Elsevier

Biological Psychiatry

Volume 48, Issue 9, 1 November 2000, Pages 940-947
Biological Psychiatry

Original article
Initial posttraumatic urinary cortisol levels predict subsequent PTSD symptoms in motor vehicle accident victims

https://doi.org/10.1016/S0006-3223(00)00896-9Get rights and content

Abstract

Background: This study was designed to examine the relationship between urinary hormone levels collected upon admission to the trauma unit following a motor vehicle accident and posttraumatic stress disorder symptomatology 1 month later.

Methods: Fifteen-hour urine samples were collected from 63 male and 36 female motor vehicle accident victims and were used to assess levels of catecholamines and cortisol reflecting peritraumatic and acute-phase posttraumatic levels. Presence of posttraumatic stress disorder symptomatology was assessed 1 month after the accident.

Results: Motor vehicle accident victims subsequently diagnosed with acute posttraumatic stress disorder excreted significantly lower levels of cortisol in 15-hour urines collected upon admission to the hospital. In addition, urinary levels of cortisol predicted a significant percentage of the variance in intrusive and avoidant thoughts 1 month after the accident.

Conclusions: The results of our study suggest that initial cortisol levels in the immediate aftermath of a traumatic event contribute, in part, to subsequent symptoms of posttraumatic stress disorder.

Introduction

The psychophysiology of posttraumatic stress disorder (PTSD) has been the focus of a number of recent studies, with research indicating that neuroendocrine levels may differentiate between those victims who meet PTSD criteria and those who do not. Despite mixed findings (for exceptions, see Lemieux and Coe 1995, Pitman and Orr 1990), the majority of studies have found lower 24-hour urinary cortisol excretion in victims with PTSD compared with victims without PTSD and normal control subjects (Mason et al 1986, Yehuda et al 1990, Yehuda and Harvey 1997, Yehuda et al 1993, Yehuda et al 1995a, Yehuda et al 1995b; for a review, see Friedman 1991), suggesting a downregulation of the hypothalamic–pituitary–adrenal (HPA) axis in PTSD. Further support for altered HPA functioning stems from findings of greater numbers of lymphocyte glucocorticoid receptors Yehuda et al 1991, Yehuda et al 1993 and exaggerated dexamethasone (.5mg dose) suppression of cortisol excretion (Halbreich et al 1989, Yehuda et al 1993; Yehuda et al 1995) in PTSD patients. Augmented adrenocorticotropic hormone response has also been noted after metyrapone treatment (Yehuda et al 1996), leading Yehuda and colleagues to suggest that PTSD is characterized by an enhanced sensitivity of the glucocorticoid negative feedback loop at the pituitary.

Studies examining basal catecholamine levels in patients with PTSD also have reported mixed results. Whereas PTSD patients and control subjects do not differ in levels of plasma norepinephrine and epinephrine Blanchard et al 1991, McFall et al 1990, Southwick et al 1994, findings concerning 24-hour urinary catecholamine excretion have suggested greater catecholamine excretion in PTSD patients versus control subjects (Lemieux and Coe 1995, Kosten et al 1987, Yehuda et al 1992; see Friedman 1991 for a review).

The majority of these studies have examined neuroendocrine levels in chronic PTSD patients who have presented with PTSD for more than 20 years, making conclusions concerning onset of altered neuroendocrine levels difficult. Determining whether neuroendocrine abnormalities reflect altered acute responses to the traumatic stressors that persist long after the event or long-term presence of PTSD is impossible. Recent research has attempted to address this issue by examining neuroendocrine levels of trauma victims during the acute phase of responding to traumatic events. Resnick et al (1995) examined plasma cortisol levels within 51 hours after a rape (mean = 12.0 ± 15.6 hours) in 37 adult female victims. Women with prior assault or rape histories demonstrated lower plasma cortisol levels and were more likely to develop PTSD than were women without similar trauma histories; however, cortisol levels did not predict PTSD diagnosis at follow-up (ranging from 17 to 157 days after the rape). McFarlane et al (1997) examined 40 motor vehicle accident (MVA) victims and found that, although victims meeting PTSD criteria 6 months after the accident had significantly lower plasma cortisol levels than did victims with major depressive disorder, cortisol levels of PTSD patients did not differ from those of victims with no diagnosis. Both of these studies examined small numbers of participants and may not have had enough power to examine the relationship between initial cortisol levels and subsequent PTSD symptoms or to detect differences in cortisol levels between PTSD and non-PTSD patients.

