The n e w e ng l a n d j o u r na l of m e dic i n e

review article

Medical Progress

Atypical Hemolytic–Uremic Syndrome

Marina Noris, Ph.D., and Giuseppe Remuzzi, M.D.

T
From the Clinical Research Center for Rare he hemolytic–uremic syndrome is characterized by nonimmune
Diseases Aldo e Cele Daccò, Mario Negri hemolytic anemia, thrombocytopenia, and renal impairment.1 The disorder
Institute for Pharmacological Research
(M.N., G.R.), and the Division of Neph- occurs most frequently in children under the age of 5 years, with an annual
rology and Dialysis, Ospedali Riuniti di incidence of 6.1 cases per 100,000 children under 5 years, as compared with an
Bergamo (G.R.) — both in Bergamo, Italy. overall incidence of 1 to 2 cases per 100,000. The presentation is generally heralded
Address reprint requests to Dr. Remuzzi
at the Mario Negri Institute for Pharma- by diarrhea, which is often bloody.2,3 Most cases (including more than 90% of those
cological Research, Via Gavazzeni 11, in children) are secondary to infection with Escherichia coli serotypes O157:H7,
24125 Bergamo, Italy, or at gremuzzi@ O111:H8, O103:H2, O123, O26, or others,1 which produce Shiga-like toxin (Stx),
marionegri.it.
and several other bacteria, such as Streptococcus pneumoniae.4
N Engl J Med 2009;361:1676-87. Approximately 10% of cases of the hemolytic–uremic syndrome are classified as
Copyright © 2009 Massachusetts Medical Society. atypical, since they are not caused by either Stx-producing bacteria or streptococci.1,5
Atypical hemolytic–uremic syndrome has a poor prognosis, with death rates as
high as 25%5 and progression to end-stage renal disease in half the patients.1,4 Re-
search has linked atypical hemolytic–uremic syndrome to uncontrolled activation
of the complement system. This article reviews current concepts about the patho-
biology of this syndrome and its diagnosis and management.

The His t ol o gic L e sion

The lesions of Stx-related hemolytic–uremic syndrome, which are indistinguishable

from those of its atypical form on the basis of standard histologic analysis, are char-
acterized by thickening of arterioles and capillaries, endothelial swelling and detach-
ment, and subendothelial accumulation of proteins and cell debris (Fig. 1).6,7 The sub-
endothelial space is widened, and platelet thrombi obstruct vessel lumina. Hemolysis
occurs, and fragmented or distorted erythrocytes are evident in blood smears. Le-
sions typically affect the kidney (mainly glomeruli and arterioles), although the brain,
heart, lungs, gastrointestinal tract, and pancreas all may be involved.

Cl a ssific at ion of Dise a se

Familial Form
Less than 20% of cases of atypical hemolytic–uremic syndrome are familial (Table 1).
Patients with the familial form of the disease have a poor prognosis, with a rate of
either end-stage renal disease or death of 50 to 80%. In 1965, a combination of hemo-
lytic anemia and azotemia was described in concordant monozygous twins.8 Since
that time, familial atypical hemolytic–uremic syndrome has been reported in chil-
dren and, infrequently, in adults. Both autosomal dominant and recessive patterns
of inheritance have been reported.9

Sporadic Form
Atypical hemolytic–uremic syndrome that develops in patients who do not have a
family history of the disease is classified as sporadic. Triggers for the sporadic form

1676 n engl j med 361;17  nejm.org  october 22, 2009

The New England Journal of Medicine

Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 medical progress

include infection with the human immunodefi- On host cells, complement activation is controlled
ciency virus, cancer, organ transplantation, preg- by both membrane-anchored and fluid-phase reg-
nancy, and the use of certain anticancer drugs, ulators, favoring the cleavage of C3b to inactive
immunotherapeutic agents (e.g., cyclosporine and C3b (iC3b) by complement factor I (CFI) (i.e., co-
tacrolimus), and antiplatelet agents (e.g., ticlopi- factor activity) and dissociating the multicompo-
dine and clopidogrel).2,6,10 nent C3 and C5 convertases (i.e., decay-acceleration
De novo atypical hemolytic–uremic syndrome activity). Without normal regulation, C3b depo-
has been reported in 3.6 to 14.0% of all kidney- sition increases by more than a factor of 2030
transplant recipients in association with humoral through the amplification loop and causes acti-
rejection and the use of calcineurin inhibitors.11-13 vation of the complement cascade, which remains
In approximately 10 to 15% of female patients with so until complement components are consumed.
atypical hemolytic–uremic syndrome, the disorder Foreign targets and injured cells that either do
develops during pregnancy or post partum.1,2 Ap- not have membrane-bound regulators or cannot
proximately 50% of sporadic cases appear to be bind soluble regulators are attacked by comple-
idiopathic. ment. On the surface of bacteria, C3b binds to
Genetic abnormalities in complement system specific receptors on neutrophils and macrophag-
proteins have been documented in the familial es, resulting in phagocytosis of complement-tagged
form of the disease14-18 and also in the sporadic bacteria.
(mainly idiopathic) form (Table 1).14,19-21 Two pa- The C3 convertases of the classic and lectin
tients with Stx-related hemolytic–uremic syndrome pathways are formed by C2 and C4 fragments,
have been reported to have mutations in comple- whereas the alternative pathway convertase cleaves
ment regulatory genes,20,22 but the frequency of C3 but not C4.29 Thus, a low serum C3 level in a
the mutations in such patients is not known. patient with atypical hemolytic–uremic syndrome
who has a normal C4 level indicates selective ac-
C ompl emen t A bnor m a l i t ie s tivation of the alternative pathway.25

Since 1974, reduced serum levels of complement Gene t ic A bnor m a l i t ie s

fraction C3 with normal levels of C4 have been
reported in patients with atypical hemolytic–ure- Complement Pathway Mutations
mic syndrome.23-25 A low C3 level reflects com- A variety of mutations in members of the comple-
plement activation and consumption (Fig. 2). Pa- ment pathway have been described in patients with
tients with the hemolytic–uremic syndrome who atypical hemolytic–uremic syndrome. These mu-
have low C3 levels have high levels of activated tations have been found to account for 50 to 60%
complement components, including C3b, C3c, and of cases (Table 2, and Fig. 2 in the Supplementary
C3d.26 Granular C3 deposits in glomeruli and ar- Appendix).
terioles during acute disease are consistent with
the activation of complement and local C3 con- Complement Factor H
sumption.24,27 C9 staining in glomeruli and small In 1981, investigators described two brothers with
arteries with intimal proliferation and thrombosis atypical hemolytic–uremic syndrome who did not
documents activation up to the final lytic C5b-9 produce complement factor H (CFH), the plasma
membrane-attack complex.28 regulator of the alternative pathway.31 The par-
The complement system consists of several ents, who were first cousins, had half-normal CFH
plasma and membrane-bound proteins that pro- levels, indicating an inherited defect. Subsequent-
tect against invading organisms.29 Three activa- ly, complete or partial CFH deficiencies have been
tion pathways — classic, lectin, and alternative reported in patients with atypical hemolytic–ure-
— produce protease complexes termed C3 and C5 mic syndrome.25,32 In 1998, a group of investiga-
convertases that cleave C3 and C5, respectively, tors headed by Goodship (Warwicker et al.33)
eventually leading to the membrane-attack com- showed an association between atypical hemolytic–
plex (Fig. 1 in the Supplementary Appendix, avail- uremic syndrome and the chromosome 1q32 lo-
able with the full text of this article at NEJM.org). cus, which contains genes for CFH and other com-
C3 hydrolysis in plasma initiates the alternative plement regulators. The investigators found a
pathway, leading to the deposition of C3b onto heterozygous CFH mutation in patients with the
practically all plasma-exposed surfaces (Fig. 2A).30 syndrome and obligate carriers in one family and

n engl j med 361;17  nejm.org  october 22, 2009 1677

The New England Journal of Medicine
Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A B

C D

Figure 1. Micrographs of Samples from Patients with Atypical Hemolytic–Uremic Syndrome.

