Different clinical risk scores for prediction of early mortality after liver resection for hepatocellular carcinoma: which is the best?

Background: Prediction of early mortality after hepatectomies for hepatocellular carcinoma is essential to identify high-risk patients and to decrease the operative mortality rate. Several post-operative clinical risk scores were developed recently to predict mortality post-hepatectomy; however, which one is the best remains undefined. Therefore, the aim of this study was to evaluate the performance of the different post-operative clinical risk scores in predicting early mortality after hepatectomies.

Methods: A total of 240 patients who underwent liver resection for hepatocellular carcinoma at our hospital between June 2011 and July 2016 were retrospectively reviewed. Post-operative clinical risk scores including 50-50 criteria, peak bilirubin >7 mg/dL, model for end-stage liver disease (MELD), risk assessment for early mortality and Hyder scores were evaluated for their performance in predicting early mortality after hepatic resection using the receiver operating characteristic (ROC) curve.

Results: The 90-day mortality rate after hepatic resection was around 2.5%. The 50-50 criteria and peak bilirubin >7 mg/dL were weak predictors of early mortality with low sensitivity (area under the ROC curve: 0.65, 0.66, respectively), whereas, Hyder, risk assessment for early mortality, and post-operative MELD were good predictors of early mortality (area under the ROC curve: 0.89, 0.91 and 0.88, respectively). Moreover, MELD score on post-operative day 3 was an independent risk factor for 90-day mortality with an odds ratio of 1.4 (95% confidence interval 1.06-1.81, P = 0.02).

Conclusions: Post-operative clinical risk scores, especially MELD, were capable of predicting early mortality after liver resection and should be used to identify high-risk patients and provide them with more intensive medical care.