• Deceased donor kidneys;
  • graft survival;
  • scoring system


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Despite the common use of diagnostic pretransplant deceased donor kidney biopsy, there is no consensus on the prognostic significance of the pathologic findings. In order to assist clinicians with interpretation we analyzed 371 pretransplant biopsies and correlated the findings with graft failure. Glomerular pathology was assessed with percent glomerulosclerosis (GS), glomerular size and periglomerular fibrosis (PGF); vascular pathology with arterial wall-to-lumen ratio (WLR) and arteriolar hyalinosis and interstitial pathology with measurement of cumulative fibrosis and presence of scar. Using two-thirds of the study population as a model-development cohort, we found that biopsy features independently associated with an increased risk of graft failure were GS ≥15%, interlobular arterial WLR ≥0.5 and the presence of PGF, arteriolar hyalinosis or scar. The Maryland Aggregate Pathology Index (MAPI), was developed from these parameters and validated on the remaining one-third of the population. Five-year actuarial graft survival was 90% for kidneys with MAPI scores between 0 and 7, 63% for scores from 8 to 11 and 53% for scores from 12 to 15 (p < 0.001). We conclude MAPI may help transplant physicians estimate graft survival from the preimplantation biopsy findings, in clinical situations similar to this study population (cold ischemia over 24 h, GS < 25%).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

The resulting severe shortage of organs for transplantation has led to continuous efforts to expand the kidney donor pool (1). One strategy has been to incorporate the use of nonideal donor kidneys that may have been considered unusable in earlier times (2–10). Biopsy of deceased donor kidneys before transplantation is commonly practiced, especially in clinical situations where histologic abnormalities are deemed likely. In the United States about 75% of kidneys from expanded criteria donors (ECD) are biopsied and 41% of these ECDs in turn discarded secondary to pathological parameters (11). Biopsy is indicated for assessment of kidneys from elderly, hypertensive or diabetic donors or those with renal dysfunction. Clinicians hope the histologic assessment will inform them about the suitability of the organ for transplantation, and help them estimate its projected survival. This, in turn, allows them to decide whether the organ should be transplanted or discarded, and to give preoperative estimates of graft success to help with patient–physician decision making. Unfortunately, there is no consensus on either the prognostic significance of biopsy findings in this setting or even on reporting standards for pretransplant kidney biopsies. These factors foster inconsistent interpretation and diminish the biopsy's diagnostic value and can potentially result in the inappropriate discard of a potentially viable kidney.

This study was performed to further clarify the clinical significance of individual pathologic findings on preimplantation kidney biopsies. We examined a large number of these biopsies from a single transplant center with both traditional and morphometric methods. The findings were correlated with graft survival, and a scoring system was developed to assist practitioners with biopsy interpretation.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Patient selection

The electronic records of the Department of Pathology at the University of Maryland Medical Center (UMMC) were searched for pretransplant biopsies of deceased donor kidneys dated between January 1999 and December 2005. Of the 549 pretransplant biopsies identified, 371 of the corresponding kidneys were transplanted as single organ transplants, the rest being either discarded or transplanted as a multiple organ procedure including dual kidney transplants. The majority of these kidneys were imported from other organ procurement organizations (OPO) after being rejected by the local centers. The biopsies and donor–recipient pairs of the single organ transplants constituted the study cohort. Clinical information was obtained from UMMC medical records after approval for the study by the University of Maryland School of Medicine Institutional Review Board.

Biopsy analysis

Each pretransplant kidney biopsy was evaluated three ways: frozen and permanent section examinations, and with morphometric measurements. Seven micron thick hematoxylin and eosin (H&E) stained frozen sections were initially examined and reported by the pathologist on call. Subsequent processing for paraffin-embedded permanent sections were also stained with H&E and Masson's trichrome stains and read by a pathologist specializing in transplantation.

Both frozen and permanent sections were evaluated for percentage of global glomerulosclerosis (GS). The degree of cortical interstitial fibrosis and tubular atrophy were subjectively graded as absent (<5%), mild (up to 25%), moderate (>25% and <50%) or severe (>50%). In addition, arterial sclerosis (concentric fibrointimal thickening and luminal narrowing) was classified by the Banff 97 grading schema as mild (<25% luminal narrowing, cv1), moderate (>25% and <50%, cv2) or severe (>50%, cv3) (12). Frozen and permanent sections were also reevaluated for the presence of any of the following: (a) periglomerular fibrosis (PGF), defined as thickening, wrinkling and reduplication of the Bowman's capsule (Figure 1A); (b) arteriolar hyalinosis, defined as amorphous, homogeneous eosinophilic deposits in the wall of arterioles (Figure 1B) and (c) scar, defined as a focus of sclerosis and renal parenchymal fibrosis and atrophy involving at least 10 tubules (Figure 1C).


