SEARCH

SEARCH BY CITATION

Keywords:

  • Acute rejection;
  • interstitial fibrosis and tubular atrophy;
  • kidney transplantation;
  • protocol kidney biopsy;
  • rapid steroid withdrawal

Abstract

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

Our aim was to study the impact of subclinical inflammation on the development of interstitial fibrosis and tubular atrophy (IF/TA) on a 1-year protocol biopsy in patients on rapid steroid withdrawal (RSW). A total of 256 patients were classified based on protocol biopsy findings at months 1 or 4. Group 1 is 172 patients with no inflammation, group 2 is 50 patients with subclinical inflammation (SCI), group 3 is 19 patients with subclinical acute rejection (SAR) and group 4 is 15 patients with clinical acute rejection (CAR). On the 1-year biopsy, more patients in group 2 (SCI) (34%, p = 0.004) and group 3 (SAR) (53%, p = 0.0002), had an IF/TA score > 2 compared to group 1 (control) (15%). IF/TA was not increased in group 4 (CAR) (20%). The percent with IF/TA score > 2 and interstitial inflammation (Banff i score > 0) was higher in group 2 (16%, p = 0.004) and group 3 (37%, p < 0.0001) compared to group 1 (3%). In a multivariate analysis, patients in groups 2 or 3 had a higher risk of IF/TA score > 2 on the 1-year biopsy (OR 6.62, 95% CI 2.68–16.3). We conclude that SCI and SAR increase the risk of developing IF/TA in patient on RSW.


Introduction

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

The steady advances in immunosuppression and immune monitoring of kidney transplant recipients over the past several years has resulted in a remarkable decrease in the risk of acute rejection, but long-term graft survival remains unchanged (1). Although several factors are implicated in long-term graft loss, progressive interstitial fibrosis and tubular atrophy (IF/TA) continues to be a significant factor (2).

The etiology of IF/TA is related to a myriad of immunologic and nonimmunologic factors (2,3). Studies of kidney transplant recipients have shown that subclinical acute rejection (SAR) is associated with an increased risk of graft fibrosis (4–6) and graft loss (7,8). In addition, subclinical inflammation occurring in association with IF/TA, even when it is below the threshold for the Banff criteria for acute rejection, is associated with inferior graft survival (7,9). IF/TA on a 1-year biopsy has been associated with inferior allograft survival (4) and may be used as a surrogate marker of long-term graft survival. All of these studies included patients on steroid-based immunosuppression.

Rapid steroid withdrawal (RSW) has been widely studied in kidney transplant recipients and found to be safe and effective (10,11). Recently, a randomized controlled trial of RSW compared to conventional chronic steroid administration demonstrated equivalent patient and graft survival and glomerular filtration rate (GFR) at 5 years (12).

The aim of this study was to look at the impact of early subclinical inflammation and SAR occurring on either 1- or 4-month protocol biopsy on the development of interstitial fibrosis and tubular atrophy (IF/TA score > 2) on a 1-year protocol biopsy in a cohort of kidney transplant recipients treated with RSW.

Materials and Methods

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

This study was approved by the Mayo Foundation Institutional Review Board. This is a retrospective study of all kidney transplant recipients treated with a RSW protocol who were transplanted at our center between July 2003 and June 2008. Multiorgan transplant recipients and patients with pretransplant donor-specific antibody (DSA) or positive flow cross-match were excluded. After transplant, patients were routinely tested for BK viremia at months 1, 4 and annually. A diagnosis of BK nephropathy required BK viremia and a biopsy demonstrating BK by in situ hybridization. Protocol biopsies were done at months 1, 4 and annually. Biopsies were performed in the outpatient radiology department with real-time ultrasound guidance using an 18 g BioPince biopsy gun (Angiotech Medical Device Technologies, Gainesville, FL) to obtain at least one core. Protocol biopsies were interpreted using the Banff ’05 criteria (13). Beginning in 2004, C4d staining was routinely done on the 1-month biopsies and beginning in 2006, C4d was done on all biopsies. A positive C4d required diffuse (greater than 50%) peritubular staining by either immunoflorescence or immunohisotochemical techniques.

