Am J Cardiovasc Drugs DOI 10.1007/s40256-014-0102-3
LEADING ARTICLE
Management of Pump Thrombosis in Patients with Left Ventricular Assist Devices John M. Stulak • Shashank Sharma Simon Maltais
•
Ó Springer International Publishing Switzerland 2015
Abstract The gradual evolution of left ventricular assist device (LVAD) therapy has resulted in a durable option for patients as either a bridge to transplantation (BTT) or a destination therapy (DT). Outcomes with current continuous-flow devices continue to demonstrate significant patient benefit, not only in enhanced survival but also in improved functional capacity and quality of life. While the lessening of adverse events through time has resulted in more widespread adoption of this therapy, there continues to be unintended consequences, including, most notably, infection, bleeding, and thrombosis. Beginning in 2011, centers and collaborative groups began to observe a significant increase in the incidence of pump thrombosis with the HeartMate II LVAD (Thoratec Corp., Pleasanton, CA, USA). However, this clinical scourge is not limited to the HeartMate II, as the HeartWare Ventricular Assist System (HVAD; HeartWare Inc., Framingham, MA, USA) has also had these same issues, which led to pump modifications and the appreciation of more strict control of blood pressure and anticoagulation with this pump design. We review the current status of the field of mechanical circulatory support in its approach to diagnosis, management, and prevention of LVAD pump thrombosis.
J. M. Stulak (&) S. Sharma Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905, USA e-mail: [email protected] S. Maltais Vanderbilt Heart and Vascular Institute, Nashville, TN, USA
1 Introduction to the Problem of Pump Thrombosis 1.1 Defining the Problem The definition of pump thrombosis has evolved as our understanding of this entity has grown. The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) is a collaborative database sponsored by the National Heart, Lung, and Blood Institute (NHLBI). Data are collected on all durable mechanical circulatory support devices approved by the US FDA and implanted in the USA, and annual reports are published [1–6]. In addition, the INTERMACS Executive Committee publishes adverse event (AE) definitions that centers utilize to ensure standardization in the reporting of AEs. The most recent version of AE definitions was approved on 15 May 2013. Relevant to the problem discussed herein are the definitions of hemolysis and pump thrombus. Hemolysis in a patient supported on left ventricular assist device (LVAD) therapy is defined as either minor or major depending on certain biochemical and clinical criteria. Minor hemolysis is defined as a plasma-free hemoglobin [20 mg/dl or a serum lactate dehydrogenase (LDH) [2.5 times the upper limits of the normal range at the implanting center. These findings should be observed after the first 72 h of implant and are in the absence of clinical symptoms, manifestations of hemolysis, or abnormal pump function. Major hemolysis includes the criteria for ‘minor’ hemolysis and associated with hemoglobinuria, anemia, hyperbilirubinemia, or pump malfunction and/or abnormal pump parameters. Pump thrombus is considered a specific case of a major device malfunction and is classified as either suspected pump thrombus or confirmed pump thrombus. With suspected pump thrombus, clinical patient condition or pump
J. M. Stulak et al.
parameters suggest thrombus on any of the blood-contacting surfaces of the pump (inflow cannula, pump itself, or outflow graft). Criteria (at least two present) should include presence of hemolysis, presence of heart failure not explained by structural disease, or abnormal pump parameters. Confirmed pump thrombus can be observed by visual inspection (at the time of device exchange, transplantation, or autopsy), irrefutable radiographic evidence, or absence of Doppler inflow or outflow signals. 1.2 The Historical and Contemporary Magnitude of the Problem Both in the clinical trial setting and in the post-approval era, numerous series have documented an unavoidable rate of pump thrombosis for both the HeartMate II (Thoratec Corp., Pleasanton, CA, USA) and the HeartWare Ventricular Assist System (HVAD; HeartWare Inc., Framingham, MA, USA) (Table 1). Concern regarding LVAD pump thrombosis has been heightened recently due to two publications: (1) an initial report by investigators who documented an abrupt increase in the incidence of confirmed pump thrombosis at 3 months with the HeartMate II from 2.2 % before March 2011 to 8.4 % after January 2013, and (2) a detailed analysis of the INTERMACS database that reported a sixfold increase in rates of pump thrombosis in the timeframe of 2011 to 2012 [8, 9]. Furthermore, the 6-month actuarial freedom from device exchange or death due to pump thrombosis has significantly fallen from 99 % in the 2008–2009 time period to 95 % in 2011 and 94 % in 2012 [9]. From a medical management standpoint, it is interesting to note a temporal relationship of these findings to reports documenting the safe adoption of lower anticoagulation levels in patients with the HeartMate II, including the safe non-use of perioperative intravenous heparin [10, 11].