Figure 1. Pathologic features used in MAPI as seen on frozen section preparations. (A) Periglomerular fibrosis, (B) arteriolar hyalinosis, (C) scar including features of interstitial fibrosis, tubular atrophy and glomerulosclerosis, and (D) measurements for arterial wall-to-lumen ratio (WLR) calculation, including the thickness of two opposing walls (T1 and T2) and the luminal diameter (LD). WLR = (T1 + T2)/LD.

Download figure to PowerPoint

Morphometric biopsy analysis was done in order to reduce subjectivity and variability in biopsy reads. Four micron thick sections were digitized under a 10× objective lens using the Automated Cellular Imaging System II (ACIS, Clarient Inc, San Juan Capistrano, CA) after staining with H&E and Masson Trichrome. All the morphometric analysis was done by using the image analysis software of the manufacturer. Glomerular pathology was quantitated with percentage of GS, glomerular size and presence of PGF. Glomerular size was determined by measuring and averaging the diameter of the four largest glomeruli. Vascular pathology was quantitated with arterial wall-to-lumen ratios (WLR) and presence of arteriolar hyalinosis. The WLR was derived by measuring the thickness of opposing two walls (T1 and T2) and the luminal diameter (LD) of the four most subjectively affected, well-oriented glomerular arterioles, interlobular arteries and large caliber arteries located in the proximity of the medulla (arcuate/interlobar–size arteries). The WLR was then calculated using the formula WLR = (T1+ T2)/LD (Figure 1D). Interstitial pathology was studied morphometrically by measuring cumulative fibrosis and visually assessed the presence of scar. Cumulative fibrosis was the percentage of the biopsy surface containing mature collagen by computerized colorimetric analysis on Masson's trichrome stain, including interstitial, glomerular and periarterial fibrosis.

Data analysis

The clinical outcome parameter of kidney graft failure was defined as the recipient's return to dialysis, or a decline in renal function to a glomerular filtration rate (GFR), of <20 mL/min. The GFR was calculated with the modification of diet in renal disease (MDRD) equation (13) using the mean of last three consecutive creatinine readings on routine bloodwork obtained after resolution of postoperative delayed graft function. Grafts lost to technical failures within 72 h of transplantation were excluded and patient deaths were treated as censored outcomes in actuarial graft survival analyses. Raw data were inspected using descriptive statistics and graphical representations to decide if logarithmic conversions were needed. The whole study population was then randomly split into model-development and model-validation cohorts at a 2:1 ratio using S-Plus 8.0 for Windows (Insightful Corporation, Seattle, WA).

A pathologic scoring system (Maryland Aggregate Pathologic Index, MAPI) was developed as follows: Cox proportional hazard methods were used to identify the pretransplant clinical and pathologic dependent variables that were associated with graft loss. Separate univariate Cox models were run for each variable. Points were assigned to the pathological variables that were statistically significant in the Cox regression by rounding their hazard ratio (HR) to the nearest integer (Table 2). Using the model-development cohort with graft failure as the outcome of interest, cut-offs for the continuous variables including GS, WLR of interlobular artery were selected using receiver operating curve analysis to maximize sensitivity and minimize the false positive rates. These cut-offs were used to categorize continuous variables. The scoring system was evaluated by Multivariate Cox Proportional Hazards Models, in order to assess the association between donor clinical variables, and the combined index (MAPI) with the outcome of graft loss. Kaplan–Meier survival curves were censored by time of death and loss to follow-up. The scoring system was validated with the model-validation cohort using area under the receiver operating curve analysis.

Table 2.  MAPI scoring system for pretransplant kidney biopsies, with HR of graft failure from Cox univariate analyses (model-development cohort)
 HR (95% CI)pMAPI points AbsentPresent
  1. WLR = wall-to-lumen ratio; CI = confidence interval; Score range from 0 to 15 MAPI points.

