The Relationship Between Forms of Childhood Maltreatment and Vascular Function Indices in Black Men and Women

Abstract

Objective:

Examine the association between types of childhood maltreatment and surrogate markers of vascular function, and to test whether sex modifies these relationships.

Methods:

Childhood maltreatment and indices of vascular function were assessed in a cohort of healthy Black adults without known CVD (n=404). Maltreatment (physical, sexual, emotional, and general trauma) domains were assessed using the Early Trauma Inventory Short Form. A trauma severity index score was calculated by summing the indexes for each domain. Outcomes of central augmentation index (cAIx) corrected for a heart rate of 75 bpm, and carotid femoral pulse wave velocity were measured as indices of wave reflections and arterial stiffness, respectively, and central pulse pressure (CPP). Associations between each domain and outcomes were assessed using multivariate-adjusted (demographics, clinical, health behaviors, and depressive symptoms) and sex-stratified linear regression models.

Results:

The cohort had a mean age of 53±10.3 years, and 61% were women. After adjustments, emotional abuse (2.68 [95% CI: −4.78, −0.58, P=.013]) and physical abuse (−2.38 [−4.26, −0.50, P=<.001]) were associated with cAIx in the overall cohort. Significant sex interactions were identified for emotional (P=.009) and physical abuse (P=.001) with cAIx and for emotional abuse with CPP (P=.021). Among women, physical abuse was associated with a higher cAIx (2.22 [95% CI: 0.70, 3.73, P=.004]), and among men, a lower cAIx (−2.59 [95% CI: −4.98, −0.19, P=.035]). Emotional abuse was also associated with higher CPP measures among women (1.31 [95% CI: 0.15, 2.46, P=.027]) with no associations identified in men.

Conclusions:

Physical and emotional abuse was associated with higher cAIx and CPP in women, but not men, suggesting sex-specific effects of early trauma on vascular function.

INTRODUCTION

Childhood traumatic events have garnered recognition as a significant risk factor for adverse mental and physical health outcomes including adult-onset cardiovascular disease (CVD).^1–3 Childhood traumatic events broadly encompass a range of adverse experiences, including emotional, physical, and sexual abuse, as well as general trauma.³ These childhood experiences have also been shown to differ by sex, leading to variations in susceptibility to acute and chronic health effects.^4–6 Existing studies examining the relationship between specific types of childhood maltreatment and subclinical markers of CVD, as well as, potential sex differences remain limited and have yielded mixed findings.⁷ A deeper understanding of which types of childhood adversity most impact cardiovascular health has important implications for developing tailored interventions and identifying key targets to mitigate the long-term cardiovascular consequences of early life exposures.

Similar to documented sex differences in CVD patterns, women are disproportionately more likely to experience childhood sexual abuse, emotional abuse, and neglect, whereas the prevalence of physical abuse seems comparable between sexes.^8,9 In women, all types of childhood maltreatment except emotional neglect have been associated with hypertensive heart disease and a greater risk of cerebrovascular disease.⁷ In men, all forms of childhood maltreatment, except sexual abuse, have been associated with hypertensive heart disease, whereas all except physical and sexual abuse have been linked to an increased risk of cerebrovascular disease.⁷ Furthermore, evidence indicates the impact of these events are stronger for women than men suggesting unequal susceptibility to future cardiovascular events.^6,7

Forms of childhood traumatic events, including physical abuse, sexual abuse, and neglect, are linked to preclinical markers of vascular dysfunction, which underlie the pathophysiology of CVD, and are accompanied by adverse hemodynamic changes.^6,10–12 Vascular dysfunction involves impairment of the endothelium, the microvascular circulation and stiffening and thickening of the large arterial walls.^13,14 In the large arteries, these vascular changes result from an imbalance between endogenous vasoconstrictive and vasodilatory factors, increased concentration of vascular reactive oxygen species and pro-inflammatory mediators, and loss of homeostatic function that ultimately reduces the distensibility and elasticity of both large and small vessels.^13,14 Clinically relevant early markers of large arterial dysfunction include measures of arterial stiffness [carotid-femoral pulse wave velocity (cfPWV)], abnormal wave reflections (augmentation index cAIx) and abnormalities in central hemodynamic parameters [central pulse pressure (CPP)].¹³ These early markers of vascular dysfunction are associated with established CVD risk factors and adverse outcomes. Measures of arterial stiffness, wave reflections, and central hemodynamics have been linked to multi-organ damage, obesity, diabetes, stroke, and Alzheimer disease.^15–17

