ماده سفید اختلالات ریزساختاری در لوب فرونتال بزرگسالان مبتلا به اختلال شخصیت ضد اجتماعی

Abstract Antisocial personality disorder (ASPD) and psychopathy involve significant interpersonal and behavioural impairments. However, little is known about their underlying neurobiology and in particular, abnormalities in white matter (WM) microstructure. A preliminary diffusion tensor magnetic resonance imaging (DT-MRI) study of adult psychopaths employing tractography revealed abnormalities in the right uncinate fasciculus (UF) (Craig et al., 2009), indicating fronto-limbic disconnectivity. However, it is not clear whether WM abnormalities are restricted to this tract or are or more widespread, including other tracts which are involved in connectivity with the frontal lobe. We performed whole brain voxel-based analyses on WM fractional anisotropy (FA) and mean diffusivity (MD) maps acquired with DT-MRI to compare 15 adults with ASPD and healthy age, handedness and IQ-matched controls. Also, within ASPD subjects we related differences in FA and MD to measures of psychopathy. Significant WM FA reduction and MD increases were found respectively in ASPD subjects relative to controls. FA was bilaterally reduced in the genu of corpus callosum while in the right frontal lobe FA reduction was found in the UF, inferior fronto-occipital fasciculus (IFOF), anterior corona radiata and anterior limb and genu of the internal capsule. These differences negatively correlated with measures of psychopathy. Also in the right frontal lobe, increased MD was found in the IFOF and UF, and the corpus callosum and anterior corona radiata. There was a significant positive correlation between MD and psychopathy scores.

Introduction 1.1. The frontal lobe theory of antisocial personality disorder (ASPD) and psychopathy The importance of the frontal lobes to social behaviour was first recognised in the 19th century following the case of Phineas Gage, in whom frontal lobe damage resulted in profound personality change associated with markedly inappropriate social behaviour [Harlow, 1993 (1869)]. A ‘frontal lobe’ syndrome was subsequently delineated based on clinical observations of the behaviour of patients with frontal lobe damage (Lishman, 1998) where symptoms included apathy, emotional lability, a lack of social awareness, unconcern for social rules, impulsivity, and reactive aggression. It is currently recognised that there is much overlap between frontal lobe syndrome and ‘functional’ or ‘non-organic’ personality disorders (PDs), particularly ASPD and psychopathy (Damasio, 2000). The definition of psychopathy has changed little since Hervey Cleckley published The Mask of Sanity in 1941 where he described the psychopath as a charming, callous, superficial individual, lacking conscience and genuine emotion ( Cleckley, 1941). The Psychopathy Checklist (PCL, Hare, 1980) and the later Psychopathy Checklist – Revised (PCL-R, Hare, 1991) were designed to operationalise Cleckley’s concept of psychopathy as a basis for diagnosing the disorder. The PCL-R consists of 20 items characterised broadly by two dimensions: Factor 1 items are primarily interpersonal or emotional traits such as remorselessness, deception, shallow affect and callousness, whereas Factor 2 items assess behavioural symptoms such as violence, criminality, and dysfunctional lifestyle. For a diagnosis of psychopathy, attributes from both of these factors need to be present. While PCL-R scores ≥30 have traditionally been used to classify an individual as having psychopathy ( Hare, 2003), more recent studies have argued for a score of ≥25 as sufficient for diagnosis ( Edens and Petrila, 2006, Edens et al., 2010 and Rutherford et al., 1999). While the related construct of ASPD in DSM-IV-TR (Diagnostic and Statistical Manual Fourth Edition – Text Revision, American Psychiatric Association, 2000) includes several traits present in psychopathy (e.g., lack of guilt/remorse, and impulsivity), diagnostic criteria can be met based entirely on antisocial behaviours (e.g., violation of social norms, irresponsibility, and criminality). Hence, the emotional deficits fundamental to psychopathy are not necessary for a diagnosis of ASPD, even if these are present in some cases. Estimates of the prevalence of the two disorders also differ, suggesting that these are non-equivalent diagnoses. While most adult psychopathic offenders meet criteria for ASPD, only approximately one third of those with ASPD are psychopathic ( Hart and Hare, 1997). Psychopathy has therefore been postulated to be a particularly severe subtype of ASPD ( Dolan and Doyle, 2007). Psychopathy and ASPD are however distinguished from behaviours secondary to frontal lobe lesions by high levels of both reactive (elicited by frustration) and instrumental (goal-directed) violence ( Blair, 2001 and Glenn and Raine, 2009). Nevertheless, overlaps between traits of both psychopathy and ASPD, and frontal lobe syndrome, have led to the suggestion that both PDs may result from frontal lobe abnormality ( Damasio, 2000). Neuroimaging studies of both people with ASPD, and of individuals with psychopathy, have provided evidence of abnormalities of frontal lobe structure and function relative to control populations, together with deficits in temporal, limbic, and other brain regions (see Table 1). Table 1. Summary of volumetry, functional and DT-MRI findings implicating fronto-limbic and other brain region abnormalities in antisocial populations. Author Method Population Comparison group Finding Region (Barkataki et al., 2006) MRI ASPD Violent and non-violent schizophrenia Reduced volume Whole brain, bilateral temporal lobe Increased volume Putamen (Laakso et al., 2000) MRI ASPD with alcoholism Healthy controls Reduced volume Right hippocampus, posterior hippocampi (Laakso et al., 2001) MRI ASPD with alcoholism Psychiatric patients Volume inversely related to PCL score Bilateral posterior hippocampus (Narayan et al., 2007) MRI ASPD Violent and non-violent schizophrenics and healthy controls Cortical thinning Medial PFC (de Oliveira-Souza et al., 2008) MRI ASPD with psychopathy Healthy controls Reduced volume Frontopolar cortex, orbitofrontal cortex, anterior temporal cortex, superior temporal sulcus, insula Volumes inversely related to psychopathy score (Raine et al., 2000) MRI Community ASPD Alcohol dependents and healthy controls Reduced volume Prefrontal grey matter (Raine et al., 2003) MRI Community ASPD with high psychopathy scores Healthy controls Increased length Corpus callosum Increased volume Reduced thickness (Tiihonen et al., 2008) MRI ASPD with alcohol dependence Healthy controls Increased WM volume Bilateral occipital lobe, bilateral parietal lobe, left cerebellum Increased grey matter volume Right cerebellum (Volkow et al., 1995) PET Violent offenders Healthy controls Reduced glucose metabolism PFC, medial temporal cortex (Raine et al., 1997) PET Murderers Healthy controls Reduced glucose metabolism PFC, corpus callosum, superior parietal gyrus, left angular gyrus (Boccardi et al., 2010) MRI Psychopathic violent offenders Healthy controls Bilateral depression Hippocampus – longitudinal axis Reduced CA1 segment Hippocampus – anterior Abnormal enlargement Hippocampus – lateral borders (Craig et al., 2009) DT-MRI Psychopaths Healthy controls Reduced FA Right UF (Glenn et al., 2010) MRI Community sample – high psychopathy scorers Community sample – low psychopathy scorers Increased volume with greater psychopathy score Striatum (Muller et al., 2008) MRI Criminal psychopaths Healthy controls Reduced volume Right superior temporal gyrus (Shamay-Tsoory et al., 2010) CT/MRI ASPD males with psychopathic traits Orbitofrontal cortex verus non-frontal lesioned males and healthy controls Impaired affective empathy performance in both orbitofrontal cortex lesioned and psychopathy group Orbitofrontal cortex (Yang et al., 2005) MRI Unsuccessful community psychopaths Successful community psychopaths and healthy controls Reduced volume Prefrontal grey matter (Yang et al., 2009a) MRI Community psychopaths Healthy controls Reduced volume Bilateral amygdala Surface deformations Amygdala nuclei: basolateral, central, cortical, lateral (Yang et al., 2009b) MRI Community psychopaths Healthy controls Cortical thinning Right frontal cortex, right temporal cortex Thinning associated with greater PCL factor 2 score Right frontal cortex, right temporal cortex (Deeley et al., 2006) fMRI Psychopaths Healthy controls Reduced BOLD activation to fearful and happy faces in emotion processing task Fusiform gyrus, extrastriate cortex Decreased, rather than increased, BOLD activation to fearful faces Fusiform gyrus (Soderstrom et al., 2002) SPECT Violent offenders with varying psychopathy scores Negative correlation between interpersonal psychopathy factor and perfusion Frontal and temporal regions, head of caudate, left hippocampus (De Brito et al., 2009) MRI Community sample of boys – high versus low callous-unemotional trait scorers Increased grey matter concentration Medial orbitofrontal cortex, anterior cingulate cortex Increased grey matter concentration and volume Bilateral temporal lobe (Kruesi et al., 2004) MRI Conduct disordered adolescents Healthy controls Reduced grey matter volume Right temporal lobe (Huebner et al., 2008) MRI Conduct disordered adolescent males comorbid with ADHD Healthy controls Reduced grey matter volume Bilateral temporal lobe Left hippocampus, left amygdala (Sterzer et al., 2007) MRI Conduct disordered adolescent males Healthy controls Reduced grey matter volume Bilateral anterior insular cortex Left amygdala (Berns et al., 2009) DT-MRI Healthy adolescents – high versus low on risk taking measure High scores positively correlated with FA and negatively with transverse diffusivity Frontal WM tracts (Marsh et al., 2008) fMRI Antisocial children with callous-unemotional traits Children with ADHD; healthy controls Reduced BOLD activation to fearful faces on emotional processing task Amygdala Reduced functional connectivity Between amygdala and vmPFC (Finger et al., 2008) fMRI Antisocial children with callous-unemotional traits Children with ADHD; healthy controls Abnormal BOLD signal to punished errors on reversal learning task vmPFC (Jones et al., 2009) fMRI Antisocial boys with callous-unemotional traits Healthy controls Reduced BOLD activation to fearful faces on emotional processing task Right amygdala (Stadler et al., 2007) fMRI Conduct disordered adolescent males Healthy controls Reduced BOLD activation to negative affective pictures in emotional processing task Right anterior cingulate cortex Table options For example, neuroimaging studies of adult psychopaths examining the frontal cortex have reported reduced grey matter volume in conjunction with reduction in the superior temporal gyrus (Muller et al., 2008), and in the prefrontal cortex (PFC) of ‘unsuccessful’ (caught) psychopaths, versus healthy controls (Yang et al., 2005). Furthermore, higher total and subfactor PCL-R scores (arrogant/deceptive, affective, and impulsive/unstable) were associated with reduced prefrontal grey matter volume (ibid). Similarly, prefrontal and temporal cortical grey matter thinning was found in psychopathic individuals, with right hemisphere reductions related to elevated PCL-R Factor 1 ‘Affective’ facet scores ( Yang et al., 2009b). Other studies have identified associations between psychopathic traits and specific subregions of the PFC. In particular, the association found in brain injured patients between ventromedial PFC (vmPFC) damage and reactive aggression (Blair and Cipolotti, 2000 and Grafman et al., 1996) is mirrored by vmPFC structural and functional impairments in psychopaths (Tiihonen et al., 2008). Functional neuroimaging studies of people with psychopathy have also provided evidence of abnormal frontal lobe perfusion and abnormalities of task-related activation in prefrontal and other brain regions on reversal learning paradigms (Table 1). Individuals with ASPD have shown similar structural and functional prefrontal abnormalities to psychopaths – for instance reduced prefrontal grey matter volume has been found in both antisocial adults (Raine et al., 2000) and conduct disordered children (Huebner et al., 2008), compared with healthy controls. Also, cortical thinning of the medial frontal lobe has been found in ASPD (Narayan et al., 2007). Nevertheless, structural and functional abnormalities in people with psychopathy and ASPD are not restricted to the frontal lobe. For example, abnormal amygdala structure and function have each been found to correlate with the emotion processing deficits observed in ASPD and psychopathic individuals (Gordon et al., 2004, Kiehl et al., 2001 and Yang et al., 2009a), as well as in regions (such as fusiform–extrastriate cortices) known to be modulated by the amygdala (Deeley et al., 2006). Reduced volume of temporal regions is also seen in ASPD (Barkataki et al., 2006). Taken together, these studies of antisocial adults and conduct disordered children (with and without psychopathic traits) illustrate reduced volume of the temporal lobe and its constituent structures, and deficits in amygdala structure and function. Apart from these fronto-temporal structures, other regions potentially relevant to psychopathy have been less extensively investigated. For instance, there have been only limited magnetic resonance imaging (MRI) studies assessing the corpus callosum, a major white matter (WM) bundle supporting interhemispheric functional integration, where abnormalities have included increases in volume and length but reduction in thickness (Raine et al., 2003). Also, while reductions in functional connectivity between prefrontal and limbic regions may contribute to antisocial traits (Table 1), their underlying microstructural basis remains unknown. Overall, the wider network of abnormalities in psychopathy has been relatively understudied although such investigation has been made possible using diffusion tensor magnetic resonance imaging (DT-MRI) (Basser et al., 1994b, Thiebaut de Schotten et al., 2012 and Catani et al., 2012) where the ‘connectivity’ of neural systems is assessed using proxy measures of microstructural integrity. 