Advertisement
American Journal of Kidney Diseases

Deceased-Donor Acute Kidney Injury and Acute Rejection in Kidney Transplant Recipients: A Multicenter Cohort

Open AccessPublished:September 30, 2022DOI:https://doi.org/10.1053/j.ajkd.2022.08.011

      ABSTRACT

      Rationale & Objective

      Donor acute kidney injury (AKI) activates innate immunity, enhances HLA expression in the kidney allograft, and provokes recipient alloimmune responses. We hypothesized that injury and inflammation manifested in deceased-donor urine biomarkers would be associated with higher rates of biopsy-proven acute rejection (BPAR) and allograft failure after transplantation.

      Study Design

      Prospective cohort.

      Setting & Participants

      862 deceased donors for 1137 kidney recipients at 13 centers.

      Exposures

      We measured concentrations of IL-18, kidney injury molecule-1 (KIM-1), and neutrophil gelatinase-associated lipocalin (NGAL) in deceased donor urine. We also used the Acute Kidney Injury Network criteria to assess donor clinical AKI.

      Outcomes

      The primary outcome was a composite of BPAR and graft failure (not from death). A secondary outcome was the composite of BPAR, graft failure, and/or de novo donor specific antibody (DSA). Outcomes were ascertained in the first post-transplant year.

      Analytical Approach

      Multivariable Fine-Gray models with death as a competing risk.

      Results

      Mean recipient age was 54±13 years and 82% received anti-thymocyte globulin. We found no significant associations between donor urinary IL-18 (subdistribution hazard ratio [sHR] for highest vs. lowest tertile 0.76; 95% CI 0.45, 1.28), KIM-1 (sHR 1.2; 95% CI 0.69, 2.07) or NGAL (sHR 1.14; 95% CI 0.71, 1.84) and the primary outcome. In secondary analyses, we detected no significant associations between a) clinically-defined AKI and the primary outcome, or between b) donor biomarkers and the composite outcome of BPAR, graft failure and/or de novo DSA.

      Limitations

      BPAR ascertained through for-cause biopsies, not surveillance biopsies.

      Conclusions

      In a large cohort of kidney recipients that were almost all induced with thymoglobulin, donor injury biomarkers were neither associated with graft failure and rejection, nor with a secondary outcome that included de novo DSA. These findings provide some reassurance that centers can successfully manage immunological complications using deceased-donor kidneys with AKI.

      Graphical abstract

      Keywords

      Introduction

      Deceased organ donors often experience acute kidney injury (AKI) either due to the circumstances of death, such as trauma or anoxia, or due to complications of subsequent treatment. Unfortunately, approximately one-third of kidneys from deceased donors with AKI are discarded, a higher rate than donors without AKI.
      • Hall I.E.
      • Schroppel B.
      • Doshi M.D.
      • et al.
      Associations of deceased donor kidney injury with kidney discard and function after transplantation.
      ,
      • Boffa C.
      • van de Leemkolk F.
      • Curnow E.
      • et al.
      Transplantation of Kidneys From Donors With Acute Kidney Injury: Friend or Foe?.
      The risk of immunological complications associated with transplanting AKI kidneys is unknown. AKI causes tissue inflammation through multiple mechanistic pathways, such as complement activation (e.g., the MB-lectin pathway) and enhanced expression of toll-like receptors (TLR) including TLR-2 and TLR-4, which are present in renal tubular epithelial cells.
      • Jang H.R.
      • Rabb H.
      The innate immune response in ischemic acute kidney injury.
      M2 macrophages and regulatory T-cells play a direct role in guiding the response to inflammation and repair following AKI.
      • Singbartl K.
      • Formeck C.L.
      • Kellum J.A.
      Kidney-Immune System Crosstalk in AKI.
      Given these inflammatory pathways, we hypothesized the recipients of AKI kidneys would experience elevated rates of acute rejection, both cellular and antibody, and formation of de novo donor specific antibody (DSA).
      The “Injury Hypothesis” was proposed over 20 years ago (with subsequent modifications) and states that oxidative stress and injury to the kidney at procurement and transplantation variably activate innate immunity in the allograft, which affects alloimmune responses and elevates the risk of rejection.
      • Land W.G.
      The role of postischemic reperfusion injury and other nonantigen-dependent inflammatory pathways in transplantation.
      Classically, this injury pathway would be expected to provoke acute cellular rejection (ACR) via alloreactive T-cells. However, B-cell responses may be affected concurrently and promote acute antibody-mediated rejection (AMR) via DSA. AMR may carry a worse prognosis than ACR, involves treatments that have not been extensively tested in clinical trials, and may cause chronic immunological injury and fibrosis.
      • Colvin R.B.
      • Smith R.N.
      Antibody-mediated organ-allograft rejection.
      To assess outcomes using AKI kidneys, we assembled a multi-center, prospective cohort that included testing deceased-donor urine for injury biomarkers and detailed chart review of recipient outcomes (the Deceased Donor Study) including biopsies. We identified AKI using both conventional serological definitions that rely on changes in serum creatinine as well as characterizing injury using sensitive urinary biomarkers including IL-18, KIM-1, and NGAL. We showed that AKI defined using serum creatinine (corresponding to ≥Stage 2 Acute Kidney Injury Network [AKIN] criteria) was present in approximately 9% of deceased-donor kidneys and that many additional donors had elevated concentrations of injury biomarkers. Our group and others have demonstrated that recipients of AKI kidneys commonly experience delayed graft function (DGF);
      • Irish W.D.
      • Ilsley J.N.
      • Schnitzler M.A.
      • Feng S.
      • Brennan D.C.
      A risk prediction model for delayed graft function in the current era of deceased donor renal transplantation.
      Donor urinary biomarkers that are generated in the setting of AKI, such as NGAL, are associated with DGF in the recipient. Nonetheless, longer-term graft survival and graft function for kidneys with AKI are comparable to kidneys without AKI.
      • Liu C.
      • Hall I.E.
      • Mansour S.
      • Thiessen Philbrook H.R.
      • Jia Y.
      • Parikh C.R.
      Association of Deceased Donor Acute Kidney Injury With Recipient Graft Survival.
      • Sonnenberg E.M.
      • Hsu J.Y.
      • Cohen J.B.
      • et al.
      Acute Kidney Injury in Deceased Organ Donors and Kidney Transplant Outcomes: A National Cohort Study using a Novel Data Source.
      • Reese P.P.
      • Hall I.E.
      • Weng F.L.
      • et al.
      Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes.
      • Heilman R.L.
      • Smith M.L.
      • Smith B.H.
      • et al.
      Long-term Outcomes Following Kidney Transplantation From Donors With Acute Kidney Injury.
      Some uncertainty remains about whether kidneys with severe, AKIN Stage 3 injury in the donor also have good long term outcomes after transplantation.
      • Boffa C.
      • van de Leemkolk F.
      • Curnow E.
      • et al.
      Transplantation of Kidneys From Donors With Acute Kidney Injury: Friend or Foe?.
      We leveraged the detailed immunological data in the Deceased Donor Study to examine whether donor kidney injury and inflammation, manifested through urinary biomarkers, were associated with allograft failure and rejection. For the subset of centers with clinical protocols for routine post-transplant assessment of DSA, we developed a study protocol to harmonize adjudication of a composite outcome that included de novo DSA within 1 year after transplant.

