Numerous magnetic resonance imaging (MRI) studies measuring the mass of limbic structures in patients with chronic non-remitting post-traumatic stress disorder (PTSD) have consistently reported smaller hippocampal (HPC) volumes (Davis, Frazier, Williford, & Newell, 2006). The HPC is a highly neuroplastic region of the brain providing negative feedback for part of the endocrine system. The HPC receives feedback from the hypothalamic-pituitary-adrenal (HPA) axis via plasma glucocorticoid levels to reduce the activation of the fight-or-flight response. Animal studies have suggested that HPC neuronal atrophy can be caused by excessive circulating glucocorticoids; without proper feedback, the fight-or-flight response is hyperactive increasing the strength of fear conditioning. Conversely, studies have also shown that HPC volume can be increased by antidepressant medications that have neurogenic properties (Sapolsky, 2000), however the physiological effects of psychotherapy (not combined with pharmacological therapy) have not been fully explored. The relationship between mind, body, and PTSD still holds many secrets.
Using structuralmagnetic resonance imaging( MRI ), Wignall et al., (2004) designed a correlational study measuring the HPC volume in recent-onset PTSD patients to further distinguish the onset timeframe of the reduced HPC volume. Two control groups were used; trauma exposed group that did not develop PTSD symptoms, and a non-trauma-exposed non-symptomatic group. Consistent with the fact not all trauma exposed individuals develop PTSD (Kessler et al., 1995), and reports of lowered glucocorticoid levels in chronic non-remitting PTSD patients, Wignall et al. (2004) hypothesized reduced HPC volume was a predisposing factor to the development of PTSD and smaller HPC volume would be measured in the recent-onset patients.
Victims of road traffic accident, assault, or occupational accident without any prior DSM-IV Axis I disorder were recruited from the Accident and Emergency Department of the Northern General Hospital in Sheffield, United Kingdom. Time between the trauma and MRIs had a mean of 158 days, +/- a standard deviation (SD) of 41 days (Wignall et al., 2004). This is important to note because without treatment, the majority of traumatic stress victims will no longer show symptoms of PTSD (meeting DMS-IV requirements for the diagnosis) 90 days post-stressor. Given the mean, SD, and assuming a normal distribution, it can be calculated that when scanned, approximately 95% of the trauma-exposed participants were more than 90 days post-trauma, thus meeting diagnostic criteria for a PTSD diagnosis.
Wignall et al. (2006) recorded the volume of the HPC, amygdala, gray matter (GM), white matter (WM), whole brain volume (WBV), and cerebral spinal fluid (CSF). Two-tailed independent samples t tests with alpha set <0.05 were performed using GM, WM, WBV, and CSF as covariates. Psychiatrists used the Clinician Administered Posttraumatic Stress Disorder Scale (CAPS) to rate the severity of patients’ PTSD symptoms. Using Pearson’s r for the correlation coefficient between the CAPS score and HPC volume, statistical analysis showed a significant reduction of the right HPC volume in PTSD diagnosed patients compared to all other subjects. The results suggest that either the smaller HPC volume predisposes the development of PTSD (in which prescreen before combat becomes increasingly more important than post-trauma treatments), or the loss of mass occurs soon after trauma (within a mean of 158 days), pointing to the importance of early interventions post-trauma to limit the release of glucocorticoids (Wignall et al., 2004). Although unlikely, it is possible that HPC volume reduction occurred between the traumatic incident and the MRI, but further longitudinal research is needed to assess if the reduced HPV volumes are a predisposing risk factor, or are caused by the excessive glucocorticoids circulating in the body post-trauma (Wignall et al., 2004).
Two limitations to the study include the small sample size and the lack of age matching between the control and traumatic-stressor groups. Age differences were controlled for using covariate analyses but given the many ways a brain will can change as it becomes older, having age matched participants will certainly yield stronger results. The types of traumas experienced by the participants should also be considered; previous studies involved victims of war and rape traumas, whereas Wignall et al., (2006) selected victims of accidents and assault. Additionally, unreported stressors such as domestic violence and rape were not screened for prior to participant inclusion into the study; these undetected traumas could unknowingly alter the psyche of participants.
Taken together with the pathophysiology frequently found in PTSD patient, cognitive behavioral therapy (CBT) has been widely used to treat the PTSD symptomatology. Resick & Schnicke (1992) were among the first to apply the basic principles of cognitive restructuring to the treatment of PTSD symptoms. Mueser et al., (2008) designed an experiment to test if CBT could benefit clients with comorbid PTSD and a severe mental illness by reducing the symptoms and severity of the PTSD. Utilizing homework assignments and tools to illicit changes in patients’ negative trauma-related beliefs Mueser et al., (2008) hypothesized that CBT will be more effective than therapy as usual (TAU) at eliminating PTSD diagnosis, reducing PTSD symptoms and negative trauma-related cognitions, and improving patients’ knowledge of PTSD.
