Portuguese Version

Year:  2003  Vol. 69   Ed. 2 - (11º)

Artigo Original

Pages: 215 to 221

PDF PT  

Audiological monitoring of cisplatin exposed patients

Author(s): Adriana P. Garcia[1],
Maria Cecília M. Iório[2],
Antônio S. Petrilli[3]

Keywords: hearing, cisplatin, audiometry.

Abstract:
Introduction: Cisplatin is an antineoplasic drug, which has ototoxicity as a side effect. The goals of this paper were to evaluate the audiological behavior in osteosarcoma patients treated with cisplatin and to verify which evaluation method is the best for early detection of drug induced hearing loss. Methods: 13 patients, that received four cisplatin cycles of 120 mg/m2/cycle divided in two days (60 mg/m2/day), were evaluated prior to start of chemotherapy, prior to each scheduled course and at the end of treatment. It was performed the pure tone audiometry (250 to 18000 Hz) and the transitory and distortion product otoacoustic emission (TOAE and DPOAE). Results: In the mean values, it was observed hearing loss, after 480 mg/m2 cumulative cisplatin dosage, beginning at 8 kHz. At the individual values, it was observed that 15,3% had mild to moderate hearing loss beginning at 3kHz, 15,3% beginning at 4 kHz, 15,3% beginning at 6 kHz and 15,3% beginning at 8 kHz. TOAE did not show changes before the audiometry. DPOAE showed smaller amplitude after the cycles of cisplatin, but this change happened together with the audiometry - not prior. Conclusion: The high frequency audiometry was more efficient to detect early ototoxicity. TOAE and DPOAE can be used as complement tests. All cisplatin exposed patients showed high frequency hearing loss, 30,6% showed hearing loss in important frequencies (3 and 4 kHz) for speech comprehension.

INTRODUCTION

Hearing loss can be congenital or acquired, in addition to presenting different grades and types. It is considered to be a disease with vocational, educational and social consequences owing to the great importance it has on human communication.

One of the causes of acquired hearing loss is the use of ototoxic drugs which are drugs that cause toxic reactions to the structures of the inner ear affecting the auditory and/or vestibular system. The drugs considered to be ototoxic are aminoglycoside, carboplatin and cisplatin, among others1.

The incidence of cisplatin-induced hearing loss according to the literature ranges from 3 to 100%2. The ototoxicity affects initially high frequencies, with posterior involvement of the lower frequencies, according to the drug cumulative dose3, 4, being that cisplatin-induced hearing loss is normally sensorineural, bilateral symmetrical and irreversible5, 6.

The drug administration mode is one of the determining factors for its ototoxic effects, since its high dose use in one presentation affects more the hearing than the same dosage administered in fractions7. In addition, the auditory affection can vary in each patient, that is, there is individual susceptibility8, 9.

Since it is effective in solid tumors, cisplatin is highly used in the treatment of bone tumors. It amounts to about 10% of the malignant neoplasm cases that affect children and adolescents, being the sixth most common type of cancer in children and the third in adolescents and young adults, second only to leukemia and lymphoma. Among the bone tumors, osteosarcoma is the most frequent one, totaling about 60% of the cases10.

The treatment available for osteosarcoma up to the 70's was surgery, limited to amputation of the entire bone. The prognosis for disease-free survival was 10 to 25%. In the 70's, some attempts to treat with adjuvant chemotherapy were made, which increased the survival of patients11. Since then, anti-neoplasm drugs have been used in the treatment of osteosarcoma, among them cisplatin and carboplatin, considered ototoxic drugs.

In patients exposed to ototoxic drugs, auditory monitoring is essential, since it results in early detection of ototoxicity induced hearing loss, enabling the review of treatment or even the choice of alternative procedures. If the measures are not possible, the family and the patient should be warned about the possible hearing loss, since the social damage of ototoxic induced hearing loss can contribute to further isolation of cancer patients.

The purposes of the present study were to assess hearing of osteosarcoma patients submitted to chemotherapy treatment using cisplatin and to check the most appropriate investigation method for early identification of auditory disorders.

