Eric S Wise, MD, Adam Sheka, MD, Keith Wirth, MD, Sayeed Ikramuddin, MD, Daniel Leslie, MD. University of Minnesota
Introduction: Open conversion (OC) is an infrequent, highly undesirable necessity during elective minimally invasive bariatric surgery. Characterization of those patients at greatest risk for OC is poor. In this study, we seek to determine those preoperative variables that may portend an increased risk of OC. We also aim to characterize the association of OC on postoperative morbidity and mortality.
Methods and Procedures: Using the robust Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program 2016 dataset, 38,907 patients (≥ 18 years of age, body-mass index [BMI]≥35 kg/m2) who underwent elective, multi-port laparoscopic Roux-en-Y gastric bypass (LRYGB) were included. Preoperative patient and surgeon factors were tested for associations with open conversion using bivariate and subsequent multivariate nominal logistic regression analysis. Bivariate analysis was subsequently used to characterize the association between OC and a panel of 30-day postoperative morbidities. The Mann-Whitney U test and chi-squared test were used for continuous and categorical data comparisons, respectively. All patient variables and outcomes were as defined by the Participant Use Data File User Guide. A criterion of P≤.05 was taken to denote statistical significance.
Results: Among the 38,907 patients, there were 79 (0.2%) OCs. On bivariate analysis, variables associated with OC were advanced age, African American race, greater BMI, gastroesophageal reflux disease, chronic obstructive pulmonary disease, hypertension requiring 3+ medications, therapeutic anticoagulation use, prior obesity/foregut surgery, chronic steroid use and non-independent functional status (P≤.05). The multivariate nominal logistic regression model considering patient variables with sufficient prevalence (n≥5 among OC patients) generated an area under the receiver-operating characteristic curve of 0.75 (n=35,008, r2=.06, chi-squared=66.3, P<.001). Independent risk factors for OC included advanced age (P<.001, odds ratio 1.04, 95% confidence interval [1.02-1.06]), higher BMI (P<.001, 1.06 [1.04-1.08]), and previous foregut/obesity surgery (P<.001, 3.9 [2.3-6.5])(Table 1). OC patients had longer lengths of stay (median 4 vs. 2 days, P<.001) and operative times (median 238 vs. 110 minutes, P<.001), as well as greater rates of perioperative transfusion (P=.03), unplanned ICU admission (P<.001), 30-day mortality (P<.001), 30-day reoperation (P=.02) and 30-day readmission (P<.001)(Table 2).
Conclusions: During elective LRYGB, OC is rare and associated with increased 30-day postoperative morbidity and mortality. We have identified several critical risk factors independently associated with OC. Derived from a robust national database, these data may guide intraoperative and postoperative expectations for the patient and bariatric surgical team.
Presented at the SAGES 2017 Annual Meeting in Houston, TX.
Abstract ID: 95118
Program Number: P095
Presentation Session: Poster Session (Non CME)
Presentation Type: Poster