Atrial fibrillation (AF), the most common sustained arrhythmia, increases stroke risk by 5-fold and mortality risk by 2-fold.1
Patients with rheumatoid arthritis (RA) face an elevated risk of cardiovascular disease (CVD) and related mortality, driven by both traditional CVD risk factors and systemic inflammation. RA also increases the likelihood of developing atrial fibrillation (AF) by 1.4-fold, along with a similar rise in ischaemic stroke risk.
There is,however, limited understanding of whether atrial fibrillation (AF) patients with rheumatoid arthritis (RA) face a higher risk of ischaemic stroke compared to those without RA, and whether they are less likely to receive oral anticoagulant (OAC) therapy for stroke prevention. To address these gaps, researchers set out to investigate these potential risks.
To conduct this population-based, retrospective cohort study, the researchers used data from the Norwegian Cardio-Rheuma Register (NCRR), which includes Norway's entire adult population (4.7 million individuals aged 18 and older.)
Individuals who had at least one hospital encounter related to AF or atrial flutter were identified as AF patients, while at least two RA-related hospital encounters within two years defined RA patients. Comorbidities such as diabetes, hypertension, atherosclerosis, CVD, congestive heart failure, and prior stroke or transient ischaemic attack were recorded if there was at least one hospital encounter for those conditions.
OAC treatment was indicated by at least one dispensed prescription within three months of the first AF diagnosis, while patients were considered treated with disease-modifying antirheumatic drugs (DMARDs) if they received at least one prescription within three months after the initial AF diagnosis.
The primary endpoint was an ischaemic stroke, with secondary endpoints including cerebral infarction due to embolism or unspecified causes.
The following findings were recorded:
- During the five-year study period, 2,750 RA patients experienced an AF incident
- During follow-up,121 (4.4%) RA patients and 5,463 (3.4%) non-RA patients experienced an ischaemic stroke, with cerebral infarction due to unspecified causes being the most common
- Mortality was high in both groups, with 29.2% (803) of RA patients and 21.5% (34,121) of controls having died within the study period
- After one year, the unadjusted cumulative incidence of ischaemic stroke was 1.8% for RA patients and 1.4% for non-RA patients
- A subgroup analysis showed that men with RA had a higher cumulative incidence of stroke compared to men without RA, while women with RA were older and had a slightly increased stroke risk (1.5% vs 1.4% at one year). This trend continued for three years, indicating ongoing increased risk among women with RA
- AF patients with RA had a 36% overall increase in ischaemic stroke risk compared to non-RA patients (HR: 1.36). This risk remained similar after adjusting for diabetes, hypertension, atherosclerotic CVD, and OAC treatment
- Univariate analysis revealed that RA patients using anti-rheumatic treatment had a 26% lower risk of ischaemic stroke (HR: 0.74; 95% CI: 0.52, 1.07)
- Finally, RA patients were less likely to receive OAC treatment than their non-RA counterparts, potentially due to concerns about interactions with RA medications or bleeding risk.
In conclusion, this study provides evidence that RA is associated with an increased risk of ischaemic stroke in AF patients, independent of diabetes, hypertension, and atherosclerotic CVD. Furthermore, RA patients with AF were less likely to receive OAC treatment, which may partly explain this increased risk.
The researchers conclude that incorporating RA into the CHA2DS2-VASc score and expanding OAC coverage for AF patients with RA may help reduce the risk of ischaemic strokes, warranting further investigation.
Limitations: This study lacked data on smoking, blood pressure measurements, alcohol use, and obesity, which may have affected the comprehensiveness of the findings. Additionally, the study population was limited to Norway and may not be generalizable to other populations.
Access the original research article