0021-972X/01/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2001 by The Endocrine Society
Vol. 86, No. 4
Printed in U.S.A.
Middle-Aged Men Show Higher Sensitivity of Sleep to
the Arousing Effects of Corticotropin-Releasing
Hormone Than Young Men: Clinical Implications
ALEXANDROS N. VGONTZAS, EDWARD O. BIXLER, ANNMARIE M. WITTMAN,
KEITH ZACHMAN, HUNG-MO LIN, ANTONIO VELA-BUENO, ANTHONY KALES,
AND GEORGE P. CHROUSOS
Sleep Research and Treatment Center, Department of Psychiatry (A.N.V., E.O.B., A.M.W., A.K.), and
Department of Health Evaluation Sciences (H.-M.L.), Pennsylvania State University, Hershey,
Pennsylvania 17033; Pediatric and Reproductive Endocrinology Branch, National Institute of Child
Health and Human Development, National Institutes of Health (K.Z., G.P.C.), Bethesda, Maryland
20892; and Department of Psychiatry, Autonomous University (A.V.-B.), Madrid, Spain
ABSTRACT
The prevalence of insomnia associated with emotional stress increases markedly in middle-age. Both the top and end hormones of the
hypothalamic-pituitary-adrenal axis, i.e. CRH and glucocorticoids,
stimulate arousal/wakefulness and inhibit slow wave (deep) sleep in
experimental animals and man. The objective of this study was to test
the hypothesis that middle-age is characterized by increased sensitivity to the sleep-disturbing effects of the hypothalamic-pituitaryadrenal axis.
We studied 12 healthy middle-aged (45.1 ⫾ 4.9) and 12 healthy
young (22.7 ⫾ 2.8) men by monitoring their sleep by polysomnography
for 4 consecutive nights, including in tandem 1 adaptation and 2
baseline nights and a night during which we administered equipotent
doses of ovine CRH (1 ␮g/kg, iv bolus) 10 min after sleep onset.
Analyses included comparisons within and between groups using
multiple ANOVA and regression analysis.
S
Although both middle-aged and young men responded to CRH with
similar elevations of ACTH and cortisol, the former had significantly
more wakefulness and suppression of slow wave sleep compared with
baseline sleep; in contrast, the latter showed no change. Also, comparison of the change in sleep patterns from baseline to the CRH night
in the young men to the respective change observed in middle-aged
men showed that middle-age was associated with significantly higher
wakefulness and significantly greater decrease in slow wave sleep
than in young age.
We conclude that middle-aged men show increased vulnerability of
sleep to stress hormones, possibly resulting in impairments in the
quality of sleep during periods of stress. We suggest that changes in
sleep physiology associated with middle-age play a significant role in
the marked increase of prevalence of insomnia in middle-age. (J Clin
Endocrinol Metab 86: 1489 –1495, 2001)
LOW WAVE SLEEP (SWS) has an inhibitory effect on the
activity of the hypothalamic-pituitary-adrenal (HPA) axis
(1–3). In a recent study we demonstrated that increased depth
of sleep after 1 night of total sleep deprivation was associated
with significantly decreased cortisol levels (4). Middle-age is
associated with a sharp decline in SWS; however, the potential
association of this physiological change with the activity of the
HPA axis is not known. The HPA axis activity changes little
with age in carefully selected healthy men (5), whereas it increases after middle-age in unselected populations of men (6),
perhaps associated with medical comorbidity.
Middle-age is associated with increased prevalence of insomnia, and reported emotional stress is the most frequent
underlying cause of insomnia complaints (7). Indeed, up to
40% of the general population report difficulties in falling or
staying asleep in middle-age in contrast to about 20% of
young subjects (7). It has been suggested that the increased
prevalence of insomnia in middle-age is due to increased life
stress during this period of life. Another, as yet unexplored
possibility is that it is due to increased sensitivity of sleep to
the arousal-producing effects of stress.
