اثر ترومای بر واکنش پذیری استرس در جوانان تهاجمی

Abstract To address gaps in the literature related to the contribution of childhood trauma on aggression, we evaluated salivary cortisol and heart rate changes to psychological challenge in aggressive children with various degrees of trauma. We hypothesized that traumatized and aggressive youths will exhibit higher responsiveness to an active challenge (Violent film—VF) than aggressive youth with no trauma but will not differ when viewing a Non-Violent film (NVF). A total of 25 children (aged 9–12; M = 15, F = 9) with history of aggression were assessed for trauma exposure. Children viewed the two films in randomized order. Four salivary cortisol and pulse measurements were obtained before (T1), 15 min after the start (T2), at the end (T3), and 15 min following the end of the movie (T4). Repeated measures Analysis of Covariance (ANCOVA) using Film (VF/NVF), Cortisol/Time at T1–T4, Group (Trauma/Non-Trauma), and Film Order were performed with age and gender as covariates. There were significant main effects for Group and Cortisol/Time for the Trauma group showing greater cortisol responsiveness than the Non-Trauma group that was most pronounced during the NVF. These results suggest that aggressive youth with personal history of trauma may exhibit unique biological characteristics, which may have important implications for classification and treatment.

. Introduction Aggressive behaviors are one of the leading causes for referrals for psychiatric evaluation of young children (Peterson et al., 1996). Ensuing in depth assessments of youth with aggression frequently establishes diagnoses like attention deficit/hyperactivity disorder (ADHD) or conduct disorder (CD); however, aggression is not a core feature of ADHD and is not necessarily present in children with CD. Further, childhood traumatization has also been linked to the development of subsequent aggressive behaviors, but similarly, aggression is not a required symptom of childhood posttraumatic syndrome. These observations point out that the relationship between aggression, disruptive behaviors and childhood trauma is not straight-forward. Aggression has been linked to states of baseline physiological under-arousal (Murray-Close et al., 2008) that may translate psychologically into either aversive emotional experiences (Coren, 1999), or experiences of reduced fear (Raine, 2002a and Raine, 2002b). In result, affected individuals appear vulnerable to engage in aggressive behaviors because the latter produce physiological arousal that can in turn alleviate aversive emotions. Alternatively, diminished experience of fear may compromise the role of fearful states as a deterrent to physical fighting and aggression. All of the above seem to negatively affect the functioning of the behavioral inhibition system, which guides individual's reactions to challenging situations (Fox et al., 2005). For instance, several studies have shown that while behavioral inhibition appears associated with increased cortisol (King et al., 1998 and Blair et al., 2004), behavioral disinhibition, viewed as a tendency to react with boldness and spontaneity to novel situations (Lopez et al., 2004 and Hirshfeld-Becker et al., 2007) has been linked to decreased cortisol during psychological challenge (Blair et al., 2004). Not surprisingly, abnormal levels of cortisol responsiveness have been observed in children with ADHD and CD, who are often characterized as behaviorally disinhibited. Attenuated cortisol responses have been reported in children with ADHD upon awakening (Blomqvist et al., 2007) or when exposed to stress such as performing timed cognitive tests (King et al., 1998 and Hong et al., 2003) compared to healthy controls. Low hypothalamic–pituitary–adrenal (HPA) axis activity, however, does not seem specific to ADHD; low basal cortisol levels have been linked to physical aggression in children with CD (McBurnett et al., 2000 and Pajer et al., 2001), and oppositional defiant disorder with or without comorbid ADHD (Snoek et al., 2004). Low basal cortisol also seems to predict more aggressive behavior in later adolescence in males regardless of diagnosis (Shoal et al., 2003). Additionally, differences in cortisol reactivity have been repeatedly documented in association with aggression in children with CD compared to nonaggressive children, but the directionality of the findings has varied across studies (van Goozen et al., 2000, van de Wiel et al., 2004 and Kempes et al., 2008). A recent meta-analysis for both basal cortisol (k = 72 studies, N = 5480) and for cortisol reactivity to a stressor (k = 29 studies, N = 2601) failed to find a relationship between these cortisol measures and externalizing behaviors in adolescents ( Alink et al., 2008). Conversely, low resting heart rate seems to be the best-replicated biological correlate of aggression in children and adolescents, which may reflect reduced noradrenergic functioning and a fearless, stimulation-seeking temperament ( Raine, 2002a, Raine, 2002b, Ortiz and Raine, 2004 and Lorber, 2004). However, some argue that heart