N-Acetylcysteine combined with prednisolone treatment shows better hearing outcome than treatment with prednisolone alone for patients with idiopathic sudden sensorineural hearing loss: a retrospective observational study

Objectives Internationally, corticosteroids are still the mainstay treatment for patients with idiopathic sudden sensorineural hearing loss (ISSHL). This is a retrospective monocentric study investing the impact of adding N-acetylcysteine (NAC) to prednisolone treatment on patients with ISSHL at a tertiary university otorhinolaryngology department. Methods 793 patients (median age 60 years; 50.9% women) with a new diagnosis of ISSHL from 2009 to 2015 were included in the study. 663 patients received NAC administration in addition to standard tapered prednisolone treatment. Univariate and multivariable analysis were performed to identify independent factors regarding negative prognosis of hearing recovery. Results Mean initial ISSHL and hearing gain after treatment in 10-tone pure tone audiometry (PTA) were 54.8 ± 34.5 dB and 15.2 ± 21.2 dB, respectively. In univariate analysis, treatment with prednisolone and NAC was associated with a positive prognosis of hearing recovery in the Japan classification in 10-tone PTA. In multivariable analysis on Japan classification in 10-tone PTA including all significant factors from univariate analysis, negative prognosis of hearing recovery were age > median (odds ratio [OR] 1.648; 95% confidence interval [CI] 1.139–2.385; p = 0.008), diseased opposite ear (OR 3.049; CI 2.157–4.310; p < 0.001), pantonal ISSHL (OR 1.891; CI 1.309–2.732; p = 0.001) and prednisolone alone without NAC treatment (OR 1.862; CI 1.200–2.887; p = 0.005). Conclusions Prednisolone treatment combined with NAC resulted in better hearing outcomes in patients with ISSHL than treatment without NAC.


Introduction
An idiopathic sudden sensorineural hearing loss (ISSHL) is a sudden onset, usually unilateral, cochlear sensorineural hearing loss of ≥ 30 dB within < 3 days in at least 3 contiguous frequencies without an identifiable cause.The incidence is estimated to be between 8-400/100,000 cases [1][2][3][4].Because of the unexplained cause, many therapies have been tried.Antivirals, thrombolytics, vasodilators, and rheologics seem to have no effect [5].Internationally, there is no standard treatment for patients with ISSHL but current therapeutic approaches are mainly focused on different forms of application of corticosteroids.If there is no improvement in hearing after systemic corticosteroid therapy, local intratympanic application may be used.Recent studies are mainly concerned with the combination of local and systemic corticosteroid administration in first-line therapy.The aim of this study was to evaluate the adding administration of N-acetylcysteine (NAC) to prednisolone treatment on patients with ISSHL at a tertiary university otorhinolaryngology department.
NAC has several effects that are thought to be beneficial to cell stress in the inner ear [6].Oxygenated radicals can damage hair cells in the inner ear by activating apoptotic cell death programs.NAC acts as a free radical scavenger and can decrease the cell's nitric oxide production by increasing the synthesis of reduced glutathione [7], thus decreasing the production of harmful nitrogen radicals [8].In addition, NAC can prevent cell apoptosis as a donor of reduced glutathione [6].In contrast to treatment for ISSHL, NAC is already a component of treatment for acute hearing loss of other etiologies.One application of NAC is hearing loss caused by aminoglycosides, which are used in tuberculosis treatment.Kranzer et al. reported of the convincing otoprotective effect of NAC in preventing aminoglycoside induced ototoxicity while tuberculosis treatment [9].In literature, some studies exist on the therapeutic outcome of NAC in combination with steroid treatment in ISSHL, but the data is limited and inconclusive [10][11][12][13].
For this purpose, the impact of administration of NAC to prednisolone treatment of 793 patients with ISSHL who were hospitalized at a department of otorhinolaryngology in a tertiary university center in the period from 2009 to 2015 were analyzed.

Ethical considerations
This retrospective study was approved by the Ethics Committee of the BLINDED (IRB No. 4755-0416).The Ethics Committee waived the requirement for informed consent of the patients because the study had a non-interventional retrospective design and all data were analyzed anonymously.

