Genetic Hydrocephalus and L1 Syndrome (CRASH Spectrum)

Monogenic congenital hydrocephalus (L1CAM, CCDC88C, MPDZ, WDR81, TRIM71, SMARCC1, AP1S2) and L1CAM-related syndrome: HSAS, MASA/SPG1 and X-linked complicated agenesis of the corpus callosum

Literature and regulatory information updated: 9 October 2026.

0. Initial Reference Table

The table brings together the entities in the two rows grouped in this article. L1 syndrome is the only form of monogenic congenital hydrocephalus with its own GeneReviews chapter; the other non-syndromic or minimally syndromic forms are described from cohorts, OMIM (Online Mendelian Inheritance in Man) and Orphanet entries and gene-specific curation. Prevalence figures measure different things and are compared in section 9.

1. Introduction

1.1. Congenital hydrocephalus and the place of monogenic forms

Hydrocephalus is accumulation of cerebrospinal fluid (CSF) within the cerebral ventricles, increasing their volume and usually intraventricular pressure. In clinical genetics the term refers mainly to the obstructive or non-communicating form, in which a point along the CSF pathway is blocked; when obstruction is at the aqueduct of Sylvius, the lateral and 3rd ventricles dilate but the 4th does not [37]. Congenital hydrocephalus develops in utero, may be acquired or hereditary, and the severity of cerebral damage depends on ventriculomegaly duration and extent [24]. Causes include central nervous system malformations such as spina bifida, congenital and postnatal infections, intraventricular haemorrhage and pathogenic variants in brain development genes [37, 38]. Fetal ventriculomegaly, defined sonographically by an atrial diameter of at least 10 mm, is the common prenatal presentation of all these causes and does not necessarily mean hydrocephalus, because it also results from cerebral tissue loss, midline malformations or destructive processes [39, 40].

For decades, the only well-established monogenic cause of isolated congenital hydrocephalus was L1 syndrome; in 2013, when an MPDZ variant was described, L1CAM was still considered “the only gene” identified, and most non-syndromic cases had no variant in that gene [14]. In 2017, Alkuraya’s group counted four Mendelian genes whose principal or sole manifestation was congenital hydrocephalus, two X-linked (L1CAM and AP1S2) and two autosomal recessive (CCDC88C and MPDZ) [15]. Trio exome sequencing in patients with sporadic congenital hydrocephalus added genes with de novo variants, such as TRIM71 and SMARCC1, between 2018 and 2024, and changed the dominant pathophysiological model: a relevant proportion of cases results from altered neurogenesis in ventricular-zone neural stem cells rather than a primary CSF circulation disturbance [16, 31, 41, 42].

1.2. L1 syndrome: from clinical description to the gene

In 1949, Bickers and Adams described hereditary stenosis of the aqueduct of Sylvius as a cause of congenital hydrocephalus in Brain, in a family with transmission consistent with X-linked recessive inheritance; the entity became known as Bickers-Adams syndrome or HSAS (hydrocephalus due to stenosis of the aqueduct of Sylvius) [3, 43]. Edwards published “sex-linked hydrocephalus” in 1961 and Holmes broadened the description in 1973 [3]. In 1974, Bianchine and Lewis delineated MASA syndrome, the English acronym for intellectual disability (mental retardation, a historical term), aphasia, shuffling gait and adducted thumbs [3]. Fryns and colleagues showed in 1991 that the same Xq28 abnormality had variable expression within a family, and linkage studies placed HSAS, MASA, X-linked agenesis of the corpus callosum and X-linked complicated hereditary spastic paraplegia type 1 (SPG1) in the same chromosomal region [3].

Molecular milestones followed between 1992 and 1995. In 1992, Rosenthal, Jouet and Kenwrick found that the X-linked hydrocephalus locus mapped to Xq28 near the neural cell adhesion molecule L1 gene, and identified abnormal L1 messenger RNA species in one family resulting from use of alternative splicing acceptors, with an intronic branch-point variant cosegregating with disease; this was the first causal variant described in the gene and led the authors to propose HSAS as a neuronal migration disorder [44]. Jouet and colleagues confirmed the defect with a missense variant in 1993 [45], and Van Camp and colleagues described a gene duplication in X-linked hydrocephalus [3]. In 1995, Jouet and colleagues reported nine new variants, including the first in fibronectin type III domains, and demonstrated that X-linked hydrocephalus, MASA syndrome and SPG1 are allelic disorders, with a spectrum varying between and within families [4]. In the same year, identifying a disabling variant in two pedigrees with apparently sporadic hydrocephalus enabled the first definitive prenatal diagnosis at 10 weeks’ gestation [46].

1.3. Nomenclature: CRASH, L1 syndrome and subtypes

In 1995, Fransen and colleagues reviewed evidence that HSAS, MASA, X-linked complicated spastic paraparesis and X-linked agenesis of the corpus callosum were all caused by variants in the L1 gene, with different phenotypes in the same family, and proposed the acronym CRASH: Corpus callosum hypoplasia, intellectual disability (Retardation), Adducted thumbs, Spastic paraplegia and Hydrocephalus [2]. The historical term, still used in OMIM and mosaicism literature [47], has been replaced in practice by “L1 syndrome” or “L1 disease”. GeneReviews, written by Connie Stumpel (Maastricht) and Yvonne Vos (Groningen), published in 2004 and last revised on 7 January 2021, groups three phenotypes: HSAS; MASA, including SPG1; and X-linked complicated agenesis of the corpus callosum [1]. Orphanet uses L1 syndrome (ORPHA:275543) as the overarching entity and treats HSAS, MASA, X-linked complicated SPG1 and X-linked complicated dysgenesis of the corpus callosum as four clinical subtypes; unlike GeneReviews, it separates SPG1 from MASA and defines it as spastic paraplegia with intellectual disability and normal brain morphology [19, 22]. Both classifications describe the same continuum, and all phenotypes may coexist within a family [1].

1.4. Definitions of the article’s entities

In a boy, L1 syndrome is defined by suggestive clinical signs and a hemizygous pathogenic L1CAM variant; it is unusual but possible in a girl with a heterozygous variant [1]. HSAS is the severe extreme, with frequently prenatal hydrocephalus, adducted thumbs, spasticity and moderate to severe intellectual disability [20]. MASA comprises mild to moderate intellectual disability, speech delay, hypotonia evolving into spastic paraplegia, adducted thumbs and mild to moderate ventricular dilatation [21]. X-linked complicated agenesis of the corpus callosum combines callosal dysgenesis, hypoplasia or aplasia, variable spastic paraplegia and mild to moderate intellectual disability [1, 23].

The non-L1 monogenic forms in row 293 are, by OMIM designation, congenital hydrocephalus 1 (HYC1, CCDC88C, 236600), congenital hydrocephalus 2 with or without brain or eye abnormalities (HYC2, MPDZ, 615219), congenital hydrocephalus 3 with brain abnormalities (HYC3, WDR81, 617967), congenital hydrocephalus 4 (HYC4, TRIM71, 618667) and susceptibility to congenital hydrocephalus 5 (HYC5, SMARCC1, 620241) [5]. OMIM entry 618570, grouped by Orphanet under congenital hydrocephalus (ORPHA:2185), is the TRIM71 gene entry, not a phenotype; the corresponding phenotype is HYC4 (618667) [5, 24]. Fried syndrome, described in 1972, and Pettigrew syndrome are historical names for the AP1S2 X-linked disorder with intellectual disability, hydrocephalus and basal ganglia calcification [6, 18]. Orphanet organises these forms by mechanism: congenital communicating hydrocephalus (ORPHA:269505; MPDZ, WDR81 and TRIM71) and congenital non-communicating hydrocephalus (ORPHA:269510; CCDC88C and WDR81) [25, 26]. The classification of MPDZ as communicating, inherited from the original family, was revised in 2025 when imaging and neuropathological reanalysis showed obstruction upstream of thalamic fusion [30].

2. Molecular Genetics and Pathophysiology

2.1. CSF physiology and hydrocephalus types

CSF passes from the lateral ventricles to the 3rd ventricle through the foramina of Monro, from there to the 4th through the aqueduct of Sylvius, and exits to the subarachnoid space through the foramen of Magendie into the cisterna magna and through the foramina of Luschka into the pontine cistern [37]. Intraventricular haemorrhage may block the foramina and produce obstructive hydrocephalus; subarachnoid haemorrhage may impair reabsorption through arachnoid villi and produce communicating hydrocephalus [37]. Gaglioti and colleagues group processes causing fetal ventriculomegaly into four categories: abnormal CSF turnover, obstructive or non-obstructive; impaired neuronal migration and proliferation; agenesis of the corpus callosum and holoprosencephaly; and destructive processes [39]. This distinction matters clinically because CSF shunting controls intracranial pressure but does not correct a primary cortical formation defect.

Genetics over the last two decades has added a third model. In 125 trios and 52 probands with congenital hydrocephalus, Furey and colleagues found a significant burden of rare damaging variants in TRIM71 (p=2.15×10⁻⁷), SMARCC1 (p=8.15×10⁻¹⁰) and PTCH1 (p=1.06×10⁻⁶) and two de novo duplications at the SHH locus; these probands represented approximately 10% of cases, and all genes regulate ventricular-zone neural stem cell fate [41]. Duy and colleagues showed in mice that neuroepithelial Trim71 deletion, or introduction of a human variant, produces prenatal hydrocephalus through premature progenitor differentiation, reduced neurogenesis, cortical hypoplasia and a more compliant cortex, with secondary ventricular enlargement without a primary CSF circulation defect [42]. In L1CAM, MPDZ and CCDC88C, by contrast, the mechanism includes a structural component (aqueductal stenosis, ependymal or choroid plexus abnormality) in addition to neuronal development abnormalities [1, 10, 13].

2.2. L1CAM: gene, protein and function

L1CAM lies at Xq28 near the long-arm telomere and encodes neural cell adhesion molecule L1 (UniProt P32004; aliases CD171, NCAM-L1, CAML1, HSAS1, SPG1, MASA, MIC5, S10) [2, 7]. The MANE Select transcript is NM_001278116.2, encoding the 1257-amino-acid isoform 1 (NP_001265045.1), the same protein encoded by NM_000425.5, the transcript most used in historical literature; NM_024003.3, a reference transcript in some databases, encodes the 1253-amino-acid isoform 2 [8]. Some reports indicate a 1238-amino-acid protein, probably the mature protein without signal peptide [8, 48]. The first systematic screen included the 28 coding exons and regulatory elements of the 5' untranslated region [49].

L1 is a transmembrane glycoprotein of the immunoglobulin superfamily, with six extracellular immunoglobulin-like domains (Ig), five fibronectin type III-like domains (FnIII), a single-pass transmembrane domain and a short cytoplasmic domain [1]. It mediates axonal growth and fasciculation, Schwann cell–axon interactions, neuronal migration and survival through homophilic (L1 with L1) and heterophilic (L1 with other proteins) interactions and signal transduction through associated proteins; GeneReviews writes “hemophilic”, a typographical error for “homophilic” [1]. Loss of function disrupts cell–cell adhesion and adhesion-dependent signalling in tyrosine kinase receptors, impairing neuronal migration, axonal growth and fasciculation and synaptogenesis; this causes corticospinal tract, corpus callosum and cerebellar malformations [7]. There is no measurable biochemical deficiency: this is a structural developmental defect without a confirmatory functional test in clinical practice [7].

2.3. L1CAM: variant spectrum and mechanism

The mechanism is loss of function. Truncating variants (nonsense, frameshift, splicing), exon or whole-gene deletions and many missense variants are pathogenic [1]. Missense variants at key residues essential for Ig and FnIII domain folding according to Bateman and colleagues are frequently pathogenic; missense variants at other residues altering domain conformation may also be pathogenic [1]. In 2010, the L1CAM variant database recorded over 200 different variants throughout the gene, approximately 35% missense; truncating variants are considered pathogenic except truncations near the stop codon, and missense variants require caution without additional evidence [50]. A ClinVar search in October 2026 returned 1879 L1CAM records, of which 487 were classified as pathogenic or likely pathogenic; these figures count aggregated submission records, not patients or unique variants [51].

Sequencing detects 99% of known pathogenic variants and targeted deletion/duplication analysis approximately 1%, including a whole-gene deletion; this proportion describes the distribution of variants detectable by method, not clinical test sensitivity in suspected patients [1]. Less obvious classes include the historical synonymous variant “C924T (G308G)”, causing HSAS and normalised as NM_001278116.2:c.924C>T p.(Gly308=), assuming NM_000425 numbering [1, 8]; the intronic branch-point variant in the first family [44]; and the deep intronic NM_001278116.2:c.1124-24T>G variant in intron 10, creating a cryptic acceptor and retaining 23 intronic base pairs, described in 2026 [52]. Contiguous deletion of L1CAM and AVPR2 causes X-linked hydrocephalus with nephrogenic diabetes insipidus [1].

The following table collects the L1CAM variants cited in the article, with original nomenclature and the normalised MANE transcript form when normalisation was performed using VariantValidator (GRCh38).

The c. numbering coincides in NM_001278116.2, NM_000425.5 and NM_024003.3 in the regions of variants c.924, c.2248 and c.2351, allowing comparison of older literature with current reports at these positions [8].

2.4. L1CAM: genotype–phenotype correlation, mosaicism and modifiers

Weller and Gärtner showed in 2001 that extracellular-domain missense variants and cytoplasmic-region variants produce milder phenotypes than extracellular-domain truncations or absence of detectable L1 protein [1]. Vos and colleagues analysed 367 referred patients and found 68 different variants in 73; among children with an identified variant and clinical data, death before 3 years occurred in 52% with a truncating variant and 8% with a missense variant (Fisher’s exact test, p=0.02), in an analysis of 33 individuals [1, 58]. There are relevant exceptions: a key-residue missense variant, p.(Tyr784Cys), caused prenatal hydrocephalus, severe delay, callosal agenesis or hypogenesis, epilepsy and adducted thumbs in two siblings and a cousin [53]. The Horiuchi case authors also cite Yamasaki’s classification, in which class 3 denotes the most severe variants, and Michaelis’s observation that FnIII-domain variants may be more severe than Ig-domain variants [48]. Mild-to-severe variability within a family carrying the same variant shows that other factors influence phenotype; the correlation is a group trend and does not predict individual outcome [1, 7].

Somatic and germline mosaicism are documented. Vits and colleagues presented evidence of both in “CRASH syndrome” [47]; Vos and colleagues demonstrated maternal germline mosaicism [58]; and Du and colleagues described somatic and germline mosaicism in an unaffected father of two heterozygous daughters [1]. In Finckh’s series, 8 of 46 identified variants had arisen de novo in recent generations [49].

