DMARD-specific risk of atrial fibrillation in rheumatoid arthritis: a new piece in the puzzle

This editorial refers to ‘New-onset atrial fibrillation in seropositive rheumatoid arthritis: association with disease-modifying anti-rheumatic drugs treatment’, by Hyung Woo Kim et al. 2023; https://doi.org/10.1093/rheumatology/kead336. Rheumatoid arthritis (RA) is a prevalent systemic inflammatory disease and is associated with an increased mortality rate [1]. Besides being a cause of joint structural damage, many diseases are associated with RA, including osteoporosis, depression, pulmonary disease, hypertension and cardiovascular disease [2]. In fact, RA is an independent risk factor for a diversity of cardiovascular (CV) conditions including atherosclerotic cardiovascular disease (ASCVD), heart failure (HF) and stroke [3]. ASCVD stands as the most common cause of death in RA and the relationship between these two entities is highly tied to the inflammatory burden [3]. In this sense, the control of systemic inflammation with disease-modifying anti-rheumatic drugs (DMARDs) is the major goal of RA treatment as inflammation represents a central trigger of both joint and cardiovascular manifestations in these individuals [3]. In addition to accelerating the development of atherosclerosis, RA is associated with increased incidence of cardiac arrhythmias, especially in women [4]. Atrial fibrillation (AF) is the most common arrythmia both in RA and the general population, and its development is promoted by advancing age, hypertension, diabetes, obesity, heart failure, coronary artery disease, inflammatory diseases and alcohol overuse [5]. Authors still debate, however, whether incident AF among individuals with RA is a direct effect of inflammation or whether AF is a consequence of poor control of cardiometabolic risk factors and the excess of AF-related diseases such as HF and ASCVD. Indeed, among 21 201 individuals with RA, there was a higher incidence of AF (incidence rate ratio, IRR 1.28), HF (IRR 1.53), stroke (IRR 1.19) and myocardial infarction (MI) (IRR 1.48) compared with non-RA individuals [6]. In this study, the undertreatment of cardiovascular (CV) risk factors explained a significant proportion of the increased risk for AF, HF and MI. Although less explored, there is also evidence suggesting a potential role of DMARDs on the risk of AF among AR individuals. Historically, it has been discussed whether hydroxychloroquine (HCQ) could predispose patients to cardiac arrhythmias (conflicting data), with scarce data on other classes [7]. In order to add a new piece into the puzzling relationship between AF and RA, Kim et al. present the study ‘New-onset atrial fibrillation in seropositive rheumatoid arthritis: association with disease-modifying anti-rheumatic drugs treatment’ in this issue of Rheumatology [8]. In this study, an extensive South Korean database was evaluated, and a nested case-control design was established to evaluate the risk factors for AF among seropositive RA (SPRA) individuals. Among the 108 085 patients with SPRA, 2652 who developed new-onset AF during follow-up (between 2010–2020) were compared with 10 500 individuals without AF matched for age, sex, follow-up time and disease duration. Having an insurance type of medical aid (aOR 1.21, 95%CI 1.02–1.42, P = 0.025), pre-existing comorbidities of hypertension (aOR 1.65, 95%CI 1.49–1.82, P < 0.001), chronic kidney disease (aOR 1.76, 95%CI 1.39–2.24, P < 0.001) and heart failure (aOR 1.93, 95%CI 1.62–2.30, P < 0.001) increased the risk of subsequent AF. Regarding the use of medications, the use of methotrexate (MTX) decreased the risk of AF (aOR 0.89, 95%CI 0.80–0.98, P = 0.019) and the use of leflunomide (LEF) increased the risk of AF (aOR 1.21, 95%CI 1.09–1.33, P < 0.001). In the subgroup aged >50 years, age, in addition to LEF, adalimumab (ADA) also increased the risk of AF. HCQ, other anti-TNF, non-anti-TNF biologicals and JAK inhibitors did not increase or decrease the risk for AF. The authors highlight potential mechanisms of toxicity arising from LEF and ADA on cardiac conduction pathways. This could be an explanation for the increased risk for AF, while the protective role of MTX would arise from its anti-inflammatory role. Bearing in mind this context, ask yourself: do we need to monitor all patients taking LEF and ADA for AF? Alternatively, should we avoid prescribing these drugs in patients at risk for AF? Although the study is full of