APLAR validation of Behcet’s disease (BD) diagnosis criteria, Iran experience.


Abstract: Considering that the majority of BD patients are from APLAR (Asia and Pacific League of Associations for Rheumatology) countries any diagnosis criteria for BD must have an adequate accuracy in these countries. The APLAR subcommittee for Behcet's Disease has set a prospective study to assess the accuracy of the existing diagnosis criteria. China, Iran, Japan, and Korea, the major countries with BD are participating to this study. The data from Iran were collected on every consecutive and new patients seen at the Behcet’s Unit, from1993. BD patients were 1084 and control patients were 472. The sensitivity was as fallow: for the Mason and Barnes criteria 60.6% ± 2.9 (confidence interval calculated at 95%). For the O'Duffy criteria 67.2% + 2.8. For the International criteria 79.8% + 2.4. For the Dilsen criteria 83% + 2.2. For the Japan criteria 84.3% + 2.2. For the traditional format of the Iran criteria 91.1% + 1.7, and for the classification tree of the Iran criteria 96.9% + 1. The specificity was as follow: for the Mason and Barnes criteria 99.8% + 0.4. For the O'Duffy criteria 99.4% + 0.7. For the international criteria 99.8% ± 0.4. For the Dillsen criteria 96.4% + 1.7. For the Japan criteria 98.5% + 1.1. For the traditional format of the Iran criteria 98.7% + 1, and for the classification tree of the Iran criteria 99.8% + 0.4. The Accuracy was as follow: for the Mason and Barnes criteria 80.2% + 2. For the O'Duffy criteria 83.3% + 1.9. For the International criteria 89.8% + 1.5. For the Dilsen criteria 89.7% + 1.5. For the Japan criteria 91.4% + 1.4. For the traditional format of the Iran criteria 94.9% +1.1, and for the classification tree of the Iran criteria 98.3% + 0.6. The Iran criteria had the highest accuracy. No statistically significant difference exists between results obtained from China, Iran, and Korea.