Prognostic models are useful for estimating disease severity and survival, and can serve as helpful medical decision-making tools with respect to guiding patient care. These models are developed using statistical methodologies that involve determining the effects of variables of interest (eg, demographic data, clinical data, and laboratory values) on specific outcomes such as death.
Several prognostic models are currently used in healthcare settings. Some focus on generalized health status such as the Acute Physiology and Chronic Health Evaluation System (APACHE III) , while others are disease specific. Examples of the latter in the field of hepatology include models for predicting survival in patients with primary biliary cirrhosis, primary sclerosing cholangitis, and alcoholic liver disease [2-5]. Two models that are used commonly in the care of patients with cirrhosis are the Child-Turcotte-Pugh score and the Model for End-stage Liver Disease (MELD) score [6-10].
This topic will review the development, use, impact, refinements, and limitations of the MELD score, particularly with regard to its use in allocating organs for liver transplantation. Other issues related to the selection of patients for liver transplantation are discussed separately. (See "Patient selection for liver transplantation".)
The Model for End-stage Liver Disease (MELD) is a prospectively developed and validated chronic liver disease severity scoring system that uses a patient's laboratory values for serum bilirubin, serum creatinine, and the international normalized ratio (INR) for prothrombin time to predict three-month survival (calculator 1 and calculator 2). In patients with cirrhosis, an increasing MELD score is associated with increasing severity of hepatic dysfunction and increased three-month mortality risk (figure 1) . Given its accuracy in predicting short-term survival among patients with cirrhosis, MELD was adopted by the United Network for Organ Sharing (UNOS) in 2002 for prioritization of patients awaiting liver transplantation in the United States. (See 'Adoption of MELD for organ allocation' below.)
Development of the MELD score — MELD was originally developed to predict three- month mortality following transjugular intrahepatic portosystemic shunt (TIPS) placement and was derived using data from a population of 231 patients with cirrhosis who underwent elective TIPS placement. The model was subsequently validated in an independent cohort of patients from the Netherlands undergoing TIPS placement . The original model included serum bilirubin, serum creatinine, INR, and etiology of the liver disease (cholestatic or alcoholic versus other etiologies).