Abstract
Objective:
Implantation of donor hearts with prolonged ischemic times is associated with worse survival. We sought to identify risk factors that modulate the effects of prolonged preservation.
Methods:
Retrospective review of the United Network for Organ Sharing database (2000–2018) to identify transplants with >5 (n = 1526) or ≤5h (n = 35,733) of donor heart preservation. In transplanted hearts preserved for >5 h, Cox-proportional hazards identify modifiers for survival.
Results:
Compared to ≤5 h, transplanted patients with >5 h of preservation spent less time in status 1B (76 ± 160 vs. 85 ± 173 days, p = .027), more commonly had ischemic cardiomyopathy (42.3% vs. 38.3%, p = .002), and less commonly received a blood type O heart (45.4% vs. 50.8%, p < .001). Longer heart preservation time was associated with a higher incidence of postoperative stroke (4.5% vs. 2.5%, p < .001), and dialysis (16.4% vs. 10.6%, p < .001). Prolonged preservation was associated with a greater likelihood of death from primary graft dysfunction (2.8% vs. 1.5%, p < .001) but there was no difference in death from acute (2.0% vs. 1.7%, p = .402) or chronic rejection (2.0% vs. 1.9%, p = .618). In transplanted patients with >5 h of heart preservation, multivariable analysis identified greater mortality with ischemic cardiomyopathy etiology (hazard ratio [HR] = 1.36, p < 0.01), pre-transplant dialysis (HR = 1.84, p < .01), pre-transplant extracorporeal membrane oxygenation (ECMO, HR = 2.36, p = .09), and O blood type donor hearts (HR = 1.35, p < .01).
Conclusion:
Preservation time >5 h is associated with worse survival. This mortality risk is further amplified by preoperative dialysis and ECMO, ischemic cardiomyopathy etiology, and use of O blood type donor hearts.
Keywords: heart transplant, ischemia/reperfusion injury, outcomes, primary graft dysfunction
1 |. INTRODUCTION
Surgical techniques for orthotopic heart transplants have changed little over the past few decades. With increasing experience, however, important advances have occurred in donor selection and optimization1 as well as post-transplant management of the recipient.2 Importantly, expanded donor criteria have been embraced by some transplant centers and mechanical circulatory assist devices have allowed for bridging to transplantation. Based on clinical experience, the generally accepted threshold for donor heart ischemic time (IT) is 4 h. An IT exceeding 6 h is certainly highly discouraged.1,3
The primary concern with prolonged IT is its association with primary graft dysfunction (PGD). PGD is a significant challenge that occurs in 10%–20% of heart transplant patients. While 1-month mortality following heart transplant is approximately 8%, PGD accounts for 39% of these early deaths.4 Donor heart characteristics are often the focus when attempting to minimize the risk of PGD, however, recipient characteristics are also critically important. A “Hostile” recipient environment can negatively impact graft function. Previously described recipient PGD risk factors include pre-transplant ventilatory support, extracorporeal membrane oxygenation (ECMO), and elevated pulmonary vascular resistance. Beyond isolated recipient and donor factors, appropriate matching for factors such as gender and body habitus are also critically important.5
With the recent adoption of ex-vivo normothermic perfusion by a limited number of centers, increases in allowable out-of-body time for the donor heart were achieved.6 However, the large majority of donor hearts are currently preserved with traditional hypothermic mechanical arrest. Achieving donor heart perfusion within 4–5 h following cold preservation is recommended guideline by multiple prior studies.7 Therefore a 5 h storage time cutoff for this study would be generally accepted to be prolonged. Despite this, there are likely factors that further modulate this risk when considering the use of donor hearts exposed to more prolonged preservation times. This study aims to identify key factors that should be taken into account when assessing a donor heart that is likely to experience an IT of greater than 5 h.
2 |. MATERIALS AND METHODS
2.1 |. Patients
Study approval was obtained from the University of Michigan Institutional Review Board with a waiver of informed consent (IRB#HUM00191544). We reviewed the United Network for Organ Sharing-Standard Transplant Analysis and Research (UNOS-STAR) database and analyzed 37,259 heart transplants from January 1, 2000 to September 30, 2018. This covers the period before the change in donor heart allocation algorithm implemented in October 2018.
