During the past several decades, left ventricular assist

Transcription

1 Advances in Heart Failure Mechanisms of Bleeding and Approach to Patients With Axial-Flow Left Ventricular Assist Devices Jorge Suarez, BS; Chetan B. Patel, MD; G. Michael Felker, MD; Richard Becker, MD; Adrian F. Hernandez, MD, MHS; Joseph G. Rogers, MD During the past several decades, left ventricular assist devices (LVADs) have become a valuable therapeutic option in the management of advanced systolic heart failure. Durable LVADs are being used to provide long-term mechanical circulatory support to a growing population of patients with end-stage systolic heart failure refractory to medical and electric therapies. This advanced stage of heart failure is associated with unacceptably high mortality when treated with medical therapy alone, and these patients are traditionally referred for cardiac transplantation. 1,2 Unfortunately, the number of heart transplantations performed in the United States is limited by the availability of donor organs to 2200 per year, a small fraction of the total patients requiring advanced heart failure care. 3 Moreover, 75% of patients with end-stage heart failure are 65 years and have severe comorbid conditions, making them ineligible for transplantation at many centers. 4 As a result of innovative trials, LVADs have become a therapeutic option in patients with end-stage heart failure as bridge-to-transplantation or destination therapy in patients deemed ineligible for cardiac transplantation. 5,6 Contemporary LVADs are continuous-flow pumps that allow for smaller pump size and improved mechanical reliability and durability. 7 In a landmark study 8 comparing the HeartMate XVE, an older first-generation pulsatile pump (Thoratec Corp; Pleasanton, CA), with the newer HeartMate II LVAD (Thoratec Corp), the continuous-flow pump outperformed the pulsatile device in overall survival, pump failure events, and infectious complications. Device technology continues to improve, and thirdgeneration continuous-flow LVADs under development use magnetic or hydrodynamic bearings for frictionless rotation of the rotor that may further enhance durability. 9 One potential limitation of continuous-flow pump technology is the relatively high nonphysiological shear stress imparted on blood components as they move through the device. This stress, coupled with the reduced pulse pressure associated with axial-flow LVADs, may cause unanticipated increases in nonsurgical bleeding events secondary to arteriovenous malformations (AVMs) in the gastrointestinal (GI) tract and the development of acquired von Willebrand disease. Bleeding, especially from a GI tract source, has recently been identified as 1 of the most common adverse events after implantation and is a major cause of morbidity in patients supported by LVAD therapy Fortunately, there have been no documented increases in mortality associated with these increased rates of GI tract bleeding. With the requirement for systemic anticoagulation after implantation of contemporary LVADs, an advanced understanding of the mechanisms underlying these bleeding risks is necessary to guide postimplantation management and therapy. In this review, we summarize the literature evaluating the prevalence, etiology, outcomes, and management of bleeding events in the LVAD population. We also identify areas where more studies are necessary to guide future therapies and interventions. Our comments will primarily focus on the population of patients treated with axial-flow devices, such as the US Food and Drug Administration approved HeartMate II, which propels blood continuously along the axis of a helical rotor known as an impeller. The limited experience with newer continuousflow pumps that use a centrifugal flow mechanism requiring lower revolutions per minute does not permit extrapolation of these comments to all continuous-flow pumps. Bleeding Events in Axial-Flow LVADs Incidence Recent data suggest that 1 of the most common adverse events within the first 30 days after LVAD implantation is nonsurgical bleeding. 7,10,12,14 16 The Table summarizes the reported bleeding events in studies that included patients treated with an axial-flow LVAD. In a multicenter study 14 demonstrating the effectiveness of the HeartMate II axialflow LVAD in patients awaiting a heart transplantation, the most common adverse event within 30 days of implantation was nonsurgical bleeding, resulting in surgery or significant blood transfusions. Across studies, the most commonly reported sources of bleeding are epistaxis, GI tract bleeding, bleeding of the mediastinum and thorax, and intracranial hemorrhage. 