Outcomes in Heart Transplant Recipients by Bridge to Transplant Strategy When Using the SherpaPak Cardiac Transport System

Abstract

The last several years have seen a rise in use of mechanical circulatory support (MCS) to bridge heart transplant recipients. A controlled hypothermic organ preservation system, the SherpaPak Cardiac Transport System (SCTS), was introduced in 2018 and has grown in utilization with reports of improved posttransplant outcomes. The Global Utilization And Registry Database for Improved heArt preservatioN (GUARDIAN)-Heart registry is an international, multicenter registry assessing outcomes after transplant using the SCTS. This analysis examines outcomes in recipients bridged with various MCS devices in the GUARDIAN-Heart Registry. A total of 422 recipients with donor hearts transported using SCTS were included and identified. Durable ventricular assist devices (VADs) were used exclusively in 179 recipients, temporary VADs or intra-aortic balloon pump (IABP) in 197, and extracorporeal membrane oxygenation (ECMO) in 14 recipients. Average ischemic times were over 3.5 hours in all cohorts. Severe primary graft dysfunction (PGD) posttransplant increased across groups (4.5% VAD, 5.1% temporary support, 21.4% ECMO), whereas intensive care unit (ICU) length of stay (18.2 days) and total hospital stay (39.4 days) was longer in the ECMO cohort than the VAD and IABP groups. A comparison of outcomes of MCS bridging in SCTS versus traditional ice revealed significantly lower rates of both moderate/severe right ventricular (RV) dysfunction and severe PGD in the SCTS cohort; however, upon propensity matching only the reductions in moderate/severe RV dysfunction were statistically significant. Use of SCTS in transplant recipients with various bridging strategies results in excellent outcomes.

In 2018, the United Network for Organ Sharing (UNOS) implemented a new organ allocation system for heart transplants designed to identify high-risk patients more accurately, with the goal of reducing wait list time for urgent cases and consequently, wait list mortality.¹ In the revised allocation policy, transplant candidates are prioritized by disease severity and type of mechanical circulatory support (MCS) in addition to duration of support.¹ In this allocation system, priority is given to patients currently using extracorporeal membrane oxygenation (ECMO,) nondischargeable biventricular ventricular assist devices (VADs), temporary, nondischargeable left VAD (LVAD), an intra-aortic balloon pump (IABP), or MCS support with significant device malfunction.¹ Accordingly, the proportion of heart transplant recipients with prior MCS utilization has increased from 10–25.4% to 41–42.6% after the 2018 changes to the UNOS heart allocation policy,^2,3 and patients admitted to US transplant centers since 2018 are about two times more likely to be treated using temporary MCS than those admitted before 2018.² Simultaneously, the current allocation policy provides patients in the highest two priority levels initial access to donor organs within 500 miles, and subsequently in expanding 500 mile radii increasing the average donor distance and therefore ischemic time donor hearts are subject to in cold ischemic storage during transit.⁴

Initial reports raised concern that this combination of increased allograft ischemic time and frequency of MCS bridging would result in more postoperative complications in heart transplant patients.⁵ A nationwide retrospective analysis including 8,902 heart transplant recipients found that, after the 2018 UNOS policy change, 6.1% of patients utilized pretransplant ECMO compared to 1.2% before allocation policy revision.⁶ These findings are consistent with increased use of ECMO as a bridge to heart transplant or long-term LVAD in patients with cardiogenic shock⁷ and increased risk of early graft failure and posttransplant mortality.^8–10 Similarly, LVAD patients with right heart failure or more than 1 year of LVAD use have an increased risk of primary graft dysfunction (PGD).^11,12

The risk of PGD in transplant recipients may be compounded by a longer allograft ischemic time, a known risk factor for PGD.^13,14 For example, the average ischemic time in patients receiving pretransplant ECMO has increased from 2.8 to 3.4 hours after the UNOS policy change.⁶ Trends in IABP use since 2018 are similar, with one single-center study showing an increased proportion of heart transplant recipients using pretransplant IABP (3–45%) and a significantly longer ischemic time (117 to 177 minutes).⁵ Thus, strategies to mitigate this potential increased risk of PGD and mortality such as organ preservation technologies that minimize the risk of damage during cold ischemic storage may improve clinical outcomes in the growing population of patients receiving pretransplant MCS.