Panels A and B show light micrographs of glomeruli from a patient with a heterozygous complement factor H
(CFH) mutation. In Panel A, intracapillary thrombosis, congestion, and thickening of the capillary wall are clearly
visible (Masson’s trichrome). In Panel AUTHOR
ICMB, there Noris/Remuzzi
is marked RETAKE and 1st
glomerular retraction wrinkling of the capillary tuft
REG F FIGURE 1a-f 2nd
(silver stain). Panel C shows renal arterioles from a patient with a heterozygous complement factor I (CFI) muta-
3rd
tion. An arteriole shows endothelialCASE
swelling and hyperplasia with narrowingRevised
TITLE of the lumen (arrow), and the lumen
EMail
of another vessel is occluded (arrowhead) (periodic acid–Schiff).
Line Panel 4-C D shows an electron micrograph of the
SIZE
glomerular capillary wall in a sampleEnon
from ARTIST:
a patientmst
with a heterozygous
H/T H/T C3 mutation. The endothelium (arrow) is
FILL Combo 33p9
detached from the glomerular basement membrane. The subendothelial space (arrowhead) is widened and occu-
AUTHOR,
pied by electron-lucent fluffy material and cell debris, withPLEASE NOTE:
a marked narrowing of the capillary lumen. Podocyte foot
Figure has been redrawn and type has been reset.
processes are focally effaced (uranyl acetate–osmium tetroxide).
Please check carefully.

JOB: 36117 ISSUE: 10-22-09

another heterozygous mutation in a patient with ogous repeats, regulates the alternative pathway by
a sporadic form of the disease, which suggested competing with complement factor B (CFB) for
that sporadic forms of the syndrome had a ge- C3b recognition by acting as a cofactor for CFI
netic basis. and by enhancing dissociation of C3 convertase
(Fig. 2A and 2B).39 These functions are located in
CFH Point Mutations the N-terminal region of CFH. In addition, CFH
More than 80 mutations in CFH have been identi- binds to glycosaminoglycans in basement mem-
fied in patients with atypical hemolytic–uremic branes and endothelium through its C-terminal
syndrome, with a mutation frequency of 40 to repeats.40 CFH down-regulates alternative-path-
45% in patients with the familial form and of 10 way activation on structures without other com-
to 20% in those with the sporadic form.14,34-38 plement regulators, such as glomerular basement
(Details about mutations in atypical hemolytic– membranes, and contributes to endothelial pro-
uremic syndrome are available at www.FH-HUS. tection when membrane-bound regulators are
org.) CFH, a plasma protein containing 20 homol- present.

1678 n engl j med 361;17  nejm.org  october 22, 2009

The New England Journal of Medicine

Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 medical progress

The vast majority of CFH mutations in patients

Table 1. Classification of Atypical Hemolytic–Uremic Syndrome.*
with atypical hemolytic–uremic syndrome are
heterozygous and cluster mainly in the C-terminal, Form of Disease Complement Abnormalities
causing amino acid changes or translation inter- Familial Mutations in CFH, 40–45%; in CFI,
ruption (Fig. 2 in the Supplementary Appendix). 5–10%; in C3, 8–10%; in MCP,
7–15%; in THBD, 9%; and in CFB,
With this abnormality there is a lower density of 1–2%.
mutant CFH on cell surfaces and diminished co- Sporadic
factor activity for C3b degradation because of low
Idiopathic Mutations in CFH, 15–20%; in CFI,
binding of mutant CFH to glycosaminoglycans on 3–6%; in C3, 4–6%; in MCP, 6–10%;
endothelial cells or to surface-bound C3b.40-42 Ly- in THBD, 2%; and in CFB, 2 cases;
sis of sheep erythrocytes is a standard test of this anti-CFH antibodies: 6–10%
system, since these cells have surface glycosamino­ Pregnancy-associated Mutations in CFH, 20%; in CFI, 15%
glycans and bind CFH. At variance with normal HELLP syndrome Mutations in CFH, 10%; in CFI, 20%;
and in MCP, 10%
serum, serum from patients with CFH mutations
lyses sheep erythrocytes through the alternative Drugs Rare CFH mutations (mostly unknown)
pathway. The addition of normal CFH stops this Organ transplantation Mutations in CFH, 15%; in CFI, 16%
lysis (Fig. 2B).43 Human immunodeficiency virus Unknown†
infection
Most mutant forms of CFH are secreted in
Cancer Unknown†
plasma. Since CFH forms oligomers,44,45 in plasma
from heterozygous patients mutant CFH seems to * HELLP denotes hemolytic anemia, elevated liver enzymes, and low platelet
interact with normal CFH and impairs its binding count.
to endothelial cells, suggesting a dominant nega- † There are no published data on the frequency of complement gene mutations
or anti-CFH autoantibodies in patients with this condition.
tive effect,42 in which the abnormal gene product
acts to block the effect of the wild-type allele.
Therefore, even though 50% of CFH molecules are cofactor activity.51 The lack of complement con-
normal, they are not sufficient for preventing trol on cells, despite control in the fluid phase,
atypical hemolytic–uremic syndrome. Homozy- mimics the course of the disease in patients with
gous CFH mutations, which account for only 15 CFH mutations. Anti-CFH autoantibodies develop
to 20% of CFH mutations in patients with this mainly in young children who, unlike their
syndrome, lead to quantitative CFH deficiency and heterozygous mothers, lack CFHR1 and CFHR3
very low C3 levels.35 because of homozygous deletions of the corre-
sponding genes.50,52,53 Interestingly, the antibod-
Genomic Abnormalities ies appear to recognize CFHR1 and CFHR3,50
A high degree of sequence identity between CFH which suggests that they may arise from an im-
and genes encoding five complement factor H– mune reaction against maternal CFHR1 and
related proteins (CFHR1 through CFHR5), which are CFHR3 (Dragon-Durey MA: personal communi-
located in tandem to CFH, may predispose to non- cation). CFHR1/3 deficiency itself may also pre-
allelic recombinations.46 In 3 to 5% of patients with dispose to atypical hemolytic–uremic syndrome,
atypical hemolytic–uremic syndrome, a heterozy- since patients with this deficiency who do not have
gous hybrid gene deriving from an uneven cross- anti-CFH antibodies have been described previ­
over between CFH and CFHR1 (which contains the ously.53,54
first 21 CFH exons and the last two CFHR1 exons47)
results in a gene product with decreased comple- Membrane Cofactor Protein
ment regulatory activity on endothelial surfaces. Mutations in the gene encoding membrane cofac-
tor protein (MCP), a widely expressed transmem-
Autoantibodies against CFH brane regulator, have been described in 10 to 15%
In approximately 6 to 10% of patients with atypi- of patients with atypical hemolytic–uremic syn-
cal hemolytic–uremic syndrome, anti-CFH autoan- drome.14-16 MCP serves as a cofactor for CFI to
tibodies develop.48-50 These antibodies bind to the cleave C3b and C4b on cell surfaces (Fig. 2A and
CFH C-terminal, reduce CFH binding to C3b, and 2B).55 Intact MCP is pivotal in preventing C3 acti-
enhance alternative-pathway–dependent lysis of vation on glomerular endothelium. In one study,
sheep erythrocytes without influencing fluid-phase an anti-MCP antibody completely blocked cofactor

n engl j med 361;17  nejm.org  october 22, 2009 1679

The New England Journal of Medicine
Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Normal Endothelial Cell