7,12,14 Data from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) have shown the most frequent locations of first bleeding episode after implantation to be mediastinal (45%), thoracic pleural space (12%), lower GI tract (10%), chest wall (8%), Received April 7, 2011; accepted September 15, From the Duke Clinical Research Institute and the Department of Medicine, Duke University School of Medicine, Durham, NC. Correspondence to Chetan B. Patel, MD, DUMC 3034, Durham, NC chetan.patel@duke.edu (Circ Heart Fail. 2011;4: ) 2011 American Heart Association, Inc. Circ Heart Fail is available at DOI: /CIRCHEARTFAILURE

2 780 Circ Heart Fail November 2011 Table. Reported Bleeding Events in Axial-Flow LVADs Source Rate of Adverse Event, % Definition of Bleeding All bleeding Pagani et al, Requiring surgery 53 Requiring 2 PRBCs Slaughter et al, Requiring surgery 81 Requiring PRBCs Pal et al, Requiring surgery GI tract bleeding Crow et al, All bleeding events that occurred past postoperative day 15 Genovese et al, Cumulative incidence by 60 d after implantation Uriel et al, Any nonsurgical bleeding requiring transfusion Crow et al, Bleeding requiring any type of intervention PRBC indicates packed red blood cell. and upper GI tract (8%), with no difference in the overall bleeding rates between axial- and pulsatile-flow devices. 18 However, the location of bleeding may vary depending on time from operation, with thoracic or mediastinal bleeding having higher, but diminishing, risk in the early postoperative period, with a delayed risk of GI tract bleeding. In a smaller study 19 of 48 patients who underwent transplantation with an axial-flow device, GI tract bleeding appeared late after implantation, with an average of 460 days of ventricular support. Small retrospective analyses 10,16 have shown significantly increased rates of GI tract bleeding with the use of axial-flow devices when compared with pulsatile-flow devices, which may be related to rates of readmission for axial-flow devices. Indeed, the INTERMACS data have also shown axial-flow devices to have overall decreased rates of readmission compared with pulsatile-flow devices, but with more axial-flow readmissions due to bleeding (9.8%) than with pulsatile devices (6.2%). 20 Therefore, the increased rates of GI tract bleeding after implantation of axial-flow devices continues to be a source of morbidity, resulting in repeated hospitalizations that may persist long after implantation. Although the INTERMACS registry is the largest experience with assist devices, providing a wealth of data on postimplantation survival and complications of LVAD therapy, we continue to acknowledge certain limitations with this growing resource. Primary among these is the evolving definition of adverse events, such as bleeding, and what constitutes a GI tract bleeding event related to LVAD therapy versus patient- or provider-driven complications, such as underlying propensity for bleeding or overanticoagulation. Methods of reporting the frequency of these events are also evolving. The INTERMACS has been instrumental in providing definitions for adverse events; however, as the field moves forward, it will remain important to provide uniform methods for reporting rates of these events as patients with LVADs survive longer and are exposed to the device and concurrent anticoagulant therapy for a longer time. This will allow improved guidance for clinicians in patient selection, postoperative screening, and anticoagulation management in the short-term and during the entire time LVAD therapy is provided. Role of Anticoagulation With the recommendation that all patients with an axial-flow device be treated with antiplatelet medications and warfarin, an obvious explanation for the increased bleeding rates is the use of anticoagulation therapy, which was not a requirement in the pulsatile VAD era dominated by the HeartMate XVE. 17 Therefore, some of the observed increase in bleeding rates must be attributed to the manufacturer s recommendation for anticoagulation. This recommendation was made after initial concerns that the devices might predispose patients to thromboembolic events and followed guidelines developed from the HeartMate II Pivotal Trial. 21 In an attempt to mitigate thrombotic risk, the HeartMate II LVAD was designed with specific features to minimize thromboembolic complications, including the use of biomaterials, such as titanium, polytetrafluoroethylene, and acrylics. 