Assessment of tobacco, alcohol and cannabinoid metabolites in 645 meconium samples of newborns compared to maternal self-reports.

Prenatal psychoactive substance exposure has significant impact on neonatal health and child development and the development of reliable biomarkers is critical. Meconium presents several advantages for detecting prenatal exposure to psychoactive substances, as it is easy to collect and provides a broad time frame of exposure (third trimester). The aim of our study was to compare the prevalence of alcohol, tobacco and/or cannabis use during the third trimester of pregnancy (using maternal self-reports) with the results of meconium testing of their metabolites in newborns (cotinine, ethyl-glucuronide (EtG) and cannabinoid metabolites). Among all deliveries (993) that occurred in all maternities in Rouen (Normandy) during a defined time period (5 consecutive weeks in August, 2010 and August, 2011), 724 mothers were included and 645 meconium samples were collected. Maternal self-reports, using the Addiction Severity Index (5th edition), and meconium samples were collected within 72 h of delivery. Cotinine detection appears highly correlated to maternal self-reports (Kappa value: 0.79; [95%CI: 0.73-0.85]). Moreover, detection in meconium seems more accurate in the prediction of neonatal consequences of prenatal tobacco exposure as compared to maternal self-reports. In contrast, we have found a lower concordance between maternal self-reports and meconium testing for EtG and cannabinoid metabolites (Kappa value: 0.13; [95%CI: 0.04-0.22] and: 0.30; [95%CI: -0.03-0.63], respectively); however the total number of EtG- and cannabinoid-positive meconium samples was small. Interestingly, meconium samples with the highest levels of EtG mainly corresponded to negative maternal self-reports. Fetal exposure to alcohol, tobacco or cannabis may also considerably differ as displayed in our pairs of dizygotic twins. Finally, a polyconsumption of these psychoactive substances was not frequently observed according to meconium testing. In conclusion, cotinine detection appears as a valuable meconium biomarker. EtG measurement in meconium samples seems interesting if there is any risk of high fetal exposure, whereas assessment of prenatal cannabis exposure, using meconium testing, needs to be improved.