Melatonin as a treatment for migraine sufferers: a systematic review – The Egyptian Journal of Neurology, Psychiatry and Neurosurgery

This study was conducted in January 2022 using a search from Pubmed according to the PRISMA protocol. The search strategy was carried out from 2010 to 2022 using 2 keywords, “migraine” AND “melatonin”. This study aims to find research based on randomized clinical trials that examine the use of melatonin as a therapy in migraine sufferers, including as prophylaxis.

The inclusion criteria were an English-language study, a randomized clinical trial, and a complete manuscript. Articles in the form of case reports, descriptive studies, cross-sectional studies, cohort studies and case–control studies were excluded. Then the titles and abstracts were screened manually according to the eligibility criteria (Table 1).

In this study, through our search strategy, we found 142 articles, then 130 were excluded because they were not randomized-controlled trials. Of the 12 articles, further screening was carried out, namely by year, where this study used studies from 2010 to 2022 and excluded 4 articles. A total of 8 articles were attained and screened further either by title, content and availability of complete manuscripts. A total of 1 article was found to assess rTMS (repetitive transcranial magnetic stimulation) for migraine and did not assess melatonin in that study, 1 article was not available in full text, and 1 article was conducted on people with epilepsy not migraine sufferers, therefore these three articles were excluded so that in the end 5 articles were collected that matched with the inclusion and exclusion criteria of researchers (Fig. 1).

Two authors (IMAO and CT) screened the titles and abstracts of all studies that met the eligibility criteria, then proceeded to read and screen the full manuscripts of these studies. The discrepancy between the two authors will be resolved by the third and fourth author.

Quality assessment for randomized-clinical trial study is the JADAD scale. A score of 0–2 is considered low quality, while a score of 3–5 is of good quality. From the 5 studies that we got, in general, the quality of the studies is quite good. Most of the studies were randomized, blinded and there was an explanation for participants dropping out (Table 2).

The studies assessed were conducted in Norway, Iran, Brazil and 2 studies in the USA. The number of participants ranged from 26 to 178 participants with a range of study periods varying from 2 to 4 months. The age of the participants in 2 studies was teenagers while the other 3 studies were over 30 years old (Table 3). 2 studies compared melatonin with placebo in migraine sufferers, 2 in addition to comparing melatonin with placebo, also compared with other drugs, while 1 study compared melatonin doses without the use of other drugs or a placebo. In most studies, a melatonin dose of 3 mg was used (Table 4).

Results

Based on our search strategy, the first study was conducted by Alstadhaug et al. in 2010 with randomization, double-blind and crossover. There were 2 groups, namely the group with 2 mg melatonin and the placebo group. Melatonin extended-release (Circadin) 2 mg given 1 h before bedtime. Then after being evaluated in 8 weeks, each group will be washed out. After 6 weeks of washout period, crossover will be carried out. There are 48 total participants with 24 participants in each group, and 2 participants dropped out. There was a decrease in the frequency of migraine attacks in the two groups, 1.4 and 1.3 in the melatonin and placebo groups, respectively, but not statistically significant (p = 0.75). In evaluating sleep quality using the Pittsburgh Sleep Quality Index (PSQI), there were significant differences in participants with baseline PSQI > 6, in the melatonin group 6.8 ± 4.0 and 9.4 ± 4.0 in the placebo group (p < 0.05). However, melatonin did not make a difference in PSQI scores compared to placebo, with the mean PSQI being 4.7 ± 3.2 and 5.6 ± 4.1, respectively (p = 0.09). In the study there were several side effects such as nightmares, nervousness, night sweats, fatigue, and so on but no further analysis was conducted [31].

In 2016, Gonçalves et al., conducted a randomized, double-blind study using placebo as a control and comparing melatonin 3 mg and amitriptyline 25 mg which are often used to treat migraine. Of the 196 participants, it turned out that there were 18 participants who lost to follow-up, so that the total participants analyzed were 178 participants who were followed for 3 months. 60 participants received melatonin, while 59 participants received a placebo and 59 participants received amitriptyline. The main result of this study was that the melatonin and amitriptyline group had fewer migraine attacks than the placebo group and a statistically significant difference with a difference of 2.7 days more in the frequency of migraine attacks on the placebo, compared to the melatonin group and 2.2 days more on the frequency of migraine attacks on the placebo, compared with the amitriptyline group. In addition, Gonçalves et al. also analyzed the mean intensity of attacks, the mean duration of attacks and the amount of analgesics used by the participants. The mean attack intensity in the melatonin group was 3.6 ± 3.5 less than the placebo (4.8 ± 3.3) with differences from the baseline data, 3.5 and 1.8, respectively. In the melatonin group, there was a significant difference in the mean duration of attacks, namely 10.9 ± 9.5 compared to placebo 16.2 ± 15.3. In this study there were some side effects, but no serious side effects. Side effects in the melatonin and placebo groups were almost similar (p > 0.05) but there were more side effects in the amitriptyline group when compared to the placebo and melatonin groups (p > 0.03). The majority of side effects were drowsiness, in addition to epigastralgia and constipation in small amounts, especially in the melatonin group [1].