All patients with both an adequate 1-year biopsy and at least one adequate biopsy done at month 1 or 4 posttransplant are the subjects of this study. To be included the biopsy had to include a minimum of seven glomeruli and one artery. SAR was defined as Banff IA (at least i2, t2) or greater on protocol biopsy and a serum creatinine within 0.3 mg/dL of the previous baseline. Subclinical inflammation (SCI) was defined as Banff interstitial inflammation score and/or Banff tubulitis score of at least 1, but below the threshold for Banff IA rejection (Banff i2, t2). If a patient had a biopsy at both 1 and 4 months, they were classified based on the biopsy with the highest Banff classification for tubular and interstitial inflammation. Using these criteria, the patients were classified into four groups:

  • • 
    Group 1 control group: no inflammation or rejection at 1 or 4 months
  • • 
    Group 2 SCI (Banff i ≥ 1 and/or t ≥ 1) but not meeting the criteria for Banff IA acute rejection on either 1- or 4-month biopsy
  • • 
    Group 3 SAR (meeting criteria for at least Banff IA acute rejection- Banff i2, t2) on either 1- or 4-month biopsy
  • • 
    Group 4 biopsy for cause showing clinical acute rejection (CAR) anytime during the first posttransplant year

The four groups are mutually exclusive. If a patient developed CAR during the first year posttransplant the patient was classified into group 4 regardless of the findings on the 1- and 4-month protocol biopsies.

Immunosuppression protocol

All patients received induction with either rabbit antithymocyte globulin (r-ATG) (6 mg/kg, usually administered in four divided doses), basiliximab (total 40 mg, given IV in two divided doses), or alemtuzumab (30 mg given SQ in a single dose). Patients who were greater than 70 years of age or with a history of prior high risk malignancy or infections received basiliximab while the remainder received r-ATG, until January 2008 when we substituted alemtuzumab for the r-ATG. Tacrolimus was started when the serum creatinine dropped by at least 30%, or by postoperative day 4. The goal for trough tacrolimus levels was 10–12 ng/mL for the first 30 days, 8–10 ng/mL between days 30 and 90 and 5–8 ng/mL after 90 days. Mycophenolate mofetil (MMF) was started at 2000 mg/day, in divided doses, and adjusted according to the individual patient's tolerance. The protocol for corticosteroid use was: methylprednisolone (MP) 500 mg IV on postoperative day (POD) 0, MP 250 mg IV POD 1, MP 125 mg IV on POD 2, prednisone (P) 60 mg orally on POD 3 and P 30 mg orally on POD 4. SAR episodes were treated with a single dose of MP 500 mg and steroids were not routinely continued, while for CAR episodes were treated with three consecutive daily doses of MP 500 mg and prednisone was continued with a tapering dose to a maintenance dose of 5 mg/day.

Statistical analysis

Comparison of the characteristics between the four groups was done using analysis of variance (ANOVA) for continuous variables and chi-square for categorical data. Groups 2 (SCI), 3 (SAR) and 4 (CAR) were each compared to the control group (group 1) separately for differences. A p-value of less than or equal to 0.05 was considered statistically significant. The primary outcome was the development of an IF/TA scores greater than 2 on the 1-year protocol biopsy. The IF/TA score is the sum of Banff chronic interstitial fibrosis (ci) score plus the tubular atrophy (ct) score (13). Univariate and multivariate logistic regression analysis was used to determine the risk factors conferred on the development of IF/TA score greater than 2 on the 1-year biopsy.

Results

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

Between July 2003 and June 2008, there were 456 consecutive patients transplanted with a kidney alone at our center and who were treated with the rapid steroid withdrawal protocol. 256 of the 456 patients (56%) met the criteria of having an adequate 1-year biopsy and at least one protocol biopsy at either 1- or 4-month posttransplant. A total of 268 of the 456 patients (59%) had a protocol biopsy done at 1-year and 256 of these biopsies (96%) were adequate for study (a minimum of 7 glomeruli and one artery). The adequacy of the biopsies done at 1-year varied between 83% and 98% but was not significantly different between the four groups. For all the 1-year biopsies included in the study, the median number of glomeruli was 17 and median number of arteries was 4 and these were not significantly different between the four groups. The reasons for not obtaining the 1-year biopsies for the remaining patients included: patient declined 36%, patient not available 24%, anticoagulation 10%, patient died or lost the graft 8% and others 22%.

