Abstract
Aims
Donation after circulatory death (DCD) has emerged as a strategy to increase the donor pool for heart transplantation (HT). Left ventricular assist device (LVAD) patients represent a discrete and unique population. We sought to explore the early outcomes of DCD‐HT compared with donation after brain death (DBD) HT in LVAD patients.
Methods and results
We obtained data from the United Network of Organ Sharing database. The main cohort consisted of adults listed for HT between 17 October 2018 and 3 July 2024, with LVAD implanted before or after listing. The primary outcome was survival within the first year post‐HT. There were 3336 patients with LVAD underwent HT during the study period (median age 55 years (interquartile range 45–62), 24% women, 29% Black, 89% DBD). The short‐term post‐HT mortality in LVAD patients who underwent DCD HT was not significantly different from DBD (adjusted hazard ratio [aHR] 1.00, 95% CI 0.70–1.42, P value > 0.9). The likelihood of transplantation within 1 year was higher at centres performing DCD (aHR 1.44, 95% CI 1.39–1.49, P < 0.001). Despite the longer donor‐recipient distance in DCD‐HT, in‐hospital outcomes (stroke and acute kidney injury requiring dialysis) were not different from DBD‐HT. A higher incidence of primary graft dysfunction (adjusted risk ratio [aRR] 3.8, 95% CI 2.5–5.7, P < 0.001), and treated rejection was observed with DCD‐HT (aRR 1.48, 95% CI 1.14–1.93, P = 0.003).
Conclusions
In LVAD patients who received DCD HT, early post‐transplant survival, stroke, acute kidney injury and length of stay were not significantly different from those who underwent DBD HT. There were increased rates of primary graft dysfunction and treated rejection among LVAD patients who underwent DCD HT. Patients in a DCD centre were significantly more likely to be transplanted earlier.
Keywords: Donation after brain death, Donation after circulatory death, Heart transplantation, Left ventricular assist device
Introduction
Heart transplantation (HT) remains the gold standard treatment for advanced heart failure, but the shortage of suitable organs persists, creating a need for further innovation. One strategy to expand the donor pool is using Maastricht category III heart donation after circulatory death (DCD). 1 This novel approach involves obtaining the heart from donors with no meaningful chance of survival but who do not meet brain death criteria. After encouraging early reports in Australia 2 and the United Kingdom, 3 DCD HT is increasingly utilized in the United States 4 , 5 and is projected to increase donor availability by up to 21–30%. 6 , 7
Fuelled by its favourable effect on waitlist outcomes, 8 implantation of a left ventricular assist device (LVAD) as a ‘bridge‐to‐transplant’ (BTT) became a cardinal strategy for hemodynamically decompensated patients. 9 , 10 , 11 Additional advances in LVAD technology and medical management, among other factors, led to improved long‐term outcomes with these devices. 12 , 13 Simultaneously, however, national data have shown that patients with BTT LVADs may fare worse after HT than their counterparts without long‐term mechanical circulatory support. 10 To compound this, the change in the allocation system enacted in October 2018 in the United States of America, which intended to refine risk stratification and improve organ access for patients with high urgency, 14 places stable BTT LVAD patients at lower priority compared with patients on temporary mechanical circulatory support (MCS) devices. 15 Moreover, HT in the BTT LVAD patient poses unique technical (e.g., explant of recipient heart, resulting in a higher transfusion requirement and a pro‐inflammatory milieu) 16 and physiological (e.g., allosensitization; ongoing/occult right ventricular dysfunction; and vasoplegia) 17 , 18 , 19 , 20 challenges that may be exacerbated by the utilization of DCD organs that have undergone considerable ischaemic injury and may result in higher rates of primary graft dysfunction 21 and rejection. 22 As such, while an augmented donor pool may benefit ‘primary’ HT and BTT LVAD patients alike, whether BTT LVAD patients have different outcomes with DCD versus donation after brain death (DBD) HT remains unknown. Accordingly, the objectives of this study were to explore the temporal trends in uptake and to ascertain the early safety of DCD‐HT in LVAD patients. We hypothesized that there is increasing uptake of DCD‐HT for BTT‐LVAD recipients as transplant centres across the United States are gaining experience with this strategy, that DCD‐HT may expedite time‐to‐transplant for eligible BTT‐LVAD by augmenting the donor pool, and that short‐term outcomes after DCD‐HT will not be different from those achieved with DBD‐HT.
Methods
Data source, study population and design
The United Network for Organ Sharing (UNOS) Standard Transplant Analysis and Research files are datasets that contain de‐identified, prospectively collected patient‐level information for transplant recipients and waiting list candidates in the United States. 23 The Institutional Review Board at the University of Washington deemed the study exempt (review date 4/28/23; study ID STUDY00017874).
Adults listed for HT between 17 October 2018 and 3 July 2024, with durable LVAD implanted before or after listing, were eligible for inclusion. The follow‐up period for this cohort ended on 3 January 2025. Patients undergoing multi‐organ transplants and patients for whom follow‐up data were unavailable were excluded from the analysis. Of the 47 492 patients (5791 with LVAD before or after listing) listed for HT during the study period, 26 697 were excluded based on the inclusion and exclusion criteria, leaving 20 795 patients (4942 with LVAD before or after listing) for the waitlist analysis, of whom 15 805 received a HT. Additionally, of the transplanted patients, 59 were excluded from the post‐HT analysis due to unknown follow‐up (Figure 1 ).
