Hemolytic Uremic Syndrome, Atypical, Susceptibility To, 1
A number sign (#) is used with this entry because susceptibility to the development of atypical hemolytic uremic syndrome-1 (AHUS1) can be conferred by variation in the gene encoding complement factor H (CFH; 134370) on chromosome 1q31.
Deficiency of the CFH-related proteins CFHR1 (134371) and CFHR3 (605336) may be associated with development of the disorder due to autoantibodies. Other genes may play a role in modifying the phenotype (see MOLECULAR GENETICS).
HUS can share overlapping clinical features with thrombotic thrombocytopenic purpura (TTP; 274150), which is caused by mutation in the von Willebrand factor-cleaving protease (VWFCP) gene (ADAMTS13; 604134).
DescriptionTypical hemolytic uremic syndrome is characterized by acute renal failure, thrombocytopenia, and microangiopathic hemolytic anemia associated with distorted erythrocytes ('burr cells'). The vast majority of cases (90%) are sporadic, occur in children under 3 years of age, and are associated with epidemics of diarrhea caused by verotoxin-producing E. coli. The death rate is very low, about 30% of cases have renal sequelae, and there is usually no relapse of the disease. This form of HUS usually presents with a diarrhea prodrome (thus referred to as D+HUS) and has a good prognosis in most cases. In contrast, a subgroup of patients with HUS have an atypical presentation (aHUS or D-HUS) without a prodrome of enterocolitis and diarrhea and have a much poorer prognosis, with a tendency to relapse and frequent development of end-stage renal failure or death. These cases tend to be familial. Both autosomal recessive and autosomal dominant inheritance have been reported (Goodship et al., 1997; Taylor, 2001; Veyradier et al., 2003; Noris et al., 2003). Noris and Remuzzi (2009) provided a detailed review of atypical HUS.
Genetic Heterogeneity of Atypical Hemolytic Uremic Syndrome
Atypical HUS is a genetically heterogeneous condition. Susceptibility to the development of the disorder can be conferred by mutations in various components of or regulatory factors in the complement cascade system (Jozsi et al., 2008). See AHUS2 (612922), AHUS3 (612923), AHUS4 (612924), AHUS5 (612925), and AHUS6 (612926). AHUS7 (see 615008) is caused by mutation in the DGKE gene (601440), which is not part of the complement cascade system.
Clinical FeaturesHagge et al. (1967) reported the hemolytic uremic syndrome in 2 sibs. Features included intravascular hemolysis, thrombocytopenia, and azotemia. One had repeated attacks ending in renal failure and death at age 8 years; the other recovered completely after one attack. Chan et al. (1969) reported HUS in 2 adopted, unrelated sibs.
Kaplan et al. (1975) reported HUS in 3 sibs and reviewed reports of 21 sibships with 2 or more affected individuals. Two groups of families could be identified among 41 analyzed. Sibs whose onset was within a short time of each other had a relatively good prognosis (19% mortality). Those whose onset was more than a year apart had a poorer prognosis (68% mortality). Kaplan et al. (1975) suggested that an environmental agent was causative in the first group and that genetic factors were important in the second. Most of the first group of families came from an endemic area, whereas most of the second group came from a nonendemic area.
Blattler et al. (1975) studied a family in which 4 sibs had died from HUS. The parents and 4 surviving sibs had normal renal function and normal platelet and fibrinogen survival. The mother and 3 sibs had an increased percentage of megathrombocytes. Two of them showed renal accumulation of Cr 51-platelet radioactivity and ultrastructural changes of the endothelium on renal biopsy.
Edelsten and Tuck (1978) reported a family with HUS inherited in an autosomal dominant pattern.
Thompson and Winterborn (1981) reported an 8-month-old Asian boy with very low levels of plasma factor H who presented with the hemolytic uremic syndrome. Complement component C3 (120700) was also depleted. A healthy 3-year-old brother had the same complement profile, suggesting activation of the alternative complement pathway. The parents, who were first cousins, had half-normal levels of factor H.