Patients
For this purpose, 920 patients were screened which were treated in the Department of Otorhinolaryngology, BLINDED, Germany, from September 2009 to December 2015.The patients were all coded according to the International Statistical Classification of Diseases and Related Health Problems, 10th revision (ICD-10) with the number H91.2 (ISSHL including acute hearing loss without further specification) [14].The exclusion criteria were: Varicellazoster virus infection, Herpes simplex virus 1/2 infection, toxic inner ear or acute otitis media, only one or no audiogram available, strong suspicion of aggravation, sarcoidosis, discontinuation of treatment, acute exacerbation of chronic otitis media, vestibular schwannoma, acute sepsis, squamous cell carcinoma of the mastoid.Finally, a total of 793 cases were included in this study.

Treatment
Standard according to the German ISSHL guideline was a tapered corticosteroid treatment: 250 mg prednisolone intravenous once daily for the first 3 days, 100 mg intravenously on the 4th day, 75 mg orally on the 5th, 50 mg orally on the 6th, and 20 mg orally on the 7th day.During the inclusion period time from 2009 to 2015 there were various treatment combinations of prednisolone according to the valid version of the clinical guideline for ISSHL treatment: hydroxyethyl starch (HAES), acetazolamide, mannitol, or pentoxifylline.Acetazolamide 500 mg was administered orally as a short infusion once daily for 7 days.Over the long inclusion period, the treatment regimen changed partially for patients with ISSHL.Hydroxyethyl starch (HAES) 6% 250 ml was administered intravenously once daily for 7 days until 2013.Mannitol 15% 250 ml was used intravenously once daily for 3 days and pentoxifylline 300 mg was administered intravenously once daily for 7 days until 2009.The experimental additional NAC treatment was given orally 600 mg two times per day for 7 days.In the absence of improvement or worsening after therapy within 4 weeks, salvage surgery in the form of an intratympanic dexamethasone therapy was offered to the patient.After tympanotomy, 4 mg dexamethasone in GELASPON ® (HEYL, Berlin, Germany) was placed into the round window niche at the latest 2 weeks later.
For diseased opposite ear classification, the mean values were calculated from the audiograms of the opposite ear.If the mean was 20 dB ISSHL or more, the ear was classified as diseased.For the calculation of the HG, the cases, each for the 6-tone PTA and the 10-tone PTA, were divided dichotomously about the median of the absolute HG.For evaluation of the outcome, the hearing improvement in Siegel classification and Japan classification was calculated additionally [16,17].For the calculations on the influence of ISSHL type, the classes were divided into ISSHL with low-frequency involvement (low frequency ISSHL and low/mid-frequency ISSHL), ISSHL with highfrequency involvement (high frequency ISSHL and high/ mid-frequency ISSHL), pantonal ISSHL, and deafness [18].The median (4 days) was chosen as the time interval for the calculations of the interval from ISSHL event to inpatient treatment initiation for the binary calculations.Shifting the interval to 2, 6, 8, and 14 days did not change the results.

Statistical analysis
Descriptive analyses were performed using SPSS ® Statistics (IBM SPSS Statistics for Windows, Version 23, Armonk, NY, USA).Biometric, anamnestic, audiometric and therapeutic data were collected in a standardized way and selected parameters were dichotomized in a SPSS database.Significance tests were performed using the chi-square test or Fisher's exact test.Subsequently, selected parameters were examined with regard to their influence on hearing recovery in a univariate analysis.Significant factors from the univariate analysis were included into multivariable regression models to identify independent risk factors for HG, respectively.Both 6-tone PTA (pure tone audiometry) and 10-tone PTA were used to evaluate the hearing findings.Absolute HG, Siegel and Japan classification were used as criteria for evaluation of recovery.Some parameters showed a very strong association with the results in the univariate analyses, the multivariable analyses were repeated again in a further modulation, excluding very strong influencing factors with p < 0.001, in order to identify other independent influencing factors.The significance level of p = 0.05 was set.

