2024; 22(1): 105-117  https://doi.org/10.9758/cpn.23.1079
The Association between Default-mode Network Functional Connectivity and Childhood Trauma on the Symptom Load in Male Adults with Methamphetamine Use Disorder
Shyh-Yuh Wei1, Tsung-Han Tsai1, Tsung-Yu Tsai1, Po See Chen1,2, Huai-Hsuan Tseng1,2, Yen Kuang Yang1,2,3, Tianye Zhai4, Yihong Yang4, Tzu-Yun Wang1
1Department of Psychiatry, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan
2Institute of Behavioral Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan
3Department of Psychiatry, Tainan Hospital, Ministry of Health and Welfare, Tainan, Taiwan
4Neuroimaging Research Branch, National Institute on Drug Abuse, Intramural Research Program, National Institutes of Health, Baltimore, MD, USA
Correspondence to: Tzu-Yun Wang
Department of Psychiatry, National Cheng Kung University Hospital, 138 Sheng Li Road, North Dist., Tainan 70403, Taiwan
E-mail: wangty@mail.ncku.edu.tw
ORCID: https://orcid.org/0000-0002-0561-9967
Received: March 29, 2023; Revised: May 25, 2023; Accepted: August 14, 2023; Published online: September 14, 2023.
© The Korean College of Neuropsychopharmacology. All rights reserved.

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Objective: The relationship between adverse childhood experiences and methamphetamine use disorder (MUD) has been shown in previous studies; nevertheless, the underlying neural mechanisms remain elusive. Childhood trauma is associated with aberrant functional connectivity (FC) within the default-mode network (DMN). Furthermore, within the DMN, FC may contribute to impaired self-awareness in addiction, while cross-network FC is critical for relapse. We aimed to investigate whether childhood trauma was associated with DMN-related resting-state FC among healthy controls and patients with MUD and to examine whether DMN-related FC affected the effect of childhood trauma on the symptom load of MUD diagnosis.
Methods: Twenty-seven male patients with MUD and 27 male healthy controls were enrolled and completed the Childhood Trauma Questionnaire. DMN-related resting-state FC was examined using functional magnetic resonance imaging.
Results: There were 47.1% healthy controls and 66.7% MUD patients in this study with adverse childhood experiences. Negative correlations between adverse childhood experiences and within-DMN FC were observed in both healthy controls and MUD patients, while within-DMN FC was significantly altered in MUD patients. The detrimental effects of adverse childhood experiences on MUD patients may be attenuated through DMN-executive control networks (ECN) FC.
Conclusion: Adverse childhood experiences were negatively associated with within-DMN FC in MUD patients and healthy controls. However, DMN-ECN FC may attenuate the effects of childhood trauma on symptoms load of MUD.
Keywords: Amphetamines; Adverse childhood experiences; Default mode network; Magnetic resonance imaging; Mediation analysis
INTRODUCTION

According to the 2021 World Drug Report, an estimated 27 million people used amphetamine-type stimulants, corresponding to 0.5 percent of the global population [1]. The use of amphetamines, especially methamphetamine, is increasing in parts of Asia and North America and is associated with substantial morbidities and psychiatric consequences [2]. The causes of drug abuse are complex and multifactorial. Childhood trauma is an important risk factor for substance use disorder [3]. The prevalence of childhood trauma is higher in people with substance use disorders [4,5]. A relationship between childhood abuse, including emotional, physical, and sexual abuse, and drug use, was found in a meta- analysis and in a prospective cohort study [6,7]. Never-theless, the underlying neural mechanism remains elusive, and to date, evidence for ameliorating interventions is limited.

Using resting-state functional connectivity (FC), network-based studies have identified core functional abnormalities across different classes of substance use disorder, especially the default-mode network (DMN) [8]. In individuals with substance use disorders, the FC of the DMN can vary between hyper- or hypo-FC, depending on multiple factors such as whether patients were abstinent during the scan, the specific substance used, and the specific region of the DMN; for example, FC involving the ventromedial prefrontal cortex (anterior DMN) tends to be decreased, while the posterior cingulate cortex (posterior DMN) tends to be increased [8]. Given that the DMN, a large-scale functional brain network, generally exhibits higher activity at rest or in tasks requiring internally directed/self-related cognition [8,9], aberrant FC within the DMN may contribute to impaired self-awareness in addiction [8]. DMN is functionally connected to other large-scale networks, such as the executive control network (ECN) and the salience network (SN), and it has effects on subcortical areas [10]. The dynamic interactions between the DMN and other networks may affect cognition, emotion regulation, attentional performance, and impulsivity [11,12]. Furthermore, disrupted DMN FC with ECN and SN is also critical for relapse [8]. Decreased anterior DMN-SN and posterior DMN-ECN connections were noted in cocaine-dependent patients [13]. There-fore, the DMN plays a central role in investigating the neurobiological mechanism underlying methamphetamine use disorder (MUD).

Childhood trauma experiences have been shown to affect brain development trajectories, leading to impaired structure and functional connectivity in several brain regions related with threat detection, emotion regulation, cognition, and reward anticipation [14]. Studies have found associations between childhood trauma and altered FC within specific functional networks of the SN [15] and ECN [16]. However, there is an increasing body of research focusing on the potential effects of maltreatment on the DMN, possibly due to its critical roles in internally directed/self-related cognition [8,9]. Additionally, a recent study using connectome-based predictive modeling to explore the relationship between functional connectome of large-scale brain networks and the influences of childhood trauma indicated that the DMN was the most implicated network in various types of childhood trauma [17]. Studies investigating the DMN in individuals exposed to maltreatment or early-life stress have also reported decreased DMN connectivity [14]. For example, neural effects of childhood trauma have been investigated using structural and functional magnetic resonance imaging (MRI) in the general population [18] or in patients with different psychiatric disorders, including major depressive disorder [19], posttraumatic stress disorder [20], and schizophrenia [21]. In general, childhood trauma is associated with dysconnectivity within the DMN in patients with schizophrenia, major depressive disorder, and posttraumatic stress disorder (PTSD), as well as in the general population [18-21]. In addition, aberrant DMN recruitment mediated or linked to the impacts of childhood trauma on the severity of psychopathology in schizophrenia [21], major depressive disorder [22], and PTSD [23]. How-ever, whether MUD patients have a similar presentation of DMN dysconnectivity under the childhood trauma experience is unknown. Further exploration of the role of the DMN in modifying the effects of childhood trauma on substance use behavior may provide insights into identifying resilient factors that can mitigate the traumatic effects.

