Share this article on:

Device Exchange in HeartMate II Recipients: Long-Term Outcomes and Risk of Thrombosis Recurrence

Levin, Allison P.*; Uriel, Nir; Takayama, Hiroo; Mody, Kanika P.*; Ota, Takeyoshi§; Yuzefpolskaya, Melana*; Colombo, Paolo C.*; Garan, Arthur R.*; Dionizovik-Dimanovski, Marija*; Sladen, Robert N.; Naka, Yoshifumi; Jorde, Ulrich P.*

doi: 10.1097/MAT.0000000000000170
Adult Circulatory Support

Successful long-term use of the HeartMate II (HM II) left ventricular assist device has become commonplace but may be complicated by mechanical failure, infection, or thrombosis necessitating device exchange (DE). A subcostal approach to device exchange with motor exchange only is less traumatic, but long-term outcomes have not been reported. A retrospective chart review of all patients who required HM II to HM II device exchange at our institution was conducted. Of the 232 HM II patients implanted between January 2008 and July 2013, 28 required 36 device exchanges during a follow-up of 33.72 ± 17.25 months. The Kaplan–Meier 1 year survival was 63% for sternotomy exchanges and 100% for subcostal exchanges. Twenty-one exchanges were performed for initial or recurring device thrombosis. Although there was no difference in the risk of subsequent thrombosis after subcostal versus sternotomy exchange, the overall risk of recurring device thrombosis after device exchange for the same was high (31%). HM II device exchange via the subcostal approach has excellent short- and long-term outcomes. Device exchange performed for thrombosis is associated with a high recurrence risk irrespective of surgical approach

From the *Division of Cardiology, Department of Medicine, Columbia University Medical Center, New York, New York; Division of Cardiology, Department of Medicine, University of Chicago, Chicago, Illinois; Department of Surgery, Columbia University Medical Center, New York, New York; §Department of Surgery, University of Chicago, Chicago, Illinois; and Department of Anesthesiology, Columbia University Medical Center, New York, New York.

Disclosures: Dr. Jorde and Dr. Naka have received consulting fees from Thoratec and Heartware; Dr. Uriel has received consulting fees from Thoratec, Heartware, and xDX. The rest of the authors have no prior or current financial relationship with any other entity. The remaining authors have no conflicts of interests to disclose.

Correspondence: Ulrich P. Jorde, Associate Professor, Division of Cardiology, Department of Medicine, Columbia University Medical Center, 622 W 168th Street, New York, NY 10032.

The morbidity and mortality of device exchange in HeartMate II (HM II) patients have become a significant concern in the wake of a recently reported increase in device thrombosis.1, 2 Although there is currently no true alternative to device exchange for refractory infection, mechanical pump failure, or pump thrombosis, a subcostal rather than sternotomy incision has successfully been used at our institution to reduce postoperative complications (including blood transfusion requirements, time to extubation, and intensive care unit [ICU] stay).3 Similarly, reduced operative morbidity and mortality has previously been described in patients exchanged via subcostal approach from the HM XVE to the HM II.4 However, long-term outcomes after subcostal device exchange or risk of recurrence of device thrombosis in cases of pump thrombosis have not previously been reported. The purpose of this study was to assess the long-term outcomes of subcostal and sternotomy device exchanges for all indications, and to assess the risk of device thrombosis recurrence after pump exchange for thrombosis.

Back to Top | Article Outline


This study was approved by the Institutional Review Board of Columbia University Medical Center. A retrospective review of medical records of patients implanted between January 2008 and July 2013 was completed. As previously described by Ota et al., we began using the subcostal approach for DEs in June 2011, unless contraindicated.3

Back to Top | Article Outline

Definitions of Indication for Exchange

Device thrombosis was defined as a positive ramp test in the setting of hemolysis or sustained power elevations, with confirmation of thrombosis upon exchange. Device malfunction was defined as any breakdown of proper device functioning that leads to compromised cardiac output, in the absence of signs for device thrombosis. In addition, device exchange was performed for driveline infection refractory to antibiotic therapy as well as suspected pocket or device infection in the presence of bacteremia.

