| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Oregon Health and Sciences University, Portland, Oregon 97239
Address all correspondence and requests for reprints to: Dr. Trevor Wheler, Department of Psychiatry, UHN-80, Oregon Health and Sciences University, 3181 SW Sam Jackson Park Road, Portland, Oregon 97239-3098. E-mail: whelert{at}ohsu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to assess concentrations of cortisol and its predominant metabolites, cortisol production rate (CPR), and glucocorticoid receptor (GR) binding characteristics in PTSD compared with normal subjects.
Design: Matched PTSD patients and control subjects had CPR determined by a stable isotope dilution technique after infusion of deuterated cortisol. Serum cortisol, urinary cortisol, and its metabolites were measured by gas chromatography/mass spectrometry. GR binding capacity (Ro) and ligand binding affinity (Kd) were measured in mononuclear leukocytes.
Setting: All subjects were tested during a 40-h admission in an inpatient clinical research center.
Patients and Participants: Ten patients with PTSD were matched by age and gender with 10 controls.
Outcome Measures: Statistical comparison was conducted for various measures of cortisol in PTSD patients and normal subjects.
Results: No statistical difference was found in mean level or circadian pattern of cortisol secretion using serum or salivary immunoassay detection methods. Although in the normal range, urinary cortisol by immunoassay showed statistically lower values over a 24-h period in PTSD patients compared with controls. This finding was not confirmed by gas chromatography/mass spectrometry determination of cortisol or its metabolites. CPR was not statistically different between these groups. GR also showed no alteration in Ro or Kd between the groups.
Conclusion: The data indicate that PTSD in the chronic and unprovoked state is not characterized by an acute biological stress response.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Posttraumatic stress disorder (PTSD) should be a prototypical stress disorder for both symptoms and stress response. Symptoms result from experiencing a severe traumatic event. Symptoms include increased sensitivity to stress, increased startle responses, severe anxiety, hypervigilance, increased autonomic response, and increased recollection of past stress as flashbacks and nightmares. These signs and symptoms suggest that PTSD could be associated with increased circulating cortisol levels. It is surprising that many reports show that cortisol levels are, in fact, low rather than high (5, 6, 7). It is noted, however, that elevated urinary cortisol has been reported in combat-related PTSD (8) and in abuse-related PTSD (9).
Tissue responsiveness to glucocorticoids is determined in part by the affinity of the intracellular glucocorticoid receptor (GR) for ligand as well as the number of receptors in a target cell. Unfortunately, studies of GR affinity and capacity in tissues from PTSD patients have yielded conflicting results. Early studies indicated that PTSD patients have substantially greater numbers of lymphocyte GRs than normal, a finding consistent with lower urinary cortisol secretion (5). In conflict with these data, recent studies have demonstrated that PTSD patients have decreased numbers of GRs in lymphocytes compared with healthy control subjects (10).
Taken together, this body of literature is incongruent in demonstrating alterations in cortisol levels (bound or unbound), HPA axis feedback regulation, or GR number or ligand binding affinity in PTSD patients. There are several plausible explanations for these disparate findings. First, immunoassay has been used as the primary tool for measurement of cortisol. This is a measurement technique subject to influence by the antibody used as well as various analytes and experimental conditions. It is also reasonable that PTSD is an individualized rather than a universal response to traumatic stress.
To address more thoroughly the status of cortisol production and metabolism in PTSD, we have studied a group of patients with active PTSD symptoms, taking no medication, in a controlled clinical study. We report here measures of the function of the hypothalamic pituitary adrenal axis as well as measurement of key adrenergic metabolites that have been implicated in PTSD (11).
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The research protocol was approved by the Oregon Health and Sciences University institutional review board. All subjects gave written informed consent before admission to the Clinical Research Center. The diagnosis of PTSD was confirmed using DSM-IV criteria, and participants’ symptoms were rated on the Clinician Administrated PTSD scale (CAPS) and the Hamilton Depression Scale (12, 13). In all PTSD subjects, the CAPS score was greater than the controls (PTSD range, 48–82; control subjects range, 0–30). Hamilton scores (H) were similar in PTSD and control subjects (PTSD H range, 2–23; control group H, <10). Twenty subjects, 10 with PTSD and 10 normal controls, completed the study. Both groups were composed of seven females and three males (average age was 40.3 yr for the PTSD subjects and 44 yr for the controls). The PTSD group had chronic symptoms, attributable to childhood trauma (five subjects), domestic violence (two subjects), or war trauma (three subjects). Subjects were excluded based on chronic illness, active drug abuse, or use of steroid medication. Subjects were free of psychotropic medications for 7 d before starting the study.
