Arthrogryposis Multiplex Congenita and Multiple Pterygium Syndrome (Escobar)

A group of congenital conditions with joint contractures in two or more body areas whose common denominator is decreased fetal movement; it includes amyoplasia, the distal arthrogryposes, the neurogenic, myopathic, and neuromuscular junction forms, the fetal akinesia deformation sequence, and the lethal and non-lethal (Escobar) multiple pterygium syndromes.

Literature and regulatory information updated: 18 October 2026.

0. Reference Summary Table

1. Introduction

1.1. Definition and terminology

The term arthrogryposis derives from the Greek arthron (joint) and gryp (curved), with the suffix -osis; multiplex means multiple and congenita, present at birth. Arthrogryposis multiplex congenita, arthrogryposis, and multiple congenital contractures are used as synonyms and denote a clinical sign, not a diagnosis, shared by more than 400 specific conditions [1, 2]. A contractured joint is one that lacks full passive range of motion, in flexion, extension, or both; a flexion contracture is a lack of full anatomical extension, and an extension contracture is a lack of the expected flexion. The limbs, spine, and jaw may be involved. Isolated contractures, such as isolated clubfoot or isolated congenital hip dislocation, fall outside the definition, which requires at least two joints in different body areas [1, 3].

In 2019, an international consensus process, comprising a literature review, identification of critical elements by eight experts, and a three-round modified electronic Delphi with 25 experts from eight countries, produced an eight-sentence definition, subsequently annotated by specialists in epidemiology, genetics, orthopaedics, neurology, plastic surgery, physiotherapy, dentistry, and occupational therapy and by people with lived experience [1, 79]. Faithfully rendered, the definition states that AMC is a term used to describe a group of congenital conditions characterised by joint contractures in two or more body areas; that, although the precise cause may be unknown in some individuals, causes are variable and may include genetic, parental, and environmental factors, as well as abnormalities during fetal development; that individuals with AMC have limited joint movement, with or without muscle weakness, in the involved areas; that contractures vary in distribution and severity and do not progress to previously unaffected joints, but may change with growth and treatment; that spinal deformities may be present at birth or develop during childhood and adolescence; that, depending on the underlying diagnosis, other systems, such as the CNS and the respiratory, gastrointestinal, and genitourinary systems, may be involved; that cognition may be affected if the CNS is involved, whereas sensation is usually normal; and that the impact on mobility, activities of daily living (ADL), and participation is variable [1].

The French PNDS adopts an equivalent formulation: joint limitations at birth at no fewer than two distinct joint levels, usually symmetric, that do not extend to unaffected joints but whose range of motion changes with growth and treatment; AMC is a symptom rather than a diagnosis, and single-level contractures and contractures acquired after birth are not part of the spectrum [4]. The available coding systems (International Classification of Diseases, 10th revision — ICD-10, Orphanet, OMIM, and the Human Phenotype Ontology — HPO) have been combined into a coding strategy for paediatric and congenital anomaly registries, because ICD-10, even with the Royal College of Paediatrics and Child Health extension, proved unsatisfactory for these conditions [21, 80].

1.2. Historical overview

In the dysmorphology literature, amyoplasia is attributed to an initial description by Paré [44]. Non-lethal multiple pterygium syndrome was described by Bussière in 1902 and delineated by Escobar and colleagues in 1978 [30, 81]; OMIM also records the siblings described by Matolcsy in 1936 (pterygia of the neck, axillae, popliteal fossae, and fingers, with cryptorchidism in the 13-year-old boy), the family reported by Norum and colleagues in 1969 (two sibships of cousins with skin "dimples" over the posterior elbows and anterior knees), and Smith's 1982 proposal of the eponym "Escobar syndrome", based on Escobar's review [69]. In 1987, Thompson and colleagues reported eleven new cases that documented the evolution of the phenotype from birth, the respiratory morbidity, and the conductive hearing loss, and confirmed recessive inheritance [69]. Lundblom described a mother and son with ulnar deviation and congenital finger flexion in 1932; Freeman and Sheldon described the "whistling face" syndrome in 1938; Hecht and Beals and, independently, Wilson and colleagues described trismus-pseudocamptodactyly syndrome in 1968, later delineated by Mabry in a large family whose index case in the United States was a girl of Dutch ancestry; and Beals and Hecht described congenital contractural arachnodactyly in 1971 [47, 50, 82, 83].

In 1974, Pena and Shokeir identified in two sisters an early lethal disorder with multiple contractures, facial anomalies, and pulmonary hypoplasia, which they considered autosomal recessive. Hall proposed that this phenotype results from decreased intrauterine movement of any cause, and in 1983 Moessinger reproduced it in rat fetuses curarised in utero, which led to the term fetal akinesia deformation sequence [28, 72]. In 2009, Hall delineated at least 20 familial types of the Pena-Shokeir phenotype, distinguishable by natural history and pathology [28]. Gillin and Pryse-Davis described three sisters with lethal multiple pterygium in 1976, and in 1982 Hall and colleagues separated the lethal form from the other pterygium syndromes [70, 84].

In 1982, Hall, Reed, and Greene delineated the distal arthrogryposes on the basis of 37 patients with congenital contractures of the distal joints, as a typical type 1 and a type 2 with additional anomalies [47, 85]. In 1996, Bamshad, Jorde, and Carey proposed a strict definition of DA, diagnostic criteria, and a revised classification expanded to types 1 to 10, as a framework for molecular classification [86]. In 1997, Hall divided AMC into three clinical groups: predominantly limb involvement; limb involvement with anomalies of other systems; and limb involvement with CNS involvement [1]. The first International Symposium on Arthrogryposis was held in Birmingham in 2007, the second in St Petersburg in 2014, and the third in Philadelphia in 2018 [87]; in 2020 the International Consortium for Arthrogryposis (IC4AMC) was established, a multinational network spanning North America, Europe, and Western Asia organised around early detection, clinical care, research, and education [88].

1.3. Milestones of molecular discovery

In 2006, Toydemir and colleagues showed that variants in MYH3, which encodes embryonic myosin heavy chain, cause Freeman-Sheldon syndrome and nearly one-third of cases of Sheldon-Hall syndrome, the most common DA; nearly all variants interfere with the catalytic activity of myosin, and congenital contractures came to be understood as disorders of the sarcomere [89]. In the same year, two groups identified variants in CHRNG, which encodes the fetal γ subunit of the nicotinic acetylcholine receptor, in Escobar syndrome and LMPS, and defined Escobar syndrome as a "prenatal myasthenia" [90, 91]. In 2008, variants in CHRNA1, CHRND, and RAPSN were described in the multiple pterygium syndromes and in fetal akinesia, and GLE1, a mediator of messenger RNA export, was identified as the gene for LCCS1, a fetal motor neuron disease [92–94]. In 2009, biallelic DOK7 variants were associated with FADS, and Freeman-Sheldon syndrome served as proof of concept that exome sequencing of a few unrelated individuals can identify Mendelian genes [3, 95]. The first DA genes had been identified in 2003 by Sung and colleagues (TPM2 in DA1A, TNNI2 in DA2B1, and TNNT3 in DA2B2), followed by MYH8 in trismus-pseudocamptodactyly syndrome in 2004 and MYBPC1 in DA1B in 2010 [48, 49, 52, 53, 62]. These were followed by ECEL1 in DA5D in 2013, PIEZO2 in DA5 (Coste and colleagues, 2013) and in Gordon and Marden-Walker syndromes in 2014, and MYH3 in autosomal dominant multiple pterygium syndrome in 2015 [57, 63, 96, 97]. More recently, MET was associated with DA11 in a four-generation Chinese family (2019), MYLPF with DA1C (2020), and ADAMTS15 with recessive DA12 (2022) [18, 19, 68].

1.4. Classification

No satisfactory or complete classification system for arthrogryposis exists; several have been developed, oriented towards clinical practice, prenatal diagnosis, research, and basic science [98]. The Bamshad and colleagues decision tree separates isolated from multiple congenital contractures and divides the latter into amyoplasia, distal arthrogryposis, and syndromic forms, with a CNS or progressive neurological aetiology [3, 99]. Hall's three-group clinical classification (1997) is the most widely used in registries and cohorts, but the French PNDS uses a different numbering, with amyoplasia as group 1, DA as group 2, and a third group subdivided into forms with and without CNS involvement; the forms without CNS involvement include myogenic, neurogenic, connective tissue, and skeletal forms [1, 4]. A third, pathophysiological approach classifies causes by the level of the motor system involved: brain, spinal cord anterior horn, peripheral nerve, neuromuscular junction, muscle, connective tissue, and extrinsic mechanical restriction [22, 31].

The 1996 DA classification remains the reference, although the PNDS considers it superseded by molecular classification; the signs of the sarcomeric forms remain the same — overlapping fingers, camptodactyly, ulnar deviation, and foot position anomalies, especially convex feet [4, 86].

GeneReviews advises against the designation DA9 for congenital contractural arachnodactyly, because it emphasises the distal contractures and downplays the marfanoid habitus, proximal contractures, and aortic and ocular involvement [39]. The PNDS appendix assigns DA10 to MYH2, an association not found in the primary sources consulted, in which DA10 (OMIM 187370) remains without a gene, as a phenotype entry with unknown molecular basis [4, 45, 67]. OMIM records that MYH2 and MYH13 were reported in 2009 in two families with DA5 and that these forms were provisionally designated "DA5A" and "DA5B", without any supporting data having been published; the designation DA5D (ECEL1) stems from this provisional numbering [57]. The historical nomenclature creates homonyms: in 1982 Hall and colleagues called Gordon syndrome "DAIIA" and the present-day DA5 "DAIIB", so DAIIA does not correspond to DA2A (Freeman-Sheldon), nor DAIIB to DA2B (Sheldon-Hall) [51, 52, 55]. Beck and colleagues (2013) found the same variant causing DA1 in one family and DA2B in another, with no significant clinical differences by locus, and proposed that DA1 and DA2B represent phenotypic extremes of the same disorder [48].

