Content uploaded by Bruce S Mcewen
Author content
All content in this area was uploaded by Bruce S Mcewen on Jun 01, 2015
Content may be subject to copyright.
Volume 338 Number 3
171
Seminars in Medicine of the
Beth Israel Deaconess Medical Center
J
EFFREY
S. F
LIER
, M.D.,
Editor
L
ISA
H. U
NDERHILL
,
Assistant Editor
Review Article
SEMINARS IN MEDICINE OF THE BETH ISRAEL DEACONESS MEDICAL CENTER
P
ROTECTIVE
AND
D
AMAGING
E
FFECTS
OF
S
TRESS
M
EDIATORS
B
RUCE
S. M
C
E
WEN
, P
H
.D.
From the Harold and Margaret Milliken Hatch Laboratory of Neuroen-
docrinology, Box 165, Rockefeller University, 1230 York Ave., New York,
NY 10021, where reprint requests should be addressed to Dr. McEwen.
©1998, Massachusetts Medical Society.
VER 60 years ago, Selye
1
recognized the
paradox that the physiologic systems activat-
ed by stress can not only protect and restore
but also damage the body. What links these seem-
ingly contradictory roles? How does stress influence
the pathogenesis of disease, and what accounts for
the variation in vulnerability to stress-related diseas-
es among people with similar life experiences? How
can stress-induced damage be quantified? These and
many other questions still challenge investigators.
This article reviews the long-term effect of the
physiologic response to stress, which I refer to as al-
lostatic load.
2
Allostasis — the ability to achieve sta-
bility through change
3
— is critical to survival.
Through allostasis, the autonomic nervous system,
the hypothalamic–pituitary–adrenal (HPA) axis, and
the cardiovascular, metabolic, and immune systems
protect the body by responding to internal and ex-
ternal stress. The price of this accommodation to
stress can be allostatic load,
2
which is the wear and
tear that results from chronic overactivity or under-
activity of allostatic systems.
O
THE PHYSIOLOGIC RESPONSE TO STRESS
Stressful experiences include major life events,
trauma, and abuse and are sometimes related to the
environment in the home, workplace, or neighbor-
hood. Acute stress (in the sense of “fight or flight”
or major life events) and chronic stress (the cumula-
tive load of minor, day-to-day stresses) can both
have long-term consequences. The effects of chronic
stress may be exacerbated by a rich diet and the use
of tobacco and alcohol and reduced by moderate ex-
ercise.
Genetic factors do not account for all the individ-
ual variability in sensitivity to stress, as evinced by
the lack of concordance between identical twins in
many disorders.
4,5
Moreover, genetic factors do not
explain the gradients of health across socioeconomic
levels in Western societies.
6
Two factors largely de-
termine individual responses to potentially stressful
situations: the way a person perceives a situation
7
and a person’s general state of physical health, which
is determined not only by genetic factors but also by
behavioral and lifestyle choices (Fig. 1). Whether
one perceives a situation as a threat, either psycho-
logical or physical, is crucial in determining the be-
havioral response — whether it is fleeing, fighting,
or cowering in fear — and the physiologic response
— calmness or heart palpitations and elevated corti-
sol levels.
The ability to adjust or habituate to repeated
stress is also determined by the way one perceives a
situation. For example, most people react initially to
the challenge of public speaking with activation of
the HPA axis. After repeated public speaking, how-
ever, most people become habituated and their cor-
tisol secretion no longer increases with the chal-
lenge. But approximately 10 percent of subjects
continue to find public speaking stressful, and their
cortisol secretion increases each time they speak in
public.
8
Others are prone to a cardiovascular stress
response, as shown by a recent study of cardiovascu-
lar responses to a stressful arithmetic test. Blood-
pressure responses to this experimental stress pre-
dicted elevated ambulatory blood pressure during
periods of perceived stress in everyday life.
9
Genetics
may also have a role in susceptibility to cardiovascu-
lar stress; many people whose blood pressure re-
mains elevated for several hours after the stress of an
arithmetic test have a parent with hypertension.
10
One’s physical condition has obvious implications
for one’s ability to mount an appropriate physiologic
response to stressful stimuli, and there may be a ge-
netic component to the response as well. In inbred
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
172
January 15, 1998
The New England Journal of Medicine
BioBreeding (BB) rats, an animal model of insulin-
dependent diabetes, exposure to repeated stress in-
creased the incidence of diabetes.
11
In children, fam-
ily instability increases the incidence and severity of
insulin-dependent diabetes.
12
Chronic stress, defined
as feelings of fatigue, lack of energy, irritability, de-
moralization, and hostility, has been linked to the
development of insulin resistance,
13
a risk factor for
non-insulin-dependent diabetes. Deposition of ab-
dominal fat, a risk factor for coronary heart disease
and diabetes,
14
is increased by the psychosocial stress
of colony reorganization in nonhuman primates
15
and may also be increased by stress in humans.
16
ALLOSTASIS AND ALLOSTATIC LOAD
In contrast to homeostatic systems such as blood
oxygen, blood pH, and body temperature, which
must be maintained within narrow ranges, allostatic
(adaptive) systems have much broader boundaries.
Allostatic systems enable us to respond to our phys-
ical states (e.g., awake, asleep, supine, standing, ex-
ercising) and to cope with noise, crowding, isola-
tion, hunger, extremes of temperature, danger, and
microbial or parasitic infection.
The core of the body’s response to a challenge —
whether it is a dangerous situation, an infection, liv-
ing in a crowded and unpleasant neighborhood, or
a public-speaking test — is twofold, turning on an
allostatic response that initiates a complex adaptive
pathway, and then shutting off this response when
the threat is past. The most common allostatic re-
sponses involve the sympathetic nervous systems and
the HPA axis. For these systems, activation releases
catecholamines from nerves and the adrenal medulla
and leads to the secretion of corticotropin from the
pituitary. The corticotropin, in turn, mediates the
release of cortisol from the adrenal cortex. Figure 2
shows how catecholamines and glucocorticoids af-
fect cellular events. Inactivation returns the systems
to base-line levels of cortisol and catecholamine se-
cretion, which normally happens when the danger is
past, the infection is contained, the living environ-
ment is improved, or the speech has been given.
