{"paper_id":"51ab12c0-1971-4c62-bf39-e1a47bfd4295","body_text":"Vol.:(0123456789)1 3\nh\nttps://doi.org/10.1007/s00228-021-03234-6\nCLINICAL TRIAL\nAdjuvant use of melatonin for pain management in dysmenorrhea \n— a randomized double‑blinded, placebo‑controlled trial\nLisa Söderman1  · Måns Edlund2 · Ylva Böttiger3 · Lena Marions1\nReceived: 14 July 2021 / Accepted: 11 October 2021 \n© The Author(s) 2021\nAbstract\nPurpose Dysmenor\nrhea is a common, recurring, painful condition with a global prevalence of 71%. The treatment regime \nfor dysmenorrhea includes hormonal therapies and NSAID, both of which are associated with side effects.\nA dose of 10 mg melatonin daily has previously been shown to reduce the level of pelvic pain in women with endometriosis. \nWe chose to investigate how this regime, administered during the week of menstruation, would affect women with dysmen-\norrhea but without any signs of endometriosis, as adjuvant analgesic treatment.\nMethods\n For\nty participants with severe dysmenorrhea were randomized to either melatonin or placebo, 20 in each group. \nOur primary outcome was pain measured with numeric rating scale (NRS); a difference of at least 1.3 units between the \ngroups was considered clinically significant. Secondary outcomes were use of analgesics, as well as absenteeism and amount \nof bleeding. Mixed model was used for statistical analysis.\nResults\n Eighteen participants com\npleted the study in the placebo group and 19 in the melatonin group. Mean NRS in the \nplacebo group was 2.45 and 3.18 in the melatonin group, which proved to be statistically, although not clinically significant.\nConclusion\n This randomized, double-blinded, placebo-contr\nolled trial could not show that 10 mg of melatonin given orally \nat bedtime during the menstrual week had better analgesic effect on dysmenorrhea as compared with placebo. However, no \nadverse effects were observed.\nClinical trials\n NCT03782740 \nregistered on 17 December 2018.\nKeywords Adjuv\nant analgesics · Dysmenorrhea · Melatonin · Menstruation · Pelvic pain · RCT  \nIntroduction\nDysmenorrhea is a common, recurring, painful condition \nwith a global prevalence of 71% with a negative impact on \nacademic performance [1] and is shown to disrupt cognitive \nperformance [2]. The treatment regime for dysmenorrhea is \nnon-steroid anti-inflammatory drugs (NSAID) and/or hor -\nmonal suppression [3]. Hormonal therapies and NSAIDs are \nboth associated with side effects motivating the need for \nadditional treatment options.\nIn primary dysmenorrhea, defined as painful menstrua-\ntion without pelvic organ pathology, leukotrienes (LT) and \nprostaglandins (PG) are released from the uterus causing \nvasoconstriction and myometrial contractions, leading to \nhypercontractility of the uterus, ischemia, and pain [4]. The \nseverity of symptoms are directly proportional to the type \nand amount of PG released into the systemic circulation \nduring sloughing of the endometrial lining [5 ]. In addition, \nwomen with dysmenorrhea have a hyper-sensitization of \npain fibers [6] and a high level of brain-derived neurotrophic \nfactor (BDNF) in serum with a positive correlation to the \nintensity of dysmenorrhea [7], suggesting that primary dys-\nmenorrhea has the characteristics of both acute and chronic \npain.\nMelatonin is a hormone regulating the circadian rhythm, \nsynthesized, and secreted mainly from the pineal gland in \nthe brain. The synthesis is synchronized with the light/dark \ncycle by photosensitive ganglion cells in the retina of the \neye [8] and blocked by light at night. Secretion reaches peak \n * Lisa Söderman \n Lisa.soder\nman@ki.se\n1 Dept of Clinical Science and Education, Södersjukhuset, \nK\narolinska Institutet, Stockholm, Sweden\n2 KBH, Karolinska Institutet, Stockholm, Sweden\n3 Dept of Biomedical and Clinical Sciences, Linköping \nU\nniversity, Linköping, Sweden\n/ Published online: 20 October 2021\nEuropean Journal of Clinical Pharmacology (2022) 78:191–196\n\n1 3\nlevels at 02–04 am night [9]. Most of its metabolism occurs \nin the liver via cytochrome P450-mediated oxygenation, \nmainly by CYP1A2, and is excreted in urine [10].