8 Despite these modifications, no material can be considered biologically inert, and any biomaterial in direct contact with circulating blood can alter immunologic and thrombotic proteins, thus activating the coagulation system. 8 Consequently, all patients are given an anticoagulation regimen, despite the previously mentioned increased risk of bleeding. Interestingly, Crow et al 10 recently concluded that the increased observed rates of bleeding in patients treated with axial flow LVADs were higher than would be expected from anticoagulation therapy alone. In fact, with current rates of GI tract and intracranial bleeding reported as high as 30% and 11%, respectively, 7,16 analyses 7,21 have suggested that the risk of bleeding is greater than that of thromboembolism. This has led to center-specific changes in anticoagulation management after both implantation and a bleeding event (Figure 1), with a shift to a decrease in target international normalized ratio (INR) at many centers. 8 For instance, to reduce GI tract bleeding risk in these patients, the Henry Ford Hospital (Detroit, Mich) has lowered the anticoagulation target to an INR of 1.8 to 2.2, with no increase in thromboembolic complications. 18 Although there have been no reports of increased thrombosis or embolic stroke events from other centers modifying their anticoagulation regimens, this has been a recent change in practice and we await longer-term follow-up. Additional efforts have also been undertaken to understand the mechanisms associated with heightened bleeding risk after the implantation of axial-flow devices. Angiodysplasia, impaired platelet aggregation, and a newly identified acquired von Willebrand disease have all been proposed as possible mechanisms responsible for these observations. 22 Acquired von Willebrand Disease With Axial-Flow LVADs The von Willebrand factor (vwf) molecule is a 250-kDa protein expressed by vascular endothelial cells that assembles into multimers that bind to exposed collagen of damaged blood vessels. It also binds to platelet receptors to induce activation, adhesion, and aggregation. 22 Homeostasis of the vwf multimer size is necessary to prevent a pathological

3 Suarez et al LVADs: Mechanisms of Bleeding and Management 781 Figure 1. Algorithm for the evaluation and treatment of gastrointestinal (GI) tract bleeding (the Duke approach): In the event that a patient supported with an axial-flow device develops GI tract bleeding, consultation from a GI specialist is of benefit to identify the bleeding source through endoscopy or use of the PillCam device. Initially, antiplatelet medication should be discontinued and bleeding should be reassessed. If bleeding continues, warfarin should be discontinued. Once the patient is hemodynamically stable, a target international normalized ratio (INR) of 1.5 to 2.0 is recommended. The medical team should consider alternative therapies to stop bleeding, including cryoprecipitate, platelets, factor VIII, octreotide, and desmopressin (DDAVP) in the event of uncontrolled hemorrhage. coagulopathy and is primarily regulated by the cleavage activity of the metalloprotease ADAMTS13. ADAMTS13 is a protease that binds to and cleaves the vwf protein at a specific site (the A2 domain) that is buried in the tertiary structure of the protein until the molecule unfolds. A deficiency of this protease is responsible for large vwf multimers, which, in turn, are responsible for the development of the prothrombotic condition, thrombotic thrombocytopenic purpura. 23 Conversely, excessive cleavage of large multimers results in vwf multimer depletion and pathological bleeding, as seen in acquired von Willebrand disease. The relationship between the loss of high-molecularweight (HMW) vwf multimers (acquired von Willebrand disease) and GI tract bleeding was first described in a phenomenon called Heyde syndrome. This syndrome affects individuals who have severe aortic stenosis and concomitantly develop a combination of acquired von Willebrand disease and GI tract bleeding. 24 Multiple mechanisms have been proposed for the development of this syndrome. One potential explanation for the acquired von Willebrand disease invokes high shear stress across the calcified aortic valve, inducing a conformational change on the vwf multimers and predisposing them to proteolytic cleavage. 