Atypical Haemolytic Uraemic Syndrome (Complement-Mediated Thrombotic Microangiopathy)

Predominantly renal thrombotic microangiopathy caused chiefly by dysregulation of the alternative complement pathway at the endothelial surface, with genetic predisposition involving CFH, CFI, CD46, C3 and CFB, rearrangements of the CFH–CFHR cluster and an autoimmune form due to anti-factor H antibodies; includes DGKE deficiency nephropathy, with a non-complement mechanism, and discusses thrombomodulin (THBD), whose association has been classified as disputed.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

1. Introduction

1.1. From haemolytic uraemic syndrome to complement disease

Haemolytic uraemic syndrome (HUS) is defined by the combination of haemolytic anaemia, thrombocytopenia and renal insufficiency caused by platelet thrombus formation in the microcirculation of the kidney and other organs, the lesion termed thrombotic microangiopathy (TMA) in histopathology [1]. The most frequent form, particularly in children, is triggered by Shiga toxin-producing Escherichia coli (STEC) strains and accompanied by prodromal diarrhoea; in contrast to this epidemic form, the adjective “atypical” has been used since the mid-1970s to designate familial and sporadic cases, eventually encompassing everything other than STEC-HUS [3, 13]. For decades, atypical HUS (aHUS) was a disease with a poor prognosis, frequent relapses and progression to end-stage renal disease (ESRD) in approximately 60% of patients with genetic aHUS before complement inhibitors [1, 26].

The milestone moving aHUS into complement genetics was the linkage study published by Warwicker and colleagues in 1998, which, in HUS families, located a 26 cM region between markers D1S212 and D1S306 at 1q32 (Zmax score 3.94) containing the factor H gene (CFH), and identified two CFH mutations: a frameshift deletion in a recurrent sporadic case, with serum factor H half normal, and an arginine-to-glycine substitution in exon 20 in one family [27]. This was followed by demonstration of missense variant clustering in the factor H C-terminal domain, the description of variants in CD46, encoding membrane cofactor protein (MCP), by Richards and colleagues in 2003, and variants in CFI and C3 [1, 28]. In 2005, Dragon-Durey and colleagues described IgG anti-factor H autoantibodies in three of 48 French children with recurrent aHUS, with acquired functional factor H deficiency despite normal gene and antigen, establishing the autoimmune form [29]. In 2006 and 2007, the CFH/CFHR1 hybrid gene, generated by non-allelic homologous recombination, and the association of CFHR1 and CFHR3 deletion with aHUS were described; in 2008, Józsi and colleagues linked anti-factor H autoantibodies to absence of FHR-1 and FHR-3 [30–32]. In 2009, Delvaeye and colleagues proposed variants in THBD, encoding thrombomodulin, as a cause of aHUS [33], and in 2013 Lemaire and colleagues identified recessive DGKE variants by exome sequencing in nine families, the first non-complement mechanism of aHUS beginning in the first year of life [34]. That same year, Legendre and colleagues published the two prospective phase 2 trials establishing eculizumab, an anti-C5 monoclonal antibody, as aHUS treatment [35].

1.2. Nomenclature and reclassifications

The designation “aHUS” became heterogeneous: it includes patients whose disease is driven by alternative complement pathway dysregulation and patients with TMA of other causes, and broad authorisation of C5 inhibitors for “aHUS” may lead to inappropriate prescribing when complement is not the mechanism [2]. The International Society of Nephrology (ISN) International Forum uses complement-mediated HUS (CaHUS) for the form with demonstrated alternative pathway dysregulation, and a National Kidney Foundation (NKF) working group proposed a roadmap for precise aetiological nomenclature using the Delphi method [2, 3]. The 2015 KDIGO (Kidney Disease: Improving Global Outcomes) controversies conference reserved “primary aHUS” for cases strongly suspected of an alternative pathway abnormality after exclusion of secondary causes [13]. The 2025 Brazilian consensus of the Rare Diseases Committee of the Brazilian Society of Nephrology (COMDORA-SBN) retained aHUS for consistency with pivotal trials and Food and Drug Administration (FDA) labelling and adopted an aetiological classification in five groups: congenital or acquired TTP; genetic or autoimmune complement-mediated aHUS; TMA due to variants in non-complement genes (DGKE, THBD) or metabolic defects (cobalamin); infection-associated TMA (STEC, Streptococcus pneumoniae); and TMA secondary to systemic disease or drugs [14]. In Orphanet, former gene-specific subtypes (ORPHA:93575 C3, 93576 MCP, 93578 CFB, 93579 CFH, 93580 CFI and 217023 THBD) are obsolete and replaced by the single entity “aHUS with a complement gene abnormality” (ORPHA:544472), listing CD46, CFH, CFI, CFB and C3 as genes with causative germline mutations and THBD only as a tested candidate gene [5]. This article uses aHUS as the main term, consistent with labelling and guidelines, and complement-mediated TMA when mechanistic distinction is necessary.

1.3. Definition of entities within scope

Genetically based complement-mediated aHUS is predominantly renal TMA in which one or more heterozygous loss-of-function variants in regulators (CFH, CFI, CD46) or gain-of-function variants in convertase components (C3, CFB), or CFH–CFHR cluster rearrangements, confer predisposition; an episode usually requires a complement-activating trigger [1]. aHUS with anti-factor H antibodies is the predominantly paediatric autoimmune form, in which autoantibodies mainly bind the factor H C-terminal domain and reproduce the effect of CFH variants; approximately 90% of cases are associated with homozygous CFHR3–CFHR1 deletion [2, 6]. DGKE deficiency nephropathy is an autosomal recessive disease, usually beginning before 1 year, with recurrent TMA, sometimes nephrotic proteinuria and membranoproliferative glomerulonephritis (MPGN), with a non-complement mechanism and no response to C5 inhibitors [8, 34]. The association of THBD with aHUS, accepted by OMIM (AHUS6) and retained in 2021 GeneReviews, was classified as disputed by ClinGen in 2024 because large cohorts and enrichment analyses failed to reproduce it [2, 9, 36]. Congenital TTP (Upshaw–Schulman syndrome, due to biallelic ADAMTS13 variants), methylmalonic aciduria with homocystinuria, cblC type (MMACHC) and C3 glomerulopathy (C3G), an allelic disease of fluid-phase alternative pathway dysregulation, are included in this article as differential diagnoses and related diseases [1, 23, 37].

2. Molecular Genetics and Pathophysiology

2.1. The alternative pathway and its regulation at the endothelial surface

The alternative complement pathway is continuously active at a low level: C3 spontaneously hydrolyses, binds factor B (FB), which factor D cleaves, and forms C3 convertase C3bBb, generating more C3b and amplifying activation; binding of additional C3b creates C5 convertase, whose action releases C5a, a pro-inflammatory anaphylatoxin, and initiates the C5b-9 membrane attack complex [23, 38]. Host surfaces escape this amplification through regulators: factor H (FH) accelerates C3bBb decay and is a cofactor for factor I (FI) in C3b degradation; FI, a circulating serine protease, cleaves and inactivates C3b and C4b in the presence of cofactors (FH and C4b-binding protein in plasma; CD46 and complement receptor 1 on cells); MCP (CD46), a widely expressed type I transmembrane glycoprotein, is an FI cofactor at the cell surface [38–40]. FH has 1,231 amino acids organised into 20 short consensus repeat modules (SCRs, or sushi domains) of approximately 60 amino acids; N-terminal SCRs 1–4 ensure fluid-phase regulation, and C-terminal SCRs 19–20 bind glycosaminoglycans and cell-surface-deposited C3b [23]. aHUS-associated CFH variants cluster in SCRs 18–20 and compromise surface protection, whereas those associated with C3G cluster in SCRs 1–4 and compromise fluid-phase regulation; this distinction summarises the pathophysiological difference between the diseases [1, 23].

In aHUS, complement hyperactivation at the glomerular and arteriolar endothelial surface produces endothelial damage, subendothelial matrix exposure, platelet activation and consumption, and microvascular thrombosis; erythrocyte fragmentation in occluded microvessels generates schistocytes and haemolysis [1]. Variants are mostly predisposing rather than sufficient: an acute episode occurs when an infection, pregnancy, drug or other condition activates complement in a carrier [1]. Penetrance is incomplete, and occurrence of more than one rare variant and common risk haplotypes increases disease probability [13, 41].

2.2. Genetic architecture and proportion of cases attributed to each gene

GeneReviews estimates that genetic aHUS accounts for approximately 60% of all aHUS and that no genetic cause is identified in approximately 40% of patients [1, 42]. The table reproduces proportions and method performance, with the warning that the method columns indicate the fraction of known variants in each gene detected by the method, rather than clinical test sensitivity for aHUS diagnosis [1].

Proportions derive chiefly from Nester and colleagues' consensus review (2015) and Bu and colleagues' genetic analysis of 400 United States patients (2018), which also proposed vitronectin (VTN) as a new gene [1, 42, 43]. In one child, complete paternal uniparental isodisomy of chromosome 1 rendered a CD46 exon 10 splice defect homozygous [1]. Series differ by population and method: in the Global aHUS Registry, with 851 patients, approximately 40% had no identified variant; in the Brazilian consensus, 33.5% of tested patients in the national registry had no variants, CFHR gene variants were more frequent than in other cohorts and no CFI variants were found in children [14].

2.3. Gene by gene: mechanism and particularities

The GeneReviews mechanism table attributes abnormal gain of function to C3 and CFB, loss of function to CD46, CFH and CFHR5 and an unknown mechanism to CFHR1 and VTN; formatting of the row grouping CFI, DGKE and THBD does not permit unequivocal attribution in the extracted version, but loss of FI and DGKE function is described in functional sources [1, 34, 39]. The ISN Forum summarises the current model: loss of CFH, CFI and CD46 function, gain of CFB and C3 function, and CFH/CFHR1/CFHR3 rearrangements and gene conversions arising from RCA cluster instability [2].

2.4. The CFH–CFHR cluster: rearrangements, hybrids and deletions

High sequence identity between CFH and downstream CFHR1–CFHR4 genes favours non-allelic homologous recombination (NAHR), generating copy number variants (CNVs) and hybrid genes; assays must be designed to detect these alterations because conventional sequencing does not identify them [1]. The CFH/CFHR1 hybrid, described by Venables and colleagues in 2006, replaces CFH C-terminal exons with CFHR1 exons and produces FH unable to protect surfaces; it accounts for approximately 3%–5% of all aHUS and is detected by multiplex ligation-dependent probe amplification (MLPA) [1, 30]. The reverse CFHR1/CFH hybrid, a CFH/CFHR3 hybrid through microhomology-mediated deletion in an Irish family, and CFHR1–CFHR4 deletions have also been described [1, 49]. In the Istituto Mario Negri cohort, rare CFH–CFHR structural variants were present in 8% of 258 patients with primary aHUS and 2 of 92 with secondary aHUS; in the CFH-involving group (14 patients: 6 CFH::CFHR1 hybrids and 7 duplications, including the reverse hybrid and internal CFH duplications), 12 of 13 acute episodes treated without eculizumab progressed to end-stage chronic kidney disease, anticomplement therapy induced remission in 4 of 4, and recurrence occurred in 6 of 7 grafts without prophylaxis and 0 of 3 with prophylaxis; in the group involving only CFHR genes (6 patients, 5 with CFHR3(1-5)::CFHR4(10) hybrid), onset was earlier and 4 of 6 remitted without eculizumab [50]. Characterisation of these rearrangements may require MLPA, molecular combing or long-read sequencing [46, 51].

The common approximately 84 kb deletion removing CFHR3 and CFHR1 results from NAHR between homologous repeats and has opposing effects depending on context: it is a risk factor for autoimmune aHUS and protective against age-related macular degeneration (AMD), where the deletion haplotype was present in 20% of control chromosomes and 8% of those from patients with neovascular AMD [31, 52, 53]. Józsi and colleagues identified anti-FH autoantibodies with absent or very low FHR-1 and FHR-3 in 16 of 147 aHUS patients (11%), all juvenile, a presentation termed DEAP-HUS (deficiency of CFHR plasma proteins and autoantibody positive) [32, 46]. Homozygous CFHR1 deletion is frequent in healthy individuals and is not diagnostic by itself [46].

2.5. The autoimmune form: anti-factor H antibodies

Anti-FH antibodies cause approximately 25% of paediatric complement-mediated HUS, particularly between 5 and 15 years, with higher prevalence in India, the Czech Republic, China and Egypt; they mainly bind the FH C-terminal domain, mimicking CFH variants, and approximately 90% of patients have homozygous CFHR1/CFHR3 deletion, in a “two-hit” model with frequent infectious prodrome [2]. In the Indian HUS Registry, 138 children had anti-FH-associated HUS, 56% of the cohort; titres were high (mean 7,054 AU/mL), inversely correlated with C3, and 60 of 68 tested children had homozygous CFHR1 deletion [54]. Besides IgG antibodies, described in approximately 10% of aHUS, IgM anti-FH antibodies have been described in 7 of 186 patients (3.8%) with TMA presenting as aHUS, without association with CFHR3–CFHR1 deletion, directed against SCR19 and inhibiting FH binding to C3b [55]. Anti-FH autoantibodies have also been described in 11% of patients with C3G [23].

2.6. Gene–disease validity

Source: ClinGen curation file downloaded on 10 October 2026 [9]. CFHR1, CFHR3, CFHR5 and C5 have no aHUS validity curation in that file [9]. THBD's disputed classification has practical consequences: a rare THBD variant in an aHUS patient should not be taken as the genetic cause or change management, although OMIM retains AHUS6 and 2021 GeneReviews includes THBD among predisposing genes [1, 2, 36]. Historical data attributed to THBD (64% penetrance, childhood onset in approximately 90%, graft recurrence) should be interpreted with this reservation [1, 26].

2.7. Modifiers, risk haplotypes and combined variants

GeneReviews identifies modifiers that increase penetrance and severity but may be unable to cause aHUS alone: the CFH-H3 (CFH-tgtgt) haplotype, defined in historical nomenclature by c.−332T, c.184G, c.1204T, c.2016G and c.2808T; the CD46 (MCP) GGAAC haplotype, defined by −547G, −261G, IVS9−78A, IVS12+638A and c.2232C; and homozygous CFHR3/CFHR1 deletion [1]. These alleles retain the historical designation published in GeneReviews. In the European consortium of 795 patients, 41% had a variant in a single gene and 3% combined variants in more than one gene; 8%–10% of CFH, C3 or CFB variant carriers and approximately 25% of CD46 or CFI variant carriers had combined variants; penetrance was 73% in carriers of combined variants with two risk haplotypes and 36% with zero or one; and patients with a combined CD46 variant reached ESRD at 3 years in 50%, versus 19% with an isolated CD46 variant [41]. The KDIGO conference states that 3%–5% of patients have more than one pathogenic variant and that penetrance of an acute episode depends on age, reaching 64% at 70 years [13].

2.8. Variant spectrum, nomenclature and C5 pharmacogenetics

CFH, CFI, CD46, C3 and CFB variants are mainly private missense variants, with some recurrent variants; GeneReviews highlights CFH c.3628C>T p.(Arg1210Cys) as a common pathogenic variant [1, 56]. The following normalisation uses MANE Select transcripts and GRCh38 assembly [57, 58].

The C5 variant NM_001735.3:c.2654G>A p.(Arg885His) does not cause disease but alters the eculizumab epitope: among 345 Japanese patients with paroxysmal nocturnal haemoglobinuria (PNH) treated with eculizumab, all 11 poor responders were heterozygous for this variant, whose prevalence was similar in patients (3.2%) and in healthy Japanese individuals (3.5%) and which also occurs in Han Chinese individuals; an Argentinian patient of Asian ancestry had NM_001735.3:c.2653C>T p.(Arg885Cys) [59]. The variant C5 is functional but is neither bound nor blocked by eculizumab; the ISN Forum recommends testing for it and states that crovalimab, which binds a different C5 epitope, should retain efficacy [2, 59]. For CFH and CFI variants described only at protein level in abstracts, such as CFH p.(Gly759Arg) and CFI p.(Gly110Arg) in a 2026 case, the published protein designation is retained, without a confirmed correspondence at DNA level [60].

2.9. Allelic and related diseases

Biallelic or heterozygous variants in C3, CD46, CFB, CFH, CFHR1, CFHR5, CFI and DGKE have been implicated in C3G [1]. FH deficiency (OMIM 609814) is associated with aHUS and basal laminar drusen (OMIM 126700); FI deficiency (OMIM 610984) and CFI variants are also associated with AMD; complete C3 and FB deficiencies are recessive and predispose to infections [23, 44, 45]. CFHR5 nephropathy (OMIM 614809), described in Greek Cypriots, is autosomal dominant C3G caused by a founder allele with internal exon 2–3 duplication: Gale and colleagues identified 26 patients from 11 apparently unrelated families and 4 of 84 Cypriots with unexplained kidney disease, and Athanasiou and colleagues described 91 patients in 16 families, with microscopic haematuria in 90% and chronic kidney disease (CKD) after 50 years in 80% of males and 21% of females [47]. Truncating THBD variants near the transmembrane region cause an autosomal dominant bleeding disorder with very high soluble thrombomodulin (ORPHA:436169), and GeneReviews mentions thrombomodulin defect-associated thrombophilia (OMIM 614486) [1, 36]. For DGKE, older literature separates aHUS type 7 and nephrotic syndrome type 7 under the same OMIM number 615008, representing a single DGKE nephropathy [8].

3. Clinical Spectrum and Manifestations by System

3.1. The acute episode

An acute episode manifests as microangiopathic haemolytic anaemia (MAHA) with schistocytes, elevated lactate dehydrogenase (LDH) and low haptoglobin, consumptive thrombocytopenia and acute kidney injury (AKI), frequently with arterial hypertension, proteinuria and haematuria [1, 14]. Besides renal findings (AKI, oedema, oligoanuria, proteinuria, haematuria and hypertension), the Brazilian consensus describes central nervous system (confusion, lethargy, seizures, coma), gastrointestinal (diarrhoea, hepatic abnormalities, pancreatitis), alveolar haemorrhage, amaurosis, cutaneous ischaemia with extremity necrosis and cardiac involvement [14]. In the Global aHUS Registry, with 851 patients assessed before eculizumab, extrarenal manifestations occurred in 19%–38% (depending on organ) in the six months after presentation, and the KDIGO conference estimates them at up to 20% [13, 61]. Hypertension and nephrotic proteinuria are frequent and do not discriminate aHUS from other TMAs [2].

Presentation is not always complete: TMA may be present on biopsy without thrombocytopenia, and subacute forms occur with hypertension and renal TMA without systemic haemolysis or thrombocytopenia, so the Brazilian consensus recommends considering TMA in any patient with kidney injury and mild haemolysis (grade 1B) [13, 14]. Approximately 30% of complement-mediated aHUS has concomitant diarrhoea or gastroenteritis, and diarrhoea does not exclude aHUS [13]. Before complement inhibitors, approximately 50% of patients remained on chronic dialysis and up to 25% died during the acute phase [14].

3.2. Triggers

GeneReviews lists non-enteric bacterial and viral infections, drugs, malignancies, transplantation, pregnancy and other medical conditions as triggers [1]. Associated drugs include chemotherapeutics (mitomycin, cisplatin, daunorubicin, bleomycin, cytarabine, gemcitabine), immunosuppressants and immunomodulators (ciclosporin, tacrolimus, muromonab-CD3, interferon), quinidine, antiplatelets (ticlopidine, clopidogrel), oral contraceptives and anti-inflammatory agents [1]. Malignancy-associated aHUS has been described in almost 6% of patients with metastatic carcinoma, half with gastric cancer; post-transplant HUS also occurs, de novo or recurrent, as does HUS in autoimmune diseases such as scleroderma, antiphospholipid syndrome (APS) and systemic lupus erythematosus (SLE) [1]. GeneReviews includes Streptococcus pneumoniae HUS among atypical presentations, with respiratory distress, neurological involvement and coma and mortality of 12.3% [1, 62]; current classifications treat it as infection-associated HUS (ORPHA:544493), distinct from complement-mediated aHUS [14]. The ISN Forum identifies infection and pregnancy as the most frequent triggers, a bimodal age distribution (early childhood and women of childbearing age) and fewer summer presentations [2].

