SEARCH

SEARCH BY CITATION

Keywords:

  • Donor age;
  • sequential stratification;
  • SRTR;
  • survival;
  • transplant benefit

Abstract

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

Nearly one-quarter of the kidney transplant waiting list is composed of repeat transplantation candidates. Survival following retransplantation using expanded criteria donor (ECD) kidneys has not been adequately studied. Using data from the Scientific Registry of Transplant Recipients, we analyzed mortality after retransplantation with ECD and non-ECD deceased-donor kidneys. Adult patients who experienced graft failure and were relisted for transplantation between 1995 and 2004 were studied (n = 9641). Follow-up began at the date of relisting and continued until death or the end of the observation period (December 31, 2004), with censoring at living-donor transplantation. Sequential stratification (an extension of Cox regression) was used to compare mortality between patients receiving an ECD retransplant and those remaining on the waiting list or receiving a non-ECD retransplant (conventional therapy). Of 2908 retransplantations, 292 used ECD kidneys. Survival after ECD retransplantation was approximately equal to that of conventional therapy, with an adjusted hazard ratio of 0.98 (p = 0.88). In contrast, non-ECD retransplant recipients experienced a significant reduction in mortality (HR = 0.44; p < 0.0001). Based on these national data, recipients of ECD retransplantation do not have a survival advantage relative to conventional therapy, whereas non-ECD retransplantation is associated with a significant survival advantage.


Introduction

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

Kidney transplantation is the desired therapy for end-stage renal disease, affording improved survival and quality of life compared with dialysis (1, 2). Unfortunately, the number of potential recipients far outweighs the number of suitable donors, with the gap continuing to increase in recent years (3). Graft failure, especially due to chronic allograft nephropathy, continues to be a major problem and results in many patients returning to dialysis and being considered for repeat transplantation.

There have been a variety of responses to the critical shortage of kidneys available for transplantation, including strategies to optimize use of deceased-donor organs. One such strategy has been the use of expanded criteria donor (ECD) kidneys. Such kidneys are procured from donors aged 60 years and older or donors aged 50–59 years with at least two of the following conditions: cerebrovascular accident as cause of death, serum creatinine greater than 1.5 mg/dL or a history of hypertension. These kidneys have a relative risk of graft loss greater than 1.70 compared with kidneys from a reference group of donors aged 10–39 years without any of the other three conditions (4). Although kidneys from such donors are associated with poorer graft survival (by definition), transplantation of marginal kidneys has been shown to provide a survival advantage to recipients compared with being on dialysis (5) or receiving conventional therapy (defined as not receiving an ECD kidney and waiting to possibly receive a non-ECD organ later) (6). Use of ECD kidneys has been growing in recent years. However, a registry-based analysis has cautioned that certain groups may not benefit from transplantation with ECD organs (7).

Despite advances in immune suppression and general care of transplant recipients, many renal allografts ultimately fail, and these patients again face decisions regarding dialysis, transplantation or end-of-life care. Though not extensively studied, repeat renal transplantation is common practice after graft failure. In 2004, 19% of the national kidney waiting list was composed of patients with failed previous grafts, making graft failure one of the most common indications for renal transplantation (8). Ojo et al. have shown that patient survival is improved by repeat renal transplantation when compared with dialysis after primary graft failure (9). Rao et al. report similar findings based on Canadian Organ Replacement Registry data (10). Survival of second and subsequent renal allografts is similar to that of primary grafts (11), though risk factors for regraft failure are not as clearly defined. Donor age, early acute rejection, short duration of primary graft function and primary allograft nephrectomy are some of the factors that have been associated with poorer outcomes in retransplant recipients (12–16).

ECD kidney transplantation now accounts for 13–30% of kidney transplantations performed in the United States. The use of ECD kidneys for retransplantation has not been rigorously studied (3). At least one group has expressed concern, finding inferior graft survival and increased expense when ECD kidneys were used for retransplantation (17). The purpose of this study is to assess the outcomes of renal retransplantation using ECD kidneys.