positive points (its size, exploring the entire therapeutic arsenal for RA, thorough analysis of the data, and design), some major points should be remembered so that we can interpret the findings and their immediate translation into clinical care. Although individuals with and without new-onset AF were matched for disease duration and all had positive rheumatoid factor (two elements that contribute to the CV risk attributed to the disease), we do not know how disease activity behaved over time. Each RA flare boosts cardiovascular event risk [9], and that can be translated into higher risk for AF. In the study by Kim et al. the association between MTX/LEF and AF was not changed when adjusted by incident HF or ASCVD [8]. However, it is not excluded that subclinical forms of these manifestations could have contributed to new-onset AF. Along the same lines, the study associates a series of previous diseases that favour the development of AF, but little is known about the diseases that occurred during the study period, which may have contributed to AF. For example, LEF has historically been associated with increased blood pressure [10]. Could the potential increase in blood pressure be responsible for the increased risk of AF observed in this group? Some typical limitations of registry studies are also present. The first is the difficulty in defining exposure. The definition of exposure used in this study, ‘more than one month’ on each drug, may overestimate drug effects by including patients with short-term exposure. Additionally, the study did not provide details on the determinants that led to the choice of intervention, such as whether patients were not exposed to MTX or whether LEF was the first choice for those with side effects related to MTX. The lack of adjustment for important variables, such as alcohol use, is a major limitation of this study. Alcohol use is a known risk factor for AF [5], and it is also a side effect of MTX and a barrier to starting MTX [11]. As a result, the study's findings may be biased. Future studies should adjust for alcohol use to provide more accurate estimates of the risk of AF associated with LEF and MTX. In parallel, the impact of having an insurance type of medical aid on AF is symptomatic of a studied event (AF) that may remain underdiagnosed depending on the access to diagnostic procedures. This means that, due to any reason, LEF or MTX users could have been screened more frequently with 24h cardiac holter monitoring than users of other DMARDs. Finally, it is not possible to rule out that findings were obtained by chance as the results were not adjusted for multiple comparisons, and it is possible that unmeasured confounding factors such as ethnicity or the absence of a diverse ethnical background may limit the generalizability of this study. In summary, the study by Kim et al. opens new doors in the understanding of cardiovascular involvement tied to RA and raises the possibility that different DMARDs may translate into putative risks for AF development. However, given the historical knowledge of RA and what is known about the etiopathogenesis of AF, it is possible that both direct (through lesion on impulse conduction pathways) and indirect (immunoinflammatory-mediated or atherosclerotic-driven) mechanisms may play a role in AF development (Fig. 1). Potential direct and indirect mechanisms that associate methotrexate (MTX) and leflunomide (LEF) with atrial fibrillation (AF) All requests for raw and analysed data and related materials will be reviewed by the Clarity Healthcare Intelligence legal department to verify whether the request is subject to any intellectual property or confidentiality obligations. Requests for patient-related data can be considered upon request (contact: contato@clarityhealth.com.br). Any data and materials that can be shared will be released via a Material Transfer Agreement. The authors had full access to all the data and can take responsibility for the integrity of the data and the accuracy of the data analysis. They consent to the submission of the manuscript as it is. This work was supported by grant 2019/09068–3 from São Paulo Research Foundation (FAPESP), grants 371/2021 and 585/2022 from Federal District Research Foundation (FAPDF), grants 437413/2018–7 and 310718/2021–0 from the Brazilian National Research Council (CNPq). The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclosure statement: The authors have declared no conflicts of interest.