MATERIAL AND METHOD

The present study was approved by the Research Ethics Committee at UNIFESP/EPM, under number 681/00.
The material of the present study included audibility thresholds of frequencies 250 to 18,000Hz dB SPL of transient otoacoustic emissions (TOAE) and distortion product otoacoustic emissions (DPOAE) of 13 patients with diagnosis of osteosarcoma submitted to chemotherapy, and referred to the Division of Pediatric Oncology, Department of Pediatrics at Federal University of São Paulo, Escola Paulista de Medicina.

The protocol developed by the Cooperative Brazilian Group of Bone Tumor Treatment (GBCTTO) was restructured based on the results obtained in a previously conducted study, named study IV5. The protocol intended to increase the survival rate of patients, reducing toxicity caused by the drugs administered in chemotherapy. Currently, the treatment of osteosarcoma patients with cisplatin is conducted before and after the surgical procedure.

Among the eligibility criteria of patients with osteosarcoma to this protocol we can point out the tumor with high degree of malignancy, which was confirmed in the biopsy and primary location.
We developed two osteosarcoma protocols: metastatic and non-metastatic tumors, with the same therapeutic basis. In the protocol of metastatic osteosarcoma there was administration of two cycles of 4m/m2 cyclophosphamide (CTX). After CTX administration, the two protocols were equal in all aspects. CTX is not considered an ototoxic drug, thus the subjects in both groups were gathered for the present study.

The protocol basis considered eight chemotherapy cycles with 21 day intervals among them. Among the 8 cycles, four were cisplatin with dosage of 120 mg/m2/cycle which was divided into 2 days, that is, 60mg/m2 per day. The total dosage of cisplatin received at the end of the treatment was 480mg/m2 in all patients. The other four cycles were of iphosphamine, which is not considered an ototoxic drug.

In the period of August 1999 to April 2001 we assessed 32 patients. We excluded the following: one patient with parietal bone tumor, since the location could compromise the inner ear structures, and those that did not present all evaluations, that is, the patients which were not submitted to all audiological assessments recommended by the protocol.

Thus, we assessed only 13 subjects that presented all its assessment whose ages varied from 7 to 20 years, mean age of 13.3 years, being six female and seven male subjects.

The patients were submitted to audiological assessment comprising pure tone audiometry (250 to 18000 Hz), transient otoacoustic emissions (TOAE) and distortion product otoacoustic emissions (DPOAE).
The audiological assessment was conducted before the beginning of the protocol, after each cycle of cisplatin and at the end of the protocol; therefore, each subject was assessed in five different situations.

The thresholds collected in the conventional pure tone audiometry in dB HL (dB hearing level) were converted into dB SPL according to the values determined by ANSI-196912, so that in this study the thresholds in all frequencies presented the same reference. The results of the audibility thresholds investigated in the right and left ears were not grouped, since there are references that cisplatin-induced hearing loss is bilateral and symmetrical13, 14.

In the assessment of results we grouped the audibility thresholds by frequency obtained in the audiometry of each assessment. Based on this group, we calculated the arithmetic mean of the audibility thresholds by frequency of each assessment. We compared the mean thresholds between the assessment using Friedman's test. When there was a statistically significant difference we employed Wilcoxon's test to determine if there were audibility thresholds alterations based on dosage use.

TOAE and DPOAE were assessed by the comparison of mean amplitude of signal/noise ratio by frequency. For such study, we used the Variance Analysis Model in Block. In the TOAE study we analyzed the significance of the absence of response as treatment evolved using Proportion Tendency Test.

All statistical tests employed in the study considered a descriptive level of 5%.

RESULTS

Based on the study of audibility thresholds (dB SPL) obtained from frequencies of 250 to 18,000Hz after administration of each cisplatin cycle (120, 240, 360 and 480 mg/m2) we calculated the mean thresholds and studied the possible differences between them, according to the cumulative dose of drug, using Friedman's test. When we detected significant differences between thresholds, we employed Wilcoxon's test to identify in which cycle the change took place (Graph 1).

