Abnormal function of the posterior cingulate cortex in heroin addicted users during resting-state and drug-cue stimulation task
LI Qiang, YANG Wei-chuan, WANG Ya-rong, HUANG Yu-fang, LI Wei, ZHU Jia, ZHANG Yi, ZHAO Li-yan, QIN Wei, YUAN Kai, Karen M. von Deneen, WANG Wei and TIAN Jie
Department of Radiology, Tangdu Hospital, The Fourth Military Medical University, Xi’an, Shaanxi 710038, China (LI Q, YANG WC, WANG YR, LI W, ZHU J, WANG W and TIAN J)
Department of Radiology, 422 Military Hospital, Zhanjiang, Guangdong, 524000, China (HUANG YF)
National Institute on Drug Dependence, Peking University, Beijing 100083, China (ZHAO L)
Life Sciences Research Center, School of Life Sciences and Technology, Xidian University, Xi’an, Shaanxi 710071, China (ZHANG Y, QIN W, YUAN K, Karen M. von Deneen and TIAN J)
LI Qiang and YANG Wei-chuan contributed equally to this work.
Correspondence to: Prof. WANG Wei, Department of Radiology, Tangdu Hospital, The Fourth Military Medical University, NO.569 Xinsi Road, BaQiao District, Xi’an, Shaanxi 710038, China (Tel.: +86 29 84777933; fax: +86 29 84777933. E-mail addresses: m_d_Wang@hotmail.com); Prof. TIAN Jie, Life Science Research Center, School of Life Sciences and Technology, P.O. Box 97, Xidian University, No.2 TaiBai South Road, Xi’an, Shaanxi 710071, China (Tel.: +86 10 62527995; fax: +86 10 62527995. E-mail addresses: tian@ieee.org).
Abstract
Background Previous animal and neuroimaging studies have demonstrated that brain function in heroin addicted users has been impaired. However, the posterior cingulate cortex (PCC) has not received much attention. The purpose of this study was to investigate whether chronic heroin use is associated with craving-related changes in the functional connectivity of the PCC of heroin addicted users.
Methods Fourteen male adult chronic heroin users and fifteen age- and gender- matched healthy subjects participated in the present study. The participants underwent a resting-state functional magnetic resonance imaging (fMRI) scan and a cue-induced craving task fMRI scan. The activated PCC was identified in the cue-induced craving task by means of a group contrast test. Functional connectivity was analyzed based on resting-state fMRI data in order to determine the correlation between brain regions. The relationship between the connectivity of specific regions and heroin dependence was investigated.
Results The activation of the PCC, the bilateral anterior cingulate cortex, caudate, putamen, precuneus, and thalamus were significantly activated in the heroin group compared to the healthy group in the cue-induced craving task. The detectable functional connectivity of the heroin users was stronger between the PCC and bilateral insula, bilateral dorsal striatum, right inferior parietal lobule (IPL) and right supramarginal gyrus (P < 0.001) compared to that of the healthy subjects in the resting-state data analysis. The strength of the functional connectivity, both for the PCC-insula (r = 0.60, P < 0.05) and PCC-striatum (r = 0.58, P < 0.05), was positively correlated with the duration of heroin use.
Conclusions The altered functional connectivity patterns in the PCC-insula and PCC-striatum areas may be regarded as biomarkers of brain damage severity in chronic heroin users.
Key words: heroin; craving; resting-state; posterior cingulate cortex (PCC)； functional magnetic resonance imaging
INTRODUCTION
Changes in neurocircuits that are implicated in reward, motivation, inhibitory control and working memory are the major reasons why drug addicts repeat compulsive drug seeking and using behavior1. Previous imaging studies determined that the activity of particular brain regions that are involved in reward, motivation and subjective urge are markedly associated with the intensity of the craving when subjects are exposed to drug-related cues. Such brain regions include the dorsolateral prefrontal cortex, orbitofrontal cortex, posterior cingulate cortex (PCC) and insula2-5. Among these regions, the PCC has not received much attention, especially with respect to heroin addiction. The PCC is a special region that is related to both drug abuse and the default mode network (DMN). Greicius6 proved that the PCC is a crucial part of the DMN and it is critical to the retrieval of episodic memories. The regions linked to the PCC in the DMN are related to the mental processes during the resting-state. Using the method of resting-state functional connectivity, Wang found that heroin users showed decreased functional connectivity between the ventral anterior cingulate cortex and PCC when compared to healthy controls7. Combining spatial and temporal information, Yuan found that heroin addicts showed significantly reduced functional connectivity between the PCC and right cerebellum and between the PCC and left dorsolateral prefrontal cortex8. Some neuroimaging studies on cue-induced craving demonstrated greater activation in the PCC in heroin/opioid-dependent individuals relative to healthy controls9, 10. However, the regions of the PCC identified in different experimental paradigms couldn’t necessarily overlap. Few studies have investigated the whole-brain functional connectivity of the PCC combining both cue-induced craving fMRI and resting-state fMRI in chronic heroin users. Exploring alteration of functional connectivity of this brain region identified through a cue-induced task-driven fMRI may provide a new perspective to understand the neural mechanism of heroin addiction.
