Study of the effect of combined interferon and ribavirin therapy on the hearing profile of hepatitis C virus patients

Objectives This study aimed to study the effect of combined pegylated interferon and ribavirin (Peg-IFN/RBV) therapy on the hearing profile of patients with hepatitis C virus (HCV). Materials and methods The auditory system of a total of 74 chronic HCV patients was assessed using pure tone audiometry (conventional and high frequency) and distortion product otoacoustic emissions (DPOAEs in the form of a DP-gram) immediately before therapy and at the end of the 12th and 24th weeks. Vestibulocochlear adverse effects including hearing loss, tinnitus, vertigo, and otalgia were also considered. Results Significant elevations in hearing thresholds were found on comparing thresholds at the 12th and 24th weeks with those at the onset of the study. The elevations were mostly at higher frequencies (3000, 4000, 6000, 8000, 9000, 10 000, 11 200, and 12 500 Hz), and did not affect speech perception. For DPOAE, significant differences were observed at all F2 frequencies on comparing both amplitudes and signal to noise ratios at the 12th and 24th weeks with those before therapy, with significance appearing earlier for higher F2 frequencies. Otologic complaints were insignificant, except for tinnitus; a significant increase in tinnitus was observed from 0 to 31.1% by the end of the study. Conclusion Significant auditory adverse effects may result from combined Peg-IFN/RBV therapy in HCV patients, highlighting the importance of prompt monitoring of auditory functions in these patients.


Introduction
Chronic hepatitis C virus (HCV) affects an estimated 300 million individuals worldwide, and in Egypt, the prevalence is estimated to be the highest, somewhere between 15% and 20% of the general population [1]. Treatment of HCV with interferon-(IFN-) 2b and ribavirin (RBV) therapy has shown to be effective for many years now, with 41-47% of patients achieving a sustained virologic response [2][3][4]. With the current pegylated interferons (Peg-IFN) and RBV combination, overall sustained virologic response rates are reported to be 54-56% [5][6][7].
Despite the current popularity of Peg-IFN and RBV therapy, many reported side effects with standard IFN-are also observed with Peg-IFN combination therapy. Flu-like symptoms such as fever, chills, muscle aches, nausea, vomiting, and fatigue are common side effects of treatment. Depression and related symptoms, such as anxiety, irritability, insomnia, and mental confusion, are not rare and may be significant in patients with a history. Common hematologic side effects that require monitoring include neutropenia and thrombocytopenia. Withdrawal rates in IFN-based combination studies because of side effects range from 6 to 7% [4][5][6][7][8].
Little is known about the effects of Peg-IFN on the auditory system. In the literature, a few patients with chronic HCV have been reported to develop sudden hearing loss because of Peg-IFN and RBV combined therapy [9][10][11]. It was suggested that the mechanisms of sudden sensorineural hearing loss in these patients included direct ototoxicity, autoimmunity, and hematological changes [9].
Other reports noted that auditory disability frequently developed in the later stages of IFN therapy for HCV patients in the form of tinnitus and/or hearing loss [12][13][14]. They described audiometry-documented SNHL, which was mostly mild and reversible after discontinuation of therapy [10,[15][16][17].
However, other studies have reported that IFN-RBV therapy does not have any significant association with hearing loss in HCV patients, and could thus be administered safely [18,19].