However, if the inactivation is inefficient (see be-
low), there is overexposure to stress hormones. Over
weeks, months, or years, exposure to increased secre-
tion of stress hormones can result in allostatic load
2
and its pathophysiologic consequences.
Four situations are associated with allostatic load
(Fig. 3). The first and most obvious is frequent
stress. For example, surges in blood pressure can trig-
ger myocardial infarction in susceptible persons,
17
and in primates repeated elevations of blood pres-
sure over periods of weeks and months accelerate
atherosclerosis,
18
thereby increasing the risk of myo-
cardial infarction.
In the second type of allostatic load (Fig. 3), ad-
aptation to repeated stressors of the same type is
lacking, resulting in prolonged exposure to stress
hormones, as was the case for some of the people
subjected to the repeated-public-speaking challenge.
8
In the third type of allostatic load (Fig. 3) there
is an inability to shut off allostatic responses after a
stress is terminated. As we have noted, the blood
Figure 1.
The Stress Response and Development of Allostatic Load.
The perception of stress is influenced by one’s experiences, genetics, and behavior. When the brain perceives an experience as
stressful, physiologic and behavioral responses are initiated, leading to allostasis and adaptation. Over time, allostatic load can
accumulate, and the overexposure to mediators of neural, endocrine, and immune stress can have adverse effects on various organ
systems, leading to disease.
Allostatic load
Environmental stressors
(work, home, neighborhood)
Major life events Trauma, abuse
Individual
differences
(genes, development, experience)
Perceived stress
(threat,
helplessness,
vigilance) Behavioral
responses
(fight or flight;
personal behavior — diet,
smoking, drinking, exercise)
Physiologic
responses
Allostasis Adaptation
Allostatic load
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
SEMINARS IN MEDICINE OF THE BETH ISRAEL DEACONESS MEDICAL CENTER
Volume 338 Number 3
173
pressure in some people fails to recover after the
acute stress of an arithmetic test,
10
and hypertension
accelerates atherosclerosis.
18
Women with a history
of depressive illness have decreased bone mineral
density, because the allostatic load of chronic, mod-
erately elevated serum cortisol concentrations inhib-
its bone formation.
19
Intense athletic training also
induces allostatic load in the form of elevated sym-
pathetic and HPA-axis activity, which results in
weight loss, amenorrhea, and the often-related con-
dition of anorexia nervosa.
20,21
The failure to turn off the HPA axis and sympa-
thetic activity efficiently after stress is a feature of age-
related functional decline in laboratory animals,
22-24
but the evidence of this in humans is limited.
25,26
Stress-induced secretion of cortisol and catechola-
mines returns to base line more slowly in some aging
animals with other signs of accelerated aging,
22-24
and
the negative-feedback effects of cortisol are reduced
in elderly humans.
26
One other sign of age-related
impairment in rats is that the hippocampus fails to
turn off the release of excitatory amino acids after
stress,
27
and this may accelerate progressive structural
damage and functional impairment (see below).
One speculation is that allostatic load over a life-
time may cause the allostatic systems to wear out or
become exhausted.
25
A vulnerable link in the regu-
lation of the HPA axis and cognition is the hippo-
campal region. According to the “glucocorticoid-
cascade hypothesis,” wear and tear on this region of
the brain leads to dysregulation of the HPA axis and
cognitive impairment.
23,28
Indeed, some but not all
aging rats have impairment of episodic, declarative,
and spatial memory and hyperactivity of the HPA
axis, all of which can be traced to hippocampal dam-
age.
29
Recent data suggest that similar events may
occur in humans.
30,31
In the fourth type of allostatic load (Fig. 3), inad-
equate responses by some allostatic systems trigger
compensatory increases in others. When one system
does not respond adequately to a stressful stimulus,
the activity of other systems increases, because the
underactive system is not providing the usual coun-
terregulation. For example, if cortisol secretion does
not increase in response to stress, secretion of in-
flammatory cytokines (which are counterregulated
by cortisol) increases.
32
The negative consequences
of an enhanced inflammatory response are seen, for
example, in Lewis rats; these animals are very suscep-
tible to autoimmune and inflammatory disturbances,
because of a genetically determined hyporesponsive-
ness of the HPA axis.
33
In another model, rats that become subordinate
in a psychosocial living situation called the “visible-
burrow system” have a stress-induced state of HPA
hyporesponsiveness.
34,35
In these rats, the response
to stressors applied by the experimenter is very lim-
ited, and concentrations of corticotropin-releasing
Figure 2.
Allostasis in the Autonomic Nervous System and the
HPA Axis.
Allostatic systems respond to stress (upper panel) by initiating
the adaptive response, sustaining it until the stress ceases, and
then shutting it off (recovery). Allostatic responses are initiated
(lower panel) by an increase in circulating catecholamines
from the autonomic nervous system and glucocorticoids from
the adrenal cortex. This sets into motion adaptive processes
that alter the structure and function of a variety of cells and tis-
sues. These processes are initiated through intracellular recep-
tors for steroid hormones, plasma-membrane receptors, and
second-messenger systems for catecholamines. Cross-talk be-
tween catecholamines and glucocorticoid-receptor signaling
systems can occur.
Allostasis
Shut-off
Stress Recovery
Catecholamine
Glucocorticoid
Physiologic and
pathophysiologic
effects
Receptor
Cell
Second-
messenger
system
Steroid
receptor
Nucleus
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
174
January 15, 1998
The New England Journal of Medicine
Figure 3.
Three Types of Allostatic Load.
The top panel illustrates the normal allostatic response, in which a response is initiated by a stressor, sustained for an appropriate
interval, and then turned off. The remaining panels illustrate four conditions that lead to allostatic load: repeated “hits” from mul-
tiple stressors; lack of adaptation; prolonged response due to delayed shutdown; and inadequate response that leads to compen-
satory hyperactivity of other mediators (e.g., inadequate secretion of glucocorticoids, resulting in increased concentrations of cy-
tokines that are normally counterregulated by glucocorticoids).