B Endothelial Cell with Dysfunctional Complement Regulation

Figure 2. Model for the Mechanisms Leading from Impaired Regulation of the Alternative Pathway to Thrombotic Microangiopathy.
In a normal endothelial cell (Panel A), complement factor H (CFH) binds to the endothelial surface and to C3b and together with mem-
brane cofactor protein (MCP) acts as a cofactor for cleavage of C3b, which is mediated by complement factor I (CFI), a process that pre-
vents its interaction with factor B. CFH also dissociates the C3 convertase of the alternative pathway (C3b). Thrombomodulin (TM) en-
hances CFI-mediated inactivation of C3b in the presence of CFH and promotes activation of the thrombin-activatable fibrinolysis
inhibitor (TAFIa), which degrades C3a and C5a. In patients with loss-of-function mutations in complement regulatory genes (CFH, CFI,
MCP, and THBD COLOR [theF Igene
G U R Eencoding thrombomodulin]) (Panel B), C3b is not degraded efficiently and forms the C3 and C5 convertases

of the alternative
Draft 4 pathway. Mutated TM cannot inactivate C5a and C3a because of reduced formation of TAFIa. C5b initiates the assem-
10/02/09
bly of the membrane-attack complex (MAC), leading to cell injury and activation, with expression of adhesion molecules, which together
Author Remuzzi
with
Fig #
C3a and
2
C5a recruit and activate leukocytes. Subsequent cell detachment results in a prothrombotic state. Aggregated platelets re-
lease
Title procoagulant platelet-derived microparticles (PMP), which facilitate the assembly of clotting enzymes. A similar situation applies
toMEpatients with gain-of-function mutations in CFB and C3. Mutant CFB forms a superconvertase that is resistant to dissociation by CFH.
Mutant
DE C3b does not bind CFH and MCP and is resistant to degradation by CFI. GAG denotes glycosaminoglycans.
Artist SBL
AUTHOR PLEASE NOTE:
Figure has been redrawn and type has been reset

activity in cell extracts.56 Most mutations are

Please check carefully
kocytes was reduced in 75% of mutants. Other
Issue date
heterozygous, though about 25% are either ho- mutants showed low C3b-binding capability and
mozygous or compound heterozygous. Such mu- decreased cofactor activity.14,57
tations usually cluster in extracellular domains
that are critical for regulation (Fig. 2 in the Sup- CFI
plementary Appendix). Expression on blood leu- CFI is a plasma serine protease that regulates the

1680 n engl j med 361;17  nejm.org  october 22, 2009

The New England Journal of Medicine

Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 medical progress

Table 2. Genetic Abnormalities and Clinical Outcome in Patients with Atypical Hemolytic–Uremic Syndrome.*

Response to Outcome
Short-Term Long-Term of Kidney
Gene Protein Affected Main Effect Frequency Plasma Therapy† Outcome‡ Transplantation
%
CFH Factor H No binding to 20–30 Rate of remission: Rate of death or Rate of recurrence:
­endothelium 60% (dose and ESRD: 70–80% 80–90%§
timing depen-
dent)
CFHR1/3 Factor HR1, R3 Anti–factor H anti- 6 Rate of remission: Rate of ESRD: 30– Rate of recurrence:
bodies 70–80% (plasma 40% 20%¶
exchange com-
bined with im-
munosuppres-
sion)
MCP Membrane cofactor No surface expression 10–15 No definitive indica- Rate of death or Rate of recurrence:
protein tion for therapy ESRD: <20% 15–20%¶
CFI Factor I Low level or low 4–10 Rate of remission: Rate of death or Rate of recurrence:
cofactor activity 30–40% ESRD: 60–70% 70–80%§
CFB Factor B C3 convertase stabi- 1–2 Rate of remission: Rate of death or Recurrence in one
lization 30% ESRD: 70% case
C3 Complement C3 Resistance to C3b 5–10 Rate of remission: Rate of death or Rate of recurrence:
inactivation 40–50% ESRD: 60% 40–50%
THBD Thrombomodulin Reduced C3b inacti- 5 Rate of remission: Rate of death or Recurrence in one
vation 60% ESRD: 60% case

* ESRD denotes end-stage renal disease.

† Remission was defined as either complete remission or partial remission (i.e., hematologic remission with renal sequelae).
‡ The long-term outcome was defined as the outcome 5 to 10 years after onset.
§ Patients in this category were eligible for combined liver and kidney transplantation.
¶ Patients in this category were eligible for single kidney transplantation.

three complement pathways by cleaving C3b and mutations reduce C3b binding to CFH and MCP,
C4b in the presence of cofactor proteins (Fig. 2A). which severely impairs degradation of mutant C3b
CFI mutations affect 4 to 10% of patients with (Fig. 2B).18
atypical hemolytic–uremic syndrome.14,58-60 All
mutations identified to date have been heterozy- Thrombomodulin
gous, and 80% cluster in the serine protease do- A recent study has shown that about 5% of pa-
main (Fig. 2 in the Supplementary Appendix). Ap- tients with atypical hemolytic–uremic syndrome
proximately 50% of mutations result in low CFI carry heterozygous mutations in THBD, the gene
levels.14,58-60 Others disrupt cofactor activity.59 encoding thrombomodulin, a membrane-bound
anticoagulant glycoprotein that facilitates comple-
CFB and C3 ment inactivation by CFI in the presence of CFH
Gain-of-function mutations can affect genes en- (Fig. 2A).61 Cells expressing these variants are less
coding the alternative pathway C3 convertase com- efficient in degrading C3b and in generating ac-
ponents, CFB and C3 (Fig. 2A).17,18 CFB mutations, tivated thrombin-activatable fibrinolysis inhibi-
which lead to chronic alternative-pathway activa- tor (TAFIa), a plasma carboxypeptidase B that
tion, occur in only 1 to 2% of patients with atyp- cleaves C3a and C5a (Fig. 2A and 2B).61
ical hemolytic–uremic syndrome (Fig. 2 in the
Supplementary Appendix).17 Mutants have excess Inc ompl e te Gene t ic Pene t r a nce
C3b affinity and form a hyperactive C3 convertase
that is resistant to dissociation, enhancing C3b Mutations in complement genes confer a predis-
formation.17 position rather than cause atypical hemolytic–
About 4 to 10% of patients have heterozygous uremic syndrome, and penetrance among carri-
mutations in C3, usually with low C3 levels.18 Most ers of CFH, MCP, and CFI mutations appears to be

n engl j med 361;17  nejm.org  october 22, 2009 1681

The New England Journal of Medicine
Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

