Donor Source

Children in all age groups receiving kidney transplants from living donors have superior short-term and long-term graft survival (see Fig. 37.3 ) compared with children receiving deceased-donor grafts. The improved outcomes for living-donor kidney transplants are thought to be due to reduced cold ischemia time, improved HLA-matching, and improved preoperative preparation of the recipient. Survival of living unrelated donor grafts has even been found to be superior to that of grafts from younger deceased donors. In addition, living donation from older relatives (e.g., a grandparent) has been found to have superior long-term outcomes in pediatric recipients compared with deceased-donor transplant. The most poorly HLA-matched grafts from the oldest living donor had survival similar to or better than any deceased donor in an analysis of USRDS. Finally, for deceased-donor kidneys, outcomes are superior for pediatric recipients of grafts from donation after brain death than donation after circulatory death due to increased risk of graft loss by 4 years in the latter.

Recipient Age

An analysis of OPTN data has shown the best long-term graft survival occurred in middle-aged adult recipients and children younger than 12 years, whereas the worst outcomes were seen in adolescent and young adult recipients. Several studies have confirmed that recipients aged 17 to 24 years have the highest risk of graft loss, prompting further research in how to best support this age group to optimize outcomes.

Donor Age and Size

Multiple reports in the literature indicate that donation from young adults improves graft survival. The “ideal” donor age derived from these analyses has been determined to be more than 6 years old and less than 45 to 50 years old. Donor age appears to be more important in determining graft survival for deceased-donor transplants than for living-donor transplants.

Interestingly, children receiving grafts from deceased donors younger than age 18 years have improved relative glomerular filtration rate (GFR) over time compared with children receiving kidneys from deceased or living adult donors. Senescence of somatic kidney cells has been postulated to be the underlying cellular mechanism responsible for this effect. In other words, renal allografts from young donors are able to adapt to the increasing metabolic demand of a growing recipient, whereas kidneys from older donors have decreased capacity for cell survival and regeneration and undergo accelerated aging under the stress of transplantation.

Transplants from very young donors (younger than age 5 years) to pediatric recipients have historically been avoided due to reports of increased risk of graft loss. In a recent analysis of the NAPRTCS database, primary graft nonfunction was found to be more frequent in kidney transplants from very young donors (3.7% compared with 0.3% in transplants from donors aged 6 to 35 years or “ideal donors”). However, longer term, the 3-year graft survival and estimated glomerular filtration rate (eGFR) of functioning grafts from very young donors was equivalent to ideal donors. In a more recent study, graft survival rate was almost 90% for recipients of infant donor kidneys implanted either en bloc or singly.

Recipient Race

African American race has been demonstrated in multiple studies to associate with worse long-term graft survival, including increased risk for delayed graft function, acute rejection, and graft loss. It is unclear whether the relative role of genetics (e.g., higher rates of FSGS) versus racial disparities explains these findings. For example, African American children are more likely to be older at the time of transplant, receive fewer living-donor transplants, and have longer duration of dialysis before transplant, all of which are associated factors contributing to worse long-term outcomes. After controlling for 19 variables, including age, primary diagnosis, preemptive transplant, rejection experience, HLA-B mismatch, immunosuppression, and gender, black patients still had higher risk for graft failure (Hazard Ratio [HR] = 1.6, 95% confidence interval [CI] 1.46–1.86).

HLA Matching

Degree of HLA mismatch is an important prognostic factor for long-term graft survival for both deceased- and living-donor transplants. Although the ideal situation would be donation from an HLA-identical sibling, this has occurred in only 3.4% of pediatric living-donor transplants reported in NAPRTCS. Most live kidney donations come from haplo-identical parents. An analysis of NAPRTCS in 2000 suggested that a six-antigen matched deceased-donor kidney had equivalent graft survival and rejection rates as haplo-identical living-donor transplants. However, long-term graft survival (5 years) was actually 10% better for the children who received six-antigen matched deceased-donor kidneys.

There is some evidence that donation from a sibling with noninherited maternal antigens confers survival benefit for the graft and may be due to bidirectional immune regulation between donor and recipient immune cells. Grafts from mothers in these studies had poorer outcomes. This has particular relevance for pediatrics because mothers represent the majority of parental donors.

There is debate as to the effect of degree of HLA mismatch on acute rejection and graft survival. An analysis of 9029 pediatric recipients in the Collaborative Transplant Study in Europe found a hierarchical relationship for an effect of increasing mismatches at the A-, B-, and DR-loci on graft survival. Similarly, there are inconsistent findings on the role of HLA-DR matching and graft survival. In 2008 Gritsch, et al. reported that children in the US receiving zero HLA-DR mismatched deceased-donor kidneys had comparable 5-year graft survival to children who had received 1- or 2-DR mismatched kidneys. A recent analysis of European pediatric kidney transplants found that whereas two HLA-DR mismatches associated with lower graft survival in children receiving transplants from 1988 to 1997, this effect was not seen in a more recent era (1998–2007). They did, however, describe a disturbing association of HLA-DR double-mismatch with increased risk of developing non-Hodgkin lymphoma. This interesting finding needs to be validated in additional cohorts of pediatric kidney transplant recipients.