The administration of CRH has been used widely as a
sensitive means to evaluate the activity of the HPA axis in
many physiological and pathological conditions (8). The effect of CRH, which is a potent arousal-promoting agent, on
human sleep has been assessed in a few studies, only in
young individuals, with inconclusive results (9, 10). On the
other hand, even though glucocorticoids have arousalpromoting properties at pharmacological doses, little is
known about the actions of endogenous cortisol elevations
around the time of sleep onset. The goal of our study was to
assess whether middle-aged, healthy men are different from
young men in terms of response to CRH administration
during sleep. We hypothesized that middle-aged men’s sleep
would be more vulnerable to the arousing effects of CRH
and/or cortisol.
Received September 13, 2000. Revision received November 29, 2000.
Accepted December 9, 2000.
Address all correspondence and requests for reprints to: Alexandros
N. Vgontzas, M.D., Sleep Research and Treatment Center, Department
of Psychiatry, Pennsylvania State University College of Medicine, 500
University Drive, Hershey, Pennsylvania 17033. E-mail: [email protected].
Subjects
Subjects and Methods
Twelve young healthy men, 20 –28 yr of age (mean ⫾ sd, 22.7 ⫾ 2.8),
and 12 middle-aged healthy men, 37–54 yr of age (45.1 ⫾ 4.9), were
recruited from the community and from the medical and technical staff
and students of the Milton S. Hershey Medical Center (Hershey, PA). A
1489
1490
JCE & M • 2001
Vol. 86 • No. 4
VGONTZAS ET AL.
thorough medical assessment, including physical history and examination, and a detailed sleep history were completed for each subject. The
subjects were in good general health, had no sleep complaints or circadian abnormalities, were not under any stress, were not taking any
medications, and were screened in the Sleep Laboratory for sleep disordered breathing, nocturnal myoclonus, or other primary sleep disorders. Also, a battery of clinical tests, including full blood count, urinalysis, thyroid indexes, and electrocardiogram, were negative for
abnormal findings. The two groups were not different in terms of body
mass index (26.2 ⫾ 2.8 vs. 25.3 ⫾ 2.2 for middle-aged and young subjects,
respectively; P ⫽ NS). The study was approved by the institutional
review board, and each subject signed a written informed consent.
Protocol
Each subject spent 4 consecutive nights in the Sleep Laboratory (1
adaptation night, 2 baseline nights, followed by the night of CRH administration). During the fourth night an indwelling catheter was inserted in the antecubital vein about 60 min before the start of the sleep
recording. The catheter was kept patent with small amounts of heparin.
During the sleep-recording period, blood collection and CRH administration took place outside the subjects’ rooms through a perforation in
the wall via 12-ft tubing, to decrease sleep disturbance from the blooddrawing technique. A standard dose of ovine CRH (1 ␮g/kg) was
administered by iv bolus during the first sleep cycle (⬃10 min after sleep
onset, as determined by standard polysomnographic criteria) (11).
ACTH and cortisol were sampled at ⫺10 min (time of sleep onset), 0 min
(time of CRH administration), 5 min, 15 min, and 30 min post-CRH and
every 30 min for the remainder of the night. Also, urinary free cortisol
was measured in two consecutive (baseline days 2 and 3) 24-h complete
urine collections. The 24-h urine collections were completed the morning
before the nighttime administration of CRH.
Sleep recordings
Sleep laboratory recording was carried out in a sound-attenuated,
light- and temperature-controlled room that had a comfortable bedroom-like atmosphere. During this evaluation, each subject was monitored continuously for 8 h by electroencephalogram (EEG), electromyogram, and electrooculogram according to standard methods. The
8-h period in bed was based on individual subject’s habitual times of
onset of sleep, which ranged between 2200 –2300 h. The average start and
stop times of sleep recordings for middle-aged and young men were
2201 ⫾ 13 vs. 2200 ⫾ 2 min and 0559 ⫾ 13 vs. 0601 ⫾ 3 min, respectively.
The sleep records were scored independently of any knowledge of the
experimental condition according to standardized criteria (11). Each
night in the laboratory, the participants completed a single seven-point
subjective questionnaire assessing levels of anxiety and stress during the
day.
