SA2 - Does the Posterior Airway Space Measured on Cephalometric Imaging Correlate With Apnea-Hypopnea Index Reduction in Patients Treated With Hypoglossal Nerve Stimulation?

Abstract: Statement of Problem: Obstructive sleep apnea (OSA) is a sleep disorder characterized by recurrent episodes of partial or complete upper airway obstruction during sleep, resulting in apnea or hypopnea. The first line of therapy for OSA is continuous positive airway pressure (CPAP), but poor patient adherence limits the effectiveness of CPAP, which has prompted increased interest in other treatment modalities, such as hypoglossal nerve stimulation (HGNS).

In the Stimulation Therapy for Apnea Reduction (STAR) trial, HGNS demonstrated decreases in apnea-hypopnea index (AHI) and improvement in sleep quality in OSA patients. Subsequent studies utilized imaging (ie, lateral cephalometry, cone-beam computed tomography) to suggest craniofacial anatomy – including soft palate thickness, hyoid position, posterior airway space (PAS) – may influence HGNS outcomes. However, substantial heterogeneity exists in literature. Variability in imaging guidelines, including preoperative acquisition, patient positioning, and the precision of measurements using properly identified critical landmarks, has made inferences difficult. Specifically, the role of the PAS, as measured on two-dimensional lateral cephalometry as a predictor for HGNS success, remains unclear. This study aims to correlate the PAS measurement as a predictor for AHI reduction in patients undergoing HGNS therapy.

Methods: A retrospective cohort study was conducted among adult OSA patients treated with HGNS within a single academic oral and maxillofacial surgery practice between 2021 and 2025. Inclusion criteria included confirmed OSA diagnosis, preoperative CBCT imaging, HGNS implantation, preoperative and postoperative sleep study. PAS was measured in a standardized method on lateral cephalometric derived from CBCT. Patients were then divided into two groups based on PAS: PAS = 10 mm and < 10 mm. Baseline and postoperative AHI or respiratory event index (REI) values were obtained from polysomnography or home sleep studies. The central apnea index, mixed apnea index, oxygen nadir, and BMI were extracted. Responder status was defined as those patients with = 50% reduction in AHI or a postoperative AHI < 15 with HGNS.

Data Analysis: The study duration spanned from 2021 to 2025 with 61 patients total with PAS = 10: n = 22 and PAS < 10: n = 39. The primary outcome is the absolute reduction in AHI from pre-HGNS baseline to postoperative follow-up when compared between the two PAS groups. Secondary outcomes included the responder rate and the relationship between BMI and AHI pre- and postoperatively. Continuous variables were compared using independent-samples t-tests assuming unequal variances and within-group changes were evaluated with paired t-tests. Statistical significance was defined as P < .05.

Results: Significant improvement in AHI after HGNS implantation was seen across the cohort. Among patients with PAS = 10 mm, mean baseline AHI reduced from 36.6 to 8.6 events/hour postoperatively. In comparison, patients with < 10 mm showed a decline from 36.3 to 11.7 events/hour postoperatively. Paired t-test demonstrated significant decreases in AHI within both groups (PAS = 10 mm: P = 1.7 x 10^-6; PAS < 10 mm: P = 1.2 x 10^-14). However, postoperative AHI between the groups was not statistically significant (P = .08). Additionally, no significant difference was seen in the absolute change in AHI between the two groups (P = .55). The PAS = 10 mm cohort did show a more favorable improvement, but it was not statically significant.

Conclusions: HGNS implantation produced clinically meaningful reductions in AHI in both PAS cohorts. Patients with PAS = 10 mm demonstrated decreased postoperative AHI and overall greater magnitude of reduction in AHI. However, PAS did not significantly predict HGNS outcomes. In retrospect, a larger PAS measurement may show a favorable airway phenotype, but it is not an independent predictor of treatment success.


References:

1.Lee CH, et al. Otolaryngol Head Neck Surg. 2021;164:1122-1127.

2.Wei Z, et al. Sleep Med Rev. 2025;80:102065.

3.Neelapu BC, et al. Sleep Med Rev. 2017;31:79-90.