Abstract
OBJECTIVE: Clinical assessment of rheumatoid arthritis (RA) based on pain and swelling and physical examination is limited by observer error and interpretation. We compared magnetic resonance imaging (MRI) and clinical examination to detect synovitis in RA. METHODS: Twelve patients with active RA were assessed according to Ritchie index, swollen joint count and score, swollen joint count of hands and wrists [2 wrists, 10 metacarpophalangeal (MCP), 10 proximal interphalangeal (PIP)], morning stiffness, pain intensity, Disease Activity Score (DAS), erythrocyte sedimentation rate, and C-reactive protein. MR images of hands and wrists were obtained with an adapted device, on T1 weighted (T1W) spin echo (SE) coronal images before and after gadolinium DTPA, TIW SE axial images with gadolinium DTPA, T2* gradient echo recall coronal and axial sequences, and assessed by 2 radiologists (O = no synovitis, 1 = synovitis). RESULTS: The swollen joint count on hands and wrists was 59 on clinical examination (mean 5.08 +/- 3.15 per patient; 20/24 wrists, 7/120 MCP, 32/120 PIP) and 162 on MRI (mean 13.50+/- 5.65; 22/24 wrists, 70/120 MCP, 70/120 PIP). Statistically significant correlations were found between MRI synovitis count and swollen joint count (p = 0.015) and score (p = 0.019), Ritchie Index (p = 0.035), DAS (p = 0.02) and morning stiffness (p = 0.07). MRI revealed synovitis significantly more often than clinical examination (162 vs 59; p = 0.00002) [2-fold in PIP (70/32) and 10-fold in MCP (70/7)]. Clinical examination and MRI were concordant for 157/264 joints (59.5%). The association of normal MRI with synovitis on clinical examination was observed in 2 cases, the opposite in 105. CONCLUSION: MRI is more sensitive than clinical examination to detect synovitis of hands and wrists in RA, especially for MCP and PIP joints, and is valuable for assessment of inflammation in hands and wrists in RA.