As to individual values, all patients presented change in high frequency thresholds as of 9kHz. In addition, hearing loss (>25 dB HL) in the following frequencies of the conventional audiometry were detected (250 to 8000Hz):

 3 kHz in two subjects (15.3%);
 4 kHz in two subjects (15.3%);
 6 kHz in two subjects (15.3%);
 8 kHz in two subjects (15.3%).

Therefore, 100% of the patients presented high frequency hearing loss and in 61.2% of the cases there was hearing loss detected in frequencies assessed by conventional pure tone audiometry.

Based on the study of signal/noise ratio (dB) obtained in frequencies 1, 2, 3, and 4 and global response of TOAE in each patient after each cycle of cisplatin, we calculated the mean values. We applied the Variance Analysis Model in Block to study whether there was change in the signal-to-noise ratio with cumulative dose of cisplatin (Graph 2). In Graphs 2 and 3 we can see the mean values of the signal/noise ratio of TOAE by frequency band and overall response, respectively.

Next, we investigated the presence and absence of TOAE in each one of the assessments conducted. To that end, we applied the Proportion Tendency Test both for the overall response and for the study by frequency 1, 2, 3, and 4 kHz (Table 1).

We calculated mean amplitudes of otoacoustic emissions by distortion product (DPOAE) by frequency (dB) after each cycle of cisplatin. To check whether there were changes in amplitude owing to the cumulative effect of cisplatin, we used the Variance Analysis Model in Block. In frequencies in which we observed statistically significant differences a new statistical study was made, expressing the values in descriptive levels by cisplatin dose (Graph 4).

Graph 1. Mean audibility thresholds (dB SPL) by frequency (Hz) by cisplatin cycle.



Graph 2. Mean signal/noise ratio (dB) of TEOAE by frequency band and cycle of cisplatin.



Graph 3. mean values of the signal/noise ratio (dB) of overall response of TEOAE in cisplatin cycles (p=0.864).


Table 1. Presence and absence of response of TOAE in overall response and by each frequency band by cisplatin cycles.




Graph 4. Mean amplitude responses of DPOAE by frequency of cisplatin cycle.

DISCUSSION

The analysis of auditory thresholds in frequencies of 250 to 18,000Hz revealed changes right after administration of the first cisplatin cycle. The modification occurred after the frequency of 3kHz. However, we detected hearing loss only after 8kHz, in which we had a mean threshold value of 40.76 dB SPL that corresponded to 27.76 dB HL. The loss occurred after a cumulative dose of 480 mg/m2 (Graph 1).
In such cases, there was predominance of high frequency impairment, with progression to affection according to cumulative dose, with no affection of medium and low frequencies.

The statistical analysis revealed statistically significant differences between mean audibility thresholds by cisplatin cycle in frequencies of 500Hz and 8000 to 18,000 Hz. In the frequency of 500Hz, the audibility thresholds were significantly better after the last dose of cisplatin (480 mg/m2). It can be explained by the fact that the patients were more used to being submitted to the test, therefore, they were more concentrated on the stimuli and improved the response.

In the frequencies of 8 to 18kHz, there was worsening of audibility thresholds. The statistically significant change occurred after the first dose of cisplatin in the frequencies of 13 to 18kHz. In the frequencies of 8 to 12 kHz, this change occurred after the dose of 240 mg/m2.

Therefore, high frequency audiometry detected ototoxicity of cisplatin in the first drug administration cycle whereas in conventional audiometry the thresholds affection was detected after the dose of 240mg/m2.

Strauss et al.15 detected hearing loss in 25% of the patients exposed to cisplatin, being that hearing loss was concentrated in the range of 4 to 8 kHz. In the study by Skinner et al.16, 35% of the subjects presented 2000Hz hearing loss. In the studies by Fausti et al.3 and Park17 there was predominance of high frequency abnormalities.

As to individual values, we observed that 61.2% of them presented hearing loss in the conventional frequency range, being that 30.6% of the patients presented hearing loss in frequencies that are important to speech understanding (3 and 4 kHz). Dishtchekenian et al.5 found similar results, in which they detected hearing loss in 66.6% of the patients in the conventional frequency range; the chemotherapy protocol that was used counted on carboplatin (also considered an ototoxic drug), in addition to cisplatin, in non-fraction doses. As to degree of loss, in mean values, Dishtchekenian et al.5 observed moderately severe loss in conventional audiometry. Conversely, our study showed mean mild to moderate hearing loss. This difference concerning affection of frequencies can be due to drug fraction or exclusive use of cisplatin, without the association with carboplatin.

