The relationship between body mass index and postoperative mortality from critical illness

Background Conflicting data exist regarding the effect of body mass index (BMI) on postoperative mortality from critical illness. Few studies have examined this issue in surgical patients specifically. We tested the hypothesis that BMI is associated with mortality from surgical critical illness. Methods Consecutive admissions to a university surgical intensive care unit (SICU) were analyzed from January 2005-August 2006. Admission BMI was analyzed as both a five-level categorical (underweight, < 18.5 kg/m(2); normal weight, 18.5-24.9 kg/m(2); overweight, 25.0-29.9 kg/m(2); obese, 30.0-39.9 kg/m(2); morbidly obese, >= 40 kg/m(2)) and dichotomous (>= 40 kg/m(2) vs. < 40 kg/m(2)) variable among all patients as well as a subgroup of patients with a SICU length of stay (ULOS) >= 4 days. Multivariable logistic regression models were fit to determine the independent effect of BMI group on SICU mortality. Results The total sample size was 946, with 490 patients admitted to the SICU for >= 4 days (51.8%). Of the variables tested, age, acute physiology and chronic health evaluation III score, gender, diabetes mellitus, and need for insulin infusion varied significantly among the five BMI groups. After adjustment for these variables, BMI was not predictive of mortality when analyzed as either a five-level categorical or dichotomous variable, regardless of ULOS. Conclusion BMI is not related to mortality of surgical critical illness. Several factors, including modern ICU care, may mitigate the risks of obesity in the SICU.