The 2021 French National Diagnosis and Care Protocol (PNDS) of the Haute Autorité de Santé (HAS) ranks triggers by age: in children, infection, particularly diarrhoea or upper respiratory tract infection, comes first; in adults, childbirth comes first, followed by the introduction of oestrogens [11]. The same protocol lists as presenting complaints asthenia, pallor, drowsiness, oedema, hypertension, haemoglobinuria with “port-wine” urine when haemolysis is massive, diarrhoea in approximately 15% of cases, the ecchymoses, petechiae and menometrorrhagia of thrombocytopenia (which can be very profound with CD46 variants), neurological or neuropsychiatric signs, the dyspnoea and nausea of oligoanuric fluid overload, and cardiac or cutaneous manifestations; the triad is not always complete and schistocytes may appear only later [11]. The proportion of 15% with diarrhoea is lower than the approximately 30% of complement-mediated aHUS with concomitant diarrhoea or gastroenteritis estimated by the KDIGO conference; both sources agree that diarrhoea does not exclude the diagnosis [11, 13].

3.3. Age, sex and reference cohorts

In the French national series of 214 patients (2000–2008), onset occurred in adulthood in 58.4% and childhood in 41.6%, and the proportion already affected by 2, 18, 40 and 60 years was 23%, 40%, 70% and 98%; 1-year mortality was 6.7% in children and 0.8% in adults, but ESRD after the first episode affected 46% of adults and 16% of children; complement variants were present in 61% [12]. In the Global Registry, the male:female ratio was 1.3:1 in paediatric onset and 1:2 in adult onset, CFI variants were more common with adult onset and CD46 variants with paediatric onset, and childhood onset was associated with lower ESRD risk (adjusted hazard ratio 0.55; 95% confidence interval [CI] 0.41–0.73) [61]. In the Istituto Mario Negri Italian cohort of 273 consecutive patients (82 familial and 191 sporadic), variants, CFH risk polymorphisms or anti-FH antibodies were found in more than 70% of sporadic and familial cases; familial cases had more CFH SCR20 variants and more severe disease, and CFH and THBD variant carriers had earlier onset and higher mortality [26]. These data describe disease treated with plasma and supportive care before C5 inhibitors, rather than pure natural history.

The 2021 French PNDS estimates the annual incidence of aHUS at approximately 0.3 per million inhabitants, states that aHUS accounts for less than 10% of HUS in children and approximately 3% of adult TMA outside pregnancy and transplantation, and reports a female:male ratio of 3:1 in adults, a family history of HUS in 20%–30% of cases and usually dominant transmission with penetrance close to 50% [11]. The 3:1 ratio is more unbalanced than the 1:2 (male:female) ratio of the Global Registry for adult onset, a difference reflecting distinct populations and inclusion criteria [11, 61]. Regarding genetic architecture, the same protocol places heterozygous CFH variants in 20%–40% of patients, both adults and children, with homozygous CFH forms more frequent in paediatric onset; in adults, approximately 10% have CD46, CFI or C3 variants; in children, CD46 variants and anti-FH antibodies follow; THBD variants are exceptional and their role debated; DGKE variants (2%–3% of aHUS) are exclusively of paediatric onset; 30%–50% of patients have no molecular complement abnormality; and plasma C3 is low in 30%–40% [11].

3.4. Phenotype and course by gene

Sources: GeneReviews and Italian and French series for age and course columns [1, 12, 26, 41]; prospective United Kingdom cohort cited by the ISN Forum, with genotype-matched historical controls, for the final column [2, 10]. Although GeneReviews states that no strict genotype–phenotype correlations have been identified, differences in severity and prognosis by gene are consistent across cohorts: CFH, C3 and hybrids are associated with poorer renal survival, CD46 with the best, and the Global Registry found the lowest ESRD-free survival with CFH and highest with CD46 [1, 61]. Within-family variability, observed for example with CFB, reflects the role of triggers and modifiers [1].

3.5. Autoimmune form

aHUS with anti-FH antibodies mainly affects children and adolescents and frequently has an infectious prodrome, sometimes with diarrhoea [2]. In the Indian Registry, 119 of 138 children required dialysis; an adverse outcome (stage 4–5 CKD or death) occurred in 36 at 3 months and in 41 at last follow-up, and relapse occurred in 14 of 122; titre above 8,000 AU/mL, low C3 and delayed plasma exchange were independent factors of poor outcome [54]. The KDIGO conference estimates relapses at approximately 20%–25% [13]. The French PNDS describes the autoimmune form mainly between 5 and 13 years of age, with very high antibody titres and frequent systemic involvement, particularly digestive with diarrhoea, which can wrongly point towards typical HUS [11].

3.6. DGKE deficiency nephropathy

HUS due to DGKE deficiency usually presents in the first year of life, with repeated episodes of MAHA, thrombocytopenia and AKI; glomerular filtration recovers, but hypertension, microscopic haematuria and proteinuria persist, and CKD is common in the second decade, worsening long after the last acute episode [1, 48]. Diagnostic clues are early onset and nephrotic syndrome 3–5 years after onset, both rare in other forms of HUS; some patients have MPGN with or without HUS [48]. In the cohort of 16 patients from the United Kingdom national aHUS service, proteinuria persisted in most, only 2 had reached ESRD 20 years after presentation, some had isolated steroid-resistant nephrotic syndrome without TMA, biopsy showed TMA and/or an MPGN-like pattern, and relapses occurred almost always before 5 years [8, 63]. In families with biallelic variants, penetrance was complete [1, 64]. The French PNDS adds that relapse frequency decreases with age and appears to cease at the end of childhood or in adolescence, that chronic proteinuria appears in the medium and long term, that progression to ESRD does not depend on the number of relapses and that there is no recurrence after kidney transplantation; it reports, however, a median renal survival of 15 years after the first episode [11]. This value diverges from the 10-year renal survival of 89%–100% and from the 2 of 16 patients with ESRD 20 years after presentation in the British cohort; the estimates come from small series, with different recruitment and length of follow-up, and should not be combined [8, 11, 63].

3.7. Pregnancy-associated aHUS

Women with a history of aHUS are at increased risk during pregnancy and at greater risk postpartum [1]. In Fakhouri's French series (2010), pregnancy-associated aHUS occurred in 21 of 100 adult women with aHUS, 79% postpartum, with complement abnormalities in 18 of 21; with plasma therapy, 62% had ESRD after 1 month and 76% at last follow-up; risk was highest in the second pregnancy; and pregnancies in women with complement abnormalities (n=44) had fetal loss in 4.8% and pre-eclampsia in 7.7% [65]. In Bruel's international retrospective cohort (France, United Kingdom and Italy; 87 patients), pregnancy aHUS occurred mainly in the first pregnancy (58%) and postpartum (76%); 71% required dialysis at diagnosis; with mean follow-up of 7.2 years, 53% reached ESRD, 19% CKD and 28% relapsed; complement variants were present in 56% (CFH 30%, CFI 9%), with severity and variant frequency similar to non-pregnancy-related aHUS [66]. Postpartum aHUS generally develops within three months of delivery [1]. Pregnancy-associated TMA occurs in approximately 1 in 25,000 pregnancies; aHUS is the rarest TMA form in pregnancy, occurs mainly in late third trimester or postpartum, with frequent AKI, mildly reduced platelets and rare neurological involvement, unlike TTP [14, 67]. It may develop as a pre-eclampsia complication and progress to HELLP syndrome (haemolysis, elevated liver enzymes and low platelets) [1].

The French PNDS places the typical episode approximately two weeks (sometimes up to four months) after delivery following an uncomplicated term pregnancy, and notes that aHUS may appear only in the second or third pregnancy after previous normal gestations [11]. In the first 7 days after delivery, the diagnosis requires exclusion of causes with a different course and treatment: TMA of placental origin (severe pre-eclampsia, which affects approximately 4% of primiparous women, and HELLP syndrome, with AKI in approximately 10%, due to acute tubular necrosis with a good prognosis, and cytolysis with alanine aminotransferase higher than aspartate aminotransferase), supported by the sFlt-1/PlGF ratio; and massive postpartum haemorrhage, which can produce anaemia, thrombocytopenia, raised LDH, low haptoglobin and AKI, with haemostasis difficult to interpret after transfusions, and in which high-dose tranexamic acid can cause renal TMA or cortical necrosis [11]. A TMA appearing 15 or more days after normal delivery of a normal term pregnancy is probably aHUS and, according to the PNDS, warrants investigation of the alternative pathway and C5 blockade [11]. TTP must always be excluded by ADAMTS13 activity: the immune form usually appears early or in mid-pregnancy and the congenital form late in pregnancy or postpartum, sometimes as a first episode in the first pregnancy, with a relapse risk close to 100% in subsequent pregnancies [11].

3.8. TMA in the transplanted kidney

Kidney transplant outcome in aHUS is determined chiefly by the genetic alteration [1]. In the French series of 57 patients and 71 transplants before eculizumab, 68% had a complement variant; death-censored graft survival was 51% at 5 years, recurrence was associated with graft loss, and recurrence risk was highest with CFH or gain-of-function C3 and CFB variants [68]. Recurrence risk was estimated at 15%–20% with CD46 variants and 50%–100% with circulating regulator variants, and de novo post-transplant HUS had genetic alterations in approximately 30% [69]. By gene, GeneReviews summarises: graft failure from recurrence in 70% of CFI patients and 50% of C3 patients and the sole CFB patient; recurrence of 30%–100% with CFH; no recurrence in 4 patients with an isolated CD46 variant, but recurrence in 2 of 3 in a French cohort; no recurrence with DGKE [1]. In the graft, TMA often presents without MAHA or thrombocytopenia and is diagnosed by biopsy [2].

3.9. Paediatric age and neonatal period

In children, the main TMA cause is STEC-HUS, followed by aHUS and pneumococcal HUS; children have lower haemoglobin and platelets, higher LDH and higher mortality than adults [14]. Below 2 years, congenital TTP, cblC and cblG and coagulation disorders must be excluded [14]. In newborns, perinatal asphyxia (low Apgar, cord blood acidosis, multiorgan failure, disseminated intravascular coagulation [DIC]) is an important differential diagnosis, and neonatal aHUS should be considered if TMA persists, recurs or there is low C3 (grade 1B) [14, 70]. In a Portuguese series from a tertiary paediatric nephrology centre over 20 years, 5 children had genetic aHUS (youngest 4 months), 4 with CFH variants; 4 received plasma exchange and 1 acute peritoneal dialysis; all remitted, 2 had more than one relapse and 1 progressed to CKD, with median follow-up of 136 months [71].

4. Clinical Recognition and Warning Signs

4.1. When to suspect the disease

Suspicion begins with the triad of MAHA, thrombocytopenia and kidney injury, or AKI with hypertension and mild haemolysis, after excluding an immediate explanation [14]. GeneReviews proposes suspecting genetic aHUS in a patient with aHUS and at least one of three elements: a relative with aHUS diagnosed at least 6 months apart, excluding exposure to a common infectious agent; HUS recurrence after complete recovery from the initial episode; or an identifiable underlying environmental factor such as drugs, systemic disease, viruses or non-Shiga-toxin-producing bacteria [1]. The Brazilian consensus recommends suspecting aHUS with TMA, ADAMTS13 activity above 10% and negative STEC testing (grade 1B) [14]. It recommends considering aHUS in stage 5 CKD of unknown cause, after pregnancy, in lupus nephritis, with TMA histology and in malignant hypertension, and the KDIGO conference requests full complement evaluation also in pregnancy-associated aHUS and de novo post-transplant TMA [13, 14].

4.2. Contexts by age

4.3. Warning signs and emergencies

TMA with AKI is an emergency: the benefit of the C5 inhibitor depends on the time to initiation, and in the retrospective paediatric eculizumab study complete response was 70% when TMA had lasted less than 2 months and nil when it had lasted more than 2 months [16]. Severity signs are oligoanuria, encephalopathy, seizures or coma, severe hypertension, cardiac or pulmonary involvement and alveolar haemorrhage [14]. With coma or seizures and strong suspicion of TTP, the KDIGO conference allows plasma exchange to be started until ADAMTS13 activity is shown to exceed 10%; outside that scenario, plasma exchange, when eculizumab is available, may delay the inhibitor and worsen outcome [13]. Haematological response to the C5 inhibitor usually appears in the first week and most renal recovery occurs in the first 30 days; absence of response within that period is a warning to reassess the diagnosis [2]. The French aHUS emergency card, distributed by the reference centres, identifies the patient's abnormality (C3, thrombomodulin, CD46, factor B, factor H, factor I, anti-FH antibodies, DGKE or none) and, when relapse is suspected, indicates a blood count with schistocytes, LDH, unconjugated bilirubin and creatinine, contact with the follow-up team and correction of hypertension and metabolic disturbances; in a patient on eculizumab with headache, fever, neck stiffness or vomiting (food refusal in infants), it instructs calling the emergency medical service (15 or 112 in France), giving cefotaxime 2 g by slow intravenous injection before lumbar puncture and then 200 mg/kg/day in 4 intravenous doses, and requesting urgent specialist advice [11]. The card does not distinguish between adults and children nor give a maximum dose: the initial 2 g dose is an adult dose, and 200 mg/kg/day in a 70 kg adult would amount to 14 g/day. The French SmPC for cefotaxime sets, in bacterial meningitis, 150–200 mg/kg/day in children up to 12 years and 9–12 g/day in adults (3 g 3 to 4 times daily), in equal doses every 6 to 8 hours, values that serve as a reference to adjust the card's regimen to weight and age [72].

4.4. Newborn screening

The sources consulted do not describe newborn screening for complement-mediated aHUS. cblC, a treatable differential diagnosis, is detected through newborn screening by propionylcarnitine, and a positive screen requires immediate parenteral hydroxocobalamin [37]. The 2017 European guideline for remethylation disorders recommends newborn screening and early cblC treatment [37, 73].

4.5. First-line management according to the French PNDS

The French PNDS includes a summary sheet for general practitioners and emergency physicians, which organises the initial approach before transfer to a centre with intensive care, dialysis and the possibility of plasma exchange [11].

Source: appendix 3 of the French PNDS [11].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. TMA and aHUS criteria

aHUS diagnosis is clinical and by exclusion: it is established with TMA after excluding TTP, STEC-HUS, cblC and secondary causes; no complement test alone confirms diagnosis, and genetic diagnosis requires identification of pathogenic variant(s) in an associated gene [1, 2, 38]. The Brazilian consensus systematises criteria in the following table [14].

The French PNDS defines TMA as the association of regenerative haemolytic anaemia (haemoglobin below 13 g/dL in men and 12 g/dL in women; reticulocytes frequently above 120×10⁹/L, raised unconjugated bilirubin and LDH, low or undetectable haptoglobin), Coombs negative (save for rare exceptions in lupus and pneumococcal infections) and of mechanical mechanism with schistocytes, with peripheral thrombocytopenia without signs of DIC (normal prothrombin time, activated partial thromboplastin time and fibrinogen) [11]. Profound anaemia and thrombocytopenia with normal leucocytes or neutrophilia are particularly suggestive, and the presence of organ failure strengthens the diagnosis and helps to exclude autoimmune cytopenias, in particular Evans syndrome [11].

5.2. Investigation algorithm

The KDIGO conference proposed the flowchart that continues to guide practice [13]. In adults, ADAMTS13 activity should be requested urgently and known before starting eculizumab to exclude TTP; in children, the result is not a prerequisite and eculizumab should not be delayed, with monitoring for non-response. STEC testing should be performed in all because approximately 5% of STEC-HUS lacks prodromal diarrhoea and approximately 30% of complement aHUS has concomitant diarrhoea. Plasma and urinary assessment for cblC deficiency is mandatory in all paediatric aHUS. Full complement evaluation, including genetics, is indicated in primary aHUS, pregnancy-associated aHUS and de novo post-transplant aHUS; in other secondary forms, evidence is insufficient to recommend routine full genetic analysis, and the decision depends on consequences, for example transplant listing [13]. Samples for complement protein assays should be collected before plasma therapy, and delayed genetic results should not postpone clinical diagnosis or treatment [13].

Sources: Brazilian consensus (Box 2) and ISN Forum (Table 1) [2, 14]. In secondary TMA, the ISN Forum adds investigation for glucose-6-phosphate dehydrogenase deficiency, Wilson disease, haemophagocytic syndrome, renovascular hypertension and glomerular disease [2].

The French algorithm has particular features with practical consequences [11]. When aHUS is suspected, and before any plasma therapy, CH50, C3, C4 and factor B (stigmata of alternative pathway activation, typically low C3 with normal C4), factor H, factor I, CD46 membrane expression and anti-FH antibodies are collected; the study of the CFH, CFI, CFB, C3, CD46, THBD and DGKE genes can be sampled later. A quantitative FH or FI deficiency is classified as homozygous (below 1%) or heterozygous (10%–60%) and is confirmed by molecular characterisation; the presence of anti-FH antibodies points towards testing for hybrid genes and deletions of the CFH–CFHR cluster by MLPA. Testing for cblC (blood and urine homocysteine and methionine, urinary amino acid chromatography or plasma methylmalonic acid) is requested in children and in adults up to 40 years of age, a wider limit than the age cut-off below 18 years of the Brazilian consensus [11, 14]. STEC testing is systematic in children and in adults with overt digestive signs. In malignant hypertension with TMA, genetic study of the alternative pathway is indicated when at least one of the following is present: age below 40 years, family history of aHUS, European ancestry, absence of chronic hypertension, low C3, glomerular TMA on biopsy or persistence of TMA despite blood pressure control. The PNDS also recommends brain MRI for any neurological sign, even minor, and allows it systematically during follow-up [11].

5.3. PLASMIC score

The PLASMIC score estimates the probability of severe ADAMTS13 deficiency while the result is unavailable; one point is assigned to each criterion: platelets below 30×10⁹/L; haemolysis (reticulocytes above 2.5%, undetectable haptoglobin or indirect bilirubin above 2 mg/dL); no active cancer; no solid organ or stem cell transplantation; mean corpuscular volume below 90 fL; INR below 1.5; creatinine below 2 mg/dL [14]. A score of 0–4 indicates low risk, 5 intermediate and 6–7 high; with a threshold of 6 or more, sensitivity was 0.85 (CI 0.67–0.94) and specificity 0.89 (95% CI 0.81–0.94) [14]. Only 5% of TTP cases have the classical pentad [14]. The French TTP PNDS also uses the French score, which applies only to TMA without an associated context (cancer, chemotherapy, pregnancy, sepsis, transplantation, disseminated intravascular coagulation) and adds two criteria: platelets below 30×10⁹/L and creatinine below 2.26 mg/dL (200 µmol/L). The probability of ADAMTS13 activity below 10% is 2% with 0 points, 70% with 1 and 94% with 2; for PLASMIC, the same protocol gives 0–4% with 0–4 points, 5–24% with 5 and 62–82% with 6–7 [20]. With a French score of 2, the PNDS starts full treatment of immune TTP immediately; with 1, it starts at least plasma exchange while awaiting the result [20].

5.4. Complement biomarkers and their limitations

During the acute phase, low C3 and FB, high C3c and Bb, low C5 and properdin and high C5a and sC5b-9 may be observed; however, C3 and sC5b-9 are normal in approximately 50% of patients and do not change consistently with treatment [2]. The KDIGO conference estimates low C3 in 30%–50% of aHUS (and up to 75% of C3G), noting that C3 also falls in STEC-HUS and pneumococcal HUS; the Brazilian consensus estimates low C3 in fewer than 20% and emphasises that normal C3 does not exclude aHUS (grade 1B) [13, 14]. C4 is typically normal, distinguishing alternative from classical pathway consumption [39]. Biomarkers do not individually distinguish complement-mediated HUS from secondary TMA [2]. The functional C5b-9 deposition assay on microvascular endothelial cells (HMEC-1) is elevated in all patients on adenosine diphosphate-activated endothelium and, on unstimulated endothelium, only during the acute phase, and its increase precedes relapse after withdrawal; it is available only in specialist laboratories and has had replication failures [2]. Leucocyte CD46 expression by flow cytometry is the functional counterpart of a CD46 variant, strongly suggestive but not confirmatory, and normal expression does not exclude a function-affecting variant [13, 40]. Antigenic and functional FI and FH assays complete assessment [39]. GeneReviews summarises initial assessment in the following table [1].

5.5. Anti-factor H antibodies

The autoimmune form is confirmed by anti-FH antibodies, generally with absent serum FHR-1 and FHR-3 due to homozygous CFHR3–CFHR1 deletion; it is distinct from a germline CFH variant [38, 46]. Assays have limited availability and are not standardised [2]. The KDIGO conference recommends confirming a positive result in a second sample collected at least 4 weeks later, testing before transplantation and, in paediatrics, measuring at diagnosis and, if positive, on days 7, 14 and 28, then monthly and at 1 year [13]. Anti-FH titres of 1,330 AU/mL or higher and free FH of 440 mg/L or lower have been associated with relapse, with limited evidence [2].

