The aminopyrine breath test has been proposed as a quantitative test of hepatic function, but its long-term prognostic value in patients with cirrhosis has not been determined. The aim of this study was to examine the usefulness of the aminopyrine breath test in assessing prognosis and to compare it with traditional methods of evaluating liver function. One-hundred eighty-seven patients with histologically confirmed cirrhosis were studied prospectively. An aminopyrine breath test was obtained at the time of inclusion in the study and results were expressed as per cent of the dose excreted in 2 hr. At inclusion, the severity of liver disease was also assessed according to the Pugh modification of the Child-Turcotte classification based on ascites, neurological status, serum albumin, serum bilirubin and prothrombin time. Mean follow-up was 844 days. During that period, 59 of 187 patients died of their liver disease. Two-year survival decreased with increasing Child-Turcotte classification score: survival was 98% in Child Class A patients (n = 62), 66% in Child Class B (n = 76) and 36% in Child Class C (n = 49) (chi 2 = 65.6, p less than 0.001). Two-year survival also decreased significantly with increasing degree of aminopyrine breath test abnormalities: survival was 90% in patients with aminopyrine breath test greater than 4% (n = 56); 78% in patients with aminopyrine breath test = 2 to 4% (n = 66), and 43% in patients with aminopyrine breath test less than 2% (n = 65) (chi 2 = 36.9, p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)