Edited by: Wei Bao, University of Iowa, United States
Reviewed by: Rade Vukovic, Mother and Child Health Care Institute of Serbia, Serbia; Tiffany M. Powell-Wiley, National Institutes of Health, United States
Specialty section: This article was submitted to Diabetes, a section of the journal Frontiers in Endocrinology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Childhood maltreatment (CM) is shown to be associated with obesity and depression. However, the relationship of CM to prediabetic state is much less studied. We tested the hypothesis that CM increases the risk for prediabetic state due to glucose intolerance, reduced insulin sensitivity, and beta cell function.
Oral glucose tolerance test (OGTT)-derived metabolic parameters of glucose tolerance, insulin sensitivity, and beta cell function were measured in 121 participants aged 19–60 years. CM exposure was measured using the Childhood Trauma Questionnaire. Blood samples were collected to measure the inflammatory factors.
After controlling for age, race, gender, education, and depression, about 15% higher glucose area under the OGTT curve was observed in the CM group. CM individuals also exhibited impaired insulin sensitivity manifested by the Matsuda index and homeostasis model assessment of insulin resistance, which were correlated with CM severity after adjusting for depression. CM group showed approximately 50% lower disposition index. C-reactive protein and tumor necrosis factor-α levels were greater in the CM group vs. the non-CM group, and both were correlated with CM severity (
These data suggest an important relationship between CM and increased risk for prediabetic state due to glucose intolerance, impaired insulin sensitivity, and beta cell function. Our findings indicate that CM appears to be an independent risk factor for developing prediabetes.
Type 2 diabetes (T2D) is preceded by prediabetes, which is characterized by impaired fasting glucose levels and/or impaired glucose tolerance (
Childhood maltreatment (CM) presents a major public health concern. CM occurs at an epidemic rate, with nearly 700,000 substantiated reports each year in the U.S., though this is likely an underestimate of the true number (
CM may be one factor that contributes to impaired glucose tolerance and thereby inducing prediabetic state and the subsequent T2D. The possible connection between CM and prediabetic state is suggested by several avenues of research, and their relationship may be mediated
Accumulating evidence has demonstrated that individuals with CM showed increased levels of C-reactive protein (CRP) and circulating cytokines, including tumor necrosis factor-α (TNFα) (
To our knowledge, evidence is lacking to link prediabetic state in adulthood as a possible long-term consequence of CM. To address this issue, experiments were designed to examine whether CM is an independent risk factor for prediabetic state and its potential mechanisms. The primary objective of this study was to investigate the relationship between CM and oral glucose tolerance test (OGTT)-derived measures of glucose tolerance, insulin sensitivity, and beta cell function in adults with adverse childhood experiences. Furthermore, we explored whether any observed associations between CM and OGTT-derived measures were dependent on obesity and the inflammatory factors.
Participants were recruited through the University of Alabama at Birmingham hospital and the community in Birmingham, AL, USA. The project was approved by the Institutional Review Board at the University of Alabama at Birmingham. Each participant was consented and provided with a copy of written consent prior to completing any research procedures. One-hundred fifty subjects were enrolled, but only 121 participants completed the study, including males and females aged 19–60 years. The exclusion criteria were endocrine disorders (e.g., hypothyroidism, hyperthyroidism, and diabetes), use of systemic corticosteroids or drugs affecting glucose metabolism and pregnancy or lactation. The subjects self-reported their age, race, medical history, and level of education. A self-reported scale, Beck Depression Inventory (BDI), was used to assess depression symptoms in participants.
The subjects completed a modified version of a Childhood Trauma Questionnaire (CTQ), which is a well-validated measure of child abuse and neglect before the age of 18 (
Participants characteristics.