5.6. Genetic testing: strategy, limitations and report

The KDIGO conference and the Brazilian consensus recommend genetic testing in all patients with suspected primary aHUS, interpreted by an experienced laboratory, and also before transplantation and when withdrawal of anticomplement therapy is being considered [13, 14]. GeneReviews proposes an age-based strategy [1]. Below 1 year of age, particularly with consanguinity or apparent recessive inheritance, DGKE is tested first, followed by a panel if two variants are not found. Above 1 year, a multigene panel with C3, CD46, CFB, CFH, CFHR5, CFI, DGKE, THBD and VTN is used, plus targeted testing for CFH/CFHR1 and CFHR1/CFH hybrids and CFHR1/CFHR4 deletions. The panel should include the modifiers (CFHR3/CFHR1 deletion and CFH-H3 and MCP-GGAAC haplotypes) and may include genes of the differential diagnosis; if negative, exome, genome or mitochondrial DNA analysis is considered. The ISN Forum adds testing for C5 p.(Arg885His) and for non-complement genes (MMACHC, MTR, MTHFD1, TSEN2, EXOSC3, DGKE; INF2, WT1, NPHS2, ACTN4 and HSD11B2 in secondary TMA) [2]. The indications for genetic testing in adults with chronic kidney disease of undetermined cause and the technical limits of renal panels are common to the hereditary nephropathies.

Test limitations must appear in the report and interpretation. Next-generation sequencing (NGS) does not detect CFH–CFHR cluster hybrids and deletions, requiring MLPA or copy number methods; the Brazilian consensus recommends MLPA, considered the copy number reference standard, if NGS is negative (grade 1B) [1, 14]. Panels vary between laboratories (CFH, CFI, CFB, C3, CD46, THBD, with or without PLG, DGKE, CFHR1–5 and MASP2), and there is no consensus on the ideal panel [14]. Approximately 40% of patients have no identified variant, and a negative result excludes neither aHUS nor C5 inhibitor indication [2, 14]. Rare complement variants have similar frequency in patients with secondary TMA and healthy people and are common in the general population, making variants of uncertain significance (VUS) frequent [2, 60]. The proportions in table 2.2 refer to the fraction of known variants detected by each method rather than the probability of a patient with aHUS testing positive [1]. Leucocyte tests do not detect exclusively germline mosaicism and may miss low-level somatic mosaicism [1].

The report should specify the gene, reference transcript with version (for example NM_000186.4 for CFH; NM_172351.3 for CD46, the MANE Select transcript), HGVS description at DNA and protein levels, zygosity, classification according to American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria, CNV method and risk haplotypes [1, 57]. VUS interpretation benefits from functional data (FH and FI levels, CD46 expression) and family segregation, remembering that incomplete penetrance limits segregation: a healthy carrier relative does not exclude pathogenicity [40, 74]. Molecular testing is recommended in all affected individuals before transplantation because it defines recurrence risk [1].

5.7. Kidney biopsy

Kidney biopsy is not mandatory; it is recommended for graft dysfunction (to distinguish TMA from rejection), suspected underlying glomerulonephritis and estimating chronicity (grade 1B) [14]. It shows acute, subacute or chronic glomerular and arteriolar TMA, thrombotic vasculopathy and, on immunofluorescence, IgM, fibrinogen and C3 deposits [14]. It may be decisive in separating aHUS from C3G, requiring immunofluorescence and electron microscopy, and in rare overlap of both diagnoses in the same or serial biopsies, described with homozygous CFH variants [1, 23].

5.8. Prenatal and preimplantation diagnosis

Prenatal and preimplantation genetic testing (PGT) are technically possible after identifying the familial variant or variants [1]. In dominant forms, incomplete penetrance and trigger effects limit predictive value; in DGKE nephropathy, recessive with complete penetrance in described families, results have high predictive value [1]. Views on using these tests differ between professionals and families, and discussion should take place before pregnancy (section 8) [1].

6. Differential Diagnosis

6.1. Comparative table

6.2. STEC-HUS and its coexistence with aHUS

Typical HUS manifests as an acute illness with a diarrhoeal prodrome, frequently bloody; in the acute phase, detection of Shiga toxins in stool (Vero cell assay or commercial immunoassays, with results within a few hours) and/or of serum antibodies against the toxin or against the lipopolysaccharides of serogroups O157, O26, O103, O111 and O145 distinguishes typical HUS from aHUS; STEC isolation and anti-lipopolysaccharide antibodies are not routinely available and take days [1]. Approximately 80%–90% of patients recover without sequelae, spontaneously in most children or after plasma therapy in severe forms and in adults, and relatives of patients with typical HUS do not require surveillance [1]. The ECULISHU trial, with eculizumab for 28 days in mild to moderate STEC-HUS, failed its primary endpoint; ECUSTEC, which included severe cases, was ongoing at the time of the ISN Forum [2]. The labels of the C5 inhibitors expressly exclude STEC-HUS from the indication [15, 17]. Coexistence is possible: in an 11-month-old boy with confirmed STEC-HUS (Stx1 and Stx2 in stool), an unexpected fall in platelets during recovery (day 13; from 305,000 to 157,000/mm³), with normal ADAMTS13 and homocysteine and C3 of 0.62 g/L, responded to eculizumab (platelets 417,000/mm³ within 7 days); he had heterozygous VUS in CFH and CFI, and withdrawal of the inhibitor because of diagnostic uncertainty was followed by relapse 3.5 months later, after a febrile respiratory infection [60]. The Brazilian consensus notes that aHUS variants have been described in STEC-HUS, in pregnancy-associated TMA, in refractory autoimmune diseases, after haematopoietic stem cell transplantation (HSCT) and in monoclonal gammopathy [14]. The French PNDS provides the epidemiological framework of the differential in France: 100 to 160 annual cases of STEC-HUS in children and at least 10 to 20 in adults; serogroup O157:H7 no longer predominates and, since 2018, O26 has prevailed, followed by O80, a hybrid pathotype emerging since 2005; progression to HUS occurs in 3%–9% of sporadic enterohaemorrhagic E. coli infections and in up to 20% in outbreaks, more than half of patients require dialysis and mortality is 1%–5% [11]. STEC testing (PCR for the stx and eae genes in stool or a rectal swab, repeated if negative, collected before antibiotics) is performed in all patients, even without diarrhoea; plasma exchange is not recommended, bactericidal antibiotics, especially quinolones, are discouraged, and the Société française de pédiatrie reserves eculizumab for severe forms with neurological or cardiac involvement [11]. The same protocol considers as poor prognostic factors severe central nervous system involvement, severe colitis or pancreatitis, neutrophilia above 20,000/mm³, haemoconcentration at diagnosis and anuria lasting more than 8 days (long-term renal prognosis); it requires early notification to Santé publique France of cases below 16 years of age; it allows azithromycin for its antitoxin effect in animal models, reserving bactericidal antibiotics for the rare septicaemic forms and for Shigella dysenteriae; and, when a reference centre indicates eculizumab in STEC-HUS, it uses the aHUS dosage for a maximum of 6 weeks, with one weekly dose, stopped as soon as there is clinical improvement [11]. Of direct genetic relevance, the PNDS recommends testing for alternative pathway variants when STEC-HUS progresses to ESRD, and regards the patient with STEC-HUS as having CKD even after complete recovery of renal function [11].

6.3. Pneumococcal HUS

S. pneumoniae-associated HUS is a severe condition, with respiratory distress, neurological involvement and coma [1, 62]. Adult plasma contains antibodies against the Thomsen–Friedenreich antigen, which may aggravate the disease; therefore, fresh frozen plasma (FFP) is contraindicated, washed erythrocyte and platelet concentrates are preferred, and there is no evidence of benefit from plasma exchange [1, 62]. The KDIGO conference notes that C3 is also low in this form [13]. The French PNDS places this form mainly below 2 years of age, explains it by pneumococcal neuraminidase, which exposes the Thomsen–Friedenreich antigen recognised by the patient's IgM, and treats a positive direct Coombs test coexisting with mechanical haemolysis as a diagnostic marker; treatment is appropriate antibiotic therapy (amoxicillin) with plasma-depleted (washed) red cell concentrates, and plasma therapy is contraindicated while the polyagglutinability test remains positive [11].

6.4. Congenital and immune thrombotic thrombocytopenic purpura

aHUS and TTP share TMA injury, but manifestations predominate in the kidney in aHUS and are more diffuse in TTP, with mainly central nervous system symptoms, although renal insufficiency may occur [1]. Approximately 80% of TTP results from ADAMTS13 activity deficiency, constitutional due to biallelic variants or acquired through an inhibitory antibody; ADAMTS13 deficiency is not typically found in HUS except in rare patients with simultaneous ADAMTS13 and CFH variants [1]. ADAMTS13 cleaves ultra-large von Willebrand factor multimers; without it, these multimers bind platelets and occlude the microcirculation [75]. Severe deficiency is much more frequently immune than congenital, and isolated low activity does not mean congenital TTP; the 2020 International Society on Thrombosis and Haemostasis (ISTH) diagnostic guideline recommends measuring the inhibitor or anti-ADAMTS13 IgG alongside activity [75, 76]. In congenital TTP, activity is severely reduced without antibody; prior plasma or recombinant enzyme raises measured activity, so sampling should occur beforehand without delaying treatment; in adults with severe deficiency without antibody, ADAMTS13 sequencing is necessary because inhibitor assays have limited sensitivity, and definitive confirmation requires biallelic variants [75].

Treatment of congenital TTP differs from that of aHUS. Recombinant ADAMTS13 (Adzynma®, initial FDA approval on 9 November 2023) is indicated as prophylactic or on-demand enzyme replacement therapy in adults and children: prophylaxis with 40 IU/kg once every two weeks, adjustable to 40 IU/kg weekly; on demand, 40 IU/kg on day 1, 20 IU/kg on day 2 and 15 IU/kg daily from day 3 until 2 days after resolution of the acute event; intravenous infusion at 2–4 mL/min, with the dose calculated from the actual vial potency [24]. In September 2026, the United States label received a boxed warning on neutralising anti-ADAMTS13 antibodies with serious outcomes, including death, which had not been observed in the trials (in patients all previously exposed to plasma); monitoring of ADAMTS13 activity and neutralising antibodies before and periodically during treatment is recommended, and there are no data in plasma-naïve patients [24]. In the European Union, Adzynma® received a positive opinion from the Committee for Medicinal Products for Human Use (CHMP) on 30 May 2024 and authorisation on 1 August 2024, under exceptional circumstances and with orphan designation of 3 December 2008, for ADAMTS13 deficiency in congenital TTP in children and adults of all ages, with the same dosage as the United States label [25]. The European SmPC allows self-administration or home administration after training and good tolerance, foresees that infants under 10 kg may need weekly doses, treats immunogenicity as a general warning (antibodies with possible loss of response) and restricts use in pregnancy to an individual benefit–risk assessment, owing to lack of data [25]. In the crossover phase 3 trial 281102 (45 patients aged 3 to 58 years, 6 months of prophylaxis with each treatment), there were 0 acute TTP events with the recombinant product versus 1 with plasma, and subacute events in 1 versus 6 patients [25]. The 2025 ISTH focused update issued a strong recommendation for recombinant ADAMTS13 over FFP in congenital TTP in remission (moderate certainty) and suggests FFP over watchful waiting if the recombinant product is unavailable (very low certainty) [78]. The 2020 ISTH treatment guideline strongly recommends prophylactic plasma infusion during pregnancy in asymptomatic congenital TTP [75, 79]. Immunosuppression (corticosteroids, rituximab) and caplacizumab belong to immune TTP, and the 2023 British Society for Haematology (BSH) guideline requires all TTP in England to be managed at designated regional centres [75, 80].

The French TTP PNDS, coordinated by CNR-MAT and published by the HAS in October 2022, devotes a specific part to the congenital form, formerly called Upshaw–Schulman syndrome [20]. About 200 ADAMTS13 variants have been described, in homozygosity or compound heterozygosity, with autosomal recessive transmission; the first episode usually occurs before 10 years of age and, in more than half of patients, right at birth. In the newborn, haemolysis and thrombocytopenia without apparent cause sometimes lead to exchange transfusion for hyperbilirubinaemia, a finding that should be sought in the child health record of an adult with suspected disease [20]. The congenital form accounts for less than 5% of severe ADAMTS13 deficiencies and less than 5% of adult TTP outside pregnancy, but for 24–66% of first TTP episodes in an obstetric context; in the summary for the attending physician, half of TTP cases in a first pregnancy are congenital [20]. In women with the congenital form, the first pregnancy almost always triggers an episode, in the third trimester in 75% of cases; with plasma, the maternal prognosis is good, but the fetal prognosis is poor once the episode sets in [20].

The first episodes resolve completely. Without sufficient prophylactic plasma, organ damage from repeated ischaemia appears after a few years: chronic kidney disease (which can lead to a misdiagnosis of HUS), stroke, myocardial infarction, atherosclerosis, ischaemic retinopathy and epilepsy [20]. There is also a chronic picture of abdominal pain, fatigue, headaches and mood disturbances, and haemolysis and thrombocytopenia tend to become moderate and persistent; diagnostic delay may reach several years [20]. The family history may reveal siblings with the same picture or deaths from neonatal haemolysis, and the parents are healthy [20]. After any TTP episode, the protocol describes cognitive sequelae in 63–88% of survivors, depending on the series (language, attention, dexterity, memory), attributed to sometimes silent cerebral infarcts, as well as severe depression, poorer quality of life and more cardiovascular disease [20].

The diagnostic algorithm of the PNDS starts from ADAMTS13 activity [20]. With activity below 10%, anti-ADAMTS13 IgG is tested: if positive, the diagnosis is immune TTP; if negative, activity is repeated in clinical remission. Activity that remains below 10%, without antibody, leads to gene sequencing (Sanger, in transition to next-generation sequencing); biallelic variants confirm the congenital form and absence of variants indicates acquired deficiency. If activity in remission rises above 10% without exceeding 30% within about a year, or without reaching 50%, sequencing can look for a monoallelic variant or a combination of polymorphisms explaining the incomplete recovery. Non-reduced activity points to complement investigation [20]. The PNDS requires activity to be measured in every TMA in children; it notes that commercial kits detect activity below 10% with 12% false negatives compared with reference tests; and, in follow-up of obstetric TTP in remission, it requests activity, antigen and anti-ADAMTS13 IgG, because the congenital and immune forms carry different risks and management in the next pregnancy [20]. Samples are collected before any plasma, with the consent of the patient or parents: a plain or citrated tube for activity and antibodies, and a 5 mL EDTA tube, frozen without centrifugation, for genetic testing [20].

Table 6.4a. Management of congenital TTP according to the French PNDS (2022) and comparison with other sources

For genetic counselling, the PNDS gives a risk per pregnancy of 1 in 4 of an affected child, 2 in 4 of a carrier and 1 in 4 of a non-carrier [20]. It does not offer routine prenatal diagnosis, because genotype does not predict prognosis and some patients have their first episode only in adulthood; it may be discussed case by case, at the parents' request. In families with an affected child, diagnosis in subsequent pregnancies can be made at birth. All siblings should have ADAMTS13 activity measured, followed by complete genotyping of the gene if it is abnormal; other cases in the family should be sought, particularly with consanguinity. Heterozygous parents are not at risk, and their activity can be measured on request or during the family investigation [20].

The sources diverge on five points. First, the PNDS predates the authorisation of recombinant ADAMTS13 and keeps plasma as the reference treatment, mentioning the recombinant product as forthcoming; the 2025 ISTH update recommends the recombinant product over FFP in the congenital form in remission and leaves plasma for when the recombinant product is unavailable; on this point, the ISTH is followed as the most recent guideline [20, 78]. Second, the PNDS text gives 10 mL/kg of plasma in the episode and in prophylaxis, and the supporting document 10–15 mL/kg; the text is the reference and the supporting document shows the range used. Third, plasma prophylaxis every 2–3 weeks is adjusted by platelets, whereas the labels of the recombinant product fix the dose at 40 IU/kg fortnightly or weekly [20, 24, 25]. Fourth, in pregnancy, the PNDS quantifies plasma volumes by trimester, the 2020 ISTH recommends prophylaxis without that degree of detail in the document cited, and the European SmPC of the recombinant product has no data in pregnancy [20, 25, 79]. Fifth, on immunogenicity, the PNDS documents no alloimmunisation under plasma, the European SmPC has a general warning and the United States label received a boxed warning in September 2026 for serious neutralising antibodies; monitoring of activity and antibodies follows the United States label [20, 24, 25].

6.5. Intracellular cobalamin defects (cblC, cblE, cblG, cblD)

cblC, caused by biallelic MMACHC variants, is the prototype and most common intracellular cobalamin defect [1, 37]. In the infantile form, HUS may be fatal unless daily hydroxocobalamin (OHCbl) is initiated promptly, and is accompanied by increased-anion-gap metabolic acidosis, ketonuria, hyperammonaemia and progressive maculopathy or retinopathy in most [37]. During the first years of life, renal TMA, HUS, pulmonary hypertension and diffuse lung disease may develop; in adolescents and adults the neuropsychiatric presentation predominates, but renal TMA, HUS, pulmonary hypertension and pulmonary thrombotic events also occur [37]. Saudubray also describes renal insufficiency, liver dysfunction, cardiomyopathy, interstitial pneumonitis, characteristic retinopathy with perimacular hypopigmentation and a hyperpigmented ring, nystagmus, microcephaly, hydrocephalus, congenital heart disease and dysmorphism [81]. Atypical HUS is uncommon in cblC and generally occurs within the first 30 days of life; a small group of newborns with TMA, anaemia and HUS died early; two siblings with late-onset TMA were successfully treated with OHCbl; and a 20-year-old man presented with malignant hypertension and renal insufficiency with arteriolar and glomerular TMA, described as eculizumab-resistant HUS associated with cobalamin C deficiency [77]. In Portugal, a 19-year-old woman with TMA due to adult-onset methylmalonic aciduria with homocystinuria was described before any central nervous system involvement, supporting homocysteine measurement also in adults [82]. Saudubray includes aHUS, nephrotic syndrome and tubulointerstitial nephropathy among presenting signs of inborn errors of metabolism and attributes HUS with nephrotic syndrome to DGKE variants [83]. HUS also occurs in cblE (MTRR) and cblG (MTR), and renal artery thrombosis, HUS and pulmonary hypertension have been described in cblD-homocystinuria [1, 37].

Source: GeneReviews on intracellular cobalamin defects [37].

The identification of MMACHC by Lerner-Ellis and colleagues in 2006 found 42 variants in 204 individuals, with c.271dupA in 40% of all alleles [84]. Severe infantile disease is associated with homozygous or compound heterozygous c.271dupA or c.331C>T, and late disease with c.394C>T and c.482G>A; c.271dupA is also associated with late onset when compound heterozygous with c.394C>T, c.347T>C, c.440G>C or c.482G>A [37, 85]. In Morel's review of 25 early-onset patients (before 6 months), 17 were homozygous for c.271dupA (n=9) or c.331C>T (n=3) or compound heterozygous (n=5); c.331C>T is a founder in Cajun and French-Canadian populations, and c.394C>T is common in patients from India, Pakistan and the Middle East [85]. Nyhan's Atlas reports more than 148 variants, c.271dupA in more than half of southern European and French-Canadian patients, and c.271dupA homozygosity uniformly associated with a severe early phenotype [77]. Normalised to MANE Select transcript NM_015506.3, the main variants are c.271dup p.(Arg91LysfsTer14), c.331C>T p.(Arg111Ter), c.394C>T p.(Arg132Ter), c.482G>A p.(Arg161Gln) and c.609G>A p.(Trp203Ter), the last common in patients of Chinese ancestry [37, 57]; the protein designation with a stop codon at position 140 appearing in some tables for c.271dup does not correspond to current normalisation.

Treatment of cblC is based on parenteral OHCbl, started immediately on clinical suspicion or positive newborn screening; cyanocobalamin and the oral route are ineffective [37]. In infants, the daily dose is 1.0 mg (approximately 0.3 mg/kg/day) intramuscularly or subcutaneously, adjusted to weight to maintain 0.3 mg/kg/day (with concentrated OHCbl up to 30 mg/mL) and titrated empirically according to clinical status and total homocysteine, methylmalonic acid and methionine, with no maximum dose defined in the source; oral betaine is added if homocysteine is elevated, starting at approximately 250 mg/kg/day in 3–4 doses and titrated by plasma total homocysteine and methionine (no maximum dose defined in the source), together with folate or folinic acid (in adults, 1.0 mg/day orally, reduced to 0.5 mg/day for maintenance; the source refers paediatric doses to reference formularies and defines no maximum) [37]. For thromboembolic complications (HUS, TMA), OHCbl and betaine are started or increased; protein restriction and methionine-free formulas intended for isolated methylmalonic aciduria are avoided [37]. In Nyhan's Atlas, OHCbl was used at 1–1.5 mg intramuscularly daily; in three newborns with cblC-related HUS, parenteral OHCbl, folic acid and betaine normalised the biochemical abnormalities, and two patients required 5 mg/day to control TMA [77]. Betaine anhydrous (Cystadane®) is FDA-approved for homocystinuria, including cobalamin cofactor defects [86]. The Brazilian consensus allows empirical emergency OHCbl while results are awaited, given that the disease is treatable and there are no serious adverse effects [14]. The ravulizumab trials excluded patients with a genetic defect of cobalamin C metabolism, and the ISN Forum classifies MMACHC, MTR and MTHFD1 among the genes whose TMA does not respond to eculizumab [2, 17].

