Amplification with hearing aids for patients with tinnitus and co‐existing hearing loss
Tinnitus is described as the perception of sound or noise in the absence of real acoustic stimulation. In the current absence of a cure for tinnitus, clinical management typically focuses on reducing the effects of co‐morbid symptoms such as distress or hearing loss. Hearing loss is commonly co‐morbid with tinnitus and so logic implies that amplification of external sounds by hearing aids will reduce perception of the tinnitus sound and the distress associated with it.
To assess the effects of hearing aids specifically in terms of tinnitus benefit in patients with tinnitus and co‐existing hearing loss.
We searched the Cochrane Ear, Nose and Throat Disorders Group Trials Register; the Cochrane Central Register of Controlled Trials (CENTRAL); PubMed; EMBASE; CINAHL; Web of Science; Cambridge Scientific Abstracts; ICTRP and additional sources for published and unpublished trials. The date of the search was 19 August 2013.
Randomised controlled trials and non‐randomised controlled trials recruiting adults with subjective tinnitus and some degree of hearing loss, where the intervention involves amplification with hearing aids and this is compared to interventions involving other medical devices, other forms of standard or complementary therapy, or combinations of therapies, no intervention or placebo interventions.
Data collection and analysis
Three authors independently screened all selected abstracts. Two authors independently extracted data and assessed those potentially suitable studies for risk of bias. For studies meeting the inclusion criteria, we used the mean difference (MD) to compare hearing aids with other interventions and controls.
One randomised controlled trial (91 participants) was included in this review. We judged the trial to have a low risk of bias for method of randomisation and outcome reporting, and an unclear risk of bias for other criteria. No non‐randomised controlled trials meeting our inclusion criteria were identified. The included study measured change in tinnitus severity (primary measure of interest) using a tinnitus questionnaire measure, and change in tinnitus loudness (secondary measure of interest) on a visual analogue scale. Other secondary outcome measures of interest, namely change in the psychoacoustic characteristics of tinnitus, change in self reported anxiety, depression and quality of life, and change in neurophysiological measures, were not investigated in this study. The included study compared hearing aid use to sound generator use. The estimated effect on change in tinnitus loudness or severity as measured by the Tinnitus Handicap Inventory score was compatible with benefits for both hearing aids or sound generators but no difference was found between the two alternative treatments (MD ‐0.90, 95% confidence interval (CI) ‐7.92 to 6.12) (100‐point scale); moderate quality evidence. No negative or adverse events were reported.
The current evidence base for hearing aid prescription for tinnitus is limited. To be useful, future studies should make appropriate use of blinding and be consistent in their use of outcome measures. Whilst hearing aids are sometimes prescribed as part of tinnitus management, there is currently no evidence to support or refute their use as a more routine intervention for tinnitus.
Derek J Hoare, Mark Edmondson‐Jones, Magdalena Sereda, Michael A Akeroyd, Deborah Hall
Plain language summary
Hearing aids for tinnitus in people with hearing loss
Tinnitus describes 'ringing', 'whooshing' or 'hissing' sounds that are heard in the absence of any corresponding external sound. About 10% of people experience tinnitus and for some it has a significant negative impact on their quality of life. Tinnitus is commonly associated with some form of hearing loss and is possibly the result of hearing loss‐related changes in brain activity. It is logical to think, therefore, that providing people who have hearing loss and tinnitus with a hearing aid will not only improve their ability to hear sound but will also reduce their tinnitus symptoms. Hearing aids increase the volume at which people hear external sounds so this may help mask or cover up the tinnitus sound. They also improve communication, which may reduce the symptoms often associated with tinnitus such as stress or anxiety. Hearing aids may also improve tinnitus symptoms by reducing or reversing abnormal types of nerve cell activity that are thought to be related to tinnitus. The purpose of this review is to evaluate the evidence from high‐quality clinical trials that try to work out the effects hearing aids have on people's tinnitus. We particularly wanted to look at how bothersome their tinnitus is, how depressed or anxious tinnitus patients are and whether hearing aid use has an effect on patterns of brain activity thought to be associated with tinnitus.
