Abstract
Deceased donor kidney transplantation (DDKT) has a serious problem of donor organ shortage, particularly in Asia. Consequently, donor kidneys with acute kidney injury (AKI) have been utilized for transplantation. This study aimed to evaluate the prognosis of DDKT using AKI kidneys. We analyzed the data of 8,523 adult patients who underwent DDKT between 2008 and 2022 using the Korean nationwide DDKT database. The association between donor AKI and death-censored graft failure was assessed using competing risk analysis, with adjustment through inverse probability of treatment weighting (IPTW). The AKI group included 3,453 patients (40.51%) and had a higher death-censored graft failure risk than the no-AKI group (hazard ratio: 1.27; 95% confidence interval: 1.05–1.53). Especially, AKI stage 3 subgroup had a higher death-censored graft failure risk than the no-AKI group. Furthermore, the AKI group had a higher death-censored graft failure risk than the no-AKI group in the subgroup with a Korean-kidney donor profile index (K-KDPI) ≥ 70 or the stable- or increasing trend of serum creatinine. In conclusion, using AKI kidneys with a K-KDPI < 70% or a decreasing creatinine trend, or at stage 1 and 2, could help in combating organ shortage, especially in areas with a long waiting time.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-026-37147-0.
Keywords: Acute kidney injury, Deceased donor kidney, Kidney transplantation, Organ shortage
Subject terms: Diseases, Medical research, Nephrology
Introduction
Although kidney transplantation (KT) is the treatment of choice for end-stage kidney disease (ESKD)1,2, the supply of donor organs remains insufficient to meet the global demand3. Deceased donation rates in Asian countries are much lower than in Western countries. For example, the donation rates in Iran and Korea were 11.50 and 8.56 per million population (pmp), whereas those of USA and Spain were 41.60 and 40.80 (pmp) in 20214. However, both ESKD incidence and the its yearly change were higher in Asian countries than in Western countries; for example, the top two increasing rates from 2010 to 2020 were 19.7 and 18.8 pmp/year in Thailand and Korea, respectively5. Therefore, donor organ shortage is a more serious problem in Asian countries and maximal utilization of deceased donors’ kidneys is very important issues to improve this problem6,7.
One of the efforts employed is the active utilization and efficient allocation of expanded criteria donor (ECD) kidneys to minimize disposal of donors’ kidneys8–10. The USA developed Kidney Donor Profile Index (KDPI) beyond dichotomous ECD criteria to define the prognostic status of a donor’s kidneys in more detail11,12. Furthermore, suitable recipient candidates whose deceased donor kidney transplantation (DDKT) using ECD kidneys had survival benefits over waiting for DDKT using standard criteria donor (SCD) were proposed to efficiently utilize the ECD kidneys13,14.
Another effort to minimize donor kidney disposal is to use kidneys that have acute kidney injury (AKI) before procurement. US studies found that donor AKI did not decrease graft survival15,16. Contrary to the above, a UK study reported that the donor’s AKI increased 1-year graft failure and mortality17, and a French study reported donor’s AKI was associated with decreased graft survival18. These conflicting results suggest that the outcomes of DDKT using AKI kidneys would be varied according to an individual country’s characteristics. Therefore, there is a need to assess the outcomes of DDKT using AKI kidneys based on each country’s own data. This Korean nationwide study aimed to investigate the impact of donor AKI based on its subgroups, such as subgroups based on AKI stages, K-KDPI categories, and trends of creatinine change, on allograft outcomes to propose criteria for adequate use of donors’ kidneys with AKI to ensure the maximal utilization and good outcomes of DDKT.
Materials and methods
Study population
We used a nationwide cohort of solitary adult DDKT patients from the Korean Network for Organ Sharing (KONOS) and the National Health Insurance Data Sharing Service (NHISS) database. A total number of 11,113 patients received DDKT between 2008 and 2022. Recipients under 19 years of age (n = 351), and those with missing or invalid information regarding height, serum creatinine, diabetes mellitus, dialysis duration (n = 2,239), were excluded from this study. Eventually, 8,523 DDKT patients were used in the final analysis (Fig. 1). This study was approved by the Institutional Review Board of Yonsei University Severance Hospital (4–2021-1358), and the review board waived the requirement for informed consent. This study was performed in accordance with the Helsinki Declaration and the Declaration of Istanbul of 200819.
Fig. 1.
Flow chart for study population. CVA cerebrovascular accident, DDKT deceased donor kidney transplantation, KT kidney transplantation.
Definitions and classification of AKI according to severity, K-KDPI, and changing trends in serum creatinine levels
Donor AKI was defined according to AKI Network (AKIN) criteria and classified as follows; stage 1, increase in serum creatinine from admission to the terminal by ≥ 0.3 mg/dL or 1.5 to < 2-fold; stage 2, 2 to < 3-fold; and stage 3, ≥ 3-fold, or terminal serum creatinine ≥ 4.0 mg/dL after an increase of ≥ 0.5 mg/dL from admission or dialysis support20. Further, donors were categorized into low and high K-KDPI groups using two different cutoffs: the Korean threshold of 70% as the primary threshold and the US threshold of 85% for sensitivity analysis15,21. A previous Korean study defined ECDs as those with a K-KDPI ≥ 70%, as this cutoff demonstrated good diagnostic performance and a significant survival difference compared with K-KDPI < 70%21.
Serum creatinine concentrations were measured at least three times before procurement, and creatinine levels at admission was considered as the baseline levels16,17,22. By changes from the peak creatinine concentrations to final creatinine concentrations just before procurement, changing trends in serum creatinine were defined as follows; stable or increasing trend (change > − 0.3 mg/dl), and decreasing trend (change ≤ −0.3 mg/dl)23,24. Estimated glomerular filtration rate (eGFR) was calculated using chronic kidney disease-epidemiology collaboration equations25.
