Acupuncture for the treatment of tinnitus: a systematic review of randomized clinical trials

Tinnitus is a common problem that affects 10-15% of the general population, causing perception of sounds without an external acoustic source. This condition can result in sleep disturbances, work impairment, and significant stress. Currently, only cognitive behavioral therapy has solid clinical evidence for improving the quality of life of patients with tinnitus. In this context, acupuncture has frequently been used as an alternative treatment.

This systematic review aimed to critically assess the available evidence on the efficacy of acupuncture in the treatment of tinnitus through the analysis of randomized clinical trials. The investigators conducted a comprehensive search across 14 databases, from inception to July 2012, including MEDLINE, EMBASE, Chinese and Korean databases, and the Cochrane Library. Randomized clinical trials that used acupuncture as the sole treatment, compared with sham acupuncture, conventional medications, or no-treatment controls, were included. Of the 382 articles initially identified, only 9 randomized clinical trials met the inclusion criteria.

These studies included a total of 567 participants and were conducted mainly in Denmark, China, Brazil, England, and Korea. The trials tested different types of acupuncture: manual acupuncture, electroacupuncture, and scalp acupuncture. The number of sessions varied dramatically, from a single session to 30 sessions, with durations of 15 seconds to 30 minutes each. Five trials compared manual acupuncture or electroacupuncture with sham acupuncture and failed to demonstrate statistically significant improvements in measures such as periodic index, annoyance, intensity, awareness of the symptom, Tinnitus Handicap Inventory, and response rate.

Two trials tested single-session scalp acupuncture compared with penetrating sham acupuncture at non-acupuncture points, demonstrating significant positive effects on subjective symptom relief on visual analog scale. Two trials compared acupuncture with conventional medications, with one showing positive effects on response rate in patients with nervous tinnitus, while the other did not show significant effects in patients with senile tinnitus. The methodological quality assessment revealed important limitations in most of the studies. Only four trials used adequate methods of random sequence generation.

Two trials had low risk of bias, but the risk of bias in sequence generation was high in two other trials. Only one trial employed adequate allocation concealment. Six of the eight trials adopted blinding of both assessor and participant, while another adopted only assessor blinding. Most of the trials had small samples, with fewer than 40 participants per group, making them prone to type II error or overestimation of the treatment effect.

No study used statistical power calculation. The acupuncture points were selected based on traditional Chinese medicine theory in eight trials, including points such as TE-3, GB-2, SI-19, GV-20, among others. The qi sensation (de qi) was considered in six trials, although no evidence was found that its presence or absence significantly influenced clinical outcomes. Adverse events were rarely reported, with only one trial mentioning two cases of significant needle pain.

The lack of adequate reporting of adverse events demonstrates inadequate reporting standards in acupuncture research. The clinical implications of this review are important. Although the evidence is not convincing enough to recommend acupuncture as a first-line treatment for tinnitus, in clinical practice, acupuncture treatment by experienced and licensed practitioners may be an option for patients with tinnitus, especially when they refuse cognitive behavioral therapy, which is the only treatment with demonstrated clinical evidence for improving quality of life. Since acupuncture is a relatively safe procedure and there is currently no treatment with proven clinical efficacy for specific tinnitus symptoms, acupuncture remains a therapeutic option for patients who request this procedure.

This review identified several important limitations affecting the conclusiveness of the results, including the scarcity and often suboptimal methodological quality of the primary data, possible publication bias, and selective reporting.