Currarino Syndrome and Sacral Agenesis (Caudal Regression)

Spectrum of caudal dysgeneses: Currarino syndrome due to MNX1 haploinsufficiency, caudal regression syndrome and sacral agenesis, sporadic or associated with maternal diabetes, sacral agenesis with vertebral anomalies due to TBXT, permanent neonatal diabetes due to biallelic MNX1 and sirenomelia as the extreme of the spectrum

Literature and regulatory information updated: 29 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Scope and definitions

Caudal dysgeneses form a group of malformations of the caudal end of the embryo that affect, in variable proportions, the sacral and lumbar spine, the terminal spinal cord, the hindgut, the urogenital tract and the lower limbs [3, 31]. Within this group, entities with different aetiological and prognostic bases are distinguished. Currarino syndrome is an autosomal dominant form of hereditary sacral dysgenesis, caused in most familial cases by heterozygous loss-of-function variants in the MNX1 gene, and is defined by the triad of sacral malformation, presacral mass and anorectal malformation [1, 20]. Caudal regression syndrome, also called caudal dysplasia sequence or, in the broad sense, sacral agenesis, is a defect of premature termination of spinal development, ranging from isolated absence of the coccyx to total sacral agenesis with missing lumbar vertebrae, accompanied by neurological deficit and visceral anomalies; it is usually sporadic and maternal pre-gestational diabetes is its main known risk factor [2, 3, 25]. Sirenomelia, with a single midline lower limb, is considered by many authors to be the extreme form of the caudal regression spectrum, although the pathogenic relationship between the two remains debated [9, 17]. The spectrum is completed by two rare monogenic entities: sacral agenesis with vertebral anomalies due to biallelic TBXT variants and permanent neonatal diabetes due to biallelic MNX1 variants, which may be accompanied by sacral agenesis and imperforate anus [13, 21]. The VACTERL association, which shares vertebral and anorectal anomalies with these entities, maternal diabetic embryofetopathy as a whole, monogenic neonatal diabetes (including Wolcott-Rallison syndrome), neural tube defects and spina bifida, and Hirschsprung disease enter the differential diagnosis. In this article, these entities are addressed only as regards the caudal spectrum.

The article groups Currarino syndrome and sacral agenesis together. Orphanet brings caudal regression syndrome and sirenomelia together in the clinical group «caudal regression–sirenomelia spectrum» (ORPHA:444941) and classifies diabetic embryopathy as a separate entity (ORPHA:1926) [32, 33]. In OMIM, entry #600145 is titled «sacral defect with anterior meningocele» (SDAM) and includes, as included entities, caudal dysgenesis syndrome, caudal regression syndrome, sacral agenesis and sirenomelia; the only phenotype–gene relationship recorded in that entry is with VANGL1 [4].

1.2. Historical background of Currarino syndrome

The first report consistent with the syndrome was published by Roger L. J. Kennedy in 1926 [20]. In 1981, Guido Currarino, Coln and Votteler described the association of anorectal, sacral and presacral anomalies as a triad with a common embryonic origin, in the article that gave the syndrome its name [20, 34]. The sacral anomaly they described, a partial agenesis of the sacrum with the first sacral vertebra intact and a sickle or scimitar shape, is considered characteristic of this syndrome [1]. In 1983, Yates and colleagues documented autosomal dominant transmission of anterior sacral defects [20]. Lynch and colleagues mapped the dominant sacral agenesis locus to 7q36 in 1995, with a maximum lod score of 4.38 with marker D7S559, in a region that includes the HPE3 holoprosencephaly locus [1]. In 1998, Ross and colleagues identified family-specific variants in the homeobox gene HLXB9, which until then had been thought to map to 1q41-q42.1 and to be expressed in lymphoid and pancreatic tissues, and proposed that a malformation of the caudal end of the notochord, around Carnegie stage 7 (16 days post-ovulation), with aberrant secondary neurulation, explains the pattern of anomalies [35]. Hagan et al. and Belloni et al. confirmed the role of the gene in larger series in 2000 [36, 37]. The gene was later given the symbol MNX1 (motor neuron and pancreas homeobox 1) [11].

1.3. Historical background of caudal regression and sirenomelia

Sirenomelia is the form of caudal dysgenesis described the longest: the first objective descriptions are attributed to Rocheus (1542) and Palfyn (1553), and Saint-Hilaire (1836) and Förster (1861) divided it into variants [3]. The modern era began with Duhamel, who in 1961 coined the expression «caudal regression syndrome» and proposed that sirenomelia and the minor forms, down to imperforate anus, were expressions of a primary deficiency of the caudal mesoderm; he distinguished a mermaid type, with fused limbs, and an ankypode type, with flexed knees, abducted hips and popliteal pterygium [3, 9]. The association with maternal diabetes was described by Rusnak and Driscoll (1965) and by Passarge and Lenz (1966), and occurrence in siblings by Finer et al. (1978) and Stewart and Stoll (1979), according to Smith [2]. In 1986, Stevenson and colleagues proposed the vitelline artery «vascular steal» theory for sirenomelia, which separated it pathogenically from caudal dysplasia [9]. Duesterhoeft et al. (2007) described five patients with caudal regression and an aberrant abdominal umbilical artery, and Thottungal et al. (2010) observed a single umbilical artery of aortic origin in cases of caudal dysgenesis and normal arteries in cases of sirenomelia, which reinforced the hypothesis of a pathogenic continuum [4, 17].

1.4. Clinical and radiological classifications in use

The classifications of caudal dysgenesis answer different questions, and several are in simultaneous use. The Renshaw classification (1978) grades vertebral agenesis: types 1 and 2 correspond to partial or complete hypoplasia of the sacrum and types 3 and 4 to absence of lumbar vertebrae; some authors add a type 5, sirenomelia [3]. The Pang classification (1993) divides sacral agenesis into five bony types and adds the morphology and level of termination of the conus medullaris, which gives it neurological prognostic value [18, 25]. Welch and Aterman (1984) proposed four clinical types, one non-familial associated with maternal diabetes and three familial, and Cama et al. (1996) five bony categories, of which only hemisacrum is associated with MNX1 variants [1, 4]. Guille et al. classified patients by walking potential, based on the relationship between the spine and the pelvis, and Stocker and Heifetz divided sirenomelia into seven types according to the bones present [3].

In Currarino syndrome, Martucciello et al. proposed in 2004 the division into complete, mild and minimal forms, and Crétolle et al. suggested in 2008 that neural tube defects (tethered cord, thickened or fatty filum, spinal lipoma and neurenteric cysts) be considered a fourth major feature [20, 38]. In an Italian multicentre series of 45 patients, 44.5% had the complete form, 42.2% the mild form and 13.3% the minimal form [39].

1.5. Nomenclature and corrections

The former symbol HLXB9 and the protein name HB9 persist in the literature before 2010 and in some reference books; the approved symbol is MNX1 [2, 11]. OMIM entry 259100 «Currarino idiopathic osteoarthropathy» refers to a primary hypertrophic osteoarthropathy, unrelated to the Currarino triad [1, 40]. The GeneReviews chapter on permanent neonatal diabetes states that no other phenotypes are known to be associated with germline MNX1 variants; this statement does not take into account Currarino syndrome, which OMIM and the literature attribute to heterozygous variants in the same gene [1, 7]. In a review article, the patient homozygous for p.(Phe248Leu) is described as female, but the table of the report produced by the same laboratories that published it indicates male sex, which is the information followed here [20, 21].

2. Molecular Genetics and Pathophysiology

2.1. Common embryology of the caudal end

The sacral spinal cord, the caudal vertebral bodies, the hindgut and part of the urogenital tract derive from an undifferentiated cell population in the primitive streak region, the caudal cell mass, which drives secondary neurulation [3]. In the human tail bud, the neural tube, notochord, somites and hindgut form from pluripotent epiblast cells at around Carnegie stage 12, which corresponds to 29–31 days of embryonic age [20]. Arrest of development of the caudal mesoderm before the fourth week, in a zone where the caudal neurons, spinal cord, hindgut and the elements of neural tube closure meet, explains why vertebral anomalies coexist with anorectal, genitourinary and neurological anomalies [3]. In Currarino syndrome, the classic hypothesis attributes the triad to a primary defect of the caudal notochord: some cells would fail to migrate from the primitive node, the notochord would be defective or split and a communication would form between the ventral gut and the dorsal neural elements, giving rise to the anterior meningocele, enteric cysts and anorectal anomaly; remnants of the primitive streak would persist in the sacrococcygeal region as a teratoma [20, 35].

2.2. MNX1: structure, expression and function

The MNX1 gene lies at 7q36.3, has three exons and encodes a canonical protein of 401 amino acids (UniProt P50219) with a homeodomain between amino acids 241 and 300 and a polyalanine tract of variable length, between 12 and 18 residues, starting at amino acid 121 [20]. The MANE Select transcript is NM_005515.4, which encodes NP_005506.3 [41]. A short isoform of 189 residues, with an alternative first exon encoding a distinct 19-amino-acid N-terminus, is annotated in UniProt, but Dworschak et al. did not detect the corresponding messenger RNA in blood, saliva or hippocampus, which makes it unlikely that variants in that exon explain cases without an identified variant [20]. The HB9 protein is a transcription factor that binds to regulatory elements of target genes and represses them; with LMO4, it establishes motor neuron identity, blocking activation of interneuron genes by the ISL1-LHX3 complex, and it is required for pancreatic development and cell fate specification [11].

In the human embryo, between Carnegie stages 12 and 21 (29–55 days), Hagan et al. detected MNX1 expression in the basal plate of the spinal cord and hindbrain, in motor neurons along the entire spinal cord, in the pharynx, oesophagus, stomach and pancreas, but not in the notochord of the tail bud [20, 36]. Mice with homozygous inactivation of Mnx1 have agenesis of the dorsal pancreas and abnormal islets of Langerhans, without sacral or hindgut anomalies, and heterozygotes have no phenotype, which shows an interspecies divergence that limits the use of these models [20]. In embryonic mouse pancreatic cells, inactivation of Mnx1 diverts beta cells towards a delta-cell-like fate [21]. The models that come closest to Currarino syndrome are non-genetic or involve other genes: administration of etretinate on day 9 of gestation produces anterior sacral meningocele and sacral defect in all embryos, with anorectal malformation in about three quarters; the recessive murine variant p.Cys470Arg in Pcsk5 and deficiency of Gdf11, a protein activated by PCSK5, reproduce a mixed VACTERL, caudal regression and Currarino phenotype, with ectopic Mnx1 expression in the tail bud; all-trans retinoic acid inhibits Pcsk5 and Gdf11 in the hindgut and induces a similar phenotype [20].

2.3. Mutational spectrum and mechanism in Currarino syndrome

The mechanism is haploinsufficiency. In the series of 50 index cases by Crétolle et al., truncating variants (frameshift and nonsense) accounted for 57% of those identified, and three cytogenetic anomalies including the gene were described for the first time [42]. Hagan et al. found, in 22 index cases with a variant, 19 intragenic variants, of which 11 were predicted to introduce a premature termination codon and 8 were missense variants clustered in the homeodomain, always with non-conservative substitution of highly conserved amino acids [36]. Merello et al. described, in 28 Italian patients, 10 novel and 4 recurrent variants, including two frameshift (p.Ser4IlefsX52 and p.Phe248SerfsX35), six missense (p.Pro27Leu, p.Gly103Arg, p.Leu254Pro, p.Leu278Pro, p.Glu282Lys and p.Arg292Gly), one nonsense (p.Lys297X) and one synonymous (p.Gln290Gln), and counted 82 variants published up to 2013 [24]. These are the published protein designations, without nucleotide description in the abstract, and none appears in ClinVar, so they are not converted to NM_005515.4 [24]. In 2021, the review by Dworschak et al. counted about 120 pathogenic heterozygous single nucleotide variants, and large deletions or 7q36 rearrangements as a less frequent cause [20]. The pLI index of MNX1 in gnomAD is 0.79, below the usual threshold of 0.9, which the authors attribute to the small number of observed variants (90% confidence interval of the observed/expected ratio between 0.04 and 0.5); there are no homozygous truncating variants in gnomAD [20].

