Highlights
• Melatonin has analgesic and cerebral blood flow-regulating properties and is associated with treating episodic migraine.
• A network meta-analysis was conducted comparing it with other treatments
• No statistical significance was found between melatonin and other types of treatment when analyzing the frequency, duration and number of analgesics used during migraine attacks.
Introduction
Migraine is a chronic neurological disorder that affects 12 to 20% of the global population. Reference Buse, Manack, Serrano, Turkel and Lipton1 Early prevention of this condition is considered a smart strategy within public health. Despite prophylaxis guidelines for migraines issued by societies in North America, South America and Europe, only 3 to 5% of patients receive adequate preventive therapy.Reference Stewart, Shechter and Lipton 2
Currently, first-line treatments for migraine prevention include anticonvulsants, antidepressants and non-sympathomimetic beta-blockers, such as metoprolol and propranolol. Reference Jackson, Kuriyama and Kuwatsuka3 However, most of these prophylactic medications do notelicit sufficient responses and may cause undesirable side effects. Reference Eigenbrodt, Ashina and Khan4 Therefore, it is paramount to investigate new medications that are genuinely effective against migraines while producing minimalor no side effects.
Melatonin, a hormone released by the pineal gland, plays a crucial role in maintaining physiological metabolism, and there is significant evidence linking its secretion to neurological disorders. One of the functions under study is its potential ability to prevent migraines, especially considering that melatonin levels tend to be lower in patients suffering from this condition. Reference Peres, Zukerman, da Cunha Tanuri, Moreira and Cipolla-Neto5 However, the use of melatonin as prophylaxis for migraines still presents conflicting results.
Methods
Search strategy and eligibility criteria
This systematic review with meta-analysis was undertaken per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline, as depicted in Figure 1, and the method used was based on the criteria established by the Cochrane Back Review Group. Reference Page, McKenzie and Bossuyt6 The International Prospective Register of Systematic Reviews (PROSPERO) was used to register this article (protocol CRD42024548292). The databases included were Pubmed, Cochrane and Embase. Studies were identified using the search terms “(Migraine OR HeadacheMigraineDisorder OR Status Migrainosus) AND (Melatonin).”
PRISMA flow diagram of study screening and selection. The search strategy in Embase, MEDLINE and Cochrane yielded 846 studies, of which 35 were fully reviewed for inclusion and exclusion criteria. Nine studies were included in the meta-analysis.
The inclusion criteria were patients who received melatonin as a treatment for migraine and presented a positive or negative response compared to the use of a placebo or other conventional therapy. The exclusion criteria were incomplete results or those that were not directly related to the initial analysis. In this arch strategy, titles and abstracts were first independently evaluated. However, when abstracts did not contain enough information, they were selected, and then the full text was evaluated. The selected articles were chosen independently by three authors, with differences resolved by consensus. Data extraction was carried out by two authors through a complete reading of the articles and extraction of baseline characteristics and pre-specified outcomes. The results of this review showed no significant association between migraine and melatonin administration.
Statistical analysis and quality assessment
Based on the pre-specified eligibility criteria for article search, data extraction and quality assessment, three authors (F.G., B.P. and A.C.) independently selected the main characteristics of the articles included in the statistical analysis, which are presented in Table 1. Disagreements were resolved by consensus. Weighted mean differences were used to group continuous results, and heterogeneity was assessed by Cochran’s Q test and the I2 statistic. As relevant values for heterogeneity, p < 0.05 and I2 > 25% were considered. A random effect model was used for outcomes with I Reference Stewart, Shechter and Lipton2 < 25% (low heterogeneity). In grouped results with high heterogeneity, the DerSimonian and Laird random effects model was used. To perform the statistical analysis, we used the R Studio version 4.3.1 program. The Cochrane tool for assessing the risk of bias in randomized trials (RoB 2) was used to define the quality of the selected studies. Three authors (F.G., D.S. and A.C.) separately evaluated the risk of bias, with disagreements solved by consensus. Each trial received a high, low or unclear risk of bias rating across 5 domains: selection, performance, detection, attrition and reporting bias. Funnel plots of study weights vs. point estimates were used to assess for evidence of publication bias. Results were interpreted considering potential biases and limitations for our final statement with the Risk of Bias (RoB 2.0) tool from Cochrane.
Baseline characteristics of included studies
M = Melatonin; C = Comparison; SD= Standard Deviation; NA = Not Available.
Results
A total of nine studies, representing diverse geographic locations such as Iraq, Iran, Norway, Brazil and the United States, were included in this network meta-analysis. These studies collectively involved 783 participants, with sample sizes ranging from 26 to 200 patients per study. The proportion of women varied widely, accounting for 22 to 88% of the study populations. Participants’ ages also varied, with mean ages spanning from 30.95 to 41.9 years across the studies. Monthly headache frequency at baseline ranged from 6.77 to 16.7 episodes among intervention groups.
