55 - Maternal self-reported disclosure is a reliable predictor of neonatal admissions with opioid withdrawal

Publication Number: 4054.55

Courtney W. Tam, Keck School of Medicine of the University of Southern California, Los Angeles, CA, United States; Olivia Venvertloh, Creighton University School of Medicine, Phoenix, AZ, United States; Maria Concepcion Mendoza, Keck School of Medicine of the University of Southern California, Torrance, CA, United States; Amy Yeh, USC/Los Angeles General Medical Center, Los Angeles, CA, United States; Rowena Cayabyab, USC/Los Angeles General Medical Center, Los Angeles, CA, United States; Hemananda Muniraman, Los Angeles General Medical Center, Los Angeles, CA, United States

Background: Screening for intrauterine substance use (IUSE) is routinely performed by maternal substance use history screening (SUHS) and maternal and neonatal urine substance screening (USS). Maternal and neonatal USS is associated with bias and disparity and may be used as a factor in determining family separation. Utility of routine USS needs thorough evaluation.

Objective: To evaluate and compare individual and combined predictive abilities of maternal SUHS, maternal USS and neonatal USS for neonatal intensive care unit (NICU) admission and pharmacologic treatment of neonatal opioid withdrawal syndrome (NOWS).

Design/Methods: Single-center retrospective study included pregnant women with positive history and/or USS for substance use during current pregnancy and their neonates with NOWS between Feb 2018-Dec 2022. Predictive abilities of maternal SUHS, maternal USS and neonatal USS of NOWS NICU admission and pharmacologic treatment were analyzed by logistic regression, sensitivity measures and area under the receiver operating characteristic curve (AUC).

Results: 952 infants with IUSE were born during the study period, of which 277 (29%) infants with mean age of 38.2 weeks were admitted for NOWS-related withdrawal/feeding issues and 176 (18.4%) required pharmacologic treatment. Odds ratio (OR) for admission was 9.9 (95% CI 7.1-13.1) with positive maternal SUHS, 8.5 ​​(95% CI 6.1-11.9) with maternal USS, and 6.2 (95% CI 4.5-8.6) with neonatal USS. Combined maternal SUHS and maternal USS increased OR to 12.0 (95% CI 8.2 -17.4), while adding neonatal USS decreased it to 10.8 (95% CI 7.3 -16.0) (Table 1, Figure 1). Multi-nominal regression analysis showed maternal history (OR 4.8) was associated with highest adjusted OR, compared with maternal USS (OR 2.9) and neonatal USS (OR 1.8, p< 0.05).

For pharmacologic treatment of NOWS, OR was 10.3 (95% CI 7.1-14.9) for maternal opioid history, 9.6 (95% CI 6.6-14.0) for maternal USS, and 7.7 (95% CI 5.3-11.2) for neonatal USS. Combined maternal SUHS and maternal USS increased OR to 17.9 (95% CI 11.0-29.2) and adding neonatal USS was associated with marginally higher OR at 18.3 (95% CI 10.5-31.6) (Table 1, Figure 2). Maternal SUHS was associated with highest adjusted OR (4.8) amongst individual screening methods on multi-nominal logistic regression.

Conclusion(s): Combined maternal SUHS and maternal USS demonstrated highest predictive ability for NOWS NICU admission. Neonatal USS did not add predictive value beyond maternal SUHS, while maternal history alone was a reliable predictor of neonatal outcomes. There is limited utility of neonatal USS when maternal history is negative.

Table 1. Accuracy and predictive measures of screening tests
Combined table delineating odds ratio (OR), 95th percentile confidence interval (CI) listed as (lower bound, upper bound), p-value, sensitivity (Sn), specificity (Sp), positive predictive value (PPV), negative predictive value (NPV), area under the receiver operating characteristic curve (AUC), AUC 95th percentile CI listed as (lower bound, upper bound) when comparing all NICU admissions and NICU admissions with medication use with maternal opiate use history, maternal USS for opiates, neonatal USS for opiates, maternal opiate history with maternal USS for opiates, and maternal opiate history with maternal USS for opiates, and neonatal USS with opiates, respectively.

Figure 1
Receiver Operator Characteristic (ROC) curve for NICU admissions for neonatal opioid withdrawal syndrome (NOWS).

Figure 2
Receiver Operator Characteristic (ROC) curve for NICU admissions for neonatal opioid withdrawal syndrome (NOWS) requiring pharmacological treatment.