Prognostic factors for postoperative recurrence of Crohn's disease

Prognostic factors for postoperative recurrence of Crohn's disease (CD) have been widely investigated but not yet clearly identified. PURPOSE: Aim of this study was, therefore, to analyze the association between demographic, clinical, laboratory, and surgical characteristics of patients and the cumulative probability of endoscopic postoperative recurrence. METHODS: The study was performed in 110 patients who were enrolled in the Italian multicenter, randomized, controlled trial on the effectiveness of 5-aminosalicylic acid (5-ASA) in the prevention of postoperative recurrence in CD. Patients had undergone their first intestinal resection for CD of the terminal ileum with or without involvement of cecum ascending colon. Recurrence was defined on the basis of endoscopy. The following variables were evaluated as potential prognostic factors: gender, age, years since diagnosis, clinical course (perforative and nonperforative), Crohn's Disease Activity Index score, white blood count, erythrocyte sedimentation rate, C-reactive protein, and orosomucoids assessed before the operation. Timing of operation (elective or urgent), type of anastomosis (end-to-end, end-to-side, side-to-end, side-to-side), and prophylactic treatment were also evaluated. Colon ileoscopy was performed at 6, 12, 24, and 36 months after operation. The association between variables and the cumulative proportion of recurrence was analyzed both by univariate analysis (life table method, log-rank test) and multivariate regression analysis (Cox's model, stepwise procedure). RESULTS. Results of this study indicate that, of the features considered before surgery, only leukocytosis (white blood count, >9,000 mi) was significantly associated with an increased risk of recurrence: (P < 0.05) at univariate analysis. This finding was not confirmed by multivariate analysis. A trend toward a higher risk of recurrence for patients who have had a resection with end-to-end anastomosis compared with those who have had a resection and other types of anastomosis was also observed. This trend reached significancy in the group of patients submitted to treatment with 5-ASA. The multivariate analysis showed that 5-ASA-treated patients with end-to-end anastomosis had a risk of recurrence more than threefold higher than those with other types of anastomosis (relative risk, 3.40; 95 percent confidence interval, 1.00-11.96; P < 0.03). CONCLUSIONS: From a practical point of view, it has been estimated that the combination of intestinal resection plus side-to-side or end-to-side anastomosis with oral 5-ASA treatment reduces by 64 percent the postoperative recurrence rate in CD at three years follow-up.