Similar to the sex-specific differences observed in cardiovascular outcomes, these vascular alterations are believed to have more severe health consequences in women than in men;⁶ however, empirical findings remain mixed. A previous study on sexual abuse found that childhood exposure among midlife women was associated with higher carotid intima-media thickness,^18,19 but there was no sex-specific difference observed in another similar study from Germany.²⁰ Prior exposure to adverse childhood events was also noted to be associated with impaired endothelial function among women with a history of childhood adversities.¹¹ Conversely, childhood sexual abuse among men was linked to a 3-fold increased risk of myocardial infarction, whereas no such association was observed in women exposed to sexual abuse.²¹ Investigation into sex differences of these associations has offered valuable insights into the impact of certain exposures on upstream preclinical CVD risk; however, evidence regarding their relationship with additional preclinical markers of CVD risk remains limited. Understanding how different forms of childhood maltreatment relate to preclinical markers of CVD may clarify underlying mechanisms of risk and inform targeted interventions to reduce future CVD burden.

To better understand the relationships between forms of childhood traumatic events of general trauma, emotional, physical, and sexual abuse, we sought to extend upon our previous findings investigating the cumulative impact of childhood traumatic events on vascular dysfunction including arterial stiffness, and surrogate markers of arterial compliance.²² Given the disparate prevalence of childhood traumatic events that are experienced among Black children compared with White children, 21% versus 11%, these disparities may partially contribute to the disproportionate burden of early-onset hypertension and CVD that is observed among Black Americans.^1,23 Furthermore, due to the scarcity of studies examining the relationship between childhood adversities and preclinical measures of CVD risk in Black populations, further research is needed to identify which specific exposures are most associated with heightened CVD risk. The aims of the current study were: (a) to investigate whether exposure to general trauma, and emotional, physical, and sexual abuse during childhood is associated with vascular function in a cohort of adult Black males and females with a low CVD risk burden; and (b) test if sex moderated these associations, as prior studies have found that men and women have differential exposures to childhood traumatic events,¹ which may have important clinical implications for risk assessment, identification, and intervention targeting.

METHODS

Study Participants

The current study was a secondary data analysis of the Morehouse-Emory Cardiovascular Center for Health Equity (MECA) Study, a cross-sectional study designed to understand the intra-racial heterogeneity of environmental, individual, and biological factors that predispose Black Americans to increased risk or resilience from CVD.²⁴ Adult participants (N=502) were recruited from 2016 to 2019, utilizing multiple sampling methods that included sampling across census tracts and convenience sampling across the greater Atlanta metropolitan area, between 30 to 70 years of age.^25,26 Comprehensive details of the primary study methods are described elsewhere.^24,26 Inclusion criteria included self-identification as African American or Black, 30 to 70 years old, and resided in their current neighborhood for at least 6 years.²⁴ Exclusion criteria included individuals with a known history of CVD, such as myocardial infarction, congestive heart failure, cerebrovascular accidents, coronary artery disease, peripheral arterial disease, atrial fibrillation, and cardiomyopathies. Additional exclusion criteria included concomitant chronic diseases (eg, cancer, human immunodeficiency virus, or lupus), self-reported substance abuse (alcohol or illicit drug), diagnosed psychiatric illness, pregnant or lactating women, and self-reported physical inability to participate in increased physical activity.^24,26 Enrolled participants completed a physical exam, blood draw, and questionnaires by trained research personnel at either Morehouse School of Medicine or Emory University School of Medicine during the study visit. All aspects of the study were approved by the Institutional Review Boards of both institutions. The analytic sample for the current study was (n=404) participants due to the exclusion of participants with missing data on the following items: Early Trauma Inventory (n=43), central augmentation index (n=36), and carotid-femoral pulse wave velocity (n=22).

