Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Validation of the ASAS Definition of a Positive MRI in a Cohort of Patients with Early SpA Followed for Eight Years.

Marzo-Ortega2,  Helena, Bennett2,  Alexander N., McGonagle2,  Dennis, Emery2,  Paul, Maksymowych1,  Walter P.

University of Alberta
University of Leeds, United Kingdom


The Assessment of SpondyloArthritis International Society (ASAS) recently proposed a definition of a positive MRI for sacroiliitis based on the finding of bone marrow edema (BME) on STIR in subchondral or periarticular bone marrow of the sacroiliac joint (SIJ). MRI is considered positive for SpA if there are two BME lesions on the same coronal slice through the SIJ or a single BME lesion present on two consecutive slices. This proposal requires validation, preferably in prospective cohorts to evaluate development of radiographic changes of sacroiliitis.


To compare the diagnostic utility of global assessment of MR scans versus the ASAS definition using a standardized approach to evaluation of the SIJ in an early SpA cohort followed over 8 years.


Four rheumatologists blinded to patient and diagnosis independently assessed MRI scans (T1W and STIR) from 37 patients with SpA according to ESSG criteria (median symptom duration 24 weeks); 13 developed radiographic sacroiliitis after 8 years, 11 controls mechanical low back pain (mLBP), 11 healthy controls. Semi-coronal slices through the SIJ were read systematically according to a standardized online training module. BME, fat infiltration, erosions, and ankylosis was recorded using an online data entry system. Schematics of the SIJ divided into quadrants allowed recording of these lesions (yes/no) in each quadrant of each coronal slice. The ASAS definition was defined as met when BME was recorded in 2 SIJ quadrants on the same slice or when a single lesion in one SIJ quadrant was present in 2 consecutive slices. Readers also answered (yes/no) whether this SIJ scan confirmed the presence of SpA. Sensitivity, specificity, and LRs for individual and concordant data (at least 2 readers) were calculated according to clinical diagnosis at baseline and according to the development of radiographic sacroiliitis.


MRI had high diagnostic utility for SpA according to global assessment by individual readers (mean for 4 readers (range) sensitivity/specificity: 66.9% (61.8–70.6)/94.4% (88.9–100), LR+ 11.9, LR- 0.2) as well as concordant data (sensitivity/specificity: 67.6%/94.4%, LR+ 12.1, LR- 0.3). By comparison, sensitivity for SpA of the ASAS definition was higher but there was reduction in specificity with 11.1% of control scans (3 mLBP) meeting the ASAS definition so that diagnostic utility was lower than for global assessment. Diagnostic utility of baseline MRI for radiographic sacroiliitis after 8 years follow up was even higher: global assessment (sensitivity/specificity: 100%/94.4%, LR+17, LR- not calculable (nc), ASAS definition (sensitivity/specificity: 100%/88.9%, LR+ 9, LR- nc).


The diagnostic utility of the ASAS definition resembles global evaluation by expert readers, the principle limitation being the finding of BME in patients with mLBP.

Global Assessment ASAS Definition


To cite this abstract, please use the following information:
Marzo-Ortega, Helena, Bennett, Alexander N., McGonagle, Dennis, Emery, Paul, Maksymowych, Walter P.; Validation of the ASAS Definition of a Positive MRI in a Cohort of Patients with Early SpA Followed for Eight Years. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1963
DOI: 10.1002/art.29728

Abstract Supplement

Meeting Menu