European epidemiological patterns of cannabis- and substance-related congenital cardiovascular anomalies: geospatiotemporal and causal inferential study.

Abstract

As prenatal and community cannabis exposures have recently been linked with congenital heart disease (CHD), it was of interest to explore these associations in Europe in a causal framework and space-time context. Congenital anomaly data from Eurocat, drug-use data from the European Monitoring Centre for Drugs and Drug Addiction, and income from the World Bank. Countries with rising daily cannabis use had in general higher congenital anomaly rates over time than those without (time: status interaction: β-Est. = 0.0267, P = 0.0059). At inverse probability-weighted panel regression, cannabis terms were positive and significant for CHD, severe CHD, atrial septal defect, ventricular septal defect, atrioventricular septal defect, patent ductus arteriosus, tetralogy of Fallot, vascular disruptions, double outlet right ventricle, transposition of the great vessels, hypoplastic right heart, and mitral valve anomalies from 1.75 × 10^-19, 4.20 × 10^-11, <2.2 × 10^-16, <2.2 × 10^-16, 1.58 × 10^-12, 4.30 × 10^-9, 4.36 × 10^-16, 3.50 × 10^-8, 5.35 × 10^-12, <2.2 × 10^-16, 5.65 × 10^-5 and 6.06 × 10^-10. At spatial regression, terms including cannabis were positive and significant for this same list of anomalies from 0.0038, 1.05 × 10^-10, 0.0215, 8.94 × 10^-6, 1.23 × 10^-5, 2.05 × 10^-5, 1.07 × 10^-6, 8.77 × 10^-5, 9.11 × 10^-6, 0.0001, 3.10 × 10^-7 and 2.17 × 10^-7. 92.6% and 75.2% of 149 E-value estimates and minimum E-values were in high zone >9; 100.0% and 98.7% >1.25. Data show many congenital cardiac anomalies exhibit strong bivariate relationships with metrics of cannabis exposure. Causal inferential modelling for the twelve anomalies selected demonstrated convincing evidence of robust relationships to cannabis which survived adjustment and fulfilled epidemiological criteria for causal relationships. Space-time regression was similarly confirmatory. Epigenomic pathways constitute viable potential mechanisms. Given exponential genotoxic dose-response effects, careful and astute control of cannabinoid penetration is indicated.