Long-term clinical course and prognostic factors in intestinal Behcet's disease

PURPOSE: The present study was aimed at evaluating the long-term course of intestinal Behcet's disease and determining predictive factors of prognosis. METHODS: This report is a retrospective study based on the records of 43 patients with intestinal Behcet's disease. The mean follow-up duration was 73 +/- GO months. We evaluated the efficacy of medical treatment for the intestinal lesion at initial eight weeks. The cumulative probabilities were calculated by using Kaplan-Meier method, and the results were compared by using the log-rank test. RESULTS: Sixteen patients (38 percent) achieved a complete remission of intestinal lesions eight weeks after medical treatment had begun. The patients who achieved a complete remission had a lower probability of receiving an operation than those who had not (13 percent at 2 and 5 years vs. 36 and 43 percent, respectively; P = 0.028). The recurrence probability of intestinal lesions was 25 percent at two years and 43 percent at fire years after complete remission with medical treatment. Patients who had a history of intestinal perforation or fistula had a higher probability of recurrence after operation than those without such history (59 vs. 33 percent at 2 rears; 88 vs. 57 percent at 5 years; P = 0.020). Patients who had taken azathioprine had a lower probability of receiving reoperation than those who did not (7 vs. 25 percent at 2 years, 25 vs. 47 percent at 5 years; P = 0.035). The length of ileal resection and whether hemicolectomy was performed had no significant effect on the recurrence or reoperation rate. CONCLUSIONS: Intestinal Behcet's disease frequently requires a surgical treatment and has a high recurrence rate. The patients a ho achieved a complete remission with medical treatment, who had no history of intestinal perforation, and who received azathioprine after operation showed better clinical courses. Resection of a short segment of bowel would be a more appropriate surgical procedure.