Abstract
OBJECTIVE: Evaluate the risk and temporal trends of heart failure (HF) with preserved (HFpEF) and reduced ejection fraction (HFrEF) in rheumatoid arthritis (RA). METHODS: We performed a retrospective, matched cohort study using Veterans Health Administration (VHA) administrative and health record data from 2000 to 2019. Patients with RA were matched up to 10 non-RA controls on age, sex, and VHA enrollment year. Incident HF and HF-related death were queried, classifying HFpEF and HFrEF using left ventricular EF data from a validated natural language processing tool. HF risk was evaluated using conditional Cox regression, adjusting for demographics, body mass index, smoking, rurality, healthcare utilization, and comorbidity burden. Trends in HF risk were evaluated, stratifying models by RA diagnosis period (2000-2005, 2006-2011, 2012-2017). RESULTS: We matched 67,850 patients with RA (mean age 62.5, 87.1 % male) to 570,933 non-RA controls (mean age 61.1, 85.8 % male). HF prevalence increased in both groups over time. Over 5,663,151 person-years of follow-up, 77,440 incident HF diagnoses occurred. RA was associated with an increased risk of HFpEF (aHR 1.51, 95 % CI 1.46-1.57) and HFpEF-related death (2.05, 1.76-2.39), as well as HFrEF (1.34, 1.30-1.38) and HFrEF-related death (1.45, 1.29-1.63). HF risk was accentuated in RA patients with elevated inflammation and seropositive RA. No improvements in HF risk were observed over time (linear p-for-trend >0.05 for all outcomes). CONCLUSION: RA was most strongly associated with HFpEF and HFpEF-related death in this national-level, observational dataset. Heightened risks of HF subtypes have not improved despite advances in RA treatment. Prospective research is needed to support the development ofprevention and management strategies to mitigate HF risk in RA.