40 to 50%.14 Healthy carriers of CFB and C3 muta- vation may lead to problems in such persons,64
tions also have been reported.17,18 Thus, the pres- since the vascular endothelium may require mul-
ence of these mutations cannot be used to predict tiple regulators for protection. Certain vascular
future cases of the syndrome. beds may be at increased risk. For example, the
About 5% of patients have combined muta- glomerular capillary bed has a fenestrated endothe-
tions, usually in CFH with either MCP or CFI, each lium, which continually exposes the subendothe-
inherited from a healthy parent.14 In some family lial matrix to a variety of circulating proteins and
pedigrees, atypical hemolytic–uremic syndrome peptides. Thus, anyone lacking complement reg-
has occurred in family members with a CFH mu- ulatory proteins is particularly vulnerable to com-
tation on one allele and two or three predisposing plement attack. Persons with gain-of-function mu-
CFH polymorphisms on the other allele, whereas tations in C3 or CFB are similarly at risk, since
the syndrome has not developed in family mem- these mutations lead to C3-convertase hyperac-
bers carrying just one affected allele.37 Similarly, tivity.
in a large pedigree with CFI and MCP mutations In the presence of events that enhance alterna-
and the risk-associated MCPggaac haplotype, the tive-pathway activation, carriers of complement
syndrome occurred only in family members who mutations undergo excess C3b formation and de-
had all the risk factors.62 In another study,63 car- position on vascular endothelium. An uncontrolled
riers of a common CFH mutation that is associated level of C3 convertase leads to more C3b mole-
with the hemolytic–uremic syndrome had disease cules and to more C5 convertase, initiating the
penetrance of 30%, and most patients also carried formation of the membrane-attack complex. Com-
at least one additional genetic risk factor. These plement-mediated endothelial injury creates a pro-
data indicate that the concurrence of both muta- thrombotic state through exposure of suben-
tions and risk polymorphisms may be required for dothelial collagen, von Willebrand factor, and
the development of the syndrome. fibrinogen (Fig. 2B). Moreover, it has been shown
Even in patients with multiple genetic risk fac- that mutant CFH has defective binding to platelets,
tors, the syndrome may not occur until middle allowing C3 and C9 deposition and platelet acti-
age, which suggests an environmental effect. In- vation.65
fections may precede clinical cases of the hemolyt-
ic–uremic syndrome in many such patients, in- Cl inic a l C our se a nd Ou t c ome
cluding in 35% of those with mutant CFH, in 50%
of those with mutant MCP, in 55% of those with Among patients with atypical hemolytic–uremic
mutant CFI, and in 22% of those with mutant syndrome who have genetic mutations, 67% are
C3.14 Pregnancy and the use of contraceptive pills affected during childhood,14,22 and the disease is
were reported to trigger disease in 8% of patients diagnosed in almost all patients with anti-CFH an-
with CFH mutations and in 20% of those with CFI tibodies before the age of 16 years.49 Acute episodes
mutations.14 manifest with severe hemolytic anemia, thrombo-
Given such data, unaffected carriers should be cytopenia, and acute renal failure. Extrarenal (i.e.,
monitored during pregnancy and episodes of in- central nervous system or multivisceral) involve-
fection, and precipitants, such as drugs that trig- ment occurs in 20% of patients.14,22
ger the hemolytic–uremic syndrome, should be Both short-term and long-term outcomes vary
avoided. according to the underlying complement abnor-
mality. About 60 to 70% of patients with CFH, CFI,
or C3 mutations and 30% of children with anti-
From C ompl emen t
A bnor m a l i t ie s t o Thrombo t ic CFH autoantibodies lose renal function or die dur-
Microa ngiopath y ing the presenting episode or have end-stage
renal disease after relapses14,22 (Table 2). CFB mu-
Several factors that contribute to complement reg- tations are associated with particularly poor renal
ulation are expressed on or bound to endothelium outcomes, with a loss of renal function in 88% of
(Fig. 2A). Normally, complement regulation re- patients in one study.17
mains intact even when the activities of one or two About 20% of patients with CFH mutations have
regulators are partially reduced. However, triggers, cardiovascular complications and excess mortal-
such as infections and pregnancy, that are asso- ity.22 Chronic complement dysregulation can con-
ciated with inflammation and complement acti- tribute to the formation of atheromatous lesions.66

1682 n engl j med 361;17  nejm.org  october 22, 2009

The New England Journal of Medicine

Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 medical progress

Long-term survival of patients with CFH mutations as hematologic normalization with renal sequelae)
(50% at 10 years) is poorer than that of patients occurred in 60% of plasma-treated patients14,22
with CFI or C3 mutations or anti-CFH autoantibod- (Table 2).
ies (80 to 90% at 10 years14,22). Plasma exchange appears to be effective in re-
Carriers of MCP mutations have a good progno- moving anti-CFH antibodies.22,48 However, the ef-
sis, with a complete remission rate of 80 to 90% fect is usually transient. The combination of plas-
(Table 2). Recurrences are frequent, but the long- ma exchange with the use of immunosuppressant
term outcome appears good, with 80% of patients drugs (e.g., corticosteroids and azathioprine or
remaining free of dialysis.14,22 Rarely, concurrent mycophenolate mofetil) and an anti-CD20 anti-
genetic abnormalities may lead to exceptionally body (rituximab) resulted in long-term dialysis-free
severe disease.14,67 survival in 60 to 70% of patients.49,51,84,85
Patients with CFI mutations have only a partial
T r e atmen t response to plasma therapy; remission occurred
in only 30 to 40% of patients.14,22,86 Since MCP is
Early reports of successful treatment with plasma a cell-associated protein, plasma exchange or in-
in atypical hemolytic–uremic syndrome date back fusion is unlikely to be effective in patients with
30 years.68,69 Since then, plasma exchange or in- MCP mutations. Indeed, 80 to 90% of such patients
fusion has been associated with a decrease in mor- have remission without plasma treatment.14,22,86
tality from 50% to 25%.70,71 Guidelines suggest Plasma exchange or infusion resulted in a response
that plasma therapy should be started within 24 in 30% of patients with CFB mutations and in 50%
hours after diagnosis, with an exchange of one to of those with C3 mutations.17,18 Possibly, patients
two plasma volumes per day or an infusion of 20 with either CFB or C3 mutations need more fre-
to 30 ml per kilogram of body weight.72,73 (For quent plasma exchanges to clear the hyperfunc-
details, see the case reports in the Supplementary tional mutant CFB and C3 (Table 2).
Appendix.) Plasma exchange allows for the provi-
sion of larger amounts of plasma than would be T r a nspl a n tat ion
possible with infusion while avoiding fluid over-
load. However, trials of plasma therapy in patients It has long been debated whether kidney transplan-
with the hemolytic–uremic syndrome are few and tation is appropriate for the treatment of end-stage
not current.74,75 A recent meta-analysis76,77 suggest- renal disease in patients with atypical hemolytic–
ing that plasma offers no significant benefit over uremic syndrome. The disease recurs in around
simple supportive therapy may be misleading, since 50% of patients who undergo transplantation, and
the trials that were evaluated did not distinguish graft failure occurs in 80 to 90% of those with
between Stx-related and atypical hemolytic–uremic recurrent disease.87,88 The type of mutation may
syndrome.74,75 predict the outcome after transplantation (Table
Since CFH is a plasma protein, plasma infu- 2). The recurrence rate after transplantation is 70
sion or exchange provides normal CFH to patients to 90% in patients with genetic abnormalities in
with homozygous mutations and complete CFH the circulating complement regulators CFH and
deficiency and induces disease remission (Ta- CFI, mainly of hepatic origin.58,87,88 Intensive long-
ble 2).14,22,78,79 However, since such patients are term plasma prophylaxis was thought to prevent
plasma-dependent, unresponsiveness may develop recurrence in one patient with a CFH mutation89
after long-term therapy.79-81 Carriers of heterozy- but failed in another patient.90 Atypical hemolytic–
gous CFH mutations usually have normal levels uremic syndrome recurred in 40 to 50% of patients
of CFH, but half is dysfunctional. The beneficial with C3 mutations.18 A patient with a CFB mutation
effect of plasma strongly depends on the amount, who underwent renal transplantation had recur-
frequency, and method of administration, with rence of disease.17
plasma exchange having been shown to be supe- The outcome of single kidney transplantation
rior to plasma infusion for remission and preven- is more favorable in patients with MCP mutations
tion of recurrences, possibly through the removal than in those with other mutations. More than
of mutant dysfunctional CFH molecules82,83 (case 80% had no recurrence, with a rate of long-
reports in the Supplementary Appendix). Overall term graft survival that was similar to that of
in patients with CFH mutations, either complete patients who underwent transplantation for oth-
remission or partial remission (which was defined er causes.22,87,88 The MCP gene product, membrane

