RESUMO
BACKGROUND: There are no data on the validity of the modified Nutrition Risk in the Critically Ill (m-NUTRIC)-score and Nutritional Risk Screening 2002 (NRS-2002)-score in Iranian intensive care unit (ICU) patients. The Malnutrition Universal Screening Tool (MUST) is still used in most Iranian ICUs. Our goal was to test the validity of these tools in the Iranian ICU population. METHODS: The association between nutrition risk scores and outcomes (longer length of stay [LOS], prolonged mechanical ventilation [MV], and 28-day mortality) was assessed using the multivariable logistic regression. The performance of nutrition risk tools to predict 28-day mortality was assessed using the receiver operating characteristic curve. A logistic regression model was used to test the interaction between nutrition risk category, energy adequacy, and 28-day mortality. RESULTS: Four hundred forty patients were included. Both the m-NUTRIC and NRS-2002 scores were significantly associated with all 3 outcomes (all P < .001). However, no significant association was identified between the MUST and all 3 outcomes (P > .05). The area under the curve for predicting 28-day mortality was 0.806 (95% CI, 0.756-0.851), 0.695 (95% CI, 0.632-0.752), and 0.551 (95% CI, 0.483-0.612) for m-NUTRIC, NRS-2002, and MUST, respectively. Greater energy adequacy was associated with a lower 28-day mortality rate in patients with high m-NUTRIC but not in those with low m-NUTRIC score (P interaction = .015). CONCLUSION: In the Iranian ICU population, the m-NUTRIC score may be a valid tool for identifying patients who would benefit from more aggressive nutrition therapy.