In the present study, we aimed to investigate whether childhood trauma was associated with DMN-related FC among healthy controls and patients with MUD. We hypothesized that possible alterations in DMN-seeded FC in MUD patients may be linked to adverse childhood experi-ences. Additionally, based on previous findings suggesting that DMN dysconnectivity may mediate the severity of psychopathology in mental disorders [21], a mediation analysis was conducted to explore the passible relationship between childhood trauma and the symptom load of MUD.

METHODS

Subjects

The research protocol was approved by the Institutional Review Board for the Protection of Human Subjects at National Cheng Kung University Hospital (NCKUH) (A- ER-106-197). The study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. The procedures were fully explained to each participant before they were asked to sign the informed consent form. MUD patients were recruited from the NCKUH addiction clinic, and healthy controls were recruited from the community. Each patient was initially interviewed and diagnosed by a board-certified psychiatrist and then screened by a research team member well trained in using the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5), criteria [24] and the Chinese version of the Mini International Neuropsy-chiatric Interview (MINI) [25]. The MINI has good reliability and has been widely used in clinical trials and epidemiological studies [26], and its interrater reliability in the Chinese version was approximately 0.75 in previous studies [27,28].

Inclusion criteria included adult male or female patients between 18 and 65 years old who met the DSM-5 criteria for current MUD. Exclusion criteria included cognitive disorder; any contraindications in relation to MRI, such as having a metal implant, a pacemaker implant, or claustrophobia; and a history of one or more uncontrolled major physical conditions. Patients with other psychiatric comorbidities, such as anxiety, depression, and antisocial personality disorder, were not excluded from the study. Healthy volunteers in the healthy control group were recruited from the community. The MINI was used to screen their psychiatric conditions. The healthy volunteers were free of present and past major and minor mental illness (affective disorder, schizophrenia, anxiety disorder, personality disorder, alcoholism, and illegal substance use disorders).

Experimental Design

After enrollment in the study, all patients received treatment as usual. Our treatment program is based on the Clinical Treatment Guideline for Schedule II Substance Users (Taiwan Ministry of Health and Welfare, 2012; https://dep.mohw.gov.tw/DOMHAOH/cp-4097-43400-107.html). Our approach is a multicomponent treatment approach and has been organized into a standard treatment protocol that includes outpatient visits, urine amphetamine tests, supportive psychotherapy, health education, case management, and service linkages, and the administration of medications was recorded. Group psychotherapy will also be given with a total of 12 sessions weekly using a cognitive behavioral therapeutic approach. All participants completed the Childhood Trauma Question-naire (CTQ) [29] to assess their adverse childhood experiences. FC was examined during the resting state using functional magnetic resonance imaging. For detailed information on the CTQ and image acquisition, please refer to our published papers [30-32].

Image Preprocessing

Preprocessing was performed using the DPARSFA toolbox V5.1 (State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University) with Statistical Parametrical Mapping 12 (SPM12, Wellcome Trust Centre for Neuroimaging; http://www.fil.ion.ucl.ac.uk/spm) in MATLAB 2016a (MathWorks Inc.). All functional images were subjected to slice timing, realignment for head-motion correction, coregistration against each individual’s anatomical image, segmentation, and normalization against the International Consortium for Brain Mapping (ICBM) space template of East Asian brains. Subjects with head motion of any volume greater than 2 mm or 2° were excluded from further processing. The images were resampled to an isotropic 2 mm3 voxel size during the normalization step and then spatially smoothed using a 3D Gaussian kernel of 6 mm full width at half maximum. Averaged time courses of the following nuisance variables or confounding artifacts were regressed out: (1) the 6 head movement parameters computed based on rigid body translation and rotation during realignment in SPM12, (2) the mean signal within the lateral ventricles, and (3) the mean signal within a deep white matter region (centrum ovale). The cerebrospinal fluid and white matter signals are thought to reflect fluctuations in nonspecific regional correlations. The time series was temporally band-pass filtered (0.01−0.1 Hz) to extract the low-frequency oscillations associated with spontaneous neuronal activity. Linear scrubbing was performed with a Jenkinson frame-wise displacement (FD) threshold of 0.35 mm, censoring the bad TR along with the previous and next TRs adjacent to it, excluding data with mean head motion greater than 0.5 mm across the entire scanning time, and discarding data if the number of total censored TRs exceeded 20% of the total TRs.

Definition of Default-mode Network Seeds and Default-mode Network Functional Connectivity Maps

Four a priori seeds (6-mm radius) of the DMN, centered at Montreal Neurological Institute (MNI) coordinates (0, 44, −8) (anterior DMN), (−39, −77, 33) (left-lateral DMN), (47, −67, 29) (right-lateral DMN), and (1, −61, 38) (posterior DMN), were identified based on the published literature [21,33]. The mean time-series activity in the seed region of each subject was extracted, and the correlation-like value between two respective mean time series was computed. The DMN-seeded FC maps were then generated. Each individual-level FC map obtained was then converted into a z-map using Fisher’s r-to-z transformation for second-level group analyses.

Statistical Analyses

The mediation analyses were performed using R software (open source & version 4.2.2), while other analyses were conducted using SPSS Statistics 20.0 (SPSS Inc.). Patients with missing data were excluded from the calculations, and the numbers of participants with missing data are indicated in the footnote of each table. The results were considered significant at p < 0.05 (two-tailed). A two-sample ttest (or a Mann–Whitney Utest, if the sample was not distributed in a Gaussian manner) was conducted to examine between-group (patients vs. healthy controls) differences in demographic characteristics and CTQ scores. A chi-square test was conducted to examine between-group differences in habitual smoking.

