Yenra : Psychology : Sleep Deprivation Facts : Deprived of sleep by the body's production of melatonin


Changes in the rhythmic action of the body producing Melatonin may be why the elderly don't sleep as well according to study published in the February 2002 edition of the American Journal of Physiology--Endocrinology and Metabolism

Getting a good night's sleep can become more difficult with age. Survey studies have shown that up to one-third of older individuals report difficulty maintaining sleep on a recurring basis and more than one-half report occasional problems with their sleep.

The depth and continuity of sleep changes with age because there is a lower percentage of sleep spent in the deepest stages of non-REM sleep, there are more frequent arousals and awakenings during the sleep episode, and the inability to sustain sleep for the desired duration frequently occurs.

As if fitful sleep wasn't bad enough, the elderly are deluged with misinformation inspired by marketing efforts that offer a "cure-all" for the problem. Too often they are told that a melatonin deficiency is the cause of their distress. Melatonin is a putative sleep-related hormone. Studies of exogenous melatonin administration have shown that melatonin can facilitate sleep onset at certain times of day. But now, a new study asserts that it is the body's inner clock -- involved with the production of melatonin -- which may be the obstacle to a good night's sleep. This challenges the marketers' notions that the problem is a deficiency of the hormone itself.


One of the most prominent changes in sleep that accompanies aging is a shift (to an hour earlier) in the timing of a nightly sleep episode. In addition to sleep timing, the rhythms of core body temperature and plasma cortisol are also known to occur at an earlier hour in older people. Age-related changes in the amplitude of circadian rhythms of hormone secretion and core body temperature have also been reported.

There may be a causal link between the age-related changes in hormone secretion and core body temperature with changes in sleep. Conversely, a single mechanism may underlie these changes. Given that the circadian timing system regulates the timing and internal organization of sleep and hormone secretion, age-related changes in this system may underlie both processes.

One of the most reliable markers of the output of the circadian pacemaker is the circadian rhythm of melatonin secretion. It has been hypothesized that melatonin secretion decreases with age and that such a decrease is causally related to the increased sleep disruption in older people. However, previous studies reveal that nocturnal plasma melatonin concentrations in most very healthy older subjects are not significantly reduced compared with those of healthy young men. Moreover, there was no significant difference in the duration of the nightly melatonin secretion time between young and older subjects. Thus, neither decreased plasma melatonin levels nor shorter duration of melatonin secretion can fully explain the age-related changes in sleep timing and consolidation observed in healthy older individuals.

One theory is that older people are not only waking up at an earlier clock hour but are also waking at a different internal circadian time. Recent findings suggest that an alteration in the relative timing between the circadian system and the nightly sleep episode may occur with aging and raise the possibility that this altered timing may contribute to the increased sleep disruption with age.

The Study

To build upon their past research efforts, a team of physiologists set out to examine the internal phase relationship between sleep-wake timing and the timing of another marker of circadian phase. The timing of the plasma melatonin rhythm is considered to be a more accurate marker of the status of the circadian timing system than that of core body temperature because it is less affected by changes in posture and sleep-wake state. Therefore, they investigated the relationship between the timing of the rhythm of plasma melatonin secretion and the timing of the habitual sleep-wake episode in healthy young and older adults.

The authors of "Peak of Circadian Melatonin Rhythm Occurs Later Within the Sleep of Older Subjects, are Jeanne F. Duffy, Jamie M. Zeitzer, David W. Rimmer, Elizabeth B. Klerman, Derk-Jan Dijk, and Charles A. Czeisler, all from the Division of Sleep Medicine, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA. Their findings are published in the February 2002 edition of the American Journal of Physiology--Endocrinology and Metabolism.


The researchers recruited 15 older men and women (mean: 67.8 +3.1 years) and 33 young men (23.4 + 3.3 years), all had participated in previous studies between 1990 and 1996. Each was in good health, as determined through medical history, clinical biochemical screening tests on blood and urine, an electrocardiogram, a physical examination, and chest radiograph (older subjects only). Subjects were also in good psychological health, which was determined by testing and an interview with a clinical psychologist.