Figure 1.
STROBE diagram illustrating the number of patients screened, those included in the time‐to‐transplant analysis, and those included in the post‐transplant analysis. It also details the number of individuals excluded and the reasons for their exclusion.
The primary outcome of the study was survival within the first year post‐HT. Secondary outcomes were post‐HT 1‐ and 6‐month survival; post‐HT stroke, acute kidney injury requiring dialysis, length of stay during the index admission, primary graft dysfunction within 24 h post‐transplant (as reported in the UNOS database), and treated rejection within 1 year from transplant in DCD versus DBD HT. Additionally, transplant likelihood in DCD versus non‐DCD centres was explored. We utilized UNOS data (time from circulatory standstill to cross‐clamp for cold cardioplegia) to distinguish procurement methods (direct procurement and perfusion [DPP] vs. normothermic regional perfusion [NRP]) as previously described. 5
The covariates reported by the transplant centre or derived from those variables and included in our analyses were the recipient's age at listing, sex at birth, self‐reported race/ethnicity, presence of diabetes mellitus, self‐reported cigarette use, diagnosis of ischaemic cardiomyopathy as an indication for transplant, estimated glomerular filtration rate (eGFR), 24 and support with veno‐arterial extracorporeal membrane oxygenation (VA‐ECMO) at transplant. Additionally, the predicted heart mass ratio was calculated as previously described. 25 Outcomes reported by the transplant programme, such as treated rejection, acute kidney injury requiring haemodialysis, and stroke during the index admission (at transplant), were included in some models.
Statistical analysis
Baseline clinical characteristics were described using absolute numbers and percentages for categorical variables and median and interquartile range (IQR) for continuous variables. Between‐group comparisons were made using the Pearson χ 2 test and Fisher's exact test for categorical variables, and the Wilcoxon rank‐sum test for continuous variables. Temporal trends in categorical data were evaluated using the Cochran–Armitage test for trend. 26 Cox proportional hazards models were built to estimate the post‐HT survival in DCD versus DBD‐HT. The post‐HT survival analysis was adjusted for a priori selected variables known to be associated with post‐transplant outcomes: recipient [age at listing, sex at birth, self‐reported race/ethnicity, diabetes mellitus, cigarette use, ischaemic cardiomyopathy, estimated glomerular filtration rate (eGFR), and support with veno‐arterial extra‐corporeal membrane oxygenation [VA‐ECMO] at transplant], transplant (predicted heart mass ratio and female donor‐male recipient mismatch) and donor (age) variables. Cox proportional hazards models using centre initiation of DCD‐HT, defined as the date when each centre performed their first DCD‐HT during the study period, as a time‐varying exposure, were built to estimate the likelihood of HT while waitlisted and tested if having LVAD was an effect modifier. This analysis was adjusted for recipient age at listing, sex at birth, self‐reported race/ethnicity, and to account for potential non‐proportional baseline hazards in certain subgroups, the model was stratified by blood type (O vs. non‐O) and medical urgency (high priority UNOS status 1–2 vs. lower priority statuses). Proportional hazard assumption was tested using Schoenfeld residuals test. 27 Lastly, Poisson regression with robust standard errors obtained from the sandwich estimator was used to test the association between DCD‐HT and length‐of‐stay, treated rejection within 1 year, as well as in‐hospital complications, including acute kidney injury requiring haemodialysis and stroke. The Poisson regression models were adjusted for recipient age, gender, race/ethnicity, diabetes mellitus, cigarette use, ischaemic cardiomyopathy, eGFR, and VA‐ECMO at transplant; donor age and LVAD type. To minimize the potential for informative censoring bias on the primary outcome, a sensitivity analysis was performed limited to patients who had completed at least 1 year of follow‐up.
Two‐sided P values <0.05 were considered significant. Analyses were performed using RStudio Version 2024.12.0+467 (2024.12.0+467).
Results
Baseline characteristics and temporal trends
A total of 15 746 patients underwent HT during the study period, including 3336 (21%) patients with durable LVAD [DBD‐HT: 2970 (89%); DCD‐HT: 366 (11%)] (Table 1 and Figure 1 ). There was a temporal trend towards proportionately higher uptake of DCD‐HT throughout the study period in both the BTT LVAD and primary transplant groups (Figure 2 ). The proportion of DCD‐HT, including in LVAD and non‐LVAD recipients, steadily increased from 0% in the earliest study period to 18.9% in the latest (P‐trend < 0.005). Similarly, LVAD patients also increased in DCD‐HT from 0% to 23.3% in similar periods (P‐trend < 0.005) (Figure 2 ). Overall, compared with patients without LVAD, patients with LVAD were more likely to undergo DCD‐HT (LVAD 11% vs. non‐LVAD 8.9%, P < 0.001).
Table 1.
