Kidney transplant represents the best option for replacement kidney therapy for patients who develop advanced chronic kidney disease. However, the supply of young donors, who are theoretically ideal, is limited, and does not cover the demand among patients on the waiting list for a kidney transplant. This imbalance has resulted in the growing use of older donors, often with associated morbidity such as hypertension and diabetes. Good clinical results are obtained with these organs provided that these grafts are suitably and receptors are adequate.1–4

Several scales have been developed with the aim of assessing the quality of donors’ kidneys and to determine their viability and post-transplant function. The Organ Procurement and Transplantation Network/United Network for Organ Sharing (OPT/UNOS) defined the concept of expanded donor criteria, but the prognostic utility of these criteria is limited, as the concept takes into consideration heterogeneous clinical profiles with different clinical outcomes.5,6 Other indexes stratify graft survival more precisely, but have a moderate discriminatory value.7–11 The donor's kidney biopsy provides information that may improve the prediction of functional outcome. However, the results that have been published are controversial and have not been widely used.12,13

At Hospital Universitario Cruces in 2008, kidney biopsy of deceased older donors was incorporated into the protocol to ensure the quality of the kidney grafts being used. This study analysed the clinical results obtained and specifically assessed the relationship between histological lesions and glomerular filtration rate one year after transplant.

Table 2 shows the clinical and demographic characteristics of the study patients. 77% of donors were ≥70 years of age; 45% had HTA, and 14% had diabetes. 89% of recipients were ≥60 years of age. The eGFR of all the donors was >60ml/min/1.73m2. Table 3Table 3 specifies the distribution of histological lesions from the preimplantation kidney biopsies. These showed a median (Q1–Q3) of 48 glomeruli (32–71) and 2 arteries (1–3). No arteries were visualised in 11 biopsies (9%). In total, 19 biopsies (15%) obtained a score of 0–3 points, 63 biopsies (51%) obtained a score of 4–5 points and 42 biopsies (34%) obtained a score of 6–8 points.

Table 4Table 4 shows both the overall results and the results obtained in each of the histological categories. Following transplantation, 68% of grafts functioned immediately, 8% never functioned and all others functioned after a mean of 10 days. A total of 8% of patients received boluses of corticosteroids due to a suspicion of acute rejection. Four patients (3%) died during the first year after the transplant. A total of 14 patients (11%) lost the graft in that period of time, in 8 cases due to arterial or venous graft thrombosis (4 from the 6–8 histological group and the other 4 from the 4–5 histological group). Four patients from the 6–8 histology group lost the graft with evidence of fibrosis, tubular atrophy and non-specific chronic vascular lesions. Overall survival for the grafts, excluding deaths, was 89% after the first and second year and 84% after the third year.

The relationship between the score obtained in the pre-implantation kidney biopsy and eGFR one year after transplant was analysed in 110 patients. A total of 14 patients were discarded from the analysis: Four due to death in the first year post-transplant, 8 because graft thrombosis, one due to haemolytic-uraemic syndrome and one owing to graft loss associated with post-operative bleeding. A total of 4 patients from the 6–8 group who lost the graft with already reported lesions from fibrosis and tubular atrophy were included. These patients were assigned a filtration rate of 10ml/min/1.73m2 one year after transplant. Eighteen patients (16%) obtained a cumulative score of 0–3 points, 56 patients (51%) obtained values of 4–5 points and 36 patients (33%) obtained an assessment of 6–8 points. The biopsies showed a median (Q1-Q3) of 48 glomeruli (33–73) and 2.3 arteries (1–3). No arteries were visualised in 9 biopsies (8%). The age of the donors with a score of 6–8 was greater than all other groups, without significant differences being recorded in any of the other baseline clinical or demographic characteristics analysed. Regarding clinical outcomes, the incidence of patients with delayed graft function was 4 (22%) in the 0–3 group, 14 (25%) in the 4–5 group and 11 (31%) in the 6–8 category (p=0.80). The time of initiation of graft function was similar in the 3 groups. Nine patients had acute rejection: one (6%), 6 (11%) and 2 (6%), in the 0–3, 4–5 and 6–8 groups respectively (p=0.72). Graft survival 3 years after transplant, excluding deaths, was 100% in the 0–3 group, 91% in the 4–5 group and 89% in the 6–8 group (Table 5Table 5).