Arteriolar hyalinosis (any)3.93 (2.02–7.64)<0.000104
PGF (any) 4.09 (1.65–10.14)0.00204
Scar (any)2.58 (1.24–5.38)0.0103
GS ≥ 15%1.87 (1.17–2.99)0.00902
WLR interlobular arteries ≥ 0.52.05 (1.21–3.47)0.00802

Statistical calculations were performed using SPSS 15.0 for Windows (SPSS, Chicago, IL), except receiver operating curve analysis, which was done with NCSS software (NCSS, Kaysville, UT). Kaplan–Meier survival curves were compared with the log-rank test. Significance for all tests was set at 0.05.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Clinical data

In the model-development group the mean donor age was 50 ± 15 years, 62% were males and 19% African-American (Table 1). Cold ischemic time was 33 ± 10 h. The terminal creatinine was 1.2 ± 0.5 mg/dL. Hypertension was present in 40% and diabetes in 7% of donors. The 259 recipients in the model-development group had a mean age of 58 ± 13 years, were 64% male and 54% African-American. The mean peak PRA values were 11 ± 25% and <1% of cases were zero-mismatch donor–recipient pairs. Seventy nine percent of the recipients had PRA detected. Hypertension and diabetes were present in 86% and 44% of recipients, respectively. Delayed graft function occurred in 52% of cases. A majority of the recipients received standard sequential immunosuppressive therapy, were as follows: anti-CD25 (84%) or antilymphocytic antibody (13%), tacrolimus (92%), mycophenolate mofetil (91%) and steroids (82%).

Table 1.  Clinical characteristics of the model-development cohort and features of the pretransplant kidney biopsies, with HR of graft failure from Cox univariate analyses
 Mean ± SD or %RangeHR (95% CI)p
  1. SD = standard deviation; CI = confidence interval; CVA = cerebrovascular accident; WLR = wall-to-lumen ratio.

  2. Bold p = values signify statistical significance <0.05.

  Donor age 50 ± 152–771.03 (1.01–1.04)0.005
  Cold ischemia (h) 33 ± 103–641.21 (0.61–2.40)0.6   
  Donor terminal creatinine (mg/dL) 1.2 ± 0.50.1–3.31.73 (1.06–2.08)0.03  
  Donor history of hypertension40%1.24 (0.79–1.94)0.3   
  CVA as cause of donor death58%1.24 (0.77–1.97)0.4   
  GS (%)12 ± 60–451.04 (1.01–1.07)0.007
  Glomerular diameter (μm)237 ± 32157–3231.00 (1.00–1.01)0.06  
  PGF (any)77% 4.09 (1.65–10.14)0.002
  WLR glomerular-size arterioles 0.5 ± 0.20.2–1.70.86 (0.34–2.18)0.8   
  WLR interlobular arteries 0.6 ± 0.30.2–1.62.74 (1.26–5.97)0.01  
  Arteriolar hyalinosis (any)68%3.93 (2.02–7.64)0.0001
  Cumulative fibrosis14 ± 62.5–401.02 (0.99–1.05)0.2   
  Scar (any)80%2.58 (1.24–5.38)0.01  
  Acute tubular necrosis (any)86%1.03 (0.53–2.02)0.9   

The analysis was done 24 months after the last transplant in the study population, with a mean recipient follow-up of 38 ± 30 months. During the follow-up period 63 grafts failed, and 39 others declined in function to a GFR <20 mL/min. Kaplan–Meier graft survivals for the 371 kidney transplants in the study were 90% at 1-year, 84% at 2 years and 65% at 5 years.

Biopsy data

The mean percentage of GS read was 9 ± 6% on the frozen sections, and 12 ± 6% on permanents. Vascular sclerosis in the frozen versus permanent sections was graded as absent in 13.2% versus 7.5%, minimal in 16.4% versus 18.9%, mild in 52.6% versus 60.1%, mild-moderate in 9.4% versus 4.3% and moderate in 8.4% versus 8.9%, on the frozen versus permanent sections, respectively. Interstitial fibrosis was graded as absent in 10.5% versus 5.1%, minimal in 21.8% versus 23.7%, mild in 54.7% versus 60.6%, mild-moderate in 6.2% versus 6.5% and moderate in 6.7% versus 3.8%, on the frozen versus permanent sections, respectively. PGF was identified in 83% of the permanent sections, arteriolar hyalinosis in 72% and scar in 81%. Biopsy morphometric data are shown in Table 1. Since 60% of specimens had no arcuate/interlobar arteries, this parameter was not used in the statistical analyses.

Statistical comparison of the frozen versus permanent section readings showed that the correlation coefficients were 0.56 for percentage of GS (p < 0.001), 0.47 for interstitial fibrosis (p < 0.001) and 0.57 for vascular sclerosis (p < 0.001). Percentage of GS increased from the frozen to the permanent sections in 43% and decreased in 10% of the cases; interstitial fibrosis was upgraded in 23% and downgraded in 25%; vascular sclerosis was upgraded in 21% and downgraded in 21%. This indicates that for all the parameters there is a statistical correlation between frozen and permanent section results.