Outcomes
The primary outcome was death-censored graft failure, defined as a reinitiating dialysis, or going through re-transplantation. Death-uncensored graft failure and all-cause mortality were used as secondary outcomes in sensitivity analysis.
Statistical analysis
The association between donor AKI and death-censored graft failure was assessed using competing risk analysis. The primary analysis was conducted using cause-specific Cox proportional hazards models with adjustment for donor-, recipient-, and transplantation-related covariates in Model 1. In Model 2, the center effect was incorporated into the multivariable analysis by using a frailty function and donor-related covariates (weight, sex, death due to cerebrovascular accident [CVA], hypertension, and donation after circulatory death [DCD]) were further adjusted using stabilized and trimmed inverse probability of treatment weighting (IPTW). The results were expressed as a hazard ratio (HR) with a 95% confidence interval (CI). Statistical significance was set at P < 0.050. The adequacy of IPTW adjustment was confirmed by assessing the absolute mean difference (Fig. S1).
Sensitivity analyses were conducted using Fine and Gray subdistribution hazard models and multiple imputations (Table S3 and S4). Multiple imputations using chained equations were performed to address missing variables, which were used as covariates in the main analysis with missingness < 20% (donor hypertension, donor diabetes mellitus, death due to CVA, previous KT history, and dialysis duration before KT). Predictive mean matching with k-nearest neighbors (k = 5) was applied, and 10 independent imputed datasets were generated. Multivariate cause-specific hazard models were then fitted across the imputed datasets, and estimates were pooled using Rubin’s rules.
All analyses were conducted using SAS (version 7.1, SAS Institute Inc., Cary, NC, USA), R (version 4.4.2, R Core Team, 2022, R Foundation for Statistical Computing, Vienna, Austria; URL: https://www.R-project.org/), and its packages (version 2024.12.0, RStudio Team, PBC, Boston, MA, USA; URL: http://www.rstudio.com). More detailed description of statistical methods was described in the supplementary methods.
Results
Comparison of clinical characteristics between AKI and no-AKI groups
Table 1 compares the clinical characteristics of donors and recipients between AKI and no-AKI groups. Among 8,523 DDKT patients, the no-AKI and the AKI groups included 5,070 (59.49%) and 3,453 (40.51%), respectively. Compared with donors without AKI, donors with AKI were older and had more males, diabetes mellitus, and hypertension. Donors in the AKI group had lower eGFR (41.75 ± 24.67 vs. 100.11 ± 31.52 ml/min/1.73 m2, P < 0.001) and higher K-KDPI (69.21 ± 32.90 vs. 53.37 ± 36.65, P < 0.001) compared to those in the no-AKI group. Serum creatinine levels at admission were lower in the AKI group compared to the no-AKI group. Recipients in the AKI group were older and had a lower proportion of previous KT history.
Table 1.
Clinical characteristics of kidney transplant patients according to presence of donor AKI.
| Variables | No-AKI (n = 5,070) | AKI (n = 3,453) | P value |
|---|---|---|---|
| Donor-related factors | |||
| Age, year | 45.25 ± 15.62 | 47.77 ± 13.61 | < 0.001 |
| sCr at admission, mg/dL | 1.06 ± 0.47 | 1.05 ± 0.62 | 0.138 |
| sCr just before procurement, mg/dL | 0.87 ± 0.41 | 2.41 ± 1.38 | < 0.001 |
| eGFR just before procurement | 100.11 ± 31.52 | 41.75 ± 24.67 | < 0.001 |
| Diabetes mellitus | 381 (7.51%) | 390 (11.29%) | < 0.001 |
| Gender, male | 3,217 (63.45%) | 2,477 (71.73%) | < 0.001 |
| Hepatitis C virus | 14 (0.27%) | 14 (0.42%) | 0.305 |
| Hepatitis B virus | 158 (3.12%) | 106 (3.07%) | 0.903 |
| K-KDPI | 53.37 ± 36.65 | 69.21 ± 32.90 | < 0.001 |
| K-KDPI ≥ 70, n | 2,188 (43.16%) | 2,151 (62.29%) | < 0.001 |
| K-KDPI ≥ 85, n | 1,522 (30.02%) | 1,669 (48.33%) | < 0.001 |
| US-KDPI | 52.13 ± 26.02 | 69.09 ± 21.03 | < 0.001 |
| DCD | 12 (0.24%) | 14 (0.41%) | 0.165 |
| Cause of death, CVA | 2,039 (40.22%) | 1,622 (46.97%) | < 0.001 |
| Donor AKI stage | < 0.001 | ||
| No AKI | 5,070 (100.00%) | NA | |
| Stage 1 AKI | NA | 1,472 (42.63%) | |
| Stage 2 AKI | NA | 733 (21.23%) | |
| Stage 3 AKI | NA | 1,248 (36.14%) | |
| CRRT | 0 (0.00%) | 452 (13.09%) | < 0.001 |
| Recipient-related factors | |||
| Age, year | 49.49 ± 10.81 | 51.52 ± 10.70 | < 0.001 |
| Gender, male | 3,007 (59.31%) | 2,135 (61.83%) | 0.019 |
| Dialysis duration, year | 6.56 ± 3.95 | 6.62 ± 3.90 | 0.475 |
| Diabetes mellitus | 4,330 (85.40%) | 2,985 (86.45%) | 0.175 |
| Hepatitis C virus | 88 (1.74%) | 61 (1.77%) | 0.915 |
| Hepatitis B virus | 338 (6.67%) | 223 (6.46%) | 0.991 |
| History of KT | 485 (9.57%) | 283 (8.20%) | 0.030 |
| Ischemic heart disease | 3,483 (68.70%) | 2,460 (71.24%) | 0.012 |
| Cerebrovascular accident | 1,853 (36.55%) | 1,280 (37.07%) | 0.624 |
| Cancer | 1,633 (32.21%) | 1,071 (31.02%) | 0.245 |
| Transplantation-related factors | |||
| Numbers of HLA mismatch | 3.69 ± 1.60 | 3.78 ± 1.54 | 0.007 |
| CIT, min | 269.00 ± 179.30 | 270.26 ± 150.33 | 0.786 |
| Positivity of PRA | 1,059 (20.89%) | 785 (22.73%) | < 0.001 |
| Outcomes | |||
| Death | 553 (10.91%) | 421 (12.19%) | 0.067 |
| Death-censored graft failure | 271 (5.35%) | 200 (5.79%) | 0.108 |
AKI acute kidney injury, CIT cold ischemic time, DCD donation after circulatory death, HLA human leukocyte antigen, K-KDPI Korean-kidney donor profile index, N number, NA not applicable, sCr serum creatinine, US-KDPI United States-kidney donor profile index.
