28 April 2011. The signature wound of modern war may be traumatic brain injury, but the picture is complicated by another invisible injury, that is, post-traumatic stress disorder. Developing as a reaction to a traumatic event, PTSD includes a cluster of symptoms ranging from nightmares, flashbacks, irritability, and hyperarousal, to emotional numbness and avoidance of places and objects that are reminders of the experience. Roger Pitman at Massachusetts General Hospital, Boston, told ARF that PTSD involves two kinds of brain changes: a conditioned fear reaction, in which certain cues incite anxiety, and sensitization to stimuli. For instance, people with PTSD startle easily. Research has also shown that people who develop PTSD may have a disturbance in their hypothalamic-pituitary system, Pitman said. Their increased number of glucocorticoid receptors corresponds to low levels of cortisol in the blood, predisposing a person to PTSD. In the big scheme of things, however, “We are still in the dark as to what the biological underpinnings of PTSD are,” said Norbert Schuff at the University of California in San Francisco.

The kinds of events that produce brain injuries are often traumatic, with the result that TBI and PTSD tend to go together. Elaine Peskind at the VA Puget Sound Health Care System and the University of Washington, Seattle, is conducting studies on a small group of veterans with TBI, as well as a small group of active-duty military with PTSD. She notes that three-fourths of these service members, all of whom were exposed to heavy combat, have both conditions. One study of returning military members concluded that PTSD and depression mediate many of the immediate poor health symptoms in soldiers with mild TBI (see Hoge et al., 2008), highlighting the importance of developing effective treatments for PTSD.

The long-term consequences of PTSD on the health of the brain remain an open question. Although TBI seems to have mechanistic links to dementia, the case to date is much weaker for PTSD. Only a couple of epidemiological studies have looked at the relationship. Both found provocative correlations. A large prospective study led by Kristine Yaffe at the San Francisco VA Medical Center and the University of California in San Francisco analyzed data from the VA’s national patient database on more than 180,000 veterans (see Yaffe et al., 2010). About 30 percent of the veterans had PTSD, and the researchers found that these veterans were more than twice as likely to develop dementia as those without PTSD over a period of about seven years. The results remained significant after adjusting for potential confounding factors such as TBI and depression. Another study found similar results, with veterans with PTSD having about twice the risk of dementia (see Qureshi et al., 2010).

What does this mean? Does PTSD influence the development of dementia, or are the two merely linked by some third factor? Some brain characteristics seen in people with PTSD, such as a smaller ventromedial prefrontal cortex, resemble features of AD, implying that the conditions could be related. Numerous studies have shown that people with PTSD have smaller hippocampi, also a hallmark of AD (see e.g., Bremner et al., 1995; Gurvits et al., 1996; Bremner et al., 1997; Hedges et al., 2003; Woodward et al., 2006). Nonetheless, Pitman urges caution in assuming that PTSD harms the brain. He points out that although PTSD may weaken some areas of the brain, other areas of the brain that mediate fear, such as the amygdala, are strengthened. “That may involve some adaptive brain plasticity that I am hesitant to call brain damage, although other people do call it brain damage,” Pitman said.

It is also possible that having a small hippocampus predisposes a person to PTSD. In other words, a small hippocampus may precede PTSD rather than following it. Pitman has conducted several studies on identical twins in which one twin had seen combat and the other had not. Some of the twins exposed to combat developed PTSD, and in agreement with other studies, these people had smaller hippocampi than combat veterans who did not get PTSD. What was interesting, though, was that the twins of the ones with PTSD also had smaller hippocampi, as well as lower cognitive performance and other subtle neurologic signs seen in people with PTSD, although these people had never seen combat and did not have the disorder (see Gilbertson et al., 2002; May et al., 2004; Pitman et al., 2006; Gurvits et al., 2006; Gilbertson et al., 2006; and Pitman, 2010). This suggests that these features are inherited risk factors for PTSD, rather than a result of a traumatic experience in war.

Along the same lines, one study claims that less intelligent people appear more likely to develop PTSD after a combat experience (see Macklin et al., 1998). Another study found no difference in hippocampal volume between veterans with and without PTSD, but saw that veterans who developed PTSD after their first exposure to a traumatic event had smaller hippocampi than those who developed the disorder after multiple traumas (see Yehuda et al., 2007). This, again, would support the idea that a smaller hippocampus is a risk factor for PTSD. Another study showed that among people with PTSD, reduced hippocampal activity was associated with more severe symptoms, implying that having a less active hippocampus can make a person more vulnerable to PTSD (see Astur et al., 2006).

Nevertheless, other studies do suggest mechanistic links between PTSD and the development of dementia. In both PTSD and AD, brain function is disturbed in the medial temporal lobe, hippocampus, and cingulate cortex, as seen by multiple forms of neuroimaging (see Tsolaki et al., 2009). Another imaging study concluded that the brains of people with PTSD showed numerous structural and physiological abnormalities in the frontal lobe and limbic regions, regions also affected by AD (see Schuff et al., 2011).

Hope, a painting by a Pakistani psychiatrist who treats PTSD. Image credit: Syed Ali Wasif via Wikimedia

A number of experiments on rodents have shown that prolonged stress can harm the hippocampus through the action of glucocorticoids (see, e.g., Sapolsky, 2000 and Dumas et al., 2010). High levels of homocysteine, a neurotoxic byproduct of cellular metabolism, have been seen in people with PTSD (see Levine et al., 2008 and Hasegawa, 2007). Since homocysteine is associated with higher dementia risk and shrinkage of the brain, this observation might provide a mechanism by which chronic stress could lead to dementia. Likewise, a recent imaging study led by Schuff showed that in people with PTSD, the hippocampus shrinks preferentially in the dentate gyrus, where new neurons are born in adulthood; this implies that “chronic stress suppresses neurogenesis and dendritic branching in these structures,” the paper concludes (see Wang et al., 2010).

“[The dentate gyrus] is one of the very few areas in the adult brain where there is regeneration of neurons,” Schuff said. “If this finding is replicated in other studies, then this gives a possibility for a pharmaceutical target to stimulate neurogenesis [that could] potentially help people who suffer from PTSD.”

Despite the link seen between PTSD and specific brain changes, “there are confounding third variables that would need to be controlled for,” Pitman cautions. “This work is all preliminary; it is suggestive. It deserves further pursuit.” Pitman observes that lower cognitive reserve, as measured by intelligence, might well predispose both to PTSD and to dementia-like brain changes, and is difficult to account for in studies. For her part, Elaine Peskind at the VA Puget Sound Health Care System, Seattle, and the University of Washington, Seattle, noted that “veterans who have PTSD have many health-related behavioral risks for dementia. [They are] more likely to smoke, more likely to be overweight, have hypertension, have sleep apnea.” Nevertheless, Yaffe points out that her study controlled for many of these factors and still saw an association between PTSD and dementia. See Part 6 for a discussion of how researchers might resolve this puzzle, and an overview of potential treatments.—Madolyn Bowman Rogers.

This is Part 5 of a six-part series. See also Part 1, Part 2, Part 3, Part 4, Part 6. See PDF of entire series.