About the MELD Score
The Model for End-Stage Liver Disease (MELD) score was originally developed to predict 3-month mortality after transjugular intrahepatic portosystemic shunting (TIPS) and was adopted by UNOS in 2002 to prioritize liver transplant allocation. It replaced the Child-Pugh score for organ allocation due to its objectivity — all variables are laboratory-based, eliminating subjective clinical assessments.
The MELD score uses three serum laboratory values — total bilirubin, INR, and creatinine — each reflecting a distinct axis of liver dysfunction: hepatocellular function (bilirubin), coagulation capacity (INR), and renal function or hepatorenal physiology (creatinine). The logarithmic transformation of each variable ensures that the score responds proportionally to changes across the physiological range rather than reacting linearly to extreme values. This mathematical design makes MELD particularly sensitive to changes in disease severity over short time periods, which is why it is used to assess for disease trajectory and urgency of transplant intervention.
Since its adoption for transplant allocation, MELD has undergone one major revision: the addition of serum sodium in 2016 to create MELD-Na. Studies from the UNOS database demonstrated that cirrhotic hyponatremia (serum sodium below 137 mEq/L) independently predicted 90-day waitlist mortality beyond what MELD alone captured — particularly in patients with ascites and diuretic-induced sodium disturbances. MELD-Na is now the standard score used in UNOS/OPTN organ allocation decisions for deceased donor liver transplantation.
How to Use This Calculator
Enter the four required laboratory values for your patient: total bilirubin (mg/dL), INR, serum creatinine (mg/dL), and serum sodium (mEq/L). All values should reflect the most recent laboratory draw taken in the context of the patient's current clinical status — not values obtained during an acute reversible event (see Limitations below). If the patient has been on dialysis at least twice in the preceding week, check the dialysis checkbox: this automatically sets creatinine to 4.0 mg/dL per UNOS/OPTN protocol.
The calculator computes both MELD and MELD-Na simultaneously. MELD-Na is the score used for UNOS organ allocation and should be the primary clinical reference when assessing transplant priority. The MELD score (without sodium) is shown for reference and remains useful for some prognosis comparisons and non-transplant clinical contexts where sodium status is separately addressed.
MELD Formula
MELD = 3.78 × ln(bilirubin) + 11.2 × ln(INR) + 9.57 × ln(creatinine) + 6.43
Floor values: bilirubin, INR, and creatinine all floored at 1.0 mg/dL. Creatinine capped at 4.0 mg/dL (also set to 4.0 if dialysis ≥2× in past week). Score rounded to nearest integer.
MELD-Na Formula
MELD-Na = MELD + 1.32 × (137 − Na) − [0.033 × MELD × (137 − Na)]
Sodium is constrained to 125–137 mEq/L. MELD-Na capped at 40. Used by UNOS since 2016 for deceased donor liver allocation — it better captures mortality risk from hyponatremia in cirrhosis.
Interpretation Guide: 90-Day Mortality Estimates
| MELD Score | 90-Day Mortality | Clinical Context |
|---|---|---|
| <10 | ~1.9% | Compensated cirrhosis; outpatient management |
| 10–19 | ~6.0% | Moderate disease; close outpatient monitoring |
| 20–29 | ~19.6% | Decompensated; transplant evaluation warranted |
| 30–39 | ~52.6% | Severe; active transplant listing consideration |
| ≥40 | ~71.3% | Critical; high-urgency listing (Status 1A/1B criteria) |
The 90-day mortality estimates above are derived from the original MELD validation study by Malinchoc et al. and subsequent UNOS waitlist mortality data. These values represent population-level estimates and individual patient outcomes may differ based on etiology of liver disease, presence of active complications (variceal bleeding, spontaneous bacterial peritonitis, hepatocellular carcinoma), nutritional status, and comorbidities. MELD trajectory — whether the score is rising or stable over weeks — is as clinically significant as the absolute value.
Transplant Listing Threshold
UNOS data suggest that the survival benefit of liver transplantation begins to exceed the surgical mortality risk at a MELD ≥15. Below this threshold, the risk of transplant may outweigh the benefit in otherwise stable patients. Most transplant centers use MELD ≥15 as the threshold for active listing consideration.
Patients with hepatocellular carcinoma (HCC) meeting Milan criteria may receive MELD exception points to reflect their malignancy-related risk not captured by laboratory MELD.
Limitations & Considerations
Not validated for acute liver failure: MELD was derived and validated in patients with chronic liver disease and cirrhosis. In acute liver failure (e.g., acetaminophen toxicity, acute viral hepatitis), the King's College Criteria or AALF criteria are more appropriate prognostic tools. MELD may not accurately reflect severity or prognosis in this setting.
Reversible laboratory abnormalities: Several conditions can transiently elevate MELD without reflecting true disease severity. Acute kidney injury (from hepatorenal syndrome, infections, nephrotoxins, or contrast) may elevate creatinine. Anticoagulation, malnutrition, and vitamin K deficiency may elevate INR. Hemolysis or biliary obstruction unrelated to hepatocellular disease may raise bilirubin. MELD should be calculated when the patient is in their clinical steady state, not during an acute, potentially reversible deterioration that has not yet been treated.
MELD exception points: This calculator computes the laboratory MELD-Na only. It does not account for MELD exception points, which transplant programs can petition UNOS to assign for conditions including hepatocellular carcinoma (within Milan criteria), hepatopulmonary syndrome, portopulmonary hypertension, and other specific diagnoses. The actual allocation score used by UNOS may differ from the laboratory-calculated MELD-Na shown here for patients with approved exceptions.
Age and non-cirrhotic liver disease: MELD was developed in adults and has not been validated in pediatric patients, for whom the PELD (Pediatric End-Stage Liver Disease) score is used. MELD is also not intended for risk stratification in non-cirrhotic chronic liver disease without evidence of portal hypertension or hepatic decompensation.
References
Malinchoc M, et al. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology. 2000;31(4):864–871.
Kim WR, et al. Hyponatremia and mortality among patients on the liver-transplant waiting list. N Engl J Med. 2008;359:1018–1026.
Kamath PS, et al. A model to predict survival in patients with end-stage liver disease. Hepatology. 2001;33(2):464–470.
UNOS/OPTN Policy 9: Allocation of Livers and Liver-Intestines. Available at: optn.transplant.hrsa.gov.