Macrolides use and the risk of sudden cardiac death

KEYWORDS:

• Macrolides
• sudden cardiac death
• arrhythmia
• coronary heart disease
• torsades de pointes

Macrolide antibiotics have commonly been used in the treatment of upper and lower respiratory infections and some sexually transmitted diseases [1,2]. Although they have been in use for decades, macrolides are proposed to have the greatest potential among antibiotics to cause QT interval prolongation, torsades de pointes (TdP), and sudden cardiac death [3,4]. Although the absolute risk of sudden cardiac death is small, widespread macrolide use may lead to a considerable excess of sudden cardiac death [5].

The most commonly known cause of sudden cardiac death appears to be severe ventricular arrhythmias, such as TdP [6,7]. The major pathophysiologic mechanism of macrolide-induced TdP involves prolongation of the QT interval as a result of impaired baseline cardiac repolarization, which is caused by blockade of the delayed rectifier potassium channel, IKr [8,9]. The prolonged QT interval may lead to early afterdepolarizations (EADs) of cardiac myocytes in the third phase of the cardiac action potential. If these EADs reach the threshold electrical potential, they can induce TdP, leading to fatal ventricular fibrillation [10]. According to the US FDA Adverse Event Reporting System, erythromycin has caused approximately half of the reported macrolide-related TdP cases; clarithromycin, 36%; and azithromycin, 11% [11]. Numerous studies have demonstrated that the association between macrolides and the risk of TdP may largely be mediated by an acute toxic mechanism, supported by both the higher risk of arrhythmia observed in current macrolide users and by the fact that ventricular arrhythmias mainly occur during short-term therapy [12,13].

However, other studies have demonstrated that an increased cardiovascular mortality risk persists for several years after discontinuation of macrolide antibiotics. Schembri and colleagues analyzed the data from two prospective cohort studies in patients admitted to the hospital with acute exacerbations of chronic obstructive pulmonary disease (COPD) and community-acquired pneumonia (CAP) [14]. They compared cardiovascular mortality at 1 year between those who were prescribed clarithromycin and those who were not. Their data suggested that an increased mortality risk may have persisted beyond the time when clarithromycin was stopped. Another double-blind, placebo-controlled trial showed that the increased mortality risk persisted for 3 years in patients with coronary heart disease after discontinuation of clarithromycin [15]. The pro-arrhythmic effects of macrolides usually do not manifest after the discontinuation of these drugs. Therefore, there appears to be a separate mechanism that may be responsible for the above-mentioned adverse cardiovascular events. Winkel and colleagues revealed that clarithromycin-activated macrophages can initiate an inflammatory cascade and, as a result, form vulnerable plaques over time. Plaque formation increases the risk of acute coronary syndrome or sudden cardiac death due to plaque rupture [16]. Other evidence, also provided by Winkle et al., further supports this theory. They noted that the incidence of cardiovascular-related deaths among those taking clarithromycin without statins was greater than that among those taking clarithromycin and statins [16]. Statins inhibit HMG-CoA reductase-mediated production of cholesterol. They have been shown to delay the development of coronary artery atherosclerosis and to maintain the stability of coronary artery lesions [17]. Therefore, the increased risk of sudden cardiac death in patients with stable coronary heart disease who are not taking statins following clarithromycin treatment suggests that plaque formation and rupture may be a result of the sudden cardiac death risk induced by macrolide antibiotics.

Thus, macrolides should be prescribed with great caution in patients with preexisting coronary heart disease because these patients may already have coronary artery plaque accumulations [18]. However, this hypothesis is inconsistent with the findings of other studies showing that macrolide antibiotics exert significantly beneficial effects on cardiovascular morbidity in patients with acute coronary syndrome. Macrolide antibiotics have been used in patients with stable coronary heart disease to eradicate Chlamydia pneumoniae from atherosclerotic plaques, rather than to control common infections [19]. It has been reported that infections have been found to play a role in atherosclerosis. C. pneumoniae has been found in atherosclerotic tissue, and infection may elicit pathogenic events in vascular wall cells and the leukocytes of atheromatous lesions, potentially leading to atherosclerosis and acute coronary syndrome [20]. Because macrolides, such as clarithromycin and azithromycin, are highly efficacious against C. pneumoniae, exposure to macrolide antibiotics may positively affect the risk of developing acute coronary disease and sudden cardiac death [21,22]. Some small trials have shown significantly beneficial effects with macrolides; however, several randomized, multicenter trials have demonstrated either no beneficial effects or an increased risk of cardiovascular events following macrolide treatment for patients with stable coronary heart disease [23–25]. Why macrolides appear to have an unfavorable effect in patients with coronary heart disease remains unclear. The plaque formation hypothesis seems to, at least in part, support this phenomenon. Furthermore, the discrepancy among trials comparing the risk of cardiovascular events following macrolide use may be related to the duration of follow-up. Follow-up of 1 year reveals that macrolides have no significant impact on cardiovascular events. However, follow-up of more than 2 years shows that antibiotics are significantly associated with increased mortality [26]. A centrally randomized, placebo-controlled, multicenter trial extended the duration of follow-up to 6 years and found that cardiovascular deaths remained higher in the clarithromycin group [27]. These results suggest that macrolides can increase the risk of potentially lethal cardiovascular events in the long term. However, this finding may not be generalizable but rather may be limited to patients with coronary heart disease.

