Triphasic Stimulation in Cochlear Implants: A Comparative Evaluation of Programming Levels, Speech Intelligibility, and Spectral-Temporal Resolution in Users With and Without Inner Ear Malformations.

Abstract

Objectives: Nonauditory or facial-nerve sensations limit the electrical dynamic range of some cochlear implant (CI) users with inner ear malformations (IEMs). Triphasic pulses, which redistribute charge over three phases, can raise comfortable current levels without provoking side effects, but their effect on everyday listening remains unclear. We asked whether triphasic stimulation improves speech perception and spectral-temporal processing in CI users with IEM and whether the same change benefits-or degrades-performance in users with normal inner ear anatomy (NIE). We hypothesized that triphasic pulses would enhance speech outcomes in the IEM group but leave the NIE group unchanged or mildly worse.

Design: Forty CI users (20 IEM, 20 NIE; mean ages ± SD = 18.1 ± 8.3 and 15.8 ± 3.4 years) completed a within subject, counterbalanced comparison of two clinical programs: standard biphasic versus triphasic stimulation. Outcome measures included average aided thresholds, CI mapping parameters (threshold, most comfortable level, stimulation rate, phase duration, dynamic range), Turkish Matrix Sentence Test (TURMatrix) in quiet and in noise, speech recognition threshold, speech discrimination score, spectral temporally modulated ripple test (SMRT), and random gap detection thresholds (RGDT). Linear mixed-effects models with Holm adjustment (α = 0.05, two-tailed) evaluated pulse shape, group, and interaction effects.

Results: Triphasic stimulation elevated most comfortable levels and widened the overall dynamic range in both cohorts, while no significant change was observed in threshold levels. Among listeners with IEM, triphasic pulses increased TURMatrix in quiet accuracy by 19%, lowered the speech reception threshold by 9.5 dB, improved the adaptive signal to noise ratio on the TURMatrix in noise by 5.8 dB, and raised the speech discrimination score by 10.8% (all Holm‑adjusted p ≤ 0.01). RGDT results improved (-6.6 msec, p = 0.02), while performance on the SMRT results remained unchanged. For NIE listeners, triphasic stimulation reduced TURMatrix in quiet accuracy by 9.6%, lowered the speech discrimination score by 6.3%, and elevated speech reception threshold by 4 dB (all p ≤ 0.03). TURMatrix in noise adaptive signal to noise ratio, RGDT, and SMRT results were unaffected.

Conclusions: Triphasic stimulation substantially enhances speech intelligibility in CI users with IEMs while offering no perceptual advantage-and a slight disadvantage-for users with NIE. Programming CIs with triphasic pulses may therefore be a valuable strategy for malformation cases, enabling greater loudness comfort and improved communication without resorting to electrode deactivation or reduced current levels.