10 Another contributing factor is increased intraluminal pressure within the vasculature and the lowered pulse pressure, 22 which is discussed in detail later in this review. Because axial-flow LVAD physiological characteristics involve high shear stress and narrow pulse pressures, it is intriguing that patients treated with contemporary LVADs have bleeding patterns similar to those described with severe aortic stenosis. Geisen et al 11 compared blood samples from heart transplant recipients and patients treated with axial-flow LVADs to examine the role of impaired hemostasis after the initiation of continuous-flow therapy. Both groups had comparable prothrombin time, platelet count, and total vwf levels. A key difference in the LVAD population was the reduced number of HMW vwf multimers, 11 an observation that has been replicated in multiple cohorts. 16,17,24,25 Although consistent across studies, the mechanisms underlying the loss in HMW vwf are under intense investigation; at this time, they remain putative. A particular focus has been the impact of shear stress on vwf cleavage, a relationship defined experimentally by Tsai et al 26 in 1994, well before the development of contemporary LVADs. By perfusing normal plasma through long capillary tubing, these investigators observed a direct relationship between the shear rate and the loss of HMW vwf multimers. They concluded that shear stress promoted vwf proteolysis by the metalloprotease ADAMTS13, leading to a decrease in multimer size. 26 By using molecular simulation, Baldauf et al 27 demonstrated that the high shear stress rates resulted in elongation and unfolding of the vwf multimers. Consequently, the enhanced exposure of the previously buried A2 domain results in the cleavage by the ADAMTS13 metalloprotease. As the shear stress rate increases, the molecule is prone to unfolding, resulting in greater cleavage rates and smaller vwf multimers. 27 Multiple studies 16,17,28 have shown that most patients who undergo implantation with an axial-flow LVAD will develop some degree of acquired von Willebrand disease. In a retrospective review 16 of 79 patients with a HeartMate II LVAD, all patients had decreased or absent HMW vwf multimers that normalized after heart transplantation. Similar findings were reported in a multicenter study involving the Duke University Medical Center and the University of Minnesota. In this prospective observational study, 17 all patients demonstrated significant HMW vwf multimer-level reductions within 30 days after LVAD implantation. An analysis of patients with different causes of heart failure has shown that no patient with normal HMW vwf multimer levels developed bleeding events, 16 thus supporting the role of HMW vwf multimer loss in the development of GI tract bleeding. An important limitation of these observations is the

4 782 Circ Heart Fail November 2011 Figure 2. Mechanisms implicated in gastrointestinal (GI) tract bleeding in patients with axial-flow left ventricular assist devices (LVADs): Patients treated with axial-flow LVADS are at increased risk of developing GI tract bleeding. There are several mechanisms implicated with these increased risks. The device leads to increased intraluminal pressure and lowered pulse pressure, resulting in hypoperfusion of the intestines. These physiological changes result in an increased risk of developing angiodysplasia. These devices also decrease the highmolecular-weight (HMW) von Willebrand factor (vwf) multimer size because of excessive cleavage by the metalloprotease ADAMTS13. This results in an acquired form of vwf disease. Although not studied in detail, vwf fragments from the breakdown of HMW multimers could be involved in the inhibition of platelet aggregation. These mechanisms work synergistically to cause GI tract bleeding. method used for assessment of HMW vwf loss and evaluation for remaining multimer size. Most studies 16,17 have used a qualitative assessment of HMW multimer loss, and no standard method for protein isolation and quantification has been successfully developed. Therefore, with current methods, risk of bleeding (or, conversely, protection from risk of thromboembolism) cannot be assessed quantitatively using vwf multimer size as an isolated marker. However, this observation is unlikely to simply be an epiphenomenon associated with bleeding. A functional assay for vwf activity (eg, the ristocetin cofactor assay) is often used to confirm the diagnosis of acquired von Willebrand disease; in many of the reported studies, 16,17 these assays demonstrate impaired vwf activity during LVAD support, with correction after removal of the device. 