Objectives: Review reports of randomized clinical trials (RCTs) in tinnitus to identify well-established treatments, promising developments, and opportunities for improvement in this area of clinical research.
Study design: Literature review of RCTs (1964-1998) identified by MEDLINE and OLD MEDLINE searches and personal files.
Methods: Studies were compared with the RCT criteria of Guyatt et al. for quality of design, performance, and analysis; "positive" results were critically examined for potential clinical relevance.
Results: Sixty-nine RCTs evaluated tocainide and related drugs, carbamazepine, benzodiazepines, tricyclic antidepressants, 16 miscellaneous drugs, psychotherapy, electrical/magnetic stimulation, acupuncture, masking, biofeedback, hypnosis, and miscellaneous other nondrug treatments. No treatment can yet be considered well established in terms of providing replicable long-term reduction of tinnitus impact, in excess of placebo effects.
Conclusions: Nonspecific support and counseling are probably helpful, as are tricyclic antidepressants in severe cases. Benzodiazepines, newer antidepressants, and electrical stimulation deserve further study. Future tinnitus therapeutic research should emphasize adequate sample size, open trials before RCTs, careful choice of outcome measures, and long-term follow-up.