Article Text

SAT0622 The Clinical Use of Digital Joint Space Width and X-Ray Radiogrammetry as Markers for Early Radiographic Progression in RA
  1. M. Platten1,
  2. Y. Kisten1,
  3. J. Kälvesten2,
  4. K. Forslind3,
  5. R. van Vollenhoven1
  1. 1ClinTRID- Unit for clinical therapy research, Inflammatory Diseases, Department of Medicine, Karolinska Institute, Stockholm
  2. 2Medicine and Health Sciences; Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping
  3. 3Medicine (Rheumatology); Clinical Sciences, Lund University, Helsingborg, Sweden


Background Early prediction of radiographic progression is beneficial in rheumatoid arthritis (RA) patient management. The van der Heijde modified Sharp score (SHS) is currently the gold standard that quantifies radiographic progression. The digital joint space width (JSW) measurement (Kälvesten et. al, submitted 2015) is a quick method which may offer a remedy to observer dependency and measurement error. Likewise, objective quantification of hand osteopenia may also offer added value in the assessment of bone involvement in early RA.

Objectives To determine the relationships between metacarpal bone mineral density (BMD) scores and the automated JSW measurements, and how they correlate with radiographic progression in SHS from baseline to 12 months.

Methods Bilateral hand BMD and JSW measurements of the metacarpals (2, 3 & 4) and SHS of early RA patient data, acquired from the SWEFOT database were studied. Computer assisted automated measurements of the MCP joint spaces were calculated from the hand x-rays of each patient using dedicated software analysis. Hand BMD was assessed by digital x-ray radiogrammetry (DXR; Sectra, Linköping, Sweden) of the same hand radiographs, scored with SHS at baseline and at 12 months. The Z-score BMD was a calculated measure adjusted for age and gender. Measurement differences of the ΔJSW, ΔBMD scores, and ΔSHS were established and correlated. IBM SPSS version 22.0 software was utilized for statistical analyses.

Results We studied 119 early RA patients (78% female), with an average age of 53.6 years. In 714 joints (MCP2, 3 & 4 bilaterally), the automated JSW showed an average narrowing (ΔJSW) of -0.0492mm from baseline to 12 months. The BMD for both hands displayed an average bone loss of -0,0238g/cm2 from baseline to 12 months. A highly significant correlation was evident for JSW and BMD averages (0.459, p<0.01; 0.551, p<0.01) at baseline and at 12-month follow-up respectively. Even the ΔJSW and ΔBMD over the 12 months demonstrated a highly significant correlation (0.417, p<0.01). BMD Z-Scores showed similar patterns with JSW (0.246, p<0.05, n=109; 0.360, p<0.01, n=106) at baseline and 12 months respectively. A positive inverse relationship emerged between automated JSW and the joint space narrowing (JSN) component of SHS (-0.319, p<0.01; -0.254, p<0.01) at baseline and 12 months respectively. The average JSW measurements of both hands (n=117) also revealed significant correlations (-0.224, p<0.05; -0.271, p<0.01) with total SHS at baseline and 12 months respectively. No significant correlation was found between JSW and SHS erosion score. The 12-month BMD displayed near significant correlations with the 12-month SHS erosion score (-0.174, p=0.060) and the 12-month total SHS (-0.157, p=0.090).

Conclusions Automated analyses of JSW and BMD were technically feasible with reproducibility and agreement with each other. Moreover, JSW displayed highly significant correlations with joint space narrowing and total SHS scores. Bone mineral density, on the other hand, did not correlate with joint space narrowing but rather approached significance with erosion and total SHS scores. Therefore, these objective digitally quantified measures of joint space narrowing and hand bone loss may be useful as complementary markers for early radiographic progression.

Disclosure of Interest M. Platten: None declared, Y. Kisten: None declared, J. Kälvesten Employee of: Sectra Imtec AB, K. Forslind: None declared, R. van Vollenhoven: 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.