Treating Psychological Trauma and PTSD
The Guilford Press Copyright © 2004 The Guilford Press
All right reserved. ISBN: 1-59385-017-4
Chapter One Allostatic versus Empirical Perspectives on Pharmacotherapy for PTSD
Matthew J. Friedman
The concept of allostatic load, as originally proposed (McEwen & Steller, 1993; McEwen, 1998) is a biological model of stability through change. In this book we have expanded that context to include intrapsychic, interpersonal, and social as well as biological domains, because allostasis is such a rich heuristic model through which to seek to understand the many complex biopsychosocial manifestations of posttraumatic stress disorder (PTSD). It is also a useful context in which to consider the specific treatments that have been tested and proposed for ameliorating the symptoms and general distress associated with this disorder. In this chapter, the focus is specifically on the many psychobiological mechanisms that are disrupted in PTSD and on the various medications that have been tested and proposed for reversing such abnormalities.
When considering allostatic load models, it is important to keep in mind that there are a number of ways in which a system can overshoot, undershoot, fail to recover, or become otherwise dysregulated because it is incapable of accurately titrating its adaptive repertoire to environmental demands (McEwen, 1998). Furthermore, even if the organism's overall responsecapability has remained intact, it may become encumbered by allostatic load because its antennae are not well calibrated to accurately assess the challenge at hand. Here the problem lies with signal detection rather than response potential so that the organism either fails to recognize all the stressors to which it must respond (e.g., false negatives) or it tends to misperceive harmless stimuli as threats to survival (e.g., false positives). In PTSD, it is well recognized that the appraisal process is biased toward perceiving danger rather than safety (e.g., false positives) and hence the response bias is toward over rather than underreaction. Such a hypervigilant, hyperreactive posture for engaging the environment is a prescription for shifting from a homeostatic to an allostatic steady state.
From an allostatic perspective, PTSD is extremely complex. As has been stated elsewhere (Friedman, Charney, & Deutch, 1995, pp. xix-xx), this is because humans who fail to meet the demands of traumatic stressors utilize and perturb the many psychobiological mechanisms that have evolved through evolution for coping, adaptation, and preservation of the species. This is why it should come as no surprise that people with PTSD exhibit abnormalities in almost every psychobiological system that has been investigated. Indeed, in the same way that many different pathological circumstances may produce the same clinical abnormality (e.g., fever or edema), many different psychobiological abnormalities may lead to PTSD. Furthermore, it is not too far fetched to anticipate that a spectrum of posttraumatic syndromes may be elucidated by future research and that each syndrome will be associated with a unique allostatic configuration. Furthermore, each syndrome may respond optimally to a different medication. But we are getting ahead of ourselves. First, we must consider current evidence that allostatic load is present in PTSD by reviewing the many different psychobiological abnormalities associated with this disorder. Next we consider a rational pharmacotherapeutic strategy based on this analysis. Then we consider how such an allostatic perspective compares with the current empirical approach to pharmacotherapy. And finally we review the decision process in pharmacotherapy and the many factors by which it is influenced.
WHAT IS THE CURRENT EVIDENCE FOR ALLOSTATIC LOAD IN PTSD?
The best evidence for allostatic load in PTSD comes from research with the two systems that have been most associated with the human stress response: the adrenergic and hypothalamic-pituitary-adrenocortical (HPA) systems. There is also evidence for allostasis in the serotonergic, opioid, and other systems.
