Trends and outcomes following total artificial heart as bridge to transplant from the UNOS database

Abstract

Introduction:

Bridge to transplantation (BTT) with a SynCardiaTotal Artificial Heart (TAH) has been gaining momentum as a therapy for patients with biventricular heart failure. Recent transplant waitlist and posttransplant outcomes with this strategy have not been comprehensively characterized. We reviewed the United Network for Organ Sharing (UNOS) database to examine BTT outcomes for the TAH system since approval.

Methods:

Adult patients listed for heart transplantation in the UNOS system between 2004 and 2020 who underwent BTT therapy with a TAH were included in the study. Trends in utilization of TAH compared with other durable mechanical support strategies were examined. The primary outcome was 1-year survival following heart transplantation following BTT with TAH. Secondary outcomes included waitlist deterioration and risk factors for waitlist or posttransplant mortality.

Results:

During the study 433 total patients underwent TAH implant as BTT therapy; 236 (54.4%) were listed with the TAH, while the remaining patients were upgraded to TAH support while on the waitlist. Waitlist mortality was 7.4%, with 375 patients (86.6%) ultimately being transplanted. Age, cerebrovascular disease, functional status, and ventilator dependence were risk factors for waitlist mortality. One-year survival following successful BTT was 80%. Risk factors for mortality following BTT included age, body mass index, and underlying diagnosis.

Conclusions:

Patients undergoing BTT with TAH demonstrate acceptable waitlist survival and good 1-year survival. While utilization initially increased as a BTT therapy, there has been a plateau in relative utilization. Individual patient and transplantation center factors deserve further investigation to determine the ideal population for this therapy.

1 |. INTRODUCTION

Mechanical circulatory support for patients with end-stage heart failure requiring biventricular support as a bridge to transplantation (BTT) remains a management challenge. While left ventricular assist devices (LVADs) have steadily increased in utilization as a BTT therapy and clinical outcomes have continuously improved over time, biventricular support remains a generally higher-risk proposition secondary to additional adverse events and higher waitlist mortality.¹ Additionally, selecting which patients may benefit from biventricular support upfront may be challenging, and transitioning from an LVAD to some configuration of biventricular assist device (BiVAD) incurs additional risks.²

The SynCardia Total Artificial Heart (TAH) system (SynCardia Systems Inc.) is a mechanical circulatory support device capable of temporary biventricular support utilized as a BTT therapy. It remains the only complete TAH system approved for use in the United States; as opposed to BiVAD support techniques utilizing two separate pump circuits anastomosed to the native heart, the TAH system is implanted in a unique manner replacing the failed ventricles in the pericardial cavity.³ Patients initially were limited to the hospital facility after implantation, but with the advent of more portable driver systems, out-of-hospital care is now possible in certain circumstances.

Although the TAH system has been used commercially in the United States from 2004, real-world data supporting the use of the TAH as a BTT more broadly has still been lacking.⁴ A previous study by Arabia et al.⁴ of all TAH implants registered with the Interagency Registry for Mechanical Assisted Circulatory Support (INTERMACS) from June 2006 to April 2017 reported 450 implants, while the most recent INTERMACS report from the Society of Thoracic Surgeons in 2020 reports 449 TAH implants from 2010 to 2020.^4,5 Multiple single institutions and small collaborative groups have pooled data to demonstrate the safe application of the TAH technology, but it remains unknown how utilization has changed broadly in the United States, and how patients fare with TAH as a BTT treatment.^2,6–8 The United Network for Organ Sharing (UNOS) database remains the central repository tracking both the transplant listing and early transplantation outcomes of all patients undergoing orthotopic heart transplant (OHT) in the United States and collects data specifically identifying the use of the TAH system and other methods of mechanical circulatory support. The aim of this study was to leverage the UNOS database to describe the trends in utilization of the TAH system over time since regulatory approval was obtained in 2004 and characterize waitlist and posttransplant outcomes of patients undergoing BTT therapy with the TAH system.

2 |. METHODS

2.1 |. Ethics statement

This study was reviewed and approved by the Institutional Review Board of the University of Pittsburgh. It was classified as exempt from informed consent requirements.

