A New Prediction Model of Postoperative Complications after Major Hepatectomy for Hepatocellular Carcinoma

Abstract

Background/Aims: Serum hyaluronic acid (HA) concentrations reflect the degree of hepatic dysfunction and may have potential for predicting postoperative complications in a major hepatectomy for hepatocellular carcinoma (HCC). Methods: Serum HA concentrations and other conventional liver function tests were measured prior to major hepatectomies in 52 patients. Independent predictors for postoperative complications were analyzed and the discriminant equation was established and validated. Results: Postoperative complications occurred in 17 patients. Serum HA concentrations and the estimated remnant liver volume ratio (remnantVol%) were recognized as independent predictors for postoperative complications (OR 1.03, 0.85; CI 95% 1.01–1.06, 0.76–0.95; p = 0.006, 0.006; respectively) and produced the discriminant equation: logit = 4.15 + 0.03 × (HA) –0.16 × (remnantVol%). The value of the area under the curve of a receiver operating characteristic analysis was 0.92. If the cut-off of the logit value was set to 0, then the predictive accuracy was 0.88. The validation accuracy performed by a leave-one-out cross-validation method was 0.83. Conclusions: The constructed discriminant equation model consisting of the preoperative serum HA concentrations and estimated remnantVol% could be useful for predicting postoperative complications in a major hepatectomy for HCC.