Kirchner et al. (1982) described this disorder in mother and daughter. The daughter's illness, characterized primarily by renal insufficiency, was most compatible with adult hemolytic uremic syndrome and the mother's illness, with prominent neurologic findings, was most compatible with thrombotic thrombocytopenic purpura. Merrill et al. (1985) reported 2 certain cases and 3 possible case in 2 generations of a North Carolina black family.
Neuhaus et al. (1997) reported clinical features of atypical D-HUS in 23 children. Features included requirement for dialysis (74%), hypertension (43%), cardiomyopathy (43%), and cerebral convulsions (48%). Only 5 patients (26%), including 4 infants, recovered completely. Six patients (32%) had 1 to 10 recurrences and 8 (42%) developed end-stage renal failure. Four children died.
Warwicker et al. (1998) reported a 36-year-old man with sporadic relapsing aHUS. He had anemia, thrombocytopenia, hypertension, and acute renal failure. Renal biopsy showed a thrombotic microangiopathy and deposition of complement component C3 in vessel walls. He had half-normal serum levels of factor H and decreased levels of C3, consistent with activation of the alternative complement pathway. HUS recurred after renal transplantation. Molecular analysis revealed a heterozygous 4-bp deletion in the CFH gene (134370.0011).
Ohali et al. (1998) reported a large consanguineous Bedouin family in which 10 infants had atypical HUS characterized by microangiopathic hemolytic anemia, acute renal failure, severe hypertension, edema, and increased serum triglycerides. All had very early onset with a median presentation at age 2 weeks. Two patients died during the first disease episode; the remaining 8 patients had a mean number of 4 relapses despite treatment. In total, 8 patients died at ages 3 weeks to 10 months. Factor H levels were low to undetectable in all 4 patients studied, and C3 levels were decreased in 9 of 10 infants tested. Four kidney biopsies showed marked arteriolar changes, including stenosis, edema, and thickening of the intima. Glomerular changes included swelling of endothelial cells with microvillus transformation and proliferation of mesangial cells with increased matrix deposition in the mesangium. Other changes included fibrotic changes in glomeruli and the tubulointerstitium, as well as C3 deposition in capillary walls.
Rougier et al. (1998) reported 6 children with complement factor H deficiency and acute glomerular disease. Five of the 6 children presented with hemolytic uremic syndrome. Two of the children were from a consanguineous family from Turkey and exhibited a homozygous deficiency characterized by absence of the 150-kD form of factor H.
Noris et al. (2003) reported a 21-year-old woman and her affected brother. Disease onset in the sister (the proband) was at age 16 months, when she developed fever, hemolytic anemia, and thrombocytopenia. At the time, renal function was normal. Thereafter she had 6 recurrences of thrombotic microangiopathy, all associated with deteriorating renal function. Treatment consisted of plasma exchange and infusions, steroids, and blood transfusions, which led to complete recovery of blood abnormalities and renal function. The last episode occurred at age 20 years and was characterized by anemia, thrombocytopenia, and severe impairment of renal function. Renal biopsy showed irreversible changes of chronic nephropathy with typical features of HUS, including diffuse narrowing/occlusion of vessels and severe glomerular ischemia. At the age of 21 years she was on chronic dialysis. The proband's brother had 2 episodes of HUS at age 9 years. Both were characterized by severe hemolytic anemia and acute renal insufficiency and resolved without plasma treatment, with no renal sequelae.
Clinical ManagementHazani et al. (1996) reported relapsing thrombotic microangiopathy in a 12-year-old girl and her 7-year-old brother. During 11 years of follow-up, the girl responded only to steroids; many other therapeutic modalities were ineffective. Following treatment with low-dose danazol, relapses became fewer and less severe, completely subsiding after 6 months. The brother's illness began with signs of hemolytic uremic syndrome, with later development of neurologic manifestations. During a 6-year follow-up he responded only to plasma exchange. Although chronic thrombocytopenia persisted during the last 3 years, the boy's clinical condition improved.
