S ensorineural hearing loss has been

Transcription

1 Am Acad Audiol 9 : (1998) Monitoring and Predicting Ototoxic Damage Using Distortion-Product Otoacoustic Emissions : Pediatric Case Study Thomas A. Littman* Amy Magruder* Douglas R. Strothert Abstract Young children undergoing cisplatin chemotherapy are known to be at risk for progressive sensorineural hearing loss. Early detection of such hearing loss is important for providing management options. However, in ill and/or young children, behavioral audiometry may not be sufficiently precise to detect the early stages of hearing loss. This case illustrates that distortion-product otoacoustic emissions (DPOAEs) may be an appropriate cross-check measure to supplement and confirm pediatric behavioral data. Perhaps more importantly, this study suggests that DPOAEs may have the potential to predict the earliest stages of progressive hearing loss before such changes are seen in audiometric thresholds. Key Words : Distortion-product otoacoustic emissions, ototoxicity, progressive hearing loss Abbreviations : DPOAEs = distortion-product otoacoustic emissions, OAEs = otoacoustic emissions, OHC = outer hair cell, TEOAEs = transient evoked otoacoustic emissions S ensorineural hearing loss has been reported in 5 percent (Brock et al, 1988) to 95 percent (Skinner et al, 199) of children who are undergoing cisplatin chemotherapy. The losses tend to progress from 6 and 8 Hz to lower frequencies (Pasic and Dobie, 1991). However, behavioral audiometry in ill and/or young children may not be sufficiently precise to detect the initial stages of hearing loss, particularly at higher frequencies. An objective and accurate cross-check measure is needed for such monitoring. Otoacoustic emissions (OAEs) might be appropriate tools for this purpose because (a) cisplatin damages cochlear outer hair cells (OHCs) (Fleishman et al, 1975) ; (b) OAEs are sensitive to cochlear hearing loss associated with OHC damage (Long and Tubis, 1988 ; Brown et al, 1989 ; Brownell, 199) ; and (c) OAEs appear to be sensitive to cisplatin ototoxicity in animals (McAlpine and ohnstone, 199) and humans (Probst et al, 1993 ; Zorowka et al, 1993). For monitoring purposes, distortion-product otoacoustic emissions (DPOAEs) would seem preferable to transient evoked otoacoustic emissions (TEOAEs) because DPOAEs provide better results at higher frequencies (Gorga et al, 1993), where cisplatin ototoxicity is first seen in children (Schell et al, 1989 ; Skinner et al, 199 ; Pasic and Dobie, 1991). The following case presentation illustrates the utility of DPOAEs for monitoring cisplatin ototoxicity in young children. Perhaps more importantly, this case demonstrates that DPOAEs may have some ability to "predict" the progression of hearing loss before the shift in pure-tone threshold occurs. CASE REPORT *Audiology Department, Texas Children's Hospital, Houston, Texas, tbaylor College of Medicine, Houston, Texas Reprint requests : Thomas Littman, Virginia Mason Medical Center, Section of Otolaryngology/Audiology, 11 Ninth Ave., P.O. Box 9, Seattle, WA he subject of this report was diagnosed with T medulloblastoma at 41/2 years of age. Total resection of the tumor was achieved via posterior fossa approach. The patient was subsequently enrolled in a Pediatric Oncology Group

