Share this article on:

Costs and Outcomes in the Care of Bi-ventricular Support as a Bridge to Cardiac Transplant

Swartz, Michael F.; Angona, Ron; Smith, Karen; Kraenzlin, Franca; Stypula, Christine M.; Joshi, Devang; Tchantchaleishvili, Vakhtang; Hicks, George L.; Massey, H. Todd

doi: 10.1097/MAT.0000000000000395
Adult Circulatory Support

Bi-ventricular (Bi-V) mechanical circulatory support is commonly used as a bridge to cardiac transplant. However, the optimal strategy is unknown. We examined the outcomes, as well as the costs in the use of Bi-V support as a bridge to cardiac transplant. From 2001 to 2014, three different Bi-V support strategies were utilized: 1) Para-corporeal ventricular assist device (PVAD-2001–2006), 2) Heartmate II left ventricular assist device in conjunction with a temporary CentriMag right ventricular assist device (HMII + CMAG-2006–2012), and the total artificial heart (TAH-2012–2014). Total costs were derived from the hospitalization at implant, and postimplant costs defined as equipment and re-hospitalizations before transplantation. Sixty-five (34 PVADs, 20 HMII + CMAG, and 11 TAHs) devices were used as a bridge for transplant. There were no differences in implant variables including age, INTERMACS score, or implant length of stay. Although the wait list mortality was not different between groups (PVAD-32%, HMII + CMAG-45%, TAH-54%; p = 0.3), the percentage of patients transplanted were highest in the PVAD group: (PVAD-55.8%, HMII + CMAG-30.0%, TAH-18.2%; p = 0.01). Total costs were not significantly different between groups (PVAD-$306,166 ± 247,839, HMII + CMAG-$278,958 ± 135,324, TAH-$321,387 ± 21,2477; p = 0.5). Despite variations in therapy, outcomes and costs for patients requiring Bi-V support as a bridge to cardiac transplant have remained constant.

From the *University of Rochester, Strong Memorial Hospital, Rochester, New York; and University of Louisville, Louisville, Kentucky.

Submitted for consideration November 2015; accepted for publication in revised form March 2016.

Disclosures: H. Todd Massey is a consultant for both Thoratec and Syncardia.

This work was funded by a honorarium from Syncardia.

Correspondence: Michael F. Swartz, Strong Memorial Hospital, Box Surg/Cardiac, 601 Elmwood Ave, Rochester, New York 14642. Email: michael_swartz@urmc.rochester.edu.

The use of mechanical circulatory support as a bridge to transplant has increased exponentially within the past decade.1–3 The combination of static organ allocation, coupled with new technology, have tripled the number of left ventricular assist device (LVAD) implants over the past 10 years.4,5 Similar to the increase in LVAD implantation, the health-care costs for those patients with advanced heart failure have only increased.6,7

As a result of the financial implications of LVAD therapy, a number of reports have compared both the clinical outcomes, and fiscal effectiveness, in the use of a ventricular assist device as a bridge to transplant.1,3 These reports have suggested that LVAD support as a bridge to transplant compared with nonbridged patients results in increased survival, but at an increased cost.1 However, these reports have focused mainly on isolated left ventricular failure, and therefore the understanding of the costs and outcomes of bi-ventricular (Bi-V) support as a bridge to transplant are unknown.

Since 2001, we have utilized three different systems in the care of adults with Bi-V congestive heart failure: para-corporeal ventricular assist device (PVAD) (Thoratec, Pleasanton, CA) Heartmate II LVAD + CentriMag (CMAG) right ventricular assist device (RVAD) (Thoratec, Pleasanton, CA), and the total artificial heart (Syncardia, Tucson, AZ). Based on these three Bi-V support systems, we sought to compare not only the clinical outcomes but also the implant and subsequent costs to the time of cardiac transplant.

Back to Top | Article Outline

Methods

Following Institutional Review Board approval, all adult patients greater than 18 years of age who received Bi-V support as a bridge to cardiac transplant were reviewed between 2001 and 2014 at the University of Rochester Medical Center. The University of Rochester Medical Center, is a quaternary cardiac care hospital, serving approximately 4.5 million people in the upstate and western New York region that encompasses approximately 16,802 of the 47,126 square miles of New York State. Exclusion criteria included those patients who required Bi-V support, and were not listed for cardiac transplant, and those patients who required isolated right or left ventricular support. Patients were divided into one of three groups based on the Bi-V support strategy: 1) PVAD 2001–2006; 2) Heartmate II LVAD (Thoratec, Pleasanton, CA) in conjunction with a temporary CMAG RVAD (Thoratec, Pleasanton, CA) 2006–2012; and 3) total artificial heart (TAH) 2012–2014.

