Glomerulonephritis (GN) is a leading cause of end-stage renal disease (ESRD) and subsequent need for renal replacement therapy.1 Despite the high prevalence of ESRD secondary to primary GN, long-term transplant outcomes, particularly in the less common glomerulonephritidies, remain inconclusive. A major cause of graft loss in patients with primary GN is posttransplant disease recurrence (PTDR), that is, the return of the original disease in the kidney allograft. The incidence of PTDR, typically a late complication, has paradoxically increased,2 as short-term graft survival has improved.3,4 The risk of resulting graft loss varies considerably depending on the type of GN,4 also within individual GNs, there is a substantial disparity in the reported rate of graft loss.5 These reported differences may be explained in part by variations in study designs.6,7 Therefore, there is the need for a large study to investigate long-term outcomes in transplant recipients with primary GN.
Furthermore, because of the genetic nature of some primary GNs, it remains unclear whether the recipients of living related donor (LRD) transplants are at a greater risk of graft loss from PTDR compared with the recipients of deceased donor (DD) transplants.7-10 Most international guidelines recommend the use of LRD transplants in patients with primary GN, though with limited supporting evidence.8-11 The conflicting evidence regarding the risk of graft loss from primary GN, in particular due to kidney donor type, makes pretransplant counselling difficult for the transplant team and the prospective donor-recipient pair.
Using the European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Registry database, we aimed, first to quantify the 15-year kidney allograft survival probability in patients with primary GN and second to compare the risk of graft failure in renal transplants in patients with primary GN in which the native renal disease can recur to those with autosomal dominant polycystic kidney disease (ADPKD) in which the kidney disease cannot recur. Third, we aimed to determine if the risk of kidney allograft loss in those with primary GN differed across donor types.
MATERIALS AND METHODS
The ERA-EDTA Registry collects data on patients undergoing renal replacement therapy from national and regional renal registries on an annual basis.12 Seventeen renal registries which provided accurate and complete data from at least 1994 were included: Austria, Dutch- and French-speaking Belgium, Denmark, Finland, Greece, Iceland, Norway, the Spanish regions of Andalusia, Asturias, Basque Country, Catalonia, Cantabria and Valencia, Sweden, the Netherlands, and Scotland (United Kingdom).
We included patients aged 18 years or older at the time of receiving a first kidney transplant, between 1991 and 2010, with one of the following primary GNs (ERA-EDTA Registry codes in brackets): immunoglobulin A nephropathy (IgAN, 12), membranoproliferative GN (MPGN) type I (13), MPGN type II (14), membranous nephropathy (MN, 15), and focal segmental glomerulosclerosis (FSGS, 17). Patients with a primary renal diagnosis where the exact GN type could not be determined, that is, “glomerulonephritis, histologically not examined” (10), “crescentic (extracapillary) glomerulonephritis” (16), and “glomerulonephritis, histologically examined, not given above” (19) formed a further group called “other.”
The primary outcome was “death-adjusted graft survival” where graft failure was the event, and the analyses were adjusted for death with a functioning graft. The adjustment for death with a functioning graft depended on the type of analyses; either by treating death as a competing risk or by censoring for death with a functioning graft (see also below).14 The secondary outcome was “graft survival” in which both death with a functioning graft and graft failure were events. In all survival analyses, the date of kidney transplantation was taken as the starting point, the end of follow-up time was set at December 31, 2011, and all analyses were censored for loss to follow-up.
Unadjusted 5-, 10-, and 15-year graft survival probabilities with 95% confidence intervals were obtained. First, the cumulative incidence competing risk method was used to calculate death-adjusted graft survival (where death with a functioning graft was treated as a competing risk, as the events graft failure and death with a functioning graft are mutually exclusive).14 Second, the Kaplan-Meier method was used to calculate graft survival probabilities.
