Clinical Predictors of OSA Treatment Success Following Implantation of a Hypoglossal Nerve Stimulation Device

Avrahan Boroosan,Anna M Salapatas,Michael Friedman
DOI: https://doi.org/10.1177/01945998221087594
Abstract:Objective: To identify prognostic indicators associated with successful hypoglossal nerve stimulation (HGNS) therapy to treat obstructive sleep apnea (OSA), focusing on patients' physiologic response to awake tongue protrusion. Study design: Retrospective chart review. Setting: Tertiary care center. Methods: We included consecutive patients with moderate-severe OSA who underwent HGNS implantation from December 2017 to December 2019. Data abstracted include standard demographics, body mass index (BMI), pre- and postoperative apnea-hypopnea index (AHI), and Friedman tongue position (FTP). Additionally, change in hypopharyngeal cross-sectional area on awake tongue protrusion was abstracted. Patients protruded their tongues, and the physician visualized change. Positive change in hypopharyngeal cross-sectional area was documented as +1 and a negative change as -1. Chi-square tests for independence and logistic regression analysis were performed to determine indicators of successful surgery. Results: Thirty-nine patients were included in this study. Mean ± SD AHI decreased significantly from 43.1 ± 17.36 to 9.18 ± 8.18. Surgical success was achieved in 79.5% of patients. Variables analyzed included BMI >32, preoperative AHI, FTP, and change in hypopharyngeal cross-sectional area on awake tongue protrusion (positive, 65.8%; negative, 34.2%). Positive predictors of success were positive change in hypopharyngeal cross-sectional area (P = .0133), severe OSA (P = .0290), and FTP IIb (P < .0001). Negative predictors were BMI >32 (P = .041) and negative change in hypopharyngeal cross-sectional area (P = .02). Conclusion: Positive change in hypopharyngeal cross-sectional area on awake tongue protrusion and severe baseline AHI were positive predictors of successful HGNS therapy. Negative change in hypopharyngeal cross-sectional area on awake tongue protrusion and BMI >32 were negative predictors.
What problem does this paper attempt to address?