Table S1 shows the comparison of the clinical characteristics among different AKI stages (AKI stage 1, n = 1,472, 42.63%; AKI stage 2, n = 733, 21.23%; and AKI stage 3, n = 1,248, 36.14%). With respect to the donor factors, there were differences in age, serum creatinine, BMI, and K-KDPI, as well as in the prevalence of diabetes mellitus, hypertension, use of continuous renal replacement therapy (CRRT), proportion of male donors, DCD, and CVA. Likewise, concerning the recipient factors, there were differences in age and history of KT. The number of HLA mismatches was higher in the AKI stage 3 group compared to stages 1 and 2.
Impact of AKI and AKI subgroups on graft outcomes
A total of 471 death-censored graft failure occurred during a median follow-up of 6.09 years. Furthermore, when AKI was subgrouped into stage 1 (n = 1,472), stage 2 (n = 733), and stage 3 (n = 1,248), death-censored graft failure occurred in 87, 44, and 69, in stages 1, 2, and 3 groups, respectively. The risk of death-censored graft failure remained higher in the AKI group compared to the no-AKI group in the multivariable analysis (Model 1, HR, 1.28; 95% CI, 1.06–1.54, P = 0.010). In the IPTW-adjusted multivariable model with incorporation of center effects, the AKI group showed a significantly higher risk of death-censored graft failure (Model 2, HR, 1.27; 95% CI, 1.05–1.53, P = 0.015) (Table 2; Fig. 2A). The variance of the random effect for center was 0.137, indicating minimal variation across transplant centers. In subgroup analysis, stage 3 groups had a higher risk for graft failure compared to the no-AKI group after adjustment for covariates, including center effects and IPTW (AKI stage 3, HR: 1.37, 95% CI: 1.04–1.81, P = 0.023) (Table 2; Fig. 2B).
Table 2.
Impact of AKI and AKI stage on death-censored graft failure.
| Group (event number) |
Event per 1000 patient-years (95% CI) | Unadjusted | Model 1 | Model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | HR | 95% CI | P value | V | ||
| No-AKI (n = 271) | 196.805 (174.068–221.685) | Ref | Ref | Ref | |||||||
| AKI (n = 200) | 191.938 (166.258–220.462) | 1.18 | 0.98–1.42 | 0.072 | 1.28 | 1.06–1.54 | 0.010 | 1.27 | 1.05–1.53 | 0.015 | 0.137 |
| No-AKI (n = 271) | 196.805 (174.068–221.685) | Ref | Ref | Ref | |||||||
| AKI 1 (n = 87) | 191.208 (153.150–235.855) | 1.12 | 0.88–1.42 | 0.361 | 1.18 | 0.93–1.51 | 0.174 | 1.20 | 0.94–1.54 | 0.140 | 0.136 |
| AKI 2 (n = 44) | 207.547 (150.804–278.622) | 1.20 | 0.87–1.65 | 0.259 | 1.28 | 0.93–1.77 | 0.134 | 1.25 | 0.90–1.73 | 0.190 | 0.136 |
| AKI 3 (n = 69) | 184.000 (143.163–232.863) | 1.26 | 0.97–1.64 | 0.086 | 1.42 | 1.08–1.87 | 0.010 | 1.37 | 1.04–1.81 | 0.023 | 0.136 |
Model 1 was adjusted for various donor-related (age, sex, BMI, diabetes mellitus, hypertension, DCD, death due to CVA cause, Hepatitis C virus), recipient-related (age, sex, diabetes mellitus, previous kidney transplantation, dialysis duration, ischemic heart diseases, CVA, cancer), transplantation-related (HLA mismatch number). Model 2 was adjusted for center effects in addition to model 1 and was further adjusted using IPTW for donor-related (sex, weight, hypertension, DCD, death due to CVA cause) covariates.
Numbers in parentheses indicate the number of events in each group.
AKI acute kidney injury, CI confidence interval, HR hazard ratio, n number, Ref reference, CVA cerebrovascular accidents, DCD donation after circulatory death, V Variance of random effect for center.
Fig. 2.
Death-censored graft failure rates according to AKI and AKI stages. A Comparison of death-censored graft failure rates between AKI group and no-AKI group (Log rank test, P = 0.072). B Comparison of death-censored graft failure rates among AKI stage 1–3 and no-AKI group (Log rank test, P = 0.086 for AKI S3 vs. no-AKI). AKI acute kidney injury, S1 stage 1, S2 stage 2, S3 stage 3.
