OBJECTIVE: To examine the clinical risk index for babies (CRIB) predictive value for hospital death; to identify the score variable with the best predictive value and to compare CRIB score capability to predict hospital mortality to birth weight, gestational age and base excess. METHODS: CRIB score was obtained through a prospective way from 100 newborns with birthweight of 1,500 g or less or gestational age less than 31 weeks, who were admitted consecutively to the Neonatal Unit of Hospital das Clínicas, Universidade Federal do Paraná. RESULTS: Fifty-five newborns were females and 45 were males, the average birthweight was 1,078 +/- 0.277 g and gestational age was 29.2 +/- 2.8 weeks. Twenty-one patients died. The mortality rate in the CRIB groups 1, 2, 3 and 4 was, respectively 6.6%; 46.2%; 87.5% and 100.0%. The score accuracy for mortality was confirmed (area under the ROC curve=0.877) and the best score variable to predict hospital death was maximum base excess (area under the ROC curve=0.795). Compared with birthweight and gestational age, CRIB was significantly better to predict mortality. CONCLUSIONS: Besides being useful to predict hospital death, CRIB was a simple score to be applied. Based on these results, we recommend its inclusion in the routine of neonatal units.
OBJECTIVE: To examine the clinical risk index for babies (CRIB) predictive value for hospital death; to identify the score variable with the best predictive value and to compare CRIB score capability to predict hospital mortality to birth weight, gestational age and base excess. METHODS: CRIB score was obtained through a prospective way from 100 newborns with birthweight of 1,500 g or less or gestational age less than 31 weeks, who were admitted consecutively to the Neonatal Unit of Hospital das Clínicas, Universidade Federal do Paraná. RESULTS: Fifty-five newborns were females and 45 were males, the average birthweight was 1,078 +/- 0.277 g and gestational age was 29.2 +/- 2.8 weeks. Twenty-one patients died. The mortality rate in the CRIB groups 1, 2, 3 and 4 was, respectively 6.6%; 46.2%; 87.5% and 100.0%. The score accuracy for mortality was confirmed (area under the ROC curve=0.877) and the best score variable to predict hospital death was maximum base excess (area under the ROC curve=0.795). Compared with birthweight and gestational age, CRIB was significantly better to predict mortality. CONCLUSIONS: Besides being useful to predict hospital death, CRIB was a simple score to be applied. Based on these results, we recommend its inclusion in the routine of neonatal units.