Abstract
Background/Aims The two most common bariatric surgical procedures are laparoscopic gastric band (LAGB) and laparoscopic gastric bypass (RYGB). Single site comparisons suggest that RYGB may result in greater weight loss at the expense of more short-term complications, whereas LAGB may be associated with less weight loss and more long-term complications. It is unclear whether these results extend to other populations.
Methods This is a multi-site retrospective cohort investigation of LAGB vs. RYGB. Outcomes are: 30-day operative complications; rehospitalization at 1 and 2 years; and weight loss at 2 years. Multivariable logistic regression, Cox proportional hazards, and repeated measures will be used to assess outcomes as a function of type of surgery.
Results From 2005–2009, there were 1,521 LAGB and 5,963 RYGB procedures at 11 sites. Mean pre-surgical BMI was 42.6 and 44.6, and mean age was 47.0 and 45.8 for each procedure respectively. RYGB patients were more likely to have diabetes, hypertension, gastric reflux, and sleep apnea. RYGB patients had more commercial insurance (89% vs. 72%), were less likely to be Caucasian (67% vs. 73%), and were 84% (vs. 82%) female. Results at 30-days for LAGB vs. RYGB respectively were: Failure to discharge, 0 (0%) vs. 8 (0.1%); mortality, 0 (0%) vs. 1 (0.02%); thrombotic diagnoses, 6 (0.5%) vs. 63 (1.1%); and reintervention, 10 (0.9%) vs. 30 (0.5%). Mean long-term follow-up BMI was 37.2 for LAGB and 33.3 for RYGB. All differences were statistically significant at P = 0.05 in bivariate models.
Conclusions Across multiple sites, individuals with higher BMI and morbidity burden were more likely to receive RYGB surgery. Preliminary results suggest that long-term weight loss and certain short-term complications are significantly greater for RYGB than LAGB procedures, and that LAGB procedures are more likely to have 30-day reintervention. Multivariable comparisons adjusting for pre-operative covariates and follow up time will better clarify risks and benefits of these procedures.