When the group with a 1-year biopsy (n = 268) is compared to the group who did not have a biopsy (n = 188), the patients in the biopsy group were more likely to be female (42% vs. 29%, p < 0.05) and less likely to be on steroids at the 1-year follow-up time point (11% vs. 23%, p < 0.05). Otherwise, the groups were similar including age at transplant, donor source, donor age, induction agent, HLA mismatching, acute rejection and serum creatinine and calculated GFR at 1-year follow-up.

When all 456 patients were included and are classified based on the earlier protocol biopsy findings the 1-year biopsy rate was not significantly different between the groups.

A comparison of the baseline characteristics for the four patient groups is shown in Table 1. There were 172 (67%) in group 1 (control), 50 (20%) in group 2 (SCI), 19 (7%) in group 3 (SAR) and 15 (6%) in group 4 (CAR). In group 2 (SCI), 2 of 50 (4%) had Banff t = 0, while the remainder were Banff t 1 or 2 and 20 (40%) had i = 0 while the remainder were i 1 to 3 on the classification biopsy done at either 1 or 4 months. Compared to the control group, the mean number of HLA mismatches was higher in group 4, but not for the other two groups. Otherwise, there were no significant differences in the demographics or transplant characteristics between the 4 groups including the donor source or donor age. Patients with a positive pretransplant flow cross-match or a DSA were excluded; therefore, the fraction of patients with a pervious transplant or with panel of reactive antibodies (PRA) greater than 0 in each group was very small.

Table 1.  Baseline patient and transplant characteristics
 Group 1, control (n = 172)Group 2, SCI only (n = 50)Group 3, SAR (n = 19)Group 4, CAR (n = 15)
  1. SCI = subclinical inflammation, SAR = subclinical acute rejection, CAR = clinical acute rejection, ECD = extended criteria donor, CIT = cold ischemia time, DGF = delayed graft function, PRA = panel reactive antibody.

  2. 1Continuous variables are shown as a mean ± standard deviation.

  3. 2CIT applies only to deceased donor kidneys only.

  4. * p < 0.05 for group 1 versus 4.

% of total67%20% 7% 6%
Age in years153.5 ± 14.051.9 ± 12.949.2 ± 15.550.2 ± 12.9
Recipient female40%36%42%47%
Recipient Race African-American8%2%11%13%
Diabetes pretransplant33%22%47%40%
Weight in kg181.6 ± 17.781.0 ± 20.982.8 ± 24.679.9 ± 19.1
Preemptive Transplant32%32%26% 7%
Deceased donor40%38%47%33%
ECD7%10%00
Donor age in years141.3 ± 13.443.4 ± 16.939.6 ± 14.637.5 ± 10.9
CIT in hours1,215.6 ± 6.6 14.1 ± 7.7 14.2 ± 4.7 12.9 ± 7.0 
DGF11%10%21%13%
Previous Tx (n)1200
Basiliximab induction18%16%32%33%
HLA mismatch13.2 ± 1.83.3 ± 1.83.2 ± 1.9 4.3 ± 1.2*
PRA > 08%6%11%13%

All patients received induction therapy. More patients received induction with basiliximab in groups 3 (32%) and 4 (33%) compared to the control group (18%) but these were not significantly different (Table 1).

Other indicators of immunosuppression management during the first year were similar for the groups (Table 2). There were no differences in the mean trough tacrolimus levels or mean daily dose of MMF at 1, 4 or 12 months after transplant between the 4 groups. Most patients were on tacrolimus as the primary immunosuppression agent at all time point.

