Abstract
Background:
We sought to derive a risk score for 1-year mortality following orthotopic heart transplantation (OHT) in patients bridged with a contemporary centrifugal left ventricular assist device (LVAD).
Methods:
Adult patients (≥18 years) in the United Network for Organ Sharing database undergoing OHT between 2010 and 2019 who were bridged with a HeartWare or HeartMate III device were included. Derivation and validation cohorts were randomly assigned with a 2:1 ratio. Threshold analysis and multivariable logistic regression were utilized to obtain adjusted odds ratios for 1-year post-OHT mortality. A risk score was generated using these adjusted odds ratios in the derivation cohort and the predictive performance of the composite index was evaluated in the validation set.
Results:
A total of 3434 patients were identified. In the derivation cohort, the mean age was 53.5 ± 12.1 years and 1758 (76.8%) were male; 1789 (78.1%) were bridged with a HeartWare device. Multivariable logistic regression revealed that recipient age ≥50 years, bilirubin level ≥2.4 mg/dl, ischemic time ≥4 h, and preoperative hemodialysis predicted 1-year post-transplant mortality. Stratification into risk groups in the validation cohort revealed significant differences in postoperative renal failure, stroke, and short-term mortality. One-year post-transplant mortality was 5%, 6.7%, and 14.8% in the low-, moderate-, and high-risk categories, respectively (p < .001).
Conclusions:
Among patients bridged to OHT with newer generation centrifugal LVADs, older age, increasing bilirubin, longer ischemic time, and pre-OHT dialysis independently predicted post-transplant mortality. The composite risk score based on these factors may assist in patient selection and prognostication in those supported with contemporary LVADs.
Keywords: heart transplantation, survival, ventricular assist device
1 |. INTRODUCTION
Technologic advances in the design of newer generation left ventricular assist devices (LVADs) and improved patient management have resulted in better outcomes when used as a bridging strategy to orthotopic heart transplantation (OHT) for many patients. Nevertheless, LVAD use has been associated with device-related adverse events, such as bleeding and thrombosis, and may increase allosensitization, thus, potentially raising the risk of rejection.1–4 Additionally, there are concerns that the newer allocation policy adopted in October 2018 has lowered rates of transplantation for those supported with durable LVADs.5 Despite this, the use of an LVAD device, as opposed to temporary mechanical circulatory support (MCS), allows for hospital discharge and further candidate optimization for OHT. Thus, it is critical to understand pretransplant factors that may inform posttransplant outcomes. Prior studies have explored the relationship between LVAD support and post-OHT outcomes.6 As an example, the use of a HeartMate II device was associated with improved long-term survival following OHT as compared with patients who did not receive LVAD support in a propensity score-matched analysis.7 In this study, we sought to derive a risk score for 1-year post-transplant mortality in patients bridged with newer generation, contemporary centrifugal LVADs, including the HeartWare ventricular assist device (HVAD) and the HeartMate III (HM3) device. Improved understanding of post-OHT outcomes in these patients will serve to guide management and individualize the risks and benefits in a patient-centered approach.
2 |. METHODS
2.1 |. Study design
Utilizing the United Network for Organ Sharing (UNOS) database, we explored adult patients (≥18 years) bridged with an HVAD or HM3 LVAD who subsequently underwent OHT. We excluded patients requiring concomitant right ventricular support as well as patients undergoing combined heart–lung, heart–kidney, or heart–liver transplantation.
The study population was randomly divided into a derivation and a validation cohort in a 2:1 fashion. The primary outcome was 1-year all-cause mortality following OHT. Secondary endpoints included postoperative renal failure requiring dialysis (i.e., new-onset dialysis during the index hospitalization in those without pre-transplant dialysis), stroke, and permanent pacemaker placement.
