235 Thrombus within the HeartMate II Left Ventricular Assist Device (LVAD): Are All Clots Created Equal?

Left ventricular assist device thrombosis in the setting of supratherapeutic international normalized ratio (INR) and bleeding

A 71-year-old female with heart failure who underwent left ventricular assist device (LVAD) placement presented for evaluation of low hemoglobin and dark stools. She also had leg pain, numbness, and weakness for which she was taking ibuprofen. She was found to have a gastrointestinal bleed, INR of 4.3, and arterial thrombi in the left leg. She was stabilized, had her anticoagulation held, and underwent mechanical thrombectomy. On hospital day 6, LVAD interrogation revealed signs of thrombosis, while subsequent labs revealed a persistently supratherapeutic INR of 5.2. The patient had the LVAD removed and underwent further hematologic workup. Her platelets remained normal throughout the admission, indicating this was not acute disseminated intravascular coagulation (DIC), thrombotic thrombocytopenic purpura (TTP), or heparin induced thrombocytopenia (HIT). Echocardiography identified it as a primary thrombus. This case illustrates the importance of appropriate anticoagulation to balance the bleeding risk with the risk of thrombi, as well as the importance of maintaining high suspicion for LVAD thrombosis regardless of INR.

Biological Response to Sintered Titanium in Left Ventricular Assist Devices: Pseudoneointima, Neointima, and Pannus

Titanium alloys have traditionally been used in blood-contacting cardiovascular devices, including left ventricular assist devices (LVADs). However, titanium surfaces are susceptible to adverse coagulation, leading to thrombogenesis and stroke. To improve hemocompatibility, LVAD manufacturers introduced powder sintering on blood-wetted surfaces in the 1980s to induce endothelialization. This technique has been employed in multiple contemporary LVADs on the pump housing, as well as the interior and exterior of the inflow cannula. Despite the wide adoption of sintered titanium, reported biologic response over the past several decades has been highly variable and apparently unpredictable-including combinations of neointima, pseudoneoimtima, thrombus, and pannus. We present a history of sintered titanium used in LVAD, a review of accumulated clinical outcomes, and a synopsis of gross appearance and composition of various depositions found clinically and in animal studies, which is unfortunately confounded by the variability and inconsistency in terminology. Therefore, this review endeavors to introduce a unified taxonomy to harmonize published observations of biologic response to sintered titanium in LVADs. From these data, we are able to deduce the natural history of the biologic response to sintered titanium, toward development of a deterministic model of the genesis of a hemocompatible neointima.

Speed Modulation of the HeartWare HVAD to Assess In Vitro Hemocompatibility of Pulsatile and Continuous Flow Regimes in a Rotary Blood Pump

Although rotary blood pumps (RBPs) sustain life, blood exposure to continuous supra-physiological shear stress induces adverse effects (e.g., thromboembolism); thus, pulsatile flow in RBPs represents a potential solution. The present study introduced pulsatile flow to the HeartWare HVAD using a custom-built controller and compared hemocompatibility biomarkers (i.e., platelet aggregation, concentrations for ADAMTS13, von Willebrand factor (vWf), and free-hemoglobin in plasma (pfHb), red blood cell (RBC) deformability, and RBC-nitric oxide synthase (NOS) activity) between continuous and pulsatile flow in a blood circulation loop over 5 h. The HeartWare HVAD was operated using a custom-built controller, at continuous speed (3282 rev/min) or in a pulsatile mode (mean speed = 3273 rev/min, amplitude = 430 rev/min, frequency = 1 Hz) to generate a blood flow rate of 5.0 L/min, HVAD differential pressure of 90 mm Hg for continuous flow and 92 mm Hg for pulsatile flow, and systolic and diastolic pressures of 121/80 mm Hg. For both flow regimes, the current study found; (i) ADP- and collagen-induced platelet aggregation, and ADAMTS13 concentration significantly decreased after 5 h (P < 0.01; P < 0.05), (ii) ristocetin-induced platelet aggregation significantly increased after 45 min (P < 0.05), (iii) vWf concentration did not significantly differ at any time point, (iv) pfHb significantly increased after 5 h (P < 0.01), (v) RBC deformability improved during the continuous flow regime (P < 0.05) but not during pulsatile flow, and (vi) RBC-NOS activity significantly increased during continuous flow (15 min), and pulsatile flow (5 h; P < 0.05). The current study demonstrated: (i) speed modulation does not improve hemocompatibility of the HeartWare HVAD based on no observable differences being detected for routine biomarkers, and (ii) the time-course for increased RBC-NOS activity observed during continuous flow may have improved RBC deformability.