Sudden sensorineural hearing loss (SSNHL)

National Guideline Centre (UK)

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Guideline Centre (UK). Hearing loss in adults: assessment and management. London: National Institute for Health and Care Excellence (NICE); 2018 Jun. (NICE Guideline, No. 98.)

Hearing loss in adults: assessment and management.

Show details

Contents

< Prev Next >

11Sudden sensorineural hearing loss (SSNHL)

11.1. Introduction

Sudden sensorineural hearing loss (SSNHL) is an ENT emergency and is defined as a loss of hearing of 30 dB HL or more, over at least 3 contiguous frequencies, that develops within 3 days. Most cases are unilateral and the commonest age group affected are adults in their 40s and 50s. In 90% of cases no underlying cause is identified and it is considered idiopathic. Idiopathic SSNHL affects approximately 5–20 per 100,000 people per year in developed countries with an equal gender distribution. The hearing loss can range from mild to profound and can be temporary or permanent. Idiopathic SSNHL significantly impacts on individuals’ lives, causing considerable disability, especially if there is a preexisting hearing deficit in the other ear.

Hearing loss is confirmed by pure tone audiometry, and serious underlying causes are excluded by history, examination, MRI and blood tests. Additional investigations may be indicated depending on the presentation.

The mainstay of treatment currently consists of early initiation of steroids with antivirals as a possible adjunct. The effectiveness of these treatments is poorly understood and there is no national standard or guidance for the management of idiopathic SSNHL, with considerable variation in practices in terms of treatment regimen employed and routes of administration. This chapter examines the most clinically and cost-effective medical treatments for idiopathic sudden sensorineural hearing loss, including routes of administration.

11.2. Review question: What is the most clinically and cost-effective treatment for idiopathic sudden sensorineural hearing loss (SSNHL)?

For full details see the review protocol in appendix C.

Table 53

PICO characteristics of review question.

11.2.1. Clinical evidence

Thirteen studies were included in the review: 8 first-line treatment comparisons⁴^,¹²^,⁴⁴^,⁹¹^,¹¹⁰^,¹¹⁵^,¹¹⁶^,¹²³; and 5 second-line treatment regimens⁶⁹^,⁷⁰^,⁹⁶^,¹²⁷^,¹²⁸, these are summarised in Table 55 below. Evidence from these studies is summarised in the clinical evidence summaries below (Table 56, Table 57, Table 58, Table 59 and Table 60). See also the study selection flow chart in appendix E, forest plots in appendix K, study evidence tables in appendix H, GRADE tables in appendix J and excluded studies list in appendix L.

Table 55

Summary of studies included in the review.

Table 56

Clinical evidence summary: Steroid (oral/IT) versus placebo [Prednisolone versus placebo].

Table 57

Clinical evidence summary: Steroid (oral/IT) versus steroid (oral) [Dexamethasone versus Prednisolone].

Table 58

Clinical evidence summary: steroid (oral) plus steroid (IT) versus steroid (oral/IT) [prednisolone plus dexamethasone versus prednisolone or dexamethasone plus placebo].

Table 59

Clinical evidence summary: Steroid (IV or oral) plus antiviral (IV or oral) versus steroid (IV or oral) [prednisolone or hydrocortisone plus acyclovir or valacyclovir versus prednisolone or hydrocortisone].

Table 60

Clinical evidence summary: Steroid (IT) versus placebo (IT) /no treatment [Dexamethasone/prednisolone versus placebo/no treatment].

Two Cochrane systematic reviews were identified: Antivirals for idiopathic sudden-onset sensorineural hearing loss⁸ and Steroids for idiopathic sudden sensorineural hearing loss.¹²¹ References from these, and other identified systematic reviews and meta-analyses, were checked and studies were included in this review if they matched our protocol.

Some papers that were excluded by the Cochrane review on Steroids for idiopathic sudden sensorineural hearing loss¹²¹ have been included in our review. The reasons for exclusion from the Cochrane review were as follows:

Battaglia 2008¹² – “impossible to determine the true effects of steroid when there was no double placebo control (placebo intratympanic injections with placebo taper)”
Lee 2011⁶⁹ – not randomised, double blind, or placebo controlled
Li 2011⁷⁰ – not double blind or placebo controlled. “impossible to determine the true effects of steroid when there was no placebo control”
Plontke 2009⁹⁶ – “Impossible to determine the true effects of steroid when there was no placebo control (participants without any systemic and local steroid treatment)”
Wu 2011¹²⁷ – “Impossible to determine the true effects of steroid when there was no double placebo control of patients who received neither systemic nor intratympanic steroid”

Our protocol included studies that lacked a placebo comparison whereas the Cochrane review did not. In addition, we believe that the Lee 2011⁶⁹ study which was excluded from the Cochrane review, is in fact, randomised and has therefore been included in our review.

