, p < 0.011. The 111-V intervave latency was also elevated [t(94) = 2.48, p < 0.021. Subsequent analyses were done by gender to avoid the possibility that the unequal number of males and females in the tinnitus group was responsible for the increased latencies. Because of the increased number of comparisons, the alpha level was adjusted ( p < 0.01) to prevent a

Table 2. Composition of comparison groups by age and gender.

Age Ranges in Years 20-29 30-39 40-49 50-59 _ _ _ _ _ _ _ _ _ _ _ M F M F M F M F

RESULTS The distribution of patients by age and gender for each of the four auditory conditions is shown in Table 2. With the exception of the ART match, NT and T groups contained the same numbers of subjects in each age group, although the N decreased as restrictions on hearing sensitivity increased. In the normal hearing group, for example, there were no normal “males” in the fifth and sixth decades because males in that age range who reported tinnitus also had hearing losses. With the exception of the 4000 Hz audiometric threshold, all subjects had auditory and acoustic reflex thresholds within + I SD. An elevation of the mean 4000 Hz thresh-

17

Matched group ART NT T 1-4 kHZ NT T 4 kHz NT T Normal NT T

5 4

2 2

8 8

8 7

4 4

4 5

1 3

3 2

2 2

2 2

4 4

6 6

2 2

3 3

3 3

5 5

2 2

1 1

2 2

4 4

3 3

1 1

2 2

1 1

1 1

1 1

3 3

4 4

2 2

1 1

+ SE for left (L) and right (R) ears. Absolute Latencies

Wave I

NT

1.80

* 0.16

t (df) T t (df)

1.86

*

1.82 k 0.17 -0.224 (33) 0.19 1.89 -C 0.15 -0.783 (33)

L R 2.08 k 0.15 2.16 0.09 -1.713 (33) 1.95 0.15 1.95+ 0.13 0.252 (33)

+

*

R

L

3.83 & 0.18 3.85 f 0.20 -0.766 (33) 3.94 f 0.22 3.91 0.22 0.627 (33) lnterwave Intervals

5.80 k 0.25 5.84 k 0.24 -0.902 (33) 6.00 f 0.25 5.93 f 0.25 1.074 (33)

IWI Ill-v

IWI I-v

+

1w11-111 NT t (df) T t (df)

R

L

R

L

Wave V

Wave 111

L

1.91

* 0.19

R

1.85 f 0.20 0.990 (33) 2.01 & 0.17 2.03 f 0.22 0.343 (33)

L

4.00

+ 0.15

4.02 -0.537 (33) 4.09 f 0.23 4.03 1.314 (33)

R

* 0.18

* 0.20

Student’s paired t-test, two-tailed. Ear and Hearing, Vol. 11, No. 1, 1990

18

lkner and Hassen

Bonferoni effect (Kirk, 1968). Using more rigorous analyses, there was a difference in wave I latencies in 1 to 4 kHz matched females [t(62) = 3.13, p 0.0051, (Table 4), but not in wave I for females in the 4000 Hz threshold match [t(26) = 1.77, p < 0.051, seen in Table 5. In the normal hearing group (Table 6), the prolonged wave I latency in female tinnitus subjects was again observed [t(30) = 2.58, p < 0.011. There was also an observable difference in female wave I11 [t(30) = 2.30, p < 0.021 that approached significance. Figure 1 is a summary graph of ABR Wave I, 111, and V values for gender by condition. Interestingly, T females have latencies for waves 1 and 111 that are similar to all males in each of the four conditions. These data indicate that the presence of tinnitus could explain the obliteration

-=

Table 3. Comparison of nontinnitus (NT) and tinnitus (T) group latencies in mean msec f SE.

Absolute Latencies WI NT T t (df)

NT T t (df)

1.76 t 0.02 1.88 f 0.02 4.42 (94)8 IWI 1-111 2.08 & 0.02 2.05 f 0.03 1.08 (94)

Wlll

wv

3.80 f 0.03 5.72 f 0.04 3.93 & 0.03 5.91 f 0.04 2.72 (94)b 3.32 (94)” lnterwave Latencies IWI Ill-v 1.92 f 0.03 2.01 f 0.03 2.48 (94)b

IWI I-v 3.98 f 0.03 4.06 f 0.03 1.87 (94)”

Difference, p < 0.00 7. bDifference,p < 0.07. Difference, p < 0.05. Student’s unpaired t-test, one-tailed.

that the presence of tinnitus could explain the obliteration of male/female ABR latency differences recently reported in patients with cochlear hearing loss (Bauch & Olson, 1987). The salient finding was the significantly different wave I prolongations in T females compared with NT females that persisted at all but the 4000 Hz condition. Therefore, further study should focus on wave I as a diagnostic indicator of tinnitus in females. Table 5. ABR latencies in mean msec f SE for males and females matched for age and 4000 Hz thresholds.

