Objective: An accelerated progression of atherosclerosis may contribute to the increased mortality due to cardiovascular disease reported in rheumatoid arthritis (RA). The aim of this study was to identify variables, related to disease onset as well as to disease progression, of importance for the presence of atherosclerosis, as diagnosed by B-mode ultrasonography, in patients with medium-term RA. The results are based on the co-analysis of retrospective data as well as cross-sectional data. The impact of RA per se on atherosclerosis was evaluated relative to age- and sex-matched controls. Methods: Thirty-nine RA patients, with a maximum age of 65 years, who had previously been included in a large retrospective cohort study, were assessed by duplex scanning after a disease duration of 19-23 years. In the present study, factors identified in the two earlier studies were assessed for their potential relationship with intima-media wall thickness (IMT) of the common carotid artery (CCA), and the presence and grade of atherosclerotic plaques of the CCA and the common femoral artery, in regression models. The candidate co-variates were: variables reflecting inflammatory activity at disease onset and at the time of ultrasound assessment, established cardiovascular risk factors, pharmacological treatment [corticosteroids, disease-modifying anti-rheumatic drugs (DMARDs)], and the presence of complications and co-morbidity identified during disease progression, as well as lipid levels, anti-lipid antibodies, haemostatic factors, and markers of immune activation measured at ultrasound assessment. Results: In patients with RA, analysis of simple linear regression models revealed those variables significantly associated with IMT-CCA to be age, tissue plasminogen activator (tPA) antigen, cholesterol, low density lipoprotein (LDL)-cholesterol, triglycerides, and atherosclerotic plaques while neither inflammatory status at disease onset, traditional cardiovascular risk factors, or pharmacological treatment during disease had any significant impact on IMT. In an estimated multiple linear regression model, variables associated with increasing log of IMT-CCA were the log of cholesterol and of soluble intracellular adhesion molecule 1 (sICAM-1), while methotrexate treatment tended to have a decreasing effect. In simple binary logistic regression, atherosclerotic plaques were associated with age, IMT-CCA, smoking, and the levels of sICAM-1, sE-selectin, interleukin-2 soluble receptor α (IL-2sRα), plasminogen activator inhibitor-1 (PAI-1) mass, cholesterol, LDL-cholesterol, and the LDL/high density lipoprotein (HDL) ratio. A multiple approach indicated that plaques were associated with age, cholesterol, and sE-selectin. Severe plaques were associated with LDL-cholesterol and disease duration. Logistic regression in the age- and sex-matched case-control study revealed that IMT-CCA was, together with the D-dimer, associated with RA per se. Conclusion: Levels of lipids and adhesion molecules were associated with the presence of atherosclerosis in RA. IMT-CCA was associated with RA per se. Disease duration could predict severe atherosclerotic plaques. Treatment with methotrexate seemed to decrease the IMT-CCA.