Physiologic Response
Normal
Time
Physiologic Response
Stress
Activity Recovery
Allostatic load
Repeated “hits”
Time
Physiologic Response
Normal response repeated over time
Lack of adaptation
Time
Normal adaptation
Prolonged response
Time
Physiologic Response
No recovery
Inadequate response
Time
Physiologic Response
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
SEMINARS IN MEDICINE OF THE BETH ISRAEL DEACONESS MEDICAL CENTER
Volume 338 Number 3
175
hormone messenger RNA in the hypothalamus are
abnormally low.
36
Human counterparts with HPA
hyporesponsiveness include adults with fibromyal-
gia
37,38
and chronic fatigue syndrome
39,40
and chil-
dren with atopic dermatitis.
41
In post-traumatic
stress disorder, basal HPA activity is also low,
42,43
al-
though reactivity to stress may not be blunted.
Feelings of anticipation and worry can also con-
tribute to allostatic load.
44
Anticipation participates
in the reflex that prevents us from blacking out
when we get out of bed in the morning
3
and is also
part of worry, anxiety, and cognitive preparation for
a threat. Anticipatory anxiety can drive the secretion
of mediators like corticotropin, cortisol, and epi-
nephrine, and for this reason, prolonged anxiety and
anticipation are likely to result in allostatic load.
44
For example, salivary cortisol concentrations increase
within 30 minutes after waking in people who are
under considerable psychological stress due to work
or family matters.
45
In a related fashion, intrusive
memories of a traumatic event (as in post-traumatic
stress disorder) can produce a form of chronic stress
and can drive physiologic responses.
46
Allostasis and allostatic load are also affected by
the consumption of tobacco and alcohol, dietary
choices, and the amount of exercise (Fig. 1). These
forms of behavior are integral to the overall notion
of allostasis — the way people cope with a challenge
— and also contribute to allostatic load by known
pathways (e.g., a high-fat diet accelerates atheroscle-
rosis and progression to non-insulin-dependent dia-
betes by increasing cortisol secretion, leading to fat
deposition and insulin resistance
47
; smoking elevates
blood pressure and accelerates atherogenesis
48
; and
exercise protects against cardiovascular disease
49
).
EXAMPLES OF ALLOSTATIC LOAD
Cardiovascular and Metabolic Systems
The best-studied system of allostasis and allostatic
load is the cardiovascular system and its links to obe-
sity and hypertension. In nonhuman primates, the
incidence of atherosclerosis is increased among the
dominant males of unstable social hierarchies and in
socially subordinate females.
50,51
In humans, lack of
control on the job increases the risk of coronary
heart disease,
52
and job strain (high psychological
demands and lack of control) results in elevated am-
bulatory blood pressure at home and an increased
left-ventricular-mass index,
53
as well as increased pro-
gression of atherosclerosis.
54
Chronic stress (feelings
of fatigue, lack of energy, irritability, and demoral-
ization) and hostility are linked to increased reac-
tivity of the fibrinogen system and of platelets, both
of which increase the risk of myocardial infarc-
tion.
55,56
Quantifying allostatic load, a major challenge, has
been attempted with the use of measures of meta-
bolic and cardiovascular pathophysiology. In a re-
cent analysis,
57
data from the MacArthur Studies of
Successful Aging were used to assess eight measures
of increased activity of allostatic systems between
1988 and 1991. Allostatic load was approximated by
determining the number of measures for which a
person had values in the highest quartile from
among the following: systolic blood pressure, over-
night urinary cortisol and catecholamine excretion,
the ratio of the waist to the hip measurement, the
glycosylated hemoglobin value, and the ratio of se-
rum high-density lipoprotein in the total serum cho-
lesterol concentration; and the number of the fol-
lowing for which the person had values in the lowest
quartile: serum concentration of dehydroepiandros-
terone sulfate and serum concentration of high-den-
sity lipoprotein cholesterol. In cross-sectional analy-
ses of base-line data, subjects with higher levels of
physical and mental functioning had lower allostatic-
load scores and a lower incidence of cardiovascular
disease, hypertension, and diabetes. During the three
years of follow-up (1988 to 1991), people in this
higher-functioning group with higher allostatic-load
scores at base line were more likely to have incident
cardiovascular disease and were significantly more
likely to have declines in cognitive and physical func-
tioning. Among women in this group, increased cor-
tisol secretion predicted a decline in memory.
31
The Brain
Repeated stress affects brain function, especially in
the hippocampus, which has high concentrations of
cortisol receptors.
58
The hippocampus participates
in verbal memory and is particularly important for
the memory of “context,” the time and place of
events that have a strong emotional bias.
59,60
More-
over, glucocorticoids are involved in remembering
the context in which an emotionally laden event
took place.
61
Impairment of the hippocampus de-
creases the reliability and accuracy of contextual
memories. This may exacerbate stress by preventing
access to the information needed to decide that a
situation is not a threat.
62
The hippocampus also
regulates the stress response and acts to inhibit the
response of the HPA axis to stress.
63,64
The mechanism for stress-induced hippocampal
dysfunction and memory impairment is twofold.
First, acute stress increases cortisol secretion, which
suppresses the mechanisms in the hippocampus and
temporal lobe that subserve short-term memory.
65,66
Stress can impair memory in the short term, but
fortunately these effects are reversible and relatively
short-lived.
67
Second, repeated stress causes the atro-
phy of dendrites of pyramidal neurons in the CA3 re-
gion of the hippocampus through a mechanism in-
volving both glucocorticoids and excitatory amino
acid neurotransmitters released during and after
stress.
68
This atrophy is reversible if the stress is short-
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
176
January 15, 1998
The New England Journal of Medicine
lived, but stress lasting many months or years can kill
hippocampal neurons.
23,69
Magnetic resonance imag-
ing has shown that stress-related disorders such as
recurrent depressive illness, post-traumatic stress dis-
order, and Cushing’s disease are associated with atro-
phy of the hippocampus.
70,71
Whether this atrophy is
reversible or permanent is not clear.
Long-term stress also accelerates the appearance of
several biologic markers of aging in rats, including
the loss of hippocampal pyramidal neurons and the
excitability of pyramidal neurons in the CA1 region
by a calcium-dependent mechanism.
72
Glucocorti-
coids may mediate these effects by enhancing calci-
um currents in the hippocampus,
73
since calcium ions
have a key role in destructive as well as plastic proc-
esses in hippocampal neurons.
