Article Text


Hepatobiliary I
PTU-071 Improved outcomes following covered TIPSS in patients admitted to intensive care following a variceal bleed: a single centre study
  1. C Corbett,
  2. D Tripathi,
  3. N Murphy,
  4. S Olliff,
  5. K Mangat
  1. Queen Elizabeth Hospital, Birmingham, UK


Introduction Variceal bleeding occurs in a third of patients with cirrhosis and varices, and accounts for 10% of all GI bleeding in the UK. The 6-week mortality of 20% is greater than that for myocardial infarction. Recent evidence suggests that early covered TIPSS can improve outcomes after variceal bleed in certain patients.

Methods This is a retrospective study of patients with a history of variceal bleeding admitted to intensive care unit over a 3-year period. Patients without no active bleeding or stigmata of variceal haemorrhage were excluded. Multiple factors surrounding the admission and therapy were analysed. All TIPSS procedures were performed using PTFE covered stents during the admission.

Results 41 patients met inclusion criteria. 56% had alcoholic liver disease, 39% and 45% had Child C and B disease respectively. Mean age was 53.7 years, and median follow-up 466 days. The mean MELD was 14.4. 18% were shocked on admission, and 83% required invasive ventilation. 91% of admissions underwent endoscopy within 24 h of admission. All patients had broad spectrum antibiotics and terlipressin during the admission. The principal endoscopic therapy was variceal band ligation (44%), with 41% undergoing balloon tamponade as 1st intervention. 86% had active bleeding at index endoscopy, with 72% having balloon tamponade placed at some point. Indication for TIPSS was salvage therapy (60%), following re-bleeding (15%) and secondary prophylaxis (25%). Overall there was 34% inpatient admission mortality. The TIPSS group (n=19) had lower mortality than the non TIPSS group (n=22) at 6 weeks (11% vs 45% p<0.05) that persisted throughout follow-up (36% vs 54%, p<0.05). Multivariate analysis revealed a MELD≥18 and SOFA ≥11as significant predictors of mortality (p<0.05). Baseline characteristics were well matched between these groups. Fewer patient in the TIPSS group reached the composite end point of rebleeding after 5 days or failure to control initial bleeding (12% vs 45%, p=0.02).

Conclusion This study shows that patients with recent severe variceal bleeding admitted to intensive care have significantly better outcomes following covered TIPSS insertion. These findings need to be validated in randomised controlled trials.

Competing interests None declared.

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.