Secondary Logo

Journal Logo

Adult Circulatory Support

Systemic Thrombolysis Versus Device Exchange for Pump Thrombosis Management: A Single-Center Experience

Oezpeker, Cenk*; Zittermann, Armin*; Ensminger, Stephan*; Kizner, Lukas*; Koster, Andreas; Sayin, Ali*; Schoenbrodt, Michael*; Milting, Hendrik*; Gummert, Jan F.*; Morshuis, Michiel*

Author Information
doi: 10.1097/MAT.0000000000000340
  • Free


Left ventricular assist devices (LVADs) are the preferred form of mechanical circulatory support in patients with end-stage heart failure.1 The progressive refinement of LVAD technology implemented this therapy either as bridge to transplant or as an alternative to heart transplantation.2,3 Therefore, careful management of device-related complications is of crucial importance for the long-term success of LVAD support.

Pump thrombosis is a life-threatening complication in LVAD patients and is the cause of up to 10% of events.4,5 There are surgical therapeutic options such as LVAD exchange6 and nonsurgical options such as systemic thrombolysis (STL).7–9 However, STL is associated with an increased risk of intracranial and gastrointestinal bleeding.10 The risk of these complications is even higher in LVAD patients because of continuous antiaggregation and anticoagulation medication.4,7 Moreover, there are concerns about a high rate of unresolved thrombus in STL-treated LVAD patients with pump thrombosis4,7,11 and the need for massive blood transfusions in case of emergency device exchange (DEx).9 Therefore, STL is not a favored therapeutic option in LVAD patients with pump thrombosis,6,8 whereas DEx has been established as the therapy of choice for the management of almost all cases of pump thrombosis.6

However, recent data12 indicate that STL may be an appropriate procedure for selected patients and should therefore still have a place in the management of pump thrombosis. Therefore, the present data analysis aimed to compare clinical outcome in patients undergoing DEx and STL, respectively. In subgroup analysis in patients with STL, we also compared the success of STL in patients supported with HW (HeartWare Inc., Framingham, MA) and HeartMate II (HMII; Thoratec Corp., Peasanton, CA), respectively.



This study was a retrospective data analysis in LVAD patients undergoing STL or DEx at our institution between August 2009 and February 2015 (Figure 1). Inclusion criterion was pump thrombosis in patients with HW and HMII implants, respectively. Patients with implants of other devices, pediatric patients, and patients with acute pump stop were excluded. During the aforementioned study period, a total number of 473 HW and HMII LVADs (HMII, n = 155; HW, n = 318) were implanted at our center. Fifty-two patients developed pump thrombosis. In 2 of these 52 patients, device thrombosis had led to acute pump stop. In these cases, DEx was the only treatment option. Thus, 50 patients could finally be included in our statistical analysis. Written informed consent for scientific use of clinical data was obtained from all patients or their relatives (in case of unconscious patients). Because of the retrospective study design, the need for an ethics committee votum was waived.

Figure 1.
Figure 1.:
Flow chart of patients with left ventricular assist device pump thrombosis. LVAD, left ventricular assist device.

Study Design

For data analysis, patients were divided into two groups: Group 1 consisted of patients who underwent STL (designated the STL group; n = 29), and group 2 consisted of patients who underwent DEx (designated the DEx group; n = 21). The decision whether STL or DEx should be performed depended on the duration of pump thrombosis. In those patients, whose pump thrombosis was longer than 24 hours, DEx was performed. Systemic thrombolysis was performed if pump thrombosis was less than 24 hours. Clinical data were obtained from the database of our mechanical circulatory support program and from the hospital’s digital patient management system. We assessed characteristics of the patients such as age, gender, height, weight, body surface area, body mass index, Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) level, sort of device, diagnosis, and concomitant diagnoses, such as diabetes mellitus, arterial hypertension, and hyperlipidemia at the time of first device implantation. At the time of pump thrombosis, we assessed the Simplified Acute Physiology II (SAPS-II) score and biochemical parameters such as creatinine, bilirubin, alanine aminotransferase (ALT), international normalized ratio (INR), C-reactive protein, white blood cell counts, red blood cell counts, and platelets. Moreover, we assessed the necessity of blood products (units of packed red blood, fresh frozen plasma, and platelets) within the first 24 hours postintervention and duration of intensive care unit stay and in-hospital stay. In addition, we assessed early therapy failure (the need for additional surgical or nonsurgical intervention because of pump thrombosis in the STL group or DEx group within 90 days postintervention), and therapy failure and mortality up to 2 years after STL and DEx, respectively. Therapy failure was assessed by de novo increase in lactate dehydrogenase (LDH) and free hemoglobin concentrations.