Two papers that were included in the Cochrane review on steroids¹²¹ have been excluded from this review (Cinamon 2001²¹ and Wilson 1980¹²⁵) due to being quasi-randomised and having unclear methodology with mixed treatment doses, respectively.

One paper included in the Cochrane review on antivirals,⁸ did not have any analysable data and was excluded (Westerlaken 2003).¹²⁴

Of the included studies, all had a population with unilateral hearing loss. The methods for excluding underlying causes of hearing loss vary, with some studies excluding patients post randomisation where pathology was later identified. Five of the studies⁴^,⁷⁰^,⁹¹^,¹²⁷^,¹²⁸ did not explicitly state that patients with autoimmune disease were actively excluded. As this means any effect of steroids could be due to their known benefits for autoimmune disorders, these studies were subject to sensitivity analysis to determine whether their findings differed from other studies within each comparison. No systematic differences were found between the results of these studies and the remaining studies that had excluded those with autoimmune disease.

The interventions that were included in the studies were prednisolone, methylprednisolone, dexamethasone, hydrocortisone, acyclovir and valacyclovir. Acyclovir was also used prerandomisation in a second-line treatment study (Xenellis 2006¹²⁸). The treatments were given intravenously (IV), orally or intratympanically (IT).

The studies reported pure tone audiometry (PTA) by change or final scores, author-defined improvement and recovery. In this report, PTA data are presented as the change or final scores as well as the author-defined recovery data as these were thought to be the most important outcomes for decision-making. Where these continuous and dichotomous outcomes are derived from the same dataset this has been highlighted to avoid double counting the data. If neither of these preferred definitions are presented by a given study, author-defined improvement was reported.

Speech discrimination was also recorded in some studies. Few studies reported adverse events, but these data have been extracted where possible. Only one study reported on quality of life, Tucci 2002,¹¹⁵ but only an overarching summary comparing the population to the USA population was available.

The definitions used to describe recovery varied between the studies and some studies included more than one definition of recovery. The definitions included as outcomes within this report were those that were most representative of clinical recovery that is important to the patient (Table 54).

Table 54

Study definitions of improvement and recovery.

Another factor to consider is the mean time between the onset of hearing loss and start of treatment and whether this would affect the end of study outcomes as this varied between the studies (Table 55).

11.2.1.1. First-line treatment for idiopathic sudden sensorineural hearing loss

11.2.1.2. Second-line treatment for idiopathic sudden sensorineural hearing loss

11.2.2. Economic evidence

11.2.2.1. Published literature

No relevant health economic studies were identified.

See also the health economic study selection flow chart in appendix F.

11.2.2.2. Unit costs

See appendix P.

11.2.3. Evidence statements

Clinical

First line

Steroid (oral/IT) versus placebo [Prednisolone versus placebo]

There was a clinically important benefit of prednisolone for recovery at 1 month (moderate quality evidence, 1 study).
There was no clinically important difference in change in PTA at day 8 and day 90 (low quality evidence, 1 study), for recovery at day 8 and day 90 and for adverse events (very low quality evidence, 1 study).

Steroid (oral/IT) versus steroid (oral) [Dexamethasone versus Prednisolone]

There was a clinically important benefit of dexamethasone for recovery of hearing to within 5% points of the contralateral SDS or within 5 dB of the contralateral PTA (very low quality evidence, 1 study).
There was no clinically important difference in PTA final score (low quality evidence, 2 studies) and in symmetrical hearing recovery (interaural hearing difference of <20 dB HL) and speech discrimination of 100% (very low quality evidence, 1 study) and in mean speech discrimination (very low quality evidence, 1 study).

Steroid (oral) plus steroid (IT) versus steroid (oral/IT) [prednisolone plus dexamethasone versus prednisolone or dexamethasone plus placebo]

There was a clinically important benefit of dual steroids (oral plus IT versus oral and versus IT) for PTA final score and mean speech discrimination (low quality evidence, 1 study).
There was no clinically important difference in recovery (very low quality evidence, 2 studies).

Steroid (IV or oral) plus antiviral (IV or oral) versus steroid (IV or oral) [prednisolone or hydrocortisone plus acyclovir or valacyclovir versus prednisolone or hydrocortisone]

There was a clinically important harm of prednisolone plus placebo (very low quality evidence, 1 study).
There was no clinically important difference in PTA final score, recovery within 10 dB of non-affected, mean speech discrimination (very low quality evidence, 1 study) and improvement (very low quality evidence, 1 study).

