Effect of Recipient Variables on Transplant Survival Following Marginal Kidney Donation: Analysis of a Mate Kidney Cohort

Introduction

End-stage kidney disease (ESKD) affects approximately 0.1% of the world’s population, and the number of patients in need of kidney replacement therapy (KRT) is rapidly growing [1]. According to data from the 2024 ISN Global Kidney Health Atlas (GKHA), chronic kidney disease (CKD) caused 1.43 million deaths worldwide in 2019, underscoring its growing impact as a global health burden [1–3]. Kidney transplantation is the best form of KRT because it is associated with lower mortality and cardiovascular risk, improved quality of life, and good cost-effectiveness [4]. According to the ERA Registry Annual Report 2019, recipients of deceased-donor kidney transplants had a significantly higher adjusted 5-year patient survival probability compared to patients who initiated dialysis during the same period (92.3% vs 42.3%) [5].

Despite this, several factors contribute to the increasing proportion of elderly and frail candidates on transplant waiting lists, including the shortage of deceased-organ donors, the aging population, advances in dialysis care, and improved management of dialysis-associated comorbidities. Moreover, allocation policies vary with respect to regional or national institutional and ethical principles, and specific populations of kidney transplant candidates.

In Eurotransplant (ET) countries, prolonged waiting times for deceased-donor kidney transplantation deceased-donor kidney transplant (DDKT) under standard allocation programs – namely, the Eurotransplant Kidney Allocation System (ETKAS) and the European Senior Program (ESP) – have led to the implementation of deviated allocation strategies [6].

Recipient-oriented extended allocation (REAL) and competitive rescue allocation (CRA) are the 2 types of deviated allocations activated when a kidney is rejected in the standard allocation process. Most kidney grafts allocated via REAL/RA are expanded-criteria donor (ECD) kidneys following the previous OPTN/UNOS definition [6,7]. ECD criteria were proposed in 2002 as an approach to expand the pool of kidney donors. ECD kidneys were defined as those with a relative risk of graft loss higher than 1.7 and included donors aged ≥60 years or 50–59 years with at least 2 of the following comorbidities: serum creatinine higher than 1.5 mg/dl, history of hypertension, or a cerebrovascular accident(CVA) as the cause of donor death [8]. Good long-term outcomes after transplantation of ECD kidneys compared to standard-donor criteria based on European transplant cohorts encourage the acceptance of kidneys in RA/REAL for specific transplant candidates [7,9]. In Germany, 76% of DDKTs are performed via REAL or CRA, leading to a significant reduction in waiting time [10].

In the United States, although the number of candidates listed for DDKT is below pre-COVID-19 pandemic levels, the discharge rate among specific subgroups has increased: 41.4% for biopsied kidneys, and ~72% for both kidneys from donors aged 65 years or older, and for kidneys with a Kidney Donor Profile Index (KDPI) of 85% or higher [11]. An ongoing challenge for allocation systems is improving the utility of deceased kidney grafts, with no current international consensus on optimal recipient and donor factors that could inform allocation decisions [12].

Mate kidney transplant studies have evaluated the impact of specific factors on transplant outcomes by attenuating donor-confounding factors [13,14]. In this retrospective study, we analyzed outcomes from 200 mate kidney recipients who received marginal kidneys allocated through non-standard allocation programs. Our donor pool met the above-mentioned ECD criteria in 66% of cases. The primary objective of our study was to identify recipient factors affecting graft survival in this population. Additionally, in a secondary analysis, we utilized the systematic implementation of zero-time biopsies at our center to evaluate the prognostic relevance of pre-implantation histological features on transplant outcomes. Given that donor-focused metrics dominate the decision-making process in current allocation systems, this mate kidney analysis aims to investigate how recipient-specific factors may influence transplant outcomes.

Material and Methods

ETHICS APPROVAL:

The study protocol was approved by the ethics committee of FAU (reference number 22-400-Br). All KTx procedures were conducted in accordance with the Declaration of Istanbul.

All transplant candidates provided informed consent at the time of inclusion in our transplant registry and again before surgery, authorizing the future use of biological materials or patient data for scientific purposes.

