Portal de Periódicos da CAPES

RECURRENT GLOMERULONEPHRITIS FOLLOWING RENAL TRANSPLANTATION

1997; Wolters Kluwer; Volume: 63; Issue: 8 Linguagem: Inglês

10.1097/00007890-199704270-00001

1534-6080

Peter Kotanko, C. D. Pusey, Jeremy Levy,

Vasculitis and related conditions

Glomerulonephritis (GN*) is the most important cause of end-stage renal disease (ESRD) leading to renal transplantation. In the United Kingdom and Ireland, 27% of renal transplant recipients have GN as their underlying disease (1). In the Eurotransplant Registry, 31-38% of patients receiving a first graft have GN as the cause of ESRD (J. De Meester and G.G. Persijn, Eurotransplant, Leiden, The Netherlands, personal communication, 1996). Recurrence of nephritis in a transplant and subsequent graft failure were first recognized in 1955, when a patient with ESRD due to vasculitis developed recurrent crescentic GN (2). Recurrent disease affects 10% of allografts, and 5% of graft loss can be attributed to recurrent GN (3). Analysis of the recent data documented in the United Kingdom Transplant Support Service Authority (UKTSSA) Renal Transplant Audit 1984-1993 suggests that currently 4% of graft loss in adults, and 7% of graft loss in children, is due to recurrent disease (1). The same rate of graft loss is seen in the Eurotransplant Registry data (J. De Meester and G.G. Persijn, Eurotransplant, personal communication, 1996), with an overall graft loss of 4.4% due to recurrent GN at 5 years after transplantation. However, the frequency of graft loss due to recurrent GN is substantially and significantly increased in grafts from living donors compared with grafts from cadaveric donors (16.2% vs. 4% at 5 years; χ2=12.5, P=0.00046). This difference is most marked in the first year after transplantation (Table 1). Posttransplantation GN in principle could be caused by truly recurrent GN, by the development of de novo GN in the transplant, or by transplanted GN (from a donor with previously unrecognized GN), and must be distinguished from other causes of posttransplantation glomerulopathy (chronic rejection and cyclosporine induced). Knowledge of the recipient's primary renal disease, the nature of the graft GN, and the absence of primary renal disease in the donor are sine qua nons for the diagnosis of "recurrent glomerulonephritis." In reality, however, all three questions can be answered only in a minority of transplant recipients. The UKTSSA Renal Transplant Audit data reveal that while 1809 graft transplants were performed in recipients with histologically confirmed GN (16% of all grafts), 1271 grafts (11%) were placed in patients with suspected but unconfirmed GN (1). For the more common nephritides, especially IgA disease, distinguishing recurrence from transplanted GN can be problematic, and occasionally de novo GN arising in the graft is more frequent than recurrence (e.g., membranous GN). A much more difficult and frequent problem is the differentiation of recurrent or de novo GN from chronic rejection. Some recurrent glomerulonephritides, especially mesangiocapillary glomerulonephritis type I and membranous nephropathy, may closely resemble the histological appearance of allograft glomerulopathy. However, the conditions should be distinguishable by the presence of interstitial and vascular abnormalities in most cases of chronic rejection, by the immunohistological (or electron microscopic) findings, and by the clinical presentation (see below). The use of grafts from living related donors (LRD) merits special consideration, because of the potential for recurrence leading to early graft loss and the increased risk of some nephropathies in families, with later consequences for both donor and recipient. Furthermore, recurrent GN causing graft loss is much more common in LRD grafts than in cadaveric grafts (see above and Table 1). In this Overview, we discuss the difficulties in diagnosing recurrent GN, and the incidence, presentation, outcome, and management of recurrence (summarized in Table 2). We will concentrate on those areas in which new data are available. PRIMARY FOCAL SEGMENTAL GLOMERULOSCLEROSIS The last few years have seen a major advance in our understanding of recurrent focal segmental glomerulosclerosis (FSGS), with the recognition of circulating factors that are capable of inducing changes in glomerular permeability. The course of untreated FSGS is usually one of progressive proteinuria and declining renal function. FSGS is an important cause of ESRD in adults, accounting for 7% to 15% of patients requiring renal replacement therapy. The overall recurrence rate has been reported to be 20-30% (4-6). However, the focal nature of the lesion in FSGS can lead to significant sampling error in kidney biopsies, and thus underreporting of recurrence. Furthermore, the histological changes characteristic of FSGS may resemble the end result of hemodynamic injury as a consequence of reduced nephron mass. Despite these problems, recurrence is generally indicated by proteinuria that develops hours to weeks after renal transplantation (4, 5). Proteinuria that develops more than a few months after renal transplantation is unlikely to be due to recurrence of FSGS. In children, the mean time to recurrence is 14 days (5). Risk of recurrence of FSGS is substantially increased in patients below the age of 15 years, in patients with a malignant course of disease with evolution to ESRD in under 3 years, in those with evidence of mesangial proliferation in the native kidney, and in those patients with recurrence of FSGS in a first graft (4, 5). Thus, in children less than 5 years of age, the recurrence rate is about 50%; in children under 15 years at onset who progress to ESRD within 3 years and also have mesangial expansion, the recurrence rate is 80-100% (4-6). If a first graft was lost due to FSGS, the likelihood of FSGS affecting a subsequent graft is about 75% to 85% (6, 7). In adults with no additional risk factors, the recurrence rate is reduced to 10-15%. Black patients seem to have a reduced risk of recurrent disease (5). Recurrence of FSGS is not benign, and leads to graft loss in 40-50% of patients with recurrence (6). Patients with recurrent disease are also more likely to have episodes of acute renal failure (ARF) and acute rejection (8). Kim et al. (8) reported ARF to occur in 16 of 26 patients with recurrent FSGS, compared with 7 of 47 patients without recurrence. One or more acute rejection episodes occurred in all patients with ARF, with or without recurrent FSGS, and in 7 of 10 patients with recurrence without ARF. Only 11 of 40 patients with neither recurrence nor ARF had acute rejection (8). The reason for this increased rate of ARF and acute rejection is unclear. However, recurrent FSGS with concomitant proteinuria and dyslipidemia may cause increased expression of cell adhesion molecules (9) and up-regulation of major histocompatibility complex and cell surface accessory molecules, and thus stimulate an immune response. The postulated circulating albuminuric factor in FSGS (see below) may also play a role in immune dysregulation. In view of the high rate of recurrence, particularly in the risk groups described above, the use of LRD grafts has to be entertained with great care. If, however, no recurrence was noted in a first graft, a subsequent renal transplantation is likely to be safe, allowing regrafting from a LRD (7). Anecdotal strategies for the management of recurrent FSGS include treatment with nonsteroidal anti-inflammatory drugs, angiotensin-converting enzyme (ACE) inhibitors, cyclosporine (CsA), FK506, dipyridamole, plasmapheresis, and protein A immunoadsorption (6). CsA has not been shown to affect the course of recurrent FSGS (6). Changing from CsA to FK506 does not improve FSGS recurrence. In pediatric patients, recurrent FSGS was successfully treated with oral cyclophosphamide (10) or an increased dose of CsA (11). ACE inhibitors and nonsteroidal anti-inflammatory drugs have been shown to decrease proteinuria in a few cases, often at the expense of a reduced glomerular filtration rate. The postulated presence of a circulating albuminuric factor has stimulated the use of plasma exchange and immunoadsorption (or protein adsorption) in recurrent FSGS (12). Beneficial effects were seen in a single-center study in patients treated with plasma exchange and CsA in the early stage of recurrence, when histological examination showed only fusion of foot processes without focal segmental sclerosis (12). Protein A immunoadsorption was shown to induce transient remission of proteinuria in patients with recurrent FSGS treated with one to three cycles of two to seven adsorption sessions, while the immunosuppressive regimens were not changed (13). In all but one patient, the effect of adsorption was limited in time, with a return to the preadsorption level of protein excretion within a maximum of 2 months. The administration of material eluted from the protein A adsorption column to rats increased their urinary albumin excretion (13). Recently, a circulating serum factor was identified in some patients with recurrent FSGS (14) that was shown to bind protein A and had a molecular mass of 50 kDa. This factor caused increased glomerular permeability to albumin in vitro in isolated rat glomeruli. The permeability activity was strongly associated with the recurrence of FSGS, such that in patients with the highest in vitro albumin permeability, the recurrence rate was 100%. The reduction of the activity of this factor by plasmapheresis was paralleled by decreased urinary protein excretion (14, 15). The discovery and partial characterization of a permeability factor involved in the recurrence of FSGS after renal transplantation provide a pathophysiological basis for the use of plasma exchange or protein A immunoadsorption in this setting. Prospective studies are needed to define the role of these procedures in the treatment of recurrent FSGS. Bilateral pretransplant nephrectomy has been advocated to reduce the likelihood of recurrence after renal transplantation in the prospective recipient at high risk for developing recurrent GN. However, for FSGS patients, Odorico et al. (16) demonstrated a detrimental effect of pretransplant nephrectomy on recurrence rates and incidence of graft loss due to recurrent disese, independent of other variables. MESANGIAL IMMUNOGLOBULIN A NEPHROPATHY Recurrence of mesangial immunoglobulin A nephropathy (IgAN) usually presents with microscopic hematuria, with or without mild proteinuria, 2 months to 4 years after transplantation. Since 1982, recurrence has been reported in 25-60% of patients (17, 18). However, these figures may be underestimates of the true recurrence rate, as clinically unaffected graft recipients rarely receive biopsies. Overall, graft survival is increased rather than decreased in IgAN compared with other renal disease (19), and graft failure due to recurrence is rare (less than 10% [20, 21]). Some patients develop anti-HLA IgA antibodies after transplantation that seem to be beneficial (20). In one report, the 2-year graft survival rate was 100% in patients with such antibodies (20). Little is known of the influence of CsA on the recurrence rate. Initial studies suggested that CsA might mitigate the clinical course and the rate of graft loss caused by IgAN recurrence (22); however, the results of a 1996 study of a large series of patients with IgAN receiving transplants revealed that CsA did not reduce the incidence or severity of recurrent disease (23). The progression rate and therapeutic responsiveness of IgAN in the native kidney seem to be influenced by the genetic polymorphism of the ACE gene. Data from two studies indicate that the homozygote DD polymorphism in the ACE gene is a risk factor for progression to chronic renal failure in IgAN (24, 25). The impact of these polymorphisms on the incidence and severity of recurrent IgAN remains to be determined. IgAN has a genetic component, and familial clusters have been reported (26). This poses a possible extra risk for recurrence in grafts from LRD. However, the risk of significant recurrence is so low that in practice LRD can be used in patients with ESRD due to IgAN (27), although a complete family evaluation and possibly a pretransplant biopsy should be considered in this setting. The increased risk of recurrence in grafts from LRD may be related to genetic susceptibility, since there are associations between HLA antigens (particularly B35 and DR4) and IgAN. No treatment is effective in recurrent IgAN, nor is it necessary given the low rate of graft failure consequent upon recurrence. Crescentic recurrent IgAN has been treated with plasma exchange and increased immunosuppression (28). HENOCH-SCHÖNLEIN PURPURA Henoch-Schönlein purpura (HSP) and IgAN probably represent two ends of the spectrum of the same disease. Between 1984 and 1993, 87 patients with HSP received transplants in the United Kingdom (0.8% of all renal cadaveric grafts) (1). No recent data have been published on recurrent disease after transplantation. IgA deposition may recur in one third to three quarters of cases, although clinical recurrence is rare. Cameron (17) reviewed nine reported patients with graft failure due to recurrent, biopsy-confirmed HSP. The graft was lost in six patients, and three patients had recurrent purpura. At least seven of the nine grafts came from LRD. Time of onset of recurrence ranged from immediately to 20 months. From these limited data, it appears that organs from LRD should probably not be used. Transplantation in those still having recurrent attacks of purpura may be hazardous (29). A disease-free interval of at least 8-12 months before transplantation reduces the risk of recurrence (20, 30). MESANGIOCAPILLARY GLOMERULONEPHRITIS TYPE I Mesangiocapillary glomerulonephritis (MCGN) is generally regarded as an immune complex disease, and can be primary or secondary to a number of systemic disorders, including systemic lupus erythematosus, cryoglobulinemia, and viral hepatitis. Most patients with MCGN type I progress relentlessly to ESRD, and recurrence of the disease is reported in 20-30% of graft recipients (17, 31). These figures may well be overestimates, as the histological differentiation from allograft glomerulopathy can be difficult. Up to 40% of patients with recurrent MCGN type I will lose their graft, and CsA treatment does not seem to prevent recurrence (32). There is no established treatment for recurrent MCGN type I, but antiplatelet agents, such as aspirin or dipyridamole, have been used with some success (17), and remission of severe recurrent disease has been achieved using steroids and cyclophosphamide (33). Recurrence in successive grafts has also been reported (17). In view of the significant graft loss due to recurrence, the use of LRD should be discouraged. MESANGIOCAPILLARY GLOMERULONEPHRITIS TYPE II MCGN type II recurs in 50-100% of all grafts. (34). Graft loss due to recurrence occurs in 10-20% of these patients (20). Male patients, and those who present with nephroticrange proteinuria or rapidly progressive GN with crescents, appear to show an increased rate (50%) of subsequent graft loss (35). The onset of MCGN type II can be as early as 3 weeks after transplantation and usually occurs within the first year, with nonnephrotic-range proteinuria. There is no proven treatment for recurrent MCGN type II, although plasma exchange has been used successfully in aggressive disease (36). In neither type of MCGN does the concentration of complement C3 or C4 in the serum correlate with recurrence. MEMBRANOUS GLOMERULONEPHRITIS Approximately 25-30% of patients with membranous nephropathy will develop progressive renal failure and ESRD. The reported frequency of recurrence of membranous glomerulonephritis (MGN) ranges widely from 3% to 26% (20, 37, 38). The estimation of the true incidence of recurrent MGN is complicated by the occurrence of de novo MGN in the graft. This type of GN is the most common de novo GN after transplantation, with an incidence of 2-9% in most series of transplant recipients after 1-2 years (37). Clinically, recurrent MGN presents earlier after transplantation (mean: 10 months, range: 1 week to 2 years) than de novo graft MGN (mean: 22 months, range: 4 months to 6 years), and in recurrent disease, proteinuria and nephrotic syndrome develop more rapidly (37). Histologically, de novo graft MGN may resemble chronic rejection, but with signs of vascular and interstitial rejection in addition to basement membrane thickening and subendothelial immune deposits (39). Despite the low absolute risk of recurrence of MGN, 50-60% of grafts with recurrent MGN fail (20, 40). Male patients and patients with very aggressive disease seem to be at a higher risk for recurrent MGN (40). The incidence of recurrence is also elevated in grafts from LRD (11.9% as compared with 3% in cadaveric grafts) (41). In 1994, an association between the TAP1B allele and MGN was demonstrated (42). A possible role of this genetic marker in recurrent MGN remains to be determined. Rapid recurrence of MGN was observed in recipients of grafts from three living related, HLA-identical donors (43), suggesting that grafts from LRD should be considered very carefully. There are only limited data on the treatment of recurrent MGN; however, a large body of literature exists on the treatment of native kidney MGN, and treatment options for recurrent disease may be derived from these studies. The use of steroids alone does not seem to be of benefit in native kidney MGN (44, 45), although anecdotal remission of nephrotic syndrome in recurrent MGN has been achieved with pulsed methylprednisolone and high-dose, alternate-day steroids (46). The currently preferred treatment of native kidney MGN is the use of cytotoxic drugs such as chlorambucil or cyclophosphamide in combination with steroids (47, 48). Patients with moderate or severe nephrotic syndrome and declining renal function seem to gain most from these treatment regimens (47). Low doses of the ACE inhibitor captopril have also been shown to reduce proteinuria in nephrotic syndrome (49). A similar approach would be reasonable in recurrent disease. ANTI-GLOMERULAR BASEMENT MEMBRANE DISEASE There have been no new data on recurrent anti-glomerular basement membrane (GBM) disease since previous reviews published as late as 1993 (3, 17, 50). The prevention of recurrence by careful timing of transplantation and the distinction of clinical recurrence from histological but insignificant recurrence remains important. The "histological" recurrence rate in transplants, defined primarily by the presence of linear IgG deposits along the GBM, is about 50% if transplantation is performed while circulating antibodies are still present (50). Only a quarter of these patients will develop anti-GBM disease clinically. Postponing transplantation for at least 6 months after antibodies have become undetectable reduces the recurrence rate to 5-15% (17). The recurrence of the disease is usually immediate, although in one case recurrence was delayed for 2 years after transplantation (51). If transplantation is performed while the patient still has circulating autoantibodies, aggressive treatment with repeated plasmapheresis and cyclophosphamide may prevent the recurrence of anti-GBM disease in the graft (52). Patients who develop clinical recurrence should probably also be treated with plasmapheresis and the addition of cyclophosphamide to their immunosuppressive regimen. Anti-GBM disease also occurs after transplantation in patients with Alport's syndrome. In this setting, the disease occurs de novo, due to the grafting of a kidney containing normal GBM into a patient with a mutation in one of the basement membrane collagens. Mutations have been defined in approximately 50% of patients with Alport's syndrome, and are found in the α3, α4, or α5 chains of type IV collagen (53). After transplantation, patients can develop anti-α3 or anti-α5 (type IV collagen) antibodies (53-55), which are deposited along the GBM but rarely cause crescentic GN (56). Confirmed nephritis should be treated in the same way as Goodpasture's disease, with cyclophosphamide and plasma exchange. SYSTEMIC LUPUS ERYTHEMATOSUS One hundred eighteen renal transplantations were performed in patients with lupus nephritis in the United Kingdom between 1984 and 1993 (≅1% of all renal grafts) (1). The development of ESRD in patients with lupus is usually associated with resolution of extrarenal symptoms and serological markers. Possibly as a result of this, the recurrence rate of systemic lupus erythematosus (SLE) is low (<1%), and only a few cases of recurrent lupus nephritis have been described (57). The clinical presentations have included extrarenal manifestations (rash and arthralgia) and nonnephrotic proteinuria. Laboratory hallmarks are rising anti-DNA antibody titers and falling or low serum complement levels. High-dose steroids, plasmapheresis, and pulse cyclophosphamide have all been used for treatment of recurrent SLE with variable success (57, 58). Overall graft survival rates are improved in patients who receive living related grafts (59). It is reasonable to ensure that SLE serology should improve or be minimally altered before transplantation. Patients with lupus anticoagulant are at risk of thromboembolic events and may require anticoagulation. HEMOLYTIC UREMIC SYNDROME Renal transplantation was performed in 65 patients with hemolytic uremic syndrome (HUS) in the United Kingdom between 1984 and 1993 (≅0.6% of all renal cadaveric grafts) (1). Recurrent disease must be distinguished from a thrombotic microangiopathy associated with either CsA or acute vascular rejection, which can develop after transplantation regardless of the initial cause of renal failure (60). Such patients develop thrombocytopenia, microangiopathic hemolytic anemia, and ARF. The recurrence rate of HUS in the transplanted kidney is 10-25%, and about half of these cases result in graft loss (61). Time of recurrence ranges from very early to 3-5 years after transplantation. Possible risk factors for recurrence include early transplantation after the acute disease (<3 months) and the use of CsA, antilymphocyte globulin, and OKT3 (40, 62). Prophylactic measures to reduce the risk of recurrence include low-dose salicylates and dipyridamole, and the avoidance of oral contraception (63). Treatment of recurrent HUS, as for the initial episode, includes plasma exchange for fresh frozen plasma or cryosupernatant, or plasma infusions. If HUS recurs in the setting of CsA treatment, alternative immunosuppressants should be considered (e.g., FK506), although one case of de novo HUS has been reported in a patient treated with FK506 (43). The use of LRD should be approached with care, especially in those (rare) cases where the history suggests a familial clustering of HUS, or recurrent disease (64). NECROTIZING CRESCENTIC GLOMERULONEPHRITIS AND SYSTEMIC VASCULITIS Necrotizing crescentic glomerulonephritis (NCGN) occurs either as an idiopathic nephropathy or as part of a systemic small vessel vasculitis-usually Wegener's granulomatosis (WG) or microscopic polyangiitis-in association with antineutrophil cytoplasmic antibodies (ANCA). Recurrence of WG has been described after transplantation in patients treated with most immunosuppressive regimens, including azathioprine, steroids, and CsA, with overall (renal and extrarenal) recurrence rates of approximately 29% and renal recurrence rates of 16% (20, 40, 43, 65, 66). Among those with renal recurrence, graft loss occurred in 40%. Exclusively extrarenal manifestations of WG recurred in 13%. Extrarenal recurrence may, however, involve the ureter, causing stenosis and obstructive nephropathy (67). The overall recurrence rate appears to be lower in patients treated with azathioprine than with CsA. In idiopathic NCGN, the recurrence rate appears to be low. Only 1 of 80 patients (1.3%) in an Australian series, 6 of 410 (1.5%) in a French series, and 13 of 750 (1.7%) in an American series had recurrent disease (20, 40). Rising ANCA titers may herald relapses of systemic vasculitis before transplantation (67-69), but longitudinal studies on the significance of rising ANCA titers are lacking in the posttransplant setting. Patients with persistent or rising ANCA titers, especially of a proteinase 3 specificity, are particularly at risk for recurrent disease (70). In a recent study, 60% of patients with persistent ANCA positivity relapsed, compared 34% of those achieving ANCA negativity (68). However, the predictive power of a rising ANCA titer is not sufficient to justify treatment on serological grounds alone (68-71), but should alert the clinician to the possibility of a clinical relapse. Recurrent WG 5 days after transplantation has been reported without increased ANCA levels (43). Transplantation should be postponed if there is any evidence of active disease, or if ANCA titers are rising. Patients with stable but persistently positive ANCA titers can receive a transplant if they have been in clinical remission for at least 6 months. Treatment of recurrent WG includes a switch from CsA or azathioprine to cyclophosphamide (65). Other potential measures include plasma exchange, intravenous immunoglobulin, and humanized antilymphocyte monoclonal antibodies. There is no contraindication to LRD in ESRD due to vasculitis or NCGN. SYSTEMIC SCLEROSIS The incidence of severe renal disease in systemic sclerosis has been markedly reduced since the advent of ACE inhibitors, and only a few patients with systemic sclerosis have received grafts. It appears from this limited experience that the rate of recurrence is approximately 20%, and that these patients tend to follow a malignant course (72). However, the histological changes in systemic sclerosis may resemble those of acute or chronic vascular rejection. The benefit of ACE inhibitors in recurrent disease after transplantation is as yet unknown. MIXED ESSENTIAL CRYOGLOBULINEMIA, AMYLOIDOSIS, NONAMYLOIDOTIC FIBRILLARY GLOMERULONEPHRITIS, AND MONOCLONAL GAMMOPATHIES Transplantation is rarely performed for mixed essential cryoglobulinemia, amyloidosis, nonamyloidotic fibrillary glomerulonephritis, and monoclonal gammopathies. However, over 60 patients with renal failure due to secondary amyloidosis have been reported to have received renal allografts. These patients usually have rheumatoid arthritis as their underlying disease, and graft survival is no different from that of a matched population. Patient survival is reduced, mostly due to infectious and vascular complications (73). Recurrence of renal amyloid occurs in 20-33% of the grafts, but rarely leads to graft loss (73, 74). Recurrent renal disease has been reported in graft recipients with multiple myeloma and light chain deposition disease (75). Recurrence of light chain deposition was found in 6 of 12 patients, with eventual graft loss in two patients despite intensive therapy with plasmapheresis, melphalan, and steroids (75). Recurrence was heralded by the onset of nephrotic-range proteinuria. Patient survival depended mainly on the evolution of the hematological malignancy and the presence of infections. Of four patients with immunotactoid glomerulopathy receiving an allograft, two experienced disease recurrence after 21 and 60 months (76). Recurrence was characterized by massive proteinuria. In mixed essential cryoglobulinemia, recurrence after transplant has been reported in approximately 50% of the grafts. Recurrent disease is often associated with significant extrarenal manifestations, such as purpura and arthralgias (77). SUMMARY Recurrent glomerulonephritis after transplantation is an important cause of graft loss, and should be considered in all at-risk patients who develop renal dysfunction or urinary abnormalities. The distinction of recurrent disease from allograft rejection has important therapeutic implications. Recent advances in the understanding of the pathophysiology of GN, particularly FSGS, may lead to novel treatments. The risk of recurrence is increased in recipients of LRD grafts, and this should be considered in the use of LRD grafts for patients with certain glomerulonephropathies. Acknowledgments. J.B.L. is a Medical Reserch Council training fellow. The authors thank Dr. J. De Meester and Dr. G.G. Persijn, Eurotransplant, Leiden, for providing data from the Eurotransplant registry.