Development of the pathologic scoring system

Results of the univariate Cox analyses using the pretransplant clinical and pathologic characteristics of the model-development cohort are shown in Table 1. Clinical characteristics having statistically significant associations with graft loss were donor age and terminal creatinine. Statistically significant pathological variables were percentage of GS, presence of PGF, WLR of interlobular arteries, presence of arteriolar hyalinosis and presence of scar. Receiver operating curve analysis of the significant continuous pathologic variables indicated that reasonable cut-offs for conversion to categorical variables were 15% for GS and 0.5 for the interlobular artery WLR that corresponds to Banff cv3 lesion. The MAPI scoring system was created using the HR of graft loss, rounded to the nearest integer, from the univariate analyses for all five categorical variables (Table 2).

Evaluation of the pathologic scoring system

The pathologic scoring system developed with the model-development cohort was tested with an independent, randomly selected 112 patient model-validation cohort. Using this latter cohort, graft survival for kidneys having biopsies either with or without each component of the MAPI score were plotted. These results showed that graft survival was reduced if any individual component of the MAPI score was found on the pretransplant biopsy (all p < 0.05 except GS, Figure 2).


Figure 2. Comparison of graft survival in the model-validation cohort for biopsies with and without each component of the MAPI scoring system. The presence of any component reduced the graft survival.

Download figure to PowerPoint

Significance of GS

It was found that the presence of ≥15% of GS was associated with a 87% increased risk in the of graft loss on univariate (HR 1.87, 95% CI 1.17–2.99, p = 0.009) in the model-development group. The 5-year graft survival for GS <15% is 64% and GS ≥15% 46%, p =  0.008.

Significance of arterial sclerosis (increased WLR) and arteriolar hyalinosis

The WLR in interlobular arteries ≥0.5 increased the risk of graft loss by 105% (HR 2.05, 95% CI 1.20–3.47, p = 0.008) in the model-development group on univariate analysis. The 5-year graft survival was 73% for WLR in interlobular arteries <0.5 and 52% for WLR in interlobular arteries ≥0.5, p = 0.007.

Presence of any degree of AH carried an increased risk for graft loss by 293% (HR 3.93, 95% CI 2.02–7.64, p = <0.0001) in the model-development group on univariate analysis. The 5-year graft survival was 48% for presence and 85% for absence of arteriolar hyalinosis, p = <0.0001.

Significance of fibrosis

The presence of the more specifically defined localized fibrotic lesions such as scar and PGF, were both associated with pronounced increase in the risk of graft loss, scar with 159% (HR 2.59, 95% CI 1.24–5.38, p = 0.01) and PGF with 309% (HR 4.09, 95% CI 1.65–10.14, p = 0.002) in the model-development group on univariate analysis. The 5-year graft survival for absence of scar was 79% and 54% for the presence of scar, p = 0.008. The 5-year graft survival for absence of PGF was 82% and 54% for presence of PGF, p = 0.001.

In order to assess the relative importance of the MAPI score and the donor clinical variables, a Cox Multivariate analysis was performed, with graft loss as the independent variable. The analysis showed the MAPI score was the only variable significantly associated with graft loss in model-validation group and for every point increase in MAPI the relative risk for graft loss increases by 21% (HR 1.21, 95% CI 1.05–1.40, p = 0.008). (Table 3). The diagnostic or discrimination ability of the MAPI score for both the model-development and model-validation cohorts were assessed using receiver operating curve analysis. As shown in Figure 3, the area under curve, which is the probability of concordance between predictions and outcomes, was similar for both the model-development and model-validation cohorts (AUC = 0.70 and 0.74, respectively). This supports the conclusion that use of the MAPI scoring system in an independent population would give results similar to those found in this study.

Table 3.  Cox Multivariate analysis showing association of MAPI score and clinical parameters to risk of graft failure (model-validation cohort)
 HR (95% CI)p
  1. Bold p = value signifies statistical significance <0.05.

MAPI1.21 (1.05–1.40)0.008
Donor age1.03 (1.00–1.07)0.096
Cold ischemia (h) 3.66 (0.77–17.40)0.102
Donor history of hypertension1.62 (0.67–3.97)0.287
Donor terminal creatinine >1.5 mg/dL1.34 (0.43–4.18)0.611
CVA as cause of donor death0.98 (0.35–2.73)0.973

Figure 3. Receiver operating curve analysis of the MAPI score for predicting graft failure for both the model-development (solid black line) and model-validation (dotted black line) cohorts. The area under the curves, which is the probability of concordance between predictions and outcomes, was similar for both the model-development and model-validation cohorts (AUC = 0.70 and 0.74, respectively). This validation supports the conclusion that use of the MAPI scoring system in another independent population would give results similar to those found in this study.

Download figure to PowerPoint

As developed here the MAPI score has a range from 0 to 15 points. Using the model-development population of 259 transplants, a ROC analysis indicated that scores of 7 and 11 were sensitive threshold cut-offs for grouping MAPI scores into groups designated as low, intermediate and high risk for graft failure. Survival curves for these three risk groups are shown in Figure 4. Five-year graft survival for kidneys with MAPI scores from 0 to 7 (low risk) was 90%, with scores from 8 to 11 (intermediate risk) was 63% and from 12 to 15 was 53%.


Figure 4. Kaplan–Meier graft survival curves for the entire study population, segregated by low, intermediate and high MAPI score ranges. Graft survival for low score group (MAPI < 7) was significantly greater than the intermediate score (MAPI 8–11, p = 0.002) or high score (MAPI 12–15, p < 0.001) kidneys and graft survival for intermediate score (MAPI 8–11) was higher than high score (MAPI 12–15, p = 0.04).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

The use of marginal donors is usually met with skepticism and subject to generalization and inappropriate discard. Indeed, this analysis exemplifies how donor kidneys that have been rejected by local centers can achieve good outcomes with more specific classification. The histological evaluation of each individual kidney is a key ingredient in determining the quality of the organ and predicting graft outcome. The variability of the histological findings in marginal donors has been well documented previously (14) as well as in our current cohort where 12% of the organs had essentially normal histology but were initially rejected based on clinical criteria.

Using a large cohort of pretransplant kidney biopsies we developed and validated an aggregate scoring system, MAPI, to predict graft survival. The score includes objective pathologic findings, which were statistically associated with graft loss in univariate analysis: interlobular arterial WLR ≥0.5, GS ≥15% and the presence of PGF, arteriolar hyalinosis or scar independent of quantity. A ‘low risk’ MAPI score of seven or less was associated with excellent graft survival (90% at 5 years), while ‘intermediate risk’ and ‘high risk’ scores were associated with lower survival rates (63% and 53%, respectively).

There are several unique aspects of this study compared to prior publications on the topic. We examined a wide range of ‘chronic tissue damage’ pathologic features and identified some that have not been studied in this setting before, but that appear to be related to graft longevity. All of these features are easily recognized on preimplantation biopsy, and may be less liable to interobserver variability than standard grading systems. We believe the scoring system's simplicity and ease of calculation would allow it to be widely used if our findings were validated by others. This is also the largest series of pretransplant kidney biopsies published to date from a single institution. The large number of cases allowed us to divide the population into model-development and model-validation cohorts, and utilize the latter to confirm the predictive value of the scoring system in the same study.

Chronic pathologic features of preimplantation kidney biopsies can be grouped into those pertaining to the glomeruli, vessels and the interstitium. With respect to the glomeruli, most authors have focused on percentage of GS partly because it is simple and straightforward for nonspecialist pathologists to identify. Some studies have emphasized the importance of GS in predicting outcome, while others have not. A numeric cut-off for an unacceptable amount remains in dispute, with some authors reporting that GS over 20% has a negative impact on graft survival (15) in contrast to no impact by others (16). This study found that graft survival was lower for kidneys with over 15% GS. It is important to note that most studies on pretransplant biopsies are limited by the fact that outcomes for kidneys with high levels of GS are generally not available. Over 80% of expanded criteria kidneys with over 20% GS are discarded in the US (11). In our study only 41 of transplanted kidneys (11% of the study population) had GS over 20%. This should be kept in mind when applying scoring systems clinically, since they are usually only applicable in situations similar to the population from which they are derived. Therefore, the MAPI scoring system should not be extrapolated to biopsies with GS over 25%, since we do not yet have data to determine how well the model would fit.

The negative impact of donor vasculopathy has been stressed in some studies (17–24) although questioned by others (25). In this study, we confirmed that narrowing of both small and medium size arterial branches in the form of a high WLR, and the presence of arteriolar hyalinosis, negatively impacted graft survival. We identified a 20% difference in 5-year graft survival from our analysis when the lumen of the interlobular arteries was reduced by ≥ 50%, which is similar to the Banff cv3 lesion in model-validation group. We found that this cv3 lesion was sensitive in predicting graft outcome. Hence, we feel that vascular changes are at least as important as GS in assessing kidney quality.

