SEARCH

SEARCH BY CITATION

Keywords:

  • Clinical research/practice;
  • glomerular filtration rate (GFR);
  • kidney (allograft) function/dysfunction;
  • kidney transplantation/nephrology

Abstract

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

Death with function (DWF) is a major cause of kidney allograft failure. Allograft dysfunction may contribute to DWF. The aim of this study was to examine the relationship between DWF and allograft function using estimated GFR (eGFR) and histology. We retrospectively analyzed 1842 kidney allografts transplanted at our center from 1996 to 2010. eGFR was estimated using the MDRD equation. Biopsies obtained 12 months posttransplant and within 1 year of DWF were analyzed. Proportional hazards models were used to examine the relationship between eGFR and DWF. During 68 ± 43 months of follow-up, 14% (n = 256) of recipients experienced DWF. Risk factors of DWF included increasing recipient age (hazard ratio [HR] = 2.07, confidence interval [CI] 1.77–2.43, p < 0.0001), diabetes (HR = 2.58, CI 1.81–3.69, p < 0.0001), prior dialysis (HR = 1.47, CI 1.05–2.06, p = 0.03) and eGFR <40 mL/min/1.73 m2 (HR 2.26 per 10 mL/min/1.73 m2 decrease in eGFR, CI 1.82–2.81, p < 0.0001). Prior to death, only 15.9% (n = 39) of DWF recipients had stage 4 chronic kidney disease (CKD) and only 4.9% (n = 12) had stage 5 CKD. Most biopsies performed within 1 year of DWF (68%) demonstrated benign histology and were comparable to biopsies from matched controls. In conclusion, allograft dysfunction is independently associated with DWF. However, the majority of DWF recipients have well-preserved allograft function and histology prior to death.


Abbreviations
CI

confidence interval

CKD

chronic kidney disease

cTNT

cardiac troponin T

DWF

death with function

eGFR

estimated GFR

ESRD

end-stage renal disease

GIF

graft interstitial fibrosis

HR

hazard ratio

TG

transplant glomerulopathy

Introduction

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

Despite marked improvement in early renal allograft survival over the past decade, long-term allograft survival has remained disappointingly stable [1]. A key step in improving long-term allograft survival is optimizing the care of patients at highest risk of graft loss [2]. Because nearly half of all graft losses are due to death with function (DWF), preventing DWF could significantly improve allograft survival [3]. Many established risk factors for DWF are not easily modified, including recipient age, cause of end-stage renal disease (ESRD) and length of pretransplant dialysis [4-6]. Recent studies, however, suggest that reduced allograft function may also contribute to DWF [5, 7-9]. This relationship is intriguing because it implies that efforts to improve allograft function could decrease rates of DWF.

Whether improving allograft function would greatly impact DWF rates depends upon several factors. First, a significant proportion of patients experiencing DWF would need to have allograft dysfunction. Previous studies, however, have suggested that the majority of patients experiencing DWF actually have preserved allograft function [9]. Second, allograft dysfunction would have to be present early enough posttransplant so that an intervention could be performed. For example, calcineurin inhibitor avoidance could be pursued if low allograft dysfunction manifests during the first posttransplant year [10]. Last, the relationship between reduced allograft function and patient survival would need to be causal. Otherwise, improving kidney function would have little impact on survival.

The goal of these analyses was to examine the relationship between allograft function and DWF using allograft histology. Our initial aim was to identify how many patients with DWF actually have reduced allograft function, defined by estimated GFR (eGFR) and/or histology. Next we examined allograft function and histology at different times posttransplant in order to determine whether low function develops early or late posttransplant in DWF recipients. Finally, we investigated whether allograft function is associated with cause of death, thereby indirectly examining whether there is a mechanistic link between graft dysfunction and mortality.

Materials and Methods

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

Patient population

The study cohort consisted of all adult kidney transplants performed at our institution between January 1, 1996 and December 31, 2010. We excluded recipients of more than one organ, positive crossmatch and blood group incompatible transplants. Patients who lost contact with our center within the first posttransplant year were considered lost to follow-up. Clinical information was abstracted from electronic databases. Our study was approved by the Mayo Clinic IRB.