As sensitivity analysis, Fine and gray competing risk analysis demonstrated that the AKI group showed a higher risk of death-censored graft failure (SHR, 1.20; 95% CI, 1.00–1.44.00.44, P = 0.047) after adjustment for covariates (Table S2, Fig. S2). Comparison between the analysis set and multiple imputation set revealed no significant differences except in donor age, donor and recipient BMI, dialysis duration, history of KT, and follow up duration (Table S3). As another sensitivity analysis, cause-specific Cox analysis after multiple imputation also showed that the risk of graft failure remained higher in the AKI group in the model 1 (HR, 1.21; 95% CI, 1.02–1.43, P = 0.026), and in the model 2 (HR, 1.15; 95% CI, 1.15–1.37, P = 0.035, Table S4).
Impact of AKI on graft outcomes according to K-KDPI groups
Donor kidneys were categorized into low and high K-KDPI groups using two different cutoffs: a threshold of 70% (low < 70%, high ≥ 70%) as the primary threshold and a threshold of 85% (low < 85%, high ≥ 85%) for sensitivity analysis. The incidence of death-censored graft failure was higher in the AKI group than in the no-AKI group, but only among recipients with high K-KDPI groups ≥ 70% (Table 3) and ≥ 85% (Table S5). In the multivariable model 1, AKI kidneys did not have a higher risk of death-censored graft failure compared to no-AKI kidneys in the low K-KDPI groups < 70% (Table 3; Fig. 3A) and < 85% (Table S5, Fig. S3A), whereas AKI kidneys had a higher risk compared to no-AKI kidneys in the high K-KDPI groups ≥ 70% (HR: 1.33, 95% CI: 1.04–1.71, P = 0.023, Table 3; Fig. 3B) and ≥ 85% (HR: 1.49, 95% CI: 1.11–1.98, P = 0.007, Table S5, Fig. S3B). In addition, when AKI subgroups according to AKI stage were compared with no-AKI, AKI stage 3 was associated with a higher risk of death-censored graft failure in the high K-KDPI group ≥ 70% (AKI stage 3, HR: 1.47, 95% CI: 1.04–2.08, P = 0.027, Table 3) and AKI stage 1 and 3 was associated with a higher risk of death-censored graft failure in the high K-KDPI group ≥ 85% (AKI stage 1, HR: 1.45, 95% CI: 1.01–2.08, P = 0.040; AKI stage 3, HR: 1.74, 95% CI: 1.17–2.57, P = 0.005, Table S5).
Table 3.
Impact of AKI and AKI stage on death-censored graft failure according to K-KDPI.
| K-KDPI (event number) |
Event per 1000 patient-years (95% CI) | Unadjusted | Model 1 | Model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | HR | 95% CI | P value | V | ||
| K-KDPI < 70 | |||||||||||
| No-AKI (n = 144) | 230.031 (193.995–270.820) | Ref | Ref | Ref | |||||||
| AKI (n = 59) | 193.442 (147.257–249.526) | 0.97 | 0.72–1.32 | 0.351 | 1.24 | 0.91–1.69 | 0.174 | 1.18 | 0.86–1.61 | 0.310 | 0.163 |
| AKI 1 (n = 27) | 200.000 (131.801–290.989) | 0.85 | 0.56–1.28 | 0.937 | 1.04 | 0.69–1.57 | 0.858 | 1.03 | 0.68–1.57 | 0.890 | 0.158 |
| AKI 2 (n = 14) | 259.259 (141.739–434.993) | 1.15 | 0.66–1.99 | 0.128 | 1.59 | 0.91–2.79 | 0.105 | 1.38 | 0.78–2.43 | 0.260 | 0.158 |
| AKI 3 (n = 18) | 155.172 (91.965–245.239) | 1.09 | 0.67–1.78 | 0.325 | 1.43 | 0.86–2.35 | 0.164 | 1.32 | 0.80–2.20 | 0.280 | 0.158 |
| K-KDPI ≥ 70 | |||||||||||
| No-AKI (n = 127) | 169.107 (140.977–201.206) | Ref | Ref | Ref | |||||||
| AKI (n = 141) | 191.316 (161.041–225.628) | 1.19 | 0.94–1.52 | 0.150 | 1.33 | 1.04–1.71 | 0.023 | 1.37 | 1.06–1.76 | 0.015 | 0.087 |
| AKI 1 (n = 60) | 187.500 (143.082–241.349) | 1.22 | 0.90–1.66 | 0.198 | 1.30 | 0.95–1.78 | 0.096 | 1.34 | 0.98–1.84 | 0.065 | 0.085 |
| AKI 2 (n = 30) | 189.873 (128.106–271.056) | 1.12 | 0.75–1.66 | 0.585 | 1.22 | 0.82–1.83 | 0.324 | 1.27 | 0.84–1.91 | 0.260 | 0.085 |
| AKI 3 (n = 51) | 196.911 (146.613–258.901) | 1.20 | 0.87–1.67 | 0.262 | 1.47 | 1.04–2.08 | 0.027 | 1.48 | 1.05–2.09 | 0.027 | 0.085 |
Model 1 was adjusted for various donor-related (age, sex, BMI, diabetes mellitus, hypertension, DCD, death due to CVA cause, Hepatitis C virus), recipient-related (age, sex, diabetes mellitus, previous kidney transplantation, dialysis duration, ischemic heart diseases, CVA, cancer), transplantation-related (HLA mismatch number). Model 2 was adjusted for center effects in addition to model 1 and was further adjusted using IPTW for donor-related (sex, weight, hypertension, DCD, death due to CVA cause) covariates. Numbers in parentheses indicate the number of events in each group.
