Background:The relationship between serum lipoproteins and cardiovascular disease risk in rheumatoid arthritis (RA) is complex1. Their levels and function may vary based on disease activity and medication use. Beneficial effects on high-density lipoprotein (HDL-C) levels, structure and behavior, in response to treatment have been described. However, the impact of HDL-C levels over time on coronary atherosclerosis progression in RA is unknown.Objectives:We here evaluated the influence of HDL-C levels over time on long-term coronary plaque formation and progression in patients with RA.Methods:One hundred one RA patients without symptoms or history of cardiovascular disease who participated in a computed tomography angiography study of coronary atherosclerosis had repeat assessments after 6.9±0.3 years to evaluate plaque progression. Clinical, laboratory and medication data were recorded at baseline and regular outpatient follow-up visits thereafter. Time-averaged HDL-C was calculated for each patient using available consecutive HDL measurements between baseline and follow-up. Robust logistic regression assessed the association between time-averaged HDL-C and likelihood of new plaque formation in segments without plaque at baseline, and transition of prevalent mixed plaque to calcified plaque. Robust multinomial logistic regression evaluated the effect of time-averaged HDL-C on likelihood of new non-calcified, mixed or calcified plaque formation in segments without plaque (compared to remaining without plaque), and non-calcified plaque regression or transition to mixed or calcified plaque at follow-up (compared to remaining non-calcified). All models accounted for clustering of coronary segments within patients and adjusted for Framingham D’Agostino risk score, proximal segment location, time-averaged CRP, cumulative prednisone dose, bDMARD duration, statin duration, waist-to-height ratio, and time-averaged triglycerides.Results:Participants were mostly female (n=87, 86.1%), with a mean ± standard deviation (SD) age of 51.5±10.3 years and time-averaged HDL-C of 51.7±13.9. Ninety-seven new plaques formed in segments without plaque at baseline; 20 were noncalcified, 21 were mixed, and 56 were calcified. Time-averaged HDL-C had no effect on new total plaque formation (adjusted odds ratio-OR 0.88 [95% CI 0.64-1.21]). However, each 1-SD increase in time-averaged HDL-C associated with a 44% reduced likelihood of new non-calcified plaque formation at follow-up (adjusted OR 0.56 [95% CI 0.35-0.92], Figure 1). In contrast, there was no effect of time-averaged HDL-C on new mixed or calcified plaque formation. Of 98 non-calcified plaques at baseline, 42 did not change at follow-up, 32 regressed (disappeared), 16 transitioned to mixed and 8 to calcified plaques. Each SD increase in time-averaged HDL-C yielded a 2.2-fold greater likelihood of non-calcified plaque regression (adjusted OR 2.21 [95% CI 1.02-4.83]). Sixteen of 52 mixed plaques present at baseline transitioned to more stable calcified lesions, and time-averaged HDL-C (per 1-SD increment) predicted a 3.5-fold increased likelihood of transition of mixed to fully calcified plaque (adjusted OR 3.56 [95% CI 1.25-10.17]).Conclusion:Higher HDL-C over time predicted regression of existing and decreased formation of new higher-risk non-calcified plaque. It also associated with transition of vulnerable mixed plaque to more stable fully calcified plaque. These effects were independent of RA treatment duration, prednisone dose and statin exposure.References:[1]Toms TE et al. Curr Vasc Pharmacol. 2010;8:301–326.Figure 1.Impact of HDL-C over time on coronary plaque progression in RADisclosure of Interests:George Karpouzas Speakers bureau: Sanofi/Genzyme/Regeneron, Consultant of: Sanofi/Genzyme/Regeneron, Grant/research support from: Pfizer, Sarah Ormseth: None declared, Elizabeth Hernandez: None declared, Matthew Budoff: None declared