Recently, arterial keton body ratio (AKBR) has attracted attention as a new indicator of liver function which is in equibilium with the ratio between oxidized and reduced forms of free nicotinamide-adenine dinucleotides (free NAD⁺/NADH ration) in the mitochondria. There are few reports on whether AKBR contributes to the hepatic energy charge in the open heart surgery with extra corporeal circulation (ECC) or not. This study was undertaken to clarify the contribution of AKBR to the hepatic energy charge during ECC and the relationship between AKBR and hepatic blood flow. AKBR was determined before, during and after ECC in the open heart surgery for 14 patients. Furthermore, the blood flow in hepatic artery, portal vein and liver microcirculation was measured before, during and after ECC in canine models. Finally, the pulsatile perfusion was performed in canine models and compared with the conventional non-pulsatile perfusion for the blood flow and AKBR. In clinical cases, AKBR was decreased in all cases during the ECC. AKBR which was measured at 2 or 3hr after weaning from the ECC was negatively correlated to the total perfusion time with -0.57 as the correlation coefficient. Six patients who were on the ECC over 180min did not show a good recovery of the AKBR after weaning from the ECC. Especially, three patients presented a clinical picture of acute hepatic failure with jaundice, elevation of the serum levels of transaminase and direct hyperbililubinemia, but only one showed hypoglycemia. These patients showed no improvement in clinical data and AKBR. The patient with improved AKBR recovered clinically. In our experiment, the blood flow in the hepatic artery, portal vein was measured by electromagnetic blood flow meter and the liver microcirculation was measured by laserdoppler flowmeter. The blood flow was decreased remakably in the non-pulsatile group at all sites of measurement: it recovered after ECC in hepatic artery and portal vein, but liver microcirculation did not improve well. AKBR was decreased during ECC and did not recover after ECC in the non-pulsatile perfusion. When the pulsatile perfusion was performed, liver circulation was maintained well, and AKBR recovered well after ECC. The above results suggest that AKBR reflects the liver microcirculation and pulsatile perfusion is beneficial to the liver microcirculation. Pulsatile circulation, however, involves several problems, hemolysis, the decrease of platelets, and so on, but these problems have been improved gradually. We think that the pulsatile perfusion will be used in clinical operations to maintain the good hepatic circulation.