Sensitization occurs with increasing HLA-mismatch and specifically the degree of HLA-DR mismatch. As stated earlier, children have longer expected remaining lifetimes after transplantation and increased expectancy for multiple transplants. Therefore development of HLA-sensitization from mismatched primary transplants has been demonstrated to be a barrier to subsequent transplant.

Presensitization

NAPRTCS reported that >5 transfusions significantly increased the frequency of delayed graft function (defined as dialysis in the first transplant week) and graft failure (HR = 1.22, p = 0.016 for living-donor transplants and HR = 1.25, p < 0.001 for deceased-donor transplants). This effect was presumed to be due to increased sensitization. The use of blood transfusion has decreased with increasing use of erythropoietin to treat anemia during ESRD, and the largest risk for sensitization in pediatric transplant candidates is prior solid-organ transplantation.

Delayed Graft Function and Thrombosis

Delayed graft function (DGF) is uncommon (approximately 6%) in pediatric kidney transplant recipients, as reported in the 2015 SRTR report. A single-center retrospective study of 126 pediatric kidney transplant recipients from 2004 to 2011 showed similar rate of 9% DGF and identified vascular cause of brain death and donor age >15 years as possible risk factors for DGF. Overall, technical causes of graft loss (vascular thrombosis, primary nonfunction, and others) contributed to 13% of all graft failures reported in NAPRTCS since 1987, suggesting that these events occur in about 3% of pediatric transplants. Several studies of pediatric transplant registries in the UK, Ireland, and the Netherlands reported similar rates of thrombosis. Risk factors for graft thrombosis in children include recipient age <6 years, donor age <6 years, cold ischemia time >24 hours, and history of peritoneal dialysis.

Focal Segmental Glomerulosclerosis

FSGS recurs in approximately 15% to 60% of children with nephrotic syndrome undergoing renal transplant and is the most common cause of graft loss due to recurrence. Although several small single-center studies in the 1990s suggested clinical features of young age of diagnosis, rapid progression to ESRD, and mesangial proliferation on native biopsy to be associated with posttransplant recurrence, many of these predictive factors have failed to be durably replicated. A recent analysis of UNOS data showed an overall recurrence rate of 15% in children receiving kidney transplant between 1988 and 2008. FSGS recurrence occurred more frequently in Caucasian and younger recipients. Although receipt of a living-donor kidney was previously described as a risk factor for FSGS recurrence, in this analysis, donor type did not remain significant in multivariate analysis. This may be explained by the confounding effect of younger recipient age because rates of living-donor transplantation increase with decreasing recipient age. The significance of young recipient age in this study may also indirectly support earlier associations between younger age at FSGS diagnosis and/or rapid progression to ESRD and FSGS recurrence. Another clinical feature recently associated with FSGS recurrence is the response to steroid therapy during native FSGS disease. In an analysis of 150 pediatric patients transplanted in Europe for ESRD due to steroid-resistant nephrotic syndrome, it was found that the patients with initial steroid sensitivity who developed steroid resistance during treatment of their native kidney disease had the highest rate of recurrence (92.9%) posttransplant compared with children who had steroid-resistance from the outset (30%). These findings suggest that the development of secondary steroid resistance during treatment of idiopathic nephrotic syndrome may predict posttransplant recurrence.

Patients with FSGS recurrence have inferior graft survival compared with other ESRD etiologies. Furthermore, the effect of FSGS recurrence on graft loss is more significant in children compared with adults. Higher rates of acute tubular necrosis (ATN) in both living-donor and deceased-donor transplants have been reported in children with FSGS compared with children with other causes of ESRD. It has been suggested that the increased rate of ATN, possibly due to early FSGS recurrence, plays a role in the decreased graft survival for children with FSGS. This observation influenced some centers to offer living-donor transplant (which has lower rates of ATN in general) with pretransplant plasmapheresis in an attempt to prevent rapid FSGS recurrence and improve graft survival; however, the use of plasmapheresis pretransplant has not proven consistently beneficial in small trials.

Recurrence of FSGS often presents early after kidney transplantation in children, with a reported median time to recurrence of 6 to 14 days, although heavy proteinuria often can be detected within hours after transplant. It is usually characterized by nephrotic-range proteinuria (protein/creatinine ratio >2 mg/mg) and hypoalbuminemia, but can present as complete nephrotic syndrome, including anasarca and hypercholesterolemia. Whereas recurrent disease typically presents within the first 2 years after transplant, the presentation of nephrotic syndrome after 2 years is generally considered to be secondary to calcineurin inhibitor (CNI) toxicity, chronic rejection, or de novo disease.

