Table 1 Neuroimaging studies of childhood maltreatment.
Study, location | Participants | Mean age (SD; range) % Female | Exposure | Study design; covariates | Primary findings |
|---|---|---|---|---|---|
Childhood maltreatment | |||||
Purcell et al. [177] United Kingdom | 300 participants from Birmingham Metropolitan area cohort | 20 (1.5; 17–23) 50.3% | Childhood Trauma Questionnaire | Cross sectional Covariates: violence exposure, race, sex, and scanner type. | Stress-elicited ventromedial PFC, dorsolateral PFC, and hippocampal activity was lower in individuals who reported sexual abuse. |
Silveira et al. [178] Canada | 392 participants from National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) | 17.36 (2.53; 12–22) 55% | Childhood Trauma Questionnaire | Cross sectional Covariates: age, sex, ethnicity, years of parental education, familial history of AUD, high-risk drinking, or frequent use of tobacco, marijuana, or other drugs | At baseline, distributed functional connectivity from hub regions in the bilateral dorsal anterior cingulate cortex, right anterior insula, right intraparietal sulcus, and bilateral pre- and postcentral gyri mediated the relationship between childhood trauma and executive dysfunction. |
Puetz et al. [22] USA | 414 participants from Duke Neurogenetics study (DNS) | 19 (1; 18–22) 60% | Childhood Trauma Questionnaire | Cross sectional Covariates: age, sex, handedness, SES, psychopathology | Childhood abuse was linked to increased activity in the ventral amygdala, while neglect was linked to increased reactivity fronto-parietal network and dorsal amygdala. |
Clausen et al. [132] USA | 577 participants | 32.25 (10.58; 18–59) 64% | Childhood Trauma Questionnaire | Cross sectional Covariates: age, sex, education, self-reported medical comorbidities, number of medications | Childhood trauma is associated with smaller regional GM volume within left superior frontal cortex and right medial cingulate cortex and higher regional GM volume within left medial cingulate cortex, right inferior insular cortex and left anterior insula. |
Luo et al. [23] Europe | 639 participants from IMAGEN 4121 participants from UK Biobank | IMAGEN: 19.06 (0.70) 50.8% UKB: 56.89 (5.02) 58.1% | Childhood Trauma Questionnaire | Cross sectional Covariates: sex, site, BMI PRS, family SES, stressful life events in the past year, birth weight, depressive symptoms, and illegal drug use | Childhood abuse was linked to obesity via prefrontal cortex. |
Gheorghe et al. [40] United Kingdom | 6751 participants from UK Biobank | 62.1 (7.2; 45–80) 58.6% | Childhood Trauma Questionnaire | Cross sectional Covariates: age, sex, handedness, ethnicity, education, depression and anxiety and head size scaling | Childhood emotional abuse was associated with smaller cerebellar and ventral striatum volumes. |
Ancelin et al. [179] France | 398 participants from ESPRIT study | 65–80 52% | Categorical presence of childhood abuse | Cross sectional Covariates: age, sex, brain volume, head injury, lifetime depression and anxiety disorder, psychiatric medication, and cardiovascular ischaemic pathologies. | Childhood adversity was associated with rostral middle frontal, lateral orbitofrontal, superior parietal, precuneus, and thalamus. |
Koyama et al. [25] Japan | 491 participants from Neuron to Environmental Impact across Generations (NEIGE) Study | 65–84 52.7% | Categorical presence of childhood abuse | Cross sectional Covariates: age, sex, ICV, prescribed medication, smoking or drinking history, BMI, depressive and cognitive score, childhood and current SES, marital status. | Individuals with 2 or more adversities had larger anterior cingulate cortex and smaller amygdala and hippocampal volumes. |
Korgaonkar et al. [180] Australia | 647 participants | 33.3 (12; 18.2–69.2) 51% | Categorical presence of childhood abuse | Cross sectional Covariates: age, sex, education, diagnosis, scan motion | Individuals who experienced abuse during childhood (but not during adolescence) had increased functional connectivity between brain networks involved in somatomotor processing and dorsal-ventral attention. |
Cohen et al. [181] Australia | 250 participants from Brain Research International Database (BRID) | 39.9 (17.2; 18–70) NR | Early Life Stress Questionnaire | Cross sectional | Adverse childhood experiences were associated with anterior cingulate cortex and caudate nucleus. |
Busso et al. [24] USA | 51 participants from a longitudinal cohort | Baseline: 15.14 (1.46) measured adversity MRI at FU1: 16.96 (1.51) Clinical assessment at FU2: 18.92 (1.50) | Childhood Trauma Questionnaire | Longitudinal Covariates: age, sex, parental education | Childhood abuse was associated with reduced cortical thickness in vmPFC, right inferior frontal gyrus, left and right parahippocampal gyri, right inferior temporal gyrus, and right middle temporal gyrus. |
