What is Alcoholic Hepatitis?
Alcoholic hepatitis is a serious inflammatory condition of the liver caused by excessive and prolonged alcohol consumption. It represents one of the most severe forms of alcohol-related liver disease, sitting on a spectrum that ranges from simple steatosis (fatty liver) to fibrosis, cirrhosis, and hepatocellular carcinoma. Unlike simple fatty liver, which is often reversible with abstinence, alcoholic hepatitis involves acute hepatocellular damage, neutrophilic infiltration, and hepatocyte ballooning degeneration.
The clinical presentation of severe alcoholic hepatitis is dramatic: patients typically present with rapid-onset jaundice, fever, tender hepatomegaly, and signs of systemic inflammation. Laboratory findings reveal markedly elevated bilirubin, prolonged prothrombin time, elevated white blood cell count, and often an AST:ALT ratio greater than 2:1. The condition carries significant short-term mortality — untreated severe alcoholic hepatitis has a 28-day mortality rate of 30–50%, making early risk stratification and treatment decisions critically important.
Risk factors include heavy alcohol use (typically more than 80 g/day for men and 60 g/day for women over many years), female sex, genetic predisposition, obesity, and nutritional deficiencies. While alcohol cessation remains the single most important intervention, patients with severe disease often require pharmacological treatment to improve survival.
Maddrey Discriminant Function
The Maddrey Discriminant Function (mDF), introduced by Maddrey and colleagues in 1978, was one of the first scoring systems developed specifically for alcoholic hepatitis. It uses a simple formula based on prothrombin time and bilirubin to identify patients with severe disease who may benefit from corticosteroid therapy.
A score of 32 or greater indicates severe alcoholic hepatitis and is the traditional threshold for initiating corticosteroid treatment. The mDF has been widely validated and remains a cornerstone of initial assessment. However, it has limitations: it is measured at a single time point and cannot assess treatment response once steroids have been started. This is precisely the gap that the Lille model was designed to fill.
Other scoring systems used in alcoholic hepatitis include the MELD score (Model for End-Stage Liver Disease), the Glasgow Alcoholic Hepatitis Score (GAHS), and the ABIC score. Each has advantages and limitations, but the Lille model uniquely addresses the question of steroid responsiveness during treatment.
Steroid Therapy in Alcoholic Hepatitis
Corticosteroids, specifically prednisolone at 40 mg/day, represent the standard pharmacological treatment for severe alcoholic hepatitis (mDF ≥ 32). The rationale for steroid use is based on the understanding that much of the liver damage in alcoholic hepatitis is immune-mediated, driven by excessive inflammatory cytokine production, particularly tumor necrosis factor alpha (TNF-α).
Multiple randomized controlled trials and meta-analyses have evaluated the efficacy of corticosteroids in severe alcoholic hepatitis. The evidence suggests that steroids provide a modest but significant reduction in short-term mortality (28-day), with a number needed to treat of approximately 5. The landmark STOPAH trial (2015), the largest RCT on this topic, confirmed a trend toward reduced 28-day mortality with prednisolone, though the benefit did not reach statistical significance at 90 days or 1 year.
The critical clinical challenge is that not all patients respond to steroids, and continued steroid use in non-responders exposes patients to risks of infection, gastrointestinal bleeding, and hyperglycemia without therapeutic benefit. This is why early identification of non-responders is essential, and it is the primary purpose of the Lille score.
The Lille Model (Louvet 2007)
The Lille model was developed and validated by Alexandre Louvet and colleagues at the University of Lille, France, and published in the journal Hepatology in 2007. The study analyzed data from 295 patients with biopsy-proven severe alcoholic hepatitis who received prednisolone 40 mg/day. The researchers aimed to create a prognostic model that could identify non-responders early in the treatment course.
The key innovation of the Lille model is its incorporation of the change in bilirubin between day 0 and day 7 of steroid treatment. This dynamic measurement captures the liver's functional response to anti-inflammatory therapy. Patients whose bilirubin decreases after one week of steroids are demonstrating a positive treatment response, while those with stable or increasing bilirubin are likely non-responders.
The model incorporates six variables: age, albumin at day 0, bilirubin at day 0, bilirubin at day 7, creatinine, and prothrombin time. These variables were selected through multivariate logistic regression analysis and collectively provide a comprehensive assessment of liver function, renal function, and treatment response. The resulting score ranges from 0 to 1, with higher values indicating poorer prognosis and lack of steroid response.