\nA recent systematic review suggests that women with \ndysmenorrhea have higher levels of oxidative stress than \nhealthy controls [11]. Melatonin has well reported anti-\noxidative properties [12]. In addition to analgesic and anti-\noxidative effects melatonin has proved to regulate contrac-\ntions of the myometrium [13], suggesting several properties \nof high interest in treating dysmenorrhea.\nThe analgesic effect of exogen melatonin has been dem-\nonstrated in both acute and chronic pain, including inflicted, \nexperimental pain [14], post-operative pain [15], fibromy -\nalgia [16], irritable bowel syndrome [17], and cluster head-\nache [18], with dosages ranging between 3 and 10 mg mela-\ntonin given orally. Ten-milligram melatonin ingested daily \nreduced dysmenorrhea in women with endometriosis and \nlowered the level of BDNF [19].\nWe chose to investigate how 10 mg melatonin daily, as an \nadjuvant analgesic during menstruation, would affect women \nwith dysmenorrhea without signs of endometriosis.\nMaterials and method\nWe conducted this randomized, double-blinded, placebo-\ncontrolled trial at Södersjukhuset, one of the largest hos-\npitals in Stockholm, Sweden. Participants were recruited \nbetween March and December 2019. Prior to enrolment, a \nwritten informed consent was obtained from the participants. \nThe trial was conducted in accordance with the principles \nexpressed in the Declaration of Helsinki.\nCall for participation was advertised on posters in the \nhospital, in gynecological outpatient clinics, in maternity \ncare outpatient clinics, and on social media. Women aged \n18–45 with regular menstruation, rating their dysmenor -\nrhea 7 or higher on a numeric rating scale (NRS) during \nthe most painful day, speaking and understanding Swed-\nish, and in good general health were screened at the clinics \nResearch Center for Womens’ Health at one of two doctors \nin charge of the trial. Screening visit included medical his-\ntory, a pregnancy test, and vaginal ultrasound to identify \nthose with manifest signs of endometriosis or other signifi-\ncant pathology for exclusion. Screening continued during \none observational menstrual cycle, during which pain was \nrecorded daily, and evaluated prior to inclusion. Exclusion \ncriteria were smoking, pregnancy, prior or current liver or \nkidney disease, endometriosis, ongoing use of melatonin, \nalteration of any medication during the last 3 months, and \nuse of opioids.\nAfter inclusion, participants were randomized to \n10 mg melatonin or placebo, each dose identical and dis-\npersed in two capsules of 5 mg melatonin or placebo (both \nmanufactured for the trial by APL, Stockholm, Sweden). The \nstudy drug was taken at bedtime daily for seven consecutive \ndays with start on the first day of menstrual bleeding. Par -\nticipants were instructed to continue with their usual pain \nmedicating regime, if needed, for the 3 months of the study \nwith no alterations. The duration of the study was one obser-\nvational cycle followed by two interventional cycles, three \nmenstrual cycles in total.\nOur primary outcome was mean value of pain recorded \ndaily during the week of menstrual bleeding. Secondary \noutcomes were use of analgesics, amount of bleeding, days \nof bleeding, days of pain, absenteeism, and potential effect \non cognition.