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.

The Impact of Positive Pathergy Test on the Performance of Diagnostic Criteria in Behcet's Disease.

Davatchi,  Fereydoun, Abdollahi,  Bahar Sadeghi, Chams-Davatchi,  Cheyda, Shahram,  Farhad, Ghodsi,  Zahra, Faezi,  Tahereh, Nadji,  Abdolhadi


No diagnostic test exists for Behcet's Disease (BD) except the pathergy test that reveals the pathergy phenomenon of the disease. Positive pathergy test (PPT) is an important criterion of many classification/diagnosis criteria. There are actually 16 sets of diagnosis/classification criteria for BD: Curth (1946), Hewitt (1969), Mason and Barnes (1969), Hewitt Revised (1971), Japan (1972), Hubault and Hamza (1974), O'Duffy (1974), Zhang (1980), Dilsen (1986), Japan Revised (1988), International Study Group (ISG) in 1990, Iran (1993), Classification Tree (1993), Dilsen revised (2000), Korea (2003), and the International Criteria for Behcet's Disease (ICBD) in 2006. The aim of this study was to find, how much PPT accounts in the overall performance of those diagnostic criteria that use PPT as a criterion. We selected 9 of the most renowned of them for this study.


Patients with BD (6607) were selected as consecutive patients from the Behcet's Disease registry, and 4292 control patients. The diagnosis of BD was made on "Expert Opinion" without taking into account any known criteria. Control patients were those sent to find if they had BD, and the expert opinion was against the diagnosis. The 9 criteria sets were tested in them, once taking into account the PPT, and then without it. The sensitivity, specificity, and accuracy (percent agreement) of them were calculated to find the impact of PPT.


Without PPT, all tested criteria lost some of their sensitivity. The largest lost was for Dilsen, ISG, and revised Dilsen criteria (17.3%, 16.1%, 16.1%). The least was for ICBD, Japan revised, and Classification Tree criteria (6.5%, 7.6%, 7.6%). The largest gain on specificity was for ICBD (3.9%) and the least for Japan and Japan revised criteria (0.1%, 0.1%). The largest lost in accuracy was for ISG, Dilsen revised, and Dilsen criteria (9.5%, 9.5%, 8.6%) and the least for ICBD, Classification Tree, Japan, and Japan revised criteria (3.6%, 4.2%, 4.6%, 4.6%). The best overall accuracy (without PPT) was ICBD, the Classification Tree, and the Japan revised criteria (93.7%, 93%, 86.1%). Details for all tested criteria are shown in Table 1.

Table 1. Performance of different Criteria set with and without pathergy test

 No61.91.2 99.80.1 76.80.8 
 No91.70.7 99.80.1 93.70.5 
 No77.31.0 97.80.4 85.40.7 
 No78.61.0 97.70.4 86.10.6 
 No64.71.2 99.70.2 78.40.8 
 No63.31.2 99.70.2 77.60.8 
 No77.21.0 97.80.4 85.30.7 
 No89.60.7 98.10.4 93.00.5 
 No76.41.0 98.80.3 85.30.7 


ICBD and Classification Tree criteria were the least dependent on PPT to classify patients as having BD. They have also the best sensitivity and the best accuracy without using the PPT, making them the best diagnostic instrument in countries where PPT is rarely encountered (countries far from the Silk Road). They are also the most performing Diagnosis/Classification criteria for countries inside the Silk Road.

To cite this abstract, please use the following information:
Davatchi, Fereydoun, Abdollahi, Bahar Sadeghi, Chams-Davatchi, Cheyda, Shahram, Farhad, Ghodsi, Zahra, Faezi, Tahereh, et al; The Impact of Positive Pathergy Test on the Performance of Diagnostic Criteria in Behcet's Disease. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :2386

Abstract Supplement

Meeting Menu