To compare the efficacy and patient acceptability of lactitol vs. lactulose in chronic recurrent portal-systemic encephalopathy (PSE), 25 cirrhotic patients with a history of repeated episodes of hepatic encephalopathy who required chronic administration of lactulose were included in a controlled cross-over clinical trial in which patients received, at random, lactitol (at an initial dosage of 10 g/6 h) or lactulose (15 ml/6 h, 66% w/v, containing 10 g of lactulose) during a 3 month period and then crossed-over to the alternative treatment for the following 3 months. Doses were adjusted to obtain two bowel movements per day. During the study period the daily protein intake was 40-60 g. Clinical and analytical data (including ammonia levels) were obtained, an EEG and the number connection test were performed and the PSE index was determined before treatment and monthly until the end of the treatment. No significant differences were found between the effects of lactitol and lactulose on the neurological and biological parameters, suggesting that the two treatments could be considered as equally effective. Lactitol was significantly better tolerated than lactulose (P = 0.02), the taste of which was assessed as being too sweet and provoking nausea. In conclusion, lactitol is a good alternative to lactulose for patients with chronic recurrent PSE, especially in those who do not tolerate the excessive sweetness of lactulose.