Trauma-based Sexually Dimorphic Changes in the Connectome and its Association with Central Sensitisation Syndromes - A Systematic Review

1. Introduction

Chronic pain syndromes represent a complex interaction of biological and psychosocial factors that are only partly understood, posing a challenge to patients and physicians[27,67]. Chronic pain, defined as the perception of pain greater than three months impacts quality of life, creating a significant global health issue with a prevalence in lower and middle-income countries thought to be 20% –60% of the population[3,7,10,13,43,44,67,73]. Sensitisation of the central nervous system may contribute to the understanding of the pathogenesis and maintenance of chronic widespread pain[10,27,49]. The term central sensitisation is a condition that is associated with neuroplastic changes in the central nervous system (CNS), whereby the perception of pain is maintained even when the initial nociceptive insult has resolved[14]. Central sensitisation is associated with CNS changes, including chronic upregulation of nociceptive receptors. One established theory is that chronically increased nociception may lead to increased levels of substance P being released, lowering thresholds for perceived pain[10]. Substance P is a neurotransmitter released from primary afferent unmyelinated C-fibres related to the sensitisation of nociceptive pathways[33]. Recent research has found that central changes affect neurotransmitter receptors, leading to decreased endocannabinoid sensitivity, for example. Such changes contribute to hyperexcitation of pain processing pathways, leading to allodynia, hyperalgesia, and cognitive, affective, and behavioural changes[33,71]. Fibromyalgia is a prototypical chronic centralised pain condition that affects the entire musculoskeletal system and is characterised by hyperalgesia and allodynia without any apparent peripheral tissue damage, combined with fatigue, sleep, cognitive, and mood problems[25,27,32,50,58,68]. Complex regional pain syndrome (CRPS) is a chronic pain disorder characterised by spontaneous or regionally evoked pain and trophic changes typically affecting the distal extremities, particularly the upper limbs. Whilst CRPS usually develops after a peripheral event, it is likely maintained by changes in the central nervous system[14]. Alterations in grey matter volume and functional neural connectivity have been demonstrated in various conditions of central sensitisation and medically unexplained pain[10,28]. Chronic somatic and visceral syndromes are often comorbid with each other and with internalising symptoms of anxiety and depression[10,27,49]. The early-life risk markers for developing central sensitisation syndromes include genetic factors, prenatal stress and birth trauma, early life adversity (ELA), female sex, injury, sleep disorders and lifestyle disorders[4,25,61,71,73,75,77,79].

ELA in childhood can come in the form of negative experiences, e.g. abuse or trauma, or the absence of experience, e.g. neglect or deprivation[5,6]. Adverse childhood experiences (ACEs) are repeated aversive experiences that represent deviations from the expected environment and require adaptation[2]. These associations include perinatal exposure to substance abuse, maternal deprivation, growing up with a depressed parent, psychological trauma, physical or verbal abuse from parents, and physical or sexual abuse by an adult[69]. Childhood maltreatment (CM) includes highly stressful and potentially traumatic events or situations that occur during childhood and/or adolescence, which include sexual abuse, psychological abuse and neglect[37]. CM moderates the association between an adult traumatic event and adult psychopathology, such that those who experienced CM have more severe symptoms after later trauma than those who did not experience maltreatment[46]. Chronic pain has also been shown to be more prevalent in individuals exposed to ELA than in the general population[4,33]. The quality of parental care, nutrition, cognitive stimulation, and socioeconomic status during early child development have been shown to affect brain morphology and functionality throughout the life course[2,26]. Understanding embryological neurodevelopment may improve understanding of the developmental origins of disease[52].

The foetal maternal environment and that of early childhood are essential modulators of brain development, with consequences throughout childhood and the lifespan[1,4,60]. Brain regions undergoing extensive neurogenesis are particularly vulnerable to insults because developmental patterns are being established[23]. Glucocorticoid exposure might induce these long term changes by acting as epigenetic modulators interfering with transcription factors. DNA methylation embeds the impact of early life experience in the genome so that environmental perturbations can modify the phenotype of the offspring[1]. Advances in neuroimaging have expanded the concept that the nervous system is a structurally interconnected and integrated network of neuronal pools, allowing the influence of one neural system over another[15]. Adults with chronic pain who have experienced ACE exposure can show different brain alterations than adults with chronic pain who have not experienced ACE[4]. These areas include the prefrontal cortex (PFC), superior temporal gyrus, insula, amygdala, hippocampus, putamen, and the anterior cerebellum[2,26]. Glucocorticoids released in response to stress bind to glucocorticoid and mineralocorticoid receptors, causing changes in DNA methylation, which correlate with an enhanced responsivity to a second stressful challenge[1]. Thus, prenatal and early life stress can render offspring more susceptible to additional environmental exposures later in life, resulting in the unmasking of psychopathology[4,23,46]. However, the maternal environment and genetic profile are not the only contribution to embryonic development. The paternal genome is demethylated faster in the first days of embryo development than the maternal genome, so preconceptual paternal stress may also significantly impact the embryo[5].