Baseline recipient with LVAD support and donor characteristics—DBD versus DCD
| All patients, N = 3336 | DBD, N = 2970 | DCD, N = 366 | P value a | |
|---|---|---|---|---|
| Recipient characteristics | ||||
| Age (years) | 55 (45, 62) | 55 (45, 62) | 54 (43, 62) | 0.4 |
| Female gender | 787 (24%) | 718 (24%) | 69 (19%) | 0.028 |
| Race/ethnicity | 0.4 | |||
| American Indian/Alaska Native | 10 (0.3%) | 10 (0.3%) | 0 (0%) | |
| Asian | 90 (2.7%) | 83 (2.8%) | 7 (1.9%) | |
| Black | 970 (29%) | 876 (29%) | 94 (26%) | |
| Hispanic | 322 (9.7%) | 288 (9.7%) | 34 (9.3%) | |
| Native Hawaiian/Pacific Islander | 6 (0.2%) | 6 (0.2%) | 0 (0%) | |
| Others | 24 (0.7%) | 21 (0.7%) | 3 (0.8%) | |
| White | 1914 (57%) | 1686 (57%) | 228 (62%) | |
| BMI (kg/m2) | 29.7 (26.0, 33.3) | 29.6 (25.9, 33.2) | 30.1 (26.3, 34.3) | 0.053 |
| Diabetes mellitus | 1084 (32%) | 965 (32%) | 119 (33%) | >0.9 |
| Estimated GFR (ml/min/m2) | 70 (54, 89) | 69 (54, 89) | 72 (58, 93) | 0.008 |
| Ischaemic cardiomyopathy | 1025 (31%) | 922 (31%) | 103 (28%) | 0.3 |
| Durable LVAD type | <0.001 | |||
| HeartMate 3 | 2219 (67%) | 1922 (65%) | 297 (81%) | |
| HeartWare | 759 (23%) | 719 (24%) | 40 (11%) | |
| HeartMate II | 344 (10%) | 316 (11%) | 28 (7.7%) | |
| Blood type | <0.001 | |||
| A | 1273 (38%) | 1158 (39%) | 115 (31%) | |
| AB | 150 (4.5%) | 144 (4.8%) | 6 (1.6%) | |
| B | 503 (15%) | 458 (15%) | 45 (12%) | |
| O | 1410 (42%) | 1210 (41%) | 200 (55%) | |
| Days on waitlist | 97 (25, 299) | 98 (25, 298) | 95 (21, 320) | >0.9 |
| UNOS status at transplant | <0.001 | |||
| 1 | 236 (7.1%) | 231 (7.8%) | 5 (1.4%) | |
| 2 | 788 (24%) | 740 (25%) | 48 (13%) | |
| 3 | 1277 (38%) | 1123 (38%) | 154 (42%) | |
| 4 | 1032 (31%) | 873 (29%) | 159 (43%) | |
| Donor characteristics | ||||
| Age (years) | 33 (26, 41) | 34 (26, 42) | 32 (25, 38) | <0.001 |
| Female gender | 877 (26%) | 823 (28%) | 54 (15%) | <0.001 |
| LVEF (%) | 60 (56, 65) | 60 (56, 65) | 62 (59, 66) | 0.002 |
| Cause of death | 0.003 | |||
| Brain anoxia | 1610 (48%) | 1447 (49%) | 163 (45%) | |
| Cerebrovascular accident/stroke | 432 (13%) | 401 (14%) | 31 (8.5%) | |
| Head trauma | 1180 (35%) | 1026 (35%) | 154 (42%) | |
| CNS tumour/other | 16 (0.5%) | 14 (0.5%) | 2 (0.5%) | |
| Transplant characteristics | ||||
| Organ travel distance (nautical miles) | 212 (91, 402) | 207 (86, 383) | 272 (127, 536) | <0.001 |
| Ischaemic time (hours) | 3.6 (2.9, 4.3) | 3.5 (2.9, 4.2) | 5.5 (3.8, 6.8) | <0.001 |
| Donor‐recipient gender mismatch | 0.005 | |||
| Match | 2734 (82%) | 2431 (82%) | 303 (83%) | |
| Female donor/male recipient | 346 (10%) | 322 (11%) | 24 (6.6%) | |
| Male donor/female recipient | 256 (7.7%) | 217 (7.3%) | 39 (11%) | |
| Predicted heart mass ratio | 0.98 (0.90, 1.08) | 0.99 (0.90, 1.08) | 0.98 (0.90, 1.08) | 0.5 |
| ECMO at the time of transplant | 33 (1.0%) | 29 (1.0%) | 4 (1.1%) | >0.9 |
Values are n/N (percentage) or median (IQR).
BMI, body mass index; CNS, central nervous system; ECMO, extra‐corporeal membrane oxygenation; DBD, donation after brain death; DCD, donation after circulatory death; GFR, glomerular filtration rate; IABP, intra‐aortic balloon pump; LVAD, left ventricular assist device; LVEF, left ventricular ejection fraction; RVAD, right ventricular assist device; UNOS, United Network for Organ Sharing.
DBD versus DCD.
Figure 2.
Heart transplants (HT) performed in patients with (dark blue and red) or without (light blue and red) an LVAD, using either the DCD (red shades) or DBD (blue shades) approach, throughout the study period.
Of the 3336 LVAD patients that underwent HT, the median age was 55 (IQR 45–62) years, 24% were women, 29% were Black and 57% White, 67% had HeartMate 3 (HM3) (Abbott, Chicago, IL) and 23% HeartWare (HVAD) (Medtronic, Minneapolis, MN) devices. Compared with the LVAD DBD‐HT group, LVAD recipients of DCD‐HT were more likely to be male, blood group O, have a HM3 device, and had marginally higher eGFR. Transplant from UNOS status 1 or 2 was less common in DCD‐HT (14% vs. 32%, P < 0.001). Notably, 43% of DCD‐HT recipients were transplanted from status 4% versus 29% for DBD‐HT, P < 0.001). In addition, DCD donors were younger [median 32 (IQR 25–38) vs. 34 (IQR 26–42) years, P < 0.001] and less likely to be female (15% vs. 28%, P < 0.001) (Table 1 ).
