Spondylocostal and Spondylothoracic Dysostoses

Spondylocostal dysostosis types 1 to 7, spondylothoracic dysostosis, TBX6-associated vertebral segmentation defects and the Jarcho-Levin eponym

Literature and regulatory information updated: 29 October 2026.

0. Initial Reference Table

The spondylocostal dysostoses (SCDO) form a class of axial skeletal malformations defined radiologically: segmentation defects of at least ten contiguous vertebrae, associated with malaligned, fused or reduced numbers of ribs, without major asymmetry of the thorax [1, 2]. Spondylothoracic dysostosis (STD) is the form caused by biallelic MESP2 variants, with symmetrical posterior rib fusion and a “crab-like” thorax [1, 3]. OMIM has added a seventh type, SCDO7, caused by biallelic DMRT2 variants and described in two newborns [4]. The following table brings together the identifiers and reference data for each entity in the class.

1. Introduction

1.1. Founding descriptions and the problem of the eponym

In 1938, Saul Jarcho and Paul Levin, of Johns Hopkins Hospital, published the description of hereditary malformations of the vertebral bodies with thoracic insufficiency caused by vertebral and rib anomalies [5, 15]. Sources differ about the original patients: OMIM refers to a Black brother and sister from Baltimore, whose disorder the authors mistakenly considered identical to Klippel-Feil syndrome, whereas Smith's textbook and the Puerto Rico group refer to Puerto Rican patients [15–17]. Almost thirty years later, in 1966, Norman Lavy and colleagues, of Indiana University, described a similar syndrome in a Puerto Rican family, with 4 of 7 children of a third-cousin marriage affected and dying of respiratory infection before 1 year of age, and John E. Moseley, in New York, first used the designation spondylothoracic dysplasia [5, 16]. In 1974, Pérez-Comas and García-Castro described six Puerto Rican cases under the designation occipito-facial-cervico-thoraco-abdomino-digital dysplasia [15, 16]. In 1996, Mortier and colleagues analysed 26 new and 115 published patients and distinguished three entities: a lethal recessive form with a symmetrical crab-like thorax, which they called Jarcho-Levin, a benign dominant SCDO and a recessive STD overlapping with SCDO [16]. Over the following decades, the Jarcho-Levin eponym came to cover almost every phenotype with vertebral segmentation defects and abnormal rib alignment, including presentations bearing no resemblance to the original patients, and also the severe Puerto Rican form now called STD [1]. This overlap of names mixed phenotypes with very different prognoses and hampered estimates of mortality and counselling [18].

The distinction between dysostosis and dysplasia is also meaningful: dysostoses result from defects in genes active mainly in early developmental stages and produce “frozen” malformations, whereas dysplasias result from genes active later, and for this reason OMIM and GeneReviews prefer the term dysostosis for these disorders [1, 16]. Current sources do not agree on the fate of the eponym, and this divergence has practical consequences for reading the literature. The atlas of bone dysplasias by Spranger and colleagues recommends abandoning the term Jarcho-Levin, designating generalised segmentation defects of proven or presumed genetic aetiology as spondylocostal dysostosis and calling sporadic localised malformations vertebral segmentation defects; it also advises against retaining the term spondylothoracic for the crab-like thorax, because the same genes cause the typical form [6]. Smith's textbook uses Jarcho-Levin specifically for STD and describes it as a severe form of SCDO type 2 [15]. Berdon and colleagues proposed reserving Jarcho-Levin for SCDO and calling STD Lavy-Moseley syndrome [5]. In 2025, De Jesús-Rojas and Cornier argued, on the contrary, that the two 1938 patients had the radiographic phenotype of STD, that STD and SCDO should be treated as separate entities and that, if the eponym is used, it should apply to STD [17], a position disputed in a letter that considered the eponym more appropriate for SCDO [19]. Orphanet retains “Jarcho-Levin syndrome” as a synonym of autosomal recessive SCDO (ORPHA:2311) [2]. For these reasons, a report using only the eponym requires review of the images and the genotype before it is assigned to an entity.

1.2. Milestones of molecular discovery

The first step was homozygosity mapping in a large consanguineous Arab-Israeli family with six affected members (lod score 6.9 at 19q13), confirmed in a Pakistani family (lod 2.4) [16, 20], and the localisation of autosomal recessive SCDO to a 7.8 cM interval at 19q13.1-q13.3, homologous to the mouse region containing Dll3, the gene mutated in the radiation-induced pudgy mutant [20, 21]. In 2000, Bulman and colleagues identified DLL3 variants in three families, two truncating variants in extracellular domains and one missense variant at a conserved glycine of the fifth epidermal growth factor (EGF)-like repeat, the first pathogenic variants in a human Delta homologue [21]. In 2003, Turnpenny and colleagues described 12 variants in a further 10 families of diverse ethnic origins, six from traditionally consanguineous communities, and proposed the term “pebble beach sign” and the designation SCD type 1 [22].

MESP2 followed, through linkage to 15q21.3-q26.1 in a consanguineous family with a four-base-pair duplication (2004) [23]; LFNG, with a missense variant at a conserved phenylalanine near the active site that produced an inactive and mislocalised enzyme (2006) [24]; and HES7 (2008) [25]. In 2008, Cornier and colleagues showed that Puerto Rican STD results from biallelic MESP2 variants, with the nonsense variant E103X present in all probands of 12 families [3]. In 2013, exome sequencing identified a TBX6 variant that abolishes the stop codon in a family with dominant SCDO, the first genetic cause of an AD form [26]. In 2015, RIPPLY2 was added, in two siblings with multiple regional defects [27], together with the compound inheritance model of TBX6 in congenital scoliosis, with a rare null allele on one chromosome and a common hypomorphic haplotype on the other [28]. In 2018 and in 2026, two newborns with biallelic DMRT2 variants and severe costovertebral anomalies led OMIM to define SCDO7 (621523) [4, 29, 30].

1.3. Current classifications

The International Consortium for Vertebral Anomalies and Scoliosis (ICVAS) proposed a classification system for vertebral segmentation defects (SDV) in 2007 and evaluated it in 2010 [31, 32]. The clinical algorithm assigns all forms of SDV to seven broad categories, combining the radiological description with specific diagnoses when the genotype or the syndrome is established; the research algorithm uses an ontology applicable to humans and animal models [1]. In this classification, SCDO is the preferred term for generalised segmentation defects with rib involvement [1]. In a pilot evaluation, seven general radiologists agreed with the consensus diagnosis in 1 of 70 responses (1.4%) before the system was explained, with 54.3% “don't know” responses, and in 47 of 70 (67.1%; 95% confidence interval, 95% CI, 55.5–77.0) afterwards, with reporting time falling from 148 to 48 minutes [32]. In 2017 ICVAS merged with the International Consortium for Scoliosis Genetics, forming the International Consortium for Spinal Genetics, Development and Disease (ICSGDD) [33].

The 2023 revision of the Nosology of Genetic Skeletal Disorders classifies STD as a “vertebral segmentation defect (congenital scoliosis) with variable penetrance”; the GeneReviews authors disagree, because congenital scoliosis usually results from defects of one or a few vertebrae in one region, whereas STD is a generalised defect in which scoliosis is not a major feature [1, 34].

1.4. Definition of the entities in this article

SCDO is defined radiologically by segmentation defects of at least ten contiguous vertebrae, usually throughout the spine, with rib anomalies (malalignment, a variable number of intercostal fusions, sometimes a reduced number of ribs) and without major asymmetry of the thorax; clinically, by a trunk short in proportion to height, a short neck and mild non-progressive scoliosis in most patients [1]. The six types recognised in GeneReviews and Orphanet correspond to the genes DLL3, MESP2, LFNG, HES7, TBX6 and RIPPLY2, and OMIM adds SCDO7, due to DMRT2 [1, 2, 4]. OMIM classifies Puerto Rican STD within SCDO2, because they share the MESP2 gene [35]. STD is defined by severe shortening of the whole spine, especially the thoracic spine, bilateral and symmetrical fusion of the ribs at the costovertebral junction with straight, aligned ribs that fan out (“crab”) and early prominence of the pedicles (tramline sign) [1, 9]. TBX6-associated congenital scoliosis (TACS) designates regional vertebral malformations, mainly low thoracic or high lumbar hemivertebrae, caused by the combination of a TBX6 null allele and a hypomorphic haplotype [1, 28]. Scoliosis of syndromic or monogenic cause in general and the Klippel-Feil anomaly are addressed here only in the differential diagnosis.

2. Molecular Genetics and Pathophysiology

2.1. Somitogenesis and the segmentation clock

The vertebrae and ribs derive from the somites, blocks of paraxial mesoderm that bud off periodically from the anterior end of the presomitic mesoderm. The periodicity depends on a molecular oscillator, the segmentation clock, in which the Notch, FGF and Wnt pathways interact with cyclically expressed genes [1, 36, 37]. The six SCDO genes encode components of the Notch pathway or its direct regulators, and the common mechanism is loss of function [1]. When the clock fails, somite boundaries form irregularly and the rostrocaudal polarity of each somite is lost; the result is a spine with fused vertebrae, hemivertebrae and bodies of variable size and shape, and ribs that arise from abnormal segments and fuse or are missing [38, 39]. Because the defect is fixed in the embryonic period, the malformations are present at birth and do not increase with age, although the curvature and respiratory function may evolve with growth [1, 10].

2.2. Function of each gene

DLL3 encodes a transmembrane ligand of the Delta family, with a signal sequence, a DSL domain, seven EGF repeats and a transmembrane domain, which, unlike DLL1, inhibits NOTCH1 signalling in the cell that expresses it (cis-inhibition) in the presomitic mesoderm [1, 38, 40]. MESP2 encodes a transcription factor with a basic helix-loop-helix (bHLH) domain expressed in the anterior presomitic mesoderm; it is a direct target of NOTCH1, defines the rostrocaudal polarity of each somite by suppressing Notch activity and reducing DLL1 in the presumptive rostral domain, activates LFNG during boundary formation and modulates the FGF pathway; in the mouse, the somite boundary forms at the interface between domains with active and repressed Notch1, created by Mesp2-mediated suppression of Notch through Lfng [1, 39, 41]. LFNG encodes a Golgi glycosyltransferase that elongates O-fucose on the EGF repeats of Notch receptors; this modification inhibits NOTCH1 activation by JAG1 and enhances activation by DLL1, and LFNG is itself a cyclic gene targeted by NOTCH1; in Lfng-null mice, Hes7 oscillations lose synchrony between cells, which shows the role of LFNG in the intercellular coupling of the clock [1, 42, 43]. HES7 encodes a 225-amino-acid transcriptional repressor, with bHLH and orange domains and the carboxy-terminal WRPW motif, of the hairy/enhancer of split family; it is a direct target of NOTCH1, an oscillating component of the clock and part of a negative feedback loop that attenuates Notch signalling [1, 44, 45]. TBX6 encodes a T-box transcription factor that specifies paraxial mesoderm, binds directly to the region upstream of MESP2 and activates its expression, activates DLL1 and influences left-right axis determination; in the mouse, Tbx6 represses Sox2 through the N1 enhancer, and in its absence the paraxial mesoderm is transformed into ectopic neural tubes [1, 46, 47]. RIPPLY2 encodes a small nuclear repressor with the WRPW tetrapeptide, which recruits Groucho/TLE corepressors; it is a transcriptional target of MESP2 and TBX6, negatively regulates MESP2 and promotes the degradation of TBX6 at the segment boundary; the human protein has 128 amino acids, and the variant p.(Arg80Ter) reduces repressor activity in vitro [1, 48, 49].

2.3. Variant spectrum, founder variants and nomenclature

The DLL3 spectrum includes nonsense, missense, frameshift and splicing variants and in-frame insertions, mainly in exons 4 to 8, predicted to truncate the mature protein in the extracellular domain or to alter conserved residues of the EGF repeats [22]. Two patients with DLL3 deletions have been described, one of exons 2 to 4 detected by multiplex ligation-dependent probe amplification (MLPA) and one of the whole gene detected by array comparative genomic hybridisation [1]. In MESP2, the Puerto Rican founder variant E103X and the other STD variants lie in exon 1 [3, 15]. The following table presents the reference variants in current Human Genome Variation Society (HGVS) notation, confirmed in VariantValidator (GRCh38, MANE Select transcripts, from Matched Annotation from NCBI and EMBL-EBI) and ClinVar [50, 51].

