Anticoagulation in simultaneous pancreas kidney transplantation - On what basis?

BACKGROUND: Despite technical refinements, early pancreas graft loss due to thrombosis continues to occur. Conventional coagulation tests (CCT) do not detect hypercoagulability and hence the hypercoagulable state due to diabetes is left untreated. Thromboelastogram (TEG) is an in-vitro diagnostic test which is used in liver transplantation, and in various intensive care settings to guide anticoagulation. TEG is better than CCT because it is dynamic and provides a global hemostatic profile including fibrinolysis. AIM: To compare the outcomes between TEG and CCT (prothrombin time, activated partial thromboplastin time and international normalized ratio) directed anticoagulation in simultaneous pancreas and kidney (SPK) transplant recipients. METHODS: A single center retrospective analysis comparing the outcomes between TEG and CCT-directed anticoagulation in SPK recipients, who were matched for donor age and graft type (donors after brainstem death and donors after circulatory death). Anticoagulation consisted of intravenous (IV) heparin titrated up to a maximum of 500 IU/h based on CCT in conjunction with various clinical parameters or directed by TEG results. Graft loss due to thrombosis, anticoagulation related bleeding, radiological incidence of partial thrombi in the pancreas graft, thrombus resolution rate after anticoagulation dose escalation, length of the hospital stays and, 1-year pancreas and kidney graft survival between the two groups were compared. RESULTS: Seventeen patients who received TEG-directed anticoagulation were compared against 51 contemporaneous SPK recipients (ratio of 1: 3) who were anticoagulated based on CCT. No graft losses occurred in the TEG group, whereas 11 grafts (7 pancreases and 4 kidneys) were lost due to thrombosis in the CCT group (P = 0.06, Fisher’s exact test). The overall incidence of anticoagulation related bleeding (hematoma/ gastrointestinal bleeding/ hematuria/ nose bleeding/ re-exploration for bleeding/ post-operative blood transfusion) was 17.65% in the TEG group and 45.10% in the CCT group (P = 0.05, Fisher’s exact test). The incidence of radiologically confirmed partial thrombus in pancreas allograft was 41.18% in the TEG and 25.50% in the CCT group (P = 0.23, Fisher’s exact test). All recipients with partial thrombi detected in computed tomography (CT) scan had an anticoagulation dose escalation. The thrombus resolution rates in subsequent scan were 85.71% and 63.64% in the TEG group vs the CCT group (P = 0.59, Fisher’s exact test). The TEG group had reduced blood product usage {10 packed red blood cell (PRBC) and 2 fresh frozen plasma (FFP)} compared to the CCT group (71 PRBC/ 10 FFP/ 2 cryoprecipitate and 2 platelets). The proportion of patients requiring transfusion in the TEG group was 17.65% vs 39.25% in the CCT group (P = 0.14, Fisher’s exact test). The median length of hospital stay was 18 days in the TEG group vs 31 days in the CCT group (P = 0.03, Mann Whitney test). The 1-year pancreas graft survival was 100% in the TEG group vs 82.35% in the CCT group (P = 0.07, log rank test) and, the 1-year kidney graft survival was 100% in the TEG group vs 92.15% in the CCT group (P = 0.23, log tank test). CONCLUSION: TEG is a promising tool in guiding judicious use of anticoagulation with concomitant prevention of graft loss due to thrombosis, and reduces the length of hospital stay.