74-76
The persistent re-
lease of the excitatory amino acid glutamate in the
hippocampus after stress in aged rats may also con-
tribute to age-related neuronal damage
27
and may
potentiate atrophy and possibly even neuronal loss.
Early stress and neonatal handling influence the
course of aging and age-related cognitive impair-
ment in animals. Early experiences are believed to set
the level of responsiveness of the HPA axis and au-
tonomic nervous system. These systems overreact in
animals subjected to early unpredictable stress and
underreact in animals exposed to neonatal han-
dling.
77
In the former condition, aging of the brain
is accelerated, whereas in the latter, aging of the
brain is reduced.
29,77
The Immune System
The immune system responds to pathogens or
other antigens with its own form of allostasis that
may include an acute-phase response as well as the
formation of an immunologic “memory.” At the
same time, other allostatic systems, such as the HPA
axis and the autonomic nervous system, tend to
contain acute-phase responses and dampen cellular
immunity.
78
However, not all the effects are suppres-
sive. Acute stress causes lymphocytes and macro-
phages to be redistributed throughout the body and
to “marginate” on blood-vessel walls and within cer-
tain compartments, such as the skin, lymph nodes,
and bone marrow. This “trafficking” is mediated in
part by glucocorticoids.
78-82
If an immune challenge
is not encountered and the hormonal-stress signal
ceases, immune cells return to the bloodstream. When
a challenge occurs, however, as is the case in delayed-
type hypersensitivity, acute stress enhances the traf-
fic of lymphocytes and macrophages to the site of
acute challenge.
83,84
The immune-enhancing effects of acute stress de-
pend on adrenal secretion and last for three to five
days. Acute stress has the effect of calling immune
cells to their battle stations, and this form of allosta-
sis enhances responses for which there is an estab-
lished immunologic “memory.”
83-85
If the immuno-
logic memory is of a pathogen or a tumor cell, the
result of stress is presumably beneficial. If, on the
contrary, the immunologic memory leads to an au-
toimmune or allergic response, then stress is likely to
exacerbate a pathologic state. When allostatic load is
increased by repeated stress, the outcome is com-
pletely different; the delayed hypersensitivity response
is substantially inhibited
86
rather than enhanced. The
consequences of suppressed cellular immunity re-
sulting from chronic stress include increased severity
of the common cold, accompanied by increased ti-
ters of cold-virus antibody.
87
In laboratory animals,
repeated stress also leads to recurrent endotoxemia,
which decreases the reactivity of the HPA axis to a
variety of stimuli and decreases production of the
cytokine tumor necrosis factor a.88
Implications of Allostatic Load in Human Society
The gradients of health across the range of socio-
economic levels6 relate to a complex array of risk fac-
tors that are differentially distributed in human so-
ciety.89,90 Perhaps the best example is offered by the
Whitehall studies of the British civil service, in which
mortality and morbidity were found to increase
stepwise from the lowest to the highest of the six
grades of the British civil service.91 Hypertension
was a sensitive index of job stress,92 particularly
among factory workers, other workers with repeti-
tive jobs and time pressures,93 and workers whose
jobs were unstable because of departmental privati-
zation (Marmot MG: personal communication). Plas-
ma fibrinogen concentrations, which predict an in-
creased risk of death from coronary heart disease, are
elevated among men in the lower British civil-service
grades.56 In less stable societies, conflict and social
instability have been found to accelerate pathophys-
iologic processes and increase morbidity and mortal-
ity. For example, cardiovascular disease is a major
contributor to the increase of almost 40 percent
in the death rate among Russian men during the so-
cial collapse that followed the fall of Communism.94
Blood-pressure surges and sustained elevation are
linked to accelerated atherosclerosis18 as well as to an
increased risk of myocardial infarction.17
Another stress-linked change is abdominal obesity
(see above), measured as an increased waist-to-hip
ratio. The waist-to-hip ratio is increased at the lower
end of the socioeconomic-status gradient in Swedish
men95 and in the lower civil-service grades in the
Whitehall studies.96 Immune-system function is also
a likely target of psychosocial stress,97 increasing vul-
nerability to such infections as the common cold.87,98
Therapeutic Implications
A consideration of allostatic load is increasingly
important in the diagnosis and treatment of many
illnesses. Allostatic load is also important in illumi-
nating the relation between disease and social insta-
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
SEMINARS IN MEDICINE OF THE BETH ISRAEL DEACONESS MEDICAL CENTER
Volume 338 Number 3 177
bility, job loss, dangerous living environments, and
other conditions that are chronically stressful. Med-
ical illness itself is a source of stress, producing anx-
iety about prognosis, treatment, disability, and inter-
ference with social roles and relationships.
Physicians and other health care providers can
help patients reduce allostatic load by helping them
learn coping skills, recognize their own limitations,
and relax. Patients should also be reminded of the
interactions of a high-fat diet and stress in athero-
sclerosis, the role of smoking in cardiovascular dis-
ease and cancer, and the beneficial effects of exercise.
But the patients themselves must change their be-
havior patterns appropriately.99,100
Beyond these obvious steps, other types of inter-
ventions must be considered. Two important causes
of allostatic load appear to be isolation101 and lack of
control in the work environment.52 Interventions
that increase social support and enhance coping pro-
long the life spans of patients with breast cancer,102
lymphomas,103 and malignant melanoma.104 Inter-
ventions designed to increase a worker’s control over
his or her job, such as the reorganization of auto
production at Volvo, have also improved health and
attitudes toward work.93
DISCUSSION
DR. JEFFREY FLIER: Is there any known correla-
tion between lifelong stress (and therefore allostatic
load) and Alzheimer’s disease?
DR. MCEWEN: There are a few anecdotes from
admissions personnel at Veterans Affairs hospitals
but nothing concrete. It is interesting, however, that
education appears to have a “protective” role against
the development of Alzheimer’s disease.105 It is not
clear, though, whether education protects against
the disease or provides more redundancy in the
brain, which delays the symptoms.106
DR. BARBARA B. KAHN: What are the important
differences between men and women in the biology
of stress?