Patients and Methods

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

Data were obtained from the Scientific Registry of Transplant Recipients (SRTR), which maintains data on all candidates for solid organ transplantation in the United States. Transplant centers submit data to the Organ Procurement and Transplantation Network (OPTN), which are then sent to the SRTR. Available patient-specific information includes demographic and clinical characteristics at wait-listing for retransplantation.

The study population (n = 9641) consisted of patients aged 18 years and older who experienced primary graft failure and were listed for a second kidney transplantation between 1995 and 2004. In an effort to increase homogeneity, patients whose first kidney transplantation occurred before January 1, 1990, were excluded. Follow-up began on the date of relisting and continued until death or the end of the observation period (December 31, 2004). To improve death ascertainment, mortality information provided by the OPTN was supplemented with that from the Social Security Administration's Death Master File.

For descriptive purposes, crude mortality rates were computed as the ratio of deaths to patient-years of follow-up. To quantify the survival benefit of ECD kidney retransplantation, a recently developed survival analysis method termed ‘sequential stratification’ was used (6). In the traditional, time-dependent Cox regression analysis (18), ECD retransplantation would be compared with remaining on the waiting list, which has at least two deficiencies. First, there is no adequate means of accounting for removals from the waiting list in the modeling, i.e., where removal corresponds to reasons other than death or transplantation. Second, a time-dependent analysis compares ECD retransplantation with remaining on the waiting list. The pertinent question from the patient's perspective is, ‘Am I better off getting an ECD kidney than remaining on the waiting list, given that I could receive a non-ECD organ in the future?’

Through sequential stratification, we are able to compare ECD retransplantation with conventional therapy, where conventional therapy entails not receiving an ECD organ but possibly receiving a non-ECD organ later. At the time of each ECD transplantation, a stratum is generated. Included in the stratum is the patient who received the ECD kidney and a set of matched conventional therapy patients (patients who were on the waiting list at the time of the ECD transplantation, matched by age to the ECD recipient). After matching, conventional therapy patients are not censored from that stratum if they subsequently receive a non-ECD kidney (which is part of conventional therapy) or if they are removed from the waiting list. Matched conventional therapy patients are only censored if they receive an ECD or living-donor (LD) kidney; we comment further on censoring issues in the next paragraph. The strata are combined and a (stratified) Cox regression model is fitted. In the Cox model, age and time after relisting are already adjusted for through the above stratification, while model covariates are employed to adjust for sex, race, primary renal diagnosis, panel reactive antibodies, calendar period, source of primary transplant (ECD, non-ECD, or living donor), time on dialysis before primary transplantation, time between primary transplantation and primary graft failure, and time between primary graft failure and relisting on the waiting list. Figure 1 presents the sequential stratification method.

image

Figure 1. Sequential stratification: selection of patients into a stratum generated by an ECD retransplant. At the time of each ECD retransplantation, age-matched patients from the waiting list are selected to create a comparison group. Over time, they may continue to wait, die, be removed from the waiting list, receive a non-ECD retransplant or receive an ECD retransplant. Other than loss to follow-up or the end of the study, censoring occurs only in the event of an ECD retransplantation, which triggers the generation of a new stratum. Note that patients who are removed from the waiting list or who receive a non-ECD transplant continue to be followed, which is important, given the comparison of interest (ECD vs. conventional therapy). Strata are created by this process sequentially as ECD retransplantations occur over time, and the strata are then combined to perform the analysis.

Download figure to PowerPoint

With regard to censoring, three issues warrant further clarification. First, LD transplantation is not of interest in the current investigation; follow-up time and death after LD transplantation are not relevant to the comparisons of interest. Hence, it is natural to censor patients if and when they receive an LD kidney, even though we have follow-up information available. The censoring of patients who receive an LD transplant is not unique to sequential stratification and would occur in a more standard analysis. Second, in the sequential stratification method employed, matched control patients within a stratum were not censored if they received a non-ECD transplant. Remaining on the waiting list, receiving a non-ECD transplant, or perhaps being removed from the waiting list are all potentially part of conventional therapy. Thus, non-ECD transplants and removals after stratum entry remain in the conventional therapy group. Such patients do not represent therapy cross-overs, given that ECD and conventional therapy are the therapies being compared. Third, matched control patients who subsequently receive an ECD kidney transplant are censored from that stratum. These patients do constitute therapy cross-overs, and their ECD transplants generate further separate strata for which matched controls will be selected.