16 Angiodysplasia Formation in the Axial-Flow LVAD Population The loss of HMW vwf multimers is not the only putative pathophysiological mechanism for GI tract bleeding shared by axial-flow LVADs and aortic stenosis. There is a strong association between aortic stenosis and the development of AVMs. 29 One proposed mechanism suggests that increased intraluminal pressures and vascular smooth muscle contractions may lead to increased sympathetic tone, causing smooth muscle relaxation, arteriovenular dilation, and, ultimately, the development of AVMs. 29 Another hypothesis is that intestinal hypoperfusion from reduced pulse pressure leads to regional hypoxia, vascular dilation, and subsequent angiodysplasia. 10 In a study 30 of 172 patients receiving axial-flow support, 31% of GI tract bleeding events were caused by AVMs, with these patients being significantly older than the other LVAD recipients. Therefore, older age may also be a contributing factor to the development of AVMs in patients receiving an assist device. Although not investigated directly with pre- and post-lvad endoscopy, the same mechanisms responsible for AVM formation in aortic stenosis are plausible in patients treated with axial-flow devices given the chronic low pulse pressure associated with the pump. Therefore, even in the absence of firm data, it remains feasible that acquired von Willebrand disease and the increased risk for AVMs in patients treated with axial flow LVADs likely work synergistically to increase the risk of GI tract bleeding. This is especially true because HMW vwf multimers are essential for platelet-mediated hemostasis in high shear stress areas, such as GI tract malformations and AVMs (Figure 2). 10,22 Unfortunately, although modification in antiplatelet and anticoagulation regimens may mitigate some of the bleeding risk associated with angiodysplasia, the rate of development of new AVMs with continuous flow has not been well established. Therefore, it is unclear whether an isolated bleeding episode from AVMs predicts future bleeding events in either the GI tract or even the cerebral circulation. These relationships should be extensively considered when assessing the competing risks of bleeding and thrombosis in the management of the LVAD patient with AVMs. Impaired Platelet Aggregation in the Axial-Flow LVAD Population In addition to the loss of HMW vwf multimers and the increased risk for angiodysplasia, patients treated with axial flow LVADs have also developed impaired platelet aggregation, which further contributes to the increased tendency to bleed. Klovaite et al 25 measured the ristocetin-induced platelet aggregation of 16 patients who had recently received a HeartMate II LVAD. Of the 16 patients, 11 had impaired ristocetin-induced platelet aggregation and a history of minor and major bleeding events. 25 In this study, HMW vwf levels were decreased in these patients, but activity remained within the defined normal range. Therefore, the investigators concluded that the loss of HMW vwf multimers was not solely responsible for the observed reduction in ristocetin-induced platelet aggregation. An alternative mechanism suggested by this group involved inhibition of platelet aggregation by vwf fragments generated from the breakdown of the HMW vwf multimers. 25 The relative contribution to bleeding risk from acquired von Willebrand

5 Suarez et al LVADs: Mechanisms of Bleeding and Management 783 disease and impaired platelet activity has yet to be studied in detail. Assessing Thromboembolic and Bleeding Risk When axial-flow LVADs were initially used clinically, the potential and observed risks of pump thrombosis and thromboembolic complications resulted in recommendations for high levels of anticoagulation and antiplatelet therapy after implantation. 21 Balancing bleeding and thrombotic complications has become a difficult clinical dilemma, made more challenging by poorly understood individual differences in patient risk profiles. Recent studies suggest that the initial aggressive anticoagulation and antiplatelet therapy recommendations resulted in excessive bleeding risk with trivial thrombotic complications. A single-center analysis showed that the HeartMate II was associated with an extremely low thromboembolic risk and required less stringent anticoagulation. In fact, in this study, only 5% of patients had a thromboembolic event, compared with 53% of patients with a bleeding event requiring 2 U of packed red blood cells. 