Objective: Faecal bifidobacteria and lactobacilli, perceived as exerting health-promoting properties, may be increased by ingestion of high-dose lactulose in humans. The effects of low and well-tolerated doses of lactulose are not well known. The aim of the study was to assess the effects of prolonged low-dose lactulose administration on faecal bifidobacteria and selected metabolic indexes potentially involved in colonic carcinogenesis. Subjects and methods: In all, 16 healthy volunteers were included in this controlled, randomised, double-blind, parallel group trial. Participants ingested lactulose or placebo (sucrose) at a dose of 5 g b.i.d. for 6 weeks. Stools were regularly collected at baseline (d0), and after 3 (d21) and 6 (d42) weeks of sugar ingestion. Tolerance was evaluated using a daily chart. Results: Faecal bifidobacterial counts were higher in lactulose than in sucrose group (P ¼ 0.03). Lactulose ingestion led to a significant increase in faecal bifidobacteria counts from d0 to d21 and d42 ((m7s.e.m.) 8.2570.53, 8.9670.40 and 9.5470.28 log colony-forming units/g wet wt (CFU/g), respectively (P ¼ 0.048)). Placebo ingestion did not lead to any faecal bifidobacterial count change. Total anaerobes, Lactobacillus and pH were not significantly changed throughout the study in the two groups. Neither faecal bile acids nor neutral sterols were modified by lactulose. Excess flatus was more common in the lactulose group (P ¼ 0.03), but was very mild. Bloating and borborygmi did not differ between both the groups. Conclusions: A measure of 10 g lactulose/day increases faecal bifidobacterial counts, and lactulose fulfils the criteria requested to be considered as a prebiotic.