Early Prediction Tool to Identify the Need for Pharmacotherapy in Infants at Risk of Neonatal Abstinence Syndrome.

OBJECTIVE

To develop a tool to predict the need for pharmacologic treatment of neonatal abstinence syndrome (NAS) within 36 hours from birth in infants at risk for opioid withdrawal.

STUDY

DESIGN

Retrospective study of infants born at gestation of ≥34 weeks with in utero exposure to opioids during two time periods from January 2013 through October 2016. Period 1 was used to develop a predictive tool for validation during period 2. Birth weight, gestational age, four categories of opioid exposure, and individual scores for 21 withdrawal symptoms from the Modified Finnegan Score at 36 hours of life were recorded. During period 1, a best subsets multiple regression analysis was performed on factors that were associated with pharmacotherapy on univariate analysis. Two tools were designed: one based on three highly predictive symptoms associated with need for pharmacotherapy for NAS and the other incorporating opioid exposure. Sensitivity, specificity, and positive and negative predictive values for the tools were calculated during period 2.

RESULTS

The study included 264 infants (period 1, n=143; period 2, n=121). Polysubstance exposure and three withdrawal symptoms present at 36 hours of life that were significantly associated with pharmacotherapy for NAS comprised the tools. The "symptoms only tool" was able to predict that infants with a score <1 would not receive pharmacotherapy, and infants with scores of ≥4 would receive pharmacotherapy with positive predictive values of 90% and 100%, respectively. When opioid exposure was included, the "symptoms + exposure tool" was able to predict that infants with a score of ≤1 would not receive pharmacotherapy and infants with scores of ≥5 would receive pharmacotherapy with positive predictive values of 94% and 86%, respectively.

CONCLUSION

An NAS prediction tool combining three clinical signs with and without category of opioid exposure had high positive predictive values for requiring and for not requiring pharmacotherapy. This tool may expedite pharmacotherapy decisions and optimize management for infants at risk for NAS.