Participants with comorbid PTSD and severe mental illness were prescreened from populations utilizing services form a community mental health facility and on some form of SSI. The screening methods used by Mueser et al., (2008) should be carefully considered; significant differences could exist between the selected population and those individuals suffering the same diagnoses yet are functioning independently in the community. After acceptance into the study, patients were randomly assigned to either the therapy as usual (TAU) control group or the experimental group to receive CBT treatment.
Randomization was stratified based on site (there were four different treatment facilities, which was shown to not be of statistical significance) to ensure the groups were equal in terms of the location they received their treatment to account for variations in the environments and mental health teams treating clients. The CBT program, consisting of twelve to sixteen sessions concentrating on psychoeducation of PTSD and how its symptoms can be expressed, cognitive restructuring, and breathing retraining was administered to the experimental group. PTSD symptom severity across all participants using the CAPS scores at three and six months posttreatment and mean scores between the two groups were compared using two-tailed independent groups t tests with alpha set at <0.05 (Mueser et al., 2008).
The CBT treatment significantly reduced the PTSD symptoms, with a particularly strong reduction of the patients’ negative trauma-related cognitions (Mueser et al., 2008). However promising, this study is limited due to the fact psychotropic medication prescribed to the participants was not added to the statistical model as a covariate. Additionally, persons prescreened for inclusion volunteered to participate in the study; they may differ in psychological measures of personality and psychiatric severity compared to patients not interested in participating in the study, and certainly compared to non-treatment seeking persons with a PTSD diagnosis.
In additional to CBT, antidepressants are commonly prescribed to treat PTSD symptoms such as depression and anxiety. Davis et al. (2006) reviewed four long-term studies published in PUBMED which focused on the effects of selective serotonin reuptake inhibitors (SSRIs) on the HPC volumes of PTSD patients. The HPC is an especially neuroplastic region of the limbic system, responding to changes in levels of endocrine hormones (which mediate changes in synapse formation and dendritic structure) and neurotransmitters. SSRIs administered to animals have been documented to regenerate HPC volume via neurogenesis thus mediating the stress induced atrophy (Sapolsky, 2000). Consistent with these findings, after treatments lasting 1-3 years of paroxetine (Paxil) or sertraline (Zoloft), Davis et al. (2006) recorded a mean HPC volume increase of ≈5% in patients with PTSD and was strongly correlated with a reduction in the psychological symptoms.
Since Sapolsky’s experiments have shown that stress can inhibit the HPA functioning and lead to HPC neuronal atrophy; an enriched environment (theoretically created via psychotherapy) should have the opposite effect and facilitate neurogenesis. Working from this theory, Lindauer et al. (2005) hypothesized that a reduction in PTSD symptomatology after psychotherapy will lead to an increase in HPC volume; similar to what had been found after paroxetine and sertraline therapy. To test, Lindauer et al. (2005) designed a randomized clinical trial involving eighteen patients with PTSD (which were randomly assigned to treatment (n=9) or waiting-list group (n=9)) and 14 traumatized control subjects without PTDS symptoms. The Structured Interview for Posttraumatic Stress Disorder (SI-PTSD) was administered for evaluation of PTSD symptom severity in participants. Finally, they received a structural MRI to obtain base-line measurements of the brain structures of interest. Consistent with previous studies, PTSD patients had significantly smaller HPC volumes at baseline than the control subjects, without presenting any variability between the treatment and wait-list groups of PTSD patients (Lindauer et al., 2005).
During treatment, the experimental PTSD group received 16 weekly individual sessions of brief eclectic psychotherapy (BEP) lasting 45–60 minutes each. Sessions included psychoeducation, imaginal exposure and psychotherapy. Post-treatment symptom severity was measured by the SI-PTSD and a second round of MRIs was completed to obtain post-treatment measurements of the HPC, amygdala, GM, WM, and CSF (Lindauer et al., 2005). BET sessions significantly reduced PTSD symptom severity for the treatment group on all clinical variables compared to the wait-list group. Surprisingly, post-treatment MRIs did not show any change in HPC volume of the experimental group. Independent t tests (with an alpha set at <0.05) and a multivariate analysis of covariance did not show any other statistically significant differences between the treatment group and wait-list group.
Flaws in the research design include that Lindauer et al. (2005) studied a relatively small sample size, measured HPC volume only over a four month period (whereas studies using pharmacological therapies evaluated patients HPC volume changes over the course of >1 year), and follow up measurements after the original psychotherapy treatments commenced were not performed. Additionally, the trauma-exposed groups with PTSD and the trauma-exposed groups without PTSD we not matched for the number and types of traumas which resulted with the PTSD group having experienced a significantly higher proportion of childhood traumas (Lindauer et al., 2005).