The French PNDS treats cblC (MMACHC, 1p34.1, autosomal recessive) as a metabolic cause of HUS to be sought in all patients, particularly with neurological signs or with negative investigation for TTP, aHUS and STEC [11]. Initial manifestations appear in the neonatal period in 50%, in childhood or adolescence in 40% and in adulthood in 10%, and the median age at the HUS episode is 6 years; megaloblastosis, leucopenia or pancytopenia and lack of response to eculizumab point to the diagnosis, which rests on hyperhomocysteinaemia greater than that explained by renal failure, raised methylmalonic acid and genetics [11]. In an emergency, the protocol indicates 1,000 µg of hydroxocobalamin intramuscularly, discourages the intravenous route because of anaphylactic reactions, and adds folic acid, betaine and L-carnitine, leaving the other products to the specialist centre [11]. The dose coincides with the GeneReviews daily dose of 1.0 mg in infants; the divergence lies in the route (GeneReviews allows the subcutaneous route, not the intravenous route, as an alternative) and in the inclusion of L-carnitine, which GeneReviews does not mention in this context [11, 37].

6.6. C3 glomerulopathy

C3G is an ultra-rare kidney disease mediated by fluid-phase alternative pathway dysregulation, rarely with simple Mendelian transmission, affecting all ages (median 23 years at diagnosis) and presenting with haematuria, proteinuria, nephritic or nephrotic syndrome and low C3 [1, 23]. It comprises C3 glomerulonephritis (C3GN) and dense deposit disease (DDD, formerly type 2 MPGN); C3 nephritic factor (C3NeF) is present in up to approximately 50% of C3GN and 80% of DDD; half of patients progress to ESRD within 10 years; acquired partial lipodystrophy occurs in 5%–17%, and retinal drusen confer late visual risk [23]. United States prevalence was estimated at 2–3 per million [23]. Definitive diagnosis requires biopsy with immunofluorescence and electron microscopy, and disease recurs in almost all grafts [23]. C3G therapy has diverged from aHUS therapy: iptacopan is approved for C3G by the FDA (March 2025) and EMA (adults, with an angiotensin-converting enzyme inhibitor [ACE inhibitor] or angiotensin receptor blocker [ARB], or if intolerant), and pegcetacoplan was FDA-approved for C3G and primary immune complex MPGN from 12 years (July 2025), with relative proteinuria reduction of 68.1% versus placebo at 26 weeks in VALIANT [44, 87–89].

6.7. Secondary TMA and mimics

In a French series of 500 TMAs from four centres, cited by the Brazilian consensus, 94% were secondary, with overlapping categories: pregnancy 35%, infection 33%, drugs 26%, malignancy 19%, transplantation 17%, autoimmune disease 9%, malignant hypertension 4% and other 6% [14]. TMA mimics include prosthetic valves, extracorporeal circulation, sickle cell crisis and tumour emboli [14]. In secondary TMA, rare complement variants have a frequency similar to healthy populations, the homozygous CD46 GGAAC haplotype is more frequent, and complement inhibitor benefit is controversial [2, 13]. In malignant hypertension, cause and consequence are difficult to separate, and HSD11B2 disease (apparent mineralocorticoid excess) shows that hypertension alone may cause TMA [2]. The randomised HYPERSHU trial compares eculizumab plus renin–angiotensin system blockade with blockade alone in HUS associated with hypertensive emergency [90].

The French PNDS defines malignant hypertension (hypertensive emergency) in adults as systolic pressure above 180 mmHg and/or diastolic pressure above 120 mmHg with ischaemic visceral damage, and notes that one third of aHUS cases are accompanied by malignant hypertension, with no clinical or histological criterion that separates the two mechanisms with certainty [11]. Glomerular capillary thrombi point towards aHUS, and the distinction rests on signs of chronic repercussion of hypertension (retina, myocardium, cerebral circulation) and on investigation of the alternative pathway; more than two thirds of malignant hypertension cases arise on essential hypertension. Treatment is given in intensive care with intravenous urapidil, nicardipine or labetalol, cautious rehydration, without diuretics in the absence of left ventricular overload, with a 20%–25% reduction in mean arterial pressure within a few hours and early introduction of renin–angiotensin system blockers after volume repletion [11]. The other causes of secondary HUS are summarised in the following table, as a differential.

Source: French PNDS, sections 5.2 to 5.7 and appendix 9 [11].

7. Treatments and Risk Management Strategies

7.1. Principles and supportive measures

All patients with aHUS are eligible for a C5 inhibitor, recommended as first line (grade 1A in the Brazilian consensus); initiation in the acute phase improves renal recovery [13, 14]. GeneReviews considers eculizumab first line when the diagnosis is unequivocal, with the potential to rescue renal function if given early [1]. Supportive measures include treatment of AKI and hypertension, dialysis when necessary, red cell concentrate transfusion if haemoglobin is below 7 g/dL and platelets only for active bleeding or an invasive procedure, with a direct Coombs test sample collected before transfusion (grade 1B) [14]. ACE inhibitors or ARBs reduce progression to ESRD and control blood pressure [1]. Triggering drugs should be withdrawn; ciclosporin and tacrolimus should be stopped if aHUS develops after exposure [1]. The French PNDS specifies phenotyped, leucodepleted red cell concentrates below 7 g/dL of haemoglobin (or earlier if poorly tolerated), transfusion simultaneous with dialysis in the presence of hyperkalaemia or overload, monitoring of anti-HLA antibodies, and no platelet transfusion without bleeding, because platelets can aggravate TMA, overload and HLA alloimmunisation in future transplant candidates; it prohibits intramuscular injections and arterial punctures and avoids potassium-containing solutions [11].

7.2. Plasma therapy

Plasma therapy is the only treatment with near-global availability and retains a place when C5 inhibitors are unavailable, in anti-FH disease and emergencies with strong TTP suspicion [1, 13]. It is initiated as soon as aHUS is suspected and continued until TMA resolves; evidence for plasma exchange superiority over infusion is minimal, and choice depends on tolerance, experience and resources, with newborns benefiting from infusion [1, 42]. GeneReviews describes the following doses: for plasma exchange, 1–2 plasma volumes (40 mL/kg) per adult session and 50–100 mL/kg in children, initially daily and then according to response, intensifiable by increasing volume, with one-volume exchange twice daily probably the option for refractory disease by minimising recycling of infused plasma; for infusion, 30–40 mL/kg initially, followed by 10–20 mL/kg/day [1]. Platelets and LDH are the most sensitive response markers, and plasma should continue until both remain normal after withdrawal; immediate exacerbation after withdrawal occurs in 29%–82%, and repeated dependency cycles justify switching to eculizumab [1]. Plasma resistance is defined as absence of remission despite treatment and dependence as relapse as soon as plasma stops [1]. The 2009 European paediatric guidelines, 2010 British guideline and 2016 international paediatric consensus guided plasma therapy; only the abstract of the last, published in a restricted-access journal, could be read, and its detailed recommendations reach this article through the KDIGO conference and Brazilian consensus [13, 14, 91–93].

The French PNDS differentiates initial management by age [11]. In adults, given the frequency of TTP and secondary TMA, daily plasma exchange with virus-inactivated plasma is the most widely used initial treatment: 60 mL/kg per session in the text and 1.5 plasma volumes (60–75 mL/kg) for at least 5 to 7 days in the algorithm of appendix 8, until an aetiological diagnosis is available; with a family history or arguments in favour of aHUS, eculizumab can be started immediately, at any age, and in complement-mediated aHUS it replaces plasma exchange even after haematological remission. In children, eculizumab is first line, and plasma therapy is used only if the inhibitor is unavailable (60 mL/kg per session while waiting), in the form with anti-FH antibodies or in the newborn, because of the complications of vascular access and the infectious risk. These doses differ from those of GeneReviews (40 mL/kg, 1–2 volumes in adults; 50–100 mL/kg in children); the choice of eculizumab as initial treatment in adults with known ADAMTS13, advocated by KDIGO, contrasts with the French priority given to plasma exchange until diagnosis [1, 11, 13].

Plasma response varied with genotype [1, 26, 28]. In patients with C3 variants, remission occurred in 57% of episodes, similar to CFH. With CFH, approximately 60% of treated patients had complete or partial remission (haematological normalisation with renal sequelae) and 20% died in the acute episode; some required weekly plasma to maintain FH levels; and in the French series, first-episode progression to ESRD was similar with or without intensive plasma therapy [1, 12]. With CFI, remission occurred in only 25% of episodes. With CD46, whose product is transmembrane, the rationale is weak: 80%–90% remitted with plasma, but all untreated patients also recovered, and decisions should depend on severity. With anti-FH antibodies, plasma induced complete or partial remission in 75% of episodes. With DGKE, plasma probably provides no benefit because relapses occurred during treatment. For THBD, remission was described in approximately 80% of episodes [1, 26]. Risks include catheter morbidity and hypervolaemia [14].

7.3. Eculizumab (Soliris® and biosimilars)

Eculizumab is a recombinant humanised monoclonal antibody against C5, preventing its cleavage by C5 convertase and blocking C5a and C5b-9 complex formation [23]. It is supplied as a 300 mg/30 mL vial for intravenous infusion, diluted to a final concentration of 5 mg/mL, infused over 35 minutes in adults (25–45 minutes in the European label) and 1–4 hours in children; if infusion is slowed, it should not exceed 2 hours in adults, and patients should be monitored for at least 1 hour afterwards [14–16]. Label doses for aHUS are as follows [15, 16].

Doses may be administered on scheduled days or up to two days later according to the United States label [15]. The Brazilian consensus table gives, for the 10 to under 20 kg band, 600 mg weekly for 2 weeks and 300 mg in week 3, differing from FDA and European Medicines Agency (EMA) labels, which prescribe a single 600 mg dose at week 1 and 300 mg at week 2; the label takes precedence [14–16]. Plasma therapy removes the antibody and requires supplementary doses [15].

The European label recommends lifelong treatment unless discontinuation is clinically indicated and allows shortening maintenance intervals to 12 days with platelet, LDH and creatinine monitoring [16]. The KDIGO conference and ISN Forum consider the approved regimen suboptimal for many patients and advocate individualisation (section 7.7) [2, 13]. Severe proteinuria increases urinary antibody excretion and may render treatment ineffective, requiring higher doses, and late pregnancy also requires higher doses and close monitoring [2].

The main adverse effects are Neisseria infections: eculizumab increases the risk of meningococcal infection approximately 550-fold, even with vaccination and antibiotic prophylaxis, and there is also a risk of disseminated gonococcal infection [2, 10, 15]. The United States label has a boxed warning, and the drug is available only through the ULTOMIRIS and SOLIRIS REMS (Risk Evaluation and Mitigation Strategy) programme [15]. Anti-eculizumab antibodies developed in 3 of 100 patients with aHUS, with low neutralising titres in 1 [16]. Vaccination may activate complement and worsen TMA signs [16]. The effect of withdrawing anticoagulation during treatment has not been established [15]. The French PNDS reproduces in appendix 7 the weight-based dosage of the label, including 600 mg in a single induction week in the 10 to less than 20 kg band, and the supplementary doses with plasma therapy, and notes that eculizumab is well tolerated and can be infused in a febrile patient; active infection with an encapsulated bacterium must be excluded before starting [11].

7.4. Ravulizumab (Ultomiris®)

Ravulizumab is a long-half-life anti-C5 antibody allowing administration every 8 weeks versus 2 weeks with eculizumab [17, 38]. In the USA, it is available in 300 mg/3 mL and 1,100 mg/11 mL vials (100 mg/mL), infused using a 0.2–0.22 µm filter at a final concentration of 50 mg/mL; the October 2025 label no longer includes the subcutaneous formulation [17]. In Brazil, only the 300 mg/3 mL presentation is available [14]. The loading dose is followed 2 weeks later by the first maintenance dose; subsequent doses allow occasional variation of ±7 days [17].

Sources: FDA and EMA labels and Brazilian consensus [14, 17, 18]. For patients on eculizumab, the United States label administers loading on the next scheduled eculizumab dose date and the European label 2 weeks after the last eculizumab infusion; maintenance begins 2 weeks after loading [17, 18]. After plasma exchange or plasmapheresis, the label requires supplementary dosing within 4 hours after each session, and after an intravenous immunoglobulin (IVIg) course a supplementary 600 mg dose [17].

aHUS treatment with ravulizumab should last at least 6 months, with subsequent duration individualised; higher-recurrence-risk patients may require chronic therapy [17, 18]. Indication differs by jurisdiction: in the USA it covers adults and children from 1 month (and 5 kg), in the European Union patients weighing 10 kg or more, inhibitor-naïve or responding to at least 3 months of eculizumab [17, 18]. The paediatric switching trial required prior eculizumab response, and there is no information on ravulizumab in eculizumab-refractory patients [17, 18]. CH50 monitoring of blockade is unreliable with ravulizumab, and the Brazilian consensus recommends clinical monitoring and serum drug measurement (grade 1A) [14]. In the Global Registry, 60 patients (43 adults and 17 children) switched from eculizumab to ravulizumab; of 49 analysed for effectiveness, 22% had a pathogenic complement variant; 20 adverse events in 13 patients were recorded, 3 related (infusion reaction, headache, fatigue), without meningococcal infections, deaths, new dialysis, transplantation or TMA events [94]. The National Institute for Health and Care Excellence (NICE) recommended ravulizumab in aHUS in 2021 in people weighing 10 kg or more, inhibitor-naïve or responding to 3 months of eculizumab, considering efficacy probably equivalent, quality-of-life improvement with infusions every 8 weeks and lower cost [95].

7.5. Eculizumab biosimilars

In the European Union, Bekemv® (Amgen Europe, 19 April 2023) and Epysqli® (Samsung Bioepis, 26 May 2023) were authorised; current summaries of product characteristics (SmPCs) for both include aHUS in adults and children at Soliris® doses, although the Bekemv® indication summary in the EMA database still mentions only PNH [88, 96, 97]. In the USA, Bkemv® (eculizumab-aeeb) was approved on 28 May 2024 and Epysqli® (eculizumab-aagh) on 19 July 2024; July 2026 labels include PNH, aHUS, generalised myasthenia gravis (gMG) and neuromyelitis optica spectrum disorder (NMOSD), with the same STEC-HUS use limitation and boxed warning [98, 99]. Biosimilar efficacy is based on PNH trials, without TMA trials [2]. United States interchangeability status was not determined in the sources consulted. The Brazilian consensus mentions Elizaria® (IBC Generium, Russia) as the first eculizumab biosimilar registered worldwide and prefers reference anti-C5 agents because of limited biosimilar experience [14].

7.6. Prevention of meningococcal infections

Meningococcal vaccination is a condition of complement inhibitor treatment, but recommendations differ on antibiotic prophylaxis [13–16].

Sources: [1, 13–17]. KDIGO's rationale for continuous prophylaxis is uncertainty about protection conferred by vaccine antibodies during C5 blockade and vaccine response during AKI, CKD and immunosuppression; the United States label acknowledges that risk persists even in patients developing antibodies [13, 15]. A 2026 review of the five FDA-approved complement inhibitors for kidney disease (eculizumab, ravulizumab, avacopan, iptacopan and pegcetacoplan) systematises the risk of encapsulated bacterial infection (N. meningitidis, S. pneumoniae, H. influenzae type b) and preventive measures [100]. Patients should be instructed to seek immediate care for fever or infection signs [15]. The KDIGO conference recommends not interrupting anticomplement therapy during intercurrent illness because of high relapse risk, unless there is suspected or documented encapsulated bacterial infection [13].

The French PNDS estimates the risk of meningococcaemia under eculizumab, before vaccination against serogroup B, at 1 case per 200 patients treated per year, and places the excess risk up to two months after withdrawal; it requires the patient to be informed of warning signs and to carry the patient card and the emergency card [11]. The protocol does not quantify the “full dose” of phenoxymethylpenicillin; the reference dose for continuous antibacterial prophylaxis, by extrapolation from pneumococcal prophylaxis in asplenia with the same drug in the French SmPC of Oracilline®, is 100,000 IU/kg/day up to 10 kg, 50,000 IU/kg/day from 10 to 40 kg without exceeding 2,000,000 IU/day, and 2,000,000 IU/day in adults, always in 2 oral doses; the SmPC gives 2,000,000–4,000,000 IU/day in adults for curative treatment [11, 101]. The French duration (60 days after withdrawal) coincides with the lower limit of the 2–3 months of KDIGO, and the extension to CH50 normalisation in children has no equivalent in the other sources [11, 13]. As for the other vaccines, the PNDS considers that the risk of an episode triggered by an infection exceeds that of one triggered by the vaccine, that antibody production appears effective under eculizumab and that indications should be defined by a reference centre [11]. For meningococcal boosters, which the PNDS makes dependent on antibody titres, ACIP recommends, in patients on a complement inhibitor, a serogroup B booster 1 year after the primary series and then every 2–3 years, and continued vaccination against serogroups A, C, W and Y and B despite incomplete protection: with endogenous complement, only 17% of patients with terminal complement deficiency or on an inhibitor had bactericidal activity after the vaccine, and there are cases of meningococcal disease in vaccinated individuals with antibodies [102]. Titres do not demonstrate protection under C5 blockade, so the ACIP recommendation serves as the reference for the booster schedule.

7.7. Monitoring blockade and individualising dose

The Brazilian consensus allows eculizumab dose spacing only after at least 3 months of improvement without disease activity (grade 2A), never without available CH50 (grade 1B) and never in kidney transplant recipients (grade 1A); the proposed algorithm measures CH50 the day before dosing and, if below 30%, extends the interval to 3 weeks and, with markers remaining normal and CH50 below 30%, to 4 weeks, without exceeding that limit (grade 2A) [14]. Supporting data come from small series: 1,200 mg every 4–5 weeks maintained concentrations above 50 µg/mL and CH50 below 10% in 11 patients, targeting 50–100 µg/mL, and in a series of 7 patients extension to 4 weeks below 90 kg and up to 6 weeks below 70 kg was possible [14]. There are no ravulizumab spacing studies, and the label is followed [14]. The ISN Forum advocates individualised maintenance at the lowest possible cumulative dose without loss of efficacy, supported by pharmacological monitoring [2].

The French PNDS assesses efficacy on day 7 after the first dose with CH50 below 20% and/or a blood eculizumab concentration of at least 500 µg/mL [11]. The CH50 threshold is less demanding than the below 10% of KDIGO, and the value of 500 µg/mL is ten times higher than the trough concentration target of 50–100 µg/mL adopted by KDIGO and the Brazilian consensus; the French text does not specify the sampling time, and the reference trough target remains 50–100 µg/mL [11, 13, 14]. CH50 cannot be used in patients with homozygous CFH variants and is uninterpretable in the neonatal period; with insufficient blockade, the causes sought are insufficient dosing (particularly with weight at the boundary between two bands), urinary loss in severe nephrotic syndrome and a C5 polymorphism preventing antibody binding (Asian variant, particularly Japanese) [11]. During prolonged treatment, CH50 can be measured before each infusion in the presence of complement reactivation factors (infection, vaccination, surgery, transplantation, pregnancy, postpartum), signs of disease reactivation, drug withdrawal or to decide the end of antibiotic prophylaxis, and free eculizumab can be measured if blockade is incomplete [11].

7.8. Duration and withdrawal of anti-C5 therapy

C5 inhibitor withdrawal is among the most debated issues. The European eculizumab label recommends lifelong continuation unless medically justified and describes TMA complications from 4 to 127 weeks after withdrawal: among 61 patients withdrawing in aHUS studies (21 children), with median follow-up of 24 weeks, there were 15 severe TMA complications in 12 patients, plus 2 in 2 patients on an unapproved reduced regimen, including severe renal deterioration, hospitalisation and progression to ESRD with dialysis, and one patient progressed to ESRD despite restarting; complications occurred irrespective of a variant, risk polymorphism or autoantibody, and monitoring may be insufficient to predict or prevent them [16]. The United States eculizumab label requires TMA surveillance for at least 12 weeks after withdrawal and ravulizumab for at least 12 months, without specific withdrawal data for the latter; the European ravulizumab SmPC cites a prospective observational study in which eculizumab withdrawal resulted in a 13.5-fold higher TMA recurrence rate [15, 17, 18].

