The diagnosis is based on behavior, but the behavior is controlled by the brain. Reliving the memory of the traumatic event can produce a form of chronic stress, making the brain, in the words of neuroscientist Bruce S. McEwen of Rockefeller University, New York City," hostage to its own imaginings."

Some 65,000 Gulf War vets have signed up for the health examination program administered by the VA. About 2,200 have been diagnosed with PTSD and are receiving treatment. Other veterans of the Gulf War still on active duty are receiving care through the Defense Department.

In a longitudinal study now in its seventh year, Steven M. Southwick, program director for the National Center for PTSD's clinical neuroscience division at the VA-Connecticut Medical Center, and colleagues are tracking the health of 84 Gulf War National Guard reservists from either a medical or military police unit who were "exposed to stressors intense enough to cause PTSD." Southwick is also a professor of psychiatry at Yale University School of Medicine.

Southwick says about 9% of the reservists have developed PTSD. An even higher percentage have trauma-related symptoms such as sleep disturbances and irritability, but don't meet all the criteria for PTSD. For comparison, about 30% of returning Vietnam War vets had been diagnosed with PTSD; 20-plus years later, the number has declined to 15%. Southwick attributes the lower rate of PTSD in Gulf War vets to a shorter traumatized period.

The symptoms of Gulf War reservists with PTSD were generally mild one month after their return home, but they hadn't improved at six months. At two years-the last year reported-symptoms, though still relatively mild, had increased in severity over those noted at one and six months. At all time points, symptoms of hyperarousal were always more severe than those of avoidance, and were more severe for medical unit reservists than for those who served as military police.

Although the data have yet to be published, Southwick tells C&EN that the four-year findings "are fairly similar to the second year, but there is a slight reduction in symptomatology over time."

Southwick's colleague, Dennis S. Charney, director of the National Center for PTSD at the VA-Connecticut Medical Center and professor of psychiatry at Yale University School of Medicine, has been focusing on the biology of stress. In particular, his research group has been studying the structure and function of the brain in Vietnam War veterans diagnosed with PTSD.

"We have found low hippocampal volume" compared with healthy controls using magnetic resonance imaging (MRI), which confirms the finding of other researchers, Charney says. The hippocampus is the site of learning and consolidated memory, "and we see learning and memory impairments on structured neuropsychological testing, which may relate to the hippocampal problem," he adds. "And now we are doing PET [positron emission tomography] scans to look at the function of the hippocampus."

In these same patients, "the brain's norepinephrine system is hyperactive, which may relate to the symptoms of PTSD, though not necessarily the memory and learning impairments," Charney says. Rather, the hyped sympathetic nervous system "may relate to what is called the hyperarousal symptoms where patients have increased heart rates, they sweat more, they have tremors, and they startle more," he explains. The startle response is a blink measurement that is significantly elevated in patients with PTSD when compared with healthy controls.

Charney, Southwick, and colleagues have been funded by the Defense Department to study Gulf War veterans in the same way they have been studying vets of the Vietnam War. Hippocampal volume and function and the sympathetic nervous system will be studied in Gulf War vets with PTSD, vets without PTSD (combat controls), and healthy subjects who did not go to the Persian Gulf region. The study, which has already begun, is expected to take two years.

The working hypothesis for why the hippocampus shrinks in volume is convoluted and contested. The belief is that there is a rapid outpouring of glucocorticoids (cortisol) from the adrenal cortex in humans at the time of stress. Very high levels of these steroid hormones have been shown to have neurotoxic effects on the hippocampus in laboratory animals. There is no proof that the high levels of corticosteroids that gush through an individual under acute stress are damaging to the hippocampus, but that is the theory.

In fact, cortisol levels in veterans with PTSD measured years after the trauma are low. Levels of the hormone could not be measured at the time of trauma, but they might have been high then. However, Rachel Yehuda, a neuroscientist at Mount Sinai School of Medicine and at the Bronx VA Medical Center's post-traumatic stress program, believes otherwise.

"I don't think hippocampal atrophy in PTSD patients is related to high cortisol levels," she says. Furthermore, "I don't necessarily believe there is hippocampal atrophy that is unrelated to substance abuse because that has not been demonstrated." Studies reporting a decrease in hippocampal volume have been done on people with more problems than PTSD, she explains. And, she notes, it is also possible that the hippocampus was smaller than normal before the traumatic stress, which predisposed the veteran to PTSD.

Yehuda argues for more definitive studies. "We need to look at these things together, to measure cortisol and hippocampal atrophy together, which studies generally have not done."

Whatever is going on with PTSD may be moot in relation to the so-called Gulf War syndrome. As Matthew L. Puglisi, the American Legion's assistant director for Gulf War veterans, points out, those vets complaining of Gulf War syndrome symptoms have not been diagnosed as having PTSD. Symptoms of Gulf War syndrome include excessive fatigue, joint pain, headaches, skin rashes, memory loss, and sleep disturbance.

One of the explanations put forth to explain the symptoms is exposure to chemical warfare agents, or to the antidotes taken to prevent the effects of these neurotoxins. Pyridostigmine bromine (PB) tablets, for example, were given to troops as a pretreatment to protect them from the toxic effects of organophosphate nerve gases.

PB, a quaternary ammonium carbamate inhibitor of acetycholinesterase, would not be expected to breach the blood-brain barrier, that wall of tightly linked endothelial cells that normally keeps lipophobic substances like PB out of the brain. But in 1991, Israeli researchers found that Israeli troops who served in the Gulf War and who had taken PB exhibited a threefold increase in central nervous system side effects-headaches, insomnia, drowsiness, nervousness. The stress of combat appeared to knock out portions of the blood-brain barrier, allowing PB to penetrate, the researchers concluded.

In 1996, another Israeli group, led by Hermona Soreq, a molecular biologist at Hebrew University in Jerusalem, conducted a follow-on study in mice that showed that PB crossed the barrier. Stressed mice were found to require 1/100th the PB dose of their unstressed brethren to inhibit the same amount of brain acetylcholinesterase. In a separate companion study, the researchers looked for the effects of PB in normal, unstressed soldiers who volunteered for the study. Unlike the symptoms noted in the combat-stressed Israeli soldiers in the 1991 study, the unstressed soldiers exhibited side effects of the peripheral nervous system-diarrhea, excess sweating, and increased salivation. PB appeared not to cross the barrier.

Soreq: could conceptually have happened

It's intriguing to speculate on the implications of this recent Israeli work as a possible explanation for some Gulf War illnesses. But as Soreq says, "The relevance is unclear. Who can measure, several years back, the extent of psychological stress to which these veterans were exposed and/or the level of their exposure to drugs, chemical warfare agents, and insecticides-all of which inhibit cholinesterases? All we can say is what could conceptually have happened." Soreq is continuing to dissect the cascade of molecular events brought on by stress.

U.S. troops were given a bottle of PB tablets and told to take three a day. Now reports of PB overdosing are coming out. It turns out that PB gave some troops a head rush, at least initially, so they started popping many more than the recommended three pills a day. It's possible that not only did PB breach the blood-brain barrier, but that it did so at elevated and harmful levels.

The Food & Drug Administration has approved PB only for the treatment of the autoimmune disease myasthenia gravis. To give U.S. troops PB, the Pentagon had to seek a waiver from FDA, which it received. Ironically, PB only protects against the Soviet-developed nerve gases soman and tabun. Saddam Hussein, however, had stockpiled sarin, a U.S.-developed agent. This March, the Pentagon said PB use would be limited in the future to conflicts in which "intelligence indicates imminent use of the nerve agents soman and tabun against U.S. forces."