The resting-state fMRI is a non-invasive technique which is easy to perform without any complicated task design. Resting-state functional connectivity, evaluated by temporal correlations of spontaneous low fluctuations (<0.08 Hz) of the blood oxygenation level-dependent (BOLD) signal across different brain regions, is regarded as a means to measure functional organization11. There have been some findings using resting-state fMRI for abnormal functional connectivity among regions that are helpful to elucidate the mechanisms underlying drug addiction8, 12, 13. Meanwhile, the abnormal activity of brain regions in experimental tasks may suggest that functional organization of neural networks involved in these brain regions have changed. In Rogers’ study14, external task-related stimuli could change the intrinsic activities, which suggest that the intrinsic and evoked brain activities are closely related. Additionally, some brain regions’ task-related activities and “resting” brain interregional functional connectivity are significantly correlated with behavioral measures or clinical performance.
We speculated that the PCC, activated in a cue-induced task, was likely to be involved in an abnormal resting network. The purpose of our study was to identify the PCC that was significantly activated during a cue-induced craving task. The significantly activated PCC during the task was employed as the seed region of interest (ROI) in the resting-state functional connectivity analysis between the heroin-addicted group and controls. In addition, the relationship between dysfunctional connectivity and heroin dependence was determined based on a correlation analysis.
METHODS
Subjects
Fourteen male heroin users (mean age, 35 ± 6.4 years; age range, 25–47 years) who fulfilled the diagnosis of heroin dependence (the 4th edition DSM) were recruited from the Drug Rehabilitation Center in Lantian, Xi’an, China. Their mean duration and dosage of heroin use was 89.3 months (range 19 to 182 months) and 0.6 grams/day (range 0.2 to 1.0 grams/day) respectively. To make sure the participants were in the detoxification phase, all of the heroin users tested negative for the presence of morphine in the urinalysis. The inclusion criteria were as follows: (1) DSM-IV criteria for opiate dependence; (2) 18-50 years old; (3) absence of cocaine or other illicit drug use; and (4) right-handed. Exclusion criteria included: (1) current or past psychiatric illness other than heroin dependence; (2) neurological signs and/or history of neurological disease; (3) history of head trauma; (4) history of cardiovascular or endocrine disease; (5) the presence of magnetically active objects in the body; and (6) claustrophobia or other medical conditions that would preclude the patient from lying in the MRI scanner for approximately 40 min.
Fifteen matched healthy controls were recruited. None of them reported a history of any drug dependence other than nicotine. None of the healthy controls had a history of injuries to the head or psychiatric disorders. All aspects of the research protocol were reviewed and approved by the Ethics Committee in Tangdu Hospital, the Fourth Military Medical University. All subjects provided written informed consent.
Experimental paradigm
This paradigm consisted of a resting-state fMRI scan and a heroin-cue-reactivity fMRI scan. For the resting-state scan, which was acquired before the cue-reactivity task, participants were instructed to fixate on a cross-hair centered on a screen, relax and move as little as possible. The resting run lasted 5 min. We employed an event-related design in the cue-induced craving task. All subjects (heroin users and healthy volunteers) participated in a 490-s run during which they were exposed to 48 pictures (24 heroin pictures and 24 neutral pictures). The task began with 10-s of a dummy scan followed by the first stimulus image. The task consisted of 48 trials and each trial included a 2-s picture (heroin-related or neutral picture) and an inter-stimulus interval (ISI) during which a crosshair was displayed. The ISI ranged from 4- to 12-s (mean time = 8-s). Trials were presented in a pseudo-randomized order. ...
WMatrixie