In animals, colonic infusion of SCFA does not affect glucagon-like peptide-1 (GLP-1) release whereas intravenous infusion does and SCFA may directly stimulate peptide YY (PYY) release. It is unknown whether SCFA and their route of administration affect human blood concentrations of GLP-1 and PYY. Our aim was to conduct a pilot study to determine the effects of intravenous and rectal acetate on blood concentrations of GLP-1, PYY, ghrelin, adiponectin and TNF-α in hyperinsulinaemic human subjects. Six hyperinsulinaemic female subjects were given 20 mmol sodium acetate intravenously, 60 mmol acetate rectally, or normal saline rectally or intravenously on four separate occasions in randomised order, with blood samples collected at 0, 10, 15, 30, 45 and 60 min. Change in plasma PYY was significantly higher after acetate and rectal infusions (9·69 and 13·78 pg/ml) compared with saline and intravenous (0·60 and − 3·1 pg/ml; P < 0·01), respectively. Change in plasma GLP-1 was increased by rectal and acetate infusions (0·25 and 0·23 mmol/l) v. intravenous and saline ( − 0·26 and − 0·19 mmol/l; P < 0·01). Acetate decreased TNF-α v. saline ( − 0·8 and 0·15 pg/ml; P < 0·05). Rectal infusions increased TNF-α and ghrelin (0·2 and 98·27 pg/ml) v. intravenous ( − 0·9 and − 40 pg/ml; P < 0·01). There was no effect of treatment on plasma adiponectin. These preliminary results suggest that acetate raises plasma PYY and GLP-1, and suppresses TNF-α. Also, distending the rectum increases PYY, GLP-1, TNF-α and ghrelin in hyperinsulinaemic females. Increasing colonic fermentation products, particularly acetate, could yield a new mechanism for modifying weight gain.

Computer-administered questionnaires have received increased attention for large-scale population research on nutrition. In Belgium-Flanders, Young Adolescents' Nutrition Assessment on Computer (YANA-C) has been developed. In this tool, standardised photographs are available to assist in portion-size estimation. The purpose of the present study is to assess how accurate adolescents are in estimating portion sizes of food using YANA-C. A convenience sample, aged 11–17 years, estimated the amounts of ten commonly consumed foods (breakfast cereals, French fries, pasta, rice, apple sauce, carrots and peas, crisps, creamy velouté, red cabbage, and peas). Two procedures were followed: (1) short-term recall: adolescents (n 73) self-served their usual portions of the ten foods and estimated the amounts later the same day; (2) real-time perception: adolescents (n 128) estimated two sets (different portions) of pre-weighed portions displayed near the computer. Self-served portions were, on average, 8 % underestimated; significant underestimates were found for breakfast cereals, French fries, peas, and carrots and peas. Spearman's correlations between the self-served and estimated weights varied between 0·51 and 0·84, with an average of 0·72. The κ statistics were moderate (>0·4) for all but one item. Pre-weighed portions were, on average, 15 % underestimated, with significant underestimates for fourteen of the twenty portions. Photographs of food items can serve as a good aid in ranking subjects; however, to assess the actual intake at a group level, underestimation must be considered.