Objective The present study aimed to assess the role of ultrasound (US) in the rapid classification of early rheumatoid arthritis (RA) by investigating whether the US features of inflammation and bone damage in early arthritis overlap with the actual clinical concept of classifying and identifying an aggressive disease.
Methods Patients with recent-onset arthritis of at least 1 peripheral joint of the hands and/or the feet were consecutively included in this study. Clinical examination, laboratory tests, the Disease Activity Score 28 (DAS28), and the 2010 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria for RA were assessed for all patients. Medication with disease-modifying antirheumatic drugs was recorded. Ultrasound assessment was performed at the following anatomical sites: wrists, metacarpophalangeal joints 2 to 5, and metatarsophalangeal joints 2 to 5 for assessing the presence/absence of synovial hypertrophy, the presence/absence of power Doppler signal, and the presence/absence of bone erosions.
The US features of inflammation and bone damage were analyzed in comparison with the DAS28, with the presence/absence of rheumatoid factor and anti–cyclic citrullinated peptide, with the fulfillment of the new 2010 ACR/EULAR classification criteria, and with the initiated disease-modifying antirheumatic drug. The prescription of methotrexate was considered a marker of an aggressive disease.
Results The US features of inflammation and bone damage correlated with the activity scores measured by the DAS28. The presence of US bone erosions overlapped with the presence of rheumatoid factor and anti–cyclic citrullinated antibodies. Synovial hypertrophy, intra-articular power Doppler signal, and bone erosions detected in at least 1 anatomical site were seen in patients fulfilling (77.7%) and in patients not fulfilling (72.7%) the 2010 ACR/EULAR classification criteria for RA. Synovial hypertrophy was found in at least 1 site in 83.3% and 58.8% of patients in whom methotrexate was prescribed and in whom methotrexate was not prescribed, respectively (P = 0.01). The US features were not correlated with the initiation of sulfasalazine or hydroxychloroquine. The patients presenting bone erosions received in significantly higher percentages the indication for methotrexate (50%) compared with sulfasalazine (20%), P = 0.03, or hydroxychloroquine (26%), P = 0.05.
Conclusions The US features of inflammation might be of help in classifying early arthritis patients despite the presence of the immune markers for RA. Together with the US features of bone damage, these might be used as an indicator of a more aggressive disease. The absence of correlation between the US findings of RA and the 2010 ACR/EULAR classification criteria indicates a possible independent contribution of US in the understanding of the future evolution of these patients.