Klin Padiatr. 2019 Sep 30. doi: 10.1055/a-1009-6671. [Epub ahead of print]
https://www.ncbi.nlm.nih.gov/pubmed/31569261
Prediction of ECMO and Mortality in Neonates with Congenital Diaphragmatic Hernia Using the SNAP-II Score.
Kipfmueller F1, Schroeder L1, Melaku T1, Geipel A2, Berg C2, Gembruch U2, Heydweiller A3, Bendixen C3, Reutter H1, Müller A1.
Author information
1Department of Neonatology and Pediatric Critical Care Medicine, University Children’s Hospital, Bonn, Germany.2Department of Obstetrics and Prenatal Medicine, University Hospital Bonn, Germany.3Department of Pediatric Surgery, University Hospital Bonn, Germany.
Abstract
BACKGROUND:
The mortality of neonates with congenital diaphragmatic hernia (CDH) ranges between 20 and 40% even in specialized high-volume centers. The Score for Neonatal Acute Physiology-II (SNAP-II Score) could facilitate the decision about supportive therapies in CDH newborns.
METHODS:
The SNAP-II score consists of the variables arterial blood pressure, pH, PaO2:FiO2, body temperature, diuresis, and seizure activity and was calculated at an age of 12 h.
RESULTS:
101 CDH newborns treated in our institution between 2009 and 2017 were included in the study. A SNAP-II score ≥ 28 was calculated as cutoff for predicting mortality (AUC 0.876; 95% CI: 0.795-0.957). The mortality rate was 52.9% with a SNAP-II score ≥ 28, and 5.9% with a SNAP-II score<28. Sensitivity and specificity for predicting mortality was 81.8 and 79.7%, the negative predicting value (NPV) was 94.0%, the positive predicting value (PPV) 52.9%. The optimal cutoff for predicting ECMO was ≥ 22 (AUC 0.895; 95% CI: 0.836-0.954). Sensitivity and specificity for predicting ECMO therapy was 90.7, and 63.8%, the NPV was 90.2%, and the PPV was 65% respectively. The SNAP-II score was independently associated with mortality [OR 1.126 (95% CI: 1.050-1.207)] and the need for ECMO therapy [OR 1.243 (95% CI: 1.106-1.397)].
CONCLUSION:
The SNAP-II score is strongly associated with mortality and the need for ECMO therapy in CDH newborns and should be implemented in the risk stratification of these infants.
© Georg Thieme Verlag KG Stuttgart · New York.PMID: 31569261 DOI: 10.1055/a-1009-6671