DR. MCEWEN: Estrogens appear to protect the
cardiovascular system, and at menopause, women’s
risk of cardiovascular disease increases to that of age-
matched men. The decline in estrogen secretion at
menopause also increases the activity of the HPA ax-
is,107 a development that has been linked to greater
cognitive decline among elderly women than among
elderly men.31 A decline in androgen secretion in
older men may affect HPA function, although to a
lesser extent. In rats, castration increases HPA activ-
ity.108 Finally, there are structural and functional dif-
ferences between the sexes in hippocampal for-
mation in rodents,109-111 and behavioral evidence
suggests functional and possibly structural sex differ-
ences in humans, as well.112 We do not yet know
whether these differences influence the vulnerability
of the hippocampus to severe stress, although a
number of studies now suggest that female rodents
and primates may be less vulnerable than males.69,113
DR. FLIER: Is there any evidence that humans are
more susceptible to the effects of stress than animals
because of the greater human capacity for cognition
and insight, as well as the human ability to feel guilt?
DR. MCEWEN: I believe that humans are more at
risk for allostatic load than animals, because of the
enormous individual differences in stress responsive-
ness and aging among humans, which relate to life
experiences, personality, and physiologic pheno-
type. However, stress responsiveness and aging also
differ among rats, so I don’t think we can be defin-
itive about the importance of cognition in our own
species.
A PHYSICIAN: What mechanisms underlie the dif-
ferences between immune responses to acute stress
and those to chronic stress?
DR. MCEWEN: These mechanisms are just begin-
ning to be understood. One key process is the re-
distribution, or trafficking, of immune cells. Acute
stress enhances this response to delayed-type hyper-
sensitivity. Chronic stress impairs delayed-type hy-
persensitivity, with the result that the blood is
depleted of fewer lymphocytes. The greater the im-
pairment of delayed-type hypersensitivity, the less
the blood is depleted of lymphocytes (Dhabhar FS:
unpublished data). Glucocorticoids are responsible
for the trafficking of lymphocytes and for the stress
enhancement of delayed-type hypersensitivity, but
they do not act alone. Various cytokines function as
more-local signals, emanating from a site of infec-
tion or challenge, and Dr. Firdaus Dhabhar at Rock-
efeller University is investigating their involvement.
Beyond that, it is well known that stress hormones
modulate immune function and influence the class
of the immune response by their ability to increase
the expression of some cytokines and decrease the
expression of others.78
I am indebted to the Health Program of the John D. and Cath-
erine T. MacArthur Foundation and its Network on Socioeconomic
Status and Health for contributions to the concepts discussed in this
article, and to Dr. Firdaus Dhabhar for assistance with Figure 3.
REFERENCES
1. Selye H. Syndrome produced by diverse nocuous agents. Nature 1936;
138:32.
2. McEwen BS, Stellar E. Stress and the individual: mechanisms leading to
disease. Arch Intern Med 1993;153:2093-101.
3. Sterling P, Eyer J. Allostasis: a new paradigm to explain arousal pathol-
ogy. In: Fisher S, Reason J, eds. Handbook of life stress, cognition and
health. New York: John Wiley, 1988:629-49.
4. Berg K. Genetics of coronary heart disease. Prog Med Genet 1983;5:
35-90.
5. Plomin R. The role of inheritance in behavior. Science 1990;248:183-8.
6. Adler NE, Boyce T, Chesney MA, et al. Socioeconomic status and
health: the challenge of the gradient. Am Psychol 1994;49:15-24.
7. Lazarus RS, Folkman S. Stress, appraisal and coping. New York: Spring-
er-Verlag, 1984.
8. Kirschbaum C, Prussner JC, Stone AA, et al. Persistent high cortisol re-
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
178 January 15, 1998
The New England Journal of Medicine
sponses to repeated psychological stress in a subpopulation of healthy men.
Psychosom Med 1995;57:468-74.
9. Matthews KA, Owens JF, Allen MT, Stoney CM. Do cardiovascular re-
sponses to laboratory stress relate to ambulatory blood pressure levels? Yes,
in some of the people, some of the time. Psychosom Med 1992;54:686-
97.
10. Gerin W, Pickering TG. Association between delayed recovery of
blood pressure after acute mental stress and parental history of hyperten-
sion. J Hypertens 1995;13:603-10.
11. Lehman C, Rodin J, McEwen BS, Brinton R. Impact of environmental
stress on the expression of insulin-dependent diabetes mellitus. Behav
Neurosci 1991;105:241-5.
12. Hagglof B, Blom L, Dahlquist G, Lonnberg G, Sahlin B. The Swedish
childhood diabetes study: indications of severe psychological stress as a risk
factor for t ype 1 (insulin-dependent) diabetes mellitus in childhood. Dia-
betologia 1991;34:579-83.
13. Raikkonen K, Keltikangas-Jar vinen L, Adlercreutz H, Hautenen A.
Psychosocial stress and the insulin resistance syndrome. Metabolism 1996;
45:1533-8.
14. Bjorntorp P. “Portal” adipose tissue as a generator of risk factors for
cardiovascular disease and diabetes. Arteriosclerosis 1990;10:493-6.
15. Jayo JM, Shively CA, Kaplan JR, Manuck SB. Effects of exercise and
stress on body fat distribution in male cynomolgus monkeys. Int J Obes
Relat Metab Disord 1993;17:597-604.
16. Moyer AE, Rodin J, Grilo CM, Cummings N, Larson LM, Rebuffe-
Scrive M. Stress-induced cortisol response and fat distribution in women.
Obes Res 1994;2:255-61.
17. Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of
onset of acute cardiovascular disease. Circulation 1989;79:733-43.
18. Kaplan JR, Pettersson K, Manuck SB, Olsson G. Role of sympathoa-
drenal medullary activation in the initiation and progression of atheroscle-
rosis. Circulation 1991;84:Suppl VI:VI-23–VI-32.
19. Michelson D, Stratakis C, Hill L, et al. Bone mineral density in women
with depression. N Engl J Med 1996;335:1176-81.
20. Boyar RM, Hellman LD, Roffwarg H, et al. Cortisol secretion and
metabolism in anorexia nervosa. N Engl J Med 1977;296:190-3.
21. Loucks AB, Mor tola JF, Girton L, Yen SSC. Alterations in the hypo-
thalamic-pituitary-ovarian and the hypothalamic-pituitary-adrenal axes in
athletic women. J Clin Endocrinol Metab 1989;68:402-11.
