Toward the Predeployment Detection of Risk for PTSD

Posttraumatic stress disorder (PTSD) is a serious public health problem in the general population that is estimated to affect more than 10 million American children and adults at some point in their lives. The National Comorbidity Survey has estimated that the lifetime prevalence of PTSD in the United States is 7.8% (1), and prevalence rates after deployment to Iraq and Afghanistan have been estimated at 6%-13% (2). Incidence rates of PTSD differ based on the type of index trauma experienced; however, only a minority of trauma victims develop PTSD. This fact, combined with research findings suggesting that early debriefing interventions targeting the prevention of PTSD are ineffective at best, and in some cases are detrimental (3), cautions against widespread intervention efforts with all trauma victims. Rather, targeting trauma victims at high risk for developing PTSD appears to be a better use of limited resources and may provide more efficacious results. However, we are currently limited in our ability to identify which trauma victims are likely to develop PTSD, leading researchers to focus on identifying reliable risk factors that predict the development of subsequent PTSD.

One promising line of research has considered possible biological abnormalities associated with PTSD. Early research into the biology of PTSD found that patients with chronic PTSD differed in levels of stress hormones from similarly traumatized victims who did not develop PTSD and normal comparison subjects, with much of this research focusing on abnormalities of the hypothalamic-pituitary-adrenal (HPA) axis and its end product, the glucocorticoid cortisol (4). Subsequent theoretical work and empirical research suggested that biological alterations observed in chronic PTSD patients often preceded the PTSD diagnosis in recently traumatized individuals and could represent risk factors for the development of the disorder. More specifically, higher heart rate levels and lower cortisol levels during and soon after trauma exposure were found to be associated with increased risk for the development of PTSD in adult trauma victims (5).

These findings led to questions about whether observed biological differences between victims who subsequently met versus did not meet PTSD criteria reflected differences with respect to reactivity to the traumatic event or whether these differences reflected pretrauma basal differences. Subsequent studies provided evidence that heightened startle responses (6) and smaller hippocampal volume (7) may be pre-trauma vulnerability factors for the development of PTSD following a traumatic event. The existence of such pretrauma vulnerability factors informs theory as to how PTSD develops, but the utility of these findings with respect to screening for individuals who are less likely to develop PTSD in high-risk occupations is limited. These measures require either scanning techniques or equipment to assess reactivity; the existence of a peripheral blood biomarker of increased vulnerability would provide a much simpler means of identifying at-risk individuals.

In this issue of the Journal, van Zuiden and colleagues (8) have found preliminary evidence for the existence of just such a biomarker of vulnerability for the development of PTSD. A large sample of soldiers participated in a prospective study in which blood was sampled prior to and at 1 and 6 months following deployment. The large sample size permitted comparison of participants reporting high levels of PTSD symptoms 6 months following deployment with matched participants reporting low levels of PTSD symptoms. Findings demonstrated that predeployment glucocorticoid receptor (GR) numbers were higher in participants reporting higher PTSD symptoms 6 months after deployment. Further, the GR number predicted inclusion in the PTSD group after controlling for childhood trauma experiences and predeployment PTSD and depression symptoms. Differences in GR numbers between the PTSD symptom groups persisted at the 1- and 6-month follow-up assessments.

The authors previously demonstrated that the predeployment GR number was positively associated with subsequent depression and fatigue (9), necessitating determination of whether observed differences in the GR number between the PTSD symptom groups in the present study was related to the presence of depression symptoms in the PTSD group. Analyses revealed that participants with versus without depression in the PTSD group did not differ in GR number, suggesting that heightened GR numbers were related specifically to PTSD. Although this may call into question their prior study results in which the impact of PTSD symptoms was not assessed, an alternative explanation is that a high pretrauma GR number may reflect a more general risk factor, representing vulnerability for a range of mental health problems. This hypothesis is not without precedent, as prior research has demonstrated that HPA axis abnormalities are associated with a number of mental and physical disorders such as PTSD, chronic stress, chronic fatigue, and a variety of medical disorders (10).

Whereas it is possible that abnormalities in the GR number may serve as a mechanistic vulnerability factor, it is also possible that high GR levels are not causally related to PTSD, but rather that they represent risk afforded by other pretrauma variables. Prior trauma history is a consistently observed risk factor for the development of PTSD following subsequent traumatization, and trauma history has been associated with alterations of HPA axis activity. It is possible that an altered GR number may represent risk afforded by prior trauma exposure. Although in the present study the GR number was not related to the number of early traumas experienced, it was significantly related in the authors' prior study (9), suggesting that high GR levels could be a marker for additional risk factors whether or not GR differences represent a biological mechanism through which PTSD develops.

In summary, the present study, combined with the authors' prior work, suggests that a higher GR number may represent a vulnerability factor for the development of post-traumatic stress. However, caution must be taken into account when interpreting these findings. Although mean GR levels differed between the high and low PTSD symptom groups, there was substantial overlap between the groups, limiting the GR number as a screener for PTSD risk. Therefore, it would be extraordinarily premature to use these data to “screen out” military cadets or trainees for other high-risk occupations. Prior research has found a number of pre-, peri-, and posttraumatic variables to be associated with increased risk for PTSD (11); however, even combined, these risk factors do not account for a large percent of the variance in PTSD. PTSD is a complex disorder, and much more work is needed to reliably identify trauma victims at risk for developing the disorder. This is a critical need in the trauma literature, as without a reliable means by which to identify victims at highest risk of developing PTSD, our attempts to intervene to prevent symptom development are hampered. Further, testing of early interventions is also made difficult by our inability to reliably identify victims at high risk for PTSD, as it is difficult to show a beneficial effect of any intervention manipulation when relatively few recruited participants develop the disorder. It is likely that a combination of risk factors will provide the best predictive utility, and further research combining the preexisting GR number with other biological predictors, cognitive appraisals, and social support moderators may provide a risk screener with high sensitivity and specificity, increasing our ability to identify high-risk victims and aiding in the testing of early interventions.