Other Studies

Emerging evidence about the effects of stress on the hippocampus suggests that a unique developmental trajectory exists. For example, two initial, separate studies conducted by De Bellis and colleagues (1999, 2002) using independent samples of children with PTSD secondary to CA indicate that the hippocampus is unaltered, at least in terms of volumetric size. More recent work has demonstrated that the hippocampus may actually be larger in children who have been abused (Tupler & De Bellis, 2006). However, studies with adults have painted a different picture, methodological limitations notwithstanding (Jelicic & Merckelbach, 2004). A decade ago, reports of reduced left hippocampal volume in adults with childhood trauma and a current diagnosis of PTSD or dissociative identity disorder first appeared in the literature (Bremner et al., 1997; Stein, 1997). Subsequently, significant reductions in hippocampal volume were observed in women with borderline personality disorder and a history of childhood abuse (Driessen et al., 2000) as well as in adult females with a history of prepubertal physical and/or sexual abuse and depression (Vythilingam et al., 2002). To our knowledge, no comparable studies have been conducted solely with adolescent youth who have CA histories.

In concert with other work found in the present book, a developmental psychopathology approach incorporating a longitudinal design is warranted to ascertain how the hippocampus, specifically, and the brain, in general, transform across childhood and adolescence toward adult maturity. One useful and important means of effecting this type of research is preclinical animal-modeling studies. In particular, rats have been successfully utilized in relatively brief time periods (i.e., days to weeks) to document brain changes longitudinally due to their protracted rate of development (Andersen, 2003). Such a model is also needed to support a causal role of the effects of stressful experiences, such as CA, on neurodevelopment. Retrospective clinical studies of CA cannot prove that alterations in brain morphology or function are actually a consequence of the abuse. Ethically, children also cannot (and should not ever!) be randomly assigned to an experimental group that is exposed to CA and compared to a group of children who do not receive such maltreatment. In contrast, animal research enables scientists to randomly assign juvenile rats to either an experimental stress exposure group or a control group and compare them at later stages of development. Specifically in rodents, the manipulation of levels of maternal care (an important species-relevant stressor) has been documented to affect brain structure and function. This type of handling has included the removal of the nursing dam from the pups for varying amounts of time (Andersen & Teicher, 2004; Kehoe, Shoemaker, Triano, Callahan, & Rappolt, 1998; Levine, 1967), the use of surrogate mothers (Harlow, Dodsworth, & Harlow, 1965), and the selection of population extremes of high- and low-quality maternal behaviors (Caldji, Diorio, & Meaney, 2000).

Recently, we conducted one such study to test the hypothesis that CA might exert an effect on neurodevelopment (i.e., smaller hippocampus) that only emerges at a much later stage of brain maturation (Andersen & Teicher, 2004). This postulate differs from ones stating that reduced hip-pocampal size is the culminating end product of years or decades of PTSD or depressive symptoms or that such a reduction is a preexisting risk factor associated with the emergence of psychopathology that persists into adulthood (Bremner, 2003; Gilbertson et al., 2002). Specifically, developing rats were stressed by periods of maternal separation and isolation that occurred between 2 and 20 days of age. Following this early stress exposure, the rats were sacrificed at weanling, peripubertal, young adult, or full adult ages. Using synaptophysin via immunohistochemistry as a marker of synaptic density, we found no significant difference in synap-tophysin measures at weanling and peripubertal ages, but a marked rise (overproduction) in synaptic density at the young adult age in normal control rats that was not observed in rats exposed to early stress. Synaptic density declined at subsequent ages, presumably as a result of pruning, but differences in synaptic density remained between normal control rats versus those exposed to early stress. In parallel, our clinical work has elucidated significant bilateral reductions in hippocampal volume in subjects aged 18 to 22 years who were exposed to repeated childhood sexual abuse (unpublished observations).

The amygdala is another brain region that has received attention from stress and trauma researchers given its crucial role in fear conditioning, the formation and recollection of emotional memory, the learning of nonverbal motor patterns, and the triggering of fight-or-flight responses (Lang & Davis, 2006; LeDoux, 1996, 2003). Structural imaging studies of amygdala volume in abuse survivors with PTSD found no differences compared to control subjects (Bremner et al., 1997; De Bellis et al., 1999; Stein, 1997). Although an initial study by Driessen et al. (2000) reported an 8% reduction in bilateral amygdaloid volume in women with a history of CA and a diagnosis of borderline personality disorder (BPD), this finding was not independently replicated in a separate study of individuals with BPD (Brambilla et al., 2004). A smaller amygdala, however, has been documented in individuals with dissociative identity disorder (Vermetten, Schmahl, Lindner, Loewenstein, & Bremner, 2006)—a condition associated with a history of severe CA. In addition, increased activation of the left amygdala was recently observed during the acquisition phase of a conditioned fear paradigm in women with CA-related PTSD (Bremner et al., 2005). Because amygdala overactivation appears to be a critical factor in PTSD (Rauch et al., 2000; Shin, Rauch, & Pitman, 2006), we have hypothesized that a smaller amygdala provides protection from the emergence of PTSD following CA or facilitates recovery from the disorder.

Since our initial EEG coherence findings, subsequent studies by other investigators have provided further evidence that CA alters cortical neuronal development in humans. The anterior cingulate cortex seems especially vulnerable, even across developmental stages. For example, in 11 children and adolescents who met DSM-IV criteria for PTSD secondary to CA, a significant reduction in the N-acetylaspartate:creatine ratio (an index of neuronal density and viability) was detected via magnetic resonance spectroscopy, which is indicative of neuronal loss and dysfunction in this brain area (De Bellis, Keshavan, Spencer, & Hall, 2000). In addition, Bremner and colleagues (2005) have documented decreased volume of the right anterior cingulate cortex in adults with CA-related PTSD as well as diminished anterior cingulate function during the extinction phase of a conditioned fear paradigm. Declines in functioning have also been seen in the right temporal pole/anterior fusiform gyrus, left precuneus, and posterior cingulate cortex of adults with histories of CA and a current diagnosis of borderline personality disorder (Lange, Kracht, Herholz, Sachsse, & Irle, 2005). Additional developmental effects were reported by Carrion et al. (2001), who found attenuated frontal lobe asymmetry and smaller total brain and cerebral volumes in 24 children (7 to 14 years of age) with a history of CA and PTSD symptoms. Together, these studies suggest that the effects of CA alter the normal developmental trajectory of the neocortex.

0 0

Post a comment