Mehraneh D Jafari, MD, Monica T Young, MD, Vinh Q Nguyen, PhD, Brian R Smith, MD, Michael J Stamos, Ninh T Nguyen, MD
University of California, Irvine
INTRODUCTION
The rapid adoption of the laparoscopic approach for bariatric operations over the past decade has been accompanied by an exponential growth in the number of procedures performed annually. Multiple studies have examined the effects of volume on surgical outcomes for bariatric surgery. However, these studies were analyzed in the era of open surgery and the absence of national accreditation centers. It has been shown that volume is an independent predictor of serious complications. Mortality associated with bariatric surgery has decreased tremendously over the past decade. The purpose of this study is to demonstrate the effect of volume on surgical outcomes in bariatric surgery within the era of laparoscopy and national accreditation.
METHODS AND PROCEDURES
Using the Nationwide Inpatient Sample, a retrospective review of elective admission of bariatric surgical cases was conducted between 2006-2010. Patient demographics, comorbidities, serious postoperative morbidity, and in-hospital mortality were reviewed. Outcomes were analyzed according to low volume (LVH, <50 cases), medium volume (MVH, 50-100 cases) and high volume hospitals (HVH, >100 cases). A multivariate analysis was conducted to estimate and test the association of volume on mortality and serious morbidity while controlling for age, gender, hospital factors (teaching, size, and location), comorbidities, and procedure type (stapling and non-stapling). Separate a priori specified models were fit to consider the effect of volume for stapling (gastric bypass and sleeve gastrectomy) and non-stapling procedures (gastric band).
RESULTS
Among the estimated 381,674 cases sampled, 74% of cases were performed in HVH. Gastric bypass and sleeve gastrectomy accounted for 72% of cases. Patient age, gender distribution, race, hospital type and comorbidity score were similar for all groups. Hospital charges were highest in the LVH, while length of stay and anastomotic leak were similar among the three groups. In-hospital mortality was higher in the LVH (0.14%) compared to HVH (0.06%). Using multivariate analysis and controlling for confounding variables, procedures performed in a LVH were associated with 2.9 fold increase in mortality rates (95% CI [1.5, 5.7]; p<0.02) and a 1.3 fold increase in serious morbidity (95% CI [1.2, 1.5]; p<0.01) compared to HVH. Stapling procedures performed in LVH were associated with a 2.9 fold increase in mortality rates (95% CI [1.4, 6.1]; p<0.04) and a 1.3 fold increase in serious morbidity (95% CI [1.1, 1.4]; p<0.01) compared to HVH. Non-stapling procedures performed in LVH are associated with a 1.6 fold increase in mortality rates (95% CI [1.2, 2.2]; p<0.01) compared to HVH.
CONCLUSION
In the era of laparoscopy, hospitals with high case volumes continue to have improved serious morbidity and mortality. We were unable to differentiate if the improved outcomes at high volume centers are related to their higher volume or their status of accreditation as centers of excellence.
Session: Podium Presentation
Program Number: S050