What is a MELD Score?
The Model for End-Stage Liver Disease (MELD) score is an objective numerical scale ranging from 6 (less ill) to 40 (gravely ill) that predicts 3-month mortality risk in patients with end-stage liver disease using three laboratory values: serum bilirubin, serum creatinine, and international normalized ratio (INR). 1
Formula and Calculation
The MELD score is calculated using the following formula: MELD = 3.8 × log(bilirubin in mg/dL) + 11.2 × log(INR) + 9.6 × log(creatinine mg/dL) + 6.4 1
- All three components are objective laboratory measurements, eliminating subjective clinical assessments 1
- The score provides a continuous scale that directly correlates with estimated 3-month survival rates from 90% (score of 6) to 7% (score of 40) 1
- Online calculators are available for ease of use 1
Original Development and Purpose
The MELD score was originally developed in 2001 to assess short-term prognosis in patients undergoing transjugular intrahepatic portosystemic shunts (TIPS) 1, 2
- The model identified serum bilirubin, INR, and serum creatinine as the best predictors of 3-month postoperative survival 1
- It has since been adopted by the United Network for Organ Sharing (UNOS) to stratify patients on liver transplantation waiting lists according to their risk of death within 3 months 1
Clinical Applications
Liver Transplantation Allocation
A MELD score ≥15 is the recommended threshold for listing patients for liver transplantation, as patients with MELD <15 may have lower 1-year survival with transplantation compared to without 1, 2
- The MELD score is the primary basis for organ allocation in many countries, including the United States 1
- Implementation in 2002 led to a sharp reduction in waiting list numbers and mortality without affecting post-transplant survival 3
Prognostic Thresholds
Different MELD cutoffs indicate varying levels of mortality risk:
- MELD ≥18: Indicates poor prognosis in alcoholic hepatitis with significant short-term mortality risk 1
- MELD >30: Patients face increased risk of mortality and morbidity after transplantation and require careful assessment 1
- MELD >35: Associated with very high post-transplant mortality risk 1
Advantages Over Child-Pugh Score
The MELD score offers several advantages compared to the Child-Turcotte-Pugh (CTP) classification 1, 2:
- Uses only objective laboratory tests rather than subjective clinical assessments of ascites and hepatic encephalopathy 1, 2
- Includes renal function (serum creatinine), which is an established prognostic marker in liver disease 1, 2
- Provides a continuous numerical scale rather than categorical classification 2
- Eliminates bedside interpretation variability inherent in clinical assessments 1
The two overlapping parameters between MELD and CTP are serum bilirubin and prothrombin time/INR, while CTP additionally includes albumin, ascites, and encephalopathy 2
Limitations and Exceptions
Laboratory Variability
Interlaboratory variability in INR measurements has the largest impact on MELD scores, with mean differences of approximately 5 MELD points between laboratories 4, 5
- Creatinine variability can account for differences up to 3 MELD points 5
- Bilirubin measurements contribute minimally, accounting for only 1 point difference 5
- Serum creatinine can be an unreliable marker of renal dysfunction in patients with cirrhosis 1
MELD Exceptions
Several conditions require "exception points" because their severity is not adequately captured by the MELD formula 1, 2:
- Hepatocellular carcinoma (HCC): Receives additional points based on tumor size, number of nodules, AFP levels, waiting time, and response to downstaging 1
- Hepatopulmonary syndrome 1
- Portopulmonary hypertension 1
- Primary hyperoxaluria 1
- Amyloidosis 1
- Recurrent cholangitis 6
Validation Limitations
The MELD score has not been validated as a predictor of survival in cirrhotic patients who are not on a liver transplantation waiting list 1, 2
- It remains unclear whether MELD is superior to Child-Pugh score for predicting survival in non-transplant cirrhotic patients 1, 2
- The MELD model favors patients with renal dysfunction, which may not always reflect liver disease severity 1
Modified MELD Scores
Several variations have been proposed to improve predictive accuracy 1, 2:
- MELD-Na: Incorporates serum sodium levels 1, 2
- Integrated MELD: Adds serum sodium and patient age 1
- Delta MELD (DMELD): Measures change in MELD over time, which may be a better predictor of mortality 1, 2
- MELD 3.0: The newest version includes patient sex and serum albumin with updated weights for other variables 7
Clinical Monitoring Recommendations
Serial calculation of MELD scores over time is justified for evaluating disease progression and adjusting transplant priority 1, 6