Resilience after 9/11: Multimodal neuroimaging evidence for stress-related change in the healthy adult brain

Abstract

Exposure to psychological trauma is common and predicts long-term physical and mental health problems, even in those who initially appear resilient. Here, we used multimodal neuroimaging in healthy adults who were at different distances from the World Trade Center on 9/11/01 to examine the neural mechanisms that may underlie this association. More than 3 years after 9/11/01, adults with closer proximity to the disaster had lower gray matter volume in amygdala, hippocampus, insula, anterior cingulate, and medial prefrontal cortex, with control for age, gender, and total gray matter volume. Further analysis showed a nonlinear (first-order quadratic) association between total number of traumas in lifetime and amygdala gray matter volume and function in the whole group. Post hoc analysis of subgroups with higher versus lower levels of lifetime trauma exposure revealed systematic associations between amygdala gray matter volume, amygdala functional reactivity, and anxiety that suggest a nonlinear trajectory in the neural response to accumulated trauma in healthy adults

Introduction

More than half of the adults in the United States experience one or more psychological traumas in their lifetime (Kessler et al., 1995). Higher levels of trauma exposure occur in certain subgroups, e.g., those who live in unsafe neighborhoods (Boothroyd and Evans, 2001), or who are exposed to terrorism (Galea et al., 2002) or war (Bramsen et al., 2000). Although only a small percentage of these people develop posttraumatic stress disorder (PTSD) (Kessler et al., 1995), trauma exposure predicts lifetime increases in mental health disorder in the general population (Brown, 1993, Kendler et al., 2003, Kessler et al., 1995). Even resilient trauma-exposed individuals are likely to show heightened cardiovascular reactivity to trauma reminders (Tucker et al., 2007) and to have greater vulnerability to PTSD with subsequent trauma exposure (Bremner et al., 1993). Resilient individuals may also experience cognitive declines (Stein et al., 2002), altered catecholamine levels (Otte et al., 2005, Young and Breslau, 2004a, Young and Breslau, 2004b), more chronic illness later in life, and decreased mean life expectancy (McFarlane, 1997). However, the neural mechanisms that may underlie the effects of trauma in this population remain unclear.

Research using animal models has provided some evidence of the effects of stressor exposure on neuronal structure and function. This research suggests that the amygdala, hippocampus, and medial prefrontal cortex may be especially vulnerable (McEwen, 2005, Mitra et al., 2005, Vyas et al., 2002). In general, uncontrollable stressors have been shown to produce hyperexcitability in the amygdala and increased fearful behavior (Adamec et al., 2005, Maier and Watkins, 1998, Rosen and Schulkin, 1998). Acute (Mitra et al., 2005) and chronic (Vyas et al., 2002) restraint stress have produced increased anxiety and increased dendritic spine density in the basolateral amygdala. Chronic restraint stress has been shown to produce hypertrophy of the dendritic arborization in the amygdala, accompanied by dendritic atrophy and decreases in spine density in medial prefrontal areas (Radley et al., 2006) and the hippocampus (Vyas et al., 2002). Severe social stress has produced similar dendritic remodeling in the hippocampus (Blanchard et al., 1993, McKittrick et al., 2000). In addition, chronic stress may reduce neurogenesis in the hippocampus (Gould et al., 1998). In general, stress-induced neuronal changes in prefrontal and hippocampal areas tend to recover with rest (Heine et al., 2004, McEwen, 2005, Vyas et al., 2004) but changes to the amygdala (and the accompanying fearful behaviors) seem to be more persistent (Adamec et al., 2005, Vyas et al., 2004).

The terrorist attacks of 9/11/01 provide a unique window into these processes in humans because closer proximity to this disaster has been shown to be associated with increased psychological distress (Blanchard et al., 2004, Galea et al., 2002) and increased amygdala activity (Ganzel et al., 2007, Sharot et al., 2007). In the present study, we measured gray matter volume and functional amygdala activity during passive viewing of fearful versus calm faces (Breiter et al., 1996) in healthy adults who were at different distances from the World Trade Center (WTC) on 9/11/01. We also measured lifetime trauma exposure, state anxiety, and current PTSD symptoms. More than 3 years after 9/11/01, adults with closer proximity to the WTC had lower gray matter volume in multiple brain regions, including the amygdala. Amygdala gray matter volume was inversely correlated with amygdala response to emotional faces. Lifetime trauma exposure was related to amygdala gray matter volume according to a first-order quadratic, with the slope of the relationship reversing at higher levels of trauma exposure. Analysis of subgroups with higher versus lower levels of lifetime trauma exposure revealed systematic associations between amygdala gray matter volume, amygdala functional reactivity, and anxiety that suggest a nonlinear trajectory in the neural response to accumulated stress.