Memory and the Hippocampus in Twins

Posttraumatic stress disorder (PTSD) is a condition of critical importance to the Veterans
Health Administration that affects up to 15% of Vietnam combat veterans.1 Studies in animals
demonstrated that stress is associated with alterations in the morphology of the
hippocampus, a brain structure involved in learning and memory, as well as memory deficits.
Patients with PTSD have smaller hippocampal volume on magnetic resonance imaging (MRI) and
deficits in memory on neuropsychological testing. Animal models also show a progressive
diminution of hippocampal volume with normal aging that is associated with a loss of memory
function. In humans, in very old age in some individuals the hippocampus becomes smaller,
with associated declines in memory function. Clinically, PTSD patients appear to have a more
rapid rate of memory decline with age, which has led to their being characterized as having
an "accelerated aging"; however there is little empiric data related to the interaction
between normal aging and PTSD and its effect on memory. In the prior period of VA Merit
funding the PI has studied memory and the hippocampus in twin pairs discordant for PTSD as
well as normal twin pairs. We have studied 42 MZ and DZ twin pairs discordant for Vietnam
theater exposure and discordant for PTSD and 28 MZ and DZ twin pairs concordant for Vietnam
theater exposure and discordant for PTSD. All twins underwent measures of hippocampal
volume, neuropsychological testing, and cortisol. A subset of 10 pairs of MZ pairs
discordant for PTSD and combat exposure underwent positron emission tomography (PET) imaging
of neural correlates of declarative memory function. Preliminary analyses are consistent
with smaller hippocampal volume and deficits in verbal declarative memory in PTSD affected
twins compared to their non-PTSD brothers who were not exposed to Vietnam combat.

We now propose to measure hippocampal volume, memory function, and neural correlates of
declarative memory, in MZ and DZ twins previously studied, who are discordant for combat and
PTSD. We will study 42 twin pairs who were studied during 1999-2003, of whom 34 were
discordant for PTSD at the Atlanta based assessment and had all study data including MRI,
neuropsychological testing, and cortisol (with PET in a subsample of 10 pairs), with repeat
measurements performed 8 years after the initial assessment, a time of life when they become
vulnerable to memory decline associated with aging (subjects not discordant for PTSD at the
Atlanta based assessment will also be studied as a comparison). These assessments will allow
a comprehensive assessment of the relative impact of shared environment (e.g. family life),
unique environment (Vietnam theater service), and genetics on the outcomes of interest. We
hypothesize that there will be an interaction between PTSD diagnosis and aging, i.e. PTSD
patients relative to non-PTSD will show smaller hippocampal volume as measured with MRI,
deficits in verbal declarative memory as measured with neuropsychological testing, lower
cortisol, and a failure of hippocampal function during memory tasks as measured with PET,
and that the rate of decline will be greater in PTSD v non PTSD. We further hypothesize that
analysis of all twin pair combinations will show a combined impact of genetics and PTSD
diagnosis on the hippocampal outcomes of interest.

Inclusion Criteria:

- All twins will be required to give written informed consent.

- Subjects will be included with a history of Vietnam Era service.

- All subjects must be free of major medical illness on the basis of history and
physical examination, lab testing, and electrocardiogram.

Exclusion Criteria:

- Subjects will be excluded with a history of shrapnel or other foreign bodies that
would preclude MRI scanning, history of loss of consciousness, or neuroleptics,
sedative or benzodiazepine medication usage within the previous four weeks.

- Subjects will also be excluded if based on the SCID interview they have comorbid:

1. schizophrenia;

2. schizoaffective disorder;

3. bulimia or anorexia, or

4. bipolar disorder.

- Subjects will also be excluded with a history of blast injury with associated
traumatic brain injury.

- In addition, subjects will be excluded with a history of alcohol or substance
abuse/dependence within the past year. We will use the strategy used in prior MRI
studies and use multivariate analyses to examine the contribution of past histories
of alcohol and substance abuse (as measured with the Addiction Severity Index) and
depression symptom level (as measured with the Hamilton Depression Scale) to the
outcomes of interest.

- Subjects will be excluded from the study if they present with a history of serious
medical or neurological illness, or as a result of routine laboratory studies,
history of asthma, steroid usage, seizure disorder, or prenatal/perinatal substance
exposure or trauma.