STATISTICAL ANALYSIS:

The chi-square test was used to compare groups for binary variables, and the paired Wilcoxon signed-rank test for numeric variables. For paired binomial data, McNemar’s test with continuity correction was performed. In this analysis, the donor was used as the unit of observation. McNemar’s test was applied to assess differences in target (dependent) variables between the 2 related groups (e.g. CIT groups, renal survival groups, and 1-month post-KTx Creatinine levels). Cox regression analysis of renal survival time was conducted for the entire cohort of kidney transplant recipients, with pairing accounted for using mixed-effects Cox regression models. The Sobel test was used for mediation analysis. Statistical significance was set at P <0.05. Statistical analyses were performed using R version 4.2.1.

Results

DEMOGRAPHIC FEATURES:

Among the two-hundred recipients, there were 144 men. The mean age of the recipients was 58 years. The mean waiting list time was 5 years (62.84±37.45 months). The mean recipient BMI averages 25.55 kg/m². In addition, 17.5% of the recipients had DM(N=35) and 28% had CVD(N=56). In our cohort, the etiology of kidney disease was attributed to diabetes and hypertension in 24% of cases. Glomerulonephritis accounted for 22% of cases, autosomal dominant polycystic kidney disease (ADPKD) for 15.5%. In approximately one-fifth of recipients, the underlying cause remained undetermined, primarily due to the absence of a diagnostic kidney biopsy. Only 7.5% were immunized, defined as a panel-reactive antibody (PRA) score >6% at the time of transplantation. The mean number of donor-recipient HLA-MMs for A-B-DR was 4.

One hundred and 29 patients received induction therapy with a monoclonal interleukin-2 receptor antibody (basiliximab). At our center, patients enrolled in the ESP program, as well all recipients with intermediate or high immunological risk received antithymocyte globulin (rATG), while the remaining recipients were administered basiliximab. Immunological Risk was assessed with Immunological Risk Index (IRI), integrates 5 parameters: HLA mismatches for A, B, DR, history of prior transplantation, presence of HLA antibodies, presence of DSA, and donor-related factors (age and cold ischemia time). The IRI score ranges from 0 to 19 points, with scores of 0–4 indicating low risk, 5–8 intermediate risk, and ≥9 high immunological risk.

Most recipients (96.5%) received triple-maintenance immunosuppressive therapy with calcineurin inhibitors (cyclosporine or tacrolimus), mycophenolate acid, and steroids. Tacrolimus represents the standard immunosuppressive therapy for most patients, whereas cyclosporine is the calcineurin inhibitor routinely used in patients enrolled in the ESP program at our center. In total, 141 patients received tacrolimus, while 54 patients were treated with cyclosporine. Average CNI-levels are according to our standard regimens which is based on the KDIGO recommendations and the results of the ELITE-Symphony trial. During the first 3 months post-transplantation, average tacrolimus trough levels ranged between 8 and 10 ng/mL, followed by 6–8 ng/mL thereafter, and 4–7 ng/mL beyond the first year. For cyclosporine, target trough levels were 120–150 ng/mL during the first 3 months, 100–120 ng/mL thereafter, and 60–80 ng/mL beyond one year.

Ninety-three recipients experienced at least one histologically proven episode of rejection. The rejections were further stratified according to the timing of occurrence after KTx: early (in the first year), late (after 12 months), and recurrent. Further stratification involved the type of rejection according to Banff classification. In 28 recipients, the rejections were graded as borderline according to the Banff Lesion Score [16].

The mean CIT was 11 h and 53 min (±4: 43), while the average WIT2 was 44 min (±19.8). The incidences of delayed graft function (DGF) and permanent non-function were 29.5% and 5.5%, respectively. The recipient characteristics are shown in Table 1.

Sixty-six donors met the criteria for ECD. The mean donor age was 58.1 years, and 53 of them were older than 60 years. The donor creatinine level was 1.02 mg/dl, and 20 donors had creatinine values higher than 1.5 mg/dl. The mean donor body mass value was 28.04 kg/m2. Thirty donors had known hypertension or cardiovascular disease, and 7 had DM2. The baseline donor characteristics are listed in Table 2.

HISTOLOGICAL FEATURES:

Pre-implantation biopsy was available for 186 recipients (132 males). The mean number of glomeruli observed was 9. Immunofluorescence microscopy was performed for 122 biopsies. ATI/N was graded as moderate in 84 biopsies and severe in 42 biopsies. The histological features of the biopsy specimens are presented in Table 1.

DIALYSIS EXPOSURE: Recipients in the longer graft survival time (103.8±42.09 months) had a significantly shorter dialysis exposure compared to those in the shorter survival group (58.10 vs 68.86 months, p=0.001, Figure 1).