Non-CM ( |
CM ( |
||
---|---|---|---|
Women/men, |
36/16 | 43/26 | 0.696 |
Age, years | 35.0 ± 1.6 | 39.2 ± 1.6 | 0.052 |
Race, C/AA | 21/31 | 35/34 | 0.248 |
Education, ≥10 years (%) | 90.3 | 87.3 | 0.889 |
BMI, kg/m2 | 31.2 ± 0.96 | 29.9 ± 0.84 | 0.451 |
Waist-to-hip ratio | 0.84 ± 0.01 | 0.86 ± 0.02 | 0.556 |
BDI | 5.88 ± 1.43 | 12.56 ± 1.55 | 0.003 |
PN | 5.3 ± 0.1 | 8.6 ± 0.4 | <0.001 |
PA | 6.2 ± 0.2 | 9.8 ± 0.6 | <0.001 |
EN | 6.6 ± 0.2 | 14.8 ± 1.5 | <0.001 |
EA | 5.9 ± 0.2 | 11.9 ± 0.7 | <0.001 |
SA | 5.2 ± 0.1 | 9.4 ± 1.5 | <0.001 |
Total | 29.1 ± 0.4 | 52.6 ± 2.2 | <0.001 |
All participants underwent a standard 2-h OGTT after an overnight fast (10–12 h) and were administered a 75 g oral glucose solution (
The Matsuda index was utilized as a surrogate estimate of insulin sensitivity. The Matsuda index is a validated insulin sensitivity index that reflects an estimate of both hepatic and muscle insulin sensitivity (
Pancreatic beta cell function was assessed using OGTT-derived insulinogenic index and disposition index (DIOGTT). The insulinogenic index correlates well with insulin secretion measured by the hyperinsulinemic-euglycemic clamp and was calculated by the ratio of the incremental response of insulin to glucose at 30 min of the OGTT (
Blood samples (10 mL) were drawn from each participant by venipuncture. The blood was centrifuged at 3,000 g for 10 min that was immediately divided into aliquots and frozen at −80°C until samples were analyzed. Plasma concentrations of CRP (minimum sensitivity = 0.50 mg/L, intra- and inter-assay CV of 6.1 and 7.5%, respectively) were analyzed using immunoassay on a Stanbio Sirrus Analyzer. Plasma levels of TNFα (minimum sensitivity = 0.20 pg/mL, intra- and inter-assay CV of 4.2 and 5.6%, respectively) were analyzed using multiplex human immunoassay kits (Meso Scale Discovery, Gaithersburg, MD, USA). All samples were run in duplicate, and the mean of the duplicate samples was reported.
Body weight and height were measured to calculate body mass index (BMI) using the Quetelet index (kilograms per square meter). Waist and hip circumferences were measured for the calculation of waist-to-hip ratio. A dual-energy X-ray absorptiometry (GE-Healthcare, Madison, WI, USA) scan was acquired with a total body scanner for each participant to determine visceral fat amount as previously described (
Data were analyzed using the Statistical Package for the Social Sciences version 23 (SPSS Inc., Chicago, IL, USA). Data were presented as mean ± SE in the text, figures, and tables.
Differences between the CM and the non-CM groups in variables of interest were compared using Chi-square test for categorical data and analysis of covariance for continuous data with adjustment for age, race, gender, education, and depression. Partial correlation coefficients controlling for age, race, gender, education, and depression were also calculated to evaluate associations between CM scores and metabolic parameters. In addition, stepwise multiple regression analyses were used to determine which, if any, risk factors for prediabetic state were most predictive of glucose area under the OGTT curve, HOMA-IR, and beta cell function (i.e., DIOGTT).
Descriptive characteristics of the participants in the non-CM and the CM groups are presented in Table
As shown in Figure
Insulin sensitivity, as estimated by the Matsuda index and HOMA-IR, was significantly different in the CM group compared with the non-CM group (Figures
The CM score was negatively associated with the Matsuda index and positively associated with the HOMA-IR after controlling for age, sex, race, and education (both
Elevated levels of CRP and TNFα were found in the CM group vs. the non-CM (Figures
Comparison of plasma concentrations of CRP
Compared with the non-CM group, visceral fat mass was significantly greater in the CM group (940 ± 86 vs. 1,202 ± 118,
The current study examined prediabetic state risk in relation to CM, and provided evidence supporting a link between CM and the risk of developing prediabetic state and the subsequent T2D in adulthood. In the present study, we found that 60 min glucose levels and glucose AUC during an OGTT were greater in individuals with CM exposure than in the non-CM group. Insulin levels were not different between the two groups. In the CM group, reduced insulin sensitivity (measured by the Matsuda index and HOMA-IR), independent of age, sex, race, education, and depression, was observed. In line with reduced insulin sensitivity, the disposition index measured using DIOGTT was lower in the CM group. To explore the underlying mechanism for the pathway from the CM to prediabetic state risk, TNFα and CRP levels were measured, and both were significantly elevated in the CM group. Positive relationships were found between both TNFα and CRP levels and CM scores. Stepwise multiple regression analyses revealed that CM was an independent contributor to variance in glucose area under the OGTT curve, HOMA-IR, and DIOGTT. Taken together, these results suggest that CM is an independent risk factor for prediabetic state because its association with metabolic parameters was independent of depression. In addition, our findings also indicate that CM plays a critical role in glucose homeostasis and insulin sensitivity in humans.