GeneReviews recognises no allelic L1CAM disorders outside the L1 spectrum. The association with Hirschsprung disease, described in at least 16 individuals, is interpreted as a modifier effect: an L1CAM variant alone does not cause Hirschsprung disease but may modify the effect of an associated gene such as RET [1, 54]. In a Spanish patient with X-linked hydrocephalus and Hirschsprung disease, the historical L1CAM variant G698R coexisted with the variant NM_020975.6(RET):c.73+9277T>C in the RET enhancer [56]. Conversely, Takenouchi and colleagues interpreted the combination as the severe extreme of the spectrum rather than a neomorphic effect [55].

2.5. L1CAM: models and experimental therapies

A mouse model carrying the p.D201N variant (published designation), in which the protein reaches the cell surface, reproduces neurological deficits, locomotor hyperactivity, enlarged ventricles and a reduced corpus callosum in hemizygous males; in cultured cerebellar neurons, L1-mimetic compounds (duloxetine, crotamiton and trimebutine) restored migration, survival and neurite outgrowth [59]. This is in vitro proof of principle, without human studies.

2.6. CCDC88C (HYC1, autosomal recessive)

CCDC88C at 14q32.11-q32.12 encodes DAPLE (also called HkRP2), a Hook-related protein that binds Dishevelled through a C-terminal PDZ-domain-binding motif, binds activated FZD7 receptor while displacing DVL1 and inhibits canonical Wnt signalling; it also functions as a non-receptor guanine nucleotide exchange factor for Gαi, activating non-canonical Wnt signalling (RAC1, PI3K-AKT) [9]. Expression is ubiquitous, highest in fetal brain, particularly pons and cerebellum [13]. Through positional cloning in a consanguineous family with autosomal recessive hydrocephalus, Ekici and colleagues identified a homozygous intron 29 splice-donor variant that excludes exon 29 and generates a non-functional truncated protein, designated p.S1591HfsX7 in the article; the variant abolishes the PDZ-binding site and is associated with increased CTNNB1 and LEF1 messenger RNA and reduced expression of other Wnt pathway genes [9, 13]. Genomic and MANE transcript notation could not be obtained because the abstract does not state the c. position [8].

The HYC1 phenotype (OMIM 236600) is severe hydrocephalus beginning in utero, with a complex hydrocephalic malformation and medial diverticulum in the original family, mild to severe intellectual disability, variable motor delay and childhood-onset seizures [9, 13]. Outcomes are not always unfavourable: the surviving patient in the original family had normal psychomotor development at 3 years and 3 months, and in a two-family series, children shunted in the first weeks of life, without multiple revisions and carrying a more distal truncating variant, achieved some developmental milestones [9, 13]. Ruggeri and colleagues concluded that biallelic variants are a rare cause of severe congenital hydrocephalus [9]. GenCC (Gene Curation Coalition) classifies the gene–disease relationship as strong [9].

The same gene causes an allelic disease with opposite inheritance, autosomal dominant spinocerebellar ataxia 40 (SCA40, OMIM 616053), with slowly progressive cerebellar ataxia and pontocerebellar atrophy; the reported heterozygous variant, normalised as NM_001080414.4:c.1391G>A p.(Arg464His), activates JNK signalling, caspase 3 cleavage and apoptosis [8, 9]. OMIM marks this association as provisional, GenCC classifies it as limited to moderate and Orphanet (ORPHA:423275) records five cases [9].

2.7. MPDZ (HYC2, autosomal recessive)

MPDZ at 9p23 encodes MUPP1, a multiple-PDZ-domain scaffold protein forming part of tight junctions and expressed early in choroid plexus and ependyma [10]. Al-Dosari and colleagues used autozygosity mapping and linkage within a 6.9 Mb interval at 9p24.1-p22.3 to identify a truncating variant in a Saudi family with severe congenital hydrocephalus described as communicating; the same founder variant was present in a stillborn infant with massive hydrocephalus from another family [14]. Shaheen and colleagues confirmed the gene with five recessive alleles and a broader phenotypic spectrum [15].

Neuropathology in fetuses with homozygous null variants shows multiple ependymal malformations along the aqueduct, 3rd and 4th ventricles and medullary central canal, multifocal rosettes and immature subependymal cells prematurely detached from the ventricular zone [10]. In Mpdz loss-of-function mice, choroid plexus epithelium becomes permeable, contrast enters ventricles, CSF protein rises up to 53-fold and transcytosis increases [10]. In 2025, reanalysis of prenatal and postnatal imaging and fetopathology in six French cases (four fetuses and two children) showed obstructive ventriculomegaly upstream of partial thalamic fusion (diencephalosynapsis), with partial 3rd ventricular atresia that may extend to the aqueduct [30]. The original family’s “communicating” designation, still reflected by Orphanet, should therefore be read as historical [25, 30].

The OMIM phenotype includes lissencephaly, a thin corpus callosum, neuronal heterotopia, iris coloboma, motor delay, sometimes intellectual disability or seizures, cardiac septal defects and non-specific dysmorphism [10]. Reports in 2024 and 2026 broadened it: prenatal Dandy-Walker malformation with vermian hypoplasia and a univentricular heart with a homozygous nonsense variant; and, in two siblings with the homozygous canonical splice variant c.2650-1G>A (transcript not stated in the source; it corresponds to a canonical acceptor site in NM_001378778.1), aortic coarctation, bicuspid aortic valve, partial anomalous pulmonary venous return, ventricular septal defect, Dandy-Walker malformation, subependymal heterotopia, megalocornea and chorioretinal lesions in the proband, and diaphragmatic hernia with a single ventricle in the deceased sibling [10, 60]. These reports suggest a multisystem phenotype that is not yet consolidated.

2.8. WDR81 (HYC3 and CAMRQ2, autosomal recessive) and WDR91

WDR81 at 17p13.3 encodes an endolysosomal membrane protein with BEACH and WD40 domains that restricts endosomal class III phosphatidylinositol kinase through BECN1, regulating endosomal fusion and transport, aggrephagy and mitochondrial organisation; it is abundant in cerebellum, corpus callosum and Purkinje cells [11]. The disorder is autosomal recessive with a broad spectrum. The mild extreme is cerebellar ataxia, intellectual disability and disequilibrium syndrome type 2 (CAMRQ2, OMIM 610185), with congenital non-progressive ataxia, cerebellar hypoplasia, intellectual disability and, in some patients, quadrupedal locomotion, compatible with adulthood; the original kindred was homozygous for NM_001163809.2:c.2567C>T p.(Pro856Leu) [8, 11]. The severe extreme is HYC3 (OMIM 617967), with prenatal ventriculomegaly, hypoplastic or absent cerebellum, holoprosencephaly or Dandy-Walker malformation, and death of most fetuses in utero or shortly after birth [11, 15]. Between these lie microcephaly and microlissencephaly with pontocerebellar hypoplasia and fibroblasts with an elevated mitotic index [11]. Lethal forms are associated with truncating or non-expressed alleles, so variant type and imaging should be interpreted together before counselling [11]. The first Chinese prenatal case, with severe hydrocephalus, cleft lip and palate, hydrops, hepatomegaly and cerebellar hypoplasia, was compound heterozygous for c.146_147insG p.(Thr52fs) and c.673delC p.(Leu225fs) (transcript not stated), suggesting extracerebral manifestations may point towards this gene [61].

WDR91, WDR81’s functional paralogue in endosomal conversion, was associated in 2026 with severe congenital communicating hydrocephalus with Dandy-Walker malformation, cerebellar hypoplasia and dysmorphism in a child homozygous for NM_014149.4:c.1245_1246del p.(Arg415SerfsTer64), with heterozygous parents; this remains a single report without an OMIM entry [62].

2.9. TRIM71 (HYC4, de novo variants)

TRIM71, the human orthologue of lin-41, is the protein with the most de novo variants and the most specific expression among congenital hydrocephalus risk genes in embryonic neuroepithelial stem cells [42]. In Jin and colleagues’ cohort, the six de novo variants clustered in the RNA-binding NHL domain, with the recurrent variants NM_001039111.3:c.1823G>A p.(Arg608His) and NM_001039111.3:c.2387G>A p.(Arg796His), each in three patients; all had white matter volume loss and corpus callosum abnormalities, and some had cranial nerve deficits (n=4), non-obstructive interhemispheric cysts (n=3), hearing impairment (n=3) and skeletal abnormalities (n=2) [16]. The variants disrupt RNA binding and produce premature differentiation of progenitors [42]. Orphanet includes TRIM71 under congenital communicating hydrocephalus, an entity it records as autosomal recessive, whereas the reported patients have heterozygous de novo variants [16, 25].

2.10. SMARCC1 (HYC5, autosomal dominant with incomplete penetrance)

SMARCC1, at 3p21.31, encodes BAF155, a scaffold subunit of the BAF (SWI/SNF) chromatin-remodelling complex, present in the neural npBAF and nBAF complexes [12]. Jin and colleagues described three de novo variants (NM_003074.4:c.1577A>C p.(His526Pro), p.(Lys891fs) and NM_003074.4:c.1571+1G>A), two transmitted loss-of-function variants (NM_003074.4:c.1723C>T p.(Gln575Ter) and p.(Val535fs); the frameshifts are published designations without a c. position in the source), one unphased loss-of-function variant and one transmitted missense variant (NM_003074.4:c.1954C>T p.(Arg652Cys)); all six imaged patients had aqueductal stenosis and corpus callosum abnormalities, three had heart disease and two skeletal abnormalities [16]. In a cohort of 2697 trios with ventriculomegaly (8091 exomes, 2016–2023) compared with 1798 control exomes from the Simons Simplex Collection, SMARCC1 reached genome-wide significance with six rare de novo variants at conserved residues; knockdown in Xenopus reproduced aqueductal stenosis and cardiac defects, was rescued by the wild-type but not the mutant protein, and altered neurogenesis genes such as NEUROD2 and MAB21L2 [31]. The authors named the disorder SMARCC1-associated developmental dysgenesis syndrome (SaDDS), a BAFopathy [12, 31]. Penetrance is incomplete: in a fetus with severe ventriculomegaly and aqueductal stenosis, terminated at 18 weeks, the loss-of-function variant was inherited from the unaffected mother; after that publication OMIM came to regard SMARCC1 as a susceptibility gene and the variant was reclassified as likely pathogenic [17]. ClinGen (Clinical Genome Resource) classifies the relationship as definitive [12]. SMARCC1 should not be described as a cause of Coffin-Siris syndrome, with which it was associated only as a candidate [12].

2.11. AP1S2 (Fried/Pettigrew syndrome, X-linked)

AP1S2, at Xp22.2, encodes the σ1B subunit of the adaptor complex AP-1, which mediates vesicular trafficking between the trans-Golgi and endosomes; the early literature called it the σ2 subunit [6, 18]. In Saudubray’s textbook, the disorder belongs to the “Golgipathies”, early-onset encephalopathies of cellular trafficking [63]. All reported variants are loss-of-function: in a four-generation French family, an intron 3 variant caused exon 3 skipping and termination at codon 64; in a Scottish family, the variant published as p.Gln66Ter (historical designation; NM_001272071.2:c.226G>T p.(Glu76Ter) in ClinVar, VCV000010781) [18]. Expressivity is highly variable and does not depend on variant type; carrier females are generally unaffected [6]. ClinGen assigns the gene a haploinsufficiency score of 3 [6].

2.12. Exome cohort genes and recent candidate genes

In the largest exome cohort published by 2020 (381 probands, 232 trios, with surgically treated primary sporadic congenital hydrocephalus and no myelomeningocele, mainly of European origin and recruited in the USA and elsewhere, including Uganda and Turkey), known familial genes (L1CAM, MPDZ, FLNA, CRB2) explained approximately 2.1% of cases, damaging de novo variants more than 17% and rare large-effect variants approximately 22% [16]. Five genes had a genome-wide significant de novo variant burden: TRIM71, SMARCC1, PTEN, PIK3CA and FOXJ1; PIK3CA, PTEN, MTOR, FOXJ1, FMN2, PTCH1 and FXYD2 together explained approximately 7.3% [16]. In a European cohort of 28 Belgian and referred probands already negative for L1CAM, karyotype and known syndromes, exome sequencing found a pathogenic or likely pathogenic diagnosis in 18% (de novo ARID1A, homozygous CRADD, KIDINS220, POMGNT1 and POMT2), candidate genes in 11% and a significant primary cilium gene burden suggesting oligogenic inheritance [64].

Reports between 2025 and 2026 described: AMOT (angiomotin, Xq23), with a first methionine codon variant producing an excessively stable protein lacking the 91 N-terminal amino acids in six males from a family with isolated X-linked congenital hydrocephalus [65]; FSD1L, with biallelic variants in 11 individuals from six families, an autosomal recessive L1 syndrome phenocopy [66]; TRIO, with de novo missense variants in Ras-GEF domains in six probands from the 2697-trio cohort (adjusted p 6.12×10⁻⁵) [67]; ARID1A, with a last-exon de novo variant, NM_006015.6:c.5259_5262dup p.(Ser1755ValfsTer2), escaping nonsense-mediated decay and carrying a BAFopathy episignature, in a fetus with severe hydrocephalus at 19 weeks [68]; and WDR91, described above [62]. AMOT, TRIO and WDR91 rely on one family or few probands; FSD1L has six families and an animal model.

3. Clinical Spectrum and Manifestations by System

3.1. L1 syndrome: overview

L1 syndrome affects almost exclusively boys. By 2010, more than 280 individuals with a pathogenic L1CAM variant had been identified [1]. The spectrum ranges from lethal prenatal hydrocephalus to mild learning difficulties and behavioural abnormalities, and all phenotypes may occur within a family carrying the same variant [1, 69]. The GeneReviews table summarises differences between the three phenotypes in male probands.

Source: GeneReviews, Table 2 [1].

3.2. HSAS: the severe extreme

Boys with HSAS are born with severe hydrocephalus, frequently prenatal in onset, and adducted thumbs [1]. GeneReviews defines hydrocephalus through two criteria: increased intraventricular fluid volume, manifested by increased head circumference and, on imaging, ventricular enlargement, sulcal effacement and transependymal CSF reabsorption; and increased intraventricular pressure, with age-dependent signs and symptoms (progressive head circumference increase, headache, nausea and vomiting, irritability) or demonstration by ultrasound or other imaging [1]. Prenatal hydrocephalus may be severe enough to prevent vaginal delivery and cause fetal or early childhood death [1, 3]. In the French neuropathological series of 138 cases, predominantly fetal, referred for L1CAM testing, the 57 with a deleterious variant had hydrocephalus in 100%, adducted thumbs in 88%, pyramidal tract agenesis or hypoplasia in 98%, aqueductal stenosis in 90% and callosal agenesis or hypoplasia in 68% [29]. Orphanet describes HSAS as the severe extreme of the spectrum with an unfavourable prognosis [20].

