AIMS To evaluate the accuracy of HVPG (direct method), ICG-R15 and LSM/SSM (non-invasive methods) in predicting the risk of PHLF and hepatic decompensation; to evaluate the correlation between hepatic functional reserve (ICG-R15) and degree of portal hypertension (HVPG) and LSM/SSM in Child Pugh 0, A and B cirrhotic patients; to evaluate the real weight of HVPG in a multivariate analysis.

METHODS: Multicentric observational prospective study. INCLUSION CRITERIA: All patients with liver cirrhosis with indication for surgical resection for hepatocellular carcinoma are considered eligible. Patients will be selected for surgery based on standard criteria of participating centers and on Child-Pugh score 0, A or B.

EXCLUSION CRITERIA: Patients undergoing emergency surgery; inability to understand informed consent.

Primary endpoint: Comparison of the predictive accuracy (evaluated as Area Under The Curve of the ROC curve, see statistics) of HVPG, ICG and Liver and Spleen Stiffness of Post-operative Hepatic Failure (PHLF, according to ISGLS).

Secondary endpoints: - Predictive accuracy of HVPG, ICG-R15, LSM and SSM on postoperative morbidity at 90 days according to the classification of Clavien-Dindo and to the Comprehensive Complication Index (CCI), and on cirrhosis decompensation at 3 months after surgery.

BACKGROUND: Most of current guidelines for the treatment of Hepatocellular Carcinoma (HCC) are based on the Barcelona Clinic Liver Cancer (BCLC) staging system1, where Portal Hypertension (PHT) alone contraindicates surgery2. However, the evolution of surgical techniques and patient management has led many centers to overcome the limits proposed by BCLC3, highlighting the need for new criteria and to consider additional factors tailored on each patient4,5. Almost all papers analyzing the prognostic role of PHT in patients undergoing liver surgery are based on retrospective data and on an indirect assessment of the presence of PHT, that have been shown to have less prognostic role7. Among the indirect methods of PHT assessment, hepatic elastography using FibroScan® has been recently consolidated. LSM is known to be related to postoperative outcomes, but its impact on the subset of PHT patients, as well as a comparison with SSM or HPVG, has never been investigated. Given the splenic congestion secondary to PHT, new diagnostic algorithms for CSPH take into account the spleen stiffness (SSM, with 46kPa cut-off)8. SSM has also been shown to be related to post-operative outcomes9. In the pre-operative evaluations on the extent of resection a key role is played by the hepatic functional reserves estimated with the blood retention at 15 'of Indocyanine Green (ICG-R15). It has been proposed a relationship between ICG blood levels and hepatic venous pressure gradient (HVPG), even if only in Child A patients6. ICG-r15' has never been compared to LSM and SSM for postoperative outcomes.

AIMS To evaluate the accuracy of HVPG (direct method), ICG-R15 and LSM/SSM (non-invasive methods) in predicting the risk of PHLF and hepatic decompensation; to evaluate the correlation between hepatic functional reserve (ICG-R15) and degree of portal hypertension (HVPG) and LSM/SSM in Child Pugh 0, A and B cirrhotic patients; to evaluate the real weight of HVPG in a multivariate analysis that takes into account patient demographics, CHILD, MELD, tumor characteristics, extent of surgery, blood loss, invasiveness of the approach, LSM, SSM.

METHODS: Multicentric prospective study. INCLUSION CRITERIA: All patients with liver cirrhosis with indication for surgical resection for hepatocellular carcinoma are considered eligible. Patients will be selected for surgery based on standard criteria of participating centers and on Child-Pugh score 0, A or B. EXCLUSION CRITERIA: Patients undergoing emergency surgery; inability to understand informed consent. Primary endpoint: Comparison of the predictive accuracy (evaluated as Area Under The Curve of the ROC curve, see statistics) of HVPG, ICG and Liver and Spleen Stiffness of Post-operative Hepatic Failure (PHLF, according to ISGLS). Secondary endpoints: - Predictive accuracy of HVPG, ICG-R15, LSM and SSM on postoperative morbidity at 90 days according to the classification of Clavien-Dindo and to the Comprehensive Complication Index (CCI), and on cirrhosis decompensation at 3 months after surgery. Statistic analysis Comparisons between groups will be performed using student's T test or Wilcoxon's rank test for continuous data, depending on the distribution of the variable. Categorical data will be compared by using Pearson's Chi Square test. The presence of a linear correlation between variables will be verified by linear regression. The accuracy of a continuous variable in predicting a categorical outcome will be evaluated using the Receiver Operating Characteristic (ROC) curves. The value of the area under the curve (AUC), will be used as an index of predictive accuracy. The comparison between the AUC of different variables will be performed by the method described by Hanley et al. A univariate and multivariate analysis will also be performed including risk factors for surgery in cirrhotic patients, including also LSM and SSM.

Inclusion Criteria:

• All adult patients with liver cirrhosis with indication for surgical resection for hepatocellular carcinoma are considered eligible. Patients will be selected for surgery based on standard criteria of participating centers and on Child-Pugh score 0, A or B

Exclusion Criteria:

• Patients undergoing emergency surgery; inability to understand informed consent.

Cirrhotic patients with HCC undergoing liver resection