Unfortunately, despite these data trends, firm conclusions are unable to be drawn as to causality, whether patient or device specific. Interesting findings were also observed by the HeartWare investigators after concern around the high incidence (0.07 events per patient year) of HeartWare HVAD pump thrombosis in their clinical trial prompted a detailed analysis [12, 13]. Interestingly, they found that the majority of pump thromboses occurred in patients with sub-therapeutic warfarin anticoagulation and taking lowdose or no antiplatelet therapy. After recommendations were made for strict patient management to adhere to an international normalized ratio (INR) of 2–3 and high-dose aspirin (325 mg), the rate of device exchange due to pump thrombosis fell by more than 50 % [12, 13]. 1.3 Potential Causes of the Problem Analysis of the available reports to date stresses the complexity of the phenomenon of LVAD pump thrombosis, and the causes can be differentiated into non-mechanical and mechanical. In relation to the former, while the trend of lowering anticoagulation that was enacted in order to lessen bleeding complications and its temporal relationship to new concerns regarding pump thrombosis is intriguing, and likely contributory to some degree, one must concede that the presence of an LVAD represents a very complex clinical arrangement. It follows then that this problem, which arises within such a complex system, is also complex in its etiology. In parallel with the relationship of maintenance anticoagulation and antiplatelet therapy, some have demonstrated a subsequent increased risk of thromboembolic events following gastrointestinal bleeding and resultant alteration in anticoagulation [14]. Aside from anticoagulation, one must recall that these pumps are afterload sensitive, with the third-generation pumps, such as HeartWare HVAD, to a greater degree. Flow through the
Table 1 Multicenter reports of rates of pump thrombosisa References
Year
Pt (n)
Incidence
Freedom from event (%) 6 months
1 year
92
Park et al. [7]
2005–2007
133
Park et al. [7]
2007–2009
281
0.038/py
Starling et al [8]
2004–2013
837
7.9 % confirmed
95
Kirklin et al. [9]
2008–2012
6,910
5.5 % (382/6,910)
97
95
Najjar et al. [12]
2008–2012
382
8.1 %, 0.08/py
96
92
Stulak et al. [32]
2004–2012
516
3 mo: 1.7 %, 1 y: 4.7 %
a
0.024/py
Pump thrombosis was defined as follows: Park et al. [7], Starling et al. [8]: Confirmed: thrombus noted on blood-contacting surfaces during urgent transplant, device exchange, or autopsy; Suspected: pump malfunction with clinical variables suggested thrombus on pump components; Kirklin et al. [9]: pump exchange for documented or probable thrombus, death with device malfunction with documented or probable thrombus; Najjar et al. [12]: alterations in pump parameters with increased markers of hemolysis and/or thrombus visualization after pump exchange; Stulak et al. [32]: Confirmed: thrombus noted on blood-contacting surfaces during urgent transplant, device exchange, or autopsy mo month, pt patient, py patient-year, y year
Management of Pump Thrombosis in LVAD Recipients
pump itself is lower in the presence of a high afterload and some have observed an association between high blood pressure and pump thrombosis [12]. Certainly, this issue requires more detailed analysis. One further non-mechanical issue that has undergone some philosophical changes is the status of the aortic valve opening following LVAD implantation. After concerns arose surrounding progression of aortic regurgitation following LVAD implant, alterations in pump management toward lower speeds in an attempt to foster aortic valve opening were suggested [15]. The benefit of this strategy was felt not only to reduce progression of aortic regurgitation but also to reduce development of gastrointestinal arteriovenous malformations and gastrointestinal bleeding. Similar to the aforementioned anticoagulation management, this strategy, aimed at decreasing complications, yielded unintended consequences, and, while permitting greater native heart function, it was found to decrease flow through the pump and increase the risk of subsequent pump thrombosis [16]. Lastly, and arguably most significantly, the unique interaction of the patient’s blood and the blood-contacting surfaces of the LVAD have been shown to result in de novo complications such as acquired Von Willebrand disease, which has been shown to predispose to gastrointestinal bleeding [17, 18], which, as mentioned above, may then increase the risk of subsequent thromboembolic events. Mechanical issues related to the pump have a demonstrated association with pump thrombosis [19]. Furthermore, the mechanism of pump thrombosis has direct implications on the success of medical therapy and may guide the clinical approach to patients. Within the HeartMate II pump, rotors generate heat and pathologically, ‘white and red’ clot is often organized and present at the inflow bearing and stator or on the rotor itself. More organized clots are likely not to respond to medical therapy, whether thrombolytics or antiplatelet therapy. The HeartWare HVAD has undergone some design modifications aimed at lowering pump thrombosis [12]. Originally, tissue ingrowth around the inflow cannula was observed, which could predispose to pump thrombus. After these observations, the inflow cannula was sintered in order to prevent tissue ingrowth, and the coring tool was increased in diameter and provided a more homogeneous surface into which the inflow cannula is inserted. Chronic tissue ingrowth around the inflow cannula is a chronic condition that is less likely to respond to medical therapy. In both pumps, surgical implantation is currently not standardized, and multiple techniques are utilized, including approach (sternotomy vs. left thoracotomy), use of cardiopulmonary bypass (on vs. off pump), position of outflow graft, pump pocket depth, and use versus non-use of the bend relief. The appreciation of the relationship
between the position and angulation of the inflow cannula and its relationship to subsequent pump thrombosis is growing significantly more recently [20]. A significantly greater acute angulation of the inflow cannula immediately post-implant and after 30 days was found in patients who had pump thrombosis. Furthermore, a smaller pump pocket depth was observed in patients with pump thrombosis. While existing reports portray some conflicting observations regarding potential causative factors for pump thrombosis, what is clear is that risk factors that place a patient at higher proclivity for this problem span across the entire landscape of the patient and the pump.
2 Management of Pump Thrombosis 2.1 Available Algorithms and Guidance No society guidelines or consensus statements exist regarding the diagnosis or management of LVAD pump thrombosis, despite the potential morbidity and mortality conferred by this clinical entity. In 2013, a working group examining the issue of LVAD pump thrombosis proposed an algorithm for the diagnosis and management of pump thrombosis [21]. Patients may present in a variety of conditions, including (1) pump power elevations, (2) isolated rise in LDH, (3) evidence of hemolysis, or (4) new heart failure symptoms. It is important to note that intermittent power elevations may often occur early postoperatively and are of unknown clinical significance. Similarly, an isolated rise in LDH may occur early in the postoperative setting as a consequence of blood transfusions or evolving resolution of elevated liver function tests from congestive hepatopathy. At this point, an echocardiogram with a ramp study can be considered, and anticoagulation should be optimized. While this algorithm was mainly created to address pump thrombosis of the HeartMate II, the clinical approach to the patient and analysis of the pump can also be extrapolated to the HeartWare HVAD with some particular additions. Specifically, the HeartWare HVAD enables the clinician to review controller log files with the ability to evaluate history of signals of power and flow [12]. Thrombotic events can present in several different characteristic manners that have a classic appearance of log file tracings. First, an acute partial or total occlusion demonstrates a sudden drop in power and flow with a new, lower set point. Second, a gradual thrombus formation demonstrates a slow increase in both power and flow in parallel. Third, resolution of a thrombus with tissue plasminogen activator (tPA) has a preceding classic spike in power and flow characteristic of thrombus formation prior to treatment, with return to normal values after treatment. This feature allows a detailed analysis of pump function
J. M. Stulak et al.
over time and has facilitated early detection with an opportunity for earlier treatment. In contrast, an isolated LDH elevation or power elevations that occur late after LVAD implantation should raise clinical concern, and an evaluation for hemolysis should immediately be undertaken. Cowger et al. [22] demonstrated that serum hemolysis marker elevations are associated with increased events in LVAD patients. Furthermore, patients who met their definition of hemolysis according to LDH definition had longer times from hemolysis to the onset of AEs and larger risk for embolism and need for device exchange than hemolysis defined by serum free hemoglobin. Certainly, if pump thrombosis is suspected early in its course, this may allow an opportunity to apply thrombolytics earlier when its success may have more efficacy. If markers of hemolysis are not evident, an echocardiogram with a ramp study to evaluate left ventricular (LV) unloading is suggested. If this is present, other investigations to rule out other causes of heart failure should be undertaken, and a second antiplatelet agent can be considered. However, if markers of hemolysis are present, the patient requires hospital admission and administration of intravenous heparin and fluid resuscitation. An echocardiogram with a ramp study should be performed to evaluate LV unloading. Once again, if LV unloading is present, then further evaluations for heart failure causes are pursued. If LV unloading is not observed with a ramp study, then a computed tomography (CT) scan can be performed to evaluate the inflow cannula and outflow graft and determine whether any mechanical impairment to contrast flow is present. If present, then consideration of surgical intervention is warranted. If none are found, then the findings of hemolysis and power elevations should be considered to be the result of pump thrombosis. After admission to the intensive care unit and the institution of aggressive heparin therapy, anti-thrombotic therapy with consideration of direct thrombin inhibitors or thrombolytics can be considered depending on the condition of the patient. If resolution occurs, then continue optimization and consider more intense anticoagulation and antiplatelet therapy. If resolution of hemolysis or pump malfunction is not obtained, then device exchange should be considered. If the patient is stable and could receive a donor heart within a short period of time, urgent transplantation can be attempted. While some patient presentations can represent a clinical dilemma, a subset of patients present with clear signs and symptoms of pump thrombosis, including red alarms and recalcitrant heart failure, with or without pump stoppage. These patients are suggested to undergo immediate and emergent device exchange without further diagnostic evaluations.
