Abstract:

Objective: A number of illness severity scores have been established to predict mortality and morbidity in neonatal intensive care units (NICUs). The objective of this study was to compare SNAPPE-II (Scores for Neonatal Acute Physiology- Perinatal Extension-II) and CRIB- II (Clinical Risk Index for Babies-II), the latest versions of European and American scoring systems, in predicting hospital mortality and overall morbidity of surviving infants. Material and Methods: Very low birthweight infants (VLBW) admitted to the neonatal intensive care unit were identified. CRIB-II and SNAP-PE-II scores were electronically calculated for each patient via the website www.sfar.org. The receiver operating characteristics (ROC) curve was used to check the accuracy of the mortality and morbidity prediction. results: A total of 189 VLBW neonates (mean CRIB-II:9.9±3.8; mean SNAP-PE-II: 45.8±25.4) were evaluated. The mean birth weight, gestational age, CRIB-II, SNAP-PE-II scores were associated with mortality. Both CRIB-II and SNAPPE-II were determined to be discriminatory for mortality, but not predictive enough for morbidity when evaluated with ROC curve analysis.conclusion: Both CRIB-II and SNAP-PE-II were found to be eligible in predicting hospital mortality of VLBW patients whereas their value was poor in predicting morbidity. CRIB-II, due to fewer parameters to evaluate, may be the preferred scoring system to predict mortality in NICUs with high patient capacity

Keywords: