Findings from a large population-based cohort study have shown that individuals with rheumatoid arthritis (RA) are at an elevated risk of venous thromboembolism (VTE) regardless of age, sex, body mass index (BMI), disease duration or the prescription of exogenous oestrogens.
Published in Rheumatology, authors Mark D Russell (King’s College London, London, UK) and colleagues state that this increased risk of VTE can be driven by a range of factors, including a systemic inflammatory state, reduced mobility and the use of corticosteroids. However, the authors state that it remains poorly understood the extent to which excess risk of VTE in patients with RA is influenced by these factors.
In 2022, findings from the Oral Rheumatoid Arthritis Trial (ORAL) raised concern over a signal for increased VTE incidence among individuals with RA treated with Janus kinase (JAK) inhibitors, leading to the issuance of a regulatory warning. Considering this, Russell and colleagues set out to contextualise any potential risk and to characterise potential contributing factors in individuals with RA.
The study population included adults who had prevalent or incident diagnosis of RA between 1999–2018 with no prior history of VTE. Patients were required to have been registered with the UK Royal College of General Practitioners Research (RCGP) Research and Surveillance Centre (RSC) database within the study period. Enrolled patients were matched 1:4 with individuals without RA and a multivariable-adjusted Cox proportional hazards model was used to compare VTE risk stratified by age, sex, BMI, disease duration and prescription oestrogen-containing contraceptives or hormone-replacement therapy (HRT).
The researchers included a total of 117,050 patients, of whom 23,410 had diagnoses of RA and 93,640 were matched controls. Average follow-up was 8.2 years. Mean age of enrolees was 59 years; 71.1% were female; mean BMI was 27.1kg/m2. Regarding years since RA diagnosis, 63.9% were included at or within two years, 7.8% within 2–5 years, 9.8% within 5–10 years, and 18.5% at 10 years or more. Individuals with RA were prescribed nonsteroidal anti-inflammatory drugs (NSAIDs) more frequently than matched controls (53.4% vs. 27.2%), as well as oral corticosteroids (26.7% vs. 5.9%) and immunosuppressant medications in a primary care setting (48.2% vs. 0.9%).
The authors report that 845 VTE events occurred during the study period in patients diagnosed with RA, compared to 2,020 events in the control group. Unadjusted absolute incidence rates for VTE were higher in people with RA (4.42 events per 1,000 person-years [py]; 95% confidence interval [CI] 4.13, 4.73) than matched controls (2.62 per 1,000py; 95% CI 2.52, 2.74). In multivariable-adjusted Cox models, the risk of VTE was higher in individuals with RA than matched controls (adjusted hazard ratio [HR] 1.46; 95% CI 1.36, 1.56; p<0.001).
Additionally, Russell et al share that the absolute incidence of VTE was higher in males and increased with age in individuals with RA. They relay that higher BMI correlated with increased risk of VTE, while disease duration did not affect VTE incidence in individuals with RA.
“Individuals with RA who are aged under 50 years and those with normal BMI remain significantly more likely to experience VTE than individuals without RA,” Russell and colleagues write. “In these circumstances, when considering additional treatments or procedures that might increase the risk of VTE, it is important to explore additional factors that could influence risk-benefit decisions, for example, a previous history of VTE, malignancy, chronic kidney disease, active smoking or corticosteroid use,” they continue.
Stating that theirs is the “first comprehensive analysis” of VTE risk associated with exposure to exogenous oestrogens in patients with RA, Russell and colleagues note that excess risk was comparable between females prescribed oestrogen-containing contraceptives and those who were not. However, the risk of VTE was greater in those prescribed HRT than those who were not.
In their Rhumatology paper, Russell and colleagues bring awareness to several limitations of their study, including the routinely collected UK primary care data which may make their findings ungeneralisable to other countries or healthcare systems. Further, they detail that there is “potential for unmeasured confounding and diagnostic misclassification” inherent to studies which utilise routinely collected health data.
Yet, acknowledging the strength of their present findings, Russell and colleagues note that their large sample size and lengthy follow-up duration can “establish a dataset which is representative of the UK’s population”, to highlight the “need to consider VTE risk in all individuals with RA”, they conclude.
