MELD 3.0

Overview

MELD 3.0 is a 2021 refinement of the Model for End-Stage Liver Disease, developed by Kim and colleagues to correct systematic sex-based and muscle-mass-based bias in the original score. It adds a female-sex term, serum sodium, and serum albumin to the classic bilirubin/INR/creatinine equation ^[1]. OPTN/UNOS adopted MELD 3.0 as the standard liver allocation score in the United States on 13 July 2023 ^[2], replacing MELD-Na, which itself had been the allocation standard since January 2016.

The original MELD equation was derived by Malinchoc and colleagues to predict survival after transjugular intrahepatic portosystemic shunt (TIPS) placement ^[3], then validated more broadly as a general prognostic score for end-stage liver disease by Kamath and colleagues, whose formulation became the basis for UNOS liver allocation in 2002 ^[4].

The central motivation for MELD 3.0 was evidence that women were systematically disadvantaged under the original formula: at a given true glomerular filtration rate, women have lower serum creatinine than men because of lower average muscle mass, so a creatinine-based estimate under-states the severity of renal impairment in female candidates and reduces their calculated priority on the waiting list ^[1]. The sex term and the addition of albumin — a marker sensitive to nutritional and muscle status more broadly — were designed to correct this disparity in access to transplantation.

As implemented in Scores2Go, MELD 3.0 applies to adult candidates only (age ≥ 18 years). OPTN uses a separate equation without the sex term for candidates aged 12–17, which is not implemented here.

Formula

MELD 3.0 = (female ? 1.33 : 0)
         + 4.56 × ln(Bilirubin)
         + 0.82 × (137 − Na)                          [sodium term]
         − 0.24 × (137 − Na) × ln(Bilirubin)           [sodium × bilirubin interaction]
         + 9.09 × ln(INR)
         + 11.14 × ln(Creatinine)
         + 1.85 × (3.5 − Albumin)                      [albumin term]
         − 1.83 × (3.5 − Albumin) × ln(Creatinine)     [albumin × creatinine interaction]
         + 6

All values in mg/dL (bilirubin, creatinine), mmol/L (sodium), g/dL (albumin).
Each input is clamped per the table below BEFORE the formula is evaluated.
Result rounded to the nearest integer, clamped 6–40.

Variables & Bounds

Variable	Floor	Ceiling	Special
Sex	—	—	Female adds a fixed +1.33 points; not a clamp
Bilirubin	1.0 mg/dL	—	—
INR	1.0	—	—
Creatinine	1.0 mg/dL	3.0 mg/dL	Set to 3.0 mg/dL if on dialysis (≥ 2 sessions in the prior week, or ≥ 24 h of CVVHD)
Sodium	125 mmol/L	137 mmol/L	—
Albumin	1.5 g/dL	3.5 g/dL	—
MELD 3.0 total	6	40	—

The creatinine ceiling is 3.0 mg/dL in MELD 3.0, down from 4.0 mg/dL in the original MELD score — this narrower band, combined with the sex and albumin terms, is part of what redistributes priority away from a pure creatinine-driven estimate of renal function.

Interpretation

MELD 3.0 Score	Severity	90-day Mortality
< 10	Low	1.9 %
10–19	Moderate	6.0 %
20–29	High	19.6 %
30–39	Very high	52.6 %
≥ 40	Very high	71.3 %

These precise 90-day mortality figures are drawn from the original UNOS validation analysis of the MELD-based allocation system by Wiesner and colleagues ^[5], and are used unchanged for the corresponding MELD 3.0 score bands.

Version History

MELD progressed through three major stages before reaching MELD 3.0. The original MELD equation, derived from a TIPS cohort ^[3] and formalised for general liver-disease prognosis ^[4], became UNOS's primary liver allocation score in 2002.

In 2006, Biggins and colleagues showed that incorporating serum sodium materially improved 90-day mortality prediction, particularly in patients with mild-to-moderate MELD elevations ^[6]. The resulting MELD-Na score — which applies the sodium term only when the calculated MELD exceeds 11, and bounds sodium to 125–137 mmol/L — was adopted by OPTN as the primary US allocation algorithm on 11 January 2016. MELD-Na is documented here for historical context only; Scores2Go does not implement it as a separate score.

MELD 3.0, published in 2021 ^[1], went into effect as OPTN policy on 13 July 2023 ^[2], replacing MELD-Na and adding the sex term and albumin described above.

Scientific Validity & Limitations

Sex disparity and creatinine/muscle mass

Creatinine at a given true GFR is lower in women because of lower average muscle mass, so the original MELD systematically under-scored female candidates and reduced their transplant priority — the direct motivation for the +1.33 sex term and for the inclusion of albumin ^[1]. The same muscle-mass bias affects sarcopenic or cachectic patients of any sex, a finding that is common in advanced cirrhosis and is only partially addressed by albumin's inclusion.