Sources: [15, 17, 18]. Prospective withdrawal studies are small and heterogeneous, with median treatment durations between 2.4 and 19.6 months [2].

These studies are not mutually comparable: they differ in design (prospective or retrospective), treatment duration before withdrawal, proportion of variant carriers and surveillance intensity, and none is randomised; interpreting a relapse rate requires its denominator and follow-up time. The convergent conclusion is that genotype stratifies risk: absence of a pathogenic variant and anti-FH antibodies makes post-withdrawal relapse very unlikely, whereas CFH variants and hybrid genes are associated with earlier, more severe relapses [2, 10, 14]. As expert opinion, the KDIGO conference proposed considering withdrawal individually after at least 6–12 months of treatment and at least 3 months of normalised or stabilised renal function, allowing withdrawal at 3 months in patients (particularly children) with a CD46 variant and rapid remission, maintaining treatment for 4–6 months after dialysis initiation before considering withdrawal, and not withdrawing in transplant recipients, particularly with prior graft loss [13]. The ISN Forum observes that most renal recovery occurs by day 90, raising the question of continuation beyond 3 months, and recommends reassessing diagnosis and considering withdrawal without renal recovery after 90 days, cautiously in advanced CKD [2]. Most relapses were identified by spontaneous patient presentation with viral symptoms rather than routine monitoring, making clear instructions and rapid drug access essential; the value of urinary dipstick self-testing and home blood pressure measurement is unproven [2]. The Brazilian consensus permits planned withdrawal only with genetic characterisation, complement assessment (CH50, sC5b-9) or drug measurement and immediate access to reintroduction (grade 1C), with shared patient decision-making (grade 1A) [14]. In neonatal aHUS, experts cited by Portuguese authors recommend at least 36 months of treatment before considering withdrawal [70].

The French PNDS sets out rules for the patient on dialysis [11]. If dialysis started less than 3 months earlier, eculizumab is started; between 3 and 6 months of dialysis, or after 3 months of eculizumab without effect, kidney biopsy is performed, and the inhibitor is started or continued if there are active lesions; 3 to 6 months of treatment are needed to conclude absence of benefit. On chronic dialysis there is no indication for prophylactic eculizumab except extrarenal TMA, and a patient on dialysis 6 months or more after HUS is considered to have ESRD, with no indication to continue eculizumab beyond the sixth month in the absence of active TMA (skin necrosis, ischaemic heart disease, stroke or persistent haemolysis justify continuation). Dose spacing or reduction is undertaken after expert advice, guided by haemolysis markers, CH50 and trough concentration, with informed consent; withdrawal takes into account the type of variant, the course and the risk of relapse, with close laboratory monitoring [11]. The French time frame of 3 to 6 months is consistent with the 4–6 months of KDIGO and longer than the 90-day reassessment proposed by the ISN Forum [2, 11, 13].

7.9. Treatment of the autoimmune form

In the form with anti-FH antibodies, plasma exchange reduces titres by 80%–85% and induces haematological remission in more than 90% of patients within one week, and maintenance immunosuppression, generally with mycophenolate mofetil, is given for 1 to 2 years [2]. In the Indian Registry, plasma exchange was used in 105 of the 138 children, IVIg in 26 and induction with prednisolone combined with intravenous cyclophosphamide or rituximab in 87; the combination of early plasma exchange with induction immunosuppression improved renal survival, with a number needed to treat of 2.6 to prevent an adverse outcome, and maintenance with prednisolone combined with mycophenolate mofetil or azathioprine significantly reduced relapses [54]. Where available, eculizumab is the initial treatment, before the antibody result is known; it can then be continued alone, combined with immunosuppression or replaced by plasma exchange with immunosuppression, based on retrospective studies; eculizumab alone may not reduce titres, with a risk of immune complex glomerulonephritis [2]. GeneReviews states that patients with anti-FH antibodies benefit from corticosteroids or other immunosuppressants combined with plasma exchange [1]. The French PNDS leaves the choice among plasma exchange, cyclophosphamide, rituximab, mycophenolate mofetil, eculizumab and corticosteroids to the specialist centre; the algorithm of appendix 8 uses an anti-FH titre above 1,000 IU/mL to direct towards plasma exchange and immunosuppression with or without eculizumab; during follow-up, antibodies and C3 are measured every 3–4 months, with a target titre below 1,000 AU/mL and normal C3, and withdrawal of treatment can be discussed after one year of stable values [11]. This schedule is less intensive than the KDIGO paediatric schedule (days 7, 14 and 28, then monthly until one year) [11, 13].

7.10. Kidney transplantation

Molecular testing should precede transplantation in all patients because it defines recurrence risk and prophylaxis need [1, 13]. The KDIGO conference recommends transplantation at least 6 months after dialysis initiation because late recovery is possible with eculizumab and requires resolution of haematological and extrarenal TMA [13]. If aHUS is probable and active, the Brazilian consensus recommends 6 months of anti-C5 before transplantation to assess recovery (grade 1C) and full genetic analysis (grade 1C) [14].

Sources: [2, 13, 14]. The three sources agree on structure and differ in details: GeneReviews, following Zuber, classifies C3, CFB and CFH variants and the CFH/CFHR1 hybrid as high risk, and anti-FH antibodies, CFI variants, VUS and absence of an identified variant as moderate risk; the Brazilian consensus also places the moderate group on prophylaxis, whereas KDIGO allows plasma exchange as an alternative [1, 13, 14].

Prophylaxis efficacy data are observational. In the French national study of 126 transplants in 116 adults (58.7% high risk), 52 received prophylaxis and recurrence occurred in 43 (34.1%), only 4 of whom were receiving prophylaxis, independently associated with lower recurrence and higher graft survival; in the French population cohort of 397 adults, the proportion transplanted among ESRD patients rose from 46.2% in 2012 to 72.3% in 2016 alongside eculizumab use [107]. In the United Kingdom, 38 transplants with prophylactic eculizumab had 1-year graft survival of 97%, versus 64% in a comparable cohort of 33 moderate- or high-risk transplants without prophylaxis [108]. Three-year recurrence-free survival by risk level, without and with eculizumab, was 72% and 100% at moderate risk and 27% and 94% at high risk [2, 107]. Rescue therapy is the alternative: in Zuber's 2012 series, TMA reversed in all 13 treated patients, but full recovery occurred only when treatment began within 7 days; in a prospective Dutch cohort, recurrence was below 30% with high-quality donors and endothelial protective measures, and of 15 rescued patients 13 responded, with 3 graft losses; comparison of the United Kingdom (routine prophylaxis, n=35) and Netherlands (rescue, n=38, of whom 18 required treatment) showed no graft survival difference, but cohorts differed in living donor proportions (20% versus 66%) and high-risk variants (36% versus 5%) [2]. In de novo post-transplant TMA, approximately 30% have an identifiable cause of complement dysregulation, and anti-C5 is indicated for recurrence and de novo TMA with demonstrated dysregulation, without a proven role in antibody-mediated rejection [2]. Late-onset post-transplant TMA has a poor prognosis despite eculizumab: in a British series of 26 patients, death-censored graft survival 12 months after eculizumab initiation was 68%, and 42.9% when presentation occurred more than 12 months after transplantation [109].

In all cases, expanded-criteria donors, donor-specific antibodies, combined mTOR and calcineurin inhibitors, high calcineurin inhibitor doses and prolonged cold ischaemia should be avoided, and cytomegalovirus prophylaxis given (grade 1B) [14]. After transplantation, the same eculizumab or ravulizumab dose is maintained, a calcineurin inhibitor used with level monitoring (grade 1B), mTOR inhibitors avoided (grade 2B), and belatacept is an alternative with limited availability in Brazil [14]; in the French pre-eculizumab series, mTOR inhibitors were associated with recurrence and calcineurin inhibitors were not [68]. The ISN Forum allows inhibitor withdrawal after transplantation in low- or moderate-risk patients with eGFR above 30 mL/min/1.73 m² and no humoral rejection or viral infection (cytomegalovirus, BK virus), and does not recommend it at high risk; probable prophylaxis duration of 3 to 6 months remains unestablished, and switching transplant recipients to ravulizumab proved safe in preliminary data [2].

The French PNDS ranks the risk of graft recurrence, highest in the first 6 months, as follows: CFH, C3, CFB, CFI, CD46 and, finally, no variant [11]. In aHUS with a genetic variant, it recommends starting eculizumab before transplantation and continuing it long term, discussing withdrawal only in patients without an identified variant; this position is more conservative than that of the ISN Forum, which allows withdrawal in low and moderate risk [2, 11]. With eculizumab, a living donor is no longer a contraindication; a related donor is accepted only if not a carrier of the recipient's variant and not at risk of aHUS, and calcineurin inhibitors need not be avoided. Antiphospholipid syndrome is the other TMA aetiology with a risk of recurrence, which contraindicates a living donor when the risk is major and requires anticoagulation in the immediate postoperative period [11].

7.11. Combined liver–kidney transplantation and nephrectomy

Because FH is produced mainly in the liver, a kidney graft does not correct the defect in CFH variants, which motivated combined liver–kidney transplantation [1]. The first two paediatric cases ended in early irreversible liver failure, with the death of one patient and recovery of the other after a second liver transplant, followed by more than 2 years without aHUS; in six patients with CFH variants and one child with CFH and CFI variants, extensive plasma exchange before surgery, to provide normal FH and protect the liver graft, allowed good renal and hepatic function at 2 years [1, 110]. In an infant with a family history of TMA and a heterozygous CFH variant, described in historical nomenclature as S1191L, repeated episodes and renal decline despite prophylactic plasma exchange led, at 4 years of age, to combined (partial) liver and kidney transplantation, with preoperative plasma exchange and enoxaparin, and both grafts maintained good function for approximately 2 years [110]. Pre-emptive plasma therapy and eculizumab should limit the indications for this high-risk procedure [69]; the KDIGO conference allows it in kidney recipients with abnormalities of liver-synthesised proteins, disease uncontrolled despite eculizumab, or for reasons of the long-term cost of eculizumab [13]. Bilateral nephrectomy is a rescue measure in selected patients with extensive microvascular thrombosis on biopsy, refractory hypertension and hypertensive encephalopathy, when conventional therapies, including plasma, fail to control the disease, with excellent results in some cases [1]. Splenectomy induced remission in some plasma-resistant patients, but is ineffective and increases morbidity and mortality in others, and antiplatelet agents, prostacyclin, heparin, fibrinolytics, corticosteroids and IVIg have been tried in plasma resistance or dependence without consistent benefit [1]. The French PNDS considers that combined liver–kidney transplantation is no longer recommended as definitive treatment, because of its morbidity and mortality compared with kidney transplantation with prolonged eculizumab, which diverges from the residual indications allowed by KDIGO [11, 13].

7.12. Pregnancy, breastfeeding and contraception

Pregnancy is a recognised trigger and, in women with known complement dysregulation, GeneReviews gives an approximately 20% pregnancy-associated aHUS risk and recommends close surveillance of every pregnancy [1]. The Brazilian consensus recommends treating pregnancy TMA with a C5 inhibitor (grade 1B), noting that without one 76% reached ESRD despite plasma exchange and that pregnancy may require greater dose or frequency because of increased distribution volume, C5 synthesis and proteinuria; it mentions more than 35 published eculizumab cases in pregnancy or postpartum with approximately 90% haematological response and renal remission [14]. In the adult ravulizumab trial subgroup analysis, 8 patients had pregnancy-triggered aHUS (TMA more than 3 days after delivery), with median age 37.7 years and 5 on dialysis; complete TMA response occurred in 7 of 8 (87.5%) at a median 31.5 days, haematological normalisation in all, and all dialysis patients stopped within 21 days, without deaths or meningococcaemia [67].

Labels differ. For eculizumab, the United States label describes pooled analysis of more than 300 exposed pregnancies with live births without a safety signal, although not excluding risk because of sample size, and contrasts risks of untreated aHUS in pregnancy (pre-eclampsia, preterm delivery, intrauterine growth restriction, fetal death, low birth weight); in mice treated with a murine anti-C5 antibody at 2 to 8 times the human dose, 2 retinal dysplasias and 1 umbilical hernia occurred among 230 pups in the high-dose group, with more moribund or dead males [15]. The European eculizumab SmPC recommends contraception during treatment and for 5 months after the last dose, reports fewer than 300 exposed pregnancy outcomes without increased malformations, notes that IgG crosses the placenta and may inhibit fetal complement, and advises use only if clearly necessary; regarding breastfeeding, limited data suggest no excretion into milk [16]. For ravulizumab, the European SmPC requires effective contraception during and for 8 months after treatment, has no clinical pregnancy data and recommends discontinuing breastfeeding during and for 8 months after treatment, a more restrictive position than eculizumab; the USA has a pregnancy exposure registry [17, 18]. A global pharmacovigilance analysis published on 9 October 2026, with reports between 21 December 2018 and 31 December 2025, identified 354 pregnancies with at least one ravulizumab dose, 72 in aHUS patients; among 106 with known outcomes, there were 68 live births (64.2%), 29 spontaneous abortions (27.4%), 3 fetal deaths (2.8%) and 6 voluntary terminations (5.7%), without unexpected signals; these are spontaneous reports with reporting bias and an unknown denominator, and a dedicated observational study is ongoing [111]. The Brazilian consensus classifies ravulizumab as category C in the Brazilian label and mentions recent efficacy reports in postpartum aHUS [14]. In congenital TTP, ISTH strongly recommends prophylactic plasma infusion during pregnancy in asymptomatic patients [79]. In cblC, OHCbl and betaine continue [37]. GeneReviews advises avoiding pregnancy where possible in women with aHUS [1]; this 2021 wording should be interpreted in light of subsequent effective C5 inhibitor treatment data during pregnancy [14, 67].

The French PNDS, also from 2021, takes a position opposite to that of GeneReviews: women with aHUS in remission, on eculizumab or after its withdrawal, can carry a normal pregnancy to term with regular monitoring; eculizumab can be given in pregnancy, apparently without fetal risk, with CH50 monitoring and frequently increased doses; hypertension, proteinuria and superimposed pre-eclampsia can complicate gestation even under treatment, especially with CKD; and pregnancies have been described in patients on dialysis or with a transplant [11]. In a woman with a pathogenic variant planning pregnancy, long-term anti-C5 therapy is accepted, and the preconception consultation addresses maternal and fetal renal prognosis (both dependent on renal function), close monitoring ideally in a level 3 maternity unit and the risk of transmitting a variant of variable penetrance. In a carrier without a history of HUS, prophylactic eculizumab is not recommended; information on warning signs and regular monitoring of the blood count and haemolysis markers until 4 months after delivery are warranted. The indication for and duration of C5 blockade in pregnancy are always a multidisciplinary decision in a specialist centre [11].

In a pregnant woman with TMA, congenital TTP must be excluded with ADAMTS13 activity and anti-ADAMTS13 antibodies, because it accounts for 24–66% of first episodes of obstetric TTP and requires plasma from conception in subsequent pregnancies; the trimester-based regimen of the French TTP PNDS is given in Table 6.4a [20].

7.13. Agents to avoid, perioperative care and transition

The triggering drugs listed in section 3.2 should be avoided, stopping ciclosporin or tacrolimus if aHUS develops after exposure [1]. FFP is contraindicated in S. pneumoniae-induced HUS [1]. In transplantation, the combination of an mTOR inhibitor with a calcineurin inhibitor and prolonged cold ischaemia are avoided [14]. In a surgical setting or intercurrent illness, anticomplement therapy should not be interrupted, because of the risk of relapse, and antibiotic prophylaxis should be maintained if applicable [13]. Until TTP has been excluded, desmopressin is avoided: the French TTP PNDS contraindicates it in any ADAMTS13 deficiency, hereditary or acquired, because it releases ultra-large von Willebrand factor multimers and reduces ADAMTS13, and can trigger an episode [20]. Transition from paediatric to adult care requires transfer of the vaccination plan, the dosing schedule and the relapse instructions; the Italian consensus identifies paediatric and adult expertise and permanent referral to reference centres among unmet needs [112]. The French PNDS adds psychological support for the child, parents and siblings from the time of disclosure, regular assessment of developmental milestones in infants (speech therapy, psychomotor therapy), adaptation of care to schooling rather than the reverse, consideration of the adolescent's educational and career plans, and therapeutic education programmes that develop knowledge, autonomy and stress management [11].

7.14. Clinical trials and evidence levels

None of these aHUS trials was randomised or placebo-controlled: all are single-arm, with historical controls where available, preventing indirect comparisons between drugs and cohorts. Response proportions are not directly comparable because populations differ in pretreatment TMA duration, dialysis proportion (51.8% in adult study 311) and age; responses were similar in patients with and without an identified variant in eculizumab trials [15, 17]. A 2026 rapid systematic review, searched through 24 April 2026, classified aHUS evidence as non-randomised and low certainty for haematological remission and renal recovery with anti-C5, with no meta-analysis possible [89]. Negative or discontinued trials (ECULISHU in STEC-HUS, narsoplimab, avacopan, cemdisiran) form part of the evidence [2, 90].

7.15. Developments over the past twelve months

The October 2025 to October 2026 window brought no new aHUS authorisations, and developments should be interpreted at their corresponding evidence level. Registry results without a located article or regulatory application: COMMUTE-p published a complete TMA response of 70.6% in the treatment-naïve paediatric cohort on ClinicalTrials.gov on 1 October 2026 [90]. Completed trial without public results: APPELHUS, with iptacopan in adults, ended on 11 April 2026 [90]. Labelling changes: the iptacopan label was updated on 16 July 2026 for IgA nephropathy, still without an aHUS indication in the USA or European Union (European Commission decision of 19 December 2025); United States eculizumab biosimilar labels added NMOSD in June 2026; and Adzynma® received a boxed warning on neutralising antibodies in September 2026 [24, 87, 88, 98, 99]. Pharmacovigilance: analysis of 354 ravulizumab-exposed pregnancies was published on 9 October 2026 [111]. Off-label case reports: a young adult with genetic TMA due to a CFH variant, nephrotic proteinuria and inadequate ravulizumab response (with low free drug concentrations) responded to crovalimab, with immediate haemolysis improvement and renal recovery after 3 months of dialysis [119]; a 21-year-old man with ravulizumab-refractory complement-mediated TMA received compassionate iptacopan, with haematological recovery, AKI resolution and discontinuation of kidney replacement therapy, and relapsed 10 months later alongside a Castleman disease diagnosis, moving to dual therapy with a C5 inhibitor, iptacopan and siltuximab; the authors describe this as the first iptacopan use in aHUS [120]; and a 59-year-old patient with mild SLE and aHUS refractory to immunosuppression, eculizumab and plasma exchange responded rapidly to iptacopan with a rituximab course, subsequently progressing to kidney failure on haemodialysis [121]. Preclinical: an oral factor B inhibitor under chemical development was described in 2026 as entering phase 3 for complement-mediated diseases, without a specific aHUS indication [90]. No aHUS regulatory application for iptacopan or crovalimab was located; a negative search does not demonstrate that such processes do not exist [90]. Pegcetacoplan, a C3 inhibitor, has not been studied in aHUS [14].

7.16. Access and organisation of care

In 2015, NICE recommended funding of eculizumab in aHUS conditional on coordination by a specialist centre, on systems monitoring the number of patients, dose and duration, on a national protocol for initiation and withdrawal for clinical reasons and on a research programme on withdrawal and dose adjustment, recognising a considerable budgetary impact [122]. The Italian consensus, built by Delphi method with scientific societies, patient organisations and regional regulators, achieved agreement of at least 70% on 51 statements on differential diagnosis, rapid recognition, referral to rare disease centres, choice of treatment, relapse prevention and management by a multidisciplinary team [112]. The ISN Forum notes that financial barriers prevent most patients worldwide from accessing diagnosis and C5 inhibitors [2]. In Brazil, eculizumab is approved by the Brazilian Health Regulatory Agency (ANVISA) and ravulizumab is available only in the 300 mg/3 mL presentation [14]. The sources consulted do not document reimbursement in Portugal. In France, the PNDS serves as the reference for the care protocol that the attending physician establishes with the health insurance medical adviser for exemption from co-payment as an off-list long-term condition; treatment is coordinated by a hospital physician linked to a reference or competence centre, the diagnosis is disclosed by a nephrologist from the centre, preferably with a psychologist, and prolonged eculizumab is infused at the prescribing hospital or in hospital-at-home care alternating with the hospital [11]. Therapeutic education programmes are authorised and evaluated by the regional health agency, and school and social support includes the file of the Maison départementale des personnes handicapées (MDPH), school support assistants (AESH), a personalised schooling project or individualised reception plan, ULIS classes, CAMPS before 6 years of age and SESSAD [11].