We also compare non-ECD retransplantation with conventional therapy using the sequential stratification approach because a time-dependent Cox model does not properly account for removals from the waiting list. Conventional therapy is defined differently for the purpose of evaluating the survival benefit of non-ECD second kidney transplantation. Similar to the ECD transplantation setting, a stratum is set up for each non-ECD transplant. Included in the stratum are the transplant recipient and patients who, at the time of the transplantation, were on the waiting list (i.e. not removed or already retransplanted) and are matched for age. The conventional therapy patients are subsequently censored if they receive any transplant (ECD, non-ECD, or living donor).

As implied by the description above, because the comparison groups are constructed differently, the model fitting proceeds separately for ECD and non-ECD transplants. First, main effects models (no interaction terms) were fitted. We then fitted models that allowed the ‘ECD versus conventional therapy’ and ‘non-ECD versus conventional therapy’ hazard ratios (HRs) to differ with time since transplantation. Next, we estimated covariate-adjusted survival curves for ECD, non-ECD and their respective conventional therapy groups. To estimate survival probability, we used models that were stratified by treatment (i.e. ECD or conventional therapy) but adjusted for age through model covariates. The stratification by therapy frees us from the assumption that the HR is constant across post-retransplantation follow-up time (assumed by the main effects models) or even that the HR is constant within various postretransplantation time subintervals (assumed by the time-dependent models). Finally, we fitted interaction models that allowed the ECD/conventional therapy and non-ECD/conventional therapy HRs to depend on time between primary transplantation and primary graft failure.

Results

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

This study included 9641 patients who were placed on the waiting list for retransplantation. Table 1 lists characteristics of the study population. For descriptive purposes, we have divided the study population into three mutually exclusive categories: patients who received an ECD retransplant, patients who received a non-ECD retransplant and patients who were not retransplanted. All characteristics were measured at the time of relisting except for time on dialysis, which was measured at the time of first transplantation. Patients who received an ECD retransplant were older at listing, on average, than patients who received a non-ECD retransplant and patients not retransplanted. More than half of patients not retransplanted (53%) were nonwhite; the corresponding percentages were 40% and 38% for ECD and non-ECD retransplant recipients, respectively. Time between initiating dialysis and the first kidney transplantation was shorter for ECD recipients (2.09 years) than for non-ECD recipients (2.40 years) or patients not retransplanted (2.91 years). Potential confounding due to these factors was addressed by including them in the regression models.

Table 1.  Descriptive statistics for study population
CharacteristicRetransplant-ted ECDRetransplan-ted non-ECDNot retransplanted
  1. ECD = expanded criteria donor; PRA = panel reactive antibody.

Mean age at relisting46.542.044.5
Race: nonwhite40%38%53%
Primary diagnosis:  diabetes13%10%17%
PRA > 047%47%40%
Mean years on dialysis2.092.402.91
First transplant: ECD16%12%13%
First transplant: living donor28%27%25%

Table 2 presents crude mortality rates. A total of 2908 (30%) of the 9641 in the study population were retransplanted. Of those retransplanted, 292 (10%) received ECD kidneys. In total, 2023 deaths were observed. The crude postretransplantation mortality rates were 77.3 and 40.0 deaths per 1000 patient-years following ECD and non-ECD retransplantation, respectively, while the crude death rate on the waiting list was 73.6.

Table 2.  Unadjusted mortality rates
GroupDeathsPatient-years (PY)Rate per 1000 PY
Waiting list (n = 9641)1,61921,98773.6
Non-ECD (n = 2616)3358,37140.0
ECD (n = 292)6989377.3
Total2,02331,25164.7