Sleep parameters assessed from the sleep recordings were grouped
into three categories: sleep efficiency measures, amount of sleep stages,
and additional rapid eye movement (REM) variables. Sleep efficiency
measures included sleep induction (sleep latency), sleep maintenance
[wake time after sleep onset (WTASO), total wake time (sum of sleep
latency and WTASO), and percentage of sleep time (total sleep time as
percentage of time in bed). The duration of sleep stages [REM, 1, 2, SWS
(3 and 4 combined)] was expressed as absolute minutes or as percentage
of total sleep time, which is calculated as the minutes in each stage as
the percentage of total sleep time. The additional REM variables included REM latency (the interval from sleep onset to the first REM
period), REM interval (average time lapsing between two consecutive
REM periods), and REM density (calculated by counting the number of
eye movements in a 40-s epoch divided by the number of epochs within
this REM period). The onset of sleep was established by the presence of
any sleep stage for 1 min or longer. However, if the initial stage of sleep
was stage 1, it had to be followed without any intervening wakefulness
by at least 60 s of stage 2, 3, 4, or REM. The distribution of wakefulness,
SWS, and REM sleep through the night was examined by halves of the
night. One half of the night was established by subtracting sleep latency
from the total amount of laboratory time and then dividing the remaining time into equal halves. Data from nights 2 and 3 were averaged
(baseline).
Hormone assays
Blood collected from the indwelling catheter was transferred to an
ethylenediamine tetraacetate-containing tube and refrigerated until centrifugation (within 3 h). The supernatant was frozen at ⫺20 C until
hormone assay. ACTH and cortisol levels were measured by specific
immunoassay techniques as previously described (12). The lower limit
of detection was 5 pg/mL for ACTH and 0.7 ␮g/dL for cortisol. The
intra- and interassay coefficients were, respectively, 4.6% and 6.0% for
cortisol and 10.0% and 12.0% for ACTH.
Statistical analysis
The basal plasma ACTH and cortisol concentrations for each individual were taken as the mean value at ⫺10 and 0 min. The average times
of blood collection for the two basal hormonal measures were 2228 ⫾ 16
and 2238 ⫾ 16 min for middle-aged men and 2227 ⫾ 16 and 2237 ⫾ 16
min for young men, respectively. The total and net amount of timeintegrated ACTH and cortisol secretion after the administration of CRH
were calculated by the trapezoid method [integrated area under the
curve (AUC)]. The total time-integrated ACTH and cortisol responses
were expressed as the area beneath the concentration-time curve from
0 – 450 min. The net integrated ACTH and cortisol responses were expressed as the area beneath the concentration-time curve from 0 – 450
min minus the area corresponding to the mean of the two baseline values
multiplied by 450 min. The peak ACTH and cortisol responses corresponded to the highest levels of ACTH and cortisol achieved after CRH
administration. Serial nighttime plasma ACTH and cortisol levels were
analyzed simultaneously using multiple ANOVA (MANOVA) for repeated measures over time (mixed effects model, with fixed effects being
time and the two groups, and random effects being the subjects), followed by the Dunnett post-hoc test. Differences in terms of sleep values
between baseline (the average across nights 2 and 3 in the sleep laboratory) and night 4 within each group were examined using paired
two-tailed Student’s t test. To assess differences between young and
middle-aged men in terms of their responses to CRH, we compared the
change from baseline to night 4 (CRH night) of the young men to the
change from baseline to night 4 of the middle-aged men using regression
analysis, where the outcome is the change in score, and the main predictor is the age group. In this analysis we also covaried the respective
baseline sleep measures to control for possible differences in the baseline. In this part of the analysis, data are presented as the least square
mean ⫾ se. The remainder of the data are expressed as the mean ⫾ se,
except for age, body mass index, and time, which are expressed as the
mean ⫾ sd.
Results
Baseline sleep profiles
The middle-aged compared with the young men demonstrated markedly and significantly lower amounts of slow
wave sleep during the 2 baseline nights (3.7 ⫾ 1.0% vs. 11.4 ⫾
1.8%; P ⬍ 0.01; Table 1). There were no differences between
the two groups in terms of overall sleep efficiency measures,
i.e. percentage of sleep time or total wake time. However,
middle-aged men compared with young men demonstrated
a shorter sleep latency (10.1 ⫾ 1.4 vs. 23.3 ⫾ 5.1 min; P ⬍ 0.05),
whereas they had more wake time after sleep onset (46.5 ⫾
8.6 vs. 25.9 ⫾ 5.0 min; P ⬍ 0.05). There were no differences
in terms of daytime anxiety/stress levels within the groups
(baseline vs. CRH night, 1.9 vs. 1.5 in the middle-aged and 1.9
vs. 2.0 in the young) or between groups during baseline (1.9
vs. 1.9) or CRH (1.5 vs. 2.0) nights.