Landau et al. (2001) described 2 patients with atypical HUS associated with factor H deficiency. One patient who underwent renal transplantation for end-stage renal disease later had an extensive nonhemorrhagic cerebral infarction on 2 occasions and died in spite of multiple plasma transfusions. A second patient, a 14-month-old boy, experienced numerous HUS episodes starting at the age of 2 weeks. Daily plasma transfusions during relapses brought about only a temporary state of remission. However, prophylactic twice-weekly plasma therapy had been successful in preventing relapses and preserving renal function. Landau et al. (2001) reported that with this regimen, serum factor H was increased from 6 mg/dL to subnormal values of 12 to 25 mg/dL (normal, greater than 60 mg/dL). Landau et al. (2001) concluded that aHUS recurs because factor H deficiency is not corrected by renal transplantation. A hypertransfusion protocol may be useful.
InheritanceConcordant monozygotic twins have been reported (Campbell and Carre, 1965). Farr et al. (1975) described a family with several affected members, including a father and his son and daughter. A common symptom was hypertension. They reviewed reports of familial occurrence. Perret et al. (1979) described this disorder in 5 members of 3 generations of a kindred and suggested genetic predisposition with a dominant gene.
In the study by Furlan et al. (1998), there were multiple patients with the familial form of HUS: patients 44 and 45 were sister and brother and had 8 sibs who had died of acute HUS; patient 46 (a male) had 3 brothers who had died of HUS; from 2 other unrelated families, patients 47 and 48 were sisters and patients 49 and 50 were brothers; 1 brother and 1 sister of patient 51 had died of HUS; patients 52 and 53 were brother and sister.
Goodship et al. (1997) stated that most familial cases are recessive but dominant pedigrees had also been reported.
In a review of thrombotic microangiopathies, Moake (2002) stated that 5 to 10% of cases of HUS are familial. The mortality rate (54%) is much higher in the familial form than in typical childhood HUS (3 to 5%). About half of the survivors have relapses and over one-third require long-term dialysis. Among patients with familial HUS who receive kidney allografts, 16% lose function in the engrafted kidney within 1 month.
PathogenesisAtypical Hemolytic Uremic Syndrome
Low levels of factor H in patients with aHUS were reported by Roodhooft et al. (1990), Pichette et al. (1994), and others. Some patients, however, may have normal levels of factor H, suggesting a dysfunction of the protein (Warwicker et al., 1998).
Reduced serum C3 levels have been reported in sporadic (Stuhlinger et al., 1974; Robson et al., 1992) and familial HUS (Zachwieja et al., 1992). In Italy, Noris et al. (1999) studied 6 families with HUS, 1 family with TTP, and 2 families with both disorders. Included in the study were 15 patients and 63 available healthy relatives as well as 25 age- and gender-matched healthy controls and 56 of their available relatives. Consanguinity was observed in 2 families. Seventy-three percent of the cases versus 16% of controls (P less than 0.001) as well as 24% of case-relatives versus 5% of control-relatives (P = 0.005) had decreased C3 levels, which were more marked in the actual cases. Factor H abnormalities were found in 4 of the 15 cases as compared with 3 of the 63 case-relatives and none of 17 healthy controls. All cases with factor H abnormalities had low C3 serum concentrations. Noris et al. (1999) concluded that reduced C3 in familial HUS is likely related to a genetically determined deficiency of factor H.
Remuzzi et al. (2002) concluded that ADAMTS13 activity does not distinguish TTP from HUS, at least in the recurrent and familial forms, and that it is not the only determinant of VWF (613160) abnormalities in these conditions.
In 41 children with D+HUS and 23 children with D-HUS, Veyradier et al. (2003) found that von Willebrand factor-cleaving protease activity was normal in over 50% of patients, but was undetectable in 1 D+HUS and 6 D-HUS children. After a 3-month remission, the D+HUS patient recovered 100% VWFCP activity, whereas the 6 D-HUS patients kept an undetectable level. In these 6 D-HUS patients, the disease was characterized by a neonatal onset and several relapses of hemolytic anemia, thrombocytopenia, acute renal failure, and cerebral ischemia. Arterial hypertension and end-stage renal failure sometimes occurred. Veyradier et al. (2003) concluded that a subgroup of patients with D-HUS is related to VWFCP and may actually correspond to TTP.