Table 3 displays covariate-adjusted HRs. As stated in the previous section, the ECD and non-ECD groups have different conventional therapy comparison groups. Patients who received an ECD kidney retransplant experienced a mortality rate approximately equal to those who remained on the waiting list and perhaps later received a non-ECD kidney retransplant. Note that HR = 1 indicates equality between the two groups. With HR = 0.98 (p = 0.88), the survival of ECD retransplant recipients was neither clinically nor statistically different from conventional therapy. Conversely, non-ECD retransplant patients experienced a significant, 56% mortality decrease (HR = 0.44; p < 0.0001) relative to wait-listed patients. The postretransplantation potential follow-up distributions (length of time between the date of retransplantation and the end of the observation period) were quite similar for ECD and non-ECD retransplant recipients (median 3.3 vs. 3.2 years, maximum 9.3 vs. 9.6 years, respectively). The failure to observe a survival benefit for the ECD group is thus unlikely to be an artifact of a high concentration of post- and perioperative follow-up among ECD versus non-ECD transplant recipients.

Table 3.  ECD and non-ECD retransplantation versus conventional therapy (CT)
ComparisonCovariate-adjusted hazard ratio (HR)*95% CIP
  1. *Adjusted for age, sex, race, primary renal diagnosis, calendar period, time on dialysis before transplant, donor source, region, PRA, time between primary transplantation and graft failure, time between graft failure and relisting.

ECD vs. CT0.980.76, 1.260.88
Non-ECD vs. CT0.440.39, 0.51<0.0001

Figures 2 and 3 show HRs for non-ECD and ECD retransplantation compared with conventional therapy based on a generalization of the model that produced the results in Table 3. Specifically, Table 3 represents the average HR across the observed post-retransplantation follow-up time, while the figures demonstrate how the HRs differ by post-retransplantation follow-up interval. The ECD/conventional therapy HR is significantly elevated (almost twofold) in the first six months following retransplantation. The HR decreases in subsequent follow-up intervals, but two differences are notable. First, post-ECD retransplantation mortality is never significantly below that of conventional therapy. Second, the HR is not monotone, in contrast to the analog of this graph in most of the published literature (2,10,19). One might expect post-ECD retransplantation mortality to stabilize and, correspondingly, the HR to drop steadily as follow-up time increased, as reported in previous studies. However, we are comparing ECD retransplantation to conventional therapy, not strictly waiting list survival. Thus, in the conventional therapy group, many patients go on to receive a non-ECD organ, making the comparison group for ECD organs a shifting mixture of waiting list candidates and non-ECD retransplant recipients. Over time, many of the conventional therapy patients do receive non-ECD kidneys, which reduces their mortality.

image

Figure 2. HR over time for ECD retransplantation compared with CT. The hazard ratio (HR) decreases initially but is never significantly less than unity; i.e. ECD retransplantation at no time confers a mortality benefit compared with conventional therapy. The HRs are generated from the sequential stratification model, adjusted for sex, race, primary renal diagnosis, time on dialysis before primary transplantation, calendar period, source of primary transplant (ECD vs. non-ECD), duration of primary transplant function, and time between primary graft failure and relisting.

Download figure to PowerPoint

image

Figure 3. HR over time for retransplantation with non-ECD kidneys compared with CT. In the immediate postoperative period, the hazard ratio (HR) is elevated but then declines in a fairly monotone fashion, showing a significant mortality advantage for non-ECD retransplantation across the follow-up period. The HRs are generated from the sequential stratification model, adjusted for sex, race, primary renal diagnosis, time on dialysis before primary transplantation, calendar period, source of primary transplant (ECD vs. non-ECD), duration of primary transplant function, and time between primary graft failure and relisting.

Download figure to PowerPoint

The pattern in Figure 3 shows that, like ECD patients, non-ECD retransplant recipients experience a mortality rate that is significantly elevated immediately after retransplantation. However, the HR for non-ECD recipients, essentially compared with waiting list patients, drops steadily as post-retransplantation follow-up time increases. For non-ECD patients, the sequential stratification unfolds quite similarly to a typical time-dependent Cox model—the only difference is the handling of removals.