22. McCarty R. Sympathetic-adrenal medullary and cardiovascular re-
sponses to acute cold stress in adult and aged rats. J Auton Nerv Syst 1985;
12:15-22.
23. Sapolsky RM. Stress, the aging brain and the mechanisms of neuron
death. Cambridge, Mass.: MIT Press, 1992.
24. McEwen BS. Re-examination of the glucocorticoid hypothesis of stress
and aging. In: Swaab DF, Hofman MA, Mirmiran M, Ravid R, van Leeu-
wen FW, eds. Progress in brain research. Vol. 93. The human hypothalamus
in health and disease. Amsterdam: Elsevier Science, 1992:365-83.
25. Seeman TE, Robbins RJ. Aging and hypothalamic-pituitary-adrenal
response to challenge in humans. Endocr Rev 1994;15:233-60.
26. Wilkinson CW, Peskind ER, Raskind MA. Decreased hypothalamic-
pituitary-adrenal axis sensitivity to cortisol feedback inhibition in human
aging. Neuroendocrinology 1997;65:79-90.
27. Lowy MT, Wittenberg L, Yamamoto BK. Effect of acute stress on hip-
pocampal glutamate levels and spectrin proteolysis in young and aged rats.
J Neurochem 1995;65:268-74.
28. Sapolsky RM, K rey LC, McEwen BS. The neuroendocrinology of
stress and aging: the glucocorticoid cascade hypothesis. Endocr Rev 1986;
7:284-301.
29. Meaney MJ, Aitken DH, van Berkel C, Bhatnagar S, Sapolsky RM. Ef-
fect of neonatal handling of age-related impairments associated with the
hippocampus. Science 1988;239:766-8.
30. Lupien S, Lecours AR, Lussier I, Schwartz G, Nair NPV, Meaney MJ.
Basal cortisol levels and cognitive deficits in human aging. J Neurosci
1994;14:2893-903.
31. Seeman TE, McEwen BS, Singer BH, Albert MS, Rowe JW. Increase
in urinary cortisol excretion and memory declines: MacArthur studies of
successful aging. J Clin Endocrinol Metab 1997;82:2458-65.
32. Munck A, Guyre PM, Holbrook NJ. Physiological functions of gluco-
corticoids in stress and their relation to pharmacological actions. Endocr
Rev 1984;5:25-44.
33. Sternberg EM, Young WS III, Bernardini R, et al. A central nervous
system defect in biosynthesis of corticotropin-releasing hormone is associ-
ated with susceptibility to streptococcal cell wall-induced arthritis in Lewis
rats. Proc Natl Acad Sci U S A 1989;86:4771-5.
34. Blanchard DC, Sakai RR, McEwen BS, Weiss SM, Blanchard RJ. Sub-
ordination stress: behavioral, brain, and neuroendocrine correlates. Behav
Brain Res 1993;58:113-21.
35. McKittrick CR, Blanchard DC, Blanchard RJ, McEwen BS, Sakai RR.
Serotonin receptor binding in a colony model of chronic social stress. Biol
Psychiatry 1995;37:383-93.
36. Albeck DS, McKittrick CR, Blanchard DC, et al. Chronic social stress
alters expression of corticotropin-releasing factor and arginine vasopressin
mRNA in rat brain. J Neurosci 1997;17:4895-903.
37. Crofford LJ, Pillemer SR, Kalogeras KT, et al. Hypothalamic-pituitary-
adrenal axis perturbations in patients with fibromyalgia. Arthritis R heum
1994;37:1583-92.
38. Heim C, Ehlert U, Hanker J, Hellhammer DH. Abuse-related post-
traumatic stress disorder and alterations of the hypothalamo-pituitar y
adrenal axis in women with chronic pelvic pain. Psychosom Med (in
press).
39. Poteliakhoff A. Adrenocortical activity and some clinical findings in
acute and chronic fatigue. J Psychosom Res 1981;25:91-5.
40. Ur E, White PD, Grossman A. Hypothesis: cytokines may be activated
to cause depressive illness and chronic fatigue syndrome. Eur Arch Psychi-
atry Clin Neurosci 1992;241:317-22.
41. Buske-Kirschbaum A, Jobst S, Psych D, et al. Attenuated free cortisol
response to psychosocial stress in children with atopic dermatitis. Psycho-
som Med 1997;59:419-26.
42. Yehuda R, Giller EL, Southwick SM, Low y MT, Mason JW. Hypotha-
lamic-pituitary-adrenal dysfunction in posttraumatic stress disorder. Biol
Psychiatry 1991;30:1031-48.
43. Yehuda R, Teicher MH, Trestman RL, Levengood RA, Siever LJ. Cor-
tisol regulation in posttraumatic stress disorder and major depression: a
chronobiological analysis. Biol Psychiatr y 1996;40:79-88.
44. Schulkin J, McEwen BS, Gold PW. Allostasis, amygdala, and anticipa-
tory angst. Neurosci Biobehav Rev 1994;18:385-96.
45. Schulz P, Kirschbaum C, Pruessner J, Hellhammer DH. Increased free
cortisol secretion after wakening in chronically-stressed individuals due to
work overload. Stress Med (in press).
46. Baum A, Cohen L, Hall M. Control and intrusive memories as possi-
ble determinants of chronic stress. Psychosom Med 1993;55:274-86.
47. Brindley DN, Rolland Y. Possible connections between stress, diabetes,
obesity, hypertension and altered lipoprotein metabolism that may result
in atherosclerosis. Clin Sci 1989;77:453-61.
48. Verdecchia P, Schillaci G, Borgioni C, et al. Cigarette smoking, ambu-
latory blood pressure and cardiac hypertrophy in essential hypertension.
J Hypertens 1995;13:1209-15.
49. Bernadet P. Benefits of physical activity in the prevention of cardiovas-
cular diseases. J Cardiovasc Pharmacol 1995;25:Suppl 1:S3-S8.
50. Manuck SB, Kaplan JR, Adams MR, Clarkson TB. Studies of psycho-
social influences on coronary artery atherogenesis in cynomolgus monkeys.
Health Psychol 1988;7:113-24.