Using in vitro expression studies, Manuelian et al. (2003) demonstrated that pathogenic mutations in the CFH gene (134370.0001; 134370.0017-134370.0018) resulted in mutant proteins with decreased binding to heparin, C3b/C3d, and human endothelial cells. The findings suggested that reduced interaction with the surface of endothelial cells is central to the pathophysiology of aHUS and that normal factor H has a protective role during tissue injury.
Stahl et al. (2008) demonstrated that aHUS-associated mutant factor H (see, e.g., 134370.0022) exhibited decreased binding to normal platelets compared to wildtype factor H. Addition of patient serum containing mutant factor H to control platelets resulted in complement activation, deposition of C3 and C9, release of platelet-derived microparticles, and platelet aggregation, indicating platelet activation. Similar findings were obtained with other aHUS-associated CFH mutations. Preincubation of normal platelets with factor H reduced these effects. The findings indicated that mutant CFH results in complement activation on the surface of platelets and platelet activation, which may contribute to thrombocytopenia.
Autoantibodies in aHUS
Dragon-Durey et al. (2005) identified serum anti-CFH IgG autoantibodies in 3 (6%) of 48 children with recurrent aHUS. Plasma CFH activity was decreased, whereas plasma CFH antigen levels were normal and CFH gene analysis was normal, indicating an acquired functional CFH deficiency. The findings indicated that aHUS may occur in the context of an autoimmune disease, and suggested that plasma exchange or immunosuppression may be a beneficial treatment.
Of 147 patients with aHUS, 121 of whom had previously been reported by Zipfel et al. (2007), Jozsi et al. (2008) identified serum anti-CFH autoantibodies in 16 (11%); 14 lacked CFHR1/CFHR3 completely and 2 showed extremely low CFHR1/CFHR3 plasma levels. These observations suggested that CFHR1/CFHR3 deficiency represents a risk factor for CFH autoantibody formation. Unaffected family members with decreased CFHR1/CFHR3 did not have CFH autoantibodies. The binding epitopes of all autoantibodies were localized to the C-terminal recognition region of factor H, which represents a hotspot for aHUS mutations. The authors thus defined a subgroup of aHUS, which they termed DEAP HUS (deficiency of CFHR proteins and CFH autoantibody positive), that is characterized by a combination of genetic and acquired factors. The findings illustrated a new combination of 2 susceptibility factors for the development of aHUS.
Dragon-Durey et al. (2009) found a deletion of 1 or both CFHR1/CFHR3 alleles in 22.7% of 144 French aHUS patients and only 8.2% of 70 healthy controls. The highest deletion frequency was in the subgroup of aHUS patients with anti-factor H autoantibodies (92.9% had 1 or 2 deleted alleles) and in the group of patients with a CFI mutation (31.8% had 1 or 2 deleted alleles). Deletion of CFHR1/CFHR3 was not significantly more frequent in those patients in whom anti-CFH antibodies or CFI mutation were excluded. The findings indicated that genomic deletion of CFHR1/CFHR3 plays a role in the development of anti-CFH autoantibodies, but likely has only a secondary role in susceptibility to aHUS.
Typical Hemolytic Uremic Syndrome
Kaplan and Drummond (1978) noted that typical HUS is triggered by specific infection. Typical HUS follows a prodrome of acute afebrile gastroenteritis, often with bloody stools. HUS and a hitherto poorly recognized condition, haemorrhagic colitis, which is clinically and pathologically similar to the prodromal bloody diarrheal phase of classic HUS, are related in a causal way to verocytotoxin-producing Escherichia coli (VTEC) infection (Karmali et al., 1985). Verocytotoxin refers to the capacity of this family of potent protein exotoxins to produce an irreversible cytopathic effect in certain cultured cell lines, especially Vero and HeLa. The toxin is also lethal to laboratory animals, especially rabbits, in minute doses. E. coli O157:H7, the most frequently isolated serotype of verotoxin-producing E. coli in the United States, is capable of causing a broad spectrum of illness, including nonbloody diarrhea, bloody diarrhea, hemolytic uremic syndrome, and thrombotic thrombocytopenic purpura (Griffin et al., 1988).