Dreschler et al.7 observed hearing affections in 80% of the cases, especially in high frequencies and pointed out that the frequencies of 12 and 14 kHz are the most important ones for early detection of cisplatin ototoxicity. Fausti et al.18 suggested the frequencies of 8, 9, 10, 11.2 and 12.5 kHz for hearing monitoring of cisplatin-exposed patients and in this case, early detection was made in 94% of the cases. These data are in accordance with our study, in which ototoxicity was detected in high frequencies prior to conventional audiometry affection. Moreover, the performance of frequencies 13 to 18kHz was the same, meaning that some of these frequencies could be excluded from the test, making it quicker and equally effective.

As to transient otoacoustic emissions (TOAE), there was a slight increase in mean signal/noise ratio of cisplatin in overall response and in frequencies of 1 and 2 kHz after a 120mg/m2 dose (Graphs 2 and 3).

This increase in mean value of TOAE is in agreement with the findings by Cevette et al.19, in which the authors described the increase in amplitude of otoacoustic emissions after a small dose of cisplatin owing to chemical changes (calcium and magnesium metabolism) that cisplatin causes in hair cells. The cell suffers an intercellular calcium increase owing to the deficit of its antagonist, magnesium, which increases the permeability of the cytoplasm membrane. The cilia mobility of outer hair cells depends on intracellular calcium. Therefore, in low doses of cisplatin there can be an increase in otoacoustic emissions, which can be an indication of damage and eventually cell death.

As to mean signal/noise ratio, the assessments after 240, 360 and 480 mg/m2 were lower than the assessment after 120 mg/m2. Thus, responses tend to increase after a small dose of cisplatin and decrease after a dose of 240 mg/m2. This behavior occurred as a global response and in frequencies of 1, 2 and 3 kHz. In frequencies of 4kHz, there was a progressive reduction of the signal/noise ratio according to the cumulative dose of cisplatin (Graph 2).

In the statistical analysis, there was no statistically significant difference between the overall response to TEOAE and frequency obtained in each assessment.

As to presence and absence of responses during treatment, we noticed that the overall response to TEOAE and frequencies of 1 and 2 kHz continued to be present throughout treatment. In frequencies of 3 and 4 kHz, there was progressive increase of absent responses during treatment, and only the frequency of 4kHz presented statistically significant difference concerning absence of response (Table 1).

Allen et al.20 observed in their study that TEOAE presented correlation with data observed in audiometry. The same was observed by Bensadon14 and Durrant, Furman, Carr21. In this study, the data were observed only in the frequency of 4kHz. Biró et al.22 did not observe changes in TEOAE in subjects that received low doses of cisplatin.

We noticed a slight decrease in mean amplitudes of distortion product otoacoustic emissions (DPOAE) in almost all frequencies after the dose of 120 mg/m2. This condition was maintained in frequencies of 1001, 1257, 1587 and 2002 Hz in subsequent assessments. In the frequency of 2515Hz the response was stable in all assessments. In the frequencies of 3174, 4004, 5042 and 6348 Hz, the highest in the test, mean amplitudes tended to decrease progressively according to the cumulative dose of drug (Graph 4).

The statistical analysis revealed statistically significant difference between mean amplitude of DPOAE in frequencies of 3174, 4004, 5042 and 6348 Hz. However, in the comparison of initial mean amplitude with the remaining ones, no statistically significant difference was observed in any of the analysis conducted. It is possible, however, to observe that there is a tendency to reduce the descriptive level according to the drug cumulative dose (Graph 4).

In the study by Ress et al.23 54% of the patients presented abnormal high frequency responses and 91% of them presented DPOAE abnormalities.

DPOAE did not detect early cisplatin ototoxicity, but they followed the modification found in the audiometry. These data are in accordance with the reports by Bensadon14 and Berg et al.24.

