Neural correlates of trauma script-imagery in posttraumatic stress disorder with and without comorbid major depression: A functional MRI investigation

Abstract

The goal of this study was to compare neural activation patterns in patients with PTSD with and without current comorbid major depression. Traumatized subjects with PTSD (n = 11), PTSD + major depression (MDD, n = 15), and subjects (n = 16) who met criterion A for PTSD but never developed the disorder were studied using the script-driven symptom-provocation paradigm adapted to functional magnetic resonance imaging (fMRI) at a 4-Tesla field strength. Both the PTSD + MDD and PTSD − MDD groups revealed decreased brain activation in the anterior cingulate gyrus (BA 24) and the right ventrolateral prefrontal cortex (BA 47). After covariation for differences in PTSD severity between these groups, the left insula (BA 13) remained more significantly activated in the PTSD − MDD group than in the PTSD + MDD group. In contrast, the PTSD + MDD group showed greater activation than the PTSD − MDD group in the bilateral anterior cingulate gyrus (BA 24) and posterior cingulate cortices (BA 23, 31). These results suggest different patterns of brain activation related to comorbid major depression occurring in the context of PTSD.

Introduction

Posttraumatic stress disorder (PTSD) may be the disorder that is most frequently associated with other comorbid Axis I psychiatric disorders. The National Comorbidity Study reveals that a lifetime history of at least one other Axis I disorder was present in 88.3% of men and 79% of women with a lifetime diagnosis of PTSD (Kessler et al., 1995). The most common psychiatric comorbidities included major depression (major depressive episode: 47.9% in men, 48.5% in women), substance use disorder (alcohol abuse/dependence: 51.9% men, 27.9% women; drug abuse/dependence: 34.5% men, 26.9% women), and other anxiety disorders (generalized anxiety disorder (GAD): 16.8% men, 15.0% women; panic disorder: 7.3% men, 12.6% women; simple phobia: 31.4% men, 29.0%; social phobia: 27.6% men, 28.4% women; agoraphobia: 16.1% men, 22.4% women) (Kessler et al., 1995; also see Wang et al., 2005).

A longstanding debate in the field of traumatic stress studies has been the question of whether major depression occurring in the context of PTSD represents a truly separate comorbid disorder or constitutes a posttraumatic mood disorder (Davidson and Fairbank, 1993, Friedman and Yehuda, 1995, Sher, 2005). Evidence suggests that PTSD with comorbid major depression differs both psychologically and biologically from PTSD without major depression. PTSD with comorbid depression has been shown to be characterized by significantly greater symptom severity and lower levels of social and occupational functioning as compared to PTSD without major depression (Moore and Boehnlein, 1991, Mintz et al., 1992, Skodol et al., 1996, Karam, 1997, Shalev et al., 1998, Koren et al., 1999, Mollica et al., 1999). In fact, a large-scale epidemiological survey reported that PTSD patients with comorbid depression were five times more likely to exhibit functional impairment as compared to those with PTSD alone (Mollica et al., 1999). Moreover, PTSD with comorbid depression has been shown to be particularly associated with the melancholic-subtype of depression, and melancholic symptoms have been correlated with PTSD symptoms of excessive guilt and emotional numbing (Constans et al., 1997). In addition, increased impulsivity, hostility and suicidality have been reported in patients with PTSD with comorbid depression as compared to those with PTSD alone (Oquendo et al., 2005). In terms of pharmacological treatment, PTSD with comorbid depression has been shown to be more treatment resistant than PTSD without major depression (Berkheimer et al., 1985, Frank et al., 1988, Reist et al., 1989). Furthermore, with regard to sleep architecture, Woodward et al. (1996) have shown that PTSD patients with comorbid depression demonstrated less slow wave sleep as well as less facial electromyographic activity as compared to PTSD patients without comorbid depression. Brain activation patterns, however, comparing the effects of comorbid depression occurring in the context of PTSD have not been investigated to date.

