POSTERS

Conclusion: in this cohort, overall mortality rate of 3rd (M3) to 6th

month (M6) follow-up was 32.9%, this one reaching 80% for Lille

score above 0.45.

In multivariate analysis, predictive factors of late death at 6 months

are the absence of alcohol withdrawal in M1 (p =0.01), the existence

of a documented infection at M1 (p =0.04) and the MELD-Na

score >27 at M1 (p =0.004).

From M3 follow-up, the causes of death are no longer related to

the initial severity of the ADT.

1361

ALCOHOL-INDUCED LIVER TOXICITY IS ASSOCIATED WITH

NEUTROPHIL DYSFUNCTION IN A NOVEL IN-VITRO MODEL OF

ACUTE LIVER INJURY

L.J. Markwick1,2, E. Palma1, A. Riva1, R. Williams1, D. Clemens3,4,

D. Shawcross2, S. Chokshi1. 1Institute of Hepatology, Foundation for

Liver Research, 2Institute of Liver Studies, King’s College London at

King’s College Hospital, London, UK; 3University of Nebraska Medical

Center, 4Veteran Affairs Medical Center, Omaha, NE, USA

E-mail: l.markwick@researchinliver.org.uk

Background and Aims: Sepsis is a major cause of mortality

in patients with alcohol-induced acute and chronic liver failure

(ALF/CLF). Neutrophils are a major innate immune cell subset

involved in the first line of defence against infection and circulating

neutrophil dysfunction has been reported in patients with ALF/CLF.

However, there is a paucity of understanding regarding the

mechanisms involved in this dysfunction. In this study we

aimed to characterise the precise relationship between neutrophil

dysfunction and alcohol-induced liver damage with a novel in

vitro model mimicking the in vivo interactions of neutrophils and

hepatocytes.

Methods: We cultured a well-characterised neutrophil-line HL-

60 either directly with ethanol or with supernatants taken from

ethanol metabolising-human hepatoma cell lines VL-17A (positive

for alcohol-dehydrogenase and CYP2E1) cultured in the presence

of safe levels (10mM) and toxic levels (250mM) of ethanol

reflecting real-life human alcohol consumption for 24 hours.

Neutrophil function was evaluated by TLR expression, chemotaxis,

phagocytosis and respiratory burst assays. Cell supernatants were

also collected for cytokine profiling and to quantitate levels of

ammonia and ethanol metabolites. The effect of ethanol on the

functional activities of neutrophils isolated from both normal and

ALF/CLF patients will also be assessed.

Results: Supernatants collected from the hepatoma line VL17A

cultured with 250mM ethanol (representative of an alcohol

binge) significantly reduced the phagocytic capacity of the HL-

60-neutrophil line (p < 0.05). This was greater than the effect of the

same concentration of ethanol applied directly to the neutrophils

(p < 0.05). Our preliminary data also suggests that the metabolised

ethanol inhibits chemotaxis of the HL60-neutrophil cells towards

a gradient of fMLP. Furthermore, we observed a reduction of TLR4

expression.

Conclusions: We describe a novel model for investigating the

correlates of dysfunctional innate immunity during acute alcohol-

induced liver injury. We identify that alcohol does impair neutrophil

function directly but this is profoundly increased after hepatocyte

alcohol metabolism implicating a causal link between liver injury

and impairment of antibacterial neutrophil functions.

1362

CLINICAL BUT NOT HISTOLOGICAL FACTORS PREDICT

LONG-TERM PROGNOSIS IN PATIENTS WITH BIOPSY PROVEN

ADVANCED ALCOHOLIC LIVER DISEASE

S. Masson1,2, I. Emmerson1, E. Henderson1, E. Fletcher1, A.D. Burt2,

C.P. Day2, S.F. Stewart3. 1Liver Transplant Unit, Freeman Hospital,2Institute of Cellular Medicine, Newcastle University, Newcastle upon

Tyne, UK; 3Mater Misericordiae University Hospital, Dublin, Ireland

E-mail: steven.masson@nuth.nhs.uk

Introduction/aim: Alcoholic liver disease (ALD) is a threat to the

health of the European population. It remains a common indications

for liver transplantation and a leading cause of death. Despite this,

the long term clinical course and predictive factors of survival in

advanced ALD have not been well described. We aimed to identify

factors that predict 15 year survival in out-patients with biopsy-

proven ALD.