2.2 Medical Versus Surgical Therapy for Pump Thrombosis: An Outcome Analysis The optimal treatment and primary approach for the patient with pump thrombosis has not been fully established. In hemodynamically stable patients, medical therapy is often employed initially, which most commonly includes highdose unfractionated heparin and fluid resuscitation with or without the addition of glycoprotein IIb/IIIA inhibitors and thrombolytics. A summary of series documenting outcomes of this approach is presented in Table 2 [8, 10, 23– 27]. While small series demonstrate a high success rate with low morbidity, an analysis of larger series demonstrate success rates ranging from 23 to 50 %, stroke from 10 to 15 %, bleeding complications of 65 %, and mortality rates of 17–52 % [8, 10, 23, 25, 27]. It is very clear that, even in the face of ‘successful medical therapy’, there is an accompanied high morbidity rate. Also included in these series reporting outcomes with medical therapy for LVAD pump thrombosis, a significant number of patients who failed medical therapy progressed to surgical therapy. Thus, in a subset of stable patients, a role for attempted medical therapy may not preclude further consideration of surgical therapy. A summary of series documenting outcomes after LVAD device exchange is presented in Table 3 [8, 28–30]. While initially considered a highly morbid procedure, improvement of peri-procedural care, surgical approach, and postoperative support in the current era has resulted in very low early mortality and low complication rates with LVAD device exchange. In the majority of cases, surgical therapy is undertaken as the primary approach in patients who present with pump malfunction or stoppage with associated hemodynamic instability. Alternate approaches are now more commonly employed that have lowered morbidity and mortality following device exchange [28, 31]. The choice of initial therapy for the patient with LVAD pump thrombosis depends on several factors, including patient presentation, surgical candidacy, and, to a large degree, institutional philosophy. There is growing appreciation of the high morbidity associated with thrombolytics when balanced with a modest success rate, especially if the onset of thrombosis is greater than 24–48 h. In this subset of patients who have a delayed presentation, the utility of thrombolytics decreases significantly. In these patients, surgical device exchange is arguably the more effective approach. However, the ideal approach remains elusive and will always depend heavily on a combination of patientand device-related factors and the weighing of risks and benefits of each approach.
Management of Pump Thrombosis in LVAD Recipients Table 2 Series reporting outcomes of medical therapya for pump thrombosis References
Pt (n)
Najjar et al. [12]
30
50
NA
17
8
38
Stroke 13
38
40
NA
NA
52 at 6 mo
2
100
GI bleed 100
0
17
23
Stroke 12, bleeding 65
41
Schlendorf et al. [23] Starling et al. [8] Al-Quthami et al. [24] Tellor et al. [25]
Success (%)
Morbidity (%)
Mortality (%)
Kamouth et al. [26]
1
100
None
0
Lenneman et al. [27]
24
38
NA
29
a
The following therapy was used for the treatment of pump thrombosis: Najjar et al. [12]: tPA, GP IIb/IIIa inhibitor, or heparin; Schlendorf et al. [23]: intraventricular thrombolytic therapy; Starling et al. [8]: intravenous anticoagulation therapy, thrombolytic agents, and antiplatelet therapy; Al-Quthami et al. [24]: GP IIb/IIIa inhibitor; Tellor et al. [25]: GP IIb/IIIa inhibitor; Kammouth et al. [26]: intraventricular tPA; Lenneman et al. [27]: GP IIa/IIIb with or without tPA GP glycoprotein, GI gastrointestinal, mo month, NA not applicable, pt patient, tPA tissue plasminogen activator Table 3 Series reporting outcomes after left ventricular assist device exchange References
Overall pts (n)
Thrombosis pts (n)
Success (%)
Early mortality (%)
Ota et al. [28]
30
19
100
0
Moazami et al. [29]
72
25
100
6.5
Starling et al. [8]
19
19
100
5.3
Stulak et al. [30]
45
15
100
3.5
pt(s) patient(s)
3 Summary and Commentary While the issue of pump thrombosis has been ever present whether considering eras, devices, or in the context of a clinical trial or in the post-approval setting, the cause has remained elusive, most likely because there is not one unifying culprit. The issue of biocompatibility will always be at the forefront of critical issues considered with the development of new pumps, but the field of mechanical circulatory support must focus on adopting a standardized approach to this therapy so that valid analyses and comparisons can be made. A structured set of principles of surveillance and therapy for pump thrombosis has been proposed. In addition, more collaborative efforts need to be embraced in order to evaluate whether observed findings are unique to that single reporting institution or if the finding is common to other practices. Recent publications regarding this ‘newly arisen’ problem of pump thrombosis highlight the power of this approach [8, 32]. Not only collaborative efforts amongst institutions, but also partnerships between institutions and industry and focused clinical trials are of paramount importance. The PREVENT (PREVENtion of HeartMate II Pump Thrombosis through Clinical Management) is an example of one such endeavor in which the incidence of HeartMate II pump thrombosis will be assessed when predefined practices for clinical management are adopted.
Numerous limitations exist in the literature, specifically small series from single institutions, aside from clinical trial data. The field is still learning how to appreciate the differences in etiology and mechanism, and subsequent implications on the most appropriate clinical approach for each device, especially according to the nature of each patient’s presentation. In order to address these deficiencies, more collaborative efforts are needed and are critical in order to compare clinical approaches and subsequent efficacy. Mehra et al. [33] eloquently capture the essence of the problem of pump thrombosis when they remind the field of the following: ‘‘The law of unintended consequences posits that a simple intervention within a complex system always creates unanticipated and often undesirable outcomes’’. Acknowledgments Simon Maltais, John Stulak, and Shashank Sharma have no financial relationships to disclose. All authors participated in major portions of this manuscript. There was no funding source for this work.