Immunosuppression

Overall, 1286 of 1842 patients (69.8%) received induction with Thymoglobulin, 270 patients (14.7%) received anti-CD25 antibodies, 90 (4.9%) received alemtuzumab and 8 (0.4%) received OKT3. Maintenance immunosuppression 1 year after transplant included mycophenolate mofetil and corticosteroids in all patients, combined with tacrolimus in 1425 patients (77.4%), cyclosporine in 233 patients (12.6%) or sirolimus in 101 patients (5.5%). Twenty-one patients (1.1%) received the combination of tacrolimus and sirolimus.

Patient and allograft follow-up

At our center, patients return regularly during the first 3 weeks posttransplant, again at month 4 and annually thereafter. Allograft function was assessed by frequent determinations of serum creatinine from which eGFR was obtained using the four-variable MDRD. Since data were collected before isotope dilution mass spectrometry-traceable creatinine values were obtained, the original MDRD equation was used: eGFR = 186 × SCr−1.154 × Age−0.203  × 0.742 [if female] × 1.212 [if black] [11]. As outlined below, eGFR was analyzed as a time-dependent covariate in the Cox proportional hazards analysis examining the relationship between eGFR and DWF. eGFR was also analyzed in three posttransplant periods: 6-month, 12-month and last. In surviving patients, last eGFR was defined as the last eGFR measured within 6 months of last follow-up. In patients with DWF, the last eGFR measured within 6 months of death was analyzed, excluding eGFRs obtained within the first 15 days posttransplant or within 30 days prior to death. Thus, eGFRs from patients who died within 45 days of transplant were excluded from analysis (n = 13). eGFRs > 125 mL/min/1.73 m2 at any time point were set to 125 mL/min/1.73 m2 for analysis (n = 5; 0.3% at 12 months).

In 1998, our institution began performing protocol allograft biopsies at the time of transplant, 4 months, 12 months, 2 years and 5 years after transplant. All biopsies were scored using Banff criteria [12-14]. In these studies, we analyzed 12-month protocol biopsies and biopsies obtained prior to last follow-up. In DWF recipients, the last biopsy obtained during the year prior to death was included. These last biopsies were compared with biopsies from a matched cohort of surviving recipients with functioning grafts. Matching was done according to the following variables: (1) recipient age (±10 years), (2) transplant year (±3 years), (3) type of donor (living vs. deceased), (4) time from transplant to biopsy (±2 years) and (5) recipient gender. Histological changes in 12-month and last biopsies were classified according to phenotypes that relate to death-censored graft survival [15], including (1) normal histology defined by the absence of any pathologic diagnosis and normal Banff scores; (2) mild graft interstitial fibrosis (GIF) defined as a “ci” score = 1 without inflammation (i.e. “i” score = 0); (3) moderate or severe GIF defined by the presence of “ci” score > 1 and “i” score = 0; (4) fibrosis associated with inflammation in nonfibrotic areas (i.e. “ci” > 0 and “i” score > 0) (GIF + “i”); and (5) transplant glomerulopathy (TG) defined by a “cg” score > 0 regardless of any other score.

Data analysis

Data were expressed as means and standard deviation or median and range as appropriate. Continuous data were compared by Student's t-test and nonparametric tests (Kruskal–Wallis) for normally distributed and skewed data, respectively. Proportions were compared with the chi-square test. Paired comparisons were done by Wilcoxon sign-rank test. Survival analysis was done using the Kaplan–Meier method. Graft failure was defined as return to dialysis or retransplantation. Patient follow-up was censored at the time of death, graft failure or date of analysis. Causes of death were determined by review of medical records and death certificates and classified as described previously.

eGFR was treated as a time-dependent covariate in Cox proportional hazards analysis examining the relationship between allograft eGFR and DWF censoring for graft failure. A smoothing spline regression model was first fit to the most recent eGFR, adjusting for age and sex, to understand the general relationship between eGFR and patient survival. Then specific terms of eGFR were fit yielding a piecewise linear function for eGFR. Multiple change points were allowed for deviation from a simple linear function and those terms selected by a stepwise selection procedure were included in the final model. For the analysis relating death with biopsy findings a nested case–control study was performed, considering controls among those individuals in the 6 months previous to the cases' time of death relative to transplant.

