Background: The decision of whether to proceed with transplantation in patients with hepatocellular carcinoma (HCC) in cirrhosis is based on clinical and imaging findings. The purpose of our study was to evaluate the accuracy of the current system.
Materials and methods: We evaluated data of 70 patients with HCC who underwent liver transplantation (LTx) at our center. We specifically analyzed the correlation between preoperative imaging studies and postoperative anatomopathologic findings. Tumor-node-metastasis, Milan, and University of California San Francisco (UCSF) classifications were used. Patients were divided in 2 groups: (1) Patients undergoing live-donor LTx (LDLTx, n=35) and (2) patients undergoing cadaveric LTx (CLTx, n=35).
Results: Only 10 (14.3%) of the 70 patients considered had tumor diameter that was correctly identified by pretransplant radiologic examinations. Twenty-four (34.2%) patients had correct identification of the number of tumors present. Fifty (71.4%) patients had incorrect measurements of tumor diameter of more than 1 cm. Sensitivity of radiologic imaging was especially poor for tumors between 1 and 2 cm and less than 1 cm (21% and 0%, respectively). Best accuracy of 60% was found for both the Milan and UCSF criteria. No significant difference was found between the two patient groups concerning the accuracy of the various systems/criteria of classification.
Conclusions: Current imaging techniques have a high incidence of false-negative and false-positive results when evaluating HCC in cirrhosis. A critical appraisal of patient characteristics together with great caution when interpreting imaging studies is recommended to determine candidacy for transplantation.