This study, conducted at the University of Pennsylvania and at the National Institutes of
Health in Bethesda, Maryland, will examine deficits in brain structure and function in
people exposed to trauma who developed post-traumatic stress disorder (PTSD) to see if these
deficits change after treatment. It also will investigate whether there is a genetic
susceptibility to PTSD.
Candidates 18 years of age and older in the following categories will be included in this
study: 1) women who have PTSD of at least 1 year's duration following sexual or non-sexual
assault; 2) healthy women (controls) who were previously assaulted but did not develop PTSD;
and 3) healthy women (controls) who were never traumatized. Candidates will be screened
with a medical history and physical examination, psychiatric evaluation, electrocardiogram
(EKG), and routine blood and urine tests.
Women with PTSD will be assigned to receive either: 1) 12 weeks of cognitive behavioral
psychotherapy either immediately upon enrollment or after a 3-month waiting period; or 2) 10
weeks of drug treatment with paroxetine (Paxil® (Registered Trademark)). Patients will be
evaluated before and after treatment with the procedures outlined below. Control subjects
will undergo the same procedures, also with a 10- to 12-week interval between evaluations.
- Neuropsychological testing: Subjects will take paper and pencil and computer tests to
evaluate memory, learning, attention and concentration, vocabulary and naming.
- Magnetic resonance imaging (MRI): Subjects will have MRI scans of the brain to examine
brain structure and blood flow while they perform two tasks. In the first task, they
will be shown a series of faces and asked to press one button for a male face and
another button for a female face. In the second task they will hear loud noises and
see colored squares. During the scan, subjects lie on a bed that slides into a narrow
tunnel (the scanner). They will wear a headset to block the noise of the scanner and
through which they will receive instructions for the tasks. Heart rate and skin
conductance (sweating) will be measured during the scan to evaluate physiologic changes
in response to the tasks.
- Eyeblink air puff test: Subjects will hear tones and will have a light puff of air
delivered to the eye. Changes in heart rate, sweat, and eyeblink will be measured with
electrodes taped to the skin on two fingers, on each side of the rib cage, and under
- Potential air puff delivery: This experiment has three parts. During each of the three
parts of this experiment, subjects will see colored lights and may or may not receive a
puff of air to the neck. Before each part they will be told that they will, will not,
or may receive an air puff to the neck. Each part will be repeated several times.
During the test, electrodes will be taped to the arms and chest to monitor skin
conductance and heart rate responses.
- Blood draw for genetic evaluation: Subjects' DNA will be examined to try to determine
if the risk of developing PTSD is inherited. The DNA will be examined for cortisol
receptor gene evaluation, to see if a form of this gene is found more often in patients
with PTSD than in healthy controls. The receptor for cortisol determines the activity
of the stress hormone cortisol, and genetic variations in the structure of this
receptor may be related to vulnerability to PTSD.
Patients taking paroxetine will be offered up to 3 months of additional drug therapy
following completion of the study and will be offered participation in other NIH studies
for evaluation and treatment of PTSD.
Posttraumatic stress disorder (PTSD) is characterized by intrusive recollections, avoidant
behavior, anxiety and exaggerated fear response. The pathophysiology of PTSD is largely
unknown. Neurophysiological testing in PTSD reveals deficits in memory and attention.
Neuroimaging studies report increased amygdala and decreased anterior cingulate activation
and reduced hippocampal volume. Clinical observations, psychophysiological measures and
animal studies suggest that facilitated fear conditioning, delayed extinction, inescapable
shock, sensitization and protracted habituation may contribute to the onset and persistence
We propose to use fMRI and the psychophysiology lab to examine the effect of treatment with
paroxetine and cognitive behavioral therapy on regional cerebral blood flow (rCBF) in brain
regions conceivably involved in evolution and maintenance of PTSD: Amygdala, anterior
cingulate and hippocampus. We will use emotional tasks that have elicited differences in
perfusion or metabolism between patients with PTSD and trauma exposed and healthy subjects.
Tasks performed in the fMRI will include 'masked' and 'unmasked' emotional faces paradigms
and differential delay conditioning. Contextual fear provocation and eyeblink trace
conditioning will be done in the psychophysiology lab. This evaluation will be performed
before and after a 10-week period of treatment with paroxetine and a 10-session course of
cognitive behavioral treatment (CBT).
Study population will comprise the following matched groups: Patients with PTSD; previously
traumatized healthy subjects who have not suffered from PTSD and non-traumatized healthy
subjects. Subjects from the 1st group will either be treated with paroxetine for 10 weeks,
undergo a 10-week 'prolonged exposure' (PE) CBT course, or will be put on the 'waiting
list'. It is assumed that variability in outcome will be observed in both treatment modes,
which will enable us to seek treatment response predictors.
It is unknown whether cerebral abnormalities in PTSD are state or trait phenomena. In
depression and OCD, functional imaging research shows improvement in malfunctioning brain
regions after clinically effective psychopharmacological and psychotherapeutic interventions
and after administration of placebo in depression. An increase in hippocampal volume was
found in patients with PTSD after treatment with paroxetine and after correction of the
endocrine abnormality in Cushing's disorder. These findings support our expectation that at
least some brain abnormalities in PTSD are state related, and that change in these
abnormalities is demonstrable by fMRI.
Men and women aged 18-65, suffering from PTSD, healthy trauma survivors and never
traumatized healthy subjects. PTSD diagnosis will be determined using the structured
clinical interview for DSM IV (SCID)(155).
Capable of providing informed consent, obtained prior to any study procedures.
Free of all psychotropic medication for at least 2 weeks, excluding short-term hypnotics.
Patients who were treated with fluoxetine will only be included after a medication free
period of at least 8 weeks.
Good physical health, confirmed by a complete physical exam (including normal vital
signs), electrocardiogram, neurological exam, and routine laboratory tests of blood and
urine. However, if patients have participated in other research studies or have had blood
work through their primary MD within the last 6 months, these results will be used instead
of repeating blood draws for inclusion into the study.
Right handed, as determined by a handedness questionnaire.
Past head trauma involving loss of consciousness, or amnesia greater than 24 hours. This
magnitude of head trauma could contribute to psychopathology and brain structure in a
manner different than PTSD thereby confounding results.
Substance abuse (alcohol or drugs) or dependence within 3 months prior to screening. The
effect of abuse/dependence on phenomenology and biology could mask and exceed PTSD effect.
All additional DSM IV Axis I comorbidity, excluding secondary diagnoses of major
depressive disorder (MDD) or anxiety disorders (AD). Given the high comorbidity of these
disorders in PTSD, and since excluding such patients would not provide the full spectrum
of the disorder, only patients in whom axis I diagnoses of MDD and AD preceded onset of
PTSD will be excluded.
Mental retardation or another pervasive developmental disorder.
Risk of homicide or suicide.
When the trauma involves assault, a continuing intimate relationship with the perpetrator.
Current or intended pregnancy. If a participant becomes pregnant during the course of the
study, her data will be removed from the analyses.
Subjects who are doing well on medication. Although we will only recruit non-medicated
patients, the decision to stop medication will be taken purely on clinical grounds. No
subject will be taken off medication solely to participate in the study.
Deafness will NOT exclude patients or healthy controls from participation. Accomodations
and protocol adjustments (e.g., abbreviation of neuropsychological battery to eliminate
tests which would be subjected to known or expected normalization violations) will be made
as necessary for this population.