Patients less than 18 years of age and patients undergoing simultaneous lung transplants were excluded from study. The patient population of interest are those with >5 h donor heart IT (n = 1526) with the control group being IT of ≤5h (n = 35,733). We also determined the propensity for different UNOS regions to transplant hearts exposed to prolonged IT and examined outcomes. The primary outcomes for the study were patient survival with secondary outcomes consisting of postoperative complications including stroke, new dialysis, permanent pacemaker implant as well as death from primary graft dysfunction and rejection.
2.2 |. Statistical methods
Categorical variables were analyzed with the Pearson X2 test. Student’s t-test or Wilcoxon rank-sum test was applied to compare continuous variables. A p < .05 was considered significant. Kaplan–Meier survival curves with Breslow (Generalized Wilcoxon) statistics as well as Cox-proportional hazards analysis with univariate analysis as well as forward and backward multivariable regression were used to analyze survival data. Statistical analysis was performed using the Statistical Package for the Social Sciences software (SPSS Inc.).
3 |. RESULTS
3.1 |. Recipient and donor demographics and comorbidities
Comparing populations with preservation times >5 versus ≤5 h respectively (Table 1), there was no difference in recipient age (51.91 ± 13.74 vs. 52.70 ± 12.55, p = .029), days spent in status 1A (1.37 ± 0.87 vs. 1.37 ± 0.93, p = .907), LVAD use (29.6% vs. 28.5%, p = .366), pre-transplant ECMO (0.7% vs. 0.6%, p = .608), nor pre-transplant hemodynamics (p > .5). No difference was seen in donor age (31.83 ± 12.16 vs. 31.76 ± 11.79, p = .829), incidence of coronary artery disease (23.9% vs. 24.9%, p = .376), creatinine (1.36 ± 1.42 vs. 1.33 ± 1.27, p = .370), diabetes (2.8% vs. 3.0%, p = .586) nor hypertension (13.5% vs. 13.9%, p = .686).
TABLE 1.
Transplant recipient and donor demographics
| Predictors of survival - Univariate | Ischemic time >5 h (n = 1526) | Ischemic time ≤5 h (n = 35,733) | p |
|---|---|---|---|
| Recipient: Demographics: Age | 51.91 ± 13.74 | 52.70 ± 12.55 | .029 |
| Male | 1177 (77.1%) | 26,733 (74.8%) | .041 |
| Weight (kg) | 82.64 ± 19.50 | 81.99 ± 17.65 | .206 |
| Height (cm) | 174.11 ± 10.49 | 173.75 ± 9.84 | .197 |
| BMI | 27.08 ± 5.13 | 27.04 ± 4.85 | .783 |
| Comorbidities: Creatinine (mg/dl) | 1.37 ± 0.87 | 1.37 ± 0.93 | .907 |
| Total Bilirubin (mg/dl) | 1.27 ± 3.08 | 1.10 ± 1.94 | .041 |
| Diabetes | 297 (19.5%) | 7400 (20.7%) | .239 |
| Dialysis | 76 (5.0%) | 1595 (4.5%) | .340 |
| Cerebrovascular disease | 73 (4.8%) | 1786 (5.0%) | .706 |
| Malignancy history | 109 (7.1%) | 2,517 (7.0%) | .882 |
| Days in status: 1A | 24.13 ± 48.05 | 25.99 ± 53.11 | .144 |
| 1B | 76.04 ± 160.05 | 85.35 ± 172.78 | .027 |
| 2 | 83.46 ± 254.32 | 82.61 ± 236.04 | .891 |
| Pre-transplant mechanical support: LVAD | 451 (29.6%) | 10,179 (28.5%) | .366 |
| RVAD | 6 (0.4%) | 62 (0.2%) | .049 |