Vascular Function Outcomes

Indices of arterial stiffness and pulse wave reflections were conducted noninvasively using applanation tonometry [SphygmoCor, Inc.]. Carotid-femoral artery pulse wave velocity (cfPWV), considered the gold standard method for the assessment of aortic stiffness,²⁷ was assessed using transcutaneous Doppler flow velocity recordings simultaneously over the common carotid artery and the femoral artery.²⁴ Peripheral pressure waveforms were recorded from the radial artery at the wrist with a high-fidelity micromanometer. After 20 sequential waveforms were acquired, a validated generalized transfer function was used to generate the corresponding central aortic pressure waveform.²⁴ Central augmentation index (cAIx) is a measure of systemic arterial stiffness derived from the ascending aortic pressure waveform and represents the degree of pressure augmentation secondary to reflected waves from the periphery during systolic ejection.^28,29 Augmentation index was normalized for a heart rate of 75 beats per minute.²⁴ Reproducibility studies in our laboratory on consecutive days on 9 subjects demonstrated a coefficient of variation of 20.3% and 3.8% for augmentation index and PWV, respectively.²⁴ Central pulse pressure (CPP), a surrogate marker of arterial compliance, was calculated as the difference between the aortic systolic and diastolic blood pressures.³⁰ These measures are considered highly reproducible and are associated with CVD events.^31,32

Childhood Traumatic Events

Childhood traumatic events were assessed using the validated 27-item Early Trauma Inventory Self-Report Short Form (ETISR-SF), which assesses exposure to traumatic life events before age 18.³³ The subdomains consisted of general trauma, physical abuse, emotional abuse, and sexual abuse. Researchers may access a review copy of the scale at https://eprovide.mapi-trust.org/instruments/early-trauma-inventory-self-report-short-form. For the current study, we focused on the individual subdomains and not the cumulative score of the ETISR-SF. A trauma severity index score was calculated by summing the indexes for each of the 4 domains.^33,34 Higher scores are indicative of greater trauma event exposure before the age of 18 years.³³ The ETISR-SF has demonstrated high reliability with a Cronbach α ranging from 0.7 to 0.91 for the scales individual domains.³³

Covariates

Covariate measures were chosen based on a priori knowledge of their association with vascular function.^35–37 Sociodemographic variables included age (years), education (≤high school, some college, or college and above), household income (<$25K, $25K to <$50K, $50K to <$75K, and ≥$75K), and marital status (married/ unmarried couple, divorce/separated/widowed, and never married). Clinical factors included systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate, and height in meters, fasting lipid panel [high-density cholesterol and low-density (HDL-c, LDL-c), triglyceride levels], and fasting blood glucose, mg/dL; medical history was self-reported and included medical history of diabetes or taking medication and hypertension or taking medication. Health behavior factors included alcohol use in the past 30 days (never; 1 to 2 times/week; 3 to 4 times/week; 5 to 6 times/week; daily; once/month; 2 to 3 times/month) and smoking status (current/quit ≤12 mo, never or quit>12 mo). Depressive symptoms were self-reported and assessed using the validated Beck Depression Inventory Index (BDI).³⁸ In these analyses, depressive symptoms were analyzed as a continuous variable to preserve the full range of symptom variability. This approach is consistent with prior evidence that even subclinical levels of depressive symptoms may influence cardiovascular and health outcomes, making categorical cutoffs less informative.³⁹

Statistical Analyses

All variables were examined for missing data, normality, and extreme values. Study variables are reported as mean (SD) for normally distributed continuous variables, median [IQR] for skewed continuous variables, and proportions for categorical variables. CAIx, CfPWV, and CPP were examined as continuous variables. Group differences between continuous variables were compared using t tests or χ² analyses for categorical variables and Mann-Whitney U test for non-normal distributed continuous variables.

Bivariate associations were conducted between each childhood traumatic event and vascular function outcomes of cAIx, CfPWV, and CPP. To facilitate interpretation and comparison across predictors, all continuous independent variables were standardized (mean=0, SD=1) before linear regression analysis. Standardized beta coefficients are reported to facilitate interpretation across predictors. Multivariate linear regression analyses were used to examine the independent relationship between each childhood traumatic event and vascular function outcomes (cAIx, CfPWV, and CPP) separately and after adjusting for potential confounding factors considered a priori for the overall cohort. Five models were specified and included, (model 1), sex; (model 2), demographic variables of age, education, household income, and marital status; (model 3), clinical factors of SBP, DBP, heart rate, height in inches, BMI, HDL-c, LDL-c, glucose, and triglycerides, medical history (diabetes, hypertension); (model 4), addition of health behaviors (smoking status and alcohol use), factors known to affect vascular function; and lastly, (model 5), the inclusion of depressive symptoms. We evaluated multicollinearity among predictors using the variance inflation factor (VIF), and all variables had VIF values below 5 indicating no evidence of multicollinearity.⁴⁰ To determine whether forms of childhood traumatic events differed by sex, we included an interaction term for each childhood traumatic event (general trauma, emotional, physical, and sexual abuse) and sex in the fully adjusted regression model 5. Significant interaction models were then stratified by sex. The significance level for main effects and interaction effects was set at P<.05. Statistical analyses were performed using STATA, version SE 19.5 (StataCorp. 2025, LP: College Station, TX) software.

