C. K. Zogg1, B. Mungo2, A. O. Lidor3, M. Stem3, K. S. Yemul1, A. H. Haider1, D. Molena2 1Johns Hopkins University School Of Medicine,Center For Surgical Trials And Outcomes Research, Department Of Surgery,Baltimore, MD, USA 2Johns Hopkins University School Of Medicine,Division Of Thoracic Surgery, Department Of Surgery,Baltimore, MD, USA 3Johns Hopkins University School Of Medicine,Department Of Surgery,Baltimore, MD, USA
Introduction: More than 1 in 3 adults in the United States, accounting for >106 million people, is obese. From a surgical perspective, the high prevalence of obesity means that operations on this population are common in everyday practice. Despite the assumption that obesity is associated with increased surgical risks, current evidence to suggest that obese patients fair worse is inconclusive. This study sought to examine associations between body-mass index (BMI) and outcomes following major resection for cancer using a nationally-validated outcomes-based database.
Methods: Data from the 2006-2012 American College of Surgeons NSQIP were queried for patients ≥18 years of age with a primary ICD-9 cancer diagnosis and corresponding CPT code for lung surgery, esophagectomy, hepatectomy, gastrectomy, colectomy or pancreatectomy. BMI calculated for included patients were categorized according to World Health Organization classification (Table). Patients were compared first via single logistic regression for differences in 30-day mortality, extended length of stay (LOS), serious morbidity, overall morbidity and isolated morbid conditions among three cohorts: normal vs. (1) underweight, (2) overweight-obese I and (3) obese II-III. Similar methodology was employed using multivariate logistic regression adjusted for clinical/demographic factors and type of resection preformed. Risk-adjusted, stratified analyses for each resection were also considered in addition to an overall propensity score-adjusted logistic analysis (Table).
Results: Consistent with the distribution of BMI in the United States, we identified 529,955 patients of whom 32.06% (169,880) were normal weight, 3.45% (18,284) underweight, 32.52% (172,355) overweight and 17.76% (93,669), 7.51% (39,820) and 4.94% (26,177) obese I-III. Unadjusted, multivariate and propensity-score adjusted logistic regression found that 30-day mortality, extended LOS and serious and overall morbidity were significantly increased in cohort 1. Overall, we did not observe worse surgical outcomes in cohort 2; although, these patients had increased risk for isolated complications such as wound infection, venous thromboembolism, prolonged mechanical ventilation and renal complications. In cohort 3, obese patients experienced a 3-9% increased odds of overall and serious morbidity. Analyses stratified by cancer-resection type reported similar trends.
Conclusion: Evidence-based assessment of outcomes following major resection for cancer suggests that obese patients should be treated according to optimal oncologic standards Surgeons should not be hindered by unproven perceptions of prohibitively increased perioperative risk in this population.