Waitlist and transplant characteristics
By the end of the study period, DCD‐HT was performed in 71 HT centres. LVAD patients listed in DCD centres waited significantly less time than those that were listed in a non‐DCD centre [85 (IQR 22–280) vs. 129 (IQR 35–371) days, P < 0.001]. The likelihood of transplantation within 1‐year from listing for both LVAD and non‐LVAD patients was 44% higher at a DCD centre [adjusted hazard ratio (aHR) 1.44, 95% CI 1.39–1.49; P < 0.001] (Figure 3 A ). This effect was more pronounced in LVAD patients (aHR 1.77, 95% CI 1.65–1.90, P < 0.001, Figure 3 B ).
Figure 3.
(A) Likelihood of heart transplant within 1 year from listing at a DCD versus non‐DCD centre, accounting for time‐varying exposure of becoming a DCD centre. (B) Likelihood of heart transplant within 1 year from listing by DCD versus non‐DCD centre status and left ventricular assist device (LVAD) versus non‐LVAD patient status. The Cox regression models with time‐varying exposure are adjusted for recipient age at listing, sex at birth, self‐reported race/ethnicity, and to account for potential non‐proportional baseline hazards in specific subgroups, the model was stratified by blood type (O vs. non‐O) and medical urgency (high priority UNOS status 1–2 vs. lower priority statuses). aHR: adjusted hazard ratio, CI: confidence interval.
For those LVAD patients who were ultimately transplanted, there was no significant difference in time on the waitlist between those who underwent DCD‐HT versus DBD‐HT [98 (IQR 25–298) vs. 95 (IQR 21–320) days, P > 0.9] (Table 1 ). Notably, DCD‐HT was associated with significantly longer donor‐recipient distance [272 (IQR 127–536) vs. 207 (IQR 86–383) nautical miles, P < 0.001] and total ischaemic time [5.5 (IQR 3.8–6.8) vs. 3.5 (IQR 2.9–4.2) hours, P < 0.001], as compared with DBD‐HT (Table 1 ).
Post‐transplant outcomes in LVAD patients
Median post‐transplant follow‐up was 731 (IQR 353–1285) days in those patients who were transplanted while on LVAD support. The estimated post‐HT 1‐year survival in the entire LVAD cohort was 89% (95% CI 87.9–90.1%) and was not different between the DCD and DBD‐HT groups (89.4% vs. 88.9%, respectively, aHR 1.00, 95% CI 0.70–1.42; P > 0.9) (Figure 4 A and Table 2 ). There was no difference in the primary outcome according to the mode of donor heart procurement (DPP vs. NRP, data available for 322/366 (88%) of LVAD patients undergoing DCD‐HT (Figure 4 B ). After multivariable adjustment, compared to DBD‐HT, DCD‐HT was not associated with a difference in post‐HT 1‐month and 6‐month survival (Table S1 ), length of hospital stay, acute kidney injury requiring haemodialysis or stroke during the index admission (Table 2 and Table S2 ). Treated rejection within 1 year [adjusted risk ratio (aRR) 1.48, 95% CI 1.14–1.93, P = 0.003) and primary graft dysfunction (aRR 3.8, 95% CI 2.5–5.7, P < 0.001) were more likely with DCD than with DBD‐HT among LVAD patients (Table 2 and Table S2 ).
Figure 4.
Kaplan–Meier shows 1‐year post‐HT survival in LVAD patients according to (A) donation after DBD (red) versus DCD (blue); and (B) to mode of donor heart procurement (DPP, direct procurement and perfusion, red versus NRP, normothermic regional perfusion, blue). The Cox regression models were adjusted for recipient age, gender, race/ethnicity, diabetes mellitus, cigarette use, ischaemic cardiomyopathy, estimated glomerular filtration rate, support with veno‐arterial extra‐corporeal membrane oxygenation, predicted heart mass, female donor‐male recipient mismatch and donor age.
Table 2.
Multivariable models for mortality and treated rejection (1‐year) and for length‐of‐stay, acute kidney injury requiring haemodialysis, stroke and primary graft dysfunction (index admission) post‐transplant
| Covariates | 1‐year | Index admission | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mortality a | Treated rejection b | Length‐of‐stay b | Haemodialysis b | Stroke b | Primary graft dysfunction b | |||||||
| HR (95% CI) | P value | RR (95% CI) | P value | RR (95% CI) | P value | RR (95% CI) | P value | RR (95% CI) | P value | RR (95% CI) | P value | |
| DCD‐HT versus DBD‐HT | ||||||||||||
| Unadjusted | 0.94 (0.66–1.33) | 0.7 | 1.42 (1.09–1.84) | 0.01 | 0.93 (0.81–1.05) | 0.2 | 1.11 (0.88–1.40) | 0.4 | 0.88 (0.55–1.40) | 0.6 | 3.92 (2.43–5.38) | <0.001 |
| Adjusted | 1.00 (0.70–1.42) | >0.9 | 1.48 (1.14–1.93) | 0.003 | 0.95 (0.83–1.08) | 0.4 | 1.16 (0.91–1.48) | 0.2 | 0.94 (0.60–1.50) | 0.8 | 3.78 (2.50–5.70) | <0.001 |
DBD, donation after brain death; DCD, donation after circulatory death; HR, hazard ratio; HT, heart transplantation; RR, risk ratio.