The International Society of Heart and Lung Transplant (ISHLT) consensus recommendation for preservation of donor hearts recommends packaging organs in preservative solutions at 4°C in three plastic bags, preventing direct contact with ice to minimize freezing injury that may lead to PGD.¹⁵ With these principles in mind, the Food and Drug Administration (FDA)-cleared and Conformitè Europëenne (CE)-marked Paragonix SherpaPak Cardiac Transport System (SCTS) was designed to maintain a stable 4–8°C environment for donor hearts while the organ remains suspended in cold storage fluid without contacting the surfaces of the container.¹⁶ The system is single use, requires no external power source, and allows for continuous remote temperature data monitoring. Recent publication of the Global Utilization And Registry Database for Improved heArt preservatioN (GUARDIAN-Heart) registry comparing posttransplant outcomes in adult patients after ischemic storage on ice (ICE) or SCTS storage have reported clinically significant benefits after SCTS utilization in an international, multicenter cohort.^17,18 In this study, we performed a retrospective subgroup analysis of the GUARDIAN-Heart registry in recipients bridged with various MCS devices. We compared clinical outcomes in transplant recipients using the SCTS system with durable and temporary VADs, IABP, and ECMO before transplantation.

Materials and Methods

We performed analysis of the GUARDIAN-Heart registry (NCT04141605) a database that contains retrospectively collected medical and demographic data on heart transplant patients utilizing standard ice cooler transport or SCTS organ preservation.^17,18 The primary objective of this study was an evaluation of the rate of PGD and right heart dysfunction postcardiac transplantation in MCS-bridged patients receiving hearts preserved in either the SherpaPak CTS or traditional ice storage. Patients were enrolled retrospectively, with all consecutive SherpaPak cases enrolled along with consecutive ice cases. Ice controls were enrolled consecutively in reverse, allowing up to 3 years prior enrollment of ice controls. The GUARDIAN-Heart registry includes up to 1 year follow-up data from 21 transplant centers, 4 in the EU (Austria, Spain, and UK) and 17 in the US (see Supplemental Digital Content Table S1, Supplemental Digital Content, http://links.lww.com/ASAIO/B199). Subjects were transplanted between October 2015 and December 2022. Each participating institution obtained informed consent where required and the study was approved by their respective institutional review or ethics boards. The GUARDIAN-Heart registry is funded and administered by Paragonix Technologies (Cambridge, MA).

We identified and analyzed patients in the GUARDIAN-Heart registry who were bridged with MCS utilization and who received a heart transplant after SCTS organ preservation. For this analysis, we included all SCTS transplant recipients with durable VAD, temporary VAD, IABP, or ECMO use before transplantation (n = 422). We compared data from GUARDIAN Registry cases using overall ice cooler (n = 354) to overall SCTS (n = 422) MCS-bridged recipients. We then further evaluated the outcomes of various bridging strategies within the SCTS subgroup by the unique cohorts of durable VAD, temporary VAD or IABP, and ECMO. The temporary VAD and IABP were combined into one cohort as temporary MCS. Baseline comparisons included donor age, donor body mass index (BMI), donor left ventricular ejection fraction (LVEF), recipient age, recipient BMI, waitlist days, recipient baseline LVEF, implantable VAD, temporary IABP, temporary ECMO/VAD, gender mismatch, predicted heart mass (PHM) mismatch, most undersized (<15%), distance to organ, total ischemic time, and heart allocation policy era (% post 2018 UNOS change). To assess posttransplant outcomes we compared the following: posttransplant MCS, new posttransplant IABP, new posttransplant ECMO/VAD, cardioversion rate, PGD rate, severe PGD rate, RV function, 24 hour LVEF, 24 hour central venous pressure (CVP), 24 hour inotrope score, posttransplant days of inotrope use, intensive care unit (ICU) length of stay (LOS), hospital LOS, LVEF at discharge, CVP at discharge, in-hospital survival, 30 day survival, and 1 year survival. Severe PGD was defined as per the 2014 ISHLT consensus statement¹⁹ as requiring MCS (excluding IABP) within 24 hours posttransplant. We calculated an overall index for mortality prediction after cardiac transplant (IMPACT) score as described by Weiss et al.²⁰ The variables used to calculate the IMPACT scores are presented in Supplemental Table S3, Supplemental Digital Content, http://links.lww.com/ASAIO/B201.

We calculated a modified donor risk score (DRS), based on the score developed by Smits et al.²¹ The registry lacked detailed doses of inotropes, therefore the scores were calculated without those variables included. The variables included were age, cause of death, donor history of either malignancy, sepsis, drug abuse, positive virology, donor history of hypertension, cardiac arrest, echocardiographic and coronary angiogram findings, and serum sodium value. The total scores assigned to each variable were as described previously.²¹

Propensity matching was performed to balance the populations by pairing cohorts based on potential sources of variance including site, ischemic time, era (pre or post the October 2018 UNOS organ allocation change), donor age, and MCS type (for multiple types, the primacy matching rules included, in order, ECMO, temporary VAD, IABP, durable VAD type) used as potential confounders. The analysis utilized 1:1 matching through a random selection process, with a matching propensity score allowable difference of 0.025 was used.

