Introduction The prognostic significance of a high body mass index (BMI) in adult intensive care unit (ICU) patients is controversial. The currently used prognostic scoring systems for ICU patients do not include BMI. We hypothesized that obesity is associated with a higher morbidity and mortality.
Methods We identified 376 patients (> 18 years) admitted to the ICU for > 3 days at OUHSC with the diagnosis of severe sepsis or trauma from January until August 2004 by review of hospital discharge logs; 56 with incomplete records were excluded. Clinical characteristics and outcome of the remaining 320 patients were determined by review of the medical record. We classified the patients as underweight (BMI < 18.5), normal (BMI: 18.5-24.9), overweight (BMI: 24.9-29.9), and obese (BMI > 30). Data are mean ± SEM.
Results Gender, ethnic background, vital signs, need for mechanical ventilation (MV), GCS and severity of illness scores were similar between obese and nonobese. Mortality was similar (31.7 vs. 32.5%, p = .8), but obese patients tended to have a longer ICU length of stay (LOS, 12.2 ± 1.1 vs. 10.4 ± 1.04, p = .17) and a longer duration of MV (7.4 ± 1 vs. 6.6 ± 1, p = .35). Trauma patients (n = 204) were younger than the medical (38.2 ± 1.2 vs. 54.6 ± 1.4 years, p < .05) and the ICU LOS was longer for obese trauma patients (12.7 ± 1.3 vs. 10.2 ± 1.3 days; p < .05), with a similar duration of MV (8.2 ± 1.3 vs. 6.9 ± 1.4 days; p = .34). In the medical group, the ICU LOS for the obese patient was 2 days longer (11.2 ± 1.9 vs. 9.4 ± 1.7; p = .2), with no difference in the number of ventilator days (7.7 ± 1.4 vs. 6.0 ± 1.3 days; p = .6). Overall, the obese group also had increased plasma creatinine and glucose levels and a lower PaO2 (p < .05).
Conclusion Obesity, which is common in the United States, may be associated with an increased ICU LOS and consequently increased health care costs.