TABLE IX.A.2.

Evidence for BMI as a criterion for surgical candidacy

Study	Year	LOE	Study design	Study groups	Clinical end-points	Conclusion
Choi et al.1486	2016	2a	Systematic review	OSA patients who underwent single level UPPP (n = 15 studies)	Surgical success as determined by AHI < 20 and AHI decrease by 50%	BMI does not affect surgical success in the immediate period; only Friedman 3 and low hyoid were negative predictors
He et al.812	2019	2a	Systematic review	Studies of patients with OSA who had single level UPPP (n = 11 studies)	Surgical success as determined by AHI < 20 and AHI decrease by 50%	BMI negatively affects long term surgical success
Camacho et al.1490	2015	2a	Systematic review	Studies of morbidly obese (BMI > 40) patients with OSA who underwent MMA or tracheostomy 1. MMA (n = 34 patients) 2. Tracheostomy (n = 14 patients)	Surgical success determined by AHI < 20 and decrease by 50%	AHI significantly improved after MMA or tracheostomy in morbidly obese patients. However, studies are limited, making it difficult to draw definitive conclusions
Browaldh et al.1487	2013	2b	RCT	Moderate to severe OSA patients with BMI < 36 kg/m2, Friedman stage 1–2. Separated by Friedman stage and BMI 30 kg/m2 1. Control group, no intervention (n = 33) 2. Intervention, UPPP (n = 32)	Primary outcome change in AHI	Intervention resulted in mean 60% improvement in AHI compared to 11% in control, and this was independent of BMI, tonsil size, or Friedman stage
Huntley et al.1489	2018	3b	Case control	OSA patients undergoing hypoglossal nerve stimulation 1. BMI < 32 kg/m2 (n = 113) 2. BMI > 32 kg/m2 (n = 40)	Surgical success, > 50% reduction in AHI and AHI < 20 postoperatively	There was no difference in surgical success between the two groups
Heiser et al.1488	2019	3b	Case control	OSA patients who underwent hypoglossal nerve stimulation in the ADHERE registry (n = 508, mean BMI 29.3)	Analysis of predictors of surgical success and failure	On posthoc analysis, for each 1 unit increase in BMI, there was a 9% decrease in odds of surgical success. Increasing age was a positive predictor for success
Shie et al.527	2013	3b	Case control	Patients with severe OSA, failed CPAP, and underwent UPPP 1. BMI < 27 kg/m2 (n = 56) 2. BMI > 27 kg/m2 (n = 61)	Surgical success, > 50% reduction in AHI or AHI < 20 postoperatively	Obesity was a significant negative predictor of treatment success
Li et al.1491	2006	3b	Case control	OSA patients with BMI < 40kg/m2 who underwent UPPP (n = 110)	Surgical success as determined by AHI < 20 and AHI decrease by 50%	Friedman staging had a significant predictive value on surgical success, and BMI was not predictive (no patients with BMI 40)
Martino et al.1492	2006	4	Case series	Obese, OSA patients with average BMI > 30 kg/m2 (mean 36 kg/m2) and tonsillar hypertrophy (n = 7)	Posttreatment reduction in AHI compared to baseline	Six of seven patients had reduction in AHI by at least 50%
Vicente et al.1493	2006	4	Case series	Severe OSA patients who underwent tongue base suspension and UPPP (n = 54, BMI < 40)	Surgical success as determined by AHI < 20, AHI decrease by 50%, and ESS < 11	78% of patients experienced surgical success at 3 years follow-up, and BMI at baseline was the only predictor of surgical success
Friedman et al.1485	2002	4	Case series	Patients with OSA undergoing UPPP (n = 151) 1. Friedman stage 1 (n = 31) 2. Friedman stage 2 (n = 79) 3. Friedman stage 3 (n = 74)	Surgical success as determined by RDI < 20 and RDI decrease by 50%	Lower Friedman stages at any BMI <40 kg/m2 did not affect outcome
Chandrashekariah et al.1494	2016	4	Case series	OSA patients who had undergone UPPP, obese (BMI ≥ 30 kg/m2, average BMI 41 kg/m2), and had persistent elevated AHI (n = 11)	Improvement in CPAP adherence	Eight of 11 patients had improvement in PAP adherence, with increased mean of 48.6 min per night