AKI acute kidney injury, CI confidence interval, HR hazard ratio, n number, ref reference, CVA cerebrovascular accidents, DCD donation after circulatory death, V Variance of random effect for center.
Fig. 3.
Death-censored graft failure rates according to AKI according to K-KDPI subgroups in Model A. A Comparison of death-censored graft failure rates between AKI group and no-AKI group in the low K-KDPI group (< 70, Log rank test, P = 0.351). B Comparison of death-censored graft failure rates between AKI group and no-AKI group in the high K-KDPI group (≥ 70, Log rank test, P = 0.150). AKI acute kidney injury, K-KDPI Korean-kidney donor profile index.
As sensitivity analysis, Fine and gray competing risk analysis demonstrated that AKI kidneys were associated with a higher risk of death-censored graft failure in the high K-KDPI group ≥ 85%, whereas AKI kidneys did not have a higher risk compared to no-AKI kidneys in the low K-KDPI group < 85% (Table S2). As another sensitivity analysis, cause-specific Cox analysis after multiple imputation also showed that the risk of death-censored graft failure in the AKI group remained higher than that in the no-AKI group in the high K-KDPI group ≥ 85% (model 1 and 2, Table S4).
Impact of AKI on graft outcomes according to changing trends of serum creatinine
In the multivariable model 1, AKI did not have a higher risk for death-censored graft failure than no-AKI in the decreasing-creatinine subgroup (HR: 1.14, 95% CI: 0.82–1.59, P = 0.443, Table 4; Fig. 4A). When each AKI stage was compared with no-AKI, no-AKI stage was associated with a higher risk (AKI 1, HR: 1.30, 95% CI: 0.82–2.07, P = 0.260; AKI 2, HR: 1.22, 95% CI: 0.69–2.16 P = 0.489; AKI 3, HR: 0.94, 95% CI: 0.57–1.56, P = 0.826, Table 4). In the stable- or increasing-creatinine subgroup, AKI showed a higher risk for graft failure than no-AKI (HR: 1.38, 95% CI: 1.09–1.73, P = 0.006, Table 4; Fig. 4B) and stage 3 AKI showed a higher risk (HR: 1.77, 95% CI: 1.27–2.45, P < 0.001, Table 4). These trends were similar in the model 2 (Table 4).
Table 4.
Impact of AKI and AKI stage on death-censored graft failure according to changing trends of serum creatinine levels.
| Creatinine trend (event number) |
Event per 1000 patient-years (95% CI) | Unadjusted | Model 1 | Model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | HR | 95% CI | P value | V | ||
| Decreasing sCr | |||||||||||
| No-AKI (n = 116) | 193.333 (159.755–231.884) | Ref | Ref | Ref | |||||||
| AKI (n = 56) | 176.100 (133.024–228.681) | 1.02 | 0.74–1.40 | 0.907 | 1.14 | 0.82–1.59 | 0.443 | 1.09 | 0.79–1.52 | 0.590 | 0.132 |
| AKI 1 (n = 23) | 200.000 (126.782–300.098) | 1.12 | 0.71–1.75 | 0.622 | 1.30 | 0.82–2.07 | 0.260 | 1.24 | 0.79–1.97 | 0.350 | 0.126 |
| AKI 2 (n = 14) | 241.379 (131.964–404.993) | 1.14 | 0.66–1.99 | 0.638 | 1.22 | 0.69–2.16 | 0.489 | 1.15 | 0.65–2.05 | 0.630 | 0.126 |
| AKI 3 (n = 19) | 131.034 (78.891–204.626) | 0.86 | 0.53–1.39 | 0.536 | 0.94 | 0.57–1.56 | 0.826 | 0.92 | 0.56–1.52 | 0.750 | 0.126 |
| Stable or increasing sCr | |||||||||||
| No-AKI (n = 155) | 199.485 (169.316–233.478) | Ref | Ref | Ref | |||||||
| AKI (n = 144) | 198.895 (167.736–234.162) | 1.27 | 1.01–1.59 | 0.040 | 1.38 | 1.09–1.73 | 0.006 | 1.36 | 1.07–1.72 | 0.012 | 0.128 |
| AKI 1 (n = 64) | 188.235 (144.964–240.372) | 1.13 | 0.84–1.51 | 0.414 | 1.18 | 0.88–1.58 | 0.270 | 1.18 | 0.88–1.60 | 0.260 | 0.127 |
| AKI 2 (n = 30) | 194.805 (131.434–278.096) | 1.24 | 0.84–1.83 | 0.280 | 1.38 | 0.93–2.06 | 0.110 | 1.36 | 0.91–2.04 | 0.140 | 0.127 |
| AKI 3 (n = 50) | 217.391 (161.352–286.603) | 1.53 | 1.11–2.11 | 0.008 | 1.77 | 1.27–2.45 | < 0.001 | 1.70 | 1.22–2.38 | < 0.001 | 0.127 |
Model 1 was adjusted for various donor-related (age, sex, BMI, diabetes mellitus, hypertension, DCD, death due to CVA cause, Hepatitis C virus), recipient-related (age, sex, diabetes mellitus, previous kidney transplantation, dialysis duration, ischemic heart diseases, CVA, cancer), transplantation-related (HLA mismatch number). Model 2 was adjusted for center effects in addition to model 1 and was further adjusted using IPTW for donor-related (sex, weight, hypertension, DCD, death due to CVA cause) covariates.
Numbers in parentheses indicate the number of events in each group.
AKI acute kidney injury, CI confidence interval, HR hazard ratio, n number, ref reference, CVA cerebrovascular accidents, DCD donation after circulatory death, sCr serum creatinine, V Variance of random effect for center.
Fig. 4.