3.3. MASA and SPG1

In MASA, including SPG1, intellectual disability is mild to moderate, speech onset is delayed and initial hypotonia evolves into spastic paraplegia; adducted thumbs may occur, and magnetic resonance imaging (MRI) shows variable abnormalities [1]. Many affected boys have no hydrocephalus: they have a narrow scaphocephalic skull, more variable intellectual disability, with intelligence quotient (IQ) between 30 and 70, spasticity, shuffling gait and aphasia [3]. Orphanet describes mild to moderate ventricular distension and neonatal onset in MASA [21] and defines X-linked complicated SPG1 as spastic paraplegia with mild to moderate intellectual disability and normal brain morphology, the mild extreme of the spectrum [22]. At the mildest reported extreme, a Dutch patient had a predominantly behavioural phenotype with mild learning difficulties, prompting the authors to recommend brain MRI and exome sequencing in patients with that profile [69].

3.4. X-linked complicated agenesis of the corpus callosum

The phenotype described by Yamasaki and colleagues in 1995 combines variable spastic paraplegia, mild to moderate intellectual disability and callosal dysgenesis, hypoplasia or aplasia [1]. Sources differ on one detail: GeneReviews Table 2 records adducted thumbs in fewer than 50% and speech delay in these patients, whereas the Orphanet definition (ORPHA:1497) states that hydrocephalus, adducted thumbs and absent speech have not been observed [1, 23]. Orphanet bases its definition on 11 cases or families, whereas GeneReviews summarises the spectrum in more than 280 patients; the discrepancy probably reflects selection in the series defining each entity [1, 23].

3.5. Central nervous system and neuroimaging

Hydrocephalus, with or without aqueductal stenosis, coexists with agenesis or hypogenesis of the corpus callosum, cerebellar hypoplasia, a small brainstem and agenesis of the medullary pyramids, that is, the corticospinal tracts [1]. Bilateral absence of the pyramids, seen on MRI or autopsy, is an almost pathognomonic finding; aqueductal stenosis is not constant [1]. Smith’s textbook adds thalamic fusion, a small brainstem and a porencephalic cyst as occasional findings [3]. A computed tomography (CT) scan of a 1-month-old boy with aqueductal stenosis, reproduced in the Oxford Desk Reference (Figure 2.15), shows marked dilatation of the 3rd and lateral ventricles, a normal 4th ventricle and periventricular hypodensity from transependymal CSF exudation under pressure [37].

3.6. Development, cognition and behaviour

The degree of intellectual disability does not necessarily track head size or hydrocephalus severity, and some boys have severe intellectual disability with normal head circumference [1]. In HSAS with severe hydrocephalus, intellectual disability is usually severe, with IQ below 30, and independent of shunting; in MASA, IQ lies between 50 and 70 (mild) and 30 and 50 (moderate) [1, 3]. All phenotypes have speech delay [1]. There are no specific behavioural problems; childhood-onset psychosis has been reported in two unrelated boys, possibly modulated by familial factors or other rare variants [1]. Epileptic seizures may occur [1, 53].

3.7. Motor system: spasticity and gait

Initial lower-limb hypotonia evolves into spasticity in the first years, with brisk tendon reflexes and Babinski signs [1]. Adult spasticity tends to be somewhat progressive, although this is not documented in a large group; it causes muscle atrophy and contractures and produces the shuffling gait from which MASA takes its name [1]. Reported orthopaedic complications include contractures, scoliosis and hip dislocation [1].

3.8. Hands: adducted thumbs

The thumb flexed across the palm (“cortical” or clasped thumb) is the most characteristic external sign. Frequency varies by series: more than 50% of boys with HSAS in GeneReviews, 50% in Table 2, 88% in Smith’s textbook and 88% of variant-positive fetuses in the French series [1, 3, 29]. Before 24 weeks’ gestation the sign is present in only approximately half of fetuses with X-linked hydrocephalus, but in most by the late second trimester [48]. The muscular mechanism is described inconsistently within GeneReviews itself: the suggestive findings section, citing Schrander-Stumpel and Fryns (1998), attributes the posture to a developmental defect of extensor pollicis longus and/or brevis, while the clinical description attributes it to absent abductor pollicis longus, the version reproduced by GeneGuidance curation [1, 7]. The extensor version is linked by GeneReviews to a primary source (Schrander-Stumpel and Fryns, 1998), also supporting the frequency above 50%; the abductor version has no dedicated reference. That primary article is not among the consulted sources, which also contain no anatomical study resolving the discrepancy; this article therefore describes the sign as hypoplasia or absence of thumb extensor and/or abductor muscles [1].

3.9. Digestive system: Hirschsprung disease and pseudo-obstruction

Hirschsprung disease, absence of ganglion cells and hypertrophic nerve trunks in the distal intestine due to failed neural crest migration, has been described in at least 16 individuals with L1CAM variants [1]. Reports include an infant with severe prenatal hydrocephalus, aqueductal stenosis, adducted thumbs and chronic constipation, confirmed by rectal biopsy [54]; a Japanese boy with severe hydrocephalus due to aqueductal stenosis, callosal hypoplasia and biopsy-confirmed Hirschsprung disease [55]; a Spanish patient with a concurrent RET enhancer variant [56]; and the Japanese newborn described in section 10 [48]. The consulted sources do not estimate the association’s frequency. Congenital idiopathic intestinal pseudo-obstruction with hydrocephalus has also been reported in an infant with an L1CAM variant [1].

3.10. Other systems

Contiguous deletion of L1CAM and AVPR2 adds nephrogenic diabetes insipidus to the neurological presentation [1]. Facial features may be somewhat coarse and asymmetric [3]. Two siblings with a deep intronic variant had hearing impairment in addition to callosal agenesis and ventriculomegaly [52].

3.11. Heterozygous females

Heterozygous females may have minor signs such as adducted thumbs or mild intellectual disability and rarely the full phenotype [1]. Smith’s textbook describes carriers as usually normal, sometimes with borderline intelligence or adducted thumbs [3]. Severe hydrocephalus has been reported in a heterozygote, and GeneReviews authors observed a girl with aqueductal stenosis, adducted thumbs and mild intellectual disability, without family history and with skewed X-inactivation [1]. Otter and colleagues described epilepsy in the daughter of a very mildly affected man, with presumed heterozygosity [1].

3.12. Natural history and treated disease

HSAS natural history includes fetal death, early childhood death and, in survivors with severe hydrocephalus, severe intellectual disability; mild to moderate ventriculomegaly is compatible with long survival [1]. Available survival data mix treated and untreated patients: the difference in mortality before 3 years between truncating (52%) and missense (8%) variants came from a laboratory referral series without uniform treatment description [58]. Shunting controls intracranial pressure but does not alter intellectual disability in HSAS with severe hydrocephalus [1, 7].

3.13. Non-L1 congenital hydrocephalus: gene-specific spectrum

The table summarises non-L1 clinical features. Data derive from small series and reports; patient numbers are those of the indicated sources.

In AP1S2 disease, walking is delayed or absent and speech limited or absent; the original Pettigrew family had iron deposition on MRI and autopsy [6]. In the FSD1L autosomal recessive phenocopy, the spectrum ranges from severe fetal hydrocephalus with callosal agenesis and absent pyramidal decussation to a neurodevelopmental disorder with severe intellectual disability, spastic tetraparesis, reduced vision, epilepsy, optic nerve hypoplasia and mild ventricular dilatation [66].

4. Clinical Recognition and Warning Signs

4.1. Fetal period

Fetal ventriculomegaly is the most frequent entry point. L1 syndrome should be suspected in a male fetus with moderate to severe or progressive ventriculomegaly, particularly with aqueductal stenosis (lateral and 3rd ventricular dilatation with a normal 4th ventricle), adducted thumbs, callosal agenesis or hypoplasia, or a maternal family history of affected boys [1, 29, 39]. The Society for Maternal-Fetal Medicine (SMFM) states that L1CAM variants cause the most common hereditary form of hydrocephalus in male fetuses [40]. Two timing details affect recognition: L1 syndrome ventriculomegaly usually begins after 20 weeks and may appear only near term or after delivery, and adducted thumbs occur in approximately half of fetuses before 24 weeks [1, 37, 48]. A normal second-trimester ultrasound in an at-risk pregnancy does not exclude disease [1].

Other fetal patterns point to specific genes: ventriculomegaly with thalamic fusion and 3rd ventricular atresia suggests MPDZ [30]; hydrocephalus with a hypoplastic or absent cerebellum, particularly with a cleft, hydrops or hepatomegaly, suggests WDR81 [11, 61]; aqueductal stenosis with callosal abnormality and heart disease suggests SMARCC1 [31]; and ventriculomegaly with a medial diverticulum pattern suggests CCDC88C [13]. A neuropathological review of fetuses without L1CAM variants showed that the aqueductal atresia or bifurcation spectrum and rhombencephalosynapsis explain 27% of suspected cases, while 20% are phenocopies with a family history suggesting autosomal recessive inheritance [29].

4.2. Newborns and infants

In infants, hydrocephalus presents with progressive macrocephaly, head circumference crossing centiles, a bulging tense anterior fontanelle and the “sunsetting” sign [37]. In a boy with hydrocephalus, examination should seek congenital adducted thumbs, thumb or radial hypoplasia (pointing to X-linked hydrocephalus with VACTERL features), spasticity signs (increased tone, exaggerated tendon reflexes, Babinski sign), spina bifida, microphthalmia and café-au-lait spots [37]. A newborn with X-linked hydrocephalus and severe constipation warrants investigation for Hirschsprung disease [1, 54, 55]. Macrocephaly with megalencephaly and polymicrogyria in a patient with hydrocephalus should suggest PIK3CA and megalencephaly-capillary malformation syndrome (MCAP): in Jin’s cohort, four of five patients with PIK3CA variants retrospectively met MCAP criteria without previous diagnosis [16].

4.3. Children and adolescents

After 2 years, hydrocephalus presents with signs and symptoms of intracranial hypertension [37]. In a boy with global developmental delay, speech delay and early hypotonia evolving into spastic paraplegia, with or without adducted thumbs and even without hydrocephalus, consider MASA/SPG1 [1]. In mild forms, hydrocephalus may be subclinical and detected only during developmental delay investigation [1]. Mild intellectual disability with a behavioural phenotype has also been described [69]. In the X-linked intellectual disability differential, hydrocephalus, adducted thumbs and neurological signs point to L1CAM [71], and in callosal agenesis MRI should be reviewed for hydrocephalus suggestive of L1CAM [72].

4.4. Warning and emergency signs

Hydrocephalus with intracranial hypertension is a neurosurgical emergency: bulging fontanelle, rapidly increasing head circumference, vomiting, irritability, drowsiness and sunsetting in infants, or headache, vomiting and altered consciousness in older children [1, 37]. In shunted patients, the same signs may indicate shunt failure; remote intracranial pressure monitoring detected failures associated with high pressure in a feasibility study [73]. Fever in a recently shunted patient raises suspicion of shunt infection, affecting up to 15% of patients [74]. In a patient with an intrathecal baclofen pump, high fever, altered mental status, rebound spasticity and rigidity suggest abrupt drug withdrawal, a potentially fatal situation [75].

4.5. Screening

There is no newborn screening for L1 syndrome or other monogenic hydrocephalus forms. Relevant screening is second-trimester obstetric ultrasound, measuring lateral ventricular atrial diameter and referring for targeted neurosonography when a central nervous system malformation or family history of hereditary malformation is suspected [76].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic criteria for L1 syndrome

There are no validated formal clinical criteria. Diagnosis is established in a boy with suggestive signs (hydrocephalus, adducted thumbs, spasticity, intellectual disability, callosal abnormalities) and a hemizygous pathogenic or likely pathogenic L1CAM variant; in a girl, particularly with global delay or hydrocephalus, by a heterozygous variant [1]. A hemizygous or heterozygous variant of uncertain significance (VUS) neither establishes nor excludes diagnosis [1]. A family history consistent with X-linked inheritance, without male-to-male transmission, supports diagnosis, but its absence does not exclude it, because approximately 40% of affected boys are simplex cases [1]. Imaging is suggestive but not diagnostic, and there is no confirmatory biochemical or functional test [7].

5.2. Pretest probability: detection rates

The probability of identifying an L1CAM variant depends on the number of typical signs and family history. Historical figures were obtained with less sensitive screening methods than current sequencing and measure detection rates in referred populations, not analytical sensitivity.

Oxford Desk Reference figures (75% and 15%) are close to Finckh’s (74.2% and 15.7%), the probable primary source [37, 49]. SMFM states that L1CAM variants explain up to 30% of boys with idiopathic X-linked hydrocephalus, citing Vos and colleagues [40]. In boys with non-syndromic congenital hydrocephalus, the proportion is 5–10% [1, 9].

5.3. Genetic testing algorithm

GeneReviews proposes two approaches [1]. When the phenotype is distinctive (hydrocephalus with adducted thumbs, spasticity and callosal abnormalities in a boy), L1CAM sequencing may be performed first, detecting small insertions and deletions and missense, nonsense and splicing variants, followed, if negative, by targeted deletion/duplication analysis using quantitative polymerase chain reaction (PCR), long-range PCR, MLPA (multiplex ligation-dependent probe amplification) or targeted microarray; alternatively, a multigene panel including L1CAM and differential diagnosis genes limits VUS, although composition and sensitivity vary between laboratories [1]. When the phenotype is indistinguishable from other disorders with hydrocephalus, spastic paraplegia or intellectual disability, exome or genome sequencing is preferred [1].

For congenital hydrocephalus generally, the Oxford Desk Reference clinical approach includes a three-generation pedigree extended on the maternal side (relatives with hydrocephalus, unexplained stillbirths, spasticity, cognitive impairment, neural tube defects), questions about consanguinity, pregnancy infections, ventriculomegaly on prenatal ultrasound, birth head circumference, perinatal history (intraventricular haemorrhage of prematurity, neonatal meningitis), developmental milestones and seizures [37]. Investigation includes MRI (periventricular leukomalacia, tumours and cysts, aqueductal stenosis, Chiari malformation, Dandy-Walker malformation, cobblestone lissencephaly, calcifications, hydranencephaly, atrophy), skull radiography if craniosynostosis is suspected, creatine kinase (CK, elevated to approximately 3000 IU/L in muscle-eye-brain disease and Walker-Warburg syndrome), ophthalmological assessment, congenital infection screening (TORCH) in infants under 6 months and genomic array, panel, exome or genome as appropriate; brain histology from affected relatives should be reviewed [37]. Jin and colleagues advocate routine exome or genome sequencing in newly diagnosed congenital hydrocephalus for diagnostic, prognostic, cancer surveillance (PIK3CA, PTEN) and recurrence-risk value [16].

5.4. Test limitations

L1CAM sequencing detects 99% of known variants and deletion/duplication analysis approximately 1%, but these figures refer to catalogued variants, not the proportion of clinically suspected patients testing positive [1]. Exome sequencing is not validated to detect all copy-number variants, so microarray should precede or accompany it [32]. Deep intronic variants escape standard exome sequencing: c.1124-24T>G was demonstrated through RNA studies in urine cells, with the mutant allele appearing in only 2% of clones and 5 of 18 RNA sequencing reads because of nonsense-mediated decay, and confirmed by minigene assay; urine-derived cells provide a non-invasive RNA source [52]. A negative maternal blood test does not exclude maternal somatic or germline mosaicism [1, 58]. A negative result does not exclude diagnosis, particularly in a family with several affected boys.

5.5. Report content and variant classification

The report should state the gene, transcript with version (NM_001278116.2, L1CAM MANE Select transcript), variant in c. and p. HGVS nomenclature, genomic coordinate, zygosity (hemizygous in a boy, heterozygous in a girl) and class according to ACMG/AMP (American College of Medical Genetics and Genomics and Association for Molecular Pathology) criteria [1, 8]. Truncating variants are classified as pathogenic except those near the stop codon, while missense variants require additional evidence [50]. Family segregation, prediction tools such as AlphaMissense and location at key Ig- and FnIII-domain residues may help reclassify VUS: a missense variant prioritised by AlphaMissense among four VUS in a fetal exome cosegregated in eight maternal relatives and was upgraded from VUS to likely pathogenic, enabling preimplantation genetic testing (PGT) [1, 57]. For autosomal genes, variant type affects prognosis: lethal WDR81 forms associate with truncating alleles [11]; in SMARCC1, inheritance from a healthy parent does not exclude pathogenicity [17].

5.6. Fetal ventriculomegaly: definitions and classification

Lateral ventricular atrial diameter is relatively constant from 14–15 to 40 weeks; the mean is 7.6 ± 0.6 mm according to Cardoza, and SMFM gives a mean of 5.4–7.6 mm, making 10 mm equivalent to 2.5–4 standard deviations above the mean [39, 40]. Measurement is inner edge to inner edge at the choroid plexus glomus, perpendicular to the cavity, and should remain below 10 mm throughout pregnancy (International Society of Ultrasound in Obstetrics and Gynecology, ISUOG, grade C) [76]. Classifications almost entirely coincide, differing at the 15 mm boundary.

Mild to moderate ventriculomegaly occurs in approximately 1% of pregnancies; it is unilateral in 50–60% and bilateral in 40–50%, and 65–75% of fetuses are male, without a known sex-based prognostic difference [40]. The Oxford Desk Reference gives mild ventriculomegaly incidence as 1.5–20 per 1000 fetuses, varying widely in the literature [39]. Progression is defined by an increase of 2 mm or more, transition from mild to moderate or unilateral to bilateral [33].

5.7. Investigation of fetal ventriculomegaly

SMFM, ISUOG and International Society for Prenatal Diagnosis (ISPD) guidelines and a 2025 Chinese consensus using GRADE (Grading of Recommendations Assessment, Development and Evaluation) agree on core investigation and differ in emphasis on MRI and exome sequencing.

Diagnostic yields vary by selection. In mild ventriculomegaly, the Oxford Desk Reference cites chromosomal abnormalities in 3–14% and, excluding maternal age-related risk, aneuploidy in approximately 3.5%; in isolated severe cases, chromosomal risk is low but amniocentesis is indicated [39]. SMFM estimates approximately 5% abnormal karyotype (mainly trisomy 21) in apparently isolated ventriculomegaly, plus 10–15% abnormal CMA, and approximately 5% congenital infection in mild to moderate cases [40]; CMV and toxoplasmosis explain 10–20% of severe ventriculomegaly [39]. The Chinese consensus reports total CMA yield of 9–16%, approximately 31% incremental exome yield after normal karyotype and CMA in structural anomalies (meta-analysis) and 45% in severe non-isolated cases, higher yield in bilateral than unilateral ventriculomegaly (16.5% vs 8.6%) and, in some cohorts, male fetuses (19.1% vs 5.5%) [33]. In a Chinese cohort of 73 fetuses with ventriculomegaly above 10 mm, copy-number variant sequencing was abnormal in 31.5% and exome provided a monogenic diagnosis in 16 of 33 negative cases (48.5%), involving genes such as TRIM71, PIK3R2 and CRB2; this is a selected tertiary-centre cohort, not a prevalence estimate [78].

Fetal MRI adds findings in 5–50% of cases (important in 1–14%), particularly callosal agenesis [40]; in one series, it changed diagnosis in 45 of 145 cases and was more useful at later gestational ages [39]. Testing for neonatal alloimmune thrombocytopenia should be considered [39]. Amniotic fluid CMV PCR sensitivity is 45–80% before 21 weeks, higher after 21 weeks or more than 6–7 weeks after primary infection, with specificity of 97–100% [40].

5.8. Prenatal and preimplantation diagnosis in L1 syndrome

With the familial variant identified, molecular prenatal diagnosis and PGT are possible; a variant in a hemizygous male fetus does not predict phenotype [1]. L1 syndrome cannot be reliably diagnosed by ultrasound alone: hydrocephalus often requires serial scans and cannot be assured before 20–24 weeks, or even the 3rd trimester [1]. Smith’s textbook suggests scans every 2–4 weeks between 16 and 28 weeks, noting that hydrocephalus may appear only after birth or never [3]. Linked-marker carrier detection, mentioned in the Oxford Desk Reference, is a historical method superseded by direct variant testing [37]. Non-invasive prenatal diagnosis using cell-free fetal DNA for monogenic disorders, including L1CAM X-linked hydrocephalus, is being assessed in a recruiting French observational study (NCT06147414) and is not routine practice [79]. Reproductive implications are developed in section 8.

5.9. Neuropathology and fetal autopsy

After pregnancy termination or fetal death, fetal DNA should be banked and autopsy with neuropathology recommended, or at least non-invasive autopsy using MRI, because ultrasound findings may not be confirmed [39]. Adle-Biassette’s series defined neuropathological criteria for fetal L1CAM testing: hydrocephalus with adducted thumbs, pyramidal tract agenesis or hypoplasia and aqueductal stenosis [29].

6. Differential Diagnosis

6.1. Principles

The differential follows three axes: cause of ventriculomegaly (obstructive, communicating, tissue loss or malformation), inheritance pattern suggested by the family and extraneurological signs. Patients with L1 syndrome and hydrocephalus have no other major physical anomalies, distinguishing it from syndromic hydrocephalus [1]. Non-syndromic congenital hydrocephalus also occurs with neural tube defects, isolated congenital aqueductal stenosis, central nervous system malformations (Chiari, Dandy-Walker, hydranencephaly, vein of Galen malformation, midline hyperplasia with fornix system malformation, congenital cysts) and posthaemorrhagic hydrocephalus, and the other monogenic causes of non-syndromic congenital hydrocephalus listed by GeneReviews are biallelic CCDC88C, MPDZ and WDR81 variants, the latter misspelt “WDR8” [1]. For MASA/SPG1 the differential is broader still, because approximately 50 genetic types of complicated hereditary spastic paraplegia and many intellectual disability forms are defined [1].

6.2. Comparative table

6.3. Principal differentials in prose

The closest differential to L1 syndrome is X-linked hydrocephalus without an L1CAM variant. In the French series of 79 fetuses with suspected X-linked hydrocephalus and no variant, 44% had hydrocephalus sometimes with callosal agenesis, 27% the aqueductal atresia or bifurcation and rhombencephalosynapsis spectrum, 9% syndromic hydrocephalus and 20% phenocopies with family history suggesting autosomal recessive inheritance [29]. AMOT offers an X-linked explanation in rare families [65], and FSD1L an autosomal recessive explanation for the full L1-like phenotype, including absent pyramidal decussation, relevant when counselling a consanguineous family or affected girl [66].

Among X-linked forms with intellectual disability and spasticity, the Oxford Desk Reference lists, besides L1CAM, Pelizaeus-Merzbacher disease, SPG2, Allan-Herndon-Dudley syndrome (SLC16A2) and genes such as ATRX, SMCX and MECP2 [71]. AP1S2 is distinguished by basal ganglia calcification, the essential parameter guiding testing [18].

According to Smith, Walker-Warburg syndrome has type 2 lissencephaly and cerebellar malformation in 100%, occipital encephalocele in 24%, Dandy-Walker malformation in 53%, hydrocephalus, usually from posterior fossa obstruction, in 53%, and ventriculomegaly even without intracranial hypertension in 95%, together with anterior chamber anomalies (91%), retinal anomalies (100%), coloboma (24%), microphthalmia (53%) and congenital muscular dystrophy (100%) [80]. The HARD±E variant (hydrocephalus, agyria, retinal dysplasia, with or without encephalocele) belongs to the same O-mannosylation group (POMT1 in Walker-Warburg, POMGNT1 in muscle-eye-brain disease) [37]. In the Saudi familial congenital hydrocephalus cohort, dystroglycanopathies explained 26% of diagnoses and ciliopathies 19% [15].

Hydrolethalus syndrome, with most cases in Finland, combines polyhydramnios, severe prenatal hydrocephalus, absent corpus callosum, septum pellucidum and olfactory structures, hypothalamic hamartoma, brainstem and cerebellar hypoplasia, a keyhole opening at the skull base (constant), micrognathia, clefts and polydactyly; 70% are stillborn and liveborn infants survive minutes to hours [81]. Hydranencephaly is distinguished by a thin cortical rim with normal cerebellum and brainstem, results from carotid disruption and may be confused with holoprosencephaly; an arachnoid cyst causes hydrocephalus through obstruction and reduced reabsorption and is generally sporadic [37, 39]. In holoprosencephaly, approximately 50% of cases are chromosomal, mainly trisomy 13, and SHH is the most frequent gene [39].

Chiari malformations enter the anatomical differential: type I (tonsillar descent) is rarely symptomatic in childhood and may accompany hydrocephalus and syringomyelia; type II (vermis, tonsils and medulla in the foramen magnum) accompanies myelomeningocele and is counselled as a neural tube defect; type III involves cerebellar descent into a posterior encephalocele and type IV is a form of cerebellar hypoplasia [37, 83]. In megalencephaly, ventricular volume should be assessed relative to its extent: in MCAP, macrocephaly frequently accompanies hydrocephalus potentially requiring shunting, while MPPH (megalencephaly-polymicrogyria-polydactyly-hydrocephalus) lacks vascular malformation and syndactyly [82]. Infantile Alexander disease causes death within the first decade [82]. In achondroplasia, head circumference should be monitored on dedicated charts every 1–2 months in the first years [37]. Finally, the extent to which neurological dysfunction is attributed to sequelae of in utero hydrocephalus governs the scope of aetiological investigation, and infants should be reassessed later to exclude age-emerging signs such as those of neurofibromatosis type 1 [37].

7. Treatments and Risk Management Strategies

7.1. Principles and team

There is no disease-modifying treatment for L1 syndrome or other monogenic hydrocephalus forms. Treatment combines neurosurgical hydrocephalus control, rehabilitation, spasticity treatment and developmental support through a multidisciplinary team including paediatrics, paediatric neurology, neurosurgery, rehabilitation medicine and clinical genetics [1]. In HSAS with severe hydrocephalus, shunting controls intracranial pressure but does not alter severe intellectual disability, and should be presented to families in those terms [1, 7].

7.2. Assessment after diagnosis

7.3. Surgical hydrocephalus treatment: principles

Hydrocephalus with intracranial hypertension and progressive hydrocephalus are treated surgically, classically with a ventriculoperitoneal (VP) shunt [37]. CSF shunting is indicated to reduce intracranial pressure; prenatal shunting has shown no advantage in L1 syndrome [1]. Genetic sources define no preferred gene-specific procedure, so choice follows general paediatric hydrocephalus guidelines and individual anatomy [7, 10].

7.4. AANS/CNS paediatric hydrocephalus guideline (2014, updated 2020)

The only systematic guideline found for surgical treatment of paediatric hydrocephalus is that of the joint American Association of Neurological Surgeons and Congress of Neurological Surgeons (AANS/CNS) Task Force, published in 2014 (search from 1966 to March 2012) and updated in 2020 (search March 2012 to November 2019) [34, 84, 85]. It uses three levels: level I (class I evidence, high certainty), level II (class II or strong class III consensus, moderate certainty) and level III (inconclusive or conflicting evidence, unclear certainty) [85]. The introduction estimates hydrocephalus in 1 in 500 children and cites X-linked aqueductal stenosis as a genetic cause, but the guideline does not address genetic testing [85]. The review scheduled five years after the update had not been published at consultation [85]. No equivalent European guideline was found.

The shunt-versus-ETV recommendation relies mainly on Kulkarni’s comparison (class III): adjusting for age and aetiology, early failure was greater with ETV (n=489, retrospective) than VP shunting (n=720, from two trials), but after 3 months ETV failure was lower; age, aetiology and prior surgery, ETV success factors, were not independently assessed, and ETV with choroid plexus cauterisation (ETV-CPC) was mentioned as an emerging technique without recommendation [34, 88]. After successful ETV, ventricular reduction may be small or absent [94].

The antibiotic prophylaxis meta-analysis combined nine studies (seven underpowered randomised trials): infection in 72 of 674 (10.7%) without prophylaxis and 38 of 643 (5.9%) with prophylaxis, relative risk (RR) 0.55 (95% CI 0.38–0.81), absolute reduction 4.8% [90]. The antibiotic-impregnated catheter meta-analysis found infection in 205 of 2396 (8.6%) with standard systems and 59 of 1078 (5.5%) with impregnated catheters, RR 0.51 (95% CI 0.29–0.89), number needed to treat 24 [91].