The proportion of patients with an identified variant depends on the family context. Hagan et al. found a variant in 20 of 21 patients with a familial form and in 2 of 7 sporadic patients, and allowed for undetected mosaicism or genetic heterogeneity in the rest [36]. Crétolle et al. estimated a detection rate of about 50%, reaching 90% in familial cases, and noted that 19 of the 24 patients without a variant had subtle phenotypic differences compared with the rest [42]. In the review of more than 60 studies, which brought together 296 unrelated cases (117 familial and 172 sporadic), no variant was identified in 9 familial (7.7%) and 117 sporadic (68%) cases, corresponding to an overall detection rate of 57.4%; of the 168 index cases with an MNX1 alteration, 108 (64.3%) belonged to families with several affected members and 55 (32.7%) were sporadic [20]. Merello et al. and Cuturilo et al. summarise these data by saying that MNX1 variants explain almost all familial cases and about 30% of sporadic cases [23, 24].

2.4. 7q36 deletions and chromosomal anomalies

7q36 deletions that include MNX1 produce Currarino syndrome as part of a broader phenotype. When they include SHH, they are associated with holoprosencephaly, and intellectual disability is frequent [43]. Hagan et al. observed that the three patients with a microdeletion were the only ones with developmental delay, and one of them, with a deletion encompassing SHH, had a single central maxillary incisor, a microform of holoprosencephaly [1, 36]. Wang et al. (1999) described a mother carrying a balanced translocation t(7;12)(q36;q24) with two children monosomic for 7q36 and sacral dysgenesis [1]. Cuturilo et al. confirmed that the 7q deletion is associated with intellectual disability, growth retardation, facial dysmorphism and additional malformations, whereas intragenic variants usually produce the triad without these signs [23]. In 2026 the first Korean case of a complete triad with an isolated 12 Mb terminal 7q35–q36.3 deletion was published, in a 9-month-old girl [44]. Outside 7q36, rare anomalies have been described in patients with a Currarino phenotype: partial trisomy of 13q and 20p, 3q duplications (a pure 7.9 Mb duplication at 3q26.32-q27.2, de novo, in a patient without an MNX1 variant, with the triad, delay, central nervous system malformations and dysmorphism), functional disomy of Xp with Xq13.2-q28 deletion and 3q25.33-q29 duplication, excess 9qh heterochromatin, de novo 18p deletion and a 0.3 Mb locus at 6q25.3 defined in patients with 6q deletions [20]. Gene dosage in distal 3q may explain some of the MNX1-negative cases [20].

2.5. Genotype–phenotype correlation, penetrance and expressivity

There is no clinically useful correlation between the type of intragenic variant and the phenotype: related and unrelated patients with the same variant may have very different presentations [20, 42]. Köchling et al. studied 23 carriers from 9 families: 10 were asymptomatic, but all had radiological changes, and only 8 had the complete triad [1]. In a three-generation family with a frameshift variant, 7 of 9 carriers had confirmed teratomas, 3 of them initially asymptomatic, and only 2 the complete triad [1]. Monozygotic twins with the same variant, NM_005515.4:c.780C>G p.(Phe260Leu), had different severities, as did two pairs of twins described previously [41, 45]. The only quantitative association comes from Costanzo et al.: among patients with an MNX1 variant, 56.5% had the complete form, vs 13% among patients without a variant (p = 0.0085), with no relationship between subtype and number of associated anomalies (p = 0.5102); the authors concluded that MNX1 mainly determines the expression and severity of the triad, and that the associated anomalies depend on other loci [39]. Proposed modifiers include the transcriptional activity of the normal allele and variants or expression differences of the antisense RNAs MNX1-AS1 and MNX1-AS2 [20].

Expression is so variable that penetrance and expressivity become blurred in practice. OMIM states that up to 33% of patients are asymptomatic [1]. The Oxford Desk Reference states that almost one third of carriers in the Crétolle series were asymptomatic [46]. Lynch et al., reviewing 205 patients, identified only two carriers (1%) without any radiological sign, and in the review by Dworschak et al. 25 carriers were negative after extensive physical examination, but 2 had constipation and 8 had not had imaging, which leads the authors to distinguish the asymptomatic carrier from the unaffected carrier [20]. Chronic constipation since childhood is often the only manifestation in relatives [20].

The polyalanine tract in exon 1 is polymorphic. In the European population of Caucasian ancestry, the (CGC)11 allele has a frequency of 90.23%, (CGC)8 0.6%, (CGC)9 7.47% and (CGC)12 1.7%; in Chinese, (CGC)11 accounts for 50%, (CGC)9 33% and (CGC)12 16.2%, with similar distributions in Japanese and Yoruba [20, 37]. These length variations are not associated with a phenotype; a single patient with a (CGC)5 contraction, absent in controls, has been described, and no pathogenic expansions of MNX1 are known [20]. In a case–control study, the length of the GCC repeat also did not modify the risk of caudal regression [12].

2.6. Other candidate genes in Currarino syndrome

Testing for SHH variants was negative in a family with eight affected members, in seven sporadic cases and in 15 patients with anorectal malformation and sacral underdevelopment, and a patient with Currarino syndrome and a microform of holoprosencephaly had a 2.7 Mb 7q36.3 deletion that did not include SHH [20]. Exome sequencing in sporadic trios suggested candidates (ETV3L, NCAPD3 and ARID5A in one series; CDH2, ITIH2, HOXB4 and TLE4 in another) that were not confirmed [20]. Conserved non-coding elements close to MNX1, with enhancer activity in zebrafish, are candidates for regulatory variants that the usual methods do not detect [20].

2.7. Permanent neonatal diabetes due to biallelic MNX1

Biallelic MNX1 variants cause permanent neonatal diabetes with autosomal recessive inheritance and account for about 1% of this disease [7]. Flanagan et al. looked for homozygous variants in 29 pancreatic transcription factor genes in 37 consanguineous patients with permanent neonatal diabetes and established MNX1 and NKX2-2 as causal genes in humans [14]. Bonnefond et al. had identified, by homozygosity mapping, the p.Phe272Leu variant in helix 2 of the homeodomain in a girl of consanguineous parents with isolated diabetes [47]. The three cases published up to 2023 are summarised in the following table; the variants are normalised to the MANE Select transcript [21, 41].

All three patients had intrauterine growth restriction, a sign of severe insulin secretion deficiency in utero [21]. A third homozygous variant, affecting the initiation codon and described in the literature as p.Met1Thr (published designation, without nucleotide description or ClinVar record; current notation of a variant at this codon: p.(Met1?)), was found in a fetus of consanguineous parents who died minutes after birth with growth restriction, cleft palate, imperforate anus, scrotal agenesis and agenesis of the lumbar and sacral spine and spinal cord; three other newborns of the couple died with a similar phenotype, and the heterozygous parents were healthy [20, 41]. The parents heterozygous for p.Phe248Leu and p.Phe272Leu had no signs of Currarino syndrome. The proposed explanation is that these recessive homeodomain missense variants do not abolish all MNX1 function, but only part of it, including beta-cell development; they have never been found in patients with Currarino syndrome [21]. The two patients with a syndromic phenotype were male, which led to the hypothesis, not demonstrated, of sex-biased MNX1 expression [21]. Diabetes is not part of Currarino syndrome, and pancreatic agenesis has never been demonstrated in these patients [20, 21].

2.8. Aetiology of caudal regression syndrome

Caudal regression is aetiologically heterogeneous and, in most cases, of unknown cause [2, 3]. MNX1 is not implicated: Merello et al. found no pathogenic variants in 48 patients with caudal regression without a Currarino phenotype, and Belloni et al. showed that, of Cama's five bony categories, only hemisacrum is associated with variants in this gene [1, 12]. Reviews describe a multifactorial and polygenic model, with an environmental contribution demonstrated by twin studies and a significant number of targets linked to the regulation of retinoic acid, whose dysregulation produces caudal malformations in animal models [3, 48]. Proposed candidate genes include CDX2, CDX1, CYP26A1 (variants F186L and C358R, published designations without a ClinVar match, with severe deficiency of retinoic acid catabolism), PLZF (ZBTB16), ID1 (paternally inherited missense variants in two of three children with non-syndromic sacral agenesis) and, in animal models, Mbtps1 [3]. The pilot exome and CNV study by Porsch et al. in four Caucasian trios found de novo variants in SPTBN5, MORN1 and ZNF330, inherited predicted-damaging variants in PDZD2 (homozygous) and CLTCL1 (compound heterozygous), variants in PTEN, GLTSCR2 and VANGL1 and two deletions (de novo 3q13.13 and homozygous 8p23.2) overlapping CNV of similar phenotypes [31]. There is potential genetic overlap between caudal regression and the VACTERL association [48].

OMIM assigns entry #600145 to VANGL1 on the basis of Kibar et al. (2007), who, studying 144 patients with neural tube defects and 106 controls, found the heterozygous V239I variant in a 10-year-old Italian girl with severe caudal regression (Pang type IV sacral agenesis), lipomyeloschisis, anorectal malformation, hydromyelia and tethered cord; the mother, without clinical signs, carried a variant that had arisen de novo in her, and the brother had a dermal sinus [4]. The variant abolishes the binding of VANGL1 to dishevelled proteins 1, 2 and 3 [49]. On the MANE Select transcript NM_138959.3, the Kibar variants are written c.715G>A p.(Val239Ile), c.821G>A p.(Arg274Gln) and c.983T>C p.(Met328Thr) [41]. Later studies showed enrichment of rare VANGL1 variants in dysraphisms (p = 7.0 × 10⁻⁵ vs controls), with incomplete penetrance and frequent inheritance from an unaffected parent; neither ClinGen nor the GenCC classification assigns VANGL1 more than limited evidence for familial caudal dysgenesis, and a heterozygous variant should be reported as a susceptibility factor [49].

2.9. Maternal diabetes and diabetic embryopathy

Maternal pre-gestational diabetes is the best-known risk factor for caudal regression [25, 31]. The mechanism is not clear: the hyperglycaemic environment simultaneously alters several teratogenic pathways, including inositol metabolism, prostaglandins and reactive oxygen species; hypoglycaemia is teratogenic in animal models, but this association has not been established in humans [50]. The risk is limited to diabetes present at conception and is not observed in gestational diabetes [50]. The magnitude and interpretation of the figures require attention to what each study measures, and are discussed in sections 8 and 9. The hypothesis that MNX1 takes part in the sacral agenesis of diabetic pregnancies, whose prevalence has been estimated as 200 to 250 times higher, found no support, because no MNX1 variant has ever been identified in these cases [20].

2.10. Sacral agenesis with vertebral anomalies due to TBXT

The TBXT gene (brachyury, 6q27) is the founding member of the T-box family and is essential for mesoderm formation and differentiation and for axial development [13]. Postma et al. described four patients from three consanguineous families with sacral agenesis, persistent notochordal canal and abnormal ossification of the vertebral bodies, and identified by homozygosity mapping a 4.1 Mb region at 6q27 and the homozygous p.His171Arg variant in the T-box domain, which on the MANE Select transcript is written NM_001366285.2:c.512A>G p.(His171Arg) [13, 41]. The mutant protein loses about half of its DNA-binding capacity, increases cell growth and alters the expression of genes for ossification, notochord maintenance and axial mesoderm [6]. Sacral agenesis and persistent notochord were detected prenatally in three patients; the tubular notochordal structure was progressively replaced by vertebral bodies on subsequent scans; two siblings died in the neonatal period, one of them with sacral and left renal agenesis, persistent cloaca with anal atresia and vertical clefts of the vertebral bodies [6]. The authors recommend assessing vertebral ossification in all patients with sacral agenesis [13]. The heterozygous TBXT variant p.Ala338Val (rs117097130, allele frequency 0.004835 in ExAC) has been found in patients with sacral agenesis, Klippel-Feil syndrome or vertebral malformations, always inherited from an unaffected parent; its role remains uncertain [20].

