Tuesday, October 01, 2002

Worried Sick

The End of Stress as We Know It

By: , Bruce S. McEwen, Ph.D., and Elizabeth Norton Lasley

Stress begins in the brain and apparently only humans can generate stress just by thinking. In this excerpt fromThe End of Stress as We Know It, an internationally known researcher on stress introduces a concept, perhaps new to most readers, that may help us to understand how our body’s mechanism for coping with threats can end up threatening our health. Allostasis (literally, changing to maintain stability) is the body’s response to danger, real or perceived—a complex interaction of brain, endocrine system, and immune system that readies us to fight or flee. The duration and intensity of this response, explains McEwen, creates allostatic load: an accumulated burden on our bodies as systems evolved to combat emergencies are kept active. When that occurs, we become vulnerable to stress-related illnesses from the common cold to heart disease and memory loss.


Excerpted from The End of Stress as We Know It by Bruce S. McEwen, Ph.D., with Elizabeth N. Lasley. Co-published by The Dana Press and Joseph Henry Press. ©2002 by Bruce S. McEwen, Ph.D., with Elizabeth N. Lasley. All rights reserved. Reprinted with permission. The full text of this book can be found at the Joseph Henry web site, www.jhpress.org. 


Allostasis sounds like a disease in its own right, but it’s a term that came into being in the early 1980s as a newer and more appreciative way to view the body’s swift and efficient methods of dealing with danger. To understand it properly we need to backtrack a bit and invoke another biological term, one that might be more familiar: homeostasis.

Homeostasis is often described as an organism’s need to maintain a steady internal state. It comes from the Greek roots homeo, meaning same, and stasis, meaning stable— remaining stable by staying the same.

The word homeostasis was first used in the mid-19th century by a French scientist named Claude Bernard. He introduced the thinking that eventually led to the concept, and science, of stress. Bernard emphasized the body’s need to maintain a constant state, what he called le milieu internel. Many bodily conditions must not only remain constant, they must stay the same, or at least remain within rigidly proscribed limits. Body temperature is one of these: a fever of 100 is a legitimate reason to stay in bed, a fever of 104 a potentially life-threatening situation. Other homeostatic systems include the blood’s acid-base balance and oxygen content and the amount of oxygen that reaches the brain. If these things change by much, we die. 

Luckily, however, only a few areas of our existence are confined within such narrow straits. Obviously our circumstances are changing constantly; we live in a world of change, and the body must be able to respond. Bernard and the scientists who followed him knew that many of the body’s systems can and do operate within an astonishingly wide range of parameters. Heartbeat, breathing, the amount of glucose in the blood, and the amount of energy currently stored as fat are all things that can change quite quickly. Some debate exists in the scientific community as to whether these systems, too, should fall under the heading of homeostasis—they do, after all, help maintain constancy of a sort. But I prefer the newer term, allostasis. It comes from the Greek root allo, meaning variable, and it emphasizes the point that allostatic systems help keep the body stable by being themselves able to change. Nowhere are these changes more dramatic than in the systems that comprise the stress response.


Allostasis is produced by a swift and intricately organized system of communication. It links the brain, which perceives a novel or threatening situation; the endocrine system (chiefly the adrenal glands), which are primarily responsible for mobilizing the rest of the body; and the immune system for internal defense. Allostasis is often thought of as the fight-or-flight response because, taken to the extreme, it prepares for just those two eventualities. The main idea is to get maximum energy to those parts of the body that need it the most.


Teleosts exemplify the true fight-or-flight response. Salmon, for example, have need of both fight and flight. Though they started off as predators, as evolution proceeded, they also became prey to larger fish, birds, bears, and humans. In preparation for any challenge from the environment, salmon have the same set of responses as humans do—involving the brain, cardiovascular system, glands, and immune system. And in salmon, as in humans, these responses pivot around the HPA axis [hypothalmic-pituitaryadrenal axis].

