Weight loss fails to resolve obstructive sleep apnea (OSA) in most patients; however, it is unknown as to whether weight loss differentially affects OSA in the supine compared with nonsupine sleeping positions. We aimed to determine if weight loss in obese patients with OSA results in a greater reduction in the nonsupine apnea/hypopnea index (AHI) compared with the supine AHI, thus converting participants into supine-predominant OSA. Post hoc analysis of data from a randomized controlled trial assessing the effect of weight loss (bariatric surgery vs. medical weight loss) on OSA in 60 participants with obesity (body mass index: >35 and <55) with recently diagnosed (<6 months) OSA and AHI of ≥ 20 events/hour. Patients were randomized to very low calorie diet with regular review (n = 30) or to laproscopic adjustable gastric banding (n = 30) with follow-up sleep study at 2 years. Eight of 37 (22%) patients demonstrated a normal nonsupine AHI (<5 events/hour) on follow-up compared to 0/37 (0%) patients at baseline (p = .003). These patients were younger (40.0 ± 9.6 years vs. 48.4 ± 6.5 years, p = .007) and lost significantly more weight (percentage weight change -23.0 [-21.0 to -31.6]% vs. -6.9 [1.9 to -17.4], p = .001). The percentage change in nonsupine AHI was greater than the percentage change in supine AHI (-54.0 [-15.4 to -87.9]% vs -33.1 [-1.8 to -69.1]%, p = .05). However, the change in absolute nonsupine AHI was not related to change in absolute supine AHI (p = .23). Following weight loss, a significant proportion (22%) of patients with obesity have normalization of the nonsupine AHI. For these patients, supine sleep avoidance may cure their OSA.