Second line

Steroid (IT) versus placebo (IT) /no treatment [Dexamethasone/prednisolone versus placebo/no treatment]

There was a clinically important benefit of steroids compared with placebo or no treatment for PTA final score (very low quality evidence, 4 studies), recovery and speech discrimination (low to very low quality evidence, 1 study) and for improvement (high quality evidence, 1 study).
There was no clinically important difference in adverse events (perforation of tympanic membrane; low quality evidence, 1 study).
There were no data on amantadine, famciclovir and ganciclovir.

Economic

No relevant economic evaluations were identified.

11.2.4. Recommendations and link to evidence

Recommendations	20. Consider a steroid to treat idiopathic sudden sensorineural hearing loss in adults.
Relative values of different outcomes	The following critical outcomes were included in this review: pure tone average, speech discrimination or recognition, health-related quality of life and hearing-specific health-related quality of life. Adverse events were included as an important outcome.
Quality of the clinical evidence	The studies reported pure tone average (PTA) by change or final scores, author-defined improvement and recovery. We analysed the change or final scores as well as the author-defined recovery data as these were thought to be the most important outcomes for decision-making. If neither of these preferred definitions was presented by a given study, author-defined improvement was reported. The majority of the evidence was of low or very low quality and based on few studies with small sample sizes. There were no data on amantadine, famciclovir and ganciclovir. There were a range of limitations of the included evidence, most notably: A lack of detail regarding how the diagnosis of ‘idiopathic’ SSNHL was determined. Some studies identified patients with causes of the SSNHL after randomisation and consequently excluded them from the study. Many studies did not describe how the patients were randomised and how allocation was concealed. Other reasons for downgrading related to blinding, attrition bias, and not specifying outcomes in the methods (post-hoc analysis). Outcomes were also downgraded due to imprecision; many studies had relatively small numbers in each treatment group. One study that provided the majority of the data about dual steroid (oral plus IT) treatments had significant baseline differences in time-to-treatment, speech discrimination score and PTA between the 3 treatment groups making the evidence difficult to interpret. Some studies did not consider high frequency losses (above 2 kHz) within their analysis. Some studies either included children but it was unclear how many, or gave no age inclusion criteria or age range to clarify whether any children had been included. However, given the mean ages and standard deviations it seems unlikely that any children were included in the later cases.
Trade-off between clinical benefits and harms	The guideline committee noted evidence for a lack of efficacy of oral steroids over placebo in first-line treatment of SSNHL, with associated adverse events. However, the adverse events were not clearly specified so it was not possible to determine the clinical importance of these. Oral steroids are currently the mainstay of treatment for this condition in clinical practice, but no evidence to support this practice was found. However, there was also insufficient evidence to change current practice. Conversely, 1 study gave moderate quality evidence that intratympanic steroids are clinically beneficial for recovery, but no other outcomes were available. When comparing dexamethasone and prednisolone for the first-line treatment of SSNHL, low and very low quality evidence showed no clinical difference for PTA final score, recovery or speech discrimination when both were given orally. However, IT dexamethasone was clinically beneficial compared with oral prednisolone for recovery and change in speech discrimination score (taking into account baseline differences). The committee noted that although there was some evidence that dual steroids (oral plus IT) were better than either alone, the majority of these data were from a study at very high risk of bias owing to baseline differences. Therefore, the committee did not have high confidence in these findings. However, they provided further evidence to support the emerging theme that IT steroids may be the most effective course of treatment in this patient group. When comparing treatment with steroid plus antiviral versus steroid alone there was no evidence of a clinical benefit of adding in the antiviral for any recorded positive outcome. There were fewer adverse events in the group given an antiviral in addition to the steroid; however, from looking at the specific adverse events reported, the committee did not believe there was a clinical explanation and it is likely to be a chance finding owing to the small sample size. There were 5 studies reporting the comparison of IT steroids versus placebo or no treatment for the second-line treatment of patients initially refractory to IV or oral steroid treatment. These all showed a clinical benefit of IT steroid in this group for PTA final score, recovery, improvement (1 study; high quality) and speech discrimination. The committee noted that it was interesting to see that second-line treatment with IT steroids is effective at a time in the natural course of the condition when spontaneous recovery would be less likely. Some studies reported individual cases of tympanic membrane perforation, but all either resolved spontaneously or were successfully repaired. As usual practice is to prescribe oral steroids, recommending IT would be a significant change in practice and would have an impact on service delivery due to the requirement of hospital-based treatment follow-up appointments and staff with expertise in this treatment. The committee discussed the possibility of a recommendation for IT steroids for second-line treatment, where there is evidence of benefit. However, it is not possible to determine whether IT steroids have a benefit by themselves or in combination with previous oral steroid treatment. Additionally, without sufficient evidence to be confident in the best route of administration for first-line treatment the committee could not specify the route of administration for second-line treatment. Therefore, it was unable to either fully support or change current treatment and the committee agreed to recommend steroid treatment without specifying the route of administration (see recommendation in following table).
Trade-off between net clinical effects and costs	No health economic evidence was identified for this question. The committee considered the clinical evidence and unit costs for the alternative treatment options. The committee noted that the costs of the most likely steroids were low, being between £3 and £13 for a course lasting up to 2 weeks. Therefore the most important resource use and cost would be the cost of administering these drugs. If oral drugs are prescribed these can be taken at home and the patient may require only 2 outpatient appointments for investigation and review. Giving drugs intratympanically would give rise to greater costs as each time drugs are administered the patient would need to come into hospital for an outpatient appointment where this would be conducted by a suitably trained clinician. This could involve 3 or 4 outpatient appointments and the first such appointment may be outside routine hours adding to the expense. An outpatient appointment for a minor ear procedure has an NHS reference cost of £110. Adopting IT as the standard first-line treatment would be a change in practice and would require greater resources. Giving drugs intravenously, by contrast, would necessitate an inpatient stay, which would be even more expensive. Given that no clinical evidence was identified favouring IV drugs, the committee did not recommend this as an option. Given the limited evidence favouring steroid treatment and their low cost, but a lack of clear evidence from the clinical review comparing different oral and IT drugs, and considering the higher costs of IT treatment, the committee was not able to support any particular drug or route of administration, but advised that the use of steroids should be considered.
Other considerations	The committee highlighted the fact that hearing aid use, audiological rehabilitation and overall management strategies for SSNHL were not considered within this review.