Three nomenclature points avoid errors of interpretation. The E230X variant of the 2008 article corresponds, in the current transcript, to p.(Glu234Ter), and the correspondence should be made only through the ClinVar record and not by manual calculation [51]. The L125V variant, described as pathogenic in 2008, is now classified as of uncertain significance in ClinVar [51]. GeneReviews identifies the third marker of the haplotype as “rs38090627”, a spelling that matches none of the identifiers in the ClinVar record of the haplotype (rs2289292, rs3809624 and rs3809627) [1, 51]. The RIPPLY2 articles use version NM_001009994.2, and the c.240-4T>G variant was classified as of uncertain significance under the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria in the report by Wegler and colleagues, although it is found in the homozygous state in three patients with an overlapping phenotype [54, 55].

2.4. Genotype–phenotype correlations

In DLL3, the radiological phenotype is very constant; two patients homozygous for missense variants in the EGF domain had somewhat milder phenotypes, attributed to proteins that fold correctly but are thermodynamically less stable, whereas other missense variants produce a picture indistinguishable from that of truncating variants [1]. In MESP2, the correlation is the clearest in the class. The c.500_503dup duplication lies after the bHLH domain and creates a premature stop codon in the last exon; the messenger RNA escapes nonsense-mediated decay (NMD) and the truncated protein retains the bHLH domain, possibly retaining some function, which is consistent with the milder SCDO2 [1]. The STD nonsense variants lie in exon 1 and are predicted to target the RNA for NMD, with reduced or absent protein [1]. Homozygotes for p.(Glu103Ter) appear to have a more severe phenotype than compound heterozygotes with at least one missense variant [15]. In 2025, the Puerto Rico group stated that it knew of no MESP2 variant producing a phenotype other than STD in any population and described variable expressivity among carriers of p.(Glu103Ter) [17]. For HES7, LFNG and RIPPLY2, no genotype–phenotype correlations have been established [1].

2.5. TBX6: gene dosage, compound inheritance and dominant and recessive forms

TBX6 is the gene in the class with the most complex inheritance mechanism, and the simple label of dominant or recessive is misleading [46]. In the study by Wu and colleagues, published in 2015, 17 of 161 Han Chinese patients with sporadic congenital scoliosis (11%) had a heterozygous TBX6 null allele (12 16p11.2 deletions, one nonsense variant and four frameshift variants), against none of 166 controls (P<3.8×10⁻⁶) [28]. Deletion carriers within the same families were discordant for the phenotype, which showed that the null allele alone is not sufficient; the second risk allele was a common haplotype, present in all 17 carriers (P<1.1×10⁻⁶), confirmed in 76 additional Chinese patients and in a multicentre series of 42 carriers of the 16p11.2 deletion [28]. The model was replicated in 94 Japanese patients, in whom the nine null-allele carriers had the haplotype in trans and one of the alleles had a dominant-negative effect [8], and in 108 Chinese patients with congenital vertebral malformations, 10 of whom (9.3%) had compound inheritance [56]. In mice generated by CRISPR-Cas9, high penetrance of the vertebral phenotype appeared only with the combination of a null allele and a hypomorphic allele, which supports a dosage threshold below 50% of normal function [46, 56]. Missense variants with loss of function (lower transcriptional activity, abnormal cellular localisation) have broadened the spectrum of null alleles; in a patient homozygous for NM_004608.4:c.418C>T p.(Leu140Phe), the deformity was more severe [52]. In a new series of 102 Japanese patients with congenital scoliosis, the null alleles were five 16p11.2 deletions, one splicing variant and three missense variants, all with the risk haplotype in trans, and, added to previous data, about 9.2% of Japanese patients had a TBX6 variant [57]. The T-C-A haplotype is common: about 44% in Asian populations, 33% in European populations and less than 1% in African populations, which leads bioinformatic pipelines to filter it out as a polymorphism and requires it to be sought explicitly [57]. Homozygosity for null alleles is probably embryonic lethal, and SCDO5 therefore corresponds to a dosage below haploinsufficiency but not zero [57].

At the two extremes of this spectrum are recessive SCDO5 due to biallelic variants, for example two missense variants without the risk haplotype in a Japanese patient with SCDO [57], with six families published up to 2023 and a phenotype almost indistinguishable from SCDO1 [1, 11], and the dominant SCDO5 of the three-generation Macedonian family, whose variant abolishing the stop codon reduced transcriptional activation to about half [1, 26, 58]. This variant was on the same chromosome as the permissive T-C-A haplotype, and the combined reduction in expression and activity, or the effect of the 81 amino acids added at the carboxy-terminal end on protein interactions, may explain the phenotype [1]. Takeda and colleagues accept that dominant SCDO and TACS may be part of the same spectrum [8]. OMIM records SCDO5 as AD and AR, and Orphanet separates the two forms (AD in ORPHA:1797; AR in ORPHA:2311) [2, 7, 46, 47]. In the replication data of Wu and colleagues, 6 of 76 additional Chinese patients (8%) had compound inheritance, and in the series of 42 carriers of the 16p11.2 deletion, 5 of the 6 with congenital scoliosis (83%) had the haplotype on the other allele [28, 47].

2.6. The proximal 16p11.2 deletion

The recurrent proximal 16p11.2 deletion (hg18 coordinates chr16:29.5–30.15 Mb) contains TBX6 and acts as a null allele [28, 59]. In a series of postnatal microarrays, the deletion was found in 146 of 33,226 cases (0.44%) against 6 of 22,246 controls (0.03%), was de novo in 33 of 47 cases with information (70.2%) and had an estimated penetrance of 46.8% (95% CI 31.5–64.2) for a clinical phenotype, mainly neurodevelopmental; the reciprocal duplication had an estimated penetrance of 27.2% (95% CI 17.4–40.7) [59]. These estimates refer to the neurodevelopmental phenotype, not to the vertebral malformations. The deletion phenotype includes motor speech, language and coordination disorders, psychiatric disturbances and autistic traits; vertebral anomalies and scoliosis appear only when the other TBX6 allele carries the hypomorphic haplotype [1]. Among the reports collected in OMIM is that of a 3-year-old boy with the 593 kb deletion, multiple hemivertebrae, absent ribs, inguinal hernia and developmental delay [60].

2.7. Allelic disorders and TBX6-associated phenotypes

Heterozygous TBX6 deletions, sometimes with the hypomorphic allele on the other chromosome, have been described in Müllerian aplasia (Mayer-Rokitansky-Küster-Hauser syndrome, MRKH, and MURCS association) and in renal malformations [1, 61]. Li and colleagues described seven patients with vertebral and rib malformations and structural renal differences, with a TBX6 deletion in trans with the hypomorphic allele or with biallelic variants [62]. Among carriers of the 16p11.2 deletion, renal and urinary tract anomalies reached about 25% in a Chinese cohort and 16% in a Caucasian/Hispanic cohort, and four of seven individuals with heterozygous TBX6 loss of function had overt renal anomalies [46]. A fetus with severe SCDO and a TBX6 nonsense variant was reported as an additional piece of the dosage model [63], and the GeneReviews authors mention an unpublished lethal form of STD with Müllerian anomalies associated with TBX6 [1]. For DLL3, HES7, LFNG and RIPPLY2, no other germline phenotypes are known [1].

2.8. Penetrance, expressivity and modifiers

The penetrance of biallelic variants in recessive SCDO appears to be complete [1]. In the extended HES7 family, the penetrance of the vertebral malformations was complete, but only three of seven affected individuals had dextrocardia with situs inversus, suggesting randomisation of left-right patterning [64]. Outside Mendelian SCDO, haploinsufficiency of Notch pathway genes may contribute to congenital scoliosis, and in mice short-term gestational hypoxia increased the penetrance and severity of vertebral defects in animals with this predisposition, through transient disruption of FGF signalling [65]. Biallelic variants in HAAO and KYNU, enzymes of the kynurenine pathway that synthesises nicotinamide adenine dinucleotide (NAD), caused NAD deficiency and vertebral, cardiac and renal malformations in four patients; in null mice, prevention of NAD deficiency during gestation prevented the malformations, a result that remains preclinical as regards prevention in humans [6, 66].

2.9. Candidate genes outside the six established types

Several genes have been associated with SCDO-like pictures without yet being included in the GeneReviews and Orphanet classification. DMRT2, a gene of the DMRT cluster at 9p24.3 expressed in the dermomyotome, already has the SCDO7 phenotype in OMIM (621523) [4, 30]. In DMRT2, a newborn with severe rib anomalies (absent, fused, bifid and hypoplastic ribs), laminar fusions and irregular ossification of the vertebral bodies, without clear segmentation defects of the bodies, had the homozygous start-loss variant NM_006557.6:c.1A>T (publication transcript, not MANE; current version NM_006557.7) and died at nine days with respiratory failure [1, 29]; a second patient, an Israeli infant born to consanguineous parents, with severe costovertebral malformations, dysmorphic features and the homozygous deletion NM_181872.6:c.129del p.(Asp43GlufsTer75), died at 3 months of respiratory failure, confirmed the association and added thymic aplasia with immunodeficiency still without a genetic explanation [4, 30]. In DLL1, Barhoumi and colleagues described in 2019 two boys with scoliosis and thoracic fusions homozygous for c.1534G>A p.(Gly512Arg), designated SCDO type 7 in a 2022 review, whereas heterozygous DLL1 variants cause neurodevelopmental disorders [67, 68]; this designation does not coincide with that of OMIM, which assigns the number 7 to DMRT2 [4]. A 37-year-old woman with SCDO and a heterozygous missense FLNA variant was presented as a possible new subtype [69], and a patient with cervicothoracic segmentation defects and a variant in SLC35A3, the gene for the Golgi UDP-N-acetylglucosamine transporter also linked to a congenital disorder of glycosylation, remains a single case [6, 70]. These findings are isolated reports and do not, by themselves, support targeted testing or risk estimates.

3. Clinical Spectrum and Manifestations by System

3.1. Common picture of spondylocostal dysostosis

By the 2023 GeneReviews update, close to 100 individuals with SCDO and biallelic variants in one of the six genes had been identified or published, and the clinical description rests on these reports [1]. The core phenotype is a body disproportion with a short trunk and, often, a short neck and limbs of normal length, present from birth and sometimes detected on prenatal ultrasound [1, 6]. Growth data are scarce: on the basis of lower limb length measurements, patients are about 10% below predicted adult height, and some have severe short stature, up to four standard deviations below the mean; the LFNG form is associated with shorter stature [1]. The neck is probably about 10% shorter, and limitation of cervical mobility has never been formally assessed [1].

Scoliosis occurs in most patients, is radiologically visible from early childhood and is generally non-progressive or slightly progressive and self-limiting [1]. In SCDO1, the vertebral changes tend to balance out between sides, which explains curvatures smaller than the vertebral disorganisation would predict [6]. Severe scoliosis requiring surgery is relatively rare, and the LFNG form has more marked and progressive curvature, mainly thoracolumbar [1]. Neurological complications due to vertebral deformity are rare in SCDO in general [6].

3.2. Respiratory system

The most important problem in the newborn is respiratory function, which may be compromised by the small thorax; respiratory failure may be the presenting feature [1]. Life-threatening failure requiring neonatal intensive care appears to be rare in SCDO, but it does occur; one patient died in the first year of respiratory failure and had, at autopsy, a membranous left hemidiaphragm, a finding not described in other patients [1]. In children who needed early respiratory support, lung growth may improve enough to allow relatively normal growth and development by about 2 years of age [1]. The most important secondary complication is chronic respiratory failure due to reduced lung capacity, with a risk of pulmonary hypertension and heart failure in patients with severe restriction from birth; there is no systematic review of susceptibility to pneumonia or of the incidence of pulmonary hypertension [1]. In the absence of pulmonary restriction, life expectancy is normal [1].

