Speech Discrimination Outcomes in Patients Implanted With Perimodiolar Versus Lateral Wall Cochlear Implant Arrays: Systematic Review and Meta-Analysis

OBJECTIVE: Provide a systematic review and meta-analysis of head-to-head comparison studies of cochlear implant outcomes from perimodiolar (PM) and lateral wall (LW) electrodes. DATABASE REVIEWED: Ovid Medline ALL, Ovid Embase, and Wiley's Cochrane Central Register of Controlled Trials. METHOD: Studies were parsed using predetermined inclusion and exclusion criteria. Risk of bias was assessed based on ROBINS-I. Preoperative and postoperative CNC and AzBio outcomes were obtained from eligible studies. Standard mean differences (SMD) were calculated, and a meta-analysis was performed. Summary effect and heterogeneity were reported. RESULTS: From 2,219 studies, data points from 11 studies were extracted for analysis. Nine studies of moderate to low risk of bias reported CNC results. Four studies reported statistically significant superiority of the perimodiolar array, with the largest mean difference between groups of 14%. Meta-analysis revealed a pooled SMD of -0.21 (-0.39; -0.02) (P=0.04), suggesting a statistically significant greater CNC scores in patients implanted with a PM array with a small effect size. When re-expressed in clinical units using the pooled standard deviation, the mean difference was 0.7% CNC percentage units between patients implanted with PM or LW arrays. The pooled SMD of -0.01 (-0.22; 0.20), P=0.81, suggests no statistically significant difference in AzBio postimplantation scores between patients implanted with LW versus PM arrays. Meta-regression suggested a propensity for studies with older patients to have improved AzBio outcomes with perimodiolar arrays (P=0.02). CONCLUSION: Analysis of combined CNC scores revealed a statistically significant improvement in post-CNC results following implantation with a PM array at ∼1 year, but this difference was slight and not clinically significant. There was no difference in post-operative AzBio scores between patients implanted with a PM array compared with those implanted with a LW array. There may be an age-related benefit to the perimodiolar electrode with regard to AzBio word understanding, as studies suggest improved AzBio scores with PM arrays included older patients. No such relationship was seen for CNC scores.