2.11. Sirenomelia: pathogenic hypotheses

Two hypotheses explain sirenomelia. In the vascular steal hypothesis, blood returns to the placenta through a single large vessel derived from the vitelline artery complex, arising from the aorta below the diaphragm, instead of the two iliac umbilical arteries; the distal aorta becomes subordinate and lacks renal and inferior mesenteric branches, and the caudal structures are deprived of flow, which explains the single limb with knees and feet in posterior alignment, the absence of the sacrum, rectum, genitalia, kidneys and bladder, and the variability between cases [9]. In the primary caudal mesoderm deficiency hypothesis, the single artery is a consequence and not a cause, and sirenomelia and caudal dysplasia are extremes of the same spectrum; this hypothesis is supported by cases of sirenomelia with normal umbilical arteries and of caudal dysplasia with an aberrant abdominal umbilical artery [9, 17]. Smith keeps the separation as controversial, and Orphanet describes the distinction between sirenomelia, caudal regression, familial caudal dysgenesis and VACTERL as debated [2, 10].

2.12. Normalisation of the variants cited

The variants of Merello et al. are kept in the protein nomenclature of the original article, without conversion to the nucleotide level [24].

3. Clinical Spectrum and Manifestations by System

3.1. Currarino syndrome: overview

The classic triad comprises a sacral malformation, a hindgut anomaly and a presacral mass [42]. The typical sacral malformation is the sickle- or scimitar-shaped hemisacrum, with the first sacral vertebra intact and partial absence of S2–S5, or total sacral agenesis below S2 [42, 46]. The hindgut anomaly ranges from anorectal malformation, often anal stenosis, to chronic constipation and chronic intestinal pseudo-obstruction without anatomical malformation [2, 27]. The presacral mass may be an anterior meningocele, a teratoma, an enteric cyst, a rectal duplication or a combination of these [2, 20]. The Oxford Desk Reference reports anorectal malformation in 57% and presacral mass in 64% of patients [26]. The triad is often incomplete, even within a family, and associated anomalies are the rule: they were present in 84.5% of 45 Italian patients and in 80% of 15 Chinese children with the complete form [39, 52].

3.2. Digestive system

Chronic constipation is the most frequent presenting symptom in cases diagnosed after the neonatal period and is often the only manifestation in carrier relatives [20, 46]. The most classic anorectal malformation is anal stenosis, but in recent series rectoperineal fistulas predominate: among 15 children in Beijing with the complete form there were 6 rectoperineal fistulas, 2 rectovestibular, 2 rectourethral, one rectovesicovaginal, one rectonavicular, one anal atresia, one cloaca and one anal stenosis [2, 52]. Hirschsprung disease and other dysganglionoses were found in half of the six patients in the Martucciello series, who considered them part of the syndrome; in the Beijing series, intraoperative biopsy of the rectal pouch showed absent ganglion cells in 2 of 15 [38, 52]. This disease must be distinguished from megacolon secondary to difficulty emptying through a stenotic or malpositioned anus, as the case in section 10 illustrates [51, 52]. Crétolle et al. described 8 of 29 patients with chronic intestinal pseudo-obstruction without anorectal malformation [27]. Other digestive anomalies described include oesophageal atresia and Meckel diverticulum [52].

3.3. Presacral mass

Anterior sacral meningocele is a herniation of the dura through the anterior sacral defect; it may communicate with the spinal canal and, through a fistula with the bowel, be a portal of entry for central nervous system infection [1, 27]. The presacral teratoma is mature in most cases. In the North American multicentre series of the Midwest Pediatric Surgery Consortium (203 resections of sacrococcygeal teratoma between 2010 and 2020, 25 with Currarino syndrome), all the teratomas of the syndrome were mature, compared with 27% immature and 12% malignant in the rest; they were smaller (median 3.3 cm vs 8.0 cm), mostly Altman type IV, that is, entirely internal (87% vs 17%), and were diagnosed after birth in 80% of cases, vs 25% [54]. Other presacral lesions described are enteric cyst, lipoma, rectal duplication and, in 2026, a myxopapillary ependymoma associated with a retrorectal teratoma and a neuroendocrine tumour in a teratoma [20, 55]. An anterior meningocele may enlarge and, in one adult, compressed the ureter of a solitary kidney, with severe hydronephrosis and rising creatinine [55, 56].

3.4. Risk of malignant transformation

Malignant transformation of the presacral teratoma in Currarino syndrome is rare but documented, in both children and adults [20, 57]. Yoshida et al. counted malignancy in 6 of about 300 published children, 4 of them after recurrence following primary excision, and in 4 adults [58]. In a three-generation family, the 22-year-old proband had malignant transformation of a presacral teratoma [1]. Two comparative studies quantified the difference from isolated sacrococcygeal teratoma. In the Dutch national study (1981–2010; 205 sacrococcygeal teratomas and 16 patients with Currarino syndrome), malignancy-free survival was 80% and 58% at one and two years in sacrococcygeal teratoma, vs 100% at two years in Currarino syndrome (p = 0.017) [59]. In the SCT-study, a global retrospective cohort from 132 institutions in 62 countries, 3388 patients were analysed (3183 with sacrococcygeal teratoma and 205 with Currarino syndrome): the proportion with malignancy at initial resection was 10.3% at one year and 31.9% at two years in sacrococcygeal teratoma, vs 4.2% at two years in Currarino syndrome (p < 0.001); the histology of recurrence was malignant in 35.4% of recurrent sacrococcygeal teratomas (n = 114) and in 5.9% of those of the syndrome (n = 1; p = 0.005), and survival was similar (94.9% vs 96.9%; p = 0.343) [60]. These data, of level of evidence III, indicate that malignancy is rare but not absent in Currarino syndrome and that the risk increases with age in isolated sacrococcygeal teratoma [60].

3.5. Nervous system

Occult spinal dysraphism is present in about 60% of the patients described and includes tethered cord, thickened or fatty filum terminale, intradural lipoma, lipomyelomeningocele and neurenteric cysts [20, 22]. The risk of tethered cord syndrome appears higher when there is a lipoma and lower when the mass is a teratoma or an anterior meningocele [22]. In a series of 95 patients, tethered cord (55.8%) was not associated with age, sex, severity of the sacral deformity, type of anorectal malformation or type of mass, which makes it necessary to look for it in all patients [45]. Manifestations include sacrococcygeal sensory deficits, motor or sensory deficits of the lower limbs, gait disturbance, anal sphincter dysfunction and focal neurological deficits [52]. The most serious neurological complication is meningitis due to ascending infection through the communication between the bowel, the mass and the subarachnoid space; Escherichia coli meningitis and fatal polymicrobial cerebritis and arachnoiditis in a 6-month-old girl have been described [1, 58]. Developmental delay is not part of the syndrome due to an intragenic variant and, when present, suggests a 7q36 deletion [20, 23].

3.6. Urinary and genital tract

Genitourinary anomalies include horseshoe kidney, duplex or solitary kidney, vesicoureteric reflux, hydronephrosis, neurogenic bladder, recurrent urinary tract infections and urinary incontinence [20]. Müllerian anomalies (bicornuate uterus, septate vagina, agenesis or hypoplasia of the uterus and vagina) reach 15% of cases [20]. In the Genoa series, urinary function was good in most: of 16 patients assessed with clinical, radiological and urodynamic study, one had detrusor overactivity, two recurrent urinary tract infections and two vesicoureteric reflux; the two with lipomyeloschisis had a neurogenic bladder, the rest voided spontaneously, and renal function was normal in all [53]. In the Beijing series, 9 of 10 children over 4 years of age had a normal score on the dysfunctional voiding symptom score (DVSS) and none required catheterisation [52]. The literature compiled by Tan et al. indicates urinary incontinence in 14–36% and neurogenic bladder in 3.4% [52].

3.7. Other manifestations

In the Beijing series, one third of the children had a heart defect, two of them requiring surgery (tetralogy of Fallot and ventricular septal defect), 20% urinary anomalies, 20% uterovaginal anomalies and 13.3% skeletal deformities [52]. The same series showed a decline in anthropometric indices with age in children over 2 years: mean Z-score for weight of −0.42 ± 1.30, for height of −0.28 ± 1.45 and for body mass index of −0.46 ± 1.50, with 24.9% underweight and worse progression above 8 years; the authors consider this finding hypothesis-generating [52]. Coexistence with caudal duplication anomaly is exceptional [56].

3.8. Age at diagnosis, sex and natural history

About 80% of cases are diagnosed before 16 years, but late diagnosis in adulthood is frequent because of phenotypic variability [20]. In surgical series, about 90% of children were diagnosed before 3 years [52]. An undiagnosed mass may manifest late: OMIM describes, under the SDAM entity, presentation in adulthood with obstructed labour, chronic constipation or meningitis [4]. The sex ratio is debated: 179 women (58.1%) and 129 men (41.9%) among 308 carriers and relatives tested, that is, a ratio of 1.39:1; Haga et al. estimated 6:1 in adults and 2:1 in children, and the female excess may reflect ascertainment bias through gynaecological and urological symptoms [20].

3.9. Functional outcomes after surgery

Long-term functional outcomes were for a long time poorly described, and Yoshida et al. acknowledged that the incidence of bowel and urinary dysfunction was underestimated [58]. In ten Japanese patients followed between 2004 and 2020, early postoperative complications reached 80%, mainly transient dysuria, more frequent when the last normal sacral vertebra was higher (p = 0.024) and in the complete form (p = 0.033); in the long term, 40% had intractable constipation requiring rectal irrigation and 30% intractable dysuria requiring clean intermittent catheterisation [61]. In 32 Beijing patients with the complete form, assessed at a median age of 9 years, the total Rintala score was 16.5, with faecal soiling in 78.1% and constipation in 50%, but high quality of life on the PedsQL 4.0 (96.8); total sacral dysplasia (β = −3.86; p = 0.026) and perioperative complications related to the presacral mass (β = −6.24; p = 0.002) predicted worse bowel function, and resection of the megarectosigmoid did not improve it and was associated with more soiling (100% vs 44.4%; p = 0.034) [62]. In the other Beijing series, 7 of 10 children over 4 years had good defecatory function, 2 fair and 1 poor (the child with Hirschsprung disease), and, in a pooled analysis of 89 patients, defecatory dysfunction was 39.3% and urinary dysfunction 11.9% [52]. In the teratoma of Currarino syndrome, urinary incontinence (44% vs 28%; p = 0.048) and constipation (76% vs 32%; p < 0.001) were more frequent than in isolated sacrococcygeal teratoma [54].

3.10. Caudal regression syndrome: clinical spectrum

The presentation depends on the level of agenesis and the morphology of the conus medullaris [3, 25]. Orthopaedic reviews simplify the spectrum into three groups: sirenomelia, complete sacral agenesis and partial sacral agenesis [3].

In complete sacral agenesis, with or without missing lumbar vertebrae (Pang types I and II), the ilia may be fused or articulate with each other or with the last vertebra; there is spinopelvic kyphosis, sometimes scoliosis, and sometimes a mobile pseudoarticulation between the spine and the pelvis [3]. The pelvis is narrow, the buttocks are flat, the intergluteal cleft is short and there are dimples on the buttocks [2, 3]. The lower limbs adopt a «Buddha» position, with flexion contractures of the knees, popliteal pterygium and abducted hips, and neurogenic clubfoot is frequent [3]. The motor deficit corresponds to the last vertebra present, the sensory deficit is patchy and less marked than the motor one, and most patients get about in a wheelchair; household walking with orthoses is only possible when all the lumbar vertebrae are present and the motor deficit is low lumbar [3]. Smith describes, in the most severe cases, a newborn with a normal upper trunk and a short lower segment, flexion and abduction of the hips and popliteal pterygia secondary to neurologically derived contractures, with equinovarus or calcaneovalgus clubfoot [2].

In partial agenesis, the spectrum ranges from asymptomatic absence of the coccyx, discovered incidentally in adolescence, to neurological deficits and orthopaedic problems at several levels [3]. Mild agenesis mainly affects the muscles innervated by S2–S5 (perineum, pelvic floor and intrinsic foot muscles), with urethral and anal sphincter dysfunction, sometimes male sexual dysfunction, pes cavus with claw toes and foot drop; with an L5–S1 level, the legs are hypoplastic and tapered; unilateral agenesis may be associated with hypoplasia of the whole limb, and symptoms may be asymmetric [3]. More than 80% of patients have limb or extravertebral bone defects, mainly shortening or hypoplasia, clubfoot, popliteal pterygium and contractures, and up to 63% have foot deformities [3]. Spinal anomalies include defects of formation and segmentation, scoliosis, abnormal kyphosis, tethered cord and abrupt termination of the cord [3]. In Pang group 2, with tethered cord, the neurological deficit may be progressive owing to traction and compression of the sacral roots during growth [3].

Urological manifestations determine much of the prognosis. Unilateral renal agenesis with vesicoureteric reflux is frequent and contributes to the risk of renal failure [2]. Absence of more than one sacral vertebra may cause neuropathic bladder, especially with defects between S2 and S4; late diagnosis is associated with recurrent urinary tract infections, incontinence and renal damage, and even isolated sacral agenesis may cause late-presenting neurogenic bladder [3]. Faecal incontinence affects about one third of patients, with imperforate anus as a risk factor [3]. Occasional manifestations described include imperforate anus, cleft lip and palate, microcephaly and myelomeningocele [2]. Cognition, hearing and speech are usually normal [3, 63].

3.11. Anorectal malformation with sacral agenesis

The Amsterdam cohort of 41 patients with anorectal malformation and sacral agenesis born between 2000 and 2024, with a median follow-up age of 11.2 years, quantifies the functional burden of this association [18]. The agenesis was mostly partial (Pang type III in 75.6%), cord termination was abnormal in 61%, there was tethered cord in 24.4%, structural urological anomalies in 48.8%, VACTERL association in 58.5% and a syndrome in 14.6% (three with Currarino syndrome) [18]. Of the 33 patients aged 4 years or more, 66.7% used transanal irrigation, started at a median of 4.6 years, and only 12.1% were continent without it; faecal incontinence was 26.3% in users and 71.4% in non-users (p = 0.015) [18]. Clean intermittent catheterisation was started in 43.9%, almost always for neurogenic bladder confirmed by urodynamics, at a median of 2.1 months; 24.4% had urological surgery, and 34.1% needed irrigation and catheterisation together, with complete continence in 64.3% of these [18]. The need for these interventions was associated with the complexity of the anorectal malformation and not with the radiological severity of the sacral agenesis [18]. In an earlier study with routine urodynamics, neurogenic bladder was present in 77% of patients with anorectal malformation and sacral agenesis, vs 18% with isolated malformation [18].

3.12. Diabetic embryopathy

In the population-based cohort from the North of England (1996–2008; 401,149 singleton pregnancies, 1677 in women with pre-existing diabetes), the rate of non-chromosomal major congenital anomalies was 71.6 per 1000 pregnancies in diabetic women (95% CI 59.6–84.9), with a relative risk of 3.8 (3.2–4.5) and a three- to sixfold increase in all common anomaly groups [64]. According to the Oxford Desk Reference, which reproduces the relative risks of that cohort, the relative risk is 170 for caudal dysplasia, 57 for laterality disorders, 5 for neural tube defects and 3 for congenital heart defects [50]. The same book stresses that, even with this relative risk, caudal dysplasia occurs in absolute numbers as often in children of non-diabetic women as in those of diabetic women, and that VACTERL and sirenomelia, although with an increased relative risk, are more frequent in children of women without diabetes [50]. Femoral hypoplasia–unusual facies is a phenocopy of diabetic embryopathy, and coexisting caudal regression points to a maternal aetiology [65, 66]. There is no concern in the literature regarding the neurodevelopment of the children of diabetic women [50].

3.13. Sacral agenesis with vertebral anomalies, MNX1-related neonatal diabetes and sirenomelia

SAVA manifests prenatally with sacral agenesis and persistent notochordal canal, and may be associated with cloaca, anal atresia and renal agenesis; two of the four patients described died in the neonatal period [6]. MNX1-related neonatal diabetes begins in the first weeks, sometimes in ketoacidosis, and may be accompanied by sacral agenesis, imperforate anus, neurogenic bladder and pulmonary hypoplasia (section 2.7) [21]. In sirenomelia, the ICBDSR series (19 surveillance systems, 249 cases) shows 47% live births, 71.2% preterm and 88.2% under 2500 g; sex was ambiguous in 47%; half the cases had genital, large bowel and urinary defects, and 10–15% lower spine defects, a single or abnormal umbilical artery, upper limb, cardiac or central nervous system defects [19]. Most fetuses are stillborn or die during or shortly after delivery, mainly from bilateral renal agenesis and cardiac malformations, and survival beyond one year is exceptional [3, 10].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age and context

In the prenatal period, suspicion arises with sacral agenesis, a cystic or solid presacral mass, bowel dilatation, ventriculomegaly with dysraphism or a single lower limb [1, 19, 25]. A cystic pelvic mass with hydrocolpos suggests cloaca, and an early cystic abdominal mass that disappears does not exclude anorectal malformation [29]. Prenatal diagnosis of Currarino syndrome is rare when the disease is not known in the family; in Lisbon, a fetal MRI requested for fetal spinal defects showed similar changes in the asymptomatic mother, which led to the diagnosis of both [67]. In pre-gestational diabetes, ultrasound detection of fetal anomalies is lower than in other pregnant women (30% vs 73% in one series), largely because of the image quality associated with maternal obesity [50].

In the newborn, any anorectal malformation requires investigation of the spine and sacrum: Heij et al. found lumbosacral anomalies on MRI in 30% of low and 50% of high anorectal malformations, and the European guideline estimates that 10–60% of patients with anorectal malformation have tethered cord or another spinal anomaly [26, 29]. Absence of a normal anus, abdominal distension with vomiting and delayed passage of meconium are the complaints that lead to the neonatal diagnosis of the complete form [52]. Signs of caudal regression in the newborn are flat buttocks with a short intergluteal cleft, gluteal dimples, a short lower segment, contractures with popliteal pterygium and clubfoot [2]. In the child of a diabetic mother, these signs point to diabetic embryopathy, and femoral hypoplasia–unusual facies is part of the same differential [65, 66].

In children and adults, Currarino syndrome should be suspected with severe constipation from early life, especially with anal stenosis, megacolon or a previous diagnosis of Hirschsprung disease without histological confirmation, a retrorectal mass palpable on digital rectal examination, recurrent perianal or gluteal abscesses, meningitis with enteric organisms, Müllerian anomalies or neurogenic bladder [1, 46, 51]. Severe early nappy dermatitis and failure of toilet training may be the first signs of faecal incontinence and neurogenic bladder due to undiagnosed sacral agenesis [3]. The co-occurrence of sacral anomaly and presacral mass is so characteristic that Merello et al. propose it as an indication for MNX1 testing [24]. A family history of chronic constipation, neonatal surgery or colostomy should be actively sought in a three-generation pedigree [26].

4.2. Warning signs and urgent situations

4.3. Screening

There is no biochemical newborn screening for these entities. The European guideline recommends that all newborns with anorectal malformation have a radiograph of the spine and sacrum, ultrasound of the spinal canal in the first month and ultrasound of the kidneys and urinary tract in the first weeks [29]. A North American observational study that aimed to screen by spinal ultrasound for sacral agenesis in asymptomatic newborns of mothers with pre-gestational diabetes (NCT05033275, 60 participants) was terminated owing to the end of funding and posted no results in the registry [68]. In relatives of a patient with Currarino syndrome, imaging of the sacrum is the practical screening test, given the variable expressivity [11, 20].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic criteria

There are no formally validated diagnostic criteria for any of these entities. Currarino syndrome is diagnosed clinically by the complete triad or by components of it in a compatible family context, and the diagnosis stands even without an identified variant [20]. Martucciello et al. defined the complete, mild and minimal forms, and Crétolle et al. recommended always characterising the four components: anorectal anomaly, sacral defect, presacral mass and spinal cord [27, 38]. Caudal regression is diagnosed radiologically by the degree of vertebral agenesis and the morphology of the conus, according to the Pang and Renshaw classifications [25]. SAVA is diagnosed by the combination of sacral agenesis and abnormal vertebral ossification with molecular confirmation of TBXT [13].

5.2. Imaging

Radiography or CT of the sacrum shows the sickle-shaped sacrum or agenesis below S2; lumbosacral and pelvic MRI is essential to demonstrate the presacral mass, its communication with the spinal canal, tethered cord and lipoma of the conus [11, 27]. For patients and their parents, Dworschak et al. propose a sequence of spinal and pelvic ultrasound, sacral radiograph and MRI when the findings are doubtful or incomplete [20]. In caudal regression, Krishnan et al. recommend fetal MRI after obstetric ultrasound, early postnatal MRI and extension of the field of view of lumbosacral MRI to the retroperitoneum, without a saturation band, to assess the kidneys, bladder and syndromic anomalies; sacral masses and filum anomalies should be identified because they may require surgery [25]. In sacral agenesis, ossification of the vertebral bodies should be assessed because of the possibility of SAVA [13]. Contrast studies and anorectal physiological assessment characterise the hindgut anomaly; suction rectal biopsy, or intraoperative biopsy of the rectal pouch, is needed to distinguish Hirschsprung disease from secondary megacolon [38, 52]. Echocardiography should be performed at the initial diagnosis [26, 52].

5.3. Genetic tests

In Currarino syndrome, the strategy of Cuturilo et al. separates two scenarios: with the classic triad and normal growth, intelligence and facies, MNX1 is sequenced first and, only if negative, chromosomal microarray is performed; with growth retardation, facial dysmorphism or intellectual disability, microarray is the first test [23]. Merello et al. added MLPA in patients without a sequence variant to exclude intragenic or whole-gene deletions [24]. Dworschak et al. suggest exome sequencing and molecular karyotyping in negative cases, to identify other genes [20]. The European guideline strongly recommends a clinical genetics consultation in every anorectal malformation with at least one other anomaly (grade B) [29]. In caudal regression with multiple anomalies, the Oxford Desk Reference recommends genomic analysis (array, panel, exome or genome) and DNA storage [26]. In neonatal diabetes, the analysis is done by a panel of all known genes; the Exeter laboratory first tested ABCC8, KCNJ11 and INS by Sanger and then a next-generation panel, and GeneReviews indicates detection of 100% of MNX1 variants by sequencing [7, 21].

5.4. Limitations of the tests

A negative MNX1 analysis does not exclude Currarino syndrome: about 68% of sporadic and 7.7% of familial cases remain without an identified variant, and the overall detection proportion is 57.4% [20]. This proportion measures the fraction of patients with a clinical diagnosis in whom a variant was found and not the analytical sensitivity of the method. Possible reasons include intronic, promoter or distant regulatory element variants, which the usual methods do not cover, mosaicism, other genes and errors in clinical diagnosis [20, 24, 36]. Sequencing alone misses whole-gene deletions and 7q36 deletions, which require MLPA or microarray [23, 24]. The polyalanine tract is a common polymorphism and length variations between 12 and 18 residues should not be reported as pathogenic [20]. In VANGL1 and for the TBXT p.Ala338Val variant, incomplete penetrance and inheritance from healthy parents prevent causality from being assigned to an isolated heterozygous variant [20, 49]. In non-syndromic caudal regression, there is no genetic test with an established yield.

5.5. Genetic report

The report should state the gene, the transcript with version (for MNX1, NM_005515.4), the nucleotide and protein HGVS description, zygosity, classification according to ACMG/AMP criteria and, for deletions, the genomic coordinates and the genes included, in particular SHH [23, 41]. Variants of uncertain significance benefit from segregation studies with sacral imaging of relatives, because many carriers have only radiological changes [1, 20]. In neonatal diabetes, the homozygosity expected in children of consanguineous parents should be confirmed by testing the parents [21].

