Background: Infliximab is a monoclonal antibody that binds and neutralises tumour necrosis factor-alpha (TNF-α) which is present in high levels in the blood serum, mucosa and stool of patients with Crohn's disease.

Objectives: To determine the efficacy and safety of infliximab for maintaining remission in patients with Crohn's disease.

Search methods: On 31 August, 2021 and 23 June, 2023, we searched CENTRAL, Embase, MEDLINE, ClinicalTrials.gov, and WHO ICTRP.

Selection criteria: Randomised controlled trials (RCTs) in which infliximab was compared to placebo or another active comparator for maintenance, remission, or response in patients with Crohn's disease.

Data collection and analysis: Pairs of review authors independently selected studies and conducted data extraction and risk of bias assessment. We expressed outcomes as risk ratios and mean differences with 95% confidence intervals. We assessed the certainty of the evidence using GRADE. Our primary outcome was clinical relapse. Secondary outcomes were loss of clinical response, endoscopic relapse, and withdrawal due to serious and adverse events.

Main results: Nine RCTs with 1257 participants were included. They were conducted between 1999 and 2022; seven RCTs included biologically-naive patients, and the remaining two included a mix of naive/not naive patients. Three studies included patients in clinical remission, five included patients with a mix of activity scores, and one study included biologic responders with active disease at baseline. All studies allowed some form of concomitant medication during their duration. One study exclusively included patients with fistulating disease. The age of the participants ranged from 18 to 69 years old. All but one single-centre RCT were multicentre RCTs. Four studies were funded by pharmaceutical companies, two had a mix of commercial and public funding, and two had public funding. Infliximab is probably superior to placebo in preventing clinical relapse in patients who have mixed levels of clinical disease activity at baseline, and are not naive to biologics (56% vs 75%, RR 0.73, 95% CI 0.63 to 0.84, NNTB = 5, moderate-certainty evidence). We cannot draw any conclusions on loss of clinical response (RR 0.59, 95% CI 0.37 to 0.96), withdrawals due to adverse events (RR 0.66, 95% CI 0.37 to 1.19), or serious adverse events (RR 0.60, 95% CI 0.36 to 1.00) because the evidence is very low certainty. Infliximab combined with purine analogues is probably superior to purine analogues for clinical relapse (12% vs 59%, RR 0.20, 95% CI 0.10 to 0.42, NNTB = 2, moderate-certainty evidence), for patients in remission, and who are not naive to biologics. We cannot draw any conclusions on withdrawals due to adverse events (RR 0.47, 95% CI 0.15 to 1.49), and serious adverse events (RR 1.19, 95% CI 0.54 to 2.64) because the evidence is very low certainty. We cannot draw any conclusions about the effects of infliximab on serious adverse events compared to purine analogues (RR 0.79, 95% CI 0.37 to 1.68) for a population in remission at baseline because the evidence is very low certainty. There was no evidence available for the outcomes of clinical relapse, loss of clinical response, and withdrawal due to adverse events. Infliximab may be equivalent to biosimilar for clinical relapse (47% vs 40% RR 1.18, 95% CI 0.82 to 1.69), and it may be slightly less effective in averting loss of clinical response (49% vs 32%, RR 1.50, 95% CI 1.01 to 2.23, low-certainty evidence), for a population with mixed/low disease activity at baseline. Infliximab may be less effective than biosimilar in averting withdrawals due to adverse events (27% vs 0%, RR 20.73, 95% CI 2.86 to 150.33, low-certainty evidence). Infliximab may be equivalent to biosimilar for serious adverse events (10% vs 10%, RR 0.99, 95% CI 0.39 to 2.50, low-certainty evidence). We cannot draw any conclusions on the effects of subcutaneous biosimilar compared with intravenous biosimilar on clinical relapse (RR 1.01, 95% CI 0.65 to 1.57), loss of clinical response (RR 0.94, 95% CI 0.70 to 1.25), and withdrawals due to adverse events (RR 0.77, 95% CI 0.30 to 1.97) for an active disease population with clinical response at baseline because the evidence is of very low certainty. We cannot draw any conclusions on the effects of infliximab compared to adalimumab on loss of clinical response (RR 0.68, 95% CI 0.29 to 1.59), withdrawals due to adverse events (RR 0.10, 95% CI 0.01 to 0.72), serious adverse events (RR 0.09, 95% CI 0.01 to 1.54) for an active disease population with clinical response at baseline because the evidence is of very low certainty. There was no evidence available for the outcome of clinical relapse.

Authors' conclusions: Infliximab is probably more effective in preventing clinical relapse than placebo (moderate-certainty evidence). Infliximab in combination with purine analogues is probably more effective in preventing clinical and endoscopic relapse than purine analogues alone (moderate-certainty evidence). No conclusions can be drawn regarding prevention of loss of clinical response, occurrence of withdrawals due to adverse events, or total adverse events due to very low-certainty evidence for both of these comparisons. There may be little or no difference in prevention of clinical relapse, withdrawal due to adverse events or total adverse events between infliximab and a biosimilar (low-certainty evidence). Infliximab may lead to more loss of clinical response than a biosimilar (low-certainty evidence). We were unable to draw meaningful conclusions about other comparisons and outcomes related to missing data or very low-certainty evidence due to serious concerns about imprecision and risk of bias. Further research should focus on comparisons with other active therapies for maintaining remission, as well as ensuring adequate power calculations and reporting of methods.