Predicting short and long-term survival in patients with ESLD still represents a challenge in everyday clinical practice. This is mainly due to several factors, such as the absence of a fixed disease course, the knowledge limitations we still have about many triggers of decompensating events, the multisystemic involvement of this disease. Moreover, the prognostic end-points we want to predict may differ across the same group of patients, according to baseline characteristics, therapeutic opportunities (e.g., overall survival, liver transplantation free survival, disease-free survival in cancer patients), and may change with time, according to superimposed or unexpected conditions (e.g., acute-in-chronic liver failure (ACLF), development of infection).

In the last decade, many efforts have been made to describe the natural history of ESLD, to stratify the risk of decompensation and/or cancer development, and to predict prognosis at any disease stage [1,2]. Given the complex landscape and its protean manifestations, the prediction of short- and long-term survival should not be assessed by a single parameter. Therefore, composite models and scores have been proposed. Some of these adopted so far (such as the CTP score) still have a major role in everyday clinical practice. Others, such as the MELD score have confirmed their strong predictive role and are widely used for organ allocation in the LT setting, with several improvements made over time. Others have recently been developed and validated in specific groups of patients with ESLD (e.g., scores for ACLF) and are now widely adopted.

Others did not stand the test of time.

In this review, we will describe the most important scores, currently used to predict prognosis of ESLD and patients with hepatocellular carcinoma (HCC).