Results

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

Patient characteristics

The study included 1842 allograft recipients with a mean follow-up of 68 ± 43 months. At the end of the follow-up, 1303 grafts (71%) were functioning, 267 (14%) were lost not due to death, 256 (14%) were lost due to DWF and 16 (1%) were lost to follow-up. The baseline characteristics of the recipients are shown in Table 1. Compared to recipients with functioning grafts, recipients with DWF were significantly older and were more likely to be diabetic and have required pretransplant dialysis. Furthermore, the DWF group included more patients with ESRD secondary to diabetes and hypertension/vascular disease and fewer patients with ESRD secondary to glomerulonephritis. In contrast, compared to recipients with functioning grafts, recipients with graft loss were younger and received more kidneys from deceased donors and older donors. Furthermore, recipients with graft loss included more re-transplants, a higher proportion of patients with ESRD secondary to glomerulonephritis and a lower proportion of patients with ESRD secondary to polycystic kidney disease (Table 1).

Table 1. Patient characteristics
Pretransplant parametersAll patients (n = 1842)Functioning grafts (n = 1303)Graft loss (n = 267)DWF (n = 256)p
  • DWF, death with function; ESRD, end-stage renal disease.

  • 1

    Analysis of variance.

  • 2

    Chi-square.

  • 3

    Polycystic kidney disease.

  • 4

    Kruskall–Wallis.

Recipient age at transplant51.6 ± 1450.9 ± 1446.9 ± 1560.9 ± 11<0.00011
Male (%)61.560.559.968.40.0542
Caucasian (%)92.392.790.692.90.530
Diabetes (%)23.120.817.240.6<0.00012
Living donor (%)73.678.758.164.4<0.0001
Donor age42.7 ± 1441.3 ± 1445.4 ± 1442.5 ± 140.002
Prior dialysis (%)60.355.765.669.5<0.00012
Cause of ESRD:    <0.00012
Glomerular disease32.234.538.917.3 
Diabetes15.711.611.734.1 
PKD313.115.37.711.8 
Hypertension/vascular10.39.57.316.4 
Retransplant10.810.218.24.5 
Other11.712.510.110.9 
Unknown6.16.46.15.0 
Follow-up (months)67.6 ± 4373.7 ± 4347.4 ± 4155.6 ± 43<0.00014

Allograft function and histology near the time of DWF

Compared to recipients with functioning allografts, eGFR at last follow-up was significantly lower in patients with DWF (DWF recipients 50.1 ± 23 mL/min/1.73 m2 vs. functioning recipients 56.3 ± 20 mL/min/1.73 m2, p = 0.001). Figure 1 shows the distribution of chronic kidney disease (CKD) stages in these two groups of recipients. Although the distribution was significantly different between groups (p < 0.0001, chi-square), it should be noted that only 39 of 245 patients with DWF (15.9%) had CKD stage 4 (eGFR between 15 and 30 mL/min/1.73 m2) and 12 (4.9%) had CKD stage 5 (eGFR <15 mL/min/1.73 m2). Lower eGFR related to DWF (hazard ratio [HR] = 0.99 [0.98–1.00], p = 0.006). However, this relationship was complex (Figure 2) as the risk of DWF was increased both in recipients with eGFR < 40 mL/min/1.73 m2 (HR 2.26 per 10 mL/min/1.73 m2 decrease in eGFR, confidence interval [CI] 1.82–2.81, p < 0.0001) and in those with eGFR > 70 mL/min/m2 (HR 1.26 per 10 mL/min/1.73 m2 increase in eGFR, CI 1.01–1.58, p = 0.041). The relationship between eGFR and DWF was statistically independent of other risk factors including recipient age, diabetes, pretransplant dialysis and cause of ESRD (Table 2). In contrast, recipient gender, recipient race, recipient BMI, donor type and induction immunosuppression were not significantly related to DWF.

image

Figure 1. Distribution of estimated GFR values at last follow-up in patients with functioning grafts (striped bars) and in patients with death with function (black bars) (p < 0.0001, chi-square).

Download figure to PowerPoint

image

Figure 2. Relationship between estimated GFR and death with function adjusted for multiple covariates.

Download figure to PowerPoint

Table 2. Analyses of factors related to DWF
VariableUnivariate modelMultivariate model
HR (CI)pHR (CI)p
  • CI, confidence interval; DWF, death with function; ESRD, end-stage renal disease; eGFR, estimated GFR; HR, hazard ratio.

  • 1

    HR for age per 10 years, eGFR per increase of 1 mL/min/1.73 m2 analyzed as a time-dependent covariate, blood pressure per increase of 1 mmHg, BMI per increase of 1 kg/m2.

  • 2

    Polycystic kidney disease.

  • 3

    Landmark analysis was performed beginning at 1 year.

