Predicts response to corticosteroid therapy by day 7; cutoff 0.45
Primary source: Louvet A et al. Hepatology 2007; 45(6): 1348-1354
The Lille Model (Louvet A, et al. Hepatology 2007) is a dynamic scoring system for assessing response to corticosteroid therapy in patients with severe alcoholic hepatitis. It is calculated on day 7 after starting prednisolone.
It uses 6 parameters: age, albumin, bilirubin at admission, bilirubin on day 7, PT, and creatinine. The result is a number between 0 and 1 (probability of death by 6 months). Cutoff 0.45: < 0.45 indicates "response" to corticosteroids (continue for up to 28 days), ≥ 0.45 indicates "no response" (discontinue corticosteroids, consider alternatives).
Lille is the only score that assesses dynamic response; the static Maddrey and MELD scores predict outcome at admission but do not answer whether therapy should continue. In the Louvet 2007 cohort, 6-month mortality was 75% in nonresponders vs 25% in responders. Major hepatology society guidelines (AASLD, EASL) include the Lille model as a required step in the management of corticosteroid therapy for severe alcoholic hepatitis.
A 52-year-old man with severe alcoholic hepatitis started on prednisolone 40 mg/day. Age 52, albumin 28 g/L, bilirubin at admission 250, day 7 bilirubin 230 µmol/L (minimal decrease), PT 21 s, creatinine 110 µmol/L.
Using the Lille formula with these parameters ≈ 0.52.
Lille ≥ 0.45 indicates no response to corticosteroids. Management: discontinue prednisolone (continuation is futile and increases infection risk), evaluate for early transplantation at a specialized center (contingent on confirmed abstinence and psychosocial assessment); N-acetylcysteine may be considered, but the evidence base is limited.
Validated scores and formulas with full interpretation, clinical examples and source references. Open access, no registration required.
65+ calculators · scores and formulas · interpretation with management notes