Figures 4 and 5 present covariate-adjusted survival curves. Recall that these are based on models that do not impose any assumptions on the ECD or non-ECD HRs—in fact, the models do not even estimate HRs for ECD or non-ECD patients. The survival curve patterns in Figures 4 and 5 are quite consistent with the results in Figures 2 and 3, respectively. For example, the ECD survival curve immediately drops below that of conventional therapy, eventually crosses and then rises above the conventional therapy curve, only to drop below it again before the six-year mark. The non-ECD survival curve diverges from the conventional therapy curve after one year and remains above it for the remainder of follow-up period.

image

Figure 4. Covariate-adjusted survival curves for ECD retransplantation group and CT comparison group. Survival for the ECD retransplantation group initially falls below that for the conventional therapy group, consistent with the elevated early hazard ratio. Despite the survival curves later crossing, they are more or less superimposed for much of the follow-up period. Survival probabilities were obtained from the sequential stratification model, adjusted for age, sex, race, primary renal diagnosis, time on dialysis before primary transplantation, calendar period, source of primary transplant (ECD vs. non-ECD), duration of primary transplant function, and time between primary graft failure and relisting.

Download figure to PowerPoint

image

Figure 5. Covariate-adjusted survival curves for non-ECD retransplantation group and CT comparison group. The initial increased mortality for the non-ECD retransplantation group is apparent but disappears early in the follow-up period. Thereafter, this group has a steadily increasing survival advantage compared to the conventional therapy group. Survival probabilities were obtained from the sequential stratification model, adjusted for age, sex, race, primary renal diagnosis, time on dialysis before primary transplantation, calendar period, source of primary transplant (ECD vs. non-ECD), duration of primary transplant function, and time between primary graft failure and relisting.

Download figure to PowerPoint

Table 4 lists results of our analysis of interactions. We examined the interaction between the retransplantation effect and several covariates, and the only statistically significant interaction was with the duration of function for the first transplant, i.e. the time between primary transplantation and primary graft failure. We fitted several models to sort out the precise nature of the interaction, and the results (not shown) were most consistent with a threshold effect. For both ECD and non-ECD retransplantation, the HR was significantly higher for patients with primary graft failure within four years of primary transplantation. In fact, even ECD retransplantation is associated with a significant, 45% mortality decrease (HR = 0.55; p = 0.03) among patients whose primary graft took longer than four years to fail.

Table 4.  Retransplantation versus CT by time until primary graft failure (GF)
ComparisonTime until primary GFHR95% Clp
ECD vs. CT0–4 years1.220.92, 1.620.17
ECD vs. CT4+ years0.550.32, 0.960.03
Non-ECD vs. CT0–4 years0.500.43, 0.59<0.0001
Non-ECD vs. CT4+ years0.320.24, 0.41<0.0001

Discussion

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

In this study of national registry data, we showed that retransplantation with ECD kidneys does not offer a significant survival benefit compared with remaining on dialysis after graft failure and awaiting possible non-ECD retransplantation. We confirmed earlier observations of the benefit of retransplantation with non-ECD kidneys (9,11).

Allograft failure continues to be an important problem faced by recipients of renal transplants. These patients are often considered for repeat renal transplantation. As with patients receiving a primary transplant, retransplantation candidates are also potential recipients of ECD kidneys. Indeed, 292 out of a total 9641 patients in our study with a history of a failed primary graft were retransplanted with an ECD kidney. While the survival benefit of receiving an ECD transplant over remaining on the waiting list has been demonstrated for first transplants (5), we could not document such a benefit for retransplantation with ECD kidneys. It is worth noting that Port et al. did not include candidates with a previous failed graft in their study that defined ECD kidneys, and Merion et al. omitted this group when trying to delineate the groups that benefit from ECD transplantation (4,7).

The lack of a significant ECD retransplantation mortality benefit contrasts with the non-ECD retransplantation benefit. While longer follow-up of the non-ECD group could potentially have explained this difference, examination of the data indicates that the follow-up time for both groups is comparable. Interestingly, the duration of primary graft function appears to have a differential effect on the outcome of ECD retransplantation. In patients with primary graft failure within a four-year period, we found no significant mortality benefit of retransplantation with an ECD kidney. However, if the primary graft functioned for at least four years, retransplantation with an ECD kidney was beneficial. Retransplantation with a non-ECD kidney, on the other hand, has a mortality benefit irrespective of the time to primary graft failure. An explanation for such an observation would be speculative, but it is possible that patients with early primary graft failure receive more aggressive immune suppression in an attempt to save the failing graft, with the attendant cumulative risks. Alternatively, patients losing their primary transplant within the first four years could represent a special group with some as-yet-unidentified risk factor that could place them at increased mortality risk.

