Postoperative mortality after hepatectomy remains high compared with other types of surgery in patients who have cirrhosis or chronic hepatitis. Although there are several useful perioperative markers of liver dysfunction, there are no standard markers for predicting postoperative liver failure. This study investigated risk factors for postoperative liver failure after resection of hepatocellular carcinoma to detect markers that could identify candidates for hepatectomy. Perioperative risk factors for liver failure after hepatectomy were analyzed in 191 patients with hepatocellular carcinoma. Multivariate logistic regression analysis was done to investigate factors with a significant independent influence among 35 variables. The ratio of serum hyaluronic acid to the maximum removal rate of technetium-99 m diethylenetriaminepentaacetic acid galactosyl human serum albumin (hyaluronate/GSA-Rmax ratio) was calculated. Liver failure occurred postoperatively in 16 patients, 3 of whom died. The hyaluronate/GSA-Rmax ratio was a risk factor for postoperative liver failure by univariate analysis and was the only risk factor according to multivariate analysis. All three patients who died had a hyaluronic acid/GSA-Rmax ratio > or = 500 mg min/dl. This ratio had a sensitivity of 88% and a specificity of 92% for predicting liver failure. To reduce postoperative liver failure, preoperative planning should employ various measures of the hepatic functional reserve, including tests of both parenchymal and nonparenchymal liver function. The hyaluronate/GSA-Rmax ratio can predict liver failure after hepatectomy, and a ratio > or = 500 mg min/dl is a relative contraindication to liver resection.