Biopsies early after recurrence often demonstrate normal histology on light microscopy with effacement of podocyte foot processes on electron microscopy. Later biopsies have characteristic segmental lesions of FSGS with endocapillary proliferation and foam cell accumulation and can progress to glomerular sclerosis and interstitial fibrosis. Unfortunately, histology classification has not been shown to be predictive of treatment response in recurrent FSGS.

The pathophysiology of primary idiopathic FSGS in native kidneys and recurrent FSGS after transplant remains unclear. It is likely a multifactorial process involving cytokines secreted by T cells, a humoral factor that alters podocyte cytoskeletal structure, and a balance between circulating permeability factors and inhibitors of such factors.

Genetic forms of FSGS have low rates of recurrence, estimated at around 3% for patients with homozygous and compound heterozygous podocin (NPHS2) mutations. Recurrence of nephrotic syndrome in children with NPHS2 mutations does not appear to be due to antipodocin antibodies. Further study is needed to clarify the pathogenesis of recurrence in this population.

A circulating factor has been proposed to cause increased albumin permeability of the slit diaphragm during FSGS recurrence, but the identity, source, and pathologic effects of such a factor are yet to be fully elucidated. Bioassays of albumin permeability have had conflicting results in predicting FSGS recurrence and have not proven to be specific or highly predictive of response to therapy or long-term renal outcome in patients with nephrotic syndrome. Furthermore, demonstration of a neutralizing effect of normal serum or urine from nephrotic patients on albumin permeability suggests that loss or deficiency of a natural inhibitor may play a role. Serum soluble urokinase receptor (suPAR) has been proposed as a circulating factor capable of causing FSGS. Two-thirds of the patients with FSGS in this study had elevated serum suPAR concentrations compared with healthy controls and patients with other glomerular diseases. Furthermore, patients who had FSGS recurrence after transplant had the highest serum concentrations of suPAR, providing hope for a clinically predictive test. However, a small prospective study found no correlation with suPAR levels and remission of proteinuria in adult patients treated for recurrent FSGS. Further research is needed in understanding the pathogenesis of FSGS recurrence and to identify predictive clinical tests.

In light of the limited understanding of the underlying pathophysiology, the treatment of recurrent FSGS is not well established. The most commonly reported therapies for recurrent FSGS are plasmapheresis or protein A immuno-adsorption therapy. A review of the literature found that 49 of 70 children (70%) receiving plasmapheresis for FSGS recurrence achieved partial or full remission. Early detection of recurrence and initiation of plasmapheresis appears to provide the best results. However, these studies likely overrepresent the benefit of plasmapheresis given small sample sizes and the use of only historical groups for comparison in a few of the studies or no comparison group in others. In general, it is recommended that children at risk should have daily monitoring of protein/creatinine ratios in the early posttransplant period to allow rapid detection of FSGS recurrence and early initiation of treatment.

Some centers perform plasmapheresis either before a planned living donor transplant or in the perioperative period of a deceased-donor transplant. It is unclear if this practice offers any benefit over early detection and treatment of established recurrence.

High-dose cyclosporine with or without adjunctive plasmapheresis and/or high-dose steroids has shown efficacy in achieving complete or partial remission of posttransplant FSGS recurrence in children. The antiproteinuric effect of CNIs has been postulated to be due to T cell suppression and inhibition of cytokine secretion thought to be injurious to podocytes, as well as to a direct effect on the stabilization of the podocyte cytoskeleton. The alkylating agent cyclophosphamide has also been reported to be efficacious in some children with recurrent FSGS, usually in combination with plasmapheresis. Reduction in proteinuria has also been reported with the use of angiotensin blockade either alone or in conjunction with plasmapheresis.

Finally, there have been anecdotal cases reporting prolonged remission of proteinuria in children with recurrent FSGS after plasmapheresis and B cell depletion with rituximab. This was first reported for a child who incidentally achieved remission of FSGS recurrence after rituximab therapy to treat posttransplant lymphoproliferative disease (PTLD). More recently, there have been small case reports demonstrating efficacy of another humanized anti-CD20 antibody, ofatumumab, in inducing remission of native kidney FSGS or FSGS recurrence in patients resistant to rituximab therapy. The true value of this approach remains to be established.

Congenital Nephrotic Syndrome

Congenital nephrotic syndrome (CNS), by definition, occurs within the first 3 months of life and is most commonly due to mutations in the NPHS1 gene, encoding nephrin, a major structural component of the slit diaphragm. Infants with CNS are frequently born prematurely, have enlarged placentas, and present with nephrotic-range proteinuria, anasarca, and hypoalbuminemia. Secondary causes (i.e., neonatal cytomegalovirus [CMV], congenital rubella, human immunodeficiency virus [HIV], hepatitis B, toxoplasmosis, syphilis, and infantile lupus) should be excluded. The genetics of CNS are proving more complex in recent years with the identification of additional genetic mutations (including genes previously only associated with FSGS).

Recurrence of nephrotic syndrome has been reported in 25% of children with CNS due to homozygous Fin-major mutations in nephrin (NPHS1) with mean time to recurrence of 12 months posttransplant (range of 5 days to 2 years). Antinephrin antibodies have been implicated in a majority of these children.

