Timing of Melatonin Start and Its Relation to Sleep Time in Older Adults under Uncontrolled and Controlled Light

The pineal gland activity is low during the day and then during the evening, it synthesizes the hormone melatonin that is then released into the blood circulation during the night. Melatonin thus serves as an endocrine signal of darkness. The timing of melatonin production is regulated through a pathway from the suprachiasmatic nuclei (SCN), the central circadian pacemaker. The SCN receives information about the environmental light–dark cycle through a pathway from the retina, and in turn, passes this information to the pineal gland [1]. Exposure to light during the biological night when melatonin is being secreted can acutely suppress melatonin [2,3].

Because circulating melatonin concentrations are low during the day and then abruptly rise in the evening when secretion begins [4]; the time at which this evening rise occurs has become a convenient measure of the central circadian phase in research studies [5,6,7,8,9]. More recently, the dim light melatonin onset (DLMO) has also been assessed in patients with circadian rhythm sleep-wake disorders in order to determine whether those patients have abnormal timing of their circadian rhythms compared to control participants [7,10,11,12]. The stability of DLMO from day to day and how measurement conditions may impact DLMO timing are important to understand when using this research tool in clinical applications [13]. Because light can have an important effect on melatonin production when assessing DLMO in patients either in the home environment [14] or in controlled laboratory conditions [15], understanding how lighting during and immediately prior to the DLMO sample collection impacts DLMO timing is necessary. In young adults, even indoor levels of light during DLMO sampling have been shown to suppress melatonin secretion and delay melatonin onset compared to collection under dim lighting [16,17]. In addition to this acute effect of light on circulating melatonin concentrations, there is also evidence from young adults of adaptation or sensitization to light, where brighter light exposure on the day or days prior to melatonin sampling changes the magnitude of responses to light exposure during sampling [18,19].

We and others have reported that the timing of the melatonin phase in healthy older adults when assessed in controlled laboratory lighting conditions is earlier than in young adults and that the relative timing between melatonin and sleep timing is significantly different between older and young adults [20,21,22]. Moreover, most previous studies of melatonin responses to light have been conducted in young adults. Therefore, the main objective of this analysis was twofold: first, to explore how DLMO is impacted by exposure to uncontrolled vs. controlled lighting early on the day of assessment; and second, to explore how melatonin onset is impacted by different indoor light levels (dim vs. room lighting) in healthy older adults.

The total number of participants included in the final analyses was 13, with 8 participants in Project A and 5 participants in Project B (see Figure 1). This included 5 women and 8 men ranging in age from 56 to 74 years (64.30 ± 5.25 years), and their average bedtime in the week prior to study while maintaining a self-selected 8 h sleep schedule was 22:57 ± 0:32 h.

For Project A, the phase angle between DLMO and bedtime was not significantly different between Night 1 and Night 2 (Night 1: −2:24 ± 1:31 h vs. Night 2: −2:01 ± 1:48 h, t = −1.48, p = 0.18), whereas for Project B, phase angle between melatonin onset and bedtime was significantly shorter on Night 2 than Night 1 (Night 1: −1:31 ± 1:41 h vs. Night 2: 0:24 ± 2:21 h, t = −4.15, p = 0.01). Furthermore, for Project A, no significant difference was found for DLMO timing between Night 1 and Night 2 (Night 1: 20:38 ± 1:35 h vs. Night 2: 21:02 ± 1:46 h, t = −1.48, p = 0.18), although the average DLMO on Night 2 was 20 min later than on Night 1. For Project B, the melatonin onset on Night 2 was significantly later than Night 1 by nearly 2 h (Night 1: 21:14 ± 1:06 h vs. Night 2: 23:10 ± 1:41 h, t = −4.15, p = 0.01, see Figure 1).

Multivariate analysis of variance (MANOVA) confirmed these findings. The phase angle of DLMO to bedtime was not significantly different on Night 1 between Project A and Project B (Night 1: Project A: −2:24 ± 1:31 h vs. Project B: −1:31 ± 1:41 h, mean difference = −0:52 h, p = 0.35), whereas on Night 2 Project A had a significantly longer phase angle between melatonin onset and bedtime than Project B (Night 2: Project A: −2:01 ± 1:48 h vs. Project B: 0:24 ± 2:21 h, mean difference = −2:25 h, p = 0.05). In addition, there was no significant difference in the timing of DLMO on Night 1 between Project A and Project B (Night 1: Project A: 20:38 ± 1:35 h vs. Project B: 21:14 ± 1:06 h, mean difference 0:36 h, p = 0.47). However, on Night 2, melatonin onset for Project A was two hours earlier than Project B (Night 2: Project A: 21:02 ± 1:46 h vs. Project B: 23:10 ± 1:41 h, mean difference 2:08 h, p = 0.05).