References 1. Kirklin JK, Naftel DC, Warner-Stevenson L, et al. INTERMACS database for durable devices for circulatory support. First annual report. J Heart Lung Transplant. 2008;27(10):1065–72. 2. Kirklin JK, Naftel DC, Kormos RL, et al. Second INTERMACS annual report: more than 1,000 primary left ventricular assist device implants. J Heart Lung Transplant. 2010;29(1):1–10.
J. M. Stulak et al. 3. Kirklin JK, Naftel DC, Kormos RL, et al. Third INTERMACS annual report: the evolution of destination therapy in the United States. J Heart Lung Transplant. 2011;30(2):115–23. 4. Kirklin JK, Naftel DC, Kormos RL, et al. The Fourth INTERMACS annual report: 4,000 implants and counting. J Heart Lung Transplant. 2012;31(2):117–26. 5. Kirklin JK, Naftel DC, Kormos RL, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant. 2013;32(2):141–56. 6. Kirklin JK, Naftel DC, Pagani FD, et al. Sixth INTERMACS annual report: a 10,000 patient database. J Heart Lung Transplant. 2014;33(6):555–64. 7. Park SJ, Milano CA, Tatooles AJ, et al. Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy. Circ Heart Fail. 2012;5:241–8. 8. Starling RC, Moazami N, Silvestry SC, et al. Unexpected abrupt increase in left ventricular assist device thrombosis. N Engl J Med. 2014;370:33–40. 9. Kirklin JK, Naftel DC, Kormos RL, et al. Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) analysis of pump thrombosis in the HeartMate II left ventricular assist device. J Heart Lung Transplant. 2014;33:12–22. 10. Slaughter MS, Naka Y, John R, et al. Post-operative heparin may not be required for transitioning patients with a HeartMate II left ventricular assist system to long-term warfarin therapy. J Heart Lung Transplant. 2010;29:616–24. 11. Boyle AJ, Russell SD, Teuteberg JJ, et al. Low thromboembolism and pump thrombosis with the HeartMate II left ventricular assist device: analysis of outpatient anti-coagulation. J Heart Lung Transplant. 2009;28:881–7. 12. Najjar SS, Slaughter MS, Pagani FD, et al. An analysis of pump thrombosis events in patients in the HeartWare ADVANCE bridge to transplant and continued access protocol. J Heart Lung Transplant. 2014;33:23–34. 13. Slaughter MS, Pagani FD, McGee EC, et al. HeartWare ventricular assist system for bridge to transplant: combined results of the bridge to transplant and continued access protocol trial. J Heart Lung Transplant. 2013;32:675–83. 14. Stulak JM, Lee D, Haft JW, et al. Gastrointestinal bleeding and subsequent risk of thromboembolic events during support with a left ventricular assist device. J Heart Lung Transplant. 2014;33:60–4. 15. Wever-Pinzon O, Selzman CH, Drakos SG, et al. Pulsatility and the risk of nonsurgical bleeding in patients supported with the continuous-flow left ventricular assist device HeartMate II. Circ Heart Fail. 2013;6:517–26. 16. Saceed O, Maybaum S, D’Alessandro D, Goldstein DJ, Patel SR. Aortic valve opening and thrombotic events with continuous-flow left ventricular assist devices. J Heart Lung Transplant. 2014;33:109–11. 17. Meyer AL, Malesha D, Bara C, et al. Acquired von Willebrand syndrome in patients with an axial flow left ventricular assist device. Circulation. 2010;3:675–81.
18. Morgan JA, Paone G, Nemeh HW, et al. Gastrointestinal bleeding with the HeartMate II left ventricular assist device. J Heart Lung Transplant. 2012;31:715–8. 19. Uriel N, Han J, Morrison KA, et al. Device thrombosis in HeartMate II continuous-flow left ventricular assist devices: a multifactorial phenomenon. J Heart Lung Transplant. 2014;33:51–9. 20. Taghavi S, Ward C, Jayarajan SN, Gaughan J, Wilson LM, Mangi AA. Surgical technique influences HeartMate II left ventricular assist device thrombosis. Ann Thorac Surg. 2013;32:S38–9. 21. Goldstein DJ, John R, Salerno C, et al. Algorithm for the diagnosis and management of suspected pump thrombus. J Heart Lung Transplant. 2013;32:667–70. 22. Cowger JA, Romano MA, Shah P, et al. Hemolysis: a harbinger of adverse outcome after left ventricular assist device implant. J Heart Lung Transplant. 2014;2014(33):35–43. 23. Schlendorf K, Patel CB, Gehrig T, et al. Thrombolytic therapy for thrombosis of continuous flow ventricular assists devices. J Card Fail. 2014;20(2):91–7. 24. Al-Quthami M, Jumean R, Kociol R, et al. Eptifibatide for the treatment of HeartMate II left ventricular assist device thrombosis. Circ Heart Fail. 2012;5:e68–70. 25. Tellor JR, Smith SM, Prasad SM, et al. The use of eptifibatide for suspected pump thrombus or thrombosis in patients with left ventricular assist devices. J Heart Lung Transplant. 2014;33:94–101. 26. Kamouth A, John R, Eckman P. Successful treatment of early thrombosis of HeartWare left ventricular assist device with intraventricular thrombolytics. Ann Thorac Surg. 2012;94(1):281–3. 27. Lenneman P, Combs S, Rhode K, et al. Management and outcomes of ventricular assist device patients with suspected pump thrombosis. J Heart Lung Transplant. 2013;32(4):S186–7. 28. Ota H, Yerebakan H, Akashi H, et al. Continuous-flow left ventricular assist device exchange: clinical outcomes. J Heart Lung Transplant. 2014;33:65–70. 29. Moazami N, Milano CA, John R, et al. Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality. Ann Thorac Surg. 2013;95:500–5. 30. Stulak JM, Cowger J, Haft HW, Romano MA, Aaronson KD, Pagani FD. Device exchange after primary left ventricular assist device implantation. Ann Thorac Surg. 2013;95(4):1262–7. 31. Rojas SV, Avsar M, Khalpey Z, Hanke JS, Haverich A, Schmitto JD. Minimally-invasive off-pump left ventricular assist device exchange: anterolateral thoracotomy. Artif Organs. 2014;38(7):539–42. 32. Stulak JM, Maltais S. A different perspective on thrombosis and the HeartMate II. N Engl J Med. 2014;370:1467–8. 33. Mehra MR, Stewart GC, Uber PA. The vexing problem of thrombosis in long-term mechanical circulatory support. J Heart Lung Transplant. 2014;33:1–11.