Absolute Latencies Wave I

Wave 111

Wave V

t (df)

1.87 f 0.06 1.85 f 0.02 0.28 (34)

4.08 t 0.06 4.01 f 0.04 0.82 (34)

6.02 & 0.05 6.09 k 0.04 1.03 (34)

Female NT T t (df)

1.80 f 0.05 1.91 f 0.04 1.77 (26)

Male NT T

Male NT T t (df) Female NT T t (df)

IWI 1-111

IWI Ill-v

IWI I-v

2.20 f 0.04 2.17 t 0.05 0.46 (34)

2.08 f 0.03 1.95 -+ 0.07 1.67 (34)

4.18 f 0.06 4.24 k 0.04 0.87 (34)

2.02 -t 0.05 1.96 f 0.05 0.85 (26)

2.01 f 0.07 1.98 & 0.05 0.46 (26)

4.03 f 0.05 3.93 f 0.07 1.24 (26)

Student’s unpaired t-test, one-tailed.

Table 4. ABR latencies in mean msec f SE for male and female age and 1000-4000 Hz threshold matched subjects.

Table 6. ABR latencies in mean msec & SE for male and female age matched, normal hearing subjects.

Absolute Latencies

Male NT T t (d9 Female NT T t (df)

Wave 111

Wave V

1.94 f 0.04 1.90 f 0.04 0.69 (42)

4.04 f 0.05 3.99 f 0.04 0.78 (42)

6.05 f 0.04 5.96 f 0.05 0.78 (42)

IWI 1-111 Male NT T t (df) Female NT T t (df) a

Absolute Latencies

Wave I

1.76 f 0.03 1.90 f 0.03 3.13 (62)”

5.77 f 0.04 3.83 f 0.03 5.81 f 0.04 3.91 f 0.04 1.83 (62) 0.74 (62) lnterwave Latencies IWI Ill-v

Male NT T t (do Female NT T t (df)

IWI I-v

2.10 f 0.05 2.12 f 0.04 0.04 (42)

2.01 & 0.05 1.97 k 0.04 0.47 (42)

4.11 f 0.05 4.09 f 0.05 0.24 (42)

2.06 f 0.04 2.01 & 0.03 0.81 (62)

1.95 f 0.04 1.90 f 0.03 0.97 (62)

4.01 f 0.04 3.91 f 0.05 1.52 (62)

Difference, p < 0.005, Student’s unpaired t-test, one-tailed.

Ear and Hearing, Vol. 11, No. 1, 1990

3.82 f 0.04 5.70 f 0.06 3.89 f 0.06 5.87 f 0.06 1.11 (26) 0.86 (26) Interwave Latencies

Male NT T t (d9 Female NT T t (d9

Wave I

Wave 111

Wave V

1.93 f 0.06 1.84 f 0.05 1.06 (10)

3.96 f 0.06 4.05 f 0.06 0.06 (10)

5.99 & 0.09 5.96 k 0.06 0.21 (10)

1.75 k 0.04 1.92 2 0.05 2.58 (30)”

3.77 f 0.04 5.70 t 0.05 5.83 f 0.07 3.94 f 0.06 2.30 (30) 1.34 (30) lnterwave Latencies

IWI 1-111

IWI Ill-v

IWI I-v

2.12 f 0.09 2.13 f 0.08 0.06 (10)

1.94 f 0.12 2.00 f 0.04 0.49 (10)

4.00 f 0.09 4.10 f 0.10 0.51 (10)

1.95 f 0.05 2.02 f 0.05 1.02 (30)

1.95 f 0.07 1.89 f 0.05 0.67 (30)

4.00 t 0.09 3.97 & 0.06 0.21 (30)

Difference, p < 0.07,Student’s unpaired t-test, one-tailed.

Tinnitus on ABR

Normal

4kHz

1QkHz

i!

