Abstract MP089: Sleep Apnea is Reduced Following a Behavioral Weight Loss Intervention but Impedes Weight Loss Outcomes

Abstract

Introduction: Several trials have documented that behavioral weight loss interventions reduce obstructive sleep apnea (OSA) severity. However, despite the known bidirectional association between body weight and OSA severity, few studies have addressed whether the presence of OSA impedes weight loss outcomes in a lifestyle intervention. Hypothesis: We hypothesized that a behavioral weight loss intervention would significantly reduce OSA severity, but that the presence of OSA would lead to poorer weight loss outcomes. Methods: Overweight and obese adults (N = 101; 50.6 ± 10.3 y, body mass index: 34.1 ± 4.6; 91.1% female, 80.2% white) who participated in a 12-mo behavioral weight loss intervention study and had OSA assessed were included in these secondary analyses. Participants wore a limited-channel home sleep testing device (ResMed ApneaLink Plus) for one night at baseline, 6 and 12 mo. Measures of OSA severity included the apnea-hypopnea index (AHI), oxygen desaturation index (ODI), and snoring index (i.e., number of snoring events divided by flow recording time). Weight change at 6 and 12 mo was expressed as percentage change from baseline. Linear mixed models were used to evaluate the effect of the intervention on OSA measures, and ANCOVA models examined the effect of weight change on OSA and the effect of OSA on subsequent weight loss. All analyses were adjusted for age, sex, and race. Results: Baseline AHI, ODI, and snoring index were 6.6 ± 7.1, 8.7 ± 8.3, and 109.9 ± 122.0, respectively; 50.5% of the sample had a baseline AHI ≥ 5. AHI and ODI, but not the snoring index, were significantly reduced by the end of the intervention (12-mo changes: -1.3 [P < .05], -2.3 [P < .01], -2.8 [P = .82], respectively). Weight loss at 6 and 12 mo were 9.0 ± 6.0% and 9.1 ± 8.3%, respectively. Participants with ≥ 5% weight loss at 6 mo had significant reductions in AHI (P = .02), ODI (P < .05), and the snoring index (P < .001) at 6 mo; adults with ≥ 5% weight loss at 12 mo had a significant reduction in the snoring index at 12 mo (P < .01), but not AHI (P = .29) or ODI (P = .39), relative to baseline. Participants with a baseline AHI ≥ 5 lost significantly less weight at 6 mo compared to those without OSA at baseline (-8.4% vs. -11.4%; P < .01), and those with an AHI ≥ 5 at 6 mo lost significantly less weight at 12 mo compared to those without OSA at 6 mo (-7.1% vs. -10.9%; P = .04). Conclusion: Behavioral weight loss interventions in overweight and obese adults lead to significant improvement in OSA severity. However, the presence of at least mild OSA is associated with less favorable weight loss in this population. Future work should explore the mechanisms underlying the blunted weight loss in overweight and obese adults with OSA.