Effect of banana consumption on faecal microbiota: a randomised, controlled trial.

Banana is a widely consumed fruit, which contains considerable amounts of potential prebiotic indigestible carbohydrates. In our randomised, controlled trial we aimed to evaluate the in vivo prebiotic effect of banana consumption on faecal microbiota. Thirty-four healthy women participated in the study, having Body Mass Index (BMI) 24-30 kg/m(2), age 19-45 years, without history of gastrointestinal disease and no antibiotic and other medication use two months prior the initiation and during the study. All women were asked to maintain their usual dietary habits for 60 days and they were randomly assigned to consume twice a day a pre-meal snack, either one medium banana, or one cup of banana-flavoured drink or one cup of water (control group). Stool samples were collected at baseline, on days 30 and 60 of intervention for enumeration of total anaerobes, bifidobacteria and lactobacilli by plate count techniques, as well as for pH and short chain fatty acids (SCFAs) measurement. Gastrointestinal symptoms were also recorded. Mean bifidobacterial levels were increased only in the banana group both at 30 and 60 days of intervention, but this change did not reach a statistical significance. No significant overall differences in the total concentrations and molar ratios of SCFAs were detected according to dietary intervention. Analysis of the gastrointestinal symptoms records revealed significantly lower bloating levels in the banana group, compared to controls, at 26-35 days (p = 0.009) and 51-60 days (p = 0.010). Banana consumption had also no adverse effects on evacuation patterns. We concluded that daily consumption of bananas is a well-tolerated eating behaviour, which may induce bifidogenesis in healthy women experiencing body weight problems.

Impact of beta-glucan on the faecal microbiota of polypectomized patients: a pilot study.

Beta-glucans are polysaccharides present in the cell walls of higher plants, in the seeds of some cereals, and certain yeasts and fungi also produce them. It is suggested that they exhibit, among many other health benefits, protective effects against carcinogenesis in the colon, but there is not enough human data to support this. The aim of the study was to determine the effect of barley-derived beta-glucan in the gut microbiota of polypectomized patients. Subjects were randomly assigned to consume 125 g of bread per day with beta-glucan (3 g/d), or without (placebo group), for 3 months. Thirty-three polypectomized men and women (mean age 57.6 years) were recruited into the study, but only 20 completed. Subjects did not consume any probiotics, prebiotics or antibiotics 2 months prior the intervention, or during the study. Stool samples were collected at baseline, on days 30 and 90 of intervention, as well as 2 weeks after the intervention, for enumeration of total aerobes and anaerobes, coliforms, E. coli, enterococci, Bacteroides spp., Clostridium perfringens, bifidobacteria, lactobacilli and Candida spp. Faecal bacterial enzyme activity (beta-glucuronidase and beta-glucosidase), pH, faecal moisture and the concentration of volatile fatty acids in the faeces were measured. Gastrointestinal symptoms were also recorded. Overall, no significant differences were observed in bacterial viable counts between the two feeding groups. Group specific analysis for ß-glucan group revealed significantly decreased total coliform counts on the 30th day of the trial compared to the baseline (p = 0.041). Clostridium perfringens concentration increased without reaching statistical significance, on the 30th day, while it decreased significantly on the 90th day of the intervention compared to the 30th day (p = 0.016). An increase was noted in the molar ratio of acetate on the 90th day of the trial compared to placebo (p = 0.018). The molar ratio of butyrate presented a trend to increase on the 30th day, which decreased (p = 0.013) on the 90th day and then increase 2 weeks after the intervention (p = 0.017) compared to placebo. A decrease was recorded in the ß-glucan group in the bloating and abdominal pain score after the 30th day of the intervention (Day 30-37) compared to placebo. During ß-glucan administration we did not observe any changes on beta-glucuronidase or beta-glucosidase activity, faecal pH, or on faecal moisture.