Factors associated with thrombotic and hemorrhagic complications in pediatric liver transplantation: A multi-center analysis from the Starzl Network for Excellence in Pediatric Transplantation
Kyle Soltys1, Xingyu Zhang1, John Bucuvalas6, Adam Griesemer1,2, Blayne Sayed3, Riccardo Superina4, Caroline Lemoine4, Sara Rasmussen5, Rene Romero7, Irini Batsis6, CJ Confair1, George Mazariegos1.
1Pediatric Transplantation, UPMC Children’s Hospital of Pittsburgh, Pittsburgh, PA, United States; 2Transplant Surgery, Columbia University, NY, NY, United States; 3Transplant Surgery, Hospital for Sick Children, Toronto, ON, Canada; 4Pediatric Surgery, Laurie Children’s Hospital, Chicago, IL, United States; 5Pediatric Surgery, Seattle Children’s Hospital, Seattle, WA, United States; 6Pediatrics, Icahn School of Medicine/ Mount Sinai Hospital, NY, NY, United States; 7Pediatric Hepatology, Children’s Hospital of Atlanta, Atlanta, GA, United States
The Starzl Network for Excellence in Pediatric Transplantation.
Introduction: Vascular thrombosis adversely impacts outcome after pediatric liver transplantation (PLT). Scant data exists regarding factors associated with vascular thrombosis and the risk or benefit of anticoagulation treatment.
Methods: Retrospective review of donor, intraoperative, recipient, and outcomes of 76 PLT recipients with 628 ± 193 days of follow- up from four PLT centers in the Starzl Network for Excellence in Pediatric Transplant. The use of individual center anticoagulation protocols was recorded with thrombotic/hemorrhagic complications and graft/patient survival. Variables analyzed included: CIT/WIT, operative time, allograft type (whole, split, reduced, living/deceased donor), donor age/weight, recipient age/weight, final perioerative coagulation profile, intraoperative PV and HA flow and intraoperative ultrasound assessment, blood loss, donor variant anatomy, pre-operative concern for PV or HA,
Results: Median age/weight were 2.25 years and 13kg respectively. 39 (51%) utilized technical variant allografts, 14 LDLT. 7 (9.2%) transplants experienced on-table HAT, five of which resolved intraoperatively. One HAT occurred on POD 8 and another on day 616. 3 had PVT on POD1. Anticoagulation was started intraoperatively in 50 LT and continued post-operatively in 66. Of 11 (14%) with post-operative hemorrhage, 6 required operative intervention. Anti-coagulation protocols were active at the time of all thrombotic and bleeding events. Two patients were re-transplanted for arterial thrombosis and two died with functioning allografts. On-table and post-operative thrombosis were statistically associated with lower HA flow (p<0.05). On-table was thrombosis associated: donor variant anatomy, split/reduced allografts and higher intraoperative blood loss. Surgeon performed intraoperative US findings did not correlate with post-operative arterial or portal venous patency. Pre-operative surgeon concern for native portal vein flow/ size correlated with PVT. Lower hepatic arterial and portal flows, higher EBL, higher PT and arterial grafts associated with hemorrhagic complications.
Conclusions: Thrombotic and hemorrhagic complications can be predicted with pre-operative assessment of donor and recipient vasculature and direct flow measurement but may not be predicted with ultrasound evaluation nor prevented with anticoagulation.