Since CM has a sustaining negative impact on the health problems, including prediabetic state and T2D, it is essential from a preventative and therapeutic perspective to learn its role. Our findings that individuals with CM experiences have impaired glucose tolerance (i.e., elevated glucose AUC during the OGTT), reduced insulin sensitivity, and impaired beta cell function would suggest a heightened risk for the development of future T2D. In addition, 60 min glucose levels during the OGTT were significantly higher in the CM group and correlated with OGTT-derived Matsuda index (data not shown). It is known that impaired glucose tolerance and reduced insulin sensitivity are present long before the onset of overt T2D (
We attempt to explore the mechanisms underlying pathways from CM to glucose intolerance and reduced insulin sensitivity because it allows for the development of more specific targets for the subsequent T2D. Previous studies indicate that childhood stress increases the tendencies for pro-inflammatory status, leading to exaggerated cytokine response to stress and decreased sensitivity to inhibitory hormonal signals and hormonal dysregulation (
Previous studies have indicated that obesity is highly related to T2D (
Our findings have some clinical implications. Due to significant high prevalence of CM and its impact on insulin sensitivity and glucose homeostasis, practitioners should assess for child abuse and neglect history with every patient. Furthermore, clinicians could educate their patients to minimize risky health behaviors and increase stress management techniques to reduce the risk of developing prediabetes and future T2D in individuals who were exposed to childhood adversity. In addition, CM could also be related to problematic diabetes outcomes (e.g., worse adherence, higher HbA1c values) (
Several limitations to this study may need to be considered when interpreting our results. First, this study was limited due to the cross-sectional methodology, leading to some reservations regarding its generalizability. However, our findings offered substantial evidence of a strong association between CM and prediabetic state risk in adulthood. Second, underreporting of adverse childhood experiences in our participants probably occurred. Due to the sensitive nature of questions about CM, the responses may represent an underreporting of actual occurrence as a result of being reluctant to report adverse experiences. Third, some health behavior data, including smoking status, diet, and physical activity, might affect our observations. Future studies should collect these data and determine their effects on the health outcomes in individuals with CM. Finally, inconsistency was reported regarding the relationship between subscales of CM (including abuse or neglect) and odds of diabetes, whereas most studies reported a positive association. In the present study, the impact of subscales of CM on prediabetic state risk was unable to be examined due to sample size. Despite the limitations of current study, CM appears to be a construct-relevant factor for prediabetes prevention and intervention.
Understanding risk factors related to the development of prediabetic state and future T2D as well as subsequent poor glycemic control is critical, and CM may be one such risk factor. Our study suggests that CM is independently associated with prediabetic state risk, possibly by impairing glucose tolerance, reducing insulin sensitivity, and beta cell function. This chain of events might begin with childhood exposure to abuse and/or neglect, leading to the development of prediabetes and subsequent T2D as a result of long-term effects of pathophysiological response to the CM. Further study is warranted to address the mediators and moderators, and underlying mechanisms for the pathways from CM to prediabetic state and the subsequent T2D, in the hope of developing specific targets, i.e., CM, to minimize the rate of prediabetic state and T2D and to improve its outcomes.
The project was approved by the Institutional Review Board at the University of Alabama at Birmingham. Written informed consent was obtained from each participant prior to completing any research procedures.
LL designed the experiments, analyzed the data, wrote and edited the manuscript. WG contributed to experimental design and the discussion, and reviewed/edited the manuscript. BG contributed to review and edit the manuscript. All authors approved the submission and subsequent publication.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors gratefully acknowledge the help of the University of Alabama at Birmingham Metabolism Core Laboratory with laboratory analyses.
AUC, area under the curve; BMI, body mass index; CM, childhood maltreatment; CRP, C-reactive protein; CTQ, Childhood Trauma Questionnaire; DIOGTT, OGTT-derived disposition index; HOMA-IR, homeostasis model assessment of insulin resistance; OGTT, oral glucose tolerance test; TNFα, tumor necrosis factor-α; T2D, type 2 diabetes.