5.6. Prenatal and preimplantation diagnosis

With a known familial variant, molecular prenatal diagnosis and preimplantation genetic diagnosis are possible [26, 51]. Without a known variant, ultrasound and fetal MRI can identify sacral agenesis, presacral mass and dysraphism, but variable expressivity prevents severity from being predicted from the genotype [1, 67]. Crétolle et al. described an ultrasound diagnosis at 22 weeks in a second pregnancy of a carrier mother, with severe ventriculomegaly and dysraphism [1]. A prenatal case initially interpreted as VACTERL association was reclassified as Currarino syndrome by MRI [69], and the prenatal ultrasound diagnosis of caudal regression in the child of a diabetic mother was described in 2025 [70]. In pre-gestational diabetes, NICE recommends an anomaly scan at 20 weeks with assessment of the fetal heart (four chambers, outflow tracts and three vessels) [28]. Maternal serum screening for neural tube defects requires informing the laboratory of the diabetes so that alpha-fetoprotein norms can be adjusted [50].

5.7. Practical algorithm

6. Differential Diagnosis

6.1. Comparative table

6.2. Currarino syndrome versus caudal regression

The distinction has direct consequences for counselling. Currarino syndrome is dominant, the hemisacrum has a characteristic shape with S1 intact, and a presacral mass and a family history of constipation point to it; caudal regression is usually sporadic, the agenesis is symmetrical or involves the lumbar spine, and there is no presacral mass [1, 2]. MNX1 variants have never been found in caudal regression without a Currarino phenotype [12]. OMIM draws attention to the overlap: some patients classified as SDAM or other familial forms of sacral agenesis may have had Currarino syndrome, since the triad is complete in only a minority of carriers [4].

6.3. Presacral mass: teratoma of the syndrome versus isolated sacrococcygeal teratoma

Isolated sacrococcygeal teratoma has an incidence of 1/35,000 to 1/40,000 live births, predominates in females (4:1), is highly vascular, may cause high-output cardiac failure with hydrops fetalis, and is classified by Altman into four types, of which types I and II, predominantly external, account for 76% [46]. The teratoma of Currarino syndrome is almost always internal (type IV), smaller, diagnosed after birth and mature, and is associated with a typical sacral malformation [54]. The risk of malignancy is much lower than in isolated sacrococcygeal teratoma (section 3.4), and identification of the syndrome changes family counselling [59, 60]. Monozygotic twins have an excess of sacrococcygeal teratoma and sirenomelia, and the less affected twin may have a milder degree of the same defect [74, 75].

6.4. Caudal regression versus sirenomelia and VACTERL

Sirenomelia is distinguished by fusion of the lower limbs and by lethality; the risk is 100 to 150 times higher in monozygotic twins, and prevalence increases with maternal age under 20 years [19, 75]. The association with bladder exstrophy, cyclopia or holoprosencephaly and acardiac twin is higher than expected [19]. The VACTERL association shares the vertebral and anorectal anomalies, but the vertebrae affected are mainly thoracic and lumbar, with hemivertebrae, and the lower limbs are spared; the Amsterdam cohort shows that more than half of patients with anorectal malformation and sacral agenesis meet VACTERL criteria, which illustrates the overlap [18, 26]. The European guideline recommends systematically looking for VACTERL anomalies in every anorectal malformation [29].

6.5. Chronic constipation and megacolon

Severe constipation with megacolon in a child with anal stenosis or a retrorectal mass should bring Currarino syndrome to mind before Hirschsprung disease is assumed; histological confirmation of the presence of ganglion cells avoids unnecessary resections [51]. Hirschsprung disease may, however, coexist with the syndrome and worsen the functional prognosis [38, 52].

7. Treatments and Risk Management Strategies

7.1. General principles and evidence base

There is no targeted pharmacological therapy for Currarino syndrome, caudal regression or SAVA; treatment is surgical, rehabilitative and supportive, and is delivered by multidisciplinary teams of paediatric surgery, neurosurgery, urology, orthopaedics, gastroenterology, genetics and specialist nursing [3, 27]. There are no specific guidelines for these entities: the applicable recommendations come from the ERN eUROGEN guideline on anorectal malformations (2024–2025, developed with ERN ERNICA using the ADAPTE method and with recommendations mostly of grade D, that is, expert opinion), from the single-centre protocol of Martucciello et al. for Currarino syndrome, from orthopaedic reviews and from retrospective series [29, 38, 76, 77]. In MNX1-related neonatal diabetes, the reference is the GeneReviews chapter on permanent neonatal diabetes, and for the prevention of diabetic embryopathy the NICE guideline NG3 [7, 28].

7.2. Currarino syndrome: surgical planning

The central message of the published series is that the four components (anorectal anomaly, sacral defect, presacral mass and spinal cord) must be defined before the first operation [11, 27]. Crétolle et al. operated on 25 of 29 patients, 7 of them in a single stage that treated the intestinal and presacral malformations and, when necessary, the spinal cord, and concluded that assessment of the four components makes this approach possible [27]. Tethered cord and communication between the mass and the spinal canal should be identified before surgery because of the neurological and infectious risk [11, 27]. Martucciello et al. propose a rational scheduling of the multidisciplinary procedures and argue that early diagnosis and treatment avoid the morbidity and mortality of an undiagnosed presacral mass [38].

In practice, the sequence varies with the anatomy. In the Beijing series of 15 children with the complete form, two thirds had staged surgery with an initial enterostomy; 20% had anoplasty first and then resection of the mass, and 13.3% anoplasty and resection in a single stage; the median age at anoplasty was 53 days, there was a single perioperative complication (cerebrospinal fluid fistula) and three patients had cord release at the same stage as the presacral neurosurgical repair [52]. The techniques used include posterior (PSARP) and anterior (ASARP) sagittal anorectoplasty, the Soave procedure and repair of lipomyelomeningocele [52]. In adults with a large anterior meningocele, a neurosurgical approach by sacral laminectomy, cord untethering, section of the filum and disconnection of the fistula between the cord and the meningocele resolved ureteric compression and symptoms in one case [56].

7.3. Presacral mass and teratoma

Resection of the mass with histological examination is part of treatment, because examination distinguishes meningocele from teratoma and identifies malignant transformation [11, 52]. Some authors recommend removal of the teratoma even in asymptomatic patients, and there are no biomarkers that predict malignancy [20]. The 2025 and 2026 data suggest that, in Currarino syndrome, teratomas are mature, resection is usually complete and recurrence low (4% vs 10% in isolated sacrococcygeal teratoma; p = 0.18), and the North American authors propose adapting surveillance to reduce unnecessary imaging [54]. The SCT-study confirms that malignancy is rare (4.2% at two years), but does occur [60]. Recurrences and enlargement of masses, in about 4% of patients, occur at older ages than the average of paediatric series, which supports periodic postoperative pelvic assessment [52]. These data are retrospective (level III) and do not define a validated surveillance interval.

7.4. Anorectal malformation: ERN eUROGEN recommendations

The Oxford Desk Reference describes the classic sequence of neonatal colostomy and later anastomosis depending on the anatomy and the continence prognosis [26]. In an exploratory pooled analysis of 89 patients with Currarino syndrome, anoplasty before one year was associated with less defecatory dysfunction (p = 0.03), but without adjustment for the severity of sacral dysplasia, tethered cord or Hirschsprung disease; the authors themselves treat this association as hypothesis-generating [52]. The European recommendation of reconstruction in the first year is concordant [76].

7.5. Bowel management

Long-term bowel management is needed for the chronic constipation and pseudo-obstruction of Currarino syndrome [11]. The tools are those of anorectal malformation: laxatives, enemas, transanal irrigation and antegrade irrigation [18, 76]. In the Amsterdam cohort, transanal irrigation was used by two thirds of patients with anorectal malformation and sacral agenesis, 86% continued it, and users had less faecal incontinence; the authors advocate early counselling that spontaneous continence is unlikely [18]. In the Currarino syndrome series, glycerine enemas and lactulose were used, and 40% of the Japanese patients needed rectal irrigation [52, 61]. Resection of the megarectosigmoid in the complete form did not improve bowel function and was associated with more faecal soiling, which favours a conservative approach [62]. Perineal dermatitis is prevented with sitz baths, zinc ointments and barrier creams [76]. Doses of laxatives and irrigation solutions follow general paediatric protocols and the respective labels; there is no dosing specific to these entities in the sources consulted.

7.6. Neurogenic bladder and renal protection

The aim is to preserve renal function [3, 18]. Esposito et al. proposed ultrasound in all newborns at risk of urological anomalies and urodynamic or videourodynamic study if abnormalities are found [3]. In the Amsterdam cohort, clean intermittent catheterisation was started at a median of 2.1 months, as a proactive renal protection regimen, and combined with anticholinergics and antibiotic prophylaxis; 53% of patients on catheterisation needed additional urological surgery (vesicostomy or ureterostomy in infants, ureteric reimplantation and bladder augmentation with a Mitrofanoff or Monti catheterisable channel in older children) [18]. The choice and dose of the anticholinergic follow the general standards for paediatric neurogenic bladder.

7.7. Neurosurgery

Release of a tethered cord and resection of myelodysplastic lesions are indicated in patients with tethering lesions, in particular Lee's «failure of regression» type, in which the deficit may worsen; associated myelomeningocele is operated on early [3]. In Currarino syndrome, neurosurgery is integrated, when possible, into the same stage as presacral and anorectal surgery [27, 52]. In the Genoa series, 8 of 16 patients were operated on for neural tube defects [53].

7.8. Orthopaedics and rehabilitation in caudal regression

The orthopaedic decision is based on walking potential and spinopelvic stability [3]. In non-ambulatory patients without significant pelvic obliquity, pressure ulcers or progression of deformity, conservative treatment is recommended, because arthrodesis eliminates the spinal mobility used in transfers; surgery (posterior instrumentation with spinopelvic fixation, preferably with sacral-alar-iliac S2AI screws, which carry a lower risk of complications) is indicated in progressive deformity and spinopelvic instability, especially in ambulatory patients, and may improve sitting in wheelchair users with pelvic obliquity [3]. The most frequent complications are implant failure, haemorrhage, delayed wound healing, dural tears and pseudarthrosis; the cervical spine should be examined to exclude atlantoaxial instability [3]. Knee flexion contractures with popliteal pterygium are the most difficult to correct and are operated on only with a functional quadriceps and walking potential, with complete correction; amputation or knee disarticulation has been proposed by some authors to improve sitting, but is poorly accepted by patients and families [3]. Hip dislocation is operated on in patients who walk or may walk (Guille group A), with frequent recurrence during growth [3]. Foot deformities are treated initially by the Ponseti method and, if necessary, surgically, with the aim of a painless plantigrade foot in ambulatory patients and a foot that fits in footwear and does not cause ulcers in the rest; worsening foot deformity should lead to tethered cord being excluded [3]. Ongoing rehabilitation in collaboration with the family is part of treatment [3]. The combination of growth hormone and rehabilitation has been described in a single case, without evidence that allows it to be recommended [3].

7.9. MNX1-related permanent neonatal diabetes

Treatment is insulin. GeneReviews reserves oral sulfonylureas for the ABCC8 and KCNJ11 forms; in the MNX1 form, a trial of gliclazide up to 1.2 mg/kg/day failed and the patient remained insulin-dependent [7, 21]. General care includes rehydration and intravenous insulin infusion soon after diagnosis, subcutaneous insulin when the infant is stable and feeding orally, a high-calorie diet to regain weight and developmental and educational support, specifically recommended in the MNX1 forms [7]. GeneReviews recommends avoiding, in young children, rapid-acting analogues (lispro and aspart) and short-acting regular insulin, except by intravenous or continuous subcutaneous infusion, because of the risk of severe hypoglycaemia [7]. The patients described used doses between 0.3 and 0.8 U/kg/day; the third case, after ketoacidosis treated according to the ISPAD guideline, moved from NPH insulin with lispro to glargine and then degludec 0.8 U/kg/day with lispro at meals, with an HbA1c of 8% at 35 months [21]. These doses are those of individual reports and do not constitute a recommended dosing regimen; adjustment follows blood glucose.

