Background/Purpose: Chronic inflammatory diseases such as rheumatoid arthritis (RA), psoriatic arthritis (PsA), and psoriasis (PsO) increase the risk of cardiovascular (CV) disease. However, it is unclear what factors, including DMARD treatment, after a first CV event may affect risk of subsequent CV events. We estimated and compared the incidence of subsequent CV events among patients with RA, PsA, or PsO who were being treated with DMARDs prior to the initial CV event. 

Methods: Patients with RA, PsA, or PsO, who experienced a major non-fatal CV event (acute myocardial infarction, stroke, or cardiac revascularization) between 1/1/2006 and 6/30/2015 were identified in an administrative claims database. Index date was defined as the hospital discharge date for the first non-fatal CV event during the study period. Eligible patients were: continuously enrolled for 12 months prior to index date; had ≥1 DMARD claim (categorized as TNF inhibitor (TNFi), conventional synthetic [csDMARD], or a non-TNFi biologic DMARD including tofacitinib) within 12 months prior to the index date; and had ≥30 days of follow-up after index date. The primary outcome was the occurrence of any CV event during the follow-up period. Incidence rates (IR) per 1,000 person-years with 95% confidence intervals (CI) were calculated and age and sex standardized to the general population. The hazard ratio (HR) and 95% CI for a subsequent CV event was estimated using Cox proportional hazard models. Patients who were not treated with any DMARD after initial CV event were not included in the analyses.

Results: We identified 8,610 patients with RA, PsA, or PsO eligible for study. After the index date, 2,924 (34.0%) patients used a TNFi, 4,813 (55.9%) used a csDMARD as monotherapy or combination, and 873 (10.1%) used a non-TNFi biologic DMARD. Median follow-up time after initial CV event was 1.6 years. Patients using non-TNFi biologic DMARDs had higher crude incidence rates of repeat CV events than those who used TNFi or csDMARDs (Table 1). The multivariate adjusted hazard ratios for subsequent CV events were 0.98 (95% CI, 0.82-1.17) for csDMARDs, and 1.16 (0.86-1.57) for other biologic DMARDs compared to TNFi (Table 2). RA diagnosis (as compared to PsO diagnosis) and heart failure diagnosis in baseline were risk factors independently associated with increased risk of subsequent CV event.

Conclusion: In this large nationwide database reflecting typical clinical care, we found that type of DMARD use after initial non-fatal CV event was not associated with risk for subsequent CV event among RA, PsA, and PsO patients. Predictors of subsequent CV event included having RA and heart failure prior to initial CV event.