Patients with rheumatoid arthritis who are treated with methotrexate (MTX) are more than twice as likely to develop venous thromboembolism (VTE) when compared with patients who use hydroxychloroquine, according to data from a propensity score–matched cohort study.
“As the effect of these medications on the risk of VTE is largely unknown, we aimed to compare the rate of incident VTE after initiating MTX versus hydroxychloroquine among older patients with RA,” wrote Mengdong He, MHS, and coauthors from Brigham and Women’s Hospital and Harvard Medical School, both in Boston. Ms. At the time of the study, Ms. He was a research specialist but is now a medical student at the University of California, Los Angeles.
The results were published in Seminars in Arthritis and Rheumatism.
Using U.S. Medicare claims data from 2008 to 2017, the researchers identified patients with RA aged 65 years and older who initiated MTX or hydroxychloroquine without prior use of any immunomodulators for at least 365 days (that is, index date). Patients who used any conventional (other than methotrexate and hydroxychloroquine), biologic, or targeted synthetic disease-modifying antirheumatic drugs (DMARDs) any time prior to the index date were excluded.
The primary outcome of interest was incident VTE, a composite endpoint of pulmonary embolism (PE) or deep vein thrombosis (DVT). Secondary outcomes were PE, DVT, and all-cause mortality.
Results
After applying the eligibility criteria, a total of 68,648 RA patients who initiated either MTX (n = 41,197) or hydroxychloroquine (n = 27,451) as their first DMARD were identified and included in the analysis.
After 1:1 propensity score matching, the cohort consisted of 26,534 matched pairs of MTX and hydroxychloroquine initiators. The mean age was 74 years (standard deviation, 7 years), and 79% of the patients were female.
During a total of 56,686 person-years of follow-up, VTE occurred in 208 MTX (incidence, 6.94 per 1,000 person-years) and 83 hydroxychloroquine initiators (incidence, 3.11 per 1,000 person-years).
Patients who initiated MTX without prior use of any DMARDs had a higher risk of PE (hazard ratio, 3.30; 95% confidence interval, 2.28-4.77) and DVT (HR, 1.53; 95% CI, 1.07-2.19) than hydroxychloroquine initiators. However, all-cause mortality did not differ between the two groups (HR, 0.91; 95% CI, 0.83-1.00).
“MTX initiators had a relative risk of VTE higher than 2 and an absolute risk increase of about 4 per 1,000 person-years, compared with hydroxychloroquine initiators,” the authors wrote. “Results from the secondary outcome analyses were consistent and subgroup analyses found no meaningful treatment effect heterogeneity.”
The researchers acknowledged that a key limitation of the study was the use of claims-based algorithms to define outcomes. As a result, outcome misclassification is possible.
Dr. Kaleb Michaud
“While the study methodology was sound, patients with RA who receive hydroxychloroquine are very different than those who receive MTX, and it’s difficult to fully account for these differences using an administrative data set,” commented Kaleb Michaud, PhD, professor of internal medicine at the University of Nebraska, Omaha.
“Most clinicians are more interested in understanding the differences in VTE risk between MTX and Jakinibs [Janus kinase inhibitors] or MTX and biologics,” Dr. Michaud said.
“More research, particularly with randomized trials including the placebo arm, is needed to determine the causal relationships between the study drugs and VTE and whether MTX elevates or hydroxychloroquine reduces the risk of VTE,” the authors concluded.
The study was funded by internal resources in the division of pharmacoepidemiology and pharmacoeconomics at Brigham and Women’s Hospital and Harvard Medical School. Several authors reported financial relationships with the pharmaceutical industry.