To determine if the use of carotid ultrasonography (US) may improve the stratification of the cardiovascular (CV) risk in rheumatoid arthritis (RA). A set of 370 consecutive patients without history of CV events were studied to assess carotid intima-media thickness (cIMT) and plaques. As previously proposed, CV risk was calculated according to the modified EULAR systematic coronary risk evaluation (mSCORE) for RA that was adapted by the application of a multiplier factor of 1.5 in those patients fulfilling ≥ 2 of 3 specific criteria. The mean disease duration was 9.8 years, 250 (68%) had rheumatoid factor/anticyclic citrullinated peptide positivity and 61 (17%) extra-articular manifestations. 43 were excluded because they had type 2 diabetes mellitus or severe chronic kidney disease. CV risk was categorised in the remaining 327 RA patients according to the mSCORE: mild (96 cases; 29.3%), moderate (201; 61.5%) and high/very high risk (30; 9.2%). Only five patients were reclassified as having high/very high CV risk when the mSCORE was applied. Severe carotid US abnormalities (cIMT >0.90 mm and/or plaques) were uncommon in patients with low mSCORE (13%). Nevertheless, in patients with moderate mSCORE, severe carotid US abnormalities were observed in 63% of cases. A model that included a chart mSCORE risk ≥ 5% plus the presence of severe carotid US findings in patients with moderate mSCORE risk (≥ 1% and <5%) yielded high sensitivity for high/very high CV risk (93 (95% CI 88 to 96)). Our results support the use of carotid US in the assessment of CV risk in patients with RA.