To evaluate the relevance of the hypothesis that the arterial ketone body ratio (acetoacetate/β-hydroxybutyrate) can reflect the liver mitochondrial redox potential, we investigated the influence of ketone body metabolism in peripheral tissues on the arterial ketone body ratio during conditions in which no ketone body was available to the systemic circulation: during total clamping of the hepatoduodenal ligament in 37 patients undergoing partial hepatectomy and during the anhepatic phase in 14 liver transplant recipients. Changes in both the arterial acetoacetate and β-hydroxybutyrate concentrations and arterial ketone body ratio varied markedly from patient to patient after clamping. Among 20 patients with hepatocellular carcinoma and associated chronic liver disease, arterial ketone body ratio 15 min after clamping was higher in 11 and lower in 9 than that before clamping. Of the remaining 17 patients with other types of tumor, arterial ketone body ratio was increased in 10 and decreased in 7 after clamping. Similar changes were observed in 14 recipients during the anhepatic phase of liver transplantation. Arterial ketone body ratio was increased in 7 recipients during the anhepatic phase compared with that at laparotomy, whereas it was lowered in the remaining 7. These results indicate that each of the two types of ketone body is metabolized in extrahepatic tissues to a different extent, varying among individual patients and disease conditions. We conclude that arterial ketone body ratio does not solely reflect the ratio of the amounts of acetoacetate to β-hydroxybutyrate produced by the liver. This may make it difficult to use arterial ketone body ratio as an accurate parameter representing the redox state in liver mitochondria. (Hepatology 1994;20:331–335.)