7.5. Application to genetic forms

The consulted sources do not compare ETV and shunting in L1 syndrome or other monogenic forms; decisions follow usual ETV success factors (age, aetiology, previous surgery) [88, 95]. In a Hydrocephalus Clinical Research Network (HCRN) study of 761 children, ETV success at 6 months was 76%, and the original ETV Success Score retained good predictive performance [95]. For ETV-CPC in infants under 2 years (191 patients, 17% with aqueductal stenosis), success was 48%, 46% and 45% at 6, 12 and 18 months [96].

In CCDC88C, Wallis’s series associated shunting in the first weeks of life and absence of multiple revisions with better development; this was a two-family observational series [9]. In SMARCC1, reported cases received shunting or ETV [12]. In MPDZ, demonstration of obstruction upstream of diencephalosynapsis changed the approach: in a Brazilian infant with diencephalosynapsis, 3rd ventricular atresia and secondary aqueductal occlusion, a ventriculoamniotic shunt was placed at 29 weeks, ETV performed on day 5 and endoscopic stoma revision at 2 years and 5 months for Liliequist membrane occlusion; at 4 years and 5 months she was VP-shunt-free with age-appropriate development [70]. This is a single case presented as proof of concept, and VP shunting remains usual MPDZ treatment [70]. Jin and colleagues speculated that, in non-obstructive congenital hydrocephalus with a thin cortical mantle and normal pressure, as in altered-neurogenesis forms, shunting might not improve neurodevelopment; this is a hypothesis, not a recommendation [16].

7.6. Fetal intervention

Prenatal shunting has shown no advantage in L1 syndrome [1]. A 2025 systematic review combined 12 studies with 172 fetuses (cephalocentesis in 29, ventriculoamniotic shunting in 131, fetal ETV attempted in 14, successful in 11). Hydrocephalus appeared prenatally isolated in 93.0%, but 11.3% of these had another postnatal anomaly; ventriculoamniotic shunting had perinatal complications in 6.9% (four in utero deaths and three preterm deliveries with death) and failed in 32.1%; among 137 children assessed at 1 to 38 months, 46.0% had normal development and 45.3% moderate or severe delay, and none of the 24 with another anomaly had normal development [97]. Evidence quality is low, without a control group, and fetal intervention remains experimental, to be reassessed in a randomised trial with fetal MRI and exome or genome sequencing [97].

7.7. Delivery

Timing and mode of delivery should follow obstetric indications, without evidence that preterm birth or caesarean section improves ventriculomegaly outcomes (SMFM, 1C); macrocephaly requiring caesarean section is rare [33, 40]. In HSAS, prenatal hydrocephalus may be severe enough to prevent vaginal delivery [3].

7.8. Rehabilitation, development and communication

GeneReviews describes the US model, which may vary between countries: from 0 to 3 years, early intervention with occupational therapy, physiotherapy, speech and feeding therapy, child mental health care and special education; from 3 to 5 years, developmental preschool with an individualised education plan; at all ages, developmental paediatric follow-up, annual education plan review, the least restrictive school environment possible and, in adolescence, transition planning, with school services until 21 years in the USA [1]. Physiotherapy aims to maximise mobility and reduce contractures, scoliosis and hip dislocation; adapted equipment includes wheelchairs, walkers, orthoses, bath chairs and pushchairs [1]. Occupational therapy addresses fine motor skills [1]. Oromotor dysfunction should be assessed at every consultation; feeding therapy may include thickening or cooling foods, and nasogastric feeding or gastrostomy are options in severe cases [1]. Augmentative and alternative communication, from low to high technology, should be assessed and does not impair speech development [1]. Autism spectrum disorders benefit from interventions such as applied behaviour analysis, attention deficit hyperactivity disorder may require medication, and severe aggression warrants child psychiatry [1].

7.9. Adducted thumbs

Splinting may reduce adduction; surgery is generally not indicated, but in milder cases tendon transfer may improve function and has been used successfully in some patients [1].

7.10. Pharmacological treatment of spasticity

GeneReviews refers hypertonia and dystonia to specialists who may consider baclofen, tizanidine, botulinum toxin, antiparkinsonian medicines or orthopaedic surgery, without specifying doses [1]. None has a specific L1 syndrome indication; use in this disorder is off-label for the specific indication when the approved indication is spasticity from another cause, and falls within the general spasticity indication when the label defines it that way. Doses below are from consulted US Food and Drug Administration (FDA) labels; European and national summaries of product characteristics were not consulted.

Intrathecal baclofen requires a prior screening-dose response; in spinal spasticity the pump is reserved for oral baclofen non-responders or intolerance, and after traumatic brain injury at least one year should elapse [75]. Above 12 years, dose requirements do not appear to differ from adults [75].

7.11. Regulatory position

No medicine is authorised for L1 syndrome or congenital hydrocephalus, and AANS/CNS advises against acetazolamide and furosemide to prevent shunting in posthaemorrhagic hydrocephalus of prematurity (level I) [34, 86]. The guideline recommends preoperative antibiotics for shunt surgery without specifying medicine or dose [90].

7.12. Trials and intervention studies

These studies cannot be compared indirectly because populations (age, aetiology, resource setting), failure definitions and follow-up differ. Relative infection reduction with impregnated catheters (approximately 60% in BASICS) corresponds to an absolute reduction of approximately 4 percentage points from a 6% rate [74]. The BASICS silver arm was negative [74], Mbabazi’s distal-slit catheter trial in a low-resource country showed no advantage [91], and Kumar’s trial found no difference between techniques [102]. None stratifies results by genetic cause, and the Ugandan study concerns post-infectious hydrocephalus and cannot be transferred to genetic forms [103].

7.13. Developments in the last 12 months (October 2025 to October 2026)

Developments in this period are mainly genetic and methodological; none yet changes treatment.

Immediately before the period, 2025 publications included AMOT (September) [65], ARID1A hydrocephalus with escape from nonsense-mediated decay (September) [68], obstructive reinterpretation of MPDZ (June) [30], the fetal surgery systematic review [97] and Chinese fetal ventriculomegaly consensus [33]. An October 2026 ClinicalTrials.gov search found no therapeutic trial targeting L1 syndrome or monogenic hydrocephalus (L1CAM results concerned oncology cell therapies); absence of registrations does not prove absence of research, and the European clinical trials registry was not consulted [107]. L1-mimetic compounds remain preclinical [59].

7.14. Agents and practices to avoid

Routine serial lumbar punctures, intraventricular thrombolytics, acetazolamide and furosemide are not recommended in posthaemorrhagic hydrocephalus of prematurity (level I) [34, 86]. Abrupt intrathecal baclofen withdrawal should be avoided [75]. Different botulinum toxin units are not interchangeable [98, 99]. Fetal shunting outside research is not recommended [1, 97].

7.15. Pregnancy, lactation, perioperative care and transition

The consulted sources describe no pregnancy in women with L1 syndrome; affected men rarely reproduce [1]. In an at-risk pregnancy, ultrasound follow-up and diagnostic options are those in section 8, and delivery follows obstetric indications [40]. Perioperative shunt surgery requires prophylactic antibiotics and antibiotic-impregnated catheters [34, 90]. In a patient with an intrathecal baclofen pump, any surgery or admission must maintain infusion continuity [75]. Transition to adult care should be planned in adolescence [1]; in adults with a spastic paraplegia phenotype, ERN-RND, with its hereditary spastic paraplegia module, is the relevant European network [36].

8. Reproductive Options and Family Counselling

8.1. L1 syndrome: X-linked inheritance

The father of an affected boy is neither affected nor hemizygous and does not require evaluation [1]. In a family with more than one affected individual, the mother is an obligate heterozygote; if a woman has more than one affected son, without other affected relatives, and the variant is not detected in her leucocytes, germline mosaicism is most likely [1]. In a simplex case, the mother may be heterozygous, the variant may have arisen de novo in the boy, or maternal somatic or germline mosaicism may exist; approximately 40% of affected boys are simplex cases, and maternal molecular testing is recommended [1]. The classic rule for lethal or non-reproductive X-linked disorders assigns, before testing, approximately a two-thirds probability that the mother of an isolated case is a carrier and one-third that the variant is de novo; this was the estimate used by the authors of the section 10 case before the mother proved heterozygous [48]. Paternal somatic and germline mosaicism has been described in an unaffected father of two heterozygous daughters [1].

8.2. Recurrence risks by scenario

The Oxford Desk Reference presents the same empirical risk in two formulations, “up to 5%” in the hydrocephalus chapter and “approximately 5%” in the ventriculomegaly chapter; both derive from the same older series and should be communicated as an order of magnitude rather than an exact value [37, 39]. When index-case information is incomplete, risk cannot be estimated precisely [39]. Recurrence of isolated ventriculomegaly generally is low but depends on diagnosis when a genetic cause exists [40].

8.3. Non-L1 forms

In autosomal recessive forms (CCDC88C, MPDZ, WDR81, WDR91, FSD1L), parents of an affected individual are obligate heterozygotes and recurrence risk is 25% per pregnancy under the Mendelian rule; sources give no gene-specific figures [9–11]. In WDR81, variant type influences expected phenotype (lethal forms with truncating alleles) [11]. SMARCC1 inheritance is autosomal dominant with incomplete penetrance: inheritance from a healthy parent does not exclude pathogenicity, and variant reclassification enabled informed reproductive decisions [17]. For de novo variants (TRIM71, de novo SMARCC1, TRIO, ARID1A), recurrence risk depends on confirming absence from both parents [16, 67]; germline mosaicism documented in L1 syndrome demonstrates that absence from parental blood does not eliminate that risk [58]. In X-linked AP1S2, a heterozygous mother transmits the variant in 50% of pregnancies; boys inheriting it are affected and girls are carriers, generally unaffected [6]. Without a proband sample, sequencing both parents may be offered to seek shared carrier status for an autosomal recessive disorder, but tissue from the previous fetus or child is preferable [32].

8.4. Carrier detection and cascade testing

Heterozygote detection requires knowledge of the familial variant or, if no affected boy is available, L1CAM sequencing followed by deletion/duplication analysis [1]. Cascade testing extends to the proband’s sisters and maternal aunts and to daughters of heterozygous women [1]. Counselling should ideally precede pregnancy and be offered to young adult heterozygotes or those at risk [1]. Carrier testing in asymptomatic minors is not indicated; in the section 10 case, the daughter born in the next pregnancy was not tested, with counselling planned for adulthood [48].

8.5. Reproductive options

With the familial variant identified, molecular prenatal diagnosis and PGT are possible, the latter only within in vitro fertilisation [1, 48]. The first molecular prenatal diagnosis was at 10 weeks’ gestation [46]. Fetal sex determination through cell-free fetal DNA in maternal plasma may avoid an invasive procedure for a female fetus; some authors propose sex determination followed by L1CAM testing only in male fetuses [48]. Non-invasive prenatal diagnosis of the variant itself is investigational [79]. Presence of the variant in a male fetus does not predict severity because intrafamilial variability is broad [1]. Perspectives on prenatal testing differ and the decision is personal [1].

8.6. Ultrasound surveillance in an at-risk pregnancy

Without molecular diagnosis or when a family declines invasive testing, ultrasound surveillance has limitations requiring explanation: ventriculomegaly usually begins after 20 weeks, may appear only near term or after delivery, and a normal scan does not exclude disease [1, 37]. Smith’s textbook proposes scans every 2–4 weeks between 16 and 28 weeks, and the Chinese consensus recommends every 2–4 weeks for identified ventriculomegaly [3, 33].

8.7. Counselling after prenatal ventriculomegaly diagnosis

Prognosis depends on cause, gestational age at diagnosis and progression. In Gaglioti’s aggregated data, outcome was normal in 94% of fetuses with isolated mild ventriculomegaly, 76% with moderate and 28% with severe disease, a small group because of pregnancy terminations [39]. SMFM recommends counselling a favourable outcome for isolated mild ventriculomegaly (10–12 mm) after full assessment (1B), and a probably favourable outcome but increased neurodevelopmental disability risk for isolated moderate disease (13–15 mm) (1B) [40]. After autopsy without genetic or chromosomal disease signs, counselling follows isolated hydrocephalus, using table 8.2 empirical risks [39]. Support organisations for decisions after prenatal diagnosis, such as the British Antenatal Results and Choices, are listed in the Oxford Desk Reference [39].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Congenital hydrocephalus: what each number measures

Published congenital hydrocephalus figures vary fourfold because they measure different populations and definitions. The table collects them without merging them.

Orphanet values are exactly those of California (5.9/10 000 = 59/100 000) and EUROCAT (4.65/10 000 = 46.5/100 000), expressed in different units [24, 27, 28]. The difference between these values, or the Oxford Desk Reference 4–8/10 000, and “1 in 500” probably reflects definition: registries measure congenital hydrocephalus recognised at birth or during pregnancy, whereas 1/500 appears to concern childhood hydrocephalus generally, including forms acquired after the neonatal period [27, 38]. Liu and colleagues estimate that genetic factors contribute to up to 40% of cases but a precise genetic cause is identified in fewer than 5% [38]; in Jin’s surgical cohort, damaging de novo variants explained more than 17% and rare large-effect variants approximately 22%, showing identified proportions depend on method and selection [16].

In California, spina bifida-associated hydrocephalus fell from 1.4 to 0.9/10 000 and obstructive hydrocephalus rose from 0.5 to 1.0/10 000 during the decade; risk factors were birthweight below 1500 g (OR 51.6; 95% CI 47.7–55.8) and between 1500 and 2000 g (OR 14.1; 12.4–16), low socioeconomic status (OR 1.5; 1.4–1.6) and male sex (OR 1.2; 1.1–1.3); Asian newborns had lower risk than white newborns (OR 0.7; 0.6–0.8), and 13% died before discharge [28]. In EUROCAT, 47% of pregnancies resulted in live birth, 5% fetal death and 48% termination (median gestational age 21 weeks); 44% had major non-cerebral malformations and 9% karyotype abnormalities; among live births, 61% had prenatal diagnosis and 34% died in the first year, mainly during the first week [27].

9.2. L1 syndrome

L1 syndrome frequency estimates are also heterogeneous. GeneReviews describes HSAS as the most common genetic congenital hydrocephalus form, with estimated prevalence of 1:30,000 and accounting for 5–10% of boys with non-syndromic congenital hydrocephalus [1]. Fernández and colleagues cite incidence of 1/30 000 male births in an introductory statement without their own method [56]. Orphanet records HSAS prevalence as 1–9/100 000, mean 1.7/100 000, not yet validated, supported by that article and GeneReviews [20]. The value of 1.7/100 000 is compatible with 1/30 000 boys converted to all births; GeneGuidance converts 1:30,000 to 3.3/100 000 without distinguishing the denominator [7]. None of these estimates rests on a dedicated population study identified in the consulted sources, and all concern HSAS, so total L1 disease frequency (including MASA and SPG1) is higher; L1 syndrome prevalence in men with complicated spastic paraplegia is unknown [1, 19, 21]. For X-linked complicated dysgenesis of the corpus callosum, Orphanet records less than 1/1 000,000 and 11 cases or families [23].

