Hydroxychloroquine: Not a Heart Breaker!

Prior to 2020, there was a lack of general awareness of the potential for hydroxychloroquine (HCQ) and chloroquine (CQ) to cause cardiac arrhythmias, as rheumatologists and dermatologists have prescribed these medications for decades with no obvious cardiac safety signal. And while HCQ and CQ are listed as drugs with a known risk of torsades de pointes (www.credibleMeds.org) due to their effect on cardiac ion channels, there was little thought of HCQ causing arrhythmias (1). However, with the recent coronavirus disease 2019 (COVID-19) pandemic, reports of HCQ use being associated with prolongation of the corrected QT (QTc) interval and torsades de pointes emerged (2, 3). Granted, azithromycin, which is another agent known to block potassium cardiac ion channels that prolong the QTc, was frequently co-administered in patients infected with COVID-19 who were also treated with HCQ. In addition, the patients receiving these medications were usually hospitalized and often quite ill. These patients often had some evidence of cardiac involvement secondary to COVID-19, which itself was found to be cardiotoxic (4, 5). All of these factors could increase the risk of cardiac arrhythmias. In this issue of Arthritis Care & Research, Gupta et al report on the reduction of atrial fibrillation (AF) incidence among patients with systemic lupus erythematosus (SLE) who were treated with HCQ (6). In this retrospective cohort analysis of adult patients with SLE from December 1, 2014 to May 30, 2017, a total of 1,647 patients were included, of which 917 were HCQ users and 730 were nonusers. Patients with previous AF were excluded from the analysis, and episodes of AF in the first year of use of HCQ were not counted as end points, allowing for a run-in period. HCQ users were older, less likely to have hypertension, and more likely to have coronary artery disease, heart failure, and diabetes mellitus, which are all risk factors for AF. All comorbidities, except for the lower incidence of hypertension, would favor a higher risk of AF in HCQ users. A total of 23 AF events were captured, including 3 in HCQ users and 20 in nonusers. Despite the more common risk factors for AF in HCQ users, even in the unadjusted analysis, administration of HCQ was associated with decreased incidence of AF. When controlling for AF risk factors with logistic regression analysis, the odds ratio for AF was 0.12, with a 95% confidence interval of 0.034–0.39 (P = 0.0005). The propensity score matching method showed essentially the same result. There were 4 episodes of ventricular arrhythmia, including 1 ventricular fibrillation in a nonuser and 2 ventricular tachycardias and 1 torsade de pointes in the HCQ user group, of which none were fatal. Although numerically higher in the HCQ users, this was not statistically significant. Still, based on the mechanism of action of HCQ on cardiac ion channels, an increase in risk of ventricular arrhythmias is plausible. While this study was intended to highlight the potential benefit of HCQ on AF incidence, it supports the concept that HCQ possesses cardiac ion channel–blocking properties and should be viewed in this light. HCQ and CQ are derivatives of quinidine. Quinidine is one of the earliest-developed anti-arrhythmic drugs that is efficacious for both atrial and ventricular arrhythmias but has a proarrhythmic risk of torsades de pointes (7). Its efficacy lies primarily in blockade of the rapid cardiac delayed-rectifier potassium current (IKr), which HCQ and CQ block as well. Blockade of this channel prolongs repolarization, leading to increased refractory time, the basis of its anti-arrhythmic effect in cardiac tissues. Not only is this channel the abnormal channel in long QT syndrome type 1 (8), but it is the same channel that is blocked by anti-arrhythmic drugs developed to treat atrial and ventricular arrhythmias, such as sotalol, dofetilide, and amiodarone. HCQ and CQ are moderate IKr blockers, similar to macrolides, antipsychotics, and antifungals in their potency to block IKr. Like anti-arrhythmic drugs, HCQ and CQ do have predictable actions on the QT interval (9). This has been observed in animal studies and now more recently in COVID-19–infected patients treated with HCQ, particularly when co-administered with azathioprine, another IKr blocker (3, 5). There are also now both case reports and larger series that show HCQ does have QT prolongation effects in individuals with SLE and those with rheumatoid arthritis in the pre–COVID-19 era (9-11). Multiple randomized and observational studies have shown that anti-arrhythmic agents that block IKr have been associated with a reduction in AF risk (12). However, these agents raise the concern for prolongation of QTc and, accordingly, increased risk of ventricular arrhythmias, including torsades de pointes and subsequent sudden cardiac death. The study by Gupta et al (6) does have a limitation of being a retrospective study, and while logistic regression and propensity score matching methods showed an association with reduced AF incidence, this is an association and not necessarily a causal relationship. Residual confounding and indication bias for HCQ use or nonuse could still be playing a role. In addition, the authors do not report QTc changes with HCQ, nor adherence to HCQ, which is a known issue (13). Still, the authors are to be commended for the evaluation of HCQ as a potential anti-arrhythmic agent and not just as a proarrhythmic agent. As lupus may be a risk for AF, and those with AF had higher mortality in 1 study, a medication that is associated with a lower incidence of AF would be beneficial in this high-risk population (14, 15). This current study adds to the data that there is a cardiac effect of HCQ. Further studies that more clearly delineate HCQ adherence and any association of blood levels of HCQ as it relates to its effect on the IKr channel will be helpful. In addition, larger data-based studies on HCQ exposure and its impact on the QTc are necessary before drawing conclusions regarding HCQ cardiac effects. At this point, it is unclear about what type of screening should be done as we follow these patients. Finally, we need to be mindful that HCQ is and has been used for years in patients with SLE and rheumatoid arthritis with clear benefits in mitigating disease activity and progression, and this benefit may far outweigh any particular risk. Both authors drafted the article, revised it critically for important intellectual content, approved the final version to be published, and take responsibility for the integrity of the data and the accuracy of the data analysis.