For patients without apparent coronary heart disease, whether macrolides have unfavorable effects related to plaque formation or rupture remain unclear. To better examine the link between macrolide treatment and plaque formation, the duration of follow-up should be at least several months because it often takes months to years for plaque formation to occur. However, for patients taking macrolides to treat common respiratory infections or soft-tissue infections, the duration of follow-up is usually no more than 1 month [28,29]. It is difficult to identify the cardiovascular death risk related to plaque rupture in such a short time. Only for patients with moderate-to-severe CAP or acute exacerbations of COPD can the duration of follow-up last 1 year. Interestingly, two studies reporting the association between clarithromycin use and hazard ratios for cardiovascular events at 1 year showed significantly higher admission rates for acute coronary heart syndrome following the administration of clarithromycin. In the first study, Schembri and colleagues analyzed two large prospectively collected data sets. One was a multicenter observational study of patients admitted to the hospital with acute exacerbations of COPD disease, and the other was the Edinberg CAP study cohort. During 1 year of follow-up, the risk of cardiovascular events, such as acute coronary syndrome and myocardial infarction, increased greatly among clarithromycin users in both patient cohorts, suggesting a potential association between macrolide treatment and the onset of ischemic cardiovascular events. The second study used an observational design to examine a database of 1323 patients with COPD and 1631 patients with CAP [30]. The hazard ratios for acute coronary syndrome at 1 year were 1.88 in patients with COPD and 1.90 in patients with CAP following clarithromycin use. There was a significant association between macrolide treatment and the risk of acute coronary syndrome. Taken together, these results suggest that even in patients without apparent coronary heart disease, macrolides should be prescribed with consideration for long-term cardiovascular safety. However, this finding may not be generalizable but rather may be limited to high-risk populations because patients with COPD or CAP are more likely to be older and have a history of other diseases.

Interestingly, mutations in the human KCNE2 gene, which encodes a cardiac and epithelial K(+) channel β subunit, result in increased genetic predisposition to coronary artery disease [31]. In vivo studies using KCNE2 knockout mice showed that KCNE2 deletion promotes atherosclerosis, increases plaque deposition more than 6-folds, and also causes sudden death [31]. A different single nucleotide polymorphism (SNP) within the KCNE2 gene has been associated with increased predisposition to cardiac arrhythmias, such as ventricular fibrillation. One variant, Q9E (a KCNE2 potassium channel gene SNP), as demonstrated by the Goldstein laboratory, is associated with macrolide-induced TdP because it increases IKr channel blockade by the antibiotic [32]. It is intriguing to consider that KCNE2 may provide a common genetic link between not only inherited TdP and atherosclerosis, but also macrolide-induced TdP and plaque rupture.

We have described two macrolide-induced mechanisms leading to sudden cardiac death. The first is ventricular arrhythmia induced by QT interval prolongation. The second is plaque formation and rupture initiated by activated macrophages. To our knowledge, the common risk factors for sudden cardiac death after macrolide administration include female sex, older age, QTc interval-prolonging drugs and inhibitors of their metabolism, hypokalemia, and preexisting cardiac disease [33]. These factors may increase the risk of arrhythmia, especially TdP. While patients, especially those with atherosclerosis or other ischemic heart diseases, may be highly susceptible to sudden cardiac death due to plaque rupture after short-term treatment with macrolides.

The implications of these findings for clinical decision-making are as follows: In young patients with low-baseline cardiovascular risks, the use of macrolides is associated only with the risk of short-term sudden cardiac death, which is possibly attributable to drug-induced fatal arrhythmia. However, in elderly patients with higher baseline cardiovascular risks, the use of macrolides is associated with higher risks of both short- and long-term cardiovascular death, with the latter possibly caused by inflammation-induced plaque rupture.

In conclusion, macrolide use may confer a significantly increased long-term risk of sudden cardiac death, which is apparently attributable to plaque rupture. Given the widespread use of macrolide antibiotics, these findings call for further confirmation in well-designed randomized clinical trials.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.