8 Likewise, in a prospective study 21 of 331 patients, the few thrombotic events were offset by the greater risk of hemorrhagic events. However, hemorrhagic events can be severe and may not only involve the GI tract. The reported incidence of intracranial hemorrhage with axial-flow LVADs has been as high as 0.05 events per patient-years of support and is an important cause of morbidity and mortality in VAD trials. 14 Moreover, the devastating impact of these events requires careful consideration of the risk-benefit ratio of anticoagulation therapy. The best available evidence (in the absence of long-term followup) suggests that thrombotic and hemorrhagic events are minimized with dose-adjusted warfarin, targeting an INR range of 1.5 to 2.5, in addition to 81 mg of aspirin therapy. 21 However, the optimal INR goal in LVAD-treated patients with other indications for anticoagulation, such as atrial fibrillation or prior thromboembolic stroke, is yet to be established. In the absence of firm data, we recommend more aggressive anticoagulation with a goal INR in the range of 2 to 3. Indeed, balancing bleeding and thrombosis risks will be a critical component of LVAD patient management moving forward, especially with the advent of novel anticoagulant agents. One example of where global risk assessment is imperative in minimizing both bleeding and thrombosis risk is in the bridge-to-transplantation VAD population. From a bleeding standpoint, numerous reports 15 suggest that repeated blood transfusions engender the development of antihuman leukocyte antigen antibodies, which may limit future candidacy for successful transplantation and increase the risk for rejection episodes. However, hemorrhagic risk must be weighed against that of a single catastrophic thromboembolic stroke, which would reduce the quality of life of the patient and possibly exclude the patient from future cardiac transplantation. Management of GI Tract Bleeding As evidenced by the preceding discussion, the management of GI tract bleeding in the LVAD population can be challenging for clinicians; therefore, we pose some recommendations. Because of the difficulty identifying the source of GI tract bleeding, it is recommended that a gastroenterologist be part of the clinical team managing the patient. Upper and lower GI tract endoscopy should be the standard of care to assess the site of bleeding after a hemorrhagic event. However, the source of bleeding is often difficult to ascertain with traditional endoscopic methods. In such cases, the wireless PillCam SB (Given Imaging, Ltd; Yoqneam, Israel) capsule has shown promise in identifying bleeding sources in these patients, particularly in the small bowel. 31 Although concerns have arisen regarding the possibility of the LVAD interfering with the signal transmission from the capsule or the capsule causing LVAD malfunction, to our knowledge, no interaction between the HeartMate II and the PillCam SB has been reported. Therefore, this technology has become an important diagnostic tool in the management of GI tract bleeding, if available at the treating center 31 ; however, limited evidence exists for benefit from routine or early endoscopy if pre- LVAD endoscopy did not demonstrate a potential bleeding source. These recommendations have been included in Figure 1, which depicts an algorithm used at our institution for the evaluation of GI tract bleeding. Future Studies An advanced understanding of the hemostatic abnormalities associated with axial-flow LVADs is necessary to guide future interventions and therapies. An understanding of the pathophysiological mechanisms associated with heightened bleeding risk is particularly important in the absence of clearly defined clinical risk factors for bleeding or thrombosis. Furthermore, carefully conducted studies, with a focus on bleeding and thrombosis as a primary end point, should be implemented as the use of these devices increases. An intriguing study might involve prospectively exploring the role of pulsatility in axial-flow pumps and its effects on GI tract bleeding. Potential outcomes to measure would include GI tract bleeding events and changes in the HMW vwf levels. Prospective documentation of anatomic GI tract changes over time, both before and after LVAD implantation, would also be helpful to define temporal trends in AVM formation. As previously mentioned, this could also be used to stratify patient bleeding risks based on preimplantation and postimplantation GI