Basal function of HPA axis
There were no differences between the two groups in
terms of average 2-day 24-h urinary free cortisol excretion
(222.5 ⫾ 18.1 nmol/24 h in the middle-aged vs. 245.6 ⫾ 26.9
CRH DISRUPTS THE SLEEP OF MIDDLE-AGED MEN
1491
TABLE 1. Comparison of baseline vs. CRH sleep patterns within young and middle-aged men
Young
Sleep efficiency
SL (min)
WTASO (min)
TWT (min)
% ST
Wake time (min)/half of night
First
Second
Sleep stages
% Stage 1
% Stage 2
% Slow wave
% REM
REM distribution
REM latency (min)
REM interval (min)
REM duration (min)
No. of REM periods
REM density
Middle-aged
Baseline nights 2–3
CRH night 4
Baseline nights 2–3
CRH night 4
23.3 ⫾ 5.1
25.9 ⫾ 5.0
49.2 ⫾ 6.8
89.7 ⫾ 1.4
24.1 ⫾ 4.9
40.5 ⫾ 8.0
64.5 ⫾ 9.3
86.6 ⫾ 1.9
10.1 ⫾ 1.4
46.5 ⫾ 8.6
56.7 ⫾ 9.1
88.1 ⫾ 1.9
22.0 ⫾ 2.1a
85.8 ⫾ 16.3b
107.8 ⫾ 16.0b
77.6 ⫾ 3.7b
11.1 ⫾ 3.7
14.8 ⫾ 2.5
11.9 ⫾ 2.3
28.5 ⫾ 7.2b
19.0 ⫾ 3.1
27.5 ⫾ 6.2
42.9 ⫾ 11.1b
42.9 ⫾ 8.9
5.3 ⫾ 0.7
63.8 ⫾ 2.3
11.4 ⫾ 1.8
19.6 ⫾ 1.5
9.8 ⫾ 2.8
61.0 ⫾ 2.7
13.3 ⫾ 2.0
16.0 ⫾ 1.7a
5.5 ⫾ 0.9
71.9 ⫾ 1.8
3.7 ⫾ 1.0
19.0 ⫾ 1.9
7.4 ⫾ 1.3
72.8 ⫾ 2.5
3.3 ⫾ 1.4
16.4 ⫾ 1.6
113.0 ⫾ 12.2
109.3 ⫾ 5.6
21.9 ⫾ 1.8
3.9 ⫾ 0.2
1.3 ⫾ 0.2
113.2 ⫾ 21.2
128.0 ⫾ 13.0
19.5 ⫾ 2.5
3.7 ⫾ 0.3
1.8 ⫾ 0.3
104.7 ⫾ 10.7
107.5 ⫾ 3.6
20.7 ⫾ 2.0
3.9 ⫾ 0.1
1.6 ⫾ 0.2
126.5 ⫾ 18.5
119.7 ⫾ 8.7
17.6 ⫾ 1.9
3.6 ⫾ 0.2
1.6 ⫾ 0.3
Data are the mean ⫾ SE; baseline values represent averaged data from nights 2 and 3. P values denote differences between baseline and
CRH nights within each age group. SL, Sleep latency; TWT, total wake time; % ST, percent sleep time.
a
P ⬍ 0.01.
b
P ⬍ 0.05.
nmol/24 h in the young; P ⫽ NS). The baseline mean plasma
ACTH and cortisol values were similar in the middle-aged
and young men [2.1 ⫾ 0.5 vs. 1.7 ⫾ 0.2 pmol/L (P ⫽ NS) and
151.4 ⫾ 44.3 vs. 94.5 ⫾ 22.6 nmol/L (P ⫽ NS) for the middleaged and young, respectively; Fig. 1].