Two models were used to examine the risk of death-adjusted graft failure (ie, censored for death with a functioning graft) and of graft failure14:
Model 1. Individual types of GN compared with ADPKD: Cox regression analysis was performed comparing the risk of both death-adjusted graft failure and graft failure for each individual GN group with the ADPKD control group considering all the donor types combined. The control group consisted of patients aged 18 years or older at the time of receiving a first kidney transplant within the same period with a primary renal diagnosis of ADPKD. As the ERA-EDTA Registry database does not include information on the causes of graft loss, we assumed that most of the additional graft loss in the GN group compared with graft loss in patients with ADPKD, in which the native kidney disease cannot recur, could be attributed to PTDR. The χ2 test and Mann-Whitney U test were used to compare the characteristics of the cases and ADPKD controls.
To verify our choice of ADPKD as an appropriate control group, a sensitivity analysis was performed by repeating the multivariable Cox regression analysis of model 1 using a different control group, consisting of all adult first kidney transplant recipients with a primary renal disease of pyelonephritis (ERA-EDTA codes, 20-25, 29), congenital anomalies of the kidney and urinary tract (30-34, 39), or tubulointerstitial nephritis (60-61, 63, 66).
Model 2. Living donor (LD) versus DD kidney transplantation by GN: A Cox regression model was used to compare the risk of both death-adjusted graft failure and graft failure between LD, LRD, and living unrelated donor (LUD) grafts to the corresponding DD grafts for each GN group individually. In all cases, the DD transplants within the GN group were used as the control (ie, IgAN LRD compared with IgAN DD grafts).
For both models 1 and 2, adjusted hazard ratios with 95% confidence intervals were calculated at 5, 10, and 15 years after transplantation with adjustments for time on dialysis, age at transplantation, sex, country, and the transplant era (1991-1995, 1996-2000, 2001-2005, 2006-2010). In model 1, additional adjustments were made for donor type (LD vs DD).
A 2-tailed P value of less than 0.05 was considered statistically significant. Analyses were performed using SAS version 9.3 and R version 3.0.2.
Within the study period, 55 741 first kidney transplants were performed in adult recipients, of which 14 383 recipients had a potential diagnosis of primary GN, 6 327 with a definitive diagnosis of primary GN, (IgAN, MPGN I, MPGN II, MN, and FSGS) were included along with 8 056 recipients in the group termed “other.” Median follow-up after transplantation was 74.3 months (interquartile range, 36–124) for transplant recipients with primary GN.
Patient characteristics are presented in Table 1. Overall, the median age of patients in the GN group was 47.5 years (interquartile range, 35.6-57.5) at the time of transplantation and 71.2% were men.
The death-adjusted graft survival and graft survival by primary GN, “other” GNs, and ADPKD are presented in Figure 1 (left and right panels, respectively). Membranoproliferative GN I and II had the lowest 15-year death-adjusted graft survival probabilities (56.7% and 56.5%, respectively), whereas IgAN had the highest 15-year death-adjusted graft survival (68.9%) of all GNs. The 15-year death-adjusted graft survival probability for those with a PRD of ADPKD was 76.6%.
Model 1: Individual Type of GN Compared to ADPKD
The 5-, 10- and 15-year adjusted relative risks of death-adjusted graft failure and graft failure for individual primary GNs and the “other” group as compared with ADPKD are presented in Figure 2 and Figure S1 (SDC,http://links.lww.com/TP/B214), respectively. All GNs had a greater risk of death-adjusted graft failure compared with ADPKD, with the exception of IgAN. For IgAN, there was no difference in the risk of death-adjusted graft failure until 10 years after transplantation, thereafter the risk of graft loss was greater in the IgAN group.