1

6.10 6.00

i

5.90 5.80 5.70

3.90 4'00 3.80

2.00

1.90

lB0

7T

"i

i

'! i . i II 4

b

ART

i 6

P

i

i

T

i, i f

*I

*Y

1.70 M

F

M

F

3

7

9

8

M 11

F 16

19

Rosenthall et al, 1985; Rowe, 1978). Because from 85 to 96% of patients with hearing loss have tinnitus (Antonelli et al, 1987), it is resonable to assume that the presence of tinnitus caused the unexplainable lengthened ABR latencies reported. Data on presence or absence of tinnitus in these patients were unavailable (Bauch & Olson, personal communication; Hyde, 1985). Two explmations have been proposed for the shorter latencies observed in female patients with normal hearing; a smaller diameter of the auditory nerve (AN) in females, causing an altered conduction velocity, and a shorter conduction distance to the AN (Stockard, Stockard, Westmoreland, & Corfits, 1979). However, nerve diameter should be less of a factor, as more rapid saltatory conduction usually occurs in nerves of greater diameter (Ross & Reith, 1985). A thicker myelin sheath in females is another anatomical possibility that could occur to facilitate conduction (Rowland, 1985). However, conduction is not facilitated for the wave I response in tinnitus females. Since the wave I response reflects first order afferents from the cochlea, tinnitus arising from cochlear damage or disruption of AN fiber activity could affect wave I. Therefore, a neuropathy that affects the onset response, such as a myelin defect, also would be reflected in slower conduction velocities (Rowland, 1985). This hypothesis, however, does not explain the reported lack of a similar latency shift in tinnitus males, a subject that requires further study in order to understand the underlying mechanism. References

M 17

F 15

Rgum 1. Summary graph by condition with number of male and female wbjects (number of ears 2M + 2F), comparing nontinnitus and tinnitus wave I, 111, and V latencies. Values are mean kSE; *, differences; p < 0.01; Student's unpaired t-test. one-tailed.

DISCUSSION

The principal finding of our study was that in women there was a measurable difference in wave I latencies if ABRs obtained from tinnitus and nontinnitus patients wrecompared. It was also observed that wave I11 latencies in female tinnitus patients were lengthened, although these Berences were not significant at the same level as wave I. Taken together, these data demonstrated that although tinnitus had no statistically demonstrable effects on the ABR latencies of male tinnitus patients, significant lengthening of latencies may occur in the female. This may be a possible explanation for the unexplained observation that females with hearing loss did not show characteristically sbortened latencies relative to males (Bauch & Olson, 1987). For example, it has been reported that subjects with cochlear hearing loss often showed lengthened ABR latencies which could not be explained by extent or slope of the hearing loss (Bauch & Olson, 1987; Hyde, 1985;

Antonelli A, Bellotto R, Grandori F. Audiologic diagnosis of central versus eighth nerve and cochlear auditory impairment. Audiology 1987;26:209-226. Bauch CD, Olson WO. Auditory brainstem responses as a function of average hearing sensitivity for 2000-4000 Hz. Audiology 1987;27:156-163. Borg E, Lofqvist L. Auditory brainstem response (abr) to rarefaction and condensation clicks in normal and abnormal ears. Scand Audio1 1982;11:227. Brackmann DE, Sheehy JI. The neuro-otologic evaluation. In Keith R, Ed. Audiology for the Physician. Baltimore: Williams & Wilkins, 1980: 202. Brix R, Ehrenberger K. Subjective and objective methods for tinnitus therapy control. In Feldman R, Ed. Proceedings 111 International Tinnitus Seminar (Munster), 1987: 170-173. Chu N. Age-related latency changes in the brainstem auditory evoked potentials. Electroencephalogr Clin Neurophysiol 1985;62:43 1-436. Coats AC, Martin JL. Human auditory nerve action potentials and brain stem evoked responses. Arch Otolaryngol 1977;103:605-622. Dobie RA. Physiological techniques used in the assessment of the auditory system. In Keith R, Ed. Audiology for the Physician. Baltimore: Williams & Wilkins, 1980 14. Gelford S . The contralateral acoustic reflex threshold. In Silman S, Ed. The Acoustic Reflex. Orlando: Academic Press, 1984: 168. Hausler R, Levine R. Brain stem auditory evoked potentials are related to interaural time discrimination in patients with multiple sclerosis. Brain Res 1980;19 1:589. Hyde M. The effect of cochlear lesions on the abr. In Jacobson JJ, Ed. The Auditory Brainstem Response. San Diego: College-Hill Press, 1985: 136. Ikner C, Hassen A, Shaver T. Objective monitoring of osteopathic manipulative treatment in tinnitus patients. J Am Osteopath Assoc 1986;(Abstr)86:671-672. Jerger J, Oliver T, Stach B. ABR testing strategies. In Jacobson JJ, Ed. The Auditory Brainstem Response. San Diego: College-Hill Press, 1985: 382-386. Kirk R. Experimental design: procedures for the behavioral sciences. Belmont, CA: Brooks/Cole, 1968. Musiek P, Gollegly K. ABR in eighth nerve and low brainstem lesions. In Jacobson JJ, Ed. the Auditory Brainstem Response. San Diego: College-Bill Press, 1985: 180-202. Musiek F, Josey A, Glasscock M. Auditory brainstem response-interwave measurements in acoustic neuromas. Ear Hear 1986;7:100-105. Northern J. Impedance audiometry. In Northern JL, Ed. Hearing Disorders. Boston: Little, Brown, and Company, 1984: 48.