Our search identified just one randomised controlled trial which evaluated 91 participants who had tinnitus for at least six months and some degree of hearing loss. It compared those receiving hearing aids to those receiving sound generators. The average age of the patients was 38 and there were 40 women and 51 men. The study took place in two centres in Italy and the USA.
The result from the single study we reviewed was not definitive and was compatible with only small differences between the effect of hearing aids and sound generators. We also found another relevant study which has not yet been completed. We believe further high‐quality trials are needed.
Quality of the evidence
The quality of this evidence is moderate to low. This review is up to date to August 2013.
Derek J Hoare, Mark Edmondson‐Jones, Magdalena Sereda, Michael A Akeroyd, Deborah Hall
Implications for practice
Hearing aids are one of a number of therapeutic options offered to tinnitus patients. However, there is currently no evidence to support or refute the provision of hearing aids as a primary intervention in the management of tinnitus in patients with co‐existing hearing loss. Provision of hearing aids for tinnitus will always have the potential consequence of reducing the distress associated with hearing loss and so any clinical improvement that is specific to tinnitus will always be difficult to estimate accurately. We identified evidence of limited quality that for patients with a particular audiological profile hearing aids seem to be as beneficial as sound generators. They may in some cases be a better option on the grounds that amplification may for some patients be more acceptable and useful than the broadband or patterned sound stimuli delivered by sound generators (Hoare 2013).
Implications for research
Future research should aspire to produce high‐quality evidence from well‐conducted RCTs which report findings to recognised standards, such as the CONSORT statement (Schulz 2010). The choice of outcomes measured in trials also needs to be carefully considered. A recent proposal for international standards for tinnitus trials (Landgrebe 2012) considers a comprehensive outcome assessment of tinnitus to include psychoacoustic measures or ratings of loudness and annoyance as well as questionnaires measuring tinnitus impact. We recommend the use of the Tinnitus Functional Index (TFI) as a core outcome measure (Meikle 2012) as it was developed to be sensitive to treatment‐related changes, unlike many tinnitus questionnaires currently in use. The TFI also sets a benchmark of what constitutes a clinically significant benefit, that is a reduction of 13 points on this 0 to 100 scale. Psychoacoustic outcome measures are also important. A case in point is seen in one of the studies excluded from this review (Schaette 2010). The conclusions in this study relate to a grouping of participants according to psychoacoustic estimates of tinnitus pitch made at baseline. Pitch was not re‐evaluated at follow‐up so their conclusion hinged on the assumption that it was stable throughout the study. As far as is feasible, future studies should routinely include psychoacoustic measures of tinnitus at study endpoints.
The single study included in this review makes one comparison (hearing aids or sound generators) for a pre‐defined subset of patients (those with hearing loss at higher frequencies) and finds no between‐group difference in outcome. It remains open to future studies to determine whether, for given populations of help‐seeking tinnitus patients, the provision of a hearing aid is superior to an education‐only intervention, no intervention (waiting list control) or to a hearing aid placebo (where a hearing aid gain is set to overcome the effects of any occlusion due to the device fitting only, with zero amplification above the normal threshold). In terms of efficacy, an important question is whether or not patients with only mild hearing loss or high‐frequency hearing loss should routinely be offered a hearing aid. Parazzini 2011, however, compared hearing aid and sound generator use in patients who might reasonably be managed with either device, i.e. patients with tinnitus, normal hearing at lower frequencies (25 dB HL at 2 kHz) and some hearing loss at higher frequencies. This patient group may or may not report hearing difficulties as a primary complaint. The effects of amplification on this patient population lends itself to a placebo‐controlled RCT in a way that would be less appropriate to patient populations who have severe co‐morbid hearing loss.
Future trials should also consider, whilst controlling for hearing loss, randomising hearing aid features that maximise hearing benefit, such as noise reduction settings, environmental steering, compression and wide dynamic range, to provide evidence about which features contribute to or reduce the tinnitus benefit a patient may experience.