Hematologic Complications Associated with Ventricular Assist Devices (VAD's) in 13 Children From a Single Center

Mechanical circulatory support systems are used for the treatment of end-stage heart failure. In children, extracorporeal membrane oxygenation systems and centrifugal pumps have been used widely in the past, but are only suitable for short term use and intensive care is obligatory. The Berlin Heart is a VAD which is an extracorporeal pulsatile pneumatic system which can be placed as a bridge to transplant or recovery and allows patients to be mobile and to lead an almost normal life. As with placement of any mechanical device, aggressive anticoagulant therapy is needed due to the increased risk of thrombosis and therefore an increased risk of bleeding associated with the treatment. Here we report the complications seen in 13 children on VAD's from a single institution.

Patient medical records were reviewed to obtain the information on complications and associated coagulation test results. Tests of hemostasis and thrombosis were performed in the authors' CAP accredited Coagulation Laboratory at the Children's Hospital of Michigan.

Thirteen children with a mean age of 42 months (range 3 weeks to 13 years) received a VAD as a bridge to transplant between 2006 and 2011. There were 7 males and 6 females, 9 were African American and 5 Caucasian. 10/13 patients were on ECMO for a short duration (1–15days) prior to VAD implantation. The duration of VAD implantation varied from 8 days to 8 months. 9/13 eventually received a heart transplant. Anticoagulation was monitored according to the Berlin Heart protocol. There were 4 deaths, all in infants <7months of age. Complications included thrombosis in 7 patients and bleeding in 5, some of whom had both. Time to thrombotic complications varied from day of implant to 14 days post implantation. Thrombotic complications included fibrin deposits in the VAD to strokes. Assessment of risk factors for thrombosis and bleeding showed that 3/10 patients were heterozygous for the MTHFR gene mutation, 1/10 had a low protein S and 2/10 had a low protein C. 5/7 patients with thrombosis had concurrent systemic or respiratory infections. ATIII was monitored and replaced to keep over 70% per protocol recommendations. Reactive thrombocytosis seen in some patients with thrombosis was associated with an increase in MA on thromboelastography despite administration of antiplatelet drugs. Bleeding complications varied from hemopericardium and intracranial hemorrhage to GI bleeding. Anticoagulation was monitored using thromboelastography with platelet mapping (monitoring antiplatelet agents), PTT and unfractionated (UFH level) or low molecular weight heparin levels respectively; UFH levels were found to be more useful for monitoring heparin anticoagulation than PTT. PTT was found to be low secondary to elevated Factor VIII activity resulting in poor correlation to heparin levels. Thromboelastography Platelet mapping assays were used to monitor aspirin and platelet ADP receptor antagonists, but did not correlate with traditional whole blood platelet aggregation studies.

While thrombosis and bleeding complications are commonly seen in adult patients on VAD's, pediatric data in this population are limited and monitoring for the complications is inadequate. At our institution, infants who received VAD's had a poor outcome, but larger studies are needed to evaluate if there are complications unique to this age group. Current monitoring strategies are not always successful in predicting the occurrence of complications. Defects in fibrinolysis and increased thrombin generation may predispose to thrombotic manifestations. Tests to detect elevated thrombin generation and evaluation of fibrinolysis using modified thromboelastography may be useful to predict an increase in thrombotic potential which may assist us in improving the outcome.

No relevant conflicts of interest to declare.