2 ournal of the American Academy of Audiology/Volume 9, Number 4, August 1998 Table 1 Medulloblastoma Treatment Protocol Dates Treatment Total Dose 9/2/95 to 11/7/95 Radiation 532 cgy 12/12/95 to 2/2/96 Cisplatin (9 mg/mg2) and 27 mg/m2 (3 courses) etoposide (3 mg/m2) 9 mg/m2 2/22/96 to 9/2/96 Cyclophosphamide (2 mg/m2) and 16, mg/m2 (8 courses) vincristine (1.5 mg/m2) 12 mg/m2 treatment protocol in the Hematology-Oncology section at Texas Children's Hospital. Doses of radiation and chemotherapy given to this patient are summarized in Table 1. The ototoxicity of cisplatin (Schweitzer, 1993) and, to a lesser extent, vincristine (Mahajan, 1981 ; Lugassy, 199) is well documented. Furthermore, prior cranial irradiation is known to potentiate cisplatin ototoxicity (Sexauer et al, 1985 ; Walker et al, 1989 ; Weatherly et al, 1991). Therefore, pre- and post-treatment audiologic monitoring was included in the treatment protocol. METHOD A udiometric evaluations included pure-tone air- and bone-conduction audiometry (using play techniques), an acoustic immittance battery, and DPOAEs (the cubic distortion product, 2fi f2). Due to patient fatigue, speech audiometry could not be completed until test 3. DPOAEs were collected with fl and f2 levels of 65 and 5 db SPL, respectively. The frequency ratio of the primaries (f2/fl) was Ear canal calibration data were monitored to ensure that adequate stimulus energy was present at each frequency. Absent or reduced amplitude DPOAEs (less than 5 db above the two standard deviation noise floor) were considered to be truly abnormal only if the ear canal sound pressure level of the primaries was within 5 db of nominal levels. DPOAEs were accepted as normal if the primaries were no more than 1 db below nominal levels and if 2fl-f2 amplitude was 5 db or more above the two standard deviation noise BASELINE ( ) RIGHT E AR LEFT EAR m a m a Distortion Product-gram Figure 1 Baseline audiometric test results. Tympanograms and acoustic reflexes were within normal limits bilaterally. For this and for subsequent DPOAE plots at the bottom of Figures 2 and 3, DPOAE amplitude (circles connected by lines) is displayed as a function of frequency. The upper shaded area represents two standard deviations above the mean noise floor. DPOAE collection parameters are detailed in the Methods section khz (F2) 258

3 Predictive Value of DPOAEs/Littman et al TEST 2 ( ) RIGHT EAR LEFT EAR Distortion Product-gram m 6 Figure 2 Test 2, following treatment radiation, risplatin, and etoposide. Tympanograms and acoustic reflexes were within normal limits bilaterally. Bone-conduction thresholds were equivalent to air-conduction thresholds. DPOAE collection parameters are detailed in the Methods section : \ : ; ; ; floor. DPOAEs were not accepted as normal whenever primary levels exceeded nominal values by more than 5 db. Emissions were collected using the Otodynamics IL92 system (software version 4.2B+). A baseline audiometric evaluation was completed after surgery but before radiation and chemotherapy. Test 2 was performed 6 months later, following radiation and three courses of cisplatin and etoposide. Test 3 was completed 4 months after completion of all therapy (11 months after test 2). RESULTS B aseline audiometry, shown in Figure 1, reflects normal sensitivity and normal DPOAEs bilaterally. In the ear canal, the amplitude of f2 at 6 Hz for the right ear was 6.8 db below the nominal level of 5 db SPL, possibly accounting for the somewhat diminished emission at that frequency. The amplitude of f2 could not be increased by refitting the probe, apparently due to idiosyncratic ear canal resonance characteristics. Nonetheless, the 6-Hz emission is considered normal by our clinic's criteria because it exceeds the two standard deviation noise level by more than 5 db. Acoustic immittance findings were normal for this evaluation and for all subsequent retests. The immittance data have been omitted from the figures to conserve space. Audiometric results following completion of cisplatin and etoposide treatment are presented in Figure 2. A bilateral high-frequency sensorineural loss is apparent. DPOAEs were absent for the right ear at and above 15 Hz and were absent or abnormal for the left ear at and above 2 Hz. Note that DPOAEs were abnormal bilaterally at 2 and 3 Hz, despite audiometric thresholds of 1 and 15 db HL for those frequencies in both ears. Boneconduction thresholds were consistent with airconduction thresholds but were omitted from the figures for the sake of clarity. The results of test 3, following completion of cyclophosphamide and vincristine therapy, are presented in Figure 3. The high-frequency sensorineural loss had progressed in both ears. Note that the loss now included 2 and 3 Hz for both ears (2 Hz for the left ear was now 15 db poorer than at baseline). DPOAEs for the right ear were much the same as for test 2, although the emission at 15 Hz appeared to