n engl j med 361;17  nejm.org  october 22, 2009 1683

The New England Journal of Medicine
Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

cofactor protein, is a transmembrane protein that would be extremely risky and for whom combined
is highly expressed in the kidney, so kidney trans- kidney–liver transplantation would be worth con-
plantation, not surprisingly, corrects the defect. sidering. They also suggested which patients would
Living-donor transplantation is contraindicated be most likely to benefit from isolated kidney
in patients with atypical hemolytic–uremic syn- transplantation (Table 2).
drome because of the high risk of recurrent dis-
ease.91,92 In addition, such procedures may be risky E volv ing A pproache s
to donors, who may be mutation carriers. For ex-
ample, de novo atypical hemolytic–uremic syn- The identification of complement genetic abnor-
drome developed in a man with a heterozygous malities may facilitate treatments that down-reg-
CFH mutation after he donated a kidney to his af- ulate complement activation. A human plasma-
fected child.92 If living-donor transplantation is derived CFH concentrate is being developed. (For
considered, the donor and recipient should under- details, see the European Medicines Agency Web
go genotyping of complement genes to identify site at www.emea.europa.eu/pdfs/human/comp/
hitherto unsuspected mutation carriers. opinion/52123506en.pdf.) A humanized anti-C5
Simultaneous kidney and liver transplantation monoclonal antibody, eculizumab, which had un-
was performed in two children with atypical dergone phase 3 trials involving patients with par-
hemolytic–uremic syndrome and CFH mutations oxysmal nocturnal hemoglobinuria,98 was reported
with the rationale of correcting the genetic defect as promising in patients with atypical hemolytic–
and preventing recurrence.93,94 However, both cas- uremic syndrome in two single-case reports.99,100
es were complicated by early failure of the trans- An 18-month-old boy with a plasma-resistant con-
planted liver. The first child recovered after a sec- genital form of the disease achieved remission af-
ond liver transplantation and had no symptoms of ter the initiation of treatment with eculizumab,99
the hemolytic–uremic syndrome for 3 years but although a possible effect of previous plasma
ultimately died from hepatic encephalopathy.93,95 therapy could not be ruled out because markers
This case offered the proof of concept that liver of hemolysis began to decrease before the ad-
transplantation cures atypical hemolytic–uremic ministration of eculizumab. A 30-year-old woman
syndrome associated with CFH mutations. The with a CFH mutation who had a recurrence of the
second case was also complicated by failure of the hemolytic–uremic syndrome in a kidney graft100
transplanted liver, with concomitant widespread had resolution of hemolysis and improved trans-
microvascular thrombosis and complement depo- plant function after receiving eculizumab. Re-
sition.94 The investigators who reported the case sults concerning 14 additional eculizumab-treated
speculated that surgical stress with ischemia– patients, 10 of whom achieved stable remission,
reperfusion induced complement activation in the have been reported recently (Nurnberger J: per-
liver that could not be regulated because of func- sonal communication). The efficacy of eculizu­
tional CFH deficiency. A modified approach that mab in the treatment of atypical hemolytic–ure-
included extensive plasma exchange before surgery mic syndrome is being evaluated in controlled trials
to provide a sufficient amount of normal CFH un­ that are either ongoing or planned (ClinicalTrials.
til the liver graft recovered synthetic function was gov numbers, NCT00844545, NCT00844428,
applied in eight subsequent cases95-97 and succeed­ NCT00838513, and NCT00844844). (For details
ed in seven; severe hepatic thrombosis and fatal regarding international registries of patients and
encephalopathy developed in the eighth patient.95 addresses of the main laboratories that perform
Thus, the substantial risks of dual kidney and genetic screening, see the Resources section in
liver transplantation require a careful assessment the Supplementary Appendix.)
of possible benefits for any candidate patient.
Screening for mutations may help patients and Sum m a r y
clinicians to make more informed decisions re-
garding listing for transplantation on the basis of During the past decade, multiple and complex
the risk of recurrence. A position paper reporting genetic abnormalities that are associated with atyp-
the conclusions of a 2007 conference on atypical ical hemolytic–uremic syndrome have been de-
hemolytic–uremic syndrome95 listed groups of scribed. Although of disparate origin, disease-
patients for whom isolated kidney transplantation associated abnormalities have in common the

1684 n engl j med 361;17  nejm.org  october 22, 2009

The New England Journal of Medicine

Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 medical progress

overactivation of the alternative complement path- Dr. Noris reports receiving consulting fees from Adienne. No
other potential conflict of interest relevant to this article was
way, which has opened perspectives for comple- reported.
ment inhibitor therapies in the future. We thank Drs. Mauro Abbate and Irene van der Meer for
Supported in part by grants from Fondazione ART La Ricerca their help in the critical reading and preparation of this manu-
Sui Trapianti, Fondazione Aiuto Ricerca Malattie Rare, Istituto script and Dr. Antonella Piccinelli for her contribution to the
Superiore di Sanità, and the Telethon Foundation. original figures.