9.3. Non-L1 forms and founder populations

Autosomal recessive forms cluster in highly consanguineous populations. The MPDZ founder variant was found in two Saudi families, and exome sequencing identified a probable cause in 21 of 27 families (78%) in the Saudi familial hydrocephalus cohort, none X-linked [14, 15]. In Jin’s cohort, consanguinity was 8.7% in cases versus 1.3% in controls [16]. In the Belgian series, the multiethnic cohort included French, Finnish, Pakistani, Belgian and Moroccan patients, and 9 of 28 probands had consanguineous parents [64]. Hydrolethalus syndrome is the European founder-effect example, with most cases in Finland [81]. No prevalence is published for CCDC88C, MPDZ, WDR81, TRIM71 or SMARCC1; Orphanet records congenital communicating and non-communicating hydrocephalus as prevalence unknown, and AP1S2 disease as less than 1/1 000,000 [6, 12, 25, 26].

9.4. Europe, Portugal and Brazil

European data derive from EUROCAT, the French neuropathological series, Groningen’s Dutch series and the Belgian cohort [27, 29, 58, 64]. Consulted sources contain no prevalence estimates for Portugal or Brazil. The only Brazilian datum identified is the São Paulo report of the first primary ETV in a liveborn infant with MPDZ hydrocephalus [70]. Absence from consulted sources does not demonstrate absence of published data.

9.5. Special populations

L1CAM- and AP1S2-heterozygous women are generally unaffected but may manifest disease through skewed X-inactivation [1, 6]. Premature infants form a separate group because posthaemorrhagic hydrocephalus has dedicated guidance and birthweight below 1500 g is the principal population risk factor [28, 86]. Surgical data from low-resource countries such as Uganda mainly concern post-infectious hydrocephalus [103].

9.6. European networks and patient associations

ERN-ITHACA (European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders) is the relevant European network for L1 syndrome and syndromic or malformation congenital hydrocephalus [35]. ERN-RND (European Reference Network on Rare Neurological Diseases), through its ataxias and hereditary spastic paraplegias module, is relevant to adolescent and adult SPG1 phenotypes [36]. Associations include the Hydrocephalus Association (USA) [1, 108], the International Federation for Spina Bifida and Hydrocephalus [109], Shine, the British spina bifida and hydrocephalus association [37, 110], and Associação Spina Bifida e Hidrocefalia de Portugal (ASBIHP) [111]. The L1CAM variant database was a reference resource for variant interpretation [1, 50]. No L1 syndrome-specific patient association or Brazilian hydrocephalus association was identified in the consulted sources.

10. Illustrative Clinical Case

10.1. X-linked hydrocephalus diagnosed in the fetal period (Horiuchi and colleagues, 2025)

A 29-year-old Japanese primigravida, whose pregnancy followed ovulation induction and without relevant personal or family history, underwent routine ultrasound at 20 weeks, showing a lateral ventricular posterior horn of 16 mm (article Figure 1: https://pmc.ncbi.nlm.nih.gov/articles/PMC12353247/figure/ccr370792-fig-0001/) [48]. During follow-up, ventriculomegaly progressed and the cortex thinned. Fetal MRI at 34 weeks showed marked lateral and 3rd ventricular enlargement with suspected aqueductal stenosis and an adducted flexed left thumb on a sagittal T2-weighted image (Figure 2: https://pmc.ncbi.nlm.nih.gov/articles/PMC12353247/figure/ccr370792-fig-0002/); the fetus was male [48]. Prenatal genetic testing was offered and declined. Biparietal diameter was 110.1 mm (+7.15 standard deviations) [48].

Elective caesarean section was performed at 36 weeks and 3 days for breech presentation. The newborn had Apgar scores of 7 and 9, weight 3054 g (+1.94 standard deviations) and head circumference 44.2 cm (+9.5 standard deviations) [48]. In neonatal intensive care on day 2, a CSF reservoir was inserted for intracranial hypertension; clinically suspected Hirschsprung disease was surgically confirmed at 38 days, and at 48 days a ventriculoperitoneal shunt was inserted, with no subsequent hydrocephalus progression [48]. The course involved severe developmental delay and spastic paraplegia under paediatric follow-up [48].

Genetic testing identified the novel hemizygous variant NM_000425.3:c.2248dup, corresponding to NM_001278116.2:c.2248dup p.(Tyr750LeufsTer36) on the MANE transcript (NC_000023.11:g.153866832dup), a frameshift in the second fibronectin type III domain, classified as likely pathogenic using ACMG/AMP criterion PVS1 [8, 48]. The mother was heterozygous (pedigree in Figure 3: https://pmc.ncbi.nlm.nih.gov/articles/PMC12353247/figure/ccr370792-fig-0003/; Sanger chromatogram in Figure 4: https://pmc.ncbi.nlm.nih.gov/articles/PMC12353247/figure/ccr370792-fig-0004/) [48]. Captions of all four figures were consulted in PubMed Central; the article has a CC BY 4.0 licence.

Before maternal testing, the authors used the classic X-linked disorder rule to estimate approximately one-third probability of a de novo variant and two-thirds that the mother was a carrier [48]. In the next pregnancy, conceived spontaneously two years later, non-invasive prenatal cell-free fetal DNA testing indicated a female fetus and avoided an invasive procedure; ultrasound showed no hydrocephalus and a healthy girl was born by repeat caesarean section at 38 weeks. The daughter’s carrier status was not tested, with counselling planned in adulthood [48].

10.2. Clinical interpretation of the case

The case brings together the central elements of this article: progressive ventriculomegaly detected at 20 weeks, late recognition of the adducted thumb (on MRI at 34 weeks, consistent with the sign occurring in only half of fetuses before 24 weeks), associated Hirschsprung disease, a truncating variant with a severe phenotype, shunting controlling hydrocephalus without preventing severe delay, and maternal counselling with a fetal sex determination strategy [1, 48, 58]. The authors discuss caesarean section for macrocephaly, citing Kuller; this should not be read as routine, because SMFM recommends delivery timing and mode according to obstetric indications, and this case had an obstetric indication (breech presentation) [40, 48]. Cell-free fetal DNA sex determination is a strategy described by the authors and does not replace molecular variant diagnosis in a male fetus [48].

11. Prognosis and Long-Term Follow-up

11.1. L1 syndrome prognosis

Prognosis depends on phenotype and, as a trend, variant type. In HSAS, prenatal hydrocephalus may cause fetal or early childhood death, and survivors with severe hydrocephalus usually have severe intellectual disability, with IQ below 30, regardless of shunting [1, 3]. Mild to moderate ventriculomegaly is compatible with long survival, and MASA IQ ranges from 30 to 70 [1, 3]. In Vos’s series, death before 3 years occurred in 52% of children with truncating variants and 8% with missense variants (p=0.02), but intrafamilial variability precludes individual prediction [1, 58]. Spasticity tends to progress slowly in adulthood, with contractures and shuffling gait [1]. Orphanet classifies HSAS as the severe extreme with an unfavourable prognosis, and complicated SPG1 as the mild extreme [20, 22].

11.2. Prognosis of non-L1 forms and congenital hydrocephalus generally

In WDR81, most HYC3 fetuses die in utero or soon after birth, whereas CAMRQ2 is compatible with adulthood [11]. In CCDC88C, outcomes are not always unfavourable, with normal development at 3 years and 3 months in the first surviving patient and milestones achieved with early shunting [9, 13]. In hydrolethalus syndrome, 70% are stillborn [81]. In EUROCAT congenital hydrocephalus without neural tube defects, 34% of liveborn infants died within the first year [27]; in California, 13% died before discharge [28]. These figures mix genetic and acquired causes and should not be applied to a specific gene.

For fetal ventriculomegaly, SMFM summarises: in isolated mild cases, survival of 93–98% and normal development in more than 90%, with delay in 7.9% in a meta-analysis (similar to the population) and previously undetected postnatal findings in 7.4%; in isolated moderate cases, survival of 80–97% and normal development in 75–93%; progression, occurring in approximately 16%, is associated with adverse outcomes in up to 44%, and regression with normal outcomes in more than 90% [40]. Approximately two-thirds of isolated mild ventriculomegaly resolves before birth [39]. The Chinese consensus reports neurodevelopmental abnormalities in approximately 8–11% of isolated mild cases, relative delay risk of 4.24 in severe versus mild disease (approximately 45% delayed) and, in a 324-case cohort, higher risk with diagnosis by 24 weeks and 6 days (OR 2.86), non-isolated presentation (OR 2.62) and intrauterine progression (OR 11.15); no validated prognostic score exists [33].

11.3. Prognosis of surgical treatment

ETV success at 6 months was 76% in children [95], and infant ETV-CPC success was 48%, 46% and 45% at 6, 12 and 18 months [96]. After 3 months, ETV failure is lower than shunt failure in the adjusted analysis cited by the guideline [34, 88]. Infection affects up to 15% of shunted patients; in BASICS, revision for infection was 6% with standard catheters and 2% with antibiotic-impregnated catheters over a median 22 months [74]. Ventricular size is not a validated success measure, so follow-up is clinically and radiologically individualised [34, 94].

11.4. Surveillance table

GeneReviews specifies no surveillance intervals for L1 syndrome, recommending neurological assessment at regular individualised intervals [1]. Table intervals without a specific source are therefore described as individualised.

In non-L1 forms, surveillance adds cardiac (SMARCC1, MPDZ), ophthalmological (MPDZ), hearing (TRIM71) and basal ganglia calcification assessment, the latter emerging in childhood and potentially progressing (AP1S2); sources do not define intervals [6, 10, 12, 16]. For patients with PIK3CA or PTEN variants identified during hydrocephalus investigation, syndrome-specific cancer surveillance is an additional reason for genetic diagnosis [16].

12. Summary and Final Considerations

L1 syndrome is the best-known monogenic congenital hydrocephalus form and the model for the others. It results from loss of function of the L1 adhesion molecule encoded by L1CAM at Xq28 (MANE transcript NM_001278116.2), and manifests along a continuum historically divided into HSAS, MASA, SPG1 and X-linked complicated agenesis of the corpus callosum, grouped since 1995 under CRASH. Recognition rests on a boy with hydrocephalus, particularly from aqueductal stenosis, adducted thumbs, spasticity, intellectual disability, callosal abnormalities and absent medullary pyramids, and a family history consistent with X-linked inheritance, although approximately 40% of cases are simplex. Variant detection probability rises from approximately 16% to 85% when three or more typical signs combine with multiple affected relatives.

Test limitations govern counselling. Sequencing and deletion/duplication analysis detect the great majority of known variants, but deep intronic variants require genome or RNA studies, missense variants require segregation and structural evidence, and maternal germline mosaicism maintains recurrence risk even when the mother tests negative. Truncating variants associate, as a group trend, with higher early mortality, but intrafamilial variability prevents individual prediction, including prenatally.

Beyond L1CAM, monogenic congenital hydrocephalus has become a heterogeneous group: autosomal recessive forms with distinctive imaging markers (medial diverticulum in CCDC88C, diencephalosynapsis with 3rd ventricular obstruction in MPDZ, cerebellar hypoplasia in WDR81), de novo or incompletely penetrant forms related to neurogenesis (TRIM71, SMARCC1) and AP1S2 syndrome with basal ganglia calcification. Exome cohorts showed that a relevant fraction of congenital hydrocephalus results from altered neurogenesis rather than CSF flow obstruction alone, supporting microarray followed by trio exome or genome sequencing in congenital hydrocephalus without an evident cause, with implications for prognosis, surveillance and recurrence risk.

Treatment continues to address hydrocephalus and its consequences: shunting or endoscopic ventriculostomy selected by general neurosurgical criteria, antibiotic prophylaxis and antibiotic-impregnated shunt catheters, rehabilitation, spasticity treatment outside the specific indication with botulinum toxin or baclofen, and educational and communication support. Shunting controls pressure but does not alter intellectual disability in severe HSAS. No disease-modifying therapies are in clinical trials; L1 mimetics remain preclinical. Developments in 2025–2026 are mainly genetic, with FSD1L as an autosomal recessive L1 syndrome phenocopy, AMOT as a rare X-linked cause, TRIO and WDR91 as candidate genes and the first successful primary ETV in an MPDZ patient.

