Even though kidney transplantation is the best treatment for end-stage renal disease, the severe shortage of transplantable organs remains an unavoidable topic [1
]. With the aging demography of China, the number of potential elder donors who die of hypertension and cerebrovascular accidents also increases. As opposed to the kidneys from standard criteria donor (SCD), the clinical use of expanded criteria donor (ECD) derived kidneys was getting more focus recently [3
The Chinese Government initiated the Donation after Citizens' Death policy in 2010, which was then promoted nationwide after a 3-year pilot implementation period [4
]. To now, it has been a major source of organs for transplant. Since it is still a young policy, there are differences in the mature international practices regarding recipients' race, organ procuring, distribution rules, organ function quality assessment, perioperative management of recipients, etc. [5
]. Corresponding clinical evidence is still urgently needed for its improvement.
This study is based on the data of kidney transplantation in our center with the Donation after Citizens' Death policy to compare the long-term outcomes of ECD and SCD organs in multi-aspects, aiming to provide a reference for the clinical use of ECD kidneys.
Materials and methods
Patients and study design
This retrospective study enrolled 415 kidney transplants from 211 donors based on the Donation after Citizens’ Death conducted between October 2011 and October 2019. The ECD meets that the donor is older than 60 years old, or between 50 and 59 years old, and meets at least two of the following criteria: 1. Final serum creatinine > 1.5 mg/dL(132 μmol/L), 2. Cerebrovascular accident as the cause of death, 3. History of hypertension [3
]. Based on this criterion, the donors, corresponding kidneys, and corresponding recipients were divided into the ECD and SCD groups for analysis.
All patients awaiting kidney transplants with end-stage renal disease were registered in the China Organ Transplant Response System (COTRS). Patients with contraindications for kidney transplantation (such as metastatic malignancy, active presence of HIV infection, and other reasons) were excluded. The CORTS algorithm was rigorously adhered to. Higher panel-reactive antibody (PRA) levels led to a lower priority for transplant, and we avoided transplantation in patients with PRA > 30%. To lower the rate of rejection following transplantation, prophylactic therapies (such as plasmapheresis, immunoadsorption, and medication therapy) were administered to enrolled patients with high immunogenicity (peak PRA > 50%).
The donor family members (spouses, adult children, and parents) consented to organ donation after death and signed the appropriate informed consent paperwork voluntarily. Donors who did not meet the usage criteria based on a needle biopsy performed prior to transplantation were ruled out [6
Data were collected from the electronic medical record system and the registry system of the organ donation database of our center, which were analyzed anonymously. Following surgery, recipients were intensively monitored during the hospitalization and then followed up by the out-patient clinic at regular intervals after discharge. Table 1
shows the patient features.
Baseline characteristics of donors and recipients
N = 52
N = 159
54.31 ± 3.96
30.99 ± 11.84
Sex, n (%)
| || |
History of hypertension, n (%)
HBV infection, n (%)
Cause of death, n (%)
| || |
Terminal Scr (μmol/L)
N = 104
N = 311
40.04 ± 11.03
38.53 ± 10.09
Sex, n (%)
| || |
Cause of renal failure, n(%)
| || |
History of hypertension, n (%)
History of diabetes, n (%)
Dialysis duration (months)
| || |
Cold ischemia time (h)
8.94 ± 2.70
8.0 ± 2.69
Negative PRA, n (%)
3.07 ± 1.14
2.82 ± 1.26
Cardiac arrest, n (%)
| || |
Warm ischemia time (min)
| || |
≤ 15 min
> 15 min
Induction therapy, n (%)
| || |
Before kidney transplantation, three pairs of the six human leukocyte antigens (HLA) –A, -B, and –DR were tested. Crossmatch testing for complement-dependent cytotoxicity (CDC) was negative in all recipients. PRA of recipients was routinely tested before transplantation by an enzyme-linked immunosorbent assay (ELISA) including PRA –I and PRA-II. We defined peak PRA < 10% as negative.