References
1. Noris M, Remuzzi G. Hemolytic ure- Familial haemolytic uraemic syndrome and 27. Barré P, Kaplan BS, de Chadarévian JP,
mic syndrome. J Am Soc Nephrol 2005; an MCP mutation. Lancet 2003;362:1542-7. Drummond KN. Hemolytic uremic syn-
16:1035-50. 16. Richards A, Kemp EJ, Liszewski MK, drome with hypocomplementemia, serum
2. Besbas N, Karpman D, Landau D, et et al. Mutations in human complement C3NeF, and glomerular deposits of C3.
al. A classification of hemolytic uremic regulator, membrane cofactor protein Arch Pathol Lab Med 1977;101:357-61.
syndrome and thrombotic thrombocy- (CD46), predispose to development of fa- 28. Landau D, Shalev H, Levy-Finer G, Po-
topenic purpura and related disorders. milial hemolytic uremic syndrome. Proc lonsky A, Segev Y, Katchko L. Familial
Kidney Int 2006;70:423-31. Natl Acad Sci U S A 2003;100:12966-71. hemolytic uremic syndrome associated with
3. Scheiring J, Andreoli SP, Zimmerhackl 17. Goicoechea de Jorge E, Harris CL, complement factor H deficiency. J Pediatr
LB. Treatment and outcome of Shiga-toxin- Esparza-Gordillo J, et al. Gain-of-function 2001;138:412-7.
associated hemolytic uremic syndrome mutations in complement factor B are as- 29. Walport MJ. Complement: first of two
(HUS). Pediatr Nephrol 2008;23:1749-60. sociated with atypical hemolytic uremic parts. N Engl J Med 2001;344:1058-66.
4. Constantinescu AR, Bitzan M, Weiss syndrome. Proc Natl Acad Sci U S A 30. Devaux P, Christiansen D, Fontaine
LS, et al. Non-enteropathic hemolytic ure- 2007;104:240-5. [Erratum, Proc Natl Acad M, Gerlier D. Control of C3b and C5b de-
mic syndrome: causes and short-term Sci USA 2007;104:10749.] position by CD46 (membrane cofactor pro-
course. Am J Kidney Dis 2004;43:976-82. 18. Frémeaux-Bacchi V, Miller EC, Liszew­ tein) after alternative but not classical com-
5. Kaplan BS, Meyers KE, Schulman SL. ski MK, et al. Mutations in complement plement activation. Eur J Immunol 1999;
The pathogenesis and treatment of hemo- C3 predispose to development of atypical 29:815-22.
lytic uremic syndrome. J Am Soc Nephrol hemolytic uremic syndrome. Blood 2008; 31. Thompson RA, Winterborn MH. Hy-
1998;9:1126-33. 112:4948-52. pocomplementaemia due to a genetic de-
6. Ruggenenti P, Noris M, Remuzzi G. 19. Le Quintrec M, Lionet A, Kamar N, et ficiency of beta 1H globulin. Clin Exp Im-
Thrombotic microangiopathy, hemolytic al. Complement mutation-associated de munol 1981;46:110-9.
uremic syndrome, and thrombotic throm- novo thrombotic microangiopathy follow- 32. Rougier N, Kazatchkine MD, Rougier
bocytopenic purpura. Kidney Int 2001;60: ing kidney transplantation. Am J Trans- JP, et al. Human complement factor H de-
831-46. plant 2008;8:1694-701. ficiency associated with hemolytic uremic
7. Benz K, Amann K. Pathological aspects 20. Fang CJ, Fremeaux-Bacchi V, Liszew­ syndrome. J Am Soc Nephrol 1998;9:2318-
of membranoproliferative glomerulone- ski MK, et al. Membrane cofactor protein 26.
phritis (MPGN) and haemolytic uraemic mutations in atypical hemolytic uremic 33. Warwicker P, Goodship TH, Donne
syndrome (HUS)/thrombocytic throm- syndrome (aHUS), fatal Stx-HUS, C3 glo­ RL, et al. Genetic studies into inherited
bopenic purpura (TTP). Thromb Haemost merulonephritis, and the HELLP syn- and sporadic hemolytic uremic syndrome.
2009;101:265-70. drome. Blood 2008;111:624-32. Kidney Int 1998;53:836-44.
8. Campbell S, Carré IJ. Fatal haemolytic 21. Fakhouri F, Jablonski M, Lepercq J, et 34. Richards A, Buddles MR, Donne RL,
uraemic syndrome and idiopathic hyperli- al. Factor H, membrane cofactor protein, et al. Factor H mutations in hemolytic
paemia in monozygotic twins. Arch Dis and factor I mutations in patients with uremic syndrome cluster in exons 18-20,
Child 1965;40:654-8. hemolysis, elevated liver enzymes, and low a domain important for host cell recogni-
9. Kaplan BS, Chesney RW, Drummond platelet count syndrome. Blood 2008;112: tion. Am J Hum Genet 2001;68:485-90.
KN. Hemolytic uremic syndrome in fami- 4542-5. 35. Dragon-Durey MA, Frémeaux-Bacchi
lies. N Engl J Med 1975;292:1090-3. 22. Loirat C, Noris M, Fremeaux-Bacchi V. V, Loirat C, et al. Heterozygous and ho-
10. Zakarija A, Bennett C. Drug-induced Complement and the atypical hemolytic mozygous factor H deficiencies associated
thrombotic microangiopathy. Semin uremic syndrome in children. Pediatr Neph- with hemolytic uremic syndrome or mem-
Thromb Hemost 2005;31:681-90. rol 2008;23:1957-72. branoproliferative glomerulonephritis: re-
11. Ruggenenti P. Post-transplant hemo- 23. Carreras L, Romero R, Requesens C, port and genetic analysis of 16 cases. J Am
lytic-uremic syndrome. Kidney Int 2002; et al. Familial hypocomplementemic hemo- Soc Nephrol 2004;15:787-95.
62:1093-104. lytic uremic syndrome with HLA-A3,B7 36. Caprioli J, Bettinaglio P, Zipfel PF, et
12. Karthikeyan V, Parasuraman R, Shah haplotype. JAMA 1981;245:602-4. al. The molecular basis of familial hemo-
V, Vera E, Venkat KK. Outcome of plasma 24. Stühlinger W, Kourilsky O, Kanfer A, lytic uremic syndrome: mutation analysis
exchange therapy in thrombotic microan- Sraer JD. Haemolytic-uraemic syndrome: of factor H gene reveals a hot spot in short
giopathy after renal transplantation. Am J evidence for intravascular C3 activation. consensus repeat 20. J Am Soc Nephrol
Transplant 2003;3:1289-94. Lancet 1974;2:788-9. 2001;12:297-307.
13. Zarifian A, Meleg-Smith S, O’Donovan 25. Noris M, Ruggenenti P, Perna A, et al. 37. Caprioli J, Castelletti F, Bucchioni S,
R, Tesi RJ, Batuman V. Cyclosporine-asso- Hypocomplementemia discloses genetic et al. Complement factor H mutations and
ciated thrombotic microangiopathy in re- predisposition to hemolytic uremic syn- gene polymorphisms in haemolytic urae-
nal allografts. Kidney Int 1999;55:2457-66. drome and thrombotic thrombocytopenic mic syndrome: the C-257T, the A2089G and
14. Caprioli J, Noris M, Brioschi S, et al. purpura: role of factor H abnormalities. the G2881T polymorphisms are strongly
Genetics of HUS: the impact of MCP, CFH, J Am Soc Nephrol 1999;10:281-93. associated with the disease. Hum Mol
and IF mutations on clinical presentation, 26. Kim Y, Miller K, Michael AF. Break- Genet 2003;12:3385-95.
response to treatment, and outcome. Blood down products of C3 and factor B in 38. Pérez-Caballero D, González-Rubio C,
2006;108:1267-79. hemolytic-uremic syndrome. J Lab Clin Gallardo ME, et al. Clustering of missense
15. Noris M, Brioschi S, Caprioli J, et al. Med 1977;89:845-50. mutations in the C-terminal region of fac-

n engl j med 361;17  nejm.org  october 22, 2009 1685

The New England Journal of Medicine
Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

tor H in atypical hemolytic uremic syn- hemolytic uremic syndrome. Blood 2007; tributes to complement deposition on plate-
drome. Am J Hum Genet 2001;68:478-84. 110:1516-8. lets and their activation. Blood 2008;111:
39. Zipfel PF, Skerka C. Complement fac- 52. Józsi M, Licht C, Strobel S, et al. Fac- 5307-15.
tor H and related proteins: an expanding tor H autoantibodies in atypical hemolytic 66. Seifert PS, Hansson GK. Complement
family of complement-regulatory pro- uremic syndrome correlate with CFHR1/ receptors and regulatory proteins in hu-
teins? Immunol Today 1994;15:121-6. CFHR3 deficiency. Blood 2008;111:1512-4. man atherosclerotic lesions. Arterioscle-
40. Jokiranta TS, Jaakola VP, Lehtinen MJ, 53. Zipfel PF, Edey M, Heinen S, et al. De- rosis 1989;9:802-11.
Pärepalo M, Meri S, Goldman A. Structure letion of complement factor H-related genes 67. Fremeaux-Bacchi V, Sanlaville D, Me-
of complement factor H carboxyl-terminus CFHR1 and CFHR3 is associated with nouer S, et al. Unusual clinical severity of
reveals molecular basis of atypical haemo- atypical hemolytic uremic syndrome. PLoS complement membrane cofactor protein-
lytic uremic syndrome. EMBO J 2006;25: Genet 2007;3(3):e41. associated hemolytic-uremic syndrome and
1784-94. 54. Koziolek MJ, Zipfel PF, Skerka C, et al. uniparental isodisomy. Am J Kidney Dis
41. Manuelian T, Hellwage J, Meri S, et al. Chronic course of a hemolytic uremic syn- 2007;49:323-9.
Mutations in factor H reduce binding af- drome caused by a deficiency of factor 68. Remuzzi G, Misiani R, Marchesi D, et
finity to C3b and heparin and surface at- H-related proteins (CFHR1 and CFHR3). al. Haemolytic-uraemic syndrome: defi-
tachment to endothelial cells in hemolytic Kidney Int 2008;74:384-8. ciency of plasma factor(s) regulating pros-
uremic syndrome. J Clin Invest 2003;111: 55. Liszewski MK, Leung M, Cui W, et al. tacyclin activity? Lancet 1978;2:871-2.
1181-90. Dissecting sites important for complement 69. Remuzzi G, Misiani R, Marchesi D, et
42. Heinen S, Józsi M, Hartmann A, et al. regulatory activity in membrane cofactor al. Treatment of the hemolytic uremic
Hemolytic uremic syndrome: a factor H protein (MCP; CD46). J Biol Chem 2000; syndrome with plasma. Clin Nephrol 1979;
mutation (E1172Stop) causes defective 275:37692-701. 12:279-84.
complement control at the surface of en- 56. Nakanishi I, Moutabarrik A, Hara T, 70. Bell WR, Braine HG, Ness PM, Kickler
dothelial cells. J Am Soc Nephrol 2007; et al. Identification and characterization TS. Improved survival in thrombotic throm-
18:506-14. of membrane cofactor protein (CD46) in bocytopenic purpura–hemolytic uremic
43. Sánchez-Corral P, González-Rubio C, the human kidneys. Eur J Immunol 1994; syndrome: clinical experience in 108 pa-
Rodríguez de Córdoba S, López-Trascasa 24:1529-35. tients. N Engl J Med 1991;325:398-403.
M. Functional analysis in serum from 57. Fremeaux-Bacchi V, Moulton EA, Ka- 71. George JN. How I treat patients with
atypical hemolytic uremic syndrome pa- vanagh D, et al. Genetic and functional thrombotic thrombocytopenic purpura-
tients reveals impaired protection of host analyses of membrane cofactor protein hemolytic uremic syndrome. Blood 2000;
cells associated with mutations in factor (CD46) mutations in atypical hemolytic ure- 96:1223-9.
H. Mol Immunol 2004;41:81-4. mic syndrome. J Am Soc Nephrol 2006; 72. Ariceta G, Besbas N, Johnson S, et al.
44. Nan R, Gor J, Perkins SJ. Implications 17:2017-25. Guideline for the investigation and initial
of the progressive self-association of wild- 58. Kavanagh D, Kemp EJ, Mayland E, et therapy of diarrhea-negative hemolytic ure-
type human factor H for complement reg- al. Mutations in complement factor I pre- mic syndrome. Pediatr Nephrol 2009;24:
ulation and disease. J Mol Biol 2008;375: dispose to development of atypical hemo- 687-96.
891-900. lytic uremic syndrome. J Am Soc Nephrol 73. Ruggenenti P, Noris M, Remuzzi G.
45. Pangburn MK, Rawal N, Cortes C, 2005;16:2150-5. Thrombotic microangiopathies. In: Wilcox
Alam MN, Ferreira VP, Atkinson MA. Poly- 59. Kavanagh D, Richards A, Noris M, et CS, ed. Therapy in nephrology & hyper-
anion-induced self-association of com- al. Characterization of mutations in com- tension: a companion to Brenner & Rector’s
plement factor H. J Immunol 2009;182: plement factor I (CFI) associated with the kidney. 3rd ed. Philadelphia: Saunders,
1061-8. hemolytic uremic syndrome. Mol Immu- 2008:294-312.
46. Heinen S, Sanchez-Corral P, Jackson nol 2008;45:95-105. 74. Loirat C, Sonsino E, Hinglais N, Jais
MS, et al. De novo gene conversion in the 60. Fremeaux-Bacchi V, Dragon-Durey MA, JP, Landais P, Fermanian J. Treatment of
RCA gene cluster (1q32) causes mutations Blouin J, et al. Complement factor I: a sus- the childhood haemolytic uraemic syn-
in complement factor H associated with ceptibility gene for atypical haemolytic drome with plasma: a multicentre ran-
atypical hemolytic uremic syndrome. Hum uraemic syndrome. J Med Genet 2004; domized controlled trial. Pediatr Nephrol
Mutat 2006;27:292-3. 41(6):e84. 1988;2:279-85.
47. Venables JP, Strain L, Routledge D, et 61. Delvaeye M, Noris M, DeVriese A, et 75. Rizzoni G, Claris-Appiani A, Edefonti
al. Atypical haemolytic uraemic syndrome al. Thrombomodulin mutations in atypi- A, et al. Plasma infusion for hemolytic-
associated with a hybrid complement gene. cal hemolytic–uremic syndrome. N Engl J uremic syndrome in children: results of a
PLoS Med 2006;3(10):e431. Med 2009;361:345-57. multicenter controlled trial. J Pediatr 1988;
48. Dragon-Durey MA, Loirat C, Cloarec 62. Esparza-Gordillo J, Jorge EG, Garrido 112:284-90.
S, et al. Anti-factor H autoantibodies as- CA, et al. Insights into hemolytic uremic 76. Michael M, Elliott EJ, Craig JC, Ridley
sociated with atypical hemolytic uremic syndrome: segregation of three indepen- G, Hodson EM. Interventions for hemo-
syndrome. J Am Soc Nephrol 2005;16:555- dent predisposition factors in a large, lytic uremic syndrome and thrombotic
63. multiple affected pedigree. Mol Immunol thrombocytopenic purpura: a systematic
49. Skerka C, Józsi M, Zipfel PF, Dragon- 2006;43:1769-75. review of randomized controlled trials.
Durey MA, Fremeaux-Bacchi V. Autoanti- 63. Martinez-Barricarte R, Pianetti G, Am J Kidney Dis 2009;53:259-72.
bodies in haemolytic uraemic syndrome Gautard R, et al. The complement factor H 77. Noris M, Remuzzi G. Thrombotic mi-
(HUS). Thromb Haemost 2009;101:227-32. R1210C mutation is associated with atyp- croangiopathy: what not to learn from a
50. Dragon-Durey MA, Blanc C, Marliot F, ical hemolytic uremic syndrome. J Am Soc meta-analysis. Nat Rev Nephrol 2009;5:
et al. The high frequency of complement Nephrol 2008;19:639-46. 186-8.
factor H related CFHR1 gene deletion is 64. Girardi G, Bulla R, Salmon JE, Tedesco 78. Cho HY, Lee BS, Moon KC, Ha IS,
restricted to specific subgroups of pa- F. The complement system in the pathophys- Cheong HI, Choi Y. Complete factor H de-
tients with atypical haemolytic uraemic iology of pregnancy. Mol Immunol 2006; ficiency-associated atypical hemolytic ure-
syndrome. J Med Genet 2009;46:447-50. 43:68-77. mic syndrome in a neonate. Pediatr Neph-
51. Józsi M, Strobel S, Dahse HM, et al. 65. Ståhl AL, Vaziri-Sani F, Heinen S, et rol 2007;22:874-80.
Anti factor H autoantibodies block C-ter- al. Factor H dysfunction in patients with 79. Licht C, Weyersberg A, Heinen S, et al.
minal recognition function of factor H in atypical hemolytic uremic syndrome con- Successful plasma therapy for atypical

