Department of General Surgery and Organ Transplantation Department of Clinical Medicine, University of Bologna, Italy.
To evaluate the efficacy of preoperative liver stiffness (LS) measurement in predicting postoperative liver failure (PLF) after hepatectomy for hepatocellular carcinoma (HCC).
Hepatectomy for HCC in cirrhosis is affected by the risk of PLF, which is not completely predictable with common biochemical tests. Transient elastography with FibroScan is used to calculate the degree of LS, and it may be applicable to patients scheduled for hepatectomy to estimate perioperative complications.
Ninety-two patients undergoing hepatectomy for HCC were prospectively evaluated with preoperative FibroScan. Accuracy of LS measurement in predicting PLF, the presence of cirrhosis, and the presence of clinical signs of portal hypertension (PH) were assessed using receiver operating characteristic (ROC) analysis.
In 2 patients, LS measurement could not be performed because of obesity; consequently, 90 patients were suitable for the study. Perioperative mortality was 2.2% (2 patients); PLF occurred in 28.9% of patients (26 patients). ROC analysis identified patients with LS value higher than or equal to 15.7 kPa as being at higher risk of PLF [area under the curve (AUC) = 0.865, 95% confidence interval: 0.776-0.928; sensitivity = 96.1%; specificity = 68.7%; positive predictive value = 55.6%; negative predictive value = 97.8%; positive likelihood ratio = 3.08; negative likelihood ratio = 0.056; P < 0.001]. Patients with LS value lower than 14.8 kPa had no PLF. LS value higher than 12.6 kPa and higher than 19.6 kPa was correlated with the presence of cirrhosis (AUC = 0.880; P < 0.001), and of PH (AUC = 0.786; P < 0.001), respectively. Multivariate analysis showed that low preoperative serum sodium levels (P = 0.012), histological cirrhosis (P = 0.024), and elevated LS (P = 0.005) were independent predictors of PLF.
LS measured with FibroScan is a valid tool for prediction of PLF in patients undergoing hepatectomy for HCC.