Melatonin supplementation improves N‐terminal pro‐B‐type natriuretic peptide levels and quality of life in patients with heart failure with reduced ejection fraction: Results from MeHR trial, a randomized clinical trial

Abstract Background Melatonin, the major secretion of the pineal gland, has beneficial effects on the cardiovascular system and might advantage heart failure with reduced ejection fraction (HFrEF) by attenuating the effects of the renin–angiotensin–aldosterone and sympathetic system on the heart besides its antioxidant and anti‐inflammatory effects. Hypothesis We hypothesized that oral melatonin might improve echocardiographic parameters, serum biomarkers, and a composite clinical outcome (including quality of life, hospitalization, and mortality) in patients with HFrEF. Methods A placebo‐controlled double‐blinded randomized clinical trial was conducted on patients with stable HFrEF. The intervention was 10 mg melatonin or placebo tablets administered every night for 24 weeks. Echocardiography and measurements of N‐terminal pro‐B‐type natriuretic peptide (NT‐Pro BNP), high‐sensitivity C‐reactive protein, lipid profile, and psychological parameters were done at baseline and after 24 weeks. Results Overall, 92 patients were recruited, and 85 completed the study (melatonin: 42, placebo: 43). Serum NT‐Pro BNP decreased significantly in the melatonin compared with the placebo group (estimated marginal means for difference [95% confidence interval]: 111.0 [6.2–215.7], p = .044). Moreover, the melatonin group had a significantly better clinical outcome (0.93 [0.18–1.69], p = .017), quality of life (5.8 [0.9–12.5], p = .037), and New York Heart Association class (odds ratio: 12.9 [1.6–102.4]; p = .015) at the end of the trial. Other studied outcomes were not significantly different between groups. Conclusions Oral melatonin decreased NT‐Pro BNP and improved the quality of life in patients with HFrEF. Thus it might be a beneficial supplement in HFrEF.


| INTRODUCTION
Heart failure with reduced ejection fraction (HFrEF) is a worldwide growing problem. Several effective medications have been developed which reduce symptoms and increase the survival of these patients. 1 However, the disease process is frequently ongoing. Underlying intrinsic mechanisms such as mitochondrial abnormalities and cardiomyocyte hypertrophy and fibrosis and extrinsic mechanisms like overactivation of the sympathetic nervous system (SNS) and renin-angiotensin-aldosterone system (RAAS) promote cardiomyocyte apoptosis and gradual deterioration of the left ventricular function. 2 This process is not entirely preventable by current heart failure (HF) treatments 2 ; thus, alternative medications targeting different mechanisms or having synergistic effects with existing drugs might improve the patients' overall health and quality of life.
Melatonin is mainly secreted by the pineal gland with the primary role of coordinating the circadian rhythm. Recently, it has been shown that melatonin has beneficial effects on the cardiovascular system by its cytoprotective and antioxidant properties and by ameliorating mitochondrial dysfunction and regulating the endocrine system, such as SNS and RAAS. [3][4][5][6] Regarding the field of HF, numerous studies demonstrated that melatonin supplementation prevents the development or progression of the disease in animal models of both ischemic and nonischemic HF. 5 Also, observational studies in humans have shown lower melatonin levels in patients with severe HF or those with hypertensive cardiomyopathy who developed HF. 7,8 However, the effect of exogenous melatonin on patients with established HFrEF is unclear, and it has rarely been assessed in clinical trials. 9 Thus, we aimed to evaluate the effect of oral melatonin supplementation on echocardiographic parameters, serum N-terminal pro-B-type natriuretic peptide (NT-Pro BNP), and clinical outcomes in patients with HFrEF.

| Study design
This report adheres to the Consolidated Standards of Reporting Trials (CONSORT). 10  The MeHR trial was a double-blinded randomized placebocontrolled clinical trial with two parallel arms, conducted from January 2019 to August 2020 in outpatient Chamran Cardiology Clinics.
The design and methods of the study are fully described elsewhere. 11

| Study participants
Participants were selected from patients with a definite diagnosis of HFrEF (left ventricular ejection fraction [LVEF] < 40%) who were symptomatic (New York Heart Association [NYHA] class II or III) and regularly attended a specialist in mentioned clinics. In addition, they had to be clinically stable for at least 3 months and on essential drugs for HFrEF according to the 2016 ESC guidelines, that is, receiving maximum tolerated doses of an angiotensin-converting enzyme inhibitor/angiotensin-receptor blocker, a beta-blocker, and spironolactone/eplerenone unless contraindicated, and diuretics if needed. The medications and other indicated treatments were prescribed by the specialists who referred the patients for participating in the study. Patients suspected to need device therapy (pacemaker or implantable cardioverter-defibrillator) within the next 6 months were not enrolled.
Details of eligibility criteria have already been declared. 11 Eligible patients were enrolled and randomized by block randomization to the placebo or melatonin groups. The participants, primary investigator, and all outcome assessors were blinded to the study groups.