RESULTS

The distributions of participant characteristics for the overall sample and by sex are shown in Table 1. The distribution of the study population was 61% women and 39% men, mean age was 53±10.3 years old. The overall cohort held some form of college education; nearly half of the cohort reported earnings of less than $25,000 annually, and the majority were unmarried. Demographic factors by sex reveal that women were slightly older and more educated than men. There were significant sex differences between men and women for anthropometric, CVD risk factors, and health behaviors such that women had a higher BMI, shorter height (meters), higher LDL-c, and lower moderate/heavy alcohol consumption, lower mean depression scores, and lower rates of smoking.

TABLE 1.

Distribution of Select Characteristics for the Overall Cohort Stratified by Sex, the Morehouse-Emory Cardiovascular Center for Health Equity (MECA) Study, (n=404)

	Overall (n=404)	Males (n=159)	Females (n=245)	P a
Age, M (SD)	53 (10.3)	52 (10.3)	54 (10.2)	.060
Education, n (%)				<.001
HS or less	123 (31)	66 (42)	57 (24)
Some college	144 (36)	50 (31)	94 (39)
College & above	136 (34)	43 (27)	93 (38)
Income, n (%)				.515
<$25K	187 (49)	75 (51)	112 (48)
$25K-<$50K	111 (29)	46 (31)	65 (28)
$50K-<$75K	52 (14)	16 (11)	36 (15)
$≥75K	33 (9)	11 (7)	22 (9)
Marital status, n (%)				.794
Married	124 (31)	46 (29)	78 (32)
Unmarried	278 (69)	112 (70)	166 (68)
Anthropometrics and CVD risk factors
BMI, kg/m2, M (SD)	32.3 (7.5)	30.1 (6.6)	33.8 (7.8)	<.001
SBP, mmHg, M (SD)	131 (18.9)	132 (17.7)	130 (19.6)	.332
DBP, mmHg, M (SD)	80 (11.6)	80.1 (11.7)	80.1 (11.6)	.976
Heart rate, bpm, M (SD)	64 (11.1)	63 (11.2)	65 (11.0)	.171
Height, m, M (SD)	1.69 (0.09)	1.77 (0.07)	1.64 (0.07)	<.001
HDL-c, mg/dl, M (SD)	58 (17)	55 (18.1)	59 (16.0)	.016
LDL-c, mg/dl, M (SD)	116 (35.3)	109 (37)	120 (33.9)	.004
Triglycerides, MG/DL, M [IQR]	91 [65, 127]	90 [63, 126]	92 [68, 127]	.364
Glucose, mg/dl, M [IQR]	93 [86, 105]	94 [88, 106]	92 [85, 104]	.559
Diabetes, n (%)	97 (24)	36 (23)	61 (25)	.597
Hypertension, n (%)	213 (53)	80 (50)	133 (55)	.385
Beck Depression Index, M (SD)	7.5 (7.7)	8.6 (8.9)	6.7 (6.8)	.022
Health behaviors, n (%)
Alcohol use				.001
none	205 (51)	71 (45)	134 (55)
low (1-2 drinks/day)	164 (41)	64 (40)	100 (41)
moderate/heavy (≥3 drinks/day)	35 (9)	24 (15)	11 (4)
Smoking status, n (%)				<.001
current	89 (22)	52 (33)	37 (15)
quit ≤12 mo	17 (4)	10 (6)	7 (3)
never/quit>12 mo	298 (74)	97 (61)	201 (82)
Childhood traumatic events, M (SD)
emotional abuse	1.6 (1.9)	1.4 (1.8)	1.7 (1.9)	.211
general trauma	4.3 (2.4)	4.6 (2.5)	4.1 (2.3)	.021
physical abuse	2.0 (1.7)	2.4 (1.7)	1.7 (1.6)	<.001
sexual abuse	1.0 (1.6)	0.5 (1.2)	1.3 (1.7)	<.001
Vascular stiffness markers
CAIx%, M (SD)	25 (12.3)	23 (13.7)	27 (11.1)	.001
CfPWV,m/s, M (SD)	7.6 (2.1)	8.0 (2.4)	7.3 (1.8)	.003
CPP,mmHg, M (SD)	42.1 (11.1)	40.1 (10.0)	43.5 (11.6)	.003