7.17. Comparative regulatory status

Sources: FDA and EMA submission and label databases accessed on 10 October 2026 [14–18, 24, 87, 88, 96–99, 123]. The French PNDS likewise places the marketing authorisation of Soliris® in aHUS in 2011, the same year as that given by the Brazilian consensus [11, 14].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks by inheritance pattern

In dominant forms (C3, CD46, CFB, CFH including hybrids, CFHR5, CFI, THBD or VTN variants), almost all patients inherited the variant from a usually unaffected heterozygous parent, and de novo variants are rare [1]. Siblings of a patient with a carrier parent have a 50% probability of inheriting the variant but may never develop aHUS, and severity and phenotype differ among carriers of the same variant through triggers and coinherited modifiers; offspring have a 50% risk of inheriting the variant [1]. If neither parent carries the variant, with paternity and maternity confirmed, the possibilities are a de novo variant or parental germline mosaicism, and sibling risk is approximately 1%; leucocyte testing may miss low-level somatic mosaicism and does not detect exclusively germline mosaicism [1]. An apparently negative family history may reflect reduced penetrance, early death or late onset [1].

In DGKE nephropathy and homozygous CFHR3–CFHR1 and CFHR1–CFHR4 deletions, inheritance is recessive: parents are obligate heterozygotes and should be tested; if only one parent is a carrier, a de novo variant, postzygotic parental mosaicism or uniparental isodisomy, described in two probands, is considered [1]. Siblings have a 25% probability of being affected, 50% of being asymptomatic heterozygous carriers and 25% of inheriting no variant; siblings with biallelic DGKE variants typically have aHUS before 1 year [1]. Heterozygotes are generally asymptomatic, with rare aHUS cases in heterozygous adults; offspring of an affected individual are obligate carriers, and parental siblings have a 50% probability of being carriers [1]. In polygenic inheritance, with two variants in different regulatory genes, typically one from each parent (although one parent may carry both and be asymptomatic, so both are tested), 75% of siblings inherit one or two variants, with increased risk, and 25% inherit none; the same applies to offspring [1].

8.2. Penetrance, individual risk and cascade testing

GeneReviews estimates penetrance at 56% for C3, 53% for CD46, 48% for CFH, 50% for CFI and 64% for THBD from Italian cohorts and the international registry, and complete penetrance in families with biallelic DGKE variants [1, 26, 28]. These estimates derive from families identified through an index case and overestimate risk in general-population carriers. The Italian family study of 257 individuals from 71 families (99 patients and 158 healthy carrier relatives) found, over 7,595 person-years, 28 carrier relatives developing aHUS, overall penetrance of 20% and a rate of 3.69 events per 1,000 person-years, higher in siblings (7.47) and offspring (6.29) than parents (2.01), and higher in causal-variant carriers (4.43) than VUS carriers (1.84) [74]. The ISN Forum summarises that approximately 60% of complement-mediated HUS patients have a heterozygous pathogenic variant, approximately 50% of relatives are carriers with penetrance below 20%, higher with C3, CFH and hybrid genes, particularly with risk haplotypes, and very low with CFI; siblings and offspring have greater risk than parents; women have greater risk; and at 30 years remaining risk falls to 30% of birth risk, with estimated penetrance of 6.8% at 35 years in carriers without additional risk haplotypes [2]. The KDIGO conference reports age-related penetrance of an acute episode up to 64% at 70 years [13].

These figures support a cautious approach: GeneReviews recommends offering molecular testing to at-risk relatives of a molecularly diagnosed patient, warning that the variant is predisposing rather than causal, predictions based on a single risk factor are unreliable and individual risk is not quantifiable [1]. The ISN Forum recommends offering genetic testing to relatives and monitoring them after triggers [2]. Identified carriers should have haemoglobin, platelets, creatinine, LDH, C3, C4 and haptoglobin monitored when exposed to triggers (severe infection, inflammation, pregnancy) and avoid precipitants; relatives of patients with typical HUS or without an identified variant require no surveillance [1]. Counselling covers the meaning of a predisposing variant, phenotype unpredictability, early recognition of TMA signs, management of infection or pregnancy and implications for kidney donation.

The French PNDS diverges from this approach: given penetrance below 50% and the psychological (anxiety, guilt, altered self-image) and social (insurance, occupational discrimination) consequences, it does not recommend systematic screening of asymptomatic relatives and reserves testing for the familial variant to relatives with symptoms (renal failure, proteinuria) or in a particular context, such as candidacy as a living donor, always with a genetics clinic consultation [11]. In France, testing of asymptomatic people with a family history is governed by decree no. 2000-570 of 23 June 2000, which requires prescription by a multidisciplinary team with clinical and genetic expertise and a protocol declared to the ministry of health; close family (direct ascendants and descendants) should be informed of the symptoms and risks to allow rapid recognition [11].

8.3. Kidney donation by relatives

Living-related transplantation should be avoided because it may precipitate disease in a healthy donor: in two families, transplantation triggered aHUS in the previously healthy donor, one with a CFH variant, and molecular testing before family donation has since been recommended, with only candidates lacking the familial variant proceeding with assessment [1, 124]. The KDIGO conference allows family donation only if causal genetic or acquired factors are clearly identified in the recipient and absent in the donor, adding that donor CFH or CD46 risk haplotypes do not contraindicate donation and donors with abnormal alternative pathway activity should be excluded [13]. The Brazilian consensus does not recommend living-related donation and, if considered, requires exclusion of donor variants and explicit discussion of post-nephrectomy aHUS risk (grade 1B) [14].

8.4. Family planning, prenatal and preimplantation testing

The optimal time to determine genetic risk and discuss prenatal testing or PGT is before pregnancy, and counselling should be offered to young adults who are affected, carriers or at risk [1]. DNA banking is recommended in probands without a molecular diagnosis [1]. Once the familial variants are identified, prenatal testing and PGT are technically possible, but in dominant forms a positive result indicates predisposition and not disease, which limits their usefulness and makes the decision particularly dependent on the family's values; the perspectives of professionals and families diverge [1]. In DGKE nephropathy, which is fully penetrant, the predictive value of a positive result is high [1]. In the recessive conditions of the differential, congenital TTP and cblC, counselling follows the classical recessive model, with a 25% risk for siblings and the possibility of targeted prenatal testing, and cblC has the particularity of being treatable from birth, which makes prenatal or neonatal diagnosis directly actionable [37, 75]. The French PNDS requires a genetics clinic consultation for any request for prenatal or preimplantation diagnosis in aHUS [11].

8.5. Counselling of children and adolescents

Testing minors is justifiable when there is a clinical action dependent on the result: in families with complement gene variants, knowledge of carrier status guides surveillance in the presence of triggers and the decision to treat a TMA early, and the molecular diagnosis defines the risk of post-transplant recurrence [1, 13]. In carrier girls and young women, information on the risk associated with pregnancy and oral contraception should be given before reproductive age, since pregnancy is one of the most frequent triggers and the risk of pregnancy-associated aHUS is approximately 20% in women with known complement dysregulation [1, 2]. The French PNDS considers, on the contrary, that presymptomatic testing in the asymptomatic minor is not indicated; the divergence sets the usefulness of targeted surveillance and early treatment, invoked by GeneReviews, against the low penetrance and the risk of psychological and social harm invoked by the French protocol [1, 11].

9. Special Considerations: Population Diversity and Epidemiology

9.1. What each figure measures

The sources consulted provide no validated country-specific prevalence estimate for genetic aHUS, and applying gene proportions to overall prevalence would produce unvalidated figures; GeneGuidance explicitly warns that the product of the CFH proportion and European prevalence should not be presented as a validated value [38]. Missing regional data mainly reflect diagnostic and registry capacity rather than absence of disease [2].

9.2. Distribution by region, ancestry and founder effect

The strongest data come from United States national registries and cohorts (400-patient Iowa series, Maga series), Italy (Istituto Mario Negri international registry, 273 and 795 patients), France (214-patient national series, national transplant study with 126 grafts, STOPECU), United Kingdom (Newcastle national service, transplant and discontinuation cohorts), Netherlands (CUREiHUS), German-speaking countries (Neumann registry) and the Global aHUS Registry, with 851 patients characterised before eculizumab and approximately 3,000 planned participants [1, 10, 12, 26, 41, 43, 61, 103, 104, 107, 125]. The proportion with an identified variant varies between 51% and 68% in these series, and approximately 40% remain without an identified cause in the Global Registry [2, 68, 103].

The autoimmune form has an unequal geographical distribution: anti-FH antibody prevalence is higher in India, the Czech Republic, China and Egypt, and in the Indian HUS Registry, anti-FH-positive children constituted 56% of the cohort [2, 54]. Pharmacogenetic variant C5 p.(Arg885His), predicting absence of eculizumab response, has approximately 3% prevalence in Japanese patients and healthy individuals and also occurs in Han Chinese individuals, making it relevant in Asian practice [59]. Differential and allelic diseases have well-defined founder effects: CFHR5 nephropathy, a C3G, results from a founder allele in Greek Cypriots, with screening of the Turkish-Cypriot renal population between 2004 and 2011 finding no new cases; in cblC, MMACHC c.331C>T is a founder in Cajun and French-Canadian populations, c.394C>T is common in patients from India, Pakistan and the Middle East, c.609G>A in those of Chinese ancestry, and c.271dup predominates in southern Europe and French Canadians [37, 47, 77, 85]. Nyhan's Atlas attributes a frequency of 16% in Italian and Portuguese alleles to one of these variants, but the sentence's referent is ambiguous in the source and could not be confirmed in the original article, so the figure is not used [77, 84].

9.3. Portugal and Brazil

The sources consulted contain no aHUS prevalence or incidence estimates for Portugal or Brazil; absence from these sources does not demonstrate absence of published data. Portuguese clinical data nevertheless exist: the 20-year series from a tertiary paediatric nephrology centre, with 5 children with genetic aHUS, 4 with CFH variants, two previously undescribed [71]; the eculizumab-treated neonatal case in section 10 [70]; and the TMA case due to adult-onset methylmalonic aciduria with homocystinuria [82]. In Brazil, the COMDORA-SBN Brazilian aHUS Registry describes predominance of women and young adults, renal involvement in the whole registered population, children with lower haemoglobin and platelets and higher LDH than adults, CFH variants and the large CFHR1–CFHR3 deletion across all ages, 33.5% of tested patients without variants, more CFHR gene variants than other cohorts and no paediatric CFI variants, motivating MLPA recommendation when NGS is negative [14]. The paediatric crovalimab COMMUTE-p trial includes Brazilian centres [90].

9.4. Special populations

In children, the distribution of TMA causes differs from adults (STEC first, followed by aHUS and pneumococcal HUS), CD46 and anti-FH forms are proportionally more frequent, and 1-year mortality exceeded adults in the French series (6.7% versus 0.8%) despite less frequent ESRD after the first episode (16% versus 46%) [12, 14, 61]. In newborns, treatment may occur outside the label indication when weight is below the approved limit [70]. In women of childbearing age, pregnancy and postpartum account for a substantial share of adult-onset episodes, and female sex was a post-eculizumab-withdrawal relapse risk factor [66, 103]. In established ESRD, relapses are usually rare and TMA frequently ceases when ESRD is reached [1, 13]. In nephrotic proteinuria, urinary antibody excretion may compromise complement blockade [2].

9.5. Reference networks, registries and patient organisations

The European reference network for rare kidney diseases is ERKNet, which maintains a working group dedicated to thrombotic microangiopathies, the ERKReg registry and an expert locator [19]. aHUS Alliance Global Action is the global patient organisation, with national affiliates, information materials on the risk of meningococcal infection and patient surveys, including on the renaming of the disease [126]. The Global aHUS Registry is an international observational registry sponsored by the marketing authorisation holder of the C5 inhibitors, with approximately 3,000 planned participants [90, 125]. Italy has the International Registry of Recurrent and Familial HUS/TTP, at the Istituto Mario Negri (Aldo e Cele Daccò Centre, Ranica), and the A.R.M.R. Foundation supports rare disease research; GeneReviews also lists MedlinePlus, the information centre of the National Institute of Diabetes and Digestive and Kidney Diseases, the American Kidney Fund and the National Kidney Foundation [1]. In the United Kingdom, care is concentrated at the National Renal Complement Therapeutics Centre in Newcastle, and TTP is managed at regional centres designated by the National Health Service [80, 108]. Brazil has the COMDORA-SBN Brazilian aHUS Registry [14]. In France, care is organised in four reference centres (MARHEA, Paris-Necker; SORARE, Toulouse; NEPHROGONES, Lyon; CNR-MAT, Paris-Saint-Antoine), with the laboratory associated with the national reference centre for E. coli (Hôpital Robert-Debré) and competence centres, under the ORKiD (rare kidney diseases) and MaRIH (rare immuno-haematological diseases) health networks (filières); alternative pathway investigation is concentrated in reference laboratories in Paris and Bordeaux, and samples for ADAMTS13 and complement are collected before any plasma [11]. The patient associations listed are AIRG-France (with a brochure and an aHUS patient group, co-founder of the aHUS Alliance), France Rein and the MAT-ADAMTS13 association, and the Maladies Rares Info Services telephone service guides patients and professionals [11]. For TTP, the 2022 PNDS was coordinated by CNR-MAT (Hôpital Saint-Antoine, Paris), with the Lariboisière (Paris) and Rouen hospitals, under the MaRIH network and with the support of the French Society of Haematology, the French National Society of Internal Medicine and the Groupe d’Étude Hémostase et Thrombose; CNR-MAT has an emergency telephone line, regional constituent and competence centres and the national ADAMTS13 laboratory at Hôpital Lariboisière [20]. The patient association listed is the Association ADAMTS13, and the PNDS also mentions the Canadian association Answering TTP; long-term condition status (ALD) can be granted for renewable 5-year periods, given the follow-up of at least 10 years [20].

10. Illustrative Clinical Case

10.1. aHUS due to a homozygous CD46 variant, with pregnancy relapse treated with ravulizumab

The following case was published in BMC Nephrology in 2026 by a Saudi Arabian team and is faithfully summarised with the chronology of article figure 1, consulted on 10 October 2026 [127].

A 27-year-old woman presented in December 2018, six months after an uncomplicated vaginal delivery, with two weeks of diarrhoea, nausea, vomiting and anorexia. She was dehydrated, with stable vital signs and normal blood pressure. Laboratory tests showed haemoglobin 90 g/L, platelets progressively falling from 90 to 40 and then 25×10⁹/L, creatinine 560 µmol/L, urea 16 mmol/L, albumin 22 g/L, pH 7.2 with bicarbonate 19 mmol/L, C-reactive protein 248 mg/L, LDH 1,376 U/L and haptoglobin below 0.058 g/L. C3 was low (0.62 g/L) and C4 normal (0.10 g/L); ADAMTS13 activity was normal (1.15 IU/mL, range 0.40–1.30); antinuclear and antineutrophil cytoplasmic antibodies were negative; factor H (548 mg/L; normal 242–759) and factor I (52.2 mg/L; normal 39–100) were normal. Renal ultrasound was normal and biopsy showed TMA without nephritis.

After TTP was excluded, she underwent ten plasma exchange sessions with full FFP replacement, transfusions and haemodialysis through a right jugular tunnelled catheter, and required mechanical ventilation for one day for severe pulmonary oedema attributed to volume overload, hypertension, transfusions and plasma exchange. Eculizumab was initiated in December 2018 after meningococcal vaccination and antibiotic prophylaxis, at 900 mg every two weeks for the first two doses and 1,200 mg every two weeks thereafter; this differs from labelled weekly induction and should not be taken as routine practice [15, 127]. Haemodialysis stopped after four weeks and renal function normalised at six months.

Exome sequencing identified a homozygous likely pathogenic CD46 variant, described in the article as NM_172359.2:c.350_351dup, p.(Glu118Thrfs*17) and confirmed by Sanger sequencing. Normalised to MANE Select, it corresponds to NM_172351.3:c.350_351dup, NP_758861.1:p.(Glu118ThrfsTer17), with genomic coordinate NC_000001.11:g.207757603_207757604dup; ClinVar record VCV 1333461 classifies it as pathogenic or likely pathogenic, with multiple submitters without conflict and latest assessment on 2 October 2025 [57, 58].

Eculizumab continued until 2021, with interruptions due to supply shortages and transport difficulties during the COVID-19 pandemic, and was then stopped with clinical surveillance. In 2022, she became pregnant unintentionally despite prior counselling on relapse risk and was monitored every three to four weeks (renal function, blood pressure, full blood count, LDH, platelets, proteinuria and C3), without prophylaxis because eculizumab was unavailable. At 25 weeks she relapsed: blood pressure 160–170/90–100 mmHg, vomiting, diarrhoea, haemoglobin 86 g/L, platelets 129×10⁹/L, creatinine 143 µmol/L, urea 6.1 mmol/L, LDH 364 U/L, haptoglobin below 0.06 g/L and C3 1.02 g/L, with negative STEC testing. With informed consent and antibiotic prophylaxis, ravulizumab was initiated at labelled doses for 60 to under 100 kg: loading 2,700 mg on day 1, 3,300 mg on day 14 and then 3,300 mg every eight weeks [17, 127]. She recovered (creatinine 120 µmol/L, LDH 169 U/L, platelets 292×10⁹/L, C3 1.24 g/L), had a spontaneous term vaginal delivery without maternal or neonatal complications and, in 2026, had remained on regular ravulizumab for approximately four years, with a healthy child.

Article figure 1, a six-panel timeline consulted on 10 October 2026 (https://pmc.ncbi.nlm.nih.gov/articles/PMC13202841/figure/Fig1/), shows the sequence: initial 2018 diagnosis with TMA and AKI, haemodialysis, plasma exchange and eculizumab initiation; recovery in 2019 with dialysis withdrawal; supply interruptions and drug withdrawal in 2020–2021, with stability under surveillance; unplanned pregnancy presentation in 2022 with monitoring without early relapse; relapse at 25 weeks treated with ravulizumab; and term delivery followed by more than three years of stable follow-up [127].

Two aspects merit comment. First, the CD46 variant, associated in heterozygosity with the mildest phenotype and best renal survival, occurred here homozygously with severe presentation requiring dialysis, reminding us that gene correlations describe cohort trends rather than individual courses [1, 61]. Second, inhibitor withdrawal and surveillance alone in a woman of childbearing age were followed by pregnancy relapse; the authors argue that withdrawal may be impractical in this group and surveillance alone should not be considered optimal, consistent with data identifying female sex and a rare variant as relapse risk factors [103, 127]. The authors also state that eculizumab remains preferred in pregnancy because of limited ravulizumab data, while the European ravulizumab SmPC has no clinical pregnancy data [18, 127].

10.2. Additional case: neonatal aHUS in Portugal without an identified variant

This second case, published by a Portuguese team, illustrates neonatal presentation and treatment access difficulties at that age [70]. A newborn girl from a 36-week monochorionic diamniotic twin pregnancy, born to non-consanguineous parents, was delivered by emergency caesarean for cord prolapse and breech presentation, weighing 2,035 g with Apgar scores 7/7/7. Within the first hour she had pallor, reduced responsiveness and grunting, with metabolic acidosis (pH 7.14; bicarbonate 15 mmol/L; lactate 12 mmol/L) and hypotension. On day one, haemoglobin was 8.8 g/dL, platelets 33×10⁹/L, aspartate aminotransferase 327 and alanine aminotransferase 81 U/L, urea 63 mg/dL and creatinine 1.61 mg/dL; she received antibiotics for suspected sepsis, with negative cultures, and transfusions. The course showed LDH 3,387 U/L, undetectable haptoglobin, negative direct Coombs, 2–4 schistocytes per field, creatinine 4.59 mg/dL, haematuria and proteinuria (urinary protein/creatinine ratio 33 mg/mg), with normal coagulation. Aetiological investigation was negative: slightly low C3c (73 mg/dL; normal 83–177), negative Shiga toxin, ADAMTS13 activity 69%, normal metabolic screening (homocysteine, amino acids and organic acids) and genetic analysis without variants in CFH, CFI, CD46, C3, CFB, THBD, CFHR1, CFHR3, CFHR4, CFHR5 and DGKE, no large CFH, CFI and CD46 deletions or duplications and no homozygous CFHR3–CFHR1 deletion.