2.2 |. Study population

Patients greater than 18 years of age listed for OHT between April 2004 and February 2020 in the UNOS registry were included in this study. Patients undergoing either TAH implantation while being listed for OHT or who had the device at the time of listing for OHT were included in the analysis. Utilization rates of durable mechanical circulatory support systems as a BTT therapy (TAH, LAVD, and BiVAD configurations) were queried and compared over time from the UNOS registry. The transition from other device support to TAH was quantified over the study period as well. Demographics and pretransplant characteristics including functional status at the time of transplant for patients undergoing BTT with TAH were queried and stratified by successful BTT versus waitlist mortality for comparison.

2.3 |. Outcomes

The primary outcomes of interest included waitlist survival to transplant and 1-year survival following transplantation utilizing the TAH system as a BTT therapy. Secondary outcomes included characterization of trends in TAH utilization since commercial approval at a BTT therapy, identification of preoperative risk factors for waitlist mortality and deterioration, and preoperative risk factors for mortality following successful BTT with TAH. Complications following OHT including stroke, pacemaker utilization, and rejection within the first year following OHT were also reported.

2.4 |. Data analysis

Device utilization rates were characterized by frequency tables and figures demonstrating absolute changes over time. Baseline clinical characteristics at the time of listing for OHT were identified and stratified by patients successfully achieving transplant and those deteriorating on the waitlist. All values in the study were reported as N (%) for categorical variables and mean (standard deviation) for continuous variables. χ² Analysis was performed for categorical comparisons, with Fischer’s exact test being used when cell values were <5; Student’s t-test was utilized for continuous variable comparison. Kaplan–Meier survival curves were constructed to display waitlist and posttransplant 1-year survival. Cox proportional hazard analysis was conducted to create models for both waitlist mortality and 1-year mortality following OHT with the TAH system; all listed baseline characteristics were included in the initial models with backward stepwise analysis arriving at the final models reported as hazard ratio and 95% confidence intervals. A p < .05 was considered significant throughout the study. Stata version 15 (StataCorp LP) was utilized for statistical analysis.

3 |. RESULTS

3.1 |. Trends in TAH utilization

TAH was utilized in 433 adult patients listed for heart transplantation in the UNOS database during the study period. The TAH was present at the time of initial listing for heart transplantation in 236 (54.5%) patients, while the remainder of patients were initially listed with either no support or an alternative mechanical support system (Table 1). The overall proportion of patients utilizing the TAH as BTT therapy has remained relatively small over time among patients who are listed for heart transplantation (Figure 1A) and who undergo successful heart transplantation (Figure 1B). The total number of patients annually who underwent heart transplantation while bridged with a TAH system peaked at 49 in 2013 but has decreased more recently (Table S1). During the study period, of the 7570 patients listed for heart transplant with an LVAD in place, 14 (0.2%) were transitioned to TAH support while listed as an upgrade in support, while 79 (1.0%) were transitioned to durable BiVAD support over the same period. Only four patients initially listed with a BiVAD in place were upgraded to a TAH before transplantation.

TABLE 1.

Initial mechanical support strategy of those undergoing cardiac transplantation with the total artificial heart system as a bridge to transplant

Mechanical support at listing	N (%)
Total artificial heart	236 (54.5%)
None	180 (41.6%)
Left ventricular support device	14 (3.2%)
Biventricular support device	2 (0.4%)
Unknown	1 (0.2%)

FIGURE 1.

(A) Relative utilization of durable mechanical circulatory support strategies in patients listed for cardiac transplantation over time. (B) Relative utilization of durable mechanical circulatory support strategies in patients who successfully underwent cardiac transplantation over time

3.2 |. Waitlist outcomes following TAH

Baseline characteristics of patients listed for transplant with a TAH implanted were stratified by those undergoing successful transplant and those experiencing death or waitlist deterioration (Table 2). Patients who were successfully transplanted were less commonly on the ventilator at time of listing, had better functional status, and spent longer time on the waitlist than those who died or deteriorated on the waitlist. A total of 32 (7.4%) patients died while on the waitlist (Figure 2), while 16 (3.7%) patients deteriorated and were made inactive while awaiting transplant (Table 3). The most common cause of death while on the waitlist was multisystem organ failure, followed by stroke (Table 4). Age, cerebrovascular disease, ventilator dependence at listing, and functional status were found to be significantly associated with death or deterioration in multivariable modeling (Table 5). The mean time on the waitlist before the UNOS allocation changes in 2018 was 4.0 [1.7–8.2] months compared to 2.8 [1.0–6.8] months following the allocation change (p = .03).