Psychotherapy having cognitive benefits and reduction of psychological symptoms associated with PTSD in absence of any physiological changes is perplexing; even more so when the psychological improvement mediated by antidepressants is accompanied with physiological improvement. Future studies should go beyond volumetric studies of limbic system structures. For example, cerebral blood flow has (CBF) been used to determine the activity and functioning on areas of the limbic system in subject groups of social phobia patients (which share the anxiety and fear found in PTSD patients) treated with CBT or citalopram (Furmark et al., 2002). Follow up studies resulted with the psychotherapy and psychopharmacology treatments producing identical physiological improvements and being correlated with significant clinical improvement. Similarities between the two disorders suggest the possible efficacy of CBF as a potential measurement for the physiological effects of psychotherapy in PTSD.
Lastly, not all studies have replicated these findings; some do not show any volumetric difference in the HPC of PTSD patients compared to controls. Ham et al. (2007) attribute the inconsistency of results to the different methods used to measure the volume from the MR images. N -acetylaspartate (NAA) is a commonly used chemical indicator to test for integrity and density of CNS neurons.Lower levels of NAA found in the HPC, amygdala, ACC and medial prefrontal cortex (mPFC) have been correlated with PTSD symptoms (Kolassa & Ebert, 2007). Measuring the NAA concentrations in the HPC and anterior cingulate cortex (ACC) of victims of a traumatic stress that developed PTSD and those that did not, Ham et al. (2007) reported statistically significant lower levels of NAA in the ACC and HPS of PTSD patients. It cannot be ruled out that perhaps the changes in HPC volume due to psychotherapy are too subtle to be detected through MR imaging; thus using NAA levels to determine the integrity of the neurons is a much needed variable in future studies (Schuff, et al., 2001).
It seems to almost impossible that the psychological effects ofabnormal HPC size in PTSD patients can be relieve with psychotherapy. Physiological differences are not absolute determinants in the expression of negative emotional responses; our minds still retain the ultimate power over our bodies. The proof CBT can mediate negative effects of pathophysiology is simply an echo of that mantra found in nearly every single psychology book: human behavior is an interaction between biology/genes, environment, and our individual perception.
Davis, L., Frazier, E., Williford, R., & Newell, J. (2006). Long-term pharmacotherapy for post-traumatic stress disorder. CNS Drugs, 20 (6), 465-76.
Furmark, T., Tillfors, M., Marteinsdottir, I., Fischer, H., Pissiota, A., & Långström, B. (2002). Common changes in cerebral blood flow in patients with social phobia treated with citalopram or cognitive-behavioral therapy. Archives of General Psychiatry, 59 (5), 425-33.
Ham, B., Chey, J., Yoon, S., Sung, Y., Jeong, D., & Kim, S. (2007). Decreased N-acetylaspartate levels in anterior cingulate and hippocampus in subjects with post-traumatic stress disorder: A proton magnetic resonance spectroscopy study. European Journal of Neuroscience, 25 (1), 324-29.
Kessler, R., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archive of General Psychiatry 52:1048 –1060.
Kolassa, I., & Elbert, T. (2007). Structural and functional neuroplasticity in relation to traumatic stress. Current Directions in Psychological Science, 16 (6), 321-25.
Lindauer, R., Vlieger, E., Jalink, M., Olff, M., Carlier, I., Majoie, C., et al. (2005). Effects of psychotherapy on hippocampal volume in out-patients with post-traumatic stress disorder: A MRI investigation. Psychological Medicine, 35 (10), 1421-1431.
Mueser, K., Rosenberg, S., Xie, H., Jankowski, M., Bolton, E., Lu, W., Hamblen, J., Rosenberg, H., McHugo, G. J., & Wolfe, R. (2008). A randomized controlled trial of cognitive-behavioral treatment for posttraumatic stress disorder in severe mental illness. Journal of Consulting & Clinical Psychology, 76 (2), 259-71.
Resick, P., & Schnicke, M. (1992). Cognitive processing therapy for sexual-assault victims. Journal of Consulting Clinical Psychology 60: 748-56.
Sapolsky, R. M. (2000): Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Archive of General Psychiatry57: 925–35.
Schuff, N., Neylan, T., Du Lenoci, M., Weiss, D., Marmar, C., & Weiner, M. (2001). Decreased hippocampal N-acetylaspartate in the absence of atrophy in posttraumatic stress disorder. Biological Psychiatry, 50: 952–59.
Wignall, E., Dickson, J., Vaughan, P., Farrow, T., Wilkinson, l., & Hunter, M. (2004). Smaller hippocampal volume in patients with recent-onset posttraumatic stress disorder. Biological Psychiatry, 56 (11), 832-36.