13. References

1. Stumpel C, Vos YJ. L1 Syndrome. 2004 Apr 28 [updated 2021 Jan 7]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1484/ PMID: 20301657.
2. Fransen E, Lemmon V, Van Camp G, Vits L, Coucke P, Willems PJ. CRASH syndrome: clinical spectrum of corpus callosum hypoplasia, retardation, adducted thumbs, spastic paraparesis and hydrocephalus due to mutations in one single gene, L1. Eur J Hum Genet. 1995;3(5):273-284. doi:10.1159/000472311. PMID: 8556302.
3. Jones KL, Jones MC, del Campo M. X-linked hydrocephalus spectrum (X-linked hydrocephalus syndrome, MASA syndrome, L1 syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 252–253.
4. Jouet M, Moncla A, Paterson J, McKeown C, Fryer A, Carpenter N, et al. New domains of neural cell-adhesion molecule L1 implicated in X-linked hydrocephalus and MASA syndrome. Am J Hum Genet. 1995;56(6):1304-1314. PMID: 7762552. Available from: https://pubmed.ncbi.nlm.nih.gov/7762552/
5. Online Mendelian Inheritance in Man (OMIM). Entries 307000, 303350, 304100, 236600, 615219, 617967, 618667, 620241, 308840, 618570, 601732. Baltimore: Johns Hopkins University. Available from: https://omim.org [consulted via NCBI E-utilities db=omim and db=gene, 2026-10-09].
6. GeneGuidance. AP1S2 — gene clinical curation. Version of 2026-09-26, consulted on 2026-10-09.
7. GeneGuidance. L1CAM — gene clinical curation. Version of 2026-09-26, consulted on 2026-10-09.
8. 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):61-68. doi:10.1002/humu.23348. Tool: https://variantvalidator.org (GRCh38) [accessed 2026-10-09].
9. GeneGuidance. CCDC88C — gene clinical curation. Version of 2026-09-26, consulted on 2026-10-09.
10. GeneGuidance. MPDZ — gene clinical curation. Version of 2026-09-26, consulted on 2026-10-09.
11. GeneGuidance. WDR81 — gene clinical curation. Version of 2026-09-26, consulted on 2026-10-09.
12. GeneGuidance. SMARCC1 — gene clinical curation. Version of 2026-09-26, consulted on 2026-10-09.
13. Ekici AB, Hilfinger D, Jatzwauk M, Thiel CT, Wenzel D, Lorenz I, et al. Disturbed Wnt Signalling due to a Mutation in CCDC88C Causes an Autosomal Recessive Non-Syndromic Hydrocephalus with Medial Diverticulum. Mol Syndromol. 2010;1(3):99-112. doi:10.1159/000319859. PMID: 21031079.
14. Al-Dosari MS, Al-Owain M, Tulbah M, Kurdi W, Adly N, Al-Hemidan A, et al. Mutation in MPDZ causes severe congenital hydrocephalus. J Med Genet. 2013;50(1):54-58. doi:10.1136/jmedgenet-2012-101294. PMID: 23240096.
15. Shaheen R, Sebai MA, Patel N, Ewida N, Kurdi W, Altweijri I, et al. The genetic landscape of familial congenital hydrocephalus. Ann Neurol. 2017;81(6):890-897. doi:10.1002/ana.24964. PMID: 28556411.
16. Jin SC, Dong W, Kundishora AJ, Panchagnula S, Moreno-De-Luca A, Furey CG, et al. Exome sequencing implicates genetic disruption of prenatal neuro-gliogenesis in sporadic congenital hydrocephalus. Nat Med. 2020;26(11):1754-1765. doi:10.1038/s41591-020-1090-2. PMID: 33077954.
17. Hourvitz N, Kurolap A, Mory A, Haratz KK, Kidron D, Malinger G, et al. SMARCC1 is a susceptibility gene for congenital hydrocephalus with an autosomal dominant inheritance mode and incomplete penetrance. Prenat Diagn. 2023;43(10):1374-1377. doi:10.1002/pd.6426. PMID: 37639281.
18. Saillour Y, Zanni G, Des Portes V, Heron D, Guibaud L, Iba-Zizen MT, et al. Mutations in the AP1S2 gene encoding the sigma 2 subunit of the adaptor protein 1 complex are associated with syndromic X-linked mental retardation with hydrocephalus and calcifications in basal ganglia. J Med Genet. 2007;44(11):739-744. doi:10.1136/jmg.2007.051334. PMID: 17617514.
19. Orphanet. L1 syndrome (ORPHA:275543). Available from: https://www.orpha.net/en/disease/detail/275543
20. Orphanet. Hydrocephalus with stenosis of the aqueduct of Sylvius (ORPHA:2182). Available from: https://www.orpha.net/en/disease/detail/2182
21. Orphanet. MASA syndrome (ORPHA:2466). Available from: https://www.orpha.net/en/disease/detail/2466
22. Orphanet. X-linked complicated spastic paraplegia type 1 (ORPHA:306617). Available from: https://www.orpha.net/en/disease/detail/306617
23. Orphanet. X-linked complicated corpus callosum dysgenesis (ORPHA:1497). Available from: https://www.orpha.net/en/disease/detail/1497
24. Orphanet. Congenital hydrocephalus (ORPHA:2185). Available from: https://www.orpha.net/en/disease/detail/2185
25. Orphanet. Congenital communicating hydrocephalus (ORPHA:269505). Available from: https://www.orpha.net/en/disease/detail/269505
26. Orphanet. Congenital non-communicating hydrocephalus (ORPHA:269510). Available from: https://www.orpha.net/en/disease/detail/269510
27. Garne E, Loane M, Addor MC, Boyd PA, Barisic I, Dolk H. Congenital hydrocephalus--prevalence, prenatal diagnosis and outcome of pregnancy in four European regions. Eur J Paediatr Neurol. 2010;14(2):150-155. doi:10.1016/j.ejpn.2009.03.005. PMID: 19410489.
28. Jeng S, Gupta N, Wrensch M, Zhao S, Wu YW. Prevalence of congenital hydrocephalus in California, 1991-2000. Pediatr Neurol. 2011;45(2):67-71. doi:10.1016/j.pediatrneurol.2011.03.009. PMID: 21763944.
29. Adle-Biassette H, Saugier-Veber P, Fallet-Bianco C, Delezoide AL, Razavi F, Drouot N, et al. Neuropathological review of 138 cases genetically tested for X-linked hydrocephalus: evidence for closely related clinical entities of unknown molecular bases. Acta Neuropathol. 2013;126(3):427-442. doi:10.1007/s00401-013-1146-1. PMID: 23820807.
30. Cabet S, Ghersi-Egea JF, Khung-Savatovsky S, Guimiot F, Putoux A, Sabatier I, et al. MPDZ Pathogenic Variants Cause Obstructive Ventriculomegaly Related to Diencephalosynapsis and Third Ventricle Atresia. Genes (Basel). 2025;16(6):707. doi:10.3390/genes16060707. PMID: 40565599.
31. Singh AK, Allington G, Viviano S, McGee S, Kiziltug E, Ma S, et al. A novel SMARCC1 BAFopathy implicates neural progenitor epigenetic dysregulation in human hydrocephalus. Brain. 2024;147(4):1553-1570. doi:10.1093/brain/awad405. PMID: 38128548.
32. Van den Veyver IB, Chandler N, Wilkins-Haug LE, Wapner RJ, Chitty LS, ISPD Board of Directors. International Society for Prenatal Diagnosis Updated Position Statement on the use of genome-wide sequencing for prenatal diagnosis. Prenat Diagn. 2022;42(6):796-803. doi:10.1002/pd.6157. PMID: 35583085.
33. Li X, Hu T, Xiao X. Expert consensus on fetal ventriculomegaly: evidence-based recommendations for 23 key clinical questions. Front Pediatr. 2025;13:1678359. doi:10.3389/fped.2025.1678359. PMID: 41181173.
34. Bauer DF, Baird LC, Klimo P, Mazzola CA, Nikas DC, Tamber MS, et al. Congress of Neurological Surgeons Systematic Review and Evidence-Based Guidelines on the Treatment of Pediatric Hydrocephalus: Update of the 2014 Guidelines. Neurosurgery. 2020;87(6):1071-1075. doi:10.1093/neuros/nyaa434. PMID: 34791462.
35. ERN-ITHACA — European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders. Available from: https://ern-ithaca.eu/ [accessed 2026-10-09].
36. ERN-RND — European Reference Network on Rare Neurological Diseases. Disease Knowledge Hub (Ataxia/HSPs). Available from: https://www.ern-rnd.eu/disease-knowledge-hub/ [accessed 2026-10-09].
37. Firth HV, Hurst JA. Hydrocephalus. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 172–175.
38. Liu XY, Song X, Czosnyka M, Robba C, Czosnyka Z, Summers JL, et al. Congenital hydrocephalus: a review of recent advances in genetic etiology and molecular mechanisms. Mil Med Res. 2024;11(1):54. doi:10.1186/s40779-024-00560-5. PMID: 39135208.
39. Firth HV, Hurst JA. Ventriculomegaly. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 802–804.
40. Society for Maternal-Fetal Medicine (SMFM); Fox NS, Monteagudo A, Kuller JA, Craigo S, Norton ME. Mild fetal ventriculomegaly: diagnosis, evaluation, and management. Am J Obstet Gynecol. 2018;219(1):B2-B9. doi:10.1016/j.ajog.2018.04.039. PMID: 29705191.
41. Furey CG, Choi J, Jin SC, Zeng X, Timberlake AT, Nelson-Williams C, et al. De Novo Mutation in Genes Regulating Neural Stem Cell Fate in Human Congenital Hydrocephalus. Neuron. 2018;99(2):302-314.e4. doi:10.1016/j.neuron.2018.06.019. PMID: 29983323.
42. Duy PQ, Weise SC, Marini C, Li XJ, Liang D, Dahl PJ, et al. Impaired neurogenesis alters brain biomechanics in a neuroprogenitor-based genetic subtype of congenital hydrocephalus. Nat Neurosci. 2022;25(4):458-473. doi:10.1038/s41593-022-01043-3. PMID: 35379995.
43. Bickers DS, Adams RD. Hereditary stenosis of the aqueduct of Sylvius as a cause of congenital hydrocephalus. Brain. 1949;72(Pt. 2):246-262. doi:10.1093/brain/72.2.246. PMID: 18136715.
44. Rosenthal A, Jouet M, Kenwrick S. Aberrant splicing of neural cell adhesion molecule L1 mRNA in a family with X-linked hydrocephalus. Nat Genet. 1992;2(2):107-112. doi:10.1038/ng1092-107. PMID: 1303258.
45. Jouet M, Rosenthal A, MacFarlane J, Kenwrick S, Donnai D. A missense mutation confirms the L1 defect in X-linked hydrocephalus (HSAS). Nat Genet. 1993;4(4):331. doi:10.1038/ng0893-331. PMID: 8401576.
46. Jouet M, Kenwrick S. Gene analysis of L1 neural cell adhesion molecule in prenatal diagnosis of hydrocephalus. Lancet. 1995;345(8943):161-162. doi:10.1016/s0140-6736(95)90170-1. PMID: 7823673.
47. Vits L, Chitayat D, Van Camp G, Holden JJ, Fransen E, Willems PJ. Evidence for somatic and germline mosaicism in CRASH syndrome. Hum Mutat. 1998;Suppl 1:S284-7. doi:10.1002/humu.1380110189. PMID: 9452110.
48. Horiuchi R, Sato H, Asai C, Hamaguchi F, Takaishi Y, Fujiwara K, et al. X-Linked Hereditary Hydrocephalus Diagnosed in the Fetal Period With Adducted Thumb and a Novel L1CAM Variant: A Case Report. Clin Case Rep. 2025;13(8):e70792. doi:10.1002/ccr3.70792. PMID: 40822607.
49. Finckh U, Schröder J, Ressler B, Veske A, Gal A. Spectrum and detection rate of L1CAM mutations in isolated and familial cases with clinically suspected L1-disease. Am J Med Genet. 2000;92(1):40-46. doi:10.1002/(sici)1096-8628(20000501)92:1<40::aid-ajmg7>3.0.co;2-r. PMID: 10797421.
50. Vos YJ, Hofstra RM. An updated and upgraded L1CAM mutation database. Hum Mutat. 2010;31(1):E1102-9. doi:10.1002/humu.21172. PMID: 19953645.
51. National Center for Biotechnology Information. ClinVar — L1CAM[gene]. Available from: https://www.ncbi.nlm.nih.gov/clinvar/?term=L1CAM%5Bgene%5D [accessed 2026-10-09].
52. Harasaki T, Miyamoto S, Yonekawa T, Isogai M, Kato M, Nakashima M, et al. Utility of Urine-Derived Cells for Characterizing Aberrant Splicing Caused by a Novel Deep Intronic L1CAM Variant. Ann Hum Genet. 2026;90(5):278-282. doi:10.1111/ahg.70046. PMID: 42411594.
53. Tüysüz B, Ercan-Sençicek AG, Özer E, Göç N, Yalçınkaya C, Bilguvar K. Severe Phenotype in Patients with X-linked Hydrocephalus Caused by a Missense Mutation in L1CAM. Turk Arch Pediatr. 2022;57(5):521-525. doi:10.5152/turkarchpediatr.2022.22070. PMID: 35950747.
54. Parisi MA, Kapur RP, Neilson I, Hofstra RM, Holloway LW, Michaelis RC, et al. Hydrocephalus and intestinal aganglionosis: is L1CAM a modifier gene in Hirschsprung disease?. Am J Med Genet. 2002;108(1):51-56. doi:10.1002/ajmg.10185. PMID: 11857550.
55. Takenouchi T, Nakazawa M, Kanemura Y, Shimozato S, Yamasaki M, Takahashi T, et al. Hydrocephalus with Hirschsprung disease: severe end of X-linked hydrocephalus spectrum. Am J Med Genet A. 2012;158A(4):812-815. doi:10.1002/ajmg.a.35245. PMID: 22354677.
56. Fernández RM, Núñez-Torres R, García-Díaz L, de Agustín JC, Antiñolo G, Borrego S. Association of X-linked hydrocephalus and Hirschsprung disease: report of a new patient with a mutation in the L1CAM gene. Am J Med Genet A. 2012;158A(4):816-820. doi:10.1002/ajmg.a.35244. PMID: 22344793.
57. Wang Z, Shen X, Xu C, Wang R, Teng C, He Y, et al. Accelerated Identification and Preliminary Validation of a Pathogenic Missense Variant in the L1CAM Gene in a Pregnant Woman With Sonographic Anomalies Using AlphaMissense. Mol Genet Genomic Med. 2025;13(12):e70169. doi:10.1002/mgg3.70169. PMID: 41423712.
58. Vos YJ, de Walle HE, Bos KK, Stegeman JA, Ten Berge AM, Bruining M, et al. Genotype-phenotype correlations in L1 syndrome: a guide for genetic counselling and mutation analysis. J Med Genet. 2010;47(3):169-175. doi:10.1136/jmg.2009.071688. PMID: 19846429.
59. Loers G, Appel D, Lutz D, Congiu L, Kleene R, Hermans-Borgmeyer I, et al. Amelioration of the abnormal phenotype of a new L1 syndrome mouse mutation with L1 mimetics. FASEB J. 2021;35(2):e21329. doi:10.1096/fj.202002163r. PMID: 33484186.
60. Stopak W, Keener KA, Misra VK. Pathogenic Variants in MPDZ are Associated with a Syndromic Neurodevelopmental Disorder: A Case Report and Review of the Literature. J Child Neurol. 2026;41(2):241-246. doi:10.1177/08830738251353005. PMID: 40717626.
61. Su J, Lu W, Li M, Zhang Q, Chen F, Yi S, et al. Novel compound heterozygous frameshift variants in WDR81 associated with congenital hydrocephalus 3 with brain anomalies: First Chinese prenatal case confirms WDR81 involvement. Mol Genet Genomic Med. 2021;9(4):e1624. doi:10.1002/mgg3.1624. PMID: 33724704.