Death-censored graft failure rates according to changing trends in serum creatinine levels. A Comparison of death-censored graft failure rates between AKI group and no-AKI group in the decreasing-creatinine group (Log rank test, P = 0.907). B Comparison of death-censored graft failure rates between AKI group and no-AKI group in the stable- or increasing-creatinine group (Log rank test, P = 0.040). AKI acute kidney injury.
As sensitivity analysis, Fine and gray competing risk analysis also demonstrated that AKI showed a higher risk for death-censored graft failure than no-AKI in the stable- or increasing-creatinine subgroup (Table S2). In contrast, AKI was not associated with a higher risk for graft failure (Table S2). Cause-specific Cox analysis in the multiple imputation dataset also showed that the risk of death-censored graft failure in the AKI group remained higher than that in the no-AKI group only in the stable- or increasing-creatinine subgroup (Table S4).
Impact of AKI on secondary outcomes
Stage 3 AKI group showed a higher risk for death-uncensored graft failure (model 1, HR: 1.19, 95% CI: 1.02–1.38, P = 0.025, Table S6, Fig. S4) in parallel with results for death-censored graft failure, although total AKI group showed only a trend of higher risk than no-AKI group (model 1, HR: 1.10, 95% CI: 0.99–1.22, P = 0.079, Table S6). However, neither model 1 nor model 2 showed a significant difference in mortality between the AKI and no-AKI groups (Table S7, Fig. S5).
Discussion
We hypothesized that DDKT using kidneys with favorable prognostic indicators would have comparable outcomes to DDKT using donor kidneys without AKI. This Korean nationwide study demonstrated that donor AKI was associated with a higher risk of death-censored graft failure; however, AKI kidneys at stage 1 or 2 and those with K-KDPI < 70% or a decreasing creatinine trend, had comparable graft outcomes to donor kidneys without AKI.
Considering the aggravating severe organ shortage in Asian countries, maximal utilization of limited deceased donor kidneys, including kidneys with AKI, is an important issue26. A previous Korean study reported that the discard rate of donors’ kidneys was approximately 0.68%, which was much lower when compared to that of the USA (17.28%) and Eurotransplant group (12.00%)27–30. Generally, donor kidneys with AKI have been used in more than half of DDKT cases (50.55%), as shown in this study, because of severe organ shortage and long waiting time for DDKT (> 10 years) in Korea. However, we need to determine the outcomes of patients who underwent DDKT using donor kidneys with AKI and define their acceptance criteria for active and safe utilization.
A US single-center study (n = 994) reported that the impact of AKI in donors’ kidneys did not decrease graft survival, irrespective of AKI stage15,20. A US multicenter study (n = 585) with a median follow-up of 4.0 years also reported that the donor’s AKI did not increase the risk for graft failure irrespective of AKI stage, thereby suggesting the active utilization of AKI kidneys16,20. However, larger European studies reported that donor AKI could significantly impact the graft survival of DDKT. For instance, a French CRISTAL registry study with a median follow-up of 5.7 years, reported that donor kidneys with AKI based on the Kidney disease: improving global outcomes (KDIGO) criteria (ongoing AKI, n = 3,373; recovery from AKI, n = 2,392; undetermined AKI/Chronic kidney disease [CKD], n = 1,745) were associated with decreased graft survival18,31. Moreover, three types of AKI were all associated with decreased graft survival (ongoing AKI, HR: 1.23; recovery from AKI, HR: 1.10; undetermined AKI/CKD, HR: 1.28)18. Likewise, a UK transplant registry study evaluated the short-term impact of AKI according to AKIN criteria as ours on DDKT17. Graft failure in 1 year was higher in all AKI stages (n = 1,869, HR: 1.20) compared to no-AKI. Despite the small degree of inferior graft survival in the AKI group, the authors suggested that the 20% increase in 1-year graft failure in the AKI group may be better than a 37% increase in graft failure in DDKT after dialysis for longer than 1 year while waiting for a donor’s kidney without AKI17. They recommended the utilization of donors’ kidneys with AKI stages 1 and 2 but advised that those with AKI stage 3 be used with caution17. Therefore, we consider that these conflicting results about the outcomes of DDKT from AKI donors might stem from differences in either cohort size and accompanying statistical power or cohort characteristics, such as the proportion of ethnicity and practice pattern of organ discard. The proportion of black donors and discard rate were 1% and 28%, respectively, in the UK cohort, whereas they were 24% and 21%, respectively, in the US cohort16,17.
This Korean nationwide study, which included a relatively large number of patients with AKI (n = 3,453) and a median follow-up of 6.38 years, demonstrated the impact of AKI severity, as defined by the AKIN stage, on graft failure. AKI, especially stage 3 AKI was associated with a higher risk of graft failure compared to the no-AKI group in parallel with the European studies.
A previous Korean multicenter study (n = 386) used the old US ECD criteria and found that AKI in the ECD group was associated with a higher risk of graft failure, whereas AKI in the SCD group was not32–34. We also analyzed subgroup analysis according to the K-KDPI score. We found that in K-KDPI < 70% or 85%, the AKI group was not associated with a higher risk of graft failure compared to the no-AKI group. However, the AKI group was associated with a higher risk of graft failure in K-KDPI ≥ 70% subgroup, especially in K-KDPI ≥ 85% subgroup. When AKI stages were analyzed along with K-KDPI, there was no significant difference in the risk of graft failure according to AKI stages in the K-KDPI < 70% or 85% subgroup. Therefore, we could use AKI donors with stage 3 only in case of low K-KDPI subgroup.