All the data about otoacoustic emissions make us believe that they can be used as an additional instrument to monitor the hearing of patients exposed to cisplatin but not as the test of choice. However, in cases in which patients can not be assessed by audiometry, it is the most effective way of monitoring hearing.

To sum up, we noticed that high frequency audiometry is a reliable test to show the progression of hearing affection owing to cisplatin ototoxicity, since the abnormalities start from high frequencies and progress to low frequencies according to the drug cumulative dose, enabling, if possible, the physician to make adjustment to the drug and counseling the family and/or patient in case it is not possible to male modifications to the chemotherapy protocol.

Otoacoustic emissions, in turn, both transient and distortion product, showed cellular abnormalities by increased and decreased amplitude, however, these data do not allow prediction of the resulting hearing loss.

CONCLUSIONS

1. Audibility thresholds are reduced right after the infusion of 120 mg/m2 of cisplatin as of frequency 8kHz;
2. Amplitude of TEOAE and DPOAE increase and decrease during the chemotherapy process as of the infusion of 120 mg/m2 of cisplatin;
3. Overall response and amplitude of frequencies 1, 2 and 3 kHz of TEOAE tend to be present until the end of treatment, except for frequency of 4kHz;
4. High frequency audiometry is the most effective method to early detect the ototoxicity of cisplatin;
5. All patients presented at the end of treatment, high frequency hearing loss but only 30.6% presented hearing loss (3 and 4 kHz) in some frequencies that were more important for speech understanding.