      Methods

      The Deceased Donor Study (ClinicalTrials.gov Identifier: NCT01848249) is an observational cohort study of deceased donors with subsequent prospective data extraction from the medical records of the recipients of kidney transplants from those donors.
      • Hall I.E.
      • Schroppel B.
      • Doshi M.D.
      • et al.
      Associations of deceased donor kidney injury with kidney discard and function after transplantation.
      ,
      • Potluri V.S.
      • Parikh C.R.
      • Hall I.E.
      • et al.
      Validating Early Post-Transplant Outcomes Reported for Recipients of Deceased Donor Kidney Transplants.
      • Doshi M.D.
      • Reese P.P.
      • Hall I.E.
      • et al.
      Utility of Applying Quality Assessment Tools for Kidneys With KDPI >/=80.
      • Hall I.E.
      • Reese P.P.
      • Doshi M.D.
      • et al.
      Delayed Graft Function Phenotypes and 12-Month Kidney Transplant Outcomes.
      • Harhay M.N.
      • Jia Y.
      • Thiessen-Philbrook H.
      • et al.
      The association of discharge decisions after deceased donor kidney transplantation with the risk of early readmission: Results from the deceased donor study.
      • Hall I.E.
      • Parikh C.R.
      • Schroppel B.
      • et al.
      Procurement Biopsy Findings Versus Kidney Donor Risk Index for Predicting Renal Allograft Survival.
      • Hall I.E.
      • Akalin E.
      • Bromberg J.S.
      • et al.
      Deceased-donor acute kidney injury is not associated with kidney allograft failure.
      Briefly, five participating organ procurement organizations (OPOs) enrolled donors between May 2010 and December 2013. Each OPO utilized set protocols for research authorization and donor management. Clinical variables were abstracted from OPO donor charts, and extensive chart reviews were performed for the subset of recipients ≥16 years at 13 participating transplant centers who received kidneys from enrolled donors. Trained site coordinators reviewed prospectively collected medical records and recorded detailed recipient data. Key outcomes including DGF (any dialysis in the week after transplant) and allograft biopsy results were reviewed by site principal investigators. The data coordinating center validated chart abstractions to confirm data accuracy (Item S1). The study also used some data from the Organ Procurement and Transplantation Network (OPTN). The OPTN data system includes data on all donor, wait-listed candidates, and transplant recipients in the US, submitted by the members of the Organ Procurement and Transplantation Network (OPTN), and has been described elsewhere. The Health Resources and Services Administration (HRSA), U.S. Department of Health and Human Services provides oversight to the activities of the OPTN contractor.
      The OPO scientific review board approved the study and authorization for research was obtained from the surrogates of the deceased donors. The institutional review boards for participating investigators approved the study and waived informed consent for transplant recipients. The clinical and research activities being reported are consistent with the Principles of the Declaration of Istanbul as outlined in the Declaration of Istanbul on Organ Trafficking and Transplant Tourism.
      The Declaration of Istanbul on Organ Trafficking and Transplant Tourism. Istanbul Summit April 30-May 2, 2008.
      All clinical investigators abided by the Ethical Principles for Medical Research Involving Human Subjects as outlined in the Declaration of Helsinki.

      Outcomes

      The primary outcome was a composite of graft failure (not due to death) or biopsy-proven acute rejection in the first post-transplant year. Kidney biopsies, pathology interpretation and treatment for rejection were performed per each center’s local protocol. In a secondary analysis, we examined the composite of graft failure, rejection or de novo DSA within the first year among a subcohort of five of the participating centers that screened recipients for de novo DSA; we did not include centers that only measured DSA due to clinical concern for rejection.
      For the binary outcome of de novo DSA, each center applied the criterion of mean fluorescence intensity (MFI) ≥1000. For patients with pre-transplant DSA, we also categorized as a positive outcome a patient with MFI ≥1000 and ≥50% higher than the pre-transplant DSA MFI. MFI was defined as antibody reactivity to the specific donor human leukocyte antigen (HLA) allele, where allele level typing was available. Where necessary, the center would choose the highest MFI associated with an HLA epitope – even if the epitope was shared among several HLA alleles, one of which corresponded to the donor HLA allele. Importantly, each transplant center followed their own clinical protocol for kidney biopsies and DSA screening (Item S2).
      Notably, we restricted outcomes to one year, because rejection and DSA development beyond 12 months would be less likely to be associated with donor injury and more likely caused by recipient clinical issues such as nonadherence.

      Donor urine injury biomarker data

      The primary exposure was donor urinary concentrations of IL-18 (pg/mL), KIM-1 (pg/mL), and NGAL (ng/dL). Prior to organ procurements, 10 mL of fresh donor urine was collected using an indwelling urinary catheter tube, transferred on ice, and then frozen. The urine was stored at -80 degrees Celsius until monthly shipments to the central study biorepository. Biomarker measurements are described thoroughly in the supplement (Item S3) and in previous work.

      Chow SCS, J.; Wang, H. Sample Size Calculations in Clinical Research. 2nd ed2008.