The predictive value of interstitial fibrosis in biopsy specimens is controversial. Whereas some groups have correlated fibrosis with poor graft outcome (22), others have not found this association (24,26,27) One explanation for these discrepant results is the difficulty in assessing interstitial fibrosis on H&E stained sections and its marked sampling variability (28). In this study we utilized a sophisticated computerized morphometric technique for assessing global fibrosis of the biopsies. This technique automatically measured the fraction of the biopsy that stained green with Masson's trichrome. We still found that this parameter was not strongly associated with graft survival, suggesting that the problem with using interstitial fibrosis for predicting outcome is not just technical but also biologic. The discrete, localized fibrotic lesions on the other hand, scar and periglomerular had a much better predictive value than interstitial fibrosis.

While biopsy scoring systems can be more difficult for practitioners to learn and use than the more familiar assessment of commonly described pathologic features, scoring also has some advantages. For example, sampling errors can lead to poor clinical decisions based on biopsies that are not representative of the kidney's overall pathology, and may be more important when there is only a limited amount of tissue available for evaluation. Objectively derived scoring systems may partially mitigate this problem by using a combination of interrelated pathological features. Scoring systems also place differential weight on specific features based on outcomes analysis in a way that may not be intuitively obvious to practitioners. A scoring system like MAPI could simplify biopsy evaluation by eliminating the need to consider features that are relatively unimportant.

Other scoring systems for predicting kidney transplant outcome have been described. Remuzzi developed a scoring system, which assigned 0 to 3 points for the degree of chronic changes in glomeruli, vessels, tubules and interstitium. They found that use of a biopsy with a scoring system gave better outcomes than transplantation with no preimplantation biopsy, because the biopsy helped physicians decide which kidneys should be used or discarded (29). Nyberg developed a scoring system based on donor clinical parameters, which did not include histology (30). Re et al. were able to correlate their results with the findings of the Nyberg scoring system, but not with the Remuzzi scoring system and graft survival (31). Our study included both donor clinical and detailed biopsy data in the regression analyses. As opposed to the Remuzzi scoring system, the weights attributed to the MAPI histological parameters were chosen based on the actual HR associated with the risk of graft failure. The MAPI score developed in this manner correlated very well with graft survival in an independent cohort.

We found that MAPI predicted graft outcome independently of the other donor characteristics in model-validation group with an increase in relative risk of 21% for graft loss for every point increase. The clinical ECD criteria of donor age, history of donor hypertension, terminal creatinine >1.5 and cerebrovascular accident as cause of death did not independently predict the outcome in this cohort when histopathology (MAPI) was considered. Several previous studies have correlated specific ECD criteria with poor graft outcome, but those were done without considering the histopathology (32–34). It should be noted that most of the kidneys in this study were imported from other donation service areas, which contributed to the mean cold ischemic time of over 33 h, 80% of them over 24 h. We were unable to demonstrate an association of ischemic time with graft loss, possibly related to the fact that there were so few organs with short preservation times. Therefore, applicability of MAPI to kidneys with shorter cold times would need to be evaluated in another study.

In conclusion, this study demonstrates the value of histological evaluation of kidneys before implantation. MAPI, which is based on five histological characteristics of the pretransplant wedge biopsy effectively stratifies the organs into low, intermediate and high risk of graft failure. For clinical situations similar to this study population (cold ischemia over 24 h, GS <25%), this scoring system may help transplant physicians estimate outcome from the preimplantation biopsy findings.