K=thousands of dollars; BMI=body mass index; SBP=systolic blood pressure; M [IQR]=median [interquartile range]; cAIx=central augmentation index; cfPWV=carotid-femoral pulse wave velocity; CPP=central pulse pressure.

^a Mann-Whitney test P-value reported for non-normally distributed continuous variables, CAIx is %.

Sex differences existed for general trauma, physical, and sexual abuse. Compared with women, men reported higher exposure to general trauma and physical abuse, whereas women had more exposure to sexual abuse. As expected, females also had significantly higher mean cAIx and CPP measures compared with males who had significantly higher mean cfPWV.

Associations Between Childhood Traumatic Events and Vascular Function

The fully adjusted associations between each childhood traumatic event (general trauma, emotional, physical, and sexual abuse) and vascular function measures for the overall cohort are shown in Table 2. Detailed models for each traumatic event and vascular function measures are presented in Table S1, Supplemental Digital Content 1, http://links.lww.com/PSYMED/B141, Table S2, SDC 2, http://links.lww.com/PSYMED/B142, Table S3, SDC 3, http://links.lww.com/PSYMED/B143, Table S4, SDC 4, http://links.lww.com/PSYMED/B144. Among the overall cohort, emotional abuse was associated with a −2.68% lower cAIx, (95% CI: −4.78, −0.58, P=.013) and physical abuse was associated with −2.38% lower cAIx (95% CI: −4.26, −0.50, P=.013) in the fully adjusted models that included the interaction terms. There were no significant associations between any of the additional childhood traumatic events and vascular function measures of cAIx, cfPWV, or CPP.

TABLE 2.

Standardized Beta Coefficients (β) and 95% CI for the Association Between Childhood Traumatic Events and Vascular Function Outcomes

	CAIx β (95% CI)	P	CfPWV β (95% CI)	P	CPP β (95% CI)	P
Emotional	−2.68 (−4.78, −0.58)	.013	0.23 (−0.09, 0.55)	.156	−1.19 (−2.69, 0.30)	.118
Female vs. male	−0.36 (−3.72, 3.00)	.832	−0.31 (−0.82, 0.21)	.243	0.76 (−1.63, 3.16)	.531
Emotionalasex	−3.89 (1.43, 6.36)	.002	−0.28 (−0.66, 0.10)	.143	2.56 (0.81, 4.32)	.004
General	−6.74 (−28.34, 14.85)	.539	−0.65 (−3.93, 2.63)	.697	−5.66 (−20.96, 9.64)	.467
Female vs. male	29.33 (−14.73, 73.39)	.191	−1.87 (−8.56, 4.82)	.582	33.45 (2.23, 64.68)	.036
Generala​sex	−18.57 (−8.77, 45.91)	.183	−0.96 (−5.11, 3.19)	.648	20.24 (0.87, 39.62)	.041
Physical	−2.38 (−4.26, −0.50)	.013	0.03 (−0.26, 0.32)	.853	−0.60 (−1.96, 0.77)	.391
Female vs. male	−1.03 (−4.39, 2.32)	.546	−0.28 (−0.80, 0.24)	.289	0.58 (−1.85, 3.01)	.640
Physicalasex	4.57 (2.16, 6.98)	<.001	−0.15 (−0.52, 0.22)	.433	1.23 (−0.52, 2.98)	.167
Sexual	0.90 (−15.22, 17.01)	.913	−0.42 (−2.87, 2.02)	.733	5.43 (−6.08, 16.94)	.354
Female vs. male	1.87 (−8.35, 12.09)	.719	−0.38 (−1.93, 1.17)	.632	−0.95 (−8.25, 6.34)a	.797
Sexualasex	−5.52 (−12.62, 23.66)	.550	−0.28 (−3.04, 2.47)	.840	−2.56 (−15.51, 10.39)	.698

All continuous independent variables were standardized before analysis. Values shown are standardized beta coefficients.

