What if melatonin could help patients with severe COVID-19?

INTRODUCTION

In March 2020, a protocol recommending the prescription of melatonin, among other sleep- and biorhythms-promoting measures, to hospitalized patients with coronavirus disease 2019 (COVID-19) with sleep problems or delirium was sent from the consultation-liaison psychiatrist to the medical staff of the Fundación Jiménez Díaz University Hospital (FJDUH) in Madrid, Spain. Several authors have suggested a potential benefit of melatonin use in COVID-19.^1–4 In addition to its circadian function, melatonin is thought to have several health-promoting properties, including immune response modulation and anti-inflammatory and antioxidant properties.⁵ We here report a retrospective analysis showing an association of melatonin with survival in a sample of 2,463 patients with COVID-19 hospitalized during the first wave of the pandemic, 265 of whom (10.75%) were given 2–6 mg of oral melatonin at 21 hours during admission (median of first day of administration was day 4, and 25% of patients received melatonin from the first day). Our work and that of Ramlall et al⁶ are the first to show real-world clinical data supporting a possible benefit of melatonin in COVID-19.

To reduce the possibility of a biased, biologically nonrelevant association of melatonin with survival, we excluded from the sample patients who died during the first 72 hours of admission without taking melatonin and patients who started on melatonin in the last 7 days of their admittance, having completed 75% of their stay. The remaining sample comprised 224 patients who received melatonin and 1,952 patients who did not receive melatonin. Both groups included patients admitted in the intensive care unit (or intermediate respiratory care unit), with the patients of the melatonin group having more probability of intensive care unit/intermediate respiratory care unit admission (Table 1). To control for baseline differences between the 2 groups we performed a propensity score matching. The melatonin group showed a much lower mortality rate (10.7% vs 23.7%) compared with the non-melatonin matched group, with an odds ratio of 0.39 (Table 1). We had data available on CURB-65 (confusion, blood urea nitrogen >19 mg/dL, respiratory rate ≥30 breaths/minute, low blood pressure, and age ≥ 65 years; a validated scale of clinical severity⁷) for 343 (76.5%) out of 448 patients in the matched groups, 179/224 in the melatonin group, and 164/224 in the non-melatonin matched group. No differences were found between the 2 groups in the distribution of their CURB-65 scores, suggesting that they were similar in terms of illness severity at admission.

One possible pathophysiological mechanism to explain age-related vulnerability to COVID-19 is the progressive loss of endogenous melatonin with aging.⁸ The circadian disruption that intensive care unit patients^9,10 and other hospitalized patients experience also likely contributes to the pathophysiology of acutely ill patients with COVID-19. The circadian rhythm strengthening of melatonin and other measures, such as appropriate daytime lighting and activity, could be of benefit not only for the management of COVID-19 but also for other diseases.

This report shows the first set of data of a bigger analysis we are performing on the effect of melatonin on the clinical evolution of patients admitted in the FJDUH throughout the pandemic. We are aware that a retrospective analysis prevents us from establishing a causal association between melatonin and survival. Prospective studies are already on their way^11,12 to assess the utility of melatonin as an adjunctive treatment for COVID-19. However, with no time to lose and given its safety profile and low cost, our data may help the clinician to consider the use of melatonin in patients with COVID-19.

DISCLOSURE STATEMENT

All authors have participated in the elaboration of this manuscript, and all of them have seen and approved its final version. Work for this article was performed at the Fundación Jiménez Díaz University Hospital in Madrid, Spain. None of the authors received any financial support for this work. The authors report no conflicts of interest.

REFERENCES