Prospective evaluation of the Whitt Neonatal Trigger Score in an 'at-risk' neonatal population

The aim of this study was to prospectively evaluate the Whitt Neonatal Trigger Score (W-NTS), determining optimum threshold scores for consideration of medical intervention and intensive care unit admission.

All neonates on the postnatal ward (PNW) with a set of pre-defined risk factors were scored on the W-NTS. Neonates were divided into three groups: 'unwell' admitted to neonatal intensive care unit (NICU); 'well', who remained on the PNW receiving standard care; and 'intervention', who received antibiotics but did not require admission to NICU.

A total of 3315 scores from 455 neonates were collected. The W-NTS area under the receiver operating characteristic curve (AUC ROC) was 0.968, with a score of 2 or more predicting NICU admission, with 82.5% sensitivity and 95.0% specificity. Adopting a cut-off score of 2 for admission would significantly improve speed to admission (11.6 vs. 6.9 h, P 0.037). A score of 0 was strongly predictive of being well enough to remain on the PNW without intervention (odds ratio 565.6, P < 0.001), and a score of 1 or more predicted the need for intravenous antibiotics with 100.0% sensitivity and 86.1% specificity (AUC ROC 0.977).

The W-NTS observation chart, previously shown to outperform existing early warning scores, acts well as an adjunct to clinical assessment on the PNW, with its simplicity allowing for the successful and safe use by non-paediatric specialists. We recommend that neonates scoring 1 should be reviewed, with a septic screen and commencement of antibiotic therapy considered, while those scoring 2 or more should be strongly considered for NICU admission for further management.