Abstract
BACKGROUND: As no direct comparison is available between drugs to prevent endoscopic postoperative recurrence (POR) in Crohn's disease (CD), hierarchizing these therapeutic options remains challenging. OBJECTIVES: We aimed to compare the effectiveness of treatments to prevent CD endoscopic POR. DESIGN: Systematic review and network meta-analysis using a random-effects model. DATA SOURCES AND METHODS: We include studies comparing treatments to prevent CD POR according to PRISMA guidelines. The primary endpoint was endoscopic POR (Rutgeerts score ⩾i2). Surface under the cumulative ranking (SUCRA) was used to hierarchize the treatments. RESULTS: Twenty studies were included (2414 patients). Overall heterogeneity was moderate (τ = 0.34). Ustekinumab (odds ratio (OR) = 0.23 (0.07-0.70); OR = 0.29 (0.08-0.99)), vedolizumab (OR = 0.17 (0.05-0.59); OR = 0.22 (0.06-0.85)), infliximab (OR = 0.18 (0.36-0.88); OR = 0.23 (0.09-0.54)), and adalimumab (OR = 0.17 (0.07-0.42); OR = 0.22 (0.08-0.59)) were more effective to prevent endoscopic POR than placebo or 5-ASA, respectively, contrary to thiopurines (OR = 0.52 (0.22-1.24); OR = 0.66 (0.25-1.76)). Adalimumab (OR = 0.33 (0.15-0.74)) and infliximab (OR = 0.34 (0.13-0.87)) were more effective than thiopurines. While no difference was observed between the four biologics, adalimumab (SUCRA = 0.81), infliximab (SUCRA = 0.80), vedolizumab (SUCRA = 0.79), and ustekinumab (SUCRA = 0.72) had the highest likelihood of being the most effective drug, contrary to thiopurines (SUCRA = 0.41), 5-ASA (SUCRA = 0.24), or placebo (SUCRA = 0.16). CONCLUSION: This network meta-analysis confirms the efficacy of anti-TNF agents, vedolizumab, and ustekinumab in preventing endoscopic CD POR without any difference between them. When a prophylactic therapy is needed, biologics should be preferred to 5-ASA or thiopurines. TRIAL REGISTRATION: PROSPERO registration number CRD42024555528.