MELD Score Calculator (Liver Disease)

MELD converts three laboratory values into a number from 6 to 40 that estimates three-month mortality in chronic liver disease. The formula, 10 times the sum of 0.957 times the natural log of creatinine, 0.378 times the log of bilirubin, 1.120 times the log of INR and 0.643, was fitted by Cox proportional hazards regression, which is why the coefficients are irregular and the inputs logarithmic: risk rises with proportional rather than absolute changes in each value. Bilirubin measures excretory failure, INR synthetic failure, and creatinine the hepatorenal circulatory failure that carries the heaviest coefficient of the three. Values below 1.0 are set to 1.0 to avoid negative logarithms, creatinine is capped at 4.0 mg/dL, and dialysis twice in the preceding week sets creatinine to 4.0. MELD-Na adds serum sodium, bounded between 125 and 137 mEq/L, because dilutional hyponatremia driven by antidiuretic hormone excess independently predicts death and can add several points to the score. Approximate three-month mortality runs about 2 percent below a score of 10, 6 percent from 10 to 19, 20 percent from 20 to 29, 53 percent from 30 to 39, and 71 percent at 40. The limitations are structural rather than incidental. MELD ignores ascites, encephalopathy and variceal bleeding, is inflated by anticoagulants that raise INR, is distorted by primary kidney disease and by the lower baseline creatinine of women and smaller patients, and describes a single moment rather than the direction of travel.

The model was not built for transplant allocation. Patrick Kamath, Michael Malinchoc and colleagues at the Mayo Clinic devised it to predict survival after transjugular intrahepatic portosystemic shunt placement, publishing the underlying risk score in 2000 and the version known as MELD in Hepatology in 2001. When the United States sought an objective replacement for a waiting list system built on Child-Pugh class and accumulated waiting time, MELD fit the requirement precisely, because every input was a laboratory number that could not be influenced by subjective assessment or by how long a patient had been listed. The United Network for Organ Sharing implemented MELD for adult deceased donor liver allocation on 27 February 2002, alongside PELD for children, and time on the list ceased to drive priority as sicker patients moved upward. Revisions followed the evidence. Ray Kim and colleagues reported in the New England Journal of Medicine in 2008 that hyponatremia independently predicted waiting list mortality, and MELD-Na became the allocation score in January 2016. In 2023 the system moved to MELD 3.0, which adds female sex and serum albumin as terms and adjusts the creatinine ceiling, correcting a documented disadvantage that earlier versions imposed on women. This calculator implements the original MELD and MELD-Na formulas.