Assessment of standard laboratory tests and rotational thromboelastometry for the prediction of postoperative bleeding in liver transplantation

作者：Doetsch T M; Dirkmann D; Bezinover D; Hartmann M; Treckmann J W; Paul A; Saner F H^*

来源：British Journal of Anaesthesia, 2017, 119(3): 402-410.

DOI：10.1093/bja/aex122

• bleeding
• coagulation factor
• critical care
• liver transplantation
• point-of-care testing

Background. Perioperative bleeding remains a major challenge in liver transplantation. We aimed to compare standard laboratory tests with thromboelastometry (ROTEM (R)) with regard to their ability to predict postoperative non-surgical bleeding.
Methods. Data from 243 adult liver transplant recipients from January 2012 to May 2014 were evaluated retrospectively. Upon admission to the intensive care unit, coagulation status was assessed using standard laboratory tests [prothrombin time (PT), activated partial thromboplastin time (aPTT), fibrinogen concentration, and platelet count] and ROTEM (R) whole blood coagulation assays. Bleeding was defined as transfusion of >= 3 units of red blood cells or reoperation for non-surgical bleeding within 48h after transplantation. Coagulation test results were analysed using receiver operating characteristics (ROC) in order to identify variables predictive of postoperative bleeding. Coagulation management was based on ROTEM (R) guided factor concentrate treatment.
Results. The overall incidence of bleeding was 12.3% (n = 30). Twenty-three (9.5%) patients underwent reoperation and seven (2.9%) received >= 3 units of red blood cells and non-operative management. Standard laboratory tests predictive of postoperative bleeding were aPTT and PT [area under the ROC curve (AUC) 0.688 and 0.623, respectively]. Tests predictive of bleeding with ROTEM (R) were CTEXTEM, CFTINTEM, A10(FIBTEM), and MCFFIBTEM, with AUCs of 0.682, 0.615, 0.615, and 0.611, respectively. Fibrinogen concentration, platelet count, and other ROTEM (R) variables failed to demonstrate predictive value for postoperative bleeding (AUC < 0.6). Dialysis-dependent kidney failure, 30 day mortality, and median model for endstage liver disease score were all significantly higher in bleeding patients.
Conclusions. Although both postoperative standard laboratory tests and ROTEM (R) assays could identify patients at risk for postoperative bleeding, ROTEM (R) assays demonstrated a greater predictive value for impaired fibrinogen polymerizationrelated coagulopathy.

• 出版日期2017-9