Background Assessment of active synovitis in rheumatoid arthritis (RA) is complex but essential for diagnosis and ongoing management. Clinical assessment is supported by serum laboratory analysis of inflammatory markers, high sensitivity C-reactive protein (hsCRP) and erythrocyte sedimentation rate (ESR). Musculoskeletal power Doppler ultrasonography (PDUS) has recently demonstrated greater sensitivity over clinical assessment and serum inflammatory markers in detecting active synovitis.
Objectives To investigate the association between findings on PDUS, clinical assessment, serum hsCRP and ESR in the assessment of active synovitis in RA.
Methods This cross-sectional study evaluated 218 consecutive RA patients (82% female) referred to ultrasound (US) clinic. Two assessors performed US examination using a 22 hand joint protocol, assessing wrists, metacarpophalangeal and proximal interphalangeal joints bilaterally (OMERACT scoring system). HsCRP and ESR were measured within 1 month of US scan. Patients with ≥3 swollen (SJ) and ≥3 tender joints (TJ) in hands were considered to have significant clinical synovitis (active or chronic).
Results Of 218 patients with RA, 100 (46%) had positive PDUS signal in at least one joint. Raised hsCRP was found in 25% and raised ESR in 45%. Seventy-two of 202 patients (36%) had ≥3 TJ and ≥3SJ (clinically active synovitis). Whilst tender joint count was not correlated with positive PDUS (OR 1.01, p=0.334), swollen joint count (SJC) demonstrated strong association, with >30% increase in odds for positive PDUS for increase in SJC by one (OR 1.3, p<0.001). Of 96 patients with positive PDUS, 46 (48%) had ≥3TJ and ≥3SJ. Of 105 patients with negative PDUS, 26 (25%) had ≥3TJ and ≥3SJ (p=0.001).
Of 196 patients with ESR and PDUS data available, 44 (47%) of PDUS positive and 45 (44%) of PDUS negative patients had raised ESR, showing no significant association (p=0.667). There was no difference in proportion of patients with raised CRP (25%) in PDUS positive vs. negative groups (p=0.748).
ESR and SJC were only very weakly correlated (Spearman's rho =0.16, p=0.03).
Of 138 patients with either clinical or serological evidence of disease activity, 73 (53%) had positive PDUS. Among asymptomatic patients with normal hsCRP and ESR, 21 (38%) were PDUS positive (p=0.058).
Conclusions Both hsCRP and ESR correlated poorly with clinical examination and PDUS detection of active synovitis. This should raise awareness among clinicians that hsCRP, ESR and clinical examination are misleading in appreciating the risk of ongoing inflammation in some patients with RA, as 38% of active RA patients in our study fell into this category. Further research into establishing the appropriate hsCRP cut-off value to distinguish between a subclinical inflammatory state and disease remission might yield better understanding of the role of hsCRP in monitoring disease activity in RA.
Disclosure of Interest None declared