Proulx et al. (2002) compared the circulating concentrations of granulocyte colony-stimulating factor (GCSF; 138970) and several chemokines in the course of E. coli O157:H7 enteritis, hemorrhagic colitis (HC), and HUS. They found that GROA (155730), CCL4 (182284), and MCP1 (158105) are produced whether or not HC or HUS develops. They also found that children with O157:H7-associated HUS may present abnormally increased circulating levels of GCSF and decreased levels of ENA78 (CXCL5; 600324). The authors concluded that leukocytes may be involved in the pathogenesis of HUS.
MappingGoodship et al. (1997) undertook a candidate gene linkage study in 2 families with autosomal dominant HUS and in 1 family with a pedigree compatible with recessive inheritance or dominant inheritance with partial penetrance. The disease segregated with the region of 1q containing the candidate HF1 gene (CFH; 134370); maximum lod = 3.94.
Warwicker et al. (1998) demonstrated that familial HUS segregated with the chromosome 1q region, bounded by the markers D1S212 and D1S306, containing the factor H gene.
Molecular GeneticsIn affected members of a large family with autosomal dominant aHUS originally reported by Edelsten and Tuck (1978), Goodship et al. (1997) and Warwicker et al. (1998) identified a heterozygous mutation in the CFH gene (134370.0001). Although none of the patients had decreased levels of plasma factor H, Warwicker et al. (1998) postulated that the mutation disrupted the structure and function of the protein. In an adult with sporadic HUS, Warwicker et al. (1998) identified a heterozygous 4-bp deletion in the HF1 gene (134370.0011).
In affected members of a Bedouin kindred with atypical HUS and factor H deficiency originally reported by Ohali et al. (1998), Ying et al. (1999) identified a homozygous mutation in the CFH gene (134370.0004). In this same family, Buddles et al. (2000) identified a different homozygous pathogenic mutation in the CFH gene (134370.0005).
In 2 children from Turkey with relapsing HUS originally reported by Rougier et al. (1998), Dragon-Durey et al. (2004) identified a homozygous mutation in the CFH gene (134370.0012). Dragon-Durey et al. (2004) also identified heterozygous mutations in the CFH gene in 2 additional patients with atypical HUS reported by Rougier et al. (1998).
Caprioli et al. (2003) analyzed the complete HF1 gene in 101 patients with HUS, 32 patients with TTP, and 106 controls in order to evaluate the frequency of HF1 mutations, the clinical outcome in mutation and nonmutation carriers, and the role of HF1 polymorphisms in the predisposition to HUS. They found 17 different HF1 mutations (16 heterozygous and 1 homozygous) in 33 HUS patients; 13 mutations were located in exons 22 and 23. No TTP patient carried HF1 mutations. HUS manifested earlier and the mortality rate was higher in mutation carriers than in noncarriers. Kidney transplants invariably failed for disease recurrences in patients with HF1 mutations, while in patients without HF1 mutations half of the grafts were functioning after 1 year. Three HF1 polymorphisms were strongly associated with aHUS: a -257T promoter allele, a 2089G allele in exon 14, and a 2881T allele in exon 19, resulting in an glu963-to-asp substitution. Two or 3 disease-associated variants led to a higher risk of HUS than 1 alone. Analysis of available relatives of mutation patients revealed a penetrance of 50%. In 5 of 9 families the proband inherited the mutation from 1 parent and 2 disease-associated variants from the other, while unaffected carriers inherited the protective variants.
