Predictive value of abdominal obesity vs. body mass index for determining risk of intensive care unit mortality

Objective: To explore whether sagittal abdominal diameter as a marker of abdominal obesity is a risk factor for death and morbidity in patients in the intensive care unit and a better outcome determinant for obese patients than body mass index. Design: Prospective, observational study from April 2008 to January 2009. Setting: Two general intensive care units, both in Toulouse University Hospitals, France. Participants: All adult patients admitted in the two intensive care units except those routinely discharged within 48 hrs or those having conditions with possible effect on anthropometric indices. Interventions: Measurement of the sagittal abdominal diameter at admission allowed us to divide the studied population into abdominally obese, underweight, and control groups. Measurements and Results: The primary outcome measure was mortality in the intensive care unit until day 60 after admission. Secondary outcomes were morbidity and length of stay in the intensive care unit. Among 503 patients admitted, 403 were included. At admission, age, diabetes, dyslipidemia, hypertension, Simplified Acute Physiology Score II, and McCabe scores were higher in the abdominally obese group (n = 109) than in the control group (n = 277). The rate of death was higher in the abdominally obese group compared to control (44% vs. 25.3%; p .01). After adjustment for age, simplified acute physiology score, II and McCabe score, a multivariate analysis showed an increased risk of death in the abdominally obese group (adjusted odds ratio, 2.12; 95% confidence interval, 1.25-3.60). A body mass index > 30 kg/m(2) was not an independent risk factor for death. During the stay in the intensive care unit, incidence of acute renal failure and abdominal compartment syndrome were higher in the abdominally obese group. Conclusion: A high sagittal abdominal diameter, and not a high body mass index, is an independent risk factor of death in critically ill patients. (Crit Care Med 2010; 38:1308-1314)