      Statistical analysis

      The supplement (Item S4) includes details about the calculation of variables. We calculated descriptive statistics as means ± standard deviation, medians (interquartile range), or frequencies (percentages). Donor, transplant, and recipient characteristics were compared by the primary outcome using Kruskall-Wallis or Chi-Square tests. Because these comparisons were by recipient outcome, we assessed donor characteristics at the level of the kidney for these analyses. We then fit a multivariable Fine-Gray regression model to determine the subdistribution hazards ratio (sHR) for donor injury biomarkers and the primary outcome, with death as a competing risk. Donor injury biomarkers were both modeled continuously after a log base 2 transformation and as tertiles.
      We used Kolmogorov-type supremum tests to evaluate proportional hazards assumptions. We accounted for the cluster effect of kidneys from the same donor going to two recipients using robust sandwich estimates. In the primary analysis, we adjusted for variables available at organ offer and collected by OPOs: the KDRI; the transplant variables cold ischemia time (hours) and number of human leukocyte antigen (HLA) mismatches; and the following recipient variables: age (years), sex, Black race, previous kidney transplant, cause of end-stage kidney disease, percent panel reactive antibody (PRA), body mass index [BMI], and preemptive transplantation
      • Reese P.P.
      • Hall I.E.
      • Weng F.L.
      • et al.
      Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes.
      ,
      • Parikh C.R.
      • Hall I.E.
      • Bhangoo R.S.
      • et al.
      Associations of Perfusate Biomarkers and Pump Parameters With Delayed Graft Function and Deceased Donor Kidney Allograft Function.
      . In the analysis of the secondary outcome (that included de novo DSA), we also adjusted for pre-transplant DSA (a binary exposure). Final models also adjusted for donor urinary creatinine concentrations.

      Exploratory analyses

      We assessed whether the following variables modified the relationship between donor injury biomarkers and the primary outcome: KDPI (cut-off: 85%), donation after circulatory determination of death (DCD), kidney machine perfusion, cold ischemia time (median value cut-off: 14 hours), DGF, donor-recipient sex combinations,
      • Hall I.E.
      • Akalin E.
      • Bromberg J.S.
      • et al.
      Deceased-donor acute kidney injury is not associated with kidney allograft failure.
      ,
      • Lepeytre F.
      • Dahhou M.
      • Zhang X.
      • et al.
      Association of Sex with Risk of Kidney Graft Failure Differs by Age.
      or donor-recipient race combinations.
      • Locke J.E.
      • Warren D.S.
      • Dominici F.
      • et al.
      Donor ethnicity influences outcomes following deceased-donor kidney transplantation in black recipients.
      ,
      • Pisavadia B.
      • Arshad A.
      • Chappelow I.
      • et al.
      Ethnicity matching and outcomes after kidney transplantation in the United Kingdom.
      Each of these Fine-Gray models used the same covariates as the primary analysis with tests for interaction between the donor biomarker and the potential modifier.
      We also fit Fine-Gray models and examined the association between donor biomarkers and the outcome of BPAR only. We then fit a Cox regression model to examine the association between donor biomarkers and the composite outcome of BPAR, graft failure, or death. Covariates were the same as for the donor biomarker models for the primary outcome. We fit Fine-Gray models to examine the association of donor clinical AKI defined as AKIN Stage 2 or greater and the primary outcome. Covariates were the same as for the donor biomarker models for the primary outcome, except that we did not adjust for urinary creatinine concentration. Finally, we examined the associations of donor clinical AKI and the outcome of BPAR only, and donor clinical AKI and the composite outcome of BPAR, graft failure, or death.

      Power

      We evaluated the statistical power by examining the association of biomarkers (highest vs. lowest tertile) within each outcome. We assumed an alpha of 5%, power of 80%, that the hazard ratio is constant throughout the study and that Cox proportional hazards regression models were used.

      Chow SCS, J.; Wang, H. Sample Size Calculations in Clinical Research. 2nd ed2008.

      ,
      • Schoenfeld D.'
      Sample Size Formula for the Proportional-Hazards Regression Model.
      For the primary outcome, we estimated the power to detect a hazard ratio of at least 0.56 (within the cohort of 1137 recipients). For the secondary outcome of BPAR, graft failure, or DSA, a hazard ratio of at least 0.47 (within the cohort of 422 recipients).
      We used SAS 9.4 for Windows (SAS Institute, Cary, NC). All statistical tests and confidence intervals were two-sided with a significance level of 0.05.

      Results

      A shown in Figure 1, the primary cohort comprised 1137 deceased-donor kidney transplant recipients at 13 centers.
      Table 1a shows that the mean recipient age was 53.7 years (+/-13.3) and 61% were male. Fourteen percent had previously received kidney transplants and 15% had estimated PRA >80%. Eighty-two percent of recipients received rabbit anti-thymocyte globulin induction therapy, 15% received basiliximab and 3% received alemtuzumab. Table S1 provides additional detail about immunosuppression. Compared to recipients who did not experience the primary composite outcome of rejection or allograft failure, recipients who did experience the outcome were more likely to be Black race (57% vs. 45%, p=0.003), have prior transplants (19% vs. 13%, p=0.04), and calculated panel reactive antibody titers (cPRA) above 80% (21% vs. 14%, p=0.05 [p-value for the association with all 4 levels of PRA]), but they were less likely to be discharged from the transplant hospitalization on tacrolimus (89% vs. 97%, p<0.001) and mycophenolate (93% vs. 97%, p=0.023). A total of 37% of recipients experienced DGF. Recipients who experienced the primary outcome were also more likely to have delayed graft function (54% vs. 35% for those without the primary outcome, p<0.001).
      Table 1aRecipient characteristics, by primary outcome status
      Recipient CharacteristicsAll (N=1137)Non-Event (Including death)*