| BIVAD or TAH | 72 (4.7%) | 1091 (3.1%) | <.001 |
| ECMO | 11 (0.7%) | 220 (0.6%) | .608 |
| IABP | 79 (5.2%) | 2142 (6.0%) | .187 |
| Heart failure etiology: Ischemic | 645 (42.3%) | 13,685 (38.3%) | .002 |
| Nonischemic | 737 (48.3%) | 20,033 (56.1%) | <.001 |
| Congenital | 91 (6.0%) | 954 (2.7%) | <.001 |
| Cardiac graft failure | 53 (3.5%) | 1061 (3.0%) | .258 |
| Simultaneous transplant: Kidney | 40 (2.6%) | 1346 (3.8%) | .021 |
| Liver | 6 (0.4%) | 232 (0.6%) | .219 |
| Catheterization: Cardiac Output (L/min) | 4.56 ± 1.51 | 4.54 ± 1.48 | .712 |
| PA Systolic (mmHg) | 40.92 ± 14.18 | 40.88 ± 13.98 | .926 |
| PA Diastolic (mmHg) | 19.78 ± 8.18 | 19.62 ± 8.46 | .492 |
| PA Mean (mmHg) | 27.70 ± 9.85 | 27.70 ± 9.99 | .995 |
| PCWP (mmHg) | 18.46 ± 8.60 | 18.32 ± 8.74 | .544 |
| Donor: Demographics: Age | 31.83 ± 12.16 | 31.76 ± 11.79 | .829 |
| Male | 1063 (69.7%) | 25,294 (70.8%) | .343 |
| Weight (kg) | 80.86 ± 19.39 | 81.88 ± 18.61 | .044 |
| Height (cm) | 173.96 ± 10.32 | 174.21 ± 9.60 | .357 |
| BMI | 26.67 ± 5.92 | 26.96 ± 5.70 | .051 |
| Blood type: A | 583 (38.2%) | 13,007 (36.4%) | .152 |
| B | 197 (12.9%) | 3845 (10.8%) | .008 |
| AB | 743 (2.1%) | 53 (3.5%) | <.001 |
| O | 693 (45.4%) | 18,138 (50.8%) | <.001 |
| Donor heart: Ischemic time (h) | 5.77 ± 0.89 | 3.06 ± 0.90 | <.001 |
| LVEF (%) | 62.58 ± 7.74 | 61.59 ± 7.23 | <.001 |
| Coronary artery disease on angiogram | 365 (23.9%) | 8904 (24.9%) | .376 |
| Comorbidities: Creatinine | 1.36 ± 1.42 | 1.33 ± 1.27 | .370 |
| Diabetes | 42 (2.8%) | 1070 (3.0%) | .586 |
| Hypertension | 206 (13.5%) | 4954 (13.9%) | .686 |
Abbreviations: BIVAD, biventricular assist device; IABP, intra-aortic balloon pump; LVAD, left ventricular assist device; LVEF, left ventricular ejection fraction; PA, pulmonary artery; PCWP, pulmonary capillary wedge pressure; RVAD, right ventricular assist device; TAH, total artificial heart.
Patients using donor hearts with >5 h of preservation time tended to have a higher total bilirubin (1.27 ± 3.08 vs. 1.10 ± 1.94, p = .041), higher incidence of total artificial heart or biventricular mechanical support (4.7% vs. 3.1%, p < .001), as well as a higher incidence of ischemic cardiomyopathy (42.3% vs. 38.3%, p = .002) and congenital heart disease (6.0% vs. 2.7%, p < .001). This group also spent less time on status 1B (76.04 ± 160.05 vs. 85.35 ± 172.78 days, p = .027), had a lower incidence of nonischemic cardiomyopathy (48.3% vs. 56.1%, p < .001), and a lower rate of utilization of blood type AB (2.1% vs. 3.5%, p < .001) and O (45.4% vs. 50.8%, p < .001) hearts (Table 1). IT in the prolonged preservation population was 5.77 ± 0.89 versus 3.06 ± 0.90h in the control group. The frequency of IT for donor hearts with ≤5h (Figure 1A) versus >5 (Figure 1B) hours are displayed as histograms (standard deviation = 1.046). While 76 patients had pre-transplant dialysis, only 40 patients had documented simultaneous kidney transplant.
FIGURE 1.
Histogram showing frequency of donor ischemic times (A) ≤5 h and (B) >5 h. Frequency distribution curve is for the entire study population (n = 37,259) with a standard deviation of 1.046
3.2 |. Postoperative outcomes
Patients transplanted with hearts preserved for >5 h had a higher rate of postoperative complications including stroke (4.5% vs. 2.5%, p < .001), new dialysis requirement (16.4% vs. 10.6%, p < .001), and permanent pacemaker implant (4.1% vs. 3.2%, p = .056). Death from PGD was also incrementally more common with prolonged preservation times (2.8% vs. 1.5%, p < .001). There were no differences in terms of death from acute (2.0% vs. 1.7%, p = .402) or chronic (2.0% vs. 1.9%, p = .618) rejection nor the rate of acute rejection overall (81.5% vs. 81.2%, p = .824, Table 2).
TABLE 2.
Postoperative outcomes with 5 h donor heart ischemic time threshold
| Postoperative outcomes | Ischemic time >5 h (n = 1526) | Ischemic time ≤5 h (n = 35,733) | p |
|---|---|---|---|
| Postoperative complication: Stroke | 69 (4.5%) | 883 (2.5%) | <.001 |
| Dialysis | 251 (16.4%) | 3804 (10.6%) | <.001 |
| Pacemaker | 62 (4.1%) | 1137 (3.2%) | .056 |
| Death from: Primary Graft Dysfunction | 42 (2.8%) | 531 (1.5%) | <.001 |
| Acute Rejection | 31 (2.0%) | 623 (1.7%) | .402 |
| Chronic Rejection | 31 (2.0%) | 663 (1.9%) | .618 |
| Acute Rejection | 1243 (81.5%) | 29,025 (81.2%) | .824 |
3.3 |. Survival analysis
The follow-up for the total study population was a median of 5.04 years and a mean of 6.19 ± 4.84 years with a 95% confidence interval of 6.14–6.24 years. There was no difference in median clinical follow-up duration for transplants utilizing a donor heart with an IT of <5 h (5.04 years) versus >5 h (5.00 years, p = .539).
As expected, preservation of donor hearts for >5 h is associated with worse survival compared to those transplanted within 5 h (p < .001, Figure 2A). The 1-, 5-, and 10-year survival respectively were 83.8%, 70.9%, and 54.7% for >5 h preservation and 89.7%, 77.7%, and 61.1% for ≤5 h preservation. In the population receiving donor heart exposed to >5 h of preservation, the most significant contributors to inferior survival are pre-transplant dialysis (p < .001, Figure 2B), transplant of O blood type donor hearts (p = .001, Figure 2C) and pre-transplant ECMO (p < .001, Figure 2D). Compared to patients transplanted with donor hearts preserved for 5.01–6h (n = 1145), IT > 6 h (n = 381) performed even worse with a trend toward higher early mortality with 1, 5, 10, and 15 years survival of 86.1% versus 80.9% (p = .055), 74.2% versus 66.9% (p = .084), 59.9% versus 48.3% (p = .104), and 40.3% versus 29.4% (p = .100), respectively.
FIGURE 2.
Survival analysis shows (A) Worse survival for donor heart ischemic times >5 h; and for donor hearts with >5 h of ischemic time, heightened risk for poorer transplant survival occurs for recipients (B) on pre-transplant dialysis, (C) transplanted with O blood type donor hearts, and (D) on pre-transplant ECMO. ECMO, extracorporeal membrane oxygenation
Univariate Cox-regression analysis (Table S1) shows: (1) Recipient factors associated with worse survival were male gender (hazard ratio [HR] = 1.25, p = .016), increasing creatinine (HR = 1.14, p < .001), increasing bilirubin (HR = 1.02, p = .011), diabetes (HR = 1.28, p = .015), dialysis (HR = 1.89, p < .001), ECMO (HR = 3.73, p < .001), ischemic cardiomyopathy (HR = 1.25, p = .005), and increased pulmonary artery systolic pressures (HR = 1.008, p = .006). (2) Recipient factors associated with improved survival included increased height (HR = 0.988, p = .002) and nonischemic cardiomyopathy (HR = 0.75, p < .001). (3) Donor factors associated with worse survival were increasing age (HR = 1.009, p = .004), and O blood type donor (HR = 1.31, p < .001). (4) Donor features associated with improved survival included male gender (HR = 0.82, p = .023).