We hypothesize how MAPI can be used to improve outcomes in marginal kidneys, one must consider the pathological features of certain immunosuppressive agents. A prominent pathologic feature of calcineurin inhibitor (CNI) toxicity is arterial hyalinosis, which results in luminal narrowing of the arteries, increased GS, and tubulointerstitial damage (35–42). These pathological changes related to CNI toxicity can ultimately lead to changes similar to chronic allograft nephropathy and graft loss. It is conceivable that suboptimal kidneys with some baseline arterial and tubulointersitial disease would fare worse with CNI-based immunosuppression given the additive pathology of these drugs. Hence, MAPI risk segregation can potentially delineate kidneys that may benefit from CNI free immunosuppression. Patients transplanted with kidneys that have intermediate or high MAPI scores can prospectively be placed on CNI free immunosuppression. This strategy allows the physician to not only select appropriate patients based on preimplantation MAPI scores, but to also tailor the immunosuppression and obtain superior long-term outcomes with the marginal kidneys.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Funding Sources: Novartis Pharmaceutical Corporation.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  • 1
    Lee CM, Scandling JD, Shen GK, Salvatierra O, Dafoe DC, Alfrey EJ. The kidneys that nobody wanted: Support for the utilization of expanded criteria donors. Transplantation 1996; 62: 18321841.
  • 2
    Hariharan S, McBride MA, Bennett LE, Cohen EP. Risk factors for renal allograft survival from older cadaver donors. Transplantation 1997; 64: 17481754.
  • 3
    Matas AJ, Gillingham KJ, Elick BA et al. Risk factors for prolonged hospitalization after kidney transplants. Clin Transplant 1997; 11: 259264.
  • 4
    Matas AJ, Gillingham KJ, Humar A, Dunn DL, Sutherland DE, Najarian JS. Immunologic and nonimmunologic factors: Different risks for cadaver and living donor transplantation. Transplantation 2000; 69: 5458.
  • 5
    McLaren AJ, Jassem W, Gray DW, Fuggle SV, Welsh KI, Morris PJ. Delayed graft function: Risk factors and the relative effects of early function and acute rejection on long-term survival in cadaveric renal transplantation. Clin Transplant 1999; 13: 266272.
  • 6
    Morris PJ, Johnson RJ, Fuggle SV, Belger MA, Briggs JD. Analysis of factors that affect outcome of primary cadaveric renal transplantation in the UK. HLA Task Force of the Kidney Advisory Group of the United Kingdom Transplant Support Service Authority (UKTSSA). Lancet 1999; 354: 11471152.
  • 7
    Ojo AO, Leichtman AB, Punch JD et al. Impact of pre-existing donor hypertension and diabetes mellitus on cadaveric renal transplant outcomes. Am J Kidney Dis 2000; 36: 153159.
  • 8
    Pfaff WW, Howard RJ, Patton PR, Adams VR, Rosen CB, Reed AI. Delayed graft function after renal transplantation. Transplantation 1998; 65: 219223.
  • 9
    Terasaki PI, Gjertson DW, Cecka JM, Takemoto S, Cho YW. Significance of the donor age effect on kidney transplants. Clin Transplant 1997; 11: 366372.
  • 10
    Tullius SG, Reutzel-Selke A, Egermann F et al. Contribution of prolonged ischemia and donor age to chronic renal allograft dysfunction. J Am Soc Nephrol 2000; 11: 13171324.
  • 11
    Sung RS, Christensen LL, Leichtman AB et al. Determinants of discard of expanded criteria donor kidneys: Impact of biopsy and machine perfusion. Am J Transplant 2008; 8: 783792.
  • 12
    Racusen LC, Solez K, Colvin RB et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: 713723.
  • 13
    Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of diet in renal disease study group. Ann Intern Med 1999; 130: 461470.
  • 14
    Papadimitriou JC, Coale R, Farney A et al. Biopsy of the marginal kidney donor: Correlation of histology with outcome. Transplant Proc 2004; 36: 742744.
  • 15
    Gaber LW, Moore LW, Alloway RR, Amiri MH, Vera SR, Gaber AO. Glomerulosclerosis as a determinant of posttransplant function of older donor renal allografts. Transplantation 1995; 60: 334339.
  • 16
    Pokorna E, Vitko S, Chadimova M, Schuck O, Ekberg H. Proportion of glomerulosclerosis in procurement wedge renal biopsy cannot alone discriminate for acceptance of marginal donors. Transplantation 2000; 69: 3643.
  • 17
    Bains JC, Sandford RM, Brook NR, Hosgood SA, Lewis GR, Nicholson ML. Comparison of renal allograft fibrosis after transplantation from heart-beating and non-heart-beating donors. Br J Surg 2005; 92: 113118.
  • 18
    Bosmans JL, Woestenburg A, Ysebaert DK et al. Fibrous intimal thickening at implantation as a risk factor for the outcome of cadaveric renal allografts. Transplantation 2000; 69: 23882394.
