Systematic review of the literature on nitinol prostheses in surgery for otosclerosis: assessment of the adequacy of statistical power.

OBJECTIVE: To perform a systematic review of observational studies reporting hearing outcome in primary stapes surgery where a heat-crimping prosthesis was used. DATA SOURCES: Medline, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, TRIP database, Clinical Trials Registry, ISI Web of Knowledge, and Web of Science. The search was performed on January 1, 2010, including articles published ahead of print. No language restrictions. STUDY SELECTION: Inclusion criteria for qualitative synthesis were a population of otosclerosis patients, intervention being primary stapes surgery with a nickel titanium alloy (Nitinol) heat-crimping prosthesis, and hearing outcome. Inclusion criteria for quantitative analysis: application of audiometry guidelines of the American Academy of Otolaryngology-Head and Neck Surgery Committee on Hearing and Equilibrium for evaluation of conductive hearing loss. DATA EXTRACTION: Strategy and reporting based on Cochrane, Quality of Reporting of Meta-analyses, and Meta-analysis of Observational Studies in Epidemiology statements. A bias assessment tool was developed according to Cochrane guidelines. DATA SYNTHESIS: A quantitative synthesis was performed, but because of the heterogeneity in postoperative follow-up periods and outcome measures reported, we were not able to pool these data. A sample size analysis was performed to indicate the sample needed to demonstrate a statistically significant difference in hearing outcome between both interventions. Hearing outcome superiority of the Nitinol heat-crimping prosthesis over manually crimping prosthesis types was not demonstrated.
CONCLUSION: Superiority could probably not be demonstrated because of insufficient sample size. Research addressing technical improvements in stapes surgery should agree on a base sample size able to detect the smallest difference that is clinically important or accept the null hypothesis. With data gathered in the Common Otology Database as basis, a sample size of at least 413 patients is needed in both the intervention and the control group. Other clinical outcome measures also should be explored.