Melatonin and the cardiovascular system in animals: systematic review and meta-analysis

Melatonin, a hormone released by the pineal gland, demonstrates several effects on the cardiovascular system. Herein, we performed a systematic review and meta-analysis to verify the effects of melatonin in an experimental model of myocardial infarction. We performed a systematic review according to PRISMA recommendations and reviewed MEDLINE, Embase, and Cochrane databases. Only articles in English were considered. A systematic review of the literature published between November 2008 and June 2019 was performed. The meta-analysis was conducted using the RevMan 5.3 program provided by the Cochrane Collaboration. In total, 858 articles were identified, of which 13 were included in this review. The main results of this study revealed that melatonin benefits the cardiovascular system by reducing infarct size, improving cardiac function according to echocardiographic and hemodynamic analyses, affords antioxidant effects, improves the rate of apoptosis, decreases lactate dehydrogenase activity, enhances biometric analyses, and improves protein levels, as analyzed by western blotting and quantitative PCR. In the meta-analysis, we observed a statistically significant decrease in infarct size (mean difference [MD], -20.37 [-23.56, -17.18]), no statistical difference in systolic pressure (MD, -1.75 [-5.47, 1.97]), a statistically significant decrease in lactate dehydrogenase in animals in the melatonin group (MD, -4.61 [-6.83, -2.40]), and a statistically significant improvement in the cardiac ejection fraction (MD, -8.12 [-9.56, -6.69]). On analyzing potential bias, we observed that most studies presented a low risk of bias; two parameters were not included in the analysis, and one parameter had a high risk of bias. Melatonin exerts several effects on the cardiovascular system and could be a useful therapeutic target to combat various cardiovascular diseases.

' BACKGROUND Melatonin (N-acetyl-5-methoxytryptamine) is a hormone produced by the pineal gland exclusively at night and is released into the bloodstream and cerebrospinal fluid in a circadian manner to regulate several physiological and neuroendocrine functions (1)(2)(3). The effects of melatonin are dependent on non-receptor-and receptor-mediated mechanisms of action. Membrane melatonin receptors (MT1, MTNR1A, MT1, and MTRN1B) are G-protein-coupled receptors, signaling through G i -G 0 or G q -G 11 transduction pathways, depending on the target organ. Melatonin secreted at night might interact with its effector and produce immediate effects when melatonin is present in the circulation (e.g., nighttime blood pressure dipping). Moreover, during the night and through several mechanisms of action, melatonin primes prospective effects (such as controlling autonomic nervous system activity) that can be observed only during the day when no pineal melatonin production occurs (3)(4)(5)(6).
Over the last 20 years, several studies have suggested that melatonin influences the cardiovascular system (7,8). Melatonin may have significant anti-inflammatory and cardioprotective properties by directly eliminating free radicals, as well as indirectly via antioxidant activity. In addition, melatonin may be involved in blood pressure regulation and have significant anti-atherogenic effects (8)(9)(10)(11)(12)(13).
In this systematic review, cardiovascular diseases such as hypertension, myocardial infarction, ischemia, and reperfusion were selected to verify the action of melatonin, as we believe that these cardiopathies currently represent a large number of cardiovascular diseases (13,14). Our study aimed ' SEARCH STRATEGIES In the present study, the search strategy was performed as described by Tawfik et al. (15). We used MEDLINE, Google Scholar, and Cochrane databases and reviewed literature published from November 2008 to June 2019; we restricted this systematic review to the last ten years, covering the latest and most relevant articles worldwide. First, we selected keywords from related articles, using Medical Subject Headings (MeSH) to identify more related keywords with similar meanings as follows: (''melatonin'') [MeSH Terms] AND (''cardiovascular system'') [MeSH Terms] [All Fields]. We then searched the three databases. Accordingly, we identified 2096 articles in PubMed using the ''other animals'' filter, 602 articles using Google Scholar filtering for keywords only in the title, and three articles using a Cochrane Library advanced search; the terms used were ''melatonin and cardiovascular system'' In addition, we reviewed retrieved articles to identify additional studies ( Figure 1). This review was conducted according to the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (16,17).
We excluded studies with cell culture experiments, as well as pre-and post-conditioning studies. The inclusion criteria were animal studies, cell culture studies, and in vivo experiments. The control group was the melatonin group in this study. The melatonin group varied in each article, as studies persistently experimented with a melatonin group related to a drug or an event.
The process of paper retrieval and titles and abstract evaluation was conducted by two independent blinded researchers capable of compiling systematic reviews (ECV and RS), following the inclusion and exclusion criteria according to the tenets of PICO (16)(17)(18)(19). The PICO was defined as patients in case the systematic review was performed in animals, interventions considering the administration of melatonin in animals using an experimental model of myocardial infarction, comparison, to compare the melatonin group with the control group receiving no melatonin, and outcome, which were results of administering melatonin. The selected articles were critically evaluated to determine their potential inclusion in the review. In the event of a disagreement between investigators regarding studies selected, a third reviewer was consulted (LCA).
In the present systematic review, data obtained from selected studies were tabulated, and the following characteristics were listed when present in the articles: authors' names, year of publication, animal type, sex (M/F), animal species, age (months), weight, induction model, and site injury (Table 1). Table 2 presents the following information: authors, sample size, number of groups, number of animals per group, melatonin administration, melatonin doses, and dependent variables. Table 3 lists the most frequent recommendations in preclinical research guidelines for in vivo animal experiments (18). Table 4 evaluates the study characteristics of selected controlled animal studies, with prior exercise and myocardial infarction as variables that showed a significant difference between the melatonin control group and the study group. These were classified as S for ''significant difference,'' and variables that did not present a significant difference were classified as NS (not significant).
RevMan (version 5.3; Cochrane Collaboration, Oxford, UK) was used to perform the meta-analysis. The randomeffects model was used to account for the heterogeneity.