11.3. Review question: What is the clinical and cost effectiveness of different routes of administration of steroids (for example oral or intratympanic) in the treatment of sudden sensorineural hearing loss (SSNHL)?

For full details see the review protocol in appendix C.

Table 61

PICO characteristics of review question.

11.3.1. Clinical evidence

For the sub-question addressing the route of steroid administration, 11 studies, reported in 13 papers were included in the review.⁴^–⁶^,¹²^,³⁶^,⁴⁰^,⁵²^,⁵³^,⁶⁵^,⁷¹^,⁷²^,⁹⁹^,¹¹² These are summarised in Table 63 below. Two of these studies were also included in the primary review.⁴^,¹² Evidence from these studies is summarised in the clinical evidence summary tables below (Table 63).

Table 63

Summary of studies included in the review.

See also the study selection flow chart in appendix E, forest plots in appendix K, study evidence tables in appendix H, GRADE tables in appendix J and excluded studies list in appendix L.

Of the included studies, all but 1¹¹² had a population exclusively with unilateral hearing loss; in this study the majority had unilateral losses. The methods for excluding underlying causes of hearing loss vary; 2 of the studies¹²^,⁹⁹ explicitly stated that patients with autoimmune disease were actively excluded.

The interventions that were included in the studies were prednisolone, methylprednisolone, dexamethasone, and acyclovir. The treatments were given intravenously (IV), orally or intratympanically (IT).

The studies reported pure tone averages (PTA) by change or final scores, author-defined improvement and recovery. In this report, PTA data are presented as the change or final scores as well as the author-defined recovery data as these were thought to be the most important outcomes for decision-making. Where these continuous and dichotomous outcomes are derived from the same dataset this has been highlighted to avoid double counting the data. If neither of these preferred definitions are presented by a given study, author-defined improvement will be reported.

Speech discrimination or word recognition and adverse events were also recorded in some studies. None of the studies reported on quality of life.

The definitions used to describe recovery varied between the studies. The definitions included as outcomes within this report were those that were most representative of clinical recovery that is important to the patient (Table 54).

Table 62

Study definitions of improvement and recovery.

11.3.1.1. First-line treatment for idiopathic sudden sensorineural hearing loss

Table 64. Clinical evidence summary: Steroid (IT) versus steroid (oral) [IT prednisolone, methylprednisolone or dexamethasone versus oral prednisolone].

11.3.2. Economic evidence

11.3.2.1. Published literature

No relevant health economic studies were identified.

See also the health economic study selection flow chart in appendix F.

11.3.2.2. Unit costs

See appendix P.