2.2 |. Statistical analysis
Continuous data are reported as mean (SD) for Gaussian data and median (interquartile range) for non-Gaussian data. Categorical data are reported as number (%). Pairwise comparison was performed utilizing Student’s t test or Mann–Whitney U test for continuous data and χ2 testing for categorical data. Univariate regression was utilized to obtain unadjusted odds for 1-year mortality in the derivation cohort. Threshold analysis was conducted to determine cut-offs for the categorization of continuous variables. Variables with p < .200 were included in the multivariable model. The c-index and Hosmer–Lemeshow goodness-of-fit tests were utilized to select the most parsimonious model. The absolute values of the odds ratios from the multivariable model were utilized to derive points for individual variables to calculate individual components of the risk score. These individual components were then summed to generate a total score. Three risk categories were defined based on clinical judgment and the distribution of scores: low, moderate, and high risk. Kaplan–Meier analysis was utilized to model 1-year survival. Statistical analyses were performed using the Stata 16 software package (Stata Statistical Software: Release 16; StataCorp). This study was approved by the Institutional Review Board at the University of Pittsburgh.
3 |. RESULTS
3.1 |. Baseline characteristics
A total of 3434 patients were identified of whom 2290 were assigned to the derivation cohort and 1144 were assigned to the validation cohort (Table 1). Overall, baseline characteristics were comparable between the cohorts. Within the derivation cohort, the mean recipient age was 53.5 ± 12.1 years and 1758 (76.8%) patients were male. Almost one-third of patients (n = 711, 31.4%) required ventilator support before transplant and one-fourth (n = 575, 25.1%) required inotropic support. Exploration of functional status revealed that most patients (n = 1278, 57.8%) required full assistance whereas only 257 (11.6%) were considered independent. The majority of patients were bridged with an HVAD device (78.1%) and a smaller proportion with a HeartMate III device (21.9%). The mean donor age was 32 ± 10.7 years and the mean donor ejection fraction was 61.7 ± 6.5%. Sex matching occurred in 1835 (80.1%) while ABO matching occurred in 2046 (89.3%). The mean ischemic time was 3.2 ± 1.1 h.
TABLE 1.
Baseline characteristics in the derivation and validation cohorts of patients bridged to orthotopic heart transplantation with a HeartMate III or HeartWare ventricular assist device (n = 3434)
| Derivation (N = 2290) | Validation (N =1144) | p value | |
|---|---|---|---|
| Recipient variables | |||
| Age (years); mean (SD) | 53.5 (12.1) | 52.9 (12.6) | .158 |
| Male sex | 1758 (76.8) | 865 (75.6) | .452 |
| Race/ethnicity | .242 | ||
| White | 1523 (66.7) | 746 (65.6) | |
| Black | 518 (22.7) | 245 (21.6) | |
| Hispanic | 158 (6.9) | 98 (8.6) | |
| Other | 83 (3.6) | 48 (4.2) | |
| Body mass index (kg/m2); mean (SD) | 28.1 (4.8) | 28.4 (4.8) | .074 |
| Ventilator support | 711 (31.4) | 336 (29.7) | .329 |
| Extracorporeal membrane oxygenation support | 11 (0.5) | 3 (0.3) | .410 |
| Intra-aortic balloon pump support | 12 (0.5) | 4 (0.4) | .601 |
| Inotropic support | 575 (25.1) | 302 (26.4) | .414 |