HLA MISMATCHING: Additionally, this group exhibited superior HLA matching (3.40 vs 3.78 MM, p=0.013). Increased mismatches in HLA-A, HLA-B, and HLA-DR were associated with worse graft survival (Figure 2).

COLD ISCHEMIA TIME, SECONDARY WARM ISCHEMIA TIME, AND PRIMARY FUNCTION:

Neither cold ischemia time (CIT: 11.89 vs 12.07 hours, p=0.967) nor secondary warm ischemia time (WIT2: 43.22 vs 45.71 minutes, P=0.2) had a significant impact on graft survival. Recipients with early graft dysfunction were more frequently represented in the shorter survival group (40% vs 27%); however, primary non-function (PNF) did not reach statistical significance in relation to graft survival.

SERUM CREATININE LEVELS 1 MONTH AND 1 YEAR AFTER TRANSPLANT:

Similarly, serum creatinine levels in the first month and 1 year after transplant did not significant differ between the groups.

CARDIOMETABOLIC RISK FACTORS:

The incidence of CVD and DM was comparable in both cohorts (P=0.503 and P=0.844, respectively).

REJECTION: While acute and chronic antibody-mediated rejection (AMR) are established predictors of graft loss [17,18], our study found no significant difference in graft survival between recipients who had at least 1 episode of rejection and those who did not. At the end of observation, 120 recipients had maintained functioning grafts, 26 recipients died with functioning grafts, and 23 had graft failure due to rejection. The presence of TPL-Failure was significantly associated with shorter graft survival (mean: 44.4±36.95 months vs 159±48.54 months; t test −5.28, P<0.001). A mediation analysis, accounting for the occurrence and type of rejection, did not reveal any mediation effect of TPL failure on renal survival (Table 3). We conducted a chi-square test to evaluate the association between basiliximab as induction therapy and the occurrence of rejection episodes; however no statistically significant association was observed (P=0.138). In addition, we conducted a logistic regression analysis to determine if the use of cyclosporine instead of tacrolimus was associated with a higher incidence of rejection episodes. The model indicates that tacrolimus is associated with a decrease in the likelihood of rejection compared to cyclosporine (P value=0.048). Despite the significant P value for the examined variable, the overall model explained only a small fraction of the variance (1.58%). Table 4 summarizes the factors associated with graft survival.

THE EFFECT OF SHORTER CIT:

We compared the DDKT recipients who were transplanted first, and thus were exposed to shorter CIT (mean 9.03±3.62 h), to their mate recipients with longer CIT (mean 14.83±3.96 h), finding that the incidence of DGF or permanent non-function did not differ between the groups (Figure 3), and no difference was observed in the graft survival time (Figure 4). Interestingly, the presence of severe ATI/N in the 0_Bx was identical between the short and long CIT groups. This suggests that, up to a certain threshold, histological ATI/N in the 0_Bx reflects donor characteristics rather than CIT duration itself. Recipients in longer CIT group were older (59.7 vs 57.3 years, P=0.020) and had more HLA mismatches (3.86 vs 3.48, P=0.008). This discrepancy may reflect more cautious clinical decision-making in these patients, possibly influencing CIT. Serum creatinine levels at 1 month and 1 year after transplantation did not significantly differ between the 2 groups ([shorter vs longer CIT 1st month]: 2.08 vs 1.99 mg/dl and ([1-month: 2.08 vs 1.99 mg/dL], [1-year: 1.49 vs 1.54 mg/dL]). Details of the associations between CIT and target variables are summarized in Table 5.

PARAMETERS AFFECTING EARLY GRAFT FUNCTION:

Mate kidney recipient pairs were stratified into 2 groups (lower and higher creatinine level) based on serum creatinine level at the 1 month after transplantation. The mean creatinine level in the lower group was 1.86 mg/dl, compared to 2.20 mg/dl in their mate recipients. Recipients in the lower creatinine group had a lower body mass index (BMI: 24.8 vs 26.29 kg/m2, P=0.011), better HLA matching (mean MM: 3.54 vs 3.85, P=0.026), and shorter cold ischemia time (CIT: 11.35 vs 12.43h, P=0.049) than their counterparts. Additionally, primary graft function and graft survival time differed significantly between the 2 groups. The incidence of DGF or primary non-function was higher among recipients with elevated creatinine levels (P=0.025), and graft survival was significant shorter in this group compared to their mates (P=0.002). The full results of analysis of variables associated with creatinine levels at 1 month are presented in Table 6.