Table 2.  Indicators of immunosuppression during the first posttransplant year
 Group 1, control (n = 172)Group 2, SCI only (n = 50)Group 3, SAR (n = 19)Group 4, CAR (n = 15)
  1. SCI = subclinical inflammation, SAR = subclinical acute rejection, CAR = clinical acute rejection, MMF = mycophenolate mofetil.

  2. 1Continuous variables are shown as a mean ± standard deviation.

  3. *p < 0.001 for group 1 versus 4.

Tacrolimus
 On tacrolimus at 1 month100%100%100%100%
 Tacrolimus level in ng/mL at 1 month110.8 ± 2.910.5 ± 3.110.4 ± 2.610.3 ± 3.3
 On tacrolimus at 4 months95%86%95%93%
 Tacrolimus level in ng/mL at 4 months17.74 ± 2.71 7.20 ± 2.867.89 ± 4.57.58 ± 3.18
 On tacrolimus at 1 year97%90%95%100%
 Tacrolimus level in ng/mL at 1 year17.67 ± 3.39 7.21 ± 2.82 8.31 ± 4.327.86 ± 3.32
Mycophenolate mofetil
 On MMF at 1 month100%100%100%93%
 MMF dose in mg/day at 1 month11750 ± 3581710 ± 3791605 ± 3941786 ± 378
 On MMF at 4 months97%92%95%87%
 MMF dose in mg/day at 4 months11464 ± 4541386 ± 4911389 ± 5021500 ± 456
 On MMF at 1 year92%86%95%87%
 MMF dose in mg/day at 1 year11261 ± 4561221 ± 4791306 ± 4581385 ± 416
Corticosteroids
 On steroids at 1 year6%16%11%47%*

A total of 24 of 256 patients (9%) were on steroids at the 1-year follow-up time point. The indications for the initiation of steroids was the occurrence if earlier acute rejection eight patients (all seven from group four who were on steroids), leucopenia in eight and other indications in eight patients. The fraction of patients on steroids at 1 year was significantly higher in group 4 (47%) (CAR) compared to 6% in the control group.

Protocol biopsy findings (Figure 1andTables 3and4)

image

Figure 1. The percent of patients with an IF/TA score greater than 2 with or without subclinical inflammation on the 1-year protocol biopsy in each group classified by the protocol biopsies findings at month 1 or 4. Group 1 is the control group with no inflammation, group 2 is patients with subclinical inflammation (SCI), but not meeting criteria for Banff IA acute rejection, group 3 is patients with subclinical acute rejection (SAR) and group 4 is patients with clinical acute rejection (CAR). The numbers in each column represent the number of 1-year biopsies with each particular finding. A greater percent of patients had a higher IF/TA score on the 1-year protocol biopsy in group 2 (SCI) (34%, p = 0.004) and group 3 (SAR) (53%, p = 0.0002) compared to group 1 (control) (15%), while the fraction of patients in group 4 (CAR) was not increased (20%). In addition, the fraction of biopsies with the combination of an IFTA score > 2 and interstitial inflammation (Banff i score > 0) was higher in group 2 (16%, p = 0.004) and group 3 (37%, p < 0.0001) compared to the control group (3%), but the difference between groups 2 and 3 is not significant.

Download figure to PowerPoint

Table 3.  Mean Banff Scores for protocol biopsies at 1 month, 4 months and 1 year
Banff component score1Group 1 controlGroup 2 SCIGroup 3 SARGroup 4 CAR
  1. 1Banff component scores: i = interstitial inflammation, t = tubulitis, ci = interstitial fibrosis, ct = tubular atrophy, cv = fibrous intimal thickening, cg = allograft glomerulopathy.

  2. * p < 0.05 compared to group 1.

  3. ∧ p < 0.05 comparing group 2 and 3.

  4. # p < 0.05 comparing group 3 and 4.

  5. & p < 0.05 comparing group 2 and 4.