      (N=978)
      Composite Event** (N=159)P
      Mean age, years53.7 (13.3)53.7 (13.3)53.7 (13.4)1.0
      Male693 (61%)597 (61%)96 (60%)0.9
      Black race528 (46%)437 (45%)91 (57%)0.003
      Hispanic ethnicity119 (10%)103 (11%)16 (10%)1.0
      Body Mass Index, Kg/m228.3 (5.7)28.2 (5.7)29.0 (5.6)0.09
      Cause of ESKDDiabetes358 (32%)315 (32%)43 (27%)0.1
      Hypertension319 (28%)271 (28%)48 (30%)
      Glomerulonephritis183 (16%)156 (16%)27 (17%)
      Graft failure91 (8%)71 (7%)20 (13%)
      Other185 (16%)164 (17%)21 (13%)
      Preemptive transplant117 (10%)105 (11%)12 (8%)0.2
      Previous kidney transplant161 (14%)130 (13%)31 (19%)0.04
      Panel Reactive Antibody (PRA)0%729 (64%)640 (66%)89 (56%)0.05
      1-20%86 (8%)70 (7%)16 (10%)
      21-80%147 (13%)127 (13%)20 (13%)
      >80%174 (15%)140 (14%)34 (21%)
      HLA mismatch level4.36 (1.33)4.32 (1.34)4.57 (1.23)0.02
      Induction ImmunosuppressionAnti-thymocyte globulin937 (82%)803 (82%)134 (85%)0.5
      Basiliximab167 (15%)148 (15%)19 (12%)0.4
      Alemtuzumab35 (3%)30 (3%)5 (3%)0.7
      Rituximab16 (1%)12 (1%)4 (3%)0.3
      Maintenance Immunosuppression At DischargePrednisone945 (84%)816 (85%)129 (82%)0.7
      Tacrolimus1087 (96%)947 (97%)140 (89%)<.001
      Cyclosporine17 (1%)11 (1%)6 (4%)0.04
      Mycophenolate1098 (97%)951 (97%)147 (93%)0.02
      Delayed Graft Function426 (37%)340 (35%)86 (54%)<.001
      * Non-event means that recipients did not experience the composite events of acute rejection or graft failure, but may have died. 36 deaths were included in the non-event group.
      ** Composite event includes acute rejection or graft failure not from death within 1 year.
      Results are presented as mean (SD) or n (%).
      ESKD, end-stage kidney disease; HLA, human leukocyte antigen.
      Induction immunosuppression was missing in <3% of recipients, and discharge immunosuppression was missing in <1% of recipients.
      Table 1bAllograft and donor characteristics, by primary outcome status
      Donor CharacteristicsAll (N=1137)*Non-event (including death)**

      (N=978)
      Composite event*** (N=159)P
      Mean age (Years)41.5 (15.3)41.0 (15.4)44.7 (14.1)0.005
      Male700 (62%)609 (62%)91 (57%)0.2
      Black race183 (16%)147 (15%)36 (23%)0.02
      Hispanic ethnicity167 (15%)148 (15%)19 (12%)0.3
      Body Mass Index, Kg/m228.3 (7.4)28.2 (7.1)29.0 (9.5)0.9
      Hypertension353 (31%)293 (30%)60 (38%)0.05
      Diabetes118 (10%)94 (10%)24 (15%)0.04
      Cause of DeathHead Trauma304 (27%)258 (27%)46 (29%)0.4
      Anoxia410 (37%)362 (38%)48 (31%)
      Stroke389 (35%)330 (34%)59 (38%)
      Other16 (1%)13 (1%)3 (2%)
      Hepatitis C seropositive30 (3%)27 (3%)3 (2%)0.5
      DCD217 (19%)192 (20%)25 (16%)0.2
      KDRI1.31 (0.43)1.3 (0.43)1.41 (0.42)<.001
      KDPI, %49.6 (27.3)48.5 (27.3)57.0 (26.0)<.001
      KDPI > 85%126 (11%)105 (9%)21 (2%)0.2
      ECD236 (21%)197 (20%)39 (25%)0.2
      Admission Creatinine (mg/dL)1.11 (0.63)1.12 (0.6)1.07 (0.75)0.02
      Terminal Serum Creatinine (mg/dL)1.21 (0.93)1.2 (0.92)1.24 (0.98)0.6
      Donor Cerebrovascular/stroke as cause of deathNo735 (65%)638 (65%)97 (61%)0.3
      Yes401 (35%)339 (35%)62 (39%)
      Donor AKI StageNo AKI827 (73%)717 (74%)110 (70%)0.2
      Stage 1184 (16%)156 (16%)28 (18%)
      Stage 269 (6%)61 (6%)8 (5%)
      Stage 351 (5%)39 (4%)12 (8%)
      Number of individual kidneys transplanted186 (8%)70 (7%)16 (10%)0.2
      21050 (92%)907 (93%)143 (90%)
      Kidney biopsied591 (52%)500 (51%)91 (57%)0.2
      Kidney pumped541 (48%)465 (48%)76 (48%)1.0
      Cold ischemia time (hours)16.34 (6.98)16.34 (7)16.31 (6.89)1.0
      * 1137 kidneys were procured from 862 total donors.
      ** Non-event means that recipients did not experience the composite events of acute rejection or graft failure, but may have died. 36 deaths were included in the non-event group.
      *** Composite event includes acute rejection or graft failure not from death within 1 year.
      Results are presented as means (SD) or n (%).
      ECD, expanded-criteria donor; DCD, donation after cardiovascular determination of death, KDRI Kidney donor risk index, KDPI kidney donor profile index
      BMI and KDRI was missing in 5 donors. Admission Serum Creatinine was missing in 6 donors.
      Table 1b shows that for the deceased kidney donors, mean (SD) terminal serum creatinine was 1.21 (0.93) mg/dL and 19% were DCD. When categorized by AKIN stages, 73% of the kidneys came from donors with no AKI, 16% from donors with Stage 1 AKI, 6% from donors with Stage 2 AKI, and 5% from donors with Stage 3 AKI.
      Figure 1 shows that 159 recipients (14%) experienced the primary composite of graft failure or BPAR during the first year. One hundred seven met the primary outcome due to BPAR. Table S2 shows time-to-event. A total of 77 of BPAR episodes were ACR only, 8 were AMR only, 12 were both ACR and AMR, and 10 could not be definitively classified (Table S3).
      Figure 2 shows distributions of donor urine IL-18, KIM-1, and NGAL concentrations. We found no significant association between urinary injury biomarkers and the primary outcome in multivariable analyses. In the fully-adjusted models with highest tertile vs. lowest tertile biomarker concentrations, the sHRs were 0.76 (95% CI 0.45, 1.28) for IL-18, 1.2 (95% CI 0.69, 2.07) for KIM-1, and 1.14 (95% CI 0.71, 1.84) for NGAL (Table 2).
      Figure thumbnail gr2
      Figure 2Donor urinary biomarker distributions, by recipients who did and did not experience the primary composite outcome of graft failure or acute rejection
      Table 2Multivariable analysis of donor urinary biomarkers and primary composite outcome, using Fine-Gray competing risks models
      BiomarkerBiomarker Range | nSubdistribution Hazard Ratio (95% CI)
      Unadjusted ModelAdjusted for KDRIModel Adjusted for KDRI and Clinical Covariates*Model Adjusted for KDRI, Urine Creatinine and Clinical Covariates*
      Lower TertileLog2-Trans(1.367, 10.501) | n=11051.00 (0.91, 1.1)1.00 (0.91, 1.11)1.00 (0.9, 1.11)0.98 (0.88, 1.09)
      Lower Tertile(2.58, 28.24) | n=3681 (ref)1 (ref)1 (ref)1 (ref)
      Mid Tertile(28.32, 78.88) | n=3691.12 (0.73, 1.72)1.19 (0.77, 1.83)1.19 (0.77, 1.84)1.1 (0.7, 1.73)
      Upper Tertile(78.9, 1448.69) | n=3680.85 (0.54, 1.34)0.88 (0.55, 1.4)0.86 (0.52, 1.41)0.76 (0.45, 1.28)
      KIM-1Log2-Trans(5.882, 15.205) | n=11051.04 (0.94, 1.16)1.05 (0.95, 1.17)1.06 (0.95, 1.18)1.04 (0.91, 1.18)
      Lower Tertile(58.96, 890.91) | n=3691 (ref)1 (ref)1 (ref)1 (ref)
      Mid Tertile(894.16, 2503.7) | n=3681.23 (0.79, 1.93)1.33 (0.84, 2.1)1.37 (0.86, 2.2)1.33 (0.83, 2.15)
      Upper Tertile(2521.91, 37759.01) | n=3681.22 (0.78, 1.9)1.29 (0.82, 2.03)1.3 (0.8, 2.11)1.2 (0.69, 2.07)
      NGALLog2-Trans(-3.322, 13.102) | n=10941.03 (0.96, 1.09)1.03 (0.97, 1.1)1.04 (0.97, 1.11)1.03 (0.96, 1.11)
      Lower Tertile(0, 20.6) | n=3681 (ref)1 (ref)1 (ref)1 (ref)
      Mid Tertile(20.7, 104.6) | n=3691.04 (0.66, 1.62)1.05 (0.67, 1.66)1.04 (0.65, 1.66)1 (0.62, 1.61)
      Upper Tertile(105, 8792.38) | N=3681.12 (0.72, 1.73)1.13 (0.72, 1.77)1.19 (0.75, 1.89)1.14 (0.71, 1.84)
      * Clinical Covariates include cold ischemia time and the following recipient variables: age (years), black race, sex, previous kidney transplant, cause of end-stage kidney disease (4 categories, other as reference), number of human leukocyte antigen mismatches, panel reactive antibody (%), body mass index (kg/m2), and pre-emptive transplant