Limbic and paralimbic structures that have most consistently been shown to be involved in the pathophysiology of PTSD using traumatic-event script-driven imagery include the medial prefrontal cortex (Bremner et al., 1999a, Bremner et al., 1999b, Lanius et al., 2001, Lanius et al., 2003, Shin et al., 2004), the anterior cingulate cortex (Liberzon et al., 1999, Bremner et al., 1999a, Bremner et al., 1999b, Lanius et al., 2001, Lanius et al., 2002, Lanius et al., 2003, Britton et al., 2005), and the amygdala (Rauch et al., 1996, Shin et al., 1997, Liberzon et al., 1999, Shin et al., 2004). In addition, changes in thalamic (Liberzon et al., 1996/1997, Bremner et al., 1999a, Lanius et al., 2001, Lanius et al., 2003) and insular (Osuch et al., 2001) activation have been reported. However, inconsistencies across studies have often been observed; studies have reported activation of different brain regions as well as variable activation within the same brain structure (Shin et al., 1997, Bremner et al., 1999a, Bremner et al., 1999b, Liberzon et al., 1999, Shin et al., 1999, Lanius et al., 2003, Shin et al., 2004, Shin et al., 2005). One possible reason underlying the variability in results of PTSD neuroimaging studies conducted to date is that studies significantly vary with respect to the degree to which PTSD samples exhibited multiple comorbidities, most notably major depression. In the script-driven-imagery neuroimaging studies published to date, between zero and 62% of PTSD subjects met criteria for current major depression (calculated from: Shin et al., 1997, Bremner et al., 1997, Bremner et al., 1999a, Shin et al., 1999, Liberzon et al., 1999, Osuch et al., 2001, Lanius et al., 2001, Lanius et al., 2002, Lanius et al., 2003, Shin et al., 2004, Britton et al., 2005); Britton et al. (2005) is the only script-driven neuroimaging study published to date that has not included PTSD subjects with current comorbid depression. In addition to the comorbidity issue, gender differences may also contribute to inconsistencies across studies. Given the higher prevalence rate of depression in women and possible gender effects on brain functioning (Luders et al., 2004), it is also worth noting that these previously studied PTSD samples were composed of between zero and 100% of female subjects.

Research has shown there to be a significant degree of overlap in the brain structures (i.e., anterior cingulate and prefrontal cortex) involved in PTSD and depression symptomatology (see Davidson et al., 2002 for review). Mayberg (1997) has proposed a cortico-limbic model of depression based on brain regions consistently identified in positron emission tomography (PET) studies of depression. In this model, disturbances in dorsal frontal-limbic systems, including dorsolateral prefrontal cortex (BA 9, 44, 46), dorsal anterior cingulate (BA 24), posterior cingulate (BA 29, 30, 31), and inferior parietal lobe (BA 40), are thought to primarily mediate the cognitive features of depression including reductions in attentional capacity, working memory, and decision-making ability. Vegetative and somatic features of depression, on the other hand, are postulated to be mediated predominantly by ventral brain systems, including the ventral prefrontal cortex (BA 47), ventral anterior cingulate (BA 25), ventral insula, hypothalamus, hippocampus, and insula. Mayberg further hypothesizes that the rostral anterior cingulate (BA 24), through its anatomical connections with both the dorsal and ventral systems, may serve important regulatory functions by facilitating information processing between these systems (Goldapple et al., 2004). Incidentally, hypermetabolism in this region has also been shown to predict a positive antidepressant response in patients with major depression (Mayberg et al., 1997).

Prior investigations examining the neuronal circuitry underlying PTSD using traumatic script-driven imagery have not adequately assessed brain activation patterns in PTSD with versus PTSD without current comorbid major depression. It is therefore unclear whether some of the changes described above are specific to PTSD, or rather are partly related to the comorbid major depressive disorder present in some of the subjects. The present study was therefore designed to compare the neural activation patterns in patients with PTSD with and without comorbid major depression. Based on the existing PTSD and depression neuroimaging literature as well as our pilot data, we hypothesized altered brain activation in the anterior (BA 24, 32) and posterior cingulate gyrus (BA 31), the medial prefrontal cortex (BA 10), the inferior-middle frontal gyri (ventrolateral prefrontal cortex, BA 47) and the insula (BA 13) in PTSD subjects with versus without major depression. Given the seminal nature of the present study, group comparisons were explored without specific hypothesis about the directionality of the effect.