The studies compared melatonin, administered at doses ranging from 0.3 mg/kg to 3 mg per day, with interventions such as amitriptyline, topiramate, naproxen, propranolol, valproic acid and placebo. Follow-up durations ranged from 8 weeks to 4 months, allowing for comprehensive observation of outcomes over clinically relevant time frames. The key outcomes assessed included monthly headache frequency, MIDAS scores, analgesic use and headache duration in hours. Below, we summarize the findings for each outcome.
Monthly headache frequency
The analysis of monthly headache frequency incorporated data from all nine included studies and seven different interventions, as depicted in Figure 2. Baseline frequencies ranged from 6.77 to 16.7 headaches per month across intervention groups, reflecting the heterogeneity of migraine severity in the study populations. The network meta-analysis revealed no statistically significant differences between melatonin and its comparators. The mean difference compared to amitriptyline, for instance, was −1.8 (95% CrI [−5.2, 1.0]), while the comparison with placebo yielded a mean difference of 0.49 (95% CrI [−1.6, 2.7]). Other comparators, such as naproxen (−0.98; 95% CrI [−5.5, 3.8]) and valproic acid (−0.60; 95% CrI [−5.0, 3.6]), also exhibited overlapping confidence intervals, indicating comparable efficacy across these treatments, as noted in Figure 3.
Network graph of migraine treatments compared to melatonin.
Comparison of monthly headache frequency and different interventions.
The rankogram analysis, in Figure 4, which evaluates the probability of each treatment being the most effective, demonstrated that amitriptyline had the highest likelihood of ranking as the best intervention for reducing monthly headache frequency. Melatonin showed an intermediate probability, surpassing comparators such as naproxen, placebo, propranolol, topiramate and valproic acid. Despite this favorable ranking, the differences between treatments were not statistically significant, highlighting the need for further research to validate these trends.
Rankogram for monthly headache frequency.
Although the analysis did not identify statistically significant differences between treatments, the probabilistic ranking analysis (Figure 4) revealed that melatonin holds an intermediate position in terms of efficacy. Notably, it consistently ranked above naproxen, placebo, propranolol, topiramate and valproic acid. While these differences did not reach statistical significance, the visual trend observed in the rankogram suggests that melatonin may offer clinically meaningful benefits, particularly for patients seeking alternatives to conventional therapies. This intermediate performance underscores the importance of further exploring melatonin’s therapeutic role, especially considering its favorable safety profile and tolerability.
These results suggest that while melatonin may offer some potential for reducing headache frequency, its efficacy is not superior to standard treatments, reinforcing the importance of individualized therapy in managing episodic migraines. As studies provided different times of follow-up, ranging from two to four months, we decided to make subanalyses, as depicted in Figures 5, 6 and 7. All of these have shown similar results, being non-statistically significant, with some of these clusters not even having melatonin in the comparison.
Monthly headache frequency among arms with 2 months of treatment.
Monthly headache frequency among arms with 3 months of treatment.
Monthly headache frequency among arms with 4 months of treatment.
MIDAS score
The analysis of MIDAS scores (migraine disability assessment) similarly revealed no significant differences (Figure 8). When compared with placebo, melatonin had a mean difference of 2.6 (95% CrI [−15.0, 11.0]). Comparisons with other therapies yielded results such as −15.0 (95% CrI [−34.0, 6.5]) for amitriptyline, 19.0 (95% CrI [−6.4, 46.0]) for propranolol and 0.24 (95% CrI [−20.0, 23.0]) for topiramate. These results suggest that melatonin is comparable to other interventions in terms of reducing migraine-related disability.
Comparison of MIDAS score and different interventions.
Analgesic use
The mean number of analgesics used was another key outcome evaluated. The results indicated no significant advantage for melatonin over the comparators. For instance, compared to naproxen, the mean difference was 1.5 (95% CrI [−1.1, 4.0]), while the comparison with placebo resulted in a mean difference of 1.4 (95% CrI [−0.071, 2.8]). These findings suggest that melatonin does not significantly reduce the reliance on analgesic medications among patients with episodic migraine (Figure 9).
Comparison of analgesic use and different interventions.
Headache duration
For headache duration, expressed in hours, the comparisons also failed to show statistically significant differences. Melatonin had a mean difference of 1.4 hours (95% CrI [0.025, 2.8]) compared to placebo and −0.62 hours (95% CrI [−2.7, 1.6]) compared to valproic acid. The confidence intervals for these comparisons encompass values close to zero, indicating no meaningful clinical differences in this outcome (Figure 10).
Comparison of headache duration and different interventions.