Image Analyses

A regression analysis was performed to determine the correlations between the DMN-seeded FC and CTQ score using SPM 12 (Wellcome Trust Centre for Neuroimaging; https://www.fil.ion.ucl.ac.uk/spm/). We entered the demeaned (in SPM) value of total CTQ as a regressor to identify brain regions with either positive or negative correlations with the DMN-seeded FC in the patients or healthy controls. To further determine whether different dimen-sions of adverse childhood experience were associated with different brain regions, we used subscales of the CTQ as regressors. Significance was thresholded at the uncorrected voxel-level p = 0.001, followed by the cluster-level familywise error rate (FWER)-corrected at p = 0.05 for whole-brain multiple comparisons.

A two-sample independent ttest was employed to analyze the FC maps. Statistical maps were computed to identify changes in the DMN-seeded FC for between-group comparisons. Significance was thresholded at the uncorrected voxel-level p = 0.001, followed by the cluster-level FWER-corrected p = 0.05.

To further present the regression results in scatterplots, we extracted the FC value in each brain region (radius = 3 mm). The extracted correlation coefficients (r) and pvalues from brain-behavior correlations will be inflated and are therefore not provided. To display 3D imaging, we used MRIcroGL for 3D rendering (Department of Psychology, University of South Carolina; http://www.mccauslandcenter.sc.edu/mricrogl).

Mediation Analyses and Models

In our mediation model, predictors were CTQ scores, mediators were FC and outcome was the symptom load of MUD. The operational definition of symptoms load of MUD was defined as the number of diagnostic criteria met for MUD [34]. The conceptual model is presented in Figure 1A.

Following the rationale provided by O’Rourke and Vazquez [35], for a simple mediation model with a count outcome, we modeled FC using linear regression and the symptom load of MUD using Poisson distributions and then multiplied coefficients from the two models to produce estimated indirect effects. To test the significance for indirect effects, we used not only Joint Significance test but also estimated 95% bias-corrected (BCa) bootstrap confidence intervals (CIs) with 10,000 bootstrap for conditional indirect effect estimates for three representative levels of CTQ scores: the low (mean −1 standard deviation [SD]), the mean and the high (mean +1SD) level at the mean of FC [36].

RESULTS

Demographic and Clinical Data

Twenty-seven male MUD patients and 27 age- and sex- matched healthy controls were enrolled in this study. There were differences between groups in terms of years of education and habitual smoking (Table 1); therefore, age, years of education, and habitual smoking were regressed out as covariates of noninterest in the following image analyses and mediation analyses.

The MUD patients in this study scored 43.9 ± 13.1 on the CTQ, and 18 (66.7%) had adverse childhood experiences (Table 1). Furthermore, the MUD patients, in comparison with the healthy controls, demonstrated higher scores and presence percentages on the subscales of sexual abuse and emotional abuse (Table 1). As there were no between-group differences in the subscales of physical abuse, emotional neglect, and physical neglect, we did not perform further correlation analysis on these subscales.

CTQ and DMN Circuitry Connectivity

Ten controls did not complete the CTQ and were excluded from this calculation. Finally, 27 male MUD patients and 17 healthy controls were included in the regression analysis. There were negative correlations between adverse childhood experiences and DMN FC, including anterior DMN-posterior parietal cortex (PPC) FC (Figs. 2A, 3A), right-lateral DMN-PPC FC (Figs. 2B, 3B) and left-lateral DMN-anterior cingulate cortex (ACC) FC (Figs. 2C, 3C) in healthy controls, and posterior DMN- PPC FC (Fig. 3D, Table 2) in MUD patients. Notably, both healthy controls and MUD patients exhibited a negative correlation between adverse childhood experiences and left-lateral DMN-visual cortex FC (Table 2), which was also correlated with emotional abuse scores in healthy controls (Table 2). The statistical brain images (i.e., t-maps) which are derived from each contrast were provided in the website (https://osf.io/8navf/).

In the MUD patients alone, there were significant negative correlations between their adverse childhood experiences and DMN-dorsolateral prefrontal cortex (dlPFC) FC (Fig. 2D) and between sexual abuse scores and DMN- SPL FC (Table 2). Therefore, we extracted the FC value in these brain regions (radius = 3 mm) and performed mediation analyses with clinical data as the outcome (i.e., the number of diagnostic criteria met for MUD).

Mediation Analysis

Twenty-seven male MUD patients were included in the mediation analysis. The mediation analysis results are presented in Figure 1B and Table 3. The Joint significant test and 95% BCa bootstrapped CIs of conditional indirect effects at representative levels of CTQ scores were consistent in supporting our mediation model. Adverse childhood experiences affected the symptom load of MUD through connectivity between the DMN-dlPFC (Table 3). Total CTQ scores had a significant positive direct effect on the number of diagnostic criteria met for MUD, while they exerted a significant negative indirect effect through connectivity between the DMN-dlPFC. Moreover, the conditional indirect effects increased in magnitude as CTQ scores elevated (Table 3).

Between-group Differences in Functional Connectivity

Twenty-seven male MUD patients and 27 healthy controls were included in the image analysis. The MUD patients, in comparison with the healthy controls, exhibited functional hypoconnectivity within the DMN, including anterior DMN-PCC FC (Fig. 4B), posterior DMN-ACC FC (Fig. 4C), and left-lateral DMN-ACC FC (Fig. 4D). Notably, there was hypoconnectivity between the anterior DMN to the frontal eye fields, medial prefrontal cortex, parahippocampus gyrus, and dlPFC (Fig. 4B, Table 4). The statistical brain images (i.e., t-maps) which are derived from each contrast were provided in the website (https:// osf.io/8navf/).

DISCUSSION

Adverse childhood experiences are reported by more than half of the study participants in the CDC-Kaiser Permanente Adverse Childhood Experiences study [37], but the prevalence is even higher (approximately 69%) in patients with substance use disorder [38]. Similarly, 47.1% of healthy controls and 66.7% of MUD patients in this study had adverse childhood experiences (Table 1). Our results showed that there were significant differences in the scores of sexual abuse and emotional abuse (Table 1), which was also consistent with previous studies: the higher the scores of sexual abuse and emotional abuse, the earlier the age of first-time MUD use [5].

Furthermore, negative correlations between adverse childhood experiences and within-DMN FC were observed in both healthy controls and MUD patients (Figs. 2, 3, and Table 2), while within-DMN FC was significantly altered in MUD patients (Fig. 4, Table 4). The mediation analysis showed that the detrimental effects of adverse childhood experiences on MUD patients may be attenuated through DMN-ECN FC independent of age, years of education and habitual smoking (Fig. 1). Moreover, such attenuation was greater in those with more adverse childhood experience (Table 3).

