Backscattered Ultrasonographic Imaging of the Tongue and Outcome in Hypoglossal Nerve Stimulation.

Objective Hypoglossal nerve stimulation (HNS) is an increasingly used therapy. However, not all patients undergoing HNS implantation benefit from the treatment, making an improved patient selection a priority. This study investigates whether backscattered ultrasonographic imaging (BUI) can predict the response to HNS therapy. Study Design Cross-sectional study. Setting Secondary and tertiary hospital. Methods In this multicenter cross-sectional study, we recruited patients who had undergone HNS implantation during their scheduled follow-up consultation. HNS therapy parameters were collected. Standardized submental ultrasonographic examination and home sleep apnea testing were performed. The primary outcome was assessing the response to HNS therapy using ultrasonographic features and preoperative patient characteristics. Results In total, 62 participants, 49 male, with a median (interquartile range [IQR]) age of 62 (55-67) and a median (IQR) body mass index of 27.6 (25.2-29.7). The follow-up was a median (IQR) of 19.5 (4.8-41.4) months after implantation. The apnea-hypopnea index (AHI) was preoperatively 40.5 (29.8-58.0) and reduced at follow-up to 21.0 (11.0-35.3). In total, 42% were responders to HNS. Preoperative AHI (34.8/hour vs 49.3/hour, r = 0.44) was significantly higher in nonresponders than in responders. The average prediction accuracy of HNS therapy based on baseline AHI alone was 71%. A lower backscatter signal, indicating less fat deposition in the tissue, was observed in the responder group. When the baseline AHI and backscatter signal were combined, the prediction accuracy of response to the HNS reached 78%. Conclusion The combination of tissue composition analyzed using the backscattered signal and the preoperative AHI is highly predictive for determining the HNS treatment response. Trial Registration ClinicalTrials.gov identifier NCT06154577.