tract pathological characteristics and to determine the burden of AVMs in the advanced heart failure population. Beyond observational analyses, a randomized trial examining bleeding and thrombosis rates in a warfarin-only group compared with a warfarin plus antiplatelet therapy group would be beneficial in modifying current anticoagulation guidelines. As we enter an era with increasing access to information, laboratory tests, and potentially other complex tests of biological signatures, we may be able to evaluate tailored approaches to anticoagulation that dynamically change during the course of treatment of patients with LVADs. Summary Axial-flow LVADs have become an integral tool in the management of end-stage heart failure. Consequently, nonsurgical bleeding has emerged as a major source of morbidity and mortality in this fragile population. The mechanisms responsible for these adverse events include acquired von Willebrand disease, GI tract angiodysplasia formation, impaired platelet aggregation, and overuse of anticoagulation therapy. Because of

6 784 Circ Heart Fail November 2011 ongoing concerns for pump thrombosis and thromboembolic events, the thrombotic/bleeding paradigm has led to a difficult clinical dilemma for those managing patients treated with axial flow LVADs. As the field progresses, advances in the understanding of the pathological mechanisms underlying bleeding/ thrombosis risk, careful risk stratification, and potential use of novel anticoagulants will all play a role in the management of the LVAD patient. Disclosures Dr Becker served as a consultant ( $10 000) to Boehringer Ingelheim, Johnson & Johnson, Bayer, Regado, BMS, AstraZeneca, and Merck; and Dr Rogers served as a consultant ( $10 000) to Thoratec and received honoraria ( $10 000) from HeartWare. References 1. Rogers JG, Butler J, Lansman SL, Gass A, Portner PM, Pasque MK, Pierson RN III. Chronic mechanical circulatory support for inotropedependent heart failure patients who are not transplant candidates: results of the INTrEPID Trial. J Am Coll Cardiol. 2007;50: Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W, Long JW, Ascheim DD, Tierney AR, Levitan RG, Watson JT, Meier P, Ronan NS, Shapiro PA, Lazar RM, Miller LW, Gupta L, Frazier OH, Desvigne-Nickens P, Oz MC, Poirier VL. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med. 2001;345: Mancini D, Lietz K. Selection of cardiac transplantation candidates in Circulation. 2010;122: Daneshmand MA, Rajagopal K, Lima B, Khorram N, Blue LJ, Lodge AJ, Hernandez AF, Rogers JG, Milano CA. Left ventricular assist device destination therapy versus extended criteria cardiac transplant. Ann Thorac Surg. 2010;89: Dembitsky WP, Tector AJ, Park S, Moskowitz AJ, Gelijns AC, Ronan NS, Piccione W Jr, Holman WL, Furukawa S, Weinberg AD, Heatley G, Poirier VL, Damme L, Long JW. Left ventricular assist device performance with long-term circulatory support: lessons from the REMATCH trial. Ann Thorac Surg. 2004;78: Hernandez AF, Shea AM, Milano CA, Rogers JG, Hammill BG, O Connor CM, Schulman KA, Peterson ED, Curtis LH. Long-term outcomes and costs of ventricular assist devices among Medicare beneficiaries. JAMA. 2008;300: Slaughter MS, Rogers JG, Milano CA, Russell SD, Conte JV, Feldman D, Sun B, Tatooles AJ, Delgado RM III, Long JW, Wozniak TC, Ghumman W, Farrar DJ, Frazier OH. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med. 2009;361: John R, Kramdar F, Liao K, Colvin-Adams M, Miller L, Joyce L, Boyle A. Low thromboembolic risk for patients with the Heartmate II left ventricular assist device. J Thorac Cardiovasc Surg. 2008;136: Gregoric ID, Cohn WE, Frazier OH. Diaphragmatic implantation of the HeartWare ventricular assist device. J Heart Lung Transplant. 2011;30: Crow S, John R, Boyle A, Shumway S, Liao K, Colvin-Adams M, Toninato C, Missov E, Pritzker M, Martin C, Garry D, Thomas W, Joyce L. Gastrointestinal bleeding rates in recipients of nonpulsatile and pulsatile left ventricular assist devices. J Thorac Cardiovasc Surg. 2009; 137: Geisen U, Heilman C, Beyeresdorf F, Benk C, Berchtold-Herz M, Schlensak C, Budde U, Zieger B. Non-surgical bleeding in patients with ventricular assist devices could be explained by acquired von Willebrand disease. Eur J Cardiothorac Surg. 2008;33: Genovese EA, Dew MA, Teuteberg JJ, Simon MA, Kay J, Siegenthaler MP, Bhama JK, Bermudez CA, Lockard KL, Winowich S, Kornos RL. Incidence and patterns of adverse event onset during the first 60 days after ventricular assist device implantation. Ann Thorac Surg. 2009;88: Stern DR, Kazam J, Edwards P, Maybaum S, Bello RA, D Alessandro DA, Goldstein FJ. Increased incidence of gastrointestinal bleeding following implantation of the HeartMate II LVAD. J Card Surg. 2010;25: Pagani FD, Miller LW, Russell SD, Aaronson KD, John R, Boyle AJ, Conte JV, Bogaev RC, MacGillivray TE, Naka Y, Mancini D, Massey HT, Chen L, Klodell CT, Aranda JM, Moazami N, Ewald GA, Farrar DJ, Frazier OH. Extended mechanical circulatory support with a continuous-flow rotary left ventricular assist device. J Am Coll Cardiol. 