TABLE 2.

Baseline clinical characteristics at the time of listing for transplant

	All patients (N = 433)	Transplanted (N = 375)	Death or deterioration on waitlist (N = 48)	p Value
Age	48.22 (13.22)	47.94 (13.38)	50.63 (12.60)	.188
Sex				.365
Female	62 (14.32)	52 (13.87)	9 (18.75)
Male	371 (85.68)	323 (86.13)	39 (81.25)
Race/ethnicity				.643
White	278 (64.50)	241 (64.44)	32 (68.09)
Black	101 (23.43)	85 (22.73)	12 (25.53)
Hispanic	32 (7.42)	30 (8.02)	2 (4.26)
Other	20 (4.64)	18 (4.81)	1 (2.13)
BMI	28.20 (5.43)	28.11 (5.35)	28.49 (6.05)	.647
History of cigarette use	183 (42.76)	159 (42.97)	19 (39.58)	.655
Creatinine	1.67 (1.22)	1.64 (1.21)	1.92 (1.29)	.134
Diabetes	95 (21.94)	83 (22.13)	8 (16.67)	.385
Cerebrovascular disease	32 (7.39)	26 (6.93)	6 (12.50)	.368
Prior malignancy	20 (4.62)	18 (4.80)	2 (4.17)	.807
Diagnosis				.627
Nonischemic cardiomyopathy	215 (49.65)	188 (50.13)	22 (45.83)
Ischemic cardiomyopathy	109 (25.17)	94 (25.07)	13 (27.08)
Congenital	8 (1.85)	8 (2.13)	0
Restrictive	23 (5.31)	21 (5.60)	1 (2.08)
Valvular	15 (3.46)	13 (3.47)	2 (4.17)
Hypertrophic cardiomyopathy	18 (4.16)	15 (4.00)	2 (4.17)
Other/unknown	45 (10.39)	36 (9.60)	8 (16.67)
Pulmonary capillary wedge pressure (mmHg)	25 [18–30]	25 [18–30]	22 [17–29]	.518
Mean pulmonary artery systolic pressure (mmHg)	33 [26–41]	32 [26–41]	36 [27–43]	.245
Ventilator at listing	29 (6.70)	20 (5.33)	8 (16.67)	.003
Functional status				.031
Independent	15 (3.59)	15 (4.14)	0
Partially dependent	128 (30.62)	116 (32.04)	8 (17.39)
Completely dependent	275 (65.79)	231 (63.81)	38 (82.61)
Days on the waitlist	110 [46–234]	119 [51–237]	61 [24–159]	.004

Abbreviation: BMI, body mass index.

FIGURE 2.

Waitlist survival for patients undergoing total artificial heart implant as a bridge to cardiac transplantation

TABLE 3.

Waitlist outcomes for patients bridged to transplant with the total artificial heart

Waitlist outcome	N (%)
Transplanted	372 (85.91)
Transferred to another center	2 (0.46)
Died	32 (7.39)
Other	3 (0.69)
Deteriorated, too sick	16 (3.70)
Transplanted at another center	1 (0.23)
Died during transplant	3 (0.69)
Still on the waitlist	4 (0.92)

TABLE 4.

Cause of death on the waitlist (n = 32)

Cause of death	N (%)
Unknown	3 (9.38)
Other	3 (9.38)
Bacteremia	3 (9.38)
Pneumonia	1 (3.12)
Aspergillosis	1 (3.12)
Cardiac arrest	1 (3.12)
Other cardiovascular	1 (3.12)
Stroke	4 (12.50)
Hemorrhagic stroke	2 (6.25)
Other cerebrovascular	1 (3.12)
Liver failure	1 (3.12)
Multiple organ failure	11 (34.38)

TABLE 5.

Multivariable Cox proportional hazard model for failure (death or deterioration) on waitlist bridged with the total artificial heart

	Hazard ratio	95% CI	p Value
Age (for every year)	1.06	1.02–1.11	.004
Cerebrovascular disease	3.56	1.03–12.30	.045
Ventilator at listing	3.94	1.27–12.25	.018
Functional status
Independent	Reference	Reference	Reference
Partially dependent	0.61	0.18–2.09	.434
Completely dependent	5.17	1.50–17.78	.009

Abbreviation: CI, confidence interval.

