This was an IRB approved (Jan 2021) prospective, pilot, case control, trial of tapering of buprenorphine in pregnancy with opioid use disorder compared with non-tapering of buprenorphine in pregnancy. Tapering of buprenorphine was done in the outpatient settings of a university obstetric clinic where pregnant women entered into a written contract with the obstetrician for tapering of a prescribed opioid (i.e., buprenorphine or buprenorphine-naloxone) for opioid use disorder (OUD). Tapering of buprenorphine for opioid use disorder was managed by the obstetrician in agreement (not the part of the research) with patients depending on several factors with regular follow-up (i.e., urine drug test at each obstetric visit along with comprehensive care with a case manager, behavioral therapist, and other community resources). The tapering group (T-case group) patients’ consent was obtained in the outpatient clinic when the patient came for antenatal visits, while the non-tapering (NT-Control group) consent was obtained when they presented for delivery at the hospital. After obtaining informed consent, demographic and relevant antenatal information was obtained from the patients and their electronic medical records. Most of our mothers had urine drug screens and cord stat done as part of hospital protocol and clinical practice when they were admitted for labor in the hospital.
All mothers were given the Edinburg Postnatal Depression Scale (EPDS), Perceived Stress Scale (PSS), and Confusion, Hubbub, and Order (CHAOS) after delivery to complete. These surveys were returned to the patient’s nurse or directly to a member of the research team prior to discharge. All newborns were monitored for the development of neonatal abstinence syndrome (NAS) or NOWS by nursing staff trained in the modified Finnegan. All newborns with worsening Finnegan scores (i.e., 10 or greater) were transferred to the NICU to observe and treat NOWS as per institutional protocol. Morphine was started as per the standard NOWS protocol of the institution (two consecutive scores of more than 10 or 3 consecutive scores of 8 or more) and weaned as per the protocol and clinical practice at the attending physician's discretion. All newborns were followed until discharge. The current study was a single center study with population and antenatal clinics and MAT clinics ted in mid-south Appalachian region where opioid prevalence rate is high.
The study population included pregnant women with OUD who received prenatal care at University affiliated clinic and surrounding medication assisted treatment centers (MAT) located in the South-Eastern State during the years 2021–2023 in a non-consecutive pattern due to pandemic related issues of available staff and hospital restriction. MAT clinics provided services related with opioid disorder while pregnancy care was taken by other area obstetrical providers with opioid use disorder pregnancy compared to university clinics where both services provided with comprehensive care as described above.
All pregnant women who have taken buprenorphine or any other medication for their opioid use disorder were eligible for the study. Newborns born before 36 weeks, very low birth weight (1800 grams or below), congenital malformation or non-English speaking patients were excluded. Eighty participants were consented at the clinic or hospital, while 151 eligible mothers were asked to enroll. However, only 142 participants (71 mother-neonate dyads) were included for this study as there was one infant with less than 36 weeks of gestational age, one record of fetal demise, and seven participants who decided to deliver at a different hospital. All the excluded participant records were removed from the dataset and were not included in the subsequent analysis. Variables of interest including socioeconomic, medical history along with opioid dosage changes of mothers and infant birth weight, APGAR, Finnegan scoring, feeding and treatment for newborns, were collected and managed using REDCap electronic data capture tool. Head circumference, birthweight, and APGAR and Finnegan scores for all newborns were measured at birth by the nursing team responsible for the post-delivery care of the newborn. Length of stay was recorded as the number of days from the newborn birth date to the discharge date. The women who underwent supervised tapering during pregnancy formed the Tapering- group (T) of case-control trial. Women who did not taper opioids during pregnancy formed the NT- group.
Intervention: Patient tailored tapering of buprenorphine in antenatal setting with comprehensive care as was done after written clinical care contract agreement which discussed the risks and benefits of tapering buprenorphine, This agreement was not part of research, and was signed by patient as a clinical care. Tapering of buprenorphine was done by university affiliated clinic provider based on many factors every 4–6 weeks with goal of lowering of buprenorphine to 2 mg or even weaning completely if tolerated and desired by the patient. A patient’s autonomy of the care along with tolerance, duration of medication with history of concurrent morbidity, presenting dose and time for care at the antenatal clinic and support during pregnancy played important role in decision to taper which varied among participants. All maternal participants in the tapering group voluntarily decided to taper their buprenorphine dosages. Patients who did not wish to taper their doses continued their medication assisted treatment (MAT) with community MAT providers.
Bias: The study, despite its non-randomized design, addressed biases such as selection and confounding. Participants were selected from a similar population of pregnant women with OUD, minimizing selection bias. However, confounding by indication was acknowledged, and statistical analyses were employed to control for variables like maternal age and substance use. The study aimed to isolate the impact of buprenorphine tapering on neonatal outcomes, particularly NOWS. Demographic information for both groups was presented transparently, and consistent exclusion criteria were applied for internal validity.
Statistical analysis
Univariate associations between the taper and non-tapered group demographic variables such as maternal age, education, marital status, maternal exposure variables, infant demographics, and infant exposure and outcome variables, was assessed by chi-square and Fisher’s exact tests for categorical variables, and two-sample t-tests and Mann-Whitney tests for continuous variables. Maternal tobacco, alcohol, substance use, and gestational age are some possible confounders by statistical association with tapering MATs. A logistic regression and a scatter plot were performed between the dependent variable (incidence of NOWS) and each one of the independent variables mother’s tapering statuses, infant gestational age, and mother’s history of smoking, alcohol, and substance use to analyze the relationship between each independent variable and the dependent variable separately, without considering the other independent variables. A multivariate logistic regression analysis was performed to find the relationship between the outcome variable and the predictor variables. The decision to use these independent variables as predictor variables was based on the findings from the literature review. A significance level of 0.05 was chosen to decide whether each independent variable was significant or not. All analyses were conducted using SAS version 9.4.