Allostasis and the Adrenergic System
It is well recognized that adrenergic reactivity is enhanced in PTSD patients. This conclusion is based on psychophysiological research on the sympathetic nervous system (SNS), which consistently shows heightened cardiovascular and acoustic startle responsivity (see the review by Pitman, Orr, Shalev, Metzger, & Mellman, 1999). Likewise, research on the adrenergic nervous system consistently indicates elevated catecholamine levels and heightened sensitivity to the adrenergic alpha-2 receptor antagonist yohimbine (see the review by Southwick et al., 1999)
Despite the heightened adrenergic reactivity observed in PSD, resting SNS and adrenergic activity is not elevated. For example, PTSD patients do not show elevated blood pressures or heart rates at rest; it is only when they are challenged by some psychological probe (e.g., trauma-related stimuli) or pharmacological probe (e.g., yohimbine) that such adrenergic abnormalities can be unmasked. In short, at rest the PTSD patient exhibits adrenergic and SNS stability. But such stability comes at a price. This price is what McEwen (1998) has termed allostatic load. With respect to PTSD, the adrenergic price of stability appears to be a reduction (or downregulation) of alpha-2 adrenergic receptors (Perry, 1994). The potential impact to the person with PTSD of excessive adrenergic stimulation is blunted by an adaptive reduction in the number of receptor sites available to react to such increased neurotransmitter levels. During the relative "quiet" of baseline function, a physiological stability is apparent that is indistinguishable from the homeostatic steady state seen in individuals without PTSD. During a stressful episode or some other provocation, however, the downregulation of adrenergic receptors is unequal to the task and therefore unable to maintain stability. Hence, under such circumstances, PTSD patients exhibit the heightened reactivity mentioned above. This is another aspect of allostatic load, another price that must be paid because the adrenergic systems of PTSD patients are inadequately equipped to cope with the demands of stress, in contrast to the systems of people without this disorder.
Allostasis and the Hypothalamic-Pituitary-Adrenocortical Systems
The case for allostasis in PTSD is easier to make with resect to the HPA system because of elegant research that has investigated the different components of HPA function more thoroughly than has been the case with the adrenergic system (see the review by Yehuda, 1999). Here the allostatic balance appears to be the reverse of that seen with adrenergic mechanisms. Whereas excessive adrenergic reactivity is partially offset by downregulation of alpha-2 receptors, in the HPA system reduced serum cortisol levels are offset by upregulation and increased sensitivity of glucocorticoid receptors. The principle is the same-only the direction of change is different. The price of stability is an adaptive change at the receptor level that can be unmasked by psychological or pharmacological probes. I have suggested elsewhere (Friedman, 1998) that, behaviorally speaking, the price of such stability is stress intolerance because people with PTSD appear less able to cope with the normal hassles and vicissitudes of life. Pharmacologically, allostatic load is evident because people with PTSD (in comparison to those without PTSD) exhibit supersensitivity or supersuppression of HPA function in response to the glucocorticoid dexamethasone (Yehuda et al., 1993).
Allostasis and the Serotonergic System
The third example of psychobiological allostasis in PTSD is admittedly much more speculative. It is worth discussing, however, because of the importance of serotonin (5-hydroxytryptamine, or 5-HT) in the human stress response and because of the recently demonstrated efficacy of drugs that modify 5-HT function in PTSD. Southwick and associates (1997) have shown that some Vietnam veterans with PTSD are especially sensitive to the 5-HT agonist m-chlorophenylpiperazine (mCPP) which interacts primarily with 5-H[T.sub.2] and 5-H[T.sub.1c] reeptors.
One interpretation of these results is that those veterans who exhibited a panic/flashback response to MCPP did so because of upregulation or supersensitivity of 5-HT receptors. If that were the case, one would predict that administering a drug that could downregulate 5-HT receptors might be an effective treatment for PTSD. Indeed, sertraline, a selective serotonic reuptake inhibitor (SSRI), does downregulate postsynaptic 5-HT receptors. Sertraline is also an effective treatment for PTSD (see below). Is this a coincidence? Or is this circumstantial evidence in support of the allostatic load hypothesis?
This speculative example is also a good place to illustrate how a rational approach to pharmacotherapy could be based on an understanding of allostasis. The pharmacological agent of choice would be a medication that reduces allostatic load by pushing the system b ack toward a homeostatic steady state. Thus SSRI-mediated downregulation of allostatically upregulated 5-HT receptors is definitely a therapeutic step in the correct homeostatic direction.
The presumption that selective reduction of allostatic load will produce clinical improvement is the guiding principle for the subsequent discussion of rational pharmacotherapy for PTSD.