Caprioli et al. (2006) identified mutations in the CFH gene in 47 (30%) of 156 patients with aHUS. Most were heterozygous, but some families had homozygous mutations. Most of the mutations were spread over the 5 exons that encode the most C-terminal part of CFH, often clustering in SCR20, which is involved in binding to surface-bound C3b. In a comparison of 14 aHUS patients with MCP mutations and 42 with CFH mutations, Caprioli et al. (2006) found that patients with MCP mutations had an overall better prognosis. Complete remission was observed in 85.7% of patients with MCP mutations compared to 17.5% of patients with CFH mutations. Only 1 patient with an MCP mutation developed end-stage renal failure and none died during the acute episode, whereas end-stage renal failure and death occurred in 22% and 30%, respectively, of patients with CFH mutations. Most (85%) of the patients with MCP mutations retained normal renal function compared to 22.5% of patients with CFH mutations.
Modifier Genes
Zipfel et al. (2007) found that an 84-kb deletion of the CFHR1 (134371.0001) and CFHR3 (605336.0001) genes was associated with an increased risk of atypical hemolytic-uremic syndrome in 2 independent European cohorts. In the first group, 19 (16%) of 121 aHUS patients had the deletion compared to 2 of 100 control individuals. Three of the patients had a homozygous deletion. All patients had normal serum factor H levels. In the second group comprising 66 patients, 28% had the deletion compared to 6% of controls. Ten percent and 2% of patients and controls, respectively, were homozygous for the deletion. In vitro functional expression studies showed that CFHR1/CFHR3-deficient plasma had decreased protective activity against erythrocyte lysis, suggesting a defective regulation of complement activation. Of 147 patients with aHUS, 121 of whom had previously been reported by Zipfel et al. (2007), Jozsi et al. (2008) identified serum anti-CFH autoantibodies in 16 (11%); 14 lacked CFHR1/CFHR3 completely and 2 showed extremely low CFHR1/CFHR3 plasma levels. The findings illustrated a new combination of 2 susceptibility factors for the development of aHUS.
Blom et al. (2008) identified an arg240-to-his (R240H) SNP in the C4BPA gene (120830) that was associated with aHUS. The heterozygous change was found in 6 of 166 patients with aHUS and in 5 of 542 healthy controls. Three of the 6 patients with this SNP carried mutations in other known aHUS susceptibility genes, including MCP and CFH. The findings were replicated in another sample. Functional expression studies showed that the C4BPA variant had impaired ability to bind C3b and to act as a cofactor in its degradation. The findings supported the hypothesis that dysregulation of the alternative complement pathway can lead to aHUS.
Associations Pending Confirmation
For discussion of a possible relationship between variation in the CFHR5 gene and atypical hemolytic uremic syndrome, see 608593.0003-608593.0005.
Population GeneticsTypical Hemolytic Uremic Syndrome
Gianantonio et al. (1968) observed 75 cases of HUS in Argentina, where the disorder seems unusually frequent, and assembled some evidence for viral etiology. Endemic areas included Argentina, South Africa, the west coast of the United States, and the Netherlands.
Tarr et al. (1989) reported an increase in the incidence of HUS in King County, Washington, during the previous 15 years. In Minnesota, Martin et al. (1990) reported an increase in mean annual incidence from 0.5 case per 100,000 child-years among children less than 18 in 1979 to 2.0 cases per 100,000 in 1988 (P = 0.000004). Of 28 patients, 13 (46%) showed E. coli O157:H7 in stool specimens. Patients were more likely than controls to attend large day-care centers, suggesting that such attendance is a risk factor for HUS. On the basis of the population-attributable risk, however, this factor could account for no more than 16% of the cases.
HistoryRemuzzi et al. (1979) suggested that deficiency of a vascular prostacyclin stimulator may underlie the disorder. Plasma from a 54-year-old woman with HUS had a low capacity to stimulate PGI2 production by rat aortic rings. Plasma treatment restored this activity. PGI2-stimulating activity was normal in 2 daughters of the proband but consistently low (20-50% of control) in both of her sons, neither of whom had a history or clinical signs of a microangiopathic disorder.