Objectives: Evidence for a bilateral cochlear implant (BiCI) advantage in quiet is inconsistent, in part because many studies compare binaural versus acutely monaural conditions within BiCI users, which cannot model long-term unilateral (UniCI) adaptation. The authors asked whether BiCI confers a measurable benefit over UniCI for speech perception in quiet after rigorously controlling selection bias, and which patient factors modulate this benefit.
Design: Single-center retrospective cohort of 184 postlingually deafened adults (109 UniCI, 75 BiCI) implanted between February 2013 and March 2025. The authors performed 1:1 propensity-score matching (PSM) using age, sex (exact match), and principal component composites of preoperative audiology to yield 120 patients (60 UniCI, 60 BiCI) with excellent balance. Speech perception in quiet (mono-, di-syllable, sentence) was assessed at 3, 6, and 12 months after activation. Longitudinal trajectories were analyzed with linear mixed-effects models, with cumulative auditory experience (Average Daily Device Use × Months Post-Activation) modeled via restricted cubic splines. Prespecified interactions tested effect modification by sex, age group (20 to 50 versus 50 to 70), cumulative use, and preoperative performance. Sensitivity analyses used a stricter PSM caliper (0.1).
Results: Before matching, BiCI recipients-despite poorer baselines-improved more steeply and achieved higher scores across time points. After PSM, BiCI showed a significant main effect for monosyllabic word recognition (β = 12.70, p = 0.005), with parallel results under stricter matching (β = 13.53, p = 0.005). Cumulative auditory experience strongly predicted improvement (both spline terms p < 0.001), but there was no significant Group × Cumulative-use interaction, indicating comparable use-performance slopes in BiCI and UniCI. Thus, the BiCI advantage cannot be attributed to greater device use. A significant Preoperative Score × Sex interaction (β = 14.49, p = 0.021) indicated that higher preoperative performance was more prognostic in males than females; this effect was specific to monosyllables. Disyllables showed a smaller yet significant BiCI effect; sentences trended in the same direction. Age group (20 to 50 versus 50 to 70) did not significantly modify the BiCI effect, and combining age groups was required to detect statistical significance, consistent with an age-independent bilateral benefit.
Conclusions: In a rigorously matched cohort with longitudinal modeling, BiCI yielded a robust advantage over UniCI for quiet monosyllabic recognition that persisted after controlling for behavioral and demographic confounders. The absence of a Group × Use interaction and the pattern restricted to low-context tasks support an interpretation most consistent with neurophysiological mechanisms (e.g., binaural summation, interaural-timing integration, cortical-level plasticity), rather than motivational or selection-related artifacts. Because this advantage was not significantly modified by age, the data suggest that bilateral stimulation offers broadly preserved benefits across adult age ranges. These findings provide clinically actionable evidence to inform counseling and strengthen policy discussions on expanding adult coverage from unilateral to bilateral implantation under public insurance frameworks. Larger, prospective cohorts with mechanistic endpoints remain essential to confirm durability, generalizability, and neural underpinnings.
Keywords: Aging; Cochlear implants; Hearing loss; Longitudinal studies; speech perception.
Copyright © 2026 The American Auditory Society.