Not All Is Lost for Relapsers: Relapsers With Low WHO Risk Drinking Levels and Complete Abstainers Have Comparable Regional Gray Matter Volumes

Abstract

Background: Reductions in substance use are associated with positive long-term treatment outcomes such as psychosocial functioning; substance use–related consequences; and mental, physical, and neurobiological health. Therefore, nonabstinent clinical trial endpoints have received growing attention from substance abuse treatment experts. Regional brain tissue volumes in alcoholism treatment seekers increase during abstinence, in parallel to cognitive performance. Methods: We examined the relationships of drinking levels in those who did not maintain abstinence after treatment with magnetic resonance imaging-derived gray matter (GM) volumes measured 8 months after baseline assessments while in outpatient treatment. The complex drinking behavior during the interval was operationalized as World Health Organization risk drinking levels (WHO-RDL), derived from the number of standard alcoholic drinks per day. We compared the volumes of addiction-relevant cortical and subcortical brain regions at long-term follow-up among abstainers and 2 groups of relapsers with low and higher WHO-RDL. Results: We found that: (i) relapsers with low WHO-RDL at follow-up, who as a group had reduced their risk levels by 2.8 units (by consuming <40 g of ethanol per day over the recovery interval), had regional brain tissue volumes indistinguishable from those of abstainers tested after the same time period and that (ii) relapsers with higher WHO-RDL, with only 1.2 units average risk-level reductions, had significantly smaller frontal GM and thalamic volumes than abstainers and relapsers with low WHO-RDL. Despite many drinking days during recovery, including several with >10 drinks per day, relapsers with low WHO-RDL at follow-up tended to perform better than those with higher WHO-RDL on cognitive domains of working memory and visuospatial skills assessed over the recovery interval. Conclusions: Nonabstinent recovery is characterized by addiction-relevant GM volumes comparable to those of complete abstainers. The WHO-RDL has meaningful structural neuroimaging correlates potentially suitable as cognitively relevant biomarkers of treatment response and general brain health, perhaps even as objective clinical trial endpoints.