Forest plots of relative odds for either severe PGD or preservation of right ventricular function by univariate logistic regression were performed to assess the risks of various bridging strategies. This analysis was performed on the overall population of MCS-bridged recipients (n = 776), where an association analysis was performed to determine whether select variables were independently associated with maintenance of RV function or risk of severe PGD.

Statistical comparative analyses were calculated using R (version 4.2.2; The R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were reported as mean ± standard deviation and analyzed by unpaired t-test with Welch’s correction, and categorical variables were reported as counts and percentages and analyzed using Fisher’s exact test. Multiple group comparisons were performed using the Chi-square test. Kaplan–Meier survival probability was analyzed by means of logistic regression. A p value of less than 0.05 was considered statistically significant.

Results

A total of 422 transplant recipients with donor hearts transported using the SCTS were included in this analysis. A comparison was made to 354 transplant patients bridged with MCS devices receiving donor hearts transported using a standard ice cooler. The distribution of VADs can be found in Supplemental Table S2, Supplemental Digital Content, http://links.lww.com/ASAIO/B200. A patient could have more than one MCS bridge before transplant, so the groups in the overall analysis are not mutually exclusive. The comparison baseline characteristics of the MCS-bridged SCTS and ICE coolers are presented in Table 1. The donor characteristics were similar with the exception of SCTS hearts traveling a greater distance for procurement than hearts on ICE (410.0 vs. 262.1 miles, p < 0.001) with a longer average total ischemic time (3.6 vs. 3.2 hours, p < 0.001). Donor risk profiles compared using a modified DRS were also similar, and detailed variables are presented in Supplemental Table S4, Supplemental Digital Content, http://links.lww.com/ASAIO/B202. More heart transplants utilizing the SCTS versus ICE were performed after the UNOS 2018 allocation changes (98.3% vs. 74.9%, p < 0.001). Recipient characteristics were similar, with the exception of more durable implantable VADs in the ICE cohort (62.1% vs. 45.7%, p < 0.001), and more IABP and temporary VADs used in the SCTS cohort (27.1% vs. 43.1%, p < 0.001, and 7.6% vs. 12.8%, p = 0.025, ICE versus SCTS, respectively). The IMPACT Score was higher in the SCTS cohort compared to ICE, 8.4 vs. 7.6, p = 0.036.

Table 1.

Baseline Characteristics of MCS-Bridged Heart Transplant Patients From the GUARDIAN-Heart Registry Utilizing Standard Cold Storage on Ice or the SCTS for Controlled Hypothermic Organ Preservation

	ICE (N = 354)	SCTS (N = 422)	p Value
Donor characteristics
Age (years)	30.3 ± 13.1	31.3 ± 11.8	0.27
BMI (kg/m2)	27.7 ± 6.7	27.3 ± 6.6	0.50
LVEF (%)	60.6 ± 6.6	61.2 ± 7.1	0.26
Distance to organ (miles)	262.1 ± 242.6	410.0 ± 323.7	<0.001
Total ischemic time (hours)	3.2 ± 0.84	3.6 ± 0.85	<0.001
F/M mismatch	49/353 (13.9%)	57/420 (13.6%)	0.92
PHM mismatch	0.05 ± 0.20	0.04 ± 0.18	0.42
Most undersized (<−15%)	39/337 (11.6%)	45/408 (11.0%)	0.82
Era (% post change)	265/354 (74.9%)	415/422 (98.3%)	<0.001
Modified donor risk score	20.5 ± 8.7	20.5 ± 8.6	0.97
Recipient characteristics
Age (years)	49.7 ± 18.4	51.0 ± 17.6	0.34
BMI (kg/m2)	27.2 ± 5.4	27.6 ± 5.3	0.30
LVEF at baseline (%)	19.1 ± 9.4	19.0 ± 9.6	0.83
Implantable VAD	220/354 (62.1%)	193/422 (45.7%)	<0.001
Temporary IABP	96/354 (27.1%)	182/422 (43.1%)	<0.001
Temporary VAD	27/354 (7.6%)	54/422 (12.8%)	0.025
Temporary ECMO	42/354 (11.9%)	40/422 (9.5%)	0.29
IMPACT score	7.6 ± 5.5	8.4 ± 5.3	0.036

Numbers in bold indicate statistically significant values.