      Secondary and exploratory analyses

      A total of 422 recipients at 5 centers comprised the subcohort with DSA screening (Figure S1). Fifty-four (13%) had pre-transplant DSA. By one year, 85 recipients (20%) experienced the composite outcome of graft failure, acute rejection and/or de novo DSA. Thirty-eight experienced the rejection outcome, 35 the de novo DSA outcome, and 12 the graft failure outcome. Table S4 shows details about de novo DSA. Twelve recipients died by one year. We found no significant association between urinary injury biomarkers and the secondary composite outcome. In the fully-adjusted models with highest tertile vs. lowest tertile biomarker concentrations, the sHRs were 0.81 (95% CI 0.42, 1.56) for IL-18, 0.9 (95% CI 0.43, 1.87) for KIM-1, and 0.66 (95% CI 0.34, 1.29) for NGAL (Table 3).
      Table 3Multivariable analysis of donor urinary biomarkers and the secondary outcome of graft failure, acute rejection, and/or de novo donor specific antibody, using Fine-Gray competing risks models
      BiomarkerBiomarker Range | nSubdistribution Hazard Ratio (95% CI)
      Unadjusted ModelAdjusted for KDRIModel Adjusted for KDRI and Clinical Covariates*Model Adjusted for KDRI, Urine Creatinine and Clinical Covariates*
      IL-18Log2-Trans(1.367, 10.501) | n=4090.89 (0.79, 1)0.89 (0.79, 1)0.91 (0.8, 1.04)0.9 (0.78, 1.05)
      Lower Tertile(2.58, 28.05) | n=1431 (ref)1 (ref)1 (ref)1 (ref)
      Mid Tertile(28.32, 78.88) | n=1220.94 (0.57, 1.58)0.94 (0.56, 1.57)0.98 (0.56, 1.71)0.99 (0.5, 1.94)
      Upper Tertile(78.9, 1448.69) | n=1440.76 (0.45, 1.28)0.77 (0.45, 1.3)0.88 (0.48, 1.63)0.81 (0.42, 1.56)
      KIM-1Log2-Trans(5.882, 15.205) | n=4090.88 (0.78, 0.99)0.89 (0.78, 1.01)0.91 (0.78, 1.05)0.92 (0.77, 1.1)
      Lower Tertile(58.96, 874.35) | n=1371 (ref)1 (ref)1 (ref)1 (ref)
      Mid Tertile(899.99, 2503.7) | n=1340.73 (0.44, 1.22)0.83 (0.49, 1.38)0.73 (0.41, 1.3)0.88 (0.46, 1.69)
      Upper Tertile(2521.91, 37759.01) | n=1380.7 (0.42, 1.17)0.74 (0.44, 1.24)0.74 (0.41, 1.36)0.9 (0.43, 1.87)
      NGALLog2-Trans(-3.322, 13.102) | n=4060.9 (0.84, 0.97)0.92 (0.85, 0.99)0.93 (0.85, 1.02)0.92 (0.83, 1.01)
      Lower Tertile(0, 20.2) | n=1581 (ref)1 (ref)1 (ref)1 (ref)
      Mid Tertile(20.8, 104.3) | n=1330.73 (0.45, 1.19)0.77 (0.47, 1.27)0.79 (0.44, 1.41)0.63 (0.33, 1.23)
      Upper Tertile(105.1, 8792.38) | n=1180.62 (0.36, 1.09)0.71 (0.4, 1.25)0.77 (0.42, 1.41)1.66 0.34, 1.29)
      * Clinical Covariates include cold ischemia time and the following recipient variables: age (years), black race, sex, previous kidney transplant, cause of end-stage kidney disease (4 categories, other as reference), number of human leukocyte antigen mismatches, panel reactive antibody (%), body mass index (kg/m2), and pre-emptive transplant
      Tables S5 & S6 show exploratory analyses of effect modification. DCD status modified the association of urinary NGAL with rejection or allograft failure. As shown in Tables S7 & S8, donor urinary biomarkers were neither significantly associated with the outcome of BPAR nor with a composite of BPAR, graft failure, or death.
      Donor AKI, defined using the AKIN scale, was also not associated with the primary or secondary outcomes or with the outcomes of exploratory analyses (Tables S9 – 12).

