Lactitol (beta-galactosido-sorbitol) has been recently compared with lactulose for the treatment of chronic hepatic encephalopathy in a few studies, each comprising a small number of patients. The results are controversial. We studied the efficiency and tolerance of both compounds by using a meta-analysis on the basis of published controlled trials. Our study only included controlled or randomized trials comprising cirrhotic patients with chronic hepatic encephalopathy. Analyzed parameters were the portosystemic encephalopathy index of Conn after treatment, the percentage of improved patients and the percentage of patients who had ill effects related to the treatment (flatulence, diarrhea). Bibliographical screening revealed five studies comparing the effects of lactitol and lactulose in chronic hepatic encephalopathy. Four crossover studies were done that included 48 patients and one parallel study that included 29 patients. The duration of the treatment ranged from 3 to 6 mo. All studies found a similar efficiency with both drugs. However, they exhibited some discrepancies in the relative frequency of adverse reactions (flatulence). Meta-analysis showed no statistical differences in the portosystemic encephalopathy index after lactitol or lactulose treatment. The percentage of improved patients after lactitol or lactulose was similar. In contrast, the analysis revealed a higher frequency (p less than 0.01) of flatulence in patients treated with lactulose compared with those treated with lactitol. In conclusion, this meta-analysis shows no statistical difference between therapeutic effects of lactitol and lactulose, but it does show a higher frequency of flatulence with lactulose. This suggests that lactitol should be preferred to lactulose for the treatment of chronic hepatic encephalopathy.