One-month biopsy (n = 233)
 i00.35*&0.29*0.69*&
 t00.48*0.50*0.77*
 Ci0.470.77*0.710.46
 Ct0.901.020.930.92
 Cv0.580.530.690.85
 Cg0000
Four-month biopsy (n = 222)
 i00.51*∧1.00*∧#0.46*#
 t00.79*0.94*#0.54*#
 Ci0.641.25*1.000.83
 Ct0.961.43*∧&1.00∧0.75&
 Cv0.500.640.410.33
 Cg0000
One-year biopsy (n = 256)
 i0.080.26*,∧0.84*,∧,#0.33*,#
 t0.110.36*0.63*0.60*
 Ci0.931.141.42*0.93
 Ct1.121.36*1.53*1.13
 Cv0.590.600.260.57
 Cg0.020.02∧0.16*,∧0.07
Table 4.  Selected findings on protocol biopsies done at 1 month and 1 year
 Group 1, control (n = 172)Group 2, SCI only (n = 50)Group 3, SAR (n = 19)Group 4, CAR (n = 15)
  1. IF/TA-sum of Banff interstitial fibrosis (ci) and tubular atrophy (ct) scores, SCI-subclinical inflammation, SAR = subclinical acute rejection, CAR = clinical acute rejection.

  2. *p < 0.05 for group 1 versus 2.

  3. # p < 0.001 for group 1 versus 2.

  4. ∧ p < 0.05 for group 1 versus 3.

  5. **p < 0.001 for group 1 versus 3.

  6. & p < 0.001 for group 1 versus 4.

One-month biopsy findings
 IF/TA score > 2 (n = 226)7%13%7%0
 + C4d (n = 216)1% (1/147)07% (1/14)8% (1/12)
One-year biopsy findings
 Banff i score >06%22%*47%**20%
 Banff t score > 08%30%#42%** 47%&
 %IF/TA > 214% 34%*53%**20%
 % of total with both IF/TA > 2 and i > 03%16%*37%**13%
 Banff cg score > 02%2%16%∧7%
 +C4d (n = 111)08%36%**0

A comparison of the selective mean Banff scores for the protocol biopsies done at 1 month; 4 months and 1 year are shown in Table 3. In general, the Banff i score (interstitial inflammation) and t scores (tubulitis) were higher in groups 2, 3 and 4 at all time points. In addition, at 1 year the Banff ci (interstitial fibrosis) was higher in group 2 and the ct scores (tubular atrophy) were higher in groups 2 and 3 compared to the control group, while the ci and ct scores were not increased in group 4.

The primary outcome for this study is the proportion of patients with an IF/TA scores > 2 with or without SCI on the 1-year biopsy (Table 4 and Figure 1). A significant greater percent of patients had a higher IF/TA score on the 1-year protocol biopsy in group 2 (SCI) (34%, p = 0.004) and group 3 (SAR) (53%, p = 0.0002) compared to group 1 (control) (15%), while the fraction of patients in group 4 (CAR) was not increased (20%). In addition, the fraction of biopsies with the combination of an IF/TA score > 2 and interstitial inflammation (Banff i score > 0) was higher in group 2 (16%, p = 0.004) and group 3 (37%, p < 0.0001) compared to the control group (3%), but the difference between groups 2 and 3 is not significant.

The percent of patients with an IF/TA score > 2 on the 1-month biopsy was not significantly different between the four groups. The fraction with diffusely positive C4d staining on the 1-month biopsy was very low and not different between the four groups (Table 4).

In addition to the IF/TA score, there were some other notable differences in the 1-year biopsy findings between the four groups (Table 4). The percent of patients with a Banff i score (interstitial inflammation) > 0 was significantly higher in group 2 (SCI) (22%) and group 3 (SAR) (47%) when compared to the control group. The percent with tubulitis (Banff t score > 0) on the 1-year biopsy was also significantly increased in groups 2 (30%), 3 (42%) as well as in group 4 (47%). The percent of patients with allograft glomerulopathy (Banff cg score > 0) was low in the four groups, but it was higher in group 3 (SAR) (16% vs. 2% in control group, p = 0.009). In addition, there was a higher percentage of biopsies with positive peritubular C4d staining in group 3 (SAR) (36% vs. 0 in control group, p < 0.001).