Numbers in bold indicate statitistically significant values. BMI, body mass index; ECMO, extracorporeal membrane oxygenation; F/M, female to male; GUARDIAN, Global Utilization And Registry Database for Improved heArt preservatioN; IABP, intra-aortic balloon pump; ICE, ice cold storage; IMPACT, index for mortality prediction after cardiac transplant; LVEF, left ventricular ejection fraction; MCS, mechanical circulatory support; PHM, predicted heart mass; SCTS, SherpaPak Cardiac Transport System; VAD, ventricular assist device.

Posttransplant outcomes are presented in Table 2. Patients receiving hearts preserved using the SCTS showed a statistically significant reduction in posttransplant rates of severe PGD compared to those with hearts transported on ICE (10.2% vs. 6.2%, p = 0.046) as well as significantly less moderate to severe right heart dysfunction (31.3% vs. 21.4%, p = 0.004). Rates of overall mild, moderate, or severe PGD trended toward a 39% reduction in the SCTS cohort (p = 0.089). There was also a trend to spend an average of 4 fewer hospital days posttransplant in those patients receiving a SCTS-preserved donor heart compared to ICE (25.4 vs. 29.6 days, p = 0.079). Survival was similar posttransplant through 1 year (Figure 1).

Table 2.

Post-Transplant Outcomes in MCS-Bridged Heart Transplant Patients From the GUARDIAN-Heart Registry Utilizing Standard Cold Storage on Ice or the SCTS for Controlled Hypothermic Organ Preservation

	ICE (N = 354)	SCTS (N = 422)	p Value
All post-Tx MCS*	98/354 (27.7%)	104/422 (24.6%)	0.37
New IABP post-Tx	45/354 (12.7%)	42/422 (10.0%)	0.25
New ECMO/VAD post-Tx†	42/354 (11.9%)	33/422 (7.8%)	0.067
PGD (mild, moderate, severe)	72/354 (20.3%)	65/421 (15.4%)	0.089
PGD severe‡	36/354 (10.2%)	26/421 (6.2%)	0.046
RV normal (not diminished)	135/300 (45.0%)	204/379 (53.8%)	0.025
RVF moderately to severely diminished	94/300 (31.3%)	81/379 (21.4%)	0.004
Cardioversion	51/351 (14.5%)	42/421 (10.0%)	0.059
LVEF (%, 24 hours)	54.8 ± 11.9	56.0 ± 12.6	0.24
CVP (24 hours)	11.2 ± 4.0	11.2 ± 3.8	0.97
Inotrope score (24 hours)	15.8 ± 15.6	14.5 ± 9.4	0.19
Inotrope use post-Tx (days)	9.9 ± 13.5	9.5 ± 8.3	0.60
ICU LOS (days)	12.5 ± 23.2	11.6 ± 13.8	0.56
Hospital LOS (days)	29.62 ± 39.9	25.4 ± 20.4	0.079
30 day survival	345/354 (97.5%)	414/422 (98.1%)	0.63
In-hospital survival	339/354 (95.8%)	409/422 (96.9%)	0.44
1 year survival	321/351 (91.5%)	359/384 (93.5%)	0.33

Numbers in bold indicate statitistically significant values.

^*All MCS categories, including MCS continued from pretransplant throughout the posttransplant period.

^†New ECMO/VAD post-Tx includes MCS use from transplant through discharge (both primary [<24 hours] and secondary [>24 hours] graft dysfunction).

^‡PGD severe defined according to ISHLT guidelines as the use of MCS (excluding balloon pump) within 24 hours posttransplant.

CVP, central venous pressure; ECMO, extracorporeal membrane oxygenation; GUARDIAN, Global Utilization And Registry Database for Improved heArt preservatioN; IABP, intra-aortic balloon pump; ICE, ice cold storage; ICU, intensive care unit; ISHLT, International Society for Heart and Lung Transplantation; LOS, length of stay; MCS, mechanical circulatory support; PGD, primary graft dysfunction; RVF, right ventricular function; SCTS, SherpaPak Cardiac Transport System; Tx, transplant; VAD, ventricular assist device.

Figure 1.

Kaplan–Meier analysis of 1 year survival probability of the two study cohorts of the full US adult population. The 95% confidence intervals of the curves are indicated by the shaded areas and the number at risk is presented below the graph. ICE, ice cold storage; SCTS, SherpaPak Cardiac Transport System.

