Predicting subsequent relapse by drug‐related cue‐induced brain activation in heroin addiction: an event‐related functional magnetic resonance imaging study

Abstract

Abnormal salience attribution is implicated in heroin addiction. Previously, combining functional magnetic resonance imaging (fMRI) and a drug cue-reactivity task, we demonstrated abnormal patterns of subjective response and brain reactivity in heroin-dependent individuals. However, whether the changes in cue-induced brain response were related to relapse was unknown. In a prospective study, we recruited 49 heroin-dependent patients under methadone maintenance treatment, a gold standard treatment (average daily dose 41.8 ± 16.0 mg), and 20 healthy subjects to perform the heroin cue-reactivity task during fMRI. The patients' subjective craving was evaluated. They participated in a follow-up assessment for 3 months, during which heroin use was assessed and relapse was confirmed by self-reported relapse or urine toxicology. Differences between relapsers and non-relapsers were analyzed with respect to the results from heroin-cue responses. Compared with healthy subjects, relapsers and non-relapsers commonly demonstrated significantly increased brain responses during the processing of heroin cues in the mesolimbic system, prefrontal regions and visuospatial-attention regions. However, compared with non-relapsers, relapsers demonstrated significantly greater cue-induced craving and the brain response mainly in the bilateral nucleus accumbens/subcallosal cortex and cerebellum. Although the cue-induced heroin craving was low in absolute measures, the change in craving positively correlated with the activation of the nucleus accumbens/subcallosal cortex among the patients. These findings suggest that in treatment-seeking heroin-dependent individuals, greater cue-induced craving and greater specific regional activations might be related to reward/craving and memory retrieval processes. These responses may predict relapse and represent important targets for the development of new treatment for heroin addiction.

Disrupted coupling of large-scale networks is associated with relapse behaviour in heroin-dependent men

“…Our procedures of longitudinal follow-up were in line with those described in a previous study. 3 Briefly, heroin-dependent men were followed for 3 months. Follow-up sessions occurred at 1, 2 and 3 months after the experimental session.…”

“…Data preprocessing using SPM 8 software (http://www.fil .ion.ucl.ac.uk/spm) and DPABI (http://rfmri.org/dpabi) included slice timing, motion correction, registration, spatial normalization to Montreal Neurological Institute (MNI) space with a resampled resolution of 3 × 3 × 3 mm 3 , and spatial smoothing (full width at half maximum 6 mm). The signals of white matter, cerebrospinal fluid and head motion parameters were removed through linear regression.…”

“…To account for testing 9 different ROIs, the critical cluster-corrected p value was set at 0.05÷9 = 0.006 based on Monte Carlo simulations. 29 This threshold corresponded to a single-voxel uncorrected significance of p < 0.001 and a minimum cluster size of 270 mm 3 . For the heroin-dependent patients, correlations between percentage of positive urine drug screens, basal craving for heroin, duration and dose of MMTP and the coupling between each 2 large-scale brain networks showing ANOVA differences were examined and corrected for multiple comparisons using a method of false-discovery rate (FDR).…”

“…2 Despite the high relapse rate, few neuroimaging studies so far have addressed the neural predictors of risk for heroin relapse. Studies using task-related functional MRI (fMRI) have shown some usefulness in identifying relapse predictors [3][4][5] or quitting predictors. 6,7 For example, a previous study by our group showed that drugrelated cues induced activity of the nucleus accumbens/ subcallosal cortex and that the cerebellum was associated with subsequent relapse in heroin-dependent individuals.…”

“…6,7 For example, a previous study by our group showed that drugrelated cues induced activity of the nucleus accumbens/ subcallosal cortex and that the cerebellum was associated with subsequent relapse in heroin-dependent individuals. 3 Brain activity associated with error processing was shown to predict relapse in cocaine-dependent patients. 4 Disrupted ventromedial prefrontal function was shown to be associated with relapse in alcohol-dependent patients.…”