Exogenous melatonin's effect on salivary cortisol and amylase: A randomized controlled trial

Abstract

This study aimed to examine the effect of acute exogenous melatonin administration on salivary cortisol and alpha‐amylase (sCort and sAA) as representatives of the HPA axis and the sympathetic nervous system, respectively. A single‐dose prolonged‐release melatonin (2 mg) or a placebo tablet was given to healthy volunteers (n = 64) at 20:00 h in a crossover design. The saliva was collected at six time points (20:00, 21:00, awakening, 30 min after awakening, 10:00, and 12:00 h) and was measured for sCort, sAA, and salivary melatonin (sMT) levels. Pulse rates and sleep parameters were also collected. Melatonin was effective in improving sleep onset latency by 7:04 min (p = .037) and increasing total sleep time by 24 min (p = .006). Participants with poor baseline sleep quality responded more strongly to melatonin than participants with normal baseline sleep quality as they reported more satisfaction in having adequate sleep (p = .017). Melatonin administration resulted in higher sCort levels at awakening time point (p = .023) and a tendency of lower sAA levels but these were not significant. Melatonin ingestion at 20:00 h resulted in a marked increase in sMT levels at 21:00 h and remained higher than baseline up to at least 10:00 h (p < .001). Melatonin increases sCort levels at certain time point with a tendency to lower sAA levels. These opposing effects of melatonin suggested a complex interplay between melatonin and these biomarkers. Also, the results confirmed the positive acute effect of a single‐dose melatonin on sleep quality.

The effect of exogenous melatonin on salivary cortisol and amylase across six time points.

Abbreviations

HPA

hypothalamic–pituitary–adrenal

PR

pulse rate

PSQI

Pittsburgh Quality Sleep Index

REM

rapid eye movement

sAA

salivary alpha‐amylase

sCort

salivary cortisol

sMT

salivary melatonin

VAS

visual analog scale

W

awakening time point

W + 30

30 min after awakening time point

1. INTRODUCTION

Melatonin is a circadian hormone released at night by the pineal gland in response to the light/dark cycle to help regulate sleep. ¹ It is well known that melatonin has a multitude of actions beyond sleep regulation. Melatonin is present in basically all kinds of living organisms including bacteria and plants. It is also released locally and acts as a local regulator in various tissues and in fact, some levels surpassed that released by the pineal gland. ² The diverse physiological effects of melatonin include providing temporal cues to adjust the circadian rhythm of various tissue organs, regulation of autonomic cardiovascular function, body temperature, immune system, retinal function, and gastrointestinal tract function. ³ Melatonin regulates numerous tissue functions due to its many modes of mechanisms of action. It can exert effects as a direct free radical scavenger, an activator of antioxidative mechanism, or through membrane‐ as well as nuclear‐receptor‐mediated mechanisms. ⁴ Due to its pleiotropic properties, it has been shown to possess beneficial effects in a variety of conditions including autoimmune diseases, neurological diseases (such as Alzheimer's disease, stroke, depression, and anxiety), liver and pancreatic diseases, pain, cancer, and even COVID‐19. ⁵ , ⁶ , ⁷ , ⁸ , ⁹

Cortisol and alpha‐amylase are widely studied biomarkers representing the hypothalamic–pituitary–adrenal (HPA) axis and the autonomic nervous system, respectively. ¹⁰ They also exhibit a diurnal pattern of release throughout the day. Cortisol levels are high upon awakening and remain low as the day progresses while alpha‐amylase levels continuously increase as the day advances. ¹¹ Under stressful conditions, both of these markers increased. They are the most studied biomarkers in the field of stress research since their levels have been reliably shown to be associated with stress; also, they can be conveniently detected in the saliva especially with real‐time measurement with alpha‐amylase. ¹⁰ , ¹²

There are many studies exploring the relationship between melatonin and cortisol. Melatonin has the ability to regulate the HPA axis. ¹³ While cortisol is also responsible for arousal and alertness ¹⁴ since it was found that sleep deprivation was associated with increased levels of cortisol which might indicate a hyperarousal state. ¹⁵ , ¹⁶ It is speculated that melatonin and cortisol mutually interact and affect each other. ¹⁷ However, the effect of melatonin on cortisol rhythm and production is still inconclusive as previous studies reported melatonin to either increase, decrease, or have no effect on cortisol levels. Zisapel et al. ¹⁸ reported that administration of exogenous melatonin for a week delayed the endogenous cortisol rhythm in the elderly with insomnia. While an experiment on fibromyalgia patients found 9 mg of melatonin administration for 10 days reduced urinary cortisol levels. ¹⁹ Cagnacci et al. ²⁰ found that high‐dose melatonin (100 mg) administered in the morning enhanced serum cortisol levels in postmenopausal women but not in younger women. Other studies found melatonin acute infusion or ingestion for at least 3 weeks to have no significant effect on cortisol secretion. ²¹ , ²² , ²³ The mechanism by which melatonin might affect cortisol production is still unclear.