3.3. Other systems

Boys with SCDO appear to have an increased risk of inguinal hernia, sometimes already in the neonatal period [1]. Cardiac and genitourinary malformations are rare (ventricular septal defect, renal aplasia or hypoplasia) [6]. Some patients have hip dysplasia [6]. In a single-centre cohort of 114 patients with a clinical and radiological diagnosis of Jarcho-Levin syndrome, followed for a mean of 6.2 years (2013–2024), 94 (82.5%) had lower limb deformities, bilateral in 73: clubfoot in 64, hip dislocation in 79 and hip flexion contracture in 75; after treatment, clubfoot recurred in 29 of 61 patients and hip dislocation in 12 of 35 [71]. This study uses the eponym in a broad sense and the abstract does not present genotypes, so the frequencies cannot be assigned to a specific type [71]. Isolated reports describe the association with Sprengel deformity [72], split cord malformation [73] and, in a fetus with SCDO1, gastric pancreatic heterotopia of uncertain significance [74]. Cognitive impairment is not part of the picture: intellectual development is normal in the available descriptions, including in STD [1, 15].

3.4. Features by type

SCDO1 (DLL3). This is the most frequent and the most homogeneous form. The whole spine is affected, with rounded vertebral bodies of smooth outline and variable size in the young child (pebble beach sign), which after mid or late childhood give way to irregular bodies and hemivertebrae that are difficult to distinguish individually; the ribs occasionally fuse distal to the costovertebral joint, and the rib cage remains globally symmetrical [1, 22]. Scoliosis is usually mild, but patients with severe curvature have been described [1].

SCDO2 (MESP2). All vertebrae show some abnormality, with a more angular appearance than in SCDO1, and the lumbar vertebrae are relatively spared compared with the thoracic ones; the ribs tend to be straight and more aligned than in other forms, with fewer points of fusion [1]. In the series of the Spranger atlas, SCD2 showed a greater tendency to asymmetry and scoliosis, and severe cases with a crab-like thorax had respiratory impact and a worse prognosis [6].

SCDO3 (LFNG). Spinal shortening and scoliosis are more severe, because all vertebral bodies have more marked segmentation defects; the lumbar vertebrae acquire a block-like appearance in childhood [1]. Reports include syringomyelia at 7 years with urinary incontinence at 10 years, distal arthrogryposis (primary or secondary to cervical cord compromise) and, in one patient, a single pelvic kidney, uterine dysgenesis, absence epilepsy and deafness [1, 75]. The original patient, of Lebanese origin and homozygous for NM_001040167.2:c.564C>A p.(Phe188Leu) (F188L), had an arm span of 186.5 cm for an adult height of 155 cm, abnormal vertebrae from the cervical to the lumbar spine, long thin fingers and camptodactyly, the latter attributed to nerve compromise secondary to the cervical malformation [24, 76]. A 9-month-old Japanese boy, with height at about −2.5 standard deviations, vertebral anomalies from cervical to sacral with a pebble beach appearance, cervical kyphosis, a hypoplastic odontoid process and absent and fused ribs, had compound heterozygous variants, one of them in the DxD motif of the active site, with loss of enzyme activity in vitro [76, 77].

SCDO4 (HES7). The radiological pattern varies between that of STD, with prominent pedicles, and that of SCDO1; all vertebrae are affected [1, 25]. It is the type associated with neural tube defects (lumbosacral myelomeningocele, spina bifida occulta, Chiari II malformation) and with dextrocardia with situs inversus: in an extended Middle Eastern family with the same homozygous variant, three of seven affected individuals had dextrocardia with situs inversus and two had neural tube malformations [1, 64]. The original patient, from a consanguineous Mediterranean family, had a prenatal diagnosis of hydrocephalus and myelomeningocele and, at birth, a short symmetrical bell-shaped thorax, an ectopic stenotic anus, clubfoot, neurogenic bladder and Chiari II malformation; the heterozygous parents and a sibling had normal stature and radiographs [25, 78]. In a non-consanguineous Italian family, two compound heterozygous siblings had 11 pairs of ribs, cervical anomalies with fusion of C1 to C2 and upward displacement of the odontoid process, and vertebral artery anomalies, which the authors considered possibly specific to the HES7 form [78, 79].

SCDO5 (TBX6). The recessive forms are radiologically almost indistinguishable from SCDO1 [1]. In the dominant family, the vertebral malformations were generalised, with a mixture of hemivertebrae and blocks, relatively spared ribs and mild mid-thoracic scoliosis, without other anomalies and with normal neurodevelopment [1, 58]. In TACS, the malformations are regional, present as congenital scoliosis and include hemivertebrae in most cases, although two Japanese patients did not have them [1, 8]. In the Chinese (345 patients), Japanese (142) and American (10) cohorts of Liu and colleagues, the 33 Chinese patients with TACS had earlier onset (median 2 versus 3 years), one or more hemivertebrae or butterfly vertebrae, malformations in the lower spine (T8 to S5), simple rib anomalies in 25 and intraspinal anomalies in only 1; the authors constructed a clinical index (TACScore) with an area under the curve of 0.9 [57]. Older dominant families, such as that described by Rimoin in 1968 (father and son with stature below 5 feet, 11 pairs of ribs, posterior rib fusions and cervical and thoracic hemivertebrae, without neurological manifestations), describe similar pictures, without molecular confirmation [1, 57].

SCDO6 (RIPPLY2). The pattern is cervical. Two siblings had defects of the posterior elements from C1 to C4, hemivertebrae and butterfly vertebrae from T2 to T7, marked cervical kyphosis at C2–C3 with spinal cord compression and mild thoracic scoliosis [1, 27]. Three patients from two families homozygous for c.240-4T>G had agenesis of the posterior and lateral elements of most cervical vertebrae, with an atlantoaxial malformation compromising the spinal cord [55]. Wegler and colleagues described three more patients with cervical and thoracic malformations, small or absent rib involvement, myelopathy with or without canal stenosis and spasticity; the boy homozygous for p.(Arg80Ter) required surgical cervical stabilisation [54, 80]. Cervical instability has acute consequences: a 7-year-old Salvadoran boy, homozygous for c.240-4T>G, suffered cardiac arrest after going down a slide and had atlanto-occipital assimilation, a hypoplastic dens, absence of the posterior elements of C2 and C3 and stenosis at C4; one of the French sisters developed spasticity at 6 months and progressive flaccid tetraplegia due to compression at C2; and two Dutch siblings homozygous for p.(Arg80Ter) had intermittent myelopathic symptoms with mechanical strain and incomplete spinal cord injury after minor trauma [80]. OMIM summarises SCDO6 as a cervicothoracic defect with agenesis of the posterior elements, cervical fusions and hypoplasia of C1 to C3, with myelopathy in most patients [80]. The GeneReviews authors accept that this form may be better classified as a variant of the Klippel-Feil anomaly [1].

SCDO7 (DMRT2). The two published patients were male newborns with severe rib anomalies (absent, fused, bifid and hypoplastic ribs), vertebral laminar fusions, dysmorphic features and respiratory failure from birth; they died at 9 days and at 3 months, and the second also had thymic aplasia and immunodeficiency [4, 29, 30]. The vertebral bodies showed no clear segmentation defects in the first patient, which distinguishes this type from the other SCDO [29]. The first patient, of North African ancestry, had cleft palate, a supernumerary nipple, an aberrant right subclavian artery, a small non-functioning left kidney and a tethered cord [81]. The second, born after severe polyhydramnios, had bilateral absence of the upper ribs, butterfly vertebrae from T8 to T10, cleft palate, ventricular septal defect, patent ductus arteriosus, duplex kidneys and bilateral inguinal hernias; newborn screening showed absent T-cell receptor excision circles (TREC), the thymus was not visible, there was profound lymphopenia with almost absent CD3+ cells, and the patient died at 3 months of respiratory failure after cytomegalovirus pneumonitis and bacterial infections [30, 81]. OMIM therefore includes a hypoplastic or absent thymus with immunodeficiency in the description of SCDO7 [81].

3.5. Spondylothoracic dysostosis

STD is the form with the highest respiratory risk. Shortening affects all segments, especially the thoracic spine, the ribs fuse posteriorly at their costovertebral origins and fan out, straight and without points of fusion along their course, and the pedicles are prominent in contrast with vertebral bodies of irregular shape [1, 9]. The vertebral bodies have a reduced anteroposterior diameter and increased width, with a sickle shape, and the number of vertebral bodies is reduced [9]. Smith's textbook describes short-trunk dwarfism of prenatal onset with a mean height of 115.7 cm, a prominent occiput in infancy that becomes flat with brachycephaly, a broad forehead, a broad nasal bridge, anteverted nares, a high palate, a short rigid neck, a low posterior hairline, pectus carinatum, an increased anteroposterior chest diameter, a protuberant abdomen and inguinal and umbilical hernias; it records neural tube defects (meningocele and myelomeningocele) in 25% to 40% and hydrocephalus [15]. Occasional anomalies include cleft soft palate, cryptorchidism, hydronephrosis, duplication of the collecting system, anal and urethral atresia, uterus didelphys, agenesis of the corpus callosum, Chiari type 1 malformation, atrial septal defect and clubfoot [15].

The natural history of STD has been described mainly in Puerto Rico, where 91% of 27 patients in the 2004 series came from small towns in the north-west of the island [35]. In the prospective series of Cornier and colleagues (27 patients), 8 of 18 patients followed prospectively died in the first 6 months (44%), of respiratory failure secondary to pneumonia and pulmonary restriction, and the remainder were aged between 4 months and 47 years; the authors concluded, with 56% survival at 6 months, that STD is not a lethal syndrome, and noted that after 6 months all patients survived with minimal medical complications, normal intelligence and a good independent quality of life [9, 35]. Smith summarises that about 60% of newborns have significant thoracic restriction with some degree of respiratory distress [15]. In a multicentre study of 28 patients with STD (8 neonatal deaths), the 20 survivors assessed were clinically well, but had a severe restrictive pattern, with a mean forced vital capacity (FVC) of 27.9% of predicted, forced expiratory volume in one second (FEV1) of 29.5% and an FEV1/FVC ratio of 0.92; lung volumes on computed tomography (CT) were 28% of expected and the rigid thorax had 24.2% of the predicted posterior length; scoliosis was minimal and there were no neurological deficits [10]. The authors interpreted this picture as a model of thoracic insufficiency syndrome (TIS) due to volume depletion, in which those who survive infancy show no progression of the anomalies and may have a good quality of life [10].

In 2024, the same team assessed 11 adults with STD confirmed radiographically and genetically, never treated surgically (mean age 32.3 years, 15–70): mean thoracic height was 16 cm, FVC 22% of predicted, FEV1 24% and the FEV1/FVC ratio 107% of predicted; the Patient-Reported Outcomes Measurement Information System (PROMIS) dyspnoea index was 40±8 points and the Early Onset Scoliosis Questionnaire (EOSQ-24) 77.3±18 points, better than described in neuromuscular populations [82]. Pulmonary function tests and thoracic height did not correlate with patient-reported outcomes [82]. Berdon and colleagues summarised that, in the Puerto Rican cohorts, up to 25% of patients survive to late childhood and adulthood [5], and the Puerto Rico group stated in 2025 that most carriers of p.(Glu103Ter) reach adulthood without surgery [17]. The two statements are not directly comparable, because they come from different samples and periods and neither is a population study. In the severe neonatal form, pulmonary hypoplasia continues to cause deaths in the first year, as in a Puerto Rican newborn with p.(Glu103Ter) who died at 4 months despite high-frequency oscillatory ventilation [19, 83] and in an infant of first-degree consanguineous parents who died at 7 months of respiratory failure with pneumonia [84]. In contrast, a newborn with a crab-like thorax, hemivertebrae, an absent rib, spina bifida and intestinal herniation through an abdominal wall defect was managed conservatively and was alive at the end of the first year [85].

3.6. Treated and untreated disease

The available data mix treated and untreated patients in a way that needs to be separated. The Puerto Rican STD series of 2007 and 2024 describe patients without thoracic expansion surgery and are the main source of natural history data on congenital TIS [10, 82]. The VEPTR series describe patients selected for surgery, with two years of follow-up, and give no information on the course without treatment [12]. In non-STD SCDO, the course described in GeneReviews is that of patients largely managed conservatively, because the manifestations do not increase with age [1].

