BACKGROUND & AIM: Increased intestinal permeability (IP) can occur in older people and contribute to the activation of the immune system and inflammation. Dietary interventions may represent a potential strategy to reduce IP. In this regard, specific food bioactives such as polyphenols have been proposed as potential IP modulator due to their ability to affect several critical targets and pathways that control IP. The trial aimed to test the hypothesis that a polyphenol-rich dietary pattern can decrease serum zonulin levels, an IP surrogate marker involved in tight junction modulation, and can beneficially alter the intestinal microbiota, and IP-associated biochemical and clinical markers in older subjects. METHODS: A randomised, controlled, cross-over intervention trial was performed. Sixty-six subjects (aged ≥ 60 y) with increased IP based on serum zonulin levels, were randomly allocated to one of the two arms of the intervention consisting of a control diet (C-diet) vs. a polyphenol-rich diet (PR-diet). Each intervention was 8-week long and separated by an 8-week wash out period. At the beginning and at the end of each intervention period, serum samples were collected for the quantification of zonulin and other biological markers. Faecal samples were also collected to investigate the intestinal microbial ecosystem. In addition, anthropometrical/physical/biochemical parameters and food intake were evaluated. RESULTS: Fifty-one subjects successfully completed the intervention and a high compliance to the dietary protocols was demonstrated. Overall, polyphenol intake significantly increased from a mean of 812 mg/day in the C diet to 1391 mg/day in the PR-diet. Two-way analysis of variance showed a significant effect of treatment (p = 0.008) and treatment × time interaction (p = 0.025) on serum zonulin levels, which decreased after the 8-week PR-diet. In addition, a treatment × time interaction was observed showing a reduction of diastolic blood pressure (p = 0.028) following the PR-diet, which was strongest in those not using antihypertensive drugs. A decrease in both diastolic (p = 0.043) and systolic blood pressure (p = 0.042) was observed in women. Interestingly, a significant increase in fibre-fermenting and butyrate-producing bacteria such as the family Ruminococcaceae and members of the genus Faecalibacterium was observed following the PR intervention. The efficacy of this dietary intervention was greater in subjects with higher serum zonulin at baseline, who showed more pronounced alterations in the markers under study. Furthermore, zonulin reduction was also stronger among subjects with higher body mass index and with insulin resistance at baseline, thus demonstrating the close interplay between IP and metabolic features. CONCLUSIONS: These data show, for the first time, that a PR-diet can reduce serum zonulin levels, an indirect marker of IP. In addition, PR-diet reduced blood pressure and increased fibre-fermenting and butyrate-producing bacteria. These findings may represent an initial breakthrough for further intervention studies evaluating possible dietary treatments for the management of IP, inflammation and gut function in different target populations. THIS STUDY WAS REGISTERED AT WWW.ISRCTN. ORG AS: ISRCTN10214981.