51. Shively CA, Clarkson TB. Social status and coronary artery atheroscle-
rosis in female monkeys. Arterioscler Thromb 1994;14:721-6.
52. Bosma H, Marmot MG, Hemingway H, Nicholson AC, Brunner E,
Stansfeld SA. Low job control and risk of coronary heart disease in White-
hall II (prospective cohort) study. BMJ 1997;314:558-65.
53. Schnall PL, Schwartz JE, Landsbergis PA, Warren K, Pickering TG.
Relation between job strain, alcohol, and ambulator y blood pressure. Hy-
pertension 1992;19:488-94.
54. Everson SA, Lynch JW, Chesney MA, et al. Interaction of workplace
demands and cardiovascular reactivity in progression of carotid atheroscle-
rosis: population based study. BMJ 1997;314:553-8.
55. Raikkonen K, Lassila R, Keltikangas-Jarvinen L, Hautanen A. Associ-
ation of chronic stress with plasminogen activator inhibitor-1 in healthy
middle-aged men. Arterioscler Thromb Vasc Biol 1996;16:363-7.
56. Markowe HLJ, Marmot MG, Shipley MJ, et al. Fibrinogen: a possible
link between social class and coronary heart disease. BMJ 1985;291:1312-
4.
57. Seeman TE, Singer BH, Rowe JW, Horwitz RI, McEwen BS. Price of
adaptation — allostatic load and its health consequences: MacArthur stud-
ies of successful aging. Arch Intern Med 1997;157:2259-68.
58. McEwen BS, De Kloet ER, Rostene W. Adrenal steroid receptors and
actions in the nervous system. Physiol Rev 1986;66:1121-88.
59. Eichenbaum H, Otto T, Cohen NJ. The hippocampus — what does
it do? Behav Neural Biol 1992;57:2-36.
60. LeDoux JE. In search of an emotional system in the brain: leaping
from fear to emotion and consciousness. In: Gazzaniga M, ed. The cogni-
tive neurosciences. Cambridge, Mass.: MIT Press, 1995:1049-61.
61. Pugh CR, Tremblay D, Fleshner M, Rudy JW. A selective role for cor-
ticosterone in contextual-fear conditioning. Behav Neurosci 1997;111:503-
11.
62. Sapolsky RM. Stress in the wild. Sci Am 1990;262:116-23.
63. Jacobson L, Sapolsky R. The role of the hippocampus in feedback reg-
ulation of the hypothalamic-pituitary-adrenocortical axis. Endocr Rev
1991;12:118-34.
64. Herman JP, Cullinan WE. Neurocircuitry of stress: central control of
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
SEMINARS IN MEDICINE OF THE BETH ISRAEL DEACONESS MEDICAL CENTER
Volume 338 Number 3 179
the hypothalamo-pituitary-adrenocortical axis. Trends Neurosci 1997;20:
78-84.
65. Kirschbaum C, Wolf OT, May M, Wippich W, Hellhammer DH.
Stress- and treatment-induced elevations of cortisol levels associated with
impaired declarative memory in healthy adults. Life Sci 1996;58:1475-83.
66. McEwen BS, Sapolsky RM. Stress and cognitive function. Curr Opin
Neurobiol 1995;5:205-16.
67. Lupien SJ, McEwen BS. The acute effects of corticosteroids on cogni-
tion: integration of animal and human model studies. Brain Res Brain Res
Rev 1997;24:1-27.
68. McEwen BS, Albeck D, Cameron H, et al. Stress and the brain: a par-
adoxical role for adrenal steroids. Vitam Horm 1995;51:371-402.
69. Uno H, Tarara R, Else JG, Suleman MA, Sapolsky RM. Hippocampal
damage associated with prolonged and fatal stress in primates. J Neurosci
1989;9:1705-11.
70. Sapolsky RM. Why stress is bad for your brain. Science 1996;273:749-
50.
71. McEwen BS, Magarinos AM. Stress effects on morphology and func-
tion of the hippocampus. Ann N Y Acad Sci 1997;821:271-84.
72. Kerr DS, Campbell LW, Applegate MD, Brodish A, Landfield PW.
Chronic stress-induced acceleration of electrophysiologic and morphomet-
ric biomarkers of hippocampal aging. J Neurosci 1991;11:1316-24.
73. Kerr DS, Campbell LW, Thibault O, Landfield PW. Hippocampal glu-
cocorticoid receptor activation enhances voltage-dependent Ca2 conduc-
tances: relevance to brain aging. Proc Natl Acad Sci U S A 1992;89:8527-
31.
74. Choi DW. Calcium-mediated neurotoxicity: relationship to specific
channel types and role in ischemic damage. Trends Neurosci 1988;11:465-
9.
75. Mills LR, Kater SB. Neuron-specific and state-specific differences in
calcium homeostasis regulate the generation and degeneration of neuronal
architecture. Neuron 1990;4:149-63.
76. Mattson MP. Calcium as sculptor and destroyer of neural circuitry. Exp
Gerontol 1992;27:29-49.
77. Meaney MJ, Tannenbaum B, Francis D, et al. Early environmental pro-
gramming; hypothalamic-pituitary-adrenal responses to stress. Semin
Neurosci 1994;6:247-59.
78. McEwen BS, Biron CA, Brunson KW, et al. The role of adrenocorti-
coids as modulators of immune function in health and disease: neural, en-
docrine and immune interactions. Brain Res Brain Res Rev 1997;23:79-
113.
79. Dhabhar FS, Miller AH, Stein M, McEwen BS, Spencer RL. Diurnal
and acute stress-induced changes in distribution of peripheral blood leu-
kocyte subpopulations. Brain Behav Immun 1994;8:66-79.
80. Dhabhar FS, Miller AH, McEwen BS, Spencer RL. Effects of stress on
immune cell distribution: dynamics and hormonal mechanisms. J Immunol
1995;154:5511-27.
81. Miller AH, Spencer RL, Hassett J, et al. Effects of selective type I and
II adrenal steroid agonists on immune cell distribution. Endocrinology
1994;135:1934-44.
82. Herbert TB, Cohen S. Stress and immunity in humans: a meta-analytic
review. Psychosom Med 1993;55:364-79.
