Differential Association of Psychosocial Comorbidities With Subclinical Atherosclerosis in Rheumatoid Arthritis

Abstract

Objective: Rheumatoid arthritis (RA) is associated with an elevated risk of cardiovascular disease (CVD) events and subclinical atherosclerosis, but the reasons for the excess risk are unclear. We explored whether psychosocial comorbidities, which may be associated with CVD in the general population, are differentially associated with subclinical atherosclerosis in RA compared to controls.

Methods: Data were from a longitudinal cohort study of 195 RA patients and 1,073 non-RA controls. Using validated scales, heterogeneity in the associations of psychosocial measures (depression, stress, anxiety/anger, support, discrimination/hassles) with measures of subclinical atherosclerosis (coronary artery calcium [CAC] and carotid intima-media thickness [IMT]/plaque) were compared in RA and non-RA groups using multivariable generalized linear models. Computed tomography and ultrasound were used to identify CAC and IMT/plaque, respectively. CAC >100 units was used to define moderate/severe CAC.

Results: In RA, per-unit higher anxiety scores (odds ratio [OR] 1.10, P = 0.029), anger scores (OR 1.14, P = 0.037), depressive symptoms (OR 3.41, P = 0.032), and caregiver stress (OR 2.86, P = 0.014) were associated with increased odds of CAC >100 units after adjustment for relevant covariates. These findings persisted despite adjustment for markers of inflammation (C-reactive protein and interleukin-6 levels) and were seen only in RA, not in controls (adjusted multiplicative interaction P = 0.001-0.077). In RA, job stress was associated with an increased risk of carotid plaque (adjusted OR = 3.21, P = 0.019), and increasing social support was associated with lower internal carotid IMT (adjusted P = 0.024).

Conclusion: Depressive symptoms, stress, anger/anxiety, and social support may preferentially affect CVD risk in RA, and screening/treatment for psychosocial morbidities in RA may help ameliorate the additional CVD burden.

Figure 1

Adjusted carotid plaque presence by job stress in those with RA compared with non-RA controls. In the RA group, job stress confers increased risk for carotid plaque presence (odds ratio [OR] 3.21 [95% confidence interval 1.21–8.25], adjusted P = 0.019, adjusted multiplicative interaction P = 0.210, and additive interaction P = 0.142). RA models adjusted for age, hypertension, education, and rheumatoid factor ≥40. Control models adjusted for hypertension and education. Adjusted multiplicative interaction P = 0.210 and adjusted additive interaction P = 0.142. Interaction models adjusted for education and hypertension. RA = rheumatoid arthritis.

Figure 2

The association between ICA IMT (logarithmically transformed scores) with the social support scale using line of fit and 95% confidence intervals in gray. In the RA group, as social support increases, log ICA IMT decreases. Adjusted P = 0.024, adjusted multiplicative interaction P = 0.206, and additive interaction P = 0.236. All models adjusted for age, sex, body mass index, education, low-density lipoprotein, and total cholesterol, and RA models adjusted additionally for past/current prednisone use and rheumatoid factor ≥40. Adjusted multiplicative interaction P = 0.206 and adjusted additive interaction P = 0.236. Interaction models adjusted for age, sex, body mass index, low-density lipoprotein, total cholesterol, and education. RA = rheumatoid arthritis; ICA = internal carotid artery; IMT = intima-media thickness; MOS = Medical Outcomes Study.