Journal Logo

Original Clinical Science—General

Kidney Transplantation After Rescue Allocation—the Eurotransplant Experience: A Retrospective Multicenter Outcome Analysis

Assfalg, Volker MD1,2; Miller, Gregor MSc3; Stocker, Felix MD1,2; van Meel, Marieke4; Groenevelt, Tiny4; Tieken, Ineke MD4; Ankerst, Donna PhD3; Renders, Lutz MD1,5; Novotny, Alexander MD1,2; Hartmann, Daniel MD1,2; Jell, Alissa MD1,2; Rahmel, Axel MD4; Wahba, Roger MD6; Mühlfeld, Anja MD7; Bouts, Antonia MD, PhD8; Ysebaert, Dirk MD, PhD9; Globke, Brigitta MD10; Jacobs-Tulleneers-Thevissen, Daniel MD, PhD11; Piros, László MD, PhD12; Stippel, Dirk MD6; Heller, Katharina MD13; Eisenberger, Ute MD14; van Laecke, Steven MD, PhD15; Weimer, Rolf MD16; Rosenkranz, Alexander R. MD17; Berger, Stefan MD18; Fischer, Lutz MD19; Kliem, Volker MD20; Vondran, Florian MD21; Sester, Urban MD22; Schneeberger, Stefan MD, MBA23; Harth, Ana MD24; Kuypers, Dirk MD, PhD25; Függer, Reinhold MD26; Arnol, Miha MD, PhD27; Christiaans, Maarten MD28; Weinmann-Menke, Julia MD29; Krüger, Bernd MD30; Hilbrands, Luuk MD31; Banas, Bernhard MD, MBA32; Hakenberg, Oliver MD33; Minnee, Robert MD, PhD34; Schwenger, Vedat MD, MHBA35; Heyne, Nils MD36; van Zuilen, Arjan MD, PhD37; Reindl-Schwaighofer, Roman MD, PhD38; Lopau, Kai MD39; Hüser, Norbert MD1,2; Heemann, Uwe MD1,5

Author Information
doi: 10.1097/TP.0000000000003964
  • Free
  • SDC
  • Infographic

Abstract

INTRODUCTION

Allocation schemes for deceased donor renal transplantation (DDRT) are based on scientifically proven risk factors for graft and recipient outcome as well as ethical principles. These algorithms rank potential recipients on the waiting list for every allocation procedure. At the Eurotransplant International Foundation (ET), immunological matching, waiting time, cold ischemia time (CIT), age, urgency, and preformed antibodies are the defining factors.1-3

Death on the ET waiting list ranges between 4.2% and 5.4% (mean 4.7%) for the last 15 y. This high mortality, which is a consequence of persistent and increasing donor shortage and high numbers of patients on the waiting list,4 resulted in the acceptance and transplantation of kidneys from comorbid donors with indefinite and disputable outcomes. During the last 2 decades, “expedited” or “rescue allocation” (RA) rules have been established and refined repeatedly by most organizations worldwide to reduce the number of discarded grafts and increase transplant numbers.1-3 Currently, 22.6% of all kidneys offered within ET are finally discarded and the median age of these donors is 61 y.

ET, the largest European organ allocation organization, defined distinct rules for RA following logistic or medical reasons that allow to deviate from the standard allocation (SA) programs “Eurotransplant Kidney Allocation System” (ETKAS) and “Eurotransplant Senior Program” (ESP).2,5 The ETKAS is destined for all candidates irrespective of their age and considers waiting time, HLA match, a regional/national bonus to favor shorter CITs, a pediatric bonus, and a high urgency bonus. However, the ESP is an alternative program only for candidates beyond 64 y of age, which abstains from HLA matching and only takes account of waiting time and preferably short CITs by regional allocation of kidneys from donors aged >64 y. Both SA programs transfer grafts to RA to prevent loss of potentially transplantable organs.5

The reasons for switching over to RA may be very inhomogeneous and can derive from different reasons:

  • repeated rejection of the offer for all candidates of 5 different centers, for example, due to donor-related reasons such as presumed inadequate quality of the graft or problems with the procurement process;
  • nonacceptance of the organ 5 h after procurement;
  • logistics do not allow for timely transplantation causing an increased CIT;
  • impending loss of the organ for transplantation; or
  • an interaction of these factors.2,6

In addition, a subsequent “cascade effect” of repeated declines has been reported in case of the subjective negative assessment of an offer and decline by 1 center.7

However, although kidneys offered via RA recently turned out to be of inferior histopathological quality8 and characteristics of RA transplants are inhomogeneous, the outcome was demonstrated to be comparable with SA in small single-center analyses.7,8

In RA, centers may self-select suitable recipients by themselves either from an ET-generated ranking list within the Recipient Oriented Extended Allocation (REAL) program, which abides by the ETKAS SA criteria, or from an in-house list for Competitive Center Offers (CCOs),2 documenting the reasons for selecting the recipient for transparency and scrutiny. The detailed regulations on RA within ET can be looked up online in reference 5.

CCOs provide centers the opportunity to allocate grafts according to the match list or by specific in-house rules, such as urgency, need, or expected transplant outcome. The potential benefits from the transplantation of kidneys from expanded criteria donors (ECD)9 to nonimmunized recipients aged >40 y with diabetes and hypertension—the most perilous comorbidity cluster—have been described repeatedly.10-12 A recent study showed that benefits of RA for selected recipients with impaired health status were most likely attributable to reduced waiting times,8 the strongest established modifiable risk factor for outcomes.13

Current increases in both the number of kidneys offered via RA and the needs for donor kidneys across most countries, particularly in Europe, force transplant physicians to identify and quantify the benefits of RA for selected target candidates. Hitherto, the outcomes from RA transplants have not been analyzed by comprehensive trials with sufficiently large case numbers. Therefore, this multicenter study was initiated to reveal the outcomes of RA from 71 ET kidney transplant centers in comparison with outcomes from SA within the same area.

MATERIALS AND METHODS

Long-term outcomes of RA kidney-only transplantations from brain death deceased donors (DDRT) within ETKAS and ESP in the ET area between January 2006 and May 2018 were investigated after approval of the study by the ET authorities (14046KAC14). During this time period, a total of 50 835 SA and 3498 RA DDRTs were performed.

All ET transplant centers were requested to return follow-up data to increase data completeness at the ET registry as previously performed by the ET community for comparable issues.14,15 The request was issued between January and September 2019 and ascertained date of last follow-up, graft loss with date of loss, patient’s death with date of death, as well as patient’s death with functioning graft (DwFG), sequence of organ transplantation, and underlying renal disease, respectively. DwFG data provide insights into the concomitant health status of the affected recipients by accounting for the number of deaths not associated with graft failure. Information on sex, age at transplant, HLA match, waiting time, transplant period, country where the transplantation was performed, and general information on the overall ET waiting list and transplantations was obtained from the ET database.

Individual records with missing follow-up were assumed to have data missing at random and removed for statistical analyses,16 other exclusions are shown in Figure 1. Missing follow-up was defined whenever no more information was available after transplantation. Cases with errors or contradictory information in the data set were excluded as well. Noninformative censoring was assumed for all time-to-event analyses.17

F1
FIGURE 1.:
Flowchart of the selection process of transplants analyzed in this study. Counts refer to number of transplants. AM, Acceptable Mismatch Program; ESP, Eurotransplant Senior Program; ET, Eurotransplant; ETKAS, Eurotransplant Kidney Allocation System.

Within the investigated period and the restricted data set, 823 patients were repeatedly transplanted, including 179 RA patients. Retransplantations were considered as independent observations. Mean (median) follow-up times for both SA and RA DDRTs were 1838.1 (1673) versus 1515.6 (1157) d, respectively. Follow-up acquisition was terminated on July 3, 2020, and reported follow-up was capped at 10 y after transplantation for all analyses.

ET data protection policy required patient and center anonymization at the ET registry department, which provided anonymized data to the study statisticians and principal investigators.

Recipient survival was counted from day of transplant to day of death and not censored for graft loss. Graft loss was defined as return to dialysis after successful transplantation. All outcome parameters were censored for patient loss to follow-up. Cumulative incidence curves were calculated for recipient death, DwFG, and graft loss, the latter 2 accounting for competing risks of each other. Censored patient survival and cumulative incidence of DwFG and graft loss were compared for all investigated subgroups defined by clinical and demographic parameters. For factors with >2 groups in this analysis, Bonferroni correction was applied to account for multiple pairwise comparisons. For patient survival, Cox proportional hazards models were used.18 For analyses of DwFG and graft loss, the Fine-Gray proportional regression model was used with semiparametric random effects for competing risks.19-21 Multivariable models for patient survival, DwFG, and graft loss included covariates previously identified22 to affect graft failure and mortality after DDRT, such as age and gender of the recipient, waiting time, CIT, diabetes, transplant count, and HLA matches for comparison between RA and SA.4,8,13-15,23-27 Both univariable and multivariable models were fit to all endpoints, with 95% confidence intervals (CIs) reported for hazard ratios.

Primary nonfunction (PNF) was assumed when graft failure was recorded within 90 d after transplantation. Patients who died on the day of transplantation (SA: n = 1; RA: n = 1) and transplants with PNF were henceforth excluded from investigations on graft loss and DwFG.

The number of HLA matches including HLA-A, -B, and -DR loci was analyzed with regards to transplant outcome and further subdivided: all matches with at least 1 -DR plus at least 1 -A or 1 -B match were assigned to the group of “favorable matches”; all others were defined as “unfavorable matches.”

To account for relevant numbers of recipients with missing follow-up, a subgroup analysis was performed to determine statistically higher rates of missing follow-up with respect to the allocation modus. The chi-square test with Monte Carlo simulations was used to test for differences in the categorical variables related to follow-up (Table S1, SDC, https://links.lww.com/TP/C358).

All analyses were performed at the 2-sided level of significance of 0.05 using the R statistical package (R Foundation for Statistical Computing, Vienna, Austria).21,28 All data ascertainments and analyses were performed in accordance with ethical standards as laid down in the Declaration of Helsinki.

RESULTS

Demographic and transplant-specific data on SA and RA transplantations are given in Table 1 and the densities of recipient age for RA and SA are depicted in Figure 2A. The steep increase in SA recipients starting at 65 y originates from the ESP. RA recipients from the ETKAS waiting list as well as from the ESP list were significantly older than recipients after SA, received organs from older donors, and had a worse HLA match and a prolonged CIT, but waiting time was shorter in each case. Notably, PNF rates were comparable between SA and RA (Table 1). Considering recipients from the ETKAS waiting list only, the mean HLA match was higher for all HLA-A, -B, and -DR, in sum, and the frequency of favorable matches was superior (Table 2).

TABLE 1. - Characteristics of transplants according to allocation
Characteristics of ETKAS and ESP transplants
Standard allocation Rescue allocation
Value N Group % Allocation % N Group % Allocation % P
Recipient sex Female 9162 36.0 90.7 935 38.6 9.3 0.010
Male 16 319 64.0 91.6 1487 61.4 8.4
Disease group Glomerulonephritis 5895 23.1 91.3 559 23.1 8.7 <0.001
Cystic disease 3429 13.5 89.8 390 16.1 10.2
Diabetes 2160 8.5 90.6 225 9.3 9.4
Other 13 997 54.9 91.8 1248 51.5 8.2
Donor sex Female 12 060 47.3 91.7 1089 45.0 8.3 0.027
Male 13 421 52.7 91.0 1333 55.0 9.0
Allocationprogram ETKAS 19 520 76.6 89.6 2258 93.2 10.4 <0.001
ESP 5961 23.4 97.3 164 6.8 2.7
Organ Left kidney 12 382 48.6 91.8 1101 45.5 8.2 0.003
Right kidney 13 099 51.4 90.8 1321 54.5 9.2
Transplantcount 1 23 549 92.4 91.4 2225 91.9 8.6 0.277
2 1811 7.1 90.6 189 7.8 9.4
≥3 121 0.5 93.8 8 0.3 6.2
Country Germany 15 064 59.1 89.1 1837 75.8 10.9 <0.001
Austria 3066 12.0 92.5 247 10.2 7.5
Belgium 2922 11.5 97.8 65 2.7 2.2
The Netherlands 1972 7.7 94.3 119 4.9 5.7
Croatia 1764 6.9 96.0 74 3.1 4.0
Slovenia 498 2.0 95.0 26 1.1 5.0
Hungary 171 0.7 76.0 54 2.2 24.0
Luxembourg 24 0.1 100.0 0 0.0 0.0
Sum of HLAmatches 0 271 1.3 70.8 112 5.4 29.2 <0.001
1 1022 4.8 75.1 338 16.3 24.9
2 3130 14.7 83.1 638 30.8 16.9
3 7150 33.5 92.3 596 28.8 7.7
4 5089 23.8 94.3 307 14.8 5.7
5 1308 6.1 95.1 67 3.2 4.9
6 3389 15.9 99.6 13 0.6 0.4
Missing 4122 92.2 351 7.8
HLA matchgrouping Favorable 18 369 86.0 93.3 1310 63.3 6.7 <0.001
Nonfavorable 2990 14.0 79.7 761 36.7 20.3
Missing 4122 92.2 351 7.8
Dead No 19 681 77.3 91.1 1923 79.4 8.9 0.016
Yes 5794 22.7 92.1 498 20.6 7.9
Missing 6 85.7 1 14.3
Failure No 18 145 71.2 91.2 1757 72.5 8.8 0.173
Yes 7336 28.8 91.7 665 27.5 8.3
DwFG No 23 754 93.2 91.3 2255 93.1 8.7 0.859
Yes 1727 6.8 91.2 167 6.9 8.8
PNF No 23 538 92.4 91.4 2216 91.5 8.6 0.130
Yes 1943 7.6 90.4 206 8.5 9.6
Characteristics of ETKAS transplants
Standard allocation Rescue allocation
Count (missing) Quartiles (range) Mean ± SD Count (missing) Quartiles (range) Mean ± SD P
Recipient age 19 520 (0) 52 (43–59) 50.2 ± 11.9 2258 (0) 58 (51–64) 56.6 ± 10.7 <0.001
Donor age 19 520 (0) 50 (41–58) 47.6 ± 13.9 2258 (0) 58 (48–68) 56.1 ± 17.1 <0.001
Cold ischemia time, min 16 323 (3197) 810 (612–1020) 832.5 ± 308 2059 (199) 1002 (771–1260) 1031.9 ± 362.1 <0.001
Waiting time, d 19 520 (0) 1867 (1006–2793) 1989.7 ± 1211.5 2258 (0) 1533.5 (823.2–2416.8) 1682.4 ± 1020.9 <0.001
Characteristics of ESP transplants
Standard allocation Rescue allocation
Count (missing) Quartiles (range) Mean ± SD Count missing) Quartiles (range) Mean ± SD P
Recipient age 5961 (0) 68 (66–71) 68.7 ± 3.4 164 (0) 69 (67–72) 69.3 ± 3.3 0.007
Donor age 5961 (0) 71 (67–74) 71.3 ± 4.8 164 (0) 76 (71–81) 76.3 ± 6.8 <0.001
Cold ischemia
time, min
5371 (590) 635 (468–822) 665.1 ± 259.9 145 (19) 880 (669–1080) 897.1 ± 285.3 <0.001
Waiting time, d 5961 (0) 1258 (810–1813) 1368.6 ± 736.6 164 (0) 815 (528.2–1409) 1058.1 ± 726.2 <0.001
DwFG, death with functioning graft; ESP, Eurotransplant Senior Program; ETKAS, Eurotransplant Kidney Allocation System; PNF, primary nonfunction.

TABLE 2. - Comparison of HLA matching between allocation types limited to ETKAS data
Standard allocation Rescue allocation
Value N % N % P
HLA-Amatches 0 1905 9.8 437 19.4 <0.001
1 9524 48.8 1188 52.6
2 8086 41.4 380 16.8
Missing 5 0.0 253 11.2
Mean 1.3 1.0
HLA-Bmatches 0 3540 18.1 879 38.9 <0.001
1 10 515 53.9 977 43.3
2 5460 28.0 149 6.6
Missing 5 0.0 253 11.2
Mean 1.1 0.6
HLA-DRmatches 0 1635 8.4 605 26.8 <0.001
1 10 744 55.0 1100 48.7
2 7136 36.6 300 13.3
Missing 5 0.0 253 11.2
Mean 1.3 0.8
Sum of HLAmatches 0 82 0.4 103 4.6 <0.001
1 530 2.7 317 14.0
2 2625 13.4 613 27.1
3 6759 34.6 594 26.3
4 4886 25.0 304 13.5
5 1252 6.4 62 2.7
6 3381 17.3 12 0.5
Missing 5 0.0 253 11.2
Mean 3.7 2.5
HLA matchgrouping Favorable 17 522 89.8 1287 57.0 <0.001
Nonfavorable 1993 10.2 718 31.8
Missing 5 0.0 253 11.2
ETKAS, Eurotransplant Kidney Allocation System.

F2
FIGURE 2.:
A, Recipient age and (B) amount of transplants between 2006 and 2018 with respect to allocation type. Percentages show the fraction of the respective period.

The numbers and proportions of RA increased markedly over the analyzed time periods (Figure 2B). Kidneys from RA were mainly allocated to candidates on the ETKAS waiting list and rarely for ESP-listed recipients (93.2% versus 6.8%). RA played a minor role in ESP- as compared to ETKAS-listed patients (2.7% versus 10.4%; Table 1). Germany had by far the most transplants within ET with respect to SA (59.1%), but especially with respect to RA (75.8%; Table 1).

With regards to cases with or without follow-up, no differences could be revealed for SA between left versus right organs, but significant differences were found for recipient age, donor age, allocation program, CIT, recipient sex, donor sex, renal disease, matching, waiting time, transplant count, transplantation period, and country. Among RA recipients, follow-up was less frequently noted in cases with unknown CIT, male donors, long waiting time, and later transplantation periods, and from Germany (Table S1, SDC, https://links.lww.com/TP/C358).

Table 3 gives an overview on patient survival, DwFG, and graft loss with regards to allocation modus and transplant-specific variables in univariate testing. Figure 3 displays the cumulative incidence curves of outcome of ETKAS-listed candidates with regards to RA versus SA. Transplant outcome after RA between the different ET member countries did not reveal any statistical differences in subgroup analyses due to low case numbers in most countries (Table 1). However, waiting time of ETKAS-listed patients was by far the longest in Germany (mean 2410 d versus <1600 d in all other ET member countries).

T3
TABLE 3.:
Univariate analysis of factors regarding survival and competing risk between DwFG and graft loss of ETKAS transplants
F3
FIGURE 3.:
Cumulative incidence curves for ETKAS patients with respect to death (A), death with functioning graft (B), and graft loss (C) according to allocation. DwFG, death with functioning graft; ETKAS, Eurotransplant Kidney Allocation System.

In univariate analyses, mortality, graft loss, and DwFG within 10 y after RA were significantly higher as compared to SA for the analyzed period (Table 3). Notably, patients with diabetes and prolonged waiting time displayed an increased mortality hazard and increased cumulative incidence of graft loss. Survival and graft loss turned out to be worse in recipients of a second graft. DDRTs in recipients with cystic disease, favorable HLA match, organs from younger donors, and with shorter CITs showed superior outcomes in all 3 categories (Table 3).

The univariate analysis of transplant-specific continuous variables and the multivariate analysis of patient survival, DwFG, and graft loss of recipients from the ETKAS waiting list with regards to known influencing variables, including the allocation modus, can be found in Table 4A and B. Remarkably, in the multivariate analysis both survival and DwFG after RA turned out to be comparable with SA (P = 0.090 and P = 0.105), whereas RA even showed reduced cumulative incidences for graft loss (subdistribution hazard ratio, 0.80; 95% CI: 0.70-0.91; P < 0.001). Diabetes and re-transplantation were associated with higher mortality as well as higher DwFG and graft loss incidence.

TABLE 4. - Univariate analysis of continuous variables regarding survival and competing risks between DwFG and graft loss for ETKAS patients (A) and multivariate analysis restricted to ETKAS (B) and ESP (C)
Mortality DwFG Graft loss
HR (95% CI) P Subdist. HR (95% CI) P Subdist. HR (95% CI) P
A. Univariate analyses of variables in ETKAS data
Donor age, y 1.02 (1.01-1.02) <0.001 1.02 (1.01-1.02) <0.001 1.01 (1.01-1.02) <0.001
Recipient age, y 1.07 (1.06-1.07) <0.001 1.07 (1.06-1.08) <0.001 1.00 (1.00-1.01) 0.011
Cold ischemia time, h 1.01 (1.00-1.02) 0.001 1.02 (1.01-1.03) <0.001 1.02 (1.01-1.03) <0.001
Waiting time, y 1.02 (1.01-1.03) <0.001 0.98 (0.96-1.00) 0.061 1.01 (1.00-1.02) 0.022
B. Multivariate analysis ETKAS
Allocation type Standard vs rescue 0.89 (0.78-1.02) 0.090 0.80 (0.62-1.05) 0.105 0.80 (0.70-0.91) <0.001
Recipient sex Female vs male 1.10 (1.02-1.19) 0.015 0.95 (0.80-1.12) 0.525 1.07 (0.99-1.16) 0.093
HLA match Favorable vs nonfavorable 1.15 (1.03-1.29) 0.011 1.19 (0.94-1.51) 0.151 1.18 (1.05-1.32) 0.004
Donor age (Continuous) 1.01 (1.01-1.01) <0.001 1.01 (1.00-1.02) 0.002 1.01 (1.01-1.02) <0.001
Recipient age (Continuous) 1.06 (1.06-1.07) <0.001 1.06 (1.05-1.07) <0.001 1.00 (1.00-1.01) 0.339
Cold ischemia time, h (Continuous) 1.01 (1.00-1.02) 0.017 1.02 (1.01-1.04) 0.006 1.02 (1.02-1.03) <0.001
Waiting time, y (Continuous) 1.06 (1.05-1.07) <0.001 1.02 (0.99-1.04) 0.137 1.02 (1.01-1.03) <0.001
Diabetes Nondiabetic vs diabetic 1.97 (1.76-2.20) <0.001 1.61 (1.29-2.02) <0.001 1.39 (1.20-1.60) <0.001
Transplant count 1 vs ≥2 1.56 (1.38-1.77) <0.001 1.33 (1.02-1.74) 0.035 1.51 (1.33-1.72) <0.001
C. Multivariate analysis ESP
Allocation type Standard vs rescue 0.61 (0.37-1.01) 0.056 0.71 (0.28-1.83) 0.484 0.93 (0.52-1.65) 0.803
Recipient sex Female vs male 1.27 (1.06-1.52) 0.009 1.23 (0.88-1.71) 0.235 1.11 (0.89-1.38) 0.368
HLA match Favorable vs nonfavorable 1.08 (0.92-1.26) 0.343 0.97 (0.72-1.30) 0.828 0.97 (0.80-1.18) 0.792
Donor age (Continuous) 1.01 (0.99-1.02) 0.418 0.98 (0.95-1.01) 0.290 1.02 (1.00-1.04) 0.037
Recipient age (Continuous) 1.06 (1.04-1.09) <0.001 1.08 (1.04-1.13) <0.001 1.04 (1.01-1.07) 0.002
Cold ischemia time, h (Continuous) 1.02 (1.00-1.04) 0.070 1.06 (1.02-1.09) 0.001 1.00 (0.98-1.03) 0.689
Waiting time, y (Continuous) 1.07 (1.03-1.12) <0.001 1.04 (0.97-1.12) 0.281 1.07 (1.02-1.13) 0.006
Diabetes Nondiabetic vs diabetic 1.66 (1.35-2.03) <0.001 1.22 (0.80-1.86) 0.367 1.52 (1.18-1.96) 0.001
Transplant count 1 vs ≥2 1.18 (0.87-1.59) 0.284 1.83 (1.13-2.94) 0.013 0.57 (0.35-0.93) 0.026
Concerning the categorical confounders of the multivariate analysis, the reported HRs and subdist. HRs refer to the second characteristic as compared to the characteristic named first.
CI, confidence interval; DwFG, death with functioning graft; ESP, Eurotransplant Senior Program; ETKAS, Eurotransplant Kidney Allocation System; HR, hazard ratio; subdist. HR, subdistribution hazard ratio.

Notably, in subanalyses for recipients from the ESP waiting list, patient survival, DwFG, and graft loss after RA were also comparable with SA (Table 4C). Furthermore, HLA match, CIT, and retransplantation were not associated with any outcomes, whereas long waiting times as well as diabetes showed a positive association with mortality and DwFG.

Finally, the respective impact of the 2 crucial factors “increasing donor age” and “prolonged CIT” on patient survival, DwFG, and graft loss was exemplarily investigated for a fictitious reference recipient: 55-y-old, nondiabetic, female, favorable HLA match, waiting time of 5 y, and first transplantation (Table 5).29 In this prediction model, the risk of a prolonged CIT was markedly less critical than an older age of the donor.

TABLE 5. - The cumulative hazard function for survival predicted by the Cox proportional hazards model and the cumulative incidence predictions29 for DwFG and graft loss from the competing risk model in percent for the time point of 5 y
Survival DwFG Graft loss
CIT (h): 5 10 15 20 25 5 10 15 20 25 5 10 15 20 25
Allocation Donor age, y
Standard 10 7.6 8.0 8.4 8.8 9.2 1.6 1.8 2.1 2.4 2.8 6.4 7.1 7.8 8.6 9.4
20 8.3 8.7 9.2 9.6 10.1 1.8 2.0 2.3 2.7 3.1 7.6 8.3 9.1 10.0 11.0
30 9.1 9.5 10.0 10.5 11.0 2.0 2.3 2.6 3.0 3.4 8.8 9.7 10.7 11.7 12.9
40 9.9 10.4 10.9 11.5 12.0 2.2 2.5 2.9 3.3 3.8 10.4 11.4 12.5 13.7 15.0
50 10.9 11.4 11.9 12.5 13.1 2.4 2.8 3.2 3.6 4.2 12.1 13.3 14.6 16.0 17.5
60 11.9 12.4 13.0 13.7 14.3 2.7 3.0 3.5 4.0 4.6 14.1 15.5 17.0 18.6 20.3
70 13.0 13.6 14.2 14.9 15.7 2.9 3.4 3.9 4.4 5.1 16.4 18.0 19.7 21.5 23.5
80 14.1 14.8 15.6 16.3 17.1 3.2 3.7 4.3 4.9 5.6 19.1 20.9 22.8 24.9 27.1
Rescue 10 6.7 7.0 7.4 7.7 8.1 1.2 1.4 1.6 1.9 2.2 4.8 5.3 5.8 6.4 7.1
20 7.3 7.7 8.1 8.4 8.9 1.4 1.6 1.8 2.1 2.4 5.6 6.2 6.8 7.5 8.3
30 8.0 8.4 8.8 9.2 9.7 1.5 1.7 2.0 2.3 2.6 6.6 7.3 8.0 8.8 9.7
40 8.7 9.2 9.6 10.1 10.6 1.7 1.9 2.2 2.5 2.9 7.7 8.5 9.4 10.3 11.3
50 9.5 10.0 10.5 11.0 11.5 1.9 2.1 2.4 2.8 3.2 9.1 10.0 11.0 12.0 13.2
60 10.4 10.9 11.5 12.0 12.6 2.1 2.4 2.7 3.1 3.6 10.6 11.7 12.8 14.1 15.4
70 11.4 11.9 12.5 13.1 13.8 2.3 2.6 3.0 3.4 3.9 12.4 13.6 14.9 16.4 17.9
80 12.4 13.0 13.7 14.3 15.0 2.5 2.9 3.3 3.8 4.3 14.5 15.9 17.4 19.0 20.8
For the other covariates we assumed the following values: female recipient, favorable match, recipient age of 55, waiting time of 5 y, nondiabetic, and first transplant. The models are equivalent to the ones in Table 4 but based on data from both ETKAS and ESP.
CIT, cold ischemia time; DwFG, death with functioning graft; ESP, Eurotransplant Senior Program; ETKAS, Eurotransplant Kidney Allocation System.

DISCUSSION

Survival of recipients after DDRT has been demonstrated to be superior to that of patients on dialysis and candidates awaiting DDRT.4 Shorter waiting time is the strongest modifiable factor for increasing transplant outcome.13 Therefore, any candidate awaiting DDRT should ideally be transplanted as soon as possible and with an adequate graft. In contrast, organ shortages and demographic changes evidently impede this desirable goal. To cope with these challenges in kidney transplant supply and maintain acceptable transplant numbers, ET implemented the ESP and RA algorithms during the past decades. In contemporary practice, transplant physicians are pushed to accept kidneys from older donors with more comorbidities. The transplant outcomes of kidneys from ECDs have been repeatedly evaluated,10-12 revealing a survival benefit in unsensitized patients aged >40 y with diabetes or hypertension, particularly due to shortened waiting times,12 but data on survival and graft loss after RA DDRTs are scarce. Kidneys transplanted after rescue allocation have been reported to originate from older donors with a higher rate of diabetes, hypertension, fulfilled ECD criteria,6,8 and both increased acute and chronic histopathological changes were observed in zero-time biopsies from RA kidneys.8 DDRTs after RA were characterized by a prolonged CIT, worse HLA matching, and increased cytomegalovirus (CMV) transmission risk, but a reduced waiting time,8 which was validated by this study.

As the proportion of DDRT after RA increased markedly over time, this option apparently acquired greater importance in the ET kidney transplant centers. We therefore performed this comprehensive long-term ET multicenter study to resolve the question of RA DDRT outcome, thus far only addressed in single-center reports.8,30,31

Demographic and Transplant-specific Characteristics of Rescue-allocated DDRTs

This ET multicenter study confirmed the previously observed significantly older age of RA DDRT recipients and donors8,30,31 in a comprehensive patient collective and even in case of distinction between ETKAS- and ESP-listed recipients. Notably, RA plays a minor role in recipients within the ESP until now (Table 1).

Considering the evidently crucial role of an “excellent donor” and a favorable HLA match for younger recipients and the shorter waiting time within the ESP, it may be assumed that centers referred to RA especially in cases with an urgent need for a transplant due to deterioration and risk of delisting. Those patients typically suffer from comorbidities like hypertension and diabetes.8,11 They are likely to be either too young to apply for the ESP (mean 57.4 y) to benefit from the shorter waiting time within this program or already qualified for the ESP, but their advanced age (mean 69.3 y; Table 1) and limiting frailty32 signal risk of imminent delisting. Considering this, transplant physicians obviously tended toward accepting RA offers, condoning increased donor age, prolonged CITs, and unfavorable HLA matching, just to escape this dilemma and shorten waiting time (Table 1). Despite the evidentially negative, though reasonable, compromises, PNF turned out to be comparable between RA and SA as previously reported,8 which additionally encourages acceptance of RA offers. The question is whether a recommendation should be made for RA kidneys to be considered for more candidates apart from older patients and those with comorbidities, frailty, and an increased risk of delisting or higher risk of mortality after transplant.33-36

Favorable HLA matching is essential for long-time graft and patient survival15,23,27 and is credited with extra allocation points in the ETKAS, but ignored in the ESP,2 which concentrates on shorter CITs by regional allocation to reduce harm to organs from older donors.2 This survey confirmed worse HLA matches and inferior HLA favorability of RA DDRTs of recipients listed within the ETKAS program (Table 2).8,30 Furthermore, less advantageous CMV-constellations were just recently identified in a single-center study.8 Taking this into account, preferring a recipient with a more favorable match in CCOs in future and assumingly better HLA matches in REAL versus CCO might even have an additional positive impact on outcome (Table 4B). Notably, right kidneys were significantly more frequent in RA, which possibly might derive from apprehended technical problems due to the shorter vein and repeated decline in different centers.7 Overall patient and graft outcome after RA including PNF was comparable with SA despite prolonged CITs, older recipient and donor age, inferior HLA matches, and assumingly higher CMV risk. This observation must be ascribed to the pivotal impact of shortened waiting times in RA.8,13

Use of Kidney Transplants From RA in the Course of Time and Among ET Countries

The increasing use of kidneys from RA, especially since 2014 (Figure 2B), correlates unambiguously with the mounting need for more grafts, which is aggravated by both the demographic change over the last decades and consecutively more comorbidities of the donors. Today, every 10th DDRT within the ET area originates from RA compared with a range of rates between 4.8% and 26.4% previously reported in single centers.7,8,30,31 Furthermore, the effect of legal regulations concerning organ donation on the use of kidneys from RA was confirmed by this survey. The opting-in approach with its specific consent of the individual and deplorably low donation rates fosters the observed significantly longer waiting times and higher rate of RA in Germany (11%), whereas countries with the opting-out approach hardly use organs from RA (Table 1). However, despite an increased use of RA kidneys, decline rates of all kidneys offered before RA was initiated were comparable between the member states.

Facing the previously identified major benefit of shortened waiting times on transplant outcome13 despite marginal grafts in RA8,10-12 repeatedly declined in different centers for various reasons,7 DDRT through RA is reasonable and should be continued especially in countries with considerably prolonged waiting times due to organ shortage.

Rescue-allocated Kidney Recipient Survival

Most encouragingly, multivariate analyses adjusting for potential confounding factors revealed comparable patient survival and DwFG in ETKAS-listed recipients of RA versus SA DDRTs (Table 4B) despite worse recipient-, donor-, and transplant-specific characteristics in RA DDRT (Table 1). These results strongly encourage transplant physicians to continue DDRT via RA and debilitate any concerns of causing harm to recipients by use of RA grafts, which might derive from the mentioned characteristics in RA and the univariate analysis (Table 3; Figure 3). According to our data, more attention should be directed to favorable HLA matching, younger donor age, and short CIT. Whenever possible, these factors should be taken into consideration and a recipient with a better HLA match should be prioritized in CCOs, especially in young recipients. Just recently, an easily practicable algorithm for acceptance of RA offers and careful selection of eligible RA recipients was demonstrated to yield excellent outcome.8 Taken together, this offers the chance to include these variables into allocation (eg, REAL), provide more safety to the centers concerning acceptance or decline, and improve RA outcome in future.

In the face of a limited pool of grafts, we urgently have to accelerate transportation and implement virtual crossmatching to reduce CITs whenever reasonable. Prospectively, even more RA grafts might be transplanted this way and allow for a reduction in waiting time—the key to reducing mortality.13

Recipients with diabetic nephropathy and recipients of a retransplantation showed inferior outcomes in multivariate analyses as previously reported10-12,15 (Table 4B). If donor numbers markedly increased and waiting times decreased, then survival of these poor prognosis patients could potentially increase.

According to the multivariate analysis, senior recipients of RA DDRT clearly profited from RA as survival and DwFG were comparable with SA. Notably, patient survival after RA was borderline significantly better compared with SA (Table 4C). These findings underline our explicit recommendation to continue and even extend RA use. In ESP-listed recipients, HLA matching and donor age had no impact on survival, but short waiting times were favorable, which facilitates the selection of appropriate RA recipients in this subgroup. Short waiting times must be expected to have a significant impact on outcome after DDRT and naturally prevent death on the waiting list in seniors.

Graft Survival After Rescue-allocated DDRTs

Fortunately, the HR of graft survival after separate abbreviations: RA DDRT was superior to SA in the multivariate analysis accounting for confounding factors, such as inferior histopathological acute and chronic tissue damages, worse HLA matching and elevated CMV risk, longer CITs, older donor age, and significantly more adverse comorbidities and fulfilled ECD criteria in RA donors as reported before.8 Therefore, the acceptance of RA kidneys should be extended especially in countries with long waiting times. With regards to the multivariate analysis (Table 4B and C) and predictions (Table 5), all efforts need to be made to avoid loss of grafts from young donors by even accepting prolonged CITs. The effect of worse HLA matching and increased CMV transmission risk in RA8 is apparently less weighty on overall graft outcome than expected.

In senior recipients, graft survival after RA was equivalent to results from ESP SA and HLA matching may equally be neglected. Adding this to the excellent survival data, RA can be recommended for senior candidates as a potentially useful tool to provide these patients with a graft before deterioration, delisting, or death on the waiting list with increasing age and comorbidities in future.13,32,34,37

Limitations

The major limitation of this study is the retrospective data assessment from a noncompulsory database. Contribution to data completeness differs between the ET member countries. Although in some countries, including the Netherlands and Belgium, data reporting to ET are compulsory, in others, such as Germany, it is up to the centers. This explains the suboptimal data completeness in some parts, for example, the high rate of SA recipients without follow-up from Hungary and Germany. However, by use of statistical censoring, missing follow-up was correctly compensated for in the analyses and thanks to the participation of 71 transplant centers, data completeness was considerable after return of the questionnaires.

Unfortunately, relevant parameters, such as delayed graft function, rejection, biopsy-proven rejections, 1-y glomerular filtration rate, concomitant diseases, and detailed donor features, were not available. However, some of these issues can be assumed to be in accordance with results from previously published data like an increased 1-y glomerular filtration rate.8,30,31 Ideally, comprehensive reporting of these parameters would allow for subgroup analyses and enable identification of distinct candidates with a maximum profit of RA kidneys and particularly suitable donor–recipient combinations.

Finally, a tool including all relevant and available parameters to predict the expected benefit of RA in every single case over continuation of dialysis would be useful. An outcome predictor might even accelerate decision making in case of an offered organ via SA and potentially antedate RA initiation, which would reduce the CIT and therefore help to improve outcomes. Furthermore, comprehensive data on discarded organs could help to identify kidneys that were unnecessarily discarded. Unfortunately, these data cannot be generated from the current ET database by now.

CONCLUSIONS

DDRTs of kidneys offered via RA should be expanded for both ETKAS- and ESP-listed recipients according to their excellent outcome in patient and graft survival, which is fully comparable with SA. The use of RA kidneys is an adequate extension of the donor pool and should be extended to increase transplant numbers and reduce waiting times. The acquiescence of longer CITs, less favorable HLA matching, and inferior histopathological renal parenchymal quality of RA kidneys are compensated by the weighty benefit of a significantly shorter waiting time. Although both ETKAS- and ESP-listed recipients profited from DDRT of RA grafts, we recommend to adhere to certain basic donor- and transplant-specific parameters such as careful consideration of proteinuria, hypertension, and diabetes of the donor and a limited donor–recipient age difference like previously recommended.8 In CCOs for younger recipients, a patient with a favorable HLA match should be preferred over a candidate with an unfavorable match and even despite a potentially prolonged CIT in case of a young donor to further increase the outcome according to our data.

In ESP recipients, however, these considerations are secondary; the shortened waiting time in RA becomes even more attractive in the race against deterioration while waiting for SA, consecutively making RA a perfect supplement to the ESP.

This study clearly indicates that a mandatory joined register to collect all data on donors and recipients, including, for example, concomitant diseases, is urgently needed to identify those candidates who do or do not profit from RA, enabling transplant physicians offered a RA kidney to separate the wheat from the chaff. Apart from these factors, our allocation procedures and organ logistics must become quicker and virtual crossmatching has to be implemented to reduce CITs and thus improve the quality of all grafts.

In the meantime, transplant centers should individually define or revise their center-specific criteria for RA transplants, if not yet done.

ACKNOWLEDGMENTS

The authors thank all ET staff members for their support with sending the questionnaires and database analyses and the contributing centers for the data supply.

REFERENCES

1. Organ Procurement and Transplantation Network (OPTN). Policies, policy 8: allocation of kidneys. 2021. Available at https://optn.transplant.hrsa.gov/media/1200/optn_policies.pdf. Accessed January 12, 2020.
2. Eurotransplant Manual Version 8.2. Chapter 4: kidney (ETKAS and ESP). Eurotransplant Foundation; 2020. Available at https://my.eurotransplant.org/wp-content/uploads/2020/11/H4-Kidney-2020.3-November-2020.pdf. Accessed January 12, 2020.
3. Kidney Advisory Group on behalf of NHS Blood and Transplant (NHSBT) AP. Kidney transplantation: deceased donor organ allocation, POL186/10. 2019. Available at https://nhsbtdbe.blob.core.windows.net/umbraco-assets-corp/16915/kidney-allocation-policy-pol186.pdf. Accessed January 12, 2020.
4. Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med. 1999;341:1725–1730.
5. Eurotransplant Manual Version 8.2. Chapter 3–allocation general. Eurotransplant Foundation; 2020. Available at https://my.eurotransplant.org/wp-content/uploads/2020/09/H3-Allocation-v2020.2-September-2020.pdf. Accessed January 12, 2020.
6. Vinkers MT, Smits JM, Tieken IC, et al. Kidney donation and transplantation in Eurotransplant 2006-2007: minimizing discard rates by using a rescue allocation policy. Prog Transplant. 2009;19:365–370.
7. Farid S, Aldouri A, Fraser S, et al. Outcomes of kidney grafts refused by one or more centers and subsequently transplanted at a single United Kingdom center. Transplant Proc. 2009;41:1541–1546.
8. Assfalg V, Misselwitz S, Renders L, et al. Kidney transplantation after rescue allocation-meticulous selection yields the chance for excellent outcome. Nephrol Dial Transplant. 2021;36:551–560.
9. Metzger RA, Delmonico FL, Feng S, et al. Expanded criteria donors for kidney transplantation. Am J Transplant. 2003;3(suppl 4):114–125.
10. Querard AH, Le Borgne F, Dion A, et al. Propensity score-based comparison of the graft failure risk between kidney transplant recipients of standard and expanded criteria donor grafts: toward increasing the pool of marginal donors. Am J Transplant. 2018;18:1151–1157.
11. Ko KJ, Kim YH, Kwon KH, et al. Kidney transplantation using expanded-criteria deceased donors: a comparison with ideal deceased donors and non-expanded-criteria deceased donors. Transplant Proc. 2018;50:3222–3227.
12. Merion RM, Ashby VB, Wolfe RA, et al. Deceased-donor characteristics and the survival benefit of kidney transplantation. JAMA. 2005;294:2726–2733.
13. Meier-Kriesche HU, Kaplan B. Waiting time on dialysis as the strongest modifiable risk factor for renal transplant outcomes: a paired donor kidney analysis. Transplantation. 2002;74:1377–1381.
14. Assfalg V, Hüser N, van Meel M, et al. High-urgency kidney transplantation in the Eurotransplant Kidney Allocation System: success or waste of organs? The Eurotransplant 15-year all-centre survey. Nephrol Dial Transplant. 2016;31:1515–1522.
15. Assfalg V, Selig K, Tolksdorf J, et al. Repeated kidney re-transplantation—the Eurotransplant experience: a retrospective multicenter outcome analysis. Transpl Int. 2020;33:617–631.
16. Hughes RA, Heron J, Sterne JAC, et al. Accounting for missing data in statistical analyses: multiple imputation is not always the answer. Int J Epidemiol. 2019;48:1294–1304.
17. Ranganathan P, Pramesh CS. Censoring in survival analysis: potential for bias. Perspect Clin Res. 2012;3:40.
18. Therneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox Model. Springer; 2000.
19. Scheike TH, Sun Y, Zhang M-J, et al. A semiparametric random effects model for multivariate competing risks data. Biometrika. 2010;97:133–145.
20. Scheike TH, Zhang M-J. Flexible competing risks regression modeling and goodness-of-fit. Lifetime Data Anal. 2008;14:464–483.
21. Scheike TH, Zhang M-J. Analyzing competing risk data using the R timereg package. J Stat Softw. 2011;38:i02.
22. Heinze G, Dunkler D. Five myths about variable selection. Transpl Int. 2017;30:6–10.
23. Yacoub R, Nadkarni GN, Cravedi P, et al. Analysis of OPTN/UNOS registry suggests the number of HLA matches and not mismatches is a stronger independent predictor of kidney transplant survival. Kidney Int. 2018;93:482–490.
24. Debout A, Foucher Y, Trébern-Launay K, et al. Each additional hour of cold ischemia time significantly increases the risk of graft failure and mortality following renal transplantation. Kidney Int. 2015;87:343–349.
25. Postalcioglu M, Kaze AD, Byun BC, et al. Association of cold ischemia time with acute renal transplant rejection. Transplantation. 2018;102:1188–1194.
26. Revanur VK, Jardine AG, Kingsmore DB, et al. Influence of diabetes mellitus on patient and graft survival in recipients of kidney transplantation. Clin Transplant. 2001;15:89–94.
27. Opelz G, Döhler B. Association of HLA mismatch with death with a functioning graft after kidney transplantation: a collaborative transplant study report. Am J Transplant. 2012;12:3031–3038.
28. Therneau TM, Lumley T, Atkinson E, et al. A package for survival analysis in S. Version 2.38. 2015. Available at https://CRAN.R-project.org/package=survival. Accessed January 12, 2020.
29. Ozenne B, Sorensen AL, Scheike T, et al. riskRegression: predicting the risk of an event using cox regression models. R J. 2017;9:440–460.
30. Wahba R, Teschner S, Stippel DL. Results of kidney transplantation after rescue allocation. Transpl Int. 2011;24:e46–e47.
31. Wahba R, Suwelack B, Arns W, et al. Rescue allocation and recipient oriented extended allocation in kidney transplantation-influence of the EUROTRANSPLANT allocation system on recipient selection and graft survival for initially nonaccepted organs. Transpl Int. 2017;30:1226–1233.
32. Haugen CE, Chu NM, Ying H, et al. Frailty and access to kidney transplantation. Clin J Am Soc Nephrol. 2019;14:576–582.
33. McAdams-DeMarco MA, Law A, Salter ML, et al. Frailty and early hospital readmission after kidney transplantation. Am J Transplant. 2013;13:2091–2095.
34. McAdams-DeMarco MA, Law A, Salter ML, et al. Frailty as a novel predictor of mortality and hospitalization in individuals of all ages undergoing hemodialysis. J Am Geriatr Soc. 2013;61:896–901.
35. McAdams-DeMarco MA, Law A, King E, et al. Frailty and mortality in kidney transplant recipients. Am J Transplant. 2015;15:149–154.
36. Pelletier RP, Pesavento TE, Rajab A, et al. High mortality in diabetic recipients of high KDPI deceased donor kidneys. Clin Transplant. 2016;30:940–945.
37. Mehdorn A-S, Reuter S, Suwelack B, et al. Comparison of kidney allograft survival in the Eurotransplant senior program after changing the allocation criteria in 2010-A single center experience. PLoS One. 2020;15:e0235680.

Supplemental Digital Content

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.