Background/Purpose: Endogenous estrogens (E) appear cardioprotective in women in the general population. Yet, proinflammatory effects were described in chronic autoimmune diseases depending on stage, cell-type involved, target organ environment and their metabolism. The impact of E on coronary atherosclerosis and cardiovascular (CV) risk in rheumatoid arthritis (RA) is unknown. We here explored the association between serum E, coronary atherosclerosis and CV risk in patients with RA.

Methods: Atherosclerosis (noncalcified, mixed or fully calcified, and low-attenuation plaque) was evaluated with computed tomography angiography in 140 patients without CV disease and reassessed in 100 after 6.9±0.4 years. Lesions with >50% stenosis were considered obstructive. Bioactive estrone (E1) and 17β-estradiol (E2) were measured at baseline and follow-up with ELISA. Multivariable robust negative binomial regression evaluated interactions of E with age and ACPAs predicting plaque numbers at baseline, and interactions of E with time-averaged C-reactive protein (TA-CRP) predicting numbers of new plaques. Multivariable Cox regression evaluated the association of E with CV risk. All models minimally adjusted for atherosclerotic CV disease (ASCVD) score while plaque progression and CV risk models further adjusted for baseline plaque.

Results: E1 associated with total plaque burden, noncalcified and low-attenuation plaques in patients older than 55 (p-interaction = 0.006, 0.006 and 0.004, Figure 1A). Higher E1 associated with fewer total, mixed and fully calcified plaques in ACPA negative (p-interaction = 0.017, 0.012 and 0.008) and more mixed plaques in ACPA positive patients (Figure 1A). The interactions of E2 with age > 55 years and ACPA status were not significant (not shown). In patients with increasing -but not decreasing- E1, higher TA-CRP(ln) associated with new total and noncalcified plaques (p-interaction = 0.002 and 0.022, Figure 1B). Likewise, higher TA-CRP(ln) associated with new noncalcified lesions only in patients with increasing E2 (p-interaction = 0.050); in contrast, in those with decreasing E2, higher TA-CRP associated with new fully calcified (low-risk) lesions (p-interaction = 0.037, Figure 1B).

E1 inversely associated with CV risk (adjusted hazard ratio- aHR 0.39, 95% confidence interval- CI 0.19-0.81 per standard deviation-SD increase in E1, Figure 2A) while E2(ln) did not (HR 0.87, 95% CI 0.50-1.51 per SD increase). Addition of E1 to ASCVD and baseline plaque enhanced CV risk prediction; area under the curve (AUC) increased from 0.81 (95% CI 0.66-0.95) to 0.85 (95% CI 0.73-0.96) and integrated discrimination improvement (IDI) = 0.06 (standard error 0.023), p = 0.008 (Figure 2B). The inverse association of E1 with CV risk was stronger in patients < 55 (HR 0.17, 95% CI 0.06-0.48) and those without obstructive plaque (HR 0.21, 95% CI 0.09-0.48, p-interaction = 0.008 and 0.002).

Conclusion: Conclusion: Effects of E on atherosclerosis vary according to age and ACPA status, and their increase over time yielded a stronger association between cumulative inflammation and plaque progression. E1 inversely associated with CV risk, suggesting additional benefits of E on plaque biology and stability.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/age-seropositivity-and-inflammation-determine-risk-versus-benefit-of-endogenous-estrogens-on-coronary-atherosclerosis-in-rheumatoid-arthritis/