  • 19
    Karpinski J, Lajoie G, Cattran D et al. Outcome of kidney transplantation from high-risk donors is determined by both structure and function. Transplantation 1999; 67: 11621167.
  • 20
    Leunissen KM, Bosman FT, Nieman FH et al. Amplification of the nephrotoxic effect of cyclosporine by preexistent chronic histological lesions in the kidney. Transplantation 1989; 48: 590593.
  • 21
    Pokorna E, Vitko S, Chadimova M, Schuck O. Adverse effect of donor arteriolosclerosis on graft outcome after renal transplantation. Nephrol Dial Transplant 2000; 15: 705710.
  • 22
    Rowshani AT, Scholten EM, Bemelman F et al. No difference in degree of interstitial Sirius red-stained area in serial biopsies from area under concentration-over-time curves-guided cyclosporine versus tacrolimus-treated renal transplant recipients at one year. J Am Soc Nephrol 2006; 17: 305312.
  • 23
    Taub HC, Greenstein SM, Lerner SE, Schechner R, Tellis VA. Reassessment of the value of post-vascularization biopsy performed at renal transplantation: The effects of arteriosclerosis. J Urol 1994; 151: 575577.
  • 24
    Wang HJ, Kjellstrand CM, Cockfield SM, Solez K. On the influence of sample size on the prognostic accuracy and reproducibility of renal transplant biopsy. Nephrol Dial Transplant 1998; 13: 165172.
  • 25
    Randhawa PS, Minervini MI, Lombardero M et al. Biopsy of marginal donor kidneys: Correlation of histologic findings with graft dysfunction. Transplantation 2000; 69: 13521357.
  • 26
    Kuypers DR, Chapman JR, O'Connell PJ, Allen RD, Nankivell BJ. Predictors of renal transplant histology at three months. Transplantation 1999; 67: 12221230.
  • 27
    Seron D, Carrera M, Grino JM et al. Relationship between donor renal interstitial surface and post-transplant function. Nephrol Dial Transplant 1993; 8: 539543.
  • 28
    Lehtonen SR, Taskinen EI, Isoniemi HM. Histopathological findings in renal allografts at time of transplantation and correlation with onset of graft function. Apmis 1999; 107: 945950.
  • 29
    Remuzzi G, Grinyo J, Ruggenenti P et al. Early experience with dual kidney transplantation in adults using expanded donor crit3:14 PMTransplant Group (DKG). J Am Soc Nephrol 1999; 10: 25912598.
  • 30
    Nyberg SL, Matas AJ, Kremers WK et al. Improved scoring system to assess adult donors for cadaver renal transplantation. Am J Transplant 2003; 3: 715721.
  • 31
    Re L, Cicora F, Petroni J, Goldberg J, Rial MC, Casadei D. Comparison between clinical and histopathological scoring in cadaveric kidney transplantation and its correlation with posttransplant evolution. Transplant Proc 2006; 38: 903904.
  • 32
    Chertow GM, Milford EL, Mackenzie HS, Brenner BM. Antigen-independent determinants of cadaveric kidney transplant failure. Jama 1996; 276: 17321736.
  • 33
    Pessione F, Cohen S, Durand D et al. Multivariate analysis of donor risk factors for graft survival in kidney transplantation. Transplantation 2003; 75: 361367.
  • 34
    Port FK, Bragg-Gresham JL, Metzger RA et al. Donor characteristics associated with reduced graft survival: An approach to expanding the pool of kidney donors. Transplantation 2002; 74: 12811286.
  • 35
    Bennett WM. Insights into chronic cyclosporine nephrotoxicity. Int J Clin Pharmacol Ther 1996; 34: 515519.
  • 36
    Bertani T, Ferrazzi P, Schieppati A et al. Nature and extent of glomerular injury induced by cyclosporine in heart transplant patients. Kidney Int 1991; 40: 243250.
  • 37
    Davies DR, Bittmann I, Pardo J. Histopathology of calcineurin inhibitor-induced nephrotoxicity. Transplantation 2000; 69 (12 Suppl): SS11SS13.
  • 38
    Myers BD, Newton L. Cyclosporine-induced chronic nephropathy: An obliterative microvascular renal injury. J Am Soc Nephrol 1991; 2(2 Suppl 1): S45S52.
  • 39
    Nankivell BJ, Borrows RJ, Fung CL, O'Connell PJ, Allen RD, Chapman JR. The natural history of chronic allograft nephropathy. N Engl J Med 2003; 349: 23262333.
  • 40
    Pascual M, Swinford RD, Ingelfinger JR, Williams WW, Cosimi AB, Tolkoff-Rubin N. Chronic rejection and chronic cyclosporin toxicity in renal allografts. Immunol Today 1998; 19: 514519.
  • 41
    Remuzzi G, Perico N. Cyclosporine-induced renal dysfunction in experimental animals and humans. Kidney Int 1995; 52(Suppl): S70S74.
  • 42
    Solez K, Vincenti F, Filo RS. Histopathologic findings from 2-year protocol biopsies from a 3:14 PMticenter kidney transplant trial comparing tarolimus versus cyclosporine: A report of the FK506 Kidney Transplant Study Group. Transplantation 1998; 66: 17361740.