Kidney Allograft Fibrosis and Atrophy Early After Living Donor Transplantation

Transcription

1 American Journal of Transplantation 2005; 5: Blackwell Munksgaard Copyright C Blackwell Munksgaard 2005 doi: /j x Kidney Allograft Fibrosis and Atrophy Early After Living Donor Transplantation Fernando G. Cosio a,, Joseph P. Grande c, Timothy S. Larson a, James M. Gloor a, Jorge A. Velosa a, Stephen C. Textor a, Matthew D. Griffin a and Mark D. Stegall b a Department of Internal Medicine, Division of Nephrology and Hypertension b Department of Surgery, Division of Transplantation Surgery and c Department of Pathology, Mayo Clinic and Foundation, Rochester, Minnesota, USA Corresponding author: Fernando G. Cosio, Cosio.Fernando@mayo.edu Kidney allograft failure is most often caused by chronic allograft nephropathy, a process of interstitial fibrosis (GIF) and tubular atrophy (TA). We assessed the pathology of living donor (LD) grafts compared to deceased donor (DD). Included are 321 recipients (245 LD; 76 DD) with protocol biopsies the first 2 years of transplant. In LD, GIF was present in 7%, 31%, 61% and 71% of grafts at 0, 4, 12 and 24 months. TA progressed in parallel to GIF. Compared to LD, more DD grafts had GIF at time 0 (29%, p = 0.002); thereafter the incidence of GIF was similar. In LD, GIF was associated with lower glomerular filtration rate (GFR) 1 year (no GIF, 62 ± 16; GIF, 49 ± 15 ml/min/m 2 iothalamate clearance, p = 0.001) and reduced graft survival (HR = 2.2, p = 0.009). GIF in LD related to acute rejection (HR = 2.6, p = 0.01), polyoma nephropathy (OR = 4.4, p = 0.02) and lower levels of GFR 3 weeks post-transplant (HR = 0.961; p = 0.03, multivariate). However, GIF also developed in 53% of recipients lacking these covariates. Thus, GIF/TA develops in the majority of LD grafts, it is often mild but is associated with reduced function and survival. GIF frequently develops in the absence of risk factors. Lower GFR post-transplant identify patients at highest risk of GIF. Key words: Chronic allograft nephropathy, kidney transplant, living donors, protocol biopsies Received 20 October 2004, revised 2 December 2004 and accepted for publication 14 December 2004 Introduction Despite recent improvements in the longevity of kidney allografts, the survival of these organs is frequently shorter than that of the patient (1,2). Consequently, failure of a kidney transplant is currently one of the most common diagnosis in patients requiring renal replacement therapy. The causes of graft failure are multiple (3) and, after the first year of transplantation, death with function and chronic allograft nephropathy (CAN) are most common. The term CAN was proposed during the first Banff meeting (4) to be used instead of the term chronic rejection. CAN is a descriptive diagnosis for certain histologic changes, including interstitial fibrosis (GIF) and tubular atrophy (TA). As such, the term CAN does not have etiologic and/or pathogenic implications. Unfortunately, in most cases it is not possible to identify the specific cause of CAN and so this term frequently becomes a default diagnosis in patients who lose their allograft after the first year post-transplant. We would argue that in part this is due to the fact that CAN is usually diagnosed late in the course of the process. In this study we sought to evaluate the pathology of allografts from living donors (LD) during the first 2 years following transplantation. Several previous studies evaluated protocol biopsies principally in allografts from deceased donors (DD) (5 12). The results of these studies are internally consistent: Interstitial fibrosis frequently develops in DD kidneys during the first year post-transplant. Although GIF is most often mild it is progressive and is associated with reduced graft survival. Certain clinical events, such as acute rejection, older donor age and delayed graft function relate to more frequent and more intense GIF. Furthermore, fibrosis of DD grafts may be related to the multiple injurious events that occur prior to transplantation. In this study we reasoned that if pre-transplant injury was a major cause of allograft fibrosis this process should be milder in LD kidney than in DD kidneys. This first systematic assessment of the pathology of LD kidney allografts show significant similarities and important dissimilarities with DD grafts. GIF develops in most LD kidneys and although the fibrosis is generally mild it correlates with reduced kidney function and shorter graft survival. Assessment of correlates of LD graft fibrosis also shows differences with the factors that are associated with the fibrosis of DD grafts. Finally, the results of this analysis suggest that the events that lead to LD fibrosis occur very early post-transplant and that those events may be reflected by lower levels of function during the first weeks of transplant, prior to the development of GIF. 1130