\nAssessments were made daily with an online question-\nnaire sent via email to the participants. Registration started \non the first day of menstrual bleeding. The worst pain of \nthe day was recorded using the NRS, a scale from 0 to 10, \nwhere 10 is the worst imaginable pain. The use of anal -\ngesics was recorded daily with specification of number of \ntablets and dosage. Absenteeism was recorded daily. Bleed-\ning was recorded daily through a pictorial blood loss assess-\nment chart (PBAC). A reminder was sent by text message \nif needed. Cognition was assessed with a cognitive assess-\nment software, CANTAB® (Cambridge Cognition 2019. \nAll rights reserved. www.\n \ncantab.\n \ncom). The cognition test \nbattery, assessing motor screening tasks, reaction time, rapid \nvisual processing, paired associates learning, and spatial \nworking memory, was performed on a tablet computer at \nthe Research Center for Women’s Health at Södersjukhuset, \nduring the observational cycle and the last treatment cycle, \nrespectively.\nDaily recording of potential adverse effects was also \nmade, and general experience of the study drug was evalu-\nated at completion of the study.\nStudy data was collected and managed using REDCap \nelectronic data capture tools (9.5.9 Vanderbuilt University, \nNashville TN, USA) hosted at Karolinska Institutet.\nTo detect a clinically significant reduction of NRS of 1.3 \nunits [20] with a power of 80% and a 2-sided alpha value \nof 0.05, 15 participants in each group were needed. We \nincluded 20 participants in each group, all in all 40 partici-\npants, to compensate for potential dropouts.\nParticipants were included consecutively upon a visit \nwith the research nurse. Randomization was made by blocks \nof 4 by the manufacturer of the study drug, who provided \nconsecutively numbered drug containers. The randomization \nkey was retrieved and opened after the last participant had \ncompleted the study, thus assuring that the study blind was \nmaintained during the treatment phase of the trial.\nThe main characteristics of the study population are \npresented in Table  1. In the placebo group two participants \nhad anxiety, two had migraine, one had vestibulitis, and \none had both hypothyroidism and depression. Fourteen \n192 European Journal of Clinical Pharmacology (2022) 78:191–196\n\n1 3\nsubjects reported no co-morbidities. In the melatonin \ngroup two participants had premenstrual syndrome and \none had celiac disease, hypothyroidism, a herniated disk, \nhyperthyroidism, polycystic ovary syndrome, and rosacea, \nrespectively. Twelve subjects reported no co-morbidities.\nA mixed model was used to test the effect of the inter -\nvention and time. The inference was made on treatment \ncycle 1 and 2 excluding day 1, since the participants \nstarted the treatment on the evening of day 1. Three dif-\nferent covariance structures were tested by comparing\n −2\n \nlog-likelihood with chi-squared test on each outcome vari-\nable: unstructured, first-order autoregressive (AR (1)), and \ncompound symmetry. The best fit was obtained with an \nunstructured model for outcome 1, with fixed effect and \nfixed intercepts. AR (1) showed the best fit for outcome 2, \nwith fixed effects and random intercept. To test the effect \nof the two groups at each separate time point, an inter -\naction effect between time and groups was tested in the \nselected model.\nUnpaired t-tests were used to compare mean days of \ndysmenorrhea, days of bleeding, amount of bleeding, and \ncognition.\nAcceptability was analysed with Fisher’s exact test.\nSPSS version 26 (SPSS, Chicago, IL) was used for data \nanalyses.\nThe Regional Ethical Review Board at Karolinska \nInstitutet approved the trial (2017/1177–21/2) on 23 \nAugust 2017. Registration at Clinicaltrials.gov was made \nin December 2018 (NCT03782740). The first participant \nwas enrolled on 4 March 2019 and last patient last visit \noccurred on 28 February 2020.