In the healthy controls, there were significant negative correlations between adverse childhood experiences and within-DMN FC (Fig. 2A−2C), indicating insufficient resilience. Our rationale is supported by previous studies showing that altered DMN-related FC was observed in trauma-exposed youth [15] and that within-DMN FC may represent a biomarker of resilience [39]. Nevertheless, hyperactivity and hyperconnectivity of the DMN are also generally associated with poor resilience; therefore, both under- and over-stimulation of DMN could result in vulnerability. Although such an association was insignificant in MUD patients (Fig. 2, Table 2), there was significant functional hypo-FC within the DMN in MUD patients (Fig. 4); therefore, a possible explanation is the floor effect of low FC within the DMN in MUD patients (Fig. 2A). Collectively, both healthy controls and MUD patients exhibited altered FC within the DMN, which was influenced by adverse childhood experiences. The MUD patients who had more childhood trauma (Table 1) exhibited hypo-FC within the DMN.

In the MUD patients, their adverse childhood experiences were positively correlated with the symptom load of MUD; nevertheless, DMN-ECN FC may ameliorate such effects and result in decreased symptoms load of MUD (Fig. 1B, Table 3). In summary, the FC alterations observed in MUD patients may result from both damage to and protective adaptation of the brain following exposure to adverse childhood experiences; the former occurs within-DMN (Fig. 4C, 4D) and the latter occurs in a cross-network manner (Fig. 4B). Our rationale is corroborated by the observations in PTSD patients, who also showed decreased FC within-DMN [40]. Additionally, increased posterior DMN-dlPFC FC was reported in PTSD patients during executive functioning tasks, implying difficulties in disengaging the DMN and the ECN [41]. These FC patterns may contribute to difficulties in shifting attention to external cognitive demands, but an inclination towards rumination in internally focused processing [42]. In contrast, we found decreased posterior DMN-dlPFC FC in MUD patients, connoting less internally focused self-referential processing and an ability to modulate executive control processing, which may attenuate the severity of symptoms in these patients.

Given that DMN FC is as a biomarker for predicting clinical outcomes in craving and relapse [8], posterior DMN-ECN FC could be a potential treatment target for neuromodulation in MUD patients. Previous studies in PTSD using trauma-focused cognitive behavioral therapy have found that longitudinal improvement in PTSD was associated with decreasing activation in posterior DMN [43]. Recently, the application of real-time fMRI neurofeedback to downregulate the posterior DMN has also found to reduce acute reliving symptoms in both PTSD patients and healthy controls [44]. As transcranial magnetic stimulation (TMS) of the DMN and ECN has been used to alleviate depression [45], our data provide insight into the potential use of internal or external neuromodulation techniques (i.e., neurofeedback or TMS) to strengthen resilience and adaptation to early adversity in MUD patients.

Interestingly, other studies found within-SN FC mediating the association between adverse childhood experiences and lower problematic substance use [46]. In addition, the disappearance of the correlation between resilience and DMN-SN cross-interaction may suggest a protective role of resilience for brain functioning [47]. Therefore, these neurocircuits may act in tandem, playing a protective role and mediating adverse childhood ex-periences. As adverse experiences early in life are associated with a higher risk of developing a substance use disorder later in life [48], understanding such neural dynamics underlying positive adaptation to early adversity will aid in the development of interventions that focus on strengthening resilience rather than mitigating already- present psychological problems [49]. Although studies have found associations between childhood trauma and altered FC within specific functional networks of the SN [15] and ECN [16], our study was limited by the sample size, and seeds from SN and ECN were not included in the analysis.

Notably, among different subcomponents of adverse childhood experiences, MUD patients exhibited the most profound differences in sexual abuse (Tables 1, 2). The MUD patients exhibited significantly higher scores of sexual abuse, and there were negative correlations between sexual abuse scores and FC encompassing the within-DMN and DMN-ECN. Our data are corroborated by previous review articles showing that the gray matter volume over the frontal cortex was also negatively correlated with sexual abuse scores [18]. Furthermore, our findings support a model in which specific trauma subtypes may lead to specific alterations in the brain [18].

Our study had limitations in its cross-sectional design, making it impossible to determine whether the abnormalities in DMN-related FC are causes or effects of the MUD. The sample size was not particularly large. Regarding the mediation analysis, small sample sizes would be one of the limitations of this manuscript. Although the MUD patients were undergoing treatment or had underlining psychiatric comorbidities that could potentially change brain circuitry, our subjects underwent fMRI at the beginning of the treatment program with minimal medication use. Furthermore, there were no more than 25% of the patients who shared the same psychiatric comorbidities. Specifi-cally, there were generalized anxiety disorder (n = 6), dysthymic disorder (n = 3), bipolar disorder (n = 3), major depressive disorder (n = 2), antisocial personality disorder (n = 2), 3,4-methylenedioxymethamphetamine (MDMA) use disorder (n = 2), ketamine use disorder (n = 1), and psychotic disorder (n = 1). We did not exclude MUD patients with comorbidities due to their high prevalence in this population and in our sample (70.3%); therefore, our findings could indicate a characteristic of psychiatric illness broadly, but are less likely to be confounded by a specifical psychiatric illness other than MUD. The MUD patients were abstinent prior to the MRI scan, as verified by urine tests; however, 3 MUD patients showed positive results in their urine tests. Although we could not rule out the acute effect of methamphetamine or other substances, the urine test results in this study were much lower than in other neuroimaging studies [50]. Additionally, our results may not be generalizable to female patients, as the MUD patients in this study comprised only men.

The number of diagnostic criteria met for MUD is part of the severity in methamphetamine use disorder [34], however, other aspects including duration, onset of age, frequency, urge questionnaire, and withdrawal question-naire, etc. may also be considered. Furthermore, the number of diagnostic criteria met for MUD was significantly associated with the MUD related problems with the highest beta value [34]. According to the DSM-5, the diagnostic severity is also based on the number of presence of symptoms (e.g., mild is presence of 2−3 symptoms). Therefore, we used the symptom load to define the MUD severity in current study. Although only presenting the diagnostic severity of MUD is a limitation, our study shed light on the association between childhood trauma, DMN, and MUD. Our preliminary report invites future studies with larger sample sizes for verification.

