The obesity paradox in surgical intensive care unit patients

Purpose: To investigate the possible impact of obesity, as assessed by body mass index (BMI), on outcome in surgical intensive care unit (ICU) patients. Methods: Prospectively collected data from all consecutive adult patients admitted to our ICU between January 2004 and January 2009 were analysed retrospectively. BMI was calculated using the formula: BMI = body weight/height(2) (kg/m(2)), and patients were grouped as underweight (<18.5 kg/m(2)), normal weight (18.5-24.9 kg/m(2)), overweight (25-29.9 kg/m(2)), obese (30-39.9 kg/m(2)) and very obese (>= 40 kg/m(2)). Results: Among the 12,938 patients who were admitted to our ICU during the study period, 9,935 (76.8%) had complete height and weight data and constituted the study group. The mean BMI was 27.1 +/- 5.0 kg/m(2). Overall, 34.4% of the study population had normal BMI, 1.8% were underweight, 41.2% were overweight, 20.8% were obese and 1.8% were very obese. The ICU mortality rate was similar among BMI subgroups, but hospital mortality was higher in underweight patients than in patients with normal BMI (17.8% versus 11.1%, P = 0.006). On multivariate Cox regression analysis, being overweight [hazard ratio (HR) 0.86, 95% confidence interval (CI) 0.74-0.99, P = 0.047] or obese (HR = 0.83, 95% CI = 0.69-0.99, P = 0.047) was independently associated with lower 60-day in-hospital mortality, with normal BMI as the reference category. Risk of death increased in very obese patients, especially after neurosurgical procedures (HR = 0.3, 95% CI = 1.06-8.48, P = 0.039). Conclusion: In this cohort of surgical ICU patients, being overweight or obese was associated with decreased risk of 60-day in-hospital mortality.