Preoperative demographics and operative details from the time of Bi-V support were determined from chart review. Total follow-up was quantified as the time from implant to either transplant, change in united network for organ sharing status to status 7, or death. Information regarding the number of subsequent hospital admissions before transplant, as well as subsequent procedures were determined. In addition, the necessary pump supplies during the follow-up period including pump replacements were quantified. Outcomes during the overall follow-up period as well one-year outcomes were determined in all cases. The University of Rochester is the only ventricular assist device/heart transplant center serving the upstate New York region, and therefore, all patients receive follow-up care at the University of Rochester.

Financial metrics were calculated from patient billing information. Our definition of health-care costs were defined as the amount of money required by the hospital to care for the patient requiring Bi-V support as a bridge to transplant. Therefore, the hospital charges were calculated, and the charges were then subtracted by the percent markup to quantify the health-care costs. We felt this definition of cost was a static measure across the implant period as the reimbursement for Bi-V support can vary by region, era, and insurance type. Costs were quantified as the amount in 2014 US dollars that were accrued during a hospitalization/procedure. Total costs were comprised from three different areas: 1) the cost of the implant hospitalization, 2) the cost of subsequent hospitalizations, and 3) the costs of subsequent pump equipment which was necessary after the time of implant. Similar to the outcomes data, one-year costs were obtained in all areas for every patient.

Back to Top | Article Outline

Statistics

Variables are reported as mean ± standard deviation, median with range, or frequency and percentage. Patient variables were evaluated for the equality of variances ensuring normal distribution, and were then compared using either analysis of variance with a post hoc Bonferroni correction between groups, or Kruskall–Wallis test with a post hoc Dunn’s comparison where appropriate comparisons made between categorical variables were done using Pearson’s χ2 analysis with post hoc Bonferroni correction. Kaplan–Meier curves were constructed and the log-rank determined, where a p value <0.05 was considered significant.

Back to Top | Article Outline

Results

Sixty-five patients received Bi-V circulatory support as a bridge to transplant from 2001 to 2014 (Figure 1). There were a total of 34 PVAD’s, 20 Heratmate II (HMII) + CMAG, and 11 TAH implants. Table 1 lists the perioperative demographics for the entire population. There were no differences in age, gender, prior surgery, liver/kidney function, or right and left ventricular performance between groups. For all groups, the acuity was high demonstrated by INTERMACS scores between 1.2 and 1.4. Despite similar acuity, the percentage of patients requiring pre- and subsequent postoperative extracorporeal membrane oxygenation (ECMO) use was highest within the TAH group. However, the operative mortality as well as the implant hospital length of stay remained nonsignificant.

Of the 20 patients requiring a HMII LVAD in conjunction with a CMAG RVAD, 95% (19/20) of the RVAD’s were explanted at 15.2 ± 8.1 days. There were no patients transplanted within the HMII group while requiring Bi-V support. Therefore, unlike the PVAD and TAH groups, the HMII patients although initially received Bi-V support, were maintained with LVAD support only until the time of transplant. Complications from the implant hospitalization as well as rehospitalizations are listed in Table 2. The number of rehospitalization was significantly greater within the HMII group, in part related to this group’s longer duration of support. The overall wait-list mortality was not different between groups (PVAD-32%, HMII + CMAG-45%, TAH-54%; p = 0.3). A total of 27/65 patients received a heart transplant at a median of 120 days (19–1,065 days). Median wait times by device demonstrated, significantly shorter wait times within the PVAD group. (PVAD-68.5; 19–619, HMII-339; 128–1,065, and TAH-327; 290–365 days; p = 0.01). Because of longer wait times, particularly within the HMII and TAH groups, we compared the 1 year survival to transplant for all patients to examine follow-up at a fixed duration. One year survival to transplant was highest within the HMII group (PVAD-63.4%, HMII-84%, TAH-45.4%; p = 0.02) (Figure 2A, B). At 1 year, within the PVAD group there were three patients remaining on support, 20 transplants, and 11 deaths. From the HMII group there were 15 patients remaining on support, two transplants, and three deaths and from the TAH group there were three patients on support, two transplants, and six deaths.

