Given the stress-related etiology of PTSD, it was expected that PTSD patients would show HPA axis abnormalities similar to those seen in depressed patients or chronically stressed animals, but this has not always been the case. An initial study (Mason et al. 1986) found that urinary free cortisol (UFC) excretion was lower in PTSD than in major depression. However, another study (Pitman and Orr 1990) found increased UFC excretion in outpatient PTSD veterans compared with combat-exposed control subjects without PTSD. Since then, there have been various findings, but the most comprehensive studies of PTSD, those by Yehuda and colleagues (for a review, see Yehuda 2002), continue to show low cortisol and enhanced cortisol suppression in response to dexamethasone in combat veterans with PTSD. Interestingly, the presence of comorbid major depression does not change the neuroendocrine picture. The main criticism of this body of work is that the sample comprised only male combat veterans and therefore is not representative, given that in the community, it is women who are most likely to experience PTSD (Breslau et al. 1991, 1995; Kessler et al. 1995). Furthermore, significant confounds with current and past alcohol and substance abuse occur in the veteran population.
Several studies have sought to address this problem, with the majority examining women with a history of childhood sexual abuse. While some studies have demonstrated increased UFC in women with PTSD or a history of abuse compared with control subjects (Lemieux and Coe 1995), others have demonstrated similar plasma cortisol (Rasmusson et al. 2001), and still others have found lower cortisol and enhanced cortisol suppression in response to dexamethasone (Stein et al. 1997). Yehuda (2002) examined Holocaust survivors, who were also predominantly exposed early in life, and observed lower UFC and enhanced cortisol suppression following dexamethasone administration in this population. The issue of comorbid depression in the PTSD population has not been well addressed, with most studies including comorbid individuals and few analyzing the data by the presence or absence of comorbid depression. The exception is the studies of Heim et al. (2001), which focused on childhood abuse and major depression and examined multiple HPA axis challenges in the same subjects. These studies found an effect of early abuse (with comorbid PTSD in 11 of 13 subjects) and major depression on stress reactivity, with both increased ACTH and cortisol response to the stressor compared with either healthy control subjects or depressed patients without childhood abuse. In this same cohort, there was a blunted response to CRH challenge in patients who had major depression with or without childhood abuse, but an increased response to CRH in those with early abuse but without major depression. The abused subjects also showed a blunted cortisol response to ACTH 1-24. Thus, childhood abuse produced an increased pituitary response with adaptive adrenal compensation, a change compatible with low or normal basal cortisol. Furthermore, lower cortisol and enhanced feedback to low-dose dexamethasone were found in the same subjects (Newport et al. 2004) regardless of the presence or absence of PTSD as the primary diagnosis, thus indicating enhanced feedback.
Epidemiologically based samples in adults have focused on natural disasters and have generally examined exposure with high and low PTSD symptoms (Anisman et al. 2001; Davidson and Baum 1986; Fukuda et al. 2000) but without diagnostic information. However, one study (Maes et al. 1998) that examined PTSD subjects recruited from community disasters demonstrated increased UFC in PTSD. In general, community-based studies suggest that exposure to disaster increases plasma (Fukuda et al. 2000) and saliva cortisol (Anisman et al. 2001) and UFC (Davidson and Baum 1986). Studies examining motor vehicle accident survivors (Hawk et al. 2000) found no difference in cortisol between those with and without PTSD 6 months later. Studies of male and female adults with exposure to mixed traumas have found either no effect of PTSD on basal cortisol (Kellner et al. 2002, 2003) or elevated basal cortisol (Atmaca et al. 2002; Lindley et al. 2004). Our analysis of recent trauma exposure in two community samples (Young and Breslau 2004a, 2004b; Young et al. 2004) found increased cortisol in those with past-year exposure to trauma, but no effect of greater than 1 year past trauma exposure and no effect of childhood abuse on basal saliva cortisol or UFC. To add further complexity, the majority of studies of trauma and PTSD included subjects with comorbid depression, and in most studies, the majority of subjects had both PTSD and major depression. The PTSD studies generally report comorbid depression in their subjects; however, studies of depression often fail to measure and report trauma histories. As a result, documented depression confounds much of the PTSD HPA axis literature, and undocumented trauma and abuse may confound some of the depression HPA axis literature.
In addition to the issue of exposure to trauma, the persistence of the neuroendocrine changes following recovery from PTSD is unclear. In an early study, Yehuda et al. (1995) reported that Holocaust survivors with past but not current PTSD demonstrated normal UFC, while later studies of offspring of Holocaust survivors (Yehuda et al. 2002) suggested that changes in cortisol may persist beyond the duration of the symptoms and thus may represent a marker of underlying vulnerability to PTSD. The large analysis by Boscarino (1996) of cortisol data from several thousand combat veterans showed a very small effect of PTSD on basal cortisol, but a very clear effect of combat exposure, with increasing levels of severity of combat exposure associated with increasingly lower cortisol.