Routine Anticoagulation

The standard anticoagulation regimen for LVAD patients in our center is phenprocomoun (INR target range: 2.3–2.8) and acetylsalicylic acid 100 mg daily. INR values are monitored daily. In our out-of-hospital patients, monitoring is accomplished by INR self-management. International normalized ratios results are send via a short daily message to our telemedicine department, where all the data are processed.13 This feedback enables our center to contact the respective patients when INR levels are out of range.

Detection and Diagnosis of Pump Thrombosis

Device thrombosis often manifests as low pump flow with an increased power consumption, hematuria, and elevated LDH (reference range: 100–600 U/L) and serum-free hemoglobin (reference range: 0–40 mg/dl) concentrations. Therefore, every patient with abnormal pump parameters or hematuria immediately contacts our 24 hour phone service and is then readmitted to our hospital for further clinical and laboratory investigations. Because computer-gated tomography and transthoracic echocardiography with or without contrast media are possible diagnostic tools,14–16 echocardiography and computer tomography were ordered at the discretion of the treating surgeon. However, only thrombus formation in parts of the outflow and inflow conduits can be visualized by these diagnostic tools. Therefore, patients with suspected pump malfunction underwent right heart catheterization to show the exact cardiac output, pulmonary capillary wedge pressure, central venous pressure, and vascular resistance. Fixed, high (>17 mm Hg) pulmonary capillary wedge pressures with simultaneously low cardiac index (<1.8 L/min/m2) despite increasing revolutions per minute of the LVAD are reliable signs of pump malfunction.

Therapy of Pump Thrombosis

The surgical strategy of DEx can be performed via median resternotomy or left subcostal thoracotomy. In our study cohort, lateral thoracotomy was performed in three patients only. The noninvasive strategy of STL was performed with either tenecteplase (100 U/10 kg body weight; n = 24) or alteplase (1 mg/kg body weight; n = 5). We performed bolus infusions over 1 hour. During infusion, we stopped heparin administration and started with it again with 600 IE/hour immediately after the infusion of tenecteplase/alteplase was finished.


The primary end-point was the probability of 90 day overall survival after the first-line therapy. Secondary end-points were the need for blood products, duration of intensive care unit, in-hospital stay, 90 day therapy failure, and 2 year therapy failure and mortality.

Statistical Analysis

Demographic, clinical, and preoperative patient characteristics, as well as postprocedural outcomes, were compared using the Mann–Whitney U test for continuous (nonparametric) variables and the χ2 test for categorical variables. Continuous data are presented as median and interquartile range (IQR). Survival and freedom from therapy failure were estimated by the Kaplan–Meier method and compared between study (sub) groups by using the log-rank test. An intention-to-treat approach was used to compare survival of the two study groups. p Values <0.05 were considered statistically significant. We applied the statistical software package SPSS, version 21 (IBM Corp, Armonk, NY) to perform the analyses.


Baseline Characteristics

In Figure 2, thrombus formation is presented in patients with HMII and HW implant, respectively. Baseline characteristics of the study groups are summarized in Table 1. Study groups differed according to sort of device, time since first LVAD implant, and the percentage of patients living at home at the onset of pump thrombosis. At the time of pump thrombosis, ALT and SAPS-II were significantly lower, and red blood cell counts were significantly higher in the STL group than in the DEx groups. Other biochemical parameters were comparable between groups. In the DEx and STL group, early postoperative pump thrombosis (within 2 weeks of LVAD implant) was 19.9% and 3%, respectively (p = 0.148). After pump thrombosis, the DEx group received either HW (n = 8) or HMII (n = 13) implants. The 29 patients in the STL group received in total 44 sessions of STL. The minimum period of time between the therapies was 48 hours. In the STL group, 15 patients did not show clinical signs of bleeding because of STL, whereas 11 patients developed minor local hematomas at the site of catheter administration. One patient developed a cerebral embolism with mild right-sided hemiplegia but showed full neurologic recovery within 1 month. Two other patients developed massive intracranial bleeding.