Results of the propensity-matched analysis are presented in Tables 3 and 4. In Table 3, the baseline comparison of the ICE control and the SCTS cohorts following matching on-site, era, ischemic time, donor age, and MCS type resulted in a similar set of study cohorts. Posttransplant, the SCTS cohort had significantly reduced rates of overall posttransplant MCS use (31.9% vs. 22.7%, p = 0.040), a trend to reduced posttransplant rates of severe PGD (10.2% vs. 5.1%, p = 0.069), and significantly reduced rates of moderate/severe right heart dysfunction (31.4% vs. 18.6%, p = 0.004). Survival was similar through 1 year (Table 4 and Supplemental Figure S1, Supplemental Digital Content, http://links.lww.com/ASAIO/B198).

Table 3.

Baseline Characteristics of the Propensity-Matched MCS-Bridged Heart Transplant Patients From the GUARDIAN-Heart Registry Utilizing Standard Cold Storage on Ice or the SCTS for Controlled Hypothermic Organ Preservation

	ICE (N = 216)	SCTS (N = 216)	p Value
Donor characteristics
Age (years)	31.3 ± 13.5	30.4 ± 12.7	0.45
BMI (kg/m2)	27.4 ± 6.2	27.0 ± 7.1	0.51
LVEF (%)	60.8 ± 6.9	61.1 ± 7.5	0.60
Distance to organ (miles)	285.4 ± 242.7	320.8 ± 276.3	0.16
Total ischemic time (hours)	3.4 ± 0.80	3.4 ± 0.78	0.96
F/M mismatch	25/216 (11.6%)	39/216 (18.1%)	0.08
PHM mismatch	0.06 ± 0.20	0.34 ± 0.20	0.14
Most undersized (<−15%)	23/209 (11.0%)	31/204 (15.2%)	0.24
Era (% post change)	209/216 (96.8%)	209/216 (96.8%)	>0.99
Modified donor risk score	20.3 ± 8.6	19.8 ± 8.7	0.54
Recipient characteristics
Age (years)	50.6 ± 17.7	49.2 ± 19.4	0.44
BMI (kg/m2)	26.8 ± 5.6	27.6 ± 5.8	017
LVEF at baseline (%)	19.9 ± 10.5	19.7 ± 10.0	0.81
Implantable VAD	110/216 (50.9%)	107/216 (49.5%)	0.85
Temporary IABP	84/216 (38.9%)	86/216 (39.8%)	0.92
Temporary VAD	21/216 (9.7%)	21/216 (9.7%)	>0.99
Temporary ECMO	27/216 (12.5%)	25/216 (11.6%)	0.88
IMPACT score	8.4 ± 6.0	8.4 ± 5.1	0.98

Matched pairs variables included ischemic time, site, era, donor age, ischemic time, and MCS bridge type (for multiple types, the primacy matching rules included, in order, ECMO, temporary VAD, IABP, durable VAD).

BMI, body mass index; ECMO, extracorporeal membrane oxygenation; F/M, female to male; GUARDIAN, Global Utilization And Registry Database for Improved heArt preservatioN; IABP, intra-aortic balloon pump; ICE, ice cold storage; IMPACT, index for mortality prediction after cardiac transplant; LVEF, left ventricular ejection fraction; MCS, mechanical circulatory support; PHM, predicted heart mass; SCTS, SherpaPak Cardiac Transport System; VAD, ventricular assist device.

Table 4.

Posttransplant Outcomes in the Propensity-Matched MCS-Bridged Heart Transplant Patients From the GUARDIAN-Heart Registry Utilizing Standard Cold Storage on Ice or the SCTS for Controlled Hypothermic Organ Preservation

	ICE (N = 216)	SCTS (N = 216)	p Value
All post-Tx MCS*	69/216 (31.9%)	49/216 (22.7%)	0.040
New IABP post-Tx	27/216 (12.5%)	18/216 (8.3%)	0.21
New ECMO/VAD post-Tx†	26/216 (12.0%)	13/216 (6.0%)	0.043
PGD (mild, moderate, severe)	44/216 (20.4%)	28/216 (13.0%)	0.052
PGD severe‡	22/216 (10.2%)	11/216 (5.1%)	0.069
RV normal (not diminished)	77/185 (41.6%)	101/194 (52.1%)	0.050
RVF moderately to severely diminished	58/185 (31.4%)	36/194 (18.6%)	0.004
Cardioversion	27/214 (12.6%)	18/216 (8.3%)	0.16
LVEF (%, 24 hours)	55.3 ± 11.2	56.8 ± 11.2	0.21
CVP (24 hours)	11.0 ± 4.1	11.3 ± 3.9	0.47
Inotrope score (24 hours)	15.4 ± 14.2	12.8 ± 7.9	0.021
Inotrope use post-Tx (days)	10.0 ± 12.2	9.4 ± 7.2	0.53
ICU LOS (days)	12.0 ± 20.2	10.6 ± 12.9	0.40
Hospital LOS (days)	29.5 ± 35.7	25.0 ± 20.6	0.11
30 day survival	213/216 (98.6%)	211/216 (97.7%)	0.72
In-hospital survival	209/216 (96.8%)	208/216 (96.3%)	>0.99
1 year survival	198/214 (92.5%)	177/190 (93.2%)	0.85

Numbers in bold indicate statitistically significant values.

