Abstract
BACKGROUND AND AIMS: We aimed to quantify the magnitude of the association between endoscopic recurrence and clinical recurrence [symptom relapse] in patients with postoperative Crohn's disease. METHODS: Databases were searched to October 2, 2020, for randomised controlled trials [RCTs] and cohort studies of adult patients with Crohn's disease with ileocolonic resection and anastomosis. Summary effect estimates for the association between clinical recurrence and endoscopic recurrence were quantified by risk ratios [RR] and 95% confidence intervals [95% CI]. Mixed-effects meta-regression evaluated the role of confounders. Spearman correlation coefficients were calculated to assess the relationship between these outcomes as endpoints in RCTs. An exploratory mixed-effects meta-regression model with the logit of the rate of clinical recurrence as the outcome and the rate of endoscopic recurrence as a predictor was also evaluated. RESULTS: In all, 37 studies [N = 4053] were included. For eight RCTs with available data, the RR for clinical recurrence for patients who experienced endoscopic recurrence was 10.77 [95% CI 4.08 to 28.40; GRADE moderate certainty evidence]; the corresponding estimate from 11 cohort studies was 21.33 [95% CI 9.55 to 47.66; GRADE low certainty evidence]. A single cohort study showed a linear relationship between Rutgeerts score and clinical recurrence risk. There was a strong correlation between endoscopic recurrence and clinical recurrence treatment effect estimates as trial outcomes [weighted Spearman correlation coefficient 0.51]. CONCLUSIONS: The associations between endoscopic recurrence and subsequent clinical recurrence lend support to the choice of endoscopic recurrence to monitor postoperative disease activity and as a primary endpoint in clinical trials of postoperative Crohn's disease.