SEVERE ATI/N IS ASSOCIATED WITH WORSE GRAFT SURVIVAL:

We found a significant correlation between severe ATI/N in the 0_Bx and renal survival (P=0.04). Table 7 summarize the result of McNemar’s analysis. In contrast, glomerulosclerosis did not emerge as a statistically significant predictor of graft survival (P=0.24) in the multivariable Cox proportional hazards model. The corresponding hazard ratio was 1.21, indicating a non-significant increase in the risk of graft loss in patients with sclerosis. The model demonstrated limited predictive power, with a concordance index of 0.52. Additionally, hyalinosis, as a categorized variable, was not significantly associated with graft survival, and the wide confidence intervals suggest considerable variability and potential model instability.

COX REGRESSION ANALYSIS FOR FACTORS AFFECTING GRAFT SURVIVAL:

Cox regression analysis revealed that the presence of severe ATI (P<0.001), recipient age at the time of transplantation (P=0.0219), number of mismatches (P=0.0114), and creatinine level at 1 year after transplant (P=0.0295) had a significant impact on the survival time after KTx. Table 8 summarizes Cox regression analysis results. Interestingly, dialysis exposure, which was significant in unadjusted analysis, did not retain statistical significance after adjustment.

Discussion

EFFECTS OF DIALYSIS EXPOSURE ON GRAFT SURVIVAL:

The duration of dialysis prior to kidney transplantation – often referred to as dialysis vintage—has been widely studied as a potential predictor of graft outcomes. In our cohort, longer pre-transplant dialysis exposure was significantly associated with reduced graft survival in paired analysis, although this association did not remain significant in multivariable analysis. These findings are consistent with previous studies suggesting that prolonged dialysis exposure can negatively affect post-transplant outcomes, including increased risks of graft loss and mortality [19–21]. Similarly, outcomes from rescue allocation (RA) in a population comparable to ours have shown that shorter waiting times improved patient survival and reduced risk of death with a functioning graft [7]. Conversely, other studies, such as that by Haller et al, did not demonstrate a significant association between pre-transplant dialysis duration and graft loss, although they did report increased recipient mortality with prolonged dialysis exposure [21]. The heterogeneity in findings likely reflects differences in patient populations, dialysis practices, and access to timely transplantation. Nonetheless, our results support that minimizing time on dialysis – particularly for recipients of marginal kidneys – can help optimize long-term graft survival.

EFFECTS OF HLA MATCHING ON GRAFT SURVIVAL:

Opelz et al, in the Collaborative Transplant Study, demonstrated a 20% improvement in graft survival for kidneys with zero HLA-A, -B, and -DR mismatches compared to those with 5 mismatches [22]. Despite advances in allocation strategies and post-transplant management, HLA matching continues to play a critical role in determining long-term graft outcomes in deceased-donor kidney transplantation [23]. A 2018 meta-analysis encompassing nearly 500000 deceased and living kidney transplants showed the prognostic significance of HLA mismatches. Notably, each HLA-DR mismatch was associated with a 12% increased risk of overall graft failure, while HLA-A mismatches were found to be less impactful and have been de-emphasized in some guideline recommendations [24]. Recent studies have further suggested that considering HLA-DQ mismatches, as well as epitope or eplet-level mismatching, could enhance predictive accuracy for graft survival and help refine allocation strategies [25]. In our cohort, approximately 50% of recipients had a maximum of 4 HLA mismatches, indicating a relatively low level of immunological disparity between donors and recipients. Despite this observation, HLA mismatching has also been recognized as a strong predictor of both short- and long-term graft survival in matched-pair and multivariable analyses.

EFFECTS OF RECIPIENT AGE ON GRAFT SURVIVAL:

In the aging ESRD population, outcomes and trends in kidney transplantation (KTx) for elderly recipients have gained increasing attention. Multiple studies have demonstrated that advanced recipient age is associated with higher post-transplant mortality [26,27]. A retrospective study evaluating the applicability of Kidney Donor Profile Index (KDPI) and Estimated Post-Transplant Survival (EPTS) scores into the ET region identified donor and recipient age as the most significant predictors of both graft and recipient survival [28,29]. Notably, there was worse 5-year overall graft survival and patient survival in recipients aged ≥60 years, although no differences were observed in graft or death-censored graft survival [30,31]. In a study comparing ESP and ETKAS population, age and HLA mismatches were the main predictors of mortality [32]. Older recipients also have a lower incidence of acute rejection, likely due to immunosenescence. Nevertheless, death with a functioning graft remains the main reason for graft loss in this population [29]. These findings, coupled with persistent organ shortages, support offering extended-criteria donor kidneys (ECD) to elderly candidates, particularly when considering the survival advantage over remaining on dialysis [7,29,33]. In our study, recipient age at transplantation was independently associated with poorer graft survival in multivariable Cox regression analysis (HR=1.02 per year, P=0.022). However, this association was likely attenuated by the narrow age distribution within our matched cohort (mean recipient age 57.09 vs 59.21 for renal survival classification groups) (Table 3). Furthermore, there was no mediation effect of death on renal survival time in our cohort.

FACTORS ASSOCIATED WITH BETTER EARLY GRAFT FUNCTION:

Serum creatinine levels at 1 month after KTx are widely recognized as independent predictors of long-term graft function and survival. Toyotome et al reported that better HLA-matched transplants exhibited lower creatinine levels during the first 20 days, which improved 1-year graft survival [34]. Similarly, a retrospective analysis including 2462 recipients showed that female sex, lower BMI, and better HLA A-B-DR-DQ locus matching were significantly associated with improved short-term graft function [35]. A meta-analysis of 138081 KTx recipients found a BMI ≥30 kg/m2 independently increased the risk of DGF [36]. In our cohort, we found that a BMI <25 kg/m2, better HLA matching, a CIT <12 hours, and primary function were significantly associated with lower serum creatinine levels at 1 month after transplant. Notably, recipients from the lower creatinine group at 1 month after transplant also had significantly longer graft survival (Table 6). Similar to our findings, previous studies have linked obesity to adverse intraoperative and early postoperative outcomes, including wound infections and increased cardiovascular complications [37]. Moreover, obesity has been associated with higher DGF risk, although without significant differences in survival [36].

EFFECTS OF CIT ON GRAFT SURVIVAL:

Evidence indicates that prolonged CIT is an independent factor for the incidence of DGF in DDKT, especially in marginal kidney transplants (with a KDPI ≥85%) [14], but the impact of CIT on graft and patient survival is not well established [38,39]. A CIT ≤18 h in deceased-donor kidneys and ≤16 h in living-donor kidneys was not associated with worse transplant outcomes [40,41]. Prolonged CIT alone does not affect the long-term outcomes of transplants. Non-marginal kidneys showed better long-term transplant survival, even with a prolonged CIT of <40 hours [42]. In our study, we did not find a correlation between CIT and graft survival, perhaps due to the relatively short CIT of our cohort.

EFFECTS OF ALLOGRAFT REJECTION ON GRAFT SURVIVAL:

Risk of graft loss due to chronic allograft injury and death with a functioning graft increases in patients who experience rejection in the first 6 months after transplant [43]. All types and timing of rejections affect long-term graft function [17]. However, irreversible rejection and AMR demonstrate deterioration of kidney function and are associated with a higher risk of graft loss due to chronic AMR [18,43]. In the present study, rejection showed no effect on graft survival when comparing recipient groups classified according to renal survival (Tables 3, 4). Further stratification revealed that 28 rejections were borderline (Table 1). Many of these borderline rejections were proven in protocol biopsies performed 12 months after transplantation with the absence of clinical rejection signs. At our center, protocol biopsy is a routine procedure for patients with intermediate or high immunological risk (IRI score). No chronic rejection was observed in our cohort. Mediation analysis did not reveal any mediation effect of rejection type on renal survival (Table 3). A possible explanation is that most of the rejection episodes were borderline or acute aCMR rejection episodes and not chronic or acute chronic ABMR episodes. Our cohort was also a non-immunized cohort, with 92% of the recipients having a PRA score <6%.

EFFECTS OF RECIPIENT CVD AND DIABETES ON GRAFT SURVIVAL:

CVD remains the most common cause of death with a functioning graft in patients after KTx [44]. Pre-KTx coronary artery disease (CAD) increases the risk of adverse cardiovascular events after KTx [45]. Despite pre-transplant CVD assessment, CVD management is under-represented in transplant candidates and kidney graft recipients [46]. Compared to non-DM recipients, those with type 2 DM have over 2 times higher long-term mortality rates and a significantly higher risk of graft failure [47]. Our study failed to detect an association between CVD prior to KTx and graft or overall survival. DM was also under-represented in our cohort. These are limitations of our study design, since we did not examine long-term cardiovascular complications.