Many ototoxic drugs exert their ototoxic effect at the highest frequencies first, and as the exposure continues,

Study of the effect of combined interferon and ribavirin therapy
Samir Asal a , Ossama Sobhy a , Omneya Ismail a , Essam Bedewy b This study aimed to study the effect of combined pegylated interferon and ribavirin audiometry (conventional and high frequency) and distortion product otoacoustic emissions (DPOAEs in the form of a DP-gram) immediately before therapy and at the end of the 12th and 24th weeks. Vestibulocochlear adverse effects including hearing loss, tinnitus, vertigo, and otalgia were also considered. and 24th weeks with those at the onset of the study. The elevations were mostly at higher F 2 frequencies on comparing both amplitudes and signal to noise ratios at the 12th and 24th weeks with F 2 frequencies. Otologic thresholds in the frequency range of 500-4000 Hz at octave intervals, as well as measurement of thresholds in the sensitive five frequency ranges, generally separated by 1/6 octave (8, 9, 10, 11.2, and 12.5  Grove Village, IL 60007) probe microphone system. Verification of the adequate quality of the probe fit was carried out using the feedback from the 'check fit' procedure of the software. The sound pressure in the ear canal was picked up by a low-level noise subminiature microphone system ER-10A, also situated inside the probe, amplified and delivered to the computer for analysis (digitizing and Fourier transformation). Patients were instructed to sit comfortably in a chair near the equipment, and they were told that no active contribution from them was required, except to sit quietly, breath regularly but as slowly and smoothly as possible, and to try not to produce any noise in the form of yawning, sneezing, or swallowing. Two pure-tone signals, F 1 and F 2 (F 1 < F 2 ; F 2 /F 1 = 1.2), were presented simultaneously as primary tone frequencies that generate 2F 1 -F 2 DP. DP-gram points were collected at 1/3 octave steps at stimulus levels so that L 1 > L 2 by 10 dB. This level is customary among many centers. The levels of the primaries were 65 dB SPL for L 1 and 55 dB SPL for L 2 . The parameters studied were DPOAE's amplitudes (dBSPL) and signal to noise ratios (SNR) (dBSPL) measured at different frequency bands centered at different values of F 2 (Hz) (i.e. at 1001, 1257, 1587, 2002, 2515. Results were also classified into one of three categories: 'pass' DP-gram, where the DP lower frequencies become involved. As hearing is traditionally only tested up to 8 kHz, most initial cases of hearing loss from ototoxic drugs are never identified by standard audiometric testing [20].
OAEs reflect the functional status of the outer hair cells, which are responsible for the exquisite sensitivity, sharp frequency selectivity, and wide dynamic range of the normal auditory system, and constitute the only noninvasive means of objective cochlear investigation [21]. Few studies in the literature have evaluated the effect of the drugs used in this study on hearing and their effect on cochlear outer hair cells.
Moreover, the Egyptian Ministry of Health guidelines for HCV therapy do not specify screening of hearing by audiometry or monitoring cochlear function in their treatment and follow-up protocol.
Considering the above data, our research aimed to study the effect of dual therapy with IFN and RBV on hearing in chronic HCV patients.
This study was carried out on 74 chronic HCV patients assigned to receive IFN and RBV who were referred to the Audiology unit, Otorhinolaryngology department at the Main University Hospital. We had no age or sex restrictions for our study. All patients were diagnosed to be HCV positive on the basis of positive viral load measurements (IU/ml); HCV-RNA levels were assessed using quantitative PCR and histopathological grading of liver biopsies was a prerequisite for commencement of therapy as well. Patients were required to have no history of chronic middle ear pathology, ear operation, or otoscopic evidence of ear drum abnormalities.
Each patient was subjected to a full assessment of history including history of any otologic complaints in the form of hearing loss, tinnitus, vertigo, and otalgia, which were assessed before starting therapy and at the end of the 12th and 24th week of therapy. All patients also underwent an otoscopic examination. An audiologic evaluation was carried out before starting therapy and at the end of the 12th and 24th week of therapy, respectively. It included the following: (1) Pure tone audiometry (PTA) (using a Madsen Astera audiometer in a sound-treated room, Part of the GN group, Denmark) in the form of standard audiometry, which included air conduction thresholds for the conventional frequencies (0.25, 0.5, 1, 2, 3, 4, 6, 8 kHz) and bone conduction ( Audiometric results ( Table 3) show that of all tested frequencies, a statistically significant difference in hearing thresholds was identified for the frequencies 3000, 4000, 6000, 8000, 9000, and 10 000 Hz in the time period of 0-24 weeks of treatment, with an apparent critical time period of 12-24 weeks. However, at higher frequencies (11 200 and 12 500 Hz), a statistically significant difference in hearing thresholds was observed in the time period 0-24 weeks of treatment, suggesting a critical time period of 0-12 weeks.