Cox proportional models were adjusted for recipient age, gender, race/ethnicity, diabetes mellitus, cigarette use, ischaemic cardiomyopathy, estimated glomerular filtration rate, support with veno‐arterial extra‐corporeal membrane oxygenation, predicted heart mass, female donor‐male recipient mismatch, and donor age.
Poisson regression models were adjusted for recipient age, gender, race/ethnicity, diabetes mellitus, cigarette use, ischaemic cardiomyopathy, eGFR, LVAD type, and VA‐ECMO at transplant; and donor age.
Sensitivity analysis for post‐HT 1‐year survival limited to a cohort of 2918 (87.5%) BTT LVAD patients who had completed 1‐year follow‐up similarly showed no difference between DCD‐HT and DBD‐HT (Figure S1 ).
Discussion
Our analysis utilizing national data from UNOS reflects the largest cohort of durable LVAD patients undergoing HT with DCD to date. In this study we found that there was no significant difference in the short‐term post‐HT survival with either DCD vs DBD HT among patients with durable LVAD that underwent HT. Second, patients with durable LVAD were more likely to receive a DCD‐HT than participants who underwent primary HT without a long‐term device. Third, the uptake of DCD organ utilization increased substantially during the study period, especially in the LVAD cohort, for whom it represented almost a quarter of all transplants in the final portion of the study period. Fourth, BTT LVAD patients listed at a centre that has initiated DCD‐HT were significantly more likely to be transplanted within 1 year. Finanlly, despite longer organ travel distance, DCD‐HT recipients with an LVAD had overall excellent outcomes, including 89.4% survival at 1 year and no penalty for index admission length‐of‐stay, AKI requiring haemodialysis or stroke, but an increase in the rate of treated rejection at 1‐year and primary graft dysfunction.
The increasing utilization of LVADs as BTT during the last decade led to almost half of the patients being transplanted in the United States in 2017 to do so from a durable LVAD status. 15 Despite improved waitlist outcomes with BTT LVAD, partly driven by improvements in technology, 13 the prior 3‐tier UNOS heart allocation policy was felt to not adequately risk‐stratify such patients. Aiming to allocate organs more equitably, the Organ Procurement and Transplantation Network revised the policy in 2018, with dischargeable LVAD patients defaulting to status 4, with a higher listing for a discretionary 30‐day period, complications, or hemodynamic compromise. This change de facto disincentivized durable LVAD utilization by centres with access to transplant but also resulted in incremental utilization of temporary mechanical circulatory support devices. 28 Furthermore, it drastically affected a large cohort of BTT LVAD patients awaiting HT who were now significantly less likely to get transplanted promptly unless they accrue complications or receive higher‐risk organs from more considerable distances, with the potential for attendant increase in post‐transplant mortality. 15 Interestingly, compared with direct transplant, BTT LVAD post‐HT outcomes appear to be worse in the new allocation system, possibly due to even longer ischaemic times and travelled distances as well as increased high‐risk donor organ utilization, among other factors. 15 Indeed, BTT LVAD patients undergoing HT in the new era are frailer and more clinically unstable, as evidenced by worse functional status and higher rates of inotropic support. 29 Moreover, LVAD recipients may be more sensitive to allograft undersizing 19 and gender mismatch, 18 and are more likely to be allosensitized, 17 , 30 all factors that can delay the identification of an optimal donor.
These idiosyncrasies in the transplantation of BTT LVAD patients, combined with the signal for worse post‐HT outcomes in the new allocation system era, underscore the need for dedicated research in this population and possibly a recalibration of their estimated risk. DCD‐HT could bridge the gap in contemporary outcomes between BTT LVAD patients and primary HT. 10 In that context, early studies of DCD‐HT have provided insight into the current and future role of DCD in the United States. An analysis of all potential DCD donors over a ~ 1 year period showed that currently, significantly selected organs (hearts were obtained from 3.9% of available DCD donors; median age 29 years, 90% male) were donated to patients that historically have long waiting times (median BMI 29.4 kg/m2, 41.7% UNOS status 4, 60.6% blood group O), with overall excellent short‐term outcomes. 4 Most recently, in the only randomized trial of DCD‐HT to date (Donors after Circulatory Death Heart Trial, NCT03831048), 21 90 patients randomized to DCD‐HT (49% LVAD, 48% status 4) had non‐inferior 6‐month survival (95% vs. 89%) as compared with overall higher urgency (52% status 2) DBD‐HT recipients. Despite exhibiting higher rates of the International Society for Heart and Lung Transplantation moderate or severe primary graft dysfunction (23% vs. 10%), no patients in the DCD‐HT group required re‐transplantation. Overall, 89% of DCD hearts that were preserved on the extra‐corporeal perfusion system (Organ Care System Heart, TransMedics) were utilized.