Recipient age at transplant12.13 (1.83–2.48)<0.00012.07 (1.77–2.43)<0.0001
Male1.37 (0.98–1.92)0.07  
Caucasian1.58 (0.74–3.37)0.24  
Diabetes3.72 (2.66–5.20)<0.00012.58 (1.81–3.69)<0.0001
Living donor0.80 (0.57–1.11)0.18  
Donor age11.19 (1.04–1.35)0.01  
Pretransplant dialysis1.53 (1.10–2.14)0.011.47 (1.05–2.06)0.03
Cause of ESRD:
Glomerular disease0.33 (0.21–0.52)<0.0001  
Diabetes1.92 (1.27–2.89)0.002  
PKD20.72 (0.44–1.20)0.21  
Hypertension/vascular1.92 (1.27–2.89)0.002  
Retransplant0.24 (0.09–0.66)0.005  
Other1.07 (0.67–1.71)0.781.65 (1.01–2.69)0.04
Unknown1.05 (0.58–1.89)0.88  
Delayed graft function2.06 (1.30–3.28)0.002  
Thymoglobulin1.29 (0.90–1.87)0.17  
eGFR (linear term)10.99 (0.98–1.00)0.006  
eGFR > 70 mL/min/1.73 m21.08 (0.86–1.35)0.501.26 (1.01–1.58)0.04
eGFR < 40 mL/min/1.73 m21.99 (1.63–2.43)<0.00012.26 (1.82–2.81)<0.0001
Acute rejection within 1 year31.41 (0.92–2.14)0.11  
Diastolic blood pressure at 1 year1, 30.99 (0.98–1.00)0.02  
BMI at 1 year1, 31.01 (0.98–1.05)0.45  

Among the 256 recipients with DWF, 62 (24%) had an allograft biopsy within 1 year of death. Of those biopsies, 47% were protocol biopsies. The remaining 53% biopsies were performed for indications such as increasing creatinine or worsening proteinuria. Overall, 68% of biopsies performed within 1 year of DWF had normal or only mild histologic changes (Figure 3) while the remainder showed more severe histologic changes that, in previous studies, have been associated with shortened allograft survival [15]. Overall, biopsies obtained within 1 year of DWF were no different than biopsies from a matched cohort of surviving recipients with functioning grafts (Figure 3).

image

Figure 3. Distribution of histologic patterns in biopsies within 1 year prior to death with function (DWF). Mild histologic changes include: normal biopsy and mild graft interstitial fibrosis (GIF); moderate/severe histologic changes include: moderate/severe GIF, GIF associated with inflammation and transplant glomerulopathy (TG).

Download figure to PowerPoint

Causes of death

The causes of death among the DWF group were cardiovascular (30.1%), infectious (13.3%), malignancy (23.4%), other (8.2%) and unknown (25.5%). Although lower last eGFR related to an overall increase in DWF (see above), this variable did not relate to an increase in any particular cause of death considering eGFR as a continuous variable or as a discontinuous variable using CKD stages (p = 0.214, chi-square). eGFR at 12 months posttransplant also did not relate to causes of death considering eGFR as a continuous variable or as a discontinuous variable using CKD stages (p = 0.945, chi-square).

Graft function and histology early posttransplant in DWF

In these studies, we investigated 6-month eGFR, 12-month eGFR and 12-month graft histology in recipients with either a functioning graft, graft loss or DWF more than 1 year posttransplant (Table 3). Compared to recipients with functioning grafts, 6-month eGFR was slightly lower in recipients who suffered graft loss or DWF beyond 1 year posttransplant. In addition, 12-month eGFR was significantly lower both in recipients with DWF and in recipients who eventually lost their grafts compared to recipients with functioning grafts (Table 3). Compared with recipients who had a 12-month eGFR > 60 mL/min/1.73 m2, the risk of DWF censored for graft loss increased progressively with decreasing eGFR (eGFR 45–60 mL/min/1.73 m2 [N = 649], HR = 1.48 [1.02–2.15], p = 0.038; eGFR 30–44 mL/min/1.73 m2 [N = 312], HR = 1.80 [1.20–2.69], p = 0.004; eGFR 15–30 mL/min/1.73 m2 [N = 46], HR = 5.49 [3.11–9.67], p < 0.0001) (Figure 4). It should be noted that the increased risk of DWF in patients with a 12-month eGFR between 30 and 60 mL/min/1.73 m2 manifests more than 5 years after transplant. In contrast, recipients with an eGFR < 30 mL/min/1.73 m2 appear to have an increased risk of DWF beginning 12 months after transplant (Figure 4). Contrary to the observations made with eGFR, the risk of DWF was not increased in patients with a 12-month eGFR > 70 mL/min/1.73 m2 (data not shown). Overall, allograft histology 12 months posttransplant was not statistically different in patients with functioning grafts (n = 851) compared with those experiencing DWF (n = 87) (Table 3). However, patients with an eGFR < 40 mL/min/m2 at 12 months were less likely to have normal or mild GIF at 12 months and more likely to have moderate/severe GIF, GIF + “i” and transplant TG (p < 0.0001) (Table 4). Patients who lost their allograft beyond 1 year posttransplant (n = 85) were significantly more likely to have abnormal histology on 12-month biopsy.