As for alpha‐amylase and melatonin, there are even fewer studies. Previous studies have reported on the association of sleep quality on alpha‐amylase levels. Since poor sleep quality is a stressful condition, it could affect alpha‐amylase response. It was found that salivary alpha‐amylase (sAA) levels were higher in winter compared with summer period. ²⁴ This study also found a weak positive correlation between seasonal differences in sAA and melatonin secretion. Another study did not find sAA to be sensitive to everyday sleep variation but seemed to correlate more with prolonged high stress as they showed lower sAA awakening response and steeper diurnal slope. ²⁵ However, there is no report on the effect of exogenous melatonin on sAA levels. Obviously, there are still many unanswered questions regarding the relationship among melatonin, cortisol, and alpha‐amylase.

Melatonin may impact both biomarkers on many levels apart from acting as a circadian pacemaker. Since melatonin exerts antioxidant, anti‐inflammatory, and immunoneuroendocrine effects, it can be speculated that melatonin may be able to reduce the level of stress hormone. It is known that stress could lead to poor sleep. We would like to explore if improving sleep quality by administering melatonin could help reduce the levels of stress biomarkers. The aim of this study was to explore the immediate effect of a single‐dose exogenous prolonged‐release melatonin on sleep parameters in association with sCort and sAA as representatives of the stress response. This is a crossover experiment assigning the healthy volunteers both a melatonin pill and a placebo pill. The sleep parameters were questioned. The saliva was collected at six time points to measure the levels of salivary stress biomarkers. This study should add to the body of knowledge on the effect of melatonin on the biomarkers' response and if it can be used to modulate stress.

2. MATERIALS AND METHODS

2.1. Subjects

The participants were healthy volunteers who fulfilled the inclusion criteria: at least 18 years old, no underlying disease, not taking any medications including oral contraception, not allergic to melatonin or lactose, not pregnant, not breastfeeding, and not having menstruation during the experiment. The exclusion criteria included melatonin administration within 24 h, alcohol drinking within 12 h, and food consumption within 1 h of saliva collection.

2.2. Procedures

The study protocol was approved by Mahidol University multi‐faculty cooperative IRB review MU‐MOU COA 2022/018.2302. The trial was registered on Thaiclinicaltrials.org with the registration number TCTR20220310009. The participants were informed about the details of the study before signing a consent form. The participants answered the questionnaires on basic personal information, information regarding sleep pattern, and Pittsburgh Quality Sleep Index (PSQI) before starting the experiment. This is a crossover study where the participants served both as control and melatonin groups. The sequence of treatment was random and blinded to the participants. The participants were given either a prolonged‐release melatonin 2 mg (Circadin®) tablet or a placebo tablet. The placebo tablets were manufactured by the Faculty of Pharmacy, Mahidol University which looked identical to the melatonin tablets in terms of size, shape, and color. The instruction was that they should choose two similar days in terms of activity and routine on the experimental days. They have to record their pulse rates using a pulse oximeter at every saliva collection time which was at 20:00 h, 21:00 h, awakening, 30 min after awakening, 10:00 h, and 12:00 h. The drug was taken at 20:00 h right after saliva collection. During the experiment, the participants recorded their sleep log and assessed their sleep quality with a visual analog scale (VAS) which ranged from 0 to 10 with 10 meaning a very sound sleep. There were several related questions to confirm the validity of the data, for example, if the information regarding onset or duration do not match with the time to sleep or time to wake up, we would check back with the participants immediately. The data were collected between April–September 2022. The study was conducted according to the principles of the Declaration of Helsinki.

