Objective To investigate the risk of congenital heart defects (CHD) and other congenital anomalies (CA) associated with first trimester use of macrolides. Design Population-based case-malformed control study. Setting Thirteen European countries. Population Data on 145,936 livebirths, stillbirths and terminations of pregnancy for CA from 15 EUROCAT registries, covering 9 million births 1995–2012. Methods Cases were babies with CHD, anencephaly, orofacial clefts, genital and limb reduction anomalies associated with antibiotic exposure in the literature. Controls were babies with other CA or genetic conditions. A meta-analysis of the literature, including this study, was conducted for CHD. Main outcome Odds ratios adjusted (AOR) for maternal age and registry, with 95% Confidence Intervals (95%CI). Results Macrolide exposure was recorded for 307 cases, 72 non-genetic controls, 57 genetic controls. AOR for CHD was not significantly raised (AOR 0.94, 95%CI: 0.70 – 1.26 vs non-genetic controls; AOR 1.01, 95%CI: 0.73 – 1.41 vs genetic controls), nor significantly raised for any specific macrolide. The risk of atrioventricular septal defect was significantly raised with exposure to any macrolide (AOR 2.98; 95%CI: 1.48 – 6.01), erythromycin (AOR 3.68, 95%CI: 1.28 – 10.61), and azithromycin (AOR 4.50, 95%CI: 1.30 – 15.58). Erythromycin, clarithromycin, azithromycin and clindamycin, were also associated with an increased risk of at least one other CA. Meta-analysis gave an overall CHD OR 1.14, 95%CI 0.90 –1.49 for macrolides. Conclusions Guidelines for macrolide use in pregnancy should consider the increased risk of specific CA. This is relevant for the potential use of azithromycin in the treatment of COVID-19.