Predicting Outcomes after Transplantation of Deceased Donor Kidneys of Marginal Quality within the Eurotransplant Service Area

The large gap between organ demand and supply in kidney transplantation necessitates the implementation of strategies to both maximize the use of donated organs and prolong the life of kidney allografts [1‒4]. On the other hand, medical advances have changed the current population structure, leading to a shift in the shape of the age pyramid. As a result, organs of inferior quality obtained from older donors or donors with multiple comorbidities are increasingly being offered to older and sicker recipients. This leads to increased rates of rejection or poorer outcomes after transplantation [5]. Many efforts have been made to address the issue of appropriate allocation of marginal quality donor kidneys. Prognostic scores based on purely clinical, purely histological, or combined clinicopathological criteria have been developed, but their performance is moderate to poor [6]. One of the most feared complications in this setting is the development of delayed graft function (DGF) as a consequence of ischemia-reperfusion injury of the transplanted kidney [7]. It has been proposed that graft function at 3 months after transplantation captures the degree of recovery after DGF and is a reliable predictor of long-term outcomes [8]. We aimed to test the performance of a recently proposed three-factor risk model by Sandal et al. [9] consisting of DGF, renal function recovery (RFR), and estimated glomerular filtration rate (eGFR) <30 mL/min per 1.73 m² after 90 days after transplantation in a European cohort of donors with kidneys of marginal quality.

We extracted data from the “Deutsche Stiftung Organtransplantation (DSO) Region Nord” and from the German transplant centers on kidneys allocated between January 2003 and March 2012. Recipient follow-up data collection was completed in December 2015. Data were analyzed between January 1, 2018, and May 31, 2020. Only adult recipients of deceased donor kidneys of lower quality were included. Recipients with dual-kidney and combined kidney transplantation were excluded. Our cohort consisted exclusively of brain death donors since donation after cardiac death is not allowed in Germany. Allocation was performed under the auspices of Eurotransplant, an international nonprofit organization responsible for the coordination and distribution of organs for transplantation among residents of 8 European countries (www.Eurotransplant.org).

The following donor parameters were used in this study: age; sex; body mass index (BMI); medical history; cause of brain death; eGFR at admission; highest, lowest, and terminal glomerular filtration rate (GFR); urine analysis at organ recovery; and development of acute kidney injury (AKI). We used four eGFR values because it was uncertain which of them was the most appropriate for assessing donor’s kidney function. For example, pre-admission eGFR is the most objective indicator of donor’s renal function but is often not available, while eGFR at organ recovery is always documented but unreliable because it reflects the cumulative kidney injury during hospitalization. On the other hand, eGFR at admission depends on pre-hospitalization events, such as out-of-hospital cardiac arrest, hypotensive episodes, or dehydration. After initial treatment admission, eGFR may increase or decline. In this case, the highest but not the lowest eGFR reflects donor’s true kidney function. For the recipients, information was sought on cardiovascular disease; immunological risk factors such as human leukocyte antigen (HLA) mismatches; mean and historical peak of plasma-reactive antibodies; peritransplant data such as cold and warm ischemia time; outcome parameters such as patient and graft survival at 1 and 3 years; and graft function at 3, 12, and 36 months.

The definition of marginality did not depend on strict criteria but was mostly based on clinical judgment at the time of procurement, considering the macroscopic appearance of the organ in combination with the donor’s medical history. Macroscopic assessment was performed on the “back table,” after removal of the perinephric fat and the clean dissection of the vessels from the surrounding tissue. It included organ quality and perfusion, both of which were graded as good, medium, or poor; likewise, atherosclerosis was characterized as no, mild, or severe. The decision was usually felt after discussion of each case between the senior surgeon of the harvesting team and the physician of the recipient’s center. Senior surgeons were accredited by the DSO and had many years of experience in transplantation.

Extended criteria donors (ECD) were classified as brain death donors >60 years of age or donors aged 50–59 years with at least 2 of the following: history of arterial hypertension, serum creatinine level higher than 1.5 mg/dL (133 µmol/L), or death due to cerebrovascular accident [10]. DGF was defined as the need for dialysis within the first 7 days after transplantation [11]. RFR is a calculated value that compares a recipient’s actual (observed) eGFR to their predicted eGFR and is expressed as a percentage (observed eGFR divided by predicted eGFR and multiplied by 100) [8]. The observed eGFR was calculated using the average of the best eGFR values within the first 3 months after transplantation. The predicted recipient eGFR was half of the donor’s terminal eGFR. In all cases, the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation was used to estimate GFR, expressed in units of mL/min per 1.73 m². Recipients were categorized in three groups according to the degree of recovery of renal function: >100%, 75–100%, and <75%. Donors were classified as low, intermediate, and high risk if they had, respectively, one, two, or three of the components of the score (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000540304) [9].

Admission, highest, lowest, and terminal eGFR were estimated using the first creatinine after admission, the lowest and highest creatinine during hospitalization or intensive care unit stay, and the last creatinine before organ recovery [12]. AKI was defined taking into consideration the first and the last or the minimum and the maximum serum creatinine value during hospitalization and classified according to the Risk, Injury, Failure, Loss, End-stage renal disease (RIFLE) criteria [12].

We explored the risk associations of DGF, RFR, and eGFR <30 mL/min/1.73 m² with death-censored graft loss, all-cause graft loss, and 1- and 3-year mortality. Death-censored graft loss was defined as the return to dialysis with the patient being alive, graft loss as patient death with a functioning graft, and patient loss as patient death from any reason [13].

The distributions of the recipient, donor, and graft characteristics at transplantation for the assessment of DGF, RFR, and eGFR were evaluated using the t test or Kruskal-Wallis test for continuous variables and the χ² or Fisher’s exact test for categorical variables. Kidney transplant recipients were censored at the time of graft loss (return to dialysis or re-transplantation), death, or the end of the study period, whichever occurred first. Three separate multivariable Cox proportional hazards models were created to assess the outcomes of graft failure, death-censored graft failure, and patient death. Nonlinear continuous variables were made categorical. Nonsignificant clinical nonrelevant variables were removed from the model using backward elimination. Wald statistics were used to assess the significance of exposure variables. The models were assessed using the Harrell C statistic and Akaike Information Criterion (AIC). A p value <0.05 was considered significant in all two-tailed tests. Statistical analysis was performed using SPSS software, v29 (IBM Corp, Armonk, NY, USA) and IBM SPSS Statistics Essentials for R.

The flowchart of the study is illustrated in Figure 1. Of the 442 kidneys recovered from 221 donors, 149 were not transplanted. For the remaining 293 transplanted kidneys, follow-up data for the actual analysis were available from 196 donor-recipient pairs.

The Kidney Donor Risk Index (KDRI) was 1.533, and 133 (67.5%) were ECD. The average age was 67 years, and 49% were females. Only 11% of donors had diabetes, and 26%, cardiovascular disease. Cerebrovascular accident was the most common cause of brain death (60%). The eGFR at recovery was 49.5 mL/min; 7% of donors showed severe proteinuria, and 41% experienced AKI during their hospitalization in the general ward or intensive care unit. The average age of the recipients was 65 years, 33% were female, 28% had diabetes, and 44% had cardiovascular disease. They had a low immunological risk profile, a relatively short cold ischemia time (CIT) of 13.1 h, and a high estimated posttransplant survival (EPTS) score. DGF was experienced by 109 of the 196 (55.6%) recipients. Twenty-five patients died, and 26 lost their grafts during a median follow-up of 43.8 months (interquartile range 19–68 months). Patient and death-censored graft survival at 1, 3 years after kidney transplantation were 90.8% and 86.2%, and 91.0% and 86.0%, respectively. Donor and recipient data are summarized in Table 1.

Recipients with DGF had a higher BMI and a longer dialysis vintage. Except for diabetes, which was less common in the DGF group, no differences were observed in the other cardiovascular risk factors, immunological profile, and peritransplant parameters. The respective donors were younger, and the recovered kidneys were of lower macroscopic quality. Although RFR at 3 months was comparable between recipients with and without DGF, chronic kidney disease (CKD) stage IV (eGFR <30 mL/min/1.73 m²) was more common in the first group (Table 1). Dialysis vintage was longer, and DGF and CKD stage IV were more common in recipients with the lowest RFR. This group received mostly (84.4%) organs from ECD (Table 2). Compared to those with eGFR ≥30 mL/min/1.73 m², recipients with CKD IV (<30 mL/min/1.73 m²) at 3 months after transplantation developed DGF more frequently, had a lower RFR, were older, and had a higher BMI and a higher EPTS score. The corresponding donors were older, had more diabetes and hypertension, were more often categorized as ECD, and had a higher KDRI (Table 3).

Recipients with DGF had a higher risk of death-censored graft failure (HR, 95% CIs: 3.058 [1.195–7.825]), and recipients with GFR <30 mL/min/1.73 m² at 3 months posttransplant had a higher risk of death censored and all-cause graft failure (HR, 95% CIs 2.457 [1.214–4.976] and 2.122 [1.129–3.990]). RFR of any degree was not found to be associated with death-censored and all-cause graft failure. All three components of the proposed score [9] (DGF, eGFR <30 mL/min/1.73 m², and RPF) were not associated with a higher risk of mortality. Finally, there were no statistically significant risk associations between any of the defined donor risk categories and death-censored graft failure, all-cause graft failure, and mortality (Table 4 and online suppl. Table 2).

The primary aim of this retrospective study was to test the performance of a simple score consisting of three easily obtained parameters of the early posttransplant period, in a population of donors of European origin with organs of marginal quality. The rather dismal clinical outcome of our cohort with 49% of recipients experiencing DGF and nearly 20% losing their grafts after 5 years highlights the unresolved issue of the appropriate allocation of marginal organs and the need to develop a reliable tool aiding physicians to select the most suitable kidney for each recipient.

In contrast to the primary publication by Sandal et al. [9], the investigated score could not predict outcomes in our cohort for three reasons. First, there were significant differences in the quality of the donated kidneys: the terminal donor eGFR was approximately half in our population, with more donors developing (AKI). Second, our recipients had a significantly longer dialysis vintage; showed a higher EPTS score, with nearly 60% of them belonging to group 5 (81–100%); and experienced CKD stage IV more often early after transplantation. Third, the lack any association of DGF, RFR, and CKD IV with patient mortality was probably due to the lower follow time of our cohort (3.7 vs. 7.3 years).

But the main difference at all was that the score of Sandal et al. [9] was developed in a population of brain death donors with kidneys of above average quality, whereas our cohort consisted exclusively of donors with marginal kidneys. We believe that donor AKI with and without recipient DGF [14] and the different creatinine metrics during the donors’ index hospitalization [15] are more important risk factors for the short-term prognosis of marginal kidneys than intermediate postoperative factors such as RFR and eGFR <30 mL/min/1.73 m² at 3 months after transplantation.

The paradoxical finding of higher CIT in the recipients with foul RFR is due to the peculiar characteristics of our donors, who were selected according to the availability of procurement biopsies. The definition of marginality in ET is not established and relies almost exclusively on subjective criteria. In this context, procurement biopsies were performed on a case-by-case basis and not according to a standardized protocol, mainly to increase the acceptability of an organ that would otherwise have been discarded. Our cohort was therefore highly selective with a considerable proportion of our donors having AKI at organ recovery. Probably, the longer CIT in recipients with RFR >100% was not decisive or was outweighed by the more favorable profile of other known risk factors for RFR, such as younger donor age, lower KDRI, and shorter time on dialysis. Beyond that, the CIT time was in the lower range of as that reported by ET (13.1 [10.8–16.4] versus 15 [12–20]) and much lower of that reported in the USA (18.2 [11.2]) [16, 17]. BMI is a known a risk factor for DGF; therefore, the higher BMI in recipients with DGF was more or less an expected and not a surprising finding [7, 18]. We suppose that the higher BMI in recipients with eGFR <30 mL/min reflects the higher prevalence of DGF in this group (Table 3; 65 vs. 49%, p = 0.030).

Limitations of our study should also be acknowledged. First, there was a bias toward marginal kidneys. Second, the performance of the discarded kidneys, which constituted more than one-third of the offered, is not known. Third, our sample size was rather small and certainly not representative of the whole population of ET donors and recipients, where the majority of organs are allocated without procurement biopsies [1, 19]. Fourth, our cohort consisted exclusively of brain death donors as donation after cardiac death is not allowed in Germany. Lastly, the recipient population was homogenous and consisted only of recipients having been on hemodialysis before transplantation. There were no patients on peritoneal dialysis or after preemptive transplantation.

In conclusion, any combination of DGF, CKD stage IV, and degree of early RFR is not a suitable parameter for the prediction of the short-term outcome after transplantation of marginal kidneys or donor kidneys with acute and chronic injury. The appropriate allocation of marginal kidneys recovered from donors after brain or cardiac death remains still an unresolved issue at the global level and in the Eurotransplant era as well. Unfortunately, a considerable proportion of these valuable organs with acceptable quality are not utilized [20].

We would thank the following German transplant centers for providing us with recipient data for the present analysis: Transplantationszentrum Uniklinik RWTH Aachen; Transplantation Berlin-Charité-Universität Berlin; Transplantationszentrum Bochum; Transplantationszentrum-Universitätsklinikum Bonn; Dr. Frans Zantvoort, Klinikum Bremen Mitte; Transplantationszentrum Uniklinik Düsseldorf; Transplantationszentrum-Uniklinikum Essen; Prof. Dr. Med. Martina Koch, Universitätsklinikum Hamburg-Eppendorf; Transplantationschirurgie-Uniklinik Innsbruck; Transplantationszentrum-Uniklinik Köln; PD Dr. Med. Antje Habicht, Transplantationszentrum München der LMU-LMU Klinikum; Prof. Dr. Med. Bernhard Banas, Universitäres Transplantationszentrum Regensburg; PD Dr. Med. Martin Nitschke, Universitätsklinikum Schleswig-Holstein; Dr. Peter Weithofer, Transplantationszentrum Süd-Niedersachsen.

This study protocol was approved by the Ethics Committee of Hannover Medical School (No. 1519-2012), which exempted the requirement for informed consent for this retrospective analysis. The study protocol was conducted in accordance with the Declarations of Helsinki and Istanbul on Organ Trafficking and Transplant Tourism.

The authors have no conflicts of interest to declare.

No funding was obtained for this study.

Christos Chatzikyrkou and Florian Gunnar Scurt conceived the study. Jan Ulrich Becker re-evaluated the biopsies. Christos Chatzikyrkou acquired recipient data. Carl-Ludwig Fischer-Fröhlich provided donor data on behalf of the DSO. Christos Chatzikyrkou wrote the manuscript. Florian Gunnar Scurt created the tables and figures and performed the statistical analysis. Peter Rene Mertens and Angela Ernst provided intellectual input and reviewed and edited the manuscript. All the authors approved the final version of the revised manuscript.

Data are available on request from the authors.