4. Clinical Recognition and Warning Signs

4.1. Prenatal period

Detailed fetal ultrasound, in experienced hands, detects multiple vertebral segmentation defects from 13 weeks of gestation, especially when the malformation is expected and specifically sought, as in a pregnancy of a couple with an affected child [1]. Outside that context, the ultrasound finding of vertebral defects has very diverse causes: in a Chinese series of 12 fetuses with vertebral defects detected in the second or third trimester (2020–2024), four had other structural anomalies, and five (41.7%) received a genetic diagnosis, three of them a de novo 16p11.2 deletion [53]. Suspicion of SCDO or STD in a fetus should prompt a targeted examination of the thorax, the whole spine, the neural tube and the heart, given the association of the HES7 form with neural tube defects and dextrocardia [1, 64].

4.2. Newborn and infant

In the newborn, suspicion arises in the presence of a short trunk and neck with normal limbs, a small or deformed thorax and scoliosis, and is confirmed by radiographs of the spine and chest [1, 6]. Tachypnoea, feeding difficulties and oxygen requirement suggest respiratory failure and warrant respiratory medicine assessment [1]. The warning signs are progressive respiratory distress, recurrent respiratory infections and signs of pulmonary hypertension, because in STD deaths in the first year result from respiratory failure with pneumonia and pulmonary restriction [1, 9]. In boys, inguinal hernia is sought from the neonatal period [1]. A lumbosacral swelling, lower limb motor deficits or sphincter disturbances point to a neural tube defect or associated spinal cord malformation [1, 15]. There is no newborn screening for these disorders; however, in the only patient with SCDO7 studied in this respect, TREC-based newborn screening for severe combined immunodeficiency was abnormal, which makes immunological assessment relevant in the presence of a severe costovertebral picture with cleft palate [30, 81].

4.3. Child, adolescent and adult

In the older child and the adult, the presentation is disproportionate short stature with a short trunk, a short neck and scoliosis, sometimes in a context of consanguinity or affected siblings [1, 86]. A regional radiological pattern (low thoracic or high lumbar hemivertebrae) in a patient with apparently isolated congenital scoliosis points to TACS and to testing for the 16p11.2 deletion and the TBX6 haplotype [1, 28]. Neck stiffness, cervical kyphosis, signs of myelopathy or spasticity in a patient with multiple cervical malformations suggest SCDO6 and require urgent cervical imaging [1, 54]. A recent change in balance or continence in a patient with SCDO3 warrants magnetic resonance imaging (MRI) of the spinal cord, as in the patient with syringomyelia at 7 years [1].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Radiological criteria

SCDO is a radiological diagnosis confirmed by genetics [1]. GeneReviews defines the suggestive findings: multiple segmentation defects most evident on the anteroposterior radiograph of the whole spine, with abnormal segmentation of at least ten contiguous vertebrae; mild scoliosis; rib anomalies with malalignment of at least some ribs, a variable number of intercostal fusions and sometimes a reduced number of ribs; absence of major asymmetry of the thorax; and a family history consistent with recessive inheritance (affected siblings or consanguinity), the absence of which does not exclude the diagnosis [1]. Current consensus reserves the term SCDO for abnormal segmentation of at least ten contiguous vertebrae, which separates these disorders from defects adjacent to a neural tube defect and from regional congenital scoliosis [1, 15].

The following table summarises the radiological signs that guide testing.

Radiography defines the phenotype but does not identify the gene, because the six types overlap [42]. CT with three-dimensional reconstruction helps to demonstrate the costovertebral fusion of STD and the cervical malformations of SCDO6 [1, 9].

5.2. Assessment after suspicion

The initial assessment recommended by GeneReviews includes radiographs of the whole spine (anteroposterior and lateral) and chest, respiratory medicine assessment, feeding assessment, examination for inguinal hernia in boys, renal and urinary tract ultrasound, MRI of the whole spinal cord in patients with SCDO3, a genetics clinic appointment and assessment of family support needs [1]. The European consensus on non-idiopathic early-onset scoliosis (EOS) recommends, with strong agreement (95%), MRI of the whole spine whenever the curvature exceeds 20° at diagnosis, and dynamic cervical radiographs (flexion-extension) before growth-friendly treatment in children with skeletal dysplasias [87]. Echocardiography is indicated when pulmonary hypertension is suspected and in SCDO4, because of the association with dextrocardia [1, 44].

5.3. Genetic strategy

The molecular diagnosis is established by biallelic pathogenic or likely pathogenic variants in one of the six genes, in a patient with suggestive radiological findings; biallelic variants of uncertain significance (VUS), or one pathogenic variant and one VUS, neither establish nor exclude the diagnosis [1]. Two approaches are possible [1]. The first is targeted testing: a multigene panel including DLL3, HES7, LFNG, MESP2, RIPPLY2, TBX6 and differential diagnosis genes, to which it is justified to add DMRT2 since the definition of SCDO7 [4], or sequential phenotype-driven testing, starting with LFNG in the presence of severe truncal shortening and with MESP2 in the presence of the STD pattern. The second is exome or genome sequencing, when the phenotype overlaps with other skeletal dysplasias.

The French study by Lefebvre and colleagues informs the order of testing. In 73 patients with SDV, the five-gene panel gave a diagnosis in 10 (13.7%), but in 8 of 10 (80%) of the subgroup meeting the SCDO criteria; the diagnoses were four biallelic TBX6, two LFNG, two DLL3, one MESP2 and one HES7, and exome sequencing in 28 patients added two (8%), one biallelic FLNB and one MEOX1 [11]. The authors recommend, after a negative microarray, reserving the SCDO panel for those who meet the criteria [11]. In a cohort of 34 patients with congenital vertebral malformations studied at three levels (microarray, panel, exome/genome), the overall yield was 12%, with most diagnoses at the last level [88]. The TACScore index, based on early onset, hemivertebrae or butterfly vertebrae of the lower spine and simple rib anomalies, helps to select patients with congenital scoliosis for TBX6 testing [57]. In TACS, the diagnosis requires a search for the null allele, including the 16p11.2 deletion by microarray or copy number variant (CNV) analysis, and genotyping of the haplotype on the other allele; no imaging finding establishes this genotype [28, 46].

5.4. Sensitivity and limitations of tests

The following table reproduces the GeneReviews estimates. The column of proportions of detectable variants refers to the known variants in each gene and is not equivalent to the clinical sensitivity of the test for the disease.

Source: [1]. The proportions are the authors' estimates based on published cases and personal communications [1].

There are specific technical limitations. DLL3 exon 8 and HES7 exon 4 are rich in guanine and cytosine and may require additives in the polymerase chain reaction (PCR) [1]. MESP2 exon 1 contains a tract of 12-nucleotide repeats (types A, B and C) beginning at nucleotide 535, with the reference sequence AABBC, and may require bidirectional Sanger sequencing [1]; ClinVar records several truncating variants in this region [51]. Exome sequencing may not cover these regions well, and detection of the 16p11.2 deletion and of exonic DLL3 deletions requires copy number methods [1]. An apparently homozygous patient may have a deletion on the other allele not detected by sequencing or uniparental isodisomy, which is clarified by parental testing [1]. A negative result does not exclude a genetic cause, because a fraction of patients with SCDO do not have variants in the known genes and because the class continues to grow (DMRT2, DLL1 and other candidates) [11, 30, 67].

5.5. Genetic report and interpretation

The report should identify the gene, the transcript with version (for example NM_001039958.2 for MESP2 and NM_004608.4 for TBX6), the variant in HGVS notation, zygosity, phase (cis or trans) when there are two variants and the ACMG/AMP classification [1, 50]. In patients with TACS, the report should explicitly mention the haplotype on the other allele, without which the isolated null allele does not explain the phenotype [28, 46]. Intronic splicing variants, such as RIPPLY2 c.240-4T>G and HES7 c.43-9T>A, may remain classified as VUS despite concordant phenotypes, and family segregation and RNA studies contribute to reclassification [54, 86].

5.6. Prenatal and preimplantation diagnosis

Once the family variants have been identified, prenatal diagnosis by molecular testing and preimplantation genetic testing are possible; molecular testing is the reference, and ultrasound from 13 weeks is a complement [1]. In the presence of fetal vertebral defects without a family diagnosis, chromosomal microarray is the first test and exome sequencing adds diagnoses in negative cases [53]. In a Vietnamese family with SCDO4, an affected pregnancy was identified by fetal ultrasound and prenatal genotyping and terminated at 13 weeks [86].

6. Differential Diagnosis

6.1. Differential diagnosis within the class

The first step is to distinguish SCDO from STD, because the respiratory prognosis differs. In STD, shortening is more severe, especially thoracic, the ribs fuse posteriorly in a bilateral and symmetrical manner and fan out without fusions along their course, and the pedicles give the tramline sign; in SCDO, the ribs are intrinsically deformed, fused at irregular points or absent, and the thorax does not have a crab-like configuration [1, 15, 17]. The second step is to distinguish generalised defects (SCDO, STD) from regional defects (TACS, sporadic defects, cervical SCDO6) [1, 6].

6.2. Sporadic segmentation defects and congenital scoliosis

Sporadic vertebral segmentation defects are more frequent than SCDO, often unilateral and restricted to a few vertebrae or one region, and are more frequently associated with cardiac, renal and urogenital malformations; they have no consistent genetic basis, although a fraction are explained by TBX6 and some cases by kynurenine pathway deficiency [6, 66]. The term congenital scoliosis is an orthopaedic description of trunk asymmetry caused by segmentation defects, rarely applicable to SCDO 1 to 4, and does not include neurogenic, neuromuscular or connective tissue scolioses [6].

6.3. Syndromes with multiple segmentation defects

The following table brings together the main differential diagnoses, based on the GeneReviews tables and the reference atlases.

6.4. Other differential diagnoses and associations

Chromosomal anomalies are rarely associated with SCDO; with the exception of trisomy 8 mosaicism, there is no consistent genomic region [1]. Families with apparently dominant SCDO without an identified gene have been published since 1968, with a variable extent of defects [1]; in some, such as the family described by Floor and colleagues, pseudodominance of SCDO1 was demonstrated in a highly consanguineous family [1, 91, 92]. Other syndromes of unknown cause with multiple vertebral defects include Casamassima-Morton-Nance syndrome (segmentation defects and urogenital anomalies), OEIS complex, Phaver syndrome, Anhalt-type spinal dysplasia, femoral hypoplasia-unusual facies syndrome, Wildervanck syndrome and the association of limb deficiency, vertebral hypersegmentation and absent thymus [1]. Orphanet has made obsolete the entries “SCDO-hypospadias-intellectual disability” (ORPHA:329252) and “SCDO-anal atresia-genitourinary malformation” (ORPHA:94095) [2].

Neural tube defects are frequently associated with severe anomalies of one or more adjacent vertebrae, but these patients do not meet the criterion of ten contiguous vertebrae and should not be classified as SCDO [1]. Jeune syndrome also causes thoracic insufficiency syndrome and was analysed alongside Jarcho-Levin syndrome in the VEPTR series, with different postoperative mortality between the two groups [93].

7. Treatment and Risk Management Strategies

7.1. Principles and sources of recommendations

There is no published clinical practice guideline for SCDO or STD; GeneReviews states this expressly, and the list of ERN BOND documents, consulted in October 2026, includes none for these disorders [1, 13]. The available recommendations are those of GeneReviews, at the level of expert opinion, those of the reference atlases and, for the vertebral deformity, the European Delphi consensus on EOS, whose scope is broader than SCDO and whose application to these disorders is an extrapolation [1, 6, 87]. There is no targeted drug and no specific registered clinical trial [1, 94]. In most patients with SCDO, management is conservative, because the vertebral and rib manifestations do not increase with age [1].

7.2. Growth and nutrition

All patients have short-trunk short stature of variable degree, and there are no specific nutritional needs beyond maintaining weight appropriate for height [1]. Growth is assessed at every visit during childhood [1].

7.3. Treatment of vertebral deformity

In the atlas by Spranger and colleagues, physiotherapy is the preferred treatment for mild asymmetry, and congenital scoliosis with marked asymmetry is almost always operated on, without solid outcome data; outside severe SCD2 with a crab-like thorax, the course is less severe than the imaging would suggest [6]. GeneReviews recommends surgery when scoliosis is significant (which is unusual) and accepts the use of VEPTR, growing rods and other devices [1].

The Delphi consensus of the EPOS Spine Study Group, published in 2026 with the participation of 22 European paediatric orthopaedic surgeons and an agreement threshold of 80%, produced 15 recommendations for non-idiopathic EOS [87]. Those that apply by extrapolation to SCDO are shown in the following table; several statements on congenital scoliosis did not reach consensus.

Source: [87] (level of evidence IV).

The 2024 review by Studer and Hasler summarises the paradigm shift: early spinal fusion has given way to the concept of growth-friendly surgery, with the efficacy of new devices still to be validated, a return of casting and orthoses and a high psychosocial burden of long treatments with many complications [95].

7.4. Thoracic insufficiency syndrome and expansion surgery

VEPTR was developed to reconstruct the chest wall in TIS. In a multicentre investigational device study with 214 patients, the 20 patients with SCDO followed for at least two years had control of the deformity (reduction of the coronal Cobb angle) in 14 (70%), 14 (70%) maintained oxygenation, 5 improved on the assisted ventilation rating scale and 14 remained on room air, while one moved from supplemental oxygen to nocturnal ventilation; mean thoracic spine growth was 0.82 cm per year and there were no deaths [12]. In the same investigational cohort, among 43 patients with Jeune or Jarcho-Levin syndrome, 4 (9%) died, all with Jeune syndrome, and life-threatening events occurred in a further 4 (9%), two with Jarcho-Levin syndrome, mostly in the first year after surgery [93]. A 2020 review notes that initial enthusiasm waned with the number of complications, including extravertebral ossification along the implants and between ribs, that motorised distraction rods have reduced the use of VEPTR outside deformities with thoracic volume depletion, that the final correction obtained is limited and that complication and reoperation rates are very high [96, 97].

In STD, the question of the indication is now explicit. STD was considered a core indication for VEPTR expansion thoracoplasty by V-osteotomy of the posterior thorax; the study of 11 adults never operated on, with severe restriction but patient-reported outcomes better than expected and no correlation between function tests and quality of life, led the authors to propose that STD should no longer be an absolute indication [82]. This proposal rests on a level IV series of survivors, which by definition excludes patients who died in childhood [9, 82]. Other techniques have been described in isolated cases, such as excision of the posterior elements, posterior column osteotomies and gradual distraction with MCGR in an 8-year-old child with a congenitally fused thoracic spine due to Jarcho-Levin syndrome, with short follow-up [98].

In TACS, a preliminary retrospective study compared 28 patients with 28 controls with congenital scoliosis without TBX6 involvement, matched by propensity score among 473 operated patients: correction of the cranial (64.9±18.6% versus 51.2±24.0%; P=0.014) and caudal (77.4±12.5% versus 65.4±22.7%; P=0.011) compensatory curves was greater and loss of cranial correction smaller (0.6±19.2% versus 26.7±50.8%; P=0.002) in the TACS group, with no differences in blood loss or revisions [99]. The authors suggest that genetic diagnosis before surgery helps to predict the outcome, a conclusion that depends on replication [99].

7.5. Respiratory management

Respiratory support, including intensive care, is adjusted to the degree of pulmonary compromise and is usually needed only in severe disease; chronic respiratory failure requires specialist management and assessment for pulmonary hypertension and heart failure [1]. In STD, vigorous treatment of respiratory infections is part of the follow-up of survivors [15]. In a newborn with STD and severe pulmonary hypoplasia, positive pressure ventilation was followed by high-frequency oscillatory ventilation, without preventing death at 4 months [19]. One report describes the combination of neurally adjusted ventilatory assist (NAVA) with VEPTR in a patient with SCDO and bronchomalacia [100]. In a series of two patients with Jarcho-Levin syndrome, tracheostomy was technically difficult because of the deformities, and the authors propose considering it early, before recurrent pneumonia and pulmonary hypertension, while acknowledging that there are no guidelines on timing [101]. Pulmonary function in EOS in general is maintained or declines over four years of surgical treatment, with significant decline in congenital scoliosis, and about a quarter of children improve by more than 8%, with no aetiology associated with this improvement [33].

7.6. Neurological and cervical manifestations

The neurological problems of SCDO3 are treated according to the findings (for example syringomyelia) [1]. In SCDO6, agenesis of the cervical posterior elements with kyphosis and canal stenosis can cause spinal cord compression and myelopathy, which makes the cervical spine the focus of surveillance and of any neurosurgical treatment [1, 54, 55]. The reports of cardiac arrest after going down a slide, of progressive tetraplegia due to compression at C2 and of incomplete spinal cord injury after minor trauma show that the cervical spine of these patients tolerates mechanical strain poorly, and two patients required surgical stabilisation [54, 80]. By analogy with the Klippel-Feil anomaly, hypermobility of the upper cervical segment warrants assessment at least annually and avoidance of impact activities [90]; this recommendation is extrapolated and was not formulated for SCDO6. The neural tube defects associated with SCDO4 and STD follow their own neurosurgical management [1, 15].

7.7. Other manifestations

Inguinal hernias are managed in the usual way by paediatric surgery, after the parents have been taught to recognise the signs and complications [1]. Lower limb deformities, frequent and recurrent in the single-centre series of 114 patients with Jarcho-Levin syndrome, require targeted assessment and prolonged follow-up after treatment [71]. In a 6-year-old child with Sprengel deformity associated with SCDO, severe downward rotation of the scapula required a modified Wilkinson procedure with Y-osteotomy, with almost complete symmetry at one year and no limitations at 18 months [72].

7.8. Drugs

There is no drug indicated for SCDO or STD, nor any drug in clinical development for these disorders in the sources consulted [1, 94]. Niacin supplementation prevented vertebral malformations in mice with NAD deficiency due to inactivation of Haao or Kynu, a mechanism distinct from Mendelian SCDO; there is no human study to support a recommendation, and the finding remains preclinical [6, 66]. The sources consulted do not identify drugs to avoid that are specific to these disorders.

7.9. Regulatory status of devices

7.10. Available evidence: series and studies

There are no randomised trials. The following table summarises the studies that support therapeutic decisions.

These series are level IV, with small groups, surgical selection and heterogeneous definitions of Jarcho-Levin syndrome, and do not allow comparison between techniques [12, 82, 93]. The absence of trials on ClinicalTrials.gov results from a search by condition in October 2026 and does not prove that no studies exist in other registries [94]. The closest registered studies are observational studies of VEPTR in TIS (NCT02241954, 7 patients, completed in 2020) and of progressive early-onset spinal deformities (NCT02299362, 2602 patients, completed in 2020) [94].

7.11. Developments in the last 12 months

7.12. Pregnancy, anaesthesia and the perioperative period

In women with SCDO, pregnancy may aggravate problems of intra-abdominal pressure due to trunk shortening, without published research, and the spinal deformity raises doubts about neuraxial anaesthesia, although spinal anaesthesia has been used successfully for a caesarean section [1, 108]. Three recent reports show the diversity of solutions planned by multidisciplinary teams: caesarean section under general anaesthesia with modified rapid sequence induction and transversus abdominis plane blocks in a woman with severe kyphoscoliosis and restrictive lung disease [109]; term vaginal delivery with epidural analgesia in a 21-year-old woman with the SCDO form, after extensive baseline functional assessment [110]; and preterm caesarean section under continuous spinal anaesthesia in a woman with costovertebral fusion, by joint decision of cardiology, otorhinolaryngology and obstetrics, given the difficult airway [111]. These reports do not constitute a rule, but indicate that the pre-anaesthetic assessment should include respiratory function, the airway and spinal imaging. The sources consulted do not address lactation.

7.13. Transition to adulthood

Patients with SCDO without pulmonary restriction have a normal life expectancy, and non-operated adults with STD may maintain reasonable function despite severe restriction, which justifies continuing respiratory and orthopaedic follow-up in adulthood and including reproductive counselling [1, 82]. The sources consulted do not describe a formal transition model for these disorders.

8. Reproductive Options and Family Counselling

8.1. Autosomal recessive forms

SCDO caused by biallelic variants in DLL3, HES7, LFNG, MESP2, RIPPLY2 or TBX6, and STD, are inherited in an autosomal recessive manner [1, 3]. The parents of a patient are presumed to be heterozygous, and molecular testing of the parents is recommended to confirm carrier status and to allow a reliable calculation of recurrence risk [1]. If only one parent is a carrier and biological parentage is confirmed, one of the variants may have arisen de novo in the patient or as a postzygotic event in a mosaic parent [1]. If the patient appears to be homozygous, a deletion on the other allele not detected by sequencing or uniparental isodisomy should be considered [1]. Carriers are asymptomatic and are not at risk of developing the disease [1].

When both parents are carriers, each sibling of a patient has, at conception, a 25% chance of being affected, a 50% chance of being an asymptomatic carrier and a 25% chance of being neither affected nor a carrier [1]. The children of a patient are obligate carriers, unless the reproductive partner is a carrier or affected in the same gene; testing of the partner is appropriate, especially if there is consanguinity, because about 75% of patients with recessive SCDO come from consanguineous families [1]. Each sibling of the patient's parents has a 50% risk of being a carrier [1]. Carrier testing in relatives requires prior identification of the family variants, and the ideal time for counselling and for discussion of prenatal diagnosis is before pregnancy, including for young adults who are affected, carriers or at risk of being carriers [1].

8.2. Pseudodominance and apparently dominant families

Some families with SCDO in several generations appear to follow dominant inheritance; in one of them, pseudodominance of SCDO1 was demonstrated, that is, a recessive disease present in two or more generations of a highly consanguineous family through marriages between carriers [1, 91, 92]. In a family with vertical transmission, this is the first hypothesis to exclude in a consanguineous context, before a dominant form is accepted [1].

8.3. TBX6: dominant, recessive and compound

Dominant SCDO5 has been demonstrated in a single three-generation family, with all affected individuals male; in this scenario, each child of an affected person has a 50% chance of inheriting the variant [1, 26, 58]. In TACS, the risk depends on the alleles of each relative and follows neither the 50% nor the 25% model [46]. An unaffected parent or sibling with only the null allele, including the 16p11.2 deletion, has a low risk of vertebral malformations, because the phenotype requires the hypomorphic haplotype from the other parent; phenotypic discordance between deletion carriers in the same family illustrates this mechanism [28, 46]. Because the T-C-A haplotype is common (about 44% in Asians and 33% in Europeans), the risk for a child of a null-allele carrier also depends on the other parent's genotype at this locus, and a family with the null allele and the frequent haplotype may mimic dominant inheritance with incomplete penetrance or pseudodominance [28, 57]. The 16p11.2 deletion has its own implications for neurodevelopment: it arises de novo in about 70% of cases with information and has an estimated penetrance of 46.8% for a clinical phenotype [1, 59].

8.4. Prenatal and preimplantation diagnosis

Once the family variants have been identified, molecular prenatal diagnosis and preimplantation genetic testing are possible [1]. Detailed ultrasound from 13 weeks complements molecular testing, which remains the reference [1]. In the presence of fetal vertebral defects of unknown cause, microarray identifies 16p11.2 deletions, and exome sequencing adds point variants in TBX6 and other genes [53]. Differences in perspective between professionals and families on the use of prenatal diagnosis should be discussed; in most centres, the decision is considered a personal one [1].

8.5. Counselling of minors and relatives

GeneReviews places the offer of counselling and carrier testing at young adults who are affected, carriers or at risk of being carriers; the sources consulted do not specifically address carrier testing in minors [1]. The diagnosis of an affected sibling not yet recognised, by contrast, has clinical consequences (respiratory assessment, hernia, cervical spine in SCDO6) and justifies clinical and radiological assessment of siblings with a suggestive phenotype [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1. What the available figures measure

There is no validated population estimate of the prevalence of any form of SCDO: Orphanet records the prevalence class as unknown for ORPHA:2311 and ORPHA:1797, without a mean value and without validation [2, 7]. The figure of 0.5–1 per 1000 births refers to vertebral segmentation defects and congenital scoliosis of all causes, mostly non-genetic, and should not be used as the prevalence of SCDO [1, 8, 46]. The proportions per gene (DLL3 ~60%, TBX6 ~20%, the others ~5% each) are relative estimates by the GeneReviews authors within recessive SCDO, based on published cases and personal communications [1]. The DLL3 carrier frequency of ~1:350 in the United Kingdom is a calculation by the same authors, assuming a period with about one million births, and not a population measurement [1].

9.2. Populations and regions

In the experience of the Exeter laboratory, 75% of patients with SCDO1 were children of consanguineous unions, mainly of Middle Eastern or Pakistani origin, and occasionally European; a small number of patients from northern Europe (England, Wales, the Netherlands and Switzerland) were compound heterozygotes [1]. A Swiss village concentrated a cluster of recessive SCDO caused by three novel DLL3 variants [112]. The mapping of DLL3 started from a consanguineous Arab-Israeli family and a Pakistani family [16]. The first MESP2 family with SCDO2 was Lebanese Arab and consanguineous [35]. The HES7 family with dextrocardia was an extended consanguineous Middle Eastern family, and other HES7 families were Mediterranean and Italian [1, 64, 78]. Among the published patients with SCDO6 there are families from El Salvador, France, Italy, Germany and the Netherlands, and the two patients with SCDO7 were of North African and Israeli ancestry [4, 80]. According to Cornier and colleagues, Puerto Rico contributes 49% of published cases [35]. STD is pan-ethnic but more frequent in Puerto Ricans of Spanish ancestry, as a consequence of the founder effect of MESP2 p.(Glu103Ter), present in all patients of the 12 Puerto Rican families studied in 2008 (10 homozygotes and 3 compound heterozygotes) and responsible for about 90% of cases in Puerto Rico according to Smith's textbook [3, 15]. The Puerto Rico group has assessed more than 81 patients with STD on the island and in other countries [17]. The dominant SCDO5 family was Macedonian [58].

TBX6 compound inheritance explained up to 11% of sporadic congenital scoliosis in a Han Chinese series, 9 of 94 patients in a Japanese series and 9.3% of 108 Chinese patients, with similar frequencies in the two Asian populations, and about 9.2% of Japanese patients with congenital scoliosis in the combined series [8, 28, 56, 57]. An American cohort of 10 patients was included in the study by Liu and colleagues, and the risk haplotype has a frequency of about 33% in European populations [57]. The sources consulted do not present the frequency of TACS in European population series. The French diagnostic cohort of Lefebvre and colleagues showed, in Europe, that TBX6 was the most frequent gene among the panel diagnoses (4 of 10) [11]. The sources consulted do not include data from Portugal or Brazil.

9.3. Resources, networks and associations

ERN BOND is the European reference network for rare bone diseases, and ERN ITHACA covers malformation syndromes [13, 14]. GeneReviews lists the UCLA International Skeletal Dysplasia Registry [1]. The Pediatric Spine Foundation, in the USA, is dedicated to early-onset spinal deformities [113]. ICSGDD brings together researchers and clinicians in spinal genetics and development [33]. The sources consulted do not identify patient associations dedicated specifically to SCDO or STD.

10. Illustrative Clinical Case

Nguyen and colleagues published in 2025 the case of a 21-year-old Vietnamese woman seen at Hanoi Medical University Hospital for short stature, a short neck and mild scoliosis [86]. She measured 1.20 m and weighed 40 kg; radiographs showed a short thorax with fused and butterfly vertebrae and slightly distorted ribs, without neural tube defects or other anomalies [86]. A 9-year-old sister was phenotypically normal (1.32 m), and in another of the mother's pregnancies an affected fetus was identified by ultrasound and prenatal genotyping, with termination at 13 weeks [86]. Exome sequencing (mean depth 150×), confirmed by Sanger sequencing, identified the homozygous variant c.43-9T>A in intron 1 of HES7, inherited from heterozygous parents and predicted in silico to cause loss of the splicing acceptor site [86]. Despite the concordant phenotype and segregation, the authors classified the variant as of uncertain significance under the ACMG/AMP criteria; the article presents in silico predictions of the effect on splicing, without study of the patient's RNA [86]. The article includes three figures: the photograph of the patient (figure 1), the radiographs with the deformed vertebrae marked (figure 2) and the pedigree with the Sanger results and the diagram of the HES7 transcripts (figure 3), available at https://pmc.ncbi.nlm.nih.gov/articles/PMC12248465/; the legends were read in the text of the article, without viewing the images [86].

The case illustrates three points of this article. SCDO4 may present in adulthood with only disproportionate short stature and mild scoliosis, without the neural tube defects or dextrocardia described in the Middle Eastern family [64, 86]. Early prenatal diagnosis was possible in a pregnancy of known risk, in keeping with the 13-week threshold described in GeneReviews [1, 86]. The ACMG/AMP classification of intronic variants may remain uncertain without RNA studies, which limits their use in prenatal and preimplantation diagnosis until reclassification [1, 86]. The report does not describe orthopaedic or respiratory treatment.

11. Prognosis and Long-term Follow-up

11.1. Prognosis

In SCDO without pulmonary restriction, life expectancy is normal, and most patients have mild non-progressive scoliosis; the risk of pulmonary hypertension and its complications has not been quantified [1]. The worst prognosis is concentrated in forms with severe thoracic restriction from birth: in STD, 8 of 18 patients followed prospectively died in the first 6 months (44%), and mortality by the end of the first year approaches 50% according to GeneReviews [1, 9]. Survivors have severe stable restriction (mean FVC of 22% to 28% of predicted in the Puerto Rican series), without progression of the anomalies, and many reach adulthood with an acceptable quality of life [10, 82]. In SCDO3, scoliosis is more progressive and stature shorter [1]. In SCDO6, the prognosis depends on cervical spinal cord compression [1, 54]. Lower limb deformities, when present, frequently recur and are associated with poorer mobility and overall function [71].

11.2. Surveillance table

11.3. Follow-up in adulthood and pregnancy

Adults continue respiratory surveillance, especially in STD and forms with severe restriction, and surveillance of the cervical spine in SCDO6 [54, 82]. Pregnancy requires joint planning by obstetrics, anaesthesiology and respiratory medicine, with assessment of respiratory function, the airway and the spine before delivery [1, 110, 111].

12. Summary and Concluding Remarks

The spondylocostal dysostoses are axial skeletal malformations resulting from failure of the embryonic segmentation clock, with defects of at least ten contiguous vertebrae and rib anomalies. Six genes of the Notch pathway and its regulators (DLL3, MESP2, LFNG, HES7, TBX6 and RIPPLY2) define types 1 to 6, and DMRT2 type 7, known from two newborns with lethal disease; DLL3 is the most frequent, and each type has its own features, such as the pebble beach appearance of SCDO1, the severe shortening of SCDO3, the neural tube defects and dextrocardia of SCDO4 and the cervical involvement of SCDO6. Spondylothoracic dysostosis, caused by MESP2 variants with the Puerto Rican founder p.(Glu103Ter), is distinguished by symmetrical costovertebral fusion with a crab-like thorax and is the form with the highest mortality in the first year, although many survivors reach adulthood with severe and stable pulmonary restriction. TBX6 illustrates a dosage mechanism in which the combination of a null allele, often the 16p11.2 deletion, with a common hypomorphic haplotype explains about one tenth of congenital scoliosis in Asian series, and in which family risk depends on each person's genotype. The Jarcho-Levin eponym designates different things depending on the source, so reading the literature requires looking at the images and the genotype.

Diagnosis is radiological, and molecular confirmation is by microarray and a targeted panel in patients who meet the criteria, or by exome and genome sequencing; the proportion of detectable variants in each gene is not equivalent to clinical sensitivity, and a negative result does not exclude a genetic cause. Treatment is orthopaedic and respiratory, without targeted drugs and without a specific guideline: surveillance of the curvature, growth, respiratory function, the cervical spine and inguinal hernia, and growth-friendly surgery reserved for significant deformities and thoracic insufficiency, with benefits and complications documented only in level IV series. Counselling follows the recessive model in most families, with attention to pseudodominance in consanguineous settings, to the dominant TBX6 form and to the compound inheritance of TACS, and prenatal and preimplantation diagnosis becomes available when the family variants have been identified and classified.