There has been an association between prenatal and childhood adversity and the development of somatic syndromes[4,10,23,33,54], predominantly in females [8,11,28,30,48,55,58,65,73,76,80]. However, the reasons for female predominance are mainly unknown[69]. Foetal sex may mediate stress responsiveness[28,33,52,56]. Prenatal trauma and ELA have been shown to put male offspring at risk of developing socialisation and externalising disorders[22]. In contrast, female offspring exposed to ELA appear to be at risk of internalising disorders[12,22,60]. This review describes the effects of prenatal and early childhood adversity on brain development. One focus is on studies characterising human brain development using MRI to identify sensitive periods during which the environment influences the adult phenotype. A further area of interest is to identify biological mechanisms contributing to vulnerability and resilience to stress, especially concerning sex assigned at birth.

Therefore, the main objective of this review is to investigate the association between sexually dimorphic changes in the connectome induced by prenatal and ELA and the development of central sensitisation syndromes.

2. Methods

Following the PRISMA framework (Figure 1), a structured search was conducted on the PubMed database using the keywords Central Sensitisation, Fibromyalgia, Complex Regional Pain Syndrome, and Neuropathic Pain combined with the keywords Prenatal Trauma, Early Life Adversity and Childhood Maltreatment. The terms Connectome and Sexual Dimorphism were then applied to all previous keywords. CRPS and fibromyalgia were included in the search terms as they are among the most common somatic medically unexplained pain syndromes encountered in practice. Pubmed was chosen because it is accessible, user friendly and uses synonymous search terms. Articles within the last ten years, from 2013 – 2023, have been included. Papers were excluded if they were animal studies, investigated tissue damage, disease processes or addiction, were conference proceedings or non-English. One paper was excluded because it used EEG procedures. Papers elicited from the search on sexual dimorphism were marked irrelevant if there was no reference to sex assigned at birth comparison in the text. Previous reviews were included to summarise evidence from different outcomes, conditions, or populations.

3. Results and Discussion

349 studies on conditions of central sensitisation, the connectome and sexual dimorphism were identified. After title, abstract, and full-text screening, 81 studies were identified as meeting inclusion.

3.1. Association of Adverse Childhood Experiences with Central Sensitisation

Published systematic reviews have demonstrated positive associations between ELA and the subsequent development of somatic and visceral syndromes, with the risk of developing somatic syndromes being higher[10]. There is a high prevalence of posttraumatic stress disorder (PTSD) (37.3%) among fibromyalgia patients, significantly higher than that observed among other chronic pain patients, for example, rheumatoid arthritis[11,26].

Chronic pain is characterised by the disruption of whole brain functional connectivity globally and the disruption of local connectivity[6,31,40,69]. The pain experience is highly subjective and top-down modulated[63]. The default mode network (DMN) is the primary network related to chronic pain and comprises the posterior cingulate cortex (PCC), medial prefrontal cortex (mPFC), and lateral parietal lobe[35,36,80]. The reward-motivation network (including the PFC, nucleus accumbens, hippocampus, and ventral tegmentum) and the descending pain modulatory system (anterior cingulate cortex (ACC), amygdala, and hypothalamus) are also implicated in vulnerability to painful conditions along with the insula and thalamus, which are involved with pain perception[2,35,81]. In addition to encoding pain intensity and duration, the structure of the neuraxis plays an integral part in developing chronic pain. The most consistent earlier results regarding perturbations of resting state point to changes in functional connectivity between the insula and the DMN[4,8] and between the insula and the mPFC[29,40]. Maltreated children show significant reductions in both global connectivity and local connectivity with reported reduced cortical thickness within regions involved in emotional regulation, including the orbitofrontal cortex (OFC), ACC and mPFC[4,10,31,56,80]. Threat and deprivation may show opposing effects where deprivation is associated with reductions in association and prefrontal cortical thickness, and threat affects the connectivity of areas involved in emotional learning, including the hippocampus, amygdala, and PFC[29]. Timing, chronicity, and maltreatment type are critical factors that likely shape neural development and behavioural outcomes[31,77], such that children who endured more extended periods of hardship and higher levels of verbal and domestic abuse present with more significant alteration[56,80]. Whether alterations in brain function precede or result from chronic pain syndromes, they might lead to a cycle of decreased resilience[69]. Connectivity is also altered with current pain states[9], with treatment induced changes demonstrated with pain reduction[20].

