Background/Purpose: Patients with rheumatoid arthritis (RA) have higher mortality than do persons of the same age and sex without RA. This is due in part to an increased risk of atherosclerosis and cardiovascular (CV) disease. The extent of atherosclerosis can be estimated non-invasively using high-resolution carotid ultrasound. In this study we examined the association of carotid atherosclerosis, as measured by carotid ultrasound, with mortality in a cohort of RA patients.

Methods: We recruited RA patients during a visit to their rheumatologist, and invited them to participate in a comprehensive clinical assessment that included a high-resolution carotid ultrasound. We also ascertained age, sex, RA duration, the CV risk factors and erythrocyte sedimentation rate (ESR), among other variables, using a predetermined protocol. After the assessment, we followed patients prospectively per protocol. We obtained a death certificate for all patients who died, from which we classified the causes of death using ICD9 codes. Deaths were classified as CV if the death certificate listed a CV condition, corresponding to ICD9 codes 390-459. We used standard time-to-event techniques to examine the association between the carotid ultrasound findings and all-cause and CV-mortality. Mortality rates are shown per 100 person-years. Rates and hazard ratios (HR) are shown with 95% confidence intervals.

Results:  We recruited 1,328 RA patients, of whom 1,197 had a carotid ultrasound. We followed them prospectively over 6,500 person-years, during which 206 deaths occurred, for a mortality rate of 3.1 (2.7, 3.6).  Death was attributed to a CV cause in 105 cases, for a CV mortality rate of 1.7 (1.4, 2.1).  The presence of carotid plaque resulted in an increased all-cause mortality rate, 4.6 (3.9, 5.47), compared to 1.6 (1.2, 2.1) with absent plaque.  The Figure shows a Kaplan Meier curve for all-cause mortality in patients with and without plaque (P < 0.0001, log rank test). Carotid plaque was also associated with an increased CV mortality rate, 2.7 (2.1, 3.4) vs. 0.8 (0.6, 1.2). The carotid intima-media thickness (IMT) was associated with increased all-cause mortality, with a HR of 2.28 per mm of IMT (2.00, 2.59), as well as CV mortality, with a HR of 2.26 per mm IMT (1.91, 2.68).  The associations of both carotid plaque and the IMT with all-cause mortality were independent of age, sex, CV risk factors and the ESR.  Their associations with CV mortality were independent of the all same variables, with the exception of the ESR.

Conclusion: Carotid atherosclerosis is significantly associated with mortality in RA, both all-cause and due to CV causes. The precise role of carotid ultrasound in mortality risk stratification and identification of candidates for intervention to reduce mortality in RA is an interesting area in need of study.