Model 2: LD Versus DD Transplants by GN
Figure 3A and Figure S2 (SDC,http://links.lww.com/TP/B214) present the 5-, 10- and 15-year adjusted risk of death-adjusted graft failure and graft failure for LD compared with DD transplants, by individual GNs, respectively. Death-adjusted graft failure for the LD transplants in the IgAN, MN, and “other” group was lower than that for the corresponding DD transplants. However, this graft survival advantage became less apparent when divided into the smaller LRD and LUD subgroups (Figure 3B and Figure S3, SDC,http://links.lww.com/TP/B214). The MPGN I and II showed no difference in death-adjusted graft failure when LD and DD transplants were compared. Within the FSGS group, LD and more specifically LRD transplants had a lower risk of death-adjusted graft failure at 5 and 10 years compared with the FSGS DD transplants but not at 15 years.
Repeating the analysis from model 1 with a control group comprising kidney transplant recipients with pyelonephritis, congenital anomalies of the kidney and urinary tract or tubulointerstitial nephritis produced similar results within the primary GN groups (Table S4, SDC,http://links.lww.com/TP/B214).
In this large cohort of 6 327 kidney transplant recipients with a PRD of primary GN and a long follow-up period, we found that all GN types had a 15-year death-adjusted graft survival probability of over 55%. Additionally, we compared graft loss in patients with various primary GNs, in which the return of the original disease in the kidney allograft can occur, to those with ADPKD. All types of primary GN had a greater risk of death-adjusted graft failure compared to ADPKD, with the exception of IgAN, which had a similar risk of death-adjusted graft failure and graft failure up to 10 years after transplantation. In addition, the expected survival advantage normally seen in LD kidney transplants over DD transplants was not apparent in the MPGN I and II groups, but present in the IgAN, MN, and FSGS groups.
Immunoglobulin a Nephropathy
It is well known that kidney transplant recipients with IgAN have a better short-term graft survival,15 possibly due to immunological benefits, though recent studies with long-term follow-up have reported that the superior graft outcomes initially enjoyed by IgAN recipients may be lost between 8 and 12 years after transplantation.13 To date, our study contains the largest number of kidney transplant recipients with IgAN and a long follow-up time. The results indicate that IgAN may eventually have a detrimental effect on long-term kidney allograft outcome.
From previous publications, it is not entirely clear whether the risk of PTDR in LD transplants is greater than that in the DD transplants in IgAN, predominantly due to the small sample sizes of the available studies.17 Although we cannot confirm the causes of graft loss in our cohort, the results of our large study clearly show that LD transplants in IgAN have a lower risk of death-adjusted graft failure compared with DD transplants in IgAN. Thus, our findings support the notion that there need be no contraindications to LD kidney transplantation in patients with IgAN.
Membranoproliferative Glomerulonephritis I and II
When comparing the LD transplant subtypes to DD transplants in both MPGN I and II, the expected survival advantage normally seen for LD transplants was no longer apparent, though the outcome was no worse than that for DD transplants. Neither MPGN I nor MPGN II is listed as a contraindication to LD kidney transplantation in a number of transplant guidelines.9,10 However, some studies still recommend caution when accepting LRD grafts for patients with MPGN II due to a higher risk of PTDR.9 The published data on PTDR in MPGN I with respect to donor source are limited and conflicting.16,18,19 In contrast to our findings, a study by Angelo et al20 based on the United Network for Organ Sharing database, reported better graft survival for LD kidney transplants. It is difficult to draw any firm conclusions or compare the results with our findings for a number of reasons. First, our study included only adult patients; differences in outcomes between adult and pediatric populations, particularly for MPGN II, have previously been described.21 Second, neither study had access to biopsy data. It has been postulated that the disease severity as manifested by the presence of crescents on the initial native kidney biopsy rather than the type of disease alone predicts transplant outcomes,22 further supporting the argument that the heterogeneity of most types of GNs makes any predictions of outcome difficult. Finally, although this study and that of Angelo et al represent the largest published cohorts of transplanted patients with MPGN II, the rarity of this disease and thus the relatively small sample size still make it difficult to compare outcomes from LRD and LUD grafts.
