625 - Respiratory Rate Patterns Identify Infants with Neonatal Opioid Withdrawal Syndrome (NOWS)

Publication Number: 4611.625

Sara Manetta, University of Virginia School of Medicine, Charlottesville, VA, United States; Sherry Kausch, UVA Childrens, Charlottesville, VA, United States; Angela Gummadi, University of Virginia School of Medicine, Charlottesville, VA, United States; Will King, Medical Predictive Science Corporation, Charlottesville, VA, United States; Colm P. Travers, University of Alabama at Birmingham, Birmingham, AL, United States; Namasivayam Ambalavanan, University of Alabama School of Medicine, Birmingham, AL, United States; Zachary A. Vesoulis, Washington University School of Medicine, St. Louis, MO, United States; Brynne Sullivan, University of Virginia School of Medicine, Charlottesville, VA, United States

Background: In utero opioid exposure leads to autonomic dysfunction that alters control of breathing during neonatal opioid withdrawal (NOWS), yet limited research has been done using respiratory data to characterize NOWS. We hypothesized that respiratory rate (RR) patterns could detect the physiologic signature of NOWS.

Objective: To examine differences in continuous respiratory rate metrics between NOWS infants with and without pharmacological treatment and control infants.

Design/Methods: We included term NOWS infants and controls (i.e., not exposed to opioids) born at 36 to 41 weeks of gestation and admitted at three academic centers. NOWS infants were further categorized into treated NOWS infants that required pharmacologic treatment and untreated NOWS infants. Infants with cardiorespiratory disorders, major anomalies, and non-opioid in-utero exposures were excluded. Respiratory rate data were sampled every 5 seconds from continuous monitoring over the 28 days after birth, excluding the first 6 hours. Respiratory rate metrics including RR mean, standard deviation (SD), and RR accelerations were calculated in 10-minute windows, smoothed over 3 hours, and averaged daily. We developed a multivariable model to detect NOWS using RR metrics. Lastly, we examined the relationship between RR metrics and Eat, Sleep, Console (ESC) assessments as a measure of withdrawal severity.

Results: We analyzed 183 NOWS infants (131 treated, 52 untreated) and 190 control infants with RR data available in the 28 days after birth (Table 1). NOWS infants had lower birthweight and longer hospital stays. Treated NOWS infants and untreated NOWS infants had higher mean, SD, and RR accelerations than controls for the first 5 days (Figure 1). In 56 NOWS infants with ESC data, mean RR were higher at times infants' ESC score contained at least 1 "no" component (score > 0) compared to times their scores contained all "yes" components (score = 0, Figure 2).

Conclusion(s): Respiratory rate patterns identify NOWS infants in the first 5 days after birth. NOWS infants with and without pharmacologic treatment have higher and more variable respiratory rates compared to non-opioid-exposed control infants. Infants with ESC scores > 0 that were unable to eat, sleep or be consoled had higher RR, suggesting this objective RR metric correlates with withdrawal severity.

Table 1. Patient Demographics and Characteristics
Patient demographics are grouped comparing controls, treated NOWS infants, and untreated NOWS infants. Categorical variables are summarized as N (%) and continuous variables as median (IQR). Groups were compared using Kruskal-Wallis rank sum tests, Pearson's Chi-squared tests and Fisher exact tests.

Figure 1. RR Metrics by NOWS Group
RR metrics over the first five days after birth by NOWS group, including control infants, treated NOWS infants (Tx NOWS), and untreated NOWS (UTx NOWS) infants. The metrics are displayed as the mean z-scores. RR accelerations are calculated by subtracting the mean RR from max RR, representing variability above the mean. Error bars represent the standard error of the mean (SEM). A z-score of 0 corresponds to the control group mean, while positive values indicate how many standard deviations a group's mean is above the control group mean. The data suggest that treated and untreated NOWS infants have higher and more variable respiratory rates than control infants. Asterisks indicate times respiratory features were significantly different between groups. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2. RR Metrics by ESC Scores
Normalized RR metric values by the ESC score. The RR metrics are displayed as z-scores. A z-score of 0 corresponds to the control group mean, while positive values indicate how many standard deviations a group's mean is above the control group mean. 56 NOWS infants had ESC scores recorded during the first 28 days of their admission. When ESC scores contained all "yes" components (blue), ESC was considered 0; when ESC scores contained at least 1 "no" component, ESC was considered > 0 (brown). In each box, the horizontal line denotes the median value; boxes extend from the 25th to the 75th percentile of each group's distribution of values; vertical extending lines denote more extreme values within 1.5 x the interquartile range; dots denote observations outside that range. The asterisk indicates that mean RR was significantly higher at times infants' ESC scores were > 0 compared to times their scores were 0 (*p < 0.05).