2009;54: Pal JD, Piacentino V, Cuevas AD, Depp T, Daneshmand MA, Hernandez AF, Felker GM, Lodge AJ, Rogers JG, Milano CA. Impact of left ventricular assist device bridging on posttransplant outcomes. Ann Thorac Surg. 2009;88: Uriel N, Pak SW, Jorde UP, Jude B, Susen S, Vincentelli A, Ennezat PV, Cappelman S, Naka Y, Mancini D. Acquired von Willebrand syndrome after continuous-flow mechanical device support contributes to a high prevalence of bleeding during long-term support and at the time of transplantation. J Am Coll Cardiol. 2010;56: Crow S, Chen D, Milano C, Thomas JW, Joyce L, Piacention V III, Sharma R, Wu J, Arepally G, Bowles D, Rogers J, Villamizar-Ortiz N. Acquired von Willebrand syndrome in continuous-flow ventricular assist device recipients. Ann Thorac Surg. 2010;90: Jessup ML, Goldstein D, Ascheim DD, Teuteberg JJ, Park SJ, Naftel DC, Gonzales-Stawirski GV, Ulisney KL, Kirklin JK, Kormos RL. Risk for bleeding after MCSD implant: an analysis of 2358 patients in INTERMACS. J Heart Lung Transplant. 2011;30(suppl I):S Morgan J, Brewer R, Nemeh H. Gastrointestinal bleeding with the Heartmate II left ventricular assist device. Abstract presented at: American Society of Artificial Internal Organs Annual Conference; June 11, 2011; Washington, DC. 20. Bonde P, Dew MA, Meyer D, Tallaj JJ, Martin T, Hollifield KA, Slaughter M, Starling R, Kormos RL. National trends in readmission (REA) rates following left ventricular assist device (LVAD) therapy. J Heart Lung Transplant. 2011;30(suppl I):S Boyle AJ, Russell SD, Teuteberg JJ, Slaughter MS, Moazami N, Pagani FD, Frazier OH, Heatley G, Farrar DJ, John R. Low thromboembolism and pump thrombosis with the HeartMate II left ventricular assist device: analysis of outpatient anti-coagulation. J Heart Lung Transplant. 2009; 28: John R, Boyle A, Pagnani F, Miller L. Physiologic and pathologic changes in patients with continuous-flow ventricular assist devices. J Cardiovasc Transl Res. 2009;2: Zhou Z, Nguyen TC, Guchhait P, Dong JF. Von Willebrand factor, ADAMTS-13, and thrombotic thrombocytopenic purpura. Semin Thromb Hemost. 2010;36: Vincentelli A, Susen S, LeTourneau T, Six I, Fabre O, Juthier F, Bauters A, Decoene C, Goudemand J, Prat A, Jude B. Acquired von Willebrand syndrome in aortic stenosis. N Engl J Med. 2003;349: Klovaite J, Gustafsson F, Mortensen SA, Sander K, Nielsen LB. Severely impaired von Willebrand factor-dependent platelet aggregation in patients with a continuous-flow left ventricular assist device (HeartMate II). JAm Coll Cardiol. 2009;53: Tsai HM, Sussman II, Nagel RL. Shear stress enhances the proteolysis of von Willebrand factor in normal plasma. Blood. 1994;83: Baldauf C, Schneppenheim R, Stacklies W, Obser T, Pieconka A, Schneppenheim S, Budde U, Zhou J, Grater F. Shear-induced unfolding activates von Willebrand factor A2 domain for proteolysis. J Thromb Haemost. 2009;7: Crow S, Milano C, Joyce L, Chen D, Arepally G, Bowles D, Thomas W, Ortiz NV. Comparative analysis of von Willebrand factor profiles in pulsatile and continuous left ventricular assist device recipients. ASAIO J. 2010;56: Letsou GV, Shah N, Gregoric ID, Myers TJ, Delgado R, Frazier OH. Gastrointestinal bleeding from arteriovenous malformations in patients supported by the Jarvik 2000 axial-flow left ventricular assist device. J Heart Lung Transplant. 2005;24: Demirozu ZT, Radovancevic R, Hochman LF, Gregoric ID, Letsou GV, Kar B, Bogaev RC, Frazier OH. Arteriovenous malformation and gastrointestinal bleeding in patients with the HeartMate II left ventricular assist device. J Heart Lung Transplant. 2011;30: Bechtel JF, Wellhoner P, Charitos EI, Bucsky B, Morshmis M, Sievers HH. Localizing an occult gastrointestinal bleeding by wireless PillCam SB capsule videoendoscopy in a patient with the HeartMate II left ventricular assist device. J Thorac Cardiovasc Surg. 2010;139:e73 e74. KEY WORDS: left ventricular assist devices bleeding rates thrombosis gastrointestinal bleeding