Melatonin for prevention of delirium in patients receiving mechanical ventilation in the intensive care unit: a multiarm multistage adaptive randomized controlled clinical trial (DEMEL)
Armand Mekontso Dessap, Jean-Damien Ricard, Damien Contou, Cyrielle Desnos, Maxens Decavèle, Romain Sonneville, Emmanuel Vivier, Nicolas Terzi, Jean-Christophe Callahan, Sébastien Jochmans, Leon Rusel, Serge Carreira, Saad Nseir, Frank Chemouni, Vincent Castelain, Muriel Paul, Jean-François Benoist, Étienne Audureau, Keyvan Razazi, Nicolas de Prost, Guillaume Carteaux, Segolène Gendreau, Gaëtan Plantefève, Olivier Pajot, Amélie Chenal, Damien Roux, Coralie Gernez, Noémie Zucman, Muriel Fartoukh, Vincent Labbé, Guillaume Voiriot, Luc Haudebourg, Morgane Faure, Alexandre Demoule, Jean-François Timsit, Lila Bouadma, Etienne Montmollin, Gael Bourdin, Audrey Large, Fanny Doroszewsky, Louis-Marie Galerneau, Guillaume Rigault, Clara Candille, Christophe Guitton, Mickaël Landais, Nicolas Chudeau, Mickael le Moal, Maryse Louistisserand, Kmar Hraiech, Tommaso Maraffi, Frédérique Schortgen, Cecila Tabra Osorio, Malo Emery, Anne Sophie le Floch, Raphaël Favory, Sébastien Préau, Julien Poissy, Jonathan Zarka, Susanna Bolchini, Antonin Michaud, Antonin Hugerot, Jean Dellamonica, Samir Jaber, Jérome Devaquet, Daniel da Silva, Jean François Georger, Didier Chevenne, Olivier Thirion, Bernard Do, Lauriane Segaux
Abstract
PURPOSE: To determine the dose of melatonin with an optimal pharmacokinetic profile and to test whether this dose reduces the prevalence of delirium in mechanically ventilated ICU patients as compared to placebo. METHODS: DEMEL, a multicenter adaptive phase 2b/3 randomized, placebo-controlled, double-blind trial included patients at 20 health centers in France from February 1st, 2019 through January 5th, 2021. Patients were randomized (1:1:1) to receive either placebo or low (0.3 mg) or high (3 mg) dose melatonin enterally at 9:00 p.m. for 14 consecutive nights or until death or ICU discharge, whichever came first. The interim primary endpoint (activity stage) was the percentage of patients who achieved an optimal melatonin pharmacokinetic profile 24 h after starting study treatment; the final primary endpoint (efficacy phase) was the percentage of patients who experienced delirium between randomization and day 14 (or until death or ICU discharge, whichever came first). Delirium was assessed twice daily using the Confusion Assessment Method for ICU. RESULTS: We randomized 355 patients and included 334 in the primary analysis. At the preplanned analysis of the activity stage performed in 75 patients, the low-dose melatonin group had the highest rate of optimal pharmacokinetic profiles (12/24, 50%) when compared with the high-dose melatonin group (6/25, 24%) and the placebo group (0/26). Therefore, the Steering Committee recommended that the high-dose melatonin group be discontinued and that the low-dose melatonin group be selected to continue in the efficacy phase along with the placebo group. At the end of the efficacy stage, there was no difference in the final primary outcome of delirium incidence between the low-dose melatonin group and the placebo group: 80/147 (54.4%) vs 85/154 (55.2%), risk ratio, 0.986 [95% CI 0.803 to 1.211]; key secondary outcomes were also similar between groups. These included sleep quality, delirium-free, coma-free, and ventilator-free days at day 28; ICU and hospital length of stay; mortality at day 28, in the ICU, and in hospital; as well as long-term outcomes such as quality of life and postintensive care syndrome at day 90. CONCLUSIONS: This randomized clinical trial found that the low-dose of melatonin (0.3 mg nightly) achieved a better pharmacokinetic profile than the high-dose (3 mg nightly), but did not change the incidence of delirium compared to placebo in mechanically ventilated critically-ill patients. TRIAL REGISTRATION: ClinicalTrial.gov website (NCT03524937).