Our study mirrors this donor and recipient population in a dedicated BTT LVAD cohort that currently reflects approximately 30% of DCD‐HT recipients in the United States, and is the only study explicitly focusing on DCD‐HT outcomes in such patients to date. Our findings support comparable in‐hospital and early outcomes of DCD‐HT in this population and demonstrate that BTT LVAD patients are transplanted faster in centres performing DCD‐HT. Arguably, this effective and efficient utilization of DCD organs does not solely benefit the index recipient but also cascades into patients that are ultimately transplanted via DBD‐HT (who had comparable waiting times in our study) and patients approaching end‐stage heart failure who may need durable LVAD support, by offloading saturated cohorts of BTT LVAD patients awaiting HT at major centres. The encouraging results of our study, combined with advances in minimizing ischaemic injury to the heart during a procurement process that involves ‘warm ischaemic time’, 31 should translate into broader acceptance of DCD‐HT nationally and increased donor risk tolerance. Nonetheless, given concerns regarding ischaemic injury incurred during the DCD procurement process and attendant primary graft dysfunction, the long‐term durability of these favourable outcomes remains to be tested. Although we found no significant difference in short‐term mortality between LVAD patients receiving DBD versus DCD HT, the confidence interval of 0.70 to 1.42 suggests that the post‐HT hazard of death with DCD in this patient population, compared to DBD, could be 30% lower or 42% higher. Therefore, these results should be interpreted with caution. It is noteworthy that we found a moderately increased rate of treated rejection at 1‐year with DCD‐HT as compared with DBD‐HT. This phenomenon, which has recently been described in the broader DCD‐HT population, may be related to the immunological substrate of the recipients, differences in procurement techniques/ischaemic time, and need for post‐operative MCS, 22 and may be of particular import to BTT LVAD patients that are known to be more allosensitized at baseline, a trait which may significantly compound the risk of rejection. 17 , 30 We also observed significantly higher rates of primary graft dysfunction among LVAD patients who underwent DCD‐HT compared with DBD. This finding is highly relevant. However, since the outcome of primary graft dysfunction in the UNOS database is not subject to a rigorous adjudication process and may vary among transplant programs, this result should be interpreted cautiously. Finally, DCD heart procurement is significantly more resource‐intensive, and limited data are available regarding the cost‐effectiveness of this strategy to date.
Our study should be interpreted within the context of significant limitations. It represents a retrospective analysis of registry data with inherent caveats. The accuracy and completeness of the UNOS database is imperfect. Despite efforts to adjust for potential confounders, the lack of randomization and various between‐group differences in donor and recipient characteristics could lead to residual bias and confounding; however, our findings are consistent with the only multicentre randomized trial to date and are representative of real‐world practice across the United States. 21 Despite earlier uptake in other countries, DCD remains a relatively new strategy in the United States, resulting in a relatively small study cohort and short follow‐up. Critical data pertaining to organ procurement strategy, including mode of donor heart preservation with respective qualitative assessment profiles, and granular time of DCD timing intervals such as ‘stand‐off’, total and functional ischaemic time, among others, were unavailable. Finally, the study included the onset and peak of COVID‐19 pandemic‐related morbidity and mortality with attendant effects on transplant performance and outcomes, which may have slowed initial DCD utilization.
In conclusion, LVAD patients listed at DCD centres nationally have higher transplantation rates. Despite the longer donor‐recipient distance, DCD‐HT in‐hospital and short‐term outcomes appear comparable to DBD‐HT in this clinically vulnerable population. As such, DCD‐HT may represent an effective and safe strategy to promptly transplant LVAD patients, optimizing outcomes for those currently on the list and reinvigorating LVAD implantation as a viable approach for future transplantation in patients with end‐stage heart failure needing advanced therapies.
Conflict of interest
JRK reports stock ownership in Abbott, AbbVie, Bristol Myers Squibb, Johnson & Johnson, Eli Lilly, Medtronic, Merck, and Pfizer. CM: Investigator/Consultant: Abbott, Abiomed. SL: Consultant for Abiomed. All remaining authors report no conflicts.
Funding
This research was supported in part by HCA Healthcare and/or an HCA Healthcare‐affiliated entity. The views expressed in this publication represent those of the authors do not necessarily represent the official views of HCA Healthcare or any of its affiliated entities.
Supporting information
Figure S1. Kaplan–Meier shows post‐heart transplant 1‐year survival of LVAD patients in the DBD (red) vs DCD (blue) groups, limited to patients that completed at least 1‐year follow‐up. The Cox regression models were adjusted for recipient age, gender, race/ethnicity, diabetes mellitus, cigarette use, ischemic cardiomyopathy, estimated glomerular filtration rate, support with veno‐arterial extra‐corporeal membrane oxygenation, predicted heart mass, female donor‐male recipient mismatch, and donor age.
Table S1. Poisson models estimates for length‐of‐stay, acute kidney injury requiring haemodialysis, stroke, and primary graft dysfunction at index admission, and treated rejection within 1‐year post‐transplant.
Table S2. Cox model estimates for 1‐month, 6‐months, and 1‐year post‐transplant mortality.
Karatasakis, A. , Silvestri, E. G. , Nair, G. G. , Zuniga, B. , Li, S. , Mahr, C. , Cheng, R. K. , Stempien‐Otero, A. S. , Dimarakis, I. , Khorsandi, M. , Pal, J. D. , Kizer, J. R. , Simon, M. A. , and Bravo, C. A. (2025) Heart transplantation outcomes with donation after circulatory death in patients with left ventricular assist device. ESC Heart Failure, 12: 3333–3342. 10.1002/ehf2.15357.