2.3. Saliva collection and analysis

The participants were instructed to pool the whole saliva and passively drooled into a 2‐mL plastic tube. Each tube was pre‐labeled to guide the timing of saliva collection. Then the saliva was kept at −20°C and returned to the researcher team the next day or as soon as possible where saliva was kept at −80°C until further analysis. To measure the biomarker levels, the saliva samples were thawed and centrifuged at 1500g for 15 min prior to each measurement. The levels of sAA were measured with a portable amylase biosensor (Nipro Co, Osaka, Japan) as described in detail previously. ¹² The biosensor can measure sAA levels up to 200 U/mL with 10.2% coefficient of variation. The levels of sCort were measured using a competitive enzyme immunoassay kit (Salimetrics, State College, PA, USA) according to the manufacturer's protocol. The sensitivity of the kit can measure between 0.007 and 3 μg/dL sCort. The intra‐ and inter‐assay coefficients of variation were 7% and 11%, respectively. The levels of salivary melatonin (sMT) were measured with a competitive enzyme immunoassay kit according to the protocol (Salimetrics). The range of measurement was 1.37–50 pg/mL. The intra‐ and inter‐assay coefficients of variation were 7.4% and 15.6%, respectively.

2.4. Statistical analysis

SPSS statistics program version 18.0 was used to analyze the data with the statistical significance level set at p < .05 (IBM, Armonk, NY, USA). All results were presented as means ± SD unless stated otherwise. Data were tested for normal distribution and homogeneity of variance using a Kolmogorov–Smirnov and Levene's test before statistical procedures were performed. Related samples Wilcoxon Signed‐Rank test was used to compare all sleep parameters between placebo and melatonin groups except for the feeling of having enough sleep and side effects where related samples McNemar test was used. A correlation study was performed with Spearman's test.

3. RESULTS

Upon 75 recruited participants, a total of 64 people completed the study. This is a crossover design experiment where all participants were given prolonged‐release melatonin (2 mg) and a placebo pill in a random sequence. The participants chose 2 days with similar routine and mood tone to closely reflect a similar condition between melatonin and placebo days. The timing between each treatment was at least 1 day apart which should be enough to wash out the effect of melatonin as its elimination half‐life was only about 2 h. ²⁶ Moreover, only 11 from 64 participants did the experiment 1 day apart and in random order, so time spacing should not affect the current results. The flow of the experiment is shown in Figure 1. The baseline demographic data are shown in Table 1. The average age was 41.11 years old with an equal number of male and female participants. Half of the participants had PSQI >5 which indicated poor sleep quality. ²⁷ The characteristics of participants with PSQI >5 and PSQI <5 were not different in terms of sex, age, and caffeine consumption.

FIGURE 1.

Flow diagram of sample allocation.

TABLE 1.

Characteristics of the participants (mean ± SD; n = 64).

Age (range)	41.11 ± 13.05 (19–75)
Female	32 (50%)
BMI	23.64 ± 3.93 kg/m2
PSQI (range)	5.66 ± 2.63 (0–11)

3.1. Sleep quality

The quality of sleep and sleep parameters are shown in Table 2. The average time to sleep was at 23:02 h in the placebo group. When the participants took melatonin, this time was significantly earlier (22:45 h; p = .031). Sleep onset latency was also shorter upon melatonin consumption (7:04 min faster compared with placebo; p = .037). However, awakening time, the number of times awake per night, the feeling of having enough sleep, or VAS were not different between the two groups. Still with earlier sleep time and less sleep onset latency, these resulted in longer total sleep time of about 30 min in the melatonin group (p = .006).

TABLE 2.

Sleep pattern (mean ± SD; n = 64).

	Placebo	Melatonin	p‐values
Time to bed	23:02 ± 1:18 h	22:45 ± 1.05 h	.031
Sleep onset latency (min)	28.06 ± 27.46	21.02 ± 17.45	.037
Time awake	6:15 ± 1.06 h	6:13 ± 1.11 h	.78
Awake latency (min)	8.66 ± 11.95	9.83 ± 13.45	.574
Number of awakening (times per night)	0.94 ± 0.89	0.86 ± 0.83	.545
Feeling of having enough sleep (% yes)	78.13%	85.94%	.092
Total sleep time (h)	6.57 ± 1.27	6.97 ± 1.15	.006
VAS	7.59 ± 1.71	7.89 ± 1.35	.185
Side effects from the drug (% yes)	12.5%	20.31%	.344

Age was negatively correlated with time to bed (r = − 0.438, p < .000) and time awake (r = − 0.433, p < .000), and was positively correlated with the number of awakening after sleep onset (r = 0.353, p = .004). That is, older participants went to sleep and woke up earlier and reported more awakenings during the night than younger participants. However, age did not affect the response to melatonin supplement, that is, older and younger participants responded similarly to melatonin (data not shown).