^a Fully adjusted models include: trauma domain, sex, age, education, income, marital status, SBP, DPB, heart rate, height (meters), HDL-c, LDL-c, glucose, triglycerides, history of DM, HTN, smoking status and alcohol use, depressive symptoms, trauma domain, and sex interaction term.

There were, however, significant trauma and sex interactions between emotional abuse and cAIx (P=.002), (Figure 1), physical abuse and cAIx (P=<.001), (Figure 2), and emotional abuse and CPP (P=.004), (Table 2, Figure 3). In sex-stratified analyses (Table 3), emotional abuse was associated with cAIx after full adjustment in men and not women. Physical abuse among men was associated with an adjusted −2.59% (95% CI: 4.98, 0.19) lower cAIx, whereas among women, it was associated with a 2.22% (95% CI: 0.70, 3.73) higher cAIx. Emotional abuse was no longer associated with CPP in models 3 to 5 after adjusting for clinical, health behaviors, and depressive symptoms for men. Among women, emotional abuse was associated with 1.31 mmHg (95% CI: 0.15, 2.46) higher CPP after full adjustment.

FIGURE 1.

Predictive margins of sex and emotion abuse interaction on central augmentation index. Fully adjusted model including trauma domain, sex, age, education, income, marital status, SBP, DPB, heart rate, height (meters), HDL-c, LDL-c, glucose, triglycerides, history of DM, HTN, smoking status and alcohol use, depressive symptoms, trauma domain, and sex interaction term. Color image is available only in online version.

FIGURE 2.

Predictive margins of sex and physical abuse interaction on central augmentation index. Fully adjusted model including trauma domain, sex, age, education, income, marital status, SBP, DPB, heart rate, height (meters), HDL-c, LDL-c, glucose, triglycerides, history of DM, HTN, smoking status and alcohol use, depressive symptoms, trauma domain, and sex interaction term. Color image is available only in online version.

FIGURE 3.

Predictive margins of sex and emotional abuse interaction on central puluse pressure. Fully adjusted model including trauma domain, sex, age, education, income, marital status, SBP, DPB, heart rate, height (meters), HDL-c, LDL-c, glucose, triglycerides, history of DM, HTN, smoking status and alcohol use, depressive symptoms, trauma domain, and sex interaction term. Color image is available only in online version.

TABLE 3.

Sex Stratified Analyses of Unadjusted and Adjusted Association Between Emotional and Physical Abuse Childhood Trauma Domains and Central Augmentation Index and Central Pulse Pressure, in Men and Women

Outcome	Childhood Maltreatment	Model 1	P	Model 2	P	Model 3	P	Model 4	P	Model 5	P
		β (95% CI)		β (95% CI)		β (95% CI)		β (95% CI)		β (95%CI)
CAIx @ 75bpm
Males	Emotional abuse	−2.66 (−4.85, −0.48)	.017	−1.74 (−4.10, 0.61)	.146	−2.05 (−4.44, 0.35)	.093	−2.40 (−4.83, 0.04)	.053	−2.93 (−5.87, 0.01)	.050
	Physical abuse	−2.82 (−4.84,−0.80)	.007	−2.18 (−4.37, 0.02)	.052	−2.19 (−4.44, 0.07)	.058	−2.60 (−4.92, −0.27)	.029	−2.59 (−4.98, −0.19)	.035
Females
	Emotional abuse	0.77 (−0.60, 2.14)	.268	1.05 (−0.36, 2.45)	.143	1.09 (−0.33, 2.51)	.133	0.91 (0.51,2.34)	.208	0.92 (−0.53, 2.37)	.211
	Physical abuse	1.82 (0.34, 3.30)	.016	2.66 (1.19, 4.12)	<.001	2.57 (1.09, 4.04)	.001	2.45 (0.95, 3.95)	.002	2.22 (0.70, 3.73)	.004
CPP
Males	Emotional abuse	−2.41 (−4.00, −0.83)	.003	−2.11 (−3.82, −0.41)	.016	−1.02 (−2.43, 0.39)	.155	−1.10 (−2.54, 0.35)	.136	−1.17 (−2.88, 0.54)	.178
Females
	Emotional abuse	0.12 (−1.32, 1.56)	.866	0.90 (−0.42, 2.22)	.178	1.56 (0.44, 2.69)	.007	1.50 (0.37, 2.62)	.009	1.31 (0.15, 2.46)	.027

Values shown are standardized beta coefficients β.

