The commonly used antibiotic azithromycin is not linked to an increased risk for ventricular arrhythmia. Rather the adverse event is more likely due to infection or patient characteristics, according to a large study (CMAJ 2017 Apr 18.
“Using an incredibly large and robust population and strong methods they show that the risk associated with azithromycin, a QTc interval–prolonging agent, is no higher than amoxicillin, a non–QTc interval–prolonging drug. I applaud the investigators of this study for showing just how low the overall risk of ventricular arrhythmia is in the general population under 85 years of age and for showing that the risk is mostly attributable to infection,” said C. Charles White, PharmD, FCP, FCCP, who was asked to comment on the findings.
Azithromycin, a macrolide, is an antibiotic commonly used to treat respiratory and urinary tract infections in people of all ages. At least one other drug in the macrolide class, erythromycin, is known to disrupt the heart’s normal rhythm, leading to ventricular arrhythmia.
Several recent studies have reported conflicting results over whether azithromycin is linked to an increased risk for death from ventricular arrhythmia in patients taking the antibiotic. To provide clarity among these conflicting findings, a team of European researchers looked at data on nearly 29 million people in health care databases from Italy, the United Kingdom, Germany, the Netherlands and Denmark to determine whether there is a link between azithromycin and ventricular arrhythmia.
Of the more than 14 million new antibiotic users, only 12,874 people (0.1%) developed ventricular arrhythmia, of whom 30 were new users of azithromycin. When compared with amoxicillin, another common antibiotic from the penicillin class of drugs, there was no increased cardiac risk in people using azithromycin. However, there was an increased risk for ventricular arrhythmia in those taking azithromycin compared with those who did not use any antibiotics.
“This finding suggests that the risk of ventricular arrhythmia is more likely to be due to a person’s poor health and caused by their infection, rather than to azithromycin itself,” said Gianluca Trifirò, MD, PhD, an assistant professor in the Department of Biomedical and Dental Sciences and Morpho-functional Imaging at the University of Messina, Italy. “This finding was confirmed in several sensitivity analyses and replicated in single databases participating in the study.”
“What makes this study unique is that they confirm what other investigators have found very clearly, that users of azithromycin are at higher risk of ventricular arrhythmias than those people without infections who receive no antibiotics, but then extend the investigation to ask, ‘Is it the azithromycin or the infection that is to blame?’” said Dr. White, who is a professor and the head of the Department of Pharmacy Practice at the University of Connecticut, in Storrs.
The researchers noted these findings might not be applied in hospital settings, as the health of patients and use of antibiotics are quite different in community settings, from which the data were drawn.
Another limitation, noted Dr. White, was that the researchers did not evaluate torsades de pointes specifically. “It may very well be that this polymorphic ventricular tachycardia is increased due to QTc interval prolongation, but the risk of monomorphic ventricular tachycardia is reduced by the same ECG effects,” he explained. “This would be consistent with Class III antiarrhythmics like sotalol and dofetilide, which reduce monomorphic ventricular tachycardia but can increase the risk of torsade de pointes.
“It also looked at a total population and not one that was prone to experience torsade de pointes. If a patient had a baseline QTc interval that was prolonged due to structural heart disease or other QTc interval–prolonging medications, the impact of overall ventricular tachycardia could have been different. In the past, I suggested being leery of starting a non-antiarrhythmic agent if the baseline QTc interval was above 470 mseconds, and I don’t think this study does anything to change that.”
The study was conducted by researchers from Erasmus University, Rotterdam, the Netherlands; the University of Messina; Italian College of General Practitioners, Florence; PHARMO Institute for Drug Outcomes Research, Utrecht, the Netherlands; Leibniz Institute for Prevention Research and Epidemiology – BIPS GmbH, Bremen, Germany; University of Bologna, Italy; Aarhus University Hospital, Aarhus, Denmark; and King’s College London.