BK nephropathy and GFR outcomes (Table 5)

Table 5.  BK nephropathy and GFR outcomes after transplantation
 Group 1, control (n = 172)Group 2, SCI only (n = 50)Group 3, SAR (n = 19)Group 4, CAR (n = 15)
  1. SCI = subclinical inflammation, SAR = subclinical acute rejection, CAR = clinical acute rejection, MDRD = Modification of diet in renal disease 4 variable for estimating glomerular filtration rate (GFR).

  2. 1 Continuous variables are shown as a mean ± standard deviation.

  3. * p < 0.05.

  4. ** p = 0.05 for group 1 versus 2.

Creatinine in mg/dL at 1 year 11.32 ± 0.381.45 ± 0.531.27 ± 0.311.28 ± 0.52
MDRD GFR in mL/min/1.73 m2 at 1 year 159.6 ± 16.755.4 ± 18.963.6 ± 22.063.4 ± 17.2
Creatinine in mg/dL at 2 years 11.34 ± 0.56 1.63 ± 1.35* 2.52 ± 2.89*1.50 ± 0.67
MDRD GFR in mL/min/1.73 m2 at 2 years 159.9 ± 19.152.1 ± 16.348.7 ± 23.357.8 ± 27.5
Creatinine in mg/dL at 3 years 11.28 ± 0.471.36 ± 0.441.67 ± 0.571.18 ± 0.13
MDRD GFR in mL/min/1.73 m2 at 3 years 159.8 ± 19.254.0 ± 19.147.0 ± 18.361.4 ± 12.1
Prevalence of BK nephropathy4%12%**00

The median follow-up for the entire cohort was 2.7 years and it was not different between the groups (data not shown). For the entire study cohort, the actuarial graft survival, including death with a functioning graft, was 98% at 3 years and 92% at 5 years after transplantation and the survival did not differ between the groups (data not shown). BK nephropathy occurred in 12% of group 2 (SCI) (p = 0.05 compared to group 1), but in none of groups 3 (SAR) or 4 (CAR) and 4% of patients in group 1 (Table 5). At 1 year after the transplant there were no differences in the serum creatinine or the GFR estimated by the modification of diet in renal disease (MDRD) formula (14) between the groups. At 2-year follow-up the serum creatinine was higher in groups 2 (SCI) and 3 (SAR), but the creatinine at 3 years and the MDRD GFR at both 2- and 3-year follow-up were not different between the patient groups (Table 5).

Risk factor analysis for the occurrence of IF/TA (Table 6)

Table 6.  Logistic regression analysis of factors associated with IF/TA score > 2 on the 1-year protocol biopsy
 Univariate analysisMultivariate analysis1
Odds ratio95% CIpOdds ratio95% CIp
  1. ECD = extended criteria donor, MMF = mycophenolate mofetil, IF/TA = sum of Banff interstitial fibrosis (ci) and tubular atrophy (ct) scores, SCI = subclinical inflammation, SAR = subclinical acute rejection, CAR = clinical acute rejection, Bx = protocol biopsy.

  2. 1Multivariate analysis adjusted for recipient age, recipient weight in kg, donor age, deceased donor source, delayed graft function, total daily MMF dose at 1-year in mg, BK nephropathy and classification to either groups 2 (SCI) or 3 (SAR) versus 1 (control).

Recipient age (per year)1.021.00–1.050.03 1.020.98–1.06NS
Recipient weight (per kg)0.980.96–1.000.03 0.980.96–1.01NS
Deceased donor source2.191.20–4.020.01 1.900.72–5.02NS
Donor age (per year)1.041.01–1.060.0011.040.99–1.07NS
ECD kidney4.741.82–12.350.002   
Delayed function2.841.27–6.330.01 4.501.23–16.60.02
MMF dose at 1 yr (per mg increase)0.990.998–0.999 0.00020.990.997–0.9990.04
BK nephropathy8.382.42–29.0 0.000519.71.71–2270.02
IF/TA > 2 at 1-month Bx5.682.04–15.8 0.00093.731.00–14.00.05
All groups (2,3,4) versus control (1)3.271.77–6.050.002   
Group 2 (SCI) versus 1 (control)3.031.47–6.340.003   
Groups 2 (SCI) or 3 (SAR) versus 1 (control)3.781.99–7.19 0.00016.622.68–16.3<0.0001

By univariate analysis, recipient age, recipient weight, deceased donor source, donor age, extended criteria donor (ECD) kidney, delayed graft function, the MMF dose at 1 year, BK nephropathy and IF/TA > 2 on the 1-month biopsy were all associated with the higher risk of IF/TA > 2 on the 1-year biopsy.

The group classification was strongly associated with the risk of IF/TA on the 1-year biopsy. Patients in either groups 2 (SCI) or 3 (SAR) were 3.78 (95% CI 1.99–7.19) times more likely to have the higher IF/TA score on the 1-year biopsy. Patients exclusively in group 2 (SCI) also had a higher risk of IF/TA at 1 year (OR 3.03, 95% CI 1.47–6.34).

The following clinical factors were not associated with a higher IF/TA score on the 1-year biopsy: recipient gender, recipient race, previous transplant, pretransplant diabetes, donor gender, preemptive transplantation, HLA mismatches, PRA > 0, induction agent used, being on tacrolimus at month 4 or 1 year, on MMF at 1 year, on steroids at 1 year or positive C4d staining on the 1-month biopsy.

We then performed a multivariate risk analysis of all the variables that were significant by the univariate analysis. We did not include ECD kidney donor in this model, since the donor age served as surrogate for this characteristic. We included classification in either groups 2 (SCI) or 3 (SAR) versus the control group in this model. In this model classification into either group 2 (SCI) or 3 (SAR) was a strong independent risk factor for a higher IF/TA score on the 1-year biopsy (OR 6.62, 95% CI 2.68–16.3).

Discussion

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

In this study of nonsensitized kidney transplant recipients treated with induction therapy and rapid steroid withdrawal, SCI below the threshold of Banff IA acute rejection occurred in 20% of patients and SAR occurred in 7% during the first 4 months. The risk of an IF/TA score > 2 at 1 year is substantially higher in patients with either SCI or SAR. In addition, the fraction of patients with 1-year biopsies showing a combination of both an IF/TA score > 2 and interstitial inflammation (Banff i score > 0) was higher in group 2 (SCI) (16%, p = 0.004) and group 3 (SAR) (37%, p < 0.0001) compared to the control group (3%). While several other clinical factors are associated with a higher IF/TA score at 1 year, in a multivariate analysis controlling for these factors, SCI or SAR occurring during the first 4 months are independently associated with a substantial higher risk of a higher IF/TA score at 1 year.

IF/TA is a marker of inferior long-term graft survival (4). However, IF/TA occurring in conjunction with inflammation has even a worse long-term graft survival when compared to patients with IF/TA alone (7,9). In a study of 292 patients with 1-year protocol biopsies Cosio at al. showed that 45% had isolated IF/TA and 18% had IF/TA with inflammation. During follow-up, the group with IF/TA and inflammation had a lower GFR and inferior graft survival compared to the control group or the group with isolated IF/TA (9). In another study, Moreso et al. also found inferior graft survival in a cohort of 435 patients with IF/TA and inflammation on a 6-month protocol biopsy (7).

The etiology of IF/TA in kidney allografts is multifactorial and includes both immunologic and nonimmunologic causes (2). The immunologic factors include the degree of HLA mismatching, acute cellular rejection or antibody mediated rejection, while nonimmunologic factors include BK nephropathy, chronic calcineurin inhibitor (CNI) nephrotoxicity, urinary tract infection, transplant ureteral stenosis and organ ischemia. There is evidence in both animal and human studies of an epithelial–mesenchymal transition in tubular epithelial cells occurring as the result of immunologic injury (15).

Previous studies of kidney transplant recipients have shown that acute cellular rejection is associated with the subsequent development of IF/TA in steroid-treated patients (3). In addition, subclinical inflammation below the threshold for Banff IA acute rejection has previously been associated with the development of IF/TA (6,9,16). In all of these studies the patients were treated with chronic corticosteroids while our study confirms similar findings in a cohort of patients on RSW. In addition, our data along with these studies would suggest that early SCI, even below the threshold for Banff IA acute rejection is associated with more IF/TA on a subsequent biopsy.