1.5. Entities in scope

Amyoplasia is a sporadic, symmetric disruptive sequence, with replacement of muscle by fibrous and adipose tissue and a characteristic limb posture, without CNS involvement [4, 44]. The distal arthrogryposes are non-progressive congenital contractures of two or more distal limb parts without primary neurological or muscle disease, in most cases dominantly inherited and caused by variants in genes encoding contractile proteins of fast-twitch myofibres or in PIEZO2 [3, 7, 96]. The neurogenic forms result from involvement of the brain, anterior horn, or peripheral nerve, and the myopathic forms from congenital myopathies or dystrophies of fetal onset [31]. The fetal akinesia deformation sequence is a descriptive diagnosis of a usually lethal condition with multiple contractures, pulmonary hypoplasia, micrognathia, FGR, short umbilical cord, and polyhydramnios, resulting from severely reduced fetal movement; it includes the lethal multiple pterygium syndromes and the lethal congenital contracture syndromes [3, 31]. The multiple pterygium syndromes combine pterygia of the neck, elbows, and knees with arthrogryposis; they are traditionally divided into a prenatally lethal form (LMPS) and a non-lethal form, Escobar syndrome, plus an autosomal dominant MYH3-related form [3, 30, 63]. Spinal muscular atrophy, congenital myasthenic syndromes and autoimmune myasthenia gravis, congenital myopathies and nemaline myopathies may present with AMC, but their management follows that of the underlying disease.

2. Molecular Genetics and Pathophysiology

2.1. Fetal movement and joint formation

The first fetal movements appear from 7 weeks of gestational age (flexion and extension of the trunk and head, dependent on the spinal cord and brainstem); before the end of week 10, movements under supraspinal control appear, such as jaw opening and movements of the upper and then the lower limbs. Mesenchymal condensation of the limb bones proceeds from 7 weeks (proximal segments of the upper limbs) to about 10 weeks (distal phalanges of the lower limbs), and the mechanotransduction generated by primitive movement guides joint differentiation and cavitation. Myoblasts appear at 8–9 weeks, a second wave at 10–15 weeks, and all fibres are present by 18 weeks [4]. In animal models, flaccid paralysis abolishes both the dynamic and static components of movement, whereas rigid paralysis, through permanent AChR activation, abolishes only the dynamic component; when very early, both prevent formation of the synovial cavity and of the vertebral fibrocartilaginous joints; once the cavity exists, flaccid paralysis leads to secondary fusion of the surfaces whereas rigid paralysis preserves the cavity, indicating that the dynamic component is essential for joint formation and the static component for joint maintenance [4].

Decreased fetal movement, common to all forms, produces increased periarticular connective tissue, joint fibrosis, skin dimples over the immobilised joint, disuse atrophy with fibrous and adipose replacement of muscle, and abnormal joint surfaces; micrognathia, mildly short limbs, FGR, pulmonary hypoplasia, and short or immature gut are common [1, 2]. Severity and distribution depend on the timing and duration of akinesia: early, brief episodes mainly affect the upper limbs, late episodes mainly the lower limbs, and prolonged episodes affect more regions with greater severity. A pterygium forms over a joint held immobile in flexion and implies prolonged immobility; in most AMC, embryonic joint development is normal and the changes are deformational [1]. Pterygia are more frequent in the myopathic and neuromuscular junction forms and rare or absent in the neuronal and anterior horn forms [4].

2.2. Genetic architecture and levels of the motor system

In the division between neurological and muscular causes, the Oxford Desk Reference attributes about 90% to the former and 5–10% to the latter, proportions with no stated denominator [3]. Genetic causes appear to account for most AMC and include point variants, aneuploidies (trisomies 21, 13, and 18), and microdeletions or microduplications; some causes always produce multiple contractures and others only at the severe end of the spectrum, and the most frequent affect the motor unit — anterior horn, axon, neuromuscular junction, and striated muscle fibre — or the CNS [1]. In 2016, 320 genes were known, grouped into 22 gene ontology categories; in 2019, 82 were added, for a total of 402 genes in 29 clusters, only 19 of which contained more than one gene [15, 100]. In a 2026 review of a French cohort of 367 patients, autosomal recessive inheritance accounted for 68.3%, and 24% of the dominant or X-linked forms were de novo [101]. In 315 undiagnosed French families, the most frequent cause was primary involvement of skeletal muscle (40%), followed by the brain (22%); recessive inheritance explained 66.3%, and among 60 sporadic cases with nonconsanguineous parents, 30 (50%) had de novo dominant or X-linked variants [102].

2.3. Sarcomeric distal arthrogryposes

About half of DA cases are caused by variants in genes encoding contractile proteins of skeletal myofibres, and the common mechanism is disruption of the contractile complex of fast-twitch fibres: the fast isoforms of troponin I (TNNI2), troponin T (TNNT3), and β-tropomyosin (TPM2), embryonic myosin (MYH3), perinatal myosin (MYH8), and slow myosin-binding protein C (MYBPC1) [3, 82, 96]. The variants are dominant gain-of-function variants: they increase calcium sensitivity, increase force, and slow relaxation in vitro; because the proteins are expressed from the embryonic stage and after birth, the "active" muscular mechanism is permanent [4]. TNNI2 variants increase the calcium sensitivity of force generation, with an effect confined to fast muscle, because the heart uses TNNI3 and TNNT2 [7].

In MYH3, three recurrent variants, described according to Human Genome Variation Society (HGVS) nomenclature on the MANE Select (Matched Annotation from NCBI and EMBL-EBI) transcript NM_002470.4:c.533C>T p.(Thr178Ile), c.2014C>T p.(Arg672Cys), and c.2015G>A p.(Arg672His), account for about 91% of MYH3-positive families with Freeman-Sheldon syndrome; Freeman-Sheldon and Sheldon-Hall syndromes affect different residues, so genotype helps predict phenotype [9, 89, 103]. In the initial series of 28 probands with Freeman-Sheldon (7 familial and 21 sporadic), 26 had an MYH3 variant: 12 p.(Arg672His), 8 p.(Arg672Cys), 3 p.(Thr178Ile), and 3 de novo missense variants at other residues; among 46 probands with DA2A (including 30 from that series), 43 had an MYH3 variant and 39 (91%) one of the three recurrent ones, and overall severity varied with genotype (p = 0.0055), with p.(Thr178Ile) the most severe (more marked foot and facial contractures and scoliosis, often congenital, only in simplex cases) and p.(Arg672Cys) the least severe and predominant in familial cases [51]. In 38 Sheldon-Hall probands negative for TNNI2 and TNNT3, MYH3 explained 5 of 12 familial cases (42%) and 7 of 26 sporadic cases (27%) [54]. DA2B3 shows incomplete penetrance: in the Swedish family with NM_002470.4:c.700G>A p.(Ala234Thr) (VCV000014145, conflicting classifications in ClinVar), the unaffected paternal grandfather was a carrier, and the same family had members classified as DA2B and as DA1; the first Asian family described had NM_002470.4:c.1160A>G p.(Tyr387Cys) (VCV000631471) [54, 103]. The dominant variants associated with DA1 and DA2 lie in the myosin head and neck, whereas two of the dominant multiple pterygium syndrome variants are located in the tail domain [63]. In the contractures, pterygia, and spondylocarpotarsal fusion syndromes (CPSFS), TGF-β signalling is also disrupted, with reduced SMAD3 phosphorylation. In the recessive form (CPSFS1B), a single truncating variant does not cause disease, because it is transmitted by unaffected parents; disease arises when the truncating allele is in trans with a hypomorphic allele, such as NM_002470.4:c.-9+1G>A (rs557849165), a splice donor variant in the 5' untranslated region intron that standard exome capture does not cover [9, 103]. A family described in 2026, with biallelic compound heterozygous variants and DA without skeletal fusions, and heterozygotes with a subclinical phenotype, shows that the classification and inheritance patterns of MYH3-related disorders (codominance or recessiveness) are still evolving [104].

In TNNI2, the DA2B1 variants include NM_003282.4:c.521G>A p.(Arg174Gln) and c.466C>T p.(Arg156Ter), described in 2003, the in-frame deletion c.524AGA[1] p.(Lys176del) (VCV000012437; published as c.526delAAG in a Swedish family and as c.523delAAG in a Chinese family with 32 affected individuals over seven generations, with expressivity ranging from mild camptodactyly to severe involvement of the hands and feet with facial anomalies), and c.493A>T p.(Ile165Phe) [52, 103]. In vitro, TNNI2 p.(Arg174Gln) and p.(Arg156Ter) and TNNT3 p.(Arg63His) increase actin-activated myosin ATPase activity, reflecting greater calcium sensitivity and increased contractility; the contractures would thus result from an active process of increased muscle tension during development rather than from a passive process [52, 53]. In TNNT3, the variant NM_006757.4:c.188G>A p.(Arg63His), often de novo, causes DA2B2 (OMIM 618435) and alters the binding of fast troponin T to tropomyosin, with protein instability on western blot [8, 105]. Codon 63 is a hotspot: c.187C>T p.(Arg63Cys) (VCV000031874) segregated in a three-generation Chinese family, and p.(Arg63His) was found in a four-generation Indian family with 18 affected individuals, classified as "DA1 or a mild variant of DA2B", and in a child with DA1 without facial involvement [53, 103]. Biallelic loss of function causes a distinct recessive entity: severe congenital myopathy with weakness, hypotonia, contractures, congenital scoliosis, and dysmorphism, with atrophy and degeneration of fast fibres, nemaline rods, and fast troponin T deficiency [8, 105]. In TPM2, dominant variants cause DA1A and DA2B4 (both in OMIM entry 108120) — in the Utah family in which DA1 was mapped, NM_003289.4:c.271C>G p.(Arg91Gly) (VCV000012460), with gain of function in vitro; in DA2B4, c.397C>T p.(Arg133Trp) (VCV000012463) in two families, one Swedish and one Korean, and c.308A>G p.(Gln103Arg) (VCV000631479) in a Chinese family [48, 103] — and congenital myopathy 23 (CMYO23, OMIM 609285, formerly nemaline myopathy 4 and cap myopathy 2, with caps, nemaline rods, or congenital fibre-type disproportion) occurs without congenital contractures, although OMIM allows that TPM2-related nemaline myopathy may be identical to DA1A [48, 106]; there is no clear relationship between variant type or position and phenotype. A homozygous null allele eliminating the skeletal β-tropomyosin isoform causes recessive nemaline myopathy with an Escobar phenotype, and a second consanguineous family had a homozygous intronic splice variant with cores, minicores, and sparse rods [6]. ClinGen classifies TPM2-related myopathy as a definitive autosomal dominant association, with a haploinsufficiency score of 1 and a triplosensitivity score of 0 [6].

MYH8 encodes perinatal myosin heavy chain. All confirmed families with trismus-pseudocamptodactyly syndrome carry the same variant, NM_002472.3:c.2021G>A p.(Arg674Gln), which arose independently in North American and European families (the three US families share a haplotype that the Dutch family does not) [62]; it is a dominant missense variant at a conserved residue of the actin-binding region of the myosin head, near the ATP-binding site. Heterozygous loss-of-function MYH8 variants do not cause the syndrome (16 individuals without the disease and about 100 in population databases), which excludes haploinsufficiency as the mechanism and calls for caution in interpreting such variants [10, 82, 103]. Unlike MYBPC3, MYBPC1 does not cause cardiomyopathy, and it shows a dosage relationship: dominant missense variants cause DA1B (OMIM 614335); de novo dominant variants in the M motif, NM_002465.4:c.788T>G p.(Leu263Arg) and c.776T>C p.(Leu259Pro), cause congenital myopathy with tremor (congenital myopathy 16, CMYO16, OMIM 618524) without congenital contractures, and the variants p.(Glu248Lys) and p.(Tyr247His) (NM_002465.4:c.742G>A and c.739T>C), also in the M motif, segregated in two multigenerational Latvian and German families with high-frequency postural tremor of myogenic origin [103, 107]; and a recessive premature stop variant that eliminates all domains causes LCCS4 (OMIM 614915), with anterior horn degeneration [12, 103]. This variant, NM_002465.4:c.952C>T p.(Arg318Ter) (VCV000039766, conflicting classifications in ClinVar), was found in the homozygous state in two consanguineous Bedouin families from southern Israel, with a shared haplotype indicating a common founder; the original DA1B family had 12 affected individuals over five generations, with complete penetrance [49, 103, 108]. A child compound heterozygous for variants in the Fn-III C7 and Ig C8 domains, with unaffected heterozygous parents, had severe early-onset DA1, which also suggests a recessive model [12].

Three additional genes complete the OMIM DA series. MYLPF (current symbol MYL11, 16p11.2), which encodes fast myosin regulatory light chain, causes DA1C: among 172 families with multiple congenital contractures negative for the DA1, DA2A, and DA2B genes, 8 families were identified, 6 with recessive inheritance (homozygous for NM_013292.5:c.470G>T p.(Cys157Phe) in families of Polish and Indian ancestry, or for c.469T>C p.(Cys157Arg) in three Pakistani probands) and 2 with dominant inheritance (c.487G>A p.(Gly163Ser) in an Ashkenazi Jewish father and son; c.98C>T p.(Ala33Val) in 5 affected individuals over three generations); the dominant and recessive phenotypes were indistinguishable, and structural analysis suggests that the more highly conserved residues Ala33 and Gly163 interact more directly with myosin heavy chain [18, 103]. MET, the hepatocyte growth factor receptor, was associated with DA11 in a four-generation Chinese family with 11 affected individuals heterozygous for NM_000245.4:c.3701A>G p.(Tyr1234Cys) (VCV000446117), which prevents receptor phosphorylation; the mouse model showed loss of fibres in muscles derived from migratory precursors (limbs and diaphragm), and OMIM keeps the association provisional because there is a single family [68, 103]. ADAMTS15 causes recessive DA12, described in 5 patients from 4 consanguineous families (Syrian, two Turkish, and one from Kosovo) with homozygous nonsense, splicing, and missense variants; the heterozygous parents were unaffected and only one of the variants was in gnomAD, in a single heterozygote [19].

2.4. PIEZO2 and ECEL1

PIEZO2 encodes a mechanosensitive cation channel expressed in the skin, the alveoli, and the sensory neurons of the dorsal root ganglia, and gives rise to two groups of disorders with opposite mechanisms [4, 11]. Heterozygous gain-of-function variants, often de novo, cause DA5, Gordon syndrome (DA3), and, in isolated cases, Marden-Walker syndrome; the variants p.Glu2727del and p.Ile802Phe, in historical numbering, accelerate recovery from inactivation and/or slow inactivation, increasing mechanically activated current [11]. In the original series, PIEZO2 variants were found in 24/29 families with DA5 (83%), 10 of which (42%) carried the recurrent deletion p.Glu2727del; in 10/12 families with Gordon syndrome (83%), nine of which carried the variant designated c.8057G>A p.(Arg2686His) and one an 8-bp deletion; and in one of two families with Marden-Walker syndrome, with the de novo variant p.Arg2686Cys at the same nucleotide as p.Arg2686His (on the MANE transcript, NM_001378183.1:c.8395C>T p.(Arg2799Cys), VCV000137630) [55, 57, 58, 103]; two probands diagnosed with DA5 without cleft palate carried p.Arg2686His and were reclassified as DA3, given the reduced penetrance of cleft palate [55]; cleft palate was associated with this variant (adjusted p <0.0001), suggesting variable expressivity of a single condition [97]. This historical numbering refers to an older transcript; on the MANE transcript NM_001378183.1 the same variant (VCV000137629) is c.8396G>A p.(Arg2799His) [103]. In a three-generation Chinese family with DA3 without cleft palate, the historical variant p.Arg2718Gln corresponds to NM_001378183.1:c.8492G>A p.(Arg2831Gln) (VCV000631524) [55, 103]. Biallelic loss-of-function variants cause DAIPT (OMIM 617146) [11, 109]: the initial reports (2016–2017) included compound heterozygous and homozygous nonsense variants, missense variants without function in HEK293 cells, and, in one patient, two homozygous deletions encompassing the gene; the heterozygous parents of a Turkish patient had distal joint laxity [59].

ECEL1 encodes a neuronal endopeptidase required for intramuscular terminal branching of motor axons and for formation of the fetal neuromuscular junction; Ecel1-deficient mice have abnormal terminal branching at the motor endplate, which identified a second, neurogenic, pathway in DA [13, 96]. DA5D is recessive; in the initial series, ECEL1 explained 5 of 7 families, about 70% of cases, and no pathogenic heterozygous ECEL1 variants were found in 20 families with dominant DA; an independent series identified seven homozygous or compound heterozygous variants in 6 of 20 families with DA, after homozygosity mapping in a Malian and a Moroccan family (combined LOD score 5.1 at 2q37) [60, 96].

2.5. Fetal neuromuscular junction: CHRNG and associated genes

The muscle AChR is a pentamer; in fetal life it contains the γ subunit, encoded by CHRNG, which is replaced by the ε subunit in the adult AChR during the third trimester, with the switch complete at about 33 weeks of gestation [29, 30]. CHRNG variants prevent correct membrane localisation of the fetal AChR, as shown in HEK cells, and expression in the prenatal mouse matches the phenotype; the γ→ε switch explains the absence of myasthenic symptoms after birth, a direct parallel with arthrogryposis caused by maternal anti-AChR antibodies [90]. The two founding studies identified 8 variants in 7 families with Escobar syndrome and 6 homozygous variants in 6 families with the lethal or non-lethal forms, both forms coexisting in one family [69]. In a review of 64 patients with CHRNG-related Escobar syndrome (7 patients from 6 families under follow-up and 57 published), the most frequent variants were c.459dup (8 patients) and p.(Arg239Cys), NM_005199.5:c.715C>T (VCV000018337; 7 patients), both in diverse ethnic groups; variants were located mainly in the extracellular domain, and neither variant type nor position correlated with severity [69, 103, 110]. The outcome is a fixed consequence of fetal akinesia rather than a progressive myasthenia, although diaphragmatic weakness, diffuse myopathy, and myasthenia-like features have been reported [5, 30]. Among 100 families with multiple pterygium syndrome or FADS, CHRNG variants were found in 11/41 families with Escobar syndrome (27%) and in 5/59 with LMPS or FADS (8%); 21/24 patients (87.5%) with a CHRNG variant had pterygia, and no variant was found in the presence of CNS anomalies; the mutational spectrum was the same in both phenotypes, and both were observed in different families with the same variant [111]. In LMPS, CHRNG accounts for about 8% and CHRNA1 and CHRND for about 3% each; neither CNS defects nor cleft palate were observed in the CHRNG-positive group [70]. CHRNG can also cause FADS without pterygia [5].

Homozygous nonsense variants in CHRNA1 and severe CHRND variants cause LMPS, because complete loss of fetal AChR function is lethal; these genes are also expressed in early somites, which may explain associated malformations [93, 112, 113]. There is a continuum of severity across the presynaptic, synaptic, and postsynaptic proteins of the motor endplate: complete disruption of the fetal AChR causes LMPS, milder alterations cause fetal hypokinesia with contractures, and others cause congenital myasthenic syndromes (CMS) later in life [3, 93]. Among 15 cases of LMPS or fetal akinesia without a CHRNG variant, a homozygous RAPSN variant, NM_005055.5:c.1177_1178del p.(Asn393GlnfsTer83), was found in a family with three children with lethal FADS; incomplete loss of rapsyn causes CMS and more severe loss causes FADS (FADS2, OMIM 618388) [31, 92, 103]. Analogously, severe biallelic loss of function of DOK7 causes lethal FADS (FADS3, OMIM 618389), and incomplete loss causes CMS [95, 114]; in a second consanguineous family, a 32-week stillborn with hydrops and pterygia was homozygous for the 1-bp duplication described as c.1263dupC in exon 7 [75]. FADS1 (OMIM 208150) results from homozygous MUSK variants: in a Dutch genetic isolate, 14 affected fetuses were homozygous for NM_005592.4:c.1724T>C p.(Ile575Thr) (VCV000190467), with reduced AChR clusters and tyrosine kinase activity at the motor endplates, and in a Swedish family 5 fetuses had a homozygous truncating variant; all died in utero or shortly after birth [73, 103]. FADS4 (OMIM 618393) is caused by NUP88 variants, and OMIM interprets the MUSK-, RAPSN-, and DOK7-related forms as extreme CMS phenotypes [73]. Among the CMS, RAPSN (15–20% of CMS) presents with hypotonia, respiratory failure at birth, episodic apnoea, and AMC; MUSK has a phenotype that includes prenatal onset with FADS; and SLC5A7 and SLC18A3 are associated with arthrogryposis, apnoeic crises at birth, and, for SLC18A3, FADS early in pregnancy [37].

2.6. Anterior horn and peripheral nerve

LCCS1 and CAAHD (congenital arthrogryposis with anterior horn cell disease, formerly called lethal arthrogryposis with anterior horn cell disease, LAAHD) are allelic and result from recessive variants in GLE1, a mediator of messenger RNA export that links LCCS1, LCCS2 (ERBB3), and LCCS3 (PIP5K1C) in a common RNA-processing pathway [72, 94]. The Finnish founder variant Fin-major is NM_001003722.2:c.433-10A>G (VCV000006462, rs386833693), an intronic variant that creates a new splice acceptor site, with insertion of nine nucleotides in the cDNA and of three amino acids (described in the source as Thr144_Glu145insProPheGln) in the coiled-coil domain; its allele frequency is 0.012 in gnomAD Finns [94, 103]. Fin-major was homozygous in 51 of 52 LCCS1 cases from 29 unrelated families (the remaining case was compound heterozygous with an exon 12 missense variant), and the carrier frequency was 2% in northeastern Finland and 1% in the general Finnish population [76]. The 12 patients with CAAHD from 9 Finnish families were all compound heterozygous for Fin-major and an exon 13 or exon 16 missense variant; in two siblings of consanguineous parents from northeastern Finland, CAAHD resulted from homozygosity for NM_001003722.2:c.2051T>C p.(Ile684Thr) (VCV000006465), the exon 16 missense variant [17, 103]. OMIM records the correlation proposed in 2018: homozygosity for Fin-major, which disrupts the coiled-coil domain, results in LCCS1 with absent movements and death in utero, whereas missense variants in other regions, even with Fin-major in trans, usually allow survival beyond the perinatal period, although many patients die in childhood [17]. The genetic LCCS series in OMIM runs from LCCS1 to LCCS12 (GLE1, ERBB3, PIP5K1C, MYBPC1, DNM2, ZBTB42, CNTNAP1, ADCY6, ADGRG6, NEK9, GLDN, and DST) [76]. It has been proposed that variants preserving the coiled-coil domain cause CAAHD and those abolishing it cause LCCS1; another group argues for a single spectrum that includes non-lethal forms, and the adjective "lethal" in CAAHD is considered inappropriate [14].

SMA with homozygous SMN1 deletion is rarely associated with AMC; when it is, it usually corresponds to SMA type 0, with low SMN2 copy number and respiratory compromise; about half of SMA type 0 cases have homozygous SMN1 deletion, and the rest, not linked to 5q, include SMA with respiratory distress (SMARD1, IGHMBP2) and Brown-Vialetto-Van Laere syndrome (SLC52A3, SLC52A2) [4, 31, 115]. X-linked infantile SMA (UBA1) combines congenital hypotonia, areflexia, anterior horn cell loss, congenital contractures, mainly digital, and fractures; only missense and synonymous variants in exon 15 have been described [16]. Somatic UBA1 variants in haematopoietic cells cause VEXAS syndrome in adult men, an allelic disorder with no clinical relationship [16]. Other genes associated with SMA and/or arthrogryposis include BICD2 (SMA-LED2A and SMA-LED2B, the latter with onset in utero and frequent prenatal death), TRPV4 (scapuloperoneal SMA and congenital non-progressive SMA with contractures, allelic with hereditary motor and sensory neuropathy 2C), ASCC1 and TRIP4 (SMA with congenital fractures), DNM2 (LCCS5), EXOSC3, TSEN54, VRK1, and RARS2 (pontocerebellar hypoplasias), and ATP7A and ZC4H2 (Wieacker-Wolff syndrome) [16, 115]. Axoglial defects with poorly myelinated motor neurons due to recessive variants in CNTNAP1, ADCY6, and ADGRG6 represent the peripheral nerve level [31]. 5q SMA is also part of this differential diagnosis.

2.7. Myopathic forms

Congenital myopathies can present prenatally or at birth with fetal akinesia, severe hypotonia, and little or no movement or respiratory effort [31]. Recessive RYR1 variants were detected in 8% of an LMPS/FADS cohort, which requires consideration of malignant hyperthermia risk in carrier parents and in siblings [31, 116]. Up to half of severe presentations of nemaline myopathy are due to recessive ACTA1 variants, and recessive NEB and KLHL40 variants have been described in FADS [31]. In the Australian cohort of 190 probands, TTN was the most frequent gene (8 cases, two with marked amyoplasia, five with variants in the triplicated repeat region poorly covered by exome sequencing) [34]. Congenital myopathies and nemaline myopathy are part of the differential diagnosis.

2.8. Normalised variants

In the Australian cohort, the 1-bp CHRNG duplication (rs774279192, c.459dup) was the most frequently observed pathogenic variant, present on five alleles (cohort allele frequency 0.0132), compared with 97 of 282,834 alleles in gnomAD (0.000343) [34]. Nomenclature discrepancies are common in this field: the article that described c.1009_1010delCA did not follow the 3' alignment rule, and the same article writes "Ter43" in the results and "Ter34" in the discussion for c.401_402del, Ter43 being the validated form [29, 103].

2.9. Oligogenicity, modifiers, and mosaicism

In a series of 48 mostly Turkish families, exome sequencing identified variants in known genes in 17 families (35.4%), with homozygous CHRNG variants in six patients and ECEL1 variants in four; three families (6.2%) had candidate genes and eight had a second locus, for an overall molecular explanation of 58.3% [117]. In one patient with a more severe phenotype, a homozygous ERCC2 variant (the gene for COFS syndrome type 2) coexisted with the homozygous CHRNG variant [117]. In the second phase of the same programme, in 89 families, 11 of 58 diagnoses (19.0%) involved more than one locus, in a setting of consanguinity with extensive regions of absence of heterozygosity; copy number variants (CNVs) and mosaicism were documented by droplet digital PCR, and monoallelic and biallelic variants in the same gene could cause similar or distinct syndromes [35]. Intrafamilial variability, especially among distant relatives, may be due to additional variants in other FADS genes that modify severity [31, 117]. Mosaicism has clinical expression: in trismus-pseudocamptodactyly syndrome, two affected children of unaffected parents indicate probable gonadal mosaicism; in Sheldon-Hall syndrome, transmission occurs from mildly affected or somatically mosaic parents [7, 82].

2.10. Allelic disorders

The allelic disorders most relevant to diagnosis are those of TPM2 (DA1A, DA2B4, congenital myopathy 23 without contractures, and recessive Escobar), TNNT3 (dominant DA2B2 and recessive congenital myopathy), MYBPC1 (DA1B, CMYO16 with tremor, and LCCS4), MYH3 (DA1, DA2A, DA2B3, CPSFS1A, and CPSFS1B), PIEZO2 (DA3, DA5, and Marden-Walker through gain of function; DAIPT through loss of function), GLE1 (LCCS1 and CAAHD), TPM2 (DA1A and DA2B4 in the same OMIM entry), MUSK (FADS1 and CMS9), MYLPF (dominant and recessive DA1C), RAPSN and DOK7 (FADS and CMS), UBA1 (X-linked SMA and somatic VEXAS), and TRPV4 (congenital SMA and hereditary motor and sensory neuropathy 2C) [6, 8, 9, 11, 12, 14, 16, 115]. Gordon syndrome of pseudohypoaldosteronism type II (WNK1, WNK4) is a homonym unrelated to Gordon DA3 (PIEZO2).

3. Clinical Spectrum and Manifestations by System

3.1. Common features of AMC

The spectrum ranges from a few contractures of the hands and feet to involvement of all limbs, the spine, and the jaw; limited mouth opening, masticatory weakness, high palate, cleft palate, and retrognathia compromise oral hygiene and feeding [1]. No new contractures arise in previously unaffected joints, and the more joints involved, the greater the impact on mobility and ADL; after birth, contractures improve, especially with early physical and occupational therapy, and passive range of motion may exceed active range because of muscle weakness; with growth they may worsen through muscle imbalance, especially during growth spurts [1]. Sensation is usually preserved, but congenital neuropathies, some with proprioceptive deficits, also cause AMC; hearing loss has been described in about 10% of DA cases in a Swedish series [1]. Intellectual disability or CNS involvement occurs in about 25% of all AMC [118]; in cohorts, brain defects were found in 21/75 (28%) in the Netherlands, 23/68 (34%) in western Sweden, and in group 3 in 33/103 (32%) in Alberta, and in a French cohort of 42 patients, with recruitment bias, groups 1, 2, and 3 accounted for 19%, 33%, and 48% [1]. Signs of CNS involvement include intellectual delay, seizures, microcephaly, ischaemic lesions, polymicrogyria, lissencephaly, hydrocephalus, and agenesis of the corpus callosum [1].

Spinal deformities include scoliosis (an unbalanced C-shaped or a compensated S-shaped curve), hyperkyphosis, and hyperlordosis. In amyoplasia and the syndromic forms they may be present at birth as "spinal contracture", which should not be called congenital scoliosis, a term reserved for vertebral malformations; curves present at birth are associated with severe contractures of all four limbs and a guarded prognosis for walking and independence, and may progress rapidly in infancy or remain stable until late childhood. Developmental curves occur in amyoplasia, in some DA (Freeman-Sheldon, Beals), and in the syndromic forms; those in DA tend to be thoracolumbar and are the most progressive [1]. In a Grenoble cohort (2007–2022) of 146 patients with radiographs, scoliosis was present in 46% of the 101 children and 78% of the 45 adults, with underlying vertebral anomalies in 41% and 53%; among children, scoliosis occurred in 47% of DA, 65% of "other causes", and 33% of amyoplasia, and Cobb angles and anomalies were greater in DA [119]. In a series of 105 children with AMC who underwent spinal magnetic resonance imaging (MRI), 42 (40%) had a tethered cord, with the conus at the level of the lower third of L2 or below, a higher proportion than in the normative population [120].

In a Norwegian cohort of 72 adults (aged 18–80 years), 57% had neurological disease or symptoms, 63% spinal or trunk involvement, 36% craniofacial malformations, 26% sensorineural hearing loss, 18% ophthalmoplegia, 53% axial weakness, 44% areflexia, and 40% hypotonia; 61% of electrophysiological studies showed neurogenic changes, and 75% of the adults required medical follow-up [121]. Pain is more frequent in adults than in children and is greater after multiple corrective procedures [122]; in a survey of 73 adults (67.3% women), 82.7% reported pain in the previous seven days, of greater intensity in women and associated with worse physical function, participation, and sleep [123].

3.2. Amyoplasia

Amyoplasia is the most common form of AMC, up to one-third of all cases, with a characteristic symmetric posture [3, 4]. In the upper limbs there are rounded, sloping, internally rotated shoulders, extended elbows, pronated forearms, flexed wrists and hands (the so-called policeman tip position), adducted thumbs, and severe flexion contractures of the metacarpophalangeal joints with mild interphalangeal contractures; in the lower limbs, flexed and often dislocated hips, knees in flexion or extension, and severe bilateral equinovarus feet; the spine is rigid and straight [4, 44]. Muscle masses are markedly atrophic or absent, replaced by fibrous bands and adipose tissue, with fusiform limbs, shallow or absent flexion creases, deep dimples over the joints, and, in most cases, a salmon patch or haemangioma in the facial midline (glabella, nasal tip) [3, 4]. The face and facial movements are normal; ptosis, trismus, and cranial nerve palsy suggest another diagnosis [3]. Intelligence is usually normal, barring birth trauma [44].

Among 560 individuals with amyoplasia, all four limbs were involved in 63%, three limbs in 5%, the upper limbs only in 17%, and the lower limbs only in about 15% [1]. Exclusive lower limb involvement is associated with an excess of boys with intestinal atresia, and exclusive upper limb involvement with an excess of girls with gastroschisis [44]. In the same series, perinatal long-bone fractures occurred in 10.5% (59/560); anomalies attributed to vascular compromise were gastrointestinal in 9.1% (51/560), trunk muscle defects in 2.7%, digit involvement in 12.1%, and constriction rings in 4.3%; and discordant monozygotic twins occurred in 6.6% (37/560; odds ratio 10.9) [124]. Seventy percent of pregnancies had first-trimester complications (bleeding, influenza, fever), breech presentation is common, and delivery may be difficult [44]. The pathogenic hypothesis is vascular, with hypotension of multiple origins; the upper limb posture, resembling C5-C6 Erb obstetric palsy, and the metameric topography of muscle involvement also suggest an early paralytic neurogenic lesion of the anterior horn [4, 44]. Heart defects have never been described in amyoplasia or in DA; renal agenesis, ureteral dilatation, and transient hydronephrosis have been mentioned [4].

The motor outcome is favourable: by age 5 years, 85% of children walk with appropriate physiotherapy, more than two-thirds require orthopaedic surgery (mean of 5.7 procedures per child), and most attend mainstream schooling at the appropriate grade level and are independent in ADL [44]. Bone growth of the affected limbs is reduced, and flexion and pterygia of the large joints may increase over time [44]. Despite the good functional prognosis, the PNDS describes severe participation restrictions in adulthood, even compared with DA [4]. In the Swedish cohort of 131 patients, muscle strength mattered more than joint range of motion for motor function in amyoplasia [125].

3.3. DA1

At birth, the vast majority have hands clenched into a fist with an adducted thumb and fingers overlapping medially, the so-called trisomy 18 position, with hypoplastic or absent flexion creases, ulnar deviation, and camptodactyly [3, 47]. In Hall's series, positional foot deformities occurred in 88% (bilateral calcaneovalgus 33%, bilateral equinovarus 25%, combinations 30%), hip involvement in 38%, mild knee flexion contractures in 30%, and shoulder stiffness in 17%; the implied denominator is the 37-patient series [47]. The hands eventually open with residual camptodactyly and ulnar deviation, and 20% of adults have straight, functional fingers; neurological examination and intelligence are normal, and response to treatment is good in all joints [47]. Expressivity is highly variable within families, and a parent may have only mild hand contractures [47]. In the Swedish cohort, 44% of the DA patients studied had muscle weakness [125]. The multigenerational Utah family in which DA1A was mapped had 15 affected individuals with highly variable expressivity, ranging from shoulder limitation and small calves without foot deformity to overlapping fingers, camptodactyly, ulnar deviation, bilateral equinovarus, and bilateral hip dislocation; the most consistent features were overlapping fingers at birth, abnormal flexion creases, and foot deformities; five had a small mouth with slightly limited opening, and all had normal intelligence [48]. In the five-generation DA1B family, 9 of the 12 affected individuals had clubfoot and 3 vertical talus, bilateral in all but two, and only 5 had hand involvement (camptodactyly with ulnar deviation in 3, extension contracture of the 4th finger in 2); biopsy showed type I fibres smaller than type II fibres [49]. In DA1C, proximal contractures (elbows, hips, knees) and short stature appear more frequent than in other DA, and in one child refractory clubfoot led to amputation, with complete absence of skeletal muscle in the foot (segmental amyoplasia) [18]. DA1 was diagnosed by ultrasound at 18 weeks in a family with two affected individuals, from the extended wrist and clenched fingers throughout the examination [48].

3.4. DA2A, Freeman-Sheldon syndrome

The face is mask-like, with a small "whistling" mouth (100%), deep-set eyes, broad nasal root, telecanthus, blepharophimosis, hypoplastic alae nasi, long philtrum, H-shaped chin dimple, high palate, small tongue, nasal voice, and dental crowding (100%); most features result from increased muscle tone [50]. Skeletal findings include ulnar deviation (91%), finger flexion (88%), equinovarus (59%), hip and/or knee contractures (73%), and scoliosis (85%) [50]. In a strictly defined cohort of 32 patients, who were only 32 of the 73 referred with the provisional diagnosis (the rest were mainly DA2B, 38 cases, and one case of multiple pterygium), severe scoliosis occurred in 85%, strabismus in 42%, and hearing loss in 30%; most infants required a nasogastric tube or gastrostomy, about 80% of the children used walking aids, about 10 operations per child were needed, all were cognitively normal, and half had anaesthetic or surgical complications, including malignant hyperthermia and hyperpyrexia [9, 51]. Walking began at a mean of 19 months in this cohort [51]. Smith's lists microcephaly (44%) and intellectual disability (31%) as occasional findings, figures probably influenced by older cases without a strict definition [9, 50]. Neonatal feeding difficulties require tube feeding in 45% and gastrostomy in 17%; the mean age at walking is 18 months (10–48), 83% use a walking aid, the mean number of surgical procedures is 10.3, and early mortality from aspiration may occur; obstructive sleep apnoea occurring exclusively in rapid eye movement (REM) sleep has been described [50]. The mean age at walking is 18 months in Smith's and 19 months in the OMIM summary, and the proportion using a walking aid 83% and about 80%, respectively; the sources do not allow determination of whether the figures derive from different counts. There are reports of whistling face without limb anomalies, of bilateral juvenile glaucoma, and of forms with CNS involvement (cerebral, cerebellar, and brainstem atrophy), the latter proposed as recessive syndromes distinct from the dominant form without neurological involvement; microstomia was treated with an oral expander 2 to 3 hours a day before orthodontics [51].

3.5. DA2B, Sheldon-Hall syndrome

It is less severe than DA2A and more severe than DA1: severe camptodactyly with ulnar deviation, clenched and overlapping fingers, equinovarus or vertical talus, a triangular face with prominent nasolabial folds, downslanting palpebral fissures, a small mouth, and a prominent chin, without pursed lips or an H-shaped chin dimple [7, 47]. Attached earlobes, mild neck webbing, and short stature have been described; contractures are most severe at birth and do not progress, and life expectancy and cognition are normal [7]. A variant in MYH3, TNNI2, or TNNT3 is found in about half of patients, and about half of cases are simplex (no other affected relatives) [7]. DA2B is considered the most common DA [52]. In the original family of 21 affected individuals over five generations, none had feeding difficulties at birth or oral surgery, both nearly universal in DA2A; in a Swedish family with DA2B1, four adults had proximal contractures, thin calves, mildly elevated creatine kinase without clinical weakness, and myopathic changes mainly in type 2 fibres [52]. In TPM2-related DA2B4, a mother and daughter aged 65 and 28 years had proximal and distal muscle weakness, hearing loss, high palate, short stature, proximal contractures, and scoliosis, without cardiac involvement; in a Korean family, the proband died at a few months of age, presumably of aspiration pneumonia [48].

3.6. PIEZO2-related forms: Gordon, DA5, Marden-Walker, and DAIPT

DA5 combines distal contractures with ptosis, ophthalmoplegia or restricted eye movements, strabismus, short stature, restrictive lung disease, sometimes with pulmonary hypertension, and variable absence of the cruciate ligaments; Gordon syndrome combines distal contractures, cleft palate, and short stature; Marden-Walker syndrome, provisionally included, has a mask-like face, blepharophimosis, micrognathia, contractures, and kyphoscoliosis [11, 47]. In the gain-of-function forms, joint dimples are constant and very extensive, and there is blepharophimosis, progressive restrictive lung disease even without spinal deformity, marked muscle hypertonia, and early musculoskeletal pain [4]. Long-term follow-up of dominant families has shown dysphagia, spinal stenosis–like symptoms, raised intraocular pressure, and progressive restrictive lung disease [11]. In Gordon's original family (1969), with 6 affected individuals over three generations, clubfoot occurred in 5, camptodactyly in 4, and cleft palate in 3; penetrance is reduced, more so in women than in men, cleft palate is the least penetrant feature (it may be limited to a bifid uvula), and there are reports of mild intellectual disability and of arthrogryposis visible on ultrasound at 13 weeks [55]. In DA5, ocular findings include ptosis, astigmatism, keratoconus or keratoglobus, pigmentary maculopathy, retinal folds, and electroretinogram abnormalities; the restriction of eye movements is attributed to fibrosis and contracture of the extraocular muscles, so the term "ophthalmoplegia" is considered misleading [57]. In one DA5 family, the proband had pulmonary hypertension due to severe restrictive disease (forced vital capacity 30% and total lung capacity 51% of predicted) with alveolar hypoventilation aggravated by altitude, and bilateral absence of the anterior cruciate ligament with knee subluxation and weak or absent Achilles and patellar reflexes have been described [57]. Marden-Walker syndrome, described in 1966 in an infant who died at 3 months, combines an immobile face with blepharophimosis, micrognathia, high or cleft palate, kyphoscoliosis, contractures, and psychomotor delay, sometimes with Dandy-Walker malformation, brainstem and cerebellar hypoplasia, agenesis of the corpus callosum, or cystic kidney disease; in a review of 22 cases most were sporadic, with two sibships and three children of consanguineous couples, and only one patient has a confirmed PIEZO2 variant [58]. A 2.84-Mb deletion at 21q22.11 has been described in a child with a Marden-Walker phenotype [58].

DAIPT manifests at birth with hypotonia, respiratory and feeding difficulties, and contractures, with motor delay, progressive scoliosis, sometimes hip dislocation and laryngomalacia, and later with markedly reduced proprioception, sensory ataxia with dysmetria worsened without visual control, loss of discriminative touch, areflexia, and sensory neuropathy on electrophysiological studies [11]. In a multicentre cohort of 42 molecularly diagnosed patients published in 2026, the 16 biallelic patients (median age at diagnosis 8 years, 0.9–14) had neonatal hypotonia, respiratory and feeding difficulties, DA, and multisystem involvement (scoliosis, hip dislocation, laryngomalacia), and the 25 monoallelic patients (median 7 years, 0.3–39) a broad spectrum; there was one prenatal diagnosis at 23 weeks in a heterozygote, and respiratory involvement was frequent in both forms [109]. In the first reports, walking was achieved between ages 6 and 7 years in two patients, at around 5 years in 10 patients from 4 consanguineous families (some never walked independently), and, in a Turkish patient, only at 16 years, after scoliosis surgery; discriminative sensation, proprioception, and vibration sense were more reduced in glabrous than in hairy skin, and there was mild cognitive impairment in 3 patients from 2 families [59]. In a series of 9 patients from 8 families (Rome and Santiago, Chile), one patient had only a maternal frameshift variant with a suggestive phenotype and another a large homozygous CNV [126].

3.7. DA5D (ECEL1)

DA5D presents with severe camptodactyly of the hands, with the 3rd to 5th fingers flexed and adducted thumbs and wrists, mild camptodactyly of the toes, clubfoot or calcaneovalgus, and extension contractures of the knees; fixed knee extension, with or without scoliosis, is the most useful clue within the DA group [13, 47]. There is muscle atrophy predominantly in the lower limbs and tongue (furrowed tongue), a round face, arched eyebrows, a bulbous upturned nose, micrognathia, and characteristic ptosis, typically unilateral or asymmetric, without ophthalmoplegia, which separates it from DA5 [4, 13]. In two sisters aged 42 and 46 years, ophthalmological follow-up showed myogenic ptosis with poor levator function, limited supraduction, lagophthalmos, refractive error, corneal scarring and vascularisation, and lower eyelid retraction [13]. In a series of 10 patients from 6 families, all had flexion contractures of the 3rd to 5th fingers and limited knee flexion; all but one had talipes valgus and hip involvement; the lower limbs were more affected than the upper limbs, with no major elbow or shoulder contractures and no pterygia; the index finger was spared or less affected; and central tongue atrophy gave it a furrowed appearance; scoliosis was frequent and 4 had restrictive respiratory insufficiency; ptosis was present in most, with normal ocular motility (except for one patient with Duane syndrome), and muscle MRI showed severe fatty replacement in the thigh, especially of the biceps femoris, sartorius, and vastus lateralis [60]. Scoliosis or kyphoscoliosis, flat or cavus foot, facial weakness, a hypophonic and nasal voice, motor axonopathy on conduction studies, and selective fatty infiltration of the thigh and leg on MRI may coexist; development and intellect are usually normal [13]. In the absence of cognitive impairment or other major malformations, lack of knee flexion even after intensive physiotherapy is a suggestive sign [4].

3.8. DA7, trismus-pseudocamptodactyly syndrome

The anomalies result from short muscles and especially short tendons: limited mouth opening, sometimes with enlarged coronoid processes, and short flexor tendons, so that wrist dorsiflexion produces partial finger flexion that disappears with wrist flexion (pseudocamptodactyly) [10, 82]. Short foot flexors (claw toes, equinovarus, calcaneovalgus, metatarsus adductus) and short hamstrings and gastrocnemii may coexist [82]. Trismus may be subtle and go unnoticed; in infancy it may progress to the point of limiting feeding to liquids, with a risk of aspiration [10]. The small mouth causes feeding, dental hygiene, and occupational problems, and tonsillectomy or intubation may pose serious problems [82]. In a large Dutch family, measurement of mouth opening and wrist angles sufficed to separate 24 affected from 19 unaffected individuals; in a six-generation pedigree, examination of 31 of the 53 affected individuals showed highly variable expressivity without reduced penetrance, and there are reports of dysphagia due to manometric abnormality and of cutaneous syndactyly [62]. A Carney complex variant with the same variant was reported in one family, but families screened subsequently had no Carney features, and the variant was absent in 49 independent cases of Carney complex [10].

3.9. DA8 and MYH3-related pterygium and fusion syndromes

Autosomal dominant multiple pterygium syndrome (formerly DA8; CPSFS1A, contractures, pterygia, and spondylocarpotarsal fusion syndrome 1A, also designated CPSKF1A; OMIM 178110) combines pterygia, camptodactyly of the hands, vertebral fusion defects, and scoliosis; it was associated with MYH3 in 3 of 4 families [47, 63]. In spondylocarpotarsal synostosis syndrome (fusion of several vertebral bodies, carpal and/or tarsal synostoses, short stature and scoliosis or lordosis, without rib anomalies), a few autosomal dominant families with heterozygous MYH3 variants have been described, MYH3 also being the gene responsible for cases of DA1, DA2, DA2B and multiple pterygium syndrome [127]. The dominant and recessive (CPSFS1B, OMIM 618469) forms share proximal and distal contractures, pterygia of the neck, axillae, elbows, or knees, vertebral fusions with scoliosis, carpal and tarsal fusions, and short stature, with substantial overlap; severe biallelic loss of function has been described with a lethal presentation [9]. The first dominant family (McKeown and Harris, 1988), a mother and three children, showed variability ranging from a DA5-like picture to an Escobar syndrome–like picture, with ptosis, severe scoliosis due to hemivertebrae, vertebral and carpal fusions, and a nasal voice; vertebral fusions progress with age, from disc space narrowing in childhood to extensive fusions with severe scoliosis in adulthood, and several patients had previously been diagnosed with spondylocarpotarsal synostosis syndrome [64]. In one family, the child had a bicuspid aortic valve with mild regurgitation, mild supravalvular stenosis, a ductus arteriosus that required surgery, and an episode of respiratory failure due to restrictive disease, whereas the mother, who carried the same variant, had normal stature and mobility [64]. In a cohort of 16 patients with contractures, pterygia, and fusions who were negative for FLNB, 4 patients from 3 families were heterozygous (CPSFS1A) and 6 patients from 4 families compound heterozygous (CPSFS1B), all with the non-coding 5' untranslated region splice donor variant as the second allele [64, 65]. A family with DA8, a father and daughter without antecubital or popliteal pterygia or vertebral fusions and with a ventricular septal defect, had no MYH3 variant and may represent a distinct entity [64].

3.10. Escobar syndrome (non-lethal multiple pterygium)

The phenotype includes short stature; ptosis with downslanting palpebral fissures, epicanthal folds, hypertelorism, micrognathia with downturned corners of the mouth, limited mouth opening, long philtrum, cleft palate, a flat, expressionless face, and low-set ears; pterygia of the neck, axillae, and antecubital, popliteal, and intercrural regions; camptodactyly, syndactyly, equinovarus or rocker-bottom feet; cryptorchidism in boys and absent labia majora in girls; scoliosis, kyphosis, vertebral fusions, rib anomalies, and absent or dysplastic patellae [30]. Occasional findings include FGR, anterior clefts of the vertebral bodies, rib fusions, radial head dislocation, vertical talus, hip dislocation, conductive hearing loss, abnormal ossicles, diaphragmatic eventration, hypospadias, and heart defects [30]. In patients with a CHRNG variant, arachnodactyly and, sometimes, congenital respiratory distress are also described [5].

The frequency of miscarriages and stillbirths is increased in affected families; most patients become ambulatory and have normal intelligence; respiratory problems (pneumonia, episodes of dyspnoea and apnoea, presumably due to kyphoscoliosis and a small thorax) cause significant morbidity and death in the first year in about 6% [30]. The features often emerge after infancy: pterygia become more evident with growth and lead to fixed contractures, camptodactyly often improves, and scoliosis appears before age 5 in most, frequently requiring spinal fusion [29, 30]. In infancy, the typical facial appearance may be absent and the capillary haemangioma fades [29]. Long-term follow-up of 7 Spanish and French patients with CHRNG variants, compared with 57 published patients, showed a stable course over the years, postnatal neuromuscular junction changes on biopsy, and, on whole-body MRI, marked reduction of muscle mass, especially of the erector spinae and gluteus maximus, with fatty infiltration limited to the deep paravertebral muscles and the distal lower limb muscles, a pattern that helps distinguish this form from other arthrogryposes [110]. The PNDS describes atrophy of the peroneal compartment in CHRNG-related AMC [4]. In the series of 7 patients from 6 families, with a mean follow-up of 10.3 years and a mean age of 15.9 years (4–40), all had pterygia (neck, axillae, and interdigital spaces) and contractures, independent walking was achieved at a median of 15 months, all required orthopaedic treatment, surgical or otherwise, and no weakness or fatigability developed; electromyography, performed in 5, showed a non-specific myopathic pattern in 2 and was normal in 3, and whole-body MRI was performed in 6 [69, 110]. The thoracic and lumbar erector spinae were always thin and, in most, severely infiltrated, more so in the thoracic region [69]. Facial hypotonia and hypomimia are frequent in Escobar syndrome [4]. Among more than 350 infants with arthrogryposis studied by Hall and colleagues in 1982, 11 had limb pterygia and congenital contractures: 7 with the non-lethal recessive form and 3 siblings with the lethal form [69]. Other findings described in isolated reports are aortic root dilatation with aortic regurgitation from birth, a tongue of peculiar shape on protrusion ("lingua cochlearis"), umbilical hernia with hypoplasia of the periumbilical skin in a large Arab family without lower limb pterygia, severe muscle atrophy, and, in mild forms, confusion with Noonan syndrome; EMG is usually normal, without myasthenic symptoms, and a relatively high frequency has been noted in Kuwait [69].

3.11. Lethal multiple pterygium syndrome

LMPS features growth deficiency of prenatal onset; epicanthal folds, hypertelorism, flat nose, cleft palate, small mouth, micrognathia, and low-set, malformed ears; flexion contractures of the elbows, shoulders, hips, knees, ankles, hands, and feet; mentosternal, cervical, axillary, antecubital, crural, popliteal, and ankle pterygia; small thorax, cryptorchidism, hypoplastic dermal ridges, nuchal oedema, and lax skin; and thin, gracile long bones with hypoplasia of the vertebrae, sacrum, ilia, ribs, clavicles, and scapulae [70]. Occasional findings include intestinal malrotation, cardiac hypoplasia, diaphragmatic hernia, hydronephrosis, and neuropathological anomalies (cerebellar and pontine hypoplasia, polymicrogyria, ventricular dilatation) [70]. All patients are stillborn or die in the immediate neonatal period, probably from pulmonary hypoplasia; polyhydramnios occurs in about one-third and hydrops in more than half [70]. In a 1999 review, the literature since 1976 totaled 43 fetuses with LMPS in 27 families (24 male and 19 female), 13 of which had only affected boys; the only family with reasonably clear X-linked transmission was that of Tolmie and colleagues (1987) (OMIM 312150) [71]. Autopsy studies suggest that LMPS combines a jugular lymphatic obstruction sequence with severe early fetal akinesia, and a lethal form has been described in siblings with malignant hyperthermia as a major complication [71]. Two forms are distinguished: an early form, with intrauterine death in the second trimester, hydrops and/or cystic hygroma, genetically heterogeneous (recessive and X-linked), and a late form, which survives into the third trimester without hydrops, with familial cases consistent with recessive inheritance [70]. Signs of fetal akinesia may be seen from 12 weeks, indicating a poor prognosis [31].

3.12. FADS and the Pena-Shokeir phenotype

The phenotype includes prenatal growth deficiency with frequently spared head circumference, a rigid, expressionless face, prominent eyes, hypertelorism, telecanthus, small, poorly folded ears, a depressed nasal tip, small mouth, high palate, and micrognathia; multiple ankyloses, ulnar deviation, rocker-bottom feet, equinovarus, camptodactyly, and absent dermal ridges and flexion creases; pulmonary hypoplasia, cryptorchidism, an apparently short neck, polyhydramnios, short gut with malabsorption, a small placenta, and a relatively short umbilical cord [72]. About 30% are stillborn and most liveborn infants die of pulmonary hypoplasia in the first month; term infants are invariably small for gestational age [72]. On ultrasound, severe akinesia produces, depending on gestational age, increased nuchal translucency or hydrops, little or no spontaneous movement, contractures (especially clubfoot), pterygia of the neck, elbows, and knees, pulmonary hypoplasia, cleft palate and micrognathia, FGR, polyhydramnios, and breech presentation [31]. Many cases with CNS changes due to vascular compromise are sporadic [28]. In a 1988 review, among about 60 published cases, only 5 had survived beyond 28 days, plus one case surviving to 20 months; in 1986 Hall distinguished at least 5 subgroups among 16 families with multiple affected individuals [73]. When the mother has anti-AChR antibodies, even if asymptomatic, the recurrence risk is high, with no healthy child described after an affected pregnancy; serum and IgG from these mothers blocked fetal AChR function by more than 90%, but not adult AChR function, which explains why the mothers are spared [73]. On ultrasound, scalp oedema and pulmonary hypoplasia distinguish FADS from trisomy 18, whereas arrhythmias, a prominent occiput, micrognathia, and omphalocele point to the latter; at autopsy, paucity of anterior horn cells, diffuse muscle atrophy, "miniature"-type adrenal hypoplasia, and perinatal long-bone fractures due to disuse osteopenia are described [73].

3.13. LCCS1 and CAAHD

LCCS1 presents with total fetal akinesia detectable from 12–13 weeks, hydrops, micrognathia, pulmonary hypoplasia, pterygia, and multiple contractures (typically elbow flexion and knee extension), with invariable death before 32 weeks; CAAHD is less severe, with fetal akinesia, arthrogryposis, and motor neuron loss [14, 26]. Histopathology shows loss of anterior horn motor neurons, thin bones, and "herringbone" ribs [31]. The original Finnish description (1985) comprised 16 cases from 10 sibships whose grandparents came from neighbouring rural communities in northeastern Finland; the 1979–1992 national study identified 40 fetuses and infants from 26 families (20 of each sex), with an estimated incidence of 1/19,000 births — 16 terminations after prenatal diagnosis of total akinesia with hydrops, 19 stillbirths, and 5 liveborn infants who died within the first hour — and the 1987–2002 study classified 39 of 115 cases of lethal arthrogryposis as LCCS, with a prevalence of 1/25,250 [76]. CAAHD, without geographic clustering in the northeast, was described in 15 infants from 11 Finnish families with a mean gestational age of 34 weeks (3 stillborn, 5 died within the first hour, 6 in the first days, one at 20 days of life), with mild hydrops and FGR when present and a macroscopically normal spinal cord; outside Finland there are survivors up to age 12 years with tracheostomy, gastrostomy, and global developmental delay, and a child who walked at age 4 [17]. The distinction between LCCS1 and CAAHD is controversial: one group considers them a genotype–phenotype correlation within a single spectrum, another considers LCCS1 a distinct entity, with prenatal death and absent fetal movements in all cases [17].

3.14. Other neurogenic, myopathic, and syndromic forms

In X-linked infantile SMA, weakness is often of prenatal onset, with polyhydramnios and reduced fetal movement; the most consistent features are anterior horn cell disease and contractures, mainly digital, with or without fractures; intellect is normal and life expectancy is greatly shortened by progressive ventilatory failure [16]. Congenital myotonic dystrophy is associated with AMC in up to 20% of cases (hips, knees, shoulders, and elbows, often without distal involvement), with polyhydramnios, severe hypotonia, and respiratory problems with diaphragmatic hypoplasia [4, 128]. Schaaf-Yang syndrome, caused by truncating variants in MAGEL2 (a maternally imprinted gene in the 15q11–q13 region), combines a phenotype overlapping that of Prader-Willi syndrome (developmental delay, hypotonia, feeding difficulties, autism) with contractures, ranging from fetal akinesia and AMC to contractures of the small finger joints, in 18 individuals from 14 families [3, 129]. ARC syndrome (arthrogryposis, renal dysfunction, and neonatal cholestasis with bile duct hypoplasia and low to normal γ-glutamyltransferase) is recessive (VPS33B, ARC1; VIPAS39, ARC2), with growth failure and death usually in the first year [45, 130, 131].

4. Clinical Recognition and Red Flags

4.1. Prenatal period

The ultrasound red flags are decreased fetal movement (hypokinesia or akinesia), persistent malposition of one or more limbs, pterygia, anhydramnios or oligohydramnios, and polyhydramnios; in very early and severe forms there may be first-trimester hydrops, and increased nuchal translucency or a cystic hygroma may be the only initial signs [4, 31]. The ultrasound picture is rarely specific, so a formal aetiological diagnosis is not possible from prenatal imaging alone; AMC is suspected on screening ultrasound in fewer than one-third of affected pregnancies [4]. In the general obstetric population, routine ultrasound practice identifies about 25% of fetuses with arthrogryposis before 24 weeks, and among 107 individuals with amyoplasia the prenatal diagnosis was missed in 73.8% [132, 133]. When the diagnosis was made, it was due to identification of bilateral clubfoot on ultrasound or to maternal perception of reduced movements, which makes these two situations the main starting points [133]. When fetal clubfoot is found, a complete assessment of the movement and range of motion of all joints is recommended to rule out arthrogryposis: in a Chinese series of 69 fetuses with AMC, 50.7% had clubfoot [134, 135].

4.2. Neonatal period and infancy

At birth, AMC is suspected when there is joint limitation at two or more distinct levels, but it may be very mild (some DA) or masked by hypotonia, malformations, seizures, or respiratory distress [4]. Nonspecific morphological signs of akinesia — low-set ears, micrognathia or retrognathia, high-arched or cleft palate, short neck — reflect the severity and early onset of the movement deficit [4]. Absent flexion creases indicate absence of prenatal movement of that joint, and skin dimples over the joints and pterygia should be sought systematically [3, 4]. In Escobar syndrome, infants may present only as AMC, with subtle pterygia and without the typical facial appearance, which emerges over the first years; the authors of the published case therefore recommend following infants with AMC [29].

4.3. Emergency signs

Neonatal respiratory distress, due to pulmonary hypoplasia, diaphragmatic weakness, kyphoscoliosis, or a small thorax, is the main cause of early death and requires monitoring in FADS, Escobar syndrome, the PIEZO2-related forms, and X-linked SMA [11, 16, 30, 72]. Sudden episodic apnea with cyanosis, stridor, and bulbar weakness suggest CMS, whose identification is a priority because specific treatments exist [4, 37]. Limited mouth opening (trismus) and micrognathia herald a difficult airway at intubation, and the trismus of DA7 may progress during infancy to the point of limiting feeding to liquids, with a risk of aspiration [3, 10]. Fractures occur at delivery in 5–10% of AMC and 10.5% of amyoplasia and may be occult: in one newborn, the sudden release of a fixed upper limb contracture during axillary hygiene revealed a spiral fracture of the humeral shaft, which healed with conservative treatment [3, 124, 136]. Joint examination must be performed with great care to avoid causing fractures, supporting the entire long bone and avoiding rotational forces [3, 136]. A history of myopathy, of RYR1 variants, or of multiple pterygium with malignant hyperthermia requires anaesthesia planning before any procedure [3, 4, 31].

4.4. Newborn screening

The sources consulted describe no newborn screening programmes targeting AMC. 5q SMA rarely presents as AMC (SMA type 0) and is detectable by newborn screening.

5. Diagnosis: Criteria, Testing, and Algorithms

5.1. Criteria and objectives

The criterion for AMC is clinical: congenital joint contractures in at least two different body areas, excluding isolated single-level contractures (isolated camptodactyly, isolated clubfoot, isolated hip dislocation), fetal hypokinesia without contractures, and noncongenital, progressive, and postnatal limitations [1, 4]. Spina bifida does not meet the definition, although it may be accompanied by clubfoot, contractures, and hip dislocation, because it is a neural tube malformation with a distinct aetiology; SMA types 0 and 2 may present with AMC, which was an exclusion criterion of the International SMA Consortium [1]. Constitutional bone and connective tissue disorders with multiple contractures meet the clinical definition, because the joint limitation is primary [4]. The diagnostic workup seeks to confirm AMC, classify it as amyoplasia, DA, or the third group, look for associated malformations, establish the causal diagnosis, guide care, enable genetic counselling, and organise psychological support; coordination is usually by a physical medicine and rehabilitation (PM&R) physician, a paediatric neurologist, or a geneticist [4]. Even after careful evaluation, a specific diagnosis is reached in only about 50% of cases in the clinical experience described in the Oxford Desk Reference [3].

The PNDS criteria for typical amyoplasia, adapted from Hall 2014, are as follows [4]:

5.2. History

The three-generation pedigree looks for clubfeet, joint dislocations, congenital hip dislocation, contractures, hyperextensibility, consanguinity, and affected siblings [3, 31]. The maternal history looks for myotonic dystrophy, myasthenia gravis (which may be asymptomatic), diabetes, uterine anomalies, infection and fever >39 °C, early amniocentesis (<14 weeks), multiple pregnancy, and drugs such as misoprostol or cocaine, which can cause intrauterine ischaemia [3, 31]. Relevant points in the pregnancy history are the perception of fetal movements, polyhydramnios (which may result from impaired fetal swallowing), oligohydramnios, or loss of amniotic fluid; at delivery, abnormal presentation, difficulty due to fixed joints, and fractures, which occur in 5–10% [3]. Recurrent affected pregnancies with no completely normal child among affected siblings may be due to transplacental antibodies from maternal myasthenia [31]. Finnish ethnicity points to GLE1 [31].

5.3. Clinical examination

Photographs document the position of the contractures at birth and over time. The resting position and the passive and active range of motion of each joint, including the jaw and spine, are examined with great care to avoid causing fractures; dimples, flexion creases, pterygia, mouth opening, tone, reflexes, and muscle texture (fibrous band, firm, or soft) are recorded, as are other anomalies such as cleft palate or jaundice; both parents are examined for contractures or absent flexion creases, and the mother for myotonic dystrophy or myasthenia [3, 31]. The location of contractures is rarely specific, with two exceptions: the upper limb pattern of amyoplasia and the lack of knee flexion even after intensive physiotherapy in ECEL1-related AMC; amyoplasia, DA, and the multiple pterygium syndromes have a distal predominance [4]. Response to physiotherapy is informative: it is rapid in oligohydramnios or anhydramnios and in most connective tissue AMC (Larsen, Ehlers-Danlos, Beals), good in amyoplasia and DA, and much more difficult in the neurogenic forms [4]. Pterygia reflect early and severe lack of movement and become more visible with long-bone growth; facial hypotonia points to FADS, Escobar, or myopathic forms, and facial hypertonia to sarcomeric DA or rare syndromic forms (CLIFAHDD, Schwartz-Jampel) [4]. Oculomotor disturbances and ptosis, sought especially in the absence of CNS involvement, are constant in gain-of-function PIEZO2 DA and in biallelic ECEL1 disease [4]. Signs of myasthenia (weak suck and swallow, ophthalmoplegia, ptosis, stridor, laryngospasm) should always be sought [4].

5.4. Nongenetic investigations

Creatine kinase (CK, or CPK) is markedly elevated (>10 times the upper limit of normal) in rare AMC due to dystrophies (especially GMPPB), anterior horn disease (BICD2), or musculocontractural Ehlers-Danlos syndrome (CHST14), and moderately elevated in TRPV4-related neuropathic AMC, TTN-related congenital myopathies, or TNNI2-related DA; a normal CK does not exclude a neuromuscular form, and because CK is physiologically elevated after delivery it should be measured or repeated well after birth [4]. The systematic search for malformations, by expert consensus, includes echocardiography, abdominopelvic ultrasound, ophthalmological examination, and skeletal radiographs; heart defects occur in syndromic forms, myopathies, dystrophies, and myotonic dystrophy [4].

Muscle MRI, preferably whole-body (1.5 T or 3 T, axial slices, at least T1-weighted), assesses the presence of muscles, fatty infiltration and its pattern, and atrophy: marked atrophy of the sartorius and biceps femoris in ECEL1, of the peroneal compartment in CHRNG, nonprogressive atrophy up to complete fibroadipose replacement of certain muscles in neurogenic distal SMA, and absence or atrophy of the deltoid, biceps brachii, and brachialis in amyoplasia [4]. In ECEL1, fatty infiltration affects the tensor fasciae latae, hamstrings, quadriceps, gluteus maximus, and leg compartments [13]. Muscle ultrasound or MRI of an affected limb documents replacement of muscle by fat and fibrous tissue when amyoplasia is being considered [3].

Brain imaging is indicated for any neurological abnormality (axial or peripheral tone, alertness): brain MRI first-line if involvement is clear, transfontanelle ultrasound first in doubtful cases and in the perinatal period (no anaesthesia, but less sensitive), and spinal MRI at the same time (expert consensus, without evidence of yield) [4]. The proportion of AMC with CNS involvement ranges from 11.4% to 48% across studies [4]. Electroneuromyography (ENMG), performed in neuromuscular centres, is a second-line test, with a yield of up to 100% in the neurogenic forms and 34% to 56% in the myopathic forms; it is proposed in severe neonatal hypotonia, is invasive and age-dependent, and is of little use in amyoplasia because of the very low muscle volumes [4]. In DA5D, conduction studies show motor axonopathy rather than a myopathic pattern [13]. In MYBPC1-related myopathy with tremor, which mimics SMA, preserved reflexes and normal conduction velocities argue against SMA [12].

Muscle biopsy has shown added value only in congenital myopathies with AMC (TPM2, MYH2, recessive TNNT3); it is normal or nonspecific in sarcomeric DA, ECEL1, and loss-of-function PIEZO2 [4]. In severe postnatal forms with a poor vital prognosis, muscle and skin biopsy are performed immediately; in the others, in the absence of a lead, they may be deferred, for example to a scheduled surgery [4]. In recessive TNNT3 forms it shows selective type 2 fibre atrophy, rods, and absent fast troponin T on western blot, and in recessive TPM2, absent β-tropomyosin; it supports, but does not confirm, the diagnosis [6, 8]. In Escobar syndrome, postnatal biopsy shows neuromuscular junction changes, but the authors of the published case consider it of little value because the γ→ε switch has already occurred [29, 110].

Metabolic testing is reserved for severe neurological presentations, together with a metabolic specialist or paediatric neurologist: plasma and urine amino acids, organic acids, lactate, pyruvate, and ammonia, and, if respiratory chain disease is suspected, MRI with spectroscopy and, as a last resort, skin or muscle biopsy [4]. Among 300 patients with confirmed respiratory chain deficiency, two had congenital contractures and one had AMC; Leigh syndrome may mimic amyoplasia before neurological signs appear, and in Gaucher disease with hydrops arthrogryposis occurs in about 19.5% [4]. TORCH serology (toxoplasmosis, others — syphilis, varicella-zoster, parvovirus —, rubella, cytomegalovirus, and herpes simplex), with enterovirus and Coxsackie virus, is indicated in the newborn or before 6 months of age; hip ultrasound is performed in the newborn if there is clinical suspicion [3].

5.5. Genetic testing

The 2019 reviews recommend that genetic studies — exome or genome in all patients and microarray in patients with intellectual disability — be preferred as first line over electromyography (EMG) and muscle biopsy [32, 118]. AMC panels achieve a yield of 40–45% depending on the clinical lead, and WES or WGS without prior analysis, more than 50% [4]. CMA is informative when there is intellectual disability, multiple malformations, or morphological variants, and should be performed at the slightest doubt or even systematically; its role is likely to be redefined in light of WGS [4]. Trisomy 18 is the most frequent chromosomal cause, with distal involvement (clubfeet or convex feet, clenched fists, overlapping fingers), and should be suspected when there is a heart defect, brain malformation, or FGR; others are trisomies 13 and 21, recurrent unbalanced translocations involving 9q and 8, and microdeletions, especially of 5q23 [4]. Mosaic trisomy 8 may require a fibroblast karyotype, and testing for chromosomal mosaicism (e.g., diploid/triploid) in skin is indicated in fetal akinesia [3, 31]. DNA should always be banked, because the list of known genes keeps growing [3].

These figures are diagnostic yields in selected cohorts, with differing case definitions, proportions of consanguinity, and periods, and not estimates of the clinical sensitivity of a test. In the Turkish cohort, the text reports the phase II yield as 65.2% (58/89) in the results and 65.1% in the discussion [35]. The yield of WES in AMC in the literature ranges from 35% to 73%, far above the roughly 5% of microarray, and WGS detects CNVs and structural and noncoding variants and may replace most tests with a shorter turnaround time [33]. In the Dutch cohort, skeletal muscle genes (DNM2, KLHL40, NEB, RYR1, SCN4A, TPM2) explained 42.9% (6/14) of group 3 diagnoses and neuromuscular junction genes (CHRNG, MUSK) 21.4% (3/14); none of the six fetuses with amyoplasia had a genetic anomaly, and reanalysis yielded no additional diagnoses [33]. In the French series, WES added value through the broader clinical spectrum of some genes and the identification of new genes (CNTNAP1, MAGEL2, ADGRG6, ADCY6, GLDN, LGI4, LMOD3, UNC50, SCN1A, and expansions in ASXL3 and STAC3), which accounted for 21% of causal genes [102].

5.6. Test limitations

A negative result does not exclude a genetic cause. The DMPK CTG expansion is not detected by next-generation sequencing and requires a specific request in all AMