Effect of dietary protein manipulation in subclinical portal-systemic encephalopathy.

Eight stable cirrhotic patients with mild or subclinical portal-systemic encephalopathy (PSE) were studied after shunt surgery when they were off all antiencephalopathic therapy. Equal amounts of mixed proteins were alternated with animal or vegetable protein in a crossover protocol under metabolic conditions for five consecutive, one week periods. The different dietary periods were not associated with either a change in the neurological impairment score or the Trailmaking Tests, which showed a learning effect. The peak frequencies of the computer analysed EEG (CAEEG) were lower during the animal (6.58 +/- 0.42 Hz) than the vegetable (7.10 +/- 0.44 Hz) diet (p 0.01). Neither arterial ammonia levels nor plasma amino acid ratios changed with the diets, whereas urinary 3-methyl-histidine excretion increased during the animal diet. During the vegetable diet the apparent nitrogen balance tended to be more positive than during either the mixed or animal diets associated with decrease in the urinary nitrogen excretion. The peak frequency of the CAEEG is the most sensitive test to monitor methods of treatment in portal-systemic encephalopathy. A vegetable protein diet, rather than overall protein restriction, should be considered in the management of this disorder, particularly when the nutritional state is poor.