7.10. Prevention of diabetic embryopathy

Periconceptional glycaemia is the main modifiable factor for congenital anomalies in women with diabetes: in the North of England cohort, the risk increased by about 30% (adjusted OR 1.3; 95% CI 1.2–1.4) for each 1% (11 mmol/mol) increase in HbA1c above 6.3% (45 mmol/mol), and pre-existing nephropathy was an independent predictor (adjusted OR 2.5) [64]. The NICE guideline NG3 recommends the following for women with diabetes planning a pregnancy [28]:

Optimisation should take place before conception, with specialist obstetric follow-up in maternal–fetal medicine [50]. These targets are intended to prevent all anomalies and there is no study quantifying the specific reduction in caudal regression.

7.11. Regulatory table

A ClinicalTrials.gov search (29 October 2026) found no therapeutic intervention trials in these entities; there were no marketing authorisation applications, CHMP opinions or regulatory decisions to record [68, 78].

7.12. Registered studies and outcome studies

None of these studies is a randomised trial, and comparisons between series from different centres are not valid. Differences in rates also reflect the type of centre, the definition of outcomes and the length of follow-up.

7.13. Pregnancy, anaesthesia and transition

In women with Currarino syndrome, the presacral mass may obstruct labour, and obstetric complications are one of the forms of presentation in adulthood [4]. Müllerian anomalies warrant gynaecological assessment, with pelvic ultrasound around menarche if there are complaints [20, 77]. Before any pelvic or sacral surgery, tethered cord and any communication between the mass and the spinal canal should be identified by MRI, because of the neurological risk [11, 27]. Transition to adult care should begin at around 12–14 years and include attention to sexuality, mental health and quality of life [77].

7.14. Developments in the last 12 months

Between the end of 2025 and October 2026 only observational studies and case reports were published, with no clinical trials or regulatory decisions. The SCT-study, the largest comparative cohort published, quantified the risk of malignant transformation in Currarino syndrome compared with sacrococcygeal teratoma (level III; invited commentary in the same issue) [60, 79]. Two retrospective studies from Beijing described bowel, urinary and growth outcomes and the lack of benefit from resection of the megarectosigmoid [52, 62]. A Chicago series extended the spectrum of presacral masses to myxopapillary ependymoma and neuroendocrine tumour [55]. A Shanghai study with discordant monozygotic twins and 95 patients documented the frequency of tethered cord (55.8%) and the absence of clinical predictors [45]. The Amsterdam cohort quantified the need for irrigation and catheterisation in anorectal malformation with sacral agenesis [18]. The first Korean case due to an isolated terminal 7q deletion was also described [44].

8. Reproductive Options and Family Counselling

8.1. Currarino syndrome

Inheritance is autosomal dominant, and each child of a carrier has a 50% probability of inheriting the variant [1, 51]. Counselling must convey that expressivity is unpredictable: the same variant produces anything from the complete triad with severe complications to isolated constipation or purely radiological changes, even between monozygotic twins, and almost one third of carriers are asymptomatic [20, 45, 46]. Lynch et al. estimated the malformation risks in heterozygotes by Weinberg's proband method in families they followed, but the figures from that study are not in the sources consulted, which include only the abstract [80]. The existence of patients with an MNX1 variant who are completely normal on examination makes it necessary to distinguish the asymptomatic carrier, who may have imaging changes, from the carrier without any sign [20].

Cascade testing is done for the familial variant when it is known and, in any case, by sacral imaging of first-degree relatives, because radiological assessment identifies carriers with minimal expression [11, 38]. Dworschak et al. recommend genetic testing of the parents of all patients, which would make it possible to know the true penetrance [20]. In an apparently sporadic case, a parent may be an asymptomatic carrier with sacral changes on imaging, as happened with the mother identified by fetal MRI in Lisbon and with the father in the case in section 10 [51, 67]. When neither parent has the variant, the recurrence risk for siblings is low, but germline mosaicism cannot be excluded; Hagan et al. allowed for undetected somatic mosaicism in sporadic cases without a variant [36]. In patients with a 7q36 deletion, the parents' karyotype should be studied to exclude a balanced translocation, which implies a high risk of recurrence with imbalance [1].

With a known variant, prenatal diagnosis and preimplantation genetic diagnosis are possible [26, 51]. Variable expressivity makes these decisions difficult, because the presence of the variant does not predict severity; fetal imaging may show sacral agenesis or a mass, but does not exclude a mild form [1, 67]. In carrier women with a presacral mass, pregnancy should be planned with prior imaging because of the risk of obstructed labour [4]. In asymptomatic minors, sacral imaging has direct clinical utility (detection of a mass or tethered cord), which justifies assessment in childhood, unlike tests without clinical consequence [20, 38].

8.2. Neonatal diabetes due to biallelic MNX1

Inheritance is autosomal recessive: the parents are heterozygous and each sibling has a 25% probability of being affected [7]. Heterozygotes for the variants described had neither diabetes nor signs of Currarino syndrome [21]. Apparently healthy siblings should be assessed to identify hyperglycaemia early [7]. The recurrence of four neonatal deaths with a severe caudal phenotype in a couple heterozygous for an initiation codon variant illustrates the 25% risk in each pregnancy [20].

8.3. Caudal regression, diabetic embryopathy and SAVA

In isolated caudal regression, most cases are sporadic and the recurrence risk is low, but affected siblings of unaffected parents have been described [2]. For isolated anorectal malformations, once syndromic and chromosomal causes have been excluded, the Oxford Desk Reference indicates an overall recurrence risk of about 1% [26]. In diabetic embryopathy, the risk in the next pregnancy depends mainly on periconceptional glycaemic control, and the intervention is prior optimisation of HbA1c with high-dose folic acid [28, 64]. In the cohort of Bell et al., the risk of major malformation was 1 in 9 with a periconceptional HbA1c of 10% and 1 in 33 with 6.1% [50]. Gestational diabetes does not increase this risk [50]. In SAVA, inheritance is autosomal recessive, with a 25% risk for each sibling, and the families described were consanguineous [6]. In families with SDAM, inheritance is dominant [4]. A heterozygous VANGL1 variant is a low-penetrance susceptibility factor and does not allow malformation to be predicted in a carrier or in a pregnancy [49]. In laterality disorders with sacral agenesis without a specific diagnosis, the empirical risk for siblings is about 5%, or 25% with consanguinity [71].

8.4. Sirenomelia

Sirenomelia is sporadic; the frequency of twinning (9% vs 1% expected) and the increase with maternal age under 20 years are the only known population factors [19]. In monozygotic pregnancies, concordance for early defects is only 5–20% [74].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Currarino syndrome

The true prevalence is unknown because of phenotypic variability and asymptomatic carriers [20]. Orphanet gives a European point prevalence of 1.0/100,000 (class 1–9/100,000), with the status «not validated» and no primary source [15, 20]. The sex ratio among carriers and relatives tested is 1.39 women per man, with possible ascertainment bias [20]. The published series come mainly from Europe (France, Italy, the Netherlands, the United Kingdom), the USA, China, Korea and Japan, and do not measure population frequency [27, 39, 45, 52, 54, 59, 61]. The distribution of the polyalanine polymorphism differs between Europeans and Asians, without clinical significance [20]. In Portugal, the source consulted is a report of prenatal diagnosis by MRI in a Lisbon hospital [67]. The sources consulted contain no Brazilian data on the syndrome.

9.2. Caudal regression and sacral agenesis

Estimates vary with the definition: about 1–2/100,000 or 1/60,000 live births for caudal regression in the clinical sense, 1–5/100,000 live births in recent reviews, and less than 0.5% when minimal sacral agenesis, which often goes unnoticed, is included [3, 18]. Orphanet classifies worldwide prevalence as unknown and records 1.75/100,000 births based on a single-centre series from Perth, Western Australia (16 cases of caudal dysgenesis and 9 of sirenomelia) [16, 17]. Welch and Aterman estimated a population frequency of 0.14% for sacral defect with anterior meningocele [4]. The sex ratio was 2.7 males per female in one series and showed no predominance in a review of 83 cases [3].

The fraction attributable to maternal diabetes varies with the study. Smith reports that 16% of cases of caudal dysplasia occur in children of diabetic mothers, and OMIM, citing Lynch et al., 15–25% [2, 4]. In the 18 EUROCAT registries (1990–2005), caudal regression sequence had an odds ratio of 26.4 (95% CI 8.98–77.64) in pregnancies with pre-gestational diabetes, but only 17% of all cases of caudal regression resulted from a pregnancy with diabetes [81]. Chan estimated that about 1% of children of diabetic mothers have defects in this group, according to the review by Jasiewicz and Kacki [3]. These figures are compatible: the relative risk is very high, but, because pre-gestational diabetes is rare, most cases occur in pregnancies without diabetes [50, 81].

9.3. Sirenomelia

The ICBDSR study, with 19 surveillance systems and 25,290,172 births, found 249 cases, or 0.98/100,000, with the highest prevalence in the Mexican registry [19]. The country values recorded in Orphanet from that study are, per 100,000 births, 1.04 in France, 0.62 in Italy, 0.64 in Spain, 0.54 in the Netherlands, 0.7 in Finland, 0.74 in the USA, 1.6 in Canada, 2.36 in Mexico and 1.36 in Latin America; Colombia has a value of 12/100,000 in another study [10]. The Orphanet value of 0.98/100,000 associated with Brazil cites the protocol of the Brazilian Rare Diseases Network, which is not a measurement of the prevalence of sirenomelia [10].

9.4. Diabetic embryopathy

Diabetic embryopathy has no validated prevalence in Orphanet [33]. The population figures available are those of Bell et al. in the North of England and of the EUROCAT registries [64, 81]. The rate of congenital anomalies does not differ between type 1 and type 2 diabetes [50, 64].

9.5. MNX1-related neonatal diabetes

The incidence of neonatal diabetes is 1/90,000 to 1/260,000 live births, about half permanent, and MNX1 explains about 1% of the permanent form [7]. Two of the three patients described were Egyptian with the same variant, which led to the proposal of a possible founder effect in Egypt, not yet studied [21].

9.6. Reference networks and patient organisations

ERN eUROGEN, which covers rare urogenital malformations, published with ERN ERNICA the European guideline on anorectal malformations, which recommends lifelong follow-up in a reference centre and referral to patient organisations [29, 77]. ERN ITHACA covers rare malformation syndromes and has a working group dedicated to spina bifida and other dysraphisms [30]. The International Sacral Agenesis/Caudal Regression Association (iSACRA), founded in 2012 in the USA and a member of NORD, connects more than 1500 people and families and took part in the narrative review by Sánchez-Romero et al. on sacral agenesis [63, 82]. The Italian association AIMAR, of patients with anorectal malformations, is a co-author of the European guideline [29].

10. Illustrative Clinical Case

Yang and Fan published in 2025 the case of a Chinese family with Currarino syndrome in which the disease went unrecognised for five decades [51]. The proband, a male infant, had from birth delayed passage of meconium, recurrent abdominal distension and constipation. Examination showed a narrow anus with explosive defecation, and the barium enema revealed megacolon. Enema and anal dilatation temporarily relieved the symptoms, but, after discharge, the child remained dependent on liquid paraffin and anal dilatations. Two months later, digital rectal examination identified a mass in the posterior wall of the rectum. The infant underwent excision of the mass and radical surgery for megacolon; histology described congenital megacolon and a mature caudal teratoma, and the diagnosis of Currarino syndrome was then established [51].

The older brother had been admitted twice in infancy for recurrent constipation, fever and distension, and had also had radical surgery for megacolon with a histological diagnosis of congenital megacolon. The family history revealed that, since the 1970s, seven relatives had suffered from chronic constipation and abdominal distension, including the grandfather, the father and a sister; three died, among them the sister, in childhood, with distension, prolonged constipation, recurrent fever and vomiting. The authors attribute these deaths, hypothetically, to ascending infection to the central nervous system through the presacral mass or a communication with the spinal canal [51].

