Cardiotoxicity of azithromycin in COVID-19: an overall proportion meta-analysis

To explore the incidence of pro-arrhythmic effects such as corrected QT interval (QTc) prolongation, arrhythmic events and myocardial injury of azithromycin as administered for the treatment of COVID-19.

We searched PubMed, the Cochrane Library and Web of Science databases from inception to 18 January 2021, as well as the medRχiv preprint database from 1 August 2020 to 18 January 2021, for studies exploring the cardiotoxicity effects of azithromycin, with or without concomitant use of hydroxychloroquine, in the context of Covid19. We performed a random effects single-arm meta-analysis of studies to calculate pooled proportion estimates for pro-arrhythmic effects. Meta-regression analyses were conducted to explain between-study heterogeneity.

Thirty-four studies with a total of 3088 patients were included. Among 12 studies, the incidence of > 60ms QTc prolongation from baseline was 13% (95% CI 9%-18%, I² = 73%), whereas, among 28 studies, the incidence of QTc ≥ 500 ms at follow-up was 8% (95% CI 6%-11%, I² = 78%). Still, the discontinuation rate due to QTc prolongation was only 3% (95% CI 2%-5%, I² = 55%). The absolute risk of Torsade de pointes and ventricular tachycardia was 0.2% and 0.8%, respectively. Increased age, male sex, presence of hypertension or diabetes mellitus, use of QTc prolonging medication, prolonged baseline QTc interval and indicators of disease severity such as death explained between-study heterogeneity.

Azithromycin, with or without hydroxychloroquine, leads to a significant risk for critical QTc prolongation in patients with Covid19. Due to its cardiotoxicity effects and its unproven efficacy in Covid19, azithromycin use should be limited to cases of bacterial co-infection.