Ovine (o) CRH-stimulated ACTH and cortisol secretion
during sleep
The mean post-CRH level of cortisol was significantly,
albeit slightly, lower in the middle-aged than in the young
men (350.4 ⫾ 22.1 vs. 375.2 ⫾ 24.8 nmol/L; P ⬍ 0.05, by
MANOVA). Also, the net amount of time-integrated (AUC)
cortisol for the entire night was significantly lower in the
middle-aged than in the young men (90,894.7 ⫾ 17,897.8 vs.
138,974.9 ⫾ 14,448.5 nmol/L䡠min; P ⬍ 0.05; Fig. 1). However,
the total AUCs of cortisol were similar in the middle-aged
and young men (150,720.7 ⫾ 9,345.1 vs. 180,115.0 ⫾ 12,250.1
nmol/L䡠min; P ⫽ NS). The net and total AUC of ACTH were
also similar in the middle-aged and young men [1,644.2 ⫾
273.5 vs. 2,159.9 ⫾ 344.5 pmol/L䡠min (P ⫽ NS); 2,509.8 ⫾
2,79.4 vs. 2,913.6 ⫾ 391.9 (P ⫽ NS), respectively; Fig. 1].
Effect of oCRH administration on sleep EEG in young and
middle-aged men
The administration of CRH in middle-aged men was associated with a significant increase in wakefulness compared
with the baseline (wake time after sleep onset and total wake
time were increased, whereas percent sleep time was decreased; all P ⬍ 0.05; Table 1 and Fig. 2A). In contrast, CRH
was not associated with a significant change in wakefulness
compared with baseline in young men (wake time after sleep
onset, total wake time, and percent sleep time; Table 1 and
Fig. 2B). The major impact in terms of wakefulness in middleaged men occurred during the first half of the night, when
CRH-stimulated ACTH and cortisol secretion peaked [42.9 ⫾
11.1 vs. 19.0 ⫾ 3.1 min in middle-aged men (P ⬍ 0.05); 11.9 ⫾
2.4 min vs. 11.1 ⫾ 3.7 min in young men (P ⫽ NS)]. Also, the
amount of SWS in middle-aged men tended to decrease
during the first half of the night (7.5 ⫾ 3.0 vs. 14.4 ⫾ 4.0 min;
P ⬍ 0.09), whereas there was no change in the amount of SWS
in young men.
In young men there was a significant suppression of percentage of REM sleep during the night of CRH administration compared with the baseline (16.0 ⫾ 1.7 vs. 19.6 ⫾ 1.5; P ⬍
0.01), whereas in middle-aged men the percentage of REM
sleep for the entire night was nonsignificantly decreased.
REM density was unaffected by the administration of CRH
in either group (Table 1).
Comparison of the effect of oCRH administration on sleep
EEG between young and middle-aged men
A comparison of the change from baseline to the CRH
night in the young men to the respective change observed in
middle-aged men after adjusting for baseline differences
showed that middle-age was associated with significantly
higher wakefulness compared with youth (Fig. 3). Specifically, middle-age was associated with a significantly higher
increase in WTASO (47.1 ⫾ 17.2 vs. 8.7 ⫾ 8.4 min; P ⬍ 0.05)
and total wake time (54.2 ⫾ 16.4 vs. 12.2 ⫾ 9.3 min; P ⬍ 0.05)
and a decrease in percentage of sleep time (⫺11.2 ⫾ 3.4% vs.
⫺2.5 ⫾ 1.9%; P ⬍ 0.05). Also, the effect on wakefulness was
stronger in the first half of the night, as indicated by the
significant increase in WTASO during the first half of the
night (26.6 ⫾ 11.1 vs. ⫺1.9 ⫾ 2.4 min; P ⬍ 0.05), but not during
the second half of the night. Furthermore, middle-aged men
demonstrated a significantly higher decrease in SWS during
the first half of the night (⫺10.7 ⫾ 4.0 vs. 10.0 ⫾ 5.0 min;
P ⬍ 0.05). No differences were found between the groups in
terms of the effects of CRH on the remaining sleep stages,
including amount of REM sleep and REM variables.