Lille Score Formula
The Lille score is calculated using the following formula. Note that laboratory values must be converted to SI units for the calculation:
Unit conversions:
- Albumin: g/dL × 10 = g/L
- Bilirubin: mg/dL × 17.1 = μmol/L
- Creatinine: mg/dL × 88.4 = μmol/L (Renal insufficiency = 1 if creatinine ≥ 115 μmol/L, else 0)
Interpreting the Score
| Lille Score | Response Category | 6-Month Survival | Recommendation |
|---|---|---|---|
| < 0.16 | Complete Responder | ~85% | Continue steroids for full 28-day course |
| 0.16 – 0.56 | Partial Responder | ~50–70% | Consider continuing steroids; close monitoring |
| > 0.56 | Null Responder | ~25% | Stop steroids; consider liver transplant evaluation |
The original Louvet study identified 0.45 as the optimal cutoff, but subsequent analyses have refined the interpretation into three response categories. The 0.16 threshold identifies complete responders with excellent prognosis, while the 0.56 threshold identifies null responders who derive no benefit from continued steroid therapy.
When to Stop Steroids
The most important clinical application of the Lille score is determining when to stop corticosteroid therapy. In patients with a Lille score greater than 0.56 at day 7, the evidence strongly supports discontinuing prednisolone. These patients are classified as null responders, and continuing steroids provides no survival benefit while increasing the risk of serious complications including sepsis, which is a leading cause of death in alcoholic hepatitis.
For complete responders (Lille < 0.16), the full 28-day course of prednisolone is recommended, followed by a taper. Partial responders (Lille 0.16–0.56) represent a gray zone; clinical judgment, patient preferences, and the overall clinical trajectory should guide decisions in this group. Some clinicians continue steroids for partial responders while closely monitoring for complications.
After stopping steroids in null responders, management focuses on supportive care, nutritional optimization, treatment of complications (ascites, hepatic encephalopathy, infections), and evaluation for early liver transplantation in appropriate candidates. Pentoxifylline was previously used as an alternative, but the STOPAH trial demonstrated no benefit, and it is no longer recommended.
Prognosis and Outcomes
The prognosis of severe alcoholic hepatitis depends heavily on the response to initial treatment, which the Lille score quantifies. Complete responders have a 6-month survival rate approaching 85%, while null responders face a grim 6-month survival of approximately 25%. The 28-day mortality for the entire cohort of severe alcoholic hepatitis patients ranges from 15% to 30%, even with treatment.
Long-term outcomes are strongly influenced by alcohol abstinence after the acute episode. Patients who maintain sobriety have significantly improved long-term survival compared to those who relapse to drinking. Liver transplantation has emerged as a life-saving option for select patients with severe alcoholic hepatitis who fail medical therapy, with carefully selected patients achieving 6-month survival rates exceeding 75% post-transplant.
Other factors influencing prognosis include the development of hepatorenal syndrome, spontaneous bacterial peritonitis, gastrointestinal bleeding, and the presence of underlying cirrhosis. The MELD score at day 0 and day 7 can provide additional prognostic information complementary to the Lille score.
Frequently Asked Questions
When should the Lille score be calculated?
The Lille score should be calculated on day 7 of corticosteroid therapy. It requires both baseline (day 0) and follow-up (day 7) bilirubin values to assess treatment response. Calculating the score earlier or later than day 7 reduces its predictive accuracy.
Can the Lille score be used without prior steroid treatment?
No. The Lille score is specifically designed to assess response to corticosteroid therapy. It requires the day 0 to day 7 bilirubin change, which only has meaning in the context of ongoing treatment. For initial severity assessment before starting treatment, use the Maddrey discriminant function, MELD score, or Glasgow Alcoholic Hepatitis Score.
What if the bilirubin increases from day 0 to day 7?
An increase in bilirubin from day 0 to day 7 is a poor prognostic sign, indicating worsening liver function despite steroid therapy. This will result in a higher Lille score, typically placing the patient in the null responder category. In such cases, steroids should be discontinued.
Is the Lille score validated for all causes of hepatitis?
No. The Lille score was developed and validated specifically for severe alcoholic hepatitis in patients receiving prednisolone. It should not be applied to other forms of hepatitis, such as viral hepatitis, autoimmune hepatitis, or drug-induced liver injury. Each of these conditions has its own prognostic tools and treatment algorithms.
What are alternatives if steroids fail?
For patients who are null responders to steroids, the main options include supportive care, nutritional support, management of complications, and evaluation for early liver transplantation. N-acetylcysteine (NAC) may have a role as adjunctive therapy. Pentoxifylline is no longer recommended based on the STOPAH trial results. Clinical trials are ongoing for newer therapies targeting specific inflammatory pathways.
How accurate is the Lille score?
The Lille score has been validated in multiple independent cohorts and demonstrates good predictive accuracy, with an area under the receiver operating characteristic curve (AUROC) of approximately 0.85 for predicting 6-month mortality. It is considered the gold standard for assessing steroid response in alcoholic hepatitis.