| Interventions
The intervention was 10 mg melatonin or placebo tablets, identical in shape and physical properties made by Sepid Teb (Sepid Teb Co.) and prescribed for at least 24 weeks, one tablet at bedtime. The drug was delivered to the patients in identical boxes containing 100 tablets and they were requested to return the unconsumed pills in the scheduled follow-ups. Patients' adherence and the adverse effects were assessed regularly by telephone calls, and patients attended the study center after 12 and 24 weeks of intervention for outcome evaluation and pill count to objectively evaluate adherence to the intervention.

| Outcomes
The primary outcomes of the study were variations in LVEF and left ventricular end-diastolic diameter (LVEDD), measured by echocardiography at Week 24 relative to the baseline, and changes in serum levels of NT-Pro BNP at Week 24. Also, a compound clinical score was calculated at Week 24 for each patient, composed from allcause mortality, any hospitalization for HF exacerbation, and changes in quality of life measured by the Minnesota Living with Heart Failure Questionnaire (MLHFQ). 12 Each component was scored according to Table S1 and the sum of scores was calculated as the compound clinical outcome.
Clinical events were collected at baseline and during the study.
Hospitalization was recorded from patients' medical documents, and mortality was recognized by verbal autopsy. An expert blinded committee of three cardiologists confirmed the events by evaluating the patient's medical records.
Other outcomes analyzed in this report were serum levels of high-sensitivity C-reactive protein (hs-CRP), lipid profile, and liver and renal function tests measured at baseline and end of the intervention covering the physical and emotional domains, and the total score ranges from 0 to 105 (from best to worst). 13 The Pittsburgh Sleep Quality Index is a valid tool for screening the sleep quality and quantity consisting of different subdomains and a total score of 0-21.
Scores more than five indicate poor sleep quality. 14 We used the trait part of Spielberger Anxiety Inventory (range: 20-80) and Beck Depression Inventory-II (range: 0-63) to evaluate the psychological status of our patients before and after the intervention, both on a Four-Likert scale. 15,16 All questionnaires were self-administered; however, a trained questioner read the items and recorded the answers for illiterate patients.

| Sample size and statistical methods
The sample size of 90 was calculated to find a 5% difference in LVEF between groups at the significance level of 0.05 (two-tailed) and the power of 90%, based on a study by Garakyaraghi et al. 9 Statistical analyses was performed by SPSS version 22 (IBM SPSS Statistics).
The data are presented as median (interquartile range) or percentage when appropriate. The independent sample T-test/Mann-Whitney U test or χ 2 test were used to compare baseline variables and the rate of adverse events between groups. The primary and secondary outcomes were analyzed using analysis of covariance, all adjusted for baseline values and other covariates as appropriate for the examined outcome. The missing data met the assumptions of missing completely at random. An intention-to-treat analysis approach was employed by including all cases with at least one measurement. A statistically significant level of less than 0.05 was acceptable for twosided tests. Pill count showed a high adherence to the intervention in participants who completed the study, ranging from 80% to 100% (97.8% in the melatonin and 98.7% in the placebo group).  Table 3 shows the EMM for primary and secondary outcomes of the trial.

| The compound clinical outcome
During the study, 12 hospitalizations occurred among patients who completed the study, of which 3 were because of HF exacerbation (one from five in melatonin and two from six in the placebo group).

| Psychological parameters
In our study, 24 weeks of melatonin supplementation did not affect the score of sleep quality, anxiety, or depression questionnaires (Table 3). Moreover, none of the subdomain scores of

| Lipid profile and liver and renal function tests
Although the mean differences of lipid profiles from baseline to difference. In addition, the renal function tests were not significantly affected by the treatment during the study. The aspartate transaminase and alanine aminotransferase levels were somewhat improved in the melatonin group, but they did not reach a statistically significant difference (Table 3).