The HPA axis makes its appearance very dramatically in teleost fish like the salmon. Earlier fish, such as the lamprey, which may be the most primitive vertebrate alive today, have a few brain regions and hormones that may be slightly similar. But in the salmon the entire system is suddenly fully operational, and the ensuing 400 million years of evolution have not made many substantive changes. A salmon has a hypothalamus, a pituitary gland, and kidneys containing adrenal-like tissue that produces cortisol. Salmon have very well defined immune systems, including white blood cells called T cells, which search out and destroy invaders such as bacteria, and they have B cells, which form antibodies for future protection. In salmon, as in humans, cortisol replenishes glucose reserves and inhibits the immune system. Salmon have the so-called autonomic nervous system, which regulates involuntary processes by sending its branches into the organs, blood vessels, and glands. They also have a rudimentary limbic system, the counterpart of the brain structures that process emotions in humans.

According to Carl Shreck, zoologist at Oregon State University, salmon present a paradigm shift in the concept of stress— from just any threat to homeostasis to a condition that specifically activates the stress response systems, particularly the HPA axis. A bacterium safeguards its homeostasis by producing heat shock proteins, but in salmon an event or situation can be called stressful only if it sets in motion the precise chain of nervous system and hormonal responses associated with stress. Salmon will avoid anything that sets off the HPA axis. “Virtually anything that can ‘scare’ the fish can cause the full range of stress responses,” says Shreck, “a bad smell or slapping the water to sound like a bird striking the surface. Humans may never know how (or even if) salmon experience fear, but if the HPA axis is triggered, the perception of risk is there as well.”

If the salmon’s fight-or-flight response is invoked repeatedly by circumstances, the final aspect of stress emerges: allostatic load, when the systems designed for protection and adaptation become overwhelmed. Overcrowding is a stressor for salmon, stimulating all of the stress responses— overactivating the adrenal glands and stifling the immune system. Overcrowding also halts the maturation process in young salmon. Salmon are born in fresh river water and swim down river to the ocean where they live as adults. The transformation of the fish from river dwellers into creatures that can live in saltwater is called smoltification. Recall that under conditions of extreme stress growth is a luxury that is postponed until the crisis has passed. Among overcrowded fish, smoltification does not occur.

The most striking example of allostatic load in salmon, though, is the one stressor that the fish do not try to avoid: the trek back upriver to the waters of their birth, where they spawn the next generation.

The journey can be as much as 1,000 miles, depending on the species. The salmon take their time, averaging 25 miles a day, but they don’t swim constantly; sometimes they rest in the river for a month or so. The trip can take up to nine months and is initiated by a rise in cortisol.

Migrating salmon show signs of severe stress. When they begin their migration, they stop feeding and the digestive tract atrophies. By the time they spawn, the high levels of cortisol have exhausted their stores of energy and devastated their immune systems; most dying salmon show massive infection. Prolonged exposure to cortisol is the presumed cause of death.


Death by cortisol is the most clear-cut example of the “bad” side of the allostatic process; it’s also the most extreme. In most species, allostasis kicks in to help the animal adjust to changes in the environment. John Wingfield, a zoologist at the University of Washington, suspects that elevations in cortisol actually produce the behavioral changes through which birds, reptiles, and mammals cope with their world. This brings us back to one of our central dogmas: allostasis is what equips us to deal with stress—a concept played out again and again in the animal kingdom.

So if allostasis is a robust dynamic coping mechanism that’s been around for 500 million years, how did humans—during the blink of an eye in geological time—manage to turn it into a recipe for stress-related disease? The issue is neatly implied in the title of a book by neuroendocrinologist Robert Sapolsky of Stanford University: Why Zebras Don’t Get Ulcers. A zebra’s stress response kicks in when the zebra is chased by a lion; when the zebra escapes (or gets eaten), the stress response shuts off. In between predation attempts, the zebra is at ease; it doesn’t flood itself with stress hormones wondering when the next lion is going to show up.