Back to Top | Article Outline

Surgical Technique

In brief, an incision is made in the left subcostal area and the soft tissue is divided to dissect out the device including the inflow conduit, pump, and the proximal part of the outflow graft. Cardiopulmonary bypass is commenced through femoral vessel cannulation. When indicated, the angle of the inflow cannula is adjusted allowing inflow cannula direction toward the mitral valve by additional dissection around the inflow cuff. The outflow graft is then clamped and the pump motor is detached from both inflow conduit and outflow graft for removal. A new pump motor is placed in the pocket and connected to inflow and outflow. The pump is then actuated and de-aired through a needle hole in the outflow graft. Cardiopulmonary bypass is then weaned and the clamp on the outflow graft is removed.3

Of note, our technique for the incision in the subcostal area has changed since the original publication by Ota et al. The original incision was parallel and close to the left costal cartilage. However, after several cases of wound dehiscence with this approach, we now use an incision more perpendicular to the costal margin. This approach preserves enough soft tissue between the incision and the costal margin to allow extension to the intercostal space and division of the costal margin. When needed, another subxiphoid vertical incision is made to gain access to the outflow graft.

The contraindications for subcostal exchange included suspected device pocket infection and the necessity to access the outflow graft or concomitant cardiac procedures. The need for sternotomy was confirmed by a computer tomography (CT) scan with contrast media and evaluation of the inflow conduit and outflow graft with transesophageal echocardiography (TEE).

Back to Top | Article Outline

Anticoagulation Protocol

At the outset of our experience, we used heparin bridging postoperatively, and prescribed dual anticoagulation therapy warfarin, Aspirin 81 mg and Persantine 75 mg three times a day, with an international normalized ratio (INR) goal of 2–3. With the recognition of gastrointestinal bleeding as a major complication in 2009–2010, we lowered our INR goal to 1.5–2 and removed Persantine.5 Concurrently, heparin bridging was discontinued after Slaughter et al. showed that its use lead to an increase in bleeding events without a reduction in short-term thrombotic or thromboembolic events.6 However, given the increased incidence of device thrombosis observed nationally beginning in 2011, we have returned to the original protocol, with the exception of Persantine use in patients more than 50 years of age. In patients with recurrent thrombosis, we use Persantine irrespective of age.

Back to Top | Article Outline

Statistical Analysis

Categorical variables are reported as frequencies, and continuous variables are reported as the mean ± standard deviation or as medians with interquartile range (IQR) when applicable. The Fisher’s exact or χ2 tests were used to analyze categorical variables for univariate analysis. Student’s t-test or the Mann–Whitney U test was used for the continuous variables based on the distribution of the data. Statistically significant was defined as p < 0.05 in a two-sided test. Kaplan–Meier analysis was used to report freedom from device exchange in the entire cohort and to compare actuarial survival after primary device exchange based on surgical technique. Additionally, the rates of secondary and tertiary device exchanges were assessed, as well as recurrent thrombosis. All statistical analyses were performed with Med Calc software (MedCalc Statistical Software version 12.7.3; MedCalc Software bvba, Ostend, Belgium).

Back to Top | Article Outline


During the study period, 232 patients were implanted with a HM II. The median time on support for the entire HM II population was 12.46 months (IQR = 6.2, 23.5), and the mean follow-up time at analysis was 33.72 ± 17.25 months. During the study period, 37 DEs (HM II to HM II) were performed on 28 patients. Of the 28 primary DEs, 12 were performed with a sternotomy approach and 16 with a subcostal approach. None of the baseline characteristics differed significantly between the sternotomy and subcostal cohorts (Table 1).

Back to Top | Article Outline

Primary Device Exchange

The overall freedom from primary device exchange was 91% at 1 year and 80% at 2 years, which represents 0.09 events per patient-year (Figure 1). For the sternotomy cohort (n = 12), the median time to primary exchange was 12.38 months (IQR: 5.76, 19.61), with an average follow-up time of 46.04 ± 15.11 months. Similarly, the median time to first device exchange for the subcostal cohort (n = 16) was 11.75 months (IQR: 3.09, 17.78) with a mean follow-up time of 33.63 ± 19.53 months.

