The diagnosis and prevalence of subclinical hepatic encephalopathy in apparently healthy, ambulant, non-shunted patients with cirrhosis.

Thirty-seven patients, all with histologic evidence of cirrhosis and with a normal neurological examination and normal mental status were evaluated by psychometric testing for subclinical hepatic encephalopathy. They were all regarded as having well compensated cirrhosis, not requiring any treatment or dietary restrictions and they were working, and many of them driving. A group of 19 patients with a history of alcoholism, or medical disorders, but without clinical or biochemical evidence of cirrhosis, served as controls. They were matched by age, sex, education, and alcohol consumption. Investigations performed were an EEG, fasting arterial ammonia, liver biochemical tests and a series of verbal and performance psychometric tests. The EEG was abnormal in 3 (8.3%) of patients, the ammonia elevated in 17 (45.9%) of patients and 26 patients (70.3%) failed 2 or more psychometric tests, as compared to 2 (10.5%) of the control group. It is concluded that 2 out of 3 patients with stable, well compensated cirrhosis were suffering from subclinical hepatic encephalopathy and that impairment of performance rather than verbal skills occurred. The digital symbol test, trail test (number connection test) and block design tests readily identified the patients with subclinical hepatic encephalopathy. The implication of these observations in patients with cirrhosis, especially those working in mechanical or skilled occupations, needs consideration.