4 ournal of the American Academy of Audiology/Volume 9, Number 4, August 1998 TEST 3 ( ) RIGH T EAR LEFT EAR 2 2 = 4 = is, a Figure 3 Test 3, following treatment with cyclophosphamide and vincristine. Boneconduction thresholds were equivalent to air-conduction thresholds. Tympanograms and acoustic reflexes were within normal limits bilaterally, except for 4 Hz, where contralateral reflexes were absent for both ears. Diagnostic speech audiometry was within normal limits bilaterally. DPOAE collection parameters are detailed in the Methods section khz (F2) khz (F2) have "recovered" somewhat and was now 3.4 db above the noise floor. The left ear emissions in Figure 3 remained abnormal for the higher frequencies, while lower frequency emission amplitude increased, relative to test 2. The left ear DPOAE at 2 Hz appeared to have recovered partially and was now 2.2 db above the noise floor (this emission might have been larger, but the noise floor at 2 Hz was greater at test 2 than at the baseline). This subject's most recent evaluation indicated that the hearing loss was stable, and the DPOAEs have shown no further recovery. The relationship between DPOAEs and puretone thresholds at 2 Hz is shown in Figure 4. This figure clearly indicates that the DPOAE amplitude dropped into the noise in test 2 before a significant threshold shift was observed at test 3. Figure 5 reflects a similar pattern at 3 Hz, where DPOAE amplitude dropped into or near the noise floor while sensitivity remained normal during test 2. Sensitivity at 3 Hz subsequently declined by 45 and 35 db for the right and left ears, respectively, at test 3. DISCUSSION n this case study, DPOAEs were sensitive to the earliest stages of ototoxicity. DPOAEs were abnormal for f2 frequencies at which sensorineural hearing loss exceeded 2 db HL, consistent with group studies in which similar recording parameters were employed (Gorga et al, 1993 ; Sun et al, 1996). DPOAEs would therefore seem to be an appropriate cross-check measure for monitoring ototoxicity due to cisplatin and other agents. Furthermore, after a complete audiometric baseline has been established, it may be feasible to monitor periodically using DPOAEs alone. Additional audiometric data would then be obtained only when DPOAEs dropped below baseline levels. In this way, monitoring might be more rapid, less expensive, and less taxing for a patient who often is not feeling well. Perhaps of greater interest, this case study suggests that DPOAEs may be more sensitive to incipient cochlear damage than behavioral thresholds. The striking pattern in Figures 4 and 26

5 Predictive Value of DPOAEs/Littman et al 2 Hz - RIGHT EAR] 25 3 Hz - RIGHT EAR I- 2s 1 a 2 a In 1 15 v O 2 L 5 ~ t O d t Z O H 4 N Z] ~ -5 C H 5 F rn tit.1 w Q d O 6 a 4 -~ -2 7 Baseline Test 2 Test 3 Baseline Test 2 Test 3 `- Pure-Tone Threshold e DPOAE Signal-to-Noise Redo Pure-Tone Threshold D DP.- Signal-to-Noise. Ratio 2 Hz - LEFT EAR Pure Tone Threshold D DPOAE Signal-to-Noise Ratio 11 Pure-Tone Threshold r3 DPOAE Signal-to-Noise Ratio Figure 4 The relationship between pure-tone thresholds at 2 Hz and DPOAEs for f2 = 2 Hz. DPOAE amplitude is displayed in terms of signal-to-noise ratio (amplitude of the emission above the two standard deviation line of the noise floor). Note that the DPOAE dropped into the noise floor at test 2, while no significant change was seen in the pure-tone threshold for the right (a) and left (b) ear. The DPOAE for the left ear "recovered" somewhat at test 3 and is now 2.4 db above the noise floor. Figure 5 The relationship between pure-tone thresholds at 3 Hz and DPOAEs for f2 = 3 Hz. DPOAE amplitude is displayed in terms of signal-to-noise ratio (amplitude of the emission above the two standard deviation line of the noise floor). Note that the DPOAE dropped into the noise floor at test 2, while no significant change was seen in the pure-tone threshold for the right ear (a). For the left ear (b), the DPOAE dropped to within 2.6 db of the noise (a drop of 18.2 db from baseline), with no significant change in pure-tone sensitivity. 5 is that DPOAE amplitudes fell into or near the noise floor before behavioral threshold changes were noted at corresponding frequencies. In this sense, the DPOAEs may be predictive, foretelling a substantial threshold shift for a given frequency prior to a measurable sensitivity loss. Because 11 months elapsed between monitoring sessions, we cannot speculate on how much time actually elapsed between the DPOAE drop and a pure-tone sensitivity shift. Warning of impending hearing loss could be useful for the oncologist, who might have the option of adjusting the chemotherapy to a potentially less ototoxic regimen. Likewise, early indicators of threshold shift would be useful for planning audiologic management and counseling. We speculate that the "predictive" drop in the DPOAE reflects the progression of hair cell damage. Thus, at the initial stage of toxicity, the chemotherapy agents may only damage one or two rows of OHCs, affecting DPOAEs but not threshold sensitivity. Subsequently, as the drugs damage the remaining OHCs in that f2 frequency region, pure-tone sensitivity is reduced. A limitation of this case study is that only two postbaseline measures were obtained. As a result, we do not have a clear picture of the time course of this phenomenon. We cannot determine the temporal interval by which the DPOAE drop precedes the sensitivity change. Similarly, we cannot comment on whether the DPOAE and pure-tone changes were gradual or precipitous. The case described herein is not an isolated one. We now have similar data from 14 children (9 undergoing cisplatin chemotherapy and

6 ournal of the American Academy of Audiology/ Volume 9, Number 4, August with congenital cytomegalovirus infection), demonstrating the predictive ability of DPOAEs. It therefore seems that DPOAEs may have the potential to predict the earliest stages of progressive cochlear hearing loss. Acknowledgment. This work was presented, in part, at the Twentieth Midwinter Research Meeting of the Association for Research in Otolaryngology, St. Petersburg Beach, FL, February 2, REFERENCES Brock P, Pritchard, Bellman S, Pinkerton CR. (1988). Ototoxicity of high-dose cis-platinum in children. Med Pediatr Oncol 16 : Brown AM, McDowell B, Forge A. (1989). Acoustic distortion products can be used to monitor the effects of chronic gentamicin treatment. Hear Res 42 : Brownell WE. (199). Outer hair cell electromotility and otoacoustic emissions. Ear Hear 11 : Fleishman RW Stadnicki SW, Ethier MF, Schaeppi U. (1975). Ototoxicity of cis-dichlorodiammine platinum (11) in the guinea pig. Toxicol Appl Pharmacol 33 : Gorga MP, Neely ST, Bergman B, Beauchaine R, Kaminski P, Peters, Schulte L, esteadt W. (1993). A comparison of transient-evoked and distortion product otoacoustic emissions in normal-hearing and hearingimpaired subjects. Acoust Soc Am 94 : Long GR, Tubis A. (1988). Modification of spontaneous and evoked otoacoustic emissions and associated psychoacoustic microstructure by aspirin consumption. Acoust Soc Am 84 : Lugassy G. (199). Sensorineural hearing loss associated with vincristine treatment. Blut 61 : Mahajan SL. (1981). Acute acoustic nerve palsy associated with vincristine therapy. Cancer 47 :244. McAlpine D, ohnstone BM. (199). The ototoxic mechanism of cisplatin. Hear Res 47 : Pasic TR, Dobie RA. (1991). Cis-platinum ototoxicity in children. Laryngoscope 11: Probst R, Harris FP, Hauser R. (1993). Clinical monitoring using otoacoustic emissions. Br Audiol 27 :85-9. Schell M, McHaney VA, Green AA, Kun LE, Hayes A, Horowitz M, Meyer W. (1989). Hearing loss in children and young adults receiving cisplatin with or without prior cranial irradiation. Clin Oncol 17: Schweitzer VG. (1993) Ototoxicity of chemotherapeutic agents. The Otolaryngology Clinics of North America. Philadelphia : WB Saunders, 26 : Sexauer CL, Khan A, Burger PC. (1985). Cisplatin in recurrent pediatric brain tumors. Cancer 56: Skinner R, Pearson AD, Amineddine HA, Mathias DB, Craft AW (199). Ototoxicity of cisplatinum in children and adolescents. Br Cancer 61: Sun X-M, ung MD, Kim DO, Robinson K. (1996). Distortion product otoacoustic emission test of sensorineural hearing loss in humans: comparison of unequaland equal-level stimuli. Ann Otol Rhinol Laryngol 15: Walker DA, Pillow, Waters KD, Keir E. (1989). Enhanced cis-platinum ototoxicity in children with brain tumors who have received simultaneous or prior cranial irradiation. Med Pediatr Oncol 17 : Weatherly RA, Owens, Catlin Fl, Mahoney DH. (1991). Cis-platinum ototoxicity in children. Laryngoscope 11: Zorowka PG, Schmitt H, Gutjar P (1993). Evoked otoacoustic emissions and pure-tone threshold audiometry in patients receiving cisplatinum therapy. Int Pediatr Otorhinolaryngol25 :73-8.