INR reproducibility and anticoagulation

The same blood sample can yield INR values that differ by roughly 20–26 % between laboratories, because thromboplastin ISI calibration is validated against warfarin-anticoagulated patients rather than against cirrhotic coagulopathy — this alone can shift the computed MELD score by several points depending on which laboratory performed the assay ^[7]. Vitamin K antagonists and direct oral anticoagulants (DOACs) also inflate INR independently of hepatic synthetic function. MELD-XI (MELD-XI = 5.11 × ln(bilirubin) + 11.76 × ln(creatinine) + 9.44), which omits the INR term entirely, was developed specifically for anticoagulated cirrhotic patients ^[8].

What MELD 3.0 still misses, and exception points

MELD 3.0 still does not take refractory ascites, hepatic encephalopathy, or recurrent cholangitis as direct inputs. Hyponatraemia itself was shown to be an independent predictor of waiting-list mortality by Kim and colleagues ^[9] — the finding that motivated adding sodium first to MELD-Na and then to MELD 3.0. Standardised MELD exception points, covering conditions such as hepatocellular carcinoma, hepatopulmonary syndrome, and familial amyloid polyneuropathy, exist within OPTN policy ^[2] precisely because laboratory MELD understates urgency for some candidates.

Eurotransplant divergence

Eurotransplant (covering Germany, Austria, the Benelux countries, Slovenia, Croatia, and Hungary) has never used MELD-Na and does not use MELD 3.0. It moved directly from the unmodified original labMELD to labReMELD-Na, a sodium-inclusive score refitted on Eurotransplant registry data because the US MELD-Na sodium coefficients did not transfer well to the Eurotransplant population ^[10]; the corresponding Bundesärztekammer Richtlinie was published on 25 March 2025 and took effect on 1 April 2025. Eurotransplant's labReMELD-Na uses a third, distinct set of variable bounds — creatinine 0.7–2.5 mg/dL (2.5 mg/dL if on dialysis), INR 1.0–2.6 (set to 1.0 for patients on vitamin K antagonists or DOACs), bilirubin 0.3–27 mg/dL, and sodium 120–138.6 mmol/L. Its coefficients are not published in the open literature — they are contained in Eurotransplant's non-public ELAS manual — so only its inputs and bounds are described here, not its formula. Clinicians in the Eurotransplant area should not treat MELD 3.0 as their region's allocation score.

Related Scores

MELD 3.0 sits alongside several related prognostic and allocation scores that are not currently implemented in Scores2Go: Child-Pugh (the original liver-severity score MELD superseded for allocation purposes), MELD-XI (the INR-free variant for anticoagulated patients, described above), PELD (the paediatric analogue used for candidates under 12), UKELD (the UK's own MELD-derived allocation score), ALBI (the Albumin-Bilirubin grade, developed primarily for hepatocellular carcinoma), and CLIF-C ACLF (a score specific to acute-on-chronic liver failure). Scores2Go's original, unmodified MELD score is implemented separately — see the MELD page.

Literature

1. Kim WR, Mannalithara A, Heimbach JK, et al. MELD 3.0: The Model for End-Stage Liver Disease Updated for the Modern Era. Gastroenterology. 2021;161(6):1887–1895.e4. PMID: 34481845.
2. OPTN Policy 9: Allocation of Livers and Liver-Intestines. Effective 13 July 2023.
3. Malinchoc M, Kamath PS, Gordon FD, Peine CJ, Rank J, ter Borg PC. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology. 2000;31(4):864–871.
4. Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology. 2001;33(2):464–470.
5. Wiesner R, Edwards E, Freeman R, et al.; United Network for Organ Sharing Liver Disease Severity Score Committee. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology. 2003;124(1):91–96.
6. Biggins SW, Kim WR, Terrault NA, et al. Evidence-based incorporation of serum sodium concentration into MELD. Gastroenterology. 2006;130(6):1652–1660. PMID: 16697729.
7. Porte RJ, Lisman T, Tripodi A, Caldwell SH, Trotter JF; Coagulation in Liver Disease Study Group. The International Normalized Ratio (INR) in the MELD Score: Problems and Solutions. Am J Transplant. 2010;10(6):1349–1353.
8. Heuman DM, Mihas AA, Habib A, et al. MELD-XI: a rational approach to "sickest first" liver transplantation in cirrhotic patients requiring anticoagulant therapy. Liver Transpl. 2007;13(1):30–37. PMID: 17154400.
9. Kim WR, Biggins SW, Kremers WK, et al. Hyponatremia and mortality among patients on the liver-transplant waiting list. N Engl J Med. 2008;359(10):1018–1026. doi:10.1056/NEJMoa0801209.
10. Goudsmit BFJ, Putter H, Tushuizen ME, et al. Refitting the Model for End-Stage Liver Disease for the Eurotransplant Region. Hepatology. 2021;74(1):351–363. doi:10.1002/hep.31677.