Stress may start outin the mind, but the body is quick to respond. As Esther Sternberg has shown, the consequences can prove deadly. Ayala Ochert reports

A packed conference room was asked: "How many of you believe, in your heart of hearts, that stress makes you ill and believing makes you well?" All hands rose.

The person posing the question, Esther Sternberg, is not a New Age guru, and her attentive audience were all scientists. "How many of you, as scientists, are afraid to admit that?" she went on. Twenty years ago there would have been many more hands in the air, but there has been a shift in attitude, thanks to the work of researchers such as Sternberg. They have battled prejudice to successfully establish a scientific underpinning for mind-body medicine.

Sternberg, a programme director at the National Institute of Mental Health outside Washington DC, had a faltering start to her research career. Her first internship after medical school left her so disillusioned with academic medicine that she cancelled her subscription to the New England Journal of Medicine and became a general practitioner. Although she loved general practice, her patients led her back to research.

ÖйúAƬ

"One day I looked at my patient roster and I saw that about 50 per cent had some sort of joint pain and 50 per cent had some sort of psychological problem. I realised that I didn't know enough about either of those areas to feel confident treating them," Sternberg says. She decided to specialise as a rheumatologist.

On Christmas Eve 1979, six months before the end of her training, she was called in to see a very sick patient - an old man with crippling epilepsy. Whenever the man attempted to do something, he would have a seizure. He had been prescribed an experimental drug that altered his levels of serotonin, a chemical produced by the brain. Although the drug helped control the seizures, the man's skin had become so inflamed that even the brush of bed sheets was agonising.

ÖйúAƬ

Sternberg trawled through research papers in a determined bid to help him - but found no answers. "There were pieces of the puzzle that fitted and pieces that were missing. I realised I had to do the research to find those missing pieces," she recalls. She took the man off the drug and conducted a detailed biochemical study. Her analysis was published in the New England Journal of Medicine, she renewed her subscription and returned to research.

She never forgot that patient, and after years of research in immunology, Sternberg announced to colleagues that she planned to investigate the effects of serotonin on the immune system. Their reaction was not quite what she had expected. At the time, the very idea that the immune system might communicate with the brain, or with any other part of the nervous system, was considered ridiculous. "I was patted on the head and asked, 'Esther, are you sure you know what you're doing? You could be destroying your career.'" But she was determined.

"That first patient was such a graphic example. You see somebody lying in bed in excruciating pain, the sheets tented up over their legs, and they're so scarred they can't move, and all they were given was a drug that changed the level of serotonin in the brain. So no matter how much anybody told me the connection between mind and body was not real, I'd seen it with my own eyes."

It was the first but not the last time that one of her patients would have a profound effect on her research. Years later, Sternberg was testing a potential arthritis drug on two strains of rat - Lewis rats, which are very prone to arthritis, and Fischer rats, which are completely resistant. At midnight on the first day of the experiment, she got a call. The rats were all dying. She had a look at them and discovered that some of her rodent patients were fading fast. But strangely, it was the Fischer rats that were dying.

Sternberg recognised their symptoms. Ten years earlier, while working in the emergency room of the Royal Victoria Hospital in Montreal, she treated a young woman with a mild bladder infection. Although she could have just sent her home with some antibiotics, something told her to keep her in for observation. "I watched as her blood pressure went through the floor," Sternberg says. Overwhelmed by the infection, the woman's immune response to her illness had caused her blood vessels to become too "leaky". Without treatment she would have been dead within hours, but Sternberg injected her with steroids and, almost as quickly as she had crashed, she was back to normal. "You never forget it when you save a life. It is the most rewarding, most moving, most frightening, tumultuous feeling," she recalls. And she would never forget the symptoms of septic shock. "It always comes back to observing patients," she says.

Had Sternberg waited until the morning to check her rats, she would have missed their symptoms of septic shock. But now she had to make sense of what she had seen. The reaction to the drug was the opposite of what she had expected, and as she left the NIH building that night her mind was racing.

Several images kept surfacing - the resistant rats that had died of septic shock, the massive dose of steroids that had saved her young patient in Montreal. And then there was that copy of an unpublished, and highly controversial paper that had been circulating round the lab, which claimed that there were immune receptors in the hypothalamus in the brain. Finally the pieces fell into place.

ÖйúAƬ

In a flash of understanding she realised that the brain and the immune system do communicate and that she had been witness to the fatal consequences of a breakdown in that communication. Under "stress" all animals produce a cascade of hormones, starting in the hypothalamus of the brain, running through the pituitary gland beneath it and ending in the adrenal gland above the kidneys. This is where the stress hormone cortisol is pumped into the bloodstream. This "stress response" can be triggered by any number of things - cold, lack of sleep, an argument. What Sternberg realised was that the immune system, too, is capable of triggering this stream of biological events. It has a direct line to the hypothalamus.

ÖйúAƬ

When someone's immune system is activated to fight an illness or infection, she reasoned, it sends a signal to the hypothalamus to produce its stress hormones, including cortisol. The flow of hormones, in turn, shuts off the immune response. This ingenious negative-feedback loop allows a short burst of immune activity, but stops the immune system from getting too carried away. With the woman in Montreal and the rats in Sternberg's lab, the loop broke down and their immune systems ran amok.

The patient produced too few stress hormones to shut down her runaway immune system. Artificial steroids had to take over. The rats had been given a drug that blocked stress hormones, so did not stand a chance.

This theory explained the paradoxical events of that evening, but it also made sense of so much more. The reason her Lewis rats had chronic inflammation of the joints was because, as she later showed, they could not produce enough of a stress response to damp down their immune systems. And while too few stress hormones can cause inflammatory disease, Sternberg now had a plausible explanation for why chronic stress would have the opposite effect and would dampen the immune system too much. She was able to explain on a molecular level how stress makes us ill.

Though she was not aware of it at the time, Sternberg was just one of a number of pioneering scientists willing to think outside the box and who was convinced that the immune and nervous systems talk to each other. In those early days, hard-nosed immunologists sneered at such suggestions. But more than two decades of accumulated research have convinced even the most sceptical, and, Sternberg says, there has since been a rapprochement between immunologists and psychologists.

In her forthcoming book, The Balance Within, Sternberg describes the history of the science of health and emotions. She notes that until recently it was taken for granted that feelings and health were intimately linked. Embodied in the medieval idea of the four humours - blood, yellow bile, black bile and phlegm - are corresponding personality types - sanguine, choleric, melancholic and phlegmatic. But that was lost with the advent of modern anatomy and its unwritten dictum, "if you can't see it, it isn't real".

There may even be something to the idea that certain personalities are prone to certain diseases. Sternberg points to her rats - the arthritis-prone Lewis rats and the resistant Fischer rats. Students never have trouble telling them apart, although both strains are white. The Lewis rats are quiet and subdued, while the Fischer rats are always tense and run frantically about their cages. "The Lewis rats are laid-back California types, and the Fischer rats are type-A, New York personalities," she jokes.

People with rheumatoid arthritis also tend to be passive and prone to depression. Research is beginning to show that psychotherapy can have a positive impact on physical health, but Sternberg is cautious: "There is probably only so much play in the system. If you're a Lewis rat, no amount of psychotherapy is going to change your low stress response."

The Balance Within: The Science Connecting Health and Emotions is published on May 1 by W. H. Freeman.

ÖйúAƬ