2 Kidney Allograft Fibrosis and Atrophy Methods Patient population Included in this analysis are 321 recipients of kidney allografts transplanted from September 1998 to May All of these patients had a graft biopsy done as per protocol 1 year following the transplant. That is, these biopsies were done as part of the routine care of kidney transplant recipients at the Mayo Clinic and not motivated by changes in graft function. The majority of these individuals (N = 245, 76%) received kidneys from living donors (LD), either genetically related (N = 158, 64% of LD) or unrelated (N = 87, 36% of LD). The remaining patients (N = 76, 24%) received kidneys from deceased donors (DD). Patients who were either ABO incompatible and/or were sensitized against the donor were excluded from this study (13). Clinical and laboratory information from donors and recipients was extracted from electronic databases and from the patient s medical record. This extraction and the reporting of these data were approved by the Institutional Review Board. Graft function was measured by serum creatinine and by glomerular filtration rate (GFR) measured by non-radiolabeled iothalamate clearance (14). GFR was measured at 3 weeks after transplantation (at the time of dismissal of the patient from the early post-transplant clinic) and yearly thereafter. In some analyses we used GFR calculated by the modification of diet in renal disease (MDRD) formula (15). Percutaneous allograft biopsies were done as per protocol approximately 30 min following implantation of the graft (time 0) and at 4, 12 and 24 months post-transplant. In 128 patients biopsies were done at all time points up to 1 year and in 63 of these the 2-year biopsy was also done. All of the remaining patients had the 1-year biopsy but did not have one or more of the biopsies at the other time points. Thus, of the 321 patients with a 1-year biopsy, 159 had a time 0 biopsy, 255 had a 4-month biopsy and 63 had a 2-year biopsy. There were no statistical differences in the percentage of LD or DD kidney recipients having biopsies at the different time points. All biopsies were evaluated by routine light microscopy by one of four dedicated nephropathologists and scored using the Banff 97 classification (16). Immunofluorescence, including C4d stains, and electron microscopy were done only if clinically indicated. The diagnosis of acute rejection was based on biopsy evidence (16). The diagnosis of polyoma-bk nephropathy was based on the demonstration of viral DNA in the biopsy by in situ hybridization (17,18), which was done on biopsies that showed cytopathic epithelial changes or interstitial inflammatory infiltrates of unclear etiology. Delayed graft function, diagnosed as the need for dialysis during the first week after transplantation, occurred in 3% of LD recipients. Slow graft function, diagnosed as a reduction in serum creatinine of less than 50% during the first 72 h, occurred in 4% of LD recipients. Because of the low numbers of patients with delayed or slow graft function patients with these diagnoses were analyzed as a single group. Patients received immunosuppression consisting of induction therapy, in 85% of cases with Thymoglobulin (1.5 mg/kg/day for 5 7 days), followed by triple oral immunosuppressive therapy with prednisone, tacrolimus (or sirolimus) and mycophenolate mofetil generally at a dose of 750 mg twice daily. The dose of tacrolimus was adjusted to achieve trough levels of ng/ml (IMX whole blood assay) during the first four months posttransplant and 6 8 ng/ml thereafter. Thirty-five patients received sirolimus rather than tacrolimus. Target sirolimus levels are from ng/ml for the first four months post-transplant and ng/ml thereafter. These immunosuppressive protocols were not different in recipients of LD or DD grafts. Data analysis Data are expressed as means and standard deviation unless specified otherwise. Means of normally distributed data were compared by Student s t-test, paired t or for more than two groups by analysis of variance (ANOVA). Data that were not normally distributed were compared by non-parametric tests. Chi square was used to compare proportions. The relationships between the presence or absence of graft interstitial fibrosis (GIF) and tubular Table 1: Patient characteristics Variables LD DD p-values N Recipient