3.3 |. Transplant outcomes following bridge with TAH

Of the 433 patients on the waitlist with TAH as BTT therapy, 375 (86.6%) underwent OHT. Posttransplant survival for patients successfully bridged with a TAH system at 30 days was 90.9% and at 1 year was 80% (Figure 3). Posttransplant complications included stroke in 31 patients (8.3%), pacemaker implantation in 11 patients (2.9%), and renal failure requiring dialysis in 111 (29.6%). There were 260 (69.3%) atrial versus 109 (29.1%) bicaval anastomosis patients with the remaining 9 being heterotopic transplants. The pacemaker rates were 4 (1.5%) for atrial versus 6 (5.5%) for bicaval anastomotic techniques (p = .003). Rejection was documented to occur in 36 patients of the 271 BTT patients with 1-year data available for review at the time of the study (13.3%). Increasing age, increasing body mass index (BMI), and diagnosis leading to heart failure were found to be associated with increasing mortality in multivariable modeling (Table 6).

FIGURE 3.

Posttransplant survival for patients undergoing total artificial heart implant as a bridge to cardiac transplantation

TABLE 6.

Multivariable Cox proportional hazard model for death following orthotopic heart transplantation bridged with the total artificial heart

	Hazard ratio	95% CI	p Value
Age (for every year)	1.04	1.02–1.07	.001
BMI	1.07	1.03–1.13	.003
Diagnosis
Nonischemic cardiomyopathy	Reference	Reference	Reference
Ischemic cardiomyopathy	1.48	0.84–2.61	.173
Congenital	8.51	1.88–38.46	.005
Restrictive	1.31	0.46–3.75	.617
Valvular	1.89	0.57–6.24	.299
Hypertrophic cardiomyopathy	1.37	0.41–4.51	.608
Other/unknown	2.48	1.14–5.41	.022

Abbreviation: BMI, body mass index; CI, confidence interval.

4 |. CONCLUSIONS

4.1 |. Study implications

The current study evaluates the trends in utilization of the TAH as a BTT therapy since commercial approval of the device utilizing the UNOS registry. We found that 54% of patients were listed for heart transplant with a TAH implanted, while the remainder were converted to TAH support while on the list as an upgrade in support. Relative use of the TAH system compared to other LVAD and BiVAD configurations in BTT therapy has remained small over time despite an increased utilization in raw numbers during the study period. Impressively, 86% of this cohort achieved cardiac transplantation with a 1-year posttransplant survival of 80%. Age, ventilator usage, cerebrovascular disease, and functional status were found to be associated with waitlist mortality; age, BMI, and cardiac diagnosis were found to be associated with mortality following transplantation in this cohort.

The SynCardia TAH system has been approved by the Food and Drug Administration for BTT therapy in the United States since 2004. As with most new innovative devices, utilization was initially low and limited to a few centers, but over the following 10 years utilization increased and spread to other centers performing OHT.⁶ Initially prospective studies identified a population of patients that benefited from BTT with TAH including patients with Class IV biventricular heart failure at risk for imminent death without significant renal or liver dysfunction and a prespecified pulmonary vascular resistance cut-off.⁹ However, the implantation technique of the TAH is unique and has previously been identified as a limiting factor for more widespread adoption.³ Additionally, identifying the ideal patient for TAH therapy as opposed to LVAD with medical management versus a durable BiVAD configuration has been somewhat elusive over time in biventricular heart failure.² While initially approved for BTT, the TAH is currently undergoing evaluation as destination therapy in adults with biventricular heart failure (Clinicaltrials.gov Identifier: NCT02232659).

The trend in the utilization of the TAH system has plateaued in recent years as demonstrated by our study. After initially steadily increasing since introduction in 2004 through 2016, a drop-off occurred in the last 3 years of the study. The reasoning for this is unclear given the limitations of the UNOS database. While donor organ allocation status changes in 2018 have affected strategies of bridging patients to transplant with both durable and temporary mechanical circulatory support, the TAH continues to have Status 2 allocation in the new scheme so this should not have necessarily decreased transplant offers.^10,11 What the change in allocation strategy did affect as demonstrated by our study was the waitlist time for TAH BTT therapy, which was significantly shorter in the TAH population after 2018. The significance of this difference will require further research. There has not been an appreciable shift to BiVAD support either, and review of the literature does not support any evidence favoring TAH or BiVAD consistently for biventricular support.¹² However, with the improvement in treatment strategies for right heart failure in the setting of LVAD support, this may partially explain the decreased reliance on both BiVAD and TAH support at a BTT strategy in patients with bi-ventricular cardiac failure.¹ Further research is indicated to better understand these trends.

