Abstract| Volume 23, SUPPLEMENT 2, A223-A224, April 2015

Associations between knee pain and synovitis on conventional and dynamic contrast-enhanced magnetic resonance imaging in obese persons with knee osteoarthritis: A cross-sectional study

      Purpose: To investigate the association between knee pain and synovitis in obese persons with knee osteoarthritis (OA).
      Methods: In a cross-sectional setting, knee synovitis was assessed using 3-tesla magnetic resonance images (MRI) and correlated to self-reported outcomes using the Knee injury and Osteoarthritis Outcome Score (KOOS).
      Synovitis was assessed: i) in the suprapatellar, lateral and medial recesses using dynamic contrast-enhanced (DCE) MRI and collapsed into one volume of interest (VOI), and ii) semi-quantitavely in 11 eleven locations using contrast-enhanced (CE) MRI generating a whole-knee synovitis score. Effusion was assessed on CE-MRI according to the Boston-Leeds OA Knee Score (BLOKS) and the combination of synovitis and effusion was scored on non CE-MRI using the MRI in OA Knee Score (MOAKS). In a voxel-by-voxel approach, the following DCE-MRI variables were automatically extracted from the VOI: i) the Initial Rate of Enhancement (IRE), i.e. the mean speed of enhancement, ii) Maximal Enhancement (ME), i.e. the mean of the highest signal intensity values and iii) Nvoxel, the sum of voxels with plateau and washout patterns, i.e. the most perfused voxels.
      The DCE-MRI variable IRExNvoxel was chosen as the primary variable in the analyses.
      Results: Valid MRI and clinical data were available in 94 persons. The typical participant was a 65-year old woman with a mean body mass index of 32.3 kg/m2, a Kellgren-Lawrence score of 2.5 and a C-reactive protein level of 2.0 mg/l.
      The DCE-MRI variable IRExNvoxel showed a statically significant bivariate correlation with KOOS pain (r= -0.34; p=0.001) and all the remaining KOOS items (-0.25 > r > -0.46; p<0.007) as was the case with the whole-knee synovitis score (Table 1).
      The non CE-MRI variable MOAKS effusion-synovitis showed statistically significant correlation with 3 of the 5 KOOS items (-0.29 > r > -0.30; p<0.005).
      Intraclass correlation coefficients ranged between 0.95-1.00 for the DCE-MRI variables and 0.80-1.00 for the static MRI variables.
      Conclusions: The results confirm an association between pain and synovitis assessed on both DCE- and CE-MRI in obese patients with knee OA. The whole-knee synovitis score on CE-MRI and most of the DCE-MRI variables were also significantly associated with the remaining KOOS items.
      DCE-MRI analyses were robust and highly reproducible and have the potential to be used to further investigate the role of inflammation and perfusion in knee OA, in a similar way it has been used in inflammatory joint diseases.
      Table 1Correlation matrix of KOOS and MRI variables.
      Nvoxel_Rel.499** (p<0.0001)1.000
      MExNvoxel.985** (p<0.0001).508** (p<0.0001)1.000
      IRExNvoxel.948** (p<0.0001).556** (p<0.0001).978** (p<0.0001)1.000
      IRExME.815** (p<0.0001).554** (p<0.0001).884** (p<0.0001).950** (p<0.0001)1.000
      CE_Synovit.794** (p<0.0001).408** (p<0.0001).823** (p<0.0001).808** (p<0.0001).769** (p<0.0001)1.000
      BLOKS_Effusion.676** (p<0.0001).182 (p=0.080).682** (p<0.0001).659** (p<0.0001).566** (p<0.0001).474** (p<0.0001)1.000
      MOAKS_Effusion.803** (p<0.0001).242* (p=0.019).802** (p<0.0001).772** (p<0.0001).673** (p<0.0001).592** (p<0.0001).824** (p<0.0001)1.000
      KOOS_Pain-.270** (p=0.008)-.167 (p=0.110)-.323** (p=0.002)-.337** (p=0.0009)-.370** (p=0.0002)-.355** (p=0.0004)-.212* (p=0.04)-.294** (p=0.004)
      KOOS_Symp-.381** (p=0.0002)-.177 (p=0.088)-.428** (p<0.0001)-.463** (p<0.0001)-.471** (p<0.0001)-.386** (p=0.0001)-.330** (p=0.001)-.290** (p=0.005)
      KOOS_ADL-.200 (p=0.053)-.200 (p=0.053)-.243* (p=0.018)-.252* (p=0.014)-.278** (p=0.007)-.278** (p=0.007)-.117 (p=0.262)-.173 (p=0.095)
      KOOS_QOL-.313** (p=0.002)-.271** (p=0.008)-.362** (p=0.0003)-.392** (p<0.0001)-.423** (p<0.0001)-.356** (p=0.0004)-.189 (p=0.068)-.298** (p=0.003)
      KOOS_SportRec-.228* (p=0.027)-.236* (p=0.022)-.278** (p=0.007)-.289** (p=0.005)-.320** (p=0.002)-.307** (p=0.003)-.127 (p=0.068)-.127 (p=0.222)
      Spearman's rho with p-values in parentheses.
      **Correlation is significant at the 0.01 level (2-tailed).
      *Correlation is significant at the 0.05 level (2-tailed).
      Nvoxel: Sum of voxels with plateau and washout enhancement patterns. Nvoxel_Rel: Nvoxel over the total number of enhancing voxels. IRE: Initial rate of enhancement; ME: Maximal enhancement; CE_Synovitis: Whole-knee synovitis score (on CE-MRI); BLOKS_Effusion: Boston-Leeds OA Knee Score effusion score (CE-MRI); MOAKS_Effusion: MRI in OA Knee Score effusion-synovitis score (non CE-MRI); KOOS: Knee injury and OA Outcome Score; Symp: Symptoms; ADL: Activity in Daily Living; QOL: Quality of Life; SportRec: Sport/Recreation.