BACKGROUND: Fasting is frequently imposed before extubation in patients in intensive care units, with the aim to reduce risk of aspiration. This unevaluated practice might delay extubation, increase workload, and reduce caloric intake. We aimed to compare continued enteral nutrition until extubation with fasting before extubation in patients in the intensive care unit. METHODS: We conducted an open-label, cluster-randomised, parallel-group, non-inferiority trial in 22 intensive care units in France. Patients aged 18 years or older were eligible for enrolment if they had received invasive mechanical ventilation for at least 48 h in the intensive care unit and received prepyloric enteral nutrition for at least 24 h at the time of extubation decision. Centres were randomly assigned (1:1) to continued enteral nutrition until extubation or 6-h fasting with concomitant gastric suctioning before extubation, to be applied for all patients within the unit. Masking was not possible because of the nature of the trial. The primary outcome was extubation failure (composite criteria of reintubation or death) within 7 days after extubation, assessed in both the intention-to-treat and per-protocol populations. The non-inferiority margin was set at 10%. Pneumonia within 14 days of extubation was a key secondary endpoint. This trial is now complete and is registered with ClinicalTrials.gov, NCT03335345. FINDINGS: Between April 1, 2018, and Oct 31, 2019, 7056 patients receiving enteral nutrition and mechanical ventilation were admitted to the intensive care units and 4198 were assessed for eligibility. 1130 patients were enrolled and included in the intention-to-treat population and 1008 were included in the per-protocol population. In the intention-to-treat population, extubation failure occurred in 106 (17·2%) of 617 patients assigned to receive continued enteral nutrition until extubation versus 90 (17·5%) of 513 assigned to fasting, meeting the a priori defined non-inferiority criterion (absolute difference -0·4%, 95% CI -5·2 to 4·5). In the per-protocol population, extubation failure occurred in 101 (17·0%) of 595 patients assigned to receive continued enteral nutrition versus 74 (17·9%) of 413 assigned to fasting (absolute difference -0·9%, 95% CI -5·6 to 3·7). Pneumonia within 14 days of extubation occurred in ten (1·6%) patients assigned to receive continued enteral nutrition and 13 (2·5%) assigned to fasting (rate ratio 0·77, 95% CI 0·22 to 2·69). INTERPRETATION: Continued enteral nutrition until extubation in critically ill patients in the intensive care unit was non-inferior to a 6-h fasting maximum gastric vacuity strategy comprising continuous gastric tube suctioning, in terms of extubation failure within 7 days (a patient-centred outcome), and thus represents a potential alternative in this population. FUNDING: French Ministry of Health. TRANSLATION: For the Chinese translation of the abstract see Supplementary Materials section.