Before leaving the 5-HT system, it is instructive to consider an additional finding in the MCPP study (Southwick et al., 1997). Whereas some veterans with PTSD exhibited panic and flashback reactions following administration of MCPP, others did not. Among those who did not were many who displayed panic and flashback reactions to the adrenergic agent yohimbine. Thus some veterans were MCPP (but not yohimbine) responders, indicating excessive serotonergic sensitivity, whereas others were yohimbine (but not MCPP) responders, indicating excessive adrenergic reactivity. Veterans without PTSD did not react to either drug. Therefore, this provocative study suggests that different people may implement different psychobiological adaptive strategies for coping with chronic stress. For some veterans allostatic load was best understood as an adrenergic abnormality, whereas for others allostatic load was best understood as a serotonergic adaptation.
Allostasis and the Opioid System
Although there is little clinical research on the opioid system in PTSD, there is abundant evidence that endogenous opioids (endorphins, dynorphins, and enkephalins) play an important role in the stress response of animals. A well-established laboratory phenomenon, stress-induced analgesia (SIA), occurs when experimental animals are exposed to stressful stimuli such as electric shock, forced swimming, or restraint. Under such circumstances, laboratory animals exhibit a reduced responsiveness to pain (e.g., SIA) that can be reversed by narcotic antagonists, thus indicating that SIA is an opioid response to stress (Stout, Kilts, & Nemeroff, 1995). There is one experiment suggesting that SIA can also be produced in humans with PTSD (Pitman, van der Kolk, Orr, & Greenberg, 1990), although these findings have never been replicated. Other studies have shown additional abnormalities in opioid function among PTSD patients (reviewed in Friedman & Southwick, 1995).
With respect to allostasis, there is one very interesting report concerning an open-label trial with a narcotic antagonist that was administered to Vietnam veterans with PTSD (Glover, 1993). The guiding hypothesis was that emotional numbing in PTSD is mediated by opioids. It was expected that by reversing opioid activity the narcotic antagonist would reduce numbing symptoms and thereby diminish PTSD severity. Indeed, several veterans responded as predicted and reported that they felt more alive, less numb, and less constricted emotionally. Unfortunately, other veterans reported that their PTSD became dramatically worse because of intolerable anxiety, panic, arousal, and even flashbacks, in some cases.
How can we understand such diametrically opposite effects among a cohort of apparently similar people (male Vietnam veterans) who all received the same medication? One explanation is that opioid-related allostatic load was balanced differently in different veterans. If we accept the hypothesis that opioid activity is an adaptive (allostatic) response to blunt/numb the excessive (adrenergic) arousal associated with this disorder, we can propose that individuals may differ in their capacity to mobilize opioid mechanisms to achieve allostatic stability. Thus, we might expect that those veterans who exhibited excessive emotional numbing (hypothetically because of excessive opioid activity) experienced relief from the narcotic antagonist because their elevated opioid function was reduced toward homeostatic levels. We might suggest that, in contrast to their "numbed out" colleagues, those veterans who had a severe anxiety reaction were those whose allostatic steady state consisted of a much smaller opioid component. They were at high risk to experience a severe anxiety reaction because the narcotic antagonist blocked what little opioid activity they had been able to mobilize to antagonize adrenergic hyperarousal.
There are several points to underscore here regarding how adaptive psychobiological strategies may differ from one individual to the next. In some cases, the difference may be (quantitatively) related to the capacity to mobilize one particular (e.g., opioid) mechanism to achieve stability. On the other hand, the yohimbine versus MCPP example suggests that different people with PTSD may utilize (qualitatively) different psychobiological allostatic strategies (e.g., adrenergic vs. serotonergic) to achieve stability. Finally, if different quantitative and/or qualitative adaptations can underlie PTSD, then different medications may be indicated for different people even though they appear to have the same DSM-IV disorder.
Allostasis and Corticotropin-Releasing Factor
Corticotropin-releasing factor (CRF) is a neuropeptide that ignites the complex cascade of adrenergic, HPA, immunological, and other psychobiological systems that participate in the human stress response. As a neurotransmitter, CRF activates adrenergic neurons in the locus coeruleus (Aston-Jones, Valentino, Van Bockstaele, & Meyerson, 1994) while as a neurohormone, CRF promotes the HPA response by releasing adrenocorticotropic hormone (ACTH) from the pituitary gland. Two studies indicate that CRF activity is increased in PTSD. CRF levels are elevated in the cerebrospinal fluid (CSF) (Bremner et al.,
Excerpted from Treating Psychological Trauma and PTSD Copyright © 2004 by The Guilford Press. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.