RESUMO
BACKGROUND & AIMS: Colonoscopy (CS) is the gold standard to assess postoperative recurrence (POR) in Crohn's disease (CD). However, CS is invasive and may be poorly tolerated by patients. The aim of this study was to prospectively assess the diagnostic accuracy of a noninvasive approach in detecting POR, using the endoscopic Rutgeerts' score (RS) as the reference standard. METHODS: Consecutive patients with CD who underwent ileo-cecal resection were prospectively enrolled in 3 referral Italian centers. Patients underwent CS and bowel ultrasound within 1 year of surgery. Uni- and multivariable analyses were used to assess the correlation between noninvasive parameters and endoscopic recurrence, defined by a RS ≥2. RESULTS: Ninety-one patients were enrolled. Sixty patients (66%) experienced endoscopic POR. The multivariable analysis identified bowel wall thickness (BWT) per 1-mm increase (odds ratio [OR], 2.43; 95% confidence interval [CI], 1.21-4.89; P = .012), the presence of mesenteric lymph nodes (OR, 15.63; 95% CI, 1.48-164.54; P = .022), and fecal calprotectin (FC) values ≥50 mcg/g (OR, 8.58; 95% CI, 2.45-29.99; P < .001) as independent predictors for endoscopic recurrence. The presence of lymph nodes or the combination of BWT ≥3 mm and FC values ≥50 mcg/g correctly classified 56% and 75% of patients, with less than 5% of patients falsely classified as having endoscopic recurrence. Conversely, the combination of BWT <3 mm and FC <50 mcg/g correctly classified 74% of patients with only 4.5% of patients falsely classified as not having endoscopic recurrence. CONCLUSIONS: A noninvasive approach combining bowel ultrasound and FC can be used with confidence for detecting POR in patients with CD without the requirement for CS.