3.2. Association of Adverse Childhood Experiences with Connectome

The quality of parental care, nutrition, cognitive stimulation, and socioeconomic status during early child development have been shown to affect brain morphology and functionality throughout the life course[4,6,26,29,31,33,56,66,70,72,74]. Graph-based network analysis is utilised to reveal information about the topography of human brain networks by characterising different brain regions as nodes and white matter tracts reconstructed through probabilistic fibre tracking as connections between the nodes as edges[35,56}. This framework allows the degree of functional segregation and integration of the network to be investigated and yields invaluable insights into normative brain development[56]. Deviations from small world brain architecture, considered the most efficient network organisation due to its dense local clusters of nodes connected by short paths facilitating quick information processing, indicate several neurodevelopmental, psychiatric and neurological disorders[56]. Studies employ a unitary measure that assesses whole-brain degree rank order disruption (kD), defined as the gradient fitted to the mean difference in nodal degree between any given subject or group of subjects about the mean nodal degree in a control population using normalised mutual information (NMI). The overall similarity of subjects to the off-site control community is determined using post-hoc analysis[6]. Changes in kD have been shown to be proportional to reported pain intensity but only once pain became chronic, approximately one year of persistent pain[40].

Adults with chronic pain who have experienced ACE exposure can show different brain alterations than adults with chronic pain who have not experienced ACE[2]. These areas include the superior PFC (sPFC), superior and inferior temporal cortex, insula, left lingual gyrus, hippocampus, putamen, and anterior cerebellum[26,39].

3.3. Parts of the Connectome Implicated in Prenatal and Childhood Trauma

3.3.1. The Cerebrum

Alterations in PFC grey matter have been reported in multiple chronic pain conditions[13,81], although the direction of this association is unknown[39]. The PFC influences the descending regulation of neuronal activity of the dorsal horn of the spinal cord, thereby influencing nociception[20]. The mPFC provides the primary source of cortical input to the periaqueductal grey (PAG) and is thought to play an integral role in descending pain modulation[34,52]. Heightened mPFC activity when processing reward or social exclusion in individuals with higher levels of ELA is associated with learned helplessness and other depression-like behavioural deficits after exposure to stress[24], with increased grey matter volumes of the mPFC observed in maltreated individuals[37]. Reward sensitivity and anticipation are negatively impacted in ELA[(4, 31] and are mediated by the OFC, influencing the perception of pain[2,6,20,56].

The dorsolateral PFC is known to perform various cognitive functions, including working memory, motivation, and attentional control[74] and the processing of pain[9,19]. Cerebral blood flow and cortical connectivity are increased within the dlPFC of patients with chronic pain[33,81], with decreased grey matter volume of the dlPFC associated with neighbourhood poverty[70]. Cortical thinning in the dlPFC provides further evidence for the involvement of the dlPFC in top-down pain modulatory processes via its connections to other pain modulatory brain areas including the PAG[20]. Other areas that might not affect the risk of developing chronic pain but are relevant to pain perception include changes in the primary and secondary somatosensory cortex, a significant area regarding the sensation and perception of pain and episodic memory retrieval. Individuals with major depressive disorder and those who have experienced early life adversity have been found to have abnormal activity in this region[2] and a significant increase in regional cerebral blood flow has been observed in the primary somatosensory cortex of migraineurs compared with healthy controls[71].The parietal lobe is a significant area in processing pain because of its role in processing sensation[2]. The interaction between stress and ELA is related to alterations during emotional processing, mainly in the middle temporal[6] and supramarginal gyri[66]. Decreased grey matter volume has been observed in the right middle temporal and left lingual gyri associated with cognitive and affective disorders[13,37].

The insula is activated by ELA and is an essential area concerning pain perception. There is increased insula activation in children exposed to violence[2,6] and increased connectivity with the insula and DMN in chronic pain[4,8]. Interoceptive awareness is mediated by the right anterior insula connectivity to structures including the amygdala, hippocampus, OFC and precuneus[15]. Pain responses in the insula seem enhanced in men with chronic pain, with insula volume negatively correlated with perceived personal control of the condition[39].

3.3.2. The Limbic System

Exogenous glucocorticoids reduce ACC activation and simultaneously increase negative affect[2,56]. Individuals with a history of severe maltreatment show lower volume in the left ACC[6,29,51,66,74]. These changes may not increase the risk of developing chronic pain but could affect how pain is perceived[2,19], especially in women[39]. The subgenual anterior cingulate cortex (sgACC) is connected to the PAG, rostroventral medulla, and mPFC, critical components in the descending pain modulation pathwayassociated with reduced temporal summation of pain and improved habituation[52]. fMRI study of static resting state functional connectivity has shown that females have greater connectivity of the sgACC with the PAG and raphe nuclei, while men have greater connectivity with the salience network[52]. Altered white matter properties in the PCC have been observed in abdominal pain conditions, including irritable bowel syndrome (IBS) and primary dysmenorrhea[8,36].The hippocampus contributes to stress regulation via the hypothalamic-pituitary-adrenal (HPA) axis by providing inhibitory feedback[70]. Oestrogen may enhance choline uptake, acetylcholine synthesis and blunt cortisol responsivity, whereas progesterone may increase HPA axis reactivity to stress, altering autonomic and neuroendocrine homeostasis[64]. Glucocorticoids are toxic to the hippocampus, which might explain the hippocampal atrophy often observed in individuals with ELA[23,37,59,74] and chronic pain[13]. Smaller hippocampus volume may increase the vulnerability to anxiety disorders and posttraumatic stress[20]. The hippocampus is recruited in anticipation of pain and is associated with internalising symptoms following maltreatment[28].

The amygdala affects affective processing and perceived pain [2,4,56]. ELA experiences have been associated with altered amygdala connectivity, including lower connectivity to the hippocampus, insula, OFC and postcentral gyrus and greater connectivity to PFC[28,31]. Volume decreases in the amygdala have also been shown in studies of adults reporting ELA[74], with increased amygdala connectivity with the hippocampus and PFC during emotion processing[80]. Increased amygdala activity has been observed in females suffering from IBS[8]and chronic pain[39]. Visceral neural circuits converge on the paraventricular nucleus of the hypothalamus (PVN), bed nucleus of the stria terminalis (BST), and the amygdala to control autonomic and neuroendocrine stress responses through viscerosensory afferents to the nucleus tractus solitarius (NST). The NST also receives viscerosensory information from the vagus and relays it to the ventral BST and PVN. Research indicates a link between childhood adversity and visceral, stress-related circuits, contributing significantly to differences in stress reactivity, affective processes and response to threat[4].The thalamus is a vital region in the integration of brain function and is believed to play a crucial role in the experience and expression of emotion and stress responses, influencing the perception of pain [2,20,74]. Abnormal corticothalamic connectivity suggests altered pain processing in migraine[78]. Functional abnormalities in the thalamus are consistently reported in CRPS[14,81].

The caudate and the putamen are thought to play a role in pain and analgesia[2,20]. The caudate is involved in reward-related and emotional processing[37]. Volume decreases in the caudate [74] and increased white matter densities in the right putamen and globus pallidus have been demonstrated in ELA[2]

3.4. Mechanisms of Connectome Alteration

ELA, including in utero, can modify the epigenome, potentially leading to changes in DNA methylation and HPA axis activity[1,41]. Dysregulation of the HPA axis may affect nociceptive processing, predisposing individuals to sensitisation and increased inflammation markers[10]. Neuroinflammation is implicated in neurodegenerative diseases, mood disorders and chronic pain [16,34,38,42]. Chronic cortisol secretion can affect dendritic and axonal sprouting in glucocorticoid-receptor-rich areas, diminishing white matter integrity[21]. The HPA axis has also been implicated in the observation that ELA accelerates telomere length shortening[18], associated with cellular damage[62]. A linear relationship has been demonstrated between parental care and telomere length, suggesting that higher parental care may protect against telomere shortening in the presence of subsequent stressors[18].

Deprivation may shape neural development via activity-dependent plasticity, leading to increased apoptosis of synaptic connections, particularly within association cortices[4]. ELA may also activate stress reactivity and nociception through hyperalgesic priming. Thereby, pain is a stressor that may generate a feedback loop impacting stress regulation[69].

