Xanthan gum and colonic function in man.

It was with interest that I read this paper, which looks at the effect of feeding xanthan gum on colonic function in man. The authors look at whether it is possible to predict the laxative effect of a polysaccharide from its fermentation characteristics in vitro, as proposed by Adiotomre et al. (1990). However, the authors have not used the methodology used in the Adiotomre paper and, therefore, can hardly be said to have refuted the acceptability of this approach. The Adiotomre experiments were designed to form the basis of simple methods to measure and compare the potential biological effects of various dietary fibres. The intention was not to produce absolute measurements of biological action to compare with grams of fibre, but to enable a hierarchy of potential biological activity to be established for a series of fibres. This would allow a fibre to be chosen for a particular function of fibre, whether the required function be retardation of nutrient absorption, influence on sterol metabolism or effect on caecal fermentation and stool weight. The mechanism by which dietary fibre increases stool weight is not fully understood, but the major mechanism is thought to be an increase in the water-holding capacity of stool, which is either due to the presence of the fibre itself or to bacterial cells which increase in number if the fibre is fermented (Stephen & Cummings, 1980). Fibres are degraded to some extent in the colon and therefore lose some of their waterholding capacity. The water-holding capacity after incubation with human faecal bacteria has been suggested to be a good indicator of their effect on stool output (McBurney et al. 1985). However, residual water-holding capacity on its own was not useful as a predictive index for stool weight; therefore, a predictive index was developed which combined the water-holding capacity of the fibre after fermentation and the production of short-chain fatty acids. We have also suggested that the prediction of how a dietary fibre would affect stool weight could be shown by a simple formula: