Dr. Rasmusson is Director of the Neuroendocrine
Laboratory within the Clinical Neurosciences Division in West
Haven, CT. Recently she has been studying the HPA axis and how it
responds under stress; her work has also uncovered some
intriguing possible differences between men and women. Her work
may point the way to more effective treatments for PTSD for both
men and women. Janet Bailey interviewed Dr. Rasmusson about her
work in August 2002.
What exactly is the HPA axis?
HPA stands for hypothalamic-pituitary-adrenal axis. The
hypothalamus is the area of the brain where stress signals get
routed. When a person is under stress, the HPA axis is
activated, and after a number of steps, the hypothalamus releases
a substance called corticotropin releasing factor (CRF). CRF then
releases adrenocorticotropin hormone (ACTH) from the pituitary
gland. Finally, ACTH releases cortisol from the adrenal gland,
along with a number of other neurosteroids. Cortisol raises
the blood sugar level and helps a stressed person’s body
respond to the increased demands of stress.
How did work on HPA axis reactivity get started?
Some early studies with male combat veterans found that
veterans with PTSD showed unusually low levels of cortisol. This
came as a surprise—cortisol is normally released in
response to stress, so we would expect PTSD patients to have
higher levels, not lower. Researchers hypothesized that cortisol
might be somehow feeding back into the brain and turning off the
HPA axis response in people with PTSD. To test this hypothesis,
they administered synthetic cortisol (dexamethasone) to both male
combat veterans and male and female Holocaust survivors.
They found that the receptors in the brain that turn off the HPA
axis did seem to be more sensitive to the cortisol negative
feedback.
How did your research with women come about?
At about this same time, a heightened interest in research on
women was developing. So in 1995, we considered doing a more
detailed study of HPA axis regulation in women with PTSD. I must
say, the idea was not entirely appealing—the research
required fairly routine endocrinological tests, and it appeared
that we were likely to simply replicate the findings of earlier
work. We did decide, though, to look at some other adrenal
neurosteroids in our research. One of these was a compound called
allopregnanolone. In the late 1980s, this substance was found to
act in the brain to reduce anxiety in a manner similar to
antianxiety medications such as Diazepam (Valium).
We designed our study to be quite rigorous in several ways.
For instance, levels of various hormones—estrogen,
progesterone, allopregnanolone, and others—fluctuate
dramatically at different points in a woman’s menstrual
cycle and can alter the results of studies. The difficulty in
controlling for the fluctuation of these hormones is one reason
that researchers had traditionally shied away from studying
premenopausal women. So, our study controlled for where the
subjects were in their menstrual cycles. We were careful to
include about the same number of smokers in each group and
required that the subjects not drink alcohol for a month before
the study. We also made sure the women were not on any
medications. This was important because many early studies had
shown that certain medications, heavy smoking, and heavy alcohol
use can all influence HPA activity.
What did you find?
We did three studies and what we found was surprising. The
women with PTSD showed much larger increases in ACTH and a higher
production of cortisol compared to the women without
PTSD—the opposite of what had been found in most of the
earlier studies! Of course, at first this appeared to be a
totally anomalous finding. However, when we looked more closely
at the earlier studies, we found that they had not always
controlled for smoking, drinking, medication use, amount of
physical activity, and other factors like the incidence of
depression which might explain the inconsistency in
findings. The bottom line was that these findings suggest
to us that there might be gender-related factors that could
contribute to an increase in HPA axis activity in women with
PTSD.
Recent results from other investigations have been mixed.
Several have confirmed our results, finding higher cortisol
output in people with PTSD. This has been especially true in all
research with premenopausal women. Two studies with male
combat veterans showed lower cortisol output, but many of these
subjects were heavy smokers or were in some stage of nicotine
withdrawal. At this point it still isn’t clear whether the
consistent finding of high cortisol output in premenopausal women
with PTSD is specific to that group, since high cortisol levels
have also been observed in a study of male veterans and in
another study of children and adolescents. More carefully
controlled studies of men and postmenopausal women must be
done.
Do you have some explanation for possible gender differences
in HPA axis reactivity?
Our most recent data show that the women with PTSD released a
lot more of a substance called
dehydroepiandrosterone, or DHEA, along with cortisol when
their adrenal glands were activated. DHEA has been shown to
increase the HPA axis’s reaction to stimulation.
Interestingly, women generally have higher levels of DHEA than
men, and younger people have higher levels than older people. A
study that is now underway is comparing DHEA and HPA axis
reactivity patterns in men and women with PTSD, and it would also
be important to look at this in postmenopausal women.
The other system we haven’t looked at yet in our study
of gender differences is the serotonin system, which is the
system that is linked to depression. Previous research has shown
that women deplete their serotonin much more rapidly than men do
and don’t replace it as quickly. This may be one factor
that contributes to women’s greater susceptibility to
depression, and may explain why the drugs that affect the
serotonin system—the so-called selective serotonin reuptake
inhibitors (SSRIs) like Paxil and Zoloft—may work better
for women. We plan to study the serotonin system in an upcoming
research project.