1.2. Disconnectivity between frontal and other regions in psychopathy DT-MRI is particularly used in the assessment of tissue (such as WM networks) where water preferentially diffuses along a particular axis aligned with the tissue’s internal structure and is predicated on the principle that microarchitectural structures, for instance, cell membranes, myelin sheaths, as well as intracellular micro-organelles, act as barriers to the diffusion and free movement of water, and thus limit the spatial motion of these molecules (Malhi and Lagopoulos, 2008). The assessment of the directional dependence of water molecule diffusion in WM is usually quantified through calculation of fractional anisotropy (FA). FA is a measure of the degree of anisotropy or directionality where values range from 0 (perfectly isotropic diffusion) to 1 (perfectly anisotropic diffusion) (Pierpaoli and Basser, 1996) – so providing a measure of tissue integrity (Horsfield and Jones, 2002 and Mori and Zhang, 2006). Mean diffusivity (MD) is another DT-MRI derived parameter used for reporting tissue differences which is calculated by division of the sum of the eigenvalues of the diffusion tensor (which correspond to the magnitude of diffusion in three orthogonal directions) by three. There are however only limited studies examining neural disconnectivity using DT-MRI in people with psychopathy and/or ASPD. A recent study by our group using DT-MRI tractography focusing on FA in the uncinate, inferior longitudinal and inferior fronto-occipital fasciculi reported a significant reduction of this measure in only the uncinate fasciculus (UF) of nine psychopaths compared with age- and IQ-matched controls (Craig et al., 2009). Additionally, a significant negative correlation was found between measures of antisocial behaviour (PCL-R Factor 2 scores) and tract volume within this WM pathway, suggesting abnormal connectivity in the amygdala–orbitofrontal cortex limbic network. However, as this study was confined to a limited number of WM tracts, it was not possible to assess WM networks on a whole brain level. Consequently, the presence of deficits affecting WM connectivity with other brain regions is yet to be established in either ASPD or psychopathy. Further, our previous study did not examine other indices of WM microstructure, such as MD. In summary, there is increasing evidence that people with ASPD and psychopathy may have significant differences in the structure and function of frontal, limbic and other brain regions. However, few studies have examined the microstructural integrity or connectivity of their WM networks. Therefore, we undertook the first DT-MRI investigation on a whole brain level of ASPD and psychopathy. We examined WM networks and tested the main hypothesis that people with ASPD and psychopathy have significant differences, based on DT-MRI derived parameters of FA and MD, in microstructural integrity and connectivity as compared to healthy matched controls. Also, we tested an additional hypothesis that within ASPD, severity of psychopathy (as measured by PCL-R) is related to differences in these WM measures.

3. Results 3.1. Group contrasts of FA using DT-MRI group mapping 3.1.1. ASPD versus controls WM FA (Fig. 1, Table 2) People with ASPD, relative to controls, had a significant reduction in WM FA; 1) bilaterally in the frontal lobe in the anterior portion of the corpus callosum (genu); 2) in the right hemisphere, only in anterior regions of the brain and in WM tracts that included the genu of corpus callosum, anterior corona radiata and anterior limb and genu of the internal capsule, and frontal course of the uncinate and inferior fronto-occipital fasciculus (IFOF); 3) in the left hemisphere in both anterior and posterior regions of the brain including respectively the genu of corpus callosum and temporo-occipital course of the inferior longitudinal and IFOF, and the retrolenticular part of the internal capsule and posterior thalamic radiation. Reduced FA in ASPD relative to healthy controls [ascending 2 mm transverse ... Fig. 1. Reduced FA in ASPD relative to healthy controls [ascending 2 mm transverse sections]. Figure options 3.1.2. ASPD versus controls WM MD (Fig. 2, Table 2) People with ASPD, relative to controls, had a significant increase in MD only in the right frontal lobe. This was localised to a cluster containing the frontal course of the IFOF and UF, and the genu of corpus callosum and anterior corona radiata. No regions of increased MD were found in the control group. Increased MD in ASPD relative to healthy controls [ascending 2 mm transverse ... Fig. 2. Increased MD in ASPD relative to healthy controls [ascending 2 mm transverse sections]. Figure options 3.1.3. Within ASPD, an analysis of Factor 1, Factor 2 and total PCL-R scores and differences in WM FA and MD In the ASPD group, there were significant correlations between PCL-R scores and both WM FA and MD. Mean FA of the cluster in the frontal lobe (Cluster 2) was negatively correlated with Factor 2 (r = −.771, p = .003, n = 12) and total PCL-R (r = −.685, p = .005, n = 15) scores. Additionally in the frontal lobe, there was a significant positive correlation between increased MD (Cluster 3) with Factor 2 scores (r = .669, p = .017, n = 12). There were no significant correlations between the cluster in the temporo-occipital cortex (Cluster 1) and total or subfactor PCL-R scores