Of the 28 primary device exchanges, 13 were performed for device malfunction, 14 for thrombosis, and one for infection (Table 2). In the sternotomy cohort, seven (59%) were performed due to electrical or mechanical device malfunction, four (33%) for device thrombosis, and one (8%) for device infection. In the subcostal cohort, six (37.5%) were performed for device malfunction and 10 (62.5%) for device thrombosis. Although the indication for primary device exchange did not differ significantly between surgical approaches, the main indication for device exchange changed from device malfunction to device thrombosis after March 2011. For implants before March 2011, 29% of all primary DEs were performed for device thrombosis, compared with 71% after March 2011 (Figure 2; p = 0.06). Although the absolute number of device exchanges for malfunction did not change significantly, the number of exchanges for device thrombosis increased markedly. In addition, the time to exchange for device thrombosis decreased markedly from before to after March 2011, with mean time to device exchange of 21.29 ± 14.88 and 5.19 ± 4.29 months, respectively (p = 0.0018). All 10 primary device exchanges performed for device thrombosis in patients with initial implants after 2011 were completed within 1 year. In contrast, the first exchange for device thrombosis for patients implanted before 2011 did not occur until 18 months after the initial implant.

Back to Top | Article Outline

Multiple Device Exchanges and Recurrent Thrombosis

After a primary device exchange for any indication, seven patients required at least one additional exchange (refer to Figure 3 for flowchart of device exchanges by indication and procedure). Overall, three patients had primary and secondary exchanges for device thrombosis and thus had recurrent thrombosis at the time of their secondary exchange (second thrombus indicated by arrows and dashed lines in Figure 3B). In addition, two patients had second and third device exchange for thrombosis, so the recurrent thrombosis occurred at their third device exchange (second thrombus indicated in Figure 3C).

Of the 18 patients who had at least one exchange for device thrombosis (14 during a primary exchange and 4 for a secondary exchange), five (31%) had at least one recurrence. The rate of re-thrombosis was 0.43 events per patient-year for patients whose first thrombosed device was exchanged by sternotomy, and 0.33 events per patient-year for those whose first thrombosed device was exchanged subcostally. The second device thrombosis was exchanged with a subcostal incision in two patients and a sternotomy in the other three (Figure 4).

All of the patients who required multiple exchanges were male, and there were no significant differences in the patients who had recurrent device thrombosis versus first time device thrombosis (Table 3). Furthermore, the average INR between the initial implant and first device exchange for all patients who had a primary exchange for device thrombosis was 1.93 ± 0.56 for patients who did not have recurrent thrombosis, and 1.94 ± 0.77 for patients who re-thrombosed (p = 0.97). In addition, for the patients who re-thrombosed, the average INR between the first and second thrombus was 2.07 ± 0.58. Although two of the patients who had a device thrombosis (at any time) did have hematological disorders, neither of them had recurrent thrombosis. The first patient was a 73 year-old male with idiopathic thrombocytopenic purpura (ITP) who had a primary sternotomy exchange for device thrombosis, who died 16 months post device exchange without any further device thromboses. The second patient was a 43 year-old female who also had a primary sternotomy exchange for device thrombosis, and died in the first month after exchange because of multiorgan failure.

Back to Top | Article Outline

Outcomes After Primary Device Exchange

After a primary device exchange, the possible outcomes were transplantation, re-exchange, death, or continued support. The 1 year survival was 62.5% for the sternotomy exchanges and 100% for the subcostal exchanges (p = 0.02; Figure 5), censored for re-exchange and transplantation. At 30 days post device exchange, the actuarial survival was 83.3% and 100% for the sternotomy and subcostal cohorts, respectively (p = 0.09). In the sternotomy cohort (5 bridge to transplant [BTT]/7 destination therapy (DT)), four patients were re-exchanged, one was transplanted, one remains on HM II support, and six died. Four of these patients died within 1 year of sternotomy device exchange, two died because of postoperative multiorgan failure, one because of an infection (bacteremia), and one because of battery removal while intoxicated. The other two patients died at 16 months and 3 years post device exchange because of multiorgan failure and sudden death at home, respectively. In contrast, none of the patients who received a subcostal exchange (11 BTT/5 DT) have died at the time of this writing. Seven (44%) were successfully transplanted with a median time to transplantation of 4.5 months (IQR = 2.1, 8.9), four (25%) were re-exchanged, and five (31%) remain on HM II support (for a mean time of 21.63 ± 7.64 months).