The main objective of this study was to explore how DLMO and phase angle between DLMO and bedtime are impacted by prior uncontrolled lighting exposure on the day of assessment compared to when lighting is controlled throughout the day of assessment, as well as to compare how melatonin onset is impacted when assessed under different light levels (dim vs. room lighting) in healthy older adults. Our results indicated that for Project B, exposure to different lighting during melatonin assessment impacted melatonin onset, with onset significantly later when assessed in room light levels than in dim light, presumably due to melatonin suppression as represented in Figure 1. These results obtained from healthy older adults are consistent with findings from young adults [16,17,18,19,23]. Moreover, the phase angle in healthy older adults in this study was comparable to young adults [23]. Interestingly, our finding of the range of DLMO timing in adults aged 56–74 coincides with a recent review analyzing DLMO across the ages [24].

Studies in young adults have shown that prior exposure to light during the day reduces the suppressive effect of light on melatonin at night Nevertheless, in our study with healthy older adults the finding that DLMO timing in Project A did not change significantly between Night 1 and Night 2 as shown in Figure 1, despite the differences in lighting during the first half of the day, is also of interest. On Day 1 these participants woke at home and spent the first 4–6 h of their day in uncontrolled lighting at home and on their travel to the laboratory for their study. They then entered controlled dim lighting for the final ~10 h of their waking day. In contrast, on Day 2 they woke in the laboratory and spent the entire 16 h wake episode in dim lighting. Understanding how exposure to brighter light early in the day might impact the timing of DLMO assessed in the evening is of particular interest for researchers and clinicians who want to assess DLMO in patients where it may be impractical to have them in controlled lighting for an entire day.

In Project A, the average DLMO on Night 2 was later than on Night 1, as would be expected given that the intrinsic period of the central circadian pacemaker is longer than 24 h and that the participants were exposed to brighter light on the morning of Day 1, although the difference (20 min) did not reach statistical significance in this small study sample. Additional studies to determine how the duration, timing, and intensity of light on the day of DLMO assessment impacts DLMO and the duration of dim light necessary to assess DLMO with accuracy should be carried out.

In summary, our results suggest that the background lighting during sample collection for DLMO assessment is important for older, as well as young adults and that lighting of even modest indoor intensity can suppress and/or delay melatonin onset in older adults. Our findings also show that brighter uncontrolled lighting on the morning of DLMO assessment minimally interferes with DLMO timing as long as DLMO is assessed in dim light.

The authors would like to thank C. O’Brien, K. Malvey, and D. McCarthy for participant recruitment and screening; the research staff from the BWH General Clinical Research Center for study execution and data collection; J.M. Ronda for bioinformatics support; and the participants of the studies.

Conceptualization, J.F.D.; methodology, J.F.D. and A.A.-P.; validation, J.F.D. and A.A.-P.; formal analysis, A.A.-P.; investigation, J.F.D.; resources, J.F.D. and C.A.C.; data curation, A.A.-P., J.-H.L. and J.F.D.; writing—original draft preparation, A.A.-P. and J.F.D.; writing—review and editing, C.A.C. and J.-H.L.; supervision, J.F.D.; funding acquisition, C.A.C. and J.F.D. All authors have read and agreed to the published version of the manuscript.

The protocols were reviewed and approved by the Partners Health Care Human Subjects Committee and the studies were conducted in accordance with the principles outlined in the Declaration of Helsinki.

Each participant gave written informed consent prior to their study.

The data presented in this paper is available from the corresponding author in compliance with Mass General Brigham data sharing policies.

