Subjective Global Assessment Can Be Performed in Critically Ill Surgical Patients as a Predictor of Poor Clinical Outcomes.

Abstract

There is limited evidence on the applicability of subjective global assessment (SGA) in critically ill patients, despite its good performance in diagnosing undernutrition in clinical and surgical patients. Our aim was to evaluate the association between SGA and anthropometric measures and the performance of SGA in predicting clinical outcomes and mortality in critically ill surgical patients. This observational prospective study was carried out with patients admitted to the surgical intensive care unit (SICU) of a Brazilian hospital. Nutrition assessment comprised anthropometric measures (weight and height for body mass index [BMI] calculation, mid-arm circumference [MAC] and calf circumference [CC]), and the SGA. Patients were followed up until hospital discharge for measurement of the following outcomes: length of stay (LOS), SICU LOS, and hospital mortality. This study included 76 surgical patients admitted to an SICU (64.9% males; 87% white ethnicity; mean age, 60.36 ± 16.24 years). According to the SGA, the prevalence of undernutrition was 60.5% (moderate undernutrition = 38.2%; severe undernutrition = 22.4%), and well-nourished 39.0%. Patients with severe undernutrition had lower values of current weight, BMI, MAC, and CC compared with well-nourished patients. The presence of undernutrition increased the risk of LOS >31 days by 2.57 (1.38-4.77) times, but it was not associated with mortality. Patients with severe undernutrition had lower current weight, BMI, MAC, and CC compared with well-nourished patients. Undernutrition did not increase the risk of death; however, it did increase the relative risk for a longer hospital stay by 2.5.