Induction therapy mainly consisted of anti-interleukin-2 receptor monoclonal antibody (basiliximab, Simulect®, Novartis) or anti-thymocyte globulin (ATG, Thymoglobuline®, Genzyme). Patients without the HLA antibodies received basiliximab, which was administered in two 20 mg doses by bolus intravenous injection. The first bolus was given within the 2 h before revascularization of the graft and the second one is on day 4 post-transplant. Patients with HLA antibodies were given single bolus ATG induction therapy at a dose of 50-75 mg. ATG was regular intravenous infusion within 6 h before graft’s revascularization and maintain 25 mg daily until 3 days post-transplant. Before starting induction therapy, 40 mg of methylprednisolone was injected intravenously to prevent the side effects of ATG and basiliximab.
Standard immunosuppressive triple therapy consists of tacrolimus (FK-506) or ciclosporin A (CsA), mycophenolate mofetil (MMF) or mycophenolate sodium (MPS), and prednisone. MMF (1 to 2 g/day) or MPS (0.72 to 1.44 g/day) was administered immediately following the transplant. The dosage was adjusted based on the blood routine examination of patients and tapered over time. The administration of CsA (6 ~ 8 mg/kg/day) or FK-506 (0.1 ~ 0.15 mg/kg/day) was initiated from day 1 of post-transplant, and the doses were adjusted according to the trough levels of the drugs.
The serum concentrations of FK-506 and CsA were routinely monitored following kidney transplantation. For FK-506, serum was obtained half an hour prior to administration (C0), and for CsA, serum was collected at C0 and 2 h following medication (C2).
The target levels were timely varying at 1 month, 1 to 3 months, 4 to 12 months, and > 1 year following transplantation. The C0 target levels of FK-506 concentrations were decreased from 8–12 ng/mL to 6–10 ng/mL, 4–10 ng/mL and 4–8 ng/mL. The C0 target levels of CsA reduced from 150–300 ng/mL to 150–250 ng/mL, 120–250 ng/mL and 80–120 ng/mL. The C2 target levels of CsA concentrations were reduced from 1000–1500 ng/mL to 800–1200 ng/mL, 600–1000 ng/mL and > 400 ng/mL, respectively. The target level was also individually modified according to the patient’s condition. Oral MMF (1 to 2 g/day) or MPS (0.72 to 1.44 g/day) also continued to be used for maintenance immunosuppressive therapy and individually modified according to the patient’s conditions. Oral prednisone was subsequently prescribed at a daily dose of 20 mg. Then the daily dose was tapered to 10 mg in 6 months.
Delayed graft function (DGF) is used to describe the status of transplanted kidneys that fail to function immediately after transplantation and is a significant complication of kidney transplantation. In this study, DGF was defined as the need for dialysis during the first week after transplantation [7
Acute rejection (AR) was suggested clinically by an unexplained rise in serum creatinine concentration of > 0.3 mg/dL or a 25% increase from baseline [8
]. The diagnosis of AR was confirmed by percutaneous kidney biopsy, and kidney pathology was classified using Banff 07 classification and its subsequent updates [9
Warm ischemia time (WIT) was defined as the time interval between the withdrawal of life support to cold perfusion.
The all-cause infections in 1-year post-transplant were also analyzed, including surgical site infection, and pulmonary, and urinary tract infections.
The estimated glomerular filtration rate (eGFR) was calculated by using the CKD-EPI eGFR equation based on gender, age, and serum creatine (Scr) [10
Results were expressed as percentages for categorical variables and numerical values for continuous variables, respectively. The baseline features of the SCD and ECD groups were compared using the Chi-square test, Mann–Whitney U test, and Student’s t-test as appropriate. The Kaplan–Meier method and Log-rank test were used to compare the graft/recipient survival, occurrence of AR and all-cause infection between the two groups and generated the survival curve. The log-rank test was used to analyze statistical differences between curves. Differences in the incidence of DGF between the two groups were determined by using the Chi-square test. Comparisons of eGFR were using the Mann–Whitney U test. Cox univariate/multivariate regression analysis was utilized to determine risk factors and the hazard ratio for graft failure. Every test was two-tailed. P values < 0.05 were regarded as statistically significant.
