Differential association of body mass index with hypoglossal nerve stimulation efficacy by pharyngeal collapse pattern in obstructive sleep apnea

IntroductionUnilateral hypoglossal nerve stimulation (HGNS) is an approved surgical therapy for obstructive sleep apnea (OSA), but its efficacy is limited by obesity and unfavorable (laterally directed) pharyngeal collapse patterns. This study tested whether the effect of body mass index (BMI) on HGNS efficacy differs by pharyngeal collapse pattern, specifically comparing laterally directed and anteroposterior (AP) collapse. MethodsWe pooled data from two independent HGNS cohorts (n=760) to evaluate the interaction between BMI and pharyngeal collapse pattern on HGNS efficacy. Collapse sites were identified via drug-induced sleep endoscopy (DISE) and grouped as laterally directed or AP directed. Multivariable mixed model regression evaluated the association between HGNS efficacy, largely based on optimal setting AHI during titration polysomnogram (primary: percent reduction in apnea-hypopnea index [AHI]; secondary: success defined as [&ge;]50% AHI reduction to <15 events/h) and the interaction between collapse pattern and BMI, adjusting for baseline AHI, partial collapse, surgical center, type of follow-up sleep study, and prior or concomitant pharyngeal surgery. ResultsGreater BMI was associated with a larger reduction in HGNS efficacy in those with lateral collapse (-19.7% [95% CI: -33.2, -6.2] per 5 kg/m{superscript 2}) than in those with AP collapse (-3.8% [-8.0, 0.36]; p-interaction=0.027). In lateral collapse, greater BMI reduced the odds of treatment success (odds ratio [95% CI]: 4.4 [1.4, 14.3] per 5 kg/m{superscript 2}), with no significant effect in AP collapse (1.1 [0.75, 1.5]; p-interaction=0.023). ConclusionThe impact of BMI on HGNS efficacy varies by collapse pattern. Tailoring BMI criteria to collapse pattern may improve HGNS candidate selection.