83. Dhabhar FS. Stress-induced enhancement of antigen-specific cell-
mediated immunity: the role of hormones and leukocyte trafficking. New
York: Rockefeller University, 1996.
84. Dhabhar FS, McEwen BS. Stress-induced enhancement of antigen-spe-
cific cell-mediated immunit y. J Immunol 1996;156:2608-15.
85. Dhabhar FS, Miller AH, McEwen BS, Spencer RL. Stress-induced
changes in blood leukocyte distribution: role of adrenal steroid hormones.
J Immunol 1996;157:1638-44.
86. Dhabhar FS, McEwen BS. Moderate stress enhances, and chronic stress
suppresses, cell-mediated immunity in vivo. Soc Neurosci 1996;22:1350.
abstract.
87. Cohen S, Tyrrell DAJ, Smith AP. Psychological stress and susceptibility
to the common cold. N Engl J Med 1991;325:606-12.
88. Hadid R, Spinedi E, Giovambattista A, Chautard T, Gaillard RC. De-
creased hypothalamo-pituitary-adrenal axis response to neuroendocrine
challenge under repeated endotoxemia. Neuroimmunomodulation 1996;3:
62-8.
89. Taylor SE, Repetti RL, Seeman T. Health psychology: what is an un-
healthy environment and how does it get under the skin? Annu Rev Psy-
chol 1997;48:411-47.
90. Lynch JW, Kaplan GA, Cohen RD, Tuomilehto J, Salonen JT. Do car-
diovascular risk factors explain the relation between socioeconomic status,
risk of all-cause mortality, cardiovascular mortality, and acute myocardial
infarction? Am J Epidemiol 1996;144:934-42.
91. Marmot MG, Smith GD, Stansfeld S, et al. Health inequalities among
British civil servants: the Whitehall II study. Lancet 1991;337:1387-93.
92. Pickering TG, Devereux RB, James GD, et al. Environmental influenc-
es on blood pressure and the role of job strain. J Hypertens Suppl 1996;
14:S179-S185.
93. Melin B, Lundberg U, Soderlund J, Granqvist M. Psycholog ical and
physiological stress reactions of male and female assembly workers: a com-
parison between two different forms of work organization. J Organizat
Psychol (in press).
94. Bobak M, Marmot M. East-West mortality divide and its potential ex-
planations: proposed research agenda. BMJ 1996;312:421-5.
95. Larsson B, Seidell J, Svardsudd K, et al. Obesity, adipose tissue distri-
bution and health in men — the study of men born in 1913. Appetite
1989;13:37-44.
96. Brunner EJ. The social and biological basis of cardiovascular disease in
office workers. In: Blane D, Brunner EJ, Wilkinson RG, eds. Health and
social organization. London: Routledge, 1996:272-313.
97. Cohen S, Kaplan JR, Cunnick JE, Manuck SB, Rabin BS. Chronic so-
cial stress, affiliation and cellular immune response in nonhuman primates.
Psychol Sci 1992;3:301-4.
98. Cohen S, Doyle WJ, Skoner DP, Rabin BS, Gwaltney JM Jr. Social ties
and susceptibility to the common cold. JAMA 1997;277:1940-4.
99. Redelmeier DA, Rozin P, Kahneman D. Understanding patients’ deci-
sions: cognitive and emotional perspectives. JAMA 1993;270:72-6.
100. Horwitz RI, Hor witz SM. Adherence to treatment and health out-
comes. Arch Intern Med 1993;153:1863-8.
101. Seeman TE, McEwen BS. The impact of social environment charac-
teristics on neuroendocrine regulation. Psychosom Med 1996;58:459-
71.
102. Spiegel D, Bloom JR, Kraemer HC, Gottheil E. Effect of psychoso-
cial treatment on survival of patients with metastatic breast cancer. Lancet
1989;2:888-91.
103. Richardson JL, Shelton DR, Krailo M, Levine AM. The effect of
compliance with treatment on survival among patients with hematologic
malignancies. J Clin Oncol 1990;8:356-64.
104. Fawzy FI, Fawzy NW, Hyun CS, et al. Malignant melanoma: effects
of an early structured psychiatric intervention, coping, and affective state
on recurrence and survival 6 years later. Arch Gen Psychiatry 1993;50:681-
9.
105. Stern Y, Alexander GE, Prohovnik I, et al. Relationship between life-
time occupation and parietal flow: implications for a reserve against Alz-
heimer’s disease pathology. Neurology 1995;45:55-60.
106. Stern Y, Alexander GE, Prohovnik I, Mayeux R. Inverse relationship
between education and parietotemporal perfusion deficit in Alzheimer’s
disease. Ann Neurol 1992;32:371-5.
107. Van Cauter E, Leproult R, Kupfer DJ. Effects of gender and age on
the levels and circadian rhythmicity of plasma cortisol. J Clin Endocrinol
Metab 1996;81:2468-73.
108. Handa RJ, Nunley KM, Lorens SA, L ouie JP, McGivern RF, Bollnow
MR. Androgen regulation of adrenocorticotropin and corticosterone se-
cretion in the male rat following novelty and foot shock stressors. Physiol
Behav 1994;55:117-24.
109. Juraska JM, Fitch JM, Henderson C, R ivers N. Sex differences in the
dendritic branching of dentate granule cells following differential experi-
ence. Brain Res 1985;333:73-80.
110. Gould E, Westlind-Danielsson A, Frankfurt M, McEwen BS. Sex dif-
ferences and thyroid hormone sensitivity of hippocampal pyramidal cells.
J Neurosci 1990;10:996-1003.
111. Roof RL. The dentate gyrus is sexually dimorphic in prepubescent
rats: testosterone plays a significant role. Brain Res 1993;610:148-51.
112. Kimura D. Sex differences in the brain. Sci Am 1992; 267:118-25.
113. Mizoguchi K, Kunishita T, Chui DH, Tabira T. Stress induces neu-
ronal death in the hippocampus of castrated rats. Neurosci Lett 1992;138:
157-60.
The New England Journal of Medicine
Downloaded from www.nejm.org at UNIVERSITY SYSTEM OF MARYLAND on September 30, 2010. For personal use only. No other uses without permission.
Copyright © 1998 Massachusetts Medical Society. All rights reserved.