To demonstrate the homogeneity of benefit, the dialysis versus retransplantation HRs over postretransplantation follow-up time was calculated for both the ECD and the non-ECD groups. There was significantly higher mortality in the first month after retransplantation followed by lower mortality than the reference group thereafter. This lower mortality was, however, not significantly different from the conventional therapy group for those patients with an ECD retransplant. The survival curves reflect this observation: the non-ECD group shows a sustained survival benefit, amplified with time, while the ECD group's survival mimics that of the comparison group. The higher mortality seen in the first month after transplantation concurs with similar observations made by Wolfe et al. in primary transplantation (2).

Merion et al. have shown that ECD transplantation confers a survival benefit only to select groups when those patients are compared with a similar cohort receiving standard therapy (7). The results of this study indicate that ECD kidneys may not be appropriate for retransplant candidates. It is, however, possible that over very long-term follow-up of these patients (i.e. beyond seven or eight years), a survival benefit of retransplantation compared with conventional therapy may become discernable.

Our results are subject to the limitations inherent in observational data. First of all, because patients are not randomly selected to receive ECD kidneys, it is possible that ECD recipients are in some unmeasured ways systematically less healthy than recipients of non-ECD kidneys. For example, we found that, compared with patients receiving non-ECD retransplants, patients who received an ECD retransplant were on average older and more likely to be diabetic. Naturally, despite being very strong mortality risk factors, age and diabetes were among the covariates adjusted for in the regression models, mitigating concerns about confounding with respect to these two factors (and other covariates included in the models). However, the systematically less favorable covariate pattern with respect to measured covariates among ECD relative to non-ECD retransplant recipients generates concern about the distribution across the two groups with respect to unmeasured covariates. Further fueling a hypothesis that ECD recipients are a select group is our own finding that, compared with non-ECD retransplant recipients, ECD retransplant recipients were more likely to have had an ECD primary transplant. Unfortunately, data are unavailable to measure, let alone remedy, any such selection bias.

The number of subjects in the study is relatively small, which could cause a type II error. This possibility is potentially problematic in two ways. First, it is possible that a significant reduction in mortality upon ECD retransplantation would be detected with a much larger sample size. Second, it is possible that the magnitude and possibly the direction of the ECD versus conventional therapy contrast depends on patient characteristics. Our ability to detect such interactions was limited by the relatively low number of ECD retransplants.

Other limitations include the fact that the comorbidities for which adjustments have been made are those reported at the time of initial listing. Also, all analyses are based on the practice of wait-listing after graft failure, and changes in the selection to the waiting list or to transplantation over time may modify the relative outcomes. In fact, future waiting times for ECD or non-ECD kidneys are unpredictable and could influence the relative benefits of the strategies we have compared. Finally, as highlighted in a recent study, there is a spectrum of quality among deceased donor kidneys that the ECD classification system oversimplifies (20). It is thus conceivable that certain organs, while meeting ECD criteria, would still confer survival benefit to recipients. The present analysis retains clinical relevance, given that current practice involves a binary decision regarding patient willingness to accept an ECD kidney. Refinement of the grading of deceased donor kidneys will no doubt be a focus of further research.

In conclusion, while significant advantage is confirmed for non-ECD retransplantation, our analysis demonstrates on average no survival benefit to performing retransplantation with ECD kidneys. Other potential benefits of retransplantation with ECD kidneys, such as improvements in health-related quality of life or cost utility, could still make this a preferable option for patients with a failed renal allograft. Such possible advantages should not be understated but are beyond the scope of the present analysis. Based on our findings on expected survival, though, use of ECD kidneys for second or subsequent transplantations should not be recommended over remaining on the waiting list until a non-ECD kidney becomes available.

