Comparison of sound processing strategies for osseointegrated bone conduction implants in mixed hearing loss: multiple-channel nonlinear versus single-channel linear processing.

Objectives: Evaluation of a single-channel linear bone conduction implant sound processor (S-BCI) and a multiple-channel nonlinear bone conduction implant sound processor (M-BCI) with objective and subjective measures in patients with mixed hearing loss.

Study Design: In total, 20 patients with mixed hearing loss were included in the study. For either sound processor aided thresholds and speech perception in quiet with monosyllables were measured. Speech perception in noise was measured with sentences. Two different configurations were used: speech and noise at 0° (S0N0) and speech at 0° and noise at 180° (S0N180). The M-BCI was tested in both omnidirectional and directional mode. Patients were first fitted with the S-BCI and evaluated 3 weeks later. The M-BCI was fitted and, again 3 weeks later, evaluated. Subjectively, patients compared both sound processors with the APHAB questionnaire.

Results: Aided thresholds were similar for both sound processors in the low- and mid-frequency range. For speech in quiet, no significant differences between both sound processors were observed. For speech in noise in the S0N0 condition, the M-BCI-thresholds were 1.7 dB (SD, 2.2dB; p = 0.002) more favorable than with S-BCI. For the S0N180 configuration, an improvement of 5.8 dB (SD, 2.8 dB; p < 0.001) was seen for the directional mode relative to S-BCI. The APHAB showed statistically significant subjective improvement with the M-BCI on all subscales relative to S-BCI.

Conclusion: Speech intelligibility in noise is better with M-BCI than with S-BCI. This was attributed to better high-frequency gain provided by the M-BCI. Improved signal processing strategies may have contributed to subjective preference for the M-BCI.