1. Publication
- Title: Melatonin for prevention of delirium in patients receiving mechanical ventilation in the intensive care unit: DEMEL, a multiarm multistage adaptive randomized controlled clinical trial
- Acronym: DEMEL
- Year & Journal: Intensive Care Medicine, published July 3, 2025 (2025;51(7):1292-1305)
- Citation: Mekontso Dessap A, Ricard JD, Contou D, et al; DEMEL Investigators. Intensive Care Med. 2025;51(7):1292-1305. doi:10.1007/s00134-025-08002-z
2. Context & Rationale
Background: Prior smaller trials and a meta-analysis (informing the 2025 SCCM PADIS update) suggested melatonin might improve sleep and prevent delirium, but evidence remained inconsistent. Optimal melatonin dosing/pharmacokinetics in critical illness were unclear.
Research Question/Hypothesis: In mechanically ventilated ICU patients, does an optimally-dosed melatonin regimen (determined via an adaptive PK-finding stage) reduce delirium prevalence vs placebo?
3. Design & Methods
- Study Type: Multicenter, adaptive, phase 2b/3, randomized, double-blind, placebo-controlled trial (multiarm multistage design)
- Setting & Centers: 20 centers, France; Feb 2019 – Jan 2021
- Population: Mechanically ventilated ICU patients
- Intervention: Enteral melatonin, low-dose (0.3mg) or high-dose (3mg), nightly at 9pm for 14 nights
- Comparator: Placebo
- Randomization: 1:1:1
- Blinding: Double-blind
- Statistical Power & Follow-Up: Interim (activity) stage: % achieving optimal PK profile at 24h. Final (efficacy) stage: % experiencing delirium.
4. Key Results
Outcome | Low-Dose Melatonin (0.3mg) | Placebo | Notes |
Delirium prevalence (primary, efficacy phase) | Not reduced | — | Low-dose melatonin failed to prevent delirium |
Optimal PK target achievement | Only 50% of low-dose patients | — | Substantial interindividual PK variability |
Note on data completeness: Exact numeric delirium-prevalence percentages/CI not accessible in the sourced text; the qualitative finding (failed to prevent delirium at the selected dose) is well-documented across multiple independent sources.
5. Internal Validity Assessment
Rigorous adaptive multiarm multistage design allowing dose selection based on PK data before efficacy testing — methodologically sophisticated. Overall: Strong, per an accompanying editorial ("rigorous and well-conducted study"), though only 50% of low-dose patients achieved the target PK profile, complicating interpretation of a true dose-response failure vs a PK-variability confound.
6. External Validity Assessment
French, 20-center, mechanically ventilated ICU population — broadly representative of general ICU populations at delirium risk.
7. Strengths & Limitations
Strengths: Novel adaptive PK-informed design; advances understanding of melatonin pharmacokinetics in critical illness.
Limitations: Only half of low-dose patients achieved optimal PK target — raises unresolved questions about interindividual dosing; efficacy phase used only the low dose.
8. Interpretation & Practice Impact
Does not support routine melatonin for delirium prevention in mechanically ventilated ICU patients at the studied dose, joining a consistent pattern of null-to-mixed melatonin trials (Pro-MEDIC, MELLOW) despite biological plausibility.
9. Controversies & Subsequent Evidence
An accompanying editorial ("Melatonin for ICU delirium: a promising molecule, a broken path") explicitly frames this as a well-conducted trial delivering "a wise message: even a biologically promising and theoretically sound intervention may fail to deliver clinical benefit," concluding bundled (non-pharmacologic) approaches remain the cornerstone of ICU delirium prevention. A related 2025 meta-analysis of 6 RCTs (2209 patients) similarly found no difference in delirium incidence with melatonin.
10. Summary & Executive Takeaway
Summary: DEMEL, an adaptive phase 2b/3 trial across 20 French ICUs, tested low- and high-dose melatonin against placebo for delirium prevention in mechanically ventilated patients. The selected low dose (0.3mg) failed to reduce delirium prevalence, with only 50% of patients achieving the target pharmacokinetic profile.
Overall Takeaway: Even with a sophisticated, PK-informed adaptive design, melatonin did not prevent delirium in mechanically ventilated ICU patients — reinforcing that bundled, non-pharmacologic approaches remain the evidence-based cornerstone of ICU delirium prevention.
11. Bibliography
- Wibrow B, Martinez FE, Myers E, et al. Prophylactic melatonin for delirium in intensive care (Pro-MEDIC). Intensive Care Med. 2022;48:414-425.
- Burry LD, Williamson DR, Detsky ME, et al. MELLOW feasibility trial.
- Melatonin for ICU delirium: a promising molecule, a broken path [editorial]. Intensive Care Med. 2025.