Abstract
BACKGROUND: While the safety and short-term weight-loss outcomes of endoscopic sleeve gastroplasty (ESG) are now well accepted, the modern uptake and its impact on healthcare utilization continue to remain poorly characterized, particularly in contrast to laparoscopic sleeve gastrectomy (LSG). METHODS: After identifying ESG and LSG cases using a combination of CPT and procedural code variables, non-parsimonious multivariable logistic regression models were conducted to identify predictors of serious complications and outpatient emergency department (ED) visits. RESULTS: A total of 506,597 patients met inclusion criteria (2285 ESG and 504,312 LSG). ESG patients were younger (42.6 ± 11.8 years versus 45.1 ± 10.7 years, p < 0.0001), had a lower BMI (39.5 ± 7.5 kg/m(2) versus 44.9 ± 7.7 kg/m(2), p < 0.0001), and were primarily female (86.2% versus 81.5%, p < 0.0001). Multivariable regression modeling revealed no differences in serious complications between ESG and LSG. ESG was protective against ED visits (OR 0.66; 95% 0.54-0.80; p < 0.0001). There were no other differences with respect to mortality or other healthcare metrics, such as outpatient rehydration, between procedures. CONCLUSIONS: Multivariable logistic regression modeling of prospectively collected 30-day outcomes in a large multi-institutional database demonstrates that ESG does not confer additional risk of 30-day serious complications compared to LSG and has lower odds off post-procedural ED utilization in the same comparison.