| Drug-related adverse effects
From 92 allocations, three did not receive the intervention (two in the melatonin group and one in the placebo group). Throughout the study, 13 of 89 patients reported at least one type of drug-related adverse effect (  In a study, Garakyaraghi et al. 9 showed that 3 mg oral melatonin for 2 months significantly improved LVEF and NYHA class in patients with HFrEF. We detected an improvement in NYHA class of the melatonin group, too; however, we found no significant change in echocardiographic parameters in any groups. This discrepancy might be due to differences in characteristics of the study participants, their inhibition of the RAAS system is controversial in different studies. 18,19 Furthermore, evidence proposes that exogenous melatonin protects the myocardium from excess epinephrine toxicity and reduces adrenergic activity in humans. 20,21 Melatonin's ability to improve mitochondrial dysfunction might alleviate myocardial function and decrease overload. In experimental The beneficial effect of melatonin was also detectable on the composite clinical outcome, including mortality, hospitalization due to HF decompensation, and quality of life. However, we had no mortality and few hospitalizations during the study. Indeed, most of the clinical benefit of melatonin was its effect on the quality of life.
Melatonin did not affect sleep quality and quantity, anxiety, and depression levels during this investigation. There are discrepancies among previous studies and recent systematic reviews regarding the effect of melatonin on sleep patterns; overall, it seems that the melatonin effect is dependent on several factors such as dose, duration of use, pharmacokinetic of the melatonin formulations, and the background characteristics of the studied population. 23,24 Furthermore, the effect of melatonin on mental health is inconsistent. Jafari-Vayghan et al. 25 found that 20 mg melatonin for 8 weeks improved the overall quality of life and physical dimension of MLHFQ scores in cachectic patients with HF but had no effect on its emotional dimension. Melatonin ameliorated depressive-like behaviors in animal models 26 ; however, the evidence for its role in human mood disorders is not conclusive. 27 Extensive trials specifically designed for these purposes would be helpful to the field of cardiac diseases.
Melatonin is supposed to improve serum lipid profile in various target populations, possibly by directly regulating lipid metabolism and reducing the detrimental effect of oxidizing lipoproteins on the cardiovascular system. 28 Our insignificant results despite substantial improvement of total cholesterol, triglyceride, and LDL might be due to low sample size or lower levels of baseline lipids in our patients, as Mohammadi-Sartang et al. 29 demonstrated that higher doses of melatonin (>8 mg) and lower baseline total cholesterol (<200 mg/dl) are associated with a significant decline in serum cholesterol. 29 We performed liver and renal function tests for the primary purpose of detecting any drug-related adverse effects in our patients.
On the other hand, several studies have shown melatonin to ameliorate nonalcoholic fatty liver disease and decrease liver transaminases in concordance with moderately lower levels of these enzymes in our melatonin group. 30 Statins are valuable drugs in HFrEF irrespective of HF etiology, and increased liver transaminases are a concern for their prescription; thus, future studies can focus on the probable positive effect of melatonin on tolerability to the statins in these patients. 31 Overall, drug-related adverse effect profiles of our patients indicated that long-term melatonin supplementation with this dose is safe for patients with HFrEF.
We found an overall 35% reduction in NT-Pro BNP concentra- However, the magnitude of difference between treatment and control groups in clinical trials leading to significant clinical benefits is not defined precisely. 35 Recent studies recommend multiple biomarkers such as serum troponin in addition to NT-Pro BNP to determine the long-term effect of interventions on HF prognosis, which might be a limitation to our study. 34 The low sample size was also a limitation of the study, which might have caused some insignificant results. This low sample size might also have affected the clinical outcomes such as death and hospitalization despite a respectively long follow-up and made any conclusion about these outcomes unreliable. However, this sample size seems enough to evaluate the primary outcomes. Also, the high number of missing in the first follow-up was another limitation imposed by the urgent COVID-19 epidemics, although none of the primary outcomes had been planned to be evaluated at the first follow-up.
Extended follow-up time relative to other clinical trials using melatonin in cardiovascular diseases is a strength of the MeHR trial.
Also, up to our knowledge, this is the first study that includes biological biomarkers, structural parameters, and patient-oriented outcomes to evaluate the melatonin effect in HFrEF.
Overall, melatonin might lower serum NT-Pro BNP and improve disease-specific health-related quality of life in patients with HFrEF.
Thus it could be a valuable supplement for these patients. Further studies in subgroups of patients with HF, such as diabetic or hypertensive cardiomyopathic patients, and sensitive evaluation methods for cardiac function might provide new information in this regard.