age (years) 49.5 ± ± Recipient sex (% males) NS Recipient BMI 27 ± 5 27 ± 3 NS Recipient race (% Caucasians) 96% 91% NS Recipient diabetes (%) NS Dialysis pre-tx (%) Donor age (years) 44 ± ± Donor sex (% males) HLA mismatch 2.9 ± ± PRA peak 1.5 ± 9 13 ± No. of transplants (% first) NS Acute rejection [N (%)] 35 (14%) 7 (10%) NS BK nephropathy [N (%)] 24 (10%) 3 (4%) NS Student s t-test. Non-parametric t-test. Chi square. atrophy (TA) and other parameters were assessed by logistic regression. Survival was analyzed by Cox and Kaplan-Meier plots. Results Patient characteristics Table 1 displays these results. Directly relevant to these studies, it should be noted that DD were significantly younger than LD. Furthermore, the incidence of acute rejection and BK nephropathy, two major causes of GIF, was not different in LD and DD recipients. During the first year post-transplant 35 LD recipients had clinical or subclinical acute rejection episodes (14%) including 28 patients with IA or IB scores, 6 recipients with either IIA or IIB and 1 with antibody-mediated rejection. In addition, 14 LD kidney recipients (6%) had borderline rejection on biopsy. There were no significant differences in either the number of rejections or their severity between DD and LD kidney recipients. Similarly, there were no significant differences between DD and LD recipients in the percentage of protocol biopsies that showed acute rejection (overall 4% and 2% of protocol biopsies taken at 4 and 12 months, respectively). Borderline rejection changes were present in the protocol biopsies with the same frequency (4% and 2% at 4 and 12 months, respectively). Chronic pathologic changes in protocol biopsies Figure 1 displays the percentage of LD recipients who had abnormal (that is a score >0) chronic Banff 97 scores. At time 0, 7% of the donor biopsies had more than 5% graft interstitial fibrosis (ci score >0, GIF). Thereafter, the percentage of patients with GIF increased progressively to 31%, 61% and 71% on biopsies taken at 4 months, 1 year and 2 years post-transplant, respectively. The percentage of biopsies with tubular atrophy (TA, ct score) closely parallel GIF but the incidence of TA was higher than that of American Journal of Transplantation 2005; 5:

3 Cosio et al. Percent of patients Months post-transplant Figure 1: Percentage of allograft biopsies, taken from LD kidney recipients at different times post-transplant, which demonstrate abnormalities (score >0) in chronic biopsy scores. Graft interstitial fibrosis ( ), tubular atrophy ( ), arteriolar hyalinosis ( ), transplant vasculopathy ( ), transplant glomerulopathy (+ +). GIF (time 0, 25% of biopsies with TA; 4 months, 48%; 1 year, 76%; and 2 years, 88%). The incidence of vascular pathology, including arteriolar hyalinosis (ah) and chronic vasculopathy (cv), were lower although the number of patients with these abnormalities increased during the first two years post-transplant. At 1 and 2 years post-transplant 5 and 8% of patients had chronic transplant glomerulopathy, respectively. Comparing the chronic pathologic scores in DD and LD kidneys the following differences were noted: (i) At time 0, GIF was more common in DD grafts than in LD grafts (29% and 7%, p =.002) and it was also more severe (0.3 ± 0.4 and 0.1 ± 0.4, p = 0.002). In contrast, at 1 and 2 years the incidence of GIF was not different in DD and LD kidneys, (ii) Transplant glomerulopathy was more common in DD than in LD kidneys at 1 year (13% vs. 5%, p = 0.01) and at 2 years (28% and 8%, p = 0.01). The severity of the transplant glomerulopathy was also significantly worse in DD than in LD at 1 (p = 0.02) and 2 years (p = 0.04) posttransplant and (iii) arteriolar hyalinosis was more common in DD than in LD kidneys at 1 year (31% and 21%, p = 0.03) and also more severe (p = 0.04). Figure 2 illustrates the changes in the severity of GIF in LD and DD kidneys at different times post-transplant. As can be seen, at time 0 a higher percentage of DD kidneys had GIF. Thereafter, in both types of kidneys, there are progressive declines in the percentage of recipients without GIF (ci = 0) and a corresponding increase in the percentage of individuals with higher ci scores. It should also be noted that the majority of these kidneys had mild GIF (ci = 1). Table 2 displays the quantitative changes over time in chronic pathologic scores in LD kidneys. By paired analysis, there is a progressive increase in the severity of GIF Figure 2: Percentage of biopsies, taken at different times post-transplant, which showed different degrees of graft interstitial fibrosis, that is, Banff 97 scores of 