The impact of AKI on graft outcomes according to changing patterns of serum creatinine concentration was also analyzed. AKI or any AKI stage was not associated with a higher risk of graft failure in the decreasing-trend group, which may be referred to as the recovering status. In contrast, AKI was associated with a higher risk of graft failure by 36% in the stable- or increasing-creatinine group. Neither stage 1 nor stage 2 was associated with higher risk, although stage 3 were associated with a higher risk of graft failure. These findings suggest that donor kidneys with AKI can be used when serial serum creatinine levels exhibit a decreasing trend, as such kidneys demonstrate a graft failure rate comparable to that of no-AKI kidneys and donors’ kidneys with stage 1 and stage 2 AKI can be used in stable- or increasing-creatinine trend as well.
The present study has several limitations. First, some detailed clinical information, including immunosuppressive agents was missing in this registry-based nationwide study; therefore, we cannot completely adjust potential confounding factors. Second, preexisting CKD could be a risk factor for AKI35,36. However, due to the lack of zero-time kidney biopsy information, history of CKD, and pre-admission serum creatinine levels, we could not determine preexisting CKD with or without AKI. Consequently, confounding AKI with undiagnosed CKD may have biased the prognostic estimates toward worse outcomes. Third, the baseline serum creatinine concentration is ideally defined as the average of all values measured within the year before admission37; however, owing to the absence of this information, we defined creatinine levels at admission as the baseline levels16,17,22. In the last, the prognostic impact of AKI etiology could not be analyzed due to insufficient information. Kidneys from donors with AKI due to interstitial nephritis may warrant more active utilization, given their higher potential for recovery compared with AKI associated with acute cortical necrosis or vasculitis38.
Nevertheless, the present study is of high significance. First, this study was a nationwide study with a large-sized, relatively long-term follow-up duration compared to previous studies. Second, we analyzed the impact of AKI on death-censored graft failure in various subgroups, such as AKI stage, K-KDPI, changing trends of serum creatinine levels. Based on this subgroup analysis, we could recommend accepting donor kidneys with all AKI stages in K-KDPI < 70 subgroup, all AKI stages in the decreasing-creatinine trend, and AKI stage 1 and 2 in stable- or increasing- creatinine trend (Fig. S6). Third, this study is the first nationwide Asian study that investigated different characteristics as against those of Western countries, such as a very long waiting time for DDKT and low discard rate of donor’s kidneys. Therefore, we believe our results would contribute to a balanced and better understanding of the impact of AKI on graft outcomes on an international basis.
In conclusion, donor AKI, especially stage 3 was associated with a higher risk for death-censored graft failure; however, subgroups of K-KDPI < 70% or a decreasing-creatinine trend, showed comparable risk of graft failure. Therefore, the active utilization of AKI kidneys with relatively favorable prognostic indicators may help address the organ shortage, particularly in Asian countries with a long waiting time for DDKT.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank the Korean Network for Organ Sharing (KONOS) and the National Health Insurance Data Sharing Service (NHISS) for sharing their database.
Author contributions
Research idea and study design: O.N., J.Y.; data acquisition: O.N., G.L., T.Y.K.; data analysis/interpretation: O.N., G.L., T.Y.K., B.K., J.Y.; statistical analysis: O.N., G.L., T.Y.K., J.H.L., H.B.K.; wrote the manuscript: O.N., J.Y. All authors read and approved of the final manuscript.
Funding
This study was supported by a grant from the National Institute of Organ, Tissue and Blood. Management (20232400C1B-00), which was not involved in the design or analysis of the study.
Data availability
Data supporting the findings of this study are available from the corresponding author uponreasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Consent for publication
All co-authors agreed to publication of the acquired data.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Omi Na and Gongmyung Lee.
References
- 1.Abecassis, M. et al. Kidney transplantation as primary therapy for end-stage renal disease: a National Kidney Foundation/Kidney Disease Outcomes Quality Initiative (NKF/KDOQI™) conference. Clin. J. Am. Soc. Nephrol.3, 471–480 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Azegami, T. et al. Efficacy of pre-emptive kidney transplantation for adults with end-stage kidney disease: a systematic review and meta-analysis. Ren. Fail.45, 2169618 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bastani, B. The present and future of transplant organ shortage: some potential remedies. J. Nephrol.33, 277–288 (2020). [DOI] [PubMed] [Google Scholar]
- 4.Transplantation IRiODa. Worldwide Actual Deceased Organ Donors Rate 2021. Web site. (accessed 21 April 2025). https://www.irodat.org/img/database/pdf/IRODaT_Newsletter_2021.pdf (2022).
- 5.USRDS. Annual Data Report, International Comparisons: Average yearly change in incidence of treated ESRD by country or region, 2010 versus 2020. Web site. (accessed 21 April 2022). https://usrds-adr.niddk.nih.gov/2022/end-stage-renal-disease/11-international-comparisons (2022).