In speech audiometry, differences in Speech Reception Threshold (dB) and WDS (%) were statistically insignificant and all patients had excellent WDS throughout the 24 weeks of treatment.
For DPOAE (Tables 4 and 5), tests of significance showed statistically significant differences on comparing both amplitudes and SNRs and in the time period 0-24 weeks of treatment, at F 2 frequencies centered at 1001, 1257, 1587, 2002, and 2515 Hz. For higher F 2 frequencies centered at 3174, 4004, 5042, and 6348 Hz, statistically significant differences were also found throughout the 24 weeks of treatment, being observed as early as 12 weeks of treatment.
Qualitative results for DPOAE (Table 6) show that before the commencement of treatment, all patients (100%) had 'pass' DP-grams. This value showed a decrease throughout the study: 93.2% at 12 weeks and down to 79.9% by the end of the study at 24 weeks. However, at 12 weeks 4.1% of the patients DP-grams were considered as 'partial pass' and 2.7% considered as 'failed', showing an increase at 24 weeks to 12.2 and 8.1%, respectively. A statistically significant period was observed in the time period 0-24 weeks of treatment. levels were above the noise floor at all frequencies by 5 dB and DP levels were more than −10 dB and the DP-gram showed replicability; 'fail' DP-gram, where all data points were embedded in the noise and did not have sufficient SNR to be considered as valid responses; and 'partial pass' DP-gram, where the DP levels were above the noise at some frequencies (four) and below at others [25].
This study was carried out on 74 chronic HCV patients assigned to combined therapy (Peg-IFN + RBV). Demographic characteristics ( Table 1) show that of 74 patients, 45 (60.8%) were men and 29 (39.2%) were women. Patients ranged in age from 20 to 59 years, with the majority (60.7%) in the age group of more than 40-59 years.
Before the commencement of therapy, a detailed assessment of history of any otologic complaints was performed to ensure that all patients fulfilled our prerequisites.
A detailed assessment of a history of otologic complaints was performed throughout the study  The backbone of HCV treatment is Peg-IFN-and RBV combined therapy. Treatment regimen includes a subcutaneous injection of Peg-IFN of 1.5 g/kg once weekly, along with a daily dose of oral RBV ranging from 800 to 1200 mg, depending on the HCV genotype and bodyweight [26]. In general, a treatment duration of 48 weeks is recommended. HCV RNA measurements at weeks 4, 12, and 24 are important for a response-guided treatment approach for Peg-IFN/RBV. If, at these time points, the viral load threshold is exceeded or detected in serum, therapy should be stopped [27,28].
Although the systemic side effects of HCV combined therapy have been well documented, little is known about its effect on the hearing system. One of the proposed side effects of therapy is ototoxicity.
In the current study, we attempted to study the proposed effect through a test battery approach that included assessment of history, speech audiometry, PTA including testing frequencies in the conventional range (250, 500, 1000, 2000, 3000, 4000, 6000, and8000 Hz) and in the extended high-frequency range (9000, 10 000, 11 200, and 12 500 Hz), and DPOAEs in the form of a DP-gram.