Regarding eGFR, the primary study outcome, statistical differences were detected. Mean (SD) eGFR for the 6–8 group (38.5 [14.1]ml/min/1.73m2) was lower than that reached in the group with a score of 4–5 (46.3 [15.7]ml/min/1.73m2 [p=0.03]) and that achieved in the 0–3 group (49.6 [12.5]ml/min/1.73m2 [p=0.04]). Similarly, a significant correlation (p=0.04) was detected between biopsy score and categorical distribution of eGFR one year after transplant. Patients with a higher score had a more unfavourable distribution (Table 6Table 6).

The incidence of patients with poor renal function one year after transplant, eGFR<30ml/min/1.73m2, was 7 (19%) in the 6–8 group, 8 (14%) in the 4–5 group and zero in patients with a score of 0–3. These differences were not statistically significant (p=0.17). In the univariate logistic regression analysis, which analysed the incidence of patients with an eGFR<30ml/min/1.73m2 one year after transplant, the variables that showed a correlation with a p-value<0.20 were dialysis method (haemodialysis vs peritoneal dialysis), delayed graft function, acute rejection, donor HTA, donor diabetes mellitus and final biopsy histological score. Tubular atrophy, interstitial fibrosis, hyaline arteriolopathy and myointimal elastosis assessed individually also had a p<0.20. Delayed graft function was the only variable that had a statistically significant association (p<0.05). In the final multivariate model (Table 7Table 7), the variables that showed a significant correlation were delayed graft function and acute rejection. The score for histological damage for the biopsies, the 6–8 group vs the 0–5 group, had a non-significant adjusted OR of 2.2 (95% CI, 0.7–7.3). In these regression analyses, the biopsy score was categorised into 2 groups, 0–5 and 6–8, owing to the fact that none of the patients in the 0–3 group had an eGFR<30ml/min/1.73m2.

We have presented the results of a follow-up study of 124 transplant recipients of a kidney from older donors in whom a preimplantation kidney biopsy had been performed. The overall results achieved in terms of mortality, graft survival, graft thrombosis, initial graft functioning, acute rejection and glomerular filtration rate were consistent with those reported in the literature in transplants performed with a similar clinical profile.3,16

The main aim of our study was to assess the relationship between the degree of severity of the histological lesions from the kidney biopsies performed and eGFR one year after transplant. The studies that have examined the prognostic value of preimplantation kidney biopsy are heterogeneous in terms of both methodology and results obtained. Glomerulosclerosis of the donor was the most evidenced initial lesion. Impairment ≥20% was associated with poorer graft clinical course.17,18 According to the data from the OPTN/UNOS, with 2096 kidneys biopsied, only the kidneys with glomerulosclerosis <6% had slightly better graft survival after 5 years than all other groups.19 Vascular and tubular-interstitial lesions also were associated with worse clinical course outcomes.20–22 With the aim of improving the prognostic utility, several scales have been proposed to produce an overall score for prior kidney lesions. The most widely used are that of Pirani–Remuzzi23 and the Maryland Aggregate Pathology Index (MAPI).24 Scales based on the Banff criteria have also been used, but at this time there is no specific proposal for this group.25–30 All of them, with their limitations and their methodological differences, show a correlation between the severity of the histological lesions and the functional evolution of grafts, in terms of both glomerular filtration rate and graft survival. However, analysis of the ROC curves from the studies conducted has shown that their discriminatory capacity with respect to graft survival is moderate, with area-under-the-curve values of 0.7–0.8.12,13,31–33 The scale by O’Valle et al.14,15 used in this study was an empirical proposal developed by pathologists from Spain and specifically designed to assess chronic histological damage from donors’ kidney biopsies. The biopsies were processed in paraffin, in a procedure that takes more time and resources than freezing but allows a better assessment of the different kidney compartments.34

