Complementarity of NUTRIC score and Subjective Global Assessment for predicting 28-day mortality in critically ill patients

Aims: To compare the prognostic power of nutritional screening (NUTRIC score) and Subjective Global Assessment (SGA), executed alone or their complementarity, for predicting 28-day mortality risk in ICU patients. We also aimed to identify the cut-off point obtained in the NUTRIC that presented the best validity parameters for predicting mortality in this population.

Methods: A sample of 159 patients was evaluated in the first 24 hours of ICU admission. Modified NUTRIC score was performed (without interleucina-6). ROC curve and Youden criterion were used to identify the best cut-off point. Poisson regression and the number needed to screen (NNS) were used to test the complementarity between the tools and their ability to predict 28-day mortality.

Results: A sample of 159 patients was evaluated (51% male, 56.6 ± 20 years) and the APACHE II, SOFA and NUTRIC score medians were 22 (IQR:15;26), 6 (IQR:2;9) and 3 (IQR: 2;5), respectively. Almost 60% of the patients were malnourished (SGA B or C) and 32.7% died during 28-day follow-up. The area under ROC curve for NUTRIC score was 0.79. Using a new cutoff (NUTRIC score ≥ 4), patients with nutritional risk have a 28-day mortality risk almost 6 times higher than subjects without nutritional risk. Patients classified as SGA C showed a 28-day mortality risk 2.19 times higher compared to nourished ones. Evaluating the complementarity of the tools, patients classified as nutritional risk (NUTRIC score ≥ 4) and SGA C showed a 28-day mortality risk 7 times higher and a lower NNS when compared to those patients with a NUTRIC < 4 and any SGA category.

Conclusions: A new cutoff value was identified for this population. Simultaneous SGA assessment in patients with nutritional risk may enhance the predictive power of 28-day mortality, providing better identification of higher risk patients who may benefit from a more aggressive nutritional therapy.