A total of 10 patients received a transplant, eight after the first device exchange, two after a second exchange, and one after a third exchange. The median follow-up time posttransplant is 13.23 months, ranging from 6.66 to 49.85 months. The 1 year freedom from death posttransplant was 87.5%, with one death at 9 days after transplant. This patient was successfully transplanted after a subcostal primary device exchange for device thrombosis, and died posttransplant because of a massive intraoperative ischemic stroke. The intraoperative TEE showed a large columnar thrombus that slowly migrated from the ascending aorta down to the descending aorta, which was suggestive of the migration of a thrombus that had existed inside of the outflow graft. Our current protocol includes screening patients with (suspected) left ventricular assist device (LVAD) thrombosis with CT angiogram to assess the outflow graft.

Back to Top | Article Outline


We examined long-term outcomes of HM II recipients requiring device exchange, as well as the risk for recurrent device thrombosis in this population. Our principal finding is that the long-term outcomes after a primary subcostal device exchange, with a 1 year survival of 100% in 16 subjects, are excellent and possibly superior to those achieved with full sternotomy. Furthermore, although the precise etiology of recurrent device thrombosis remains elusive, the frequency of 31% in this series is of great concern and warrants particularly careful monitoring.

The HM II is the only currently approved device for destination therapy in the United States and has an excellent overall performance record with a nearly 65% 2 year survival in patients otherwise highly unlikely to survive at all.7, 8 However, the recently reported significant increase in the rate of device thrombosis has raised concerns regarding the overall benefit of its use.1 Historically, device thrombosis has been an infrequently observed event diagnosis of which required substantial hemolysis (plasma free hemoglobin > 40, lactate dehydrogenase > 1,000) as well as overt device malfunction.7 More recently, many centers, including our own, use surveillance LDH screening and echocardiography to detect device thrombosis early.9 Whether the recently reported “spike” in device thrombosis is a true increase or the result of enhanced screening and “preemptive” device exchange remains subject of intense discussion in the LVAD community at the time of this writing. Furthermore, device exchange is a clinical endpoint equivalent to major stroke or death in prior as well as ongoing mechanical circulatory support trials (ENDURANCE).10, 11 Therefore, development of a successful management approach for device thrombosis and understanding of its impact on overall outcomes are necessary.

Although device exchange is costly and has historically been associated with major morbidity and possibly mortality, our data support the use of an early subcostal device exchange as a favorable solution to device thrombosis. Although the difference in survival between sternotomy and subcostal approach in our experience at first glance is striking (and similar to those previously reported by Moazami et al.),12 a report of 45 primary continuous flow device exchanges performed by sternotomy by Stulak et al. with 95% actuarial survival rate at 1 year and 77% at 18 months underscores that the sternotomy approach might also yield outstanding results.13 Therefore, the outcome differences we report may be because of selection bias, as patients with a suspicion of inflow cannula or outflow graft thrombosis will undergo device exchange via sternotomy. With that said, we believe that shorter operative and bypass times would ultimately translate into better outcomes in a head-to-head comparison, and that the substantially less invasive nature of subcostal exchange prohibits randomized study. Furthermore, although subcostal exchange is less traumatic, we do not feel that this resulted in a lower threshold for surgical intervention in either cohort. Although the introduction of subcostal exchange may lead to more widespread adoption of early device exchange algorithms, our indication for device exchange has been unchanged for several years. As described in one of our earlier publications, we routinely proceed to device exchange once device thrombosis is suspected based on LDH elevation as well as positive ramp study, and end-organ dysfunction has ensued.9, 14 This algorithm has been used regardless of the surgical technique that was used, and as such we do not believe this could have biased our results.

The risk of recurrent thrombosis caused by surgical and mechanical factors must also be considered. One concern inherent in the use of subcostal exchange (i.e., a simple “motor replacement”) is the chance of missing a thrombus when the outflow graft and inflow cannula are not removed.10 However, we did not detect a difference in recurrent thrombosis rate between subcostal (i.e., motor exchange only) and sternotomy approach (with replacement of inflow cannula and outflow graft). Of note, multiple steps are taken in the operating room to reduce the chance of missing a thrombus. After motor removal, the inlet stator is inspected for clot to confirm the clinical diagnosis of device thrombosis. Clot is most commonly found in this location. Second, brisk back-bleed should be expected from an uncompromised inflow conduit and outflow graft. Absence of clot in conjunction with poor back bleeding would dictate conversion to sternotomy with replacement of inflow cannula and outflow graft. Furthermore, an abbreviated ramp test with intraoperative TEE is sometimes performed after the exchange to ensure that the left ventricle (LV) decompresses with increased speed.9