To the best of our knowledge, the current work is the first study to examine DMN FC in relationship with adverse childhood experiences and MUD use. We found negative correlations between adverse childhood experiences and within-DMN FC in both healthy controls and MUD patients. DMN-ECN connectivity may attenuate the effects of childhood trauma on MUD symptom load. Our findings provide one plausible neural underpinning of the relationship between adverse childhood experiences and MUD and may be generalized to other substance use disorders.

ACKNOWLEDGEMENTS

The authors thank all the participants in this study and extend particular appreciation to Chien Ting Lin from National Cheng Kung University for his technical assistance. The authors would like to thank Miss Chien-Yu Tseng, Kuan Yu Chen, Hung-Yi Chan, Rou An Chen, and Yu Ting Hung for their assistance in the preparation of the manuscript.

Conflicts of Interest

No potential conflict of interest relevant to this article was reported.

Author Contributions

TYW designed the study and wrote the protocol. TYW, TYT, HHT, YKY, and PSC recruited participants. SYW and THT analyzed the data. SYW wrote the first draft of the manuscript. YY and TZ reviewed the literature and contributed to the discussion. All authors contributed to and reviewed the final version of the manuscript.

Figures
Fig. 1. Visualization of mediation model of childhood trauma, functional connectivity and the number of diagnostic criteria met for methamphetamine use disorder. (A) In our mediation model, the predictor was scores of Childhood Trauma Questionnaire, the mediator was standardized functional connectivity and outcome was the number of diagnostic criteria met for methamphetamine use disorder. We calculated the product of paths a and b, which represented indirect effects of adverse childhood experience on diagnostic criteria met for methamphetamine use disorder. (B) The simple mediation model of childhood adverse events predicting the number of diagnostic criteria met for methamphetamine use disorder through functional connectivity between the posterior default-mode network and dorsolateral prefrontal cortex. Age, years of education and habitual smoking were regressed out as covariates of non-interest.
*p < 0.05, **p < 0.01, ***p < 0.001.
Fig. 2. Default-mode network functional connectivity covaries with adverse childhood experiences. Functional connectivity (FC) within the default-mode network (DMN) was negatively correlated with adverse childhood experiences in healthy controls, including (A) anterior DMN to left-lateral DMN, (B) right-lateral DMN to left-lateral DMN, and (C) left-lateral DMN to anterior DMN. In the methamphetamine use disorder (MUD) patients alone, there was a significant correlation (D) between adverse childhood experiences and posterior DMN-dorsolateral prefrontal cortex (dlPFC) FC. The coordinates of the peak voxel are presented in Table 2. The extracted correlation coefficients (r) and pvalues from brain-behavior correlations will be inflated and are therefore not provided. Significance was thresholded at the uncorrected voxel level p = 0.001, followed by the familywise error rate-corrected cluster level p = 0.05.
PPC, posterior parietal cortex; ACC, anterior cingulate cortex; CTQ, Childhood Trauma Questionnaire.
Fig. 3. Brain regions whose functional connectivity with default-mode network seeds is correlated with the childhood trauma questionnaire in healthy controls or methamphetamine use disorder patients. Functional connectivity (FC) within the default-mode network (DMN) was negatively correlated with adverse childhood experiences in healthy controls, including (A) anterior DMN to left-lateral DMN, (B) right-lateral DMN to left-lateral DMN, and (C) left-lateral DMN to anterior DMN. In the methamphetamine use disorder patients alone, there were significant correlations (D) between the total Childhood Trauma Questionnaire results and posterior DMN-dorsolateral prefrontal cortex FC. Significance was thresholded at the uncorrected voxel level p = 0.001, followed by the familywise error rate-corrected cluster level p = 0.05. The color bar denotes the tscores. Figures are displayed according to neurological convention (left = left).
Fig. 4. The methamphetamine use disorder patients exhibited dysconnectivity within the default-mode network and cross-network. (A) Default- mode network (DMN) seeds (green dot) including the anterior DMN, posterior DMN, right-lateral DMN, and left-lateral DMN. There was hypoconnectivity in the (B) anterior DMN-posterior DMN, (C) posterior DMN-anterior DMN, and (D) left-lateral DMN-anterior DMN functional connectivity in the methamphetamine use disorder patients. Each region’s coordinates are listed in Table 4. Significance was thresholded at the uncorrected voxel level p = 0.001, followed by the familywise error rate-corrected cluster level p = 0.05. The color bar denotes the tscores. Figures are displayed according to neurological convention (left = left).
Tables

Demographic data and baseline information

Variable MUD (n = 27) Controls (n = 27) pvalue
Age (yr) 31.9 ± 7.5 32.2 ± 9.1 0.897
Education (yr) 13.5 ± 2.7 16.5 ± 2.7 < 0.001*
Smoking 18 (66.7) 4 (14.8) < 0.001*
Criteria numbers 6.0 ± 2.6 -
CTQa
Emotional abuse 8.7 ± 3.5 6.5 ± 1.9 0.009*
Physical abuse 6.9 ± 2.5 7.2 ± 3.1 0.720
Sexual abuse 7.0 ± 3.4 5.1 ± 0.5 0.008*
Emotional neglect 12.7 ± 5.5 10.4 ± 3.6 0.102
Physical neglect 8.6 ± 3.4 7.1 ± 1.7 0.054
Total 43.9 ± 13.1 36.3 ± 6.5 0.014*
CTQ presencea
Emotional abuse 11 (40.7) 3 (17.6) 0.109
Physical abuse 6 (22.2) 6 (35.3) 0.343
Sexual abuse 9 (33.3) 1 (5.9) 0.034*
Emotional neglect 18 (66.7) 8 (47.1) 0.198
Physical neglect 16 (59.3) 5 (29.4) 0.054
Total 18 (66.7) 8 (47.1) 0.198

Values are presented as mean ± standard deviation or number (%).

MUD, methamphetamine use disorder; CTQ, Childhood Trauma Questionnaire.

aTen controls did not complete the CTQ and were excluded from this calculation.