The Australia and New Zealand Dialysis and Transplant Registry reported 12.5% renal graft loss from recurrent MN at 10 years post transplant.2 Accordingly, we would have expected a higher risk of death-adjusted graft failure in the MN group compared with the ADPKD group, which was indeed the case.
Since the identification of M-type phospholipase A2 receptor autoantibodies (PLA2R) as potentially pathogenic in idiopathic MN,23 several studies have implicated PLA2R in the pathogenesis of recurrent idiopathic MN.24 In a study of 23 transplant recipients with idiopathic MN, Andrésdóttir and Wetzels25 reported a cumulative rate of PTDR at 3 years of 70% in LRD and 21% in LUD transplants. They hypothesized that genetic factors, such as PLA2R, may therefore play an important role in disease recurrence. Based on this hypothesis, we would expect to see a worse graft survival in LRD transplants compared with DD transplants in patients with MN; however, we found that at 5 and 10 years, LRD transplants in the MN group had a lower risk of death-adjusted graft failure. From our data, it is not possible to distinguish idiopathic MN from secondary MN. The possibility that some cases of secondary MN were reported as idiopathic MN, and the small sample size in this study may have masked any potential graft survival outcomes from differing LD types.
Focal Segmental Glomerulosclerosis
The FSGS is the most frequently recurring primary GN after kidney transplantation, developing in approximately 30% of transplants.26 Our study found that those with a first kidney transplant for ESRD secondary to FSGS were, by 15 years, 77% more likely to experience death-adjusted graft failure compared with ADPKD.
Transplantation guidelines do not exclude LRD transplants in patients with FSGS but suggest caution and, in particular, avoidance in those known to be at high risk of recurrence.9 Many studies report a better graft survival in LD transplants compared with DD transplants in patients with FSGS,27-29 particularly in pediatric cohorts.30 The majority of studies examining PTDR in FSGS include only pediatric or mixed adult and pediatric cohorts, whereas our study comprised only adult recipients. Our results suggest that in an adult European population, LD, and in particular LRD kidney transplants have a reduced risk of death-adjusted graft failure compared with DD transplants in patients with FSGS.
Within this study, we report on the long term graft survival of a large cohort of transplant recipients with primary GN. We additionally investigated the risk of graft loss between patients with ESRD secondary to primary GN, in which the native kidney disease can recur in the transplanted kidney and in patients with ADPKD in which the native kidney disease cannot recur in the transplanted kidney. We do not have the causes of graft loss in our cohort. Therefore, we assumed that most of the additional graft loss in the GN group compared with the ADPKD group could be attributed to PTDR. It should be noted, however, that other factors could be responsible for the increased risk of graft loss, such as the type of immunosuppression used and the pretransplantation clinical status of the patients. However, information regarding these potential confounding factors are not available within the ERA-EDTA Registry data. Some studies have observed better kidney allograft outcomes in patients with ADPKD compared with other causes of ESRD,31 which may have resulted in overestimation of the relative risk of graft loss when comparing patients with primary GN to those with ADPKD. Using different control groups, we obtained similar results supporting the assumption that the ADPKD group is suitably representative of a control group with no possibility of PTDR.
A large proportion of the patients had a diagnosis of GN but not of 1 of the 5 primary GNs (IgAN, MPGN I, MPGN II, MN, and FSGS), on which this study focused. These patients were categorized as “other” on the basis that the assigned PRD codes (“glomerulonephritis, histologically not examined,” “crescentic (extracapillary) glomerulonephritis,” and “glomerulonephritis; histologically examined, not given above”) did not provide any information as to the underlying type of GN. Although little can be said with regards to the clinical significance of this group, the fact that the results of this large group were not dissimilar to the other primary GN groups is reassuring.