^*All MCS categories, including MCS continued from pretransplant throughout the posttransplant period.

^†New ECMO/VAD post-Tx includes MCS use from transplant through discharge (both primary [<24 hours] and secondary [>24 hours] graft dysfunction).

^‡PGD severe defined according to ISHLT guidelines as the use of MCS (excluding balloon pump) within 24 hours posttransplant.

CVP, central venous pressure; ECMO, extracorporeal membrane oxygenation; GUARDIAN, Global Utilization And Registry Database for Improved heArt preservatioN; IABP, intra-aortic balloon pump; ICE, ice cold storage; ICU, intensive care unit; ISHLT, International Society of Heart and Lung Transplant; LOS, length of stay; LVEF, left ventricular ejection fraction; MCS, mechanical circulatory support; PGD, primary graft dysfunction; RVF, right ventricular function; SCTS, SherpaPak Cardiac Transport System; Tx, transplant; VAD, ventricular assist device.

A hazard analysis was performed to assess the relative odds of severe PGD posttransplant (see Figure 2). Univariate regression showed that use of the SCTS in transplants for patients bridged to with MCS had a significantly reduced risk of severe PGD posttransplant, with an odds ratio [OR] of 0.58 compared to ICE (95% confidence interval [CI] = 0.34–0.98, p = 0.043), driven largely by a significantly reduced risk in the subgroup of patients with a durable VAD, with an OR of 0.35 (95% CI = 0.16–0.77, p = 0.009). In a hazard analysis to assess the relative odds of preservation of posttransplant RV function (Figure 3), the use of SCTS again showed a significant benefit, with an odds ratio of overall MCS bridging of 1.42 (95% CI = 1.05–1.93, p = 0.023), and an odds ratio in the temporary VAD or IABP cohort of 1.83 (95% CI = 1.13–2.96, p = 0.014).

Figure 2.

Forest plot of a univariate logistic regression analysis of relative odds of developing posttransplant severe PGD. Odds ratio and 95% CI for severe PGD for selected recipient bridging strategies. CI, confidence interval; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; ICE, ice cold storage; MCS, mechanical circulatory support; PGD, primary graft dysfunction; SCTS, SherpaPak Cardiac Transport System; VAD, ventricular assist device. *p ≤ 0.05.

Figure 3.

Forest plot of a univariate logistic regression analysis of relative odds of preservation of posttransplant RV function. Odds ratio and 95% CI for RV function preservation for selected recipient bridging strategies. CI, confidence interval; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; ICE, ice cold storage; MCS, mechanical circulatory support; RV, right ventricular; SCTS, SherpaPak Cardiac Transport System; VAD, ventricular assist device. *p ≤ 0.05.

To examine the impact of various MCS bridging strategies on heart transplant outcomes when the SCTS was used for donor heart preservation, only exclusive MCS cohorts were examined to minimize confounders. Inclusion of only exclusive MCS bridging device strategies reduced the total N from 422 to 390 (Supplemental Table S5, Supplemental Digital Content, http://links.lww.com/ASAIO/B203). This resulted in a comparison of outcomes in 179 recipients bridged with a durable VAD, 197 recipients with a temporary VAD or IABP (temporary MCS), and 14 recipients with an ECMO circuit in place at the time of transplant. The baseline characteristics of these cohorts are presented in Supplemental Table S5, Supplemental Digital Content, http://links.lww.com/ASAIO/B203. The donor baseline characteristics were similar, except there were more female donors in the cohort of ECMO-bridged recipients (50% vs. 23–30% in the durable VAD and temporary MCS groups, p = 0.044). The average ischemic times were similar, between 3.5 and 3.8 hours. The recipient characteristics differed in younger recipients on ECMO (43.6 years, with durable VAD and temporary MCS recipients average ages 49.4 and 53.9 years, respectively, p = 0.011). The recipients’ BMIs were significantly larger in the durable VAD cohort (28.8 kg/m²), followed by temporary MCS (26.0 kg/m²) then ECMO (23.6 kg/m², p < 0.001). Left ventricular ejection fraction pretransplant was lowest in recipients on temporary MCS (17.5%) followed by ECMO (19.4%) and VAD (20.7%). Average waitlist days were longest in the durable VAD-bridged patients (281 days) followed by ECMO (108 days) and temporary MCS (66 days, p < 0.001). Finally, the ECMO-bridged patients had the highest IMPACT score (13.9) followed by temporary VADs (8.8) and durable VAD (6.2, p < 0.001).

