Rheumatoid arthritis (RA) is a systemic inflammatory condition associated with increased cardiovascular risk. This is not fully explained by traditional risk factors, but direct vascular inflammation and aortic stiffening may play a role. We hypothesized that patients with RA exhibit aortic inflammation, which can be reversed with anti-tumor necrosis factor-α therapy and correlates with aortic stiffness reduction. Aortic inflammation was quantified in 17 patients with RA, before and after 8 weeks of anti-tumor necrosis factor-α therapy by using (18)F-fluorodeoxyglucose positron emission tomography with computed tomography coregistration. Concomitantly, 34 patients with stable cardiovascular disease were imaged as positive controls at baseline. Aortic fluorodeoxyglucose target-to-background ratios (TBRs) and aortic pulse wave velocity were assessed. RA patients had higher baseline aortic TBRs in comparison with patients who have cardiovascular disease (2.02±0.22 versus 1.74±0.22, P=0.0001). Following therapy, aortic TBR fell to 1.90±0.29, P=0.03, and the proportion of inflamed aortic slices (defined as TBR >2.0) decreased from 50±33% to 33±27%, P=0.03. Also, TBR in the most diseased segment of the aorta fell from 2.51±0.33 to 2.05±0.29, P<0.0001. Treatment also reduced aortic pulse wave velocity significantly (from 9.09±1.77 to 8.63±1.42 m/s, P=0.04), which correlated with the reduction of aortic TBR (R=0.60, P=0.01). This study demonstrates that RA patients have increased aortic (18)F-fluorodeoxyglucose uptake in comparison with patients who have stable cardiovascular disease. Anti-tumor necrosis factor-α therapy reduces aortic inflammation in patients with RA, and this effect correlates with the decrease in aortic stiffness. These results suggest that RA patients exhibit a subclinical vasculitis, which provides a mechanism for the increased cardiovascular disease risk seen in RA.