FACTORS ASSOCIATED WITH BETTER GRAFT FUNCTION 1 YEAR AFTER TRANSPLANTATION:

In agreement with our findings, previous studies have identified serum creatinine values 1 year after KTx as a strong predictor of long-term kidney transplant survival [48] (Table 8).

DONOR EFFECTS:

Data on the impact of ATI/N on post-transplant outcomes remain controversial. Donor kidneys with histologically severe ATI/N have previously been associated with an increased risk of DGF and primary graft non-function risk, but not with inferior long-term graft survival [49,50]. This finding confirms the results of a 2022 observational study and emphasized that the predictive performance improved when ATI grading was integrated with clinical parameters into a composite DGF risk index [50]. There was no correlation between renal function and arterial hyalinization [49]. Other studies revealed no significant difference in graft failure for kidneys with versus without ATN [51]. Assesing the impact of histological features of pre-implantation biopsy in elderly recipients, ATN and glomerulosclerosis were significantly associated with renal outcomes after KTx, with severe ATN remaining the strongest independent predictor for renal function in this population in multivariate analysis [52]. The most robust finding of our study was the independent association between the presence of severe ATI/N on pre-implantation biopsy and long-term graft survival. This association was consistently demonstrated in the matched-pair analysis using McNemar’s test and in the multivariable Cox regression model (Tables 7, 8). This underscores the importance of detailed histological assessment in pre-implantation biopsies, particularly in marginal-donor scenarios where the risk of ATI/N is higher due to factors like advanced donor age. Conversely, there was no significant correlation between primary function and severe ATI. Moreover, no significant associations were identified between glomerulosclerosis or arteriolar hyalinosis on pre-implantation biopsy and graft survival, suggesting that these histological features have limited prognostic value in this context.

LIMITATIONS OF OUR STUDY:

This study has 5 limitations. First, it had a retrospective design. Second, it reflects the experience of a single transplant center. The population of recipients was also distinctive. Few of the recipients in this cohort were immunized, approximately the half were ESP, and most were allocated through deviated allocation programs. Consequently, generalizability to broader transplant populations, particularly those involving standard allocation pathways or highly sensitized patients, may be limited, especially given that only with only a small percentage of the study population was immunized. Third, the recipient population, in which diabetes and hypertension – commonly the leading global causes of end-stage renal disease (ESRD) – were under-represented. However, this distribution may reflect regional epidemiology and center-specific referral patterns. While the incidence of ESRD due to diabetes has generally increased in recent decades, it continues to vary considerably across countries [53]. For instance, data from the ERA-EDTA Registry in 2016 showed that only 23% of patients initiating renal replacement therapy (RRT) had diabetes mellitus as the primary cause of ESRD, which is comparable to our own findings. In contrast, the proportions of glomerulonephritis (22%) and autosomal dominant polycystic kidney disease (ADPKD) (15.5%) observed in our cohort may also reflect such regional or temporal variation [54]. Fourth, the mean cold ischemia time (CIT) in our cohort was relatively short, at 12 hours, which prevents comparisons with cohorts with longer CIT times. Fifth, a methodological limitation involves the histopathological classification of pre-implantation biopsies. The Banff criteria for zero-time biopsies were not formally established until 2017. Therefore, biopsies performed prior to that time may have been misclassified.

Conclusions

This study is one of the largest paired analyses of mate kidney recipients receiving grafts from marginal donors. Key strengths include a distinctive recipient profile, characterized by a high proportion of Eurotransplant Senior Program participants and a predominantly non-immunized population. The use of a matched-pair design and comprehensive evaluation of multiple recipient-related parameters strengthen the internal validity of the findings.

We identified several recipient-related factors independently associated with long-term graft survival, including pre-transplant dialysis time, higher degrees of HLA mismatching, older recipient age, and elevated 1-year post-transplant serum creatinine levels. In contrast, cold ischemia time (CIT) did not significantly influence long-term outcomes, nor did it differ meaningfully between mate kidneys with or without primary function, likely reflecting the relatively short CIT observed in our cohort. Notably, the detailed histopathological assessment of pre-implantation biopsies, particularly the identification of severe acute tubular injury/necrosis (ATI/N), provided prognostic insights. These findings may inform post-transplant management strategies, support more efficient utilization of marginal kidneys, and encourage further research into the prognostic significance of ATI/N in kidney transplantation.