- 6.Ng, Y-M. et al. Comparison of clinical outcomes among deceased donor kidney transplant recipients before and after utilizing estimated posttransplant survival score for kidneys allocation in Malaysia. Transpl. Proc.54, 278–281 (2022). [Google Scholar]
- 7.Manla, Y. et al. Temporal trends in brain-death organ donation in asia: results from the global observatory on donation and transplantation. Transpl. Proc.54, 233–236 (2022). [Google Scholar]
- 8.Wu, D. A. et al. Global trends and challenges in deceased donor kidney allocation. Kidney Int.91, 1287–1299 (2017). [DOI] [PubMed] [Google Scholar]
- 9.Glorioso, J. M. Kidney allocation policy: past, present, and future. Adv. Kidney Dis. Health. 28, 511–516 (2021). [Google Scholar]
- 10.Sethi, S. et al. Allocation of the highest quality kidneys and transplant outcomes under the new kidney allocation system. Am. J. Kidney Dis.73, 605–614 (2019). [DOI] [PubMed] [Google Scholar]
- 11.Rao, P. S. et al. A comprehensive risk quantification score for deceased donor kidneys: the kidney donor risk index. Transplantation88, 231–236 (2009). [DOI] [PubMed] [Google Scholar]
- 12.Kidney Donor Profile Index Calculator. Organ Procurement and Transplantation Network. Web site. https://optn.transplant.hrsa.gov/data/allocation-calculators/kdpi-calculator/
- 13.Merion, R. M. et al. Deceased-donor characteristics and the survival benefit of kidney transplantation. JAMA294, 2726–2733 (2005). [DOI] [PubMed] [Google Scholar]
- 14.Massie, A. B. et al. Survival benefit of primary deceased donor transplantation with high-KDPI kidneys. Am. J. Transpl.14, 2310–2316 (2014). [Google Scholar]
- 15.Heilman, R. L. et al. Long-term outcomes following kidney transplantation from donors with acute kidney injury. Transplantation103, e263–e272 (2019). [DOI] [PubMed] [Google Scholar]
- 16.Hall, I. E. et al. Deceased-donor acute kidney injury is not associated with kidney allograft failure. Kidney Int.95, 199–209 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Boffa, C. et al. Transplantation of kidneys from donors with acute kidney injury: friend or foe? Am. J. Transpl.17, 411–419 (2017). [Google Scholar]
- 18.Lenain, R. et al. Association between deceased donor acute kidney injury assessed using baseline serum creatinine back-estimation and graft survival: results from the French National CRISTAL registry. Am. J. Kidney Dis.79, 164–174 (2022). [DOI] [PubMed] [Google Scholar]
- 19.International Summit on Transplant T, Organ T. The declaration of Istanbul on organ trafficking and transplant tourism. Kidney Int.74, 854–859 (2008). [DOI] [PubMed] [Google Scholar]
- 20.Mehta, R. L. et al. Acute kidney injury network: report of an initiative to improve outcomes in acute kidney injury. Crit. Care. 11, R31 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Koo, T. Y. et al. A prognostic index for deceased donor kidneys and criteria for identifying suitable candidates for kidney transplantation from expanded criteria donors with prolonged waiting times. Kidney Dis. (Basel). 11, 143–153 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Liu, C. et al. Association of deceased donor acute kidney injury with recipient graft survival. JAMA Netw. Open.3, e1918634–e1918634 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lassnigg, A. et al. Minimal changes of serum creatinine predict prognosis in patients after cardiothoracic surgery: a prospective cohort study. J. Am. Soc. Nephrol.15, 1597–1605 (2004). [DOI] [PubMed] [Google Scholar]
- 24.Chertow, G. M., Burdick, E., Honour, M., Bonventre, J. V. & Bates, D. W. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J. Am. Soc. Nephrol.16, 3365–3370 (2005). [DOI] [PubMed] [Google Scholar]
- 25.Meeusen, J. W., Kasozi, R. N., Larson, T. S. & Lieske, J. C. Clinical impact of the refit CKD-EPI 2021 creatinine-based eGFR equation. Clin. Chem.68, 534–539 (2022). [DOI] [PubMed] [Google Scholar]
- 26.Chan-On, C. & Sarwal, M. M. A comprehensive analysis of the current status and unmet needs in kidney transplantation in Southeast Asia. Front. Med. (Lausanne). 4, 84 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hart, A. et al. OPTN/SRTR 2014 annual data report: kidney. Am. J. Transpl.16 (Suppl 2), 11 (2016). [Google Scholar]
- 28.Kidney: donation, waiting list and transplants. Table 5.1 (ii) Deceased donors / kidneys in Eurotransplant in 2015. Eurotransplant Annual Report. (2015). https://www.eurotransplant.org/wp-content/uploads/2019/12/AR2015.pdf
- 29.Kim, K. H. et al. Current status of discarded grafts in Korean organ transplantation. Transpl. Proc.51, 1478–1480 (2019). [Google Scholar]
- 30.Mohan, S. et al. Factors leading to the discard of deceased donor kidneys in the united States. Kidney Int.94, 187–198 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Khwaja, A. KDIGO clinical practice guidelines for acute kidney injury. Nephron Clin. Pract.120, c179–184 (2012). [DOI] [PubMed] [Google Scholar]
- 32.Park, W. Y. et al. Impact of acute kidney injury in expanded criteria deceased donors on post-transplant clinical outcomes: multicenter cohort study. BMC Nephrol.20, 1–11 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Metzger, R. A. et al. Expanded criteria donors for kidney transplantation. Am. J. Transpl.3, 114–125 (2003). [Google Scholar]
- 34.Port, F. K. et al. Donor characteristics associated with reduced graft survival: an approach to expanding the pool of kidney donors. Transplantation74, 1281–1286 (2002). [DOI] [PubMed] [Google Scholar]
- 35.Grams, M. E. et al. A meta-analysis of the association of estimated GFR, albuminuria, age, race, and sex with acute kidney injury. Am. J. Kidney Dis.66, 591–601 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Singh, R. et al. The impact of trimethoprim-sulfamethoxazole as Pneumocystis Jiroveci pneumonia prophylaxis on the occurrence of asymptomatic bacteriuria and urinary tract infections among renal allograft recipients: a retrospective before-after study. BMC Infect. Dis.16, 90 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Siew, E. D. et al. Estimating baseline kidney function in hospitalized patients with impaired kidney function. Clin. J. Am. Soc. Nephrol.7, 712 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bonomini, V., Stefoni, S. & Vangelista, A. Long-term patient and renal prognosis in acute renal failure. Nephron36, 169–172 (1984). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data supporting the findings of this study are available from the corresponding author uponreasonable request.