It is difficult to compare studies of protocol biopsy findings due to the wide variations in patient demographics, the utilization of induction therapy, the timing of the biopsy and the selection of calcineurin inhibitor. Anil Kumar et al. studied protocol biopsy findings in a cohort of 256 patients treated with RSW (17). Their cohort was 60% African American which is associated with a higher immunologic risk (18). All patients received basiliximab induction with tacrolimus with either MMF or sirolimus. The prevalence of SCR was 17% at 1 month and 8% at 6 months. Borderline changes were not reported (17). In our study of recipients on RSW we found SCI in 20% and SCR in 7% of patients on either the 1- or 4-month biopsy.

All of the other reported studies of early protocol biopsy findings are in patients treated with chronic corticosteroids. Nankivell et al. studied 119 patients treated without induction therapy and with various combinations of immunosuppression agents (5). In biopsies done at 3 months, the prevalence of SCI was 23.4% and SAR was 6.9%. Scholten et al. studied 126 patients randomized to tacrolimus or cyclosporine with MMF and steroids and no induction therapy. At 6 months, the prevalence of SCI in the entire cohort was 23.4% and SAR was 7.4% (19). Finally, Rush et al. studied 218 low immunologic-risk recipients treated with tacrolimus, MMF and steroids and no induction. In this prospective study, the risk of SAR was unexpectedly low with a prevalence of 4.6% at 6 months (20). Based on all these studies, it is difficult to make any conclusions regarding the impact of RSW on the occurrence of SAR and SCI.

Perhaps surprisingly, we found that patients with CAR (group 4) were not associated with an increased risk of IF/TA on the 1-year biopsy (Table 3). There are a number of possible explanations for this unexpected finding. Our protocols for treating subclinical and CAR are different. Patients with subclinical rejection receive a single dose of methylprednisolone and the maintenance immunosuppression is optimized, but steroids are not continued. While in patients with CAR, our protocol calls for three doses of methylprednisolone and prednisone is continued chronically. This practice is reflected in the finding that more patients in group 4 are on steroids at 1 year compared to the control group (Table 2). It is feasible that the chronic steroid use may suppress inflammation and modulate the development of IF/TA.

We excluded patients with pretransplant donor-specific antibodies (DSA’s) or a positive flow cross-match. In more than 90% of patients in our study, the PRA was zero at the time of transplant, and there were no significant differences between the four groups in this characteristic (Table 1). But there were differences in the occurrence of peritubular C4d staining between the groups (Table 4). On the 1-year biopsy, group 3 (SAR) was more likely to be C4d positive. In addition, group 3 (SAR) was more likely to have allograft glomerulopathy (Banff cg changes). It is possible that de novo DSA's and antibody mediated endothelial injury contributed to the development of IF/TA in this group (21) but our data cannot directly address this question. We do not have data on the development of de novo DSA's in our patients.

There are several limitations in this study. Although we have a clinical protocol for treating acute rejection, the treatment of the SAR on the biopsies was not uniform. In addition, concordance between interpretations of the protocol biopsy findings between the two pathologists reading the biopsies was not measured. We did not determine the total inflammation score (Banff ti score) in this study (22). It is feasible that the total burden of inflammation, including in the areas of scaring, may contribute to the progression of IF/TA, but our data cannot directly address this question (6).

We conclude that either subclinical inflammation or SAR occurring during first 4 months after kidney transplantation is associated with a substantial higher risk of IF/TA score > 2 and persistent inflammation on the 1-year protocol biopsy in a cohort of patients on RSW. Our data would suggest a valuable role for early allograft biopsies in patients on rapid steroid withdrawal, even in patients with stable graft function, to monitor for subclinical inflammation. Controlled studies to determine the impact of the treatment of subclinical inflammation on the subsequent development of IF/TA are needed.

References

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