Subjects were drug free as verified by a toxicological urine analysis of their urine and were without significant sleep complaint by history and questionnaire. Older subjects underwent an overnight polysomnographic sleep screening examination before the study to rule out those individuals with clinically significant sleep apnea and/or periodic limb movements. To ensure that the circadian timing system of each subject was adapted to his or her daily routine, only subjects who denied a history of night shift work within the past 3 years and transmeridian travel (>1 time zone) within the past three months were studied.

Key elements of the study included:

Protocol, where each study began with three baseline days and nights, with 8-h sleep episodes scheduled at the subject's habitual times as determined from the sleep logs from the week immediately before the study commenced. The baseline segment was followed by a constant routine (CR) to assess the endogenous phase and amplitude of the subject's circadian rhythms of plasma melatonin and core body temperature; and

(2) Data analysis, which included recording and averaging habitual wake and bed times from the sleep diary for the seven nights immediately before the study began. The phase of the melatonin secretion pattern was defined as the midpoint between the upward and downward crossing of the 24-hour mean value.

To further explore potential age-related changes in the sleep-melatonin phase relationship, plasma Melatonin rhythm was analyzed in four additional ways to determine the timing of plasma elatonin onset and offset. Plasma melatonin onset was defined as (1) the dim-light melatonin onset (DLMO), defined as the time at which plasma levels reached 10 pg/ml, and (2) the time at which melatonin levels rose to 25 percent of the nightly peak.


The average wake time and bedtime of the older subjects occurred greater or equal to one hour earlier than did those of the young subjects. The circadian phase of MELmid also occurred at a significantly earlier hour in the older subjects. When the plasma melatonin rhythm was examined in more detail, the findings indicated that the earlier midpoint of the overall rhythm among the older subjects was reflected in a significantly earlier onset of the rhythm Although the offset of the melatonin rhythm also occurred at an earlier clock hour in the older subjects, this did not reach statistical significance.

The 24-hour mean melatonin values of the older and young subjects were not significantly different nor was the duration of melatonin secretions different. There was a significant correlation between habitual wake time (HW) and melatonin phase (MELmid) in both the older and young groups of subjects. A linear regression fitted to both data sets indicates that the nature of this relationship is different between the two groups, with an earlier wake time associated with an even earlier circadian phase in the older subjects. Other key findings included:

A comparison of the habitual wake time and melatonin phase between the age groups by use of a general linear model found a significant effect of age, a significant effect of melatonin phase and a significant interaction between melatonin phase and age.

An examination of the phase relationship between the average wake time and the timing of the midpoint of the circadian rhythm of plasma melatonin secretion revealed a significantly shorter interval between these two measures in the older subjects.

When the relationship between the melatonin rhythm and habitual sleep times was examined in more detail, the altered phase relationship in the older subjects was particularly evident when the offset of melatonin secretion was considered. This altered phase relationship between melatonin secretion and habitual sleep times in the older subjects was also evident when the onset of melatonin secretion was considered, although this did not reach statistical significance. Thus the phase relationship between habitual sleep times and the melatonin secretion pattern in the older subjects was such that they were going to bed earlier with respect to melatonin onset and waking earlier with respect to melatonin offset.

These results show that the older subjects were going to bed and waking up at an earlier clock hour, and these earlier bed and wake times were also at an earlier internal circadian phase.


The study revealed that the timing of the circadian rhythm of plasma melatonin secretion occurred at a significantly earlier clock hour in older subjects than in young adults, a finding consistent with previous reports of earlier circadian rhythms in older subjects in general. This earlier timing was evident regardless of whether circadian phase was estimated using the overall pattern of melatonin secretion or using only the onset or offset of the melatonin rhythm.

These findings do not support a causal role for melatonin phase in the sleep disruption associated with aging. Understanding the mechanisms underlying the age-related change in the relative timing between the circadian system and the habitual sleep-wake episode may aid in the development of chronobiological treatments for the sleep disruption and early morning awakening that affect so many older people.