Methods: Patients (n = 134) with biopsy-proven advanced (stage

III OR IV) ALD were followed-up for fifteen years or until death

or transplantation. At baseline, clinical and laboratory data were

collected. On biopsy, the presence of cirrhosis and histological

features (fat severity, lymphocyte and neutrophil infiltration) were

scored semi-quantitatively.

Results: Median age was 51 (29–67) and the majority (72%) were

male. All had history of alcohol excess (>80g/day [M], 50 g/d [F]).

Patients were followed until death (n =99; median 62m), OLT (n =5;

median 96m) or are still alive (n = 33; median 187m). Overall,

the 5, 10 and 15-year survival was 64, 40 and 26%, respectively.

Baseline characteristics are shown according to outcome (Table 1).

In multivariate analysis age (p = 0.01), smoking (p =0.01), persistent

drinking (p < 0.01) and serum albumin at baseline (p = 0.02) were

associated with significantly increased risk of death. No histological

features correlated with prognosis.

Table: Characteristics according to 15 year survival

Characteristic Alive (n = 32) Dead (n =104) OR (95%CI) p value

Age 48 (44–54) IQR 53 (47–58) IQR 1.08 (1.02–1.13) <0.01

Smoker 12 (50%) 72 (73%) 2.53 (1.10–5.83) 0.04

Subsequent abstinence 13 (42%) 17 (17%) 0.28 (0.12–0.69) <0.01

Ascites 2 (7%) 20 (21%) 3.42 (0.75–15.64)

0.11

Platelet count 171 (146–214) 125 (89–191) 1.00 (0.99–1.00) 0.25

Serum Albumin 44 (42–47) 39 (32–44) 0.83 (0.75–0.92) <0.01

Child-Pugh 9 (9–10) 10 (9–12) 1.76 (1.19–2.59) <0.01

MELD 7 (6–9) 11 (8–15) 1.18 (1.06–1.32) <0.01

UKELD 47 (47–50) 51 (49–55) 1.25 (1.09–1.45) <0.01

Conclusion: In out-patients with biopsy-proven advanced ALD,

clinical but not histological factors determine prognosis. Age,

persistent alcohol intake, smoking habit and serum albumin are

independent poor prognostic factors. Abstinence from alcohol

and smoking cessation should be the priorities in the long-term

management of ALD.

1363

METABOLIC PROFILES ASSOCIATED WITH THE SEVERITY OF

ALCOHOLIC HEPATITIS AND THE RESPONSE TO TREATMENT.

PRELIMINARY RESULTS

J. Michelena1, C. Alonso2, J. Barr2, J. Altamirano1, I. Martinez

Arranz2, R. Bataller1, M.L. Martinez Chantar3, A. Castro2, J.M. Mato3,

J. Caballeria1. 1Liver Unit, Hospital Clınic, IDIBAPS, CIBERehd,

Barcelona, 2OWLGenomics, 3CICbioGUNE, CIBERehd, Derio, Spain

E-mail: caballer@clinic.ub.es

Background and Aims: Alcoholic hepatitis (AH) includes a

spectrum of diseases that range from mild injury to severe, life

threatening injury. Corticosteroids are the only recommended

therapy for severe AH. However, a percentage of patients fail

to respond to corticosteroid treatment and, on the other hand,

complications, especially infections, occur in almost 25% of patients

during corticosteroid treatment. The aim of this ongoing study was

Journal of Hepatology 2012 vol. 56 | S389–S548 S535