The influence of lactulose on organic acid fermentation, nitrogen metabolism and excretion in the colon associated with its mechanism of action on hepatic encephalopathy was investigated. Orally administered lactulose in increasing amounts ( 0 to 20 to 40 to 80 to 160 &day) to 12 healthy volunteers decreased ammonia production in 16.6% fecal homogenates incubated 6 hr and 24 hr at 37" C (mean f S.E.M.: from 7 f 1 to 0 f 0 and from 13 f 2 to 0 f 0 mmolb, respectively). Every dose of lactulose was given for 3 days with intervals of 1 to 2 wk, and 24-hr stools were collected on day 3. Fecal concentrations of ammonia decreased (from 50 & 9 to 11 f 3 mmolb), but ammonia excretions increased (from 6 f 2 to 17 f 4 mmo1/24 hr). Total fecal concentrations of nitrogen decreased (from 1,043 -+ 78 to 300 f 136 mmolb), but excretions of nitrogen increased fourfold (from 111 f 21 to 457 f 113 mmoL/24 hr) because of the increase in stool mass. Fecal pH declined (from 6.9 f 0.1 to 4.9 f O.l), but total organic acids (short-chain fatty acids and DL-lactate; range = 105 to 148 mmolb) and osmolality in feces (417 to 450 mOsmb) did not change, although the colonic fermentation of lactulose had a major impact on the proportions between the nontoxic acetate (increased from 65% f 2% to 89% f 3%) and the potentially neurotoxic 3-6-carbon fatty acids (decreased from 35% f 2% to ll% & 2%). The effects of lactulose on bacterial ammonia assimilation, protein degradation and fermentation in the colonic contents in uiuo are central in the mechanism of action for the increased excretion of nitrogen. However, ammonia excretion per se is not important because it only constitutes 4% to 5% of fecal nitrogen. (HEPATOLOGY 1992; 16~21350-1356.) When lactulose was first used to treat hepatic encephalopathy approximately 25 yr ago, it was assumed that its beneficial effects were mediated by way of changes in the enteric flora (1, 2). More recently, it has become apparent that the nonabsorbable disaccharides exert not