Mean eGFR and incidence of patients with an eGFR<30ml/min/1.73m2 one year after transplant were the primary outcomes of our study. These variables have already been used previously in other publications, as they are prognostic factors with a major impact on the survival of kidney grafts in the medium to long term.29–33,35–38 In our results, grafts with a score of 6–8 reached a mean glomerular filtration rate one year after transplant lower than that achieved in the other 2 groups. This was also the group that recorded a greater incidence of grafts with a low eGFR<30ml/min/1.73m2 one year after transplant, although this difference was not statistically significant. The association between the severity of the histological impairment from the preimplantation biopsy and evolutionary eGFR of the graft has already been reported in other publications.29–33 In our analysis, delayed graft function and acute rejection were significantly associated with poor renal function one year after transplant, eGFR<30ml/min/1.73m2. Patients from the 6–8 group had a slightly greater incidence of delayed graft function, which could have influenced this group's worse renal function. Some studies have shown a correlation between preimplantation kidney lesions of the donor, incidence of delayed function and evolution of the glomerular filtration rate of the graft.21,22,27 The incidence of acute rejections was low, and very similar, in the 3 groups analysed.

In interpreting the results obtained, some considerations and limitations should be identified: 1. The histological criteria used in graft acceptance ruled out the use of kidneys with more severe histological impairment; this hindered assessment of their impact on the results. 2. In the analysis of glomerular filtration rate one year after transplant, patients who died or lost the graft in the first year after the transplant were excluded, with the exception of 4 graft losses that were considered to be related to the quality of the kidneys implanted, with non-specific lesions of fibrosis and tubular atrophy in the post-transplant biopsy. These patients were assigned an eGFR of 10ml/min/1.73m2. Graft thromboses were considered to be technical complications not directly related to preimplantation parenchymal lesions. 3. The variability between the diseases that the biopsies analysed was not controlled. 4. Arteries were not observed in 9 biopsies. Hence, these biopsies were not scored in terms of myointimal elastosis. This points to a potential error in classification that may have included some of these transplants in a lower histological group than the one to which they belonged. This fact could have reduced the clinical differences found between the groups with higher and lower histological severity. 5. The number of patients analysed limited the power of the statistical analyses performed.

With the results of the present study, it may be concluded that overall transplant outcomes of older kidney donors, with the acceptance and allocation criteria used, were adequate. Kidneys with greater structural damage in the preimplantation biopsy, a histological score of 6–8, had a worse renal function one year after transplant. This may lead to worse medium and long term graft survival, as has already been reported in previous studies.30–33 Delayed graft function was a significant variable that determine the incidence of grafts with poor renal function one year after transplant. This is a matter that calls for intervention to improve the results obtained. Acute rejection also had an impact on subsequent function of grafts, but the impact in our series was very low.

In our opinion, the use of kidneys with chronic structural damage has negative effects on transplant outcomes, but these negative effects may be acceptable in many cases, especially in older recipients. It is difficult to determine the threshold for histological damage that precludes transplant. In our study, we established a precautionary threshold similar to that used in other histopathological scales.23,28,29 Allowing kidneys with more histological impairment would increase the number of grafts available, but with some presumably worse overall results. Pre-implantation biopsy has its limitations. It requires locally available trained pathologists, its individual functional prognostic value is moderate and it may rule out viable kidneys. However, despite these weaknesses, preimplantation kidney biopsy makes it possible to determine the severity of the donor's kidney lesions. This is information that we consider to be useful, especially in older donors and donors with associated vascular morbidity. It complements the usual clinical–laboratory donor assessment and macroscopic graft assessment, and aids in decision-making regarding acceptance and allocation of kidneys to be transplanted with more criteria. Future initiatives should rigorously validate one of the existing scales or modifications thereof and arrive at a consensus on their use as a guarantor of the quality of our activities.

The authors declare that they have no conflicts of interest.