Hanson et al. [39] USA | 106 participants from a longitudinal cohort | Baseline: 13.67 (11.88–15.45) Follow up: 13.77–18.25 48.1% | Childhood Trauma Questionnaire | Longitudinal Covariates: age, time between scans, sex, depressive and anxiety symptoms. | Emotional neglect was associated with blunted development of reward-related ventral stratum activity. |
Hein et al. [35] USA | 167 participants from Fragile Families and Child Wellbeing Study (FFCWS) | Adversity at ages 3, 5, 9 MRI at 15 53.9% | Parent-Child Conflict Tactics Scale Mother’s report on partner/community violence and support | Longitudinal Covariates: sex, internalising psychopathology, and current life stress | Childhood violence exposure was associated with increased amygdala activation to angry faces in adolescence, whereas childhood deprivation was associated with decreased ventral striatum activation to happy faces in adolescence. |
Goetschius et al. [182] USA | 178 participants from Fragile Families and Child Wellbeing Study (FFCWS) | Adversity at ages 3, 5, 9 MRI at 15 56% | Parent-Child Conflict Tactics Scale Mother’s report on partner/community violence and support | Longitudinal Covariates: sex, race, pubertal development, adolescent life stress, maternal educational level and marital status at the child’s birth | Childhood violence exposure was associated with reduced rsFC density, with fewer salience network connections and salience network-default mode connections. |
Ganella et al. [42] Australia | 91 participants from a longitudinal cohort | Adversity measures at 15.02 (0.43; 13–15) MRI at 16.45 (0.51; 13–15) and 18.80 (0.44; 17–20) years 46% | Childhood Trauma Questionnaire | Longitudinal Covariates: age, sex, SES, diagnosis | Childhood maltreatment was associated with accelerated pituitary gland development in females. |
Paquola et al. [26] Australia | 123 participants Follow Up: 52 participants | 19 (3; 14–28) 64.2% | Childhood Trauma Questionnaire | Longitudinal | Childhood maltreatment was associated with significantly stunted right hippocampal growth. |
Farrow et al. [43] Australia | 129 participants from Families and Childhood Transitions Study (FACTS) | Baseline: 8.4 (8–9.09) Follow Up: 9.9 (9.4–11.1) 52.7% | Lifetime Incidence of Traumatic Events Multidimensional Neglectful Behaviour Scale | Longitudinal Covariates: age, sex, SES, ICV | Childhood neglect was associated with greater baseline anterior pituitary volume, that was stable over the follow-up period. |
Gehred et al. [183] New Zealand | 861 participants of the Dunedin Study | Adversity accessed 7 times between 3 to 15 years. At 38 years retrospective account of adversity. MRI at 45 years. 49.3% | CDC–Kaiser Permanente Adverse Childhood Experiences Questionnaire | Longitudinal Covariates: Prenatal complications, Neurodevelopmental differences, perceived adult stress. | Childhood adversity, both prospectively and retrospectively accessed, was linked to smaller total surface area, thinner average cortex and smaller subcortical GMV. Stronger and more widespread association was observed for prospectively ascertained childhood adversity. |
Hidalgo et al. [184] Netherlands | 2993 participants from Generation R cohort | 10.1 (0.6; 8.72–11.9) 50.8% | Life events and difficulty schedule. | Longitudinal Covariates: age at MRI, sex, total intracranial volume, maternal national origin, highest household education, and maternal prenatal alcohol use and smoking. | Childhood adversities (but not prenatal adverse events experienced by the mother) were related to global brain volume differences at age 10 years |
Institutionalised care | |||||
Tottenham et al. [48] USA | 38 PI, 40 control | 8.91 (2; 4.9–15.7) 25% | Institutionalised children | Covariates: age, cortical size | Late adoption was associated with larger corrected amygdala volumes, poorer emotion regulation. |
Olsavsky et al. [49] USA | 33 PI, 34 control | 10.5 (3.5; 4–17) 46.5% | Institutionalised children | Covariates: age at adoption, age at scan, IQ | Previously institutionalised exhibited reduced amygdala discrimination between mothers and strangers. These effects correlated with age-at-adoption. |
Herzberg et al. [50] USA | 44 PI, 30 control | 12.93 (0.58; 11.75–14.09) 67.5% | Institutionalised children | Covariates: age, sex, IQ, ICV | Later-adopted participants had decreased prefrontal volume and made fewer risky decisions. |
Hodel et al. [47] USA | 110 PI, 62 control | 13.01 (0.55; 12.04–14.15) 66.3% | Institutionalised children | Covariates: age, sex, ICV | Hippocampal volumes showed an association with duration of institutional care, with later-adopted children showing the smallest volumes. |
Sheridan et al. [51] Romania | 136 PI, 72 control from Bucharest Early Intervention Project | Institutionalised: 0.5–2.75 Assessment: 11.14–14.68 49.6% | Institutionalised children | Covariates: age, sex | Prolonged institutional rearing leads to deficits in reward responsivity and implicit learning. |