Table 3. eGFR and graft histology during the first year posttransplant in recipients with functioning grafts, DWF or graft loss more than 1 year posttransplant
VariablesAll recipientsFunctioning graftsDWFGraft lossp
  • DWF, death with function.

  • 1

    Analysis of variance; by Bonferroni post hoc analysis, estimated GFR (eGFR) in all groups was significantly different from one another.

  • 2

    GIF = graft interstitial fibrosis, GIF + “i” = fibrosis associated with inflammation in nonfibrotic areas, TG = transplant glomerulopathy.

  • 3

    Chi-square comparing the histology of grafts that were eventually lost with the histology of grafts that were functioning at the end of follow-up.

6-month eGFR54.3 ± 16 (n = 1494)56.2 ± 16 (n = 1160)51.0 ± 15 (n = 190)43.6 ± 16 (n = 141)<0.0011
12-month eGFR53.7 ± 16 (n = 1273)55.7 ± 15 (n = 987)49.9 ± 17 (n = 161)41.9 + 16 (n = 124)<0.00011
12-month histology2    <0.00013
Normal510 (49.9%)462 (54.3%)31 (35.6%)17 (20%) 
Mild GIF284 (27.8%)240 (28.2%)30 (34.5%)14 (16.5%) 
Moderate/severe GIF73 (7.14%)50 (5.9%)7 (8.1%)16 (18.8%) 
GIF + “i”119 (11.6%)80 (9.4%)15 (17.2%)24 (28.2%) 
TG37 (3.6%)19 (2.2%)4 (4.6%)14 (16.4%) 
image

Figure 4. Relationship between estimated GFR (eGFR) measured 12 months posttransplant and death with function (Kaplan–Meier plots). Lines represent: eGFR > 60 mL/min/1.73 m2 (—); eGFR 45–60 (+ — +); eGFR 30–44 (Δ — Δ); and eGFR < 30 (○ — ○) (log rank p < 0.0001).

Download figure to PowerPoint

Table 4. A 12-month allograft histology in patients according to eGFR
12-month histology112-month eGFR < 40 mL/min/m2 (n = 189)12-month eGFR ≥ 40 mL/min/m2 (n = 1041)
  • 1

    GIF = graft interstitial fibrosis, GIF + “i” = fibrosis associated with inflammation in nonfibrotic areas.

  • 2

    Chi-square comparing the proportion of histologic findings in patients with 12-month estimated GFRs (eGFRs) < 40 mL/min/m2 or ≥40 mL/min/m2 (p < 0.0001).

Normal52 (27.5%)548 (52.6%)2
Mild GIF46 (24.3%)300 (28.8%)2
Moderate/severe GIF36 (19.1%)56 (5.4%)2
GIF + “i”34 (18.0%)104 (10.0%)2
Transplant glomerulopathy21 (11.1%)33 (3.2%)2

Discussion

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

The first aim of our study was to identify how many patients with DWF have reduced allograft function. In our cohort, DWF occurred in 14% of patients and accounted for approximately half of all graft losses during 68 ± 43 months of follow-up. The majority of DWF recipients had preserved allograft function with a mean eGFR of 50.1 ± 23 mL/min/1.73 m2 during the 6 months prior to death. In fact, only 15.9% of DWF recipients had stage 4 CKD and only 4.9% had stage 5 CKD in the months before death. While patients with low eGFR were more likely to have abnormal allograft histology, the majority of allograft biopsies performed within 1 year of DWF exhibited only mild histologic abnormalities. In the patients who underwent allograft biopsies within 1 year of DWF, histology was comparable to histology from matched recipients with functioning grafts. Ours is the first study to examine both allograft function and histology within the year prior to DWF. Our finding that allograft function is preserved in the majority of DWF recipients suggests that rates of death-censored graft loss are accurate and do not underestimate the burden of allograft dysfunction.

