Introduction The European Association for the Study of the Liver guidelines recommend the use of liver biopsy to confirm alcoholic steatohepatitis (ASH) in patients suspected to be suffering from alcoholic hepatitis and who are classified as high risk after prognostic assessment. Despite this it was only in 2014 that Altamirano et alpublished the first prognostic liver biopsy based scoring system for the evaluation of ASH (the Alcoholic Hepatitis Histological Score or AHHS). This work sought to validate their scoring system and further explore the utility of the liver biopsy in ASH.

Method Two independent histopathologists, blinded to treatment and outcome, centrally reviewed liver biopsies of patients with clinically high-risk alcoholic hepatitis who had been recruited to the STOPAH trial.

Results 93 (47%) of the 208 biopsies received were both adequate in quality and taken between admission and day 5 of trial treatment. 88% (82/93) had histological features diagnostic of ASH. 89% (32/36) of cases obtained as routine clinical practice were diagnostic of ASH compared to 79% (15/19) biopsied for diagnostic uncertainty. Clinically more severe liver disease was associated a higher rate of ASH diagnosis (82% of GAHS ≤8 vs. 97% of GAHS >8).

65% (53/82) of biopsy proven cases of ASH were classified as severe by AHHS. This group had a significantly higher 28 day mortality rate than those classified as mild/moderate (18% vs. 0%, Fisher’s exact p = 0.02). AHHS severity positively correlated with baseline Maddrey’s Discriminant Function and Glasgow Alcoholic Hepatitis Score (r = 0.2, p = 0.045 and r = 0.3, p = 0.01 respectively).

Clinical markers of severe disease positively correlated with biopsy features of severe disease including:

- serum bilirubin with bilirubinostasis (r = 0.5, p= <0.0001)

- serum white cell and neutrophil count with lobular inflammation (r = 0.4, p = <0.001)

Elevation of serum alkaline phosphatase (ALP) and bilirubin are usually associated with more severe liver disease, however in this study they were seen to negatively correlate with certain biopsy features of more severe disease:

- serum ALP negatively correlated with ductular changes (r = -0.2, p = 0.04)

- serum bilirubin negatively correlated with Laennec fibrosis grading (r = -0.3, p = 0.01)

Conclusion This work goes some way towards validating the AHHS classification. The work also highlights the parallels between clinical and histological parameters and documents negative correlations seen in other liver diseases but not previously noted in ASH.

Disclosure of interest None Declared.