rate variability is different for different aggression types so that disinhibited or reactive aggression appears linked to decreased heart rate variability whereas premeditated aggression seems associated with increased heart rate variability ( Scarpa et al., 2010). Of particular interest to the topic of this paper is the relationship between childhood traumatization and the consequential development of aggression, which has been documented in youth who either experienced (Connor et al., 2003 and Hazen et al., 2006) or witnessed abuse (Hazen et al., 2006 and Chemtob et al., 2008). These trauma-related consequences may further persist over the course of development, supported by the finding that physically abused children are significantly more likely to be arrested for both non-violent and violent offenses as adolescents (Lansford et al., 2007). Also of interest is the observation that irritability in adult trauma survivors, which is similar to “reactive aggression” in children, is one of the diagnostic criteria of posttraumatic stress disorder (PTSD). However, as important as it is to investigate the interaction between psychosocial and biological processes in the development and maintenance of aggression (Raine et al., 1997, Raine, 2002a and Raine, 2002b) the purported biological pathways leading from childhood trauma to aggression are complex and remain poorly understood. Trauma in children alone may result in heightened basal cortisol levels and cortisol reactivity (Carlson and Earls, 1997, Cicchetti and Rogosch, 2001a, Gunnar et al., 2001 and Gunnar et al., 2009). These effects of traumatization on the relation between aggression and HPA system have been linked to more pronounced dysregulation of cortisol among aggressive children with history of trauma (Cicchetti and Rogosch, 2001a and Cicchetti and Rogosch, 2001b). This thesis is further supported by findings that the combination of heightened cortisol reactivity to provocation and experiences of victimization (physical abuse and community violence exposure) is associated with high levels of aggression (Raine, 2002a, Raine, 2002b and Scarpa and Ollendick, 2003) and that adverse parenting, family conflicts, and acute life events may contribute to an increased cortisol responsiveness in children with ADHD and externalizing problems (Freitag et al., 2009). One study that specifically examined the association between relational and physical aggression and maltreatment reported results suggesting that physiological correlates of aggression may be different for these two forms of aggression and may also differ among maltreated vs. nonmaltreated youths (Murray-Close et al., 2008). Less is known about the link between traumatization, aggression and heart rate changes in children outside of one study suggesting that increased heart rate seems to predict later development of PTSD in children exposed to traumatic injury (Kassam-Adams et al., 2005). Taken together, these reports clearly show that the relationship between traumatization, aggression, and cortisol and heart rate responsiveness is not one-dimensional and requires further investigations. While it appears that physiological under-arousal is more consistently associated with childhood aggression in particular, the effects of childhood traumatization on the association between aggression and biological correlates such as cortisol and heart rate responsiveness remain controversial. This interplay between traumatic exposure and disruptive behaviors poses the question as to whether aggressive behaviors in childhood that are underpinned by trauma may represent a biologically distinct subgroup. Given the fact that traumatic stress has a strong influence on both the HPA and autonomic nervous system, illustrated by altered responsiveness of cortisol and heart rate (Gunnar and Quevedo, 2008 and Chrousos, 2009), it is reasonable to examine if particular changes in these biological measures may differentiate aggressive youths with prior trauma exposure from aggressive counterparts with no trauma. Toward this purpose we designed an experimental paradigm that followed well established methodology for the assessment of physiological responsiveness to active and neutral stimuli indexed by the changes in salivary cortisol secretion and heart rate. The primary goal of this study was to collect preliminary data on measures of HPA and autonomic nervous system responsiveness to an active (Violent film) and neutral (Non-Violent film) psychological challenge in aggressive children with various degrees of trauma (i.e. Trauma vs. Non-Trauma youths). This protocol tested the hypothesis that traumatized and aggressive youths will exhibit higher levels of responsiveness to the active psychological challenge (Violent film) than aggressive youth with no trauma whereas the responsiveness to the neutral psychological challenge (Non-Violent film) will not differ between the two groups. We did not examine a group of nonaggressive children as we wished to focus on subtypes within a specific clinical group of young children (i.e. aggressive youth with and without trauma).

