Neural Dysfunction during Decision-Making as a Predictor of Cocaine Relapse
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Cocaine dependence is a costly disorder characterized by recurrent relapse events. The current investigation used functional magnetic resonance imaging (fMRI) during a decision-making task to predict relapse to drug use in a cocaine-dependent sample. Forty-five treatment-seeking cocaine-dependent subjects, two to four weeks abstinent, and 23 healthy control subjects underwent 3T fMRI. The Response Reversal Task was administered in the scanner to elicit decision-making processes. Individuals were followed for up to six months post-discharge from inpatient substance use treatment to determine time-to-relapse. Seventy-eight percent of the patient sample relapsed an average of 35 days after treatment; ten individuals did not relapse during the follow-up period. No group differences were found between healthy control and cocaine-addicted groups in activation patterns or behavioral measures of decision-making performance. Mean percent BOLD signal change in the patient group was used to identify regions of interest (ROIs) for discriminant analyses to classify patients by short- and long-term relapse. The fMRI activation patterns in the precuneus, bilateral orbitofrontal cortex, left insula, right dorsolateral prefrontal cortex, paracingulate, left hippocampus, and bilateral amygdala correctly classified 71% of patients by short-term and long-term relapse. This investigation suggests that neuroimaging may be a valid predictor of cocaine relapse, which could allow for better individual tailoring of treatment options for improving long-term abstinence.