1492
VGONTZAS ET AL.
JCE & M • 2001
Vol. 86 • No. 4
FIG. 1. ACTH and cortisol responses after the administration of 1 ␮g/kg oCRH 10 min after sleep onset in young (䡺) and middle-aged (f) healthy
men. Inset, Net integrated nighttime AUC in young (䡺) and middle-aged (f) men.
Correlation between sleep disturbance and ACTH and
cortisol elevations
Correlations between indexes of sleep disturbance, i.e.
change in SWS (baseline minus CRH night) and peak concentrations of cortisol after CRH administration, showed a
significant correlation between change in SWS and peak
cortisol levels in middle-aged men (rxy ⫽ 0.6; P ⬍ 0.05), but
not in young men (rxy ⫽ 0.1; P ⫽ 0.7; Fig. 4). Correlations
between WTASO and cortisol peaks were in the same direction. No correlations were observed between sleep disturbance and peak ACTH levels.
Discussion
The administration of CRH was associated with a significantly increased wakefulness in the middle-aged, but not the
young, men. This increase was more pronounced in the first
half of the night, when CRH-induced secretion of cortisol and
ACTH peaked. Also, the administration of CRH in middleaged, but not in young, men caused a significant reduction
of SWS during the early part of the night. These findings
suggest that the sleep of middle-aged men compared with
that of the young is more vulnerable to the arousing/waking
effect of equipotent doses of exogenous CRH. The latter
probably exerts its arousing effect directly after crossing into
the central nervous system at sites where this is possible
(periventricular organs) and/or indirectly through elevations of plasma cortisol. CRH is a potent stimulus of the locus
caeruleus, and its central administration in experimental animals is associated with increased wakefulness and inhibition of SWS (13, 14). Increased wakefulness is the most frequent placebo-controlled side-effect of glucocorticoids used
on a short-term basis at therapeutic doses (15).
Our observations of the effect of CRH on sleep are consistent with the sleep disturbances (decreased SWS, increased wakefulness) observed in hypercortisolemic patients
with melancholic depression (16, 17) and in patients with
Cushing syndrome (18) and with an early report that decreased adrenal corticosteroids in normal subjects or patients
with Addison’s disease were associated with a significant
increase in ␦ (deep) sleep (19). In humans the nighttime
administration of human CRH either in a pulsatile mode
(four injections of 50 ␮g or hourly injection of 10 ␮g through
the night) or as a constant infusion (30 ␮g/h) in young
healthy men either resulted in a small decrease in SWS (9) or
had no effect on any sleep parameters (10, 20). Two recent
studies showed either no significant effect on sleep (50 ␮g
oCRH, iv bolus) (21) or a significant decrease in SWS and
sleep efficiency in young males (100 ␮g hCRH, iv bolus) (22).
CRH DISRUPTS THE SLEEP OF MIDDLE-AGED MEN
1493
FIG. 2. A, Sleep histogram in a middle-aged man at baseline (night 2, top) and during the night of CRH administration (night 4, bottom). Note
the marked decrease in SWS and increase in wakefulness, particularly during the first half of the night. B, Sleep histogram in a young man
at baseline (night 2, top) and during the night of CRH administration (night 4, bottom). Note that sleep structure, with the exception of reduced
REM sleep, is not altered.
Thus, the majority of the studies, including ours, despite the
differences in type of CRH used (ovine vs. human), dose,
timing, and type of administration (bolus vs. constant infusion) suggest that the sleep of young individuals is rather
resistant to the arousing effects of CRH.
Blood-drawing procedures may have a mild disturbing
effect on healthy subjects’ sleep (23) (our unpublished data).
Thus, it is possible that in this study the presence of the
catheter and the awareness of iv injection of CRH might have
had an impact on the subjects’ sleep. This impact was more
apparent in the middle-aged group, who showed an increase
in sleep latency. However, the facts that 1) there was a positive correlation between sleep disturbance and elevation of
cortisol in the middle-aged, but not in the young, and 2) sleep
disturbances were more pronounced in the middle-aged
men, although the cortisol elevation was smaller than that in
young men support our conclusion that middle-aged men
showed increased vulnerability to stress hormones.