Sexual dimorphism may be due to different gene expressions because of variability in the hormonal environment of the two sexes[43]. The transient receptor potential (TRP) channel family controls cellular differentiation by regulating gene expression by activating calcium-dependent transcription factors. DNA methylation of TRP channels is implicated in the pathology of pain syndromes. Steroids and TRP channels also closely interplay with oestradiol positively and androgens negatively, enhancing TRP expression. Therefore, TRP channels may function as a modulator of sexual dimorphism in pain perception[1,7].

3.5. Sexual dimorphism

Prenatal and ELA may influence the epigenome in a sex-dependent manner, which might underpin the higher rates of chronic pain in females[28,29]. ELA has been associated with sexually dimorphic altered connectivity within the brain's fear-regulatory circuit, including the hippocampus, amygdala, and sgACC, as a component of the ventromedial PFC (vmPFC). The vmPFC inhibits hyperactivity of the amygdala with a consequent expression of fear responses, whereas the hippocampus inhibits fear responses via connections to both the amygdala and vmPFC[28,33].

3.6. Sex Assigned at Birth and the Limbic System

There are more significant growth rates for both the amygdala and hippocampus in females during the first several years of life, with a more extended period of growth in the amygdala for males[28]. As periods of rapid brain maturation are susceptible to ELA, there may be a more significant neural impact of maltreatment in females by affecting the connectivity of the amygdala and hippocampus with the sgACC[23]. This alteration in amygdala connectivity to the cingulate cortex may explain higher internalising symptoms in females[5]. Sex differences in sgACC functional connectivity have also been observed in chronic pain patients[52].

However, exposure to trauma has a relatively more significant effect on hippocampal volume in males than in females, as oestrogen exhibits neuroprotective effects in the hippocampus[28]. Males show greater structural connectivity between the vmPFC and hippocampus, which may offer protection against CM. However, adult trauma has revealed increased hippocampus–vmPFC connectivity, with failure to increase this connectivity associated with PTSD. This observation emphasises the significance timing of trauma has on the connectome, with age-dimorphic effects also revealed in the literature[52,60].

The interaction of the HPA, sympathetic nervous system and immune system can also contribute to the effects of ELA. Women show greater activation of the HPA[10,23,46,64], higher sympathetic tone and greater levels of inflammation[17,23,47,52,65]. Table 1 summarises examples of sexually dimorphic experiences of central sensitisation syndromes.

3.7. Sex and Adverse Childhood Experiences

Sexual dimorphic effects of trauma may also be unrelated to biological differences in neural development. Females and males may be exposed to different types or higher rates of stressors, which may correspond to the sex differences noted in patients with chronic pain and comorbid depression[5,22,33,45,57]. Dysmenorrhea may also predispose women to a chronic pain state. Altered white matter integrity of the cingulum associated with dysmenorrheic pain may lead to spontaneous communication within the pain connectome, amplified by nociceptive input, so pain becomes a learned behaviour[11]. Examples of sexually dimorphic experience of adverse childhood experiences revealed in the literature are summarised in Table 2.4. Conclusions

The main objective of this review was to investigate the association between sexual dimorphic epigenetic changes in the connectome induced by prenatal and ELA. A secondary objective was to examine any potential association of ELA with the development of central sensitisation syndromes. The results showed that brain regions undergoing extensive neurogenesis are particularly vulnerable to insults. However, the type of adversity, e.g., deprivation and threat, and the age at which it was experienced would impact the connectome differently and potentially affect the sexes differently. Alterations in the connectome associated with trauma can include grey matter volume decrease, higher or lower integrity of white matter tracts and higher or lower gyrification indices. The brain is programmed for survivability, so increasing tracts associated with, for example, hypervigilance can be seen as advantageous whilst placing the individual at risk for anxiety disorders. Nevertheless, this is still a new field of study. Extensive literature reviews are invaluable for comparison, collaboration, and assessment of information. Characterising the effects of sex on brain circuitry is essential in developing effective personalised pain treatments because therapies designed with research based primarily on one sex can potentially be less effective in the other sex.

5. Limitations

Despite the structured search strategy, this review has limitations. An inherent recall and response bias is associated with self-reported documentation of childhood trauma, which may be unstable over time[46]. Not all gender identifications and sexual orientations are acknowledged or assessed. Despite the strengths of graph theory, another limitation of concern to developmental neuroscientists is the reliability of connectivity measures obtained from MRI scans interpreting graph metrics, which have shown limited test-retest reliability[29]. The origin of the information in the attached tables has not been well defined. The search was limited to ten years because of resource constraints.

6. Further Research

The influence of prenatal and childhood adversity on the developing connectome and its association with the adult phenotype is an emerging field, and further research will be required to prove or disprove these theories.