Animals can experience allostatic load in a quite “human” fashion when subjected to repeated or ongoing stress. In Hans Selye’s experiments of the 1930s, rats exposed to a variety of stressful treatments got ulcers, and monkeys and dogs can get them too. Animals can show stress-related wear and tear even in the wild. But in general they tend not to experience allostatic load because once a stressful situation is over, the stress response subsides. For the most part, only humans can keep the HPA axis going indefinitely. We can do this because of how our faculties of perception, thought, and emotion are produced in the brain and how they are connected to the stress response.


Stress begins in the brain. As far as we know, only humans can become stressed out from things that exist in idea only, such as by performing a mental rehash of an old argument until it sparks the release of adrenaline and activates the autonomic nervous system, causing a rise in fibrinogen and aggravating a case of clogged arteries.


Chronic stress takes its toll most immediately on the heart. When the fight-or-flight response swings into action, one of the most important things the animal (or person) must do is to move quickly. So the first of the stress hormones, adrenaline, courses through the sympathetic nervous system to step up the heart rate, increasing blood pressure to drive more oxygen to the large muscles of the arms and legs. But these sudden surges, when activated too often, cause damage to the blood vessels in the coronary arteries. The sites of damage are places where the arteries become clogged with the sticky buildup that sets the stage for atherosclerosis.

In humans, too many sudden escalations in blood pressure can trigger myocardial infarctions (heart attacks) in blood vessels that have become clogged. Studies with primates show that stress can aggravate the clogging-up process. Jay Kaplan of Bowman Gray University made into an experiment the social disruption the USDA had proposed for laboratory monkeys. After weeks and months of shuffling the groups around, forcing the monkeys repeatedly to scramble for position, the resulting elevations in blood pressure sped up the process of atherosclerosis and increased the risk of heart attacks. (Sapolsky and other scientists cited Kaplan’s work when protesting the USDA’s suggestion.)

Studies of social disruption in monkeys have parallels in the world of human doings. A striking example was observed in the British Civil Service (a fount of information about stress-related diseases) during the years of privatization under then-Prime Minister Margaret Thatcher. In a classic series called the Whitehall studies, blood pressure was found to be lowest among the highest grade of employees and highest among the rank and file (food for thought for those who believe that stress plagues only fast-track yuppies). But one study focused on a department that was undergoing privatization during the years of Margaret Thatcher; in this department researchers found increases in body mass index, need for sleep, incidence of stroke, and cholesterol.


There are also situations in which, though the stress itself is not lengthy or severe, the body responds in a way that is inappropriate. Often we perceive a situation as stressful not because it imperils our survival but because it is unfamiliar and a bit challenging—a new job, perhaps, or a newly assumed position of leadership in the community. Once the alarming becomes commonplace, we cease to activate the fight-or-flight sequences. In some people, though, when an initially threatening experience becomes business as usual, the stress response doesn’t get the message.

Clemens Kirschbaum and colleagues at the University of Trier subjected a group of 20 male volunteers to some of the most threatening events imaginable—public speaking and mental arithmetic in front of an audience for five consecutive days. The investigators measured the cortisol in saliva samples taken from the subjects on each day. Most of the men became comfortable “on stage” by the second day, and their cortisol levels diminished accordingly. But seven of them showed no discernible difference in their reported discomfort or in their cortisol levels between days 1 and 2 and only a very slight decrease by day 5. Perhaps not coincidentally, a personality questionnaire revealed that the still-stressed men had low self-confidence and self-esteem. In short, the men were not able to habituate to an experience that was no longer novel. Although no particular diseases or symptoms were matched up to this group (which Kirschbaum dubbed “high responders”), it’s likely they were overexposing their bodies to stress hormones under many circumstances in daily life that other people might not consider stressful. And we know, for example, that frequent jumps in blood pressure are a risk for cardiovascular disease.