Participants with poor baseline sleep quality (PSQI >5) reported 14.59 ± 6.30 min faster sleep onset latency upon melatonin administration while participants with normal baseline sleep quality (PSQI <5) reported no difference in sleep onset latency (0.50 ± 2.12 min); however, the difference between these groups was not significant (p = .063). Nevertheless, participants with poor baseline sleep quality did report more satisfaction with melatonin administration as they reported to have adequate sleep more than participants with normal baseline sleep quality (p = .017; Table S1). PSQI was also correlated with total sleep time (r = − 0.359, p = .004) and VAS (r = − 0.385, p = .002) as participants with higher PSQI reported less total sleep time and less VAS.

Thirteen people reported side effects from taking melatonin supplement as shown in Table 3. Some reported more than one symptom. The most common side effect was sleepiness. On the other hand, eight participants reported side effects from taking a placebo pill. All symptoms were mild and transient. There was no significant difference on side effects between melatonin and placebo groups.

TABLE 3.

Side effects.

	Placebo (n = 8)	Melatonin (n = 13)
Poor sleep	2	1
Dizziness	2	4
Sleepy	1	6
Dry mouth	1	1
Awakening at night	1	1
Palpitation	0	1
Feverish	0	1
Dream	3	0
Headache	1	0

3.2. Salivary markers

The effects of exogenous melatonin on sCort and sAA levels are shown in Figure 2 (and Table S2). The levels of sCort were not different between melatonin and placebo groups at all time points except upon awakening when melatonin supplement resulted in significantly higher sCort levels compared with the placebo group (p = .023). Participants with PSQI >5 had higher sCort levels at 10:00 h time point with exogenous melatonin compared with placebo group (p = .032; Table S3). Age had no effect on sCort levels at all time points.

FIGURE 2.

The levels of salivary cortisol and alpha‐amylase in placebo and melatonin groups at 20:00, 21:00, awakening time, 30 min after awakening, 10:00, and 12:00 h (n = 64). Data are presented as mean ± SEM. *p < .05 compared with placebo.

The levels of sAA were not significantly different between melatonin and placebo groups, even though the melatonin group tended to have lower sAA levels at all time points examined. Age and PSQI did not affect sAA levels in response to melatonin administration (Table S4, S5).

The levels of sMT are shown in Table 4. Consumption of melatonin supplement significantly increased sMT levels at all time points examined (21:00, W, and 10:00 h) compared with placebo group (all p < .001). The highest level of sMT was observed at 21:00 h, 1 h after ingestion of melatonin. While in the placebo group, the highest level of sMT was observed at awakening time point. At 10:00 h, the levels of sMT dropped down close to before melatonin ingestion at 20:00 h but the melatonin group still showed higher sMT level than in the placebo group suggesting the persistence of melatonin's effect at this time point. The technical problem we encountered was that the levels of melatonin increased in the thousand pg/mL range which is beyond the range of the kit even after much dilution. This could result in some errors in the actual melatonin levels and are the limitation of the current study.

TABLE 4.

Salivary melatonin levels (pg/mL).

	Median	Interquartile range	p‐values
Melatonin group
20:00 h (n = 61)	8.20	3.70–16.56
21:00 h (n = 60)	890.85	412.88–2407.75
Awakening (n = 58)	63.04	34.24–103.99
10:00 h (n = 54)	19.36	8.00–32.00
Placebo group
20:00 h (n = 63)	7.37	3.05–15.46	.827*
21:00 h (n = 63)	10.34	5.86–21.42	< .001
Awakening (n = 63)	23.92	12.80–37.44	< .001
10:00 h (n = 51)	3.82	0.20–12.39	< .001

^*Compared with melatonin group.

Participants with poor or normal baseline sleep quality had similar baseline melatonin levels (at 20:00 h before melatonin consumption). After melatonin consumption, participants with poor baseline sleep quality (PSQI >5) had higher increase in melatonin levels compared with participants with normal sleep quality at all time points examined as shown in Table 5.

TABLE 5.

Salivary melatonin difference between placebo and melatonin administration (pg/mL).

	Median	Interquartile range	p‐values
PSQI <5
20:00 h (n = 32)	0	−6.78–10.47
21:00 h (n = 30)	770.39	292.97–1235.72
Awakening (n = 29)	34.43	8.57–54.98
10:00 h (n = 28)	6.47	−1.41‐18.74
PSQI >5
20:00 h (n = 32)	0	−3.97–7.14	.830*
21:00 h (n = 31)	1305.46	421.23–4451.73	.028
Awakening (n = 30)	60.74	18.32–92.60	.044
10:00 h (n = 26)	17.28	4.36–26.79	.045

^*Compared with PSQI <5 group.