Table 1.
Table 1.:
Baseline Characteristics of the Study Groups
Figure 2.
Figure 2.:
Pump thrombus in HeartMate II (A) and HeartWare (B) device.

Primary End-Points

Ninety-day survival is illustrated in Figure 3, broken down by study group. Data indicate similar survival in both groups. The 90 day survival was 89.3% in the STL group and 91.0% in the DEx group. Six patients died in the STL group. Causes of death were massive intracerebral hemorrhage (n = 2) and multiorgan failure with sepsis (n = 4). All four patients who died in the DEx group were preoperatively on high dosages of inotropic support and had multiorgan failure before the intervention. In all patients, causes of death were multiorgan failure. Three patients also experienced sepsis. None of the patients received heart transplantation during 90 day follow-up, neither in the DEx group nor in the STL group.

Figure 3.
Figure 3.:
Kaplan–Meier estimates of survival up to 2 years after treatment of pump thrombosis. DEx, device exchange; STL, systemic thrombolysis.

Secondary End-Points

The average use of blood products was significantly higher, and intensive care unit stay tended to be higher in the DEx group than in the STL group (Table 2). In-hospital stay was also lower in the STL group than in the DEx group. Freedom from 90 day therapy failure is illustrated in Figure 4 and was 89.1% in the DEx group and 60.7% in the STL group (p = 0.027). In detail, therapy failure occurred in 2 patients with HW implants out of the 21 patients in the DEx group and in 11 out of the 29 patients in the STL group. In the STL group, therapy failure occurred in 4 of the 8 patients with HMII implants and in 7 of 21 patients with HW implants (p = 0.460). In STL patients with therapy failure, median duration until DEx was 6 days (IQR: 2–13 days; range: 1–34 days). In the STL patients with and without 90 day therapy failure, duration of intensive care unit stay was 1 day (IQR: 1–1 day) and 2 days (IQR: 1–5 days), respectively (p = 0.173). The corresponding values for in-hospital stay were 9.5 days (IQR: 8–20 days) and 20 days (16–52 days), respectively (p = 0.031). Survival up to 2 years was significantly lower in STL patients with therapy failure than in STL patients without therapy failure (45.5% vs. 94.4%, p = 0.001).

Table 2.
Table 2.:
Postintervention Outcomes in Patients With Systemic Thrombolysis or Device Exchange
Figure 4.
Figure 4.:
Kaplan–Meier estimates of freedom from therapy failure up to 2 years after treatment of pump thrombosis. DEx, device exchange; STL, systemic thrombolysis.

In the DEx and STL group, freedom from therapy failure up to 2 years was 55.2% and 18.5%, respectively (p = 0.006) (Figure 4). Two-year survival was 61.9% (DEx group) and 75.9% (STL group) (p = 0.267) (Figure 3).


Pump thrombosis is a life-threatening complication in LVAD patients.5 Several case series have already described STL treatment,7 but none of these publications compared STL with DEx. Because STL can be associated with high risk of bleeding complications4,7 and DEx is a safe surgical procedure, the latter is currently considered as the gold standard.

