Calculate Maddrey's Discriminant Function (MDF) and Model for End‑Stage Liver Disease (MELD) scores for my patient.

Calculating MDF and MELD Scores

To calculate these prognostic scores, you need specific laboratory values from your patient: prothrombin time (PT), total bilirubin, INR, and serum creatinine. Without these values, the scores cannot be computed, but I will provide you with the exact formulas and interpretation thresholds below.

Maddrey Discriminant Function (MDF) Formula

MDF = 4.6 × (Patient's PT - Control PT) + Total Bilirubin (mg/dL) 1

Key Components:

• Patient's PT: Your patient's prothrombin time in seconds 1
• Control PT: The laboratory's control prothrombin time, which varies by institution (typically 12-14.8 seconds) 2
• Total Bilirubin: Measured in mg/dL 1

Critical Interpretation:

• MDF ≥ 32 indicates severe alcoholic hepatitis with 30-50% mortality at one month and warrants consideration of corticosteroid therapy 1, 3
• MDF < 32 suggests lower short-term mortality risk 1

Important Caveat:

The choice of control PT significantly impacts the MDF calculation and treatment decisions. Using a control PT of 12 seconds versus 14.8 seconds can change whether a patient meets the threshold for steroid therapy (MDF ≥ 32), though this does not appear to impact actual survival 2. Most institutions use a control PT between 12-13.5 seconds 2.

Model for End-Stage Liver Disease (MELD) Formula

MELD = 3.8 × log(bilirubin mg/dL) + 11.2 × log(INR) + 9.6 × log(creatinine mg/dL) + 6.4 1, 4, 5

Key Components:

• Serum bilirubin: In mg/dL, reflecting hepatic excretory function 4, 5
• INR: International normalized ratio, measuring coagulation function 4, 5
• Serum creatinine: In mg/dL, assessing renal function 4, 5

Critical Interpretation:

• MELD ≥ 18-21 indicates poor prognosis in alcoholic hepatitis with significant 90-day mortality 1
• MELD > 20 is the threshold for considering corticosteroid therapy in severe alcoholic hepatitis 3
• MELD ≥ 15 is the general threshold for liver transplant listing 4
• MELD > 26 in severe alcoholic hepatitis warrants urgent transplant evaluation 3

MELD Score Ranges and Mortality:

• MELD 6-11: Low risk 6
• MELD 11-18: Intermediate risk 1
• MELD 18-21: High risk, approximately 20% 90-day mortality 1
• MELD > 21: Very high risk, 20-30% three-month mortality 4

Comparative Performance

Both MDF and MELD perform similarly in predicting short-term mortality in alcoholic hepatitis, with MELD showing slightly superior predictive accuracy 7, 8, 2. The MELD has demonstrated better ability to predict 30-day survival compared to MDF regardless of which control PT is used (AUROC 0.767 vs 0.735) 2.

Advantages of Each Score:

MDF Advantages:

• Disease-specific for alcoholic hepatitis 1
• Simpler calculation requiring only PT and bilirubin 1
• Well-validated threshold (≥32) for treatment decisions 1, 3

MELD Advantages:

• Entirely objective with no subjective parameters 4, 5
• Incorporates renal function (creatinine), a critical prognostic marker 4, 5
• Continuous scale (6-40) providing more granular risk stratification 4
• Better validated across multiple liver disease etiologies 4
• Superior predictive accuracy for 30-day mortality 2

Clinical Application Algorithm

For patients with suspected alcoholic hepatitis, calculate BOTH scores 1, 3:

1. Obtain required labs: PT, INR, total bilirubin, serum creatinine 1, 4
2. Calculate MDF using your institution's control PT (document which control PT you used) 2
3. Calculate MELD using the logarithmic formula or online calculator 1
4. Interpret severity:
   • Severe disease: MDF ≥ 32 OR MELD > 20 3
   • Very severe disease: MELD > 26 3
5. Consider corticosteroid therapy (prednisolone 40mg or methylprednisolone 32mg daily) if MDF ≥ 32 or MELD > 20 and no contraindications 3
6. Reassess at day 7 using Lille score to determine corticosteroid response 3
7. Consider transplant evaluation if MELD > 26 with good insight and social support 3

Common Pitfalls to Avoid

• Do not use MDF or MELD in isolation—both scores have excellent negative predictive values (86-100%) but poor positive predictive values (17-50%), meaning they are best at identifying low-risk patients 7
• Be aware that different control PT values will change MDF results and treatment decisions, though not necessarily outcomes 2
• MELD can be artificially elevated in patients on warfarin or with Fontan circulation due to elevated INR independent of liver function 4
• Serum creatinine may overestimate renal dysfunction in sarcopenic patients or underestimate it in fluid-overloaded patients 4
• Neither score captures clinical parameters like ascites or hepatic encephalopathy, which significantly impact prognosis 4, 5