Most of us can probably remember repeatedly rehashing an argument or other stressful scene, getting worked up all over again until our friends grew tired of hearing about it. For some people this condition is chronic; they continue to mount an allostatic response long after the stressful event has ended, and here there is evidence that genes may play a role. Bill Gerin and Tom Pickering of the Hypertension Center at Cornell University gave an arithmetic test to more than 500 undergraduate volunteers, measuring heart rate and blood pressure before, during, and after the test. They were interested in seeing whether differences in cardiovascular response could be attributed to the students’ race, sex, or parental history of hypertension. It turned out that none of these factors had anything to do with cardiovascular response while the test was going on. But afterwards, a certain percentage of the students still showed elevated blood pressure. Even though the stress was over, their systems were not able to recover and return to baseline. Most of these subjects had two parents with hypertension, suggesting they were genetically ill disposed to let go of a stressful situation.

Failure to shut off the sympathetic nervous system and the hypothalamic-pituitary-adrenal (HPA) axis when appropriate may also be a function of aging, as animal studies show, although there is less evidence for this in humans. In some aging laboratory animals, stress-induced secretions of both cortisol and adrenaline return to baseline more slowly than in normal animals. In humans the “negative feedback” effects of cortisol (by which it tells the brain to ease off on some aspects of the stress response) don’t work as well in the elderly.


All of the above three scenarios involve long-term overexposure to adrenaline and cortisol, whether it’s because the stress itself goes on too long, because the system cannot accommodate the fact that the situation should no longer be stressful, or because the shutoff processes are not functioning. Stress hormones do more than raise blood pressure and stifle the immune system. For example, cortisol is often chronically elevated in depression, and some women with a history of depressive illness have decreased bone mineral density. This is because bone formation is one of those long-term luxuries that get played down as part of the fight-or-flight response; cortisol actually interferes with the processes by which bone is formed. Prolonged exposure to cortisol has been shown to create allostatic load in women undergoing intense athletic training; though exercise may not seem stressful to the athlete, when carried to an abusive extreme it elevates both the sympathetic nervous system and the HPA axis. Results can include weight loss, lack of menstruation, and anorexia, a condition often related to exercise extremism.

Chronically elevated cortisol can also dampen the effects of insulin, and indeed chronic stress—defined as feelings of fatigue, lack of energy, irritability, demoralization, and hostility—has been linked to the development of insulin resistance, a risk factor for type II or non-insulin-dependent diabetes.

It’s also possible that overactivation of the stress response over a lifetime may undermine the whole process of allostasis itself, causing the systems to wear out and become exhausted. One particularly vulnerable link is the hippocampus, which helps turn off the HPA axis after stress and is also a nexus of memory and cognition. Because the hippocampus is rich in receptors for cortisol, using levels of this hormone to play its “checks-and-balances” role in the stress response, it is one of the first targets when levels of cortisol get too high. According to the so-called glucocorticoid cascade hypothesis, when the hippocampal region of the brain is flooded with cortisol, the resulting wear and tear leads to both an improperly functioning HPA axis and cognitive impairment.


The idea of checks and balances in the stress response brings us to the final way in which the protective systems of allostasis can trigger the damage of allostatic load: when the stress response is insufficient, resulting in underproduction of the stress hormones, particularly cortisol, wear and tear can also result.

How can this be? Surely if there are no stress hormones, there must be no stress and consequently no stress-related illness. But like most of human physiology, it isn’t quite that simple. Cortisol acts somewhat like a thermostat; in fact, it clamps down on its own production. It slows the production of the two hormones that touch off the HPA axis: corticotropin-releasing factor in the hypothalamus and adrenocorticotropic hormone in the pituitary. Cortisol also reins in the immune system and reduces inflammation and swelling from tissue damage.