In this study, we also investigated the risk of graft loss between patients with ESRD secondary to primary GN, who received an LD transplant compared with a DD transplant. Despite the very large sample size, the smaller LD subgroups and in particular the lack of distinction between LRD and LUD in a proportion of our data mean that in some cases firm conclusions cannot be drawn. Some types of GN represent histological patterns rather than well-defined diseases, and therefore we are unable to distinguish between certain GN subgroups using the current coding system.
Furthermore, for a subset of patients, a presumed diagnosis of ESRD secondary to GN is likely to have been applied without histological confirmation. The diagnosis of a specific form of GN may have been confirmed histologically once PTDR had occurred resulting in a change of the ERA-EDTA code. The reclassification of these patients could potentially have increased the number of patients in the cohort with graft loss secondary to GN, leading to an overestimation of the risk of graft loss.
This study has detailed the actual 15-year kidney allograft survival outcomes in patients with primary GN within a number of European countries. We found that all GN types had a 15-year death-adjusted graft survival probability of over 55%. Kidney transplants in patients with all types of primary GN, in which the native disease could recur, including IgAN, eventually have a greater risk of 15-year death-adjusted graft failure compared to ADPKD in which the original disease cannot recur. The expected graft survival advantage in LD over DD kidney transplants was apparent in patients with IgAN, MN, and FSGS, although in FSGS, the advantage was observed in LRD, while not in LUD, which may be due to the small sample size. In MPGN I and II, no survival benefit of LD versus DD grafts was seen. This study confirms that the reluctance to use LRD in some primary GNs remains unfounded. The results of this study are important to aid nephrologists in the pretransplant counselling of potential renal transplant recipients and donors about risk of graft loss, particularly when considering donor type.
The authors thank the patients and staff of all the dialysis and transplant units who have contributed data via their national and regional renal registries. Furthermore, the authors gratefully acknowledge the following registries and persons for their participation in the data collection: R. Kramar [Austrian Dialysis and Transplant Registry (OEDTR)]; R Alonso de la Torre, JR Quirós, and RERCA working group (Asturian Renal Registry); Á. Magaz, J. Aranzabal, M Rodrigo, I Moina (Basque Country Renal Registry); E. Arcos, J. Comas, and J Tort [Catalan Renal Registry (RMRC) and Catalan Transplant Organization (OCATT)]; J-M. des Grottes and F. Collart (French-speaking Belgium Registry); G.A. Ioannidis (Greek national Renal Registry); O Zurriaga Llorens and M Ferrer Alamar (Valencian Renal Registry).
1. Matas AJ, Smith JM, Skeans MA, et al. OPTN/SRTR 2011 Annual Data Report: kidney. Am J Transplant
. 2013;13(Suppl 1):11–46.
2. Briganti EM, Russ GR, McNeil JJ, et al. Risk of renal allograft loss from recurrent glomerulonephritis. N Engl J Med
3. Chang SH, Russ GR, Chadban SJ, et al. Trends in kidney transplantation in Australia and New Zealand, 1993–2004. Transplantation
4. Chadban S. Glomerulonephritis recurrence in the renal graft. J Am Soc Nephrol
5. Choy BY, Chan TM, Lo SK, et al. Renal transplantation in patients with primary immunoglobulin A nephropathy. Nephrol Dial Transplant
6. Floege J. Recurrent glomerulonephritis following renal transplantation: an update. Nephrol Dial Transplant
7. Ponticelli C, Traversi L, Feliciani A, et al. Kidney transplantation in patients with IgA mesangial glomerulonephritis. Kidney Int
8. European Renal Best Practice Transplantation Guideline Development Group. ERBP Guideline on the Management and Evaluation of the Kidney Donor and Recipient. Nephrol Dial Transplant
. 2013;28(Suppl 2):ii1–ii71.
9. Andrews PA, Burnapp L, Manas D, et al. British Transplantation Society; Renal Association. Summary of the British Transplantation Society/Renal Association U.K. guidelines for living donor kidney transplantation. Transplantation
11. Knoll G, Cockfield S, Blydt-Hansen T, et al. Kidney Transplant Working Group of the Canadian Society of Transplantation. Canadian Society of Transplantation consensus guidelines on eligibility for kidney transplantation. CMAJ
12. Pippias M, Stel VS, Abad Diez JM, et al. Renal replacement therapy in Europe: a summary of the 2012 ERA-EDTA Registry Annual Report. Clin Kidney J
13. Moroni G, Longhi S, Quaglini S, et al. The long-term outcome of renal transplantation of IgA nephropathy and the impact of recurrence on graft survival. Nephrol Dial Transplant
14. Noordzij M, Leffondré K, van Stralen KJ, et al. When do we need competing risks methods for survival analysis in nephrology? Nephrol Dial Transplant
15. Andresdottir MB, Haasnoot GW, Doxiadis II, et al. Exclusive characteristics of graft survival and risk factors in recipients with immunoglobulin A nephropathy: a retrospective analysis of registry data. Transplantation
16. Karakayali FY, Ozdemir H, Kivrakdal S, et al. Recurrent glomerular diseases after renal transplantation. Transplant Proc
17. Wang AY, Lai FM, Yu AW, et al. Recurrent IgA nephropathy in renal transplant allografts. Am J Kidney Dis
18. Andresdottir MB, Assmann KJ, Hoitsma AJ, et al. Recurrence of type I membranoproliferative glomerulonephritis after renal transplantation: analysis of the incidence, risk factors, and impact on graft survival. Transplantation
19. Lorenz EC, Sethi S, Leung N, et al. Recurrent membranoproliferative glomerulonephritis after kidney transplantation. Kidney Int
20. Angelo JR, Bell CS, Braun MC. Allograft failure in kidney transplant recipients with membranoproliferative glomerulonephritis. Am J Kidney Dis
21. Nasr SH, Valeri AM, Appel GB, et al. Dense deposit disease: clinicopathologic study of 32 pediatric and adult patients. Clin J Am Soc Nephrol
22. Little MA, Dupont P, Campbell E, et al. Severity of primary MPGN, rather than MPGN type, determines renal survival and post-transplantation recurrence risk. Kidney Int
23. Beck LH Jr, Bonegio RG, Lambeau G, et al. M-type phospholipase A2 receptor as target antigen in idiopathic membranous nephropathy. N Engl J Med
24. Debiec H, Martin L, Jouanneau C, et al. Autoantibodies specific for the phospholipase A2 receptor in recurrent and De Novo membranous nephropathy. Am J Transplant
25. Andrésdóttir MB, Wetzels JF. Increased risk of recurrence of membranous nephropathy after related donor kidney transplantation. Am J Transplant
26. Pardon A, Audard V, Caillard S, et al. Risk factors and outcome of focal and segmental glomerulosclerosis recurrence in adult renal transplant recipients. Nephrol Dial Transplant
27. Abbott KC, Sawyers ES, Oliver JD 3rd, et al. Graft loss due to recurrent focal segmental glomerulosclerosis in renal transplant recipients in the United States. Am J Kidney Dis
28. Nehus EJ, Goebel JW, Succop PS, et al. Focal segmental glomerulosclerosis in children: multivariate analysis indicates that donor type does not alter recurrence risk. Transplantation
29. Maas RJ, Deegens JK, van den Brand JA, et al. A retrospective study of focal segmental glomerulosclerosis: clinical criteria can identify patients at high risk for recurrent disease after first renal transplantation. BMC Nephrol
30. Baum MA, Stablein DM, Panzarino VM, et al. Loss of living donor renal allograft survival advantage in children with focal segmental glomerulosclerosis. Kidney Int
31. Jacquet A, Pallet N, Kessler M, et al. Outcomes of renal transplantation in patients with autosomal dominant polycystic kidney disease: a nationwide longitudinal study. Transpl Int