[Predictive value of SNAP and SNAP-PE for neonatal mortality]

OBJECTIVE: To evaluate the Score for Neonatal Acute Physiology and the Score for Neonatal Acute Physiology Perinatal Extension as neonatal mortality predictors in our neonatal intensive care unit, and to compare their results. METHODS: All newborn infants admitted to our neonatal intensive care unit from March 1997 through December 1998 were prospectively evaluated just at completion of 24 hours of life for SNAP and SNAP-PE. Exclusion criteria were: death or discharge from the neonatal intensive care unit in the first 24 hours of life, congenital malformations incompatible with life, and outborn infants. RESULTS: 553 newborn infants were included in the study and 54 died. The median SNAP and SNAP-PE values were higher in those who died. Infants were allocated to five different raising ranges of SNAP and SNAP-PE severity. SNAP: up to 6, 7-11, 12-15, 16-24, higher than 24 (mortality: 3%, 11%, 29%, 48%,75%, respectively). SNAP-PE: up to 11, 12-23, 24-32, 33-50, higher than 50 (mortality: 3%, 10%, 53%, 78%, 83%, respectively). The optimal cut off points based on ROC curve were 12 for SNAP, and 24 for SNAP-PE. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for death were figured out. SNAP 12: sensitivity 79.6%, specificity 71.7%, PPV 23.4%, NPV 97%. SNAP-PE 24: sensitivity 79.6%, specificity 80%, PPV 30%, NPV 97.3%. The area under the curve (Az) were 81.4% for SNAP, and 85.1% for SNAP-PE, both statistically significant. There were no statistical differences between the two areas under the curve. CONCLUSIONS: SNAP and SNAP-PE are excellent predictors of neonatal survival. Therefore, we recommend their use in Neonatal Intensive Care Units.