Due to the severe imbalance between the supply and demand of transplantable kidneys, some patients remain on dialysis and experience long waits on the waitlist. In 2020, only about a quarter of waitlisted patients received a deceased donated kidney transplant within 5 years [1
]. While the Donation after Citizens' Death policy becomes an effective way to address this urgent need [4
]. As one of the pilot centers, our center has performed kidney transplantation with this new policy since 2011. However, there are few long-term studies on it due to its short implementation period. There are still many rules that need to be reconsidered and improved. Corresponding long-term clinical evidence is needed. Despite the potential risks associated with ECD-derived organ use, the clinical benefits cannot be denied [11
]. Increasing the use of ECD-derived kidneys is a way to expand the donor pool but is also a result of the aging demography of China. So, evaluating and using ECD-derived kidneys is of great significance.
This study presented the results of 415 kidney transplants at our center since the implementation of the Donation after Citizens' Death policy. Its primary outcomes were satisfying and comparable to established international practices [12
]. This study used the conventional definition of ECD, which makes the result comparable with results published internationally. Previous studies have shown poor survival in ECD-derived kidneys [13
], as demonstrated in our study (Log-rank test, P
= 0.013; Multivariate cox regression, P
= 0.044, HR = 2.27). However, the results of this study were slightly higher than those of similar studies in terms of patient and graft survival. This may be because our center is cautious in the selection of donated kidneys, and the age of both donors and recipients (especially the ECD group) is slightly lower than that of them, leading to a better prognosis. In addition, in the results of this study, the eGFR of ECD-derived kidney recipients was significantly worse than that of SCD-derived kidney recipients at 3 months, 6 months, 1 year, 3 years, and the highest value posttransplant. That has a similar result to the cohort from Nagaraja et al. [14
]. The statistical difference in eGFR turned negative at 5 years after transplantation may be due to the tendency of kidney loss with poor renal function.
The results of postoperative complications in this study showed that ECD-derived kidney recipients were more frequent to develop AR after surgery, while there was no difference in DGF and all-cause infection. Similar results can be observed in other clinical practices [16
]. The overall age of the recipients in this study was 40.04 ± 11.03 years old, a young group. In organ transplantation practice, young recipients have a more robust immune response to antigens. While due to proper T-cell effector immune response with an intact regulatory and memory T-cell response, the aged recipients may be weaker immune responses [18
]. A recent study by Iske et al. [19
] found that the increased content of free mitochondrial DNA in the organs of elderly donors would activate CD11c and DC cells of the recipients, thereby promoting the proliferation of helper T cells and the secretion of IL-17A. This immune response was more robust in young recipients and more likely to lead to AR occurrence. This may also be one of the mechanisms contributing to this result.
The value of ECD-derived kidney clinical use is still controversial [20
]. A study comparing ECD recipients with dialysis patients published in JAMA showed that only diabetic recipients and candidates older than 40 years with long waiting times could benefit from receiving an ECD donor kidney [11
]. Thus, ECD-derived kidneys may provide a benefit, but it is limited. So, finding an appropriate way to use and allocate these marginal kidneys became a tremendously meaningful topic. At present, the mainstream view prefers to allocate ECD-derived kidneys to older recipients. Since 1999, European countries have recommended the implementation of the European Advanced Transplant Program, a so-called "old-to-old" allocation system, so that kidneys from donors over 65 years old prefer to be allocated to recipients over 65 years old to reduce the wait time for organs [21
]. Similarly, In 2014, a new kidney allocation system was implemented in the United States. They expanded the donation criteria and included matching the recipients' life expectancy with organ life expectancy in allocate rules. In this way, the donor pool is expanded, and more elderly patients have access to organ transplants [22
]. Based on the above references, the allocation rules of the Donation after Citizens' Death policy for organs from the elderly donor can be reconsidered and further defined.
This study also showed that the occurrence of AR, more mismatches HLA loci, and uncontrolled cardiac arrest, prolonged dialysis duration as predictors of graft loss. Among them, AR occurrence and more HLA mismatch may cause immunogenic injury, uncontrolled cardiac arrest leads to longer ischemia time and ischemia–reperfusion injury, the adverse effects of the prolonged period of dialysis on kidney graft survival have been demonstrated in clinical studies [4
]. These risk factors had been well explored in clinical research, and results could guide clinical work.
There were still several limitations to our study. Some factors have been confirmed to be independent risk factors for kidney allograft loss (such as DGF occurrence [24
] et al.), and no significant results were obtained in our study, which may be due to the statistical bias caused by median sample size. Besides, inclusion bias and omitted confounding factors may also influence the results.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.