Model 1: unadjusted.

Model 2: (model 1) + age, education, income, marital status.

Model 3: (model 2) + SBP, DPB, heart rate, height (meters), HDL-c, LDL-c, glucose, triglycerides, history of DM, HTN.

Model 4: (model 3) + smoking status and alcohol use.

Model 5: (model 4) + depressive symptoms.

DISCUSSION

In the current study, we sought to investigate the relationship between types of childhood maltreatment and markers of vascular function, and test whether sex modified these relationships in a cohort of Black adults. The main findings were that physical abuse was associated with lower cAIx in men, but higher cAIx in women. Emotional abuse was associated with higher central hemodynamics of CPP in women, with no relationship observed in men. A higher cAIx is indicative of increased pulse wave reflections or reduced arterial compliance, that in previous studies has been associated with adverse cardiovascular outcomes.⁴¹ To our knowledge, these findings are novel in Black adults, a group with elevated exposure to childhood trauma and limited research on its impact on subclinical CVD risk. Our findings build upon our previous observations that cumulative childhood trauma is associated with elevated cAIx and CPP, with sex significantly altering these relationships.²² We extend these prior findings by conducting an in-depth investigation of each type of childhood trauma on multiple indices of vascular function. Collectively, the current findings demonstrate that exposure to specific types of traumatic experiences during childhood may have important long-term clinical implications, especially for Black women.⁶

Emotional Abuse, Physical Abuse, and Central Augmentation Index

The emergence of an association between emotional abuse, physical abuse, and cAIx only after inclusion of the interaction term indicates emotional abuse and physical abuse do not independently predict cAIx. Moreover, our findings for the overall cohort contrast with prior investigations that have linked emotional abuse and physical abuse to an increased risk of subclinical and late-life CVD.^18,42,43 Although sex differences were initially observed in the associations between emotional abuse, physical abuse, and cAIx, sex-stratified analyses revealed that emotional abuse was only associated with cAIx in men and not women. However, physical abuse was associated with cAIx in both males and females, with divergent patterns observed by sex. Females exposed to childhood physical abuse exhibited higher cAIx values, whereas males showed lower cAIx values. Our findings among females align in part with prior studies, which have reported that midlife women exposed to physical abuse had higher carotid intima-media thickness and increased carotid plaque.¹⁸ Similarly, in a population-based study of adults aged 30 to 64, women exposed to traumatic events demonstrated higher augmentation index values, an association not observed in men.⁴⁴ Additional study is warranted to elucidate why physical abuse on cardiovascular health is stronger and more consequential in Black women compared with Black men, despite reporting higher exposure to physical abuse than women.

Emotional Abuse and Central Pulse Pressure

Sex differences were also observed between emotional abuse and CPP. Sex-stratified analyses revealed an association between emotional abuse and higher CPP among women that was absent in men. Our findings are in alignment with prior evidence suggesting the impact of certain forms of childhood adversity is larger for women than men.⁷ Factors that may contribute to these relationships in men require further investigation as our findings seem to be counterintuitive to current evidence.

SEX DIFFERENCES

The cause of these sex differences remains less clear. The cumulative stress model posits that distinct early adversities have additive effects on health, and the high burden of risk factors observed in women in this study may reflect stress-related mechanisms underlying behavioral, cognitive, affective, and biological sequalae.^45,46 A notable paradox was that depressive symptoms were higher in males than females, even though depression is known to exert stronger cardiovascular effects in females.^47,48 Although we did not observe adverse cardiovascular effects from abuse exposures among males, this should not be interpreted as absence of risk, as Black males often experience underdiagnosed chronic conditions and disproportionate CVD morbidity and mortality.⁴⁹ These patterns may reflect distinct coping strategies used to navigate cumulative stressors and societal expectations.⁵⁰ Nonetheless, additional study is warranted to better disentangle these relationships.

