Utility of Lille Score in Predicting 30-Day Survival in Steroid-Treated Alcohol-Associated Hepatitis

Marcus A. Healey,Geetha Ramalingam,Yiwei Hang,Ekaterina Smirnova,Amon Asgharpour,Vaishali Patel,Hannah Lee,Velimir Luketic,Scott Matherly,Mohammad Siddiqui,Joel Wedd,Arun Sanyal,Richard K. Sterling

DOI: https://doi.org/10.1007/s10620-024-08479-6

2024-06-20

Digestive Diseases and Sciences

Abstract:In alcohol-associated hepatitis (AH), the Lille score is used to assess futility of steroids. However, the ability of the Lille score to predict 30-day survival in AH is not well-defined. Our aim is to compare the utility of the Lille score in predicting 30-day survival in those with AH treated with steroids.

gastroenterology & hepatology

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to evaluate the utility of the Lille score in predicting the 30 - day survival rate of patients with alcohol - related hepatitis (AH), especially in patients receiving corticosteroid treatment. The research background points out that although the Lille score is often used to evaluate the effectiveness of corticosteroid treatment, its ability to predict the 30 - day survival rate has not been fully defined. Therefore, the study aims to compare the utility of the Lille score in predicting the 30 - day survival rate of AH patients receiving corticosteroid treatment.

Specifically, the study retrospectively analyzed the medical records of 882 AH patients hospitalized between January 1, 2012 and December 30, 2019, and screened out 201 patients who met the criteria for severe AH and received corticosteroid treatment. These patients were divided into two groups according to their Lille scores (< 0.45 or ≥ 0.45) on day 4 or day 7, and the survival rates within 30 days of these two groups were compared. In addition, there was also a concurrent control group, namely 110 severe AH patients who did not receive corticosteroid treatment, for comparative analysis.

The main findings of the study include:
- The 30 - day survival rate of patients with a Lille score < 0.45 was significantly higher than that of patients with a Lille score ≥ 0.45 (94.9% vs. 80.72%, p = 0.002).
- For a Lille score < 0.45, the sensitivity in predicting the 30 - day survival rate was 95%, the specificity was 19%, the positive predictive value (PPV) was 63%, and the negative predictive value (NPV) was 73%.

These results indicate that the Lille score has a high sensitivity in predicting the 30 - day survival rate of AH patients receiving corticosteroid treatment, but performs poorly in terms of specificity. This suggests that in the future, it may be necessary to combine other indicators or methods to improve the accuracy of prediction.