A couple of weeks ago, in one of those moments that make all community pharmacists fear a deluge of patient enquiries the following morning, Mail Online screamed: "Antibiotic taken by millions is linked to heart deaths: Treatment used for bacterial infections linked with 76 per cent higher risk compare [sic] with alternative medicine." Being more than aware of the Mail's propensity for alarmist headlines on health stories (and after I'd finished laughing at the quality of the Mail's sub-editing), I thought I'd download the BMJ paper and have a look myself.
Let me be clear - the 76 per cent figure in the headline, which is also quoted in C+D's coverage, is correct. In this Danish cohort study of adults aged between 40 and 74 years of age who received seven-day courses of clarithromycin, roxithromycin or penicillin V, those taking clarithromycin were 76 per cent more likely to die from a cardiac-related cause than those taking penicillin V. The authors report that there were an extra 37 deaths per 1 million courses of clarithromycin when compared with 1 million courses of penicillin V. To present this figure another way, this means that for every 27,027 courses of clarithromycin or penicillin V prescribed, we would expect to see one extra cardiac death in the clarithromycin group. A single extra death is, of course, a tragedy - but, while the increase in relative risk is marked, it appears much less "impressive" when reported as absolute numbers.
While the authors attempted to control some variables, they were unable to control some factors known to increase the risk of cardiac death such as smoking and body mass index
However, the excess deaths observed in patients taking clarithromycin cannot be causally attributed to the drug itself. As the study design was not randomised, the excess deaths could have been the result of any number of confounding factors. As no data are included on presentation, it could be the case that clarithromycin was prescribed to a cohort of "sicker" patients than those prescribed penicillin, for example. While the authors attempted to control some variables such as age and sex to minimise the chances of confounding, they were unable to control some important factors that are known to increase the risk of cardiac death, such as smoking and body mass index.
The risk of cardiac toxicity with macrolides is well known. It is believed that macrolide-associated cardiac deaths occur because of prolongation of the QT interval resulting in fatal arrhythmias, which is why the BNF advises caution in their use in patients with a predisposition to QT interval prolongation - meaning that the likelihood of those patients who are most at risk of cardiac death receiving a prescription for clarithromycin is reduced. The amount of data examined during this research - more than 5 million seven-day treatment courses - makes this a useful addition to the evidence base on the association between macrolides and cardiac death. But it doesn't justify the revelatory, "scary" headlines (with all their potential negative consequences) seen in the tabloid press.
Joseph Bush is a senior lecturer in pharmacy practice at Aston Pharmacy School