LEADING ARTICLE
Management of Pump Thrombosis in Patients with Left Ventricular Assist Devices John M. Stulak • Shashank Sharma Simon Maltais
•
Ó Springer International Publishing Switzerland 2015
Abstract The gradual evolution of left ventricular assist device (LVAD) therapy has resulted in a durable option for patients as either a bridge to transplantation (BTT) or a destination therapy (DT). Outcomes with current continuous-flow devices continue to demonstrate significant patient benefit, not only in enhanced survival but also in improved functional capacity and quality of life. While the lessening of adverse events through time has resulted in more widespread adoption of this therapy, there continues to be unintended consequences, including, most notably, infection, bleeding, and thrombosis. Beginning in 2011, centers and collaborative groups began to observe a significant increase in the incidence of pump thrombosis with the HeartMate II LVAD (Thoratec Corp., Pleasanton, CA, USA). However, this clinical scourge is not limited to the HeartMate II, as the HeartWare Ventricular Assist System (HVAD; HeartWare Inc., Framingham, MA, USA) has also had these same issues, which led to pump modifications and the appreciation of more strict control of blood pressure and anticoagulation with this pump design. We review the current status of the field of mechanical circulatory support in its approach to diagnosis, management, and prevention of LVAD pump thrombosis.
J. M. Stulak (&) S. Sharma Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905, USA e-mail: [email protected] S. Maltais Vanderbilt Heart and Vascular Institute, Nashville, TN, USA
1 Introduction to the Problem of Pump Thrombosis 1.1 Defining the Problem The definition of pump thrombosis has evolved as our understanding of this entity has grown. The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) is a collaborative database sponsored by the National Heart, Lung, and Blood Institute (NHLBI). Data are collected on all durable mechanical circulatory support devices approved by the US FDA and implanted in the USA, and annual reports are published [1–6]. In addition, the INTERMACS Executive Committee publishes adverse event (AE) definitions that centers utilize to ensure standardization in the reporting of AEs. The most recent version of AE definitions was approved on 15 May 2013. Relevant to the problem discussed herein are the definitions of hemolysis and pump thrombus. Hemolysis in a patient supported on left ventricular assist device (LVAD) therapy is defined as either minor or major depending on certain biochemical and clinical criteria. Minor hemolysis is defined as a plasma-free hemoglobin [20 mg/dl or a serum lactate dehydrogenase (LDH) [2.5 times the upper limits of the normal range at the implanting center. These findings should be observed after the first 72 h of implant and are in the absence of clinical symptoms, manifestations of hemolysis, or abnormal pump function. Major hemolysis includes the criteria for ‘minor’ hemolysis and associated with hemoglobinuria, anemia, hyperbilirubinemia, or pump malfunction and/or abnormal pump parameters. Pump thrombus is considered a specific case of a major device malfunction and is classified as either suspected pump thrombus or confirmed pump thrombus. With suspected pump thrombus, clinical patient condition or pump
J. M. Stulak et al.
parameters suggest thrombus on any of the blood-contacting surfaces of the pump (inflow cannula, pump itself, or outflow graft). Criteria (at least two present) should include presence of hemolysis, presence of heart failure not explained by structural disease, or abnormal pump parameters. Confirmed pump thrombus can be observed by visual inspection (at the time of device exchange, transplantation, or autopsy), irrefutable radiographic evidence, or absence of Doppler inflow or outflow signals. 1.2 The Historical and Contemporary Magnitude of the Problem Both in the clinical trial setting and in the post-approval era, numerous series have documented an unavoidable rate of pump thrombosis for both the HeartMate II (Thoratec Corp., Pleasanton, CA, USA) and the HeartWare Ventricular Assist System (HVAD; HeartWare Inc., Framingham, MA, USA) (Table 1). Concern regarding LVAD pump thrombosis has been heightened recently due to two publications: (1) an initial report by investigators who documented an abrupt increase in the incidence of confirmed pump thrombosis at 3 months with the HeartMate II from 2.2 % before March 2011 to 8.4 % after January 2013, and (2) a detailed analysis of the INTERMACS database that reported a sixfold increase in rates of pump thrombosis in the timeframe of 2011 to 2012 [8, 9]. Furthermore, the 6-month actuarial freedom from device exchange or death due to pump thrombosis has significantly fallen from 99 % in the 2008–2009 time period to 95 % in 2011 and 94 % in 2012 [9]. From a medical management standpoint, it is interesting to note a temporal relationship of these findings to reports documenting the safe adoption of lower anticoagulation levels in patients with the HeartMate II, including the safe non-use of perioperative intravenous heparin [10, 11].
Unfortunately, despite these data trends, firm conclusions are unable to be drawn as to causality, whether patient or device specific. Interesting findings were also observed by the HeartWare investigators after concern around the high incidence (0.07 events per patient year) of HeartWare HVAD pump thrombosis in their clinical trial prompted a detailed analysis [12, 13]. Interestingly, they found that the majority of pump thromboses occurred in patients with sub-therapeutic warfarin anticoagulation and taking lowdose or no antiplatelet therapy. After recommendations were made for strict patient management to adhere to an international normalized ratio (INR) of 2–3 and high-dose aspirin (325 mg), the rate of device exchange due to pump thrombosis fell by more than 50 % [12, 13]. 1.3 Potential Causes of the Problem Analysis of the available reports to date stresses the complexity of the phenomenon of LVAD pump thrombosis, and the causes can be differentiated into non-mechanical and mechanical. In relation to the former, while the trend of lowering anticoagulation that was enacted in order to lessen bleeding complications and its temporal relationship to new concerns regarding pump thrombosis is intriguing, and likely contributory to some degree, one must concede that the presence of an LVAD represents a very complex clinical arrangement. It follows then that this problem, which arises within such a complex system, is also complex in its etiology. In parallel with the relationship of maintenance anticoagulation and antiplatelet therapy, some have demonstrated a subsequent increased risk of thromboembolic events following gastrointestinal bleeding and resultant alteration in anticoagulation [14]. Aside from anticoagulation, one must recall that these pumps are afterload sensitive, with the third-generation pumps, such as HeartWare HVAD, to a greater degree. Flow through the
Table 1 Multicenter reports of rates of pump thrombosisa References
Year
Pt (n)
Incidence
Freedom from event (%) 6 months
1 year
92
Park et al. [7]
2005–2007
133
Park et al. [7]
2007–2009
281
0.038/py
Starling et al [8]
2004–2013
837
7.9 % confirmed
95
Kirklin et al. [9]
2008–2012
6,910
5.5 % (382/6,910)
97
95
Najjar et al. [12]
2008–2012
382
8.1 %, 0.08/py
96
92
Stulak et al. [32]
2004–2012
516
3 mo: 1.7 %, 1 y: 4.7 %
a
0.024/py
Pump thrombosis was defined as follows: Park et al. [7], Starling et al. [8]: Confirmed: thrombus noted on blood-contacting surfaces during urgent transplant, device exchange, or autopsy; Suspected: pump malfunction with clinical variables suggested thrombus on pump components; Kirklin et al. [9]: pump exchange for documented or probable thrombus, death with device malfunction with documented or probable thrombus; Najjar et al. [12]: alterations in pump parameters with increased markers of hemolysis and/or thrombus visualization after pump exchange; Stulak et al. [32]: Confirmed: thrombus noted on blood-contacting surfaces during urgent transplant, device exchange, or autopsy mo month, pt patient, py patient-year, y year
Management of Pump Thrombosis in LVAD Recipients