\nResults\nForty participants were randomized, 20 in each group, to \neither melatonin or placebo (Fig.  1). Clinical and demo-\ngraphic data of the two groups were similar (Table  1) \nwith no statistically significant differences. No differences \nbetween the groups were seen in the tests assessing cogni -\ntion (data not shown).\nITT analysis showed that the mean NRS of both treat-\nment cycles was 0.73 units lower in the placebo group which \nproved to be statistically significant using mixed model anal-\nysis (Table  2). The difference of mean NRS between the \nstudy groups of the baseline cycle was 0.74 units, with no \nstatistical significance using unpaired t-test. Mean NRS was \nlower in the placebo group in both treatment cycles (Fig. 2).\nSimilar results were seen in the per-protocol analyses \nwith the mean difference of 0.47 units, P\n \n=\n \n0.09, CI\n \n−1.00 \nto\n −0.07 (mean 2.64, sd\n =\n 2.84 in t\nhe placebo group, and \nmean 3.10, sd\n =\n 3.36 in t\nhe melatonin group). Adjusting \nfor weight did not affect the results.\nNo significant differences were seen between the groups \nregarding amount of bleeding (PBAC), use of analgesics, \nlevel of absenteeism, or in any of the cognition tests.\nNo adverse effects were reported. There were no \nobserved differences regarding acceptability between the \ngroups. Two participants used combined oral contraceptive \npill (COCP), which has been shown to result in a higher \nlevel of serum melatonin compared with controls [21]. They \nwere both randomized to the melatonin group and reported \nno adverse effects. Both reported a good experience with \nthe study drug.\nTable 1  Main c haracteristics \nof the study population and \nobservations during the 7-day \nbaseline cycle \na P ain refers to mean value of numeric rating scale (NRS) of the 7 days of observation. One participant in \nthe placebo group failed to report the level of dysmenorrhea for every day in the baseline cycle\nPlacebo Melatonin\nn Mean (SD) n Mean (SD)\nAge 20 28.45 (7.16) 20 26.95 (5.20)\nWeight, in kg 20 72.80 (16.67) 19 68.84 (13.12)\nNumber of pregnancies 18 0.78 (1.22) 20 0.45 (1.0)\nPain,  meana 19 3.61 (.96) 20 4.35(1.71)\nDays of pain 20 4.40 (1.18) 20 4.80 (1.70)\nTotal amount of analgesics in mg 20 4695.00 (3880.38) 20 4887.50 (5715.10)\nDays of bleeding 20 4.80 (0.95) 20 5.35 (1.09)\nTotal PBAC 20 120.15 (59.72) 20 125.55 (122.20)\nContraceptive method n n\nCondom 5 8\nHormonal IUD 2 1\nCupper IUD 2 1\nCOCP 0 2\nNone 11 8\n193European Journal of Clinical Pharmacology (2022) 78:191–196\n\n1 3\nA majority in both groups wished to continue with the treat-\nment, 61% in the placebo group and 63% in the melatonin \ngroup. Fifty-three percent in the melatonin group would rec-\nommend the treatment to a friend, 39% in the placebo group. \nNo statistically significant differences in the acceptability of \nthe study drugs were identified.\nThere were 20 days of missing data, 10 in each treatment \ngroup, all representing the last days of the cycle with no pain \nand no bleeding, suggesting no impact on the results.\nDiscussion\nTen-milligram melatonin given during the menstrual week \nshowed no clinically significant difference in the level of \ndysmenorrhea compared with placebo. We compared mean \nNRS for 7 days irrespectively of the number of days with \ndysmenorrhea, which gives seemingly low values of mean \nNRS considering most participants had fewer days of dys-\nmenorrhea than 7.\nNo differences were observed in the secondary outcomes, \nuse of analgesic drugs, number of days of pain, and bleeding, \nrespectively, or the amount of bleeding. No adverse effects \nwere reported, and no one reported daytime sleepiness after \nreceiving melatonin treatment. Our assessment of cognition \nincluding reaction time and rapid visual processing showed \nno difference between the groups. In concordance with prior \nstudies, tolerability and acceptance was good.\nThe study drug was ingested at bedtime to harmonize with \nthe cyclicity of endogenous melatonin. For many participants \nthe pain was already manifest when the treatment was initi-\nated, on the evening of the first day of menstruation. This may \nFig. 1  Flo wchart showing \nrecruitment and progress \nthrough the study. Intent-to-treat \n(ITT) analyses included 19 par-\nticipants in the melatonin group \nand 18 in the placebo group. \nPer-protocol (PP) analyses were \nmade on 16 participants in each \ngroup\nTable 2  Mean s tudy outcomes of the two 7-day cycles (ITT)\na Anal yzed with mixed model analysis\nb Anal yzed with unpaired t-test, cycle 3\nOutcomes Treatment n Adjusted mean (SD) Adjusted mean \ndifference\n95% confidence intervals P-value\nPain,  meana Placebo 18 2.45 (2.94)  −.73  −1.30 to  −.16 .015\nMelatonin 19 3.18 (3.37)\nAmount of analgesics (mg)a Placebo 18 464.032 (986.20)  −115.31  −497.64 t o 267.02 .505\nMelatonin 19 579.342 (1192.00)\nDays with \n dy\nsmenorrheab Placebo 18 3.89 (1.28)  −.53  −1,56-.49 .149\nMelatonin 19 4.42 (1.74)\nDays with \n bleedingb Placebo 18 4.78 (.94)  −.17  −.93-.59 .329\nMelatonin 19 5.03 (1.15)\nPBAC — pictorial blood loss \nassessment  c\nhartb\nPlacebo 18 127.17 (48.66)  −29.78  −118.26-58.70 .246\nMelatonin 19 156.95 (178.48)\n194 European Journal of Clinical Pharmacology (2022) 78:191–196\n\n1 3\nhave been too late to give a pain reduction. Considering that \nthe time to maximal concentration (tmax) of melatonin gener-\nally occurs around 50 min after ingestion, and also taking into \naccount its short half-life in humans (20–40 min) [22], a pro-\nphylactic regime, given 45 min prior to pain onset, might have \nbeen favorable, but too challenging to administer due to the \nrelatively unpredictable nature of dysmenorrhea. We did not \nevaluate plasma concentrations of melatonin, which would have \nprovided valuable information, since bioavailability is low at \n15% and is associated with high inter-subject variability [22].\nIf there is a correlation between pain regulation and mela-\ntonin, it should be possible to find an alteration in the endog-\nenous melatonin concentration and the perception of pain. \nNelson et al. [23] showed a decrease in melatonin concentra-\ntion in saliva within 5 min after an acute pain stimulus. In addi-\ntion, Almay et al. found lower levels of endogenous melatonin \nin serum and urine in patients with chronic pain [24]. Studies \nhave shown conflicting results regarding a suggested cyclic \nalteration of melatonin levels following the menstrual cycle \n[25–27]. To our knowledge there are no studies measuring the \nendogenous levels of melatonin in women with dysmenorrhea.\nDose, administration route, and timing are of interest for \nfuture studies. Perhaps a prophylactic regime of melatonin \nadministered with a mode of longer duration such as trans-\ndermal application, assessed with serum levels of melatonin \nas well as measuring its clinical effect on pain, could provide \ninformation on how to treat dysmenorrhea with melatonin.\nThe strengths of the study include the comparison of mel-\natonin and placebo in a double-blinded, placebo-controlled \ntrial; the low level of dropouts and missing data; and the \nlow risk for recollection bias with a daily questionnaire. The \nstudy design offers high internal validity. We conducted this \ntrial according to the CONSORT guidelines [28].\nThe study limitations involve study design as well as \nmethod. By including only self-selected women in good \nhealth, non-smokers with regular periods our study popula-\ntion is lacking in diversity with a possibly low external valid-\nity. The standard deviations and confidence intervals suggest a \nlarge variance which might suggest the study population being \ntoo small.\nThe first day is often the most painful, but due to not \nknowing the exact day the menstruation would commence \nwe administered the first dose of melatonin in the evening \nof the first day of menstruation. Resulting, unfortunately, in \nthat the first day of menstruation remained untreated.\nConclusion\nOur study could not show that 10 mg of melatonin given orally \nat bedtime during the menstrual week had better analgesic \neffect on dysmenorrhea as compared with placebo. However, \nno adverse effects were observed.\nAuthor contribution Lena Mar ions: conceptualization, funding acqui-\nsition, supervision. Lisa Söderman, Lena Marions: clinical investiga-\ntion. Lisa Söderman: data curation, formal analysis, project admin -\nistration, resources, software, roles/writing — original draft. Lisa \nSöderman, Lena Marions, Måns Edlund, Ylva Böttiger: methodology, \nstudy design, writing — review and editing.\nFunding\n Open access funding pr\novided by Karolinska Institute. This \nstudy was supported by grants provided by the Stockholm County \nCouncil (ALF project 20180306) and AFA insurance (project number \n170157). The founding sources had no role in study design, collection \nof data, analysis, or interpretation of data, nor in decision to submit \nthe article for publication.\nAvailability of data and material\n A\nll data are stored with REDCap, \nsoftware.\nFig.\n \n2\n  Dy\nsmenorrhea day by \nday. Mean level of dysmenor-\nrhea reflected by numeric rating \nscale (NRS) days 2–7 for the \ntwo study groups during the two \ntreatment cycles, respectively\n195European Journal of Clinical Pharmacology (2022) 78:191–196\n\n1 3\nDeclarations \nEthics approval  The R egional Ethical Review Board at Karolinska \nInstitutet approved the trial (2017/1177–21/2) on 23 August 2017.\nConsent to participate\n Inf\normed consent was obtained from all indi-\nvidual participants included in the study.\nConflict of interest\n L\nS, YB, and LM report no conflict of interest. ME \nholds a full-time position as Medical Director for SOBI AB.\nOpen Access\n  This ar\nticle is licensed under a Creative Commons Attri-\nbution 4.0 International License, which permits use, sharing, adapta-\ntion, distribution and reproduction in any medium or format, as long \nas you give appropriate credit to the original author(s) and the source, \nprovide a link to the Creative Commons licence, and indicate if changes \nwere made. The images or other third party material in this article are \nincluded in the article's Creative Commons licence, unless indicated \notherwise in a credit line to the material. If material is not included in \nthe article's Creative Commons licence and your intended use is not \npermitted by statutory regulation or exceeds the permitted use, you will \nneed to obtain permission directly from the copyright holder. To view a \ncopy of this licence, visit http://\n \ncreat\n \niveco\n \nmmons.\n \norg/\n \nlicen\n \nses/\n \nby/4.\n \n0/.\nReferences\n 1 .  A rmour M et al (2019) Self-care strategies and sources of knowl -\nedge on menstruation in 12,526 young women with dysmenorrhea: \na systematic review and meta-analysis. PLoS One 14(7):e0220103\n 2\n. K\neogh E et al (2014) The effects of menstrual-related pain on \nattentional interference. Pain 155(4):821–827\n 3\n. A\nCOG Committee Opinion No (2018) 760: Dysmenor -\nrhea and endometriosis in the adolescent. Obstet Gynecol \n132(6):e249–e258\n 4.\n Da\nwood MY (2006) Primary dysmenorrhea: advances in patho-\ngenesis and management. Obstet Gynecol 108(2):428–441\n 5\n. Chan \nWY, Dawood MY, Fuchs F (1981) Prostaglandins in primary \ndysmenorrhea. Comparison of prophylactic and nonprophylactic \ntreatment with ibuprofen and use of oral contraceptives. Am J \nMed 70(3):535–41\n 6\n. Ia\ncovides S et al (2013) Women with dysmenorrhea are hyper -\nsensitive to experimental deep muscle pain across the menstrual \ncycle. J Pain 14(10):1066–1076\n 7\n. Han F e\nt al (2019) Correlation between thalamus-related func-\ntional connectivity and serum BDNF levels during the periovula-\ntory phase of primary dysmenorrhea. Front Hum Neurosci 13:333\n 8\n. Cipolla-N\neto J, Amaral FGD (2018) Melatonin as a hormone: new \nphysiological and clinical insights. Endocr Rev 39(6):990–1028\n 9\n. Andersen LP (2016) \nThe analgesic effects of exogenous melatonin \nin humans. Dan Med J 63:10\n 1\n0. Chen WW\n, Zhang X, Huang WJ (2016) Pain control by mela-\ntonin: physiological and pharmacological effects. Exp Ther Med \n12(4):1963–1968\n 1\n1. Szmidt MK e\nt al (2020) Primary dysmenorrhea in relation to oxi-\ndative stress and antioxidant status: a systematic review of case-\ncontrol studies. Antioxidants (Basel) 9:10\n 1\n2. R\neiter RJ et al (2016) Melatonin as an antioxidant: under promises \nbut over delivers. J Pineal Res 61(3):253–278\n 1\n3. A\nyar A et al (2001) Melatonin inhibits spontaneous and oxytocin-\ninduced contractions of rat myometrium in vitro. Neuro Endo-\ncrinol Lett 22(3):199–207\n 1\n4. S\ntefani LC et al (2013) A phase II, randomized, double-blind, \nplacebo controlled, dose-response trial of the melatonin effect on \nthe pain threshold of healthy subjects. PLoS One 8(10):e74107\n 15.\n Caumo \nW et al (2007) The clinical impact of preoperative mela-\ntonin on postoperative outcomes in patients undergoing abdominal \nhysterectomy. Anesth Analg 105(5):1263–1271\n 1\n6. de\n Zanette SA et al (2014) Melatonin analgesia is associated with \nimprovement of the descending endogenous pain-modulating \nsystem in fibromyalgia: a phase II, randomized, double-dummy, \ncontrolled trial. BMC Pharmacol Toxicol 15(1):40–40\n 1\n7. L\nu WZ et al (2005) Melatonin improves bowel symptoms in \nfemale patients with irritable bowel syndrome: a double-blind pla-\ncebo-controlled study. Aliment Pharmacol Ther 22(10):927–934\n 1\n8. Leone M e\nt al (2016) Melatonin versus placebo in the prophylaxis \nof cluster headache. Cephalalgia 16(7):494–496\n 19.\n Sc\nhwertner A et al (2013) Efficacy of melatonin in the treatment \nof endometriosis: a phase II, randomized, double-blind, placebo-\ncontrolled trial. Pain 154(6):874–881\n 2\n0. T\nodd KH et al (1996) Clinical significance of reported changes in \npain severity. Ann Emerg Med 27(4):485–489\n 21.\n Hilli J e\nt al (2008) The effect of oral contraceptives on the phar -\nmacokinetics of melatonin in healthy subjects with CYP1A2 g.-\n163C>A polymorphism. J Clin Pharmacol 48(8):986–94\n 2\n2. Har\npsøe NG et al (2015) Clinical pharmacokinetics of melatonin: \na systematic review. Eur J Clin Pharmacol 71(8):901–909\n 2\n3. N\nelson FA, Farr LA, Ebadi M (2001) Salivary melatonin response \nto acute pain stimuli. J Pineal Res 30(4):206–212\n 2\n4. Alma\ny BG, von Knorring L, Wetterberg L (1987) Melatonin in \nserum and urine in patients with idiopathic pain syndromes. Psy-\nchiatry Res 22(3):179–191\n 25.\n Br\nzezinski A et al (1988) The circadian rhythm of plasma mela-\ntonin during the normal menstrual cycle and in amenorrheic \nwomen. J Clin Endocrinol Metab 66(5):891–895\n 2\n6. K\nostoglou-Athanassiou I et al (1998) Neurohypophysial hormone \nand melatonin secretion over the natural and suppressed men-\nstrual cycle in premenopausal women. Clin Endocrinol (Oxf) \n49(2):209–216\n 2\n7. S\nhechter A, Varin F, Boivin DB (2010) Circadian variation of \nsleep during the follicular and luteal phases of the menstrual \ncycle. Sleep 33(5):647–656\n 2\n8. Sc\nhulz KF, Altman DG, Moher D (2010) CONSORT 2010 State-\nment: updated guidelines for reporting parallel group randomised \ntrials. BMJ 340:c332\nPublisher's Note\n Spr\ninger Nature remains neutral with regard to \njurisdictional claims in published maps and institutional affiliations.\n196 European Journal of Clinical Pharmacology (2022) 78:191–196","source_license":"CC0","license_restricted":false}