The next randomized study found in this search was the study conducted in Iran by Ebrahimi-Monfared et al. in 2017 in a double-blind and placebo as a control. In addition to melatonin 3 mg, Ebrahimi-Monfared et al. also analyzed the administration of valproate 200 mg. Each group consisted of 35 participants and was matched by age, sex, and clinical characteristics. Then participants will be evaluated twice, at the end of the first and second months. The mean age of the participants was 38.9 ± 9.2 years of which 51.42% were women. Based on this study, the majority of migraine experienced by participants occurred in the morning with the most characteristic being throbbing. About 65.57% were triggered by stress and pain complaints improved with sleep (70.47%). In the post hoc test analysis, significant results were obtained in terms of attack frequency, duration, severity, and MIDAS (Migraine Disability Assessment Score). In the melatonin group, the frequency of attacks was lower (2.5 ± 1.3) when compared to the placebo group (3.8 ± 1.1) with a difference of 1.7 and 0.4 from the baseline, respectively. In terms of duration and severity of attacks, the placebo group was much higher than the melatonin group (14.1 ± 8.1, 6.0 ± 3.2 and 8.7 ± 12.4, 3.5 ± 2.6). Changes in MIDAS were also found in the melatonin and valproic acid groups with p = 0.005 and p = 0.001, respectively. This change in MIDAS was not found in the placebo group (p = 0.44). In the melatonin group there were 2 participants who experienced side effects, fatigue and confusion, but only 1 participant experienced side effects on the placebo, namely dry mouth. However, there are no analytical data for side effect outcomes in this study [30].

The study conducted by Gelfand et al. was a randomized, double-blind study comparing melatonin and placebo. This study consisted of 31 participants, but 5 participants experienced fewer migraine attacks so that only 26 participants were continued in this study. The first group consisted of 13 people and was given melatonin 3 mg, while the second group consisted of 13 people and was given a placebo. The baseline characteristics in the two groups were almost the same except that headache-related disability was higher in the melatonin group. Headache-related disability was calculated using the Pediatric Migraine Disability Assessment scoring. Of the 26 participants, there were 3 participants who withdrew (2 people from the first group, and 1 person from the second group). The mean migraine attack day by day was 3.6 ± 0.9 in the melatonin group compared with 4.9 ± 1.7 in the placebo group. These results suggest that migraine attacks were fewer in the melatonin group, but not statistically significant (difference 1.3 with 95% KI − 5.1 to 2.6). No serious side effects were found on giving melatonin 3 mg, namely the effect of daytime fatigue in 2 people in the first group [29].

Then there was a more recent randomized study conducted in 2020. However, this study did not compare melatonin with a placebo or other drugs. This study compared high doses of melatonin (4–8 mg) with low doses (1–2 mg). A total of 84 participants were randomized by stratification based on body weight, namely body weight above equal to 40 kg and body weight less than 40 kg. Because the average age of the participants was 11.8 years, and there were children, the melatonin in this study used milk chocolate-based melatonin. In addition to being edible, this formula has a faster onset of action than tablets, so it is hoped that headache relief will occur more quickly. This study was conducted by giving headache diaries to participants so it is a home-based trial. Participants will be observed every 2 weeks with pain improvement if there is a change in the scale ≥ 3 cm. A total of 38 participants were excluded (loss to follow-up = 10; withdrew = 10 in the low-dose melatonin group and loss to follow-up = 12; withdrew = 6 in the high-dose melatonin group). The basic characteristics in both groups were almost the same, both for pain intensity, photophobia, phonophobia, nausea and vomiting. At 2 h of melatonin administration, there was a change in pain intensity of 2.3 ± 2.1 in the low-dose group, and 2.7 ± 2.1 in the high-dose group, with a p value of 0.581. In these two groups there was only a difference of 0.4 cm with a 95% KI value: − 1.17 to 1.92 which was not statistically significant, so there was no significant difference between the administration of low doses of melatonin and high doses. This study also evaluated side effects with the majority of participants sleeping after 2 h of high-dose melatonin (81%) and 68% in the low-dose group (p value = 0.352) [22].

Of the 5 studies, only 2 studies state that melatonin is better for migraine treatment than placebo. In both studies, a dose of melatonin of 3 mg was used and the results obtained that melatonin was superior to placebo in terms of attack frequency, attack duration, intensity and severity of attacks. In addition, based on a study conducted by Ebrahimi-Monfared et al. there was a significant decrease in MIDAS scores with the use of 3 mg melatonin. In addition, in 2 studies comparing melatonin with other drugs, melatonin did not give significantly better results than valproic acid and amitriptyline.

Of the 5 studies, only 3 studies discussed the side effects of giving melatonin and placebo. Side effects that occur vary from drowsiness, nausea, pruritus, dizziness, weight gain and constipation with the majority of side effects being drowsiness. However, only 1 study (Gelfand et al.) analyzed the incidence of drowsiness, but found no significant results (p = 0.209) (Table 4). General view of this systematic review can be seen in Fig. 2.

Brief results of melatonin as a treatment for migraine sufferers