References
- 1. Thuong M, Ruiz A, Evrard P, Kuiper M, Boffa C, Akhtar MZ, et al. New classification of donation after circulatory death donors definitions and terminology. Transpl Int 2016;29:749‐759. doi: 10.1111/tri.12776 [DOI] [PubMed] [Google Scholar]
- 2. Chew HC, Iyer A, Connellan M, Scheuer S, Villanueva J, Gao L, et al. Outcomes of donation after circulatory death heart transplantation in Australia. J Am Coll Cardiol 2019;73:1447‐1459. doi: 10.1016/j.jacc.2018.12.067 [DOI] [PubMed] [Google Scholar]
- 3. Messer S, Cernic S, Page A, Berman M, Kaul P, Colah S, et al. A 5‐year single‐center early experience of heart transplantation from donation after circulatory‐determined death donors. J Heart Lung Transplant 2020;39:1463‐1475. doi: 10.1016/j.healun.2020.10.001 [DOI] [PubMed] [Google Scholar]
- 4. Madan S, Saeed O, Forest SJ, Goldstein DJ, Jorde UP, Patel SR. Feasibility and potential impact of heart transplantation from adult donors after circulatory death. J Am Coll Cardiol 2022;79:148‐162. doi: 10.1016/j.jacc.2021.10.042 [DOI] [PubMed] [Google Scholar]
- 5. Kwon JH, Ghannam AD, Shorbaji K, Welch B, Hashmi ZA, Tedford RJ, et al. Early outcomes of heart transplantation using donation after circulatory death donors in the United States. Circ Heart Fail 2022;15:e009844. doi: 10.1161/CIRCHEARTFAILURE.122.009844 [DOI] [PubMed] [Google Scholar]
- 6. Jawitz OK, Raman V, DeVore AD, Mentz RJ, Patel CB, Rogers J, et al. Increasing the United States heart transplant donor pool with donation after circulatory death. J Thorac Cardiovasc Surg 2020;159:e307‐e309. doi: 10.1016/j.jtcvs.2019.09.080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Suarez‐Pierre A, Iguidbashian J, Stuart C, King RW, Cotton J, Carroll AM, et al. Appraisal of donation after circulatory death: how far could we expand the heart donor pool? Ann Thorac Surg 2022;114:676‐682. doi: 10.1016/j.athoracsur.2022.01.042 [DOI] [PubMed] [Google Scholar]
- 8. Truby LK, Garan AR, Givens RC, Takeda K, Takayama H, Trinh PN, et al. Ventricular assist device utilization in heart transplant candidates: nationwide variability and impact on waitlist outcomes. Circ Heart Fail 2018;11:e004586. doi: 10.1161/CIRCHEARTFAILURE.117.004586 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Trivedi JR, Cheng A, Singh R, Williams ML, Slaughter MS. Survival on the heart transplant waiting list: impact of continuous flow left ventricular assist device as bridge to transplant. Ann Thorac Surg 2014;98:830‐834. doi: 10.1016/j.athoracsur.2014.05.019 [DOI] [PubMed] [Google Scholar]
- 10. Truby LK, Farr MA, Garan AR, Givens R, Restaino SW, Latif F, et al. Impact of bridge to transplantation with continuous‐flow left ventricular assist devices on posttransplantation mortality. Circulation 2019;140:459‐469. doi: 10.1161/CIRCULATIONAHA.118.036932 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Colvin M, Smith JM, Hadley N, Skeans MA, Uccellini K, Goff R, et al. OPTN/SRTR 2018 annual data report: heart. Am J Transplant 2020;20:340‐426. doi: 10.1111/ajt.15676 [DOI] [PubMed] [Google Scholar]
- 12. Mehra MR, Uriel N, Naka Y, Cleveland JC Jr, Yuzefpolskaya M, Salerno CT, et al. A fully magnetically levitated left ventricular assist device ‐ final report. N Engl J Med 2019;380:1618‐1627. doi: 10.1056/NEJMoa1900486 [DOI] [PubMed] [Google Scholar]
- 13. Mehra MR, Goldstein DJ, Cleveland JC, Cowger JA, Hall S, Salerno CT, et al. Five‐year outcomes in patients with fully magnetically levitated vs axial‐flow left ventricular assist devices in the MOMENTUM 3 randomized trial. JAMA 2022;328:1233‐1242. doi: 10.1001/jama.2022.16197 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Goff RR, Uccellini K, Lindblad K, Hall S, Davies R, Farr M, et al. A change of heart: preliminary results of the US 2018 adult heart allocation revision. Am J Transplant 2020;20:2781‐2790. doi: 10.1111/ajt.16010 [DOI] [PubMed] [Google Scholar]
- 15. Mullan CW, Chouairi F, Sen S, Mori M, Clark KAA, Reinhardt SW, et al. Changes in use of left ventricular assist devices as bridge to transplantation with new heart allocation policy. JACC Heart Fail 2021;9:420‐429. doi: 10.1016/j.jchf.2021.01.010 [DOI] [PubMed] [Google Scholar]
- 16. Khoshbin E, Schueler S. Pre‐transplant ventricular assist device explant. Ann Cardiothorac Surg 2018;7:160‐168. doi: 10.21037/acs.2018.01.04 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Drakos SG, Stringham JC, Long JW, Gilbert EM, Fuller TC, Campbell BK, et al. Prevalence and risks of allosensitization in HeartMate left ventricular assist device recipients: the impact of leukofiltered cellular blood product transfusions. J Thorac Cardiovasc Surg 2007;133:1612‐1619. doi: 10.1016/j.jtcvs.2006.11.062 [DOI] [PubMed] [Google Scholar]