Patient's characteristics, treatment characteristics and hearing characteristics
The distribution of patient characteristics, treatment characteristics, and hearing characteristics is shown in Tables 1  and 2. The median age at diagnosis was 60 years and half of patients were women (404 women, 50.9%).Most patients had a pantonal ISSHL (24.8%), followed by low frequency ISSHL (21.1%) and deafness (18%).Nearly all patients received prednisolone (97.7%), in various combinations.A combination with NAC was given to 83.6% of the patients.Slightly more than half of patients started the treatment < 4 days after onset (54.4%), and slightly less than half of the patients had a pre-existing diseased opposite ear.The majority of patients was treated additionally with NAC (83.6%).Mean initial ISSHL in 6-tone PTA and 10-tone PTA was 53.8 ± 34.9 dB and 54.8 ± 34.5 dB, respectively.Mean HG in 6-tone PTA and 10-tone PTA after treatment was 15.5 ± 21.7 dB and 15.2 ± 21.2 dB, respectively.According to the Japanese classification, most of the patients were assigned to type IV (39.8%).One third of the patients

Univariable analysis
The results of univariate analyses are shown in Table 3.
Patients with ISSHL treated with NAC in addition to prednisolone were close to those without NAC administration regarding the median of absolute HG.The tendency for NAC treatment to perform better, but not significantly, continued in the Siegel classification (frequency independent) and in the 6-tone PTA of the Japan classification.Looking at the Japan classification in 10-tone PTA, NAC treatment was a significant factor of hearing recovery (p = 0.027).No combination of prednisolone with another drugs than NAC had no significant influence on the hearing recovery (all p > 0.05).Japan classification I/II was assigned to 46.5% (308) of patients treated with NAC, whereas only 26.9% (35) of patients who did not receive NAC treatment were assigned as Japan classification I/II.Age, comorbidities, diseased opposite ear and pantonal ISSHL (in Japan classification) were very strong influencing factors (all p < 0.001).Permanent diseases such as hypertension, diabetes mellitus, coronary artery disease (CAD), or vascular risk were significant factors regarding a negative prognosis of hearing recovery in Siegel and Japan classification (all p < 0.05).

Multivariable analysis
In multivariable logistic regression analysis, all variables which were significant in univariate analysis besides NAC treatment were included (Table 4).The multivariable analyses were repeated again in a further modulation as model 2 (Table 5) and model 3 (