Our study found a strong and independent relationship between allograft function and mortality. The relationship between last eGFR and DWF was complex given that the risk of death was increased both in patients with an eGFR < 40 mL/min/1.73 m2 and in patients with an eGFR > 70 mL/min/1.73 m2. These results are strikingly similar to those of previous studies [7] involving a large, multicenter group of kidney transplant recipients followed for approximately 10 years. Kasiske et al [7] noted that recipients with a 12-month eGFR of 15–29 mL/min/1.73 m2 were 68% more likely to experience DWF compared with recipients with an eGFR of 60–89 mL/min/1.73 m2. In addition, recipients with an eGFR > 90 mL/min/1.73 m2 had a 61% increased risk of death [7], a finding similar to our study. Another large study that examined the relationship between allograft function and patient survival also noted an association between low eGFR and DWF. However, they did not find a relationship between high GFR (eGFR ≥ 75 mL/min/1.73 m2) and death [8]. It is postulated that the association between high eGFR and death does not reflect an intrinsic property of allograft function itself, but rather the relationship between low creatinine and decreased muscle mass. Comorbidities contributing to low muscle mass may explain the increased risk of death [7, 8, 16].

The second aim of our study was to examine whether allograft function relates to particular causes of death. Prior research has established a relationship between decreasing GFR and cardiovascular death, both in the general population and in kidney transplant recipients [8, 17-19]. In a recent study [20], we noted a relationship between allograft function and elevated cardiac troponin T (cTNT), a marker of underlying myocardial ischemia and cardiac pathology. While cTNT normalizes within the first posttransplant month in the majority of kidney transplant recipients, recipients with an eGFR < 30 mL/min/1.73 m2 are more likely to have a persistently elevated cTNT and high mortality. In fact, the risk of death in patients with an elevated cTNT early posttransplant is quantitatively similar to the risk of death in patients from our cohort with a 12-month posttransplant eGFR < 30 mL/min/1.73 m2. These findings support a causal relationship between allograft dysfunction and cardiovascular disease. Although we did not find an association between reduced eGFR and particular etiologies of death in our current cohort, the proportion of patients with DWF and low eGFR was quite small and this relationship deserves further study.

Finally, the last aim of our study was to examine whether low allograft function develops early or late posttransplant in DWF recipients. We found that the relationship between reduced eGFR and DWF begins to develop within the first posttransplant year. By 6 months posttransplant, DWF recipients have a slightly lower eGFR than surviving recipients with functioning grafts. Progressively lower 12-month eGFRs were associated with an increasing risk of DWF. Interestingly, the timing of DWF differed according to level of 12-month eGFR (Figure 4). Thus, patients with a 12-month eGFR between 15 and 30 mL/min/1.73 m2 experienced a rapidly increasing risk of DWF beginning 12 months posttransplant. The increased risk was so pronounced that approximately 50% of these patients died within the first 5 years following transplant. In contrast, recipients with higher 12-month eGFRs did not experience an increased risk of DWF until more than 5 years after transplant.

Our study does not specifically address whether improving eGFR would decrease DWF. Further studies are needed to determine whether transplanting better organs or avoiding calcineurin-based immunosuppression in high-risk recipients will improve survival. However, we would be cautious about the widespread implementation of calcineurin inhibitor avoidance given that it may improve short-term graft function but potentially compromise long-term graft survival [21]. Furthermore, preventing death in DWF recipients with stages 4 or 5 CKD in our cohort would improve allograft survival by only 2%.

In conclusion, our studies show that allograft dysfunction is independently associated with DWF. DWF recipients develop lower allograft function within the first 12 months posttransplant. Despite the relationship between graft dysfunction and death, only a minority of DWF recipients actually have low function. In fact, the vast majority of patients experiencing DWF have well-preserved allograft function prior to death. In addition, biopsies obtained shortly prior to death showed benign histology. Our findings suggest that efforts to improve allograft function will have little impact on DWF given that the majority of recipients do not have allograft dysfunction prior to death.

Acknowledgments

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

We would like to thank the Mayo Clinic transplant coordinators for their tireless efforts in patient follow-up and data collection.

Disclosure

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

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

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