Exome sequencing of the two brothers identified the heterozygous variant NM_005515.4:c.852+1G>A in MNX1, which alters the splicing donor site of intron 2 and is classified in ClinVar as likely pathogenic; Sanger sequencing confirmed it in the father [41, 51]. Lumbosacral CT showed three different degrees of dysplasia in the same family: in the proband, sacral hypoplasia with a normal S1 (Pang type III); in the brother, hemisacral hypoplasia (type IV); and in the father, only coccygeal hypoplasia (type V) (figure 1 of the article: https://pmc.ncbi.nlm.nih.gov/articles/PMC12049736/figure/F1/) [51]. Immunohistochemistry of the resected colons of the two brothers showed calretinin-positive ganglion cells, which indicates that the megacolon was secondary to obstruction and not Hirschsprung disease [51]. The family expressed the intention of using in vitro fertilisation with preimplantation genetic diagnosis [51].

The case brings together several points of this article: constipation as the dominant and sometimes only manifestation, variable expressivity within the family (from a severely dysplastic sacrum to isolated coccygeal hypoplasia in an apparently healthy father), the risk of severe infection and the misdiagnosis of Hirschsprung disease, which led to resections that early recognition of the syndrome and treatment of the anal stenosis could have avoided [20, 51]. The management described does not correspond to the planning recommended by the reference series, which call for characterisation of the four components by MRI and histological confirmation of aganglionosis before surgery [27, 38].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

In Currarino syndrome, the vital prognosis is good when the presacral mass is treated: survival in the SCT-study was 96.9% [60]. Avoidable mortality results mainly from central nervous system infection and, rarely, from malignant transformation, and early diagnosis is the main modifiable factor [38, 58]. Functional morbidity is frequent and lasting: defecatory dysfunction in about 40% of operated patients, faecal soiling in 78% in a series of complete forms, urinary incontinence in 14–36% and neurogenic bladder in about 3%, with worse bowel function when sacral dysplasia is total or there were perioperative complications of the mass [52, 62]. Reported quality of life was high in a Chinese series [62].

In caudal regression, prognosis depends on the level of agenesis, the associated anomalies and renal function; in the most severe forms it is poor, and the great majority of survivors need permanent urological and orthopaedic follow-up [2, 63]. Cognition is usually normal, and the aim of treatment is the best possible function and quality of life, bearing in mind that problems may worsen with growth [3]. In anorectal malformation with sacral agenesis, spontaneous continence is the exception and most patients need structured bowel and bladder management programmes [18]. In MNX1-related neonatal diabetes, prognosis depends on the extrapancreatic anomalies: one of the three patients described died at 10 months [21]. Sirenomelia is lethal in most cases [10].

11.2. Surveillance

There are no validated surveillance intervals for Currarino syndrome or caudal regression. The table brings together the recommendations of the sources, indicating whether they are a guideline, a centre protocol or practice described in series.

The French study NCT00780117 evaluated annual lumbosacral MRI as a strategy in Currarino syndrome, with no results posted in the registry, and the 2025 North American data suggest that imaging surveillance may be less intensive than in isolated sacrococcygeal teratoma [54, 78].

12. Summary and Final Considerations

Currarino syndrome and caudal regression belong to the same embryological territory, but they are different entities. The first is an autosomal dominant disorder due to MNX1 haploinsufficiency, with a characteristic hemisacrum, presacral mass and anorectal anomaly, and expressivity so variable that many carriers have only constipation or imaging changes; the variant is found in almost all familial cases and in about one third of sporadic cases, and 7q36 deletions add intellectual disability and, sometimes, holoprosencephaly. The second is usually sporadic, multifactorial and associated with pre-gestational diabetes, which multiplies the risk but explains only a minority of cases; MNX1 is not involved, and VANGL1 and other genes act, at most, as susceptibility factors. At the periphery of the spectrum lie SAVA due to TBXT, recessive and recognisable by abnormal vertebral ossification, neonatal diabetes due to biallelic MNX1 and sirenomelia.

In clinical practice, every anorectal malformation requires assessment of the sacrum, spinal cord and kidneys in the first month, and every case of severe constipation with anal stenosis, a retrorectal mass or suspected Hirschsprung disease should bring Currarino syndrome into consideration. Before surgery, the four components (anorectal, sacral, presacral and spinal cord) should be characterised by MRI, because communication of the mass with the spinal canal and tethered cord cause meningitis and neurological deficit. The teratoma of the syndrome is almost always mature and the risk of malignant transformation is much lower than that of isolated sacrococcygeal teratoma, but it is not zero. Long-term bowel and urinary morbidity is frequent and justifies lifelong multidisciplinary follow-up, with structured bowel management, renal protection and organised transition. Genetic testing begins with MNX1 sequencing with dosage analysis, or with microarray if there is intellectual disability or dysmorphism, and a negative result does not rule out the diagnosis. Counselling must explain dominant inheritance with unpredictable expressivity and offer sacral imaging to relatives. In caudal regression associated with diabetes, prevention depends on glycaemic control before conception and high-dose folic acid. There are no targeted therapies or ongoing trials, and the available evidence is retrospective; registries and European reference networks are the way to improve knowledge of these disorders.

