BACKGROUND AND AIMS: Alcohol Use Disorder (AUD) is a psychiatric disorder characterized notably by gut microbial dysbiosis and insufficient dietary fiber (DF) intake. This study aims to investigate the effect of DF placebo-controlled intervention in patients suffering from AUD during a three-week period of alcohol withdrawal, in order to discover microbial-derived metabolites that could be involved in metabolic and behavioral status. METHODS: A randomized, double-blind, placebo-controlled study was performed with 50 AUD patients supplemented with inulin (prebiotic DF) or maltodextrin (placebo) during 17 days. Fecal microbiota composition, plasma and fecal metabolomics (liquid chromatography coupled to mass spectrometry), blood markers of inflammation and hepatic alterations, and psychological assessment (questionnaires) were analyzed before and after the intervention. RESULTS: Fecal metabolomics revealed 14 metabolites significantly modified by inulin versus placebo treatment (increased N8-acetylspermidine and decreased indole-3-butyric acid, 5-amino valeric acid betaine (5-AVAB) and bile acids). Thirteen plasma metabolites differentiated both treatments (higher levels of long-chain fatty acids, medium-chain acylcarnitines and sphingomyelin species, and reduced 3-methylhistidine by inulin versus placebo). Fecal Lachnoclostridium correlated with 6 of the identified fecal metabolites, whereas plasma lipidic moieties positively correlated with fecal Ruminococcus torques group and Flavonifractor. Interestingly, parameters reflecting liver alterations inversely correlated with sphingomyelin (SM 36:2). CONCLUSIONS: Three weeks of inulin supplementation during alcohol withdrawal leads to specific and different changes in the plasma and fecal metabolome of AUD patients, some of these gut microbiota-related metabolites being correlated with liver function. TRIAL REGISTRATION: NCT03803709, https://clinicaltrials.gov/ct2/show/NCT03803709.