Impaired working memory performance in opioid-dependent patients is related to reduced insula gray matter volume: a voxel-based morphometric study

Eur Arch Psychiatry Clin Neurosci. 2021 Aug;271(5):813-822. doi: 10.1007/s00406-019-01052-7. Epub 2019 Aug 17.


Opioid-dependent patients frequently show deficits in multiple cognitive domains that might impact on their everyday life performance and interfere with therapeutic efforts. To date, the neurobiological underpinnings of those deficits remain to be determined. We investigated working memory performance and gray matter volume (GMV) differences in 17 patients on opioid maintenance treatment (OMT) and 17 healthy individuals using magnetic resonance imaging and voxel-based morphometry. In addition, we explored associations between substance intake, gray matter volume, and working memory task performance. Patients on OMT committed more errors during the working memory task than healthy individuals and showed smaller insula and putamen GMV. The duration of heroin use prior to OMT was associated with working memory performance and insula GMV in patients. Neither the substitution agent (methadone and buprenorphine) nor concurrent abuse of illegal substances during the 3 months prior to the experiment was significantly associated with GMV. Results indicate that impaired working memory performance and structural deficits in the insula of opioid-dependent patients are related to the duration of heroin use. This suggests that early inclusion into OMT or abstinence-oriented therapies that shorten the period of heroin abuse may limit the impairments to GMV and cognitive performance of opioid-dependent individuals.

Keywords: FMRI; Gray matter volume; Methadone; Opioid addiction; Voxel-based morphometry; Working memory.

MeSH terms

  • Gray Matter* / diagnostic imaging
  • Gray Matter* / pathology
  • Humans
  • Magnetic Resonance Imaging
  • Memory Disorders* / physiopathology
  • Memory, Short-Term / physiology
  • Opioid-Related Disorders* / physiopathology
  • Organ Size