Heterogeneity and quality of evidence
The heterogeneity across studies was assessed using the I2 statistic, with results indicating low to moderate heterogeneity for most outcomes. This suggests that the included studies were reasonably consistent in their findings. Quality assessment using the Cochrane RoB 2 tool revealed that most studies had a low risk of bias in selection and reporting domains. However, some studies showed an unclear risk in performance and detection bias, which may have influenced the reliability of the results. This happened especially in the performance and detection domains, in which authors identified subjects such as data being reported in different manners across different patient groups. Hence, if different patients from the same study had their results measured at different times or with different tools, the results could have been influenced toward the opposite direction of true results (Figure 11).
Summary of Findings
Overall, the results of this meta-analysis did not reveal statistically significant differences between melatonin and other interventions in reducing migraine-related outcomes. While melatonin did not demonstrate superior efficacy, its favorable safety profile and low incidence of adverse effects make it an attractive option for further investigation, particularly for patients who are unable to tolerate or respond to standard treatments.
These findings underscore the need for additional high-quality, randomized controlled trials to explore the potential mechanisms of melatonin in migraine prevention. Future studies should focus on optimizing dosing regimens, evaluating patient subpopulations and investigating possible synergistic effects with other therapies. Such research could help elucidate the specific role of melatonin in migraine management and its broader implications for neurological and pain disorders.
Discussion
This systematic review and network meta-analysis evaluated the efficacy of melatonin as a preventive treatment for episodic migraines, comparing it to placebo and other standard treatments, including amitriptyline, naproxen, valproic acid, topiramate and propranolol. The analysis incorporated data from nine randomized clinical trials involving 783 patients, with outcomes such as monthly headache frequency, MIDAS scores, analgesic use and headache duration assessed across treatment arms.
The results consistently showed no statistically significant differences between melatonin and other interventions across all evaluated outcomes. For example, in terms of monthly headache frequency, melatonin demonstrated mean differences compared to amitriptyline (−1.8; 95% CrI [−5.2, 1.0]) and placebo (0.49; 95% CrI [−1.6, 2.7]), with similarly non-significant findings for the other comparators. Likewise, no significant differences were observed in MIDAS scores (e.g., compared to placebo: 2.6; 95% CrI [−15.0, 11.0]), analgesic use (e.g., compared to naproxen: 1.5; 95% CrI [−1.1, 4.0]) or headache duration (e.g., compared to placebo: 1.4; 95% CrI [0.025, 2.8]).
These findings challenge the presumed efficacy of melatonin as a robust migraine prophylactic. Despite the lack of significant results, melatonin’s pharmacological profile – characterized by its neuromodulatory, analgesic and cerebrovascular effects – warrants further investigation. Potential factors such as variations in dosage, timing of administration, patient subpopulations and underlying migraine pathophysiology may influence its effectiveness. Notably, while this analysis could not substantiate melatonin’s role in reducing migraine burden, its favorable safety profile remains a compelling reason for further exploration, particularly in patients unresponsive to conventional therapies.
The limitations of this study must be acknowledged. The small number of included trials, heterogeneity in study designs and variability in reported outcomes may contribute to the inconclusive findings. Studies did not provide information such as the timing patients took melatonin. Knowing that this product is already used as a sleep aid and that it does interfere with the circadiancycle, one could say that the way patients use it, ifqpmorqpm, would be relevant data. Additionally, the confidence intervals for several comparisons were wide, reflecting uncertainties in effect estimates. Lastly, the potential for publication bias and the limited representation of diverse populations further highlight the need for more rigorous and well-powered studies to clarify melatonin’s role in migraine management.
Conclusion
In this systematic review and network meta-analysis, no significant differences were observed in reducing monthly headache frequency, MIDAS scores, analgesic use or headache duration between melatonin treatment and other commonly used migraine treatments, including amitriptyline, naproxen, valproic acid, topiramate, propranolol and placebo. These findings challenge the expected efficacy of melatonin in episodic migraine prevention. However, given melatonin’s known neuromodulatory and cerebral blood flow-regulating properties, further studies are warranted to investigate its influence on migraine pathophysiology and explore potential mechanisms that may benefit patients unresponsive to standard therapies. Thus, physicians should wait for more robust and positive evidence of melatonin use for headache if deciding to pursue its use in clinical practice.
Risk of bias.
Authors contribution
F.A.G: Writing – review &editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. D.A.S: Writing – original draft, Data curation. B.R.P: Writing – original draft, Data curation. L.F.V: Writing – original draft, Data curation; C.L.N.C: Writing – original draft, Data curation; A.C.C: Writing – original draft, Data curation
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Funding statements
This project has not received any type of funding.
Target article
Melatonin Compared to Other Treatments for Episodic Migraine: A Systematic Review and Network Meta-Analysis
Related commentaries (1)
Reviewer Comment on Araujo Gouhie et al. “Melatonin Compared To Other Treatments For Episodic Migraine: A Systematic Review and Network Meta-Analysis”