These are the first data linking maternal buprenorphine dose, buprenorphine and metabolite concentrations in meconium, and neonatal outcomes in a cohort of women who received controlled dosing with buprenorphine. In addition, to our knowledge this is the first comparison of neonatal meconium with maternal urine data collected three times a week throughout gestation to estimate frequency and/or magnitude of illicit drug use. Evaluation of maternal buprenorphine dose and neonatal outcomes provided valuable information on buprenorphine safety in treating opioid dependence during pregnancy, and the inclusion of meconium analysis yielded the first data on the disposition of buprenorphine and metabolites disposition in this neonatal biological matrix. The ability of these meconium biomarker concentrations to predict many neonatal outcomes was also evaluated for the first time.
There was no relationship between the total or cumulative third trimester dose of buprenorphine, or dose at delivery, and free or total buprenorphine or norbuprenorphine concentrations in the meconium of the neonate. This is the first study to examine the relationship between maternal buprenorphine dose and buprenorphine and metabolite concentrations in the meconium of the neonate; previous research has evaluated this relationship for methadone and cocaine. A positive relationship was reported between maternal methadone dose and methadone concentrations in the meconium.13
Ostrea et al.
reported a statistically significant relationship between maternal cocaine dose and cocaine concentrations in meconium in rats.14
Buprenorphine, norbuprenorphine, and glucuronide conjugates were identified in meconium specimens, indicating passage of analytes across the placenta and/or creation through placental or fetal metabolism. Factors affecting the distribution of buprenorphine and its metabolites to the fetus include maternal plasma concentration, placental transfer and metabolism, and fetal metabolism. Nanovskaya et al.
reported <10% buprenorphine transfer to the fetus, and ~5% metabolized to norbuprenorphine by the placenta.15
Low placental permeability to glucuronidated compounds was documented.16
There are no data available on phase II placental buprenorphine metabolism; however, enzymes capable of buprenorphine glucuronidation were identified in first trimester as well as full-term placentas.17–19
Variability in fetal metabolism may also influence buprenorphine and metabolite concentrations in meconium. CYP3A4 and CYP2C8 account for 95% of the metabolism of buprenorphine to norbuprenorphine, although CYP3A5, CYP3A7, CYP2C18, and CYP2C19 also contribute to the process.20,21
CYP3A7 is the principal fetal hepatic enzyme, with CYP3A7 and CYP3A5 expression becoming evident as early as 42 days after fertilization.22
CYP3A7 expression decreases during gestation, and hepatic metabolism shifts primarily to CYP3A4 within days of birth.22,23
There are also large interindividual variations in CYP expression, and the data available on fetal hepatic uridine diphospho–glucuronosyltransferase expression and activity are limited. Significant hepatic morphine glucuronidation was shown to occur in fetal baboon liver, but the glucuronidation of other xenobiotics was limited.23,24
Sulfation also appears to be more prevalent in fetal liver.23
Buprenorphine and norbuprenorphine concentrations in meconium taken from 16 infants whose mothers had been exposed to 1–16 mg/day buprenorphine during gestation have been popiureviously reported.25
Data were obtained from buprenorphine assays for urine, serum, or plasma, with no details relating to analysis of meconium and no mention of whether the data represented free or total concentrations. After maternal doses of 1–16 mg/day for variable time periods during pregnancy, buprenorphine and norbuprenorphine concentrations in a single infant were 107 and 295 ng/g, respectively,25
and mean concentrations for 15 others were 122 and 176 ng/g, respectively. 26
These data relating to buprenorphine concentrations in meconium are similar to those in this study, whereas the norbuprenorphine concentrations reported were much lower. In this study, meconium was analyzed with and without hydrolysis, allowing estimation of the extent of conjugation. Glusulase hydrolysis frees both glucuronide and sulfate conjugates.27
Drug analysis in meconium samples is an alternative to maternal self-report and maternal urine testing for monitoring the use of illicit drugs during pregnancy. Maternal self-report underestimated in utero
drug exposure by up to 44% when compared with data from meconium analyses.28,29
Two previous studies reported higher sensitivity for meconium data than for maternal urine data for identifying in utero
cocaine exposure; one investigation showed equal sensitivity.30–32
In this study, the thrice-weekly analysis of maternal urine specimens followed by analysis of infant meconium permitted a comparison of efficacies in documenting cocaine, benzodiazepine, and opiate use during pregnancy. These are the first data on timing and frequency of illicit drug use, as monitored by urine testing in the mothers and detection in neonate meconium. Meconium begins to form at ~12–13 weeks of gestation, and this would suggest that meconium drug concentrations are cumulative until birth. Our data suggest that drug use during the third trimester is more likely to be reflected in meconium than is second-trimester use. Meconium testing failed to identify four infants born to women who used cocaine in the second trimester; three of the women had >15% cocaine-positive urine specimens. All participants had opiate-positive urine specimens in the second trimester because morphine was administered at enrollment to transfer the participants onto treatment medication; however, only three infants had opiate-positive meconium specimens. These findings are supported by maternal methamphetamine self-report data,33
where detection rates were 68% when women reported third-trimester use, and ≤10% with only first or second-trimester exposure. Our data show that time of last drug use, as documented by a positive urine test, is another important factor influencing incorporation of the drug in meconium. It is possible that meconium can be contaminated by infant urine, although only cocaine or opiate use within ~72 hours of birth would be reflected.