Our data indicate that STL is a feasible treatment alternative for selected LVAD patients with pump thrombosis. Although STL is less invasive, it is often not considered because of concerns about serious complications such as massive intracranial bleeding or gastrointestinal hemorrhages. DEx after failed STL may amplify such complications, and the concern is that these complications could become even more serious if the device has to be exchanged subsequently to STL.8,9 However, the plasma half-life of tenecteplase and alteplase is only 18 and 14 minutes, respectively.17 Therefore, we deem that subsequent DEx can be safely performed 12 hours after STL. There are also concerns about the high rate of therapy failure.8 Our results are in general agreement with these earlier data. Systemic thrombolysis should therefore be considered as first-line therapy. In agreement with this strategy, the STL group was slightly less sick than the DEx group at the time of pump thrombosis, as indicated by SAPS-II score values and some biochemical parameters. Device exchange remains an important treatment option for pump thrombosis. Moreover, because of the risk of pericardial tamponade, STL should not be performed within the first month of ventricular assist device implantation. During that period, massive adhesions are still absent. Therefore, DEx can easily be performed. Device exchange in LVAD patients, especially with pump stop, is a very satisfactory therapy, especially in patients who are not in severe cardiogenic shock.6,18 Device exchange through a lateral subcostal thoracotomy in patients supported with HMII18 seems to be easier to perform than that in HW patients because of better access to the pump head lying in the diaphragmatic pocket. This can probably further reduce the surgical risk of DEx because of avoidance of resternotomy. Alternatively, minimally invasive DEx without cardiopulmonary bypass can be performed through a left subcostal thoracotomy to reduce bleeding risks.19

However, there are situations in which STL may be preferable. This therapy is minimally invasive and can be performed in any cardiologic center with minimal effort if the patient is hemodynamically stable. Systemic thrombolysis usually does not increase the risk of pump exchange, and for that reason, it is worth performing in selected cases. Compared with DEx, the burden of STL for the patient is much lower. In addition, it is still possible to perform DEx after STL, if necessary, whereas STL cannot be performed immediately after DEx because of the risk of pericardial tamponade. Furthermore, in some countries, such as the United Kingdom and Brazil, reimbursement of LVAD therapy is very limited.19,20 It is however noteworthy that STL should only be performed in stable patients, whereas DEx is definitively necessary in patients who are in need of resuscitation because of pump stop. In our view, timing of STL is crucial and influences its success. Thrombolysis should be performed as soon as possible after the onset of clinical symptoms of pump thrombosis, within 24 hours at the latest. Longer duration of untreated device thrombosis allows the thrombus to become organized within the specific parts of the rotor and bearings, making resolution nearly impossible.7 Pump design probably also plays a role in the efficacy of STL. The HMII pump contains a cup and a ball bearing. Any thrombus that forms within these components can become very solid and firm because of protein degeneration caused by heat generation through higher friction, designated as white thrombus. In addition, low pump flows in HMII patients caused by hypovolemia or malposition of the inflow conduit can facilitate fresh thrombus formation at the inflow stator of the pump head (Figure 2). In the HW device, the rotor is suspended on a fluid hydrodynamic film, which allows more complete washout of blood (Figure 2). It may well be that the standard alteplase dose for patients with acute stroke we used in our patients (1 mg/kg body weight) has contributed to the success of the therapy and also to the fatal cases of intracerebral hemorrhage. Notably, many others use lower doses of alteplase (e.g., 30 mg) in LVAD patients with thrombosis. Another potential limitation of STL (unless CT scan is done in all patients) is that any outflow graft issues or pump placement issues are not address with STL. Also, it is often unknown how much clot is present until the device is removed. Therefore, in some patients STL is an ultimate cure, but in other patients the thrombus may be only partially resolved.

This study is limited by its small number of patients, its retrospective study design, and the large variability of the clinical condition of the patient population. Nevertheless, the results of our investigation allow us to make several tentative recommendations to assist in guiding the decision to be made between DEx and STL: Pump thrombosis is an emergency situation needing an immediate intervention. Systemic thrombolysis is a reliable treatment option in LVAD patients with pump thrombosis. Systemic thrombolysis should be performed as soon as possible after the onset of pump thrombosis.

In summary, our data indicate that STL and DEx reveal similar 90 day and 2 year survival rates in LVAD patients experiencing pump thrombosis. In selected patients, STL appears to be a relatively safe, less invasive, and less expensive therapy when compared with DEx.


