Objectives: A review was undertaken to evaluate the peer-reviewed literature on three areas of tinnitus management for the following Key Questions (KQs): (1) measures used to assess patients for management needs (KQ1); (2) effectiveness of treatments (KQ2); and (3) identification of prognostic factors (KQ3).
Data sources: MEDLINE®, Embase®, CINAHL®, PsycINFO®, AMED©, and Cochrane CENTRAL were searched from January 1970 to June 2012. An extensive grey literature search, which included documents from regulatory and tinnitus-related organizations, was also undertaken.
Review methods: Standardized systematic review methodology was employed. Eligibility criteria included English-language studies of adults with subjective idiopathic (nonpulsatile) tinnitus; excluded studies involved tinnitus as the result of middle-ear pathologies or focused on methods to determine psychosomatic tinnitus. For KQ2, all pharmacological/food supplement, medical/surgical, sound/technological, and psychological/behavioral interventions aimed at ameliorating tinnitus symptoms were eligible (except stapedectomy or tympanoplasty). Randomized controlled trials with placebo controls or head-to-head trials were eligible for all KQs.
Results: From 9,725 citations, 52 eligible publications were extracted for data. None were eligible for KQ1 or KQ3. From the 52 publications eligible for KQ2, 17 evaluated pharmacological interventions; 11 evaluated medical interventions (low-level laser, acupuncture, transcranial magnetic stimulation); 5 evaluated sound technologies; and 19 evaluated psycholocal/behavioral interventions. Data on adverse effects were generally poorly collected and reported.
Conclusions: There is low strength of evidence (SOE) indicating that cognitive behavioral therapy interventions improve tinnitus-specific quality of life relative to inactive controls. For pharmacological interventions, SOE is low for improvements to subjective loudness from neurotransmitter drugs versus placebo; insufficient for antidepressants, other drugs, and food supplements with respect to subjective loudness; and insufficient for all other outcomes. There is insufficient SOE to suggest that medical interventions improve outcomes relative to inactive controls; sleep and global quality of life were not evaluated for medical interventions. The SOE for the adverse effect of sedation in pharmacological studies was judged insufficient. Future research should address the substantial gaps identified for KQ1 and KQ3. For KQ2, future research should concentrate on improving collection of adverse effects, calculating sample size, and specifying doses for interventions.