Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.

Comparing 2 Referral Strategies to Diagnose Axial Spondyloarthritis: The Recognizing and Diagnosing Ankylosing Spondylitis Reliably Study.

Sieper1,  J., Srinivasan2,  S., Zamani3,  O., Mielants4,  H., Choquette5,  D., Pavelka6,  Karel, Loft7,  Anne Gitte

Charitè Berlin, Campus Benjamin Franklin, Berlin, Germany
Barzilai Medical Centre, Ashkelon, Israel
Stabilimento Ospedaliero Misericordia, Prato, Italy
G.T. Popa Center for Biomedical Research, Iasi, Romania
Scientific Institute of Rheumatology, Moscow, Russia
National Institute of Rheumatic Diseases, Pie[scaron][squ] any, Slovakia
Northwick Park Hospital, Harrow, United Kingdom
Univ of Cambridge/Clin Med, Cambridge, United Kingdom
Merck, Sharp & Dohme Corporation, Brussels, Belgium
Merck, Sharp & Dohme Corporation, Rahway, NJ
Rheumazentrum Favoriten, Wien, Austria
University Hospital, Gent, Belgium
Institute of Rheumatology of Montreal, Montreal, QC
1Institute of Rheumatology, Department of Experimental Rheumatology, 1st Faculty of Medicine, Charles University in Prague, Prague, Czech Republic
Vejle Hospital, Vejle, Denmark
Semmelweis University, Budapest, Hungary
Christian Medical College, Vellore, India


Improved referral from primary care (PC) allows shortening the time to diagnosis of axial spondyloarthritis (SpA), which may lead to better treatment and improved prognosis. The objective of the RecognizIng and Diagnosing Ankylosing Spondylitis Reliably(RADAR) study was to determine which of 2 referral strategies, when used by physicians for chronic back pain (CBP) patients, is superior in diagnosing definite axial SpA by rheumatologists.


Primary-care referral sites in 16 countries were randomized (1:1) to refer CBP patients to a rheumatologist according to 1 of 2 referral strategies: Strategy I, 1 of 3 criteria—inflammatory BP (IBP), HLA-B27+, sacroiliitis on imaging (SI); or Strategy II, 2 of 6 criteria—IBP, HLA-B27+, SI, family history, good response to NSAIDs, extra-articular manifestations (EAMs). The rheumatologist then established a diagnosis. The primary analysis compared the proportion of patients diagnosed with axial SpA depending on the referral strategy used. Reported are: usage of referral criteria in PC, concordance with rheumatologist, sensitivity, specificity, predictive values (PV), and likelihood ratio of the criteria to make a diagnosis.


Subjects had CBP of unknown origin >3 months, onset before age 45, and no diagnosis of axial SpA or ankylosing spondylitis established yet. Strategy I was used at 135 primary-care sites to refer 504 subjects of which 35.6 were diagnosed with axial SpA. Strategy II was used at 143 primary-care sites to refer 568 subjects of which 39.8% were diagnosed with axial SpA. (For both strategies: delta: 4.40%, 95%CI: –7.09% to 15.89%, ns.) IBP was the most used referral criterion, and rheumatologist concordance was high (Table). PV for SI, HLA-B27, or EAMs judged positive by PC physicians were 72.5, 76.1, and 55.6%, respectively, and +/– 40% for other criteria. Negative PV and sensitivity of IBP judged by rheumatologists was >85%, but positive PV and specificity was <50%. NPV and/or sensitivity of any other single criterion was <80%. Analysis of alternative strategies showed that referral on 2 of 2 criteria performed poorly, whereas many 2 of 3 strategies, most of which included SI, performed well. But only the strategy with SI, HLA-B27, and IBP had sensitivity, specificity, NPV, and PPV >80%.

Referral Criteria and Rheumatologist Results

CriterionUsage by Referring PhysicianConcordance with Rheumatologist
Strategy 1 (%)Strategy 2 (%)Combined (%)Strategy 1 (%)Strategy 2 (%)Combined (%) 
SI137 (27.2)204 (35.9)341 (31.8%)90 (68.2)120 (61.5)210 (64.2)
HLA-B2787 (17.3)97 (17.1)184 (17.2)82 (97.6)87 (96.7)169 (97.1)
IBP469 (93.1)546 (96.1)1015 (94.7)376 (84.5)425 (86.2)801 (85.4)
EAMs 112 (19.7%)  98 (89.1) 
Family history 66 (6.2)  42 (76.4) 
NSAID response 428 (75.4)  240 (60.2) 


This is the first international randomized study to show that a referral strategy for CBP based on 3 criteria performs as well as one with 6 and leads to diagnosis of axial SpA in >35% of patients. IBP was nearly always used, showed good concordance with rheumatologists, and had high sensitivity and NPV. Combining IBP with other criteria such as HLA-B27 and SI increases the likelihood of diagnosis.

To cite this abstract, please use the following information:
Sieper, J., Srinivasan, S., Zamani, O., Mielants, H., Choquette, D., Pavelka, Karel, et al; Comparing 2 Referral Strategies to Diagnose Axial Spondyloarthritis: The Recognizing and Diagnosing Ankylosing Spondylitis Reliably Study. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :517

Abstract Supplement

Meeting Menu