Discussion
Our results show that the transplant outcomes using kidneys from deceased donors with AKI are excellent. There are no differences in chronic changes on biopsy or eGFR's at 1 year between the AKI and non-AKI groups. Moreover, molecular phenotyping of the 4-month protocol biopsies by global gene expression profiling demonstrates complete resolution of the tissue injury gene signatures clearly present at the 1-month time point in the AKI group. It is important to note that this cohort represents kidneys from donors with severe AKI. Seventy-one percent of the SCD AKI kidneys and 65% of the ECD AKI kidneys met the criteria for most severe stage of AKI (AKIN stage 3).
Our study group represents a selected sample of AKI donor kidneys. Although the majority of the kidneys had AKIN stage 3 AKI, a transplant surgeon, nephrologist, and pathologist carefully selected each donor. In the early period of our experience, the selection criteria were more stringent, but with increased experience, we have liberalized the criteria. Currently, we only exclude kidneys with cortical necrosis or moderate to severe chronic changes on pre-implantation biopsy or concerning perfusion pump parameters, defined as terminal flow <80 mL/min. or resistance index of >0.40. However, the validity of using pulsatile pump parameters for the selection of kidneys for transplantation needs further investigation. Despite this liberalization of the selection criteria, our results do not show an adverse impact on the outcomes in the later cohort.
Since ECD kidney donor classification has been shown to be a significant risk factor for inferior outcomes, we stratified the cohort by ECD status. ECD is defined by UNOS as a donor >60 years old or donor >50 years old with at least two of the following criteria: death from cerebral vascular accident, history of hypertension or creatinine >1.5 mg/dL. More recently, the kidney disease profile index (KDPI) has been adopted as a new stratification schema for deceased donor kidneys in the US. The KDPI scores were not significantly different between the AKI and non-AKI cohort after they were stratified by ECD/SCD status (Table 1).
Results of previous studies have suggested that DGF may be a risk factor for early acute rejection. Despite the higher rate of DGF in the AKI cohort, we did not see an increased risk of rejection (Table 3). The cumulative rejection rate at 30 days was 2.5% in the AKI cohort and 4.4% in the non-AKI cohort. The lack of impact of the DGF on early rejection may be related to our practice of using induction therapy and starting tacrolimus early (postoperative day 2) and obtaining adequate drug levels within the first several days regardless of the DGF status. In addition, despite the higher rate of DGF in the AKI group, the length of hospital stay is similar in the two groups. This may be related to protocol driven processes for discharge and careful outpatient follow up at our center.
AKI in the nontransplant clinical setting has been associated with increased risk of developing CKD and end stage renal disease. The mechanism of this transition from AKI to CKD is thought to be related to the abnormal healing response, which may involve several cell types including tubular epithelial cells, fibroblasts, pericytes, myofibroblasts, fibrocytes, and immune cells. In the transplant kidney, there are a number of other potential etiologies for graft fibrosis other than ischemia reperfusion injury, including both the alloimmune response and nonimmune factors (e.g. drug toxicities, infections, and obstructive uropathy). These additional factors may mask any potential effect of an AKI-related abnormal healing response to cause allograft fibrosis. This may explain the apparent lack of impact of AKI donor kidneys on graft fibrosis at 1 year in our study. Although, it is reassuring to see no apparent negative impact on allograft fibrosis and renal function at 1 year, longer follow-up with a larger cohort is needed to confirm our results.
Our analysis showed that graft survival for the AKI cohort was similar to the control groups (Figure 1); however, the length of the follow-up period is limited. The median follow up was 19.6 months for the SCD–AKI cohort and 12.3 months for the ECD–AKI cohort. The number at risk in the survival analysis beyond 4 year for the SCD group and beyond 1 year for the ECD group is limited; therefore, it is not possible to make firm conclusions regarding graft survival beyond these time points. In addition, graft survival for patients who received a kidney from donors with stage 3 AKIN was not significantly different than the control group (Table 4); however, the follow-up period may not be adequate to demonstrate a difference. A longer follow-up period may be needed to demonstrate the impact of donor AKI and the severity of the AKI injury (i.e. AKIN stage 3 vs. <3), which are both associated with higher rates of DGF, on graft survival.
There are additional limitations to this study, including the retrospective, single center design, which might have introduced important biases in the donor and recipient selection, which were not controlled for in the analysis. The criteria for the selecting AKI donor kidneys were not consistent during the entire study period. Additionally, the biopsy rate at 1 year posttransplant was only 56%, therefore, the findings may not accurately reflect the changes in the entire group.
There have been several previous reports of single center experiences with transplanting kidneys from select donors with AKI. Similar to our findings, these studies have shown a higher rate of DGF, but good patient and graft survival and GFR outcomes with AKI donor kidneys. Our study includes a larger proportion of donors with severe AKI and we included a broader spectrum of AKI kidneys, including kidneys with diffuse glomerular thrombi, donors with rhabdomyolysis, donors with oligoanuria, and donors on RRT.
A unique aspect of our study is the data from the 1-month biopsies of AKI kidney recipients showing many differentially expressed genes associated with cell death/stress, inflammation, and disrupted metabolic processes typical of kidney injury. However by 4 months, these signals have resolved. These results offer additional reassurance regarding the safety of transplanting properly selected AKI donor kidneys. We also note that over 800 genes were significantly differentially expressed at 1 month in the AKI group despite no significant differences in the serum creatinine or Banff histology scores with the non-AKI group. The role of injury and immune inflammation has been documented in AKI by others. It has been shown that the immune response in AKI involves both the innate and adaptive immune systems. The detailed analyses of biopsies from patients with AKI unequivocally showed the presence of mononuclear leukocytes (some CD3+ T cells) and neutrophils using immunoperoxidase staining. Similarly, Gomez et al have proposed a unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury that goes beyond the classical attribution of all AKI to ischemia on the basis of macrohemodynamic changes. The metabolic shift in AKI that we observed have also been observed in a recent metabolomics study, where the metabolome of plasma, kidney cortex, and medulla were profiled in a mouse model of renal ischemia/reperfusion surgery using gas chromatography/mass spectrometry (GC/MS) and liquid chromatography/mass spectrometry (LC/MS) platforms. The results showed changes in specific metabolites associated with early injury and a shift of energy source from glucose to lipids as well as inflammatory changes and a subsequent late phase kidney recovery. These results reveal the complementary clinical value of molecular phenotyping by global gene expression profiling of biopsies and underline how insensitive serum creatinine can be as a metric of significant kidney transplant injury and inflammation.
Our study suggests that there is a significant opportunity to increase the utilization of kidneys from AKI donors. When analyzing all the AKI kidney offers from a recent 1-year period and applying our current acceptance criteria, we estimated that 31% of discarded SCD kidneys and 22% of discarded ECD kidneys could be acceptable for transplantation. We believe utilization of these kidneys is now safe and feasible. The objective of this report is to support new efforts to increase awareness among the dialysis, transplant, and OPO communities of the excellent results that can be achieved with transplanting selected AKI kidneys and thus increase national efforts to efficiently place these organs.