For pulse rate (PR), it shows diurnal variation with the highest levels at 20:00 h and a dip afterward with a trough at awakening then it increased during the day (Figure 3). PRs were not different upon melatonin ingestion. The quality of sleep (as determined with PSQI) had no effect on PR. However, among participants older than 51 years, the PR at noon was significantly lower upon melatonin ingestion (5.42 ± 7.19 bpm less than when placebo was taken) compared with younger participants that showed no change in PR after melatonin ingestion (0.07 ± 8.77 bpm difference from placebo ingestion; p = .011). Also, see Table S6.

FIGURE 3.

Pulse rates in placebo and melatonin groups at 20:00, 21:00, awakening time, 30 min after awakening, 10:00, and 12:00 h (n = 64). Data are presented as mean ± SEM.

4. DISCUSSION

This study shows for the first time the acute effect of exogenous prolonged‐release melatonin on sCort and sAA levels. The sleep quality, sMT levels, and PR were also examined to confirm the existing literature on melatonin administration.

The results showed that melatonin facilitated faster sleep time, and faster sleep onset latency leading to longer total sleep time. It had no significant effect on awakening time, feeling of having enough sleep, and VAS. Recent meta‐analyses have concluded significant beneficial effects of both short‐ and long‐term melatonin on sleep quality, sleep onset latency, and total sleep time without significant side effects. ²⁸ , ²⁹ , ³⁰ However, most of the clinical trials on melatonin lasted at least a week; the effect of a single‐dose melatonin was not widely explored. We found a previous study on a one‐time 1 mg melatonin in healthy volunteers to increase sleep time, sleep efficiency, and REM sleep latency. ³¹ These results suggested that the beneficial impact of melatonin on sleep parameters was evident even only after a low single dose.

Baseline sleep quality could be a predictor of melatonin's response as PSQI was negatively correlated with total sleep time and VAS. There was a high proportion of poor sleepers in our cohort as 50% of the participants reported baseline PSQI >5 indicating poor sleep quality. ²⁷ These poor sleepers responded more strongly to melatonin as shown with more pronounced faster sleep onset latency compared with participants with normal baseline sleep quality. They also reported more satisfaction in taking melatonin. It could be that because participants with low PSQI already have good sleep so they did not see much effect from melatonin but people with poor sleep could notice the difference easily and they reported to have adequate sleep more than participants with normal baseline sleep quality (Table S1). We did not find a significant difference between melatonin and placebo groups in terms of side effects. A previous meta‐analysis also confirmed the safety of high‐dose melatonin (>10 mg) with a very good tolerability profile. ³²

For stress biomarkers, both sCort and sAA exhibited the diurnal patterns similar to previous studies. ¹¹ The diurnal pattern of sCort went up with a peak at 30 min after awakening and declined throughout the day with the lowest levels at night time. While the diurnal pattern of sAA had an opposite trend showing a trough level at 30 min after awakening and progressively increasing throughout the day.

To our surprise, our results did not support the notion that melatonin could decrease sCort levels in either normal sleepers or poor sleepers. In fact, melatonin administration increased sCort levels upon awakening time point. However, we would like to emphasize that such an effect was modest. As mentioned earlier that the effect of melatonin supplement on sCort levels is contradictory. There was only one previous study that found a very high‐dose melatonin (100 mg) administration to increase serum cortisol during the day at some time points only in postmenopausal women but not in younger women. ²⁰ The mechanism is still unclear. Moreover, participants with poor baseline sleep quality had a more pronounced increase in sCort levels following melatonin consumption at 10:00 h compared with participants with normal baseline sleep quality. A previous study has found that sleep deprivation led to enhanced plasma cortisol in the following evening. ¹⁵ Melatonin might not simply affect sCort levels in terms of stress reduction but it could also modulate the rhythm of sCort secretion or cortisol awakening response or cortisol slope throughout the 24‐h diurnal pattern. ¹⁸ As the change in circadian rhythm affected cortisol awakening response, ³³ it could be possible that melatonin administration at a particular time point as in the current study shifted the diurnal rhythm of sCort. Also, an increase in sCort level does not always mean more stress as an increase in cortisol awakening response possibly reflects better sensitivity to coping with stress during the day. ³⁴ More detailed monitoring of sCort levels at more time points and for more than 1 day should yield more information in this aspect.

