Lactitol (β-galactosido-sorbitol) is a nonabsorbable disaccharide available as a powder which, in open comparison, is as effective as lactulose in the treatment of chronic hepatic encephalopathy, but is better tolerated. Twenty-five cirrhotic patients experiencing 28 episodes of acute hepatic encephalopathy were randomized blindly to treatment with either lactitol (n = 15) or lactulose (n = 13). The sugars were dispensed in solutions identical in appearance, taste and pH and of similar osmolarity, which contained either 66.7 gm per 100 ml lactitol or 66.7 ml (44.5 gm) per 100 ml lactulose syrup. The initial dose of 0.75 ml per kg was adjusted to produce two semisoft stools per day. Patients were assessed every 12 hr for 5 days. There were no significant differences in sex ratio, age, body weight, clinical status, duration and extent of coma, etiology of liver disease or of heptic encephalopathy between the two groups of patients on entry to the trial. An adequate catharsis was obtained with an equivalent mean (± 1 S.D.) daily dose of 26 ± 5 gm lactitol or 31 ± 7 ml (21 ± 5 gm) lactulose syrup. During the trial, significant improvements occurred in clinical status and psychometric performance and in the electroencephalogram mean cycle frequencies in the majority of patients in both groups. At the end of the trial, 67% of the patients in the lactitol group and 69% of the lactulose group were clinically normal. However, patients treated with lactitol responded significantly more quickly than patients treated with lactulose. Thus, at 24, 36, 48 and 72 hr, significantly greater mean percentage improvements were seen in PSE Indices in patients receiving lactitol than in patients receiving lactulose. Both sugars are effective in the treatment of acute hepatic encephalopathy in cirrhotic patients, but patients treated with lactitol respond significantly more quickly. Lactitol may be the treatment of choice for this condition.