Crohn's disease activity index does not correlate with endoscopic recurrence one year after ileocolonic resection

Background: Crohn's disease clinical trials utilize the Crohn's Disease Activity Index (CDAI) to measure primary endpoint assessments of clinical recurrence and remission. We evaluated the extent of agreement between clinical recurrence/remission as defined by the CDAI and endoscopic recurrence 1 year after intestinal resection for Crohn's disease (CD). Methods: Twenty-four CD patients who had been randomly assigned to a postoperative clinical trial had 1 year clinical, endoscopic, and histological assessment for disease recurrence. The primary endpoint was the extent of agreement between endoscopic recurrence and clinical recurrence 1 year after intestinal resection for CD. Secondary endpoints were extent of agreement between endoscopic recurrence and the surrogate markers of CD activity, i.e., histological activity, sedimentation rate, and C-reactive protein (CRP). Results: Twelve of the 24 patients (50%) were in endoscopic remission (i0, i1) and 12 (50%) had endoscopic recurrence (i2, i3, or i4). There was good agreement between endoscopy and histological activity scores (intraclass correlation coefficient = 0.53, kappa coefficient = 0.58). In contrast, there was little to no relationship between endoscopy and CDAI scores; median CDAI scores for endoscopy scores of i0/i1, i2, i3, and i4 were 118, 76, 156, and 78, respectively (P for trend = 0.88). The kappa coefficient (of agreement) between endoscopy score ± 2 and CDAI score ± 150 was 0.12 (exact P = 0.68), indicating poor agreement. Similarly, there was no consistent association observed between endoscopy scores and mean CRP and ESR values at week 54. Conclusions: The CDAI shows poor agreement with endoscopic recurrence 1 year after intestinal resection. Endoscopic recurrence should be the primary endpoint of future postoperative studies and ileocolonoscopy the gold standard test to detect postoperative recurrence.