OBJECTIVE The arterial ketone body ratio (AKBR) and the cellular adenosine triphosphate (ATP) concentration have been proposed as indicators of liver function. However, recent studies of the utility of the AKBR as a biochemical marker have been called into question. Furthermore, there is no practical data defining the relationship between ATP concentration and ischemia-reperfusion (IR) changes during liver surgery. METHODS The relationship of the AKBR and arterial ATP concentration to IR during hepatectomy was investigated. In 20 patients who underwent hepatectomy, arterial acetoacetate, beta-hydroxybutyrate, and ATP concentrations were measured. The ratio of acetoacetate to beta-hydroxybutyrate (AKBR) was calculated before and after vascular occlusion. RESULTS The AKBR 15 minutes after clamping was lower than the preclamping values in all of the patients. It increased after unclamping, returning toward the preclamping levels. An AKBR of less than 0.5 prior to clamping did not correlate with preoperative hepatocellular function. An AKBR of less than 0.7 throughout IR was not a consistent risk factor for postoperative complications and liver dysfunction. The arterial ATP concentration did not correlate with the changes during IR or with preoperative hepatocellular function. CONCLUSIONS Although the AKBR changed during IR as a general indicator of cellular activity, the absolute value of the AKBR was not an accurate predictor of liver function. The arterial ATP concentration also was not a suitable clinical biochemical marker of hepatic function.