REFERENCES

1. Schweitzer VG. Ototoxicity of Chemotherapic Agents. Otolaryngol Clin North Am 1993;26 (5):759-789.
2. Ruiz L, Gilden J, Jaffe N, Robertson R & Wang YM. Auditory Function in Pediatric Ostesarcoma Patients Treated with Multiple Doses of cis-Diamminedichloroplatinum (II). Cancer Res 1989;49:742-44.
3. Fausti SA, Henry JA, Schaeffer HI, Olson DJ, Frey RH & Bagby GC. High-Frequency Monitoring for Early Detection of Cisplatin Ototoxicity. Arch Otolaryngol Head Neck Surg 1993;119:661-666.
4. Pasic TR & Dobie RA. Cis-Platinum Ototoxicity in Children. Laryngoscope 1991;101:985-991.
5. Dishtchekenian A, Iório MCM, Petrilli AS & Azevedo MF. Monitorização Auditiva na Ototoxicidade. In: Barros APB, Arakawa L, Tonini MD & Carvalho VA. Fonoaudiologia em Cancerologia. Fundação Oncocentro de São Paulo - Comitê de Fonoaudiologia em Cancerologia; 2000. p. 260-269.
6. Iório MCM, Dishtchekenian A, Garcia AP& Petrilli AS. Ototoxicity of Cisplatin in Osteosarcoma Patients. Anais do 12nd American Academy of Audiology Annual Convention. Chicago; 2000. p. 146.
7. Dreschler WA, VD Hulst RJAM, Tange RA & Urbanus NAM. Role of High-Frequency Audiometry in Early Detection of Ototoxicity. Audiology 1989;28:211-220.
8. Laurell G & Jungnelius U. High-Dose Cisplatin Treatment: Hearing Loss and Plasma Concentrations. Laryngoscope 1990;100:724-734.
9. Brock PR, Bellman SC, Yeomans EC, Pinkerton CR & Pritchard J. Cisplatin Ototoxicity in Children: a Practical Grading System. Med Pediatr Oncol 1991;19:295-300.
10. Link MP & Eilber F. Osteosarcoma. In: Pizzo, P.A. & Poplack, D.G. Principles and Practice of Pediatric Oncology. 3rd edition. Philadelphia: Lippincott Raven Publishers; 1997. p. 889-920.
11. Macedo CR & Petrilli AS. Tumores Ósseos Malignos na Criança e no Adolescente. Pediatr Mod 1999;35 (8):600-608.
12. American National Standard Institute - American National Standards Specification for Audiometers, American Standard, S3.6, 1969.
13. Kopelman J, Budnick AS, Sessions RB, Kramer MB & Wong GY. Ototoxicity of High-Dose Cisplatin by Bolus Administration in Patients with Advanced Cancers and Normal Hearing. Laryngoscope 1988;98:858-863.
14. Bensadon RL. Estudo Clínico e Audiológico da Ototoxicidade da Cisplatina. Comparação entre os Exames de Audiometria Tonal Convencional, Audiometria de Altas Freqüências e Emissões Otoacústicas. [Tese Doutorado]. Faculdade de Medicina. Universidade de São Paulo, 1998.
15. Strauss M, Towfighi J, Lord S, Lipton A, Harvey HA & Brown B. Cis-platinum Ototoxicity: Clinical Experience and Temporal Bone Histopathology. Laryngoscope 1983;93:1554-1559.
16. Skinner B, Pearson ADJ, Amineddine DB, Mathias DB & Craft AW. Ototocixity of Cisplatinum in Children and Adolescents. Br J Cancer 1990;61:927-31.
17. Park KR. The Utility of Acoustic Reflex Threshold and Other Conventional Audiologic Tests for Monitoring Cisplatin Ototoxicity in the Pediatric Population. Ear Hear 1996;17:107-115.
18. Fausti SA, Henry JA, Helt WE, Phillips DS, Frey RH, Noffsinger D, Larson VD & Fowler CG. An Individualized, Sensitive Frequency Range for Early Detection of Ototoxicity. Ear Hear 1999;20:497-505.
19. Cevette M J, Drew D, Webb T & Marion MS. Cisplatina Ototoxicity, Increased DPOAE Amplitudes, and Magnesium Deficiency. J Am Acad Audiol 2000;11:323-329.
20. Allen GC, Tiu C, Koike K, Ritchey K, Kurs-Lasky M & Wax MK. Transient-Evoked Otoacoustic Emissions in Children After Cisplatin Chemotherapy. Otolaryngol Head Neck Surg 1998;118 (5):584-588.
21. Durrant JD, Furman JM & Carr B. DPOAE in Monitoring of Cisplatin Ototoxicity, 1997. Disponível em: URL:HTTP:// www.ausp.memphis.edu/ierasg/abstract/durrant.txt.
22. Biró K, Baki M, Büki B, Noszek L & Jókúti L. Detection of Early Ototoxic Effect in Testicular Cancer Patients Treated with Cisplatin by Transiently Evoked Otoacoustic Emission: a Pilot Study. Oncology 1997;54:387-399.
23. Ress BD, Sridhar KS, Balkany TJ, Waxman GM, Stagner BB & Lonsbury-Martin BL. Effects of cis-platinum Chemotherapy on Otoacustic Emissions: The Development of na Objective Screening Protocol. Otolaryngol Head Neck Surg 1999;121:693-701.
24. Berg AL, Spitzer JB & Garvin JH. Ototoxicity Impact of Cisplatin in Pediatric Oncology Patients. Laryngoscope 1999;109:1806-1814.
25. Baum ES, Gaynon P, Greenberg L, Krivit W & Hammond D. Phase II Study of cis-Dichlorodiammineplatinum (II) in Childhood Osteosarcoma: Children's Cancer Study Group Report. Cancer Treat Rep 1979;63 (9-10):1621-1627.

---

1 Speech therapist and audiologist, Master in Human Communication Disorders, Federal University of São Paulo - Escola Paulista de Medicina.
2 Joint Professor, Discipline of Auditory Disorders, Department of Otorhinolaryngology and Human Communication Disorders,
Federal University of São Paulo - Escola Paulista de Medicina.
3 Joint Professor, Department of Pediatrics, General Director of Institute of Pediatric Oncology/ GRAAC,
Federal University of São Paulo - Escola Paulista de Medicina.
Address correspondence to: Adriana Pontin Garcia - R. Aníbal dos Anjos Carvalho, 42 Interlagos
Tel (55 11)5666-2216 - E-mail: apgarcia@osite.com.br
Article submitted on July 26, 2002. Article accepted on March 27, 2003.

Print: