Melatonin Use in the ICU: A Systematic Review and Meta-Analysis.
Journal Articles
Overview
Research
Identity
View All
Overview
abstract
OBJECTIVES: Melatonin has wide-ranging effects on the body, including the regulation of circadian rhythm, and potentiation of cellular immune and antioxidant activities. In critically ill patients, endogenous melatonin has been shown to be markedly deranged and reduced. Therefore, the purpose of this systematic review and meta-analysis was to determine if exogenous supplementation of melatonin improves patient-centered outcomes. DATA SOURCES: We searched five electronic databases. STUDY SELECTION: Randomized clinical trials (RCTs) that compared melatonin to no melatonin in adults admitted to the ICU were identified. DATA EXTRACTION: We aggregated data as relative risks, mean differences (MDs), and standard mean differences (SMDs) using a random-effects model. Supporting evidence for each effect was evaluated for certainty using the Grading Recommendations, Assessment, Development, and Evaluations approach. DATA SYNTHESIS: In total, 32 RCTs (n = 3895 patients) were included. We found that melatonin may reduce delirium (relative risk [RR] 0.72; 95% CI, 0.58-0.89; low certainty), may slightly reduce ICU length of stay (MD -0.57 d; 95% CI, -0.95 to -0.18 d; low certainty), and may improve reported sleep quality (SMD 0.54; 95% CI, 0.01-1.07; low certainty). Melatonin may result in a slight reduction in the frequency of adverse events (low certainty). Evidence was uncertain with regards to the frequency of sleep awakenings, anxiety level, agitation, and post-traumatic stress disorder incidence (all very low certainty), as well as to ICU mortality and post-ICU functional status (both low certainty). CONCLUSIONS: Our findings suggest that melatonin administration in the critically ill may improve perceived sleep and reduce delirium, without increasing adverse effects. Certainty of evidence was negatively affected by the risk of bias and inconsistency. Future RCTs should focus on identifying optimal dosing, administration timing, improving measurements of sleep outcomes, and target populations.