When one of the participants in a checks-and-balances arrangement isn’t doing its job, the others may go overboard in doing theirs. In some people, allostatic load takes the form of a sluggish response by the adrenals and a subsequent lack of sufficient cortisol. The most immediate result is that the immune system, without cortisol’s steadying hand, runs wild and reacts to things that do not really pose a threat to the body. Allergies are one example of this process. In most people the immune system does not put things like dust and cat dander on a par with pathogenic (disease-causing) bacteria. But in people prone to allergies, the immune system goes on red alert in the presence of such usually innocuous substances, throwing everything it’s got at the irritants: uncontrollable sneezing to expel the invaders, mucous secretion to entrap them, swelling caused by the influx of white blood cells to the infected area, pain, redness, and general misery. All of these symptoms are reduced by the action of cortisol.

Asthma is another example in which the small tubes called bronchioles in the lungs swell and constrict. Once again, the oversensitized system is trying to ward off things that are not actually harmful (such as dust, cold, and exercise), in this case by barring the portals of access to the lungs. Allergies and asthma are both considered inflammatory diseases, and they are classic signs of the type of allostatic load signaled by the underproduction of cortisol. People who suffer from these conditions notice that their symptoms worsen when they are under stress. Other kinds of inflammatory disorders are the so-called autoimmune diseases, in which the immune system fails in its prime directive—distinguishing “self” from “nonself”—and goes after the person’s own body tissue. These conditions, too, are normally prevented by cortisol (and often treated with cortisol by doctors).

Rashes are a prime example of the immune system attacking healthy skin; one type, atopic dermatitis, in children is a sign of both stress and an underresponsive HPA axis. Other autoimmune disorders, often exacerbated by stress, are rheumatoid arthritis, in which the joints are chronically inflamed, and multiple sclerosis, a degenerative disease in which the immune system destroys a part of the nervous system known as the myelin sheath.

A feeble HPA response can often manifest itself in conditions not always immediately associated with the immune system. Fibromyalgia, for example, is a condition of chronic pain that most doctors consider psychosomatic (and some consider imaginary, though the patients certainly don’t). The connection with the immune system and cortisol becomes clear when we consider that pain is a part of the inflammatory response; pain warns us that there’s a problem and encourages us to leave the affected area alone until the problem is resolved. But in many chronic pain states, as with other inflammatory disorders, there is no apparent threat. Rather, the system is responding in a maladaptive way, which the available supply of cortisol is too low to prevent.


It’s important to remember that allostatic load is more than the experience of being under stress. It also reflects our lifestyle and ways of coping with daily life. What we eat, if we smoke, how well we sleep, and whether we exercise all feed into the final common path that is the production of cortisol, adrenaline, and other cast members in the allostatic scenario.

To a certain degree, we ourselves can determine whether allostasis will slide into allostatic load by the ways in which we cope with stress. If we make poor choices, we can tilt the scales in favor of stress-related illness. For example, smoking (a front-line defense for many) elevates blood pressure and accelerates clogging of the coronary arteries, thereby raising the risk of both heart attack and stroke. Finding solace in high-fat snacks, such as doughnuts or potato chips, can also lead to health problems. A high-fat diet accelerates atherosclerosis and increases cortisol secretion. Increased cortisol, in turn, steps up the accumulation of body fat, which is a risk factor for cardiovascular disease, stroke, and diabetes.

On the other hand, if we counteract stress with a brisk walk or a visit to the health club, we can increase the odds in our favor.

Exercise prevents the buildup of body fat, protects against cardiovascular disease, and reduces chronic pain and depression. We can also protect ourselves by seeking company and support. Sheldon Cohen, who studies the relationship between stress and upper respiratory disease, has reported that people with many social connections get fewer colds. Ronald Glaser and Janice Kiecolt-Glaser of Ohio State University have shown that isolation can undermine the activity of the immune system.

With so many ways of going wrong, it may sound as if the fight-or-flight response is a fragile thing, but actually it’s quite resilient. In the past 10 years, research into the effects of stress on the cardiovascular, immune, and nervous systems has shown in detail what can happen when allostasis gets out of kilter, while at the same time indicating just how resilient these systems really are.  

About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Helen Mayberg, M.D., Icahn School of Medicine at Mount Sinai 
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine
Charles Zorumski, M.D., Washington University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.