Comparison of nutritional screening and diagnostic tools in diagnosis of severe malnutrition in critically ill patients

Rationale: While various nutritional assessment tools have been proposed, consensus is lacking with respect to the most effective tool to identify severe malnutrition in critically ill patients. Methods: We conducted a retrospective study in an adult general intensive care unit (ICU) comparing four nutritional assessment tools: Nutrition Risk Screening (NRS), Nutrition Risk in Critically Ill (NUTRIC), and malnutrition criteria proposed by European Society of Clinical Nutrition and Metabolism (ESPEN) and American Society for Parenteral and Enteral Nutrition (ASPEN). These criteria were tested for their sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) in diagnosis of severe malnutrition, defined as Subjective Global Assessment (SGA) C. Results: Hospitalization records for 120 critically ill patients were analyzed. 60 (50%), 17 (14.2%) and 43 (35.8%) patients were classified as SGA A, B, and C, respectively. The sensitivity in diagnosis of severe malnutrition was 79.1%, 58.1%, 65.1%, and 65.1%, and specificity was 94.8%, 74.0%, 94.8%, and 98.7% for NRS, NUTRIC, ESPEN, and ASPEN, respectively. NRS, ESPEN, and ASPEN had higher PPV (89.5%, 87.5%, and 87.5%, respectively) and NPV (89%, 83%, and 83.5%, respectively) than NUTRIC (PPV 55.6% and NPV 76%). NUTRIC showed the highest correlation with mortality, but none of the tools retained their correlation with mortality after adjustment for potential confounding factors. Conclusions: NRS showed the highest sensitivity and high specificity, PPV, and NPV. NUTRIC had least effective overall performance in diagnosis of severe malnutrition in an ICU setting. A larger population may be required to explore the association between mortality and these nutritional assessment tools. (C) 2020 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.