Posttransplant outcomes are presented in Supplemental Table S6, Supplemental Digital Content, http://links.lww.com/ASAIO/B204, where the analysis demonstrated that patients with a baseline durable VAD had the lowest rate of severe PGD (4.5%), followed by temporary MCS (5.1%) and ECMO (21.4%, p = 0.025), as well as the fewest need for any posttransplant MCS support, 12.3%, compared to those on ECMO (28.6%) and temporary MCS (35.0%, p < 0.001). Posttransplant ICU LOS was significantly longer in ECMO-bridged patients (18.2 days), followed by durable VAD-bridged patients (12.3 days), then temporary MCS patients (9.3 days, p = 0.008). Total hospital stay was significantly longer in patients bridged with ECMO at 39.4 days, compared to 26.2 days for durable VAD patients and 23.2 days for temporary MCS patients (p = 0.009). Survival was statistically similar and excellent in all cohorts, with in-hospital survival ranging from 95.5% to 100% and 1 year survival 91.0% to 100%

Discussion

Historically, the donor hearts for transplant have been transported while the organ is immersed in a cold preservative solution surrounded by ice. This technique is inconsistent and does not afford control of cooling or final temperature obtained with myocardial temperatures commonly falling below 1°C within 60 minutes of ice cooler packing.²² These temperatures are below the range recommended by the ISHLT consensus statement (4–8°C).¹⁵ As a result, ice storage risks freezing injury to the myocardium. Studies have shown that prolonged exposure to freezing temperatures during ischemic storage increases cellular edema, ischemic cellular injury, reperfusion injury, and the risk of adverse posttransplant outcomes such as PGD.^13,14,22,23 In addition to variable cooling, ice static storage may predispose donor hearts to heterogenous warming during the implant operation with different temperature warming occurring in different sizes and cooled hearts further amplifying the cold injury described experimentally.^22,23

In contemporary heart transplantation, multiple factors may amplify the risk of adverse posttransplant outcomes associated with prolonged ischemic storage at sub-optimal temperatures. First, after 2018 changes to the UNOS allocation policy the average ischemic time for donor hearts has significantly increased.^6,7 Second, donors with higher risk profiles appear to be common, as evidenced by the high scores observed in the calculation of a modified DRS. The average scores in both ice and SCTS cohorts were similar in both overall and propensity-matched analyses, ranging from 19.8 to 20.5, which indicates a higher risk donor selection, where donors with scores greater than 17 are considered higher risk.²¹ Considering three variables were not included in calculating the score, this seems to demonstrate an evolution of donor selection toward higher-risk donors, since the score was initially developed utilizing transplant data from 2005 to 2008, at least 7 years earlier than the transplants in the GUARDIAN registry.²¹ Third, more transplant recipients in the current era undergo surgery after utilizing bridging MCS such as ECMO, IABP, or VAD.^2,3,6 Longer cold ischemic storage and MCS use before transplant are associated with a higher risk of early graft failure, particularly in patients utilizing ECMO/IABP because of more severe diseases^6,7 that are given a higher transplant priority in the new allocation system.¹ Thus, the increased ischemic time and prior MCS utilization in modern transplant patients have led to some concern that the rate of transplant failure may increase due to these compounding risk factors.

While some of these risks are related to disease etiology that may not be modified (ie, MCS requirements before transplant), the hypothermic preservation environment during heart transport is a controllable element that may be optimized to mitigate the risks associated with heart transplant. Controlled hypothermic organ storage with the Paragonix SherpaPak CTS¹⁶ preserves hearts at temperatures that maximize posttransplant cardiac function while minimizing potential freezing injury.^24–26 Previous studies have demonstrated that the SCTS utilization can improve intraoperative and postoperative outcomes, within complex recipient subsets and with extended ischemic times compared to ICE storage.^17,27 When compared to patients receiving transplant after cold ICE, SCTS utilization reduced the incidence of severe PGD (16.1% ICE vs. 5.7% SCTS) despite a significantly lower proportion of patients in the SCTS cohort with durable MCS (45.1% ICE vs. 31.5% SCTS) and a significantly higher proportion of patients in the SCTS cohort utilizing temporary IABP at baseline (13.7% ICE vs. 29.0% SCTS).¹⁸ Moreover, a propensity-matched analysis in this same cohort showed a 67% reduction in severe PGD after SCTS transport and a 9.8% increase in 1 year survival.¹⁸ Given the increasing number of heart transplant patients in the modern era utilizing pretransplant MCS bridging strategies, the efficacy of SCTS transport in this patient population is an important clinical consideration.

In this study, we assessed the efficacy of cardiac preservation using the SCTS for controlled hypothermic preservation in patients with MCS bridge to transplant utilization. We found that when these patients receive hearts utilizing the controlled hypothermic preservation SCTS, the risk of severe PGD and moderate to severe RVD is significantly reduced compared to patients undergoing transplants using hearts transported in an ice cooler. This reduced morbidity was despite a significantly greater procurement distance traveled and longer ischemic time in the SCTS transplants. The hazard analysis further identified that the overall risk of severe PGD in MCS-bridged patients had an odds ratio that was significantly lower for SCTS preservation compared to ice cooler storage. Posttransplant RV function was also found to be significantly better in patients receiving hearts preserved in an SCTS versus an ice cooler.

To minimize the baseline differences and provide a more similar set of comparative cohorts, a propensity-matched analysis was performed. The selection of the matching variables included those typically considered associated with differences in posttransplant morbidity and mortality (ischemic time, donor age) as well as study-related potential biases (site and transplant era), as well as the specific sub-types of MCS bridges due to the variability of the patient profiles and risks specifically associated with these devices. The results of the propensity matching resulted in a preservation of the advantages observed in the overall analysis; however, the 50% reduction in severe PGD was only trending following propensity matching (p = 0.069).

Since the impact of controlled moderate hypothermic preservation using the SCTS demonstrated significant advantages despite longer ischemic times and a slightly sicker recipient population, we further examined the outcomes stratified by the specific types of bridging strategies using this preservation system. When the SCTS cohort was stratified by unique MCS bridging strategies, it was not surprising to find certain differences among the recipient demographics, such as younger ECMO patients who tended to be more often female. The IMPACT scores were also highest in the ECMO cohort. As a result of the 2018 UNOS Allocation changes, the average days spent on the waitlist was significantly longer for those with a durable VAD than those on temporary MCS support, with ECMO-supported patients predictably with the shortest wait times. An examination of posttransplant outcomes revealed that the risk of severe PGD was lowest in the durable VAD cohort, followed by temporary support and ECMO. In fact, the risk of severe PGD in the VAD cohort was very low at 4.5%, despite ischemic times averaging 3.5 hours. These findings are very encouraging in such a complex patient group.

There are several important limitations to consider in this analysis. First, the registry was retrospective, therefore there may be inherent biases related to treatment decisions that were not controlled. Additionally, the ICE controls were typically enrolled in a reverse consecutive manner, which often meant that the traditional ice storage transplant cases were performed up to 3 years earlier, which could add bias in terms of center experience, practice protocols, and temporal UNOS organ allocation changes. However, the inclusion of site and era variables in the propensity matching should have minimized this potential bias. Second, although limiting the analysis to unique uses of specific types of MCS support may have limited confounders of compounded risks, this also led to a smaller cohort size available for analysis, in particular in the ECMO cohort, which limits the rigor of the findings. Third, severe PGD is a known risk factor for mortality, yet in our analysis the survival was similar. This finding is especially interesting in the ECMO-bridged cohort, which had a very high survival despite a greater risk of severe PGD. This observed survival benefit could be related to improved management strategies for severe PGD, as the ECMO cohort did have the longest ICU and total hospital lengths of stay. Additionally, the GUARDIAN Registry does not allow further analysis of the hemodynamic course in PGD with granular hemodynamic data to define the mechanism of the observed benefit in posttransplant RV function. Finally, although the lack of randomization could be considered a limitation, in fact the nature of the registry in allowing an “all comers” design allows for more robust subgroup analyses which would be limited in a highly controlled, randomized trial.

Conclusion

Our analysis suggests that the use of SCTS controlled hypothermic preservation of donor hearts for use in heart transplants for patients bridged with various MCS bridging strategies can lead to significantly reduced rates of severe PGD and improved right ventricular function posttransplant, as well as reducing the total hospital stays. Although the 50% reduction in severe PGD did not reach significance following propensity matching, these results appear to suggest that when the donor and recipient risks are compounded as in the setting of longer ischemic times, the use of the SCTS may attenuate the posttransplant morbidities typically observed when transplants are performed using hearts transported in ice coolers. Further studies are needed to define the mechanism and extent of the observed benefits of this technology in heart transplant recipients.

Acknowledgment

The authors gratefully acknowledge the medical writing support of Aarti Urs of ALKU, Andover, MA, Julia Kobe of Paragonix Technologies for statistical support, and Mary V. Jacoski of Paragonix Technologies for assistance with the analysis and editorial support.