Abstract
Background: Hypercoagulability can lead to serious thromboembolic events. The aim of this study was to assess the perioperative coagulation status in liver transplant recipients with a tendency to hypercoagulability. Methods: In a prospective observational study (South African Cochrane Registry 201405000814129), 151 potential liver transplant recipients were screened for thrombophilic factors from October 2014 to June 2017, and 57 potential recipients fulfilled the inclusion criterion of presenting two or more of the following thrombophilic factors: low protein C, low protein S, low anti-thrombin, increased homocystein, increased antiphospholipid IgG/IgM antibodies, increased lupus anticoagulant, and positive Factor V Leiden mutation. Seven patients were excluded from the study because they fulfilled the exclusion criteria of cancelling the liver transplantation, oral anticoagulation, or intraoperative treatment with rFVIIa. Accordingly, 50 patients were included in the final analysis. Thromboelastometry (ROTEM) (EXTEM, INTEM and FIBTEM) and conventional coagulation tests (CCT) were performed preoperatively, during the anhepatic phase, post reperfusion, and on postoperative days (POD) 1, 3 and 7. ROTEM was used to guide blood product transfusion. Heparin was infused (60-180 U/kg/day) postoperatively for 3 days and then was replaced by low-molecular-weight heparin (20 mg/12 h). Results: FIBTEM MCF significantly increased postoperatively above reference range on POD 7 despite normal fibrinogen plasma concentrations (p < 0.05). Both EXTEM and INTEM demonstrated significant changes with the phases of transplantation (p < 0.05), but with no intra- or postoperative hypercoagulability observed. INTEM CT (reference range, 100-240 s) normalized on POD 3 and 7 (196.1 +/- 69.0 and 182.7 +/- 63.8 s, respectively), despite prolonged aPTT (59.7 +/- 18.7 and 46.4 +/- 15.7 s, respectively; reference range, 20-40 s). Hepatic artery thrombosis (HAT) and portal vein thrombosis (PVT) were reported in 12.0% and 2.0%, respectively, mainly after critical care discharge and with high FIBTEM MCF values in 57% on POD 3 and 86% on POD 7. Receiver operating characteristics curve analyses of FIBTEM MCF were significant predictors for thromboembolic events with optimum cut-off, area under the curve and standard error on POD 3 (>23 mm, 0.779 and 0.097; p = 0.004) and POD 7 (>28 mm, 0.706 and 0.089; p = 0.020). Red blood cells (mean +/- SD, 8.68 +/- 5.81 units) were transfused in 76%, fresh frozen plasma (8.26 +/- 4.14 units) in 62%, and cryoprecipitate (12.0 +/- 3.68 units) in 28% of recipients. None of the recipients received intraoperative platelet transfusion or any postoperative transfusion. Main transplant indication was hepatitis C infection in 82%. 76% of recipients included in this highly selected patient population showed increased lupus anticoagulant, 2% increased antiphospholipid IgG/IgM antibodies, 20% increased homocysteine, 74% decreased anti-thrombin, 78% decreased protein C, 34% decreased protein S, and 24% a positive Factor V Leiden mutation. Overall 1-year survival was 62%. Conclusion: A significant postoperative step-wise increase in FIBTEM MCF beyond the reference range was observed despite normal fibrinogen plasma concentrations, and FIBTEM MCF was a predictor for thromboembolic events in this study population, particularly after POD 3 and 7 on surgical wards when CCTs failed to detect this condition.
However, the predictive value of FIBTEM MCF for postoperative HAT and PVT needs to be confirmed in a larger patient population. A ROTEM-guided anticoagulation regime needs to be developed and investigated in future studies.