Vascular thrombosis and death from infection (with a functional graft) have been described more often in children with CNS after transplant compared with children with other primary diseases. Hypercoagulability due to urinary losses of antithrombin III (ATIII) along with younger age at the time of transplant likely contribute to the increased risk of these complications.

Diffuse mesangial sclerosis (DMS) can also present as nephrotic syndrome in early infancy and is associated with mutations of the Wilm tumor suppressor gene 1 (WT1). Denys-Drash syndrome consists of progressive glomerulopathy (DMS) and male pseudohermaphroditism (although genotypic females have also been described) and is due to heterozygous mutations within exon 8 or 9 of WT1. Children with Denys-Drash syndrome are at increased risk of developing Wilm tumor; therefore bilateral nephrectomy is often performed once they develop ESRD. Frasier syndrome presents with normal female genitalia with streak gonads, XY karyotype, and progressive glomerulopathy (DMS on histology) and is due to splice mutations within exon 9 of WT1. Children with Frasier syndrome are at risk of developing gonadoblastoma and therefore should undergo oophorectomy. There have also been reports of isolated DMS in children with WT1 mutations without other syndromic features who may not be at risk for malignancy.

Other than one report of membranoproliferative glomerulonephritis (MPGN) in a child with Denys-Drash, no other studies report recurrence of nephrotic syndrome in children with WT1 mutations undergoing kidney transplant. Overall, children with Denys-Drash have comparable patient and graft survival as children with other causes of ESRD.

Alport Syndrome

Alport syndrome (AS) is a clinically and genetically heterogeneous nephropathy characterized by glomerular basement membrane (GBM) defects due to alterations in the type IV collagen matrix. AS presents with persistent microscopic hematuria and proteinuria that can progress to renal failure and is often associated with sensorineural hearing loss and ocular abnormalities. It is the prototype of inherited nephritis and accounts for 2% of children with ESRD in the US. The most common and severe form is inherited in an X-linked recessive pattern and is associated with early progression to ESRD (50% of boys by age 25 years). Autosomal recessive and autosomal dominant forms have also been described.

The genetics of X-linked AS were defined more than 20 years ago as a mutation in the gene encoding the α5 chain of type IV collagen (COL4A5). Patients with mutations resulting in a truncated protein (i.e., large rearrangement, premature stop, or frame-shift mutations) tend to progress to ESRD earlier (50% by age 19 years). Female carriers of COL4A5 mutation are considered to have a less severe course, but some do develop ESRD in late adulthood. Development of hearing loss and progression of proteinuria appear predictive of a more severe course in female carriers. Evaluation of potential living donors, especially female relatives, should include evaluation for hematuria and proteinuria, as the heterozygous carrier state carries a risk of developing ESRD after donation.

Autosomal recessive forms involving mutations in the α3 (COL4A3) and α5 (COL4A4) chains have also been described. Rare autosomal dominant mutations in COL4A3 and COL4A4 have also been reported with less severe renal disease, fewer cases of hearing loss, and absence of reported ocular changes.

Although Alport syndrome itself does not recur, development of de novo antibodies directed to the glomerular basement membrane (anti-GBM) has been reported to occur in approximately 3% to 5% of AS males within the first year after transplant. Late occurrence has also been described.

Antibodies are generated against the type IV collagen α5 or α3 chains. Patients with mutations resulting in complete absence or severe truncation of these proteins appear to be at highest risk of developing anti-GBM antibodies. It should be noted that the anti-GBM antibodies in these patients are specific for different epitopes of the noncollagenous domains of the α3α4α5(IV) collagen network than the Goodpasture allo-epitope, and therefore may not be detectable in the serum using the clinically available enzyme-linked immunosorbent assay (ELISA) used to diagnose spontaneous anti-GBM nephritis in Goodpasture disease.

Early reports described rapidly progressive anti-GBM disease with nearly 90% to 100% graft loss. With better recognition of this entity, reports of subclinical anti-GBM (i.e., demonstration of linear IgG deposits along the GBM in transplant biopsies without graft dysfunction) suggest a broader spectrum of clinical disease. Despite the poor graft outcome described during anti-GBM nephritis in AS, the severe form does not occur frequently enough to affect overall graft survival statistics for AS patients. Chronic rejection is the most common cause of graft loss in this population. Patients who have lost a prior graft to anti-GBM disease are at higher risk of recurrence after a subsequent transplant, although some have reported successful retransplantation.

MPGN/C3 Glomerulopathy

MPGN describes a pattern of glomerular injury with common histologic features of glomerular capillary wall thickening (membrano-) and hypercellularity in the glomerular tufts (-proliferative). MPGN was historically classified into three morphologic types based on electron microscopy findings: type I with presence of immune deposits in the subendothelial and mesangial areas, type II with electron-dense deposits within the basement membrane, and type III with complex GBM formation with subendothelial and subepithelial electron-dense deposits that are bridged by intramembranous deposits. A unifying characteristic of all types of MPGN is hypocomplementemia (low C3). This traditional system resulted in overlap among MPGN type I and type III while considering MPGN type II, known as dense deposit disease (DDD), as a separate entity because it has unique pathogenic and clinical features.

Recent advances in our understanding of MPGN pathogenesis has led to a new classification system for the diagnosis and management of these disorders. In the new system, MPGN is classified as either mediated by immune complexes (immune complex mediated MPGN) or by complement dysregulation and activation of the alternative complement pathway, now termed C3 glomerulopathy. This differentiation is made based on immunofluorescence microscopy. Immune complex mediated MPGN is defined by the presence of immunoglobulin staining on immunofluorescence microscopy and is most commonly associated with underlying autoimmune disease or infections. Therefore recurrence and management are dependent on the underlying systemic disease resulting in renal pathology. In contrast, complement-mediated MPGN (i.e., C3 glomerulopathy) results from dysregulation and persistent activation of the alternative complement pathway with deposition of complement products in the glomeruli and the mesangium with predominantly bright C3 and minimal immunoglobulin staining on immunofluorescence microscopy. Genetic defects in the proteins involved in the alternative complement cascade and/or autoantibodies have been identified in patients with C3 glomerulopathy, which is further classified into DDD, C3 glomerulonephritis (C3GN), and CFHR5 nephropathy. DDD is distinguished from C3GN on electron microscopy by the presence of large, sausage-shaped, osmophilic dense deposits within the GBMs. In contrast, in C3GN the deposits tend to be mesangial and subendothelial.

The major defect in DDD and C3GN is excessive activation of the alternative complement pathway. The alternative complement pathway is normally autoactivated by spontaneous cleavage of the C3 to C3b which leads to the formation of the C3 convertase by the binding factor B and properdin. Normally the activity of this pathway is tightly regulated by the activity of the C3 convertase, which is inhibited by factor H. In patients with DDD and C3GN, increased activity of C3 convertase is most commonly caused by (1) stabilization and reduced degradation of C3 convertase enzyme by autoantibodies (called C3 nephritic factor; C3NeF) or (2) deficiency or autoantibody directed against the natural inhibitor of C3bBb, factor H. Low levels or defective factor B, factor I, and C3 have also been described in DDD and C3GN.

C3 glomerulopathies have a progressive course with development of ESRD in up to 50% of patients and high recurrence after transplantation.

Historic reports of outcomes for patients with MPGN type I and III are challenging to apply to patients diagnosed under the new classification system.

An analysis of UNOS data showed MPGN type I has a significant negative effect on graft survival compared with other forms of glomerulonephritis. Disease recurrence was the most common cause of graft loss (14.5%) for patients with MPGN type I in this study. Graft loss has been reported in about 50% of patients with recurrent MPGN type I. Low complement levels (C3) at the time of transplant and younger age at time of transplant have been associated with increased risk of recurrence. Another report of serial protocol biopsies identified that some patients have asymptomatic recurrence of MPGN type I.

An analysis of the NAPRTCS database showed significantly worse 5-year graft survival (50.0%) for children with MPGN type II (DDD) compared with the database as a whole (74.3%). The most common cause of graft failure in children with DDD was recurrent disease (14.7% of graft losses). Of the children with DDD who had posttransplant biopsies ( n = 18) in the registry, 67% had recurrent DDD. No correlation was found between pretransplant presentation or C3 levels with the risk of recurrence or graft loss. In a single-center retrospective study, DDD was reported to recur in 60% of children, whereas the rate of graft loss was similar to other children, transplanted at that center. Finally, an analysis of the UNOS database, including children and adults, reported graft loss due to recurrence in 30% of patients with DDD and significantly worse graft survival compared with patients with other forms of glomerulonephritis.

There is minimal data about transplant outcomes for patients with C3GN. Zand et al. evaluated 21 adult and pediatric renal transplant recipients with a history of C3GN as a cause of the ESRD. C3GN recurred in 66.7% of the patients with graft failure in 50% of recurrent C3GN at median time 28 months. The median time to graft failure posttransplant was longer in patients with C3GN compared to DDD (6–7 years vs. 2.5 years, respectively).

Recurrent MPGN presents with hematuria, progressive proteinuria, and deteriorating graft function. Low or low-normal C3 levels have been reported. There are no randomized trials to determine efficacy of treatment of posttransplant recurrence of C3GN and DDD. Nonspecific treatments, including the use of angiotensin blockade, steroids, anticoagulation, or antiplatelet therapy, have been reported in the literature with variable success. In patients with genetic deficiency in factor H, plasma infusion can be used to correct the deficiency. Plasma exchange has been reported in small case series for removal of circulating autoantibodies like C3 nephritic factor or antifactor H antibodies and to replace defective proteins in the case of activating mutations in C3. Rituximab has been reported in cases involving autoantibodies. More recently, the use of an anti-C5 antibody (eculizumab) has been proposed to block the downstream effects of uncontrolled C3 convertase activity, but the clinical utility, long-term implications, and identification of patients who would benefit most from this therapy still need to be defined.

IgA and Henoch-Schonlein Purpura

IgA nephropathy accounts for 3% of incident cases of ESRD in children in the US. Histologic recurrence in adult transplant recipients with IgA nephropathy has been reported to be 30% to 35%. Transplant patients with abnormal urinalysis (hematuria or proteinuria) are more likely to show histologic evidence of recurrence. The role of recurrence in graft loss is variable in these reports. In some single-center studies, as many as 40% to 50% of patients with recurrence have been reported to have graft failure; however, a recent analysis of patients in East Asia found that chronic rejection had a larger effect on long-term graft survival rate in patients with IgA nephropathy.

Henoch-Schonlein purpura (HSP) nephritis is a rare cause of ESRD in children, accounting for only 0.4% of incident cases in the US. Histologic recurrence of HSP has been reported in as high as 70% of patients within 2 years after transplant, with clinically evident recurrence (hematuria, moderate proteinuria, and hypertension) in 15% to 35%. Long-term graft survival, however, does not appear to be greatly affected by recurrence. An analysis of adolescents and young adults in the UNOS database revealed graft loss due to recurrent disease in 13.6% of patients with HSP.

Hemolytic-Uremic Syndrome

Hemolytic-uremic syndrome (HUS) accounts for approximately 1.5% of children with ESRD in the US. HUS is characterized by the clinical triad of microangiopathic hemolytic anemia, thrombocytopenia, and renal failure, and it can be due to secondary causes (infectious, drug-induced, auto-antibodies) or primary genetic defects in complement regulatory components that lead to persistent activation of the alternative pathway of the complement cascade.

The most common cause of HUS in children (accounting for approximately 90% of cases) is associated with colitis due to shiga-toxin producing bacteria ( E. coli, Shigella dysenteriae, others). Epidemiologic studies estimate that 5% to 10% of children with shiga-toxin producing E. coli (STEC) infections develop HUS, with children under the age of 5 years carrying the highest risk. Of the children that develop HUS, about two-thirds develop oliguric renal failure. ESRD occurred in 10% to 15% of children with HUS in a meta-analysis of long-term prognosis (>1 year) after diarrhea-associated HUS. The presence of anuria or prolonged oliguria has been associated with higher risk for long-term sequelae, including CKD, proteinuria, and hypertension in as many as 30% of HUS survivors. The risk of disease recurrence after transplant in this particular group appears to be very low (<1%), and graft survival is similar to children with nonglomerular primary disease.

Infections with bacteria other than STEC that produce shiga- or shiga-like toxins, including Shigella dysenteriae and Citrobacter sp, have also been implicated in the etiology of HUS in children. Furthermore, urinary tract infections with STEC have also been found in children with HUS presenting without diarrhea.

Finally, invasive Streptococcus pneumoniae infections have also been linked with a rare, but severe, form of HUS that has higher rates of progression to ESRD than the classical diarrhea-associated disease. Cases typically have large bacterial burden with empyema and bacteremia, but cases with meningitis or pericarditis have also been described. Bacterial neuraminidase exposes a crypt-antigen known as the Thomsen-Freidenrich (or T-) antigen on erythrocytes, platelets, and endothelium, and this is thought to play a role in endothelial activation and subsequent microvascular thrombus formation. T-antigen exposure can be confirmed on patient erythrocytes using the lectin Arachis hypogea. Recurrence of pneumococcal HUS after transplantation has not been reported for these patients.

HUS that cannot be associated with infection or other secondary causes (often referred to as atypical HUS) accounts for 5% to 10% of all cases in children and carries a higher risk to progress to ESRD without treatment. Genetic or acquired disorders of complement regulation are identified in about 60% to 70% of these cases. They also have higher risk of recurrence and graft loss posttransplant. Therefore identifying these cases is paramount for (1) preemptive treatment to prevent progression to ESRD and (2) planning for successful transplantation.

In recent years, a clear link has been established between disordered regulation of the alternative pathway of the complement system and atypical HUS (aHUS). Mutations have been described in three important regulatory proteins of the alternative pathway: complement factor H (CFH; 20%–30% of aHUS registry cases), complement factor I (CFI; 2%–12%), and membrane cofactor protein (MCP; 10%–15%). Gain-of-function mutations in genes encoding complement factor B (CFB; 1%–2%) and complement C3 (10%) and loss-of-function mutations in thrombomodulin (THBD gene) have also been associated with aHUS. Finally, autoantibodies to factor H are detectable in 5% to 10% of patients with aHUS, but up to 40% of these patients also carry a mutation in CFH, CFI, MCP, or C3.

Atypical HUS is also increasingly recognized as a complex, polygenic disease. First, incomplete penetrance is common in most of these mutations, suggesting that “multiple hits” contribute to a predisposition to aHUS. Second, various polymorphisms in genes encoding CFH, MCP, CFH-related protein (CFHR1), and C4b-binding protein (C4b-BP) have been associated with aHUS. Furthermore, bigenic abnormalities have been described in about 10% of patients; therefore mutational analysis of all complement components is recommended during the workup of patients suspected to have aHUS. Finally, cases of de novo thrombotic microangiopathy (TMA) after kidney transplant in patients with non-HUS causes of ESRD have also been found retrospectively to have CFH or CFI mutations.

Historically, studies on the outcomes of children with HUS have used simple delineation of cases based on the presence or absence of prodromal diarrhea. This delineation is not always clinically straightforward because shiga-toxin-related cases may present without diarrhea, and 20% to 30% of cases of aHUS due to genetic complement disorders are preceded by a diarrheal illness (including STEC infection). Furthermore, as mentioned previously, there are nondiarrheal bacterial infections associated with HUS (STEC urinary tract infection [UTI] and Pneumococcus ). A recent guideline has been published in an effort to standardize diagnostic workup and treatment for children with atypical HUS. Children without diarrheal prodrome or pneumococcal infection or those with recent diarrhea and certain clinical characteristics associated with increased risk of genetic predisposition ( Box 37.1 ) should undergo a full diagnostic evaluation for the cause of HUS, including investigation for shiga-toxin-producing bacteria as well as mutational analysis.

Recurrence after transplant can present as graft thrombosis or graft failure with hematologic signs of HUS (microangiopathic hemolytic anemia and thrombocytopenia). The risk of recurrence after transplant varies depending on the genetic mutation identified ( Table 37.2 ), and before the development of anticomplement therapy was approximately 60% overall. The Global aHUS Registry (NIH clinical trials identifier NCT01522183 ), of which approximately 20% have received kidney transplant, may provide additional information on outcomes with current therapeutic approaches. High rates of recurrence (80%) and graft loss (80%–100%) have been reported for patients with factor H or factor I mutations. In theory, patients with isolated mutations in the membrane-bound MCP protein should not develop recurrent HUS because the allograft would have wild-type, functional MCP protein. An analysis of the International Registry of Recurrent and Familial HUS/Thrombotic Thrombocytopenic Purpura (TTP) included three patients with documented MCP mutations, all with excellent graft function at 3 to 13 years posttransplant. However, HUS recurrence has been reported in a few patients with MCP mutation. Possible explanations include endothelial microchimerism, in which endothelial cells of recipient origin (expressing the mutated form of MCP) repopulate the transplanted kidney, or additional, undiagnosed, genetic susceptibility of the recipient (in circulating or fluid phase complement components). Living donation is not advised for children with potential aHUS given the high risk of recurrence and the uncertain effects of gene-gene interactions even from related donors found not to carry the same mutation as the recipient.

Plasma therapy has historically been the cornerstone of treatment for children with aHUS with CFH mutations. Infusion of fresh frozen plasma (FFP) can provide functional factor H, factor I, and C3 for patients with deficiencies in these factors, whereas plasma exchange withdraws anti-CFH antibodies and mutated forms of factor H. Preemptive plasma exchange (PE) initiated before and continuing for some time after transplant has been successful in preventing aHUS recurrence in a small number of patients with CFH mutation, but delayed recurrence can occur with tapering of therapy or during infections (especially CMV infection). Furthermore, several patients with CFH or CFI mutations have been reported to have graft loss after HUS recurrence despite PE, although these cases did not receive preemptive PE. Finally, patients with MCP mutation do not appear to benefit from plasma therapy. Liver transplant alone or in combination with kidney transplant has been reported for children with factor H mutation with the rationale that the transplanted liver would provide wild-type factor H. Although early attempts resulted in acute thrombotic events and high mortality rates, several centers in Europe have reported improved outcomes with ancillary plasma exchange and anticoagulation therapies. However, the risk of morbidity and mortality has limited the use of this approach. Successful transplantation in patients with complement factor H autoantibodies have been reported with the use of preemptive plasma exchange with or without rituximab.

Although CNIs have been associated with de novo HUS after solid-organ transplant, avoidance of CNIs has not been shown to affect the risk of recurrence in genotyped patients with aHUS.

Finally, the use of the anti-C5 monoclonal antibody, eculizumab, to prevent membrane attack complex formation holds promise for both the prevention and treatment of recurrence in children with aHUS undergoing renal transplantation. Moreover, the use of eculizumab may prevent progression to ESRD and obviate the need for renal transplantation. Optimal dosage and interval between eculizumab infusions remain to be fully defined, and the high cost of this agent may limit access. Although studies of long-term outcomes are not available, a requirement for life-long therapy is anticipated.

Membranous Nephropathy

Membranous nephropathy is rare in children at only 0.5% of incident ESRD in the US, so risk of recurrence in children after transplant is not clear.

Systemic Lupus Erythematosis

Data on lupus nephritis recurrence in children are scarce. This is likely owing to the late presentation of recurrence, which for most patients who are transplanted as adolescents would occur in adulthood. An analysis of the NAPRTCS registry reported similar patient and graft survival for children with systemic lupus erythematosis (SLE) compared with matched controls. There was an increased incidence of recurrent rejection episodes in SLE patients receiving living donor transplants for which there is no current explanation.

c-ANCA and p-ANCA-Positive Glomerulonephritis

Pauci-immune glomerulonephritis associated with antineutrophil cytoplasmic antibody with cytoplasmic (c-ANCA) or perinuclear (p-ANCA) staining patterns are a rare cause of ESRD in children, estimated at 2% of incident ESRD cases in the US. Recurrence rates of the small vessel vasculitides (SVV) in the adult literature are low at about 5% to 6%, and recurrence of granulomatosis with polyangiitis (or GPA; formerly known as Wegener granulomatosis) is rare. Average time to recurrence is 31 months, but it can occur within weeks to many years after transplant. Graft loss due to recurrence has been reported in 2% to 7% of adults transplanted with diagnosis of SVV.

Renal recurrence is often heralded by microscopic hematuria and proteinuria with focal or diffuse pauci-immune necrotizing glomerulonephritis seen on biopsy. The ANCA pattern or titers at the time of transplant do not appear to be predictive of disease recurrence. Similar to SLE, a waiting period of 6 to 12 months of inactive disease is recommended before transplant. Persistently positive ANCA serologies, however, should not preclude transplant because they are not an accurate marker of disease activity. There are anecdotal reports of the use of cyclophosphamide, corticosteroids, mycophenolate, and plasmapheresis for treatment of recurrence in patients with GPA and p-ANCA associated glomerulonephritis.

Primary Hyperoxaluria Type I (Oxalosis)

Primary hyperoxaluria type I (PH1, also known as oxalosis) is a rare autosomal recessive disorder caused by a defect in hepatic alanine:glyoxylate aminotransferase (AGT), which catalyzes the conversion of glyoxylate to glycine. AGT deficiency results in overproduction of oxalate, resulting in massive renal excretion of insoluble calcium oxalate leading to nephrolithiasis and nephrocalcinosis. As GFR declines with progressive renal involvement, oxalate accumulates and results in systemic oxalosis. More than 100 different mutations have been described, and there is considerable phenotypic heterogeneity, even within family members with identical mutations.

The most severe form presents in infancy with renal failure necessitating dialysis. In children with less severe presentation and early diagnosis, conservative management with pyridoxine (thought to reduce oxalate production in a subset of B6-responsive patients), increased fluid intake, and citrate treatment might delay progression of kidney disease. Some centers advocate for preemptive liver transplant or combined liver–kidney transplant because liver transplantation corrects the underlying metabolic disorder. The use of hemodialysis before transplant has also been advocated to reduce systemic oxalate levels to mitigate injury to the allograft. Data from the NAPRTCS registry indicate poor patient and graft survival for children with oxalosis after kidney-only transplant with high rates of recurrence and death from sepsis. Recent longitudinal studies from Europe suggest improved outcomes in recent years, with the best outcomes in children who were diagnosed early and underwent combined kidney–liver transplant.

Nephropathic Cystinosis

Cystinosis is a rare, autosomal recessive disease due to a defect in the lysosomal cystine transporter (encoded by the cystinosin gene), resulting in intracellular accumulation of cystine, proximal tubule dysfunction (renal Fanconi syndrome), and progressive kidney disease. Since the development of the cystine-depleting drug, cysteamine, progression to ESRD may be delayed. Although nephropathic cystinosis does not recur posttransplant, protocol biopsies have shown interstitial deposition of cystine crystals without apparent clinical consequences. Graft survival rates for children with cystinosis are comparable or even improved compared with peers with other ESRD etiologies. Extrarenal manifestations of continued cystine accumulation (i.e., visual impairment, hypothyroidism, endocrine pancreatic insufficiency, and myopathy) have become more apparent as the lifespan of patients with cystinosis increases and likely are postponed by continued cysteamine treatment posttransplant.

Medical Evaluation of Issues Related to ESRD

Evaluation of Extrarenal Disease

Surgical Evaluation

Children undergoing kidney transplant evaluation often need multiple surgical interventions, including bladder augmentation, placement of Mitrofanoff continent urinary diversion, vesicostomy closure, gastrostomy tube placement, or native nephrectomy before or at the time of transplant. Therefore it is vital that a surgical plan be established among pediatric surgeons, transplant surgeons, and urologists before kidney transplant in children to coordinate surgical approaches and spare or maintain the vascular supplies that are unique for each procedure.

Neurologic Development

Psychosocial Issues

Evaluation Updates

Pediatric candidates for kidney transplant should be reevaluated at regular intervals (generally every 6–12 months) to identify any changes in their medical condition or psychosocial status that might alter the risk of commencing with transplantation. A simplified version of the initial medical evaluation is appropriate for update visits.