. Results 3.1. Demographics The two groups did not differ for age. The Non-Trauma group comprised 11 males and four females and the Trauma group had five males and five females. Ethnic distribution for the Non-Trauma group was: African-American = 8, Hispanic = 5, and Caucasian = 2; the distribution for the Trauma group was: African-American = 7, Hispanic = 2, and Caucasian = 1. T-scores on the clinical ratings are presented in Table 1. Table 1. Participants' demographics. Abbreviations: PTSRI — Posttraumatic Stress Reaction Index; CBCL — Child Behavior Checklist; DSM — Diagnostic and Statistical Manual; measures for CBCL aggression, CBCL delinquent, Conners' ADHD index, DSM hyperactivity, DSM inattention, DSM total index reflect T-scores; measures for PTSRI-Parent and PTSRI-Child reflect raw scores since T-scores for these instruments have not been estimated. Factors Traumatized youth (n = 10) Non-Traumatized Youth (n = 15) p-value Age, mean ± S.D., years 10.5 + 1.4 10.3 + 1.0 0.28 PTSRI-Parent 34.5 + 11.5 10.2 + 10.8 0.001* PTSRI-Child 30.8 + 10.6 13.2 + 10.9 0.001* CBCL aggression 59.0 ± 34.2 61.9 ± 14.3 0.74 CBCL delinquent 71.0 ± 11.8 64.3 ± 7.51 0.24 Conners' ADHD index 72.9 ± 17.2 66.5 ± 12.6 0.33 DSM hyperactivity 76.0 ± 17.8 70.9 ± 14.2 0.46 DSM inattention 69.5 ± 19.1 64.7 ± 13.9 0.51 DSM total index 73.8 ± 19.1 68.1 ± 13.7 0.43 Gender (M/F) 5/5 11/4 0.23 The asterisks indicate significant differences on scores of PTRSI-Parent and PTSRI-Child between the two groups (Traumatized vs. Non-Traumatized youths). Table options 3.2. Self-reports The self-reports of the children's experience during the movie viewing did not differ between the groups with most children rating feelings of “Upset/Sad” (F = 2.00, d.f. = 1/23, p = 0.503), “Mad/Angry” (F = 3.05, d.f. = 1/23, p = 0.426), and “Bored” (F = 0.47, d.f. = 1/23, p = 0.62) in the range of 1–2 out of 10. 3.3. Biological measures The two groups did not significantly differ on the baseline measures of cortisol and heart rate obtained before viewing each of the film segments. The main effect for group was registered for all time points during the Non-Violent film for the Trauma group at T1 (F = 4.690, d.f. = 1/23, p = 0.041), T2 (F = 16.358, d.f. = 1/23, p = 0.001), T3 (F = 8.172, d.f. = 1/23, p = 0.009) and T4 (F = 17.512, d.f. = 1/23, p = 0.0001) and also during the Violent film at T3 (F = 4.556, d.f. = 1/23, p = 0.044), indicating significantly heightened cortisol response in Trauma youth compared to Non-Trauma counterparts. There was a significant three-way interaction for Film Type × Cortisol/Time × Trauma (F = 3.085, d.f. = 3/69, p = 0.034) suggesting that the psychological challenge and trauma status of the children interacted and influenced the changes in the cortisol measures during the two movie viewings. The overall pattern of decreased cortisol secretion over the course of the film viewings is in accordance with other reports of decreased cortisol reactivity to stress in aggressive youth with ADHD. However, post hoc analyses showed a main effect for Cortisol/Time for the Trauma children (F = 2.964, d.f. = 3/69, p = 0.050) reflecting significantly different cortisol measures at the different time points of the two films and suggesting that HPA responsiveness was most pronounced during the Non-Violent film ( Fig. 1). In contrast, there was a main effect for film type for the Non-Trauma children only (F = 6.529, d.f. = 1/23, p = 0.023) ( Fig. 1), suggesting that for this group the Violent film elicited significantly higher overall HPA response. There was a significant interaction for cortisol collected at baseline Cortisol (T1) × Film Order (F = 8.454, d.f. = 3/15, p = 0.012) and a significant three-way interaction between Cortisol (T1) × Film Order × Film Type (F = 4.516, d.f. = 3/15, p = 0.008) for the non-traumatized group only, suggesting that the HPA system for the two groups responded differently in relation to the order in which the films were presented ( Fig. 2). There were no significant main effects for film type or any significant interactions for heart rate for either group. No significant main effect for gender was ascertained. There were also no significant interactions between measures of aggression and cortisol and pulse measures. Changes of salivary cortisol at different collection points during the movie ... Fig. 1. Changes of salivary cortisol at different collection points during the movie viewing. Time points of cortisol collections correspond to before the start of the film (Time 1), 15 min within the film viewing (Time 2), at the end of the each film (Time 3) and after a 15 min “wash-out” period (Time 4). Error bars reflect the standard error of the mean. Figure options Baseline cortisol at Visit 1 and Visit 2 of the experiment. The two groups ... Fig. 2. Baseline cortisol at Visit 1 and Visit 2 of the experiment. The two groups showed similar baseline cortisol levels at Visit 1 although these were film specific. Non-traumatized children (n = 15) exhibited predominantly habituation of the HPA responsiveness from Visits 1 to 2 whereas the traumatized youth (n = 10) demonstrated sensitization of the HPA axis response. Error bars reflect the standard error of the mean.