CRH administration was associated with a suppression of
the amount of REM sleep, which was significant in young
men, whereas REM density was unaffected. These findings
are consistent with previous reports that showed a reduction
in the amount of REM sleep, but not REM density, induced
by exogenously administered CRH (9), prednisone (24), or
FIG. 3. Comparisons of changes in wakefulness and SWS from baseline (nights 2 and 3) to the night of CRH administration (night 4)
between young (䡺) and middle-aged (f) healthy men after adjusting
for baseline value. Values represent the least square mean ⫾ SE.
ⴱ, P ⬍ 0.05.
1494
JCE & M • 2001
Vol. 86 • No. 4
VGONTZAS ET AL.
with increased sensitivity to arousal-producing stress hormones, such as CRH and cortisol. This suggests that a middle-aged individual compared with a young individual is at
a significantly higher risk of developing insomnia when
faced with equivalent stressors. Second, our findings may
explain at least partially why major depression, a condition
of increased production of CRH and cortisol, is associated in
middle-age with insomnia, whereas in the young it is frequently associated with sleepiness (16, 17). Third, our study
explains the common experience that widely used stimulants, i.e. caffeinated beverages, have a stronger sleepdisturbing effect in middle-age than in the young.
Acknowledgments
We thank the nursing staff of the General Clinical Research Center at
Pennsylvania State College of Medicine and Elaine Mallios, R.N., for
their technical assistance, and Barbara Green for her assistance with
word processing and overall preparation of the manuscript.
References
FIG. 4. Correlations between change in SWS (baseline minus CRH
night) and peak cortisol response post-CRH in middle-aged men (bottom) and young (top) men.
hydrocortisone (25). We have previously demonstrated that
24-h urinary free cortisol excretion correlated significantly
and positively with the amount of REM sleep and not with
REM density in normal volunteers, suggesting that REM
sleep is associated with an endogenous activation of the HPA
axis (26). The reduction of REM sleep after CRH administration may be the result of elevated cortisol levels on REM
sleep as previously described (24, 25), possibly through a
negative feedback regulation loop. An alternative explanation is that the REM reduction may reflect the fact that in
young men a small increase in wakefulness occurred primarily during the second half of the sleep period when REM
is more prevalent.
Our findings have several clinical implications. First, the
increased prevalence of insomnia in middle-age may, in fact,
be the result of deteriorating sleep mechanisms associated
1. Weitzman ED, Zimmerman JC, Czeisler CA, Ronda J. 1983 Cortisol secretion
is inhibited during sleep in normal man. J Clin Endocrinol Metab. 56:352–358.
2. Bierwof C, Struve K, Marshall L, Born J, Fehm H. 1997 Slow wave sleep drives
inhibition of pituitary-adrenal secretion in humans. Neurophysiology.
9:479 – 484.
3. Scheen AJ, Byrne MM, Plat L, Leproult R, Van Cauter E. 1997 Relationships
between sleep quality and glucose regulation in normal humans. Am J Physiol.
271:E261–E270.
4. Vgontzas AN, Mastorakos G, Bixler EO, Kales A, Gold PW, Chrousos GP.
1999 Sleep deprivation effects on the activity of the hypothalamic-pituitaryadrenal and growth axes: potential clinical implications. Clin Endocrinol (Oxf).
51:205–215.
5. Pavlov GP, Harman S Mitchell, Chrousos GP, Loriaux L, Blackman MR. 1986
Responses of plasma adrenocorticotropin, cortisol, and dehydroepiandrosterone to ovine corticotropin-releasing hormone in healthy aging men. J Clin
Endocrinol Metab. 62:767–772.
6. Van Cauter E, Leproult R, Kupfer DJ. 1996 Effects of gender and age on the
levels of circadian rhythmicity of plasma cortisol. J Clin Endocrinol Metab.
81:2468 –2473.
7. Kales A, Kales JD. 1984 Evaluation and treatment of insomnia. New York:
Oxford University Press.
8. Chrousos GP, Gold PW. 1992 The concept of stress and stress system disorders. JAMA. 267:1244 –1252.
9. Holsboer F, von Bardeleben U, Steiger A. 1988 Effects of intravenous corticotropin-releasing hormone upon sleep-related growth hormone surge and
sleep EEG in man. Neuroendocrinology. 48:32–38.