13. References

1. Turnpenny PD, Sloman M, Dunwoodie S. Spondylocostal Dysostosis, Autosomal Recessive. 2009 Aug 25 [updated 2023 Aug 17]. In: Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK8828/
2. Orphanet. Autosomal recessive spondylocostal dysostosis (ORPHA:2311). Paris: INSERM; 2026 [Orphadata data of 2026-06-23]. Available from: https://www.orpha.net/en/disease/detail/2311
3. Cornier AS, Staehling-Hampton K, Delventhal KM, Saga Y, Caubet JF, Sasaki N, et al. Mutations in the MESP2 gene cause spondylothoracic dysostosis/Jarcho-Levin syndrome. Am J Hum Genet. 2008;82(6):1334-41. doi:10.1016/j.ajhg.2008.04.014
4. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 604935. DOUBLESEX- AND MAB3-RELATED TRANSCRIPTION FACTOR 2; DMRT2. Baltimore (MD): Johns Hopkins University; last edited 2026-04-30. Available from: https://omim.org/entry/604935
5. Berdon WE, Lampl BS, Cornier AS, Ramirez N, Turnpenny PD, Vitale MG, et al. Clinical and radiological distinction between spondylothoracic dysostosis (Lavy-Moseley syndrome) and spondylocostal dysostosis (Jarcho-Levin syndrome). Pediatr Radiol. 2011;41(3):384-8. doi:10.1007/s00247-010-1928-8
6. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Spondylocostal dysostoses. In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford: Oxford University Press; 2018. p. 857–862.
7. Orphanet. Autosomal dominant spondylocostal dysostosis (ORPHA:1797). Paris: INSERM; 2026 [Orphadata data of 2026-06-23]. Available from: https://www.orpha.net/en/disease/detail/1797
8. Takeda K, Kou I, Kawakami N, Iida A, Nakajima M, Ogura Y, et al. Compound Heterozygosity for Null Mutations and a Common Hypomorphic Risk Haplotype in TBX6 Causes Congenital Scoliosis. Hum Mutat. 2017;38(3):317-323. doi:10.1002/humu.23168
9. Cornier AS, Ramírez N, Arroyo S, Acevedo J, García L, Carlo S, et al. Phenotype characterization and natural history of spondylothoracic dysplasia syndrome: a series of 27 new cases. Am J Med Genet A. 2004;128A(2):120-6. doi:10.1002/ajmg.a.30011
10. Ramírez N, Cornier AS, Campbell RM Jr, Carlo S, Arroyo S, Romeu J. Natural history of thoracic insufficiency syndrome: a spondylothoracic dysplasia perspective. J Bone Joint Surg Am. 2007;89(12):2663-75. doi:10.2106/JBJS.F.01085
11. Lefebvre M, Dieux-Coeslier A, Baujat G, Schaefer E, Judith SO, Bazin A, et al. Diagnostic strategy in segmentation defect of the vertebrae: a retrospective study of 73 patients. J Med Genet. 2018;55(6):422-429. doi:10.1136/jmedgenet-2017-104939
12. Ramirez N, Flynn JM, Emans JB, Betz R, Smith JT, Price N, et al. Vertical expandable prosthetic titanium rib as treatment of thoracic insufficiency syndrome in spondylocostal dysplasia. J Pediatr Orthop. 2010;30(6):521-6. doi:10.1097/BPO.0b013e3181e78e6c
13. ERN BOND — European Reference Network on Rare Bone Diseases. Available from: https://ernbond.eu/
14. ERN ITHACA — European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders. Available from: https://ern-ithaca.eu/
15. Jones KL, Jones MC, del Campo M. Jarcho-Levin syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 848–851.
16. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 277300. SPONDYLOCOSTAL DYSOSTOSIS 1, AUTOSOMAL RECESSIVE; SCDO1. Baltimore (MD): Johns Hopkins University; last edited 2026-04-21. Available from: https://omim.org/entry/277300
17. De Jesús-Rojas W, Santiago Cornier A. Reply to Shan et al.: Bone-Specific Manifestations of Spondylothoracic Dysostosis in Jarcho-Levin Syndrome. Am J Respir Crit Care Med. 2025;211(3):524. doi:10.1164/rccm.202412-2418LE
18. Cornier AS, Ramirez N, Carlo S, Reiss A. Controversies surrounding Jarcho-Levin syndrome. Curr Opin Pediatr. 2003;15(6):614-20. doi:10.1097/00008480-200312000-00012
19. Shan H, Liu P, Wang X. Bone-Specific Manifestations of Spondylothoracic Dysostosis in Jarcho-Levin Syndrome. Am J Respir Crit Care Med. 2025;211(3):523. doi:10.1164/rccm.202410-1891LE
20. Turnpenny PD, Bulman MP, Frayling TM, Abu-Nasra TK, Garrett C, Hattersley AT, et al. A gene for autosomal recessive spondylocostal dysostosis maps to 19q13.1-q13.3. Am J Hum Genet. 1999;65(1):175-82. doi:10.1086/302464
21. Bulman MP, Kusumi K, Frayling TM, McKeown C, Garrett C, Lander ES, et al. Mutations in the human delta homologue, DLL3, cause axial skeletal defects in spondylocostal dysostosis. Nat Genet. 2000;24(4):438-41. doi:10.1038/74307
22. Turnpenny PD, Whittock N, Duncan J, Dunwoodie S, Kusumi K, Ellard S. Novel mutations in DLL3, a somitogenesis gene encoding a ligand for the Notch signalling pathway, cause a consistent pattern of abnormal vertebral segmentation in spondylocostal dysostosis. J Med Genet. 2003;40(5):333-9. doi:10.1136/jmg.40.5.333
23. Whittock NV, Sparrow DB, Wouters MA, Sillence D, Ellard S, Dunwoodie SL, et al. Mutated MESP2 causes spondylocostal dysostosis in humans. Am J Hum Genet. 2004;74(6):1249-54. doi:10.1086/421053
24. Sparrow DB, Chapman G, Wouters MA, Whittock NV, Ellard S, Fatkin D, et al. Mutation of the LUNATIC FRINGE gene in humans causes spondylocostal dysostosis with a severe vertebral phenotype. Am J Hum Genet. 2006;78(1):28-37. doi:10.1086/498879
25. Sparrow DB, Guillén-Navarro E, Fatkin D, Dunwoodie SL. Mutation of Hairy-and-Enhancer-of-Split-7 in humans causes spondylocostal dysostosis. Hum Mol Genet. 2008;17(23):3761-6. doi:10.1093/hmg/ddn272
26. Sparrow DB, McInerney-Leo A, Gucev ZS, Gardiner B, Marshall M, Leo PJ, et al. Autosomal dominant spondylocostal dysostosis is caused by mutation in TBX6. Hum Mol Genet. 2013;22(8):1625-31. doi:10.1093/hmg/ddt012
27. McInerney-Leo AM, Sparrow DB, Harris JE, Gardiner BB, Marshall MS, O'Reilly VC, et al. Compound heterozygous mutations in RIPPLY2 associated with vertebral segmentation defects. Hum Mol Genet. 2015;24(5):1234-42. doi:10.1093/hmg/ddu534
28. Wu N, Ming X, Xiao J, Wu Z, Chen X, Shinawi M, et al. TBX6 null variants and a common hypomorphic allele in congenital scoliosis. N Engl J Med. 2015;372(4):341-50. doi:10.1056/NEJMoa1406829
29. Bouman A, Waisfisz Q, Admiraal J, van de Loo M, van Rijn RR, Micha D, et al. Homozygous DMRT2 variant associates with severe rib malformations in a newborn. Am J Med Genet A. 2018;176(5):1216-1221. doi:10.1002/ajmg.a.38668
30. Rips J, Mor-Shaked H, Shamriz O, Somech R, Omar RA, Eventov-Friedman S, et al. A Confirmatory Case of Severe Spondylocostal Dysostosis Caused by Biallelic Loss-of-Function of DMRT2. Am J Med Genet A. 2026;200(2):496-501. doi:10.1002/ajmg.a.64270
31. Turnpenny PD, Alman B, Cornier AS, Giampietro PF, Offiah A, Tassy O, et al. Abnormal vertebral segmentation and the notch signaling pathway in man. Dev Dyn. 2007;236(6):1456-74. doi:10.1002/dvdy.21182
32. Offiah A, Alman B, Cornier AS, Giampietro PF, Tassy O, Wade A, et al. Pilot assessment of a radiologic classification system for segmentation defects of the vertebrae. Am J Med Genet A. 2010;152A(6):1357-71. doi:10.1002/ajmg.a.33361
33. Wu N, Hadley-Miller N, Gray R, Schauer S, Redding G, Cheng T, et al. Perspectives from the 2024 International Consortium for Spinal Genetics, Development and Disease (ICSGDD). Genet Med Open. 2025;3:103466. doi:10.1016/j.gimo.2025.103466
34. Unger S, Ferreira CR, Mortier GR, Ali H, Bertola DR, Calder A, et al. Nosology of genetic skeletal disorders: 2023 revision. Am J Med Genet A. 2023;191(5):1164-1209. doi:10.1002/ajmg.a.63132
35. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 608681. SPONDYLOCOSTAL DYSOSTOSIS 2, AUTOSOMAL RECESSIVE; SCDO2. Baltimore (MD): Johns Hopkins University; last edited 2023-07-17. Available from: https://omim.org/entry/608681
36. Dequéant ML, Glynn E, Gaudenz K, Wahl M, Chen J, Mushegian A, et al. A complex oscillating network of signaling genes underlies the mouse segmentation clock. Science. 2006;314(5805):1595-8. doi:10.1126/science.1133141
37. Dequéant ML, Pourquié O. Segmental patterning of the vertebrate embryonic axis. Nat Rev Genet. 2008;9(5):370-82. doi:10.1038/nrg2320
38. GeneGuidance. DLL3 — Spondylocostal dysostosis 1. Version of 2026-09-26.
39. GeneGuidance. MESP2 — Spondylocostal dysostosis 2. Version of 2026-09-26.
40. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 602768. DELTA-LIKE CANONICAL NOTCH LIGAND 3; DLL3. Baltimore (MD): Johns Hopkins University; last edited 2026-03-20. Available from: https://omim.org/entry/602768
41. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 605195. MESODERM POSTERIOR BASIC HELIX-LOOP-HELIX TRANSCRIPTION FACTOR 2; MESP2. Baltimore (MD): Johns Hopkins University; last edited 2020-11-12. Available from: https://omim.org/entry/605195
42. GeneGuidance. LFNG — Spondylocostal dysostosis 3. Version of 2026-09-26.
43. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 602576. LFNG O-FUCOSYLPEPTIDE 3-BETA-N-ACETYLGLUCOSAMINYLTRANSFERASE; LFNG. Baltimore (MD): Johns Hopkins University; last edited 2020-11-12. Available from: https://omim.org/entry/602576
44. GeneGuidance. HES7 — Spondylocostal dysostosis 4. Version of 2026-09-26.
45. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 608059. HES FAMILY bHLH TRANSCRIPTION FACTOR 7; HES7. Baltimore (MD): Johns Hopkins University; last edited 2021-03-03. Available from: https://omim.org/entry/608059
46. GeneGuidance. TBX6 — Spondylocostal dysostosis 5 and congenital scoliosis. Version of 2026-09-26.
47. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 602427. T-BOX TRANSCRIPTION FACTOR 6; TBX6. Baltimore (MD): Johns Hopkins University; last edited 2025-04-10. Available from: https://omim.org/entry/602427
48. GeneGuidance. RIPPLY2 — Spondylocostal dysostosis 6. Version of 2026-09-26.
49. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 609891. RIPPLY TRANSCRIPTIONAL REPRESSOR 2; RIPPLY2. Baltimore (MD): Johns Hopkins University; last edited 2026-10-23. Available from: https://omim.org/entry/609891
50. 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. Tool: https://variantvalidator.org (GRCh38, MANE Select)
51. National Center for Biotechnology Information. ClinVar [Internet]. Bethesda (MD): NCBI; records VCV000005184, VCV000005183, VCV000005186, VCV000005185, VCV000100633, VCV000188053, VCV000427808, VCV000427809. Available from: https://www.ncbi.nlm.nih.gov/clinvar/
52. Chen W, Lin J, Wang L, Li X, Zhao S, Liu J, et al. TBX6 missense variants expand the mutational spectrum in a non-Mendelian inheritance disease. Hum Mutat. 2020;41(1):182-195. doi:10.1002/humu.23907
53. Xie W, Zhou L, Hu A, Chen J, Pang J, Xi H, et al. Ultrasound and genetic findings in a case series of fetuses presenting vertebral defects. BMC Pregnancy Childbirth. 2025;25(1):858. doi:10.1186/s12884-025-07920-6
54. Wegler M, Roth C, Schumann E, Kogan J, Totten E, Guillen Sacoto MJ, et al. Congenital cervical spine malformation due to bi-allelic RIPPLY2 variants in spondylocostal dysostosis type 6. Clin Genet. 2021;99(4):565-571. doi:10.1111/cge.13916
55. Serey-Gaut M, Scala M, Reversade B, Ruaud L, Cabrol C, Musacchia F, et al. Congenital posterior cervical spine malformation due to biallelic c.240-4T>G RIPPLY2 variant: A discrete entity. Am J Med Genet A. 2020;182(6):1466-1472. doi:10.1002/ajmg.a.61549
56. Yang N, Wu N, Zhang L, Zhao Y, Liu J, Liang X, et al. TBX6 compound inheritance leads to congenital vertebral malformations in humans and mice. Hum Mol Genet. 2019;28(4):539-547. doi:10.1093/hmg/ddy358
57. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 122600. SPONDYLOCOSTAL DYSOSTOSIS 5; SCDO5. Baltimore (MD): Johns Hopkins University; last edited 2025-04-10. Available from: https://omim.org/entry/122600
58. Gucev ZS, Tasic V, Pop-Jordanova N, Sparrow DB, Dunwoodie SL, Ellard S, et al. Autosomal dominant spondylocostal dysostosis in three generations of a Macedonian family: Negative mutation analysis of DLL3, MESP2, HES7, and LFNG. Am J Med Genet A. 2010;152A(6):1378-82. doi:10.1002/ajmg.a.33471
59. 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.
60. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 611913. CHROMOSOME 16p11.2 DELETION SYNDROME, 593-KB. Baltimore (MD): Johns Hopkins University; last edited 2020-01-28. Available from: https://omim.org/entry/611913
61. Sandbacka M, Laivuori H, Freitas É, Halttunen M, Jokimaa V, Morin-Papunen L, et al. TBX6, LHX1 and copy number variations in the complex genetics of Müllerian aplasia. Orphanet J Rare Dis. 2013;8:125. doi:10.1186/1750-1172-8-125
62. Li G, Strong A, Wang H, Kim JS, Watson D, Zhao S, et al. TBX6 as a cause of a combined skeletal-kidney dysplasia syndrome. Am J Med Genet A. 2022;188(12):3469-3481. doi:10.1002/ajmg.a.62972
63. Errichiello E, Arossa A, Iasci A, Villa R, Ischia B, Pavesi MA, et al. An additional piece in the TBX6 gene dosage model: A novel nonsense variant in a fetus with severe spondylocostal dysostosis. Clin Genet. 2020;98(6):628-629. doi:10.1111/cge.13854
64. Sparrow DB, Faqeih EA, Sallout B, Alswaid A, Ababneh F, Al-Sayed M, et al. Mutation of HES7 in a large extended family with spondylocostal dysostosis and dextrocardia with situs inversus. Am J Med Genet A. 2013;161A(9):2244-9. doi:10.1002/ajmg.a.36073
65. Sparrow DB, Chapman G, Smith AJ, Mattar MZ, Major JA, O'Reilly VC, et al. A mechanism for gene-environment interaction in the etiology of congenital scoliosis. Cell. 2012;149(2):295-306. doi:10.1016/j.cell.2012.02.054
66. Shi H, Enriquez A, Rapadas M, Martin EMMA, Wang R, Moreau J, et al. NAD Deficiency, Congenital Malformations, and Niacin Supplementation. N Engl J Med. 2017;377(6):544-552. doi:10.1056/NEJMoa1616361
67. Barhoumi T, Nashabat M, Alghanem B, Alhallaj A, Boudjelal M, Umair M, et al. Delta Like-1 Gene Mutation: A Novel Cause of Congenital Vertebral Malformation. Front Genet. 2019;10:534. doi:10.3389/fgene.2019.00534
68. Umair M, Younus M, Shafiq S, Nayab A, Alfadhel M. Clinical genetics of spondylocostal dysostosis: A mini review. Front Genet. 2022;13:996364. doi:10.3389/fgene.2022.996364
69. Cai H, Huang X, Chen H, Zhao J, Sun H, Huang Y, et al. A Novel Subtype of Spondylocostal Dysplasia Associated With a Heterozygous Missense FLNA Variant. Orthop Surg. 2025;17(6):1890-1896. doi:10.1111/os.70026
70. Jaeken J, Morava E. Congenital disorders of glycosylation, dolichol and glycosylphosphatidylinositol metabolism. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 43, section 43.1.
71. Abay B, Ozel A, Alatas I. Prevalence and Recurrence of Lower-Limb Deformities in Jarcho-Levin Syndrome: A Single-Center Retrospective Cohort Study. Children (Basel). 2026;13(8). doi:10.3390/children13081002
72. Obataya Y, Tsukagoshi Y, Ikezawa Y, Homma Y. Sprengel Deformity Associated with Spondylocostal Dysostosis Treated Using a Modified Wilkinson Procedure: A Case Report. J Orthop Case Rep. 2026;16(6):519-523. doi:10.13107/jocr.2026.v16.i06.7530
73. Demelash K, Belay S, Alebachew H, Hailay G, Yishar F, Weldetsadik A. Spondylocostal dysostosis with split cord malformation in a resource-limited setting: A case report. Radiol Case Rep. 2025;20(9):4727-4730. doi:10.1016/j.radcr.2025.06.013
74. Darouich S, Darouich S, Khemiri A, Bellamine H. Spondylocostal Dysostosis-1 Associated With Pancreatic Heterotopia: Coincidence or True Association? Pediatr Dev Pathol. 2026;29(3):239-242. doi:10.1177/10935266251411748
75. Schuhmann S, Koller H, Sticht H, Kraus C, Krumbiegel M, Uebe S, et al. Clinical and molecular delineation of spondylocostal dysostosis type 3. Clin Genet. 2021;99(6):851-852. doi:10.1111/cge.13952
76. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 609813. SPONDYLOCOSTAL DYSOSTOSIS 3, AUTOSOMAL RECESSIVE; SCDO3. Baltimore (MD): Johns Hopkins University; last edited 2025-03-14. Available from: https://omim.org/entry/609813
77. Otomo N, Mizumoto S, Lu HF, Takeda K, Campos-Xavier B, Mittaz-Crettol L, et al. Identification of novel LFNG mutations in spondylocostal dysostosis. J Hum Genet. 2019;64(3):261-264. doi:10.1038/s10038-018-0548-2
78. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 613686. SPONDYLOCOSTAL DYSOSTOSIS 4, AUTOSOMAL RECESSIVE; SCDO4. Baltimore (MD): Johns Hopkins University; last edited 2013-12-18. Available from: https://omim.org/entry/613686
79. Sparrow DB, Sillence D, Wouters MA, Turnpenny PD, Dunwoodie SL. Two novel missense mutations in HAIRY-AND-ENHANCER-OF-SPLIT-7 in a family with spondylocostal dysostosis. Eur J Hum Genet. 2010;18(6):674-9. doi:10.1038/ejhg.2009.241
80. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 616566. SPONDYLOCOSTAL DYSOSTOSIS 6, AUTOSOMAL RECESSIVE; SCDO6. Baltimore (MD): Johns Hopkins University; last edited 2026-02-13. Available from: https://omim.org/entry/616566
81. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 621523. SPONDYLOCOSTAL DYSOSTOSIS 7, AUTOSOMAL RECESSIVE; SCDO7. Baltimore (MD): Johns Hopkins University; last edited 2026-02-27. Available from: https://omim.org/entry/621523
82. Ramírez N, Monroig-Rivera C, De Jesús-Rojas W, Rosado E, Arciniegas Medina NJ, Cornier AS, et al. Long-Term Follow-up of Untreated Adult Patients with Spondylothoracic Dysostosis (Jarcho-Levin Syndrome). J Bone Joint Surg Am. 2024;106(6):501-507. doi:10.2106/JBJS.23.00800
83. García-Irizarry KD, Perez-Ortiz JJ, Meléndez Montañez JM, De Jesús-Rojas W. Spondylothoracic Dysostosis: Crab-like Rib Cage Configuration in a Neonate with Jarcho-Levin Syndrome. Am J Respir Crit Care Med. 2024;210(6):e1-e2. doi:10.1164/rccm.202402-0382IM
84. Atitallah S, Othmen WB, Rabeh RB, Missaoui N, Bouyahia O, Mazigh S, et al. Jarcho-Levin Syndrome With Fatal Respiratory Failure. Ochsner J. 2025;25(2):119-122. doi:10.31486/toj.24.0111
85. Sahu UP, J R, Kumar S, Vatsana N. Spondylo-Thoracic Dysplasia: Survival, a Rare Occurrence. Cureus. 2024;16(12):e74894. doi:10.7759/cureus.74894
86. Nguyen HM, Lien NTK, Tran TH, Nguyen NL, Nguyen SBT, Bui THC, et al. Novel Splice Variant in the HES7 Gene in Vietnamese Patient with Spondylocostal Dysostosis 4: A Case Report and Literature Review. Diagnostics (Basel). 2025;15(13). doi:10.3390/diagnostics15131587
87. Ilharreborde B, Yazici M, Yüksel S, Demir P, Helenius I. Early onset scoliosis: Can Best Practice Guidelines Be Provided in Europe? J Child Orthop. 2026;20(1):83-95. doi:10.1177/18632521251398742
88. Szoszkiewicz A, Sowińska-Seidler A, Wnuk-Kłosińska A, Bukowska-Olech E, Biel K, Matuszewska K, et al. Deciphering the Genetic Basis of Congenital Vertebral Malformations Through a Stepwise Diagnostic Approach. Int J Mol Sci. 2026;27(4). doi:10.3390/ijms27041752
89. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Spondylocarpotarsal synostosis syndrome. In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford: Oxford University Press; 2018. p. 344–347.
90. Jones KL, Jones MC, del Campo M. Klippel-Feil sequence. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 882–883.
91. Floor E, De Jong RO, Fryns JP, Smulders C, Vles JS. Spondylocostal dysostosis: an example of autosomal dominant transmission in a large family. Clin Genet. 1989;36(4):236-41. doi:10.1111/j.1399-0004.1989.tb03196.x
92. Whittock NV, Ellard S, Duncan J, de Die-Smulders CE, Vles JS, Turnpenny PD. Pseudodominant inheritance of spondylocostal dysostosis type 1 caused by two familial delta-like 3 mutations. Clin Genet. 2004;66(1):67-72. doi:10.1111/j.0009-9163.2004.00272.x
93. Betz RR, Mulcahey MJ, Ramirez N, Flynn JM, Smith JT, St Hilaire T, et al. Mortality and life-threatening events after vertical expandable prosthetic titanium rib surgery in children with hypoplastic chest wall deformity. J Pediatr Orthop. 2008;28(8):850-3. doi:10.1097/BPO.0b013e31818e19a4
94. ClinicalTrials.gov. Search by condition “Spondylocostal Dysostosis”, “Jarcho Levin”, “Spondylothoracic Dysplasia”, “Thoracic Insufficiency Syndrome”. Available from: https://clinicaltrials.gov/
95. Studer D, Hasler CC. Diagnostic and therapeutic strategies in early onset scoliosis: A current concept review. J Child Orthop. 2024;18(2):113-123. doi:10.1177/18632521241228141
96. Studer D, Hasler CC. Long term outcome of vertical expandable prosthetic titanium rib treatment in children with early onset scoliosis. Ann Transl Med. 2020;8(2):25. doi:10.21037/atm.2019.09.158
97. Parnell SE, Effmann EL, Song K, Swanson JO, Bompadre V, Phillips GS. Vertical expandable prosthetic titanium rib (VEPTR): a review of indications, normal radiographic appearance and complications. Pediatr Radiol. 2015;45(4):606-16. doi:10.1007/s00247-014-3128-4
98. Trask M, Yamaguchi JT, Redding G, Yaszay B, Browd S, White KK. Posterior Column Release and Lengthening with a Magnetic Growing Rod Construct in Severe Congenital Thoracic Fusion: A Report of 2 Cases. JBJS Case Connect. 2024;14(4). doi:10.2106/JBJS.CC.24.00256
99. Lin G, Yang Y, Chen Z, Zhao S, Niu Y, Du Y, et al. Preliminary study assessing the long-term surgical outcomes of TBX6-associated congenital scoliosis (TACS) patients using the propensity score matching method: exploring the clinical implications of genetic discoveries in congenital scoliosis. Orphanet J Rare Dis. 2025;20(1):32. doi:10.1186/s13023-024-03471-9
100. Pons-Odena M, Verges A, Arza N, Cambra FJ. Combined use of Neurally Adjusted Ventilatory Assist (NAVA) and Vertical Expandable Prostethic Titanium Rib (VEPTR) in a patient with Spondylocostal dysostosis and associated bronchomalacia. BMJ Case Rep. 2017;2017. doi:10.1136/bcr-2016-217027
101. Zapata-Valentin M, Soto PM, Nater LM, Urban JC, Justiniano VO. Surgical management of the airway in Jarcho-Levin syndrome: a case series. J Surg Case Rep. 2026;2026(6):rjag405. doi:10.1093/jscr/rjag405
102. Ministry of Health, Labour and Welfare (Japan). Document s0607-10g (2007), VEPTR regulatory data (FDA HDE H030009, 24 August 2004). Available from: https://www.mhlw.go.jp/shingi/2007/06/dl/s0607-10g.pdf
103. Children's Hospital of Philadelphia. Vertical Expandable Prosthetic Titanium Rib (VEPTR). Available from: https://www.chop.edu/treatments/vertical-expandable-prosthetic-titanium-rib-veptr
104. U.S. Food and Drug Administration. 510(k) premarket notification K142587: Vertical Expandable Prosthetic Titanium Rib (DePuy Synthes Spine, Inc.). Decision date 2014-11-18. Available from: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K142587
105. Niu Y, Zhou H, Huang X, Shen J. Advances in the etiology of congenital scoliosis: from morphology to spatiotemporal molecular mechanisms (2015-2025). EFORT Open Rev. 2026;11(9):1005-1014. doi:10.1530/EOR-2026-0006
106. Wang X, Zhang Q, Ji E, Sui Y, Zhu Q, Li K, et al. An Evidence-Tiered Developmental Framework for Congenital Vertebral Malformations: Spatiotemporal Thresholds in Genotype-Phenotype Interpretation. Genes (Basel). 2026;17(10). doi:10.3390/genes17101309
107. Varney S, Vernau K, Brown C, Toedebusch C, Vo J, Bannasch D. A Splice Acceptor Variant in DLL3 Is Associated with Spondylocostal Dysostosis in a Litter of Mixed-Breed Dogs. Genes (Basel). 2026;17(2). doi:10.3390/genes17020131
108. Dolak JA, Tartt S. Spinal anesthesia for Cesarean delivery in a parturient with spondylocostal dysostosis. Can J Anaesth. 2009;56(2):172-3. doi:10.1007/s12630-008-9029-y
109. Bellamy J, Pye S. Anesthesia for Cesarean Delivery in a Patient With Spondylocostal Dysostosis: A Case Report. A A Pract. 2024;18(12):e01881. doi:10.1213/XAA.0000000000001881
110. Song E, Romanelli E. Epidural analgesia for a successful vaginal delivery in a parturient with Jarcho-Levin syndrome: case report. BMC Anesthesiol. 2025;25(1):157. doi:10.1186/s12871-025-02978-3
111. Ramírez-Rodríguez Á, Zayas-Vélez JJ, Domínguez D, Carrero E, Crespo MJ. Anesthetic Considerations for and Management of a Parturient With Jarcho-Levin Syndrome: A Case Report of Continuous Spinal Anesthesia for Preterm Cesarean Section. Cureus. 2025;17(8):e90303. doi:10.7759/cureus.90303
112. Bonafé L, Giunta C, Gassner M, Steinmann B, Superti-Furga A. A cluster of autosomal recessive spondylocostal dysostosis caused by three newly identified DLL3 mutations segregating in a small village. Clin Genet. 2003;64(1):28-35. doi:10.1034/j.1399-0004.2003.00085.x
113. Pediatric Spine Foundation. Available from: https://www.pediatricspinefoundation.org/