Statistical analysis
Mean values and standard deviation between studies, presented as the mean difference (MD) of post-intervention values after calculating the inverse variance, were employed to verify the magnitude of the protection afforded by melatonin (19). In addition, heterogeneity was assessed using Cochran's Q and I 2 tests, followed by visual inspection of the graph. The analyses were performed using the RevMan software (version 3.3.1) (20).
In Table 5, we analyzed the study characteristics of selected controlled animal studies. Accordingly, we obtained the following results according to each experiment performed in articles examined in this systematic review. Table 5 presents experiments in which melatonin significantly improved the investigated variable (marked as S), as well as those where melatonin showed no significant improvements (NS).  (Table 5).
Furthermore, melatonin showed beneficial effects on autophagosome evaluation, lactose dehydrogenase measurements, angiotensin, and aldosterone, nitric oxide levels, and mitochondrial analysis, as determined by Zhang et al.

38.88%
Table 5 -Study characteristics of selected controlled animal studies assessing melatonin and the cardiovascular system. The meta-analysis revealed a statistically significant decrease in infarct size (MD -20.37 [-23.56, -17.18]). However, there was no statistical difference in systolic pressure between articles analyzed (MD -1.75 [-5.47, 1.97]). In articles analyzing lactate dehydrogenase, a statistically significant decrease in the levels of this enzyme was noted in animals in melatonin groups (MD -4.61 [-6.83, -2.40]). With regard to the ejection fraction, two articles showed improvement in melatonin-treated groups. Another study analyzed the influence of melatonin in infarcted animals with the same ejection fraction; however, this parameter was not statistically significant in the meta-analysis (MD -8.12 [-9.56, -6.69]) ( Figure 3).
In terms of selection bias, the results were well-balanced between low risk, no clear risk, and high risk of bias. All studies presented a low risk of bias in the baseline variable characteristics. On analyzing allocation concealment, most selected articles had a high risk, and a little less than half presented a low risk of bias. The randomization parameter was also fairly balanced between low risk, no clear risk, and high risk of bias. On analyzing random outcome assessment, most studies (more than 50%) had a low risk of bias, and some presented an unclear risk of bias. On analyzing blinding bias, most articles were unclear as to whether investigators were blinded. The articles presented a low risk of bias in the results of incomplete outcome data (Figure 4 and 5).

' DISCUSSION
This systematic review revealed that melatonin has various beneficial effects on the cardiovascular system; these effects include decreased infarct size, improved cardiac function and cellular oxidation functions, reduced apoptosis, and healthier cellular histomorphology.
Reportedly, melatonin is an important anti-apoptotic agent in various tissues, reducing calcium uptake, mitigating reactive oxygen species generation, and decreasing the levels of pro-apoptotic proteins, such as Bax (39). In addition, melatonin destabilizes hypoxia-induced hypoxia-inducible factor (HIF)-1a protein expression. Moreover, melatonin suppresses HIF-1a transcriptional activity under hypoxic conditions, resulting in vascular endothelial growth factor expression (40). Melatonin also confers anti-inflammatory effects on the cardiovascular system (41). Furthermore, a systematic review and recent meta-analysis have identified      10 Melatonin and cardiovascular system Veiga ECA et al.
The main novelty of this study is that it highlights the benefits assimilated by melatonin in experimental models of myocardial infarction, such as improved ejection fraction. Apart from limitations such as differences between animal organisms and humans, experimental research in Brazil is often restricted due to limited funding for animal studies when compared with human trials. In addition, results from animal studies fail to precisely correlate with the experience of testing melatonin or other substances in an environment that differs from the human body. Another limitation that must be considered is the nature of systematic reviews, which examine non-published data and data previously published by other authors, thus hindering novel scientific findings.

' CONCLUSION
Notably, this systematic review is based on animal experiments. Melatonin may impact the cardiovascular system, including experimental myocardial infarction, and further studies are necessary to determine its use in clinical settings for treating cardiovascular diseases.

' AUTHOR CONTRIBUTIONS
Veiga ECA contributed substantially to the study conception and design, definition of intellectual content, was involved in literature search, data analysis, statistical analysis, and manuscript preparation, drafting and critical review for important intellectual content, and approved the final manuscript version to be published. Simões RS, Caviola LL, Abreu LC and Cavalli RC were involved in data analysis and statistical analysis, manuscript drafting and critical review for important intellectual content, and approved the final manuscript version to be published. Cipolla-Neto J, Baracat EC and Soares Junior JM substantially contributed to the study conception and design, definition of intellectual content, were involved in manuscript preparation, drafting and critical review for important intellectual content, and approved the final manuscript version to be published.