Our research results support previous findings that there is no relationship between maternal mean daily buprenorphine dose, dose in mg/kg, or total dose and NAS intensity or total number of infants requiring treatment for NAS.11
In a second investigation comparing methadone with buprenorphine, no association was observed between mean dose of medication at delivery and NAS intensity.3
The largest study to date, of 159 women treated with 0.4–24 mg/day of buprenorphine, showed no significant interaction of dose with NAS intensity.8
Together, these data suggest that up to 24 mg/day of buprenorphine can be safely administered to the mother without increasing the risk of NAS in the infant.
These data are the first from a prospective clinical study to evaluate the value of meconium buprenorphine and metabolite concentrations in predicting neonatal outcomes. Although there is no apparent relationship between maternal buprenorphine dose and neonatal NAS, meconium concentrations reflect infant exposure, taking into account differences in total metabolism and placental transfer, and may therefore provide a better neonatal outcome predictor than the maternal drug dose. The research was limited by its small sample size, but preliminary data suggest that several relationships should be further investigated. Higher levels of buprenorphine concentration and percentage of free buprenorphine in meconium were associated with lower head circumference in the infant. In our study, women received prenatal care and a wide variety of support measures that helped control for socioeconomic and psychosocial factors that negatively affect fetal growth. However, 78% of the women smoked tobacco and 67% used cocaine during pregnancy; both of these are associated with lower head circumference in infants.34–36
Therefore, additional research is needed in a larger sample to confirm the effect of maternal exposure to buprenorphine on the infant head circumference.
The retrospective study by Marquet et al
., discussed earlier, compared buprenorphine and norbuprenorphine concentrations in the meconium of infants who experienced withdrawal (n
= 9) with those in infants who did not experience withdrawal (n
The meconium of infants who experienced withdrawal tended to have higher buprenorphine concentrations than those who did not (P
= 0.53), but no relationship was noted for norbuprenorphine concentrations (P
= 0.13). In this study, a negative correlation was observed between the percentage of free norbuprenorphine in meconium and the time-to-peak NAS score. In addition, as the free buprenorphine/free norbuprenorphine ratio increased, time to NAS onset decreased. These data suggest that increased prenatal drug exposure may exceed placental and fetal phase I and phase II metabolism capacity. There was a strong relationship between total buprenorphine concentrations and buprenorphine/norbuprenorphine ratios (free and total) and the percentage of NAS scores >4. Increased buprenorphine exposure may increase the frequency of occurrence of NAS symptoms as a result of the fetus’ inability to effectively metabolize buprenorphine to less active compounds. Measurement of free buprenorphine and free norbuprenorphine may allow clinicians to predict the onset and severity of NAS.
The labor-intensive and costly hydrolysis step required for quantifying total drug did not improve predictive value. Although one infant’s buprenorphine concentrations in meconium were below the LOQ, an analysis of free norbuprenorphine documented buprenorphine exposure. The measurement of free drug concentration alone enables laboratories to report results more rapidly, with savings in cost.
Only limited data are available on the effects of maternal exposure to nicotine, either alone or in combination with other drugs, on NAS in infants. Infants with no NAS or only mild NAS were born to women who reported smoking fewer cigarettes, whereas the mothers of infants with moderate NAS reported higher levels of tobacco use.11
Infants of opioid-dependent women receiving methadone had significantly lower peak NAS scores (P
= 0.014) and significantly shorter (P
= 0.016) time-to-peak scores when mothers smoked ≤10 cigarettes/day (n
= 16) as compared to ≥20 cigarettes/day.37
The protocol of this study provided for analysis of nicotine concentrations in the meconium of infants in relation to self-reported cigarette use in the mothers. No relationship was found between nicotine concentrations in meconium and neonatal outcomes. The application of a 10 ng/g cutoff for nicotine, cotinine, or OH-cotinine in meconium to define maternal tobacco smoking was able to identify the self-reported smokers. It also identified one participant who was observed smoking, although she denied use.
There are several limitations to this study. The small sample size and large number of statistical comparisons limit the robustness of the results; however, these data are among the first to examine the associations among maternal urine test results throughout gestation, drug concentrations in neonatal meconium, and neonatal outcomes to identify the important factors responsible for producing positive meconium test results. Also, although there was a 10-week variability in the duration of study enrollment, all of the women participated throughout their third trimesters. Similar results were obtained for evaluations of total cumulative dose and third-trimester dose. Also, there was no relationship between duration of study enrollment and drug concentrations in meconium or neonatal outcomes. Concomitant drug use is a confounder that must be considered when interpreting outcome measures.
Despite its limitations, this comprehensive treatment program allowed close monitoring of study compliance and illicit drug use. Thereby, accurate and precise information on buprenorphine doses received throughout gestation and in the third trimester and consistent evaluation of NAS for 10–14 days after birth were available. The lack of correlation between maternal buprenorphine dose and neonatal outcomes highlights the safety of buprenorphine pharmacotherapy during pregnancy. If buprenorphine and metabolite concentrations in meconium prove to accurately predict the onset and frequency of occurrence of NAS symptoms, clinicians could better identify and treat neonates with negative sequelae. Comparison with maternal urine tests suggests that meconium may not accurately identify infants exposed to drugs in utero in the second trimester, a finding with implications for prenatal drug exposure diagnosis and research. Moreover, these data suggest that measurement of free buprenorphine concentrations alone should be sufficient, thereby eliminating the need for costly and labor-intensive analysis of total drug concentrations in meconium.