      Discussion

      In this large and well-phenotyped cohort, we found no association between donor kidney injury and inflammation biomarkers and a composite outcome of graft failure and acute rejection. In a subcohort, we also found no association between these biomarkers and a composite outcome that also included DSA. A secondary analysis also detected no association between clinical AKI and the primary outcome. These findings contradict our hypothesis. We propose that donor AKI may have provoked inflammation in the allograft, but contemporary immunosuppression may have been sufficient to ameliorate immunological consequences of inflammation after transplantation. Taken together with other studies, this analysis provides new evidence that transplant centers can successfully manage complications and achieve good outcomes using AKI kidneys.
      • Liu C.
      • Hall I.E.
      • Mansour S.
      • Thiessen Philbrook H.R.
      • Jia Y.
      • Parikh C.R.
      Association of Deceased Donor Acute Kidney Injury With Recipient Graft Survival.
      • Sonnenberg E.M.
      • Hsu J.Y.
      • Cohen J.B.
      • et al.
      Acute Kidney Injury in Deceased Organ Donors and Kidney Transplant Outcomes: A National Cohort Study using a Novel Data Source.
      • Reese P.P.
      • Hall I.E.
      • Weng F.L.
      • et al.
      Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes.
      ,
      • Hall I.E.
      • Akalin E.
      • Bromberg J.S.
      • et al.
      Deceased-donor acute kidney injury is not associated with kidney allograft failure.
      Deceased-donor kidneys with AKI are frequently discarded due to concerns about early clinical complications such as primary nonfunction and longer-term risks of allograft fibrosis.
      • Hall I.E.
      • Schroppel B.
      • Doshi M.D.
      • et al.
      Associations of deceased donor kidney injury with kidney discard and function after transplantation.
      Yu et al. examined kidney non-procurement among deceased donors in the US from 2000 to 2018. Compared to donors with terminal creatinine <1.00 mg/dL, those with values between 1.00-1.49 mg/dL and between 1.50-2.00 mg/dL (for AKI or any reason) were 48% and 300% more likely to have no kidneys procured, respectively.
      • Yu K.
      • King K.
      • Husain S.A.
      • et al.
      Kidney nonprocurement in solid organ donors in the United States.
      It is clear that donor AKI increases the risk of recipient DGF.
      • Hall I.E.
      • Schroppel B.
      • Doshi M.D.
      • et al.
      Associations of deceased donor kidney injury with kidney discard and function after transplantation.
      ,
      • Heilman R.L.
      • Smith M.L.
      • Smith B.H.
      • et al.
      Long-term Outcomes Following Kidney Transplantation From Donors With Acute Kidney Injury.
      However, studies from diverse data sources have demonstrated that recipients of AKI kidneys still usually experience longer-term graft survival and allograft function similar to kidneys without AKI.
      • Liu C.
      • Hall I.E.
      • Mansour S.
      • Thiessen Philbrook H.R.
      • Jia Y.
      • Parikh C.R.
      Association of Deceased Donor Acute Kidney Injury With Recipient Graft Survival.
      ,
      • Sonnenberg E.M.
      • Hsu J.Y.
      • Cohen J.B.
      • et al.
      Acute Kidney Injury in Deceased Organ Donors and Kidney Transplant Outcomes: A National Cohort Study using a Novel Data Source.
      ,
      • Heilman R.L.
      • Smith M.L.
      • Smith B.H.
      • et al.
      Long-term Outcomes Following Kidney Transplantation From Donors With Acute Kidney Injury.
      ,
      • Hall I.E.
      • Akalin E.
      • Bromberg J.S.
      • et al.
      Deceased-donor acute kidney injury is not associated with kidney allograft failure.
      ,
      • Kwon J.A.
      • Park H.
      • Park S.J.
      • et al.
      Factors of Acute Kidney Injury Donors Affecting Outcomes of Kidney Transplantation From Deceased Donors.
      For example, Sonnenberg et al. examined a national US cohort of recipients of kidneys in which donor AKI was ongoing at procurement (terminal creatinine >1.5 mg/dL); one-third of these kidneys met criteria for AKI Stage 3. All-cause graft failure rates by 3 years were 15.5% for recipients of AKI vs. 15.1% for recipients of non-AKI kidneys. In multivariable adjustment, AKI kidneys were associated with only slightly higher risk of all-cause graft failure (aHR 1.05, 95% CI:1.01-1.09).
      • Sonnenberg E.M.
      • Hsu J.Y.
      • Cohen J.B.
      • et al.
      Acute Kidney Injury in Deceased Organ Donors and Kidney Transplant Outcomes: A National Cohort Study using a Novel Data Source.
      Prior studies from our DDS cohort examined AKI using both creatinine-based criteria and donor urinary biomarkers, which can detect subclinical AKI located in the distal tubule or other compartments of the nephron.
      • Hall I.E.
      • Akalin E.
      • Bromberg J.S.
      • et al.
      Deceased-donor acute kidney injury is not associated with kidney allograft failure.
      We found that higher donor NGAL was associated with recipient DGF (highest vs. lowest NGAL tertile relative risk 1.21; 95% CI 1.02, 1.43). Yet, analyses of 6-month recipient eGFR revealed that NGAL and liver fatty acid binding protein were associated with only modestly lower eGFR, and this association was restricted to recipients without DGF.
      • Reese P.P.
      • Hall I.E.
      • Weng F.L.
      • et al.
      Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes.
      While a study from the United Kingdom reported higher primary nonfunction rates for Stage 3 AKI kidneys (9% vs. 4%, p=0.04) and advised caution about accepting these kidneys,
      • Boffa C.
      • van de Leemkolk F.
      • Curnow E.
      • et al.