FFP at 10 mL/kg/day was initiated on day five because plasma exchange was unfeasible at 2 kg, and eculizumab on day fifteen, delayed ten days pending ethics committee authorisation because there was then no dose recommendation below 1 month and 5 kg: 300 mg, repeated the following week and then every three weeks. LDH normalised within seven days, creatinine fell more than 25% within seven days and normalised after 13 weeks, anaemia resolved within ten weeks, eGFR increased 61 mL/min/1.73 m² by week 27 and proteinuria and haematuria disappeared within 90 days. Prophylaxis included intravenous ampicillin followed by oral amoxicillin and authorised off-label vaccination against meningococcus B on day twenty and at 2, 4 and 6 months, boosted at 13 months, and MenACWY on day twenty and at 2 and 4 months, boosted at 13 months, with close contacts also vaccinated. At 14 months there was transient renal deterioration, with creatinine 0.45 mg/dL, cystatin C 1.8 mg/L and eGFR 69 mL/min/1.73 m² by the Schwartz formula, attributed to probable viral illness and recovering within three weeks. At 20 months, during eculizumab, full blood count and renal function were normal, with growth between the 15th and 50th centiles and appropriate development.

Article figure 1, consulted on 10 October 2026 (https://pmc.ncbi.nlm.nih.gov/articles/PMC8221835/figure/FI210009-1/), shows haemoglobin, platelet and creatinine evolution from day one to approximately day 370: creatinine rises to approximately 4.6 mg/dL between days six and nine, falls partially with plasma and normalises below 0.5 mg/dL after eculizumab; platelets recover with plasma, peaking near 900×10³/µL around day 28; haemoglobin stabilises above 13 g/dL after approximately 13 weeks of eculizumab; and the graph marks transfusions, daily plasma infusion, eculizumab regimen and antibiotic prophylaxis [70]. The case confirms that absence of an identified variant does not exclude aHUS, treatment in a very-low-weight newborn may require off-label decisions, and the ten-day administratively imposed delay in inhibitor initiation contrasts with the recommendation to begin as early as possible [2, 70].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis before and after C5 inhibitors

In the plasma therapy era, approximately 60% of genetic aHUS patients progressed to ESRD, with marked variation by gene: 60%–80% ESRD or death with CFH, more than 60% with C3, 70% with CFB, 58% with CFI and more than 50% with THBD, whereas 60%–70% of CD46 patients remained dialysis-free even after multiple relapses [1, 26]. In the French series, first-episode ESRD affected 46% of adults and 16% of children, and 1-year mortality was 0.8% in adults and 6.7% in children [12]. In pregnancy-associated aHUS, 62% had ESRD after 1 month and 76% at last follow-up, and in postpartum aHUS ESRD or death occurred in 50%–60% [1, 65]. These values describe disease treated with plasma and support rather than untreated natural history.

Comparison with genotype-matched controls in the United Kingdom national cohort quantifies C5 blockade's effect: cumulative 5-year ESRD-free survival was 85.5% with eculizumab versus 39.5% in controls, with a hazard ratio of 4.95 (95% CI 2.75–8.90) and number needed to treat 2.17 (1.81–2.73) to prevent one ESRD case; the ISN Forum summary table gives 78% versus 39.5% for all complement-mediated HUS and a hazard ratio of 2.92 (2.09–4.11), values from different analyses of the same cohort [2, 10]. Even with eculizumab, 20% do not recover renal function and 30% have stage 3 to 5 CKD at 6 months, and prognosis is worse with first presentation in adulthood; predictors of eGFR above 60 mL/min/1.73 m² at 6 months were lower creatinine, lower platelets and lower systolic blood pressure at presentation, younger age and shorter time to first dose [2, 10]. Time dependence is consistent with label data: in the retrospective paediatric study, complete response was 70% when TMA lasted less than 2 months and zero beyond that period because delay reflects irreversible renal damage [16]. Renal recovery is concentrated within the first 30 to 90 days [2].

The French PNDS draws attention to long-term neurological sequelae in children, described more frequently after aHUS than after STEC-HUS, although with very variable frequencies between studies: delayed psychomotor and language development, neurocognitive or neuropsychological difficulties (school failure, aggressiveness, depression), tremor, microcephaly, hemiparesis, hydrocephalus and visual disturbances (strabismus, blindness), sometimes in children without neurological involvement in the acute phase [11].

11.2. Relapses

Relapses are frequent even after complete recovery from the first episode and depend on genotype [1]. In the French series, 92% of children with CD46 variant-associated HUS relapsed after the first year versus approximately 30% in other subgroups [12]. After inhibitor withdrawal, relapse rates ranged from 22% to 23% in prospective and observational studies and reached 58% in a Brazilian unplanned-withdrawal cohort, with risk concentrated in pathogenic-variant carriers (section 7.8) [14, 103, 105]. Patients with established ESRD do not usually relapse [1].

11.3. Surveillance

aHUS surveillance intervals originate as GeneReviews authors' opinion rather than trial results, and the ISN Forum observes that most relapses were identified by spontaneous patient presentation rather than scheduled monitoring, shifting the weight of surveillance towards patient education and rapid treatment access [1, 2]. Monitoring may be insufficient to predict or prevent severe post-withdrawal TMA complications [16].

12. Summary and Final Considerations

Atypical haemolytic uraemic syndrome is predominantly renal thrombotic microangiopathy caused in most cases by inability to contain alternative complement pathway amplification at the endothelial surface. Approximately 60% of patients have an identifiable genetic basis, mainly heterozygous loss-of-function CFH, CFI and CD46 variants, gain-of-function C3 and CFB variants, or CFH–CFHR cluster rearrangements, including the CFH/CFHR1 hybrid, undetected by conventional sequencing and requiring MLPA or another copy number technique. Approximately 40% have no identified variant, excluding neither diagnosis nor treatment indication. The autoimmune form, due to anti-factor H antibodies associated with homozygous CFHR3–CFHR1 deletion, is predominantly paediatric and has its own treatment, with plasma exchange and immunosuppression in addition to C5 blockade. DGKE deficiency nephropathy, recessive with first-year onset, has a non-complement mechanism and does not respond to C5 inhibitors, a distinction with direct therapeutic consequences. The THBD association, accepted in OMIM, was classified as disputed by ClinGen in 2024, and a rare variant in this gene should not be taken as the disease explanation.

Penetrance is incomplete and an episode depends on a trigger, chiefly infection and pregnancy. A variant found in a relative therefore indicates predisposition rather than disease: observed penetrance in carrier relatives was approximately 20%, higher in siblings and offspring than parents, and individual risk is not quantifiable. This underpins counselling, combining an offer of cascade testing with avoidance of individual prognostication, surveillance directed towards trigger exposure periods and exclusion of living-related donors carrying the familial variant, after two cases in which donation precipitated disease in the donor.

Diagnosis is clinical and by exclusion, and its urgency is therapeutic. ADAMTS13 activity should be known before C5 inhibition in adults but is not a prerequisite in children; STEC testing is performed in all because diarrhoea does not exclude aHUS and approximately 5% of STEC-HUS cases lack it; homocysteine, methylmalonic acid and methionine are mandatory in paediatrics because cblC is treatable with parenteral hydroxocobalamin and delay may be fatal. Complement biomarkers support but do not decide: C3 is normal in a substantial proportion and also falls in STEC-HUS and pneumococcal HUS, sC5b-9 may remain abnormal in remission, and no non-molecular test confirms genetic diagnosis.

Treatment changed prognosis. With plasma therapy, approximately 60% reached end-stage kidney disease; with C5 blockade, 5-year ESRD-free survival rose from 39.5% to 78%–85.5% in a national cohort with genotype-matched controls, with a number needed to treat of approximately 2. Benefit is time-dependent: renal recovery is concentrated in the first 30 to 90 days, and weeks of delay reflect irreversible damage. Eculizumab and ravulizumab, plus eculizumab biosimilars, have approved indications, with well-defined weight-based dosing and practical jurisdictional differences such as the 1-month and 5 kg limit in the United States versus 10 kg in the European Union for ravulizumab. Iptacopan and crovalimab have no aHUS indication in either jurisdiction: APPELHUS completed without published results, COMMUTE-p published paediatric results only in the registry, and described use in refractory disease is off-label. Labelling requires meningococcal vaccination and risk mitigation programmes; guidelines go further by recommending antibiotics throughout treatment, a divergence that should be explained to patients given the approximately 550-fold increase in meningococcal infection risk.

Two decisions remain without definitive answers. First is treatment duration: withdrawal studies, all small and non-randomised, show relapse in approximately one-quarter of patients, concentrated in pathogenic-variant carriers and virtually absent without a rare variant, and labels retain conservative positions, with surveillance for 12 weeks to 12 months after withdrawal. Planned withdrawal is defensible with genetic characterisation, blockade monitoring and immediate access to reintroduction, and discouraged in transplant recipients and women planning pregnancy. Second is transplant management, where genotype-based risk stratification has consensus on structure but differing details between KDIGO, ISN Forum and Brazilian consensus, and prophylaxis and early rescue have not been compared randomly. In both cases, the practical determinant is organisational: reference centres, timely accessible genetics, patients informed of relapse signs and immediate drug availability. Funding barriers preventing most patients worldwide from accessing diagnosis and C5 inhibitors are now the main obstacle to outcomes shown possible by trials and cohorts.