0 (open bar), 1 (stippled bar), 2 (stripped bar), 3 (black bar). Top: LD kidney recipients; Bottom: DD kidney recipients. and TA overtime in LD kidney recipients. The severity of vascular and glomerular pathology also increase with time but these changes did not reach statistical significance. The development of chronic histologic changes correlated with each other although GIF correlated most closely with TA (r = 0.76, p < 0.001). The relationships between GIF and other pathologic scores were also statistically significant but weaker: GIF versus arteriolar hyalinosis (r = 0.16, p = 0.009); GIF versus chronic vasculopathy (r = 0.29, p < 0.001) and GIF versus chronic glomerulopathy (r = 0.12, p = 0.02). Relationship between chronic histologic scores, graft function and survival At 1 year, LD grafts without GIF have better function than those with GIF (GFR: 62 ± 16 vs ± 15 ml/min/m 2, p < (t-test); serum creatinine: 1.4 ± 0.3 vs. 1.7 ± 0.7 mg/dl, p = by Mann-Whitney and calculated GFR 53 ± 10 vs. 46 ± 15 ml/min/m 2,p< (t-test)). Furthermore, increasing GIF severity is associated with progressively lower GFR at 1 year (r = 0.35, p < ) and also at 2 years (r = 0.28, p = 0.002). For example, at 1 year GFR values are: ci = 0, 62 ± 16 ml/min/m 2 ; ci = 1, 1132 American Journal of Transplantation 2005; 5:

4 Kidney Allograft Fibrosis and Atrophy Table 2: Changes in chronic biopsy scores in LD kidneys with time post-transplant. For this analysis the scores of individuals grafts at different points of time post-transplant were compared with themselves by paired analysis Time (months) ah cg ci ct cv ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.7 Significantly different from time 0 value. Significantly different from 4-month value. Significantly different from 1-year value. All statistics done by non-parametric paired t-test. ah = arteriolar hyalinosis; cg = chronic glomerulopathy; ci = interstitial fibrosis; ct = tubular atrophy; cv = chronic vasculopathy. 50 ± 14; ci = 2, 47 ± 17; ci = 3, 39.5 ± 16. A similar relationship is present between more severe TA and lower GFR at 1 year (r = 0.225, p = 0.001) and at 2 years (r = 0.217, p = 0.016). These relationships are also significant in DD kidney recipients. For example, GFR at 1 year correlates with the severity of GIF (r = 0.557, p < ) and of TA (r = 0.431, p = 0.001). It is of interest to note the evolution of the GFR from 3 weeks to 2 years in LD grafts with or without GIF at 1 year (Table 3). The donor GFRs are not significantly different between these two groups of recipients. However, 3 weeks after transplantation patients later shown to have GIF have significantly lower GFR. This difference increases at 1 year and persists at 2 years. Furthermore, in patients without GIF, GFR increases from 3 weeks to 1 year (p = 0.005, paired t). In contrast, in patients later shown to have GIF, GFR declines during the same interval (p = 0.007, paired t). The follow-up period after transplantation was 33 ± 16 months. The percentage of grafts lost not due to patient death is higher in patients with GIF (no GIF: 2.2%, GIF: 8%, p = 0.05). By Cox regression, the severity of GIF correlates significantly with death-censored graft survival (HR = 2.2; p = 0.009). Figure 3 displays this latter relationship in Kaplan-Meier plots of death-censored graft survival. In this latter analysis patients with ci = 2 or greater are considered as a single group given the limited number of patients with these ci scores. The presence of GIF at 1 year does not cor- Table 3: Changes in graft function (GFR) in LD kidneys with or without GIF at 1 year post-transplant Variables No GIF GIF p-values GFR donor 100 ± ± 17 NS GFR at 3 weeks 58 ± ± GFR at 1 year 62 ± ± 15 < GFR at 2 years 60 ± ± t-test. Significantly higher GFR than at 4 months in patients without GIF (p = 0.005, paired t). Significantly lower GFR than at 4 months in patients with GIF (p = 0.007, paired t). relate with patient survival (no GIF: 2.2% of patients died vs. GIF: 5.8%). Relationship between GIF and clinical parameters Table 4 compares clinical variables in LD recipients with GIF or without GIF at 1 year post-transplant. Table 5 displays the relationships between these variables and GIF. By univariate analysis (Table 5) GIF relates to acute rejection, BK nephropathy and lower graft function at 3 weeks measured either as GFR, serum creatinine or as calculated GFR. The number of HLA mismatches have a statistically weak correlation with GIF. In contrast, the following variables do not relate significantly with the presence of GIF in LD kidneys: Donor age, race, BMI or sex; recipient age, sex, race, or BMI; history of dialysis pre-transplant; diabetes; the average CNI level during the first year post-transplant; PRA level; transplant number; history of delayed graft function or other post-transplant complications including infections, wound problems, lymphoceles and others. It is of particular interest to note in Table 4 that up to 44% of patients with GIF had no identifiable antecedent post-transplant complication. In other words, in 86 recipients of young Graft survival Months post-transplant Figure 3: Kaplan-Meier graft survival plots of LD kidneys with no graft interstitial fibrosis: GIF ( ), mild GIF ( ) ormoderate to severe GIF ( ) (p = by log rank). 60 American Journal of Transplantation 2005; 5:

5 Cosio et al. Table 4: Clinical variables in LD kidney recipients with or without GIF at 1 year Variables No GIF GIF p-values N(%) 93 (39) 147 (61) Donor age (years) 43 ± ± 12 NS Donor age >45 years (%) NS Delayed graft function (%) NS Acute rejection # (%) Acute rejection (+ borderline) # (%) Number of acute 0.11 ± ± rejection episodes HLA mismatches 2.6 ± ± 1.6 NS PRA peak 0.35 ± ± 12 NS BK nephropathy (%) No complications (%) NS Tacrolimus level (ng/ml, average first year) 8.9 ± ± 4 NS Chi square. t-test. Mann-Whitney non-parametric test. Percentage of patients without identifiable complications following transplantation. # Includes all biopsy-proven episodes of acute rejection that occurred during the first year. (<45 years old) LD kidneys who had neither acute rejection nor BK nephropathy GIF developed in 45 (53%). The presence of GIF at 1 year related to certain histologic findings on the 4-month biopsy including: ct score (OR = 2.1, p = 0.002); ci score (OR = 4.7, p < ) and cv score (OR = 1.71, p = 0.03). Table 5: Relationship between clinical variables and GIF at 1 year in LD kidney recipients. Values represent results of univariate and multivariate logistic regression analyses comparing patients with ci scores of 0 with those with scores >0 Univariate analysis Multivariate analysis Variables OR p-values OR p-values Donor age NS NS Donor GFR pre-tx NS NS Acute rejection BK nephropathy HLA mismatches NS GFR 3 weeks post-tx Creatinine < weeks post-tx Calculated GFR 3 weeks post-tx < Multivariate analysis including acute rejection, BK and GFR. Multivariate analysis including acute rejection, BK and serum creatinine. Multivariate analysis including acute rejection, BK and calculated GFR (15). Figure 4: Incidence of GIF in LD kidney biopsies taken at time 0 (open bars), 4 months (stripped bars) and 1 year (black bars) after transplantation in recipients classified according to their GFR measured 3 weeks after transplantation (X axis). Grafts with lower levels of function 3 weeks posttransplant have a higher incidence and more severe GIF at 1 year. Furthermore, the relationship between graft function and GIF is independent of other variables (Table 5). Figure 4 displays the incidence of GIF at different times post-transplant in LD kidney recipients classified according to their GFR at 3 weeks. As can be seen, at time 0 biopsies all groups of patients have a similar incidence of GIF. However, at 4 months and at 1 year more grafts have GIF and that increase is particularly notable in those patients with the lower levels of function. Compared to LD kidney recipients without rejection, those with rejection have GIF more commonly (56% and 81%, p = 0.001) and the fibrosis is more severe (ci scores, 0.73 ± 0.77 and 1.15 ± 0.77, p = 0.001). Of interest, the severity of GIF is not different in patients with borderline rejection and those with histologically more severe rejection (1.0 ± 0.67 and 1.2 ± 0.8, respectively, NS). For this latter analysis grades of rejection higher than borderline are analyzed as a group given the limited numbers of patients within each group. Discussion The results of this study show that GIF and TA develop in the majority of LD kidney grafts during the first two years of transplant. Somewhat surprisingly, at 1 year, the incidence and severity of GIF was similar in LD and in DD kidneys, challenging the postulate that GIF is largely due to kidney injury that occurs during donor death and during prolonged preservation. Consistent with previous studies these analyses showed that acute rejection and BK nephropathy are associated with more severe GIF. However, the variables that correlate with the development of GIF/TA were absent in approximately half of the patients who develop fibrosis. Furthermore, in this study we were unable to find a relationship between CNI use or levels and the development 1134 American Journal of Transplantation 2005; 5:

6 Kidney Allograft Fibrosis and Atrophy of fibrosis. However, these results do not rule out a participation of these drugs in the fibrogenic process. In fact, recent prospective studies indicate that avoiding CNI is associated with a lower incidence of interstitial fibrosis (19). Although the incidence of GIF and TA are similar in LD and DD kidney at 1 year there are notable differences in the histology of these allografts. First, despite the fact that LD donors were older than DD donors, the incidence of GIF at time 0 was higher in the latter. This likely reflects the selection process of LD that excludes individuals with kidney functional abnormalities. Furthermore, compared to LD kidneys, DD kidneys are more likely to develop arteriolar hyalinosis and transplant glomerulopathy. The higher incidence of transplant glomerulopathy in DD kidneys may relate to the higher PRA levels found in those recipients (20). The severity of GIF and TA demonstrated in LD kidneys is most often mild, that is, a Banff score of 1, indicating that the percentage of the renal cortex occupied by fibrous tissue is between 6 and 25% (4). It should be noted that the Banff scores are based on a subjective estimate of the amount of fibrosis in the kidney. Accordingly, it is not surprising that there is significant variability in scoring by different pathologists (21). Despite this consideration the histologically mild GIF that develops in kidney allografts during the first year is associated with changes in function and prognosis and thus it deserves close scrutiny from both the clinical and research point of views. Still, more precise estimates of interstitial fibrosis would be helpful in further defining the severity, cause and consequences of GIF (22,23). Previous studies on protocol biopsies of DD allografts agree that interstitial fibrosis correlates with function and with graft survival (5 12). These results are consistent with those observations. In addition, this analysis shows that GIF/TA develops more frequently and it is more severe in patients with lower levels of GFR at 3 weeks posttransplant. This observation is consistent with the postulate that the graft injury that leads to lower levels of function and eventually to fibrosis occurs very early during transplantation. Although our protocol does not include biopsies at 3 weeks, examination of 4- and 12-month biopsies indicate that fibrosis progresses throughout the first year (see Figure 4) particularly in individuals with lower GFR. It would be inappropriate to interpret this finding as indicating that early graft dysfunction leads to GIF/TA because, the levels of GFR associated with an increased risk of fibrosis are generally considered as very good graft function. The relationship between GFR and risk of GIF was shown in LD and in DD kidneys and either including or excluding patients with delayed graft function. This study shows other associations between graft function and fibrosis. Thus, in contrast to grafts without GIF, those with GIF at 1 year not only fail to increase their GFR during the first year but lose function during that interval. Whether this is an effect of the initial insult that causes the relatively low GFR or whether this is a consequence of the developing fibrosis cannot be clarified by these data. Additional injurious events may come into play during the first year post-transplant and greatly magnify the development of GIF/TA. Those events may involve inflammation (e.g. rejection, BK nephropathy), drug toxicity (e.g. calcineurin inhibitors) and others. These results highlight the increasing relevance of polyoma-bk nephropathy in kidney transplantation and suggests the need for aggressive approaches for the detection and treatment of this infection because early diagnosis is associated with better prognosis (18). These results suggest that the mild GIF that develops early posttransplant may be progressive. However, it is likely that the rates of progression are variable among patients. Unfortunately, aside from the presence of persistent inflammation (24), we know little of the variables associated with progressive fibrosis. Recent studies suggest that the striking reduction in the incidence of acute rejection achieved in recent years has not been followed by an equally significant improvement in long-term graft survival (2). The results of this study suggest that the recent advances in immunosuppression have not controlled the development of kidney allograft fibrosis and atrophy thus may not improve graft survival much further. The identification of markers of fibrosis risk and the use of protocol biopsies may allow us to address this issue directly and early enough after transplantation when remedies may be effective. Of interest, up to 40% of transplant recipients do not develop GIF or TA during the first year post-transplant. Thus, the process of graft fibrosis and atrophy is by no means inevitable. Acknowledgments We thank Ms. Cynthia Handberg for excellent secretarial assistance. This work was supported in part by grants from the Mayo Clinic Nephrology Division. Results of this study were presented in part at the American Transplant Congress, Boston, MA, May, References 1. Hariharan S, Johnson CP, Breshnahan BA, Taranto SE, McIntosh MJ, Stablein D. Improved graft survival after renal transplantation in the United States, 1988 to N Engl J Med 2000; 342: Meier-Kriesche HU, Schold JD, Srinivas TR, Kaplan B. Lack of improvement in renal allograft survival despite a marked decrease in acute rejection rates over the most recent era. Am J Transplant 2004; 4: Matas AJ, Humar A, Gillingham KJ et al. Five preventable causes of kidney graft loss in the 1990s: a single-center analysis. Kidney Int 2002; 62: Solez K, Axelsen RA, Benediktsson H et al. International standardization of criteria for the histologic diagnosis of renal allograft rejection: the Banff working classification of kidney transplant pathology. Kidney Int 1993; 44: American Journal of Transplantation 2005; 5:

7 Cosio et al. 5. Isoniemi H, Taskinen E, Hayry P. Histological chronic allograft damage index accurately predicts chronic renal allograft rejection. Transplantation 1994; 58: Rush DN, Jeffery JR, Gough J. Sequential protocol biopsies in renal transplant patients. Transplantation 1995; 59: Nicholson ML, McCulloch TA, Harper SJ et al. Early measurement of interstitial fibrosis predicts long-term renal function and graft survival in renal transplantation. Br J Surg 1996; Seron D, Moreso F, Bover J et al. Early protocol renal allograft biopsies and graft outcome. Kidney Int 1997; Solez K, Vincenti F, Filo RS. Histopathologic findings from 2-year protocol biopsies from a U.S. multicenter kidney transplant trial comparing tacrolimus versus cyclosporine. Transplantation 1998; 66: Nickerson P, Jeffery J, Gough J et al. Identification of clinical and histopathologic risk factors for diminished renal function 2 years posttransplant. J Am Soc Nephrol 1998; 9: Kuypers DRJ, Chapman JR, O Connell PJ, Allen RDM, Nankivell BJ. Predictors of renal transplant histology at three months. Transplantation 1999; 67: 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: Gloor JM, DeGoey SR, Pineda AA et al. Overcoming a positive crossmatch in living-donor kidney transplantation. Am J Transplant 2003; 3: Wilson DM, Bergert JH, Larson TS, Liedtke RR. GFR determined by nonradiolabeled iothalamate using capillary electrophoresis. Am J Kidney Dis 1997; 30: Levey AS, Greene T, Schluchter MD et al. Glomerular filtration rate measurements in clinical trials. Modification of Diet in Renal Disease Study Group and the Diabetes Control and Complications Trial Research Group. J Am Soc Nephrol 1993; 4: Racusen LC, Solez K, Colvin RB et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: Nickeleit V, Hirsch HH, Binet IF et al. Polyomavirus infection of renal allograft recipients: from latent infection to manifest disease. J Am Soc Nephrol 1999; 10: Buehrig CK, Lager DJ, Stegall M et al. The influence of surveillance renal allograft biopsy on diagnosis and prognosis of Polyomavirus-associated nephropathy. Kidney Int 2003; 64: Flechner SM, Kurian SM, Solez K et al. De novo kidney transplantation without use of calcineurin inhibitors preserves renal structure and function at two years. Am J Transplant 2004; 4: Sijpkens YW, Joosten SA, Wong MC et al. Immunologic risk factors and glomerular C4d deposits in chronic transplant glomerulopathy. Kidney Int 2004; 65: Furness PN, Taub N. International variation in the interpretation of renal transplant biopsies: Report of the CERTPAP Project. Kidney Int 2001; 60: Grimm PC, Nickerson P, Gough J et al. Computerized image analysis of Sirius Red-stained renal allograft biopsies as a surrogate marker to predict long-term allograft function. J Am Soc Nephrol 2003; 14: Diaz Encarnacion MM, Griffin MD, Slezak JM et al. Correlation of quantitative digital image analysis with the glomerular filtration rate in chronic allograft nephropathy. Am J Transplant 2004; 4: Shishido S, Asanuma H, Nakai H et al. The impact of repeated subclinical acute rejection on the progression of chronic allograft nephropathy. J Am Soc Nephrol 2003; 14: American Journal of Transplantation 2005; 5: