Type 2 Collagen Dysplasias (COL2A1)

Allelic family of chondrodysplasias caused by COL2A1 variants, ranging from prenatally lethal achondrogenesis to early hip osteoarthritis in adults, with a comparative subsection on achondrogenesis IA (TRIP11), IB (SLC26A2) and II (COL2A1) and on COL2A1-related Perthes-type proximal femoral epiphyseal dysplasia.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

Table 0.1 brings together the data common to the group; Table 0.2 describes each entity within the scope. The abbreviations AD (autosomal dominant), AR (autosomal recessive), SED (spondyloepiphyseal dysplasia) and SMD (spondylometaphyseal dysplasia) are used throughout the article. OMIM (Online Mendelian Inheritance in Man) numbers follow the 2023 nosology and the 2026 ClinGen (Clinical Genome Resource) curations; ORPHA codes follow the Orphanet index [1–3].

Table 0.1. Data common to the group of type II collagenopathies

Table 0.2. Entities within the scope

AD, autosomal dominant; AR, autosomal recessive; SED, spondyloepiphyseal dysplasia; SMD, spondylometaphyseal dysplasia.

1. Introduction

1.1. Founding descriptions

The history of the type II collagenopathies begins with independent radiological descriptions of chondrodysplasias that were shown to be allelic only decades later. In 1952, Wilhelm Kniest described a girl with disproportionate short stature, bulky joints with restricted mobility and vertebral changes, which he distinguished from «enchondral dysostosis» and chondrodystrophy; the patient was re-evaluated at 42 years of age, with short stature, restricted joint mobility and blindness, but with an active life and full mental capacity [15, 32, 33]. In 1966, Spranger and Wiedemann delineated spondyloepiphyseal dysplasia congenita (SEDC) on the basis of six new cases and fourteen from the literature, and in 1970 Spranger and Langer established its radiological semiology [34–36]. Achondrogenesis type II was described by Langer and colleagues in 1969 and by Saldino in 1971, in the context of lethal short-limbed dwarfism, and was delineated as a separate entity by Chen (1981) and Borochowitz (1986), who discussed whether hypochondrogenesis was a distinct disease or the less severe end of the same condition [37–39]. SEMD, Strudwick type, was described by Murdoch and Walker in 1969 and characterised by Anderson and colleagues in 1982 in a series of 14 patients; for years it was considered autosomal recessive because of two siblings born to unaffected parents [40, 41]. Spondyloperipheral dysplasia was described by Kelly in 1977, Czech dysplasia by Kozlowski and colleagues in 2004, and platyspondylic dysplasia, Torrance type, was first grouped with San Diego chondrodysplasia among the «platyspondylic types of short-limbed dwarfism», before the latter was recognised as a phenotype of thanatophoric dysplasia caused by FGFR3 variants [13, 20, 42].

1.2. Milestones of molecular discovery

In 1988, Godfrey and Hollister identified abnormal collagen II in tissue from patients with achondrogenesis II and hypochondrogenesis [39]. In 1989, Lee and colleagues described the first COL2A1 variant, in a family with spondyloepiphyseal dysplasia: a heterozygous single-exon deletion that removed 36 amino acids from the helical domain, so that about 90% of homotrimers contained at least one abnormal chain, a direct demonstration of the dominant-negative effect [43]. In the same year, Vissing and colleagues identified a glycine-to-serine substitution in the triple-helical domain in a perinatally lethal form, with post-translational overmodification, intracellular retention and reduced secretion of collagen [44]. In 1993, Winterpacht and colleagues showed that the same 28-base-pair deletion at the exon 12–intron boundary caused Kniest dysplasia in a 2-year-old girl and a Stickler phenotype in her mother, with maternal somatic mosaicism, which linked Kniest and Stickler to the same gene [45]. In 1994, Spranger, Winterpacht and Zabel proposed the concept of «type II collagenopathies» as a spectrum of chondrodysplasias [46]. In 1995, Tiller and colleagues demonstrated heterozygous COL2A1 variants in three unrelated patients with SEMD Strudwick, which reclassified the disease as dominant, and Mortier and colleagues described a case of achondrogenesis II with a glycine-to-arginine substitution, inclusions of dilated rough endoplasmic reticulum in chondrocytes and a paucity of collagen fibres [41, 47]. The attribution of Torrance dysplasia to dominant-negative C-propeptide variants (Nishimura 2004; Zankl 2005), of spondyloperipheral dysplasia to truncating C-propeptide variants (Zankl 2004) and of Czech dysplasia to the p.(Arg275Cys) variant (Hoornaert 2007) completed the current map of the spectrum [21, 22, 48, 49].

1.3. Nomenclature and reclassifications

The 2023 nosology of the International Skeletal Dysplasia Society uses dyadic names that link the phenotype to the gene: «achondrogenesis, COL2A1-related» replaces achondrogenesis type II or Langer-Saldino type, and «SED with metatarsal shortening, COL2A1-related» replaces Czech dysplasia [1, 6]. In the same revision, mild SED with premature osteoarthritis and Namaqualand-type hip dysplasia are included under COL2A1-related SEDC, and «proximal femoral epiphyseal dysplasia, COL2A1-related» (OMIM 150600 and 608805) groups the Perthes-type forms and familial avascular necrosis of the femoral head, with the caveat that not all cases of these phenotypes are due to COL2A1 and that Meyer dysplasia of the hip is not associated with this gene [1]. The term «achondrogenesis» was historically used for the condition now called Grebe dysplasia, which does not belong to any of the achondrogenesis types [30]. The «Luton type» is phenotypically identical to the Torrance type [13]. Dysspondyloenchondromatosis (enchondroma-like metaphyseal lesions with irregular vertebrae) and SEMD, Algerian type (metaphyseal irregularities, marked genu varum and scoliosis) were later identified as type II collagenopathies [17, 18]. The 2026 ClinGen curations bring together under a single «COL2A1-related SED» relationship SEDC, SEMD Strudwick, SMD Algerian type (OMIM 184253), SMD «corner fracture» type (OMIM 184255), SED Stanescu type (OMIM 616583), Czech dysplasia (OMIM 609162), osteoarthritis with mild chondrodysplasia (OMIM 604864) and dysspondyloenchondromatosis, keeping separate achondrogenesis II/hypochondrogenesis, platyspondylic dysplasia Torrance type, Kniest dysplasia, spondyloperipheral dysplasia, Stickler syndrome type 1 and proximal femoral epiphyseal dysplasia [2]. Orphanet links OMIM number 184253 to an «SMD, Schmidt type» (ORPHA:93316), whereas the Spranger atlas and ClinGen assign it to the Algerian form [2, 17, 50]. For SED Stanescu type, the Spranger atlas gives the number 616538, and Orphanet and ClinGen the number 616583; for SEMD Strudwick, Orphanet, the nosology and ClinGen use 184250. The numbers of the nosology and of ClinGen are adopted here [2, 24, 51].

1.4. Definition of the entities and ranking by severity

Collagen II is the main fibrillar component of the cartilage extracellular matrix, essential for endochondral ossification, linear growth and joint function, and is required for the development and function of the eye and inner ear [6]. The type II collagenopathies share, to varying degrees, skeletal dysplasia, ocular manifestations (myopia, cataract, lens subluxation, vitreous anomalies and retinal detachment), hearing loss and orofacial changes (flat midface, cleft palate, Robin sequence) [6, 9]. GeneReviews groups the entities into four levels of severity: the most severe, often perinatally lethal (COL2A1 achondrogenesis, COL2A1 hypochondrogenesis and platyspondylic dysplasia Torrance type); the severe or moderately severe with neonatal presentation (Kniest dysplasia, SEDC and COL2A1 SEMD); the intermediate, presenting in the neonatal period, childhood or adolescence (spondyloperipheral dysplasia, SED with metatarsal shortening and COL2A1 Stickler syndrome); and the mild form of adolescence or adulthood (mild SED with premature osteoarthritis) [6]. The SDMC consensus ranks in decreasing order of severity achondrogenesis II, hypochondrogenesis, Kniest dysplasia, SEDC, Stickler syndrome type I and mild SED with premature osteoarthritis, with 100% panel agreement [9]. COL2A1 achondrogenesis and hypochondrogenesis form a phenotypic continuum: cases with the most severe radiological and pathological findings receive the first label and the less severe the second [6, 39]. There are no formal clinical diagnostic criteria for any of the entities; the specific diagnosis rests on the combination of clinical and radiological phenotype with genetic testing [6].

1.5. Scope and related entities

This article covers in full the entities in Table 0.2 caused by COL2A1, including COL2A1-related Perthes-type proximal femoral epiphyseal dysplasia, and compares in a single table the three classic achondrogeneses (IA due to TRIP11, IB due to SLC26A2 and II due to COL2A1), with odontochondrodysplasia as the non-lethal allelic form of achondrogenesis IA. Achondrogenesis IB and the other SLC26A2 dysplasias enter the differential diagnosis. Stickler syndrome type 1 and the isolated ocular form are members of the COL2A1 spectrum and appear in the comparison tables. Prenatal diagnosis of the lethal forms and Robin sequence are addressed insofar as they concern the COL2A1 spectrum.

2. Molecular Genetics and Pathophysiology

2.1. The COL2A1 gene and collagen II

The COL2A1 gene is located at 12q13.11 and encodes the pro-α1(II) chain; the MANE Select transcript is NM_001844.5, which encodes the protein NP_001835.3 [4, 5]. Three identical α1(II) chains form the [α1(II)]3 homotrimer, the main fibrillar collagen of hyaline cartilage, nucleus pulposus and vitreous humour, which gives the matrix its resistance to compression [4]. The chain has an N-propeptide, a central triple-helical domain formed by repeats of the Gly-X-Y motif, and a non-collagenous C-propeptide [6, 52]. Collagen II co-assembles with collagens IX and XI in the heterotypic cartilage fibril; collagen XI is a heterotrimer formed by chains encoded by COL11A1, COL11A2 and COL2A1 itself, and this structural relationship explains the phenotypic overlap between Kniest dysplasia, Stickler syndrome and oto-spondylo-megaepiphyseal dysplasia (OSMED) [10, 53]. Exon 2 of COL2A1 undergoes alternative splicing and is expressed mainly in the eye; variants in this exon produce a predominantly ocular phenotype, and in one series 90% of carriers of exon 2 variants had significant eye disease by the age of 20 years [9, 10, 54]. COL2A1 transcription is directly regulated by the transcription factor SOX9, through a SOX9-dependent enhancer in intron 6, and polymorphisms in regulatory elements and in modifier genes have been proposed to explain part of the intrafamilial variability [6, 55].

2.2. Molecular mechanisms

The severity spectrum reflects the degree of disruption of the fibril network: qualitative protein abnormalities give more severe phenotypes and quantitative abnormalities the mildest [9]. Four mechanisms account for most entities. The first is the dominant-negative effect of glycine substitutions in the helical domain: replacement of glycine by a bulkier amino acid interferes with triple-helix formation, the chains become overmodified, are retained in the endoplasmic reticulum and secretion decreases; this mechanism predominates in the severe forms (achondrogenesis II, hypochondrogenesis, SEDC, SEMD) [6, 44, 47]. The second is haploinsufficiency due to nonsense and frameshift variants with a premature termination codon and nonsense-mediated messenger RNA decay, which reduces the amount of normal collagen II without producing abnormal protein; this is the mechanism of Stickler syndrome type 1, in which the variant prevents incorporation of the chain into the trimer [6, 53, 56]. The third is exon skipping or a small in-frame deletion in the helical domain, which generates shortened chains incorporated into heterotrimeric molecules with normal chains; this mechanism is characteristic of Kniest dysplasia and explains its greater severity compared with SEDC [2, 14, 52]. The fourth involves the C-propeptide: dominant-negative variants or deletions in this region and in the adjacent helical domain cause Torrance dysplasia, and truncating C-propeptide variants cause spondyloperipheral dysplasia; both are variants or numerical aberrations of the non-collagenous C-propeptide of the α1(II) chain, which explains the overlap between the two entities [13, 20, 21, 49]. C-propeptide variants can also cause a syndrome of vitreoretinal anomalies with mild phalangeal dysplasia, normal stature and coxarthrosis in adulthood, in which the vitreous loses its usual lamellar organisation and predisposes to retinal detachment, even in the absence of myopia [20]. The accumulation of glycoproteins in chondrocytes described by Stanescu as the hallmark of the SED that bears his name appears to be a non-specific sign of endoplasmic reticulum stress [24].

2.3. Gene–disease validity and dosage sensitivity

The ClinGen Skeletal Disorders Gene Curation Expert Panel published the COL2A1 curations in 2026; it is the gene with the most splits of gene–disease relationships in the set assessed [2]. Table 2.1 summarises them.

Table 2.1. COL2A1 gene–disease relationships according to ClinGen (2026)

Source: [2].

ClinGen assigns COL2A1 a haploinsufficiency score of 3 (sufficient evidence) and a triplosensitivity score of 0; there is no ClinGen actionability assertion for this gene, for either adults or children, and surveillance recommendations come from GeneReviews and the SDMC consensus [4].

2.4. Genotype–phenotype correlations

GeneReviews stresses that there is no clear genotype–phenotype correlation and that phenotypic overlap is large, but there are general rules supported by cohorts [6]. In the Japanese series of Nishimura (2005; 56 families, 38 variants in 41 families), the 22 missense variants and the in-frame deletion in the helical region all fell within the SED spectrum; glycine-to-serine substitutions produced alternating zones of more severe and milder phenotypes along the helix, whereas glycine substitutions by other amino acids gave only more severe phenotypes; the gradient of skeletal severity did not correlate with the extraskeletal manifestations; the nine truncating or splicing variants in the helical region or the N-propeptide gave Stickler syndrome type I or Kniest dysplasia, always with extraskeletal manifestations; and the six C-propeptide variants gave atypical skeletal phenotypes with ocular changes but no ear, nose and throat changes [52]. In the European cohort of Terhal (2015; 93 patients from centres in the Netherlands, Belgium, France and the United Kingdom with SEDC or a related phenotype), 68 patients (73%) had helical glycine substitutions, glycine to serine being the most frequent, followed by glycine to aspartic acid and to arginine; nine had splicing variants; there was no linear gradient of severity from the amino to the carboxyl end (p.Gly744Ser and p.Gly945Ser gave mild SED resembling multiple epiphyseal dysplasia, p.Gly702Ser gave severe SEDC); patients with splicing variants had a more severe ocular and auditory phenotype, and carriers of arginine-to-cysteine substitutions (p.Arg719Cys, p.Arg989Cys) had no ocular problems [7]. In the French series of Barat-Houari (2016; 136 probands selected by decision tree: 71 with a Stickler phenotype, 21 with SEDC, 11 with Kniest and 34 with other dysplasias), 66 different variants were found in 71 positive patients, 44 of them novel; 86% of the variants lay in the helical domain, glycine substitutions clustered in the severe phenotypes and arginine-to-cysteine substitutions in the moderate ones, and 46% of patients with a Stickler phenotype had a COL2A1 variant [57]. Splicing variants appear to confer a higher risk of ophthalmological complications and hearing loss [6, 7].

Arginine-to-cysteine substitutions form a subgroup with a constant, site-specific phenotype that is never lethal: in the series of Hoornaert (2006; 11 probands, six variants), R75C, R519C and R1076C (historical helical numbering; R519C = NM_001844.5:c.2155C>T p.(Arg719Cys); R1076C is the published designation, with no alias match in ClinVar) caused spondyloarthropathy with normal stature and no ocular involvement, with the short third and fourth toes characteristic of R75C and brachydactyly with enlarged finger joints in R1076C; R704C (historical numbering, = NM_001844.5:c.2710C>T p.(Arg904Cys)) caused Stickler syndrome with brachydactyly, R365C (historical numbering, = NM_001844.5:c.1693C>T p.(Arg565Cys)) classic Stickler syndrome and R789C SEDC [58, 59]. The R75C variant corresponds to p.(Arg275Cys) in current numbering and defines Czech dysplasia; R789C corresponds to p.(Arg989Cys), a recurrent SEDC variant described in a Brazilian series of six patients from the skeletal dysplasia group of the State University of Campinas [22, 59, 60]. For growth, the Terhal curve (2012; 79 patients with helical glycine substitutions and 27 with other defects) showed that glycine-to-cysteine substitutions lie in the upper part of the curve, glycine-to-serine or valine substitutions between +1 and −1 standard deviation (SD), substitutions in the carboxy-terminal third below the amino-terminal ones and splicing variants relatively high, with significant intrafamilial differences [61].

2.5. Variant nomenclature

The literature before about 2005 numbers residues from the start of the helical domain; current numbering starts from the initiator methionine of the prepro-collagen in NM_001844.5. In the three pairs confirmed by alias in ClinVar, the difference is 200 codons (R75C = p.Arg275Cys; R789C = p.Arg989Cys; G691R = p.Gly891Arg) [59]. Liu (2005) and Su (2008) already use the prepro-collagen numbering, and the 2023 nosology writes p.R275C, p.G393S, p.G717S and p.G1170S in current numbering [1, 27, 62]. Some published notations do not match the reference sequence: the Spranger atlas writes the Czech variant as «c.823C>T of the COL2A2 gene» with «p.arg823Cys», whereas the gene is COL2A1 and the resulting protein change is p.(Arg275Cys); the «c.3190C>T» of Girisha (2020) gives a reference error in NM_001844.5 and corresponds in ClinVar to c.3397C>T p.(Arg1133Cys), a difference of 207 nucleotides consistent with numbering on an isoform lacking exon 2; the atlas's «c.2349delGTC p.Val783del» does not match the reference sequence (TGG), and deletion of one codon in this region normalises to c.2349_2351del p.(Gly784del), without the original notation being confirmable [5, 8, 23, 59]. Table 2.2 lists the variants cited in the article with normalised Human Genome Variation Society (HGVS) nomenclature.

Table 2.2. COL2A1 variants cited, normalised to NM_001844.5 (GRCh38)

Sources: [5, 8, 17, 20, 59, 63–66]. PPRD, progressive pseudorheumatoid dysplasia.

2.6. Penetrance, expressivity and mosaicism

Penetrance is high, probably complete, with rare cases of apparent reduced penetrance; the SDMC consensus describes inheritance as autosomal dominant with complete penetrance [6, 9]. The mild forms have age-dependent manifestations and great inter- and intrafamilial variability, but severe and mild forms do not coexist in relatives with the same variant: the specific diagnosis tends to recur within the family, with variable expressivity [6]. The mother–daughter family with Torrance dysplasia documented in the Spranger atlas illustrates this expressivity, with more severe platyspondyly in the daughter than in the mother at the same age, and the SEDC figures show a child with well-ossified femoral necks at 4 years and the adult mother with flattened femoral heads but in normal position [13, 16]. Mosaicism is a central element in the genetics of these diseases. The SDMC consensus estimates somatic mosaicism in 6–10% of parents of children with an apparently germline variant; the mosaic parent often has subtle signs, such as a Stickler phenotype or mild SED, and their recurrence risk may reach 50% per child [9, 45, 67]. Individuals with somatic mosaicism for C-propeptide variants may have minimal bone changes and children with Torrance dysplasia [20]. Cases of sibling recurrence due to paternal and maternal somatogonadal mosaicism and to gonadal mosaicism have been published [68–71]. Somatic mosaicism in the patient may attenuate the phenotype: Yilmaz (2026) described the p.(Gly513Ser) variant, associated with lethal phenotypes, with an allele ratio of 36:17 in a patient with marked skeletal asymmetry and a non-lethal phenotype [66].

2.7. Autosomal recessive inheritance

Recessive inheritance due to biallelic hypomorphic variants is documented in several families, mainly with SEDC or severe SEMD [8, 72–76]. In the Barat-Houari family, the proband was homozygous for p.Pro458Leu; the heterozygous father had severe myopia and skeletal complaints and the heterozygous mother had no phenotype [64, 73]. In the Tham and Barat-Houari families, one heterozygous father had high myopia, lower-limb asymmetry and average stature, one heterozygous mother measured 154 cm and the other two heterozygotes had normal stature [6]. Girisha (2020) described six patients from four families with SEMD of intermediate to variable severity, homozygous for p.(Arg1379Cys) or p.(Arg1133Cys) [8]. Two cases of severe SEDC with a homozygous variant are also recorded in Smith's textbook [36].

2.8. Allelic disorders and modifiers

Besides the entities covered, the allelic spectrum of COL2A1 includes Stickler syndrome type 1 and the isolated ocular form, «corner fracture» SMD, which is heterogeneous and more often caused by FN1, and the syndrome of vitreoretinal anomalies with phalangeal dysplasia due to C-propeptide variants [2, 20, 28]. Partnerships between people of short stature may give rise to double heterozygosity for two dominant dysplasias, with distinct and sometimes severe phenotypes, such as the combination of pseudoachondroplasia and SEDC [6, 77, 78]. Modifiers of expressivity have not been identified; environmental factors, polymorphisms in modifier genes and variation in SOX9-dependent regulatory elements have been proposed [6].

2.9. Genetics of achondrogenesis IA and IB

TRIP11-related achondrogenesis (formerly type IA) is autosomal recessive and results from loss of function of TRIP11, which encodes GMAP-210, a microtubule-associated golgin of the Golgi apparatus [1, 29]. In the mouse model with a nonsense variant in Trip11, Golgi architecture is altered in several tissues, including cartilage; chondrocytes show endoplasmic reticulum dilatation and stress, abnormal differentiation and cell death; Golgi-dependent glycosylation is altered; and there is intracellular accumulation of perlecan but not of collagen II or aggrecan; loss-of-function variants were found in all ten unrelated patients with achondrogenesis IA studied [29]. Odontochondrodysplasia is caused by hypomorphic TRIP11 variants and is the non-lethal counterpart of achondrogenesis IA (null phenotype): variants with residual function maintain partial Golgi integrity and normal global protein secretion, which are lost in achondrogenesis IA, and both conditions share a chondrocyte maturation defect due to abnormal glycan processing and deficient proteoglycan secretion [31]. SLC26A2-related achondrogenesis (formerly type IB or Fraccaro type, OMIM 600972) is autosomal recessive and belongs to group 4 of the nosology [1]. In contrast, achondrogenesis II is dominant and almost always de novo, with possible recurrence due to parental mosaicism, which decisively changes counselling after a prenatal diagnosis of achondrogenesis (section 8) [30, 39].

3. Clinical Spectrum and Manifestations by System

3.1. Overview of the spectrum

Table 3.1 reproduces the GeneReviews clinical and radiological comparison, which serves as the common thread for the subsections by entity. The three most severe forms can be very difficult to distinguish prenatally [6].

Table 3.1. COL2A1 entities: age at diagnosis, ossification, stature and distinguishing features

Source: [6].

3.2. COL2A1 achondrogenesis and hypochondrogenesis

COL2A1 achondrogenesis is the most severe form and usually presents prenatally with extreme micromelia, a narrow chest with pulmonary hypoplasia, a short trunk with a protuberant abdomen, flat midface, micrognathia, often cleft palate or Robin sequence, and oedema with a hydropic appearance; polyhydramnios is frequent [6, 30]. Birth length is between 27 and 36 cm; the calvaria is large, with wide fontanelles, and cranial ossification is normal, unlike in achondrogenesis IA [39]. On radiographs, the vertebral bodies do not ossify or ossify very late, the sacrum does not ossify, the chest is barrel-shaped with short ribs without fractures, the ilia are small with crescent-shaped medial and inferior margins, the pubis and ischium do not ossify and the tubular bones are very short, with metaphyseal flaring and cupping in severe cases; the fibula is disproportionately long and there is metaphyseal irregularity of the distal ulna [30, 39]. Occasional anomalies have been described, such as soft-palate cleft, microtia, postaxial polydactyly of the feet, cystic hygroma, proximal small-bowel diverticulosis, atrial septal defect and atrioventricular canal defect [39]. Most do not reach term: preterm births, stillbirths or death in the first hours from cardiorespiratory failure and pulmonary hypoplasia; one child survived to 3 months [6, 39].

Hypochondrogenesis shares short limbs, a small chest, a flat facial profile and Robin sequence, with less severe skeletal involvement and course; the vertebral bodies are small and ovoid, unossified in the cervical region, the pubis does not ossify, the ilia are hypoplastic and the long bones short and relatively broad [6]. Most are born alive and die from cardiorespiratory failure between the first days and 3 months [9, 79]. The figures of the Spranger atlas document the gradient: a newborn with hypochondrogenesis died at 3 weeks from respiratory failure; a stillborn infant at 38 weeks had flat, unossified vertebral bodies in the cervical and sacral spine; another, born at term, died at 4 weeks; in a stillborn infant at 39 weeks, the ischia were ossified but coronal clefts separated the anterior and posterior ossification centres of some vertebral bodies [30]. The SDMC panel did not reach consensus (27.3% agreement) on the statement that hypochondrogenesis is uniformly lethal: in the original description by Maroteaux (1983, four cases), one patient survived and was reclassified as SEDC, which creates an arbitrary distinction based solely on survival, and the panel recognises survivors with the radiology of hypochondrogenesis [9]. The mildest cases of hypochondrogenesis and the most severe cases of SEDC may be indistinguishable [16, 30].

3.3. Achondrogeneses: comparison of types IA, IB and II

The term «achondrogenesis» designates three distinct genetic diseases that converge on a lethal prenatal phenotype of extreme micromelia and deficient ossification. The distinction has direct consequences for counselling, because achondrogenesis II is dominant and usually de novo, whereas achondrogeneses IA and IB are recessive, with a 25% recurrence risk (section 8). Table 3.2 compares them.

Table 3.2. Comparison of achondrogeneses IA, IB and II

Sources: [1, 6, 29–31, 39, 80]. MED, multiple epiphyseal dysplasia.

3.4. Platyspondylic dysplasia, Torrance type

The infant has short-limbed dwarfism, a narrow chest, a protuberant abdomen, a disproportionately large head and coarse facies with a depressed nasal bridge and prominent forehead; polyhydramnios is frequent and short limbs are detected on prenatal ultrasound [13]. Neonatal radiographs show variable platyspondyly, with wafer-like, disc-shaped or ovoid vertebral bodies, a chest of reduced height with short ribs with cupped anterior ends, hypoplasia of the lower ilia, broad ischium and pubis, and short, broad tubular bones with flared, cupped metaphyses and straight femora [13]. Most die at or shortly after birth, but at least two patients reached adulthood with normal psychomotor development and a height of 127 cm; in the adult, the spine has vertebral bodies of reduced height with irregular endplates, the capital femoral epiphyses are small, there is premature coxarthrosis, carpal dysplasia and epimetaphyseal dysplasia of the short tubular bones with delta-shaped metaphyses [13, 81]. In the atlas's mother–daughter family, the mother had no major spinal anomaly at about 7 years and at 26 years showed anterior flattening of L2–L3, dorsal wedging of L1 and L4, narrow intervertebral spaces, persistent hypoplasia of the basilar portion of the ilia and asymmetric flattening of the femoral heads [13]. Okamoto (2012) described two siblings with Torrance dysplasia due to a helical-domain variant, born to phenotypically normal parents [71].

3.5. Kniest dysplasia

Kniest dysplasia is the most severe type II collagenopathy compatible with prolonged survival [6, 9]. The face shows a flat midface, depressed nasal bridge and sometimes shallow orbits with prominent eyes; the trunk is short, with dorsal kyphosis, accentuated lumbar lordosis and sometimes late thoracic scoliosis; the chest is short and broad; the extremities are short, with prominent joints and restricted mobility; cleft palate occurs in about 50% of patients and clubfoot in some [14]. The head is of normal size, relatively large for height; the enlarged joints have variable pain and stiffness, there are hip flexion contractures and the child cannot make a fist because of bony enlargement and swelling of the interphalangeal joints [15]. Radiographs show platyspondyly with anterior wedging, coronal clefts in the lumbar vertebral bodies in infancy, broad ilia with basilar hypoplasia, very broad and short femoral necks, late-ossifying capital femoral epiphyses (2nd–3rd year in mild cases, only in adolescence in severe cases) that end up large and flattened, and short tubular bones with broad metaphyses and large epiphyses (dumbbell-shaped femora); by 3 years the pelvis becomes cup-shaped and cloud-like radiodensities appear on both sides of the growth plates [14, 15]. Extra ossification centres at the distal ends of the proximal phalanges of the 2nd to 5th fingers are typical of Kniest and occasionally seen in SEDC [14].

The course is variable: some patients die in the neonatal period from respiratory failure, others live with mild disproportionate short stature and kyphoscoliosis [15]. Walking may be delayed until 2–3 years; chronic otitis media linked to cleft palate, conductive and sensorineural hearing loss, myopia, retinal detachment and cataract are the main complications; intelligence is normal despite motor and speech delay; and final height is between 106 and 145 cm, a range consistent in the Spranger and Smith atlases [14, 15]. Other manifestations include inguinal and umbilical hernias, short clavicles, tracheomalacia, lens dislocation and glaucoma [15]. Hypoplasia of the dens of the axis predisposes to cervical instability and spinal cord compression, and progressive cervical kyphosis is often associated with stenosis [6, 82]. In the ophthalmological series of Sergouniotis (2015) and in the SDMC review, Kniest is associated with myopia, lens subluxation, retinal detachment and other vitreous anomalies, and an early description by Siggers (1974) reported deafness in 75% [9, 83]. Hu (2026) estimates that fewer than 150 cases have been published since 1952 [65].

3.6. Spondyloepiphyseal dysplasia congenita

SEDC presents at birth with disproportionate short-trunk dwarfism, a short neck, barrel chest, pectus carinatum, genu valgum (less often varum), flat face, sometimes widely spaced eyes, hands and feet of normal size in most, and occasionally cleft palate or clubfoot [16]. Smith's textbook adds malar hypoplasia, muscle weakness with easy fatigability, hypoplasia of the abdominal muscles and, occasionally, hip dislocation [36]. Infants are hypotonic, motor development is delayed and mental development is normal; walking is late and waddling; there may be morning stiffness, usually without disproportionate joint pain in childhood [16, 36]. Severe short stature may not be evident until 2–3 years [10]. On infant radiographs, the ossification centres of the pubis and of the knee epiphyses are absent, the upper cervical vertebral bodies do not ossify and the thoracic and lumbar bodies are small, ovoid or pear-shaped; in childhood, the vertebral bodies are flat with anterior defects, the dens of the axis is hypoplastic and late to ossify, the acetabular roofs are horizontal, the femoral head and neck ossify late, with coxa vara in many, and the carpus and tarsus ossify late; in the adult there is a short spine with usually moderate kyphoscoliosis, accentuated lumbar lordosis, hypoplasia of the dens or failure of fusion with the body of C2 (os odontoideum appearance), coxa vara with large high trochanters and small, deformed femoral heads that are nevertheless in normal position [16]. Short stature is proportional to the severity of the epiphyseal and metaphyseal changes [16].

Adult height differs between sources: 85–145 cm in the Spranger atlas and 94–132 cm in Smith's textbook; the Terhal curve (2012), built from longitudinal measurements of patients with glycine substitutions, gives a mean stature of −2.6 SD at birth, −4.2 SD at 5 years and −5.8 SD in adulthood (World Health Organization reference), with a mean adult height of 138.2 cm and wide dispersion [16, 36, 61]. The frequency of ocular complications also diverges: the Spranger atlas reports myopia in about half of patients and states that it rarely leads to detachment, whereas Smith's textbook gives retinal detachment in 50%; the Terhal cohort, the only one with an explicit denominator, found myopia in 45% and detachment in 12% (section 3.13) [7, 16, 36]. The Brazilian series of Silveira (2015) followed six patients with the recurrent p.(Arg989Cys) variant clinically and radiologically [60].

3.7. SEMD, Strudwick type, and metaphyseal forms

SEMD, Strudwick type, is identical to SEDC in the newborn and is distinguished during the first year by the appearance of metaphyseal flaring and irregularity («dappled metaphyses»), absent in SEDC [6, 41]. Clinical features include disproportionate dwarfism with a short spine, progressive scoliosis, pectus carinatum, genu valgum, flat face, myopia in about half of patients, cleft palate in about 20% and occasionally clubfoot [17]. In childhood, coxa vara appears with marked metaphyseal irregularities, multicentric ossification of the femoral necks and deep metaphyseal defects that produce marked genu valgum, while the short tubular bones are relatively spared; short stature and deformities are proportional to the metaphyseal changes [17]. Progressive scoliosis, malalignment and leg-length discrepancy of the lower limbs are frequent, retinal detachment is rare and adult height is between 85 and 137 cm [17]. The risk of cervical instability and spinal cord compression is equal to or greater than in SEDC [6, 84]. In the Terhal cohort, five patients had SEMD Strudwick, four with glycine substitutions (p.Gly519Ser, p.Gly1005Ser, p.Gly1053Val, p.Gly1122Arg) and one with a splicing variant [7]. The association of enchondroma-like metaphyseal lesions with irregular vertebral bodies was described as dysspondyloenchondromatosis, and metaphyseal irregularities with marked genu varum and scoliosis as SEMD, Algerian type; both were later recognised as type II collagenopathies, and the atlas illustrates a 10-year-old patient with p.(Gly753Asp) with asymmetric cystic structures and dense trabeculae in the metaphyses [17, 18]. «Corner fracture» SMD, phenotypically and genetically heterogeneous, presents with a waddling gait as the first sign and progressive developmental coxa vara with continuous slippage of the femoral neck; it is sometimes due to COL2A1 and more often to FN1 [28].

3.8. Spondyloperipheral dysplasia

Spondyloperipheral dysplasia presents at birth with relatively proportionate short stature, short hands and feet, midface hypoplasia, myopia, inconstant cleft palate and clubfoot; growth remains below the 3rd centile and adult height is between about 130 and 150 cm [20]. Infant radiographs show delayed ossification of the lower pelvis and pubis, short and broadened tubular bones and ovoid vertebral bodies; in childhood, type E brachydactyly, short ulnae, mildly flattened vertebral bodies, accentuated lumbar lordosis and flat, irregular capital epiphyses appear [6, 20]. Premature hip osteoarthritis causes joint pain [20, 21]. One of the atlas patients, with p.(Cys1438Ter), had flat capital epiphyses at 14 years but well-ossified femoral necks with a normal angle, unlike most patients with SEDC, and a bone age advanced by about 1 year [20]. In the Terhal cohort, two patients with carboxy-terminal frameshift variants had this phenotype [7]. Bedeschi (2011) described the prenatal presentation and management of a mother and son with a C-propeptide variant [85].

3.9. SED with metatarsal shortening (Czech dysplasia)

Czech dysplasia presents in late childhood with severe joint pain in the lower limbs, broad knees and shortening of the postaxial toes (usually the third and/or fourth), with average stature and without ocular or orofacial changes [6, 23]. Radiographs show mild platyspondyly with irregular endplates, narrowed intervertebral spaces, deformed femoral heads, a dysplastic pelvis with irregular acetabula and shortening of the metatarsals and sometimes the metacarpals [6]. Knee and hip pain appears in adolescence and leads to early arthroplasty [23]. Sensorineural hearing loss is part of the phenotype; it was discovered in young adults, but audiometry should be performed earlier [23]. The disease is due almost exclusively to the c.823C>T p.(Arg275Cys) variant, identified in the original patients and in three additional ones; in a third original patient it was excluded and p.(Tyr1391Cys) in the C-propeptide was found, with a spondyloperipheral phenotype, a variant also described in Torrance dysplasia, which supports a Torrance–spondyloperipheral continuum [22]. Only one patient with a proven p.(Arg275Cys) variant had short stature [23]. In a Brazilian family with 12 carriers of c.823C>T (Moreira 2024), half met the 2010 American College of Rheumatology and European League Against Rheumatism classification criteria for rheumatoid arthritis, with arthritis of the hands, sonographic synovitis and abnormal inflammatory markers, and all had moderate to severe functional disability on the Health Assessment Questionnaire-Disability Index; dominant inheritance, platyspondyly, sensorineural hearing loss and short metatarsals distinguished the dysplasia from rheumatoid arthritis [86]. Tzschach (2008) delineated the phenotype in a large family and Burrage (2013) described the presentation in early childhood [23, 87].

3.10. SED, Stanescu type

SED, Stanescu type, presents between 3 and 6 years with walking difficulty, joint pain and progressive contractures, mainly of the hips, knees and fingers, with normal stature and sometimes mild scoliosis [24]. Radiographs show flat vertebral bodies with irregular endplates and marked anterior ossification defects, flat capital femoral epiphyses, coxa valga, broad femoral necks and slender metacarpals and phalanges with swollen ends and large, rounded epiphyses, with normal bone age [24]. It was associated with heterozygous COL2A1 variants by Hammarsjö (2016), and Jurgens (2015) described p.(Gly207Arg) with a phenotype resembling progressive pseudorheumatoid dysplasia and SED Stanescu [63, 88].

3.11. Mild SED with premature osteoarthritis

This is the mildest form of the spectrum: progressive joint pain and limited mobility of the hips and knees from adolescence or adulthood, with epiphyseal dysplasia and early osteoarthritis on radiographs, and usually normal stature, vision, hearing and orofacial structures [6, 25]. Early degenerative osteoarthritis often necessitates arthroplasty before the age of 50 years [9, 25]. Kannu (2010) proposed premature arthritis as a distinct collagen II phenotype [25].

3.12. COL2A1-related proximal femoral epiphyseal dysplasia (genetic Perthes)

COL2A1 variants are associated with a phenotype resembling Legg-Calvé-Perthes disease, more precisely with dysplastic proximal femoral epiphyses [6]. In three Taiwanese families with autosomal dominant avascular necrosis of the femoral head linked to 12q13, Liu (2005) identified the p.Gly1170Ser variant (exon 50) in two families with different haplotypes and p.Gly717Ser (exon 33) in a third; no variants were found in sporadic cases [62]. Su (2008) studied a five-generation Chinese family (42 members, 16 affected) in which p.Gly1170Ser co-segregated with three age-dependent presentations relative to closure of the femoral epiphysis: Perthes disease before closure (6–14 years, five patients), avascular necrosis during closure (15–18 years, six patients) and isolated early hip osteoarthritis afterwards (21–34 years, five patients) [27]. Kannu (2011) described two children with bilateral Perthes-type hip dysplasia and novel COL2A1 variants, and highlighted the implications for prognosis, surveillance and reproductive counselling [89]. In the Terhal cohort, one patient with p.Gly744Ser had a Perthes-type multiple epiphyseal dysplasia phenotype, with an enlarged and flattened left femoral epiphysis after avascular necrosis at 12.5 years [7]. Namaqualand-type hip dysplasia belongs to the same group [1, 2]. In Istanbul, Tüysüz (2026) used exome sequencing to study 40 children from 36 families with bilateral femoral head dysplasia mimicking bilateral Perthes disease and achieved a diagnosis in 12 families (33.3%), six of them with COL2A1 variants; seven families had a phenotype consistent with mild SED and six with Stickler syndrome, and three children with closely located COL2A1 variants shared short stature with bilateral irregular femoral heads [90]. Conversely, COL2A1 screening in 49 cases of idiopathic osteonecrosis of the femoral head found no variant, which limits the yield of testing in sporadic unilateral cases [9].

In type II collagenopathy, ossification of the femoral head, normally visible at 4–10 months, is markedly delayed and, when it appears, is small, stippled and irregular, mimicking the fragmentation phase of bilateral, symmetrical and slowly progressive Perthes disease; the child is usually asymptomatic, unlike in Perthes disease, and magnetic resonance imaging (MRI) or scintigraphy do not show the avascular phase [9]. Perthes disease is bilateral in only 10–15% of cases and, in that case, asynchronous; synchronous and symmetrical involvement favours a dysplasia [6, 9]. The femoral head may undergo true osteonecrosis, especially in Stickler syndrome, which is treated like Perthes disease [9].

3.13. Manifestations by system and natural history

The Terhal cohort (2015) is the main source of frequencies with a denominator: 93 European patients (51 female) with SEDC (64), SEMD (5), Kniest (7), spondyloperipheral dysplasia (2), Torrance-type phenotype (2) and 13 of normal stature with mild SED or a Stickler-type or multiple epiphyseal dysplasia-type phenotype; 80 had short stature and the median age was 17 years (4 months to 70 years) [7]. These are data from patients followed in genetics centres, collected retrospectively, and are not equivalent to population prevalences; Table 3.3 summarises them.

Table 3.3. Frequencies in the Terhal cohort (n = 93, Europe, published in 2015)

Source: [7]. CI, confidence interval; IQR, interquartile range.

Respiratory system. The manifestations affect the entire airway, from the midface and oropharynx to the parenchyma and chest, and the earlier the presentation the greater the likelihood of respiratory problems [9]. In severe cases, a small chest, pulmonary hypoplasia, tracheobronchomalacia and central apnoea due to cervical instability combine; kyphoscoliosis may compromise respiratory function with growth [9]. Adults of short stature have sleep-disordered breathing, to which the sequelae of cleft repair (obstructive pharyngeal apnoea) and tracheobronchomalacia contribute; an unrecognised unstable cervical spine may present as central apnoea [9]. Patients with predominant shortening of the trunk, as in SEDC, may have insufficient pulmonary reserve, hypoxaemia and sleep hypoventilation [91].

Spine. Hypoplasia of the dens of the axis is inversely related to height and, together with ligamentous laxity and hypotonia, predisposes to atlantoaxial dislocation and spinal cord compression at the C2–C3 level; increased fatigability and abnormal respiratory control may be the first symptoms of compression [16]. The incidence of atlantoaxial instability with myelopathy in severe SEDC is estimated at between 28% and 43%, depending on the study [9]. Kyphoscoliosis is common, often progressive and may appear in the first years, with a risk of spinal cord compression with upper motor neurone signs and bladder and bowel dysfunction; spinal cord injury after limb surgery has been described in these children [9]. In a set of five published cases of high thoracic spinal cord infarction after prolonged anaesthesia, four had mucopolysaccharidosis and one SEDC [82].

Eye. The ocular spectrum includes myopia, vitreous changes, retinal detachment, lens subluxation, cataract and blindness [9]. In Kniest and Stickler syndrome the risk is high; in SEDC it is intermediate (myopia 45%, detachment 12%); in Czech dysplasia, mild SED and the forms with arginine-to-cysteine substitution it is usually absent [6, 7, 58]. In the Terhal cohort, two patients with a normal examination at 6 weeks and 4 months had myopia above 10 dioptres at 6 years [7].

Hearing and orofacial region. Hearing loss occurs in Kniest, SEDC and Stickler syndrome, and may be sensorineural, conductive or mixed; it is reported in 60% in Stickler syndrome, 75% in an early description of Kniest and 37% in the cohort of 93 patients [7, 9]. Cleft palate (soft, with or without hard-palate involvement, bifid or short uvula) and Robin sequence are characteristic of the group [9].

Joints and limbs. Early osteoarthritis presents with pain and limitation before joint destruction, mainly in the hip and knee, rarely with swelling, warmth or acute-phase reactants; some patients become wheelchair-dependent [9]. In disproportionate short stature, coxa vara with limited abduction, flexion and external rotation contractures, a Trendelenburg gait that worsens with fatigue and severe hip osteoarthritis in the 3rd–4th decades are typical, and genu valgum is very common [9]. In Stickler syndrome (Rose 2001), 79% of adults had hip pain and one third hip osteoarthritis [9, 92].

Other systems. Children with type II collagenopathies do not have intellectual disability [9]. Mitral valve prolapse was described as more common in Stickler syndrome, but a study of patients with molecular confirmation showed no increase in prolapse or other cardiac anomalies [9].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age and context

In the fetus, suspicion arises in the presence of shortened long bones with less ossification than expected, especially of the spine, knee epiphyses and pubis, a small chest, micrognathia, cleft palate, polyhydramnios or hydrops [6, 9]. Lethal dysplasias tend to show limb shortening below the 2.3rd centile between 16 and 18 weeks, whereas in SEDC the shortening becomes evident at around 21–24 weeks; the discrepancy between very short limbs and relatively preserved fetal weight favours SEDC over a lethal dysplasia [93]. In the prenatal series reviewed by Larissa (2026), SEDC was detected between 11 and 28 weeks, with non-specific manifestations [64].

In the newborn, type II collagenopathy is suggested by disproportionate short-trunk short stature, flat midface, cleft palate, Robin sequence, clubfoot and, on the radiograph of a term newborn, vertebral coronal clefts, an occipital defect posterior to the foramen magnum, absent ossification of the distal femoral epiphysis and absent ossification of the pubis [9]. Every newborn with Robin sequence should have a detailed ophthalmological and audiological examination to look for type II collagenopathy; in a retrospective series of 115 patients with Robin sequence, 21 (18.3%) had Stickler syndrome [9].

In the child, suspicion rests on the combination of disproportionate short stature, radiological spondyloepiphyseal dysplasia, palatal anomalies, early or high myopia and hearing loss [9]. Bilateral Perthes disease, especially if symmetrical and synchronous, or an asymptomatic child with small, stippled and irregular femoral heads, should prompt investigation for a dysplasia, including type II collagenopathy [6, 9]. In adolescents and adults of average stature, early hip or knee osteoarthritis, an inflammatory-looking arthritis with unimpressive markers, short toes (Czech dysplasia) or a family history of retinal detachment, cleft palate and early osteoarthritis point to the mild forms [6, 9, 86]. The family history should cover three generations and ask about cleft palate, joint pain or deterioration, sudden loss of vision or retinal detachment, and hearing loss [6].

4.2. Warning and emergency signs

Table 4.1. Warning signs and management

4.3. Newborn screening

There is no biochemical newborn screening for type II collagenopathies, because there is no measurable biomarker [4]. In newborns with SEDC or a related phenotype, the Terhal cohort recommends otoacoustic emissions in the neonatal period or, failing that, auditory brainstem evoked potentials, and the SDMC craniofacial consensus recommends hearing assessment at birth or at diagnosis [7, 91].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Consensus criteria and clues

There are no formal clinical diagnostic criteria; the specific diagnosis should be confirmed by genetic testing [6]. The SDMC consensus approved, with 100% agreement, that the diagnostic clues are disproportionate short stature, radiological spondyloepiphyseal dysplasia, palatal anomalies, myopia and hearing loss, and that the prenatal presentation is characterised by the lack of expected ossification [9]. A complete radiological skeletal survey is essential for diagnosis in most cases, and in the classic severe forms it is often sufficient for a clinical diagnosis, which molecular testing confirms [4, 9]. The physical examination records height, weight, head circumference and body proportions, craniofacial morphology (flat profile, hypertelorism, cleft palate, retrognathia) and the joints (enlargement, hypermobility, contractures) [6].

5.2. Radiological signs

In the term newborn, the distal femoral and proximal tibial ossification centres should be visible; their delay characterises achondrogenesis II, hypochondrogenesis, Kniest and SEDC and is absent in the milder forms [9]. Absent ossification of the pubis occurs in the lethal and moderately severe forms, sometimes with absent ossification of the sacrum and ischia [9]. Coronal clefts (vertical lucency in the vertebral body on the lateral view) occur in Kniest and Stickler syndrome and usually not in SEDC; they also occur in chondrodysplasia punctata, Desbuquois dysplasia, metatropic dysplasia and Larsen syndrome [9, 82]. The inferior occipital defect posterior to the foramen magnum is not described in the literature and is based on the panel's experience in newborns with SEDC; the statement that these signs are «highly specific» did not reach consensus (63.6%) [9]. Platyspondyly is common to pseudoachondroplasia, metatropic dysplasia and the type II collagenopathies, but anterior vertebral beaking is typical of pseudoachondroplasia and the mucopolysaccharidoses and, if present, should prompt leucocyte enzyme assays and measurement of urinary glycosaminoglycans [82]. Kenis (2026) retrospectively analysed the pelvic radiographs of 135 patients with a pathogenic COL2A1 variant (SEDC, multiple epiphyseal dysplasia, Kniest and Stickler) and found multiple ossification centres of the superior pubic ramus in 20 (15.5%), all with an SEDC phenotype: sensitivity of 36.4%, specificity of 99.1% and accuracy of 96.5% for SEDC compared with the other COL2A1 phenotypes and with historical general-population data; the authors call for cautious interpretation because of the limitations of historical controls [94]. In SEDC, bone age is markedly delayed; in Kniest, the dysplasia table of the Oxford Desk Reference describes it as advanced, and in the atlas's spondyloperipheral patient it was advanced by about 1 year [10, 20, 95].

5.3. Prenatal imaging

The ultrasound predictors of lethal dysplasia cited by the Oxford Desk Reference are early and severe shortening of the long bones, a femur length/abdominal circumference ratio below 0.16, a small chest with short ribs and a high cardiothoracic ratio, and marked bowing or fractures; clubfoot is an ultrasound sign of SEDC [80]. In SEDC, short limbs (below the 5th centile) and delayed ossification of the spine, knee epiphyses and pubis make ultrasound diagnosis difficult [9]. In the review by Larissa (2026; six studies, seven fetuses with confirmed COL2A1 SEDC), five cases were familial and two sporadic, prenatal molecular diagnosis was made in three (chorionic villus sampling at 12 weeks with a nuchal translucency of 4 mm in a familial context; amniocentesis at 19+2 and 19+3 weeks), all variants were missense variants identified by Sanger sequencing, there was no reliable prenatal genotype–phenotype correlation and diagnosis required molecular confirmation or a family history [64]. In a Mexican case without a family history, ultrasound at 30+2 weeks showed a femur below the 1st centile, a thoracic circumference index of 0.65, a femur/foot ratio of 0.72, a cisterna magna above 12 mm and reduced fetal movements; after fetal MRI at 35 weeks and caesarean section at 36+5 weeks, the newborn had respiratory compromise, delayed ossification of the pubis, S5 and the patella, and the stop-loss variant c.4462T>G p.(Ter1488GluextTer34), classified as likely pathogenic [64]. Ueki (2026) stresses the importance of population biometric charts: the same fetal weight corresponded to the 26.5th centile on Japanese charts and to the 8th centile on INTERGROWTH-21st charts [93]. Yazici (2010) described fetal MRI in Kniest dysplasia [96].

5.4. Molecular testing

The SDMC consensus recommends COL2A1 molecular testing whenever clinical or radiological features suggest type II collagenopathy (90.9% agreement), because of its value for anticipatory guidance, recurrence counselling, prenatal diagnosis, identification of parents with mosaicism and exclusion of other conditions [9]. The Definitive ClinGen curations support the inclusion of COL2A1 in diagnostic skeletal dysplasia panels [2]. The strategy depends on the context. When the clinical picture and family history point to COL2A1, single-gene sequencing detects missense, nonsense and splicing variants and small insertions and deletions; if negative, targeted deletion and duplication analysis follows [6]. In diagnostic uncertainty, as in prenatal assessments, a multigene panel including COL2A1, the genes of the differential diagnosis and deletion and duplication analysis is preferred; the genes included and the sensitivity vary between laboratories [6]. Exome or genome sequencing is an alternative when the phenotype is atypical; it was the method used in the Istanbul series of bilateral Perthes-type dysplasias and in the Kniest case of Hu (2026), the latter as a trio [65, 90]. There is no confirmatory biochemical test [4].

5.5. Limitations of testing

A negative result does not exclude the diagnosis. Sequencing usually does not detect deletions or duplications of one or more exons or of the whole gene [6]. Deep intronic variants with aberrant splicing are well recognised in COL2A1, especially in Stickler syndrome, and escape sequencing limited to exons and junctions [10]. Analysis of leucocyte DNA may fail to detect low-level somatic mosaicism and does not detect exclusively gonadal mosaicism; in parents, a negative result does not exclude recurrence risk [6]. In the proband, somatic mosaicism may explain asymmetric or attenuated phenotypes and require allele-ratio analysis or study of another tissue [66]. Apparent homozygosity in a recessive form should prompt consideration of an undetected exon deletion on the other allele or of uniparental isodisomy [6]. The proportion of patients with a suggestive phenotype in whom a variant is found differs from the clinical sensitivity of the test: in the French series of Barat-Houari, 46% of patients with a Stickler phenotype had a COL2A1 variant, because some of them had another aetiology, and in the Istanbul series the overall yield in bilateral Perthes-type dysplasias was 33.3% [57, 90].

5.6. Interpretation and report content

The report should state the gene, the transcript with version (NM_001844.5), the variant in HGVS nomenclature at the complementary DNA and protein levels, the zygosity, the classification according to the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria with the criteria applied, and, where relevant, the historical helical nomenclature (Table 2.2) [5, 59]. Glycine substitutions in the helical domain receive additional points in the ClinGen curations [2]. Confirmed de novo occurrence is a strong pathogenicity criterion in a dominant disease with high penetrance; in the case of Hu (2026), the c.4121A>G p.(Gln1374Arg) variant was classified with the PS2, PM2, PP3 and PP4 criteria [65]. Variants of uncertain significance require segregation studies, phenotypic correlation and, sometimes, RNA studies; in the Istanbul series, the authors proposed variants of uncertain significance in collagen genes as causal in 5 of 36 families (13.9%), a proposal that does not amount to confirmation [90]. Some variants have conflicting classifications in ClinVar, such as the biallelic variants described by Girisha [8, 59].

5.7. Algorithm

Table 5.1. Practical diagnostic algorithm

CNV, copy-number variation.

5.8. Prenatal and preimplantation diagnosis

When the familial variant is known, prenatal diagnosis is made by COL2A1 analysis of chorionic villi or amniocytes (in the cases reviewed by Larissa, biopsy at 12 weeks and amniocentesis at 19 weeks), and preimplantation genetic testing is possible [6, 16, 64]. Without a known variant, prenatal diagnosis rests on ultrasound, possibly supplemented by fetal MRI, and on panel molecular testing of amniotic fluid or villi [6, 64]. The reproductive options are detailed in section 8.

6. Differential Diagnosis

6.1. Comparative table

Table 6.1 brings together the differentials of GeneReviews, supplemented by those of the Spranger and Smith atlases and the Oxford Desk Reference, ordered by the severity level of the collagenopathy with which they are confused.

Table 6.1. Differential diagnosis of the type II collagenopathies

6.2. Main differentials

The first differential problem is prenatal and neonatal: distinguishing achondrogenesis II from achondrogeneses IA and IB and from the other lethal dysplasias. Normal cranial ossification and ribs without fractures favour achondrogenesis II, thin ribs with fractures and a poorly ossified skull favour IA, and loss of longitudinal orientation of the tubular bones favours IB (Table 3.2) [6, 30, 39]. Thanatophoric dysplasia is distinguished by the acorn-shaped radiolucency of the proximal femur, and hypophosphatasia by the lack of cranial ossification and low alkaline phosphatase [6, 30].

Within the COL2A1 spectrum, the distinction between neighbouring entities rests on radiological signs and on time. SEDC differs from hypochondrogenesis only quantitatively; Torrance dysplasia has distinctive changes of the short tubular bones, but some patients resemble SEDC and can be distinguished only molecularly [16]. Kniest is distinguished from SEDC by the very broad and short femoral neck, the coronal clefts in infancy and the involvement of the tubular bones of the hands [14, 16]. SEMD Strudwick separates from SEDC only during the first year, when the metaphyseal irregularities appear; SEDC has smooth metaphyseal margins and less severe deformities of the spine and legs [6, 17]. Spondyloperipheral dysplasia is distinguished from Czech dysplasia because the metacarpals are also short [23].

Outside the spectrum, OSMED and Kniest have a similar early phenotype, but Kniest has more severe platyspondyly, tubular shortening and metaphyseal flaring, and ocular changes that do not occur in OSMED [101]. Mucopolysaccharidosis IV resembles SEDC clinically, but presents later, involves the hands and feet, and causes corneal opacities and keratan sulfaturia; the SDMC panel did not reach consensus (54.6%) on routinely testing for the mucopolysaccharidoses in patients with finger stiffness, but anterior vertebral beaking should prompt this investigation [9, 16, 82]. Progressive pseudorheumatoid dysplasia lacks the coxa valga and the anterior endplate defects of SED Stanescu type [24].

In the forms with average stature, the differential is rheumatological and orthopaedic. Czech dysplasia may meet the classification criteria for rheumatoid arthritis, and dominant inheritance, platyspondyly, sensorineural hearing loss and short metatarsals are the elements that set it apart [86]. Bilateral Perthes disease, especially if symmetrical and synchronous, in an asymptomatic child should prompt investigation for a dysplasia; in addition to type II collagenopathy, multiple epiphyseal dysplasia, pseudoachondroplasia and the mucopolysaccharidoses can also be mistaken for bilateral Perthes disease, and bilateral involvement of the femoral heads, even at different stages, warrants investigation [6, 9]. In familial Perthes disease with apparent dominant inheritance and incomplete penetrance, other familial causes of dysplastic femoral heads, such as multiple epiphyseal dysplasia, should be excluded [10]. Stickler syndrome type 1 is distinguished from types 2 to 5 by the membranous vitreous and the higher risk of detachment [6, 53].

7. Treatments and Risk Management Strategies

7.1. Principles and initial assessment

There is no treatment that corrects the collagen II defect; management is symptomatic, preventive and multidisciplinary, and rests on the 2019 SDMC consensus on type II collagenopathies, supplemented by the same group's consensus statements on the spine (2020), craniofacial aspects (2021), the perioperative period (2017) and pregnancy (2018) [9, 82, 91, 104, 105]. These guidelines are expert consensus statements developed by the modified RAND-UCLA Delphi method, without GRADE grading; the strength of each recommendation is expressed as the percentage of panel agreement [9]. The 2019 panel brought together 11 experts (six paediatric orthopaedic surgeons, four geneticists and one paediatric radiologist, with a mean of 20 years of experience and more than 500 patients followed); in the first round (September 2017), 12 of 26 statements reached consensus, and in the second 22 of 24, with consensus defined a priori as 80% [9]. Table 7.1 lists the assessment recommended after diagnosis.

Table 7.1. Initial assessment after diagnosis

7.2. Cervical spine

The SDMC consensus recommends routine imaging and surveillance of the cervical spine in SEDC and Kniest (100% agreement), with flexion–extension radiographs in all patients, with or without symptoms [9]. Dynamic radiographs are taken as soon as the child can cooperate, typically at 2–3 years; until then, surveillance is clinical and, if the neurological examination is abnormal, findings are inconclusive or os odontoideum is suspected, flexion–extension MRI is performed under sedation or anaesthesia, and dynamic computed tomography (CT) improves bone assessment [82]. In young children with delayed ossification of the upper cervical spine, flexion–extension MRI replaces radiography [9]. Without instability, reassessment is performed every 2–3 years or when symptoms occur; with instability and reduced space available for the cord (SAC), flexion–extension MRI is performed [9]. Mild instability, defined as movement of 5 mm or less, SAC greater than 10 mm, absence of myelomalacia and absence of symptoms, is kept under careful annual surveillance; spinal cord compression, with or without myelomalacia, is an indication for surgery [9]. The spine consensus adds that prophylactic C1–C2 arthrodesis is not indicated in the absence of spinal cord compression or myelopathy, that clinical myelopathy requires urgent assessment and management, that the anatomy of the vertebral artery and upper cervical spine is variable and requires advanced imaging before surgery, and that several stabilisation techniques are effective (all with 100% agreement) [82]. In progressive myelopathy, instability that reduces without residual compression is treated with instrumented fusion without decompression, and compression that persists after reduction requires decompression before fusion; instrumentation improves the fusion rate in children, and uninstrumented techniques with a halo are less effective but sometimes necessary [82]. Surgery should be performed by a specialist in rare skeletal dysplasias with spinal involvement [6]. The Terhal cohort recommends flexion–extension radiographs at 3 years and before any surgery, neurological referral and MRI if abnormal, and repetition at 7 years if there is doubt about stability [7]. Cervical kyphosis may progress and requires repeated reassessment, and stenosis may develop at any cervical level [82].

7.3. Kyphoscoliosis and the thoracolumbar spine

Progressive kyphoscoliosis requires clinical and radiological surveillance (100%) [9]. In progressive scoliosis, standing radiographs are taken every 6 months; in small non-progressive curves, annually; with symptoms or signs of compression, MRI of the whole spine; and before any lower-limb surgery in a child with kyphoscoliosis, MRI and consideration of spinal cord monitoring [9]. GeneReviews recommends clinical examination of the spine every 6–12 months depending on severity [6]. Treatment uses bracing for mild curves and serial casting if the curve progresses in young children with flexible scoliosis; in severe progressive scoliosis during growth, growing rods or other growth-preserving techniques are used, with definitive instrumentation and fusion at skeletal maturity [9, 82]. In progressive thoracolumbar kyphosis, with neurological deficit or refractory pain, stabilisation is indicated, its success depending on achieving sagittal alignment and balance; high thoracic kyphosis is associated with spinal cord injury during procedures under anaesthesia, advanced imaging is strongly recommended before instrumentation and the surgical complication rate is higher than in the general population [82]. Routine whole-spine MRI in the surveillance of patients with known vertebral disease did not reach consensus (70%) [82]. Increased lumbar lordosis is associated with hip flexion contracture, and hip realignment may improve sagittal alignment [82]. No drug has been shown to reduce vertebral disease in the dysplasias; the anticipatory measures are physiotherapy, weight reduction and appropriate non-steroidal anti-inflammatory drugs (NSAIDs) [82].

7.4. Airway, breathing and sleep

Significant respiratory problems are common and should be assessed at all ages (81.8%) [9]. Assessment may include physical examination, oximetry, bronchoscopy, spirometry, a sleep study and chest imaging; a sleep study with end-tidal carbon dioxide and continuous oximetry may reveal reduced pulmonary reserve [9]. Treatment may include surgery for Robin sequence, tracheostomy and mechanical ventilation, continuous positive airway pressure (CPAP), supplemental oxygen, bronchodilators and prophylactic vaccination against influenza and pneumococcus; central apnoea attributable to unrecognised cervical instability requires assessment and treatment of the spine [6, 9]. The SDMC craniofacial consensus recommends assessing signs of obstruction and sleep-disordered breathing at every visit, polysomnography in the presence of snoring or symptoms, adenoidectomy and/or tonsillectomy as first line in obstructive apnoea, polysomnography before adenotonsillectomy and overnight monitoring afterwards, non-invasive ventilation as an option, and imaging and/or laryngeal visualisation in the presence of stridor or hoarseness [91]. Adults of short stature should have lifelong follow-up by an adult respiratory physician, with monitoring of physical activity and lung capacity [9].

7.5. Eye

Routine ophthalmological assessment and surveillance are necessary (100%), by an experienced ophthalmologist, with dilated examination from diagnosis, because treatments are more effective before advanced disease or blindness [9]. Myopia is corrected with spectacles; the patient and family should know the symptoms of retinal detachment and seek immediate assessment [6]. Prophylactic retinopexy was the recommendation that did not reach consensus in the SDMC (72.7%): the panel recommends referring patients with Stickler syndrome and a COL2A1 variant to a centre that offers it, and the disagreement concerned making it routine in all patients with Stickler type I and extending it to the other type II collagenopathies [9]. The supporting data come from Stickler syndrome: in the Cambridge prophylactic cryotherapy protocol (487 patients followed over 36 years), the detachment rate was 9% in the 293 treated eyes and 53.6% in the 194 observed eyes, without long-term complications; in a series of 204 patients with Stickler type I, detachment occurred in 78% of observed eyes and in 8% and 10% after bilateral and unilateral cryotherapy; a systematic review considered cryotherapy and 360° laser effective compared with observation [9, 106]. The Oxford Desk Reference summarises the same protocol as an approximately tenfold risk reduction, and notes that, without prophylaxis, half of patients with Stickler type 1 have a detachment of the second eye within 4 years of the first [10]. GeneReviews and GeneGuidance support prophylaxis because surgical repair of established detachment in Stickler syndrome often fails and redetachment is common; in the series cited by the SDMC, complete reattachment was achieved in 78.6% of 30 eyes in one centre and in 100% of 16 eyes with multiple surgeries in another [4, 6, 9]. For SEDC and Kniest, no source consulted provides comparative data on prophylactic retinopexy; in the Kniest case of Hu (2026), prophylactic barrier laser was performed in the fellow eye (section 10) [65].

7.6. Hearing, cleft palate, ear, nose and throat and dental care

Routine hearing assessment and surveillance are necessary (100%), given the possibility of conductive hearing loss [9]. The craniofacial consensus recommends hearing assessment at birth or at diagnosis and at 5 years (84.6%), full audiological assessment in the presence of speech delay, suspected hearing loss or otitis (100%), tympanostomy tubes for unilateral or bilateral otitis media with effusion unlikely to resolve (type B tympanogram or duration of 3 months or more), investigation for persistent middle-ear disease in recurrent acute otitis media and treatment of acute otitis according to general-population guidelines; annual hearing assessment up to 5 years did not reach consensus (30.8%) [91]. Children with skeletal dysplasia should be considered «at risk» according to the American Academy of Otolaryngology-Head and Neck Surgery guideline on otitis media with effusion; in a screening programme of 112 individuals with dysplasias, more than 25% of children and more than 50% of adults failed hearing screening [91]. Hearing loss is treated with hearing aids and/or surgery when indicated [6]. Cleft palate is corrected surgically, and velopharyngeal insufficiency is monitored [6]. Palatal, midfacial, dental and mandibular anomalies are more prevalent in the dysplasias, and specialised dental and orthodontic care should begin in early childhood [91].

7.7. Hip, knee and lower limbs

Type II collagenopathy may mimic bilateral Perthes disease and present as early-onset arthritis (both with 100%) [9]. True osteonecrosis of the femoral head is treated like Perthes disease, with rest, mobilisation, analgesia and physiotherapy [9]. In disproportionate short stature, coxa vara and genu valgum are frequent (100%) [9]. Valgus-extension osteotomy with internal rotation improves gait in coxa vara (100%): in a 2008 series, support osteotomy with an external fixator improved pain, waddling gait, mobility and the Harris score; in a 2017 series in SEDC, valgus-extension osteotomy with a blade plate reduced hip flexion contracture from 16° to 3° and lumbar lordosis from 81° to 70°, with low recurrence, a benefit on sagittal alignment not demonstrated in achondroplasia [9, 82]. In severe genu valgum, guided growth (while growth remains) and realignment osteotomies are effective (90.9%); in a hemiepiphysiodesis series of 29 patients with dysplasias (four SED, two Kniest, one Stickler), all but one of the subgroup improved mechanical axis deviation, with failure in a 14-year-old patient with SED and an unknown recurrence risk; knee correction fails if not combined with hip correction [9]. Hip replacement is beneficial in adolescents and young adults with early arthritis (100%), after assessment of comorbidities, anatomical limitations and perioperative issues [9]. At the Hospital for Special Surgery in New York (Derocher 2026; 55 patients with dysplasias of various types, not only COL2A1; 103 arthroplasties, 75 hip and 28 knee), custom components were needed in 49% of hips and 54% of knees; implant survival was 86.6% (hip) and 88.7% (knee) at 5 years, 86.6% and 59.1% at 10 years and 57.7% for the hip at 20 years; postoperative complications affected 38.7% of hips and 53.6% of knees, and severity decreased after the introduction, in 2018, of a dedicated multidisciplinary preoperative clinic [107]. In the Terhal cohort, 54% of patients had undergone surgery, with femoral osteotomy the most frequent operation in childhood and hip replacement after the age of 20 years [7]. In Czech dysplasia, knee and hip pain in adolescence leads to early arthroplasty [23]. Clubfoot is corrected early [16].

7.8. Pain, rehabilitation, mobility and psychosocial support

Joint problems (laxity, contractures, pain due to early osteoarthritis) are treated with orthopaedics, physiotherapy, occupational therapy and analgesics, favouring low-impact activities such as swimming and cycling, and walking aids when necessary [6]. A mobility device for community distances should be considered (100%), such as a powered wheelchair or scooter, when early arthritis, pain and deformity limit walking; driving depends on cognitive readiness rather than age, and physical activity should be maintained as much as possible to avoid a sedentary lifestyle and obesity [9]. Obesity should be avoided even in patients of normal weight, with dietitian support [6]. Short stature, dysmorphism, low vision, hearing loss and pain warrant psychological support and contact with patient associations [6].

7.9. Pharmacological therapy

No medicine modifies the disease, and no source consulted gives specific doses for type II collagenopathies. Analgesia and NSAIDs are used according to the general indications and dosages for musculoskeletal pain, with no specific regimen described in the guidelines [6, 82]. Vaccination against influenza and pneumococcus and bronchodilators are part of respiratory treatment according to the general schedules and indications [9].

Growth hormone (somatropin) is the only drug with published data on stature. The SDMC consensus stated, with 100% agreement, that there is no evidence that growth hormone increases final stature; in the first round, 72.8% of experts disagreed that it was appropriate, and the evidence cited amounts to two patients with SED treated for one year by Kanazawa, with no change in stature or growth velocity [9]. Zhan (2025) described, in a series of seven Chinese children with SEDC, two patients treated with growth hormone who gained 0.76 SD in 3.5 years and 0.27 SD in 3 years, without significant adverse effects; these are two cases without a control group, without final-height data and with a dose not reported in the abstract consulted, and the authors themselves call for multicentre studies [108]. The two sources are not contradictory regarding final stature, but differ in their reading of benefit: somatropin treatment in SEDC is off-label use, and the available evidence does not support its use outside research.

Table 7.2. Regulatory status

7.10. Trials, studies and research

Table 7.3. Relevant trials and studies

Comparisons between these studies are not legitimate: ROCKET is observational and serves as a basis for future trials, the retinal prophylaxis data come from non-randomised cohorts with internal or historical controls, and the growth hormone series has two patients without a control [9, 108, 110]. The relative reduction in detachment in the Cambridge protocol should be read alongside the absolute rates (9% vs 53.6% over decades) and applies to Stickler syndrome, not to SEDC or Kniest [9]. The only registered randomised trial of retinal prophylaxis was withdrawn without enrolment [112]. A ClinicalTrials.gov search on 10 October 2026 with the terms COL2A1, spondyloepiphyseal dysplasia, Kniest, type II collagenopathy, achondrogenesis and Stickler identified no pharmacological therapeutic trial in type II collagenopathies, which does not prove that none exists in other registries [110].

Developments in the last 12 months (October 2025 to October 2026). The ClinGen curations classified seven COL2A1 gene–disease relationships as Definitive (scientific publication) [2]. Kenis proposed multiple ossification centres of the superior pubic ramus as a supporting sign for the diagnosis of SEDC (retrospective study) [94]. Larissa reviewed the prenatal spectrum of SEDC (review and two cases) [64]. Tüysüz showed a 33.3% diagnostic yield of exome sequencing in bilateral Perthes-type dysplasias (cohort) [90]. Derocher updated the results of arthroplasty in skeletal dysplasias (retrospective cohort) [107]. Ueki described a monitored term delivery in a woman with SEDC and Hu a retinal detachment in an adolescent with Kniest (case reports) [65, 93]. Yilmaz documented the attenuating effect of somatic mosaicism (case report with review) [66]. The OSC/SS trial began recruitment and the French cerclage trial was withdrawn (registries) [111, 112]. For INS-101, a gene therapy targeting the growth plate cartilage developed by Innoskel for SEDC and the type II collagenopathies, there are only proof-of-concept data in a mouse model disclosed by the company and press reports (Series A financing of 20 million euros in 2020); no official 2025–2026 press release or registered interventional trial was found [109]. The 2020 spine consensus mentions «several emerging therapies» in the dysplasias, none with demonstrated benefit on vertebral disease [82].

7.11. Situations to avoid, perioperative care, pregnancy and transition

Activities and manoeuvres to avoid. In cervical instability, extreme neck extension and flexion and contact sports should be avoided; the Terhal cohort discourages contact sports in all patients with SEDC or a related phenotype, because of the dual risk of retinal detachment and cervical instability, and recommends teaching the warning signs of myelopathy (frequent falls, changes in micturition or defecation, altered sensation) [6, 7]. Activities that overload the joints should be avoided [6].

Perioperative care and anaesthesia. Before general anaesthesia, the cervical spine should be assessed by imaging and the neck handled with care [6, 7]. The SDMC perioperative consensus (13 experts; 19 of 22 recommendations with consensus above 80%) addresses preoperative pulmonary assessment, polysomnography, cardiac and neurological assessment, cervical spine imaging and management of the difficult airway at intubation and extubation; only the abstract was accessible [104]. The spine consensus defines the «spine at risk» (foramen magnum stenosis, atlantoaxial instability, cervical stenosis or kyphosis, cervicothoracic kyphosis, thoracic stenosis, thoracolumbar kyphosis at the level of the cord, syrinx and cord signal changes on MRI), with the highest risk in the upper thoracic spine, and recommends considering neuromonitoring in all surgical procedures in these patients (100%), understood as applying to prolonged anaesthesia, above 1 hour, and not to short procedures such as tympanostomy tubes or endoscopy [82]. Before lower-limb surgery in a child with kyphoscoliosis, MRI should be performed and spinal cord monitoring considered [9]. Airway surgical risk is higher in the dysplasias [91]. In the case of Ueki (2026), caesarean section was performed under spinal anaesthesia after exclusion of tracheomalacia, and in the case of Hu (2026) the ophthalmic technique was chosen partly to shorten general anaesthesia given the respiratory risk [65, 93].

Pregnancy. In women with a small pelvis, caesarean section should be considered, and each case should be assessed by an obstetrician familiar with skeletal dysplasias [6]. The SDMC prenatal consensus (13 experts, 54 statements, 40 with final consensus) addresses the assessment and management of the pregnant woman with a dysplasia and of the fetus with a suspected dysplasia; only the abstract was accessible [105]. Ueki's review brought together five deliveries in women with SEDC (1991–2026; maternal heights between 100 and 134 cm), all by caesarean section, because of maternal dyspnoea or cephalopelvic disproportion; three of the four previous deliveries were preterm (35–36 weeks), two of them due to maternal respiratory compromise, and the authors hypothesise, on the basis of a very small number of cases, that delivery tends to occur earlier in mothers shorter than about 122 cm [93]. Monitoring of oxygen saturation in the third trimester, pelvimetry and airway assessment before anaesthesia were the elements of management in that case (section 10); the oxygen therapy threshold (peripheral oxygen saturation, SpO2, below 93%, target above 96%) was chosen by the authors, with no established threshold [93]. Bedeschi (2011) described the prenatal and delivery management of a mother with spondyloperipheral dysplasia and her affected son [85]. The sources consulted provide no data on lactation in these diseases.

Transition to adulthood. The adult of short stature continues follow-up with an adult respiratory physician, ophthalmological surveillance continues throughout life, and hip and knee osteoarthritis, with arthroplasty often before the age of 50 years, makes adult orthopaedics with experience in dysplasias a central element of the transition [7, 9]. Reproductive counselling should be offered to affected adolescents and young adults before pregnancy [6].

7.12. Guideline recommendations and divergences

Table 7.4. Final recommendations of the SDMC consensus on type II collagenopathies (2019)

Source: [9].

Table 7.5. Divergences between sources

8. Reproductive Options and Family Counselling

8.1. Dominant inheritance: risks by scenario

Most type II collagenopathies are autosomal dominant [6]. Most severe forms result from de novo variants, but the overall proportion of de novo cases is unknown; in the Terhal cohort, the variant was de novo in 51% of patients and inherited in 33%, including one case of maternal gonadal mosaicism [6, 7]. Many mild forms have an affected parent. In an isolated case, testing of both parents is recommended, because mildly affected relatives may go unnoticed [6]. Table 8.1 summarises the risks.

Table 8.1. Recurrence risks in the autosomal dominant forms

Sources: [6, 9, 77, 78].

Parental mosaicism is the reason why no recurrence should be considered nil after a de novo case. A parent with somatic and gonadal mosaicism may have a minimal phenotype, such as myopia, a Stickler phenotype or mild SED, and blood analysis may not detect it; exclusively gonadal mosaicism is not detectable in blood [6, 9, 45, 67]. Sibling recurrences of achondrogenesis II and hypochondrogenesis, Torrance and severe collagenopathies due to parental mosaicism have been published [30, 39, 68–71]. Individuals with somatic mosaicism for C-propeptide variants may have minimal bone changes and children with Torrance dysplasia [20]. Because the specific diagnosis tends to recur within the family, the expected severity in a child of an affected parent approximates that of the parent, with variable expressivity [6].

8.2. Recessive inheritance

In the recessive forms, the parents are presumed heterozygous, which should be confirmed by testing; if only one parent is a carrier, a de novo variant on the proband's other allele or post-zygotic parental mosaicism is considered [6, 113]. Siblings have a 25% probability of being affected, 50% of being heterozygous and 25% of not being carriers; heterozygotes are unaffected or mildly affected, and homozygotes resemble the proband with variable expressivity [6]. The offspring of a proband with a recessive form are obligate heterozygotes, unless the partner is a carrier [6]. In a severe or atypical presentation, both parents should be assessed, because the biallelic hypomorphic variants described (p.Pro458Leu, p.Arg1379Cys, p.Arg1133Cys) may go unnoticed as mild changes on a single allele [8, 64, 73].

8.3. Achondrogeneses: implications of the differential diagnosis

After a pregnancy with achondrogenesis, identification of the gene determines the risk: in achondrogenesis II, almost always de novo, the risk is low but not nil because of parental mosaicism; in achondrogeneses IA and IB, which are recessive, the risk is 25% in each pregnancy [1, 6, 39]. Molecular diagnosis in the fetus, on fetal tissue or stored DNA, is therefore essential for counselling about future pregnancies.

8.4. Cascade testing and minors

The genetic status of older or younger asymptomatic at-risk relatives should be clarified in order to start surveillance early [6]. In Stickler syndrome, predictive testing serves to guide prophylactic retinopexy, and all first-degree relatives should have a specialist eye assessment [10]. In families with COL2A1-related proximal femoral dysplasia, haplotype analysis and sequencing make it possible to identify carriers before symptoms appear [62]. In forms with ocular or cervical risk, testing of minors is justified by the direct benefit of surveillance; in mild adult-onset forms, the decision should weigh the benefit of surveillance against the minor's future autonomy.

8.5. Prenatal diagnosis, preimplantation diagnosis and family planning

The risk should be determined and discussed before pregnancy, and counselling should be offered to affected or at-risk young adults [6]. Once the variant has been identified, prenatal diagnosis by chorionic villus sampling or amniocentesis and preimplantation genetic testing are possible [6, 16]. In the Larissa review, two pregnancies with a family history and molecular confirmation were terminated and five resulted in live births [64]. In the Ueki case, maternal genetic diagnosis during pregnancy made it possible to communicate the 50% fetal risk and to anticipate SEDC in the newborn [93]. In Stickler syndrome, prenatal diagnosis is technically possible but rarely requested [10].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

The prevalence of the type II collagenopathies as a group is unknown [6, 9]. Table 9.1 summarises the available data and what each one measures.

Table 9.1. Epidemiological data

The Stickler syndrome figures are not COL2A1 prevalences, because they include the types caused by other genes, and the case counts for the rare forms measure the published literature rather than population frequency [4, 12]. Case counts probably underestimate the mild forms, which are mistaken for common osteoarthritis, rheumatoid arthritis or Perthes disease [86, 90]. No Orphanet prevalence data were obtained for achondrogenesis IA and odontochondrodysplasia.

9.2. Geographical distribution and populations studied

The reference cohorts are European and Asian. The Terhal cohort (93 patients) was assembled in centres in the Netherlands, Belgium, France and the United Kingdom; the Barat-Houari series (136 probands) is French; the Nishimura series (56 families) is Japanese [7, 52, 57]. The families with avascular necrosis of the femoral head due to p.Gly1170Ser and p.Gly717Ser were described in Taiwan and mainland China, and COL2A1 variants in familial avascular necrosis are concentrated in the East Asian literature; the p.Gly1170Ser variant arose in two Taiwanese families with different haplotypes, which suggests independent occurrence rather than a single founder effect [9, 27, 62]. The series of bilateral Perthes-type dysplasias is Turkish and the series on pubic ossification is Russian [90, 94]. In the USA, the most recent data are the arthroplasty data from the Hospital for Special Surgery, which pool several dysplasias [107]. No COL2A1 founder variants were identified in the sources consulted; the recurrent variants (p.Arg275Cys in Czech dysplasia, p.Arg989Cys in SEDC) occur in different populations [22, 60, 93].

In Brazil, the skeletal dysplasia group of the State University of Campinas described six patients with SEDC due to the recurrent p.(Arg989Cys) variant, with clinical and radiological follow-up, and a family from Brasília with 12 carriers of the Czech dysplasia variant, half of whom met criteria for rheumatoid arthritis [60, 86]. The sources consulted contain no data from Portugal on type II collagenopathies, which does not demonstrate their absence from the literature.

9.3. Special populations

Pregnant women with SEDC are at risk of respiratory compromise in the third trimester and have a narrow pelvis, which led to caesarean section in all the published deliveries reviewed by Ueki [93]. Young children with delayed cervical ossification require dynamic MRI instead of radiography [9]. Adults of average stature with mild forms often reach rheumatology or orthopaedics without a diagnosis, and the sensorineural hearing loss of Czech dysplasia may go unnoticed until adulthood [23, 86]. The literature on adults of average stature is limited; the experience of the SDMC panel suggests fewer respiratory problems in this group [9].

9.4. Reference networks and patient associations

The relevant European reference network is ERN BOND (European Reference Network on Rare Bone Diseases, https://ernbond.eu); no ERN BOND guideline specific to type II collagenopathies was found, and the reference guideline remains that of the SDMC [9]. The Skeletal Dysplasia Management Consortium (skeletaldysplasia.org) and the International Skeletal Dysplasia Registry of the University of California, Los Angeles, are clinical and research resources [6]. Patient associations include, in Portugal, ANDO Portugal (Associação Nacional de Displasias Ósseas, https://www.andoportugal.org), with branches in Madeira and the Azores and a section for the Portuguese-speaking African countries; in the USA, Little People of America (https://www.lpaonline.org) and Stickler Involved People (https://www.stickler.org); in the United Kingdom, Little People UK (https://littlepeopleuk.org), with a page dedicated to SEDC, the Restricted Growth Association (https://restrictedgrowth.co.uk) and Stickler Syndrome UK; and, in Australia, Short Statured People of Australia [6].

10. Illustrative Clinical Case

10.1. Term pregnancy in a woman with SEDC and transmission to her son (Ueki 2026)

Ueki and colleagues, from Shinshu University (Japan), described a 30-year-old primigravida 128 cm tall (below the 1st centile) and weighing 37 kg, with a clinical diagnosis of SEDC since the age of 8 years and parents of normal stature [93]. She had mild myopia without retinal detachment or vitreous anomalies, and had no cleft palate or hypermobility. The pregnancy was spontaneous; the patient was referred at 11 weeks and, after counselling, genetic testing at 13 weeks identified the heterozygous variant NM_001844.5:c.2965C>T p.(Arg989Cys), the recurrent SEDC variant known in historical helical numbering as R789C, and the 50% fetal risk was communicated [59, 93]. At 29 weeks, the fetal long bones were close to the 1st centile and the chest was barrel-shaped, with no platyspondyly visible on ultrasound. During the third trimester, continuous oxygen saturation monitoring was performed, and pelvimetry showed a narrow pelvis (transverse diameter of the pelvic inlet 11.1 cm and obstetric conjugate 9.58 cm), so an elective caesarean section was planned [93].

The patient was admitted at 33 weeks. At 34 weeks, asymptomatic nocturnal desaturation down to 91% appeared; contrast-enhanced CT and echocardiography excluded pulmonary thromboembolism and cardiac dysfunction, and the hypoxia was attributed to compression of the lungs by the uterus, treated with nocturnal oxygen at 1 L/min, with saturation above 96% [93]. Caesarean section was performed at 37+3 weeks under spinal anaesthesia, after an assessment that excluded tracheomalacia. The male newborn weighed 2500 g (26.5th centile) and measured 46.0 cm (22.4th centile), with Apgar scores of 8 and 8; he required intubation and ventilation for 4 days because of thoracic hypoplasia, no longer needed oxygen by day 16 and was discharged on day 29. The neonatal radiographs (Figure 2 of the article, https://pmc.ncbi.nlm.nih.gov/articles/PMC13371809/figure/jog70409-fig-0002/) show flattened vertebrae with pear-shaped lumbar bodies (panel A), maxillary hypoplasia (B), hypoplasia of the rib cage (C), metaphyseal flaring of the distal humerus (D) and shortening of the long bones with metaphyseal flaring of the distal femur (E) [93]. Automated auditory evoked potentials were normal bilaterally, and the newborn carried the same variant, confirming SEDC. The mother came off oxygen on the first postoperative day and was discharged on day 7 [93].

The case brings together several points of this article: maternal molecular diagnosis turned a theoretical risk into an anticipated fetal diagnosis; the narrow pelvis and third-trimester respiratory compromise determined the route and timing of delivery, as in the four previous published deliveries in women with SEDC; airway assessment and neuraxial anaesthesia followed the perioperative concerns of the dysplasias; and the newborn illustrated neonatal respiratory compromise due to a small chest, present in 26% of the Terhal cohort [7, 93]. The authors acknowledge that they did not perform pulmonary function tests or blood gas analysis, and the oxygen therapy threshold they used is not established in any guideline; their management is an example and not a protocol [93]. Assessment of the maternal cervical spine before anaesthesia, recommended by GeneReviews before general anaesthesia, is not described in the report [6, 93].

10.2. Retinal detachment in an adolescent with Kniest dysplasia (Hu 2026)

Hu and colleagues (Foshan, China) described a 17-year-old boy measuring 139 cm (below −5 SD), weighing 75 kg and with bilateral myopia of −12.0 dioptres, who presented with painless visual loss in the left eye for 4 days, with no family history [65]. Corrected visual acuity was 20/30 in the right eye and 20/1000 in the left; the right eye had lattice degeneration and the left a rhegmatogenous detachment with proliferative vitreoretinopathy extending to the macula, inferotemporal atrophic holes and subretinal proliferation. He had brachydactyly, and CT showed scoliosis, platyspondyly, endplate irregularities, horizontal acetabula with degenerative changes and a narrow pelvis; pulmonary function tests showed a moderate restrictive pattern (he slept sitting up), and audiometry, tympanometry and echocardiography were normal. Trio exome sequencing identified the de novo variant c.4121A>G p.(Gln1374Arg), absent from population databases, classified by the authors as pathogenic (PS2, PM2, PP3, PP4); the location in the C-propeptide is atypical for Kniest, whose usual mechanism is exon skipping [2, 65]. He underwent prophylactic barrier laser in the right eye and urgent scleral buckling in the left (external drainage, cryotherapy, encircling band and silicone oil tamponade), a technique chosen partly to shorten general anaesthesia given the respiratory risk; he declined polysomnography, orthopaedic assessment and vitrectomy. At 6 months, acuity was 20/30 in the right eye and 20/400 in the left, with a partially detached retina, and he continued annual retinal screening [65].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis by entity

Table 11.1. Prognosis by entity

In the Terhal cohort, dominated by SEDC, joint pain began at a median of 8.8 years, 54% of patients underwent surgery (first surgery at a median of 10.5 years) and 10 of 93 used a wheelchair permanently or daily; retinal detachment occurred in 12%, at a median of 14 years, and atlantoaxial instability in 28% of the 18 patients with dynamic radiographs [7]. These figures describe patients followed in specialised centres at a time when cervical and ophthalmological surveillance was already practised; they do not represent untreated natural history or lifetime risks. The sources consulted contain no mortality or life-expectancy data for the non-lethal forms. Children do not have intellectual disability, and functional prognosis depends mainly on the spine, hip, vision and hearing [9]. The prognosis for each family is guided by the specific diagnosis, which tends to recur among relatives [6].

11.2. Surveillance

Table 11.2 consolidates the surveillance recommendations. Where the sources diverge, both are given (see also Table 7.5).

Table 11.2. Surveillance plan

The intervals coincide with those in section 7: cervical imaging every 2–3 years without instability and annually in mild instability; spinal radiographs every 6 or 12 months depending on progression; annual ophthalmological examination after the first year; hearing assessment at diagnosis and at 5 years, adjusted to severity.

12. Summary and Final Considerations

The type II collagenopathies are an allelic family of chondrodysplasias caused by variants of COL2A1, the gene for the α1 chain of collagen II, and extend from prenatally lethal achondrogenesis to early hip osteoarthritis in an adult of normal stature. Severity depends on the mechanism: dominant-negative glycine substitutions in the helical domain cause the severe forms (achondrogenesis II, hypochondrogenesis, SEDC, SEMD); exon skipping causes Kniest dysplasia; C-propeptide variants cause Torrance and spondyloperipheral dysplasias; haploinsufficiency causes Stickler syndrome type 1; and arginine-to-cysteine substitutions give non-lethal, site-specific phenotypes, such as Czech dysplasia due to p.(Arg275Cys). ClinGen classifies seven COL2A1 gene–disease relationships as Definitive.

Inheritance is dominant, with a high proportion of de novo cases in the severe forms, and parental mosaicism (6–10% of parents of apparently de novo cases) means that the recurrence risk cannot be considered nil. There are rare recessive forms due to biallelic hypomorphic variants. Achondrogenesis II is distinguished from achondrogeneses IA (TRIP11) and IB (SLC26A2), which are recessive with a 25% risk, by normal cranial ossification, ribs without fractures and preserved orientation of the tubular bones. COL2A1-related proximal femoral epiphyseal dysplasia shows that the same gene can produce Perthes disease, avascular necrosis or hip osteoarthritis depending on age, and bilateral, symmetrical and synchronous Perthes disease should prompt investigation for a dysplasia.

Diagnosis rests on the radiological survey, ophthalmological and audiological examination and molecular confirmation, bearing in mind that sequencing does not detect exon deletions, deep intronic variants or low-level mosaicism. There is no disease-modifying treatment or authorised medicine; growth hormone has not shown a final height gain. Preventive management, systematised by the SDMC consensus, focuses on the cervical spine (flexion–extension radiographs from 2–3 years and MRI when indicated), kyphoscoliosis, the airway and sleep, regular ophthalmological examination and education about the symptoms of retinal detachment, hearing, hip and knee reconstruction and early arthroplasty. Anaesthesia and pregnancy require planning with cervical imaging, airway assessment and teams experienced in dysplasias. Therapeutic research is at the preclinical and natural-history stage, and the main gaps are the absence of population data on prevalence and mortality, the scarcity of data from Portugal and the lack of comparative trials of retinal prophylaxis outside Stickler syndrome.

13. References

1. 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. PMID: 36779427. PMC10081954.
2. Webb RF, McCurry H, Girod A, Hughes M, Wilcox E, Patel M, et al. Evidence-based classification of genes implicated in skeletal disorders using the ClinGen curation framework. J Bone Miner Res. 2026;41(7):717-728. doi:10.1093/jbmr/zjaf183. PMID: 41313243. PMC13192282.
3. Orphanet. Type 2 collagen-related bone disorder (ORPHA:93421). https://www.orpha.net/en/disease/detail/93421
4. GeneGuidance. COL2A1 – gene record and clinical curation [database; version of 26 September 2026].
5. 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. Service: https://rest.variantvalidator.org (GRCh38, NM_001844.5 MANE Select; accessed 2026-10-10).
6. Gregersen PA, Savarirayan R. Type II Collagen Disorders Overview. 2019 Apr 25 [updated 2024 Oct 24]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK540447/
7. Terhal PA, Nievelstein RJ, Verver EJ, Topsakal V, van Dommelen P, Hoornaert K, et al. A study of the clinical and radiological features in a cohort of 93 patients with a COL2A1 mutation causing spondyloepiphyseal dysplasia congenita or a related phenotype. Am J Med Genet A. 2015;167A(3):461-75. doi:10.1002/ajmg.a.36922. PMID: 25604898.
8. Girisha KM, Bhavani GS, Shah H, Moirangthem A, Shukla A, Kim OH, et al. Biallelic variants p.Arg1133Cys and p.Arg1379Cys in COL2A1: Further delineation of phenotypic spectrum of recessive Type 2 collagenopathies. Am J Med Genet A. 2020;182(2):338-347. doi:10.1002/ajmg.a.61414. PMID: 31755234.
9. Savarirayan R, Bompadre V, Bober MB, Cho TJ, Goldberg MJ, Hoover-Fong J, et al. Best practice guidelines regarding diagnosis and management of patients with type II collagen disorders. Genet Med. 2019;21(9):2070-2080. doi:10.1038/s41436-019-0446-9. PMID: 30696995.
10. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 528–529.
11. European Medicines Agency. Medicines output: orphan designations. https://www.ema.europa.eu/en/documents/report/medicines-output-orphan_designations-json-report_en.json (accessed 2026-10-10; 3310 records; none for COL2A1/SEDC/Kniest/achondrogenesis/Stickler).
12. Orphanet. Orphadata: Epidemiological data of rare diseases (en_product9_prev.xml). https://www.orphadata.com/data/xml/en_product9_prev.xml (accessed 2026-10-10).
13. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 68–71 (4.2 Platyspondylic dysplasia, Torrance type).
14. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 85–89 (4.5 Kniest dysplasia).
15. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 506–507.
16. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 72–79 (4.3 Spondyloepiphyseal dysplasia congenita).
17. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 80–84 (4.4 Spondylo-epi-metaphyseal dysplasia, Strudwick type).
18. Merrick B, Calder A, Wakeling E. Dysspondyloenchondromatosis (DSC) associated with COL2A1 mutation: clinical and radiologic overlap with spondyloepimetaphyseal dysplasia-Strudwick type (SEMD-S). Am J Med Genet A. 2015;167A:3103-7. doi:10.1002/ajmg.a.37282. PMID: 26250472.
19. Orphanet. Dysspondyloenchondromatosis (ORPHA:85198). https://www.orpha.net/en/disease/detail/85198
20. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 92–94 (4.7 Spondyloperipheral dysplasia).
21. Zankl A, Zabel B, Hilbert K, Wildhardt G, Cuenot S, Xavier B, et al. Spondyloperipheral dysplasia is caused by truncating mutations in the C-propeptide of COL2A1. Am J Med Genet A. 2004;129A(2):144-8. doi:10.1002/ajmg.a.30222. PMID: 15316962.
22. Hoornaert KP, Marik I, Kozlowski K, Cole T, Le Merrer M, Leroy JG, et al. Czech dysplasia metatarsal type: another type II collagen disorder. Eur J Hum Genet. 2007;15(12):1269-75. doi:10.1038/sj.ejhg.5201913. PMID: 17726487.
23. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 95–96 (4.8 Spondyloepiphyseal dysplasia with short metatarsals).
24. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 90–91 (4.6 Spondyloepiphyseal dysplasia, Stanescu type).
25. Kannu P, Bateman JF, Randle S, Cowie S, du Sart D, McGrath S, et al. Premature arthritis is a distinct type II collagen phenotype. Arthritis Rheum. 2010;62:1421-30. doi:10.1002/art.27354. PMID: 20131279.
26. Orphanet. Legg-Calvé-Perthes disease (ORPHA:2380). https://www.orpha.net/en/disease/detail/2380
27. Su P, Li R, Liu S, Zhou Y, Wang X, Patil N, et al. Age at onset-dependent presentations of premature hip osteoarthritis, avascular necrosis of the femoral head, or Legg-Calvé-Perthes disease in a single family, consequent upon a p.Gly1170Ser mutation of COL2A1. Arthritis Rheum. 2008;58(6):1701-6. doi:10.1002/art.23491. PMID: 18512791.
28. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 228–229 (7.18 Spondylometaphyseal dysplasia, corner fracture type).
29. Smits P, Bolton AD, Funari V, Hong M, Boyden ED, Lu L, et al. Lethal skeletal dysplasia in mice and humans lacking the golgin GMAP-210. N Engl J Med. 2010;362(3):206-16. doi:10.1056/NEJMoa0900158. PMID: 20089971. PMC3108191.
30. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 65–67 (4.1 Achondrogenesis II, hypochondrogenesis).
31. Wehrle A, Witkos TM, Unger S, Schneider J, Follit JA, Hermann J, et al. Hypomorphic mutations of TRIP11 cause odontochondrodysplasia. JCI Insight. 2019;4(3):e124701. doi:10.1172/jci.insight.124701. PMID: 30728324. PMC6413787.
32. Kniest W. Zur Abgrenzung der Dysostosis enchondralis von der Chondrodystrophie. Z Kinderheilkd. 1952;70:633-40.
33. Spranger J, Winterpacht A, Zabel B. Kniest dysplasia: Dr. W. Kniest, his patient, the molecular defect. Am J Med Genet. 1997;69(1):79-84. doi:10.1002/(sici)1096-8628(19970303)69:1<79::aid-ajmg15>3.0.co;2-l. PMID: 9066888.
34. Spranger J, Wiedemann HR. Dysplasia spondyloepiphysaria congenita. Helv Paediatr Acta. 1966;21:598.
35. Spranger JW, Langer LO Jr. Spondyloepiphyseal dysplasia congenita. Radiology. 1970;94(2):313-22. doi:10.1148/94.2.313. PMID: 5412797.
36. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 502–505.
37. Langer LO Jr, Spranger JW, Greinacher I, Herdman RC. Thanatophoric dwarfism. A condition confused with achondroplasia in the neonate, with brief comments on achondrogenesis and homozygous achondroplasia. Radiology. 1969;92(2):285-94. doi:10.1148/92.2.285. PMID: 4885523.
38. Saldino RM. Lethal short-limbed dwarfism: achondrogenesis and thanatophoric dwarfism. Am J Roentgenol Radium Ther Nucl Med. 1971;112(1):185-97. doi:10.2214/ajr.112.1.185. PMID: 5582025.
39. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 462–463.
40. Anderson CE, Sillence DO, Lachman RS, et al. Spondylometepiphyseal dysplasia, Strudwick type. Am J Med Genet. 1982;13:243-56. doi:10.1002/ajmg.1320130304. PMID: 6817637.
41. Tiller GE, Polumbo PA, Weis MA, Bogaert R, Lachman RS, Cohn DH, et al. Dominant mutations in the type II collagen gene, COL2A1, produce spondyloepimetaphyseal dysplasia, Strudwick type. Nat Genet. 1995;11(1):87-9. doi:10.1038/ng0995-87. PMID: 7550321.
42. Kozlowski K, Marik I, Marikova O, Zemkova D, Kuklik M. Czech dysplasia metatarsal type. Am J Med Genet A. 2004;129A(1):87-91. doi:10.1002/ajmg.a.30132. PMID: 15266623.
43. Lee B, Vissing H, Ramirez F, Rogers D, Rimoin D. Identification of the molecular defect in a family with spondyloepiphyseal dysplasia. Science. 1989;244(4907):978-80. doi:10.1126/science.2543071. PMID: 2543071.
44. Vissing H, D'Alessio M, Lee B, Ramirez F, Godfrey M, Hollister DW. Glycine to serine substitution in the triple helical domain of pro-alpha 1 (II) collagen results in a lethal perinatal form of short-limbed dwarfism. J Biol Chem. 1989;264(31):18265-7. PMID: 2572591.
45. Winterpacht A, Hilbert M, Schwarze U, Mundlos S, Spranger J, Zabel BU. Kniest and Stickler dysplasia phenotypes caused by collagen type II gene (COL2A1) defect. Nat Genet. 1993;3(4):323-6. doi:10.1038/ng0493-323. PMID: 7981752.
46. Spranger J, Winterpacht A, Zabel B. The type II collagenopathies: a spectrum of chondrodysplasias. Eur J Pediatr. 1994;153:56-65. doi:10.1007/BF01959208. PMID: 8157027.
47. Mortier GR, Wilkin DJ, Wilcox WR, Rimoin DL, Lachman RS, Eyre DR, et al. A radiographic, morphologic, biochemical and molecular analysis of a case of achondrogenesis type II resulting from substitution for a glycine residue (Gly691-->Arg) in the type II collagen trimer. Hum Mol Genet. 1995;4(2):285-8. doi:10.1093/hmg/4.2.285. PMID: 7757081.
48. Nishimura G, Nakashima E, Mabuchi A, Shimamoto K, Shimamoto T, Shimao Y, et al. Identification of COL2A1 mutations in platyspondylic skeletal dysplasia, Torrance type. J Med Genet. 2004;41:75-9. doi:10.1136/jmg.2003.013722. PMID: 14729840.
49. Zankl A, Neumann L, Ignatius J, et al. Dominant negative mutations in the C-propeptide of COL2A1 cause platyspondylic lethal skeletal dysplasia, Torrance type, and define a novel subfamily within the type 2 collagenopathies. Am J Med Genet A. 2005;133A:61-7. doi:10.1002/ajmg.a.30531. PMID: 15643621.
50. Orphanet. Spondylometaphyseal dysplasia, Schmidt type (ORPHA:93316). https://www.orpha.net/en/disease/detail/93316
51. Orphanet. Spondyloepiphyseal dysplasia, Stanescu type (ORPHA:459051). https://www.orpha.net/en/disease/detail/459051
52. Nishimura G, Haga N, Kitoh H, Tanaka Y, Sonoda T, Kitamura M, et al. The phenotypic spectrum of COL2A1 mutations. Hum Mutat. 2005;26(1):36-43. doi:10.1002/humu.20179. PMID: 15895462.
53. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 97–100 (4.9 Stickler dysplasia).
54. Snead MP, McNinch AM, Poulson AV, et al. Stickler syndrome, ocular-only variants and a key diagnostic role for the ophthalmologist. Eye (Lond). 2011;25:1389-400. doi:10.1038/eye.2011.201. PMID: 21921955.
55. Bell DM, Leung KK, Wheatley SC, Ng LJ, Zhou S, Ling KW, et al. SOX9 directly regulates the type-II collagen gene. Nat Genet. 1997;16:174-8. doi:10.1038/ng0697-174. PMID: 9171829.
56. Richards AJ, Laidlaw M, Whittaker J, Treacy B, Rai H, Bearcroft P, et al. High efficiency of mutation detection in type 1 Stickler syndrome using a two-stage approach: vitreoretinal assessment coupled with exon sequencing for screening COL2A1. Hum Mutat. 2006;27(7):696-704. doi:10.1002/humu.20347. PMID: 16752401.
57. Barat-Houari M, Dumont B, Fabre A, Them FT, Alembik Y, Alessandri JL, et al. The expanding spectrum of COL2A1 gene variants IN 136 patients with a skeletal dysplasia phenotype. Eur J Hum Genet. 2016;24(7):992-1000. doi:10.1038/ejhg.2015.250. PMID: 26626311. PMC5070901.
58. Hoornaert KP, Dewinter C, Vereecke I, Beemer FA, Courtens W, Fryer A, et al. The phenotypic spectrum in patients with arginine to cysteine mutations in the COL2A1 gene. J Med Genet. 2006;43(5):406-13. doi:10.1136/jmg.2005.035717. PMID: 16155195. PMC2564515.
59. National Center for Biotechnology Information. ClinVar [database]. Bethesda (MD): NLM. VCV000017368, VCV000017366, VCV000547252, VCV000017394, VCV000017393, VCV000017372, VCV000439509, VCV000451774, VCV001679885, VCV004888334. https://www.ncbi.nlm.nih.gov/clinvar/ (accessed 2026-10-10).
60. Silveira KC, Bonadia LC, Superti-Furga A, Bertola DR, Jorge AA, Cavalcanti DP. Six additional cases of SEDC due to the same and recurrent R989C mutation in the COL2A1 gene--the clinical and radiological follow-up. Am J Med Genet A. 2015;167A(4):894-901. doi:10.1002/ajmg.a.36954. PMID: 25735649.
61. Terhal PA, van Dommelen P, Le Merrer M, Zankl A, Simon ME, Smithson SF, et al. Mutation-based growth charts for SEDC and other COL2A1 related dysplasias. Am J Med Genet C Semin Med Genet. 2012;160C(3):205-16. doi:10.1002/ajmg.c.31332. PMID: 22791362.
62. Liu YF, Chen WM, Lin YF, Yang RC, Lin MW, Li LH, et al. Type II collagen gene variants and inherited osteonecrosis of the femoral head. N Engl J Med. 2005;352(22):2294-301. doi:10.1056/NEJMoa042480. PMID: 15930420.
63. Jurgens J, Sobreira N, Modaff P, et al. Novel COL2A1 variant (c.619G>A, p.Gly207Arg) manifesting as a phenotype similar to progressive pseudorheumatoid dysplasia and spondyloepiphyseal dysplasia, Stanescu type. Hum Mutat. 2015;36:1004-8. doi:10.1002/humu.22839. PMID: 26183434.
64. Larissa LR, Felipe LL, Salvador JC, Manuel GJ, Jose RM, Angelica CD, et al. Prenatal spectrum of COL2A1-related spondyloepiphyseal dysplasia congenita: a review and two case reports. Prenat Diagn. 2026;46(10):1637-47. doi:10.1002/pd.70201. PMID: 42363587.
65. Hu X, Li P, Liang W, Peng W, Kong X. Rhegmatogenous retinal detachment in an adolescent with Kniest Dysplasia: A case report. Am J Ophthalmol Case Rep. 2026;42:102502. doi:10.1016/j.ajoc.2025.102502. PMID: 41732752. PMC12925474.
66. Yilmaz RST, Cruz L, Robinson C, Bale AE, Carpenter TO. Somatic Mosaicism Modulating Clinical Severity in COL2A1-Related Skeletal Dysplasia: A Case of Profound Skeletal Asymmetry. Calcif Tissue Int. 2026;117(1). doi:10.1007/s00223-026-01599-3. PMID: 42758201.
67. Spranger J, Menger H, Mundlos S, Winterpacht A, Zabel B. Kniest dysplasia is caused by dominant collagen II (COL2A1) mutations: parental somatic mosaicism manifesting as Stickler phenotype and mild spondyloepiphyseal dysplasia. Pediatr Radiol. 1994;24(6):431-5. doi:10.1007/BF02011911. PMID: 7700721.
68. Morrison PJ, Znaczko A, Jansson M. Paternal somatogonadal COL2A1 mosaicism causing recurrence of severe type 2 collagenopathy. Am J Med Genet A. 2020;182:2191-4. doi:10.1002/ajmg.a.61763. PMID: 32648279.
69. Yamamoto K, Kubota T, Takeyari S, Kitaoka T, Miyata K, Nakano Y, et al. Parental somatogonadal COL2A1 mosaicism contributes to intrafamilial recurrence in a family with type 2 collagenopathy. Am J Med Genet A. 2020;182(3):454-60. doi:10.1002/ajmg.a.61422. PMID: 31854518.
70. Nagendran S, Richards AJ, McNinch A, Sandford RN, Snead MP. Somatic mosaicism and the phenotypic expression of COL2A1 mutations. Am J Med Genet A. 2012;158A:1204-7. doi:10.1002/ajmg.a.35303. PMID: 22496037.
71. Okamoto T, Nagaya K, Asai H, Tsuchida E, Nohara F, Hayashi T, et al. Platyspondylic lethal dysplasia Torrance type with a heterozygous mutation in the triple helical domain of COL2A1 in two sibs from phenotypically normal parents. Am J Med Genet A. 2012;158A:1953-6. doi:10.1002/ajmg.a.35509. PMID: 22711552.
72. Tham E, Nishimura G, Geiberger S, Horemuzova E, Nilsson D, Lindstrand A, et al. Autosomal recessive mutations in the COL2A1 gene cause severe spondyloepiphyseal dysplasia. Clin Genet. 2015;87:496-8. doi:10.1111/cge.12466. PMID: 25060605.
73. Barat-Houari M, Baujat G, Tran Mau Them F, Fabre A, Geneviève D, Touitou I. Confirmation of autosomal recessive inheritance of COL2A1 mutations in spondyloepiphyseal dysplasia congenita: Lessons for genetic counseling. Am J Med Genet A. 2016;170A(1):263-5. doi:10.1002/ajmg.a.37374. PMID: 26358419.
74. Al-Sannaa NA, Hoornaert KP, Van Laer L, Al-Abdulwahed HY, Mortier G. Spondylo-epiphyseal dysplasia in two sibs due to a homozygous splicing variant in COL2A1. Eur J Med Genet. 2020;63:104059. doi:10.1016/j.ejmg.2020.104059. PMID: 32896647.
75. Zhang Q, Yao R, Li Q, Li X, Feng B, Chang G, et al. A novel homozygous variant of COL2A1 in a Chinese male with type II collagenopathy: a case report. BMC Med Genomics. 2021;14:201. doi:10.1186/s12920-021-01048-0. PMID: 34380476.
76. Tüysüz B, Kasap B, Sarıtaş M, Alkaya DU, Bozlak S, Kıykım A, et al. Natural history and genetic spectrum of the Turkish metaphyseal dysplasia cohort, including rare types caused by biallelic COL10A1, COL2A1, and LBR variants. Bone. 2023;167:116614. doi:10.1016/j.bone.2022.116614. PMID: 36400164.
77. Unger S, Korkko J, Krakow D, Lachman RS, Rimoin DL, Cohn DH. Double heterozygosity for pseudoachondroplasia and spondyloepiphyseal dysplasia congenita. Am J Med Genet. 2001;104:140-6. doi:10.1002/ajmg.10062. PMID: 11746045.
78. Flynn MA, Pauli RM. Double heterozygosity in bone growth disorders: four new observations and review. Am J Med Genet A. 2003;121A:193-208. doi:10.1002/ajmg.a.20143. PMID: 12923858.
79. Castori M, Brancati F, Scanderbeg AC, Dallapiccola B. Hypochondrogenesis. Pediatr Radiol. 2006;36:460-1. doi:10.1007/s00247-005-0068-z. PMID: 16432703.
80. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 784–787.
81. Neumann L, Kunze J, Uhl M, et al. Survival to adulthood and dominant inheritance of platyspondylic skeletal dysplasia, Torrance-Luton type. Pediatr Radiol. 2003;33:786-90. doi:10.1007/s00247-003-1055-x. PMID: 12961049.
82. White KK, Bober MB, Cho TJ, Goldberg MJ, Hoover-Fong J, Irving M, et al. Best practice guidelines for management of spinal disorders in skeletal dysplasia. Orphanet J Rare Dis. 2020;15(1):161. doi:10.1186/s13023-020-01415-7. PMID: 32580780. PMC7313125.
83. Sergouniotis PI, Fincham GS, McNinch AM, Spickett C, Poulson AV, Richards AJ, Snead MP. Ophthalmic and molecular genetic findings in Kniest dysplasia. Eye (Lond). 2015;29:475-82. doi:10.1038/eye.2014.334. PMID: 25592122.
84. Walter K, Tansek M, Tobias ES, Ikegawa S, Coucke P, Hyland J, et al. COL2A1-related skeletal dysplasias with predominant metaphyseal involvement. Am J Med Genet A. 2007;143A:161-7. doi:10.1002/ajmg.a.31516. PMID: 17163530.
85. Bedeschi MF, Bianchi V, Gentilin B, et al. Prenatal manifestation and management of mother and child affected by spondyloperipheral dysplasia with a C-propeptide mutation in COL2A1. Orphanet J Rare Dis. 2011;6:7. doi:10.1186/1750-1172-6-7. PMID: 21356074.
86. Moreira LA, Carvalho DR, Santos SCL, Silva CCE, Ferreira BSA, Cunha BMD, et al. Czech dysplasia mimicking rheumatoid arthritis: Case series and literature review. Mod Rheumatol. 2024;34(4):705-710. doi:10.1093/mr/road070. PMID: 37489771.
87. Tzschach A, Tinschert S, Kaminsky E, et al. Czech dysplasia: report of a large family and further delineation of the phenotype. Am J Med Genet A. 2008;146A:1859-64. doi:10.1002/ajmg.a.32389. PMID: 18553548.
88. Hammarsjö A, Nordgren A, Lagerstedt-Robinson K, et al. Pathogenic variant in the COL2A1 gene is associated with spondyloepiphyseal dysplasia type Stanescu. Am J Med Genet A. 2016;170A:266-9. doi:10.1002/ajmg.a.37387. PMID: 26420734.
89. Kannu P, Irving M, Aftimos S, Savarirayan R. Two novel COL2A1 mutations associated with a Legg-Calvé-Perthes disease-like presentation. Clin Orthop Relat Res. 2011;469(6):1785-90. doi:10.1007/s11999-011-1850-x. PMID: 21442341. PMC3094608.
90. Tüysüz B, Güneş N, Yıldırım T, Karakaş H, Kasap B, Onur H, et al. Diagnostic Yield of Genetic Disorders in Children with Hip Dysplasia Mimicking Bilateral Legg-Calvé-Perthes Disease. Diagnostics (Basel). 2026;16(14). doi:10.3390/diagnostics16142293. PMID: 42510156. PMC13409581.
91. Savarirayan R, Tunkel DE, Sterni LM, Bober MB, Cho TJ, Goldberg MJ, et al. Best practice guidelines in managing the craniofacial aspects of skeletal dysplasia. Orphanet J Rare Dis. 2021;16(1):31. doi:10.1186/s13023-021-01678-8. PMID: 33446226. PMC7809733.
92. Rose PS, Ahn NU, Levy HP, Magid D, Davis J, Liberfarb RM, et al. The hip in Stickler syndrome. J Pediatr Orthop. 2001;21:657-63. PMID: 11521037.
93. Ueki M, Ono M, Kamiya M, Yokokawa Y, Tanaka Y, Asaka R, et al. Closely Monitored Successful Full-Term Delivery in a Woman With Spondyloepiphyseal Dysplasia Congenita: A Case Report and Literature Review. J Obstet Gynaecol Res. 2026;52(7):e70409. doi:10.1111/jog.70409. PMID: 42455566. PMC13371809.
94. Kenis V, Markova T, Melchenko E, Gorodilova D. Multiple Ossification Centers of the Pubic Bone as a Supportive Radiographic Feature for COL2A1-Related Congenital Spondyloepiphyseal Dysplasia. Diagnostics (Basel). 2026;16(13). doi:10.3390/diagnostics16131955. PMID: 42449737. PMC13359935.
95. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 853–855.
96. Yazici Z, Kline-Fath BM, Laor T, Tinkle BT. Fetal MR imaging of Kniest dysplasia. Pediatr Radiol. 2010;40:348-52. doi:10.1007/s00247-009-1444-x. PMID: 20020120.
97. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 765–766 (20.3 Spondylo-megaepiphyseal-metaphyseal dysplasia).
98. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 101–102 (4.10 Fibrochondrogenesis).
99. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 117–123 (5.2 Mucopolysaccharidosis IV).
100. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 129–134 (5.4 Mucolipidosis III).
101. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 103–108 (4.11 Oto-spondylo-megaepiphyseal dysplasia).
102. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 199–204 (7.12 Schwartz-Jampel syndrome).
103. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 222–227 (7.17 Progressive pseudo-rheumatoid dysplasia).
104. White KK, Bompadre V, Goldberg MJ, Bober MB, Cho TJ, Hoover-Fong JE, et al. Best practices in peri-operative management of patients with skeletal dysplasias. Am J Med Genet A. 2017;173(10):2584-2595. doi:10.1002/ajmg.a.38357. PMID: 28763154.
105. Savarirayan R, Rossiter JP, Hoover-Fong JE, Irving M, Bompadre V, Goldberg MJ, et al. Best practice guidelines regarding prenatal evaluation and delivery of patients with skeletal dysplasia. Am J Obstet Gynecol. 2018;219(6):545-562. doi:10.1016/j.ajog.2018.07.017. PMID: 30048634.
106. Fincham GS, Pasea L, Carroll C, et al. Prevention of retinal detachment in Stickler syndrome: the Cambridge prophylactic cryotherapy protocol. Ophthalmology. 2014;121:1588-97. doi:10.1016/j.ophtha.2014.02.022. PMID: 24793526.
107. Derocher CE, Mulcrone JE, Fitzgerald G, Carter EM, Bostrom MPG, Figgie MP, et al. Arthroplasty in Individuals With Skeletal Dysplasias: Updated Surgical Outcomes and Complications Following Multidisciplinary Preoperative Screening. Arthroplast Today. 2026;41:102149. doi:10.1016/j.artd.2026.102149. PMID: 42830780. PMC13633928.
108. Zhan S, He Q, Yuan J, Xu X, Huang K, Dong G, et al. Clinical Features of Seven COL2A1 Variations in Chinese Children With Type II Collagen Disorders. Acta Paediatr. 2025;114(7):1720-1730. doi:10.1111/apa.70029. PMID: 39953747. PMC12147410.
109. Clinical Trials Arena. Innoskel enrols first patient in study of collagen disorders in children [news]. 14 Jul 2022. Available from: https://www.clinicaltrialsarena.com/news/innoskel-collagen-disorders-children/ (accessed 10 Oct 2026).
110. Innoskel. An International Prospective Natural History Study in Children With a Type II Collagen Disorder With Short Stature (ROCKET). ClinicalTrials.gov NCT05408715. https://clinicaltrials.gov/study/NCT05408715 (last updated 2023-10-02; accessed 2026-10-10).
111. Helen Keller Eye Research Foundation. Ora Secunda Cerclage Laser Retinopexy To Prevent Retinal Detachment In Stickler Syndrome (OSC/SS). ClinicalTrials.gov NCT07146516. https://clinicaltrials.gov/study/NCT07146516 (updated 2026-07-01; accessed 2026-10-10).
112. Assistance Publique - Hôpitaux de Paris. Scleral Buckling for Retinal Detachment Prevention in Genetically Confirmed Stickler Syndrome: a Randomized Controlled Trial. ClinicalTrials.gov NCT04465188. https://clinicaltrials.gov/study/NCT04465188 (Withdrawn; updated 2026-05-06; accessed 2026-10-10).
113. Jónsson H, Sulem P, Kehr B, et al. Parental influence on human germline de novo mutations in 1,548 trios from Iceland. Nature. 2017;549:519-22. doi:10.1038/nature24018. PMID: 28959963.