Comparison of the NUTRIC-PNI score and NUTRIC score to assess nutritional risk and predict 28-day mortality in critically ill patients: a prospective cohort study in Vietnam

Background: Adequate nutritional support is crucial in enhancing clinical outcomes for critically ill patients. The nutrition risk in the critically ill (NUTRIC) score serves as a risk assessment tool to identify critically ill patients at high risk of malnutrition. This score incorporates Interleukin-6 (IL-6) as a key component, which may be costly and less accessible in low- and middle-income countries. Methods: A prospective cohort study of patients admitted to tertiary hospital. IL-6 in the NUTRIC score was replaced by the Prognostic Nutritional Index (PNI) to establish the NUTRIC score incorporating the PNI (NUTRIC-PNI) score. The outcomes are determining the concordance in identifying nutritional risk and comparing the prognostic value between NUTRIC-PNI and NUTRIC scores in predicting 28-day mortality in critically ill patients. Results: A total of 123 patients were analyzed with a median age of 69±16 years old and 72.4% of males. The Cohen's Kappa coefficient of 0.869 indicates strong agreement between the two scoring systems. The NUTRIC-PNI and NUTRIC scores demonstrated comparable predictive accuracy for 28-day mortality, each achieving an area under the curve of 0.68 (95% CI, 0.59-0.78 for NUTRIC-PNI and 0.59-0.77 for NUTRIC; P<0.001. The logistic regression analysis demonstrated that the NUTRIC-PNI and its high-risk classification were independent predictors of 28-day mortality (OR, 1.38; 95% CI, 1.05-1.81; P=0.021 and OR, 4.21; 95% CI, 1.55-11.43; P=0.005, respectively). Conclusions: The concordance and performance of the NUTRIC-PNI score are comparable to the NUTRIC score. Furthermore, the NUTRIC-PNI score and its high-risk classification were independent predictors of 28-day mortality in critically ill patients. In addition to the modified NUTRIC score, the NUTRIC-PNI score is feasible and beneficial for routine clinical practice, particularly in resource-limited settings.