The cost of the pumps contributed, in most cases, to the majority of the implant costs and were highest within the PVAD group ($128,656), and lowest within the HMII + CMAG group ($94,500). Costs at the time of implant were greatest within the TAH group ($313,783 ± 203,950) and lowest within the HMII + CMAG group ($222,482 ± 123,798; Figure 3A). However, the large standard deviations prevented any statistical significance. In contrast, despite the large standard deviations, the subsequent re-hospitalization and pump supply costs were significantly greater with the HMII group (Figure 3B, C). Interestingly, the re-hospitalization costs exceeded $100,000 in many cases from within the HMII group over the study period. However, despite greater costs within specific areas, the total costs over the study period were not significantly different between the different pump strategies (PVAD-$306,166 ± 247,839, HMII + CMAG-$278,958 ± 135,324, TAH-$321,387 ± 212477; p = 0.5; Figure 3D).

Variation in costs over the first year of support was identical to the total costs for the TAH and PVAD groups because of the shorter duration of support. From within the HMII group, the average re-hospitalization costs were $19,108 ± 25,678 over the first year, and this group did not incur any equipment charges during the first year. Therefore, there were no significant differences when comparing the total costs at one year (Figure 4).

Back to Top | Article Outline

Discussion

The use of Bi-V support as a bridge to cardiac transplantation has dramatically changed over the past decade. With constantly changing organ allocation and listing restrictions and regulations,1–5 the ability to transplant patients from Bi-V support devices has dramatically changed. The PVAD group had the shortest wait times and subsequently the highest transplant rate when compared with the HMII and TAH devices. However, PVAD’s were implanted between 2001 and 2006 and benefited from a time where organs could be quickly allocated to those requiring mechanical circulatory support. Data from patients requiring Bi-V support as a bridge to transplant from other groups have demonstrated similar results with close to 50% of patients transplanted after only 100 days of support.8 The majority of those patients remained in the hospital until the time of transplant, and their re-hospitalization costs were low. The supply costs for those patients was limited to pump malfunction shortly after implant, and in part is related to the PVAD as an early generation pump.

The Heartmate II LVAD in conjunction with a CMAG RVAD began the third generation of heart pumps and provided nonpulsatile flow to patients for a longer duration of support. In the majority of cases, the HMII pump is able to provide afterload reduction to the right ventricle, and therefore in the absence of sustained ventricular arrhythmia’s can be explanted within 2–4 weeks. We have previously demonstrated an 88% survival to discharge when using this technique.9 Unfortunately for this group, in the absence of device malfunction, or recurrent arrhythmia it is very difficult to transplant these patients. As a result they had the longest duration of support, and despite having an implant in some patients 5 years earlier, are still listed as a UNOS 1B awaiting transplant. As a consequence the health-care costs associated with this group were some of the highest, in part relating to readmissions for stroke, bleeding, infection, and heart failure. In addition, this increased duration for support also resulted in a significant increase in pump-related supplies after implant.

The TAH system, because of its size requirements,10 was implanted only in men, and therefore this group had a larger body surface area; however, this was not statistically significant. This is similar to the experience of other centers with a larger number of implants, where 85% of patients were male with a mean weight of 88 kg.10 In addition, the TAH was associated with an increased use of ECMO and may in part relate to the overall greatest costs within this population. Although the device itself is more expensive than the HMII, the subsequent readmission, and supply costs were the low. The TAH system is unique in that it allows the surgeon to recover the patient from end-organ dysfunction, discharge the patient home, and then after the patient is stable readmit them to the hospital as a 1AA patient. Therefore, the patient and surgeon must discuss the options in using either a permanent HMII LVAD with temporary RVAD, which may result in longer wait times, or using permanent Bi-V support with the hope of obtaining earlier organ allocation. Therefore, this pump may be optimal in underpopulated areas such as upstate and western New York where the duration to heart transplant may be extend beyond the national average due to limited organ allocation. In contrast, super-populated areas such as southern California, which historically have shorter wait-list times,11 may be less suited to the TAH platform. Longer wait times, may have in part influenced the higher mortality observed from within our population. In many cases, patients were previously listed, and we were unable to obtain a suitable heart for transplant before the patient developed unstable hemodynamics and subsequently requiring Bi-V mechanical circulatory support.

The health-care costs associated Bi-V support as a bridge to cardiac transplant continues to grow. As the technology only improves, the indications for implantation have become less stringent. There is clearly a benefit in the use of Bi-V support as a bridge to cardiac transplant, lowering the transplant mortality, and improving the overall outcomes.1–5 However, the financial impact of these devices is dramatic.12,13 Our costs were lower to previous reports in which isolated LVAD therapy was associated with costs of $1,000.00/day. This may be related to the methods by which costs are calculated, as our definition of cost was related to the cost incurred by the hospital, and not that charged by the hospital. We felt that the costs incurred by the hospital provided a realistic financial metric for the services provided. Hospital charges (money billed to private insurance and Medicare) are a commonly used financial metric, however, charges for the same procedure often vary from hospital to hospital because of differences in charging structure. In addition, charges are only reimbursed at a rate of approximately 25% for Medicare and 38% for private insurance14,15

Unfortunately, based on both the financial metrics and clinical outcomes, there was no clear Bi-V support strategy that demonstrated clear superiority from within this cohort of patients. This suggests that surgeons can freely implant the device they feel most comfortable, without concern that only one strategy results in superior clinical outcomes with limited costs.

