A new model of human secretory diarrhoea using cholera toxin.

Secretory diarrhoea is a major cause of morbidity and mortality worldwide. However, there is no biologically relevant test system in man for assessing new anti-diarrhoeal therapies prior to clinical trial. We have used highly purified cholera toxin in combination with the triple lumen jejunal perfusion technique to establish a subclinical model of cholera in man. Cholera toxin was administered either by mouth with sodium bicarbonate or directly into a 30 cm 'open' or 'closed' (isolated between two inflated balloons) jejunal segment in healthy adult volunteers. Both oral dosing and direct delivery into an 'open' jejunal segment failed to produce consistent secretion of water and electrolytes. In contrast 15 micrograms or 25 micrograms of cholera toxin elicited secretion of water and sodium 3 h after instillation into the balloon occluded 'closed' jejunal segment (P less than 0.05 vs. controls). The rate of secretion was constant over the maximal period studied (4.5 h) and was similar to that reported in human cholera. None of the subjects experienced troublesome diarrhoea. We believe this model offers a relevant test system for assessing anti-diarrhoeal therapy in man.