[NEMS: a new predictor of mortality in the critical patient?]

Numerical scales are commonly used in intensive care units to predict hospital mortality and to assess the therapeutic effort and care that critically ill patients require. The aim of this work was to study whether the NEMS value can be used as a predictor of mortality, comparing it with the APACHE II. A prospective study in a 24 intensive care unit beds was conducted. The APACHE II and NEMS values were stratified into three levels. Demographic data and the first 24 hours values of APACHE II and NEMS scales were collected. A total of 1257 patients were included, 16.4% of whom died. 69.6% were surgical; median stay was 2 days (1-4). Median age was 66 years (50-77), 59.3% were men. The median APACHE II and NEMS for the living and the dead in the subsequent course was 10 (6-20) versus 22.5 (17.25 to 29) (p <0.001) and 24 (18-29) versus 34 (25 to 39.7) (p<0.001) respectively. The correlation between both scales was rho=0.457 (p<0.01). Logistic regression controlled for age, sex and APACHE II showed an OR of 3.1 (95% CI: 1.5-6.6) only for high NEMS, compared to the lowest level. According to the results NEMS should not be used as a predictor of mortality, although the risk of death increases by three times with high NEMS.