Network Meta-Analysis Will Compare Acupuncture Modalities for Sudden Sensorineural Hearing Loss
Sudden sensorineural hearing loss remains an otologic emergency because auditory function can fall abruptly and the window for recovery is often narrow. Corticosteroids remain the conventional backbone of treatment, yet the clinical response is incomplete for many patients, leaving a persistent need for adjunctive approaches that can improve hearing without adding unacceptable harm. A protocol published in BMJ Open sets out to examine whether acupuncture-related therapies can fill part of that gap, and whether one modality performs better than another when the available randomized evidence is pooled and compared across treatment nodes.
The planned work is not a treatment trial; it is a systematic review and network meta-analysis designed to evaluate relative efficacy among multiple acupuncture-related interventions used for SSNHL. That distinction matters because the study aims to organize an already fragmented literature rather than generate new patient-level outcomes. The protocol is registered under CRD42024525763, giving the project a prespecified framework before analysis begins.
Pure-Tone Threshold Change Anchors the Primary Clinical Question
Investigators will search nine databases and five clinical trial registries for randomized controlled trials of acupuncture-related therapies in SSNHL. The primary endpoint is the change in pure-tone audiometric thresholds, a clinically direct measure of whether sound detection improves after treatment. In an emergency disorder defined by abrupt auditory loss, that endpoint carries more weight than broad surrogate claims because it reflects the actual hearing deficit that brings patients to care.
Secondary outcomes extend beyond the audiogram. The planned analysis will also examine the proportion of recovered patients, tinnitus severity, psychological symptoms, and adverse events. Those outcomes are relevant because SSNHL rarely affects hearing in isolation; tinnitus can remain intrusive, emotional distress can follow the hearing loss, and any adjunctive therapy must still clear a basic safety bar. A therapy that improves thresholds while worsening tolerability would have limited bedside value, especially when patients are already navigating uncertainty about recovery.
Bayesian Methods Will Map Relative Efficacy Across Multiple Acupuncture Approaches
The analysis plan relies on a Bayesian network meta-analysis, a design well suited to clinical questions where several competing interventions exist but head-to-head trials are sparse. In practical terms, this approach allows comparison of more than two acupuncture modalities within the same analytical framework, provided the evidence network is sufficiently connected. That structure is particularly relevant here because acupuncture-related care is not a single uniform exposure; it includes multiple forms that may differ in technique, intensity, and theoretical basis.
Two independent reviewers will handle study selection, data extraction, and methodological quality assessment. The protocol specifies use of the Cochrane risk of bias tool, version 2.0, which should help standardize judgments about randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting. A study set with inconsistent methods can quickly distort pooled estimates in a network analysis, so the quality appraisal is central rather than decorative.
The review will also apply the GRADE approach to judge the overall certainty of evidence. That step is essential because a statistically ordered ranking of therapies is not the same thing as strong clinical evidence. Network estimates can look tidy even when the underlying trials are small, heterogeneous, or methodologically uneven, and the GRADE framework is intended to keep those limits visible.
Safety, Tinnitus, and Psychological Symptoms Remain Clinically Relevant Endpoints
A trial program focused only on hearing thresholds would miss the broader burden of SSNHL. Tinnitus can be persistent and distressing, and psychological symptoms may compound the functional loss produced by abrupt deafness. By including those secondary outcomes, the review is positioned to capture more than a narrow audiologic signal and may show whether any acupuncture-related therapy influences the symptom cluster patients actually experience at the bedside.
Adverse events are also built into the protocol, which is necessary for any intervention being considered alongside standard care. Acupuncture is often discussed in terms of tolerability, but systematic safety assessment still needs to be explicit rather than assumed. The protocol does not forecast what the safety profile will look like; it simply makes clear that adverse events will be extracted and considered in the comparison.
What the Protocol Can and Cannot Answer
The study is structured to clarify comparative efficacy, not to settle every question in SSNHL management. Because it is a review of previously published trials, it cannot resolve confounding that was never controlled in the source studies, and it cannot rescue poor trial design after the fact. Its value lies in synthesis: if the evidence base is large enough and sufficiently connected, the analysis may identify which acupuncture-related approaches appear most effective for hearing recovery and which, if any, carry a clearer safety signal.
At the same time, the protocol’s own precision is a reminder of the evidence gap. The investigators had to specify a wide search across multiple databases and registries because the field remains scattered, and they had to choose network methods because direct comparative trials are likely limited. In SSNHL, where delayed or incomplete recovery can leave lasting auditory disability, that fragmentation is itself clinically important.
No ethical approval is required because the project uses data already published in the literature. The findings will be disseminated in a peer-reviewed journal, where the strength of the final message will depend on the quality and coherence of the randomized trials that enter the network. If the evidence base proves thin or uneven, the analysis will reflect that rather than manufacturing certainty where little exists.