Steroids for Idiopathic Sudden Sensorineural Hearing Loss

Abstract

IDIOPATHIC SUDDEN SENSORINEURAL HEARING LOSS, OR sudden deafness, is the acute onset of hearing loss of at least 30 dB in 3 connected frequencies occurring instantaneously or over a period of up to 3 days. Sudden hearing loss is thought to be relatively rare, occurs mainly in patients between the ages of 30 and 60 years, and can be devastating for the patient. Therefore, sudden hearing loss should be treated as a medical emergency. The natural course is characterized by spontaneous recovery among approximately 50% of patients. Accurate estimates of incidence are difficult to obtain because a significant number of patients who spontaneously recover do not seek medical attention. As the term idiopathic implies, the etiology and pathophysiology of hearing loss are unknown. Possible causes include infections (especially viruses), autoimmune disease, circulatory problems, and neurological disease, including multiple sclerosis. Treatment options include oral steroids, intratympanic steroid injections, oral antivirals, vasodilators, diuretics, hyperbaric oxygen therapy, and vitamins. As a result of the high rate of spontaneous resolution, multiple potential causes, and absence of validated tests for etiological identification, the evaluation of treatment effectiveness is difficult. In this issue of JAMA, Rauch and colleagues describe the findings from the first multi-institutional study of oral vs intratympanic corticosteroid therapy for hearing loss. In this study, 2443 adult patients were assessed for eligibility at 16 different academic and community-based otology practices. Of these patients, 250 (10%) agreed to participation and were randomized to receive either 60 mg/d oral prednisone for 14 days with a 5-day taper (n=121) or four 1-mL doses of 40 mg/mL of methylprednisolone injected into the middle ear over a 2-week period (n=129). Audiometry and safety monitoring were performed at screening, and after 1 week, 2 weeks, 2 months, and 6 months of treatment. The primary study hypothesis was that intratympanic methylprednisolone is inferior to oral prednisone for the treatment of hearing loss. The investigators established an average of more than 10 dB as the clinically significant amount of audiometric difference between the 2 study groups. Subgroup analyses of patients based on the amount of hearing loss ( 90-dB or 90-dB pure-tone average [PTA]) were specified a priori. The primary finding of the study was a 2.0-dB difference in PTA between the 2 treatment groups at 2 months with a maximum difference (upper limit of the confidence interval) of 6.6 dB. This value of 6.6 dB was below the required value of 10 dB to demonstrate clinical significance and reject the null hypothesis of noninferiority (ie, intratympanic treatment is worse than oral prednisone). Among the patients who presented with hearing loss of 90 dB or greater and dizziness, came for treatment in fewer than 7 days from onset, or had no prior steroid use, the possibility of a clinically meaningful difference in hearing recovery between the 2 groups could not be ruled out. Overall, the 2 treatments were well tolerated, and the number and severity of serious adverse events and adverse events were similar between groups, as had been anticipated from prior studies. However, of the patients randomized to receive intratympanic steroids, 3.1% had persistent tympanic membrane perforation and 4.7% had persistent otitis media vs 0% and 0.8% in the oral group, respectively. At present, the emergency administration of a tapering dose of systemic steroids after hearing loss onset is widely considered standard treatment. Several uncontrolled studies suggested that intratympanic steroids have equivalent efficacy as oral steroids, while minimizing the potential adverse systemic effects of oral steroids. The practice of prescribing a tapering dose of oral steroids is largely based on the results of 1 randomized clinical study and a retrospective study. In the study by Wilson et al, patients in the steroid group had a recovery rate (61%) that was similar (65%) to untreated patients reported by Mattox and Sim-