- 18. Healy AH, Stehlik J, Edwards LB, McKellar SH, Drakos SG, Selzman CH. Predictors of 30‐day post‐transplant mortality in patients bridged to transplantation with continuous‐flow left ventricular assist devices‐‐an analysis of the International Society for Heart and Lung Transplantation Transplant Registry. J Heart Lung Transplant 2016;35:34‐39. doi: 10.1016/j.healun.2015.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Schumer EM, Black MC, Rogers MP, Trivedi JR, Birks EJ, Lenneman AJ, et al. Donor oversizing results in improved survival in patients with left ventricular assist device. ASAIO J 2016;62:571‐577. doi: 10.1097/MAT.0000000000000399 [DOI] [PubMed] [Google Scholar]
- 20. Ali JM, Patel S, Catarino P, Vuylsteke A, Pettit S, Bhagra S, et al. Vasoplegia following heart transplantation and left ventricular assist device explant is not associated with inferior outcomes. J Thorac Dis 2020;12:2426‐2434. doi: 10.21037/jtd.2020.03.53 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Schroder JN, Patel CB, DeVore AD, Bryner BS, Casalinova S, Shah A, et al. Transplantation outcomes with donor hearts after circulatory death. N Engl J Med 2023;388:2121‐2131. doi: 10.1056/NEJMoa2212438 [DOI] [PubMed] [Google Scholar]
- 22. Li SS, Funamoto M, Osho AA, Rabi SA, Paneitz D, Singh R, et al. Acute rejection in donation after circulatory death (DCD) heart transplants. J Heart Lung Transplant 2023; doi: 10.1016/j.healun.2023.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. OPTN ‐ data . https://optn.transplant.hrsa.gov/data. Accessed 1 June 2025
- 24. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604‐612. doi: 10.7326/0003-4819-150-9-200905050-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Kransdorf EP, Kittleson MM, Benck LR, Patel JK, Chung JS, Esmailian F, et al. Predicted heart mass is the optimal metric for size match in heart transplantation. J Heart Lung Transplant 2019;38:156‐165. doi: 10.1016/j.healun.2018.09.017 [DOI] [PubMed] [Google Scholar]
- 26. Agresti A, Kateri M. Categorical data analysis. In: Lovric M, ed. International encyclopedia of statistical science. Berlin, Heidelberg: Springer Berlin Heidelberg; 2011:206‐208. [Google Scholar]
- 27. Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994;81:515‐526. doi: 10.1093/biomet/81.3.515 [DOI] [Google Scholar]
- 28. Varshney AS, Berg DD, Katz JN, Baird‐Zars VM, Bohula EA, Carnicelli AP, et al. Use of temporary mechanical circulatory support for management of cardiogenic shock before and after the United Network for Organ Sharing donor heart allocation system changes. JAMA Cardiol 2020;5:703‐708. doi: 10.1001/jamacardio.2020.0692 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Li SS, Funamoto M, Osho A, Wolfe S, Kubi B, Singh R, et al. Effect of the 2018 heart allocation system on patients with durable left ventricular assist devices. J Thorac Cardiovasc Surg 2022;167:217‐230.e9. doi: 10.1016/j.jtcvs.2022.09.037 [DOI] [Google Scholar]
- 30. Chau VQ, Feinman J, Fullin K, Mahmood K, Oliveros E, Mitter SS, et al. De novo human leukocyte antigen allosensitization patterns in patients bridged to heart transplantation using left ventricular assist devices. Transpl Immunol 2022;72:101567. doi: 10.1016/j.trim.2022.101567 [DOI] [PubMed] [Google Scholar]
- 31. Schroder JN, Shah A, Pretorius V, Smith J, Daneshmand M, Geirsson A, et al. Expanding heart transplants from Donors After Circulatory Death (DCD) ‐ results of the first randomized controlled trial using the Organ Care System (OCS™) heart ‐ (OCS DCD Heart Trial). J Heart Lung Transplant 2022;41:S72. doi: 10.1016/j.healun.2022.01.165 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Kaplan–Meier shows post‐heart transplant 1‐year survival of LVAD patients in the DBD (red) vs DCD (blue) groups, limited to patients that completed at least 1‐year follow‐up. The Cox regression models were adjusted for recipient age, gender, race/ethnicity, diabetes mellitus, cigarette use, ischemic cardiomyopathy, estimated glomerular filtration rate, support with veno‐arterial extra‐corporeal membrane oxygenation, predicted heart mass, female donor‐male recipient mismatch, and donor age.
Table S1. Poisson models estimates for length‐of‐stay, acute kidney injury requiring haemodialysis, stroke, and primary graft dysfunction at index admission, and treated rejection within 1‐year post‐transplant.
Table S2. Cox model estimates for 1‐month, 6‐months, and 1‐year post‐transplant mortality.