The TAH has been shown to be effective at BTT via select small institutional studies, but specific risk factors for waitlist mortality were importantly identified in these reports. Copeland and colleagues found in a series of over 100 TAH implants in critically ill patients that 68% of patients were successfully bridged to transplantation and they enjoyed a 77% 1-year survival. Multiple organ failure was a leading cause of death withTAH implantation, with increased age and ventilator utilization being high-risk predictors as we found in our study.⁶ Kirsch et al.² echoed these findings regarding risk factors in their successful series utilizing TAH as a BTT. When looking at pooled interagency registry data, increased age and need for dialysis/renal failure were risk factors for death in patients implanted with the TAH.⁴ Again, early infection and respiratory failure were found to be primary early pathways to death in this cohort. Renal failure specifically should be avoided in this patient population when possible secondary to these findings, with specific risk factors among TAH patients having been identified.¹³ With that being said, there can be renal recovery associated with TAH implantation in some settings, and utilization of the TAH system as a bridge to combined heart and kidney transplantation in selected patients has been successfully reported.¹⁴

Perhaps the largest series to date is the recent multicenter (six sites) series of 217 patients in North America undergoing BTT with TAH.⁷ They were able to successfully bridge 63.5% of patients to transplant over the course of the study, and found the primary risk factor driving mortality on device therapy was once again patient age. Overall cohort survival was 75% at 1 year, with posttransplant survival 88% at 1-year follow-up. While most studies agree that patients undergoing heart transplants following BTT with a TAH do generally well (including our current data), the length of time on the device does correlate with posttransplant complications including pericarditis, infectious issues, and bleeding complications suggesting prolonged support duration is not without risks.⁸ A previous study of the UNOS database comparing TAH to BiVAD support in an earlier era (2004–2014) demonstrates similar survival and morbidity between biventricular failure patients undergoing BTT with either device strategy with 78% versus 83% 1-year posttransplant survival for each device, respectively.¹⁵

Identifying the proper patient for TAH BTT therapy seems to be the single largest outstanding factor after completion of our study and review of the current data.^1,4 One particular population that may benefit is those undergoing retransplantation for rejection, thus allowing the explant of the offending organ while awaiting a new organ.¹⁶ Evolving risk-score development utilizing preoperative data to predict posttransplant survival may be of value to biventricular failure patients in the future.¹⁷ This may help time appropriately the implantation of this increased level of support, minimizing time on the device while optimally protecting end-organ function which is key. Utilizing similar predictive modeling to identify and mitigate failure on device therapy will also contribute to optimizing outcomes in this critically ill heart failure patient population.¹⁸

4.2 |. Limitations

This study is a retrospective analysis of the UNOS database and is therefore subject to the biases inherent in such a study. As the UNOS database only captures patients listed for heart transplantation in the United States, outcomes of patients who underwent implantation of a TAH system but were never subsequently listed for heart transplantation are not captured in this study and may influence results presented here. Biventricular support strategies and timing of biventricular support vary significantly among centers, and the inability to capture these trends/data limits the generalizability of these results.

5 |. CONCLUSIONS

Patients undergoing OHT using the SynCardia TAH as BTT support for biventricular heart failure demonstrate acceptable waitlist survival and good 1-year survival following OHT. While over the first 10 years utilization of the TAH system increased as a BTT therapy, over the last 4 years there has been a dip and plateau in relative utilization compared to LVAD and even BiVAD strategies. Individual patient and transplantation center factors that play a role in these observed trends and outcomes deserve further investigation to determine the ideal population for this unique therapeutic option.

Abbreviations:

BiVAD

bi-ventricular assist device

BTT

bridge to transplantation

LVAD

left ventricular assist device

OHT

orthotopic heart transplantation

TAH

total artificial heart