874 The Journal of Rheumatology 2007; 34:4
Computer Based Methods for Measurement of Joint
Space Width: Update of an Ongoing OMERACT Project
JOHN T. SHARP, JANE ANGWIN, MAARTEN BOERS, JEFF DURYEA, GABRIELE von INGERSLEBEN,
JAMES R. HALL, JOOST A. KAUFFMAN, ROBERT LANDEWÉ, GEORG LANGS, CÉDRIC LUKAS,
JEAN-FRANCIS MAILLEFERT, HEIN J. BERNELOT MOENS, PHILIPP PELOSCHEK, VIBEKE STRAND,
and DÉSIRÉE van der HEIJDE
ABSTRACT. Computer-based methods of measuring joint space width (JSW) could potentially have advantages over
scoring joint space narrowing, with regard to increased standardization, sensitivity, and reproducibility.
In an early exercise, 4 different methods showed good agreement on measured change in JSW over time
in the small joints of the hands and feet. Despite differences in measurement values between methods,
measurement of within-joint change over time showed no systematic differences. The within-method
variation was small, with intra-operator variation being smaller than inter-operator variation. Although
this initial study was limited in terms of the number of patients and timepoints (total 10), the number of
joints was relatively high (340 joints), so the results were considered strong evidence supporting the
validity of computer-based JSW measurements to continue the study of the potential value of JSW by
comparison of measurements to manual scoring of joint space narrowing using the COBRA trial
images. (J Rheumatol 2007;34:874–83)
Key Indexing Terms:
JOINT SPACE MEASUREMENT COMPUTER-BASED
RELIABILITY PRECISION SENSITIVITY TO CHANGE
From the University of Washington School of Medicine, Seattle,
Washington, USA; GlaxoSmithKline, Middlesex, England; VU University
Medical Center, Amsterdam, The Netherlands; Brigham and Women’s
Hospital, Harvard University, Boston, Massachusetts, USA; Synarc, San
Francisco, California, USA; Snoqualmie, Washington, USA; University of
Twente, Enschede, The Netherlands; University Hospital Maastricht,
Maastricht, The Netherlands; Graz University of Technology, Graz,
Austria; Dijon University Hospital, University of Burgundy, Dijon,
France; University Medical Center, Utrecht, The Netherlands; University
of Vienna, Vienna, Austria; and Stanford University, Palo Alto, California,
USA.
J.T. Sharp, MD, Affiliate Professor of Medicine, University of Washington;
J. Angwin, MA, Respiratory and Inflammation Centre of Excellence in
Drug Discovery, GlaxoSmithKline R&D, UK; M. Boers, MSc, MD, PhD,
Professor of Clinical Epidemiology, VU University Medical Center;
J. Duryea, PhD, Assistant Professor, Brigham and Women’s Hospital,
Harvard University; J.A. Kauffman, Signals and Systems Group, Faculty
of Electrical Engineering, Mathematics and Computer Science, University
of Twente; R. Landewé, MD, PhD, Associate Professor of Rheumatology,
University Hospital Maastricht; C. Lukas, MD, Rheumatologist,
University Hospital Maastricht; J.F. Maillefert, MD, PhD, Associate
Professor of Rheumatology, University of Burgundy; H.J. Bernelot Moens,
MD, Rheumatologist, Ziekenhuis Groep Twente, Hengelo, The
Netherlands; G. Langs, Institute for Computer Graphics and Vision, Graz
University of Technology; P. Peloschek, MD, Radiologist, Medical
University of Vienna; D.M.F.M. van der Heijde, MD, PhD, Professor of
Rheumatology, University Hospital Maastricht; G. von Ingersleben,
Radiologist, Synarc; J.R. Hall, Software Engineer, Snoqualmie, WA, USA.
Address reprint requests to Dr. J.T. Sharp, 8387 N.E. Sumanee Place,
Bainbridge Island, WA 98110, USA. E-mail: johntsharp@comcast.net.
Introduction
A subcommittee within the OMERACT imaging committee
was formed after OMERACT 6 to test reliability, sensitivity,
and feasibility of computer-based methods for measuring radi-
ographic damage in the small joints of the hands and feet in
patients with rheumatoid arthritis (RA). Initial efforts concen-
trated on measurement of joint space width (JSW). A report
described the rationale and objectives of the committee in
detail
1
. Briefly, measurement of damage on a continuous met-
ric scale would be preferable to scoring damage on an ordinal
scale if the measurements were highly reproducible, especial-
ly if they were more sensitive to change. Further, the difficul-
ty of standardizing scoring by different readers even within a
single center is an obstacle to comparing results across stud-
ies, and even limits of agreement are specific to a given study
and cannot be generalized to others
2
. Metric measurements
could be a means of reducing this obstacle.
To test reliability, sensitivity, and discriminating ability,
investigators participated in a series of exercises to measure
JSW in the finger, wrist, and toe joints, the joints that are most
regularly involved in RA. The results of measurements on
radiographs of 2 patients with RA at 2 timepoints using 3 dif-
ferent computer programs were presented at OMERACT 7
and encouraged more studies. Developers of 4 different meth-
ods and 10 different readers participated in a second exercise,
called exercise A, involving radiographs from 4 patients at a
total of 10 different timepoints, 2 or more for each patient
3-10
.
The study determined variation of repeated JSW measure-
ments by the same operator, between operators, and between
methods, and the extent to which any differences in JSW
methods affected the measurement of JSW change over time.
The results of the analysis of reliability of repeated measure-
ment and agreement between methods, which are reported
here, were considered sufficiently encouraging for the com-
mittee to agree that a more comprehensive study employing a
Personal non-commercial use only. The Journal of Rheumatology Copyright © 2007. All rights reserved.