Acknowledgments

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

This research was presented, in part, as a free communication at the 2006 World Transplant Congress in Boston, MA. The statistical methodology development and analysis for this investigation was supported by National Institutes of Health grant R01 DK-70869 to the second author. The SRTR is funded by contract number 234-2005-37009C from the Health Resources and Services Administration (HRSA), U.S. Department of Health and Human Services. The views expressed herein are those of the authors and not necessarily those of the U.S. Government. This study was approved by HRSA's SRTR project officer. HRSA has determined that this study satisfies the criteria for the IRB exemption described in the ‘Public Benefit and Service Program’ provisions of 45 CFR 46.101(b)(5) and HRSA Circular 03.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Russell JD, Beecroft ML, Ludwin D, Churchill DN. The quality of life in renal transplantation—A prospective study. Transplantation 1992; 54: 656660.
  • 2
    Wolfe RA, Ashby VB, Milford EL et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1992; 341: 17251730.
  • 3
    OPTN/SRTR Annual Report 1994–2003. HHS/HRSA/HSB/DOT; UNOS; URREA. 2004.
  • 4
    Port FK, Bragg-Gresham JL, Metzger RA, et al. Donor characteristics associated with reduced graft survival: An approach to expanding the pool of kidney donors. Transplantation 2002; 74: 12811286.
  • 5
    Ojo AO, Hanson JA, Meier-Kriesche H, et al. Survival in recipients of marginal cadaveric donor kidneys compared with other recipients and wait-listed transplant candidates. J Am Soc Nephrol 2001; 12: 589597.
  • 6
    Schaubel D, Wolfe RA, Port FK. A sequential stratification method for estimating the effect of a time-dependent experimental treatment in observational studies. Biometrics 2006; 62: 910917.
  • 7
    Merion RM, Ashby VB, Wolfe RA, et al. Deceased-donor characteristics and the survival benefit of kidney transplantation. JAMA 2005; 294: 27262733.
  • 8
    Cohen DJ, St Martin L, Christensen LL, Bloom RD, Sung RS. Kidney and pancreas transplantation in the United States, 1995–2004. Am J Transplant 2006; 6: 11531169.
  • 9
    Ojo A, Wolfe RA, Agodoa LY et al. Prognosis after primary renal transplant failure and the beneficial effects of repeat transplantation: Multivariate analyses from the United States Renal Data System. Transplantation 1998; 66: 16511659.
  • 10
    Rao PS, Schaubel D, Wei G, Fenton S. Evaluating the survival benefit of retransplantation. Transplantation 2006; 82: 669674.
  • 11
    Hirata M, Terasaki PI. Renal retransplantation. Clin Transpl 1994: 419433.
  • 12
    Abouljoud MS, Deierhoi MH, Hudson SL, Diethelm AG. Risk factors affecting second renal transplant outcome, with special reference to primary allograft nephrectomy. Transplantation 1995; 60: 138144.
  • 13
    Gjertson DW. A multi-factor analysis of kidney regraft outcomes. Clin Transpl 2002: 335349.
  • 14
    Ishikawa N, Tanabe K, Tokumoto T et al. Risk factors affecting the long-term results of renal retransplantation. Transplant Proc 1999; 31: 28582859.
  • 15
    Arndorfer JA, Meier-Kriesche HU, Ojo AO et al. Time to first graft loss as a risk factor for second renal allograft loss. Transplant Proc 2001; 33: 11881189.
  • 16
    El-Agroudy AE, Wafa EW, Bakr MA et al. Living-donor kidney retransplantation: Risk factors and outcome. BJU Int 2004; 94: 369373.
  • 17
    Sellers MT, Velidedeoglu E, Bloom RD et al. Expanded-criteria donor kidneys: A single-center clinical and short-term financial analysis—Cause for concern in retransplantation. Transplantation 2004; 78: 16701675.
  • 18
    Cox D. Regression models and life tables. J Royal Stat Soc 1972; 34: 187220.
  • 19
    Rabbat CG, Thorpe KE, Russell JD, Churchill DN. Comparison of mortality risk for dialysis patients and cadaveric first renal transplant recipients in Ontario, Canada. J Am Soc Nephrol 2000; 11: 917922.
  • 20
    Schold JD, Kaplan B, Baliga RS, Meier-Kriesche H-U. The broad spectrum of quality in deceased donor kidneys. Am J Transplant 2005; 5: 757765.