AAP, JHL, and JFD declare no conflicts of interest. CAC serves as the incumbent of an endowed professorship provided to Harvard Medical School by Cephalon, Inc., Frazer, PA, USA and reports institutional support for a Quality Improvement Initiative from Delta Airlines, Atlanta, GA, USA and Puget Sound Pilots, Seattle, WA, USA; education support to Harvard Medical School Division of Sleep Medicine and support to Brigham and Women’s Hospital from: Jazz Pharmaceuticals PLC Inc., Dublin, Ireland, Philips Respironics, Inc., Murrysville, PA, USA, Optum, Eden Prairie, MN, USA and ResMed, Inc, San Diego, CA, USA.; research support to Brigham and Women’s Hospital from Axome Therapeutics, Inc., New York, NY, USA, Dayzz Ltd., Tel Aviv, Israel, USA, Peter Brown and Margaret Hamburg, Washington, DC, USA, Regeneron Pharmaceuticals, Tarrytown, NY, USA, Sanofi SA, Paris, France, Casey Feldman Foundation, Philadelphia, PA, USA, Summus, Inc., New York, NY, USA, Takeda Pharmaceutical Co., Ltd., Tokyo, Japan, Abbaszadeh Foundation, Los Angeles, CA, USA, CDC Foundation, Atlanta, GA, USA; educational funding to the Sleep and Health Education Program of the Harvard Medical School Division of Sleep Medicine from ResMed, Inc., San Diego, CA, USA, Teva Pharmaceuticals Industries, Ltd., Tel Aviv, Israel, and Vanda Pharmaceuticals, Washington, DC, USA; personal royalty payments on sales of the Actiwatch-2 and Actiwatch-Spectrum devices from Philips Respironics, Inc., Murrysville, PA; USA; personal consulting fees from Axome, Inc., New York, NY, USA, Bryte Foundation, Silicon Valley, CA, USA. With Deep, Inc., New York, NY, USA and Vanda Pharmaceuticals, Washington, DC, USA; honoraria from the Associated Professional Sleep Societies, LLC, Darien, IL, USA, for the Thomas Roth Lecture of Excellence at SLEEP 2022, from the Massachusetts Medical Society, Waltham, MA, USA, for a New England Journal of Medicine Perspective article, from the National Council for Mental Wellbeing, Washington, DC, USA, from the National Sleep Foundation, Washington, DC, USA, for serving as chair of the Sleep Timing and Variability Consensus Panel, for lecture fees from Teva Pharma Australia PTY Ltd., New South Wales, Australia, and Emory University, Atlanta, GA, USA, and for serving as an advisory board member for the Institute of Digital Media and Child Development, the Klarman Family Foundation, Boston, MA, USA, and the UK Biotechnology and Biological Sciences Research Council, Swindon, UK. CAC has received personal fees for serving as an expert witness on a number of civil matters, criminal matters, and arbitration cases, including those involving the following commercial and government entities: Amtrak, Washington, DC, USA; Bombardier, Inc., Montreal, Canada; C&J Energy Services, Houston, Texas; Dallas Police Association, Dallas, TX, USA; Delta Airlines/Comair, Atlanta, GA, USA; Enterprise Rent-A-Car, San Luis, MO; FedEx, Memphis, TN; Greyhound Lines, Inc./Motor Coach Industries/FirstGroup America, Dallas, TX, USA; PAR Electrical Contractors, Inc., Kansas City, MO, USA; Puget Sound Pilots, Seattle, WA; and the San Francisco Sheriff’s Department, San Francisco, CA, USA; Schlumberger Technology Corp., Houston, TX, USA; Union Pacific Railroad, Omaha, NE; United Parcel Service, Atlanta, GA, USA; Vanda Pharmaceuticals, Washington, DC, USA. CAC has received travel support from the Stanley Ho Medical Development Foundation, Macau, China for travel to Macao and Hong Kong; equity interest in Vanda Pharmaceuticals, Washington, DC, USA, with Deep, Inc., New York, NY, USA, and Signos, Inc., Burlingame, CA, USA; and institutional educational gifts to Brigham and Women’s Hospital from Johnson & Johnson, New Brunswick, NJ, USA, Mary Ann and Stanley Snider via Combined Jewish Philanthropies, Boston, MA, USA, Alexandra Drane, DR Capital, Harmony Biosciences, LLC, Plymouth Meeting, PA, USA, San Francisco Bar Pilots, San Francisco, CA, USA, Whoop, Inc., Boston, MA, USA, Harmony Biosciences LLC, Plymouth, PA, USA, Eisai Co., Ltd., Tokyo, Japan, Idorsia Pharmaceuticals Ltd., Allschwil, Switzerland, Sleep Number Corp., Minneapolis, MN Apnimed, Inc., Cambridge, MA, USA, Avadel Pharmaceuticals, Lyon, France, Bryte Foundation, Silicon Valley, CA, USA, f.lux Software, LLC, Boulder, CO, USA, Stuart F. and Diana L. Quan Charitable Fund, Tucson, AZ, USA. Dr Czeisler’s interests were reviewed and are managed by the Brigham and Women’s Hospital and Mass General Brigham in accordance with their conflict-of-interest policies.

This research was funded by the US National Institutes of Health (NIH) grant numbers P01 AG09975, R01 AG06072, and R01 AG044416. The studies were carried out at the BWH General Clinical Research Center, which was supported by NIH grant M01 RR02635. AAP was supported by the Sara Elizabeth O’Brien Trust, Bank of America, N.A. Fellowship. Support for the analyses included here was also provided by NIH grant R01 HL148704.