Variability in the Management of Idiopathic Sudden Sensorineural Hearing Loss

Subjects and Methods

In January 2010, a multiple-choice survey (see appendix, online at otojournal.org) was mailed nationwide to 500 NONs and 500 ONs selected randomly from the membership of the American Academy of Otolaryngology—Head and Neck Surgery. ONs identified their “subspecialty” as either “otology” or “neurotology.” The survey was approved by the Institutional Review Board (IRB #HM12436). Response rate was 24% (240/1000 received, 120/500 from each group). No follow-up attempts were made to contact nonrespondents.

The 14-item survey addressed physician demographics (state of practice, years in practice, specialty, type of practice), patient demographics (number of patients per year, sources of referral, duration between hearing loss and initial visit), definition of ISSNHL (level of dB loss, number of consecutive frequencies affected, and duration of hearing loss). The survey presented a 2-part case of 2-day unilateral hearing loss and tinnitus. The survey addressed frequency of diagnostic tests used, assessed by Likert scale, and initial treatment strategies written in by respondents (systemic steroids, intratympanic steroids, antivirals, other or combination of each). The survey required specification of medication, dosage/concentration, route, frequency, duration, taper, number of treatments prior to reevaluation, and definition of end point. Respondents were then given a scenario of partial responses and asked if they would continue (and, if so, modify) treatment. Respondents were also asked to comment on their comfort levels in both the diagnosis and treatment of ISSHNL.

Thirty-eight states were represented with New York, California, Texas, and Virginia as the most represented states. Of the respondents, 57.5% (NON 50.8%, ON 64.2%) reported having 0 to 20 years of experience in practice, whereas 42.5% (NON 49.1%, ON 35.8%) reported having greater than 20 years. In addition, 46.3% (NON 3.33%, ON 89.2%; P < .0001) reported fellowship training. Of the respondents, 77.9% (NON 90.8%, ON 65%) characterized their practice as private practice, and 19.2% (NON 5.83%, ON 32.5%) practiced in an academic hospital setting.

Data were compiled in Excel (Microsoft, Bellevue, Washington) and tabulated into 3 groups: NON, ON, and total. Data were analyzed by χ², 2-tailed Fisher exact test, and analysis of variance (ANOVA) techniques using SAS version 9.2 (SAS Institute, Cary, North Carolina).

Results

Initial Diagnosis and Assessment

Participants were asked to define diagnostic parameters for ISSNHL: level of dB loss, consecutive frequencies lost, and duration of hearing loss. Of the total respondents, 60.4% (NON 56.7%, ON 64.2%) characterized a ≤20-dB loss as a sufficient criterion to diagnose ISSNHL ( Figure 1 ), 39.6% (NON 43.3%, ON 35.8%) required a ≥30-dB loss for diagnosis, 68.8% (NON 67.50%, ON 70%) required a ≥3 consecutive frequency loss to define ISSNHL, and 23.8% (NON 21.7%, ON 25.8%) characterized a 2-frequency loss as sufficient to diagnose ISSNHL. For rapidity of hearing loss sufficient to define ISSNHL, a range of responses was observed ( Figure 2 ). Of the respondents, 14.6% (NON 15.0%, ON 14.2%) noted hearing loss occurring over 1 day or less as the maximum duration sufficient for diagnosis, 16.3% reported progressive loss over 2 days as the maximum duration sufficient to diagnose ISSNHL (NON 23.3%, ON 9.17%; P < .004), 52.9% required a progressive loss over 3 or fewer days to diagnose ISSNHL (NON 40.8%, ON 65%; P < .05), and 16.2% believed hearing loss progressing over 7 days would also be considered ISSNHL (NON 20.8%, ON 11.6%).

OPEN IN VIEWER

Figure 1.

Diagnostic criteria: dB hearing loss. NON, nonotologist/neurotologist; ON, otologist/neurotologist.

OPEN IN VIEWER

Figure 2.

Diagnostic criteria: number of days for hearing loss. NON, nonotologist/neurotologist; ON, otologist/neurotologist.

Participants were asked to rate the type of diagnostic tests by the frequency of their use (from 1-5, with 1 being “never used” and 5 being “always used”) during the initial assessment of hearing loss. For initial workup, 100% “always used” pure-tone audiometry and speech perception scores, 75.4% (NON 80%, ON 70.8%) “always used” tympanogram, and 50.4% (NON 34.2%, ON 66.7%; P = .0001) always ordered magnetic resonance imaging (MRI). For those respondents who did not routinely obtain MRI, rates of auditory brainstem response (ABR) use were no different. Other tests that were used less frequently by both groups included complete blood count (CBC), antigen serologic, fluorescent treponemal antibody-absorption (FTA-ABS), basic metabolic panel, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), and antinuclear antibody (ANA) tests. Their use demonstrated no statistically significant patterns.