Discussion
The effect on NAC treatment on patients with ISSHL has been rarely investigated worldwide.Current therapeutic approaches on ISSHL are mainly focused on different forms of application of corticosteroids.The aim of this retrospective study was to evaluate the adding administration of NAC to prednisolone treatment on patients with ISSHL.NAC combined to prednisolone treatment was associated with improved hearing outcome on patients with ISSHL according to Japan classification.In multivariable analysis treatment without NAC had an increased odds ratio than with NAC treatment for Japan classification in 10-tone PTA.
In literature, some studies exist on the therapeutic outcome of NAC in combination with steroid treatment in ISSHL, but the data is limited and inconclusive.Kranzer et al. reported in a review of the convincing otoprotective effect of ACC when used with aminoglycosides [9].A similar conclusion was reached by Kocygit et al., who investigated the otoprotective effect of NAC during administration of ototoxic amikacin in dialysis-associated peritonitis [19].Kocygit et al. concluded that NAC mainly protects the higher frequency range.The effect of NAC was significant from the fourth week onward.The otoprotective effect of NAC is also being tested in noise-induced hearing loss by an Italian study.Lorito et al. exposed in their study rats to defined noise.A dose-dependent protection of the cochlea by NAC treatment was found.The rats that received high doses of NAC were better protected [20].Lin et al. investigated the effect of NAC treatment in noise-induced temporary shift on male workers.The authors concluded that the administration of 1200 mg NAC resulted in a significantly reduced "temporary threshold shift" [13].However, in a randomized, prospective, double-blind, placebo-controlled study from 2015, no benefit of NAC in noice-induced hearing loss was found.For this, soldiers with noice-induced hearing loss were divided into an NAC group (2700 mg per day, starting before a shooting exercise) and a placebo group.After the exercise, their hearing was assessed.In contrast to the post-hoc analysis, there was no advantage for the NAC group when the study was evaluated [12].Chen et al. investigated the effect of NAC on hearing loss from sudden deafness confined to the inner ear [21].For this purpose, 35 patients with sudden deafness of unclear origin were treated with NAC 600 mg two times per day for two days and were then discharged with a 3-month consecutive medication, while the control group received a combination treatment of corticosteroid (1 mg/ kg), dextran and ginkgo.The group treated with NAC had a significantly greater mean hearing gain than the comparison group with combination treatment (NAC treatment: 43 ± 27 dB vs. combination treatment: 21 ± 28 dB) [21].Angeli et al. also reported an improvement in hearing recovery of patients with ISSHL with the addition of oral NAC to corticosteroid treatment compared to single therapy without NAC.NAC treatment was given at a dose orally 1200 mg three times daily for two weeks.After 6 months, the NAC group showed an average improvement of 26.1 dB in pure-tone threshold at 500-400 Hz compared to 15.1 dB in single therapy group [22].In addition, Bai et al. investigated the efficacy of a combination treatment of oral NAC and intratympanic dexamethasone in patients with ISSHL.NAC treatment was given orally 600 mg two times daily for two weeks.There was no improvement in average hearing gain in pure tone audiometry, but a significant hearing gain at 8000 Hz in the NAC group was evident [10].Chen et al. also reported a significant improvement at 8000 Hz between the NAC group and the non-NAC group.NAC treatment was given orally 600 mg two times daily for at least 1 month.The NAC results were better than the non-NAC group in mean hearing level gain, speech reception threshold gain and speech discrimination score gain, but these differences were not significant [11].
However, the effect of NAC treatment for patients with ISSHL and even with noise exposure is still ambiguous in literature.Lin et al. [14] reported a significant improvement with NAC, while Kopke et al. [9] found no difference.Chen et al. concluded that a greater hearing gain can be achieved with additional NAC administration [21].In addition Bai et al. and Chen et al. reported of significant improvement at 8000 Hz of a NAC treatment, which is consistent with our findings of improvement in hearing recovery with NAC treatment.But to our knowledge a direct comparison with our study is difficult due to difference of treatment, treatment duration and different classification of hearing recovery.
The present study has due to his retrospective design some limitations.The retrospective design cannot guarantee sufficient information and standardized treatment decision.Causal connections are only traceable to a limited extent.The results from our study showed that the addition of NAC has an impact on the hearing recovery for patients with ISSHL For a better understanding of the role of NAC in treatment of ISSHL, clinical studies for the in a prospective design are needed to provide an adequate evidence.

Conclusions
This retrospective monocentric study investing the effect of adding NAC to prednisolone treatment on 793 patients with idiopathic sudden sensorineural hearing loss (ISSHL) according to absolute hearing gain, Siegel and Japan classification between 2009 and 2015.In summary, significant factors regarding a negative prognosis of hearing recovery were higher age, diseased opposite (> 20 dB), pantonal ISSHL and prednisolone treatment without additional NAC application.Treatment without addition of NAC had an increased odds ratio than with NAC for Japan classification in 10-tone pure tone audiometry.The results from our study showed that NAC has an important impact on hearing recovery on patients with ISSHL.However, the results of the positive effects of NAC on hearing recovery need to be verified by further analyses in a prospective study.

Table 1
Patients' characteristics

Table 6
) after exclusion of excluding very strong influencing factors like diseased opposite ear and age (both p < 0.001).The odds ratio (OR) and

Table 2
Treatment characteristics and hearing characteristics

Table 3
Univariate analyses of association of patient's characteristics, treatment characteristics and hearing characteristics on absolute hearing gain, and recovery due to Siegel and Japan clas-

Table 4
Multivariable analyses on hearing outcome according to Japan classification (I/II vs. III/IV) model 1 for worse outcome

Table 6
Multivariable analyses on hearing outcome according to Japan classification (I/II vs. III/IV) model 3 for worse outcome Significant p-values (p < 0.05) in bold OR odds ratio, CI confidence interval, NAC N-acetylcysteineneed to obtain permission directly from the copyright holder.To view a copy of this licence, visit http:// creat iveco mmons.org/ licen ses/ by/4.0/.
Significant p-values (p < 0.05) in bold OR odds ratio, CI confidence interval, NAC N-acetylcysteine