      Transplantation of Kidneys From Donors With Acute Kidney Injury: Friend or Foe?.
      those results contrast with findings from multiple other single and multi-center studies that have described favorable outcomes after kidney transplantation with AKI kidneys.
      • Sonnenberg E.M.
      • Hsu J.Y.
      • Cohen J.B.
      • et al.
      Acute Kidney Injury in Deceased Organ Donors and Kidney Transplant Outcomes: A National Cohort Study using a Novel Data Source.
      ,
      • Heilman R.L.
      • Smith M.L.
      • Smith B.H.
      • et al.
      Long-term Outcomes Following Kidney Transplantation From Donors With Acute Kidney Injury.
      ,
      • Hall I.E.
      • Akalin E.
      • Bromberg J.S.
      • et al.
      Deceased-donor acute kidney injury is not associated with kidney allograft failure.
      The present analysis provides fresh data by focusing on immunological outcomes of acute rejection and de novo DSA. We suggest potential explanations for the lack of association between donor injury biomarkers and our composite outcome. First, many scientific insights related to HLA upregulation due to ischemic injury were derived in the ischemia-reperfusion setting at implantation and may not apply to the earlier event of donor AKI.
      • Land W.G.
      The role of postischemic reperfusion injury and other nonantigen-dependent inflammatory pathways in transplantation.
      ,
      • Kreimann K.
      • Jang M.S.
      • Rong S.
      • et al.
      Ischemia Reperfusion Injury Triggers CXCL13 Release and B-Cell Recruitment After Allogenic Kidney Transplantation.
      Indeed, we previously made the observation that among kidney transplant recipients with DGF, recipients of kidneys from donors with elevated injury markers actually experienced better 6-month graft function vs. recipients of kidneys with low levels of injury. As a result, our group speculated that donor AKI might provoke ischemic preconditioning and upregulation of molecular mechanisms that protect against ischemia-reperfusion.
      • Reese P.P.
      • Hall I.E.
      • Weng F.L.
      • et al.
      Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes.
      Next, we note that eighty-two percent of recipients received anti-thymocyte globulin and nearly all received tacrolimus and mycophenolate. This regimen may have been sufficient to mitigate immunological responses caused by AKI.
      We acknowledge limitations. It is possible that an association between AKI and subclinical rejection exists but was undetected because of limited power or because surveillance biopsies were not part of center protocols. We also did not measure novel genetic biomarkers of rejection such as cell-free DNA or others that may reflect gene expression. On the other hand, our findings suggest that even if AKI caused subclinical rejection, the clinical consequences were limited, perhaps due to the robust immunosuppression regimen. From that perspective, we emphasize that our results do not need to be interpreted as contradicting the “Injury Hypothesis.”
      • Land W.G.
      The role of postischemic reperfusion injury and other nonantigen-dependent inflammatory pathways in transplantation.
      Second, it is possible that centers only accepted AKI kidneys with otherwise favorable characteristics. We acknowledge that subsequent studies, for instance with a higher proportion of AKIN Stage 3 kidneys, might find an association between severe AKI and risk of recipient rejection. Nonetheless, we adjusted for a wide range of characteristics relevant to immunological outcomes, including HLA mismatch, PRA, and recipient age. We call attention to a recent study using this cohort in which donor AKI was associated with reduced risk of BK virus infection and BK nephropathy-associated graft failure.
      • Hall I.E.
      • Reese P.P.
      • Mansour S.G.
      • et al.
      Deceased-Donor Acute Kidney Injury and BK Polyomavirus in Kidney Transplant Recipients.
      This finding suggests the possibility that specific and, thus far undefined, pathways of immunological activation in a donor AKI kidney might be protective against viral infection. The present study also has the limitation that DSA assessment took place at each center’s laboratory. However, all five centers in the subcohort used the same screening platform and single antigen beads to characterize DSA. The investigators then applied uniform criteria to the binary outcome of de novo DSA. An additional limitation is that all participating centers were academic medical centers. We also did not adjust for induction therapy or perfusion pumping, because of concerns about confounding-by-indication for kidneys at risk of injury. We also emphasize the study’s strengths in that the population was large, ethnically diverse, and the kidney transplant recipients were treated with the most common immunosuppressive regimens used nationally and experienced outcomes such as rejection and graft failure at rates similar to the national experience.
      • Tanriover B.
      • Jaikaransingh V.
      • MacConmara M.P.
      • et al.
      Acute Rejection Rates and Graft Outcomes According to Induction Regimen among Recipients of Kidneys from Deceased Donors Treated with Tacrolimus and Mycophenolate.
      In this multi-center study with close follow-up of recipients, donor injury biomarkers were neither associated with the primary outcome of graft failure and rejection, nor with a secondary outcome that included de novo DSA. These results should be confirmed in other cohorts. For transplant centers trying to develop greater experience with transplanting donor AKI kidneys, these findings provide initial evidence that accepting deceased-donor kidneys with AKI will not substantially increase risks of acute rejection under a regimen of robust immunosuppression.