Ear and Hearing, Vol. 11, No. 1, 1990

20

lkner and Hassen

Rosenhall GU, Bjorkman G, Pedersen RA. Brain-stem auditory evoked potentials in different age groups. Electroencephalogr Clin Neurophysiol I985;62:426-430. Ross M, Reith E. Histology: A Text and Atlas. New York: Harper & Row, 1985: 247. Rowe JM. Normal variability of the brain-stem auditory evoked response in young and old adult subjects. Electroencephalogr Clin Neurolphysiol 1978;44:46 1 . Rowland L. Diseases of the motor unit: the motor neuron, peripheral nerve, and muscle. In: Kandel E. Schwartz J, Eds. Principles of Neural Science. New York: Elsevier, 1985: 204-205. Rubin W. Tinnitus evaluations: aids to diagnosis and treatment. In Shulman A, Ballantyne J, Eds. Proceedings o f the I1 International Tinnitus Seminar. J Laryngol Otol 1984;(SuppI 9): 179. Seabea P. The importance of measuring electrode impedence. Oxford, Florida Observer Newsletter. Clearwater: Oxford Medilog. Inc., 1985. Shea T, H a d M. Management of tinnitus aurium with lidocaine and carbamazepine. Laryngoscope 1978;(Pt. 1):1477-1484. Shulman A. Electro-acoustics and tinnitus (electrodiagnostic cochleo-vestibular test battery). In Feldman H, Ed. Proceedings 11 International Tinnitus Seminar (Munster), 1987: 190. Shulman A, Goldstein B. Neurotologic classification of tinnitus. In Shulman A, Ballantyne J. Eds. Proceedings of the 11 International Tinnitus Seminar. J Laryngol Otol 1984;(Suppl9):147-149. Shulman A, Seitr M. Central tinnitus-diagnosis and treatment: observations of simultaneous binaural auditory brain responses with monaural stimulation in the tinnitus patient. Laryngoscope 1981 :92:2025-2036. Silman S, Gelford S. The relationship between magnitude of hearing loss and acoustic reflex threshold levels. J Speech Hear Disord 198 1;46:312-3 16.

Ear and Hearing, Vol. 11, No. 1, 1990

Squires N, 0110 C, Jordan R. Auditory brain stem responses in the mentally retarded: audiometric correlates. Ear Hear 1986;7:90. Stockard JE, Stockard JJ, Westmoreland B, Corfits J. Brainstem auditory-evoked responses, normal variation as a function of stimulus and subject characteristics. Arch Neurol 1979;36:328-83 1. Stockard JJ, Stockard JE. Sharbrough FW. Nonpathologic factors influencing brainstem auditory evoked potentials. Am J EEG Techno1 1978: 18: 177-200. Tomasula RA, Peele PB. A new technique for interpreting the BAER in cochlear diesease. Ann Neurol 1988;23:204-206. Vernon J. Reliefoftinnitus masking treatment. In English GM. Ed. Otolaryngology. Philadelphia: Harper and Rowe, 1983: 1-20. Weber BA. Interpretation problems and pitfalls. In Jacobson JJ. Ed. The Auditory Brainstem Response. San Diego: College-Hill Press, 19x5: 100.

Acknowledgments:The authors express their appreciation to Richard S. Tyler, Ph.D. and Kathleen C. M. Campbell, Ph.D. for their helpful comments on an earlier version of this paper, and to Abraham Shulman, M.D. for his encouragement which led to implementation of this project. Address reprint requests to C. L. Ikner, WVSOM Health Center, Inc., 400 N. Lee St., Lewisburg, WV 24901. Presented in part to the American Auditory Society 13th Annual Meeting, November 20, 1986 in Detroit. Received July 28, 1988; accepted August 11, 1989.