Multivariable Cox-regression analysis (Table 3) demonstrated that worse survival was associated with recipient ischemic cardiomyopathy (HR = 1.36, p < .001), pre-transplant dialysis (HR = 1.84, p = .002), pre-transplant ECMO (HR = 2.36, p = .089), and increased pulmonary artery systolic pressure (HR = 1.007, p = .021). Recipient factors associated with superior survival include increased height (HR = 0.989, p = .009) and nonischemic cardiomyopathy (HR = 0.74, p = .001). Donor factors associated with poorer survival include increasing age (HR = 1.01, p = .003) and O blood type (HR = 1.35, p = .001).
TABLE 3.
Multivariable Cox proportional hazards analysis for transplant survival following >5 h ischemic times in donor hearts
| Predictors of survival - Multivariate | B | SE | Wald | df | p | HR |
|---|---|---|---|---|---|---|
| Recipient: Height | −0.011233 | 0.004281 | 6.883685 | 1.000000 | 0.008699 | 0.988830 |
| Ischemic Cardiomyopathy | 0.308311 | 0.088877 | 12.033756 | 1.000000 | 0.000522 | 1.361125 |
| Nonischemic Cardiomyopathy | −0.297908 | 0.088904 | 11.228464 | 1.000000 | 0.000806 | 0.742370 |
| Pre-Transplant Dialysis | 0.612378 | 0.192967 | 10.070978 | 1.000000 | 0.001506 | 1.844814 |
| Pre-Transplant ECMO | 0.860093 | 0.506079 | 2.888386 | 1.000000 | 0.089220 | 2.363381 |
| PA Systolic (mmHg) | 0.006859 | 0.002979 | 5.299931 | 1.000000 | 0.021326 | 1.006882 |
| Donor: Age | 0.010366 | 0.003498 | 8.784351 | 1.000000 | 0.003038 | 1.010420 |
| O blood type | 0.297026 | 0.088935 | 11.154456 | 1.000000 | 0.000838 | 1.345851 |
Abbreviation: ECMO, Extracorporeal membrane oxygenation.
The univariable survival analysis for recipients of donor hearts exposed to ≤5 h of IT is shown in Table S2. While many of the risk factors for mortality are similar on multivariable analysis (Table S3), the HR tends to be lower in the IT ≤ 5 h group. For example, HR for the >5 and ≤5 h respectively for pre-transplant dialysis was 1.84 versus 1.29 (p = .002) while for pre-transplant ECMO it was 2.36 versus 2.15 (p = .041). Furthermore, donor O blood type was not a risk factor for mortality for donor heart IT ≤ 5 h.
3.4 |. UNOS region-specific challenges with prolonged allograft ischemic times
Region 6 has geographic challenges from its location in the northwestern United States where donor hearts often travel over wider geographic distances. Therefore, this region has the highest proportion of donor hearts (16.1%) experiencing preservation times greater than 5 h (p < .001) with an average IT of 5.84 ± 0.83 h (Table 4). In the ≤5 h preservation range, region 6 still had the highest mean preservation time (3.33 ± 0.96 h) compared to other geographical areas (Table 4). In contrast, the proportion of transplants performed with donor hearts >5 h were mostly between 2% and 5% in other UNOS regions.
TABLE 4.
Tendency for prolonged donor heart ischemic times beyond 5 h in different UNOS regions
| UNOS Region | ≤5 h (frequency) | p value versus region 6 | ≤5 h (mean) | p value versus region 6 | >5 h (frequency) | p value versus region 6 | >5 h (mean) | p value versus region 6 |
|---|---|---|---|---|---|---|---|---|
| 1 | 1445 (98.0%) | <.001 | 3.06 ± 0.82 | <.001 | 30 (2.0%) | <.001 | 6.04 ± 1.07 | .665 |
| 2 | 4475 (96.5%) | <.001 | 3.00 ± 0.88 | <.001 | 160 (3.5%) | <.001 | 5.58 ±0.60 | <.001 |
| 3 | 4265 (94.1%) | <.001 | 3.08 ± 0.89 | <.001 | 269 (5.9%) | <.001 | 5.99 ± 1.06 | .375 |
| 4 | 3829 (96.4%) | <.001 | 3.01 ± 0.93 | <.001 | 141 (3.6%) | <.001 | 5.61 ±0.78 | .001 |
| 5 | 5707 (95.2%) | <.001 | 3.19 ± 0.88 | <.001 | 290 (4.8%) | <.001 | 5.70 ± 081 | <.001 |
| 6 | 962 (83.9%) | 3.33 ± 0.96 | 184 (16.1%) | 5.84 ± 0.83 | ||||
| 7 | 3349 (95.9%) | <.001 | 3.07 ± 0.92 | <.001 | 142 (4.1%) | <.001 | 5.66 ± 0.77 | .013 |
| 8 | 2126 (97.6%) | <.001 | 2.83 ± 0.91 | <.001 | 53 (2.4%) | <.001 | 5.78 ± 1.09 | .166 |
| 9 | 2237 (96.4%) | <.001 | 3.09 ± 0.90 | <.001 | 83 (3.6%) | <.001 | 5.73 ± 0.96 | .045 |
| 10 | 3146 (98.5%) | <.001 | 2.97 ± 0.86 | <.001 | 49 (1.5%) | <.001 | 5.71 ± 0.75 | .100 |
| 11 | 4192 (97.1%) | <.001 | 3.03 ± 0.90 | <.001 | 125 (2.9%) | <.001 | 5.83 ± 1.03 | .055 |
Abbreviation: UNOS, United Network for Organ Sharing.
Survival for ischemic time >5 h was better for region 6 compared with other regions. Respective survival at 1, 5, and 10 years was 90.2%, 85.3%, and 71.0% versus 82.8%, 68.9%, and 52.5% (p < .001). Compared with other combined regions, region 6 had a much lower proportion of recipient with ischemic heart failure etiologies (61/184 [33.2%] vs. 584/1342 [43.5%]) with concomitant younger recipient age (50.03 ± 13.34 vs. 52.17 ± 13.78 years, p = .048). However, there was no difference in pre-transplant ECMO status (1/184 [0.5%] vs. 10/1342 [0.7%], p = .762), pre-transplant dialysis (25/184 [13.6%] vs. 226/1342 [16.8%], p = .264), donor age (32.49 + 12.43 vs. 31.74 + 12.13 years, p = .431) or donor O blood type (85/184 [46.2%] vs. 608/1342 [45.3%], p = .820). We also performed Cox regression multivariable analysis using other variables significantly associated with transplant survival as indicated in Table S1. Region 6 was still associated with superior survival compared with other composite UNOS regions with an HR = 0.544 (p < .001).
4 |. DISCUSSION
While the negative impact of prolonged donor heart IT on transplant survival is well recognized,6 factors that modulate transplant risk at the higher range of preservation times are less studied. The impetus for accepting hearts preserved for >5 h may be varied including dire recipient clinical status as well as unanticipated procedural or transport delays. The less time spent in status 1B, higher preoperative bilirubin levels, and greater use of total artificial heart or biventricular support in the prolonged preservation group supports the former explanation.
We demonstrate that UNOS region 6 with donors distributed over a wide geographical area are more accepting of longer preservation times with 16.1% of donor heart preserved for >5 h. Other UNOS regions also have a notable incidence (2%–5%) of donor hearts exposed to longer ischemic times. Despite this, centers in region 6 achieved superior transplant survival utilizing donor hearts with >5 h of preservation time. The significantly lower proportion of recipients with ischemic heart disease may partly explain this but center experience with recipient and donor matching and management of PGD also likely play a role. Offer of donor hearts from distant facilities is becoming more widespread with the recent changes in cardiac donor allocation algorithm. The wider regional donor availability coupled with a trend toward traveling further distances makes this topic particularly relevant.4
Prolonged preservation times >5 h was associated with worse survival and greater likelihood of death from PGD consistent with prior studies.8 Inferior survival may be a result of postoperative complications such as stroke and dialysis which are associated with PGD and its management modalities such as mechanical support.9,10 Furthermore, prolonged myocardial preservation may disproportionally impact the conduction system given the increased need for pacemaker implant. However prolonged IT did not impact the incidence of death from acute or chronic rejection nor affect the rate of acute rejection. The lack of impact of prolonged donor heart IT on the rate of acute or chronic rejection was consistent with findings by other investigators.11,12
Interestingly, increasing IT beyond 6 h leads to the further detriment to survival compared to 5.01–6 h. However, the cold ischemic threshold beyond which survival plateaus is unclear and will likely not be revealed given necessary clinical prudence. Recent advances with ex-vivo normothermic perfusion have doubled the allowable out-of-body times of donor hearts.6 However, the incidence of PGD has not been altered with this technology thus far.6,13 The cardiac responses to ex-vivo perfusion are currently not well understood but may bear some resemblance to cold reservation given the need for hypothermic cardiac arrest at the commencement and end of ex-vivo perfusion.
In this group of patients with prolonged donor heart preservation, multivariate analysis identified recipient factors such as increased pulmonary pressures (HR = 1.01) and increasing donor age (HR = 1.01) to be associated with poorer survival. This is consistent with prior studies showing elevated pulmonary pressure14 and older donor age15 to be associated with inferior survival. However, factors exerting the greatest negative impact are pre-transplant ECMO (HR = 2.36), pre-transplant dialysis (HR = 1.84), ischemic cardiomyopathy (HR = 1.36), and O donor blood type (HR = 1.35). Although only 40/76 (52.76%) of pre-transplant dialysis patients had documented simultaneous kidney transplant, it is possible that a subsequent kidney transplant was staged and not performed concomitantly with the heart transplant. Pre-transplant venoarterial ECMO,16 ischemic cardiomyopathy,17 and use of O donor hearts18,19 were previously reported to be associated with poorer transplant survival. Interestingly, transplants utilizing donor hearts with IT < 5 h had similar risk factors for mortality. However, key risk factors such as recipient pre-transplant dialysis and ECMO were associated with a lower mortality risk. Furthermore, donor O blood group was not associated with mortality in the IT ≤ 5 h group.
Although prolonged IT of donor hearts is associated with PGD and worse survival, these are specific factors that should be taken into consideration when faced with preservation times >5 h. Based on our findings, recipient factors such as preoperative ECMO support, dialysis, and an ischemic etiology should deter the use of donor hearts stored for longer periods. This is consistent with the impact of the “hostile recipient” environment is recognized to impact donor heart function and transplant survival.5 Furthermore, a preference should be given to donor hearts from younger donors with a non-O blood type. Importantly, the stacking of multiple aforementioned risk factors should be discouraged to minimize the risk of PGD.
Limitations to this study include its retrospective nature from an international registry that has limited data granularity. Numerous other factors that can affect survival and graft function are likely not considered such as operative blood loss, post-transplant hemodynamics, and outpatient transplant management to highlight a few. Furthermore, there are selection biases based on established management paradigms for donor organs within transplant programs which modulate outcomes and may dilute the impact of specific factors in our study population. Furthermore, there may be differences in the ability to rescue patients from PGD associated complications amongst centers of varying transplant volumes.
5 |. CONCLUSIONS
When evaluating donor hearts that are likely to be exposed to greater than 5 h of IT, we recommend exercising caution in recipients on pre-transplant ECMO, receiving pre-transplant dialysis, or with an ischemic heart failure etiology. Donor hearts from younger donors are preferred with avoidance of O blood type hearts. Development of programmatic management strategies for significant PGD which may include early utilization of mechanical circulatory support (e.g., ECMO) in at-risk transplants may promote favorable outcomes while utilizing donor heart exposed to prolonged ischemic times.
Supplementary Material
Abbreviations:
- CMO
extracorporeal membrane oxygenation
- IT
ischemic time
- LVAD
left ventricular assist device
- PA
pulmonary artery
- PGD
primary graft dysfunction
- UNOS
United Network for Organ Sharing
Footnotes
CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the supporting information tab for this article.
REFERENCES
- 1.John R Donor management and selection for heart transplantation. Semin Thorac Cardiovasc Surg. 2004;16(4):364–369. [DOI] [PubMed] [Google Scholar]
- 2.Kittleson MM. Recent advances in heart transplantation. F1000Research. 2018;7:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Del Rizzo DF, Menkis AH, Pflugfelder PW, et al. The role of donor age and ischemic time on survival following orthotopic heart transplantation. J Heart Lung Transplant. 1999;18(4):310–319. [DOI] [PubMed] [Google Scholar]
- 4.Cogswell R, John R, Estep JD, et al. An early investigation of outcomes with the new 2018 donor heart allocation system in the United States. J Heart Lung Transplant. 2020;39(1):1–4. [DOI] [PubMed] [Google Scholar]
- 5.Iyer A, Kumarasinghe G, Hicks M, et al. Primary graft failure after heart transplantation. J Transplant. 2011;2011:175768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schroder JN, D’Alessandro D, Esmailian F, et al. Successful utilization of extended criteria donor (ECD) hearts for transplantation – Results of the OCS™ Heart EXPAND trial to evaluate the effectiveness and safety of the OCS heart system to preserve and assess ECD hearts for transplantation. J Heart Lung Transplant. 2019;38(4):S42. [Google Scholar]
- 7.Valero-Masa MJ, González-Vílchez F, Almenar-Bonet L, et al. Cold ischemia >4 hours increases heart transplantation mortality. An analysis of the Spanish heart transplantation registry. Int J Cardiol. 2020;319:14–19. [DOI] [PubMed] [Google Scholar]
- 8.Singh A, Vathulya M, Agrawal A, Verma R, Mittal SK, Kumar B. Primary graft dysfunction after heart transplantation: a thorn amongst the roses. Heart Fail Rev. 2019;11:237–240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Prieto D, Correia PM, Batista M, Antunes MJ. Primary graft failure after cardiac transplantation: prevalence, prognosis and risk factors. Interact Cardiovasc Thorac Surg. 2018;27(5):765–772. [DOI] [PubMed] [Google Scholar]
- 10.Hulman M, Artemiou P, Ondrusek M, et al. Short-term mechanical circulatory support for severe primary graft dysfunction following orthotopic heart transplant. Interact Cardiovasc Thorac Surg. 2018;27(2):229–233. [DOI] [PubMed] [Google Scholar]
- 11.Scheule AM, Zimmerman GJ, Johnston JK, Razzouk AJ, Gundry SR, Bailey LL. Duration of graft cold ischemia does not affect outcomes in pediatric heart transplant recipients. Circulation. 2002;106 (12 suppl 1):I163–I167. [PubMed] [Google Scholar]
- 12.Yeen W, Polgar A, Guglin M, et al. Outcomes of adult orthotopic heart transplantation with extended allograft ischemic time. Transplant Proc. 2013;45(6):2399–2405. [DOI] [PubMed] [Google Scholar]
- 13.Ardehali A, Esmailian F, Deng M, et al. Ex-vivo perfusion of donor hearts for human heart transplantation (PROCEED II): a prospective, open-label, multicentre, randomised non-inferiority trial. Lancet. 2015;385(9987):2577–2584. [DOI] [PubMed] [Google Scholar]
- 14.Vakil K, Duval S, Sharma A, et al. Impact of pre-transplant pulmonary hypertension on survival after heart transplantation: a UNOS registry analysis. Int J Cardiol. 2014;176(3):595–599. [DOI] [PubMed] [Google Scholar]
- 15.Gupta D, Piacentino V, Macha M, et al. Effect of older donor age on risk for mortality after heart transplantation. Ann Thorac Surg. 2004;78(3):890–899. [DOI] [PubMed] [Google Scholar]
- 16.Fukuhara S, Takeda K, Kurlansky PA, Naka Y, Takayama H. Extracorporeal membrane oxygenation as a direct bridge to heart transplantation in adults. J Thorac Cardiovasc Surg. 2018;155(4):1607–1618. [DOI] [PubMed] [Google Scholar]
- 17.Aziz T, Burgess M, Rahman AN, Campbell CS, Yonan N. Cardiac transplantation for cardiomyopathy and ischemic heart disease: differences in outcome up to 10 years. J Heart Lung Transplant. 2001;20(5):525–533. [DOI] [PubMed] [Google Scholar]
- 18.Jawitz OK G Jawitz N, Yuh DD, Bonde P. Impact of ABO compatibility on outcomes after heart transplantation in a national cohort during the past decade. J Thorac Cardiovasc Surg. 2013;146(5):1239–1245. [DOI] [PubMed] [Google Scholar]
- 19.Tang PC, Haft JW, Lei I, et al. Impact of donor blood type on outcomes after prolonged allograft ischemic times. J Thorac Cardiovasc Surg. 2021;S0022–5223(21):00020–00029. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.