Calculate the Lille Model score for corticosteroid response in severe alcoholic hepatitis. Assesses 7-day bilirubin response to guide steroid discontinuation.
The Lille Score assesses response to corticosteroid therapy in patients with severe alcoholic hepatitis, typically after 7 days of prednisolone treatment. Developed from a French multicenter cohort, it uses age, albumin, Day 0 and Day 7 bilirubin values, creatinine, and prothrombin time to predict 6-month mortality and identify patients who are non-responders to steroids.
Severe alcoholic hepatitis (Maddrey discriminant function ≥32) carries 30-50% short-term mortality without treatment. Prednisolone 40mg daily for 28 days is the current standard of care, improving 28-day survival by approximately 15-20%. However, approximately 40% of patients do not respond to steroids, and continuing futile immunosuppression in non-responders increases infection risk.
The Lille score at Day 7 identifies three groups: complete responders (Lille <0.16, ~15% 6-month mortality), partial responders (0.16-0.56, ~25%), and non-responders (≥0.56, ~75% mortality). Non-responders should have steroids discontinued. Check the example with realistic values before reporting. Use the steps shown to verify rounding and units. Cross-check this output using a known reference case.
Without the Lille score, clinicians face a difficult dilemma: continue steroids (risking infection) or stop them (risking loss of benefit). The Lille model provides an evidence-based, quantitative answer to this question at the 7-day mark.
In the landmark STOPAH trial, Lille remained the strongest predictor of outcome regardless of initial treatment assignment. Early identification of non-responders enables redirection of care toward transplant evaluation, infection monitoring, and supportive measures.
Lille Score = exp(-R) / (1 + exp(-R)) R = 3.19 − 0.101 × age + 0.147 × albumin(g/L) + 0.0165 × Δbilirubin(μmol/L) − 0.206 × renal insufficiency − 0.0065 × bilirubin Day 0(μmol/L) − 0.0096 × PT(seconds) Renal insufficiency = 1 if creatinine >1.3 mg/dL Δbilirubin = Day 0 − Day 7 bilirubin (μmol/L)
Result: Lille 0.18 — Partial Responder
Bilirubin decreased from 12 to 8 mg/dL (33% drop), with Lille 0.18 indicating partial response. Steroids should be continued for the full 28-day course. The bilirubin decline is a favorable sign, and continued monitoring for infection and hepatorenal syndrome is essential.
The typical management pathway: (1) Confirm diagnosis (clinical, labs, consider biopsy if uncertain), (2) Calculate Maddrey DF — if ≥32 or MELD ≥21, patient has "severe" AH, (3) Screen for infection (contraindication to immediate steroids), (4) Start prednisolone 40mg daily + NAC, (5) Calculate Lille at Day 7, (6) If Lille <0.56: continue steroids for 28 days total; if ≥0.56: discontinue steroids, (7) Refer non-responders for transplant evaluation if appropriate.
The STOPAH trial (largest AH treatment trial, n=1,103) confirmed a modest benefit of prednisolone on 28-day mortality (OR 0.72) but no benefit at 90 days or 1 year. Pentoxifylline showed no benefit. This has motivated research into new therapies: granulocyte colony-stimulating factor (G-CSF), fecal microbiota transplantation, IL-22, and obeticholic acid are under investigation.
The 2011 Mathurin study showed that highly selected patients with severe AH and steroid non-response who underwent early transplantation had 77% 6-month survival vs 23% in controls. Strict selection criteria (first liver decompensation, no prior treatment attempts, strong social support, favorable addiction psychiatry assessment) are essential.
Day 7 is defined as 7 full days after starting prednisolone. Draw the morning bilirubin on Day 7 of treatment (not Day 7 of hospitalization, which may differ). Some centers also use Day 4 bilirubin for earlier assessment using modified Lille models.
The Lille model is probabilistic, not absolute. If there is clear clinical improvement (decreasing ascites, improving encephalopathy, falling bilirubin after Day 7), clinical judgment may override the Lille score. However, non-response by Lille remains a strong prognostic indicator.
Yes. Early liver transplantation for severe alcoholic hepatitis (without the traditional 6-month sobriety requirement) has shown 70-80% 5-year survival in highly selected patients at specialized centers. Lille non-response is one of the criteria prompting transplant evaluation in recent protocols.
The Maddrey DF = 4.6 × (PT − control PT) + bilirubin (mg/dL). DF ≥32 defines "severe" alcoholic hepatitis and is the threshold for considering corticosteroid therapy. The Lille score is only applicable to patients who have started steroids for severe AH (typically DF ≥32).
The STOPAH trial showed pentoxifylline has no benefit in severe AH. N-acetylcysteine (NAC) combined with prednisolone reduced 1-month infection and hepatorenal syndrome rates compared to prednisolone alone. Current practice at many centers: prednisolone + NAC for 28 days, with Lille assessment at Day 7.
Younger age, lower baseline bilirubin, preserved renal function, and higher albumin are associated with better Lille responses. Infection at presentation and ACLF grade 3 predict non-response. Early aggressive nutritional support (35-40 kcal/kg/day) may also improve steroid response rates.