Childhood Maltreatment and Pulse Wave Velocity

There was no evidence of sex differences in the relationships between any of the childhood traumatic events and arterial stiffness, measured as cfPWV. These findings were unexpected, given the substantial body of evidence indicating that arterial stiffness precedes the development of hypertension.^30,51 Augmentation index, a surrogate marker of arterial stiffness, is higher in females than males, largely but not entirely, due to shorter height in women. Other factors known to impact cAIx and CPP include heart rate, BP, smoking, and, LDL-c^16,52 of which LDL-c was higher among females. Elevated LDL-c, a component of metabolic syndrome that is closely associated with obesity, is also considered a co-occurring comorbidity of post-traumatic stress disorder (PTSD).⁵³ Beyond the scope of the current study, further research is needed to explore the presence of co-occurring PTSD symptoms and childhood trauma exposure, to understand their role in the subclinical CVD pathway, and how they may contribute to sex differences.

STRENGTHS AND LIMITATIONS

Our study had limitations that warrant comment. First, we assessed childhood traumatic events in a cohort of adults who self-identified as African American/Black and were from a southern geographic region. Thus, the results may not be generalizable to other racial/ethnic populations or different geographical regions of the United States. Second, our study was cross-sectional, limiting our ability to determine causal inferences. We acknowledge that multiple factors including age, genetics, lifestyle, and access to health care substantially influence both the risk and management of CVD.⁵⁴ Accordingly, we used appropriate statistical methods to control for relevant covariates. Values for cAIx, cf-PWV, and CPP were analyzed as continuous variables and fell within the normal range based on established guidelines.^30,55,56 However, we acknowledge that due to the single time point of data collection, we were unable to determine any temporal deviations or abnormalities in these measures. Nonetheless, further longitudinal research is warranted to explore these associations in greater depth, particularly earlier in the life course, to capture additional determinants that may affect vascular function. Third, we retrospectively assessed childhood traumatic events introducing the potential for recall bias and under-reporting of trauma as traumatic events rarely occur in isolation suggesting individuals are likely to experience multiple events simultaneously.⁵⁷ In this regard, retrospective assessment of early life exposures through questionnaires is widely conducted and as such, individuals are likely to recall traumatic experiences that has occurred in their lifetime.⁵⁸ Given the established connection between exposure to traumatic events and sex differences of PTSD,⁵⁹ we were unable to investigate these relationships and their effect on vascular function. We acknowledge that our cohort included a wide age range; however, stratified analyses among participants under and over 50 years of age revealed that the associations remained relatively unchanged. In addition, we did neither control for menstrual sex hormones in relation to the menstrual cycle, nor occupational or physical activity on the day of vascular assessment, as these data were not collected as part of the parent study design. Lastly, we examined these associations using multiple surrogate markers of arterial stiffness, which may be less sensitive or informative than other indicators of subclinical disease for assessing the impact of childhood maltreatment on CVD risk.

Our study also had key strengths, including the assessment of traumatic-specific events utilizing a valid measure with clinically relevant, vascular function outcomes in a large cohort of Black adults, for which there is a paucity of evidence regarding these relationships, despite the unequal exposure to trauma and CVD burden in this population.^60,61 After excluding individuals with CVD, substance use, and psychiatric illness, all of which have been linked to early life maltreatment,⁶² we still observed associations between specific types of childhood exposures and indices of vascular function. We also leverage the use of incorporating noninvasive yet highly reproducible measures of arterial stiffness and surrogate markers of vascular function. Lastly, we were able to investigate the relationship between childhood traumatic events and their link to adult-onset cardiovascular health in a cohort of Black adults with a low CVD risk burden.

In conclusion, emotional abuse and physical abuse among women were associated with worse vascular function, measured as CPP and cAIx, suggesting lower arterial compliance. Men had better vascular function despite exposure to emotional and physical abuse. These findings suggest that further study is needed to identify the underlying mechanisms that may explain these sex-based disparities and CVD risk in Black adults. The findings highlight domains associated with CVD risk, particularly in Black women, and may inform screening practices in ambulatory care as well as secondary prevention strategies targeting modifiable risk factors to reduce CVD events.

Supplementary Material

psy-88-61-s001.docx ^{(60.7KB, docx)}

psy-88-61-s002.docx ^{(47.4KB, docx)}

psy-88-61-s003.docx ^{(44.9KB, docx)}

psy-88-61-s004.docx ^{(48.9KB, docx)}