The comparison of cardiovascular disease risk prediction scores and evaluation of subclinical atherosclerosis in rheumatoid arthritis: a cross-sectional study

Objectives Primary objectives estimated prevalence of traditional cardiovascular disease (CVD) risk factors and compared different CVD risk prediction algorithms in an Indian rheumatoid arthritis (RA) population. Secondary objectives evaluated associations between carotid intima-media thickness (CIMT) and subclinical atherosclerosis (SCA) with CVD risk factors and CVD risk scores. Methods The presence of CVD risk factors were recorded, and 10-year CVD risk was predicted using Framingham risk scoring (FRS) using lipids (FRS-Lipids), FRS using body mass index (FRS-BMI), QRISK-2, SCORE, and the algorithm recommended by ACC/AHA (ASCVD). CIMT was measured on the far-wall of the common carotid artery. Subclinical atherosclerosis was defined as CIMT > 0.9 mm or the presence of carotid plaque. Results A total of 332 patents were enrolled, 12% had diabetes mellitus, 21.4% hypertension, and 6.9% were current/past smokers. Proportions of RA with predicted 10-year CVD risk > 10% varied from 16.2 to 41.9% between scores. Highest magnitude of risk was predicted by FRS-BMI. Agreement between scores in predicting risk was moderate in general. Mean CIMT was 0.70 ± 0.15 mm. Age, male sex, and extra-articular manifestations associated with greater CIMT. All risk scores except SCORE moderately correlated with CIMT. About one-seventh had SCA defined as CIMT > 0.9 mm or the presence of carotid plaques, associated with increasing age, male gender, or higher ratio of total cholesterol to high-density lipoprotein cholesterol. ASCVD and QRISK-2 scores had maximum area under curve for distinguishing SCA. Conclusion Individual CVD risk scores predict 10-year CVD risk differently in Indian patients with RA, and require validation for predicting hard end points (CVD events, mortality). Key Points • Diabetes mellitus and hypertension are the most prevalent cardiovascular disease risk factors in Indian patients with RA. • Individual cardiovascular risk prediction scores predict risk differently in Indian patients with RA, highest risk being predicted by the FRS-BMI. • Carotid intima-media thickness in RA associated with increasing age, male sex and extra-articular manifestations. • 14% RA had subclinical atherosclerosis, associated with increasing age, male sex, and higher total cholesterol to HDL-C ratio, best distinguished by ASCVD and QRISK-2 scores.