Article Text


GAHS TAHP: the glasgow alcoholic hepatitis score in the treatment of alcoholic hepatitis in practice
  1. H Lafferty *1,
  2. E Watson1,
  3. R Hamid1,
  4. J Winter1,
  5. J Morris1,
  6. A Stanley1,
  7. E Forrest1
  1. 1Glasgow Royal Infirmary, Glasgow, UK


Introduction Alcoholic hepatitis (AH) continues to be a frequent cause of alcohol related mortality. The Glasgow Alcoholic Hepatitis Score (GAHS) has not only been proposed as an accurate measure of prognosis, but also as a guide to therapeutic intervention. However these studies have been retrospective. The authors assessed the use of the GAHS in the management of unselected patients with AH.

Methods 182 patients with a clinical diagnosis of AH were assessed prospectively (Group 1), and then compared with a historical control group (Group 2). All patients had an admission serum bilirubin greater than 80 μmol/l. No patients were Hepatitis C PCR positive or had presented with evidence of upper gastro-intestinal haemorrhage. The unit protocol was for treatment with prednisolone 40 mg or pentoxifylline 400 mg three times a day at the discretion of the responsible consultant when the GAHS was ≥9 in Group 1. No Group 2 patients received either prednisolone or pentoxifylline (Group 2).

Results Overall 28-day survival figures for Group 1 and Group 2 were 150 (82.4%) and 129 (73.7%) respectively (p=0.03; HR 1.64 (1.06–2.56)). The respective 84-day survival rates were 126 (69.2%) and 119 (68%) (p=0.41; HR 1.17 0.806–1.70)). In Group 1, 96 patients developed a GAHS ≥9 during the first week of their hospital admission (58%). On comparison Group 1 and 2 patients with a GAHS<9 had similar rates of survival at day 28 and 84. However differences were seen between patients with a GAHS ≥9. At 28 days the survival rates of Group 1 and Group 2 were 71% and 41% respectively (p=0.0002; HR 2.86 (1.64–3.18)). At 84 days the survival rates of the Group 1 and Group 2 were 54% and 37% respectively (p=0.008; HR 1.95 (1.19–3.18)). The improvement in outcome was more clearly seen in GAHS ≥9 patients who received treatment as per protocol (80 of 96: 83%). Reasons given for not receiving treatment were as follows: 10 suspected sepsis (confirmed: 3 SBP; 3 chest sepsis; 1 secondary peritonitis), 1 upper gastro-intestinal haemorrhage after admission and 1 patient's condition deteriorated too rapidly to allow treatment to start. In 4 cases no reasons were given.

Conclusion The GAHS can be used in clinical practice. A GAHS ≥9 identifies patients who may benefit from treatment.

  • alcoholic hepatitis

Statistics from


  • Competing interests None.

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.