      References

        • Hall I.E.
        • Schroppel B.
        • Doshi M.D.
        • et al.
        Associations of deceased donor kidney injury with kidney discard and function after transplantation.
        Am J Transplant. 2015; 15: 1623-1631
        • Boffa C.
        • van de Leemkolk F.
        • Curnow E.
        • et al.
        Transplantation of Kidneys From Donors With Acute Kidney Injury: Friend or Foe?.
        Am J Transplant. 2017; 17: 411-419
        • Jang H.R.
        • Rabb H.
        The innate immune response in ischemic acute kidney injury.
        Clin Immunol. 2009; 130: 41-50
        • Singbartl K.
        • Formeck C.L.
        • Kellum J.A.
        Kidney-Immune System Crosstalk in AKI.
        Semin Nephrol. 2019; 39: 96-106
        • Land W.G.
        The role of postischemic reperfusion injury and other nonantigen-dependent inflammatory pathways in transplantation.
        Transplantation. 2005; 79: 505-514
        • Colvin R.B.
        • Smith R.N.
        Antibody-mediated organ-allograft rejection.
        Nat Rev Immunol. 2005; 5: 807-817
        • Irish W.D.
        • Ilsley J.N.
        • Schnitzler M.A.
        • Feng S.
        • Brennan D.C.
        A risk prediction model for delayed graft function in the current era of deceased donor renal transplantation.
        Am J Transplant. 2010; 10: 2279-2286
        • Liu C.
        • Hall I.E.
        • Mansour S.
        • Thiessen Philbrook H.R.
        • Jia Y.
        • Parikh C.R.
        Association of Deceased Donor Acute Kidney Injury With Recipient Graft Survival.
        JAMA Netw Open. 2020; 3e1918634
        • Sonnenberg E.M.
        • Hsu J.Y.
        • Cohen J.B.
        • et al.
        Acute Kidney Injury in Deceased Organ Donors and Kidney Transplant Outcomes: A National Cohort Study using a Novel Data Source.
        Ann Surg. 2020 Nov 13; https://doi.org/10.1097/SLA.0000000000004597
        • Reese P.P.
        • Hall I.E.
        • Weng F.L.
        • et al.
        Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes.
        Journal of the American Society of Nephrology : JASN. 2016; 27: 1534-1543
        • Heilman R.L.
        • Smith M.L.
        • Smith B.H.
        • et al.
        Long-term Outcomes Following Kidney Transplantation From Donors With Acute Kidney Injury.
        Transplantation. 2019; 103: e263-e272
        • Potluri V.S.
        • Parikh C.R.
        • Hall I.E.
        • et al.
        Validating Early Post-Transplant Outcomes Reported for Recipients of Deceased Donor Kidney Transplants.
        Clin J Am Soc Nephrol. 2016; 11: 324-331
        • Doshi M.D.
        • Reese P.P.
        • Hall I.E.
        • et al.
        Utility of Applying Quality Assessment Tools for Kidneys With KDPI >/=80.
        Transplantation. 2017; 101: 1125-1133
        • Hall I.E.
        • Reese P.P.
        • Doshi M.D.
        • et al.
        Delayed Graft Function Phenotypes and 12-Month Kidney Transplant Outcomes.
        Transplantation. 2017; 101: 1913-1923
        • Harhay M.N.
        • Jia Y.
        • Thiessen-Philbrook H.
        • et al.
        The association of discharge decisions after deceased donor kidney transplantation with the risk of early readmission: Results from the deceased donor study.
        Clin Transplant. 2018; 32e13215
        • Hall I.E.
        • Parikh C.R.
        • Schroppel B.
        • et al.
        Procurement Biopsy Findings Versus Kidney Donor Risk Index for Predicting Renal Allograft Survival.
        Transplant Direct. 2018; 4: e373
        • Hall I.E.
        • Akalin E.
        • Bromberg J.S.
        • et al.
        Deceased-donor acute kidney injury is not associated with kidney allograft failure.
        Kidney Int. 2019; 95: 199-209
      1. The Declaration of Istanbul on Organ Trafficking and Transplant Tourism. Istanbul Summit April 30-May 2, 2008.
        Nephrol Dial Transplant. 2008; 23: 3375-3380
        • Parikh C.R.
        • Hall I.E.
        • Bhangoo R.S.
        • et al.
        Associations of Perfusate Biomarkers and Pump Parameters With Delayed Graft Function and Deceased Donor Kidney Allograft Function.
        Am J Transplant. 2016; 16: 1526-1539
        • Lepeytre F.
        • Dahhou M.
        • Zhang X.
        • et al.
        Association of Sex with Risk of Kidney Graft Failure Differs by Age.
        J Am Soc Nephrol. 2017; 28: 3014-3023
        • Locke J.E.
        • Warren D.S.
        • Dominici F.
        • et al.
        Donor ethnicity influences outcomes following deceased-donor kidney transplantation in black recipients.
        J Am Soc Nephrol. 2008; 19: 2011-2019
        • Pisavadia B.
        • Arshad A.
        • Chappelow I.
        • et al.
        Ethnicity matching and outcomes after kidney transplantation in the United Kingdom.
        PLoS One. 2018; 13e0195038
      2. Chow SCS, J.; Wang, H. Sample Size Calculations in Clinical Research. 2nd ed2008.

        • Schoenfeld D.'
        Sample Size Formula for the Proportional-Hazards Regression Model.
        Biometrics. 1983; 39: 499-503
        • Yu K.
        • King K.
        • Husain S.A.
        • et al.
        Kidney nonprocurement in solid organ donors in the United States.
        Am J Transplant. 2020; 20: 3413-3425
        • Kwon J.A.
        • Park H.
        • Park S.J.
        • et al.
        Factors of Acute Kidney Injury Donors Affecting Outcomes of Kidney Transplantation From Deceased Donors.
        Transplant Proc. 2019; 51: 2575-2581
        • Kreimann K.
        • Jang M.S.
        • Rong S.
        • et al.
        Ischemia Reperfusion Injury Triggers CXCL13 Release and B-Cell Recruitment After Allogenic Kidney Transplantation.
        Front Immunol. 2020; 11: 1204
        • Hall I.E.
        • Reese P.P.
        • Mansour S.G.
        • et al.
        Deceased-Donor Acute Kidney Injury and BK Polyomavirus in Kidney Transplant Recipients.
        Clin J Am Soc Nephrol. 2021; 16: 765-775
        • Tanriover B.
        • Jaikaransingh V.
        • MacConmara M.P.
        • et al.
        Acute Rejection Rates and Graft Outcomes According to Induction Regimen among Recipients of Kidneys from Deceased Donors Treated with Tacrolimus and Mycophenolate.
        Clin J Am Soc Nephrol. 2016; 11: 1650-1661