Pump thrombosis redux.

Left ventricular assist devices (LVADs) are now a durable mainstay of therapy for medically refractory advanced heart failure (HF). Continuous-flow LVADs demonstrated increased longevity and enhanced quality of life in a series of landmark trials, ushering in an era where more LVADs are now implanted annually than hearts transplanted. Less than 50% of current LVAD recipients are listed for heart transplantation at the time of implant, and many stand-alone VAD programs, outside of transplant centers, are well established. Enthusiasm grew in the early part of this decade to investigate expanded indications for LVADs in ambulatory patients before transition to inotrope dependence. Even as adoption of this technology accelerated, concerns about a rise in pump malfunction due to device thrombosis were reported in the most widely implanted LVAD platform worldwide, the HeartMate (HM) II (Thoratec Inc, Pleasanton, CA). Two separate reports in 2013 documented a spike in HM II pump thrombosis rates beginning in 2011, several years after U.S. Food and Drug Administration approval of the device. These reports estimated the risk of thrombosis at 6 months to be between 6% and 12%, at least 3-fold higher than noted in the registration trials. Another report suggested that this problem was not confined to the HM II device but was also observed with the HeartWare HVAD (HeartWare International Inc, Framingham, MA) pump. Fueling more uneasiness, thrombosis rates appeared to be increasing with each subsequent year, threatening the viability of mechanical support in expanded indications. Although rising thrombosis rates were not initially linked to a decline in overall post-LVAD survival, the diagnosis of pump thrombosis conferred substantial morbidity, need for pump exchange, and increased cost of care. Awareness of the effect of pump thrombosis on patient outcomes prompted the United Network of Organ Sharing in 2013 to allow emergency priority listing for patients with impending or manifest pump thrombosis, whereas no bailout option was available to lifetime (“destination”) therapy patients who were not eligible for transplant. The increase in thrombosis risk also led directly to the indefinite suspension of the National Heart, Lung, and Blood Institute (NHLBI) sponsored Randomized Evaluation of VAD Intervention before Inotropic Therapy (REVIVE-IT) trial of LVAD therapy with the HM II pump in earlier-stage, non–inotrope-dependent patients after equipoise was lost to randomize New York Heart Association Functional Classification III patients to the HM II device. A vigorous debate ensued as multiple explanations for increased LVAD thrombosis rates were posited. A mechanical defect in the device could not be confirmed, and prevailing theories centered around the device-patient interface, including unfavorable angulation of the inflow cannula relative to the septum and a constellation of alterations in clinical management. These included a relaxation of anti-coagulation targets in the face of gastrointestinal bleeding and acquired von Willebrand factor deficiency, a reduced emphasis on post-operative bridging anti-coagulation, and the lowering of pump speeds to facilitate aortic valve opening to reduce the risk of aortic insufficiency and maintain transaortic pulsatile flow. After the initial reports on the pump thrombosis problem, most centers rapidly adopted risk-mitigation strategies to emphasize bridging anti-coagulation, vigilant monitoring for intravascular hemolysis as a harbinger of pump thrombosis, and early aggressive treatment of suspected thrombosis with augmented anti-coagulation and pump exchange. In early 2015, the NHLBI commissioned 2 separate analyses of the most current HM II data to explore the evolving effect of pump thrombosis on patient care. These updated analyses, which are highlighted in this issue of the Journal, used data on HM II devices implanted between April 2008 and June 2014 at 146 centers participating in the Interagency Registry of Mechanically Assisted Circulation Support (INTERMACS). Kirklin et al analyzed 9,808 primary HM II implants and confirmed an increasing risk of thrombosis from 2009 through 2013, followed by an observed decrease in risk in the first half of 2014. By http://www.jhltonline.org