Our recent studies of cortisol in PTSD from two epidemiological samples (Young and Breslau 2004a, 2004b; Young et al. 2004) demonstrated normal UFC and saliva cortisol in community-based individuals with "pure" and comorbid PTSD. The studies also demonstrated a clear effect of lifetime comorbid major depression on cortisol, showing increased HPA axis activation in the late afternoon/evening in patients with both major depression and PTSD. Furthermore, the elevated HPA drive demonstrated by increased evening cortisol levels was greater in the comorbid group than the elevation already documented in pure major depression.
Studies examining the response to low-dose dexamethasone in PTSD veterans found enhanced feedback to dexamethasone in veterans who met criteria for PTSD, regardless of the presence of comorbid major depression (Yehuda et al. 2002). Similar enhanced cortisol suppression in response to dexamethasone administration has been found in Holocaust survivors with PTSD and their offspring.
In the studies of Yehuda (2002) as well as the report by Stein et al. (1997), the enhanced suppression was also paired with low baseline cortisol, although other studies did not replicate this finding (Kellner et al. 2004a, 2004b).
In a CRH challenge study in combat-related PTSD, there was a normal to increased plasma cortisol at the time of challenge (Smith et al. 1989) and a decreased ACTH response in subjects with high baseline cortisol. Another study of women with PTSD and a history of childhood abuse (Rasmusson et al. 2001) showed enhanced cortisol response to CRH and to exogenous ACTH infusion, as well as a trend toward higher 24-hour UFC. Interestingly, all the women with PTSD had either past or current major depression, so comorbidity was the rule. In the study by Heim et al. (2001) examining response to CRH in women with major depression with and without childhood abuse, 14 of 15 major depressive disorder patients with childhood abuse also met criteria for PTSD. This group with comorbid major depression and PTSD demonstrated a blunted ACTH response to CRH challenge similar to that observed in major depression without PTSD. The abused groups also demonstrated lower baseline and stimulated cortisol both in response to CRH challenge and following ACTH infusion. These same groups of women showed a significantly greater HPA response to the Trier Stress Test, despite smaller responses to CRH challenge (Heim et al. 2000).
Several additional studies have evaluated response to stressors. Our early study (Liberzon et al. 1999) using combat noise versus white noise in male veterans with PTSD showed elevated basal and postprovocation cortisol compared with combat controls but no real evidence of a difference between the combat and white-noise days. A study by Bremner et al. (2003) of PTSD subjects of both sexes used a stressful cognitive challenge and found elevated basal saliva cortisol and continued higher cortisol for 60 minutes postchallenge. Eventually the saliva cortisol of the PTSD group returned to the same level as that of controls, raising the issue of whether the "basal" samples were truly basal or were influenced by the anticipatory challenge. Similar data were found in a study (Elzinga et al. 2003), using trauma scripts, in women with childhood abuse and PTSD compared with abused women with no PTSD. In that study, salivary cortisol was again significantly elevated at baseline, increased in response to the challenge (whereas controls showed no response), and then greatly decreased following the stressor, compatible with "basal" levels already reflecting exaggerated stress sensitivity in this group. Using a 1-minute cold pressor test, a recent study (Santa Ana et al. 2006) compared the ACTH and cortisol response in PTSD subjects with either childhood trauma or adult trauma with that of control subjects and found lower basal cortisol in the childhood abuse group. However, their data do not support an actual change in ACTH or cortisol in response to the stressor in any group, so it is difficult to interpret their findings as reflecting differences in stress response. In addition, sampling was very infrequent and therefore inadequate to characterize the time course to a very brief stressor. Overall, the existing stress data suggest an exaggerated stress response in PTSD.
Furthermore, the challenge studies certainly suggest that the picture is complicated in PTSD with comorbid depression; the findings of some studies look like depression while others look quite different—for example, showing a smaller response to ACTH infusion whereas patients with major depression show an augmented response. Age of trauma exposure may be one reason for contradictory data. Finally, one study by Yehuda (2002) of combat veterans with PTSD demonstrated greater rebound ACTH secretion compared with controls following administration of metyrapone in the morning, indicating that increased CRH drive is present in the morning but is normally restrained by cortisol feedback. The other two studies examining metyrapone challenge in PTSD found a normal ACTH response to afternoon or overnight metyrapone as well as a normal response to cortisol infusion in PTSD subjects and panic disorder subjects (Kanter et al. 2001; Kellner et al. 2004a, 2004b). In summary, these data suggest that there may be no simple relationship between diagnostic categories and specific HPA axis abnormalities. Timing of trauma or of onset of depression or anxiety disorders may differentially affect the HPA axis profile, although definitive studies have not been done.
Was this article helpful?
Stop Letting Anxiety Rule Your Life And Take Back The Control You Desire Right Now! You don't have to keep letting your anxiety disorder run your life. You can take back your inner power and change your life for the better starting today! In order to have control of a thing, you first must understand it. And that is what this handy little guide will help you do. Understand this illness for what it is. And, what it isn't.