It is difficult to measure acute physiologic responses to traumatic stress in naturalistic studies because traumatic stressors often involve a number of stressful events and do not have clear start and end points. For example, an MVA can include a number of potentially traumatic events: initial realization that a person is going to be in an accident, actual impact, pain experienced, hospital procedures, body disfigurement (facial scars), and so forth. Plasma cortisol levels provide an approximation of cortisol levels over the last hour or two (Baum and Grunberg 1995), and the acute time frame encompassed in a single blood draw may not be sufficient to examine more general systemic alterations in HPA hormones and their effects on the encoding and consolidating of these various traumatic memories. Urine samples average hormone levels across a longer time frame (the duration of the collection period) than do plasma samples and may provide a better measure of hormone levels throughout the acute phase of responding to a traumatic event (Baum and Grunberg 1995).

In our study, 15-hour urine samples were collected from MVA victims beginning upon admission to the trauma unit after their accidents. Urine samples were used to examine MVA victims’ peri- and acute-phase posttraumatic neuroendocrine responses to their accident and to assess their efficacy in predicting PTSD symptomatology 1 month after the accident.

Although 24-hour urine samples would have provided hormone levels across the full circadian cycle, lesser injured victims were often not hospitalized for a full 24 hours, and the shorter sampling period allowed us to examine a wider range of patients.

Section snippets

Participants

Participants consisted of 63 male and 36 female motor vehicle (20% motorcycle) accident victims admitted to the level 1 trauma center of a Midwestern community hospital. Victims with Glasgow Coma Scale scores of less than 14 were excluded. Among the subjects, 73% were drivers. The average age of participants was 37.3 years (SD = 17.7, range = 18–84), and the sample was 85% European American, 13% African American, and 2% Asian American.

Procedures

Upon admission to the trauma center, patients are routinely

Statistical analyses

Initial one-way analyses of variance (ANOVAs) and chi-square tests were conducted to examine differences in demographics between PTSD and non-PTSD patients. Pearson product correlations were conducted to determine covariates for analyses. Differences between PTSD groups in urinary hormone levels were examined using ANOVAs and analyses of covariance (ANCOVAs). Finally, the relationship between PTSD symptomatology and urinary hormone levels were analyzed with hierarchical multiple regression

Discussion

Numerous recent reports have demonstrated reliable differences in levels of urinary catecholamines and cortisol in PTSD and non-PTSD patients who were similarly exposed to a traumatic event. Nonetheless, these studies have typically examined victims presenting with chronic (often decades long) PTSD, making it impossible to conclude whether altered neuroendocrine levels reflect long-term presence of psychopathology or altered or persistent response to the initial traumatic event. Our study

References (37)

  • R Yehuda et al.

    Glucocorticoid receptor number and cortisol excretion in mood, anxiety, and psychotic disorders

    Biol Psychiatry

    (1993)
  • R Yehuda et al.

    Increased pituitary activation following metyrapone administration in posttraumatic stress disorder

    Psychoneuroendocrinology

    (1996)
  • R Yehuda et al.

    Predicting the development of posttraumatic stress disorder from the acute response to a traumatic event

    Biol Psychiatry

    (1998)
  • Association for the Advancement of Automotive Medicine (1990): The Abbreviated Injury Scale. Des Plaines, IL:...
  • A Baum et al.

    Emotional, behavioral, and physiological effects of chronic stress at Three Mile Island

    J Consult Clin Psychol

    (1983)
  • Baum A, Grunberg N (1995): Measurement of stress hormones. In: Cohen S, Kessler RC, Gordon LU, editors. Measuring...
  • E.B Blanchard et al.

    The psychophysiology of motor vehicle accident related post-traumatic stress disorder

    Biofeedback Self Regul

    (1991)
  • Charney DS, Southwick SM, Krystal JH, Deutch AY, Murburg MM, Davis M (1994): Neurobiological mechanisms of PTSD. In:...
  • Cited by (0)

    View full text