13. Bibliographic References

1. Noris M, Bresin E, Mele C, Remuzzi G. Genetic Atypical Hemolytic-Uremic Syndrome. 2007 Nov 16 [updated 2021 Sep 23]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1367/
2. Kavanagh D, Ardissino G, Brocklebank V, Bouwmeester RN, Bagga A, Ter Heine R, et al. Outcomes from the International Society of Nephrology Hemolytic Uremic Syndromes International Forum. Kidney Int. 2024;106(6):1038-50. doi:10.1016/j.kint.2024.09.012
3. Nester CM, Feldman DL, Burwick R, Cataland S, Chaturvedi S, Cook HT, et al. An expert discussion on the atypical hemolytic uremic syndrome nomenclature—identifying a road map to precision: a report of a National Kidney Foundation Working Group. Kidney Int. 2024;106(3):326-36. doi:10.1016/j.kint.2024.05.021
4. Orphanet. Atypical hemolytic uremic syndrome (ORPHA:2134). Available from: https://www.orpha.net/en/disease/detail/2134
5. Orphanet. Atypical hemolytic uremic syndrome with complement gene abnormality (ORPHA:544472). Available from: https://www.orpha.net/en/disease/detail/544472
6. Orphanet. Atypical hemolytic uremic syndrome with anti-factor H antibodies (ORPHA:93581). Available from: https://www.orpha.net/en/disease/detail/93581
7. Orphanet. Hemolytic uremic syndrome with DGKE deficiency (ORPHA:357008). Available from: https://www.orpha.net/en/disease/detail/357008
8. GeneGuidance. DGKE — clinical gene curation (version of 26 September 2026).
9. Clinical Genome Resource. Gene-disease validity curations (download file created 2026-10-10). Available from: https://search.clinicalgenome.org/kb/gene-validity
10. Brocklebank V, Walsh PR, Smith-Jackson K, Hallam TM, Marchbank KJ, Wilson V, et al. Atypical hemolytic uremic syndrome in the era of terminal complement inhibition: an observational cohort study. Blood. 2023;142(16):1371-86. doi:10.1182/blood.2022018833
11. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Le syndrome hémolytique et urémique (SHU). Centres de référence SORARE, NEPHROGONES, MARHEA et CNR des microangiopathies thrombotiques; filières ORKiD et MaRIH. Saint-Denis La Plaine: HAS; February 2021 (published 16 February 2021). Available from: https://www.has-sante.fr/jcms/p_3689435
12. Frémeaux-Bacchi V, Fakhouri F, Garnier A, Bienaimé F, Dragon-Durey MA, Ngo S, et al. Genetics and outcome of atypical hemolytic uremic syndrome: a nationwide French series comparing children and adults. Clin J Am Soc Nephrol. 2013;8(4):554-62. doi:10.2215/CJN.04760512
13. Goodship TH, Cook HT, Fakhouri F, Fervenza FC, Frémeaux-Bacchi V, Kavanagh D, et al. Atypical hemolytic uremic syndrome and C3 glomerulopathy: conclusions from a "Kidney Disease: Improving Global Outcomes" (KDIGO) Controversies Conference. Kidney Int. 2017;91(3):539-51. doi:10.1016/j.kint.2016.10.005
14. Vaisbich MH, Andrade LGM, Barbosa MINH, Castro MCR, Miranda SMC, Poli-de-Figueiredo CE, et al. Recommendations for diagnosis and treatment of Atypical Hemolytic Uremic Syndrome (aHUS): an expert consensus statement from the Rare Diseases Committee of the Brazilian Society of Nephrology (COMDORA-SBN). J Bras Nefrol. 2025;47(2):e20240087. doi:10.1590/2175-8239-JBN-2024-0087en
15. Alexion Pharmaceuticals. SOLIRIS (eculizumab) injection, for intravenous use. Prescribing information. BLA 125166. Revised 2026 (SPL version 44, effective 2026-07-08). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=ebcd67fa-b4d1-4a22-b33d-ee8bf6b9c722
16. European Medicines Agency. Soliris (eculizumab): EPAR — product information (summary of product characteristics). Amsterdam: EMA; last updated 2026-02-19. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/soliris
17. Alexion Pharmaceuticals. ULTOMIRIS (ravulizumab-cwvz) injection, for intravenous use. Prescribing information. BLA 761108. Revised 2025 (SPL version 19, effective 2025-10-09). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=a9a590d9-0217-43c7-908d-e62a71279791
18. European Medicines Agency. Ultomiris (ravulizumab): EPAR — product information. Amsterdam: EMA; last updated 2026-07-21. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/ultomiris
19. European Rare Kidney Disease Reference Network (ERKNet). Available from: https://www.erknet.org/
20. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Purpura thrombotique thrombocytopénique. Centre de référence des microangiopathies thrombotiques (CNR-MAT), filière MaRIH. Saint-Denis La Plaine: HAS; 7 October 2022. Available from: https://www.has-sante.fr/jcms/p_3375723
21. Orphanet. Congenital thrombotic thrombocytopenic purpura (ORPHA:93583). Available from: https://www.orpha.net/en/disease/detail/93583
22. Orphanet. Methylmalonic acidemia with homocystinuria, type cblC (ORPHA:79282). Available from: https://www.orpha.net/en/disease/detail/79282
23. Martín B, Smith RJH. C3 Glomerulopathy. 2007 Jul 20 [updated 2018 Apr 5]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1425/
24. Takeda Pharmaceuticals U.S.A. ADZYNMA (ADAMTS13, recombinant-krhn) for injection. Prescribing information. STN 125795. Revised 9/2026. Available from: https://www.fda.gov/vaccines-blood-biologics/adzynma
25. European Medicines Agency. Adzynma (rADAMTS13): EPAR — product information (summary of product characteristics). EMEA/H/C/006198. Amsterdam: EMA; first authorised 2024-08-01; page updated 2026-10-02. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/adzynma
26. Noris M, Caprioli J, Bresin E, Mossali C, Pianetti G, Gamba S, et al. Relative role of genetic complement abnormalities in sporadic and familial aHUS and their impact on clinical phenotype. Clin J Am Soc Nephrol. 2010;5(10):1844-59. doi:10.2215/CJN.02210310
27. Warwicker P, Goodship TH, Donne RL, Pirson Y, Nicholls A, Ward RM, et al. Genetic studies into inherited and sporadic hemolytic uremic syndrome. Kidney Int. 1998;53(4):836-44. doi:10.1111/j.1523-1755.1998.00824.x
28. Caprioli J, Noris M, Brioschi S, Pianetti G, Castelletti F, Bettinaglio P, et al. Genetics of HUS: the impact of MCP, CFH, and IF mutations on clinical presentation, response to treatment, and outcome. Blood. 2006;108(4):1267-79. doi:10.1182/blood-2005-10-007252
29. Dragon-Durey MA, Loirat C, Cloarec S, Macher MA, Blouin J, Nivet H, et al. Anti-factor H autoantibodies associated with atypical hemolytic uremic syndrome. J Am Soc Nephrol. 2005;16(2):555-63. doi:10.1681/ASN.2004050380
30. Venables JP, Strain L, Routledge D, Bourn D, Powell HM, Warwicker P, et al. Atypical haemolytic uraemic syndrome associated with a hybrid complement gene. PLoS Med. 2006;3(10):e431. doi:10.1371/journal.pmed.0030431
31. Zipfel PF, Edey M, Heinen S, Józsi M, Richter H, Misselwitz J, et al. Deletion of complement factor H-related genes CFHR1 and CFHR3 is associated with atypical hemolytic uremic syndrome. PLoS Genet. 2007;3(3):e41. doi:10.1371/journal.pgen.0030041
32. Józsi M, Licht C, Strobel S, Zipfel SL, Richter H, Heinen S, et al. Factor H autoantibodies in atypical hemolytic uremic syndrome correlate with CFHR1/CFHR3 deficiency. Blood. 2008;111(3):1512-4. doi:10.1182/blood-2007-09-109876
33. Delvaeye M, Noris M, De Vriese A, Esmon CT, Esmon NL, Ferrell G, et al. Thrombomodulin mutations in atypical hemolytic-uremic syndrome. N Engl J Med. 2009;361(4):345-57. doi:10.1056/NEJMoa0810739
34. Lemaire M, Frémeaux-Bacchi V, Schaefer F, Choi M, Tang WH, Le Quintrec M, et al. Recessive mutations in DGKE cause atypical hemolytic-uremic syndrome. Nat Genet. 2013;45(5):531-6. doi:10.1038/ng.2590
35. Legendre CM, Licht C, Muus P, Greenbaum LA, Babu S, Bedrosian C, et al. Terminal complement inhibitor eculizumab in atypical hemolytic-uremic syndrome. N Engl J Med. 2013;368(23):2169-81. doi:10.1056/NEJMoa1208981
36. GeneGuidance. THBD — clinical gene curation (version of 26 September 2026).
37. Sloan JL, Carrillo N, Adams D, Venditti CP. Disorders of Intracellular Cobalamin Metabolism. 2008 Feb 25 [updated 2018 Sep 6; revised 2021 Dec 16]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1328/
38. GeneGuidance. CFH — clinical gene curation (version of 26 September 2026).
39. GeneGuidance. CFI — clinical gene curation (version of 26 September 2026).
40. GeneGuidance. CD46 — clinical gene curation (version of 26 September 2026).
41. Bresin E, Rurali E, Caprioli J, Sanchez-Corral P, Fremeaux-Bacchi V, Rodriguez de Cordoba S, et al. Combined complement gene mutations in atypical hemolytic uremic syndrome influence clinical phenotype. J Am Soc Nephrol. 2013;24(3):475-86. doi:10.1681/ASN.2012090884
42. Nester CM, Barbour T, Rodriguez de Cordoba S, Dragon-Durey MA, Fremeaux-Bacchi V, Goodship TH, et al. Atypical aHUS: state of the art. Mol Immunol. 2015;67(1):31-42. doi:10.1016/j.molimm.2015.03.246
43. Bu F, Zhang Y, Wang K, Borsa NG, Jones MB, Taylor AO, et al. Genetic analysis of 400 patients refines understanding and implicates a new gene in atypical hemolytic uremic syndrome. J Am Soc Nephrol. 2018;29(12):2809-19. doi:10.1681/ASN.2018070759
44. GeneGuidance. C3 — clinical gene curation (version of 26 September 2026).
45. GeneGuidance. CFB — clinical gene curation (version of 26 September 2026).
46. GeneGuidance. CFHR1 — clinical gene curation (version of 26 September 2026).
47. GeneGuidance. CFHR5 — clinical gene curation (version of 26 September 2026).
48. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 35, section 35.1.5 Diacylglycerol Kinase Epsilon (DGKE) Deficiency.
49. Francis NJ, McNicholas B, Awan A, Waldron M, Reddan D, Sadlier D, et al. A novel hybrid CFH/CFHR3 gene generated by a microhomology-mediated deletion in familial atypical hemolytic uremic syndrome. Blood. 2012;119(2):591-601. doi:10.1182/blood-2011-03-339903
50. Piras R, Valoti E, Alberti M, Bresin E, Mele C, Breno M, et al. CFH and CFHR structural variants in atypical hemolytic uremic syndrome: prevalence, genomic characterization and impact on outcome. Front Immunol. 2023;13:1011580. doi:10.3389/fimmu.2022.1011580
51. Tschernoster N, Erger F, Walsh PR, McNicholas B, Fistrek M, Habbig S, et al. Unraveling structural rearrangements of the CFH gene cluster in atypical hemolytic uremic syndrome patients using molecular combing and long-fragment targeted sequencing. J Mol Diagn. 2022;24(6):619-31. doi:10.1016/j.jmoldx.2022.02.006
52. Hughes AE, Orr N, Esfandiary H, Diaz-Torres M, Goodship T, Chakravarthy U. A common CFH haplotype, with deletion of CFHR1 and CFHR3, is associated with lower risk of age-related macular degeneration. Nat Genet. 2006;38(10):1173-7. doi:10.1038/ng1890
53. GeneGuidance. CFHR3 — clinical gene curation (version of 26 September 2026).
54. Sinha A, Gulati A, Saini S, Blanc C, Gupta A, Gurjar BS, et al. Prompt plasma exchanges and immunosuppressive treatment improves the outcomes of anti-factor H autoantibody-associated hemolytic uremic syndrome in children. Kidney Int. 2014;85(5):1151-60. doi:10.1038/ki.2013.373
55. Cugno M, Berra S, Depetri F, Tedeschi S, Griffini S, Grovetti E, et al. IgM autoantibodies to complement factor H in atypical hemolytic uremic syndrome. J Am Soc Nephrol. 2021;32(5):1227-35. doi:10.1681/ASN.2020081224
56. Martínez-Barricarte R, Pianetti G, Gautard R, Misselwitz J, Strain L, Frémeaux-Bacchi V, et al. The complement factor H R1210C mutation is associated with atypical hemolytic uremic syndrome. J Am Soc Nephrol. 2008;19(3):639-46. doi:10.1681/ASN.2007080923
57. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-8. doi:10.1002/humu.23348
58. National Center for Biotechnology Information. ClinVar (VCV000127071, VCV000016558, VCV001333461). Bethesda (MD): NLM. Available from: https://www.ncbi.nlm.nih.gov/clinvar/
59. Nishimura J, Yamamoto M, Hayashi S, Ohyashiki K, Ando K, Brodsky AL, et al. Genetic variants in C5 and poor response to eculizumab. N Engl J Med. 2014;370(7):632-9. doi:10.1056/NEJMoa1311084
60. Ardissino G, Bernardi S, Dato L, Mancuso MC, Rossetti D, Tamburini G, et al. Typical or atypical hemolytic uremic syndrome? That is the question. Pediatr Nephrol. 2026 Jul 25. doi:10.1007/s00467-026-07473-6. Online ahead of print
61. Schaefer F, Ardissino G, Ariceta G, Fakhouri F, Scully M, Isbel N, et al. Clinical and genetic predictors of atypical hemolytic uremic syndrome phenotype and outcome. Kidney Int. 2018;94(2):408-18. doi:10.1016/j.kint.2018.02.029
62. Copelovitch L, Kaplan BS. Streptococcus pneumoniae-associated hemolytic uremic syndrome. Pediatr Nephrol. 2008;23(11):1951-6. doi:10.1007/s00467-007-0518-y
63. Brocklebank V, Kumar G, Howie AJ, Chandar J, Milford DV, Craze J, et al. Long-term outcomes and response to treatment in diacylglycerol kinase epsilon nephropathy. Kidney Int. 2020;97(6):1260-74. doi:10.1016/j.kint.2020.01.045
64. Azukaitis K, Simkova E, Majid MA, Galiano M, Benz K, Amann K, et al. The phenotypic spectrum of nephropathies associated with mutations in diacylglycerol kinase epsilon. J Am Soc Nephrol. 2017;28(10):3066-75. doi:10.1681/ASN.2017010031
65. Fakhouri F, Roumenina L, Provot F, Sallée M, Caillard S, Couzi L, et al. Pregnancy-associated hemolytic uremic syndrome revisited in the era of complement gene mutations. J Am Soc Nephrol. 2010;21(5):859-67. doi:10.1681/ASN.2009070706
66. Bruel A, Kavanagh D, Noris M, Delmas Y, Wong EKS, Bresin E, et al. Hemolytic uremic syndrome in pregnancy and postpartum. Clin J Am Soc Nephrol. 2017;12(8):1237-47. doi:10.2215/CJN.00280117
67. Gäckler A, Schönermarck U, Dobronravov V, La Manna G, Denker A, Liu P, et al. Efficacy and safety of the long-acting C5 inhibitor ravulizumab in patients with atypical hemolytic uremic syndrome triggered by pregnancy: a subgroup analysis. BMC Nephrol. 2021;22(1):5. doi:10.1186/s12882-020-02190-0
68. Le Quintrec M, Zuber J, Moulin B, Kamar N, Jablonski M, Lionet A, et al. Complement genes strongly predict recurrence and graft outcome in adult renal transplant recipients with atypical hemolytic and uremic syndrome. Am J Transplant. 2013;13(3):663-75. doi:10.1111/ajt.12077
69. Zuber J, Le Quintrec M, Sberro-Soussan R, Loirat C, Frémeaux-Bacchi V, Legendre C. New insights into postrenal transplant hemolytic uremic syndrome. Nat Rev Nephrol. 2011;7(1):23-35. doi:10.1038/nrneph.2010.155
70. Gomes SM, Teixeira RP, Rocha G, Soares P, Guimaraes H, Santos P, et al. Neonatal atypical hemolytic uremic syndrome in the eculizumab era. AJP Rep. 2021;11(2):e95-e98. doi:10.1055/s-0041-1731057
71. Maximiano C, Silva A, Duro I, Branco T, Correia-Costa L, Teixeira A, et al. Genetic atypical hemolytic uremic syndrome in children: a 20-year experience from a tertiary center. J Bras Nefrol. 2021;43(3):311-7. doi:10.1590/2175-8239-JBN-2020-0199
72. ANSM. CEFOTAXIME NORIDEM 2 g, poudre pour solution injectable/pour perfusion — Résumé des caractéristiques du produit (CIS 60386439), base mise à jour le 03/11/2026. Base de données publique des médicaments. Available from: https://base-donnees-publique.medicaments.gouv.fr/medicament/60386439/extrait
73. Huemer M, Diodato D, Schwahn B, Schiff M, Bandeira A, Benoist JF, et al. Guidelines for diagnosis and management of the cobalamin-related remethylation disorders cblC, cblD, cblE, cblF, cblG, cblJ and MTHFR deficiency. J Inherit Metab Dis. 2017;40(1):21-48. doi:10.1007/s10545-016-9991-4
74. Ardissino G, Longhi S, Porcaro L, Pintarelli G, Strumbo B, Capone V, et al. Risk of atypical HUS among family members of patients carrying complement regulatory gene abnormality. Kidney Int Rep. 2021;6(6):1614-21. doi:10.1016/j.ekir.2021.03.885
75. GeneGuidance. ADAMTS13 — clinical gene curation (version of 26 September 2026).
76. Zheng XL, Vesely SK, Cataland SR, Coppo P, Geldziler B, Iorio A, et al. ISTH guidelines for the diagnosis of thrombotic thrombocytopenic purpura. J Thromb Haemost. 2020;18(10):2486-95. doi:10.1111/jth.15006
77. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. Chapter 4, Cobalamin C, D, F, G diseases; methylmalonic aciduria and variable homocystinuria; p. 34–41.
78. Zheng XL, Al-Housni Z, Cataland SR, Coppo P, Geldziler B, Germini F, et al. 2025 focused update of the 2020 ISTH guidelines for management of thrombotic thrombocytopenic purpura. J Thromb Haemost. 2025;23(11):3711-32. doi:10.1016/j.jtha.2025.06.002
79. Zheng XL, Vesely SK, Cataland SR, Coppo P, Geldziler B, Iorio A, et al. ISTH guidelines for treatment of thrombotic thrombocytopenic purpura. J Thromb Haemost. 2020;18(10):2496-502. doi:10.1111/jth.15010
80. Scully M, Rayment R, Clark A, Westwood JP, Cranfield T, Gooding R, et al. A British Society for Haematology guideline: diagnosis and management of thrombotic thrombocytopenic purpura and thrombotic microangiopathies. Br J Haematol. 2023;203(4):546-63. doi:10.1111/bjh.19026
81. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 28, section 28.2 Disorders of Intracellular Utilisation of Cobalamin.
82. Navarro D, Azevedo A, Sequeira S, Ferreira AC, Carvalho F, Fidalgo T, et al. Atypical adult-onset methylmalonic acidemia and homocystinuria presenting as hemolytic uremic syndrome. CEN Case Rep. 2018;7(1):73-6. doi:10.1007/s13730-017-0298-6
83. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, section 1.6.9 Nephrology (Table 1.40).
84. Lerner-Ellis JP, Tirone JC, Pawelek PD, Doré C, Atkinson JL, Watkins D, et al. Identification of the gene responsible for methylmalonic aciduria and homocystinuria, cblC type. Nat Genet. 2006;38(1):93-100. doi:10.1038/ng1683
85. Morel CF, Lerner-Ellis JP, Rosenblatt DS. Combined methylmalonic aciduria and homocystinuria (cblC): phenotype-genotype correlations and ethnic-specific observations. Mol Genet Metab. 2006;88(4):315-21. doi:10.1016/j.ymgme.2006.04.001
86. GeneGuidance. MMACHC — clinical gene curation (version of 26 September 2026).
87. Novartis Pharmaceuticals. FABHALTA (iptacopan) capsules, for oral use. Prescribing information. NDA 218276. Revised 2026 (SPL version 6, effective 2026-07-16). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=a76b5845-6e21-4d3b-ad07-cd8df1b60bee
88. European Medicines Agency. Medicines output — report of authorised medicines (Soliris, Ultomiris, Bekemv, Epysqli, Fabhalta, Piasky, Voydeya). Available from: https://www.ema.europa.eu/en/documents/report/medicines-output-medicines_json-report_en.json
89. Wu Q, Yu Z, Wu Z, Yang L, Lin X. Complement-targeted therapies for C3 glomerulopathy and atypical hemolytic uremic syndrome: a time-limited rapid systematic review with narrative synthesis. Front Med (Lausanne). 2026;13:1867718. doi:10.3389/fmed.2026.1867718
90. ClinicalTrials.gov. Studies for “atypical hemolytic uremic syndrome” (API v2; NCT04889430, NCT05795140, NCT05935215, NCT04861259, NCT04958265, NCT07308574, NCT06312644, NCT07399730, NCT05876351, NCT05726916, NCT05684159, NCT03205995, NCT03303313, NCT03999840, NCT02574403, NCT01522183). Bethesda (MD): National Library of Medicine. Available from: https://clinicaltrials.gov/
91. Ariceta G, Besbas N, Johnson S, Karpman D, Landau D, Licht C, et al. Guideline for the investigation and initial therapy of diarrhea-negative hemolytic uremic syndrome. Pediatr Nephrol. 2009;24(4):687-96. doi:10.1007/s00467-008-0964-1
92. Taylor CM, Machin S, Wigmore SJ, Goodship TH. Clinical practice guidelines for the management of atypical haemolytic uraemic syndrome in the United Kingdom. Br J Haematol. 2010;148(1):37-47. doi:10.1111/j.1365-2141.2009.07916.x
93. Loirat C, Fakhouri F, Ariceta G, Besbas N, Bitzan M, Bjerre A, et al. An international consensus approach to the management of atypical hemolytic uremic syndrome in children. Pediatr Nephrol. 2016;31(1):15-39. doi:10.1007/s00467-015-3076-8
94. Schaefer F, Al-Dakkak I, Anokhina K, Cohen D, Greenbaum LA, Ariceta G. Global aHUS Registry analysis of patients switching to ravulizumab from eculizumab. Kidney Int Rep. 2024;9(9):2648-56. doi:10.1016/j.ekir.2024.06.020
95. National Institute for Health and Care Excellence. Ravulizumab for treating atypical haemolytic uraemic syndrome. Technology appraisal guidance TA710. London: NICE; 2021 Jun 23. Available from: https://www.nice.org.uk/guidance/ta710
96. European Medicines Agency. Bekemv (eculizumab): EPAR — product information. Amsterdam: EMA; last updated 2026-09-15. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/bekemv
97. European Medicines Agency. Epysqli (eculizumab): EPAR — product information. Amsterdam: EMA; last updated 2026-08-07. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/epysqli
98. Amgen. BKEMV (eculizumab-aeeb) injection, for intravenous use. Prescribing information. BLA 761333. Revised 2026 (effective 2026-07-02). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=002f9748-b0e8-4b6e-a38b-56930a839491
99. Teva Pharmaceuticals USA. EPYSQLI (eculizumab-aagh) injection, for intravenous use. Prescribing information. BLA 761340. Revised 2026 (effective 2026-07-06). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=1fa1aa80-1c59-44d4-9841-b6c52882651b
100. Java A, Subramanian A, Nester C. Practical considerations for infection prevention with the clinical use of complement inhibitors. Clin J Am Soc Nephrol. 2026 Feb 6. doi:10.2215/CJN.0000001020. Online ahead of print
101. Agence nationale de sécurité du médicament et des produits de santé. Oracilline 1,000,000 UI, comprimé sécable (phénoxyméthylpénicilline): résumé des caractéristiques du produit. Saint-Denis: ANSM; mis à jour le 04/08/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/medicament/67151899/extrait
102. Mbaeyi SA, Bozio CH, Duffy J, Rubin LG, Hariri S, Stephens DS, MacNeil JR. Meningococcal vaccination: recommendations of the Advisory Committee on Immunization Practices, United States, 2020. MMWR Recomm Rep. 2020;69(9):1–41. doi:10.15585/mmwr.rr6909a1. PMID:33417592.
103. Fakhouri F, Fila M, Hummel A, Ribes D, Sellier-Leclerc AL, Ville S, et al. Eculizumab discontinuation in children and adults with atypical hemolytic-uremic syndrome: a prospective multicenter study. Blood. 2021;137(18):2438-49. doi:10.1182/blood.2020009280
104. Bouwmeester RN, Duineveld C, Wijnsma KL, Bemelman FJ, van der Heijden JW, van Wijk JAE, et al. Early eculizumab withdrawal in patients with atypical hemolytic uremic syndrome in native kidneys is safe and cost-effective: results of the CUREiHUS study. Kidney Int Rep. 2023;8(1):91-102. doi:10.1016/j.ekir.2022.10.013
105. Ariceta G, Fakhouri F, Sartz L, Miller B, Nikolaou V, Cohen D, et al. Eculizumab discontinuation in atypical haemolytic uraemic syndrome: TMA recurrence risk and renal outcomes. Clin Kidney J. 2021;14(9):2075-84. doi:10.1093/ckj/sfab005
106. Menne J, Delmas Y, Fakhouri F, Licht C, Lommelé Å, Minetti EE, et al. Outcomes in patients with atypical hemolytic uremic syndrome treated with eculizumab in a long-term observational study. BMC Nephrol. 2019;20(1):125. doi:10.1186/s12882-019-1314-1
107. Zuber J, Frimat M, Caillard S, Kamar N, Gatault P, Petitprez F, et al. Use of highly individualized complement blockade has revolutionized clinical outcomes after kidney transplantation and renal epidemiology of atypical hemolytic uremic syndrome. J Am Soc Nephrol. 2019;30(12):2449-63. doi:10.1681/ASN.2019040331
108. Glover EK, Smith-Jackson K, Brocklebank V, Wilson V, Walsh PR, Montgomery EK, et al. Assessing the impact of prophylactic eculizumab on renal graft survival in atypical hemolytic uremic syndrome. Transplantation. 2023;107(4):994-1003. doi:10.1097/TP.0000000000004355
109. Glover EK, Montgomery EK, Wong EKS, Johnson S, Malina M, Marchbank KJ, et al. Late onset thrombotic microangiopathy in kidney transplants; poor outcome despite eculizumab treatment. Transpl Int. 2025;38:15404. doi:10.3389/ti.2025.15404
110. Saland JM, Shneider BL, Bromberg JS, Shi PA, Ward SC, Magid MS, et al. Successful split liver-kidney transplant for factor H associated hemolytic uremic syndrome. Clin J Am Soc Nephrol. 2009;4(1):201-6. doi:10.2215/CJN.02170508
111. Carrillo Infante C, Mujeebuddin A, Koneru M, Williams C. Safety of ravulizumab use in pregnancy: insights from a global pharmacovigilance analysis. Adv Ther. 2026 Oct 9. doi:10.1007/s12325-026-03816-9. Online ahead of print
112. Stea ED, Pugliano M, Gualtierotti R, Mazzucato M, Santangelo L, Annicchiarico G, et al. Multidisciplinary consensus on the diagnosis and management of patients with atypical Hemolytic Uremic Syndrome (complement-mediated TMA): Recommendations from Italian scientific societies, patient associations and regulators. Pharmacol Res. 2025;216:107714. doi:10.1016/j.phrs.2025.107714
113. Rondeau E, Scully M, Ariceta G, Barbour T, Cataland S, Heyne N, et al. The long-acting C5 inhibitor, ravulizumab, is effective and safe in adult patients with atypical hemolytic uremic syndrome naïve to complement inhibitor treatment. Kidney Int. 2020;97(6):1287-96. doi:10.1016/j.kint.2020.01.035
114. Barbour T, Scully M, Ariceta G, Cataland S, Garlo K, Heyne N, et al. Long-term efficacy and safety of the long-acting complement C5 inhibitor ravulizumab for the treatment of atypical hemolytic uremic syndrome in adults. Kidney Int Rep. 2021;6(6):1603-13. doi:10.1016/j.ekir.2021.03.884
115. Ariceta G, Dixon BP, Kim SH, Kapur G, Mauch T, Ortiz S, et al. The long-acting C5 inhibitor, ravulizumab, is effective and safe in pediatric patients with atypical hemolytic uremic syndrome naïve to complement inhibitor treatment. Kidney Int. 2021;100(1):225-37. doi:10.1016/j.kint.2020.10.046
116. Dixon BP, Kavanagh D, Aris ADM, Adams B, Kang HG, Wang E, et al. Ravulizumab in atypical hemolytic uremic syndrome: an analysis of 2-year efficacy and safety outcomes in 2 phase 3 trials. Kidney Med. 2024;6(8):100855. doi:10.1016/j.xkme.2024.100855
117. ClinicalTrials.gov. NCT05876351 — Eculizumab in pediatric and adult participants with aHUS (complement inhibitor treatment-naïve), results posted. Available from: https://clinicaltrials.gov/study/NCT05876351
118. Kavanagh D, Greenbaum LA, Bagga A, Karki RG, Chen CW, Vasudevan S, et al. Design and rationale of the APPELHUS phase 3 open-label study of factor B inhibitor iptacopan for atypical hemolytic uremic syndrome. Kidney Int Rep. 2023;8(7):1332-41. doi:10.1016/j.ekir.2023.04.029
119. Aigner C, Prohászka Z, Szilágyi Á, Böhmig GA, Arbeiter K, Schmidt A, et al. Crovalimab rescue therapy in a case with genetic complement mediated thrombotic microangiopathy. Kidney Med. 2025;8(1):101185. doi:10.1016/j.xkme.2025.101185
120. Nguyen MD, Ciurea S, Desai S, Kalebasty AR, Tran MH, Gee C, et al. Iptacopan in C5 blockade refractory atypical hemolytic uremic syndrome with associated Castleman's disease: case report. BMC Nephrol. 2025;26(1):585. doi:10.1186/s12882-025-04487-4
121. Benvenuti F, Proglio M, Rigno M, Sorbo D. Iptacopan/LNP023 and rituximab as rescue therapy in a patient with systemic lupus erythematosus-associated atypical haemolytic uraemic syndrome. J Nephrol. 2025;38(8):2435-40. doi:10.1007/s40620-025-02425-z
122. National Institute for Health and Care Excellence. Eculizumab for treating atypical haemolytic uraemic syndrome. Highly specialised technologies guidance HST1. London: NICE; 2015. Available from: https://www.nice.org.uk/guidance/hst1
123. Genentech. PIASKY (crovalimab) injection. Prescribing information. BLA 761388. Effective 2025-11-11. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=2597efc2-a97b-487c-b500-a67fd282a73b
124. Donne RL, Abbs I, Barany P, Elinder CG, Little M, Conlon P, et al. Recurrence of hemolytic uremic syndrome after live related renal transplantation associated with subsequent de novo disease in the donor. Am J Kidney Dis. 2002;40(6):E22. doi:10.1053/ajkd.2002.36938
125. Global aHUS Registry. Available from: https://ahusregistry.com/
126. aHUS Alliance Global Action. Available from: https://www.ahusallianceaction.org/
127. Alkhaldi A, Elawad M, Al Dalbhi S, Alquraini B, Alshannan A. A case report on the effective and safe use of ravulizumab in atypical hemolytic uremic syndrome during pregnancy. BMC Nephrol. 2026;27(1):337. doi:10.1186/s12882-026-04950-w