Table 2

Associations between baseline MRI inflammation and radiographic SHS progression over the first year

Early arthritis patients (n=113)Subgroup of 2010 RA patients (n=53)
Univariableβ (95% CI)R2p Valueβ (95% CI)R2p Value
BME0.06 (0.02 to 0.10)0.13<0.0010.06 (0.01 to 0.11)0.140.006
Synovitis0.14 (0.05 to 0.22)0.13<0.0010.12 (−0.02 to 0.23)0.110.02
Tenosynovitis0.11 (0.03 to 0.20)0.080.0020.07 (−0.05 to 0.21)0.040.19
Total inflammation0.04 (0.02 to 0.06)0.16<0.0010.04 (0.01 to 0.07)0.150.004
Multivariableβ (95% CI)R2ap Valueβ (95% CI)R2ap Value
BME0.05 (0.01 to 0.09)0.160.0030.04 (−0.01 to 0.10)0.270.04
Synovitis0.13 (0.03 to 0.21)0.180.0010.11 (−0.03 to 0.21)0.260.04
Tenosynovitis0.10 (0.01 to 0.17)0.140.0110.02 (−0.10 to 0.13)0.190.76
Total inflammation0.04 (0.01 to 0.06)0.18<0.0010.03 (−0.01 to 0.06)0.260.04
  • Linear regression analysis was used. Multivariable: adjusted for age, gender, RF +/−, CCP +/− CRP level, 66 SJC. R2 represents the explained variability in radiographic progression, by the model. For the multivariable model the R2 adjusted (R2a) is presented, which adjusts for the amount of variables included in the model. Total inflammation=BME + synovitis + tenosynovitis.

  • BME, bone marrow oedema; RA, rheumatoid arthritis; RF, rheumatoid factor; CRP, C-reactive protein; SJC, swollen joints count; CCP, cyclic citrullinated peptide; SHS, Sharp–van der Heijde scoring.