Introduction:

The NUTRIC score is one of the tools proposed for nutritional risk assessment in critically ill patients. This score uses two severity scores: Acute Physiology and Chronic Evaluation II (APACHE II) and Sequential Organ Failure Assessment score (SOFA), which generates a high risk for collinearity.

Methods:

The goal of the present study is to eliminate the collinearity between the APACHE II and the SOFA used in the NUTRIC score, optimizing such tool. Two alternative scales were developed: modified NUTRIC with an adjustment for APACHE II, by eliminating the points given by the APACHE II and obtaining an adjustment coefficient through linear regression, and the NUTRIC without APACHE II. ROC curves were built to determine the performance of both scores for detecting nutritional risk.

Results:

Two hundred twenty five patients where included in the study. Linear regression between the conventional NUTRIC score and the NUTRIC without APACHE II showed a Pearson coefficient of 0.918, r(2) = 0.843, p < 0.05, m = 1.368, b = 0, and the modified NUTRIC, a Pearson coefficient of 0.918, r(2) = 0.843, p < 0.05. Both ROC curves resulted in equal areas under the curve of 0.951, p < 0.05.

Conclusion:

By eliminating the APACHE II of the NUTRIC score, collinearity is avoided, obtaining two cutoff points of 3 and 5 to detect low risk (sensibility = 99%) and high risk (specificity = 99%), respectively, and a score of 4 to perform an advanced assessment, optimizing its time.

        · resumo em Português | Espanhol     · texto em Espanhol     · Espanhol ( pdf )