On the other hand, melatonin did not affect sAA levels even though there was a tendency toward lower sAA levels upon melatonin ingestion at all time points. This was partly because of a very large variation in sAA secretion. Probably with longer term treatment, a more pronounced effect might be observed. To our knowledge, this is the first study to examine the effect of melatonin supplement on diurnal sAA levels.

In the current study, we chose to measure sMT levels as salivary levels were found to be highly correlated with serum levels. ³⁵ Our diurnal baseline sMT levels were within the same ranges as others as well as from the kit's information. ³⁶ As expected, ingestion of melatonin at 20:00 h significantly increased sMT levels up to at least 10:00 h (the time point 12:00 h was not examined). The peak sMT level was observed at 21:00 h within 1 h of ingestion. The levels went up too high beyond the kit's measurement range even though we have diluted the sample to a hundredth‐fold. So we think that this is the limitation of the salivary kit in case of melatonin ingestion as the levels could go up very high. However, the sMT levels in the current study was in the same range of serum melatonin in a previous study. ²⁶

The increases in sMT levels after melatonin ingestion was more pronounced among participants with poor baseline sleep quality (PSQI >5) compared with participants with normal sleep quality (Table 5). This suggested that poor sleepers were more sensitive to exogenous melatonin's effect. To our surprise, age had no effect on sMT levels both in the placebo group and after melatonin ingestion. Older participants (>51 years) did not exhibit lower sMT levels and responded similarly to younger participants. Generally, it is well known that melatonin production decrease with age. ³⁷ It could be that because the measurement of sMT levels was performed outside the peak secretion period of melatonin which is during the night time (2:00–3:00 h), the levels outside this peak window were generally lower so the effect from age was not apparent. ³⁵

We measured PR to check if they were in agreement with the stress biomarkers. PR also displayed a diurnal rhythm as already known. ³⁸ Exogenous melatonin ingestion and baseline sleep quality had no impact on PR; however, participants older than 51 y displayed more pronounced decrease in PR at 12:00 h after melatonin supplement. This implied that melatonin had a tendency to decrease PR and the effect was more pronounced in older people.

There are several limitations to this study. First, the administration of melatonin was only a single dose. Longer term treatment could possibly yield different effects. As the levels of these biomarkers could vary greatly, performing the experiment for many days to average the results would increase the reliability of the data. More insight could be gained if the saliva collection was performed at more time points throughout the day and night to better reflect the impact of melatonin administration on diurnal rhythm of sCort and sAA. Finally, the format of melatonin used was prolonged‐release, the immediate release could have a different dynamic.

In summary, a single‐dose administration of prolonged released melatonin was found to impact sleep quality as demonstrated from time to go to sleep, sleep onset latency, and total sleep time. These results confirmed the positive acute effect of even a single low dose melatonin. For saliva study, melatonin administration increased sCort levels at 10:00 h but not at other time points as well as a trend toward lower sAA at all time points examined. The reason behind these opposing effects on sCort and sAA is unclear. Long‐term studies would give more insight into the mechanism of melatonin on stress biomarkers.

AUTHOR CONTRIBUTIONS

Praewpat Pachimsawat: Conceptualization, Data processing, Formal analysis, Writing − review & editing. Piyanee Ratanachamnong: Investigation, Methodology, Data processing, Validation, Writing − review & editing. Nattinee Jantaratnotai: Investigation, Methodology, Data processing, Funding acquisition, Writing − original draft, Writing − review & editing. All of the authors have read and approved the final version of the manuscript.

FUNDING INFORMATION

This work was supported by Specific League Funds from Mahidol University.

CONFLICT OF INTEREST STATEMENT

The authors declare that there are no conflicts of interest. The funder had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

ETHICS STATEMENT

The study protocol was approved by Mahidol University multi‐faculty cooperative IRB review MU‐MOU COA 2022/018.2302. The trial was registered on Thaiclinicaltrials.org with the registration number TCTR20220310009. The study was conducted according to the principles of the Declaration of Helsinki.

Supporting information

Tables S1 –S6. xxx.

Pachimsawat P, Ratanachamnong P, Jantaratnotai N. Exogenous melatonin's effect on salivary cortisol and amylase: A randomized controlled trial. Pharmacol Res Perspect. 2024;12:e1205. doi: 10.1002/prp2.1205

DATA AVAILABILITY STATEMENT

The results obtained for all experiments performed are shown in the manuscript and Supplementary Materials; the raw data will be provided upon request.