pump itself is lower in the presence of a high afterload and some have observed an association between high blood pressure and pump thrombosis [12]. Certainly, this issue requires more detailed analysis. One further non-mechanical issue that has undergone some philosophical changes is the status of the aortic valve opening following LVAD implantation. After concerns arose surrounding progression of aortic regurgitation following LVAD implant, alterations in pump management toward lower speeds in an attempt to foster aortic valve opening were suggested [15]. The benefit of this strategy was felt not only to reduce progression of aortic regurgitation but also to reduce development of gastrointestinal arteriovenous malformations and gastrointestinal bleeding. Similar to the aforementioned anticoagulation management, this strategy, aimed at decreasing complications, yielded unintended consequences, and, while permitting greater native heart function, it was found to decrease flow through the pump and increase the risk of subsequent pump thrombosis [16]. Lastly, and arguably most significantly, the unique interaction of the patient’s blood and the blood-contacting surfaces of the LVAD have been shown to result in de novo complications such as acquired Von Willebrand disease, which has been shown to predispose to gastrointestinal bleeding [17, 18], which, as mentioned above, may then increase the risk of subsequent thromboembolic events. Mechanical issues related to the pump have a demonstrated association with pump thrombosis [19]. Furthermore, the mechanism of pump thrombosis has direct implications on the success of medical therapy and may guide the clinical approach to patients. Within the HeartMate II pump, rotors generate heat and pathologically, ‘white and red’ clot is often organized and present at the inflow bearing and stator or on the rotor itself. More organized clots are likely not to respond to medical therapy, whether thrombolytics or antiplatelet therapy. The HeartWare HVAD has undergone some design modifications aimed at lowering pump thrombosis [12]. Originally, tissue ingrowth around the inflow cannula was observed, which could predispose to pump thrombus. After these observations, the inflow cannula was sintered in order to prevent tissue ingrowth, and the coring tool was increased in diameter and provided a more homogeneous surface into which the inflow cannula is inserted. Chronic tissue ingrowth around the inflow cannula is a chronic condition that is less likely to respond to medical therapy. In both pumps, surgical implantation is currently not standardized, and multiple techniques are utilized, including approach (sternotomy vs. left thoracotomy), use of cardiopulmonary bypass (on vs. off pump), position of outflow graft, pump pocket depth, and use versus non-use of the bend relief. The appreciation of the relationship
between the position and angulation of the inflow cannula and its relationship to subsequent pump thrombosis is growing significantly more recently [20]. A significantly greater acute angulation of the inflow cannula immediately post-implant and after 30 days was found in patients who had pump thrombosis. Furthermore, a smaller pump pocket depth was observed in patients with pump thrombosis. While existing reports portray some conflicting observations regarding potential causative factors for pump thrombosis, what is clear is that risk factors that place a patient at higher proclivity for this problem span across the entire landscape of the patient and the pump.
2 Management of Pump Thrombosis 2.1 Available Algorithms and Guidance No society guidelines or consensus statements exist regarding the diagnosis or management of LVAD pump thrombosis, despite the potential morbidity and mortality conferred by this clinical entity. In 2013, a working group examining the issue of LVAD pump thrombosis proposed an algorithm for the diagnosis and management of pump thrombosis [21]. Patients may present in a variety of conditions, including (1) pump power elevations, (2) isolated rise in LDH, (3) evidence of hemolysis, or (4) new heart failure symptoms. It is important to note that intermittent power elevations may often occur early postoperatively and are of unknown clinical significance. Similarly, an isolated rise in LDH may occur early in the postoperative setting as a consequence of blood transfusions or evolving resolution of elevated liver function tests from congestive hepatopathy. At this point, an echocardiogram with a ramp study can be considered, and anticoagulation should be optimized. While this algorithm was mainly created to address pump thrombosis of the HeartMate II, the clinical approach to the patient and analysis of the pump can also be extrapolated to the HeartWare HVAD with some particular additions. Specifically, the HeartWare HVAD enables the clinician to review controller log files with the ability to evaluate history of signals of power and flow [12]. Thrombotic events can present in several different characteristic manners that have a classic appearance of log file tracings. First, an acute partial or total occlusion demonstrates a sudden drop in power and flow with a new, lower set point. Second, a gradual thrombus formation demonstrates a slow increase in both power and flow in parallel. Third, resolution of a thrombus with tissue plasminogen activator (tPA) has a preceding classic spike in power and flow characteristic of thrombus formation prior to treatment, with return to normal values after treatment. This feature allows a detailed analysis of pump function
J. M. Stulak et al.
over time and has facilitated early detection with an opportunity for earlier treatment. In contrast, an isolated LDH elevation or power elevations that occur late after LVAD implantation should raise clinical concern, and an evaluation for hemolysis should immediately be undertaken. Cowger et al. [22] demonstrated that serum hemolysis marker elevations are associated with increased events in LVAD patients. Furthermore, patients who met their definition of hemolysis according to LDH definition had longer times from hemolysis to the onset of AEs and larger risk for embolism and need for device exchange than hemolysis defined by serum free hemoglobin. Certainly, if pump thrombosis is suspected early in its course, this may allow an opportunity to apply thrombolytics earlier when its success may have more efficacy. If markers of hemolysis are not evident, an echocardiogram with a ramp study to evaluate left ventricular (LV) unloading is suggested. If this is present, other investigations to rule out other causes of heart failure should be undertaken, and a second antiplatelet agent can be considered. However, if markers of hemolysis are present, the patient requires hospital admission and administration of intravenous heparin and fluid resuscitation. An echocardiogram with a ramp study should be performed to evaluate LV unloading. Once again, if LV unloading is present, then further evaluations for heart failure causes are pursued. If LV unloading is not observed with a ramp study, then a computed tomography (CT) scan can be performed to evaluate the inflow cannula and outflow graft and determine whether any mechanical impairment to contrast flow is present. If present, then consideration of surgical intervention is warranted. If none are found, then the findings of hemolysis and power elevations should be considered to be the result of pump thrombosis. After admission to the intensive care unit and the institution of aggressive heparin therapy, anti-thrombotic therapy with consideration of direct thrombin inhibitors or thrombolytics can be considered depending on the condition of the patient. If resolution occurs, then continue optimization and consider more intense anticoagulation and antiplatelet therapy. If resolution of hemolysis or pump malfunction is not obtained, then device exchange should be considered. If the patient is stable and could receive a donor heart within a short period of time, urgent transplantation can be attempted. While some patient presentations can represent a clinical dilemma, a subset of patients present with clear signs and symptoms of pump thrombosis, including red alarms and recalcitrant heart failure, with or without pump stoppage. These patients are suggested to undergo immediate and emergent device exchange without further diagnostic evaluations.