During assessment of history, we ensured that all patients fulfilled our selection criteria; none of them (0%) complained of hearing loss, tinnitus, vertigo, or otalgia. By the end of the 12th week of therapy, some patients started to develop tinnitus (23%) and otalgia (8.1%), which, by the end of the 24th week of therapy, increased to 31.1 and 16.2%, respectively. The presence of a statistically significant difference in the number of patients complaining of tinnitus throughout the time period studied suggests that tinnitus might represent the subjective counterpart of ototoxicity. Tinnitus has been reported to be a common side effect of many ototoxic drugs, and there is a common assumption of tinnitus as an early indicator of ototoxicity [29]. The increase in otalgia as a complaint was found to be statistically insignificant, which may indicate that its presence may have been coincidental. An increase in patients' complaints of hearing loss (14.9%) by the end of the 24th week of therapy was found to be statistically insignificant.  Mean ± SD 2.14 ± 5.55 1.88 ± 6.02 1.18 ± 7.05 P 2 <0.001* P 1 , P P 2 , P P Speech audiometry showed statistically insignificant changes that were expected as those complaining of hearing loss showed threshold shifts at frequencies that were not significantly impairing speech perception.
On analyzing audiometric results, therapy was found to show statistically insignificant differences in the frequencies 250-2000 Hz. Significant changes in hearing thresholds started to show at 3000-10 000 Hz on comparing the 24th week thresholds with those at pretreatment as well as on comparing those at the 24th week to thresholds at the 12th week of therapy. However, differences were found to be insignificant on comparing thresholds at the 12th week with those at pretreatment, which signifies a critical time period of 12-24 weeks during which a hearing threshold shift takes place.
At the frequencies of 11 200 and 12 500 Hz, statistically significant changes in thresholds were observed on comparing thresholds at the 12th week with those at pretreatment, and thresholds at the 24th week with those at the 12th week and with those at pretreatment. This highlights the importance of those two frequencies in the early detection of a hearing threshold shift as they were the only frequencies that were affected in the time period of 0-12 weeks.
In the literature, extended high-frequency audiometry was reported to show earlier threshold changes than conventional audiometry and was proposed as a more sensitive screening tool for ototoxicity [30]. Also, changes were reported to appear as early as the second week of therapy [31].
DPOAEs amplitudes and SNRs at F 2 frequencies centered at 1001, 1257, 1587, 2002, and 2515 Hz showed a similar pattern to the audiometric frequencies in the range 3000-10 000 Hz, where the significant critical time period occurred between the 12th and the 24th week. Results for F 2 frequencies centered at 3174, 4004, 5042, and 6348 Hz showed a pattern similar to the last two frequencies in the extended highfrequency audiometry, showing a critical time period of 0-12 weeks of therapy.
Frequency-specific DPOAEs always showed significant changes throughout the course of treatment, unlike the PTA, where lower frequencies reflecting speech (250-2000 Hz) showed no significant change at all. This would lead us to consider frequency-specific DPOAEs to be a more reliable tool in the early detection of the ototoxic effect of HCV combined therapy.
On assessing DPOAEs qualitatively using the criteria of 'pass', 'partial pass,' and 'failed', a statistically significant change in the criteria was found in the therapeutic time period of 0-24 weeks, which indicates that DPOAE, if assessed qualitatively, may be a partially less sensitive tool in the detection of ototoxicity.  In conclusion, this study indicates an ototoxic effect of Peg-IFN/RBV combined therapy on hearing, starting with a perceived complaint of patients with tinnitus, which was correlated with a shift in hearing thresholds mainly at high frequencies, in both PTA and DPOAE.
In the current study, we propose the critical time period of ototoxicity to be in the therapeutic time period of 0-12 weeks, at which the frequencies 11 200 and 12 500 Hz and DPOAE at F 2 frequencies centered at 3174, 4004, 5042, and 6348 Hz have been proven to be the most sensitive in early detection. However, qualitative DPOAE has shown to be less effective in early detection of the ototoxic effect.
Also, we should raise awareness of the importance of tinnitus as a complaint in this proposed critical time period, where it could be used as a simple screening tool.
Although patients' daily life activities were not altered by the ototoxic effect of therapy as speech perception was not affected significantly, we believe that the auditory function of patients receiving Peg-IFN/RBV therapy should be monitored periodically.