Article Text

AB0759 A direct comparison of web-based and real-life ultrasound assessments of rheumatoid arthritis hands involving the same ultrasonographers
  1. V. Vlad1,
  2. M. Micu2,
  3. L. Varzaru3,
  4. M. Bojinca4,
  5. M. Milicescu4,
  6. F. Berghea5,
  7. D. Predeteanu5,
  8. R. Ionescu5
  1. 1Sf. Maria Hospital, Bucharest
  2. 2Rheabilitation Clinical Hospital, Cluj-Napoca
  3. 3Center of rheumatic diseases Ion Stoia
  4. 4Dr. Ion Cantacuzino Hospital
  5. 5Carol Davila University Of Medicine And Pharmacy, Bucharest, Romania


Background Ultrasound(US) is intensely tested for becoming a real outcome measure in Rheumatoid Arthritis (RA) and to be used in clinical trials. The method has proven good reliability in recent over-the-web exercises (LUMINA Study –results not published yet) offering simmilar results with published real-life exercises (1). However no direct comparison between over-the-net and real life exercises with the same rheumatologists involved was done so far.

Objectives To test inter and intraobserver agreement of US grading on RA patients in two settings (over-the-web and real-life) with the same assessors involved in both settings.

Methods 4 patients suffering from active RA (clinically swollen hands) were selected from our database, after having given their consent to participate in the study. For each patient we selected 100 joint sites to be examined - carpal, MCPs 2-5, PIPs 2-5 (both dorsal and volar, bilaterally), extensor compartments 2,4, and 6, and flexor tendons 2-5 (total 400 scans for all patients). In the real-life exercise, 5 rheumatologists with different experience (but more than 5 years of US practice each) from different rheumatology clinics took turn in scanning patients on 2 US machines (Essaote MyLab 25 and Essaote MyLab 70) for 6-7 hours, during the same day. The joints were graded with semiquantitative scale (0,1,2,3) and tendons with binar scale (0/1). In the over-the-web exercise, from each investigator’s recorded movies, 20 movies were selected and resent to them by internet (on youtube channel) for the second grading. Values of agreement of 0–0.20 were considered poor, >0.20–0.40 fair, >0.40–0.60 moderate, >0.60–0.80 good and >0.80–1 excellent Statistical analyses were performed using SPSS 16.0.

Results For web-based exercise, interobserver agreement coeficient was 0.774 and 0.786 (in two different exercises 7 days apart). For real-life-exercise the results have been quite simmilar with an interobserver agreement coeficient of 0.680. In both cases the agreement was better for joints than for tendons (simmilar with very recent published studies -1). Intraobserver agreement was between 0.707 and 0.856 (web-based). We analysed various sub-groups:joints, tendons, joint different recesses.

Conclusions This study demonstrates simmilar results of over-the-web and real-life US grading of hand joints in RA patients opening the possibility of using internet in large colaborative trials involving US.

  1. Esperanza Naredo et al on behalf of the OMERACT Ultrasound Task Force- Reliability of a consensus-based ultrasound score for tenosynovitis in rheumatoid arthritis, AnnRheumDis 2012;0:1-7

  2. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull.1979;86:420–428.

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.