s34_ab3

Validation of the classification criteria commonly used in Korea and a modified set of preliminary criteria for Behçet's disease: A multi-center study

H.K. Chang¹, S.S. Lee², H.J. Bai¹, Y.W. Lee³, B.Y. Yoon³, C.H. Lee³, Y.H. Lee⁴, G.G. Song⁴, W.T. Chung⁵, S.W. Lee⁵, J.Y. Choe⁶, C.G. Kim⁶, D.K. Chang⁷

¹Department of Internal Medicine, Dankook University, Cheonan; ²Department of Internal Medicine, Chonnam National University Medical School, Gwangju; ³Department of Internal Medicine, Inje University, Koyang; ⁴Department of Internal Medicine, Korea University, Seoul; ⁵Department of Internal Medicine, Dong-A University, Pusan; ⁶Department of Internal Medicine, Catholic University of Daegu, Daegu; ⁷Department of Internal Medicine, Eulji University, Seoul, South Korea.

ABSTRACT
Objective
Recently we have proposed a modified set of criteria to settle the questions raised regarding the International Study Group (ISG) criteria for Behçet's disease (BD). The aim of the present study was to validate the two pre-existing criteria sets commonly used in Korea, the ISG criteria and the criteria of the Behçet's Disease Research Committee of Japan (Japanese criteria), as well as the proposed modified criteria.

Methods
The study population included 155 consecutive patients with BD and 170 controls with non-Behçet's rheumatic diseases. Detailed data for all of the subjects were recorded prospectively by the participating physicians on a standard form that listed the clinical features of BD. The sensitivity, specificity, and accuracy of each set of the criteria were measured.

Results
Of the three criteria sets employed, the modified criteria were the most accurate, with an accuracy of 96.3%. The ISG criteria often failed to classify the following patients with BD: patients with only oral and genital ulcerations, certain patients with intestinal ulcerations, patients who did not manifest oral ulcerations, and patients with acute disease but fewer than three recurrent oral ulceration relapses in a 1-year period. The Japanese criteria also failed to categorize the following patients with BD: patients with oral and genital ulcerations, and patients with oral ulcerations, skin lesions, and a positive pathergy reaction. In addition, the Japanese criteria misclassified some of the control subjects with non-Behçet's uveitis as having BD.

Conclusions
The results of this study suggest that there are some points that need to be reconsidered in the clinical application of the two pre-existing sets of criteria. Although the modified criteria were the most accurate, further validation studies will be required in other ethnic populations.

Key words
Behçet's disease, criteria, validation.

This research was supported by the research fund of Dankook University in 2003.
Please address correspondence and reprint requests to: Prof. Hyun Kyu Chang, MD, Division of Rheumatology, Department of Internal Medicine, College of Medicine, Dankook University, 16-5 Anseo-Dong, Cheonan, Chungcheong Nam Do, 330-715, South Korea.
E-mail: hanks22@dankook.ac.kr