1686 n engl j med 361;17  nejm.org  october 22, 2009

The New England Journal of Medicine

Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.
 medical progress

hemolytic uremic syndrome caused by fac- uremic syndrome. J Am Soc Nephrol 2007; 94. Remuzzi G, Ruggenenti P, Colledan
tor H deficiency owing to a novel mutation 18:2392-400. M, et al. Hemolytic uremic syndrome:
in the complement cofactor protein domain 87. Bresin E, Daina E, Noris M, et al. Out- a fatal outcome after kidney and liver trans-
15. Am J Kidney Dis 2005;45:415-21. come of renal transplantation in patients plantation performed to correct factor H
80. Nathanson S, Frémeaux-Bacchi V, De- with non-Shiga toxin-associated haemo- gene mutation. Am J Transplant 2005;5:
schênes G. Successful plasma therapy in lytic uremic syndrome: prognostic signifi- 1146-50.
hemolytic uremic syndrome with factor H cance of genetic background. Clin J Am 95. Saland JM, Ruggenenti P, Remuzzi G.
deficiency. Pediatr Nephrol 2001;16:554-6. Soc Nephrol 2006;1:88-99. Liver-kidney transplantation to cure atyp-
81. Nathanson S, Ulinski T, Frémeaux- 88. Loirat C, Fremeaux-Bacchi V. Hemo- ical hemolytic uremic syndrome. J Am Soc
Bacchi V, Deschênes G. Secondary failure lytic uremic syndrome recurrence after Nephrol 2009;20:940-9.
of plasma therapy in factor H deficiency. renal transplantation. Pediatr Transplant 96. Saland JM, Emre SH, Shneider BL, et
Pediatr Nephrol 2006;21:1769-71. 2008;12:619-29. al. Favorable long-term outcome after liver-
82. Davin JC, Strain L, Goodship TH. 89. Olie KH, Goodship TH, Verlaak R, et kidney transplant for recurrent hemolytic
Plasma therapy in atypical haemolytic al. Posttransplantation cytomegalovirus- uremic syndrome associated with a factor H
uremic syndrome: lessons from a family induced recurrence of atypical hemolytic mutation. Am J Transplant 2006;6:1948-
with a factor H mutation. Pediatr Nephrol uremic syndrome associated with a factor 52.
2008;23:1517-21. H mutation: successful treatment with 97. Jalanko H, Peltonen S, Koskinen A, et
83. Lapeyraque AL, Wagner E, Phan V, et intensive plasma exchanges and ganciclo- al. Successful liver-kidney transplantation
al. Efficacy of plasma therapy in atypical vir. Am J Kidney Dis 2005;45(1):e12-e15. in two children with aHUS caused by a
hemolytic uremic syndrome with comple- 90. Olie KH, Florquin S, Groothoff JW, et mutation in complement factor H. Am J
ment factor H mutations. Pediatr Nephrol al. Atypical relapse of hemolytic uremic Transplant 2008;8:216-21.
2008;23:1363-6. syndrome after transplantation. Pediatr 98. Brodsky RA, Young NS, Antonioli E,
84. Dragon-Durey MA, Frémeaux-Bacchi Nephrol 2004;19:1173-6. et al. Multicenter phase 3 study of the
V. Atypical haemolytic uraemic syndrome 91. Chan MR, Thomas CP, Torrealba JR, complement inhibitor eculizumab for the
and mutations in complement regulator et al. Recurrent atypical hemolytic uremic treatment of patients with paroxysmal noc-
genes. Springer Semin Immunopathol syndrome associated with factor I mutation turnal hemoglobinuria. Blood 2008;111:
2005;27:359-74. in a living related renal transplant recipi- 1840-7.
85. Kwon T, Dragon-Durey MA, Macher ent. Am J Kidney Dis 2009;53:321-6. 99. Gruppo RA, Rother RP. Eculizumab
MA, et al. Successful pre-transplant man- 92. Donne RL, Abbs I, Barany P, et al. Re- for congenital atypical hemolytic–uremic
agement of a patient with anti-factor H au- currence of hemolytic uremic syndrome af- syndrome. N Engl J Med 2009;360:544-6.
toantibodies-associated haemolytic urae- ter live related renal transplantation associ- 100. Nürnberger J, Philipp T, Witzke O,
mic syndrome. Nephrol Dial Transplant ated with subsequent de novo disease in the Opazo Saez A, et al. Eculizumab for atyp-
2008;23:2088-90. donor. Am J Kidney Dis 2002;40(6):E22. ical hemolytic–uremic syndrome. N Engl
86. Sellier-Leclerc AL, Fremeaux-Bacchi 93. Remuzzi G, Ruggenenti P, Codazzi D, J Med 2009;360:542-4. [Erratum, N Engl J
V, Dragon-Durey MA, et al. Differential et al. Combined kidney and liver trans- Med 2009;360:2487.]
impact of complement mutations on clin- plantation for familial haemolytic urae- Copyright © 2009 Massachusetts Medical Society.
ical characteristics in atypical hemolytic mic syndrome. Lancet 2002;359:1671-2.

powerpoint slides of journal figures and tables

At the Journal’s Web site, subscribers can automatically create PowerPoint slides.
In a figure or table in the full-text version of any article at NEJM.org, click
on Get PowerPoint Slide. A PowerPoint slide containing the image, with its title
and reference citation, can then be downloaded and saved.

n engl j med 361;17  nejm.org  october 22, 2009 1687

The New England Journal of Medicine
Downloaded from nejm.org at KAOHSIUNG MEDICAL UNIVERSITY on July 29, 2018. For personal use only. No other uses without permission.
Copyright © 2009 Massachusetts Medical Society. All rights reserved.