2.2 Medical Versus Surgical Therapy for Pump Thrombosis: An Outcome Analysis The optimal treatment and primary approach for the patient with pump thrombosis has not been fully established. In hemodynamically stable patients, medical therapy is often employed initially, which most commonly includes highdose unfractionated heparin and fluid resuscitation with or without the addition of glycoprotein IIb/IIIA inhibitors and thrombolytics. A summary of series documenting outcomes of this approach is presented in Table 2 [8, 10, 23– 27]. While small series demonstrate a high success rate with low morbidity, an analysis of larger series demonstrate success rates ranging from 23 to 50 %, stroke from 10 to 15 %, bleeding complications of 65 %, and mortality rates of 17–52 % [8, 10, 23, 25, 27]. It is very clear that, even in the face of ‘successful medical therapy’, there is an accompanied high morbidity rate. Also included in these series reporting outcomes with medical therapy for LVAD pump thrombosis, a significant number of patients who failed medical therapy progressed to surgical therapy. Thus, in a subset of stable patients, a role for attempted medical therapy may not preclude further consideration of surgical therapy. A summary of series documenting outcomes after LVAD device exchange is presented in Table 3 [8, 28–30]. While initially considered a highly morbid procedure, improvement of peri-procedural care, surgical approach, and postoperative support in the current era has resulted in very low early mortality and low complication rates with LVAD device exchange. In the majority of cases, surgical therapy is undertaken as the primary approach in patients who present with pump malfunction or stoppage with associated hemodynamic instability. Alternate approaches are now more commonly employed that have lowered morbidity and mortality following device exchange [28, 31]. The choice of initial therapy for the patient with LVAD pump thrombosis depends on several factors, including patient presentation, surgical candidacy, and, to a large degree, institutional philosophy. There is growing appreciation of the high morbidity associated with thrombolytics when balanced with a modest success rate, especially if the onset of thrombosis is greater than 24–48 h. In this subset of patients who have a delayed presentation, the utility of thrombolytics decreases significantly. In these patients, surgical device exchange is arguably the more effective approach. However, the ideal approach remains elusive and will always depend heavily on a combination of patientand device-related factors and the weighing of risks and benefits of each approach.
Management of Pump Thrombosis in LVAD Recipients Table 2 Series reporting outcomes of medical therapya for pump thrombosis References
Pt (n)
Najjar et al. [12]
30
50
NA
17
8
38
Stroke 13
38
40
NA
NA
52 at 6 mo
2
100
GI bleed 100
0
17
23
Stroke 12, bleeding 65
41
Schlendorf et al. [23] Starling et al. [8] Al-Quthami et al. [24] Tellor et al. [25]
Success (%)
Morbidity (%)
Mortality (%)
Kamouth et al. [26]
1
100
None
0
Lenneman et al. [27]
24
38
NA
29
a
The following therapy was used for the treatment of pump thrombosis: Najjar et al. [12]: tPA, GP IIb/IIIa inhibitor, or heparin; Schlendorf et al. [23]: intraventricular thrombolytic therapy; Starling et al. [8]: intravenous anticoagulation therapy, thrombolytic agents, and antiplatelet therapy; Al-Quthami et al. [24]: GP IIb/IIIa inhibitor; Tellor et al. [25]: GP IIb/IIIa inhibitor; Kammouth et al. [26]: intraventricular tPA; Lenneman et al. [27]: GP IIa/IIIb with or without tPA GP glycoprotein, GI gastrointestinal, mo month, NA not applicable, pt patient, tPA tissue plasminogen activator Table 3 Series reporting outcomes after left ventricular assist device exchange References
Overall pts (n)
Thrombosis pts (n)
Success (%)
Early mortality (%)
Ota et al. [28]
30
19
100
0
Moazami et al. [29]
72
25
100
6.5
Starling et al. [8]
19
19
100
5.3
Stulak et al. [30]
45
15
100
3.5
pt(s) patient(s)
3 Summary and Commentary While the issue of pump thrombosis has been ever present whether considering eras, devices, or in the context of a clinical trial or in the post-approval setting, the cause has remained elusive, most likely because there is not one unifying culprit. The issue of biocompatibility will always be at the forefront of critical issues considered with the development of new pumps, but the field of mechanical circulatory support must focus on adopting a standardized approach to this therapy so that valid analyses and comparisons can be made. A structured set of principles of surveillance and therapy for pump thrombosis has been proposed. In addition, more collaborative efforts need to be embraced in order to evaluate whether observed findings are unique to that single reporting institution or if the finding is common to other practices. Recent publications regarding this ‘newly arisen’ problem of pump thrombosis highlight the power of this approach [8, 32]. Not only collaborative efforts amongst institutions, but also partnerships between institutions and industry and focused clinical trials are of paramount importance. The PREVENT (PREVENtion of HeartMate II Pump Thrombosis through Clinical Management) is an example of one such endeavor in which the incidence of HeartMate II pump thrombosis will be assessed when predefined practices for clinical management are adopted.
Numerous limitations exist in the literature, specifically small series from single institutions, aside from clinical trial data. The field is still learning how to appreciate the differences in etiology and mechanism, and subsequent implications on the most appropriate clinical approach for each device, especially according to the nature of each patient’s presentation. In order to address these deficiencies, more collaborative efforts are needed and are critical in order to compare clinical approaches and subsequent efficacy. Mehra et al. [33] eloquently capture the essence of the problem of pump thrombosis when they remind the field of the following: ‘‘The law of unintended consequences posits that a simple intervention within a complex system always creates unanticipated and often undesirable outcomes’’. Acknowledgments Simon Maltais, John Stulak, and Shashank Sharma have no financial relationships to disclose. All authors participated in major portions of this manuscript. There was no funding source for this work.
