P0206 : Prevention of bleeding following invasive procedures in cirrhotic patients: A single center experience

Abstract

Background: Risk of bleeding in cirrhosis has predominantly been associated with coagulopathy and thrombocytopenia due to impaired liver function and splenomegaly. The evaluation of the actual bleeding risk in cirrhotic patientsundergoing invasiveprocedures is a critical point to optimizemanagement in termsof platelet (Plt) or plasma prophylactic transfusions. Methods: During 2013, 480 cirrhotic patients underwent diagnostic or therapeutic invasive procedures at our Liver Unit (Fondazione IRCCS Ca Granda, Ospedale Maggiore Policlinico, Milan, Italy). Plt transfusion was performed when Plt count less than 50,000/mmc while fresh frozen plasma was infused when international normalized ratio (INR) >1.5, except those undergoing paracentesis. Low-molecular-weight-heparin was discontinued 24h before any procedure while antiaggregant therapy was interrupted five days before any procedure except oesophageal varices band ligation and paracentesis. Major hemorrhagic events were those requiring hospitalisation or blood transfusion; minor events was a haemoglobin decline >1.5 g/dl post procedure without clinical relevance. Results: 174 Transarterial Chemo-Embolization (TACE), 16 Radio-Frequency Termal Alblation (RFTA), 214 paracentesis, 59 oesophageal varices banding, 6 trans-jugular liver biopsy (LB) and 11 Percutaneous Ethanol Injection (PEI) were performed. Overall, 61 procedures met the criteria for plasma or Plt infusion and major bleeding complications occurred in 3 patients (0,6%). In 2 patients, anemia-related paracentesis was treated by blood transfusions whereas one patient following variceal band ligation had to be hospitalized for severe anemia. Major complications rate was 1.6% in Plt/plasma infusion exposed patients versus 0.47% in unexposed (p=0.28).Minor eventswith a>1.5 g/dl haemoglobindecline occurred in 17 patients (15 paracentesis and 2 band ligations), rate was 3.2% in Plt/plasma infusion exposed patients versus 3.5% in unexposed (p=0.9). Conclusion: Pre-treatment platelet transfusion in cirrhotic patients with Plt count 1.5, was associated with a negligible risk of bleeding and appeared as a safe, cost/effective strategy.