We’ve also begun to look into possible genetic bases for
variability in HPA axis reactivity and the risk for PTSD
development. Some of these gene variations have a greater
influence in women than in men. However, it’s important to
remember that many systems are involved in the stress response
and in PTSD, so there are many possible genes, gene variations,
and genetic factor interactions to consider. But, at least we
know some of the important neurobiological systems we should
start looking at.
And, of course, the effects of the menstrual cycle must be
considered. Women who are in the late luteal phase of the
menstrual cycle—the days just prior to
menstruation—show higher heart rates and other responses to
stress. We think that hormone and neurosteroid fluctuations at
this time might, in some way, contribute to PTSD symptom
development, particularly since premenstrual symptoms and PTSD
symptoms are somewhat similar: irritability, anxiety, and
difficulty sleeping, for instance. Or, it’s possible
that the neurobiological state seen during the late luteal phase
of the menstrual cycle gets mimicked somehow in PTSD. Levels of
progesterone, allopregnanolone, and other related neurosteroids
drop dramatically at this time. Work by others has demonstrated
that decreases in these neurosteroids affect the function of
certain important anxiety-reducing receptors in the brain.
What other avenues of research have you been pursuing?
We have begun to look into the balance between the excitatory
steroids, like DHEA, and the ones that are inhibitory or have a
tranquilizing effect in the brain, like allopregnanolone.
A high level of DHEA may be very useful for a soldier standing
guard all night, for instance, but at the same time it could
contribute to common PTSD symptoms like sleep disturbances or
hyperarousal. We think there may be an imbalance between the
excitatory and inhibitory steroids in people suffering from PTSD
or perhaps that there are some people who are predisposed to
tipping that balance in one direction or another.
It’s also possible that DHEA may be
protective—DHEA antagonizes cortisol, and may in some ways
protect the brain from damage due to cortisol or other
neurotransmitters. This is important when you consider that
DHEA levels are very low in children before the ages of 6 or 7,
and then begin dropping steadily after age 30 and into old age.
This may be one reason why children exposed to abuse at a very
young age are at greater risk of later psychiatric complications,
including PTSD.
I think it’s also interesting that levels of
allopregnanolone, which has a tranquilizing effect, have been
shown to fall after age 50 in men but not women. It makes me
wonder if that contributes to why we’re seeing veterans at
that age with onset of delayed PTSD syndromes.
How do you find your research subjects?
We recruit from the population in our area by advertising in
the local newspapers, and we pay people for their participation.
We advertise in papers as far away as Hartford or Waterbury, and
we just started advertising in the New York area. We find that
people are usually willing to travel a little to participate in
our research, since they often are looking for solutions
too.
Recruitment of people with PTSD for our studies is quite
challenging, since in most cases we can’t study people who
are already on medication—we just can’t make sense of
the data. So we look for people who may not yet have gotten into
treatment for their PTSD, or who have found currently available
medications to be unhelpful. Indeed, some people have learned to
cope in a somewhat reasonable way with the symptoms they have and
don’t want to take medications.
Our research team views our role in the lives of our PTSD
research subjects as more than just researchers. We can help them
by making accurate diagnoses, educating them about their PTSD
symptoms, and steering them into a type of treatment in which
they may be interested. And if we encounter people who have PTSD
or other psychiatric problems that are so disabling that they are
at risk of losing a job, or need hospitalization, or are placing
themselves or others at risk, we refer them for treatment
immediately.
How will your research affect treatment of PTSD for veterans
and others?
We hope to identify the key points in the production of the
various neurosteroids that can be adjusted with medication.
Sometimes when people experience repeated stress, their system
doesn’t go back to baseline. Therefore, one possible
treatment goal would be to find medications that help an
activated stress system return to normal. Another approach might
be to figure out how the balance between the excitatory and
inhibitory substances works, in order to find a medication
regimen that can restore the balance. Either of these approaches
could involve development of new drugs or just better targeting
of existing drugs.
Look at smoking, for instance. What is it about nicotine that
is helpful in handling stress? We know that nicotine use blocks
the HPA axis reaction, and that people with PTSD have an
unusually difficult time quitting smoking—in part because
quitting is associated with an increase in PTSD-like symptoms. If
we can isolate the reasons for this, it could help point us in
the direction of new treatments—and also help people with
PTSD quit smoking.
Our work might also be useful in conjunction with other forms
of therapy. For instance, one common and often potent form of
treatment for PTSD is exposure therapy, where the patient is
prompted to talk about the traumatic experience, remember the
details, and hopefully reduce the fear response over time. But a
fair number of people find exposure therapy so difficult to
tolerate that they either fail to return for a second session or
they return to alcohol or drug use in an effort to reduce their
now overwhelmingly increased symptoms. Indeed, this
isn’t an unusual response in combat veterans with PTSD,
when they finally come in for treatment.
It would be good to see how much this avoidance behavior
correlates with levels of DHEA, allopregnanolone, and cortisol
during exposure therapy. If we could treat the underlying
problem, we might be able to help people complete the exposure
therapy successfully.
We’re also looking into possibilities for PTSD
prevention. Studies by colleagues in Israel have shown that
everyone shows pretty much the same initial reactions immediately
after severe trauma. After about four months, though, some people
have returned to normal and others have developed PTSD. If we can
understand what happens in that window of time, we might be able
to figure out how to prevent PTSD from developing—which
would be the best result of all.