The methodology for conducting isotopic dilution studies and calculating CPR has been described (14). Briefly, subjects were admitted to the Oregon Health and Sciences University Clinical Research Center and allowed to ambulate. On the day of admission, an iv line was started in each arm and at 0200 h, d3-cortisol prepared in normal saline was infused at a rate calculated to administer 20 µg deuterated cortisol each hour, for a total of 480 µg/24 h. Starting at 0800 h on the morning of the second day, 3-ml blood samples were withdrawn from the opposite iv line at 30-min intervals. Blood sampling continued for 24 h, until 0800 h on the morning of the third day. Samples for salivary cortisol determination were collected every 2 h (2200–0800 h), and urine was collected over the course of the 24-h sampling period. In addition, 3-ml blood samples were collected every 4 h (0800–2200 h) and every 2 h (2200–0800 h) for ACTH analysis.
Detection of urinary cortisol and its metabolites
Gas chromatography-mass spectrometry was used to determine the pattern of cortisol and its predominant metabolites in pooled 24-h urine samples (courtesy of C. Shakleton, Children’s Hospital Oakland Research Institute, Oakland, CA).
GR ligand binding characteristics
Scatchard analysis of [3H]dexamethasone binding to the GR in circulating mononuclear leukocytes from PTSD patients and controls was conducted as described (15).
Measurement of cortisol, ACTH, and adrenergic metabolites in biological fluids
The following assays were performed in the General Clinical Research Center Core Laboratory at Oregon Health and Sciences University: serum cortisol (by automated Immulite chemiluminescent assay from Diagnostic Products Corp., Los Angeles, CA), ACTH (by ELISA from American Laboratory Products Co., Windham, NH), urine metanephrine, urine normetanephrine (by RIA from American Laboratory Products), and urinary free cortisol (with Coat-A-Count RIA, Diagnostic Products) after dichloromethane extraction. Salivary cortisol was determined by Saliva Testing and Reference Laboratory, Inc. (Seattle, WA).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
. No statistical differences were found in these parameters with the exception of urinary free cortisol, which showed a slightly lower concentration in PTSD compared with control subjects. Despite this statistical difference in urinary cortisol between PTSD and controls, mean urinary free cortisol levels in both groups were within the normal range.
|
Urinary normetanephrine and metanephrine in PTSD and control subjects were within normal reference ranges. The expectation was to find multi-fold differences, similar to what is found for the diagnosis of pheochromocytoma. Power analysis (power = 0.8; 
= 0.05) using this group size (n = 10) indicated that an 80% difference in means would be sufficient to detect a statistically significant difference between PTSD and control subjects for either of these catecholamines. Interestingly, assuming that the difference in means found in this study is maintained over a larger set of measurements, a group size of more than 20 subjects would be needed to attain statistical significance.
RIA data demonstrating the circadian variation of serum cortisol over a 24-h time period in PTSD patients and control subjects are shown in Fig. 1. Again, no significant difference in mean 24-h serum cortisol levels was demonstrable.
|
-tetrahydrocortisol, - and ß-cortolone, - and ß-cortisol, 20- and 20ß-hydrocortisol, 20- and 20ß-dehydrocortisone, and 6ß-hydroxycortisol. |
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Mean values for cortisol-associated parameters, with the exception of CPR, appeared to be lower in PTSD. These mean differences were not statistically significant and were not even close to P < 0.05, indicating that a much larger group of subjects would be needed to show potential significance. The only parameter that differed statistically in PTSD patients compared with control subjects was decreased urinary cortisol. However, we found no difference in distribution of cortisol or its metabolites in urine that could have explained this finding. It remains possible that lower urinary cortisol values found in patients with PTSD could be a result of differences in clearance rates of cortisol in these two groups. One possibility is that cortisol levels at certain time points may exceed cortisol binding globulin capacity and lead to increased cortisol clearance. This might explain the apparent lower cortisol levels in PTSD between 0400 and 0800 h or 2000 and 0800 h (Fig. 1
). However, we tested this segment of the curve and found no statistical difference in the integrated area under the curves during this time period. It remains possible that there is a phase shift in cortisol levels between PTSD and controls during this time period similar to what has been found in depression (16). Additional studies will be necessary to determine whether this is the case. It remains possible that alterations in adrenergic responsiveness may underlie PTSD symptomatology. However, in this limited study, we found no significant alterations in urinary metanephrine or normetanephrine.
The absence of demonstrable alterations in cortisol parameters clearly indicates that PTSD in the chronic unprovoked state is not characterized by an acute biological stress response. Interestingly, provocative stress, for example using personalized trauma scripts (a reading of past trauma experience to the patient), has shown that PTSD symptoms correlate with post-stress but not with baseline cortisol levels (17). Clarifying this may be important for understanding the pathogenesis of PTSD and the appropriate treatment at various times during its course. In conclusion, stress-stimulated cortisol levels may be a better measure of stress in PTSD than baseline cortisol levels widely studied previously.
|
Footnotes |
|---|
First Published Online June 20, 2006
Abbreviations: CPR, Cortisol production rate; GR, glucocorticoid receptor; HPA, hypothalamic-pituitary-adrenal axis; PTSD, posttraumatic stress disorder.
Received January 11, 2006.
Accepted June 9, 2006.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Read all eLetters
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