62. Shriyan R, Kachhadiya T, Sharma R, Patel S, Vaniawala S, Tamhankar VP, et al. Report of a Homozygous Nonsense Variant in the WDR91 Gene Associated With Severe Communicating Hydrocephalus, Dandy Walker Malformation, and Cerebellar Hypoplasia. Am J Med Genet A. 2026. doi:10.1002/ajmg.a.70228. PMID: 42311177.
63. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 44, Disorders of Cellular Trafficking (sections 44.1.3 and 44.2.1).
64. Jacquemin V, Versbraegen N, Duerinckx S, Massart A, Soblet J, Perazzolo C, et al. Congenital hydrocephalus: new Mendelian mutations and evidence for oligogenic inheritance. Hum Genomics. 2023;17(1):16. doi:10.1186/s40246-023-00464-w. PMID: 36859317.
65. Hastar N, Daum H, Kardos-Török N, Ganz G, Obendorf L, Vajkoczy P, et al. A pathogenic variant of AMOT leads to isolated X-linked congenital hydrocephalus due to N-terminal truncation. J Clin Invest. 2025;135(17):e179438. doi:10.1172/jci179438. PMID: 40892511.
66. Serpieri V, Vezain-Mouchard M, Orsi A, Lecointre M, Mazzotta C, Marguet F, et al. Bi-allelic variants in FSD1L cause a neurodevelopmental disorder overlapping with L1 syndrome. Am J Hum Genet. 2026;113(3):600-615. doi:10.1016/j.ajhg.2026.01.014. PMID: 41720098.
67. Mehta NH, Dennis E, Allington G, Mekbib KY, Hale AT, Davalan WC, et al. De Novo TRIO Missense Variants Disrupt Ras-GEF Domains and Cause Congenital Ventriculomegaly and Hydrocephalus. Hum Mutat. 2026;2026:8870037. doi:10.1155/humu/8870037. PMID: 42051466.
68. Tanaka Y, Yamada M, Miya F, Otani T, Kasuga Y, Suzuki H, et al. ARID1A gene variants and fetal hydrocephalus: First evidence of mRNA decay escape. Eur J Med Genet. 2025;78:105048. doi:10.1016/j.ejmg.2025.105048. PMID: 40962111.
69. Otter M, Wevers M, Pisters M, Pfundt R, Vos Y, Nievelstein RJ, et al. A novel mutation in L1CAM causes a mild form of L1 syndrome: a case report. Clin Case Rep. 2017;5(8):1213-1217. doi:10.1002/ccr3.1038. PMID: 28781826.
70. Cavalheiro S, da Costa MDS, Suzuki FS, Dastoli PA, de Melo Oliveira DP, Verges REG, et al. Endoscopic third ventriculostomy for diencephalosynapsis-related hydrocephalus in MPDZ gene syndrome: a case report and review of the literature. Childs Nerv Syst. 2026;42(1):380. doi:10.1007/s00381-026-07483-7. PMID: 42825918.
71. Firth HV, Hurst JA. Intellectual disability with apparent X-linked inheritance. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 194–197.
72. Firth HV, Hurst JA. Structural intracranial anomalies (agenesis of the corpus callosum, septo-optic dysplasia, and arachnoid cysts). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 320–323.
73. Malpas SC, Wright BE, Guild SJ, Heppner P, Gallichan RJ, Leung DP, et al. Long-term brain pressure monitoring via a discrete microimplant; a first-in-human safety and initial efficacy trial in adults and children with hydrocephalus. Nat Commun. 2026;17(1):3158. doi:10.1038/s41467-026-70864-8. PMID: 41927547.
74. Mallucci CL, Jenkinson MD, Conroy EJ, Hartley JC, Brown M, Dalton J, et al. Antibiotic or silver versus standard ventriculoperitoneal shunts (BASICS): a multicentre, single-blinded, randomised trial and economic evaluation. Lancet. 2019;394(10208):1530-1539. doi:10.1016/s0140-6736(19)31603-4. PMID: 31522843.
75. Gablofen (baclofen injection) intrathecal — Prescribing Information. NDA 022462. DailyMed/openFDA set_id 00d3e846-dd92-448d-9ab8-6a07be823cc1 (version effective 2026-05-07). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=00d3e846-dd92-448d-9ab8-6a07be823cc1 [accessed 2026-10-09].
76. Malinger G, Paladini D, Haratz KK, Monteagudo A, Pilu GL, Timor-Tritsch IE. ISUOG Practice Guidelines (updated): sonographic examination of the fetal central nervous system. Part 1: performance of screening examination and indications for targeted neurosonography. Ultrasound Obstet Gynecol. 2020;56(3):476-484. doi:10.1002/uog.22145. PMID: 32870591.
77. Paladini D, Malinger G, Birnbaum R, Monteagudo A, Pilu G, Salomon LJ, et al. ISUOG Practice Guidelines (updated): sonographic examination of the fetal central nervous system. Part 2: performance of targeted neurosonography. Ultrasound Obstet Gynecol. 2021;57(4):661-671. doi:10.1002/uog.23616. PMID: 33734522.
78. Chenyue Z, Huiqin X, Jingbo G, Min G, Hao Y, Rong G, et al. Genetic etiology of ventriculomegaly in 73 fetuses identified by High-Throughput sequencing. Sci Rep. 2025;15(1):23622. doi:10.1038/s41598-025-06714-2. PMID: 40603987.
79. Assistance Publique – Hôpitaux de Paris. Evaluation of the Diagnostic Performance of Non-Invasive Prenatal Diagnosis for Single Gene Disorders. ClinicalTrials.gov NCT06147414. Available from: https://clinicaltrials.gov/study/NCT06147414 [accessed 2026-10-09].
80. Jones KL, Jones MC, del Campo M. Walker-Warburg syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 256–257.
81. Jones KL, Jones MC, del Campo M. Hydrolethalus syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 254–255.
82. Firth HV, Hurst JA. Macrocephaly. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 224–227.
83. Firth HV, Hurst JA. Cerebellar anomalies. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 92–95.
84. Flannery AM, Mitchell L. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 1: Introduction and methodology. J Neurosurg Pediatr. 2014;14 Suppl 1:3-7. doi:10.3171/2014.7.peds14321. PMID: 25988777.
85. Congress of Neurological Surgeons. Pediatric Hydrocephalus: Systematic Literature Review and Evidence-Based Guidelines (2014; text edited to reflect the 2020 update), Parts 1–10. Available from: https://www.cns.org/guidelines/browse-guidelines-detail/pediatric-hydrocephalus [accessed 2026-10-09].
86. Mazzola CA, Choudhri AF, Auguste KI, Limbrick DD Jr, Rogido M, Mitchell L, et al. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 2: Management of posthemorrhagic hydrocephalus in premature infants. J Neurosurg Pediatr. 2014;14 Suppl 1:8-23. doi:10.3171/2014.7.peds14322. PMID: 25988778.
87. Flannery AM, Duhaime AC, Tamber MS, Kemp J, Pediatric Hydrocephalus Systematic Review and Evidence-Based Guidelines Task Force. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 3: Endoscopic computer-assisted electromagnetic navigation and ultrasonography as technical adjuvants for shunt placement. J Neurosurg Pediatr. 2014;14 Suppl 1:24-29. doi:10.3171/2014.7.peds14323. PMID: 25988779.
88. Limbrick DD Jr, Baird LC, Klimo P Jr, Riva-Cambrin J, Flannery AM, Pediatric Hydrocephalus Systematic Review and Evidence-Based Guidelines Task Force. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 4: Cerebrospinal fluid shunt or endoscopic third ventriculostomy for the treatment of hydrocephalus in children. J Neurosurg Pediatr. 2014;14 Suppl 1:30-34. doi:10.3171/2014.7.peds14324. PMID: 25988780.
89. Baird LC, Mazzola CA, Auguste KI, Klimo P Jr, Flannery AM, Pediatric Hydrocephalus Systematic Review and Evidence-Based Guidelines Task Force. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 5: Effect of valve type on cerebrospinal fluid shunt efficacy. J Neurosurg Pediatr. 2014;14 Suppl 1:35-43. doi:10.3171/2014.7.peds14325. PMID: 25988781.
90. Klimo P Jr, Van Poppel M, Thompson CJ, Baird LC, Duhaime AC, Flannery AM, et al. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 6: Preoperative antibiotics for shunt surgery in children with hydrocephalus: a systematic review and meta-analysis. J Neurosurg Pediatr. 2014;14 Suppl 1:44-52. doi:10.3171/2014.7.peds14326. PMID: 25988782.
91. Klimo P Jr, Thompson CJ, Baird LC, Flannery AM, Pediatric Hydrocephalus Systematic Review and Evidence-Based Guidelines Task Force. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 7: Antibiotic-impregnated shunt systems versus conventional shunts in children: a systematic review and meta-analysis. J Neurosurg Pediatr. 2014;14 Suppl 1:53-59. doi:10.3171/2014.7.peds14327. PMID: 25988783.
92. Tamber MS, Klimo P Jr, Mazzola CA, Flannery AM, Pediatric Hydrocephalus Systematic Review and Evidence-Based Guidelines Task Force. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 8: Management of cerebrospinal fluid shunt infection. J Neurosurg Pediatr. 2014;14 Suppl 1:60-71. doi:10.3171/2014.7.peds14328. PMID: 25988784.
93. Kemp J, Flannery AM, Tamber MS, Duhaime AC, Pediatric Hydrocephalus Systematic Review and Evidence-Based Guidelines Task Force. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 9: Effect of ventricular catheter entry point and position. J Neurosurg Pediatr. 2014;14 Suppl 1:72-76. doi:10.3171/2014.7.peds14329. PMID: 25988785.
94. Nikas DC, Post AF, Choudhri AF, Mazzola CA, Mitchell L, Flannery AM, et al. Pediatric hydrocephalus: systematic literature review and evidence-based guidelines. Part 10: Change in ventricle size as a measurement of effective treatment of hydrocephalus. J Neurosurg Pediatr. 2014;14 Suppl 1:77-81. doi:10.3171/2014.7.peds14330. PMID: 25988786.
95. Verhey LH, Kulkarni AV, Reeder RW, Riva-Cambrin J, Jensen H, Pollack IF, et al. A re-evaluation of the Endoscopic Third Ventriculostomy Success Score: a Hydrocephalus Clinical Research Network study. J Neurosurg Pediatr. 2024;33(5):417-427. doi:10.3171/2023.12.peds23401. PMID: 38335514.
96. Riva-Cambrin J, Kestle JRW, Rozzelle CJ, Naftel RP, Alvey JS, Reeder RW, et al. Predictors of success for combined endoscopic third ventriculostomy and choroid plexus cauterization in a North American setting: a Hydrocephalus Clinical Research Network study. J Neurosurg Pediatr. 2019;24(2):128-138. doi:10.3171/2019.3.peds18532. PMID: 31151098.
97. Murray C, Nicolaides K, Bleil C, Zebian B, Hart AR. Surgical interventions for fetal hydrocephalus: systematic review. Ultrasound Obstet Gynecol. 2025;66(4):413-421. doi:10.1002/uog.29310. PMID: 40825578.
98. Allergan/AbbVie. BOTOX (onabotulinumtoxinA) for injection — Prescribing Information. BLA 103000. DailyMed/openFDA set_id 33d066a9-34ff-4a1a-b38b-d10983df3300 (version effective 2023-11-18). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=33d066a9-34ff-4a1a-b38b-d10983df3300
99. Ipsen. Dysport (abobotulinumtoxinA) for injection — Prescribing Information. BLA 125274. DailyMed/openFDA set_id 71313a04-1349-4c26-b840-a39e4a3ddaed (version effective 2024-10-24). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=71313a04-1349-4c26-b840-a39e4a3ddaed [accessed 2026-10-09].
100. OZOBAX DS (baclofen oral solution) — Prescribing Information. NDA 208193. DailyMed/openFDA set_id d8ddcec5-debe-42b4-a803-e65917cc6a54 (version effective 2026-03-13). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=d8ddcec5-debe-42b4-a803-e65917cc6a54 [accessed 2026-10-09].
101. Tizanidine Tablets, USP — Prescribing Information. ANDA 076416. DailyMed/openFDA set_id 03fa03a6-fd05-4f17-872a-3c123330935d (version effective 2023-11-09). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=03fa03a6-fd05-4f17-872a-3c123330935d [accessed 2026-10-09].
102. Kumar A, Rashi R, Sinha AK, Ali MM, Ahmad S, Kumar B. Laparoscopic-assisted versus Open Placement of Ventriculoperitoneal Shunts in the Pediatric Population with Congenital Hydrocephalus: A Randomized Controlled Trial. J Indian Assoc Pediatr Surg. 2026;31(5):762-768. doi:10.4103/jiaps.jiaps_45_26. PMID: 42799526.
103. Rattani A, Spencer JC, Agwu CK, Dewan MC, Kabachelor EM, Ssenyonga P, et al. Five-Year Cost-Effectiveness Analysis of Endoscopic Third Ventriculostomy With Choroid Plexus Cauterization Versus Ventriculoperitoneal Shunting in Postinfectious Hydrocephalus in a Low-Income Setting. Neurosurgery. 2026. doi:10.1227/neu.0000000000004218. PMID: 42757834.
104. Long-term intracranial pressure monitoring microimplant (Malpas et al.). ClinicalTrials.gov NCT06402786. Available from: https://clinicaltrials.gov/study/NCT06402786
105. Cardoso LJC, Ferreira MY, Mitre LP, Brenner LBO, Bishay AE, Porto Junior S, et al. Enhancing reporting standards and cumulative evidence in ventriculoperitoneal shunt studies: a systematic review and reporting guideline proposal. J Neurosurg. 2026;144(4):821-833. doi:10.3171/2025.8.jns251448. PMID: 41569691.
106. Wu C. Editorial. Introducing the VPS Reporting Guideline as a framework to improve evidence in hydrocephalus care. J Neurosurg. 2026;144(4):819-820. doi:10.3171/2025.8.jns252040. PMID: 41569737.
107. ClinicalTrials.gov API v2 — condition searches «L1CAM», «L1 syndrome», «X-linked hydrocephalus», «congenital hydrocephalus», «fetal ventriculomegaly», «fetal hydrocephalus», «MPDZ», «aqueductal stenosis». Available from: https://clinicaltrials.gov [accessed 2026-10-09].
108. Hydrocephalus Association. Available from: https://www.hydroassoc.org/ [accessed 2026-10-09].
109. International Federation for Spina Bifida and Hydrocephalus (IF SBH). Available from: https://ifglobal.org/ [accessed 2026-10-09].
110. Shine — Spina bifida • Hydrocephalus • Information • Networking • Equality. Available from: https://shinecharity.org.uk/ [accessed 2026-10-09].
111. Associação Spina Bifida e Hidrocefalia de Portugal (ASBIHP). Available from: https://asbihp.pt/ [accessed 2026-10-09].