[Infants of drug-addicted mothers: pitfalls of replacement therapy].

Maternal drug addiction can cause problems for the fetus and the newborn, and hamper long-term development. The prevalence of drug addiction during pregnancy varies from 1 % to more than 10 % depending on the country and the maternity unit. Management of these mothers can be further complicated by medical, social and psychological problems. Compared to methadone, heroin replacement therapy with buprenorphine provides better stabilization of the mother and causes fewer withdrawal symptoms in the newborn. Despite numerous publications on the effects of this partly preventive medication, data on buprenorphine pharmacology at birth are scarce. In this study, 20 newborns of mothers using oral buprenorphine were observed until the end of the withdrawal syndrome, when present. Buprenorphine plasma levels were determined with HPLC and mass spectrometry in the mother at delivery and in the newborn at birth (cord blood), 24 and 48 hours. Fifteen newborns were born at term (mean +/- SD birth weight 3029 +/- 273 g), and the other five between 32 and 36 weeks. All Apgar scores were > or =7. Withdrawal symptoms were observed in 8 of the 15 infants born to mothers taking buprenorphine alone, and lasted between 5 and 35 days. The newborns were classified in three groups. Groups I (N8) and II (N7) comprised newborns with and without withdrawal symptoms, respectively. In group III (N5), the mothers were polyintoxicated (as shown by urinary drug or neurotropic substance screening) and the newborns were symptomatic for 1 to 69 days. Buprenorphine plasma levels in the mothers ranged from 0 to 2.9 microg/L, suggesting large differences in adherence. At birth there was no significant difference in the mean plasma buprenorphine level between newborns with and without withdrawal symptoms; the respective values were 0.7 (0.4-1.3) and 0.5 (0-0.6) microg/L. In asymptomatic newborns (group II), buprenorphine was no longer detectable at 48 h, whereas in symptomatic newborns (group I), the mean level rose from 0.7 microg/l at birth to 1.5 microg/L at 48 h (+114 %). In the absence of breastfeeding, this increase appears to be related to tissue release of this strongly lipophilic compound. The difference in plasma buprenorphine kinetics between groups I and II might be explained by genetic polymorphism of drug-metabolizing enzymes. The paradoxically high plasma buprenorphine levels at 48 hours in infants with withdrawal symptoms are intriguing. One possibility is that the mothers missed one or several doses of buprenorphine around the time of delivery, in the same way that smoking mothers tend to cut down during the last days of their pregnancy. If buprenorphine plasma levels at birth appear to reflect maternal adherence, cord blood levels do not predict the risk of a withdrawal syndrome. In contrast, the level at 48 h might help to discriminate between high- and low-risk newborns. Pregnant women on opiate replacement therapy must be delivered in maternity units with adequate neonatal facilities.