Background/Purpose: Previous studies showed that obesity in patients with rheumatoid arthritis (RA) is associated with higher disease activity. Beyond obesity, cardiometabolic disturbances may impact the prognosis of RA. We aimed to evaluate the impact of cardio-metabolic clustering (CMC) and of its components at the time of RA diagnosis on disease activity and radiographic progression through the 3 first years of evolution.

Methods: 494 patients from the French early arthritis cohort ESPOIR, fulfilling the ACR/EULAR 2010 criteria and with data available for CMC assessment, were included. CMC was defined as ≥ 2 abnormalities among low levels of HDL cholesterol, elevated levels of triglycerides, blood pressure ≥130/85 mmHg, elevated glycemia and insulin-resistance (Wildman RP et al. Arch Intern Med. 2008;168(15):1617-1624.).  Patients were split into 4 categories: non-obese with or without (w/o) CMC and obese patients (BMI≥30 kg/m²) w/o CMC. Baseline characteristics were compared using Chi² and Kruskal Wallis tests. The evolution of DAS28-ESR and total Sharp score at 6 months, 1, 2 and 3 years was compared between the different groups using univariate and multivariate mixed models. Multivariate mixed models including CMC components were used to assess parameters associated with outcomes. Multivariate models including obesity, age, gender, rheumatoid factor or ACPA positivity, CRP (for DAS28 model) and DAS28 (for total Sharp model) were used to assess independent associations between HDL, glycemia and RA outcomes.

Results: 192 and 229 non-obese patients respectively with and without CMC and 55 and 18 obese patients respectively with and without CMC were included. At baseline, age, gender, rheumatoid factor positivity and DAS28 were significantly different between the 4 groups. In univariate mixed models, DAS28 and HAQ evolution during the 3 first years were significantly different between the 4 groups (p<0.001), with higher values in obese patients and in patients with CMC.  CMC was associated with higher DAS28 and HAQ through the 3 years in non-obese patients (p=0.003 and p<0.001), but not in obese patients. CMC was also associated with a higher total Sharp score only in non-obese patients (p=0.02). In a multivariate model, the only components of CMC associated with outcomes were HDL cholesterol level and glycemia. Low HDL cholesterol level at baseline was an independent predictor of DAS28 during the 3 first years after multiple adjustment (p=0.003). Similarly, hyperglycemia at baseline was associated with a higher total Sharp score (p=0.035).

Conclusion: In early RA, low HDL cholesterol level and hyperglycemia are independently associated with subsequent higher DAS28 and more severe radiographic progression, respectively. Beyond cardiovascular complications, systematic screening of the CMC in RA patients will help to delineate more accurately disease prognosis.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/dyslipidemia-and-hyperglycemia-two-cardiometabolic-parameters-independently-predict-poor-outcome-in-early-rheumatoid-arthritis-results-from-espoir-cohort/