Although multiple papers have commented on the relative safety of device exchange, none have highlighted a high rate of recurrent thrombosis. In our series, 31% of patients with device thrombosis developed a second thrombosis despite strict monitoring of anticoagulation; none of these patients had a known hematological disorder. This is much higher than the previously reported rate of multiple exchanges of 8.3% by Moazami et al., in which only six of the 72 patients required multiple device exchanges (note: this is for all indications).12

Although the recurrent thrombosis cohort is not large enough to warrant any definitive conclusions, it is clear that this issue requires further investigation of potentially contributing clinical, surgical, and manufacturing factors. Since the five patients who experienced recurrent device thrombosis required a total of 17 devices, it is apparent that this phenomenon has considerable financial and clinical implications.

As with any retrospective analysis with relatively small cohort sizes, this investigation has its limitations. For example, it must be noted that the subcostal and sternotomy exchanges were used in different eras, so that familiarity and understanding of device exchanges were higher when the subcostal exchanges began. Also, the high rate of re-exchange as compared to prior analyses must be evaluated in the context of increased vigilance for the detection of device thrombosis. Lastly, it is acknowledged that the direct comparison of surgical technique to long-term outcomes is not perfect, and that many clinical confounding factors likely exist.

In conclusion, this study shows excellent long-term outcomes after subcostal device exchange, as well as the significant risk of re-thrombosis after device exchange, irrespective of operative approach. The subcostal exchange is without question less traumatic and costly than a sternotomy exchange with lower, if any, blood transfusion requirements, shorter time to extubation, and shorter ICU stay.3 Lastly, patients can be successfully exchanged multiple times, although recurrent device thrombosis may render multiple device exchanges a costly and ultimately futile effort despite excellent technical results. This finding further underscores the urgent need to study the etiology of device thrombosis.

Back to Top | Article Outline


1. 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
2. Kirklin JK, Naftel DC, Kormos RL, et al. Intermacs analysis of pump thrombosis in the HeartMate II left ventricular assist device. J Heart Lung Transplant. 2014;33:12–22
3. Ota T, Yerebakan H, Akashi H, et al. Continuous-flow left ventricular assist device exchange: clinical outcomes. J Heart Lung Transplant. 2014;33:65–70
4. Gregoric ID, Bruckner BA, Jacob L, et al. Clinical experience with sternotomy versus subcostal approach for exchange of the HeartMate XVE to the HeartMate II ventricular assist device. Ann Thorac Surg. 2008;85:1646–1649
5. Uriel N, Pak SW, Jorde UP, et al. 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;56:1207–1213
6. 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;29:616–624
7. Rose EA, Gelijns AC, Moskowitz AJ, et al.Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Study Group. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med. 2001;345:1435–1443
8. 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;5:241–248
9. Uriel N, Morrison KA, Garan AR, et al. Development of a novel echocardiography ramp test for speed optimization and diagnosis of device thrombosis in continuous-flow left ventricular assist devices: the Columbia ramp study. J Am Coll Cardiol. 2012;60:1764–1775
10. Pagani F, Rogers J. A Clinical Trial to Evaluate the HeartWare® Ventricular Assist System (ENDURANCE). 2010 Unique identified: NCT01166347.
11. Slaughter MS, Rogers JG, Milano CA, et al.HeartMate II Investigators. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med. 2009;361:2241–2251
12. Moazami N, Milano CA, John R, et al.HeartMate II Investigators. Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality. Ann Thorac Surg. 2013;95:500–505
13. Stulak JM, Cowger J, Haft JW, Romano MA, Aaronson KD, Pagani FD. Device exchange after primary left ventricular assist device implantation: indications and outcomes. Ann Thorac Surg. 2013;95:1262–1267 discussion 1267
14. 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;33:51–59

CF-LVAD; Exchange; LVAD; Malfunction; Subcostal; Sternotomy; Thrombosis

Copyright © 2015 by the American Society for Artificial Internal Organs