10. Born J, Späth-Schwalbe E, Schwakenhofer H, Kern W, Fehm HL. 1989 Influences of corticotropin-releasing hormone, adrenocorticotropin, and cortisol
on sleep in normal man. J Clin Endocrinol Metab. 68:904 –911.
11. Rechtschaffen A, Kales A. 1968 A manual of standardized terminology, techniques, and scoring system for sleep stages of human subjects. Bethesda: NIH;
publication 204.
12. Yanovski JA, Cutler GB, Chrousos GP, Nieman LK. 1993 Corticotropinreleasing hormone stimulation following low-dose dexamethasone administration: a new test to distinguish Cushings syndrome from pseudo-Cushings
states. JAMA. 269:2232–2238.
13. Ehlers C, Reed TK, Henriksen SJ. 1986 Effects of corticotropin-releasing factor
and growth hormone-releasing factor on sleep and activity in rats. Neuroendocrinology. 42:467– 474.
14. Opp M, Obál Jr F, Krueger JM. 1989 Corticotropin-releasing factor attenuates
interleukin 1-induced sleep and fever in rabbits. Am J Physiol. 257:R528 –R535.
15. Chrousos GP, Kattah JC, Beck RW, Cleary PA, Optic Neuritis Study Group.
1993 Side effects of glucocorticoid treatment. JAMA. 269:2110 –2112.
16. Gold PW, Goodwin FK, Chrousos GP. 1988a Clinical and biochemical manifestations of depression: relationship to the neurobiology of stress (part 1).
N Engl J Med. 319:348 –353.
17. Gold PW, Goodwin FK, Chrousos GP. 1988b Clinical and biochemical manifestations of depression: relationship to the neurobiology of stress (part 2).
N Engl J Med. 319:413– 420.
18. Friedman TC, Garcia-Borreguero D, Hardwick D, et al. 1994 Decreased
␦-sleep and plasma delta-sleep-inducing peptide in patients with Cushing
syndrome. Neuroendocrinology. 60:626 – 634.
CRH DISRUPTS THE SLEEP OF MIDDLE-AGED MEN
19. Gillin JC, Jacobs LS, Snyder F, Henkin RI. 1974 Effects of decreased adrenal
corticosteroids: changes in sleep in normal subjects and patients with adrenal
cortical insufficiency. Electroencephalogr Clin Neurophysiol. 36:283–289.
20. Kellner M, Yassouridis A, Manz B, Steiger A, Holsboer F, Wiedemann K.
1997 Corticotropin-releasing hormone inhibits melatonin secretion in healthy
volunteers: a potential link to low-melatonin syndrome in depression? Neuroendocrinology. 65:284 –290.
21. Mann K, Roschke J, Benkert O, et al. 1995 Effects of corticotropin-releasing
hormone on respiratory parameters during sleep in normal men. Exp Clin
Endocrinol Diabetes. 103:233–240.
22. Tsuchiyama Y, Uchimura N, Sakamoto T, Maeda H, Kotorii T. 1995 Effects of hCRH
on sleep and body temperature rhythms. Psychiatry Clin Neurosci. 49:299–304.
1495
23. Kerkhofs M, Linkowski P, Mendlewicz J. 1989 Effects of intravenous catheter
on sleep in healthy men and in depressed patients. Sleep. 12:113–119.
24. Gillin JC, Jacobs LS, Fram DH, Snyder F. 1972 Acute effect of a glucocorticoid
on normal human sleep. Nature. 237:398 –399.
25. Garcia-Borreguero D, Schwartz PE, Barker C, Barbato G, Wehr T. 1994 Effects
of suppressing the daily cortisol rhythm on the REM-sleep circadian rhythm.
Preliminary findings. Sleep Res. 23:495.
26. Vgontzas AN, Bixler EO, Papanicolaou DA, et al. 1997 Rapid eye movement
sleep correlates with the overall activities of the hypothalamic-pituitaryadrenal axis and sympathetic system in healthy humans. J Clin Endocrinol
Metab. 82:3278 –3280.