*p < 0.05

Functional connectivity of the default-mode network seeds co-varying with Childhood Trauma Questionnaire

Group Direction Seed Region BA Cluster pvalue tscore Peak coordinate

x y z
CTQ total
Controls Negative Anterior DMN Left PPC 39 200 0.030 3.58 −44 −54 28
Controls Negative Left-lateral DMN ACC 32 217 0.021 4.24 −6 48 8
Left-lateral DMN Visual cortex 19 318 0.003 5.93 −30 −88 30
Controls Negative Right-lateral DMN Left PPC 39 201 0.030 3.50 −26 −70 34
Right-lateral DMN Fusiform 37 317 0.003 4.77 50 −62 −16
MUD Negative Left-lateral DMN Visual cortex 19 499 < 0.001 4.22 −26 −50 −4
MUD Negative Posterior DMN Left PPC 39 198 0.037 4.51 −36 −52 36
Posterior DMN Left dlPFC 46 311 0.005 4.32 −40 40 4
Sexual abuse
MUD Negative Posterior DMN Right SPL 7 639 < 0.001 5.58 34 −54 48
Posterior DMN Left dlPFC 9 332 0.003 3.88 −42 22 26
Posterior DMN Left PPC 39 934 < 0.001 4.28 −44 −46 3
Posterior DMN Left premotor 6 936 < 0.001 4.72 −32 −2 52
Emotional abuse
Controls Negative Right-lateral DMN Visual cortex - 205 0.029 5.23 32 −74 2
Controls Negative Posterior DMN Cerebellum - 708 < 0.001 5.19 −6 −54 -8

ACC, anterior cingulate cortex; MUD, methamphetamine use disorder; BA, brodmann area; dlPFC, dorsolateral prefrontal cortex; DMN, default-mode network; PPC, posterior parietal cortex; SPL, superior parietal lobe; CTQ, Childhood Trauma Questionnaire.

As there was no between-group differences in the subscale of physical abuse, emotional neglect and physical neglect, we did not perform further correlation analysis in these subscales. Ten controls did not complete the CTQ and were excluded from this calculation. Peak coordinates refer to the Montreal Neurological Institute space. Significance was thresholded at the uncorrected voxel level p = 0.001, followed by the familywise error rate-corrected cluster level p = 0.05.

Mediation analysis of adverse childhood experiences, default-mode network functional connectivity, and the number of diagnostic criteria met for methamphetamine use disorder

CTQ Functional connectivity Mediation model

Unstandardized estimates (SE) Boostrapped indirect effect


a b c c' ab
Total scores pDMN-dlPFC −0.016 (0.004)*** 0.819 (0.386)* 0.011 (0.006) 0.025 (0.009)** −0.013
Conditional indirect effects (SE)

ab 95% BCa CI
High (mean +1SD) −0.098 (0.140) −0.5635, −0.0130
Mean −0.071 (0.097) −0.3926, −0.0114
Low (mean −1SD) −0.052 (−0.066) −0.2689, −0.0091

BCa CI, bias-corrected bootstrap confidence intervals; CTQ, Childhood Trauma Questionnaire; dlPFC, dorsolateral prefrontal cortex; pDMN, posterior default-mode network; SD, standard deviation; SE, standard error.

To test the significance of mediation with a count outcome, we used not only Joint Significance test but also estimated 95% BCa bootstrapped CI of conditional indirect effect (ab) for representative levels of predictor at mean of mediator: the low (mean −1SD), the mean and the high (mean +1SD). Joint Significance test required effects of predictors on the mediators (a) and effects of mediators on outcomes controlling for predictors (b) were statistically significant to support the claim of mediation. The 95% CI suggested a significant conditional indirect effects if it did not include zero. In this mediation model, the predictor was CTQ score, the mediator was functional connectivity and the outcome was the number of diagnostic criteria met for methamphetamine use disorder. Age, years of education, and habitual smoking were covariates.

*p < 0.05, **p < 0.01, ***p < 0.001.

Peak Montreal Neurological Institute coordinates for the regions exhibiting significant resting-state functional connectivity with the default-mode network seeds with between-group differences

Contrast Seed Region BA Cluster pvalue tscore Peak coordinate

x y z
MUD < controls Anterior DMN PCC 23 1,058 < 0.001 4.60 8 −46 24
Anterior DMN Left FEF 8 286 < 0.001 4.73 −20 44 44
Anterior DMN Left dlPFC 9 - - 3.41 −26 56 36
Anterior DMN mPFC 10 229 < 0.001 4.88 −6 54 4
Anterior DMN Parahippocampus - 191 < 0.001 4.24 28 −28 −18
MUD < controls Posterior DMN ACC 24 444 0.001 3.74 0 34 6
MUD < controls Left-lateral DMN ACC 32 222 0.022 4.40 8 30 −8

ACC, anterior cingulate cortex; MUD, methamphetamine use disorder; BA, brodmann area; dlPFC, dorsolateral prefrontal cortex; DMN, default-mode network; FEF, frontal eye fields; mPFC, medial prefrontal cortex; PCC, posterior cingulate cortex; MNI, Montreal Neurological Institute.

Peak coordinates refer to the MNI space. Significance was thresholded at the uncorrected voxel level p = 0.001, followed by the familywise error rate-corrected cluster level p = 0.05. No higher functional connectivity was found in the MUD.

References
  1. United Nations Office on Drugs and Crime. World Drug Report 2021 [Internet]. United Nations Office on Drugs and Crime; 2021 [cited at 2023 Mar 29]. https://www.unodc.org/unodc/en/data-and-analysis/wdr2021.html.
  2. Gonzales R, Mooney L, Rawson RA. The methamphetamine problem in the United States. Annu Rev Public Health 2010;31:385-398.
    Pubmed KoreaMed CrossRef
  3. Mirhashem R, Allen HC, Adams ZW, van Stolk-Cooke K, Legrand A, Price M. The intervening role of urgency on the association between childhood maltreatment, PTSD, and substance-related problems. Addict Behav 2017;69:98-103.
    Pubmed KoreaMed CrossRef
  4. Santo T Jr, Campbell G, Gisev N, Tran LT, Colledge S, Di Tanna GL, et al. Prevalence of childhood maltreatment among people with opioid use disorder: A systematic review and meta-analysis. Drug Alcohol Depend 2021;219:108459.
    Pubmed KoreaMed CrossRef
  5. Huang C, Yuan Q, Zhang L, Wang L, Cui S, Zhang K, et al. Associations between childhood trauma and the age of first-time drug use in methamphetamine-dependent patients. Front Psychiatry 2021;12:658205.
    Pubmed KoreaMed CrossRef
  6. Norman RE, Byambaa M, De R, Butchart A, Scott J, Vos T. The long-term health consequences of child physical abuse, emotional abuse, and neglect: A systematic review and meta- analysis. PLoS Med 2012;9:e1001349.
    Pubmed KoreaMed CrossRef
  7. Hadland SE, Werb D, Kerr T, Fu E, Wang H, Montaner JS, et al. Childhood sexual abuse and risk for initiating injection drug use: A prospective cohort study. Prev Med 2012;55:500-504.
    Pubmed KoreaMed CrossRef
  8. Zhang R, Volkow ND. Brain default-mode network dysfunction in addiction. Neuroimage 2019;200:313-331.
    Pubmed CrossRef
  9. Yeshurun Y, Nguyen M, Hasson U. The default mode network: Where the idiosyncratic self meets the shared social world. Nat Rev Neurosci 2021;22:181-192.
    Pubmed KoreaMed CrossRef
  10. Raichle ME. The brain's default mode network. Annu Rev Neurosci 2015;38:433-447.
    Pubmed CrossRef
  11. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 2005;102:9673-9678.
    Pubmed KoreaMed CrossRef
  12. Shannon BJ, Raichle ME, Snyder AZ, Fair DA, Mills KL, Zhang D, et al. Premotor functional connectivity predicts impulsivity in juvenile offenders. Proc Natl Acad Sci U S A 2011;108:11241-11245.
    Pubmed KoreaMed CrossRef
  13. Liang X, He Y, Salmeron BJ, Gu H, Stein EA, Yang Y. Interactions between the salience and default-mode networks are disrupted in cocaine addiction. J Neurosci 2015;35:8081-8090.
    Pubmed KoreaMed CrossRef
  14. Teicher MH, Samson JA, Anderson CM, Ohashi K. The effects of childhood maltreatment on brain structure, function and connectivity. Nat Rev Neurosci 2016;17:652-666.
    Pubmed CrossRef
  15. Marusak HA, Etkin A, Thomason ME. Disrupted insula-based neural circuit organization and conflict interference in trauma-exposed youth. Neuroimage Clin 2015;8:516-525.
    Pubmed KoreaMed CrossRef
  16. Silveira S, Shah R, Nooner KB, Nagel BJ, Tapert SF, de Bellis MD, et al. Impact of childhood trauma on executive function in adolescence-mediating functional brain networks and prediction of high-risk drinking. Biol Psychiatry Cogn Neurosci Neuroimaging 2020;5:499-509.
    Pubmed KoreaMed CrossRef
  17. Zhang J, Zhao T, Zhang J, Zhang Z, Li H, Cheng B, et al. Prediction of childhood maltreatment and subtypes with personalized functional connectome of large-scale brain networks. Hum Brain Mapp 2022;43:4710-4721.
    Pubmed KoreaMed CrossRef
  18. Cassiers LLM, Sabbe BGC, Schmaal L, Veltman DJ, Penninx BWJH, Van Den Eede F. Structural and functional brain abnormalities associated with exposure to different childhood trauma subtypes: A systematic review of neuroimaging findings. Front Psychiatry 2018;9:329.
    Pubmed KoreaMed CrossRef
  19. Wang L, Dai Z, Peng H, Tan L, Ding Y, He Z, et al. Overlapping and segregated resting-state functional connectivity in patients with major depressive disorder with and without childhood neglect. Hum Brain Mapp 2014;35:1154-1166.
    Pubmed KoreaMed CrossRef
  20. Bluhm RL, Williamson PC, Osuch EA, Frewen PA, Stevens TK, Boksman K, et al. Alterations in default network connectivity in posttraumatic stress disorder related to early-life trauma. J Psychiatry Neurosci 2009;34:187-194.
  21. Dauvermann MR, Mothersill D, Rokita KI, King S, Holleran L, Kane R, et al. Changes in default-mode network associated with childhood trauma in schizophrenia. Schizophr Bull 2021;47:1482-1494.
    Pubmed KoreaMed CrossRef
  22. Wang X, Liu Q, Fan J, Gao F, Xia J, Liu X, et al. Decreased functional coupling within default mode network in major depressive disorder with childhood trauma. J Psychiatr Res 2022;154:61-70.
    Pubmed CrossRef
  23. Tursich M, Ros T, Frewen PA, Kluetsch RC, Calhoun VD, Lanius RA. Distinct intrinsic network connectivity patterns of post-traumatic stress disorder symptom clusters. Acta Psychiatr Scand 2015;132:29-38.
    Pubmed CrossRef
  24. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 5th edition (DSM-5). American Psychiatric Association;2013.
    CrossRef
  25. Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59 Suppl 20:22-33; quiz 34-57.
  26. Ritchie K, Artero S, Beluche I, Ancelin ML, Mann A, Dupuy AM, et al. Prevalence of DSM-IV psychiatric disorder in the French elderly population. Br J Psychiatry 2004;184:147-152.
    Pubmed CrossRef
  27. Chie WC, Chang KJ, Huang CS, Kuo WH. Quality of life of breast cancer patients in Taiwan: Validation of the Taiwan Chinese version of the EORTC QLQ-C30 and EORTC QLQ- BR23. Psychooncology 2003;12:729-735.
    Pubmed CrossRef
  28. Lung FW, Cheng CT, Chang WT, Shu BC. Anxiety and mood disorder in young males with mitral valve prolapse. J Multidiscip Healthc 2008;1:89-92.