1. Kirklin JK, Naftel DC, Pagani FD, et al.: Sixth INTERMACS annual report: A 10,000-patient database. J Heart Lung Transplant 2014.33: 555564.
2. Kirklin JK, Naftel DC, Pagani FD, et al.: Long-term mechanical circulatory support (destination therapy): On track to compete with heart transplantation? J Thorac Cardiovasc Surg 2012.144: 584603; discussion 597.
3. Park SJ, Milano CA, Tatooles AJ, et al.; HeartMate II Clinical Investigators: Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy. Circ Heart Fail 2012.5: 241248.
4. Najjar SS, Slaughter MS, Pagani FD, et al.; HVAD Bridge to Transplant ADVANCE Trial Investigators: An analysis of pump thrombus events in patients in the HeartWare ADVANCE bridge to transplant and continued access protocol trial. J Heart Lung Transplant 2014.33: 2334.
5. 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: 1222.
6. Ota T, Yerebakan H, Akashi H, et al.: Continuous-flow left ventricular assist device exchange: Clinical outcomes. J Heart Lung Transplant 2014.33: 6570.
7. Bartoli CR, Ailawadi G, Kern JA: Diagnosis, nonsurgical management, and prevention of LVAD thrombosis. J Card Surg 2014.29: 8394.
8. Schlendorf K, Patel CB, Gehrig T, et al.: Thrombolytic therapy for thrombosis of continuous flow ventricular assist devices. J Card Fail 2014.20: 9197.
9. Hohner E, Crow J, Moranville MP: Medication management for left ventricular assist device thrombosis. Am J Health Syst Pharm 2015.72: 11041113.
10. Viegas LD, Stolz E, Canhão P, Ferro JM: Systemic thrombolysis for cerebral venous and dural sinus thrombosis: A systematic review. Cerebrovasc Dis 2014.37: 4350.
11. Tang GH, Kim MC, Pinney SP, Anyanwu AC: Failed repeated thrombolysis requiring left ventricular assist device pump exchange. Catheter Cardiovasc Interv 2013.81: 10721074.
12. Kiernan MS, Pham DT, DeNofrio D, Kapur NK: Management of HeartWare left ventricular assist device thrombosis using intracavitary thrombolytics. J Thorac Cardiovasc Surg 2011.142: 712714.
13. Koertke H, Zittermann A, Mommertz S, El-Arousy M, Litmathe J, Koerfer R: The Bad Oeynhausen concept of INR self-management. J Thromb Thrombolysis 2005.19: 2531.
14. Paluszkiewicz L, Schulte-Eistrup S, Körtke H, Morshuis M, Gummert J: Thrombosis of the LVAD inflow cannula detected by transthoracic echocardiography: 2D and 3D thrombus visualization. Echocardiography 2011.28: E194E195.
15. Paluszkiewicz L, Gürsoy D, Spiliopoulos S, et al. HeartMate II ventricular assist device thrombosis-an echocardiographic approach to diagnosis: Can Doppler evaluation of flow be useful? Am Soc Echocardiogr 2011.24: 350.e1350.e4.
16. Raman SV, Sahu A, Merchant AZ, Louis LB IV, Firstenberg MS, Sun B: Noninvasive assessment of left ventricular assist devices with cardiovascular computed tomography and impact on management. J Heart Lung Transplant 2010.29: 7985.
17. Melandri G, Vagnarelli F, Calabrese D, Semprini F, Nanni S, Branzi A: Review of tenecteplase (TNKase) in the treatment of acute myocardial infarction. Vasc Health Risk Manag 2009.5: 249256.
18. Gregoric ID: Exchange techniques for implantable ventricular assist devices. ASAIO J 2008.54: 1419.
19. Birks EJ: The comparative use of ventricular assist devices: Differences between Europe and the United States. Tex Heart Inst J 2010.37: 565567.
20. Moreira LF, Benício A: Mechanical circulatory support: A great gap in Brazilian cardiac surgery. Rev Bras Cir Cardiovasc 2010.25: XXII.

left ventricular assist device; pump thrombosis; systemic thrombolysis; device exchange

Copyright © 2016 by the American Society for Artificial Internal Organs