13. References

1. OMIM. #176450 Currarino syndrome. Online Mendelian Inheritance in Man. Baltimore: Johns Hopkins University. Available from: https://omim.org/entry/176450
2. Jones KL, Jones MC, del Campo M. Caudal dysplasia sequence (caudal regression syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 896–897.
3. Jasiewicz B, Kacki W. Caudal Regression Syndrome-A Narrative Review: An Orthopedic Point of View. Children (Basel). 2023;10(3):589. doi:10.3390/children10030589 PMID: 36980147
4. OMIM. #600145 Sacral defect with anterior meningocele (SDAM). Available from: https://omim.org/entry/600145
5. Orphanet. Sacral agenesis-abnormal ossification of the vertebral bodies-persistent notochordal canal syndrome (ORPHA:397927). Available from: https://www.orpha.net/en/disease/detail/397927
6. OMIM. #615709 Sacral agenesis with vertebral anomalies (SAVA). Available from: https://omim.org/entry/615709
7. De León DD, Pinney SE. Permanent Neonatal Diabetes Mellitus. 2008 Feb 8 [updated 2024 Nov 14; revised 2025 Sep 25]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1447/
8. OMIM. *142994 Motor neuron and pancreas homeobox 1; MNX1. Available from: https://omim.org/entry/142994
9. Jones KL, Jones MC, del Campo M. Sirenomelia sequence. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 894–895.
10. Orphanet. Sirenomelia (ORPHA:3169). Available from: https://www.orpha.net/en/disease/detail/3169
11. GeneGuidance. MNX1 — clinical curation of the gene and the condition Currarino syndrome (version of 2026-09-26).
12. Merello E, De Marco P, Mascelli S, Raso A, Calevo MG, Torre M, et al. HLXB9 homeobox gene and caudal regression syndrome. Birth Defects Res A Clin Mol Teratol. 2006;76(3):205-9. doi:10.1002/bdra.20234 PMID: 16498628
13. Postma AV, Alders M, Sylva M, Bilardo CM, Pajkrt E, van Rijn RR, et al. Mutations in the T (brachyury) gene cause a novel syndrome consisting of sacral agenesis, abnormal ossification of the vertebral bodies and a persistent notochordal canal. J Med Genet. 2014;51(2):90-7. doi:10.1136/jmedgenet-2013-102001 PMID: 24253444
14. Flanagan SE, De Franco E, Lango Allen H, Zerah M, Abdul-Rasoul MM, Edge JA, et al. Analysis of transcription factors key for mouse pancreatic development establishes NKX2-2 and MNX1 mutations as causes of neonatal diabetes in man. Cell Metab. 2014;19(1):146-54. doi:10.1016/j.cmet.2013.11.021 PMID: 24411943
15. Orphanet. Currarino syndrome (ORPHA:1552). Available from: https://www.orpha.net/en/disease/detail/1552
16. Orphanet. Caudal regression syndrome (ORPHA:3027). Available from: https://www.orpha.net/en/disease/detail/3027
17. Thottungal AD, Charles AK, Dickinson JE, Bower C. Caudal dysgenesis and sirenomelia-single centre experience suggests common pathogenic basis. Am J Med Genet A. 2010;152A(10):2578-87. doi:10.1002/ajmg.a.33599 PMID: 20734338
18. Theeuwes JN, de Beaufort CMC, McDonald CJ, Bakker DP, van Schuppen J, Kuijper CF, et al. Bowel and bladder outcomes in patients with anorectal malformations and sacral agenesis: a retrospective cohort study. Pediatr Surg Int. 2026;42(1):244. doi:10.1007/s00383-026-06471-x PMID: 42207205
19. Orioli IM, Amar E, Arteaga-Vazquez J, Bakker MK, Bianca S, Botto LD, et al. Sirenomelia: an epidemiologic study in a large dataset from the International Clearinghouse of Birth Defects Surveillance and Research, and literature review. Am J Med Genet C Semin Med Genet. 2011;157C(4):358-73. doi:10.1002/ajmg.c.30324 PMID: 22002878
20. Dworschak GC, Reutter HM, Ludwig M. Currarino syndrome: a comprehensive genetic review of a rare congenital disorder. Orphanet J Rare Dis. 2021;16(1):167. doi:10.1186/s13023-021-01799-0 PMID: 33836786
21. Aly HH, De Franco E, Flanagan SE, Elhenawy YI. MNX1 mutations causing neonatal diabetes: Review of the literature and report of a case with extra-pancreatic congenital defects presenting in severe diabetic ketoacidosis. J Diabetes Investig. 2023;14(4):516-521. doi:10.1111/jdi.13968 PMID: 36586106
22. Kole MJ, Fridley JS, Jea A, Bollo RJ. Currarino syndrome and spinal dysraphism. J Neurosurg Pediatr. 2014;13(6):685-9. doi:10.3171/2014.3.PEDS13534 PMID: 24745342
23. Cuturilo G, Hodge JC, Runke CK, Thorland EC, Al-Owain MA, Ellison JW, et al. Phenotype analysis impacts testing strategy in patients with Currarino syndrome. Clin Genet. 2016;89(1):109-14. doi:10.1111/cge.12572 PMID: 25691298
24. Merello E, De Marco P, Ravegnani M, Riccipetitoni G, Cama A, Capra V. Novel MNX1 mutations and clinical analysis of familial and sporadic Currarino cases. Eur J Med Genet. 2013;56(12):648-54. doi:10.1016/j.ejmg.2013.09.011 PMID: 24095820
25. Krishnan V, Jaganathan S, Jayappa S, Glasier C, Choudhary A, Ramakrishnaiah R, et al. Clinical and radiological evaluation of caudal regression syndrome. Pediatr Radiol. 2024;54(9):1451-1461. doi:10.1007/s00247-024-05945-1 PMID: 38750326
26. Firth HV, Hurst JA. Anal anomalies (atresia, stenosis). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 58–60.
27. Crétolle C, Zérah M, Jaubert F, Sarnacki S, Révillon Y, Lyonnet S, et al. New clinical and therapeutic perspectives in Currarino syndrome (study of 29 cases). J Pediatr Surg. 2006;41(1):126-31; discussion 126-31. doi:10.1016/j.jpedsurg.2005.10.053 PMID: 16410121
28. National Institute for Health and Care Excellence. Diabetes in pregnancy: management from preconception to the postnatal period (NICE guideline NG3). Publicada 25 fev 2015; atualizada 16 dez 2020. Available from: https://www.nice.org.uk/guidance/ng3
29. Aubert O, Irvine WFE, Aminoff D, de Blaauw I, Cascio S, Cretolle C, et al. ERN eUROGEN Guidelines on the Management of Anorectal Malformations Part I: Diagnostics. Eur J Pediatr Surg. 2025;35(2):104-111. doi:10.1055/s-0044-1791250 PMID: 39393414
30. European Reference Network ITHACA (Intellectual disability, TeleHealth, Autism and Congenital Anomalies) — European Network for Rare Malformation Syndromes; working group WG13 Spina Bifida & other Dysraphisms. Available from: https://ern-ithaca.eu/
31. Porsch RM, Merello E, De Marco P, Cheng G, Rodriguez L, So M, et al. Sacral agenesis: a pilot whole exome sequencing and copy number study. BMC Med Genet. 2016;17(1):98. doi:10.1186/s12881-016-0359-2 PMID: 28007035
32. Orphanet. Caudal regression-sirenomelia spectrum (ORPHA:444941). Available from: https://www.orpha.net/en/disease/detail/444941
33. Orphanet. Diabetic embryopathy (ORPHA:1926). Available from: https://www.orpha.net/en/disease/detail/1926
34. Currarino G, Coln D, Votteler T. Triad of anorectal, sacral, and presacral anomalies. AJR Am J Roentgenol. 1981;137(2):395-8. doi:10.2214/ajr.137.2.395 PMID: 6789651
35. Ross AJ, Ruiz-Perez V, Wang Y, Hagan DM, Scherer S, Lynch SA, et al. A homeobox gene, HLXB9, is the major locus for dominantly inherited sacral agenesis. Nat Genet. 1998;20(4):358-61. doi:10.1038/3828 PMID: 9843207
36. Hagan DM, Ross AJ, Strachan T, Lynch SA, Ruiz-Perez V, Wang YM, et al. Mutation analysis and embryonic expression of the HLXB9 Currarino syndrome gene. Am J Hum Genet. 2000;66(5):1504-15. doi:10.1086/302899 PMID: 10749657
37. Belloni E, Martucciello G, Verderio D, Ponti E, Seri M, Jasonni V, et al. Involvement of the HLXB9 homeobox gene in Currarino syndrome. Am J Hum Genet. 2000;66(1):312-9. doi:10.1086/302723 PMID: 10631160
38. Martucciello G, Torre M, Belloni E, Lerone M, Pini Prato A, Cama A, et al. Currarino syndrome: proposal of a diagnostic and therapeutic protocol. J Pediatr Surg. 2004;39(9):1305-11. doi:10.1016/j.jpedsurg.2004.05.003 PMID: 15359381
39. Costanzo S, Spaccini L, Pio L, Mattioli G, Virgone C, Dall'Igna P, et al. Currarino syndrome: does the presence of a genetic anomaly correlate with a more severe phenotype? A multicentre study. J Pediatr Surg. 2017;52(10):1591-1596. doi:10.1016/j.jpedsurg.2017.06.012 PMID: 28689883
40. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Hypertrophic osteoarthropathy, autosomal recessive (MIM 259100). In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 653–655.
41. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-8. doi:10.1002/humu.23348. Available from: https://variantvalidator.org (accessed 29 Oct 2026).
42. Crétolle C, Pelet A, Sanlaville D, Zérah M, Amiel J, Jaubert F, et al. Spectrum of HLXB9 gene mutations in Currarino syndrome and genotype-phenotype correlation. Hum Mutat. 2008;29(7):903-10. doi:10.1002/humu.20718 PMID: 18449898
43. Firth HV, Hurst JA. Deletions and duplications (microdeletions). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 660–665.
44. Jung JH, Jeong J, Kim SH. Currarino syndrome associated with an isolated 7q terminal deletion in Korea: a case report. J Yeungnam Med Sci. 2026;43:8. doi:10.12701/jyms.2026.43.8 PMID: 41486773
45. Wang X, Min L, Bao L, Wei J, Wang X. Clinical Diversity in Currarino Syndrome: Insights From Monozygotic Twins and a Single-Center Retrospective Study. Congenit Anom (Kyoto). 2026;66(1):e70046. doi:10.1002/cga.70046 PMID: 41871837
46. Firth HV, Hurst JA. Congenital cystic lung lesions, Currarino syndrome, and sacrococcygeal teratoma. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 712–713.
47. Bonnefond A, Vaillant E, Philippe J, Skrobek B, Lobbens S, Yengo L, et al. Transcription factor gene MNX1 is a novel cause of permanent neonatal diabetes in a consanguineous family. Diabetes Metab. 2013;39(3):276-80. doi:10.1016/j.diabet.2013.02.007 PMID: 23562494
48. Warner T, Scullen TA, Iwanaga J, Loukas M, Bui CJ, Dumont AS, et al. Caudal Regression Syndrome-A Review Focusing on Genetic Associations. World Neurosurg. 2020;138:461-467. doi:10.1016/j.wneu.2020.03.057 PMID: 32200015
49. GeneGuidance. VANGL1 — clinical curation of the gene (version of 2026-09-26).
50. Firth HV, Hurst JA. Maternal diabetes mellitus and diabetic embryopathy. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 756–757.
51. Yang G, Fan W. Familial constipation and death: one family with Currarino syndrome. BMJ Case Rep. 2025;18(4):e260455. doi:10.1136/bcr-2024-260455 PMID: 40306754
52. Tan X, Li G, Jin Y, Li S, Zhang Y, Liao J, et al. Follow-up of defecation, urination, growth and development in children with Currarino syndrome. Transl Pediatr. 2026;15(5):175. doi:10.21037/tp-2026-1-0153 PMID: 42292623
53. Pio L, Piatelli G, Rossi A, Scarsi P, Merello E, Capra V, et al. Urological outcome in patients with Currarino syndrome. J Pediatr Surg. 2014;49(11):1643-6. doi:10.1016/j.jpedsurg.2014.05.038 PMID: 25475810
54. Srivatsa S, Gil L, Zhang Y, Rymeski B, Gavulic A, Mak G, et al. Sacrococcygeal Teratomas in Currarino Syndrome: A Multicenter Review of Tumor Characteristics, Surgical Outcomes, and Recurrence. J Pediatr Surg. 2025;60(9):162420. doi:10.1016/j.jpedsurg.2025.162420 PMID: 40541662
55. Patel RA, Du R, Rowell E, Grabowski JE, Jamshidi P, Wadhwani NR, et al. Atypical presacral masses in Currarino syndrome: a case series. Childs Nerv Syst. 2026;42(1):389. doi:10.1007/s00381-026-07486-4 PMID: 42864507
56. Shah KH, Tigchelaar SS, Khalafallah AM, Levi AD. Neurosurgical management of a giant anterior sacral meningocele in a patient with combined Currarino syndrome and caudal duplication anomaly: illustrative case. J Neurosurg Case Lessons. 2025;9(23):CASE2572. doi:10.3171/CASE2572 PMID: 40489955
57. Rod J, Cretolle C, Faivre L, Jacquot C, Yacoub O, Ravasse P, et al. Malignant transformation of presacral mass in Currarino syndrome. Pediatr Blood Cancer. 2019;66(6):e27659. doi:10.1002/pbc.27659 PMID: 30740898
58. Yoshida A, Maoate K, Blakelock R, Robertson S, Beasley S. Long-term functional outcomes in children with Currarino syndrome. Pediatr Surg Int. 2010;26(7):677-81. doi:10.1007/s00383-010-2615-4 PMID: 20473613
59. Dirix M, van Becelaere T, Berkenbosch L, van Baren R, Wijnen RM, Wijnen MH, et al. Malignant transformation in sacrococcygeal teratoma and in presacral teratoma associated with Currarino syndrome: a comparative study. J Pediatr Surg. 2015;50(3):462-4. doi:10.1016/j.jpedsurg.2014.07.014 PMID: 25746708
60. van Heurn LJ, Derikx JPM, Hall NJ, AbouZeid AA, de Campos Vieira Abib S, Chirdan LB, et al. Malignant transformation of sacrococcygeal teratoma versus presacral teratoma in Currarino syndrome: Results of 'The SCT-study'. J Pediatr Surg. 2026;61(6):162848. doi:10.1016/j.jpedsurg.2025.162848 PMID: 42250912
61. Sakurai T, Nakamura M, Endo N. Postoperative complications and long-term outcomes in Currarino syndrome. Pediatr Surg Int. 2021;37(12):1773-1781. doi:10.1007/s00383-021-04984-1 PMID: 34462793
62. Cao Y, Pang W, Zhang Z, Huang X, Wang K, Yan J, et al. Medium-long term bowel function and quality of life in complete Currarino syndrome. Pediatr Surg Int. 2026;42(1):303. doi:10.1007/s00383-026-06548-7 PMID: 42455303
63. Sánchez-Romero M, Tlaxcala-Castillo L, Pichardo-Rojas PS, Valencia-Melo MA, Paz-López ÁA, Sánchez-Sagastegui F, et al. Sacral Agenesis. Pediatr Neurol. 2025;163:27-34. doi:10.1016/j.pediatrneurol.2024.10.020 PMID: 39642685
64. Bell R, Glinianaia SV, Tennant PW, Bilous RW, Rankin J. Peri-conception hyperglycaemia and nephropathy are associated with risk of congenital anomaly in women with pre-existing diabetes: a population-based cohort study. Diabetologia. 2012;55(4):936-47. doi:10.1007/s00125-012-2455-y PMID: 22314812
65. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Femoral-facial syndrome (MIM 134780). In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 876–879.
66. Firth HV, Hurst JA. Short limbs. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 784–787.
67. Ferreira C, Santos AP, Fonseca J. Currarino syndrome - a pre and post natal diagnosis correlation: case report and literature review. J Matern Fetal Neonatal Med. 2022;35(25):5224-5226. doi:10.1080/14767058.2021.1876021 PMID: 33618589
68. ClinicalTrials.gov. NCT05033275 — Screening for Sacral Agenesis in Offspring of Mothers With Diabetes in Pregnancy. Available from: https://clinicaltrials.gov/study/NCT05033275
69. Tian T, Jiang W, Chen H, Hu T, Luo H. Initially suspected case of prenatal VACTERL association ultimately diagnosed as Currarino syndrome via magnetic resonance imaging: a case description. Quant Imaging Med Surg. 2026;16(4):333. doi:10.21037/qims-2025-1-2639 PMID: 41972029
70. Ge T, Song X, Li C, Deng X, Lin X. Prenatal diagnosis of caudal regression syndrome with maternal diabetes mellitus via ultrasound: a case description. Quant Imaging Med Surg. 2025;15(8):7613-7617. doi:10.21037/qims-2025-4 PMID: 40785861
71. Firth HV, Hurst JA. Laterality disorders (heterotaxy, isomerism). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 206–207.
72. Jones KL, Jones MC, del Campo M. Fanconi pancytopenia syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 452–455.
73. OMIM. #607864 Caudal duplication anomaly. Available from: https://omim.org/entry/607864
74. Jones KL, Jones MC, del Campo M. Monozygotic twinning and structural defects—general. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 914–921.
75. Firth HV, Hurst JA. Limb reduction defects. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 212–215.
76. Aubert O, Irvine WFE, Aminoff D, de Blaauw I, Cascio S, Cretolle C, et al. European Reference Network eUROGEN Guidelines on the Management of Anorectal Malformations, Part II: Treatment. Eur J Pediatr Surg. 2025;35(2):112-119. doi:10.1055/s-0044-1791257 PMID: 39299249
77. Aubert O, Irvine WFE, Aminoff D, de Blaauw I, Cascio S, Cretolle C, et al. ERN eUROGEN Guidelines on the Management of Anorectal Malformations Part III: Lifelong Follow-up and Transition of Care. Eur J Pediatr Surg. 2025;35(2):120-127. doi:10.1055/s-0044-1791249 PMID: 39299248
78. ClinicalTrials.gov. NCT00780117 — Characterization of At-risk Population and Prognosis Factors for Sacro-coccygeal Teratoma in Currarino Syndrome. Available from: https://clinicaltrials.gov/study/NCT00780117
79. Ehrlich PF, Petroze RT. Invited commentary re: "Malignant transformation of sacrococcygeal teratoma (SCT) versus presacral teratoma in Currarino syndrome (CS): Results of 'The SCT-study,'". J Pediatr Surg. 2026;61(6):162992. doi:10.1016/j.jpedsurg.2026.162992 PMID: 41651110
80. Lynch SA, Wang Y, Strachan T, Burn J, Lindsay S. Autosomal dominant sacral agenesis: Currarino syndrome. J Med Genet. 2000;37(8):561-6. doi:10.1136/jmg.37.8.561 PMID: 10922380
81. Garne E, Loane M, Dolk H, Barisic I, Addor MC, Arriola L, et al. Spectrum of congenital anomalies in pregnancies with pregestational diabetes. Birth Defects Res A Clin Mol Teratol. 2012;94(3):134-40. doi:10.1002/bdra.22886 PMID: 22371321
82. International Sacral Agenesis/Caudal Regression Association (iSACRA). Available from: https://isacra.org/