Back to Top | Article Outline

Limitations

This is a retrospective review and therefore has the limitations inherent to that approach. In addition, the Bi-V support for each pump was indicative of a specific era’s both in pump generation, as well as transplant. Certainly, if PVAD support was required beyond one year in the majority of patients, this approach would appear inferior. In contrast, if the HMII + CMAG group and/or the TAH group could be transplanted within months after implant, these results might appear superior. The HMII + CMAG group although received Bi-V support at the time of implantation, the RVAD was weaned and explanted, and therefore, those patients were followed and subsequently transplanted supported only with an LVAD. Finally, these data are representative from one program within an underpopulated area of the United States and may not be reflective of all cardiac transplant centers.

Back to Top | Article Outline

Conclusions

Over the past 15 years, there have been several pump strategies used in the support of Bi-V failure. Despite the variations in Bi-VAD (bi-ventricular assist device) strategy data from our program suggest no significant difference in clinical outcomes between groups. Further, financial data suggest no clear strategy is optimally cost–effective. Therefore, the choice of the Bi-VAD strategy utilized can rest entirely with the surgeon, and their comfort level with each pump, as no one pump demonstrates clear clinical or financial superiority.

Back to Top | Article Outline

References

1. Alba AC, Alba LF, Delgado DH, et al: Cost-effectiveness of ventricular assist device therapy as a bridge to transplantation compared with nonbridged cardiac recipients. Circulation 2013.127: 2424–2435.
2. Lampropulos JF, Kim N, Wang Y, et al: Trends in left ventricular device use and outcomes among medicare beneficiaries, 2004–2011. Open Heart 2014.1: e000109.
3. Khazanie P, Hammill BG, Patel CB, et al: Trends in the use and outcomes of ventricular assist devices among medicare beneficiaries 2006 through 2011. J Am Coll Cardiol 2014.63: 1395–1404.
4. Mulloy DP, Bhamidipati CM, Stone ML, et al: Orthotopic heart transplant versus left ventricular assist device: A national comparison of cost and survival. J Thorac Cardiiovasc Surg 2013.145: 566–574.
5. Rose EA, Gelijns AC, Moskowitz AJ, et al: Randomized evaluation of mechanical assistance for the treatment of congestive heart failure. Study group: Long-term mechanical left ventricular assistance for end stage heart failure. N Engl J Med 2001.345: 1435–1443.
6. Roger VL, Go AS, Lloyd-Jones DM, et al: Heart disease and stroke statistics-2012 update. A report from the American Heart Association. Circulation 2012.125: e2–220.
7. Heidenreich PA, Trogdon JG, Khavjou GA, et al: Forecasting the future of cardiovasciular disease in the united states: A policy statement form the American Heart Association. Circulation 2011.123: 933–944.
8. Brehm K, Heilmann C, Siepe M, et al: Thoratec paracorporeal biventricular assist device therapy: The Freiburg experience. Eur J Cardiothorac Surg 2012.41: 207–212.
9. Lazar JF, Swartz MF, Schiralli MP, et al: Survival after left ventricular assist device with and without temporary right ventricular support. Ann Thorac Surg 2013.96: 2155–2159.
10. Copeland JG, Copeland H, Gustafson M, et al: Experience with more than 100 total artificial heart implants. J Thorac Cardiovasc Surg 2012.143: 727–734.
11. Colvin-Adams M, Smith JM, Heubner BM, et al. OPTN/SRTR 2013 Annual data report: Heart. Am J Transplant 15(suppl 2): 1–28, 2015.
12. Hernandez AF, Shea AM, Milano CA, et al: Long-term outcomes and costs of ventricular assist devices among medicare beneficiaries. JAMA 300: 2398–2406, 2008.
13. Patel SR, Sileo A, Bello R, et al: Heart transplantation versus continuous-flow left ventricular assist device: Comprehensive cost at 1 year. J Card Fail 2015.21: 160–166.
15. Reinhardt UE: The pricing of US hospital services: Chaos behind a veil of secrecy. Health Aff 2006.25 1)57–69.

ventricular assist; cardiac transplant; bi-ventricular assist; mechanical circulatory support

Copyright © 2016 by the American Society for Artificial Internal Organs