Article Text

AB0283 Predictors of rheumatoid arthritis (RA): Quantitative and semi-quantitative sonographic measurements of peripheral joints
  1. F.S. Machado,
  2. R.N.V. Furtado,
  3. R.D. Takahashi,
  4. A.L.P. Buosi,
  5. J. Natour
  1. Rheumatology, Universidade Federal De Sao Paulo, Sao Paulo, Brazil


Background Sonographic quantitative and semi-quantitative measurements in peripheral joints of normal subjects are yet to be defined, as are sonographic predictors of RA.

Objectives To estimate quantitative and semi-quantitative ultrasound measurements predictors of RA in small, medium and large joints.

Methods A cross-sectional study was carried out involving 78 healthy volunteers (HV) and 60 patients with RA (ACR), matched for age group and gender. A “blind” radiologist used a My Lab 60 XVision machine (Esaote, Biomedica - Genova, Italy) with a linear trnasducer (6-18MHz) to evaluate 6.348 joint recesses. Quantitative measurements of synovial recess (QSR) (mm) and semi-quantitative measurements of synovial hyperplasia (SSH), Power Doppler (SPD) and bone erosion (SBE) (scores 0-3) were performed. To determine the chance to detect RA, ROC curve analysis for QSR measurements were performed (specificity 98.7%) and, for the semi-quantitative measures, an univariate logistic regression (expressed in odds ratio - OR) was carried out. P value <0.05 was set as significant.

Results The mean age (± SD) was 46.48 (9.14) and 43.89 (9.09), respectively for the HV and RA groups. The sample was homogeneous for gender, age group and skin color. RA group: mean disease duration was 7.89 years (± 6.76) and DAS-28 4.20 (± 1.71). Statistical difference was observed between groups for QSR (p<0.013) for most of the joint recesses. Mean (±SD) (mm) of QSR, respectively for HV and RA groups (HV/RA), were: radiocarpal: 2.07 (0.56)/3.24 (1.24); distal radioulnar: 1.45 (0.37)/2.28 (1.11); ulnocarpal: 1.37 (0.59)/2.74 (1.76); dorsal 2nd metacarpophalangeal (MCP): 1.06 (0.53)/1.51 (0.96); palmar 2nd MCP: 0.88 (0.60)/1.40 (1.01); dorsal 3rd MCP: 0.81 (0.62)/1.24 (0.99); dorsal 2nd proximal interphalangeal (PIP): 0.46 (0.25)/0.76 (0.64); dorsal 3rd PIP: 0.44 (0.32)/0.83 (0.56); palmar 3rd PIP: 0.83 (0.27)/1.11 (0.55); coronoid fossa: 0.97 (1.06)/2.18 (2.27); olecranean fossa: 1.51 (1.17)/2.79 (2.65); posterior GH recess: 2.43 (0.45)/3.03 (1.29); knee: 2.21 (1.65)/3.95 (2.96); talocrural: 2.38 (1.13)/3.34 (1.99); talonavicular: 2.67 (1.10)/3.56 (1.50); subtalar: 2.15 (1.13)/3.07 (1.71); dorsal 5th MTP: 0.72 (0.70)/1.47 (1.11). Cutoff values of QSR specific of RA (AUC>0,700) were: radiocarpal 3.78mm; ulnocarpal 3.07mm; distal radioulnar 2.21mm; dorsal 3rd PIP 1.19mm; knee 6.7mm and dorsal 5th MTP 2.33mm. For semi-quantitative measurements, progression from score 0 to 3, at the joint recesses with greater chance to detect RA were: SSH: ulnocarpal (OR=100, p=0.000); radiocarpal (OR=70, p=0.000); distal radioulnar (OR=43, p=0.000) and knee (OR=28, p<0.000); SPD: radiocarpal (OR=66, p=0.000); SBE: radiocarpal (OR=324, p=0.000); lateral 5th MTP (OR=100, p=0.000); 2nd MCP (dorsal and radial) (OR=92, p=0.000) and ulnocarpal (OR=48, p=0.000). Inter-observer reliability (r) for quantitative and semi-quantitative measures ranged from 0.563 to 0.872 and 0.341 to 0.823, respectively.

Conclusions Values of QSR were higher in RA patients compared with HV for most studied joint recesses. Also, quantitative measures specific of RA were found in almost all recesses. Semi-quantitative measurements analysis showed that the worst scores found at radiocarpal, ulnocarpal and lateral 5th MTP recesses increases the chance to detect RA.

  1. K Ellegaard, 2007.

  2. M Szkudlarek, 2003.

  3. DG Disler, 2000.

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.