    Pubmed KoreaMed CrossRef
  29. Bernstein DP, Ahluvalia T, Pogge D, Handelsman L. Validity of the childhood trauma questionnaire in an adolescent psychiatric population. J Am Acad Child Adolesc Psychiatry 1997;36:340-348.
    Pubmed CrossRef
  30. Tsai TY, Tseng HH, Chi MH, Chang HH, Wu CK, Yang YK, et al. The interaction of oxytocin and social support, loneliness, and cortisol level in major depression. Clin Psychopharmacol Neurosci 2019;17:487-494.
    Pubmed KoreaMed CrossRef
  31. Wei SY, Tseng HH, Chang HH, Lu TH, Chang WH, Chiu NT, et al. Dysregulation of oxytocin and dopamine in the corticostriatal circuitry in bipolar II disorder. Transl Psychiatry 2020;10:281.
    Pubmed KoreaMed CrossRef
  32. Hsieh YT, Wu R, Tseng HH, Wei SY, Huang MC, Chang HH, et al. Childhood neglect is associated with corticostriatal circuit dysfunction in bipolar disorder adults. Psychiatry Res 2021;295:113550.
    Pubmed CrossRef
  33. Felger JC, Li Z, Haroon E, Woolwine BJ, Jung MY, Hu X, et al. Inflammation is associated with decreased functional connectivity within corticostriatal reward circuitry in depression. Mol Psychiatry 2016;21:1358-1365.
    Pubmed KoreaMed CrossRef
  34. Moallem NR, Courtney KE, Ray LA. The relationship between impulsivity and methamphetamine use severity in a community sample. Drug Alcohol Depend 2018;187:1-7.
    Pubmed KoreaMed CrossRef
  35. O'Rourke HP, Vazquez E. Mediation analysis with zero-inflated substance use outcomes: Challenges and recommenda-tions. Addict Behav 2019;94:16-25.
    Pubmed CrossRef
  36. Geldhof GJ, Anthony KP, Selig JP, Mendez-Luck CA. Accom-modating binary and count variables in mediation: A case for conditional indirect effects. Int J Behav Dev 2018;42:300-308.
    CrossRef
  37. Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev Med 1998;14:245-258.
    Pubmed CrossRef
  38. Liebschutz J, Savetsky JB, Saitz R, Horton NJ, Lloyd-Travaglini C, Samet JH. The relationship between sexual and physical abuse and substance abuse consequences. J Subst Abuse Treat 2002;22:121-128.
    Pubmed CrossRef
  39. Miyagi T, Oishi N, Kobayashi K, Ueno T, Yoshimura S, Murai T, et al. Psychological resilience is correlated with dynamic changes in functional connectivity within the default mode network during a cognitive task. Sci Rep 2020;10:17760.
    Pubmed KoreaMed CrossRef
  40. Akiki TJ, Averill CL, Wrocklage KM, Scott JC, Averill LA, Schweinsburg B, et al. Default mode network abnormalities in posttraumatic stress disorder: A novel network-restricted topology approach. Neuroimage 2018;176:489-498.
    Pubmed KoreaMed CrossRef
  41. Daniels JK, McFarlane AC, Bluhm RL, Moores KA, Clark CR, Shaw ME, et al. Switching between executive and default mode networks in posttraumatic stress disorder: Alterations in functional connectivity. J Psychiatry Neurosci 2010;35:258-266.
    Pubmed KoreaMed CrossRef
  42. Aupperle RL, Stillman AN, Simmons AN, Flagan T, Allard CB, Thorp SR, et al. Intimate partner violence PTSD and neural correlates of inhibition. J Trauma Stress 2016;29:33-40.
    Pubmed CrossRef
  43. Garrett A, Cohen JA, Zack S, Carrion V, Jo B, Blader J, et al. Longitudinal changes in brain function associated with symptom improvement in youth with PTSD. J Psychiatr Res 2019;114:161-169.
    Pubmed KoreaMed CrossRef
  44. Nicholson AA, Rabellino D, Densmore M, Frewen PA, Steryl D, Scharnowski F, et al. Differential mechanisms of posterior cingulate cortex downregulation and symptom decreases in posttraumatic stress disorder and healthy individuals using real-time fMRI neurofeedback. Brain Behav 2022;12:e2441.
    Pubmed KoreaMed CrossRef
  45. Liston C, Chen AC, Zebley BD, Drysdale AT, Gordon R, Leuchter B, et al. Default mode network mechanisms of transcranial magnetic stimulation in depression. Biol Psychiatry 2014;76:517-526.
    Pubmed KoreaMed CrossRef
  46. Rakesh D, Allen NB, Whittle S. Longitudinal changes in within-salience network functional connectivity mediate the relationship between childhood abuse and neglect, and mental health during adolescence. Psychol Med 2023;53:1552-1564.
    Pubmed CrossRef
  47. Brunetti M, Marzetti L, Sepede G, Zappasodi F, Pizzella V, Sarchione F, et al. Resilience and cross-network connectivity: A neural model for post-trauma survival. Prog Neuropsycho-pharmacol Biol Psychiatry 2017;77:110-119.
    Pubmed CrossRef
  48. Choi NG, DiNitto DM, Marti CN, Choi BY. Association of adverse childhood experiences with lifetime mental and substance use disorders among men and women aged 50+ years. Int Psychogeriatr 2017;29:359-372.
    Pubmed CrossRef
  49. Iadipaolo AS, Marusak HA, Paulisin SM, Sala-Hamrick K, Crespo LM, Elrahal F, et al. Distinct neural correlates of trait resilience within core neurocognitive networks in at-risk children and adolescents. Neuroimage Clin 2018;20:24-34.
    Pubmed KoreaMed CrossRef
  50. Hu Y, Salmeron BJ, Gu H, Stein EA, Yang Y. Impaired functional connectivity within and between frontostriatal circuits and its association with compulsive drug use and trait impulsivity in cocaine addiction. JAMA Psychiatry 2015;72:584-592.
    Pubmed CrossRef


This Article

Close ✕


Cited By Articles

Author ORCID Information

Funding Information
  • Ministry of Science and Technology
      10.13039/501100004663
      MOST-107-2314-B-006-067, MOST-109-2314-B-006-056
  • National Cheng Kung University Hospital
      10.13039/501100004844
      NCKUH-11004011, NCKUH-11001003, NCKUH-11204032
  • Ministry of Health and Welfare
      10.13039/100008903
      MOHW107-TDU-B-211-123003, MOHW108-TDU-B-211-133003
  • Integrated Drug Addiction Treatment Center of the Jianan Psychiatric Center
     
     
  • National Institute on Drug Abuse
      10.13039/100000026
     

Services
Social Network Service

e-submission

Archives