References 1. Kirklin JK, Naftel DC, Warner-Stevenson L, et al. INTERMACS database for durable devices for circulatory support. First annual report. J Heart Lung Transplant. 2008;27(10):1065–72. 2. Kirklin JK, Naftel DC, Kormos RL, et al. Second INTERMACS annual report: more than 1,000 primary left ventricular assist device implants. J Heart Lung Transplant. 2010;29(1):1–10.
J. M. Stulak et al. 3. Kirklin JK, Naftel DC, Kormos RL, et al. Third INTERMACS annual report: the evolution of destination therapy in the United States. J Heart Lung Transplant. 2011;30(2):115–23. 4. Kirklin JK, Naftel DC, Kormos RL, et al. The Fourth INTERMACS annual report: 4,000 implants and counting. J Heart Lung Transplant. 2012;31(2):117–26. 5. Kirklin JK, Naftel DC, Kormos RL, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant. 2013;32(2):141–56. 6. Kirklin JK, Naftel DC, Pagani FD, et al. Sixth INTERMACS annual report: a 10,000 patient database. J Heart Lung Transplant. 2014;33(6):555–64. 7. Park SJ, Milano CA, Tatooles AJ, et al. Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy. Circ Heart Fail. 2012;5:241–8. 8. Starling RC, Moazami N, Silvestry SC, et al. Unexpected abrupt increase in left ventricular assist device thrombosis. N Engl J Med. 2014;370:33–40. 9. Kirklin JK, Naftel DC, Kormos RL, et al. Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) analysis of pump thrombosis in the HeartMate II left ventricular assist device. J Heart Lung Transplant. 2014;33:12–22. 10. Slaughter MS, Naka Y, John R, et al. Post-operative heparin may not be required for transitioning patients with a HeartMate II left ventricular assist system to long-term warfarin therapy. J Heart Lung Transplant. 2010;29:616–24. 11. Boyle AJ, Russell SD, Teuteberg JJ, et al. Low thromboembolism and pump thrombosis with the HeartMate II left ventricular assist device: analysis of outpatient anti-coagulation. J Heart Lung Transplant. 2009;28:881–7. 12. Najjar SS, Slaughter MS, Pagani FD, et al. An analysis of pump thrombosis events in patients in the HeartWare ADVANCE bridge to transplant and continued access protocol. J Heart Lung Transplant. 2014;33:23–34. 13. Slaughter MS, Pagani FD, McGee EC, et al. HeartWare ventricular assist system for bridge to transplant: combined results of the bridge to transplant and continued access protocol trial. J Heart Lung Transplant. 2013;32:675–83. 14. Stulak JM, Lee D, Haft JW, et al. Gastrointestinal bleeding and subsequent risk of thromboembolic events during support with a left ventricular assist device. J Heart Lung Transplant. 2014;33:60–4. 15. Wever-Pinzon O, Selzman CH, Drakos SG, et al. Pulsatility and the risk of nonsurgical bleeding in patients supported with the continuous-flow left ventricular assist device HeartMate II. Circ Heart Fail. 2013;6:517–26. 16. Saceed O, Maybaum S, D’Alessandro D, Goldstein DJ, Patel SR. Aortic valve opening and thrombotic events with continuous-flow left ventricular assist devices. J Heart Lung Transplant. 2014;33:109–11. 17. Meyer AL, Malesha D, Bara C, et al. Acquired von Willebrand syndrome in patients with an axial flow left ventricular assist device. Circulation. 2010;3:675–81.
18. Morgan JA, Paone G, Nemeh HW, et al. Gastrointestinal bleeding with the HeartMate II left ventricular assist device. J Heart Lung Transplant. 2012;31:715–8. 19. Uriel N, Han J, Morrison KA, et al. Device thrombosis in HeartMate II continuous-flow left ventricular assist devices: a multifactorial phenomenon. J Heart Lung Transplant. 2014;33:51–9. 20. Taghavi S, Ward C, Jayarajan SN, Gaughan J, Wilson LM, Mangi AA. Surgical technique influences HeartMate II left ventricular assist device thrombosis. Ann Thorac Surg. 2013;32:S38–9. 21. Goldstein DJ, John R, Salerno C, et al. Algorithm for the diagnosis and management of suspected pump thrombus. J Heart Lung Transplant. 2013;32:667–70. 22. Cowger JA, Romano MA, Shah P, et al. Hemolysis: a harbinger of adverse outcome after left ventricular assist device implant. J Heart Lung Transplant. 2014;2014(33):35–43. 23. Schlendorf K, Patel CB, Gehrig T, et al. Thrombolytic therapy for thrombosis of continuous flow ventricular assists devices. J Card Fail. 2014;20(2):91–7. 24. Al-Quthami M, Jumean R, Kociol R, et al. Eptifibatide for the treatment of HeartMate II left ventricular assist device thrombosis. Circ Heart Fail. 2012;5:e68–70. 25. Tellor JR, Smith SM, Prasad SM, et al. The use of eptifibatide for suspected pump thrombus or thrombosis in patients with left ventricular assist devices. J Heart Lung Transplant. 2014;33:94–101. 26. Kamouth A, John R, Eckman P. Successful treatment of early thrombosis of HeartWare left ventricular assist device with intraventricular thrombolytics. Ann Thorac Surg. 2012;94(1):281–3. 27. Lenneman P, Combs S, Rhode K, et al. Management and outcomes of ventricular assist device patients with suspected pump thrombosis. J Heart Lung Transplant. 2013;32(4):S186–7. 28. Ota H, Yerebakan H, Akashi H, et al. Continuous-flow left ventricular assist device exchange: clinical outcomes. J Heart Lung Transplant. 2014;33:65–70. 29. Moazami N, Milano CA, John R, et al. Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality. Ann Thorac Surg. 2013;95:500–5. 30. Stulak JM, Cowger J, Haft HW, Romano MA, Aaronson KD, Pagani FD. Device exchange after primary left ventricular assist device implantation. Ann Thorac Surg. 2013;95(4):1262–7. 31. Rojas SV, Avsar M, Khalpey Z, Hanke JS, Haverich A, Schmitto JD. Minimally-invasive off-pump left ventricular assist device exchange: anterolateral thoracotomy. Artif Organs. 2014;38(7):539–42. 32. Stulak JM, Maltais S. A different perspective on thrombosis and the HeartMate II. N Engl J Med. 2014;370:1467–8. 33. Mehra MR, Stewart GC, Uber PA. The vexing problem of thrombosis in long-term mechanical circulatory support. J Heart Lung Transplant. 2014;33:1–11.