| Calculated panel reactive antibodies | 0 (0–10) | 0 (0–7) | .433 |
| Dialysis dependence | 41 (1.8) | 29 (2.5) | .146 |
| Diabetes | 688 (30.1) | 328 (28.7) | .398 |
| Serum creatinine (mg/dl); mean (SD) | 1.2 (0.4) | 1.2 (0.4) | .998 |
| Serum albumin (mg/dl); mean (SD) | 3.6 (0.6) | 3.6 (0.6) | .768 |
| Serum bilirubin (mg/dl); mean (SD) | 0.6 (0.4–0.9) | 0.6 (0.4–0.9) | .586 |
| Mean pulmonary artery pressure (mmHg); mean (SD) | 29.3 (10.5) | 29.9 (10.6) | .115 |
| Pulmonary capillary wedge pressure (mmHg); mean (SD) | 19.3 (9.1) | 19.9 (9.3) | .047 |
| Transpulmonary gradient (mmHg); mean (SD) | 10.1 (5.7) | 10.1 (5.4) | .940 |
| Functional status | .939 | ||
| Full assistance | 1278 (57.8) | 626 (57.1) | |
| Moderate assistance | 678 (30.6) | 340 (31.0) | |
| Independent | 257 (11.6) | 130 (11.9) | |
| Left ventricular assist device | .653 | ||
| HeartWare | 1789 (78.1) | 886 (77.5) | |
| HeartMate III | 501 (21.9) | 258 (22.6) | |
| Donor variables | |||
| Age | 32.0 (10.7) | 32.2 (10.8) | .636 |
| Smoking history | 246 (10.9) | 140 (12.5) | .157 |
| Diabetes mellitus | 67 (2.9) | 42 (3.7) | .231 |
| Hypertension | 799 (35.0) | 401 (35.1) | .932 |
| Left ventricular ejection fraction (%); mean (SD) | 61.7 (6.5) | 61.8 (6.7) | .499 |
| Mechanism of death | .974 | ||
| Trauma | 1052 (45.9) | 519 (45.4) | |
| Cerebrovascular accident | 361 (15.8) | 182 (15.9) | |
| Cardiovascular, aspiration, natural causes | 373 (16.3) | 184 (16.1) | |
| Other | 504 (22.0) | 259 (22.6) | |
| Transplant variables | |||
| Race-matched | 1243 (54.3) | 620 (54.2) | .963 |
| Human leukocyte antigen (HLA)-matched | 1943 (95.8) | 967 (97.5) | .018 |
| ABO-matched | 2046 (89.3) | 1032 (90.2) | .433 |
| Cytomegalovirus-matched | 994 (43.6) | 492 (43.3) | .857 |
| Sex-matched | 1835 (80.1) | 916 (80.1) | .966 |
| Donor-to-recipient body mass index ratio | 1.00 (0.24) | 1.01 (0.25) | .235 |
| Ischemic time (h); mean (SD) | 3.2 (1.1) | 3.1 (1.0) | .144 |
Note: Data are shown as number (%) except where indicated.
3.2 |. Univariate and multivariable analysis
Univariate analysis of variables with a p < .200 is shown (Table 2). By univariate analysis, age ≥50 years (odds ratio [OR], 2.11; p < .001), BMI 25–34.9 (OR, 1.70; p = .005), preoperative extracorporeal membrane oxygenation (OR, 6.03; p = .019), preoperative intra-aortic balloon pump (OR, 6.46; p = .006), dialysis (OR, 3.58; p = .001), diabetes (OR, 1.51; p = .009), total bilirubin ≥2.4 mg/dl (OR, 3.40; p < .001), pulmonary capillary wedge pressure (OR, 0.98; p = .031), hospitalization at the time of transplant (OR, 2.13; p < .001), need for intensive care unit care before transplant (OR, 2.63; p < .001), donor hypertension (OR, 0.71; p = .046), center volume (OR, 0.97; p = .043), and ischemic time ≥4 h (OR, 1.61; p = .008) were associated with 1-year post-OHT mortality. Multivariable logistic regression revealed that age ≥50 years (OR, 1.84; p = .006), the need for hemodialysis pre-operatively (OR, 2.92; p = .024), total bilirubin ≥2.4 mg/dl (OR, 2.42; p = .038), and ischemic time ≥4 h (OR, 1.63; p = .017) were independently associated with 1-year mortality following OHT. The final model had a c-index of 0.68 in the validation cohort and the Hosmer–Lemeshow goodness-of-fit test had a nonsignificant p value (.353) indicating appropriate fit.
TABLE 2.
Generation of risk stratification score by univariate and multivariable logistic regression for one-year mortality
| Univariate analysis; OR (95% CI) | p value | Multivariable analysis; OR (95% CI) | p value | Points assigned | |
|---|---|---|---|---|---|
| Age (years) | |||||
| <50 | Reference | Reference | Reference | Reference | 0 |
| ≥50 | 2.11 (1.45–3.08) | <.001 | 1.84 (1.17–2.86) | .006 | 2 |
| Body mass index (kg/m2) | |||||
| <25 | Reference | Reference | Reference | Reference | - |
| 25–34.9 | 1.70 (1.17–2.47) | .005 | 1.36 (0.90–2.05) | .149 | - |
| ≥35 | 1.48 (0.80–2.73) | .214 | 1.37 (0.67–2.81) | .394 | - |
| Preoperative mechanical ventilation | 7.97 (0.50–128.00) | .143 | 2.70 (0.08–89.24) | .578 | - |
| Preoperative ECMO | 6.03 (1.34–27.15) | .019 | 3.18 (0.53–19.06) | .205 | - |
| Preoperative IABP | 6.46 (1.72–24.27) | .006 | 2.12 (0.24–18.54) | .497 | - |
| Dialysis dependence | |||||
| Yes | Reference | Reference | Reference | Reference | 0 |
| No | 3.58 (1.68–7.65) | .001 | 2.92 (1.15–7.42) | .024 | 3 |
| Diabetes | 1.51 (1.11–2.06) | .009 | 1.19 (0.82–1.73) | .362 | - |
| Bilirubin | |||||
| <2.4mg/dl | Reference | Reference | Reference | Reference | 0 |
| <2.4mg/dl | 3.40 (1.83–6.32) | <.001 | 2.42 (1.05–5.58) | .038 | 2 |
| Pulmonary capillary wedge pressure | 0.98 (0.96–0.99) | .031 | 1.00 (0.97–1.02) | .696 | - |
| Hospitalized | 2.13 (1.53–2.96) | <.001 | 1.48 (0.90–2.45) | .124 | - |
| ICU before transplant | 2.63 (1.70–4.08) | <.001 | 1.66 (0.86–3.22) | .132 | - |
| Donor hypertension | 0.71 (0.51–0.99) | .046 | 0.76 (0.52–1.11) | .154 | - |
| HLA-matched | 4.00 (0.97–16.52) | .055 | 3.96 (0.93–16.95) | .064 | - |
| Sex-matched | 1.37 (0.91–2.05) | .128 | 1.28 (0.82–2.01) | .281 | - |
| Center volume (increasing volume) | 0.97 (0.94–0.99) | .043 | 0.99 (0.96–1.02) | .390 | - |
| Ischemic time (h) | |||||
| <4 | Reference | Reference | Reference | Reference | 0 |
| ≥4 | 1.61 (1.13–2.29) | .008 | 1.63 (1.09–2.44) | .017 | 2 |
Abbreviations: CI, confidence interval; ECMO, extracorporeal membrane oxygenation; HLA, human leukocyte antigen; IABP, intra-aortic balloon pump; ICU, intensive care unit; OR, odds ratio.
3.3 |. Generation of the risk score and stratification of outcomes
Utilizing the absolute values of the odds ratios for the significant variables, a final risk score was generated with 2 points allotted to age ≥50 years, 3 points for dialysis, 2 points for total bilirubin ≥2.4 mg/dl, and 2 points for ischemic time ≥4 h for a total of 9 possible points (Table 3). The distribution of total risk scores for the derivation and validation cohorts demonstrates that the majority of patients scored between 2 and 3 total points (Figure 1). The mean score in both the derivation and validation cohorts was 1.9 ± 1.3 points. Three risk stratification groups were generated: low risk (0 risk factors), moderate risk (1 risk factor), and high risk (≥2 risk factors; Table 4). The predicted overall mortality rates were 10.1%, 14.8%, 24.8% as compared with the observed mortality rates in the validation cohort of 9.6%, 11.8%, and 21.7% in the three risk groups, respectively. Among the three risk groups, new-onset renal failure requiring dialysis (9.9%, 11.4%, 20.6%; p = .009) and rates of stroke (2.8%, 2.8%, 9.7%; p = .001) were significantly different (Table 5). The median length of stay was similar at 16 days in the low and moderate-risk groups and 17 days in the high-risk group (p = .191). Thirty-day (1.4%, 2.6%, 7.4%), 90-day (2.8%, 4.4%, 12.5%), and 1-year (5.0%, 6.7%, 14.8%) mortality were significantly different between the groups. Kaplan–Meier analysis among the groups demonstrated a log-rank p <.001 indicating significant differences in mortality up to 1 year (Figure 2).
TABLE 3.
Final risk score for patients bridged with a HeartWare or HeartMate III device undergoing heart transplantation
| Risk factor | Points assigned |
|---|---|
| Age (years) | |
| <50 | 0 |
| ≥50 | 2 |
| Dialysis | |
| No | 0 |
| Yes | 3 |
| Bilirubin | |
| <2.4 mg/dl | 0 |
| <2.4 mg/dl | 2 |
| Ischemic time (h) | |
| <4 | 0 |
| ≥4 | 2 |
| Total potential points | 9 |
FIGURE 1.
Bar graph demonstrating the distribution of total risk scores in the derivation and validation cohorts
TABLE 4.
Risk group categorization for patients bridged with a HeartWare or HeartMate III device undergoing heart transplantation
| Low risk (0 risk factors) | Moderate risk (1 risk factor) | High risk (≥2 risk factors) | |
|---|---|---|---|
| Number in derivation group | 545 (23.8) | 1368 (59.7) | 377 (16.5) |
| Number in validation group | 282 (24.7) | 686 (60.0) | 176 (15.4) |
| Total number | 827 (24.1) | 2054 (59.8) | 553 (16.1) |
| Predicted overall mortality | 55 (10.1) | 202 (14.8) | 77 (20.4) |
| Observed overall mortality in validation cohort | 27 (9.6) | 81 (11.8) | 38 (21.7) |
TABLE 5.
Outcomes stratified by risk categories in the validation cohort
| Low risk (N = 282) | Moderate risk (N = 686) | High risk (N =176) | p value | |
|---|---|---|---|---|
| Renal failure requiring dialysis | 28 (9.9) | 78 (11.4) | 36 (20.6) | .009 |
| Stroke | 8 (2.8) | 19 (2.8) | 17 (9.7) | .001 |
| Pacemaker | 9 (3.2) | 16 (2.3) | 7 (4.0) | .645 |
| Rejection requiring treatment | 61 (27.4) | 98 (19.0) | 24 (19.2) | .031 |
| Median length of stay | 16 (11–23) | 16 (12–26) | 17 (12–29) | .191 |
| Mortality 30-day | 4 (1.4) | 18 (2.6) | 13 (7.4) | .001 |
| Mortality 90-day | 8 (2.8) | 30 (4.4) | 22 (12.5) | <.001 |
| Mortality 1-year | 14 (5.0) | 46 (6.7) | 26 (14.8) | <.001 |
FIGURE 2.
Kaplan–Meier analysis of 1-year mortality in the low-, moderate-, and high-risk groups
4 |. CONCLUSIONS
MCS has emerged as an important management strategy in the approach to patients with end-stage heart disease. In particular, the use of LVADs allows for outpatient management and greater independence among those patients being bridged before heart transplantation. Nevertheless, the impact of pre-transplant characteristics on post-OHT outcomes among patients who are bridged with an LVAD is not well defined. In a study of patients bridged with newer generation centrifugal LVADs (i.e., HVAD, HM3) utilizing the UNOS database, we derived a risk score to predict 1-year post-OHT mortality. We found that older age, end-stage renal disease, higher total bilirubin levels, and a longer organ ischemic time were predictive of post-OHT death. The risk score also correlated with post-OHT rates of new-onset renal failure requiring dialysis as well as postoperative stroke. Thus, an improved understanding of key factors impacting post-OHT mortality in patients bridged with a newer generation LVAD will enhance informed consent discussions and also serves to guide further exploration into risk modification to improve outcomes under this management approach.
Among our three risk groups, 1-year mortality ranged from 5.0% to 14.8%. New-onset renal failure as well as stroke may represent contributors to these elevated rates of death. Though risk estimation should ideally be patient-specific, these estimates provide a relative reference when counseling patients regarding potential outcomes. Prior studies have explored and validated risk scoring tools for post-OHT mortality, yet, it is unclear whether these scores are applicable to all patient populations undergoing transplantation, particularly among those requiring MCS, including newer generation devices.8–10 The relationship between LVAD infection and post-OHT outcomes has also been examined. Tong et al.11 found no significant differences in the likelihood of transplant, survival to transplant, and post-transplant death among patients with and without an LVAD infection. Despite this, patients requiring prolonged LVAD support have been demonstrated to have lower post-transplant survival at 3 years (68% vs. 88%).3 Interestingly, we did observe a correlation between the risk categories and rates of rejection requiring treatment. Multiple studies have described the allosensitization associated with MCS, though this has not been definitively linked to post-transplant survival.4,12 Taken together, it is clear that both the potential advantages and adverse effects of a bridge-to-transplant approach with a centrifugal LVAD warrant specific inquiry into post-transplant mortality. The development of this risk score represents an initial step toward improving understanding of a subset of patients who might derive the greatest benefit from pre-transplant HVAD or HM3 implantation. This is particularly relevant in the modern era in which allocation policy makes it more challenging to transplant stable patients with durable LVADs.5
Increased recipient age was a critical component of our risk score and has previously been explored in the context of outcomes following both LVAD implantation and OHT. In a study of patients ≥60 years bridged with an LVAD device, post-OHT survival at 30 days and 1 year was significantly lower among those patients who were bridged with an LVAD as compared with patients requiring inotropic agents or in those who did not require pre-transplant support.13 These populations were not matched, however, thus it is plausible that those requiring MCS had a worse cardiac function. Nevertheless, others have shown that increased age is associated with inferior post-transplant outcomes which calls into question whether the use of an LVAD can potentially modify these risks.14,15 In selected older patient populations, however, favorable transplant outcomes can be achieved, and furthermore, studies have shown comparable mortality among various age groups undergoing LVAD implantation.16,17 Utilizing our scoring system, patients ≥50 years immediately fall into the moderate-risk category and the presence of any other risk factor would categorize them as high risk thus almost tripling their predicted 1-year mortality compared with the lowest risk group. A prospective exploration of this risk scoring system would elucidate its utility in clinical decision-making. Organ ischemic time is one factor that should be considered at the time of organ offer, particularly in these elderly patients.
Risk scores such as this one can be applied to the iterative process of assessing the potential outcomes following OHT in patients bridged with an LVAD. Prior work has associated the duration of LVAD support with post-OHT outcomes, finding that patients requiring a longer duration of support tend to have increased post-transplant mortality.3 Though the durability of LVAD devices, particularly with evolving technology, makes them an appealing choice for long-term management, prolonged support should be accompanied by frequent reassessment of transplant risks including discussions with the patients on any significant perceived changes in perioperative risk.18 This is particularly critical given the potential for shifts in organ allocation policy, which may decrease organ availability in these patients requiring HVAD or HM3 support.19,20 Thus, it is plausible that the recalculation of a risk score may uncover a change in risk categorization. Targeted inquiry into the incorporation of risk estimation for post-OHT mortality in the decision for either a bridge-to-transplant or destination therapy approach may help to clarify the management of patients with borderline risk.
Improved understanding of the relative risks of post-OHT mortality and relevant risk factors can also be applied to guide patient selection for bridge-to-transplant. To date, there have been few studies exploring the relative effectiveness of currently available MCS devices and thus device selection and timing of implantation may be subject to bias.21 It is plausible that patients who fall under the high-risk category may derive benefit from continued medical management and LVAD optimization rather than proceeding with OHT. An interesting and provocative concept would be to guide policy changes in further organ allocation schema using such risk stratification tools. This is critically important given that the most recent change to the UNOS allocation policy has resulted in organ sharing from greater distances with associated increases in organ ischemic time.22 As the latter was found to be predictive of outcomes, calculation of estimated ischemic time as a factor in postoperative outcomes should be considered.23 Furthermore, the investigation into ex vivo organ preservation with greater ischemic times may dampen the effects of the insult on overall mortality outcomes.24
This study has several limitations. Due to the retrospective nature of the study design, the study was limited and may be subject to data entry errors. Additionally, we were unable to obtain granular data on LVAD or postoperative complications which may further elucidate the influence of the risk components on ultimate outcomes, including the causes of death. We were also unable to assess the influence of right ventricular function in these patients nor were we able to assess the chronicity of renal failure in those requiring dialysis preoperatively. The 2018 allocation policy change may have impacted patient selection and practice patterns for LVAD implantation; this could not be fully accounted for. Finally, detail regarding decision-making with device timing and selection as well as the timing of OHT listing was not able to be assessed. It is possible that the selection of patients for HVAD or HM3 implantation may have been biased and thus these results should be interpreted with caution when considering all patients with end-stage heart failure who require support.
In conclusion, we report the results of the derivation of a risk score to predict 1-year post-transplant mortality among patients bridged to transplant with a HeartWare or HM3 LVAD. In particular, we found that older recipient age, end-stage renal disease, elevated bilirubin, and a longer ischemic time were independently predictive of elevated mortality. Thus, patients with these risk factors should be counseled about the anticipated postoperative course following OHT. Additionally, this study guides future investigations into risk modification and patient selection for the use of centrifugal LVADs before OHT.
Footnotes
This manuscript was accepted for presentation at the International Society of Heart and Lung Transplantation 2020 Annual Meeting.
CONFLICT OF INTERESTS
Dr. Arman Kilic serves on a Medical Advisory Board for Medtronic, Inc. The remaining authors declare that there are no conflict of interests.
DATA AVAILABILITY STATEMENT
The data are available upon request from the corresponding author.
REFERENCES
- 1.Truby LK, Garan AR, Givens RC, et al. Ventricular assist device utilization in heart transplant candidates: nationwide variability and impact on waitlist outcomes. Circ Heart Fail. 2018;11:1–9. 10.1161/CIRCHEARTFAILURE.117.004586 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Levine A, Gass A. Third-generation LVADs: has anything changed? Cardiol Rev. 2019;27:293–301. 10.1097/crd.0000000000000268 [DOI] [PubMed] [Google Scholar]
- 3.Takeda K, Takayama H, Kalesan B, et al. Outcome of cardiac transplantation in patients requiring prolonged continuous-flow left ventricular assist device support. J Heart Lung Transplant. 2015;34: 89–99. 10.1016/j.healun.2014.09.007 [DOI] [PubMed] [Google Scholar]
- 4.Chiu P, Schaffer JM, Oyer PE, et al. Influence of durable mechanical circulatory support and allosensitization on mortality after heart transplantation. J Heart Lung Transplant. 2016;35:731–742. 10.1016/j.healun.2015.12.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kilic A, Hickey G, Mathier MA, et al. Outcomes of the first 1300 adult heart transplants in the United States after the allocation policy change. Circulation. 2020;141:1662–1664. [DOI] [PubMed] [Google Scholar]
- 6.Truby LK, Farr MA, Garan AR, et al. Impact of bridge to transplantation with continuous-flow left ventricular assist devices on posttransplantation mortality: a propensity-matched analysis of the United Network of Organ Sharing Database. Circulation. 2019;140: 459–469. 10.1161/CIRCULATIONAHA.118.036932 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Donneyong M, Cheng A, Trivedi JR, et al. The association of pretransplant HeartMate II left ventricular assist device placement and heart transplantation mortality. ASAIO J. 2014;60:294–299. 10.1097/MAT.0000000000000065 [DOI] [PubMed] [Google Scholar]
- 8.Kilic A, Allen JG, Weiss ES. Validation of the United States-derived Index for Mortality Prediction After Cardiac Transplantation (IMPACT) using international registry data. J Heart Lung Transplant. 2013;32:492–498. 10.1016/j.healun.2013.02.001 [DOI] [PubMed] [Google Scholar]
- 9.Singh TP, Almond CS, Semigran MJ. Piercey G, Gauvreau K. Risk prediction for early in-hospital mortality following heart transplantation in the United States. Circ Heart Fail. 2012;5:259–266. 10.1161/CIRCHEARTFAILURE.111.965996 [DOI] [PubMed] [Google Scholar]
- 10.Weiss ES, Allen JG, Arnaoutakis GJ, et al. Creation of a quantitative recipient risk Index for Mortality Prediction After Cardiac Transplantation (IMPACT). Ann Thorac Surg. 2011;92:914–922. 10.1016/j.athoracsur.2011.04.030 [DOI] [PubMed] [Google Scholar]
- 11.Tong MZ, Smedira NG, Soltesz EG, et al. Outcomes of heart transplant after left ventricular assist device specific and related infection. Ann Thorac Surg. 2015;100:1292–1297. 10.1016/j.athoracsur.2015.04.047 [DOI] [PubMed] [Google Scholar]
- 12.Joyce DL, Southard RE, Torre-Amione G, Noon GP, Land GA, Loebe M. Impact of left venticular assist device (LVAD)-mediated humoral sensitization on post-transplant outcomes. J Heart Lung Transplant. 2005;24:2054–2059. 10.1016/j.healun.2005.06.028 [DOI] [PubMed] [Google Scholar]
- 13.Allen JG, Kilic A, Weiss ES, et al. Should patients 60 years and older undergo bridge to transplantation with continuous-flow left ventricular assist devices? Ann Thorac Surg. 2012;94:2017–2024. 10.1016/j.athoracsur.2012.06.009 [DOI] [PubMed] [Google Scholar]
- 14.Hong KN, Iribarne A, Worku B, et al. Who is the high-risk recipient? Predicting mortality after heart transplant using pretransplant donor and recipient risk factors. Ann Thorac Surg. 2011;92:520–527. 10.1016/j.athoracsur.2011.02.086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wever-Pinzon O, Edwards LB, Taylor DO, et al. Association of recipient age and causes of heart transplant mortality: implications for personalization of post-transplant management–an analysis of the International Society for Heart and Lung Transplantation Registry. J Heart Lung Transplant. 2017;36:407–417. 10.1016/j.healun.2016.08.008 [DOI] [PubMed] [Google Scholar]
- 16.Crespo-Leiro MG, Paniagua-Martín MJ, Muñiz J, et al. Long-term results of heart transplant in recipients older and younger than 65 years: a comparative study of mortality, rejections, and neoplasia in a cohort of 445 patients. Transplant Proc. 2005;37:4031–4032. 10.1016/j.transproceed.2005.09.158 [DOI] [PubMed] [Google Scholar]
- 17.Huang R, Deng M, Rogers JG, et al. Effect of age on outcomes after left ventricular assist device placement. Transplant Proc. 2006;38: 1496–1498. 10.1016/j.transproceed.2006.02.115 [DOI] [PubMed] [Google Scholar]
- 18.Quader MA, Wolfe LG, Kasirajan V. Heart transplantation outcomes in patients with continuous-flow left ventricular assist device-related complications. J Heart Lung Transplant. 2015;34:75–81. 10.1016/j.healun.2014.07.015 [DOI] [PubMed] [Google Scholar]
- 19.Hanff TC, Harhay MO, Kimmel SE, et al. Trends in mechanical support use as a bridge to adult heart transplant under new allocation rules. JAMA Cardiol. 2020;5:728–729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Uriel N, Jorde UP, Woo Pak S, et al. Impact of long term left ventricular assist device therapy on donor allocation in cardiac transplantation. J Heart Lung Transplant. 2013;32:188–195. 10.1016/j.healun.2012.11.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Theochari CA, Michalopoulos G, Oikonomou EK, et al. Heart transplantation versus left ventricular assist devices as destination therapy or bridge to transplantation for 1-year mortality: a systematic review and meta-analysis. Ann Cardiothorac Surg. 2018;7: 3–11. 10.21037/acs.2017.09.18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Vucicevic D, Lum C, Rahman J, et al. Transplant survival and competing outcomes after UNOS allocation policy change. J Heart Lung Transplant. 2020;39:S72. [Google Scholar]
- 23.Russo MJ, Iribarne A, Hong KN, et al. Factors associated with primary graft failure after heart transplantation. Transplantation. 2010;90:444–450. 10.1097/TP.0b013e3181e6f1eb [DOI] [PubMed] [Google Scholar]
- 24.Sponga S, Ius F, Ferrara V, et al. Normothermic ex-vivo perfusion for donor heart preservation in transplantation of patients bridged with ventricular assist devices. J Heart Lung Transplant. 2020;39:S245. 10.1016/j.healun.2020.01.926 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data are available upon request from the corresponding author.