Article Text

Download PDFPDF

ATU-7 Incident acute kidney injury has a worse prognosis than baseline in severe alcoholic hepatitis
  1. Luke Tyson1,
  2. Professor Mark Thursz1,
  3. Ewan Forrest2,
  4. Michael Allison3,
  5. Nikhil Vergis1,
  6. Stephen Atkinson1
  1. 1Imperial College London, London, UK
  2. 2NHS Greater Glasgow and Clyde, Glasgow, UK
  3. 3Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK


Introduction Alcoholic hepatitis (AH) is the most severe alcohol-related liver disease. Acute kidney injury (AKI) is associated with increased mortality. AKI may be present at the time of presentation (baseline) or develop subsequently (incident). We used data from the STOPAH (STeroids Or Pentoxifyline for Alcoholic Hepatitis) trial to describe the prevalence of AKI, its association with mortality and risk factors for its development.

Methods The primary endpoint in analysis was 90-day mortality. Patients from the STOPAH trial were classified as having a baseline or incident AKI (within 7 days of starting treatment; D7). AKI was defined as any of: i) creatinine ≥ 26.5 micromol/L above or ≥ 1.5x the lowest recorded creatinine; ii) creatinine ≥ 133 micromol/L; iii) renal replacement therapy. The effect of AKI on 90-day mortality was tested by Kaplan-Meier survival analysis. Factors associated with incident AKI were compared by Student’s t-test, Mann-Whitney U test or Chi-squared test as appropriate.

Results Baseline creatinine was recorded in 1051 patients; 282 patients with a normal creatinine at baseline were alive at D7 but did not have a second creatinine recorded so were excluded from survival analysis. Baseline AKI was present in 198/1051 (19%) patients while 119/1051 (11%) developed an incident AKI. Baseline AKI was associated with increased D90 mortality compared to patients without baseline or incident AKI. Incident AKI was associated with the highest mortality (Figure 1). There was no difference in mortality between patients with a baseline AKI that resolved by D7 or persisted (Breslow Chi-square 0.227, p = 0.633). Patients with incident AKI had significantly higher bilirubin (mean 374 mmol/L vs 281, p < 0.001), INR (2.0 vs 1.8, p = 0.001), and neutrophil count (8.1 vs 7.1, p = 0.031) than those without baseline or incident AKI. Prednisolone treatment was associated with a reduced risk of incident AKI (odds ratio 0.53, 95% confidence interval 0.34 - 0.81, p = 0.003). Age, gender, baseline observations and hepatic encephalopathy were not associated with incident AKI.

Abstract ATU-7 Figure 1

Kaplan-Meier survival curves for patients with a baseline acute kidney injury (n = 198, 63 events), incident acute kidney injury (n = 119, 56 events) and normal baseline and day seven creatinine (n = 452, 115 events)*Beslow (Generalised Wilcoxon) p < 0.05; **p < 0.001

Conclusions Incident AKI at D7 confers a worse prognosis than either no or baseline AKI. This highlights the need for proactive monitoring and treatment of factors predisposing to AKI in patients with AH, particularly for patients with markers of severe disease.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.