Of all respondents, 87.1% (NON 85%, ON 89.2%) rated that they were “very comfortable” in diagnosing ISSNHL.

Initial Management

Participants were surveyed for management strategies. Initial treatment regimens included (alone or in combination) steroids (oral or intratympanic), antivirals, and others (diuretics, histamines, or vasodilators) ( Figure 3 ). Generally, 26.7% of respondents (NON 35.0%, ON 18.3%; P < .0001) preferred oral systemic steroids alone, 22.1% (NON 11.7%, ON 32.5%; P < .0001) preferred a combination of oral and intratympanic steroids, and 15% (21% of NON, 9% of ON) preferred a combination of oral steroids and antiviral therapy.

OPEN IN VIEWER

Figure 3.

Initial management. I, intratympanic steroids; NON, nonotologist/neurotologist; ON, otologist/neurotologist; S, oral steroids; V, antivirals.

For oral steroid usage, participants were surveyed for steroid of choice, dose, frequency, and duration. Of the respondents, 93.2% (NON 95.2%, ON 91.1%) preferred prednisone as the systemic steroid of choice (over methylprednisone and dexamethasone). Among the NON and ON groups, variable prednisone dosing (1 mg/kg, 20-60 mg) was reported ( Figure 4 ); 65.9% (NON 65.0%, ON 66.7%) preferred to dose prednisone at 60 mg. For dosing frequency, 62.2% (NON 61.9%, ON 62.5%) preferred to prescribe steroids to be taken once daily. A wide range of oral steroid duration was reported ( Figure 5 ), with 16.1% (NON 14.3%, ON 17.9%) preferring dosing systemic steroids for 7 days, 33.2% (NON 34.3%, ON 32.1%) preferring dosing systemic steroids for 10 days, and 32.2% (NON 29.5%, ON 34.8%) preferring dosing for 14 days.

OPEN IN VIEWER

Figure 4.

Initial treatment: dosage of oral steroid. NON, nonotologist/neurotologist; ON, otologist/neurotologist.

OPEN IN VIEWER

Figure 5.

Initial treatment: duration of oral steroid therapy. NON, nonotologist/neurotologist; ON, otologist/neurotologist.

For intratympanic steroid usage, participants were surveyed for steroid of choice, frequency of injections, and number of injections prior to hearing assessment after therapy. Of the respondents, 63.2% (56.5% NON, 69.9% ON) preferred dexamethasone, and 19% (37% NON vs 1.0% ON; P < .0001) preferred prednisone. Of the respondents, 60.3% (61.5% NON, 59% ON) reported weekly injections with a range of frequencies reported ( Figure 6 ); 59.5% (58% NON vs 61% ON) reported injecting steroids once prior to rechecking hearing, whereas other respondents perfused patients multiple times prior to formal audiologic reevaluation.

OPEN IN VIEWER

Figure 6.

Initial treatment: frequency of intratympanic steroid injections. NON, nonotologist/neurotologist; ON, otologist/neurotologist.

Secondary Management

Of the respondents, 68.6% (NON 57.5%, ON 80.0%; P = .0003) would continue with additional treatment after partial response to initial treatment (as defined as 20-dB improvement in pure-tone average). A small number of respondents specified the amount of elapsed time between initial treatment and partial response in which they would consider further treatment, a variable intentionally left blank. These responses ranged from 1 day to 8 weeks. For secondary treatment, regimens included (alone or in combination) steroids (oral or intratympanic), antivirals, and others (diuretics, histamines, or vasodilators) ( Figure 7 ). Of those who would retreat, 20.8% would retreat with oral steroids alone (NON 33.3%, ON 8.3%; P < .005), 46.6% would retreat with intratympanic injections (NON 35.8%, ON 57.3%; P < .05), and 69.2% were “very comfortable” in treating ISSNHL (NON 45.8%, ON 92.5%; P = .0001).

OPEN IN VIEWER

Figure 7.

Secondary management. I, intratympanic steroids; NON, nonotologist/neurotologist; ON, otologist/neurotologist; S, oral steroids; V, antivirals.

Discussion

ISSNHL can have severe and long-lasting effects on those unfortunate patients who suffer from this condition. Unfortunately, despite decades of study, much remains unknown regarding the etiology and natural course of this disease. As a result, data are lacking as a guide for management.

The main aim of this study was to investigate whether management styles vary in the diagnosis and treatment of ISSNHL. These data confirm that disparate methods of diagnosing and treating ISSNHL are employed by both general and subspecialty otolaryngologists, although some differences between nonotologists/neurotologists and otologists/neurotologists exist. Otologists/neurotologists saw on average more patients per year and were more likely to use MRI as a diagnostic adjunct, less likely to use intratympanic prednisone, more likely to offer additional treatment after partial response, and more likely to transfuse steroids following partial response. The reasons for this are open for debate. The vast majority of respondents who identify themselves as “otologists/neurotologists” received subspecialty (fellowship) training and, by seeing more patients, presumably had initially or subsequently developed an interest and/or inclination to see patients with ISSNHL. Whether this correlates with “aptitude” is difficult to say as no standardized management paradigm exists. Interestingly, roughly one-third of both NONs and ONs prescribed antivirals at initial presentation, despite several prospective randomized studies disproving their efficacy.^6-9 In addition, despite widespread use of steroids (systemic and intratympanic), the Cochrane Review and other studies of steroid use in the treatment of ISSNHL were inconclusive in the efficacy of steroids in improving outcomes.^3-5

In 2000, Loughran ¹⁰ surveyed 100 otolaryngologists in the United Kingdom, inquiring about diagnostic and management practices. The study demonstrated varying practices as well and, most notably, showed that although 98.5% would use steroids as part of initial management, the administration (oral vs intratympanic) and use of steroids in combination with other treatment options (ie, antivirals) widely varied. More recently, Shemirani et al ¹¹ compared management practices between general otolaryngologists and primary care physicians, showing that although a higher percentage of otolaryngologists use steroids as part of an initial treatment regimen when compared with general practitioners, the dose, duration, and timeframe to consider treatment varied among respondents. In this study, similar such variability in steroid use was found. In addition to the above studies, we demonstrate a range of opinions among both general otolaryngologists and otologists/neurotologists toward diagnostic criteria, initial workup and use of intratympanic steroids, and comfort levels.

Like many surveys, this study is not without its limitations. The response rate, although seemingly low, is comparable or higher than other mailed surveys of otolaryngologists performed recently in the United States. ¹¹ Selection bias for those with more interest or experience in ISSNHL may skew the results slightly, but one might expect that this would create more and not less uniformity. Therefore, these findings of variable management styles may or may not be extrapolated to the majority of otolaryngologists who did not respond but are suggestive of practice disparities. In addition, we have no way of confirming that responses actually correlate with practice. As no standard diagnostic or treatment paradigm exists, no standard “mean” could be created around which standard deviation and variance could be calculated. However, the purpose of this study was simply to prove a variety of management styles rather than a divergence from a clinical meaningful norm.

Ultimately, the true challenge of ISSNHL lies within its name: idiopathic. Without a clear understanding of the pathophysiology, outcome measurements become unreliable as distinct processes may be accounting for similar symptomatology. Likewise, failure to standardize diagnostic criteria can result in inappropriate inclusion and/or exclusion of subjects. Disparate treatment strategies also reflect a lack of evidence to support any particular modality or algorithm. The central function of this study is not merely to document that “different physicians treat ISSNHL differently” but rather to precisely highlight these differences to serve a broader purpose: (1) to spur practitioners to seek out existing evidence-based studies and possibly tailor their practice accordingly and (2) to spur researchers (and those that fund them) to engage in more and continued high-level study of this disorder.

Conclusion

Variance exists in the diagnostic workup and management of ISSNHL. Management styles differ between nonotologists/ neurotologists and otologists/neurotologists. Otologists are more likely to use MRI in their workup and more likely to employ combination therapy in the treatment of ISSNHL. Disparity exists between comfort level in diagnosis and treatment of ISSNHL. Only through continued research can we begin to better address the lives of our patients afflicted with this disease.

Author Contributions

Daniel H. Coelho, idea conception, study design, data analysis, manuscript drafting, revising/editing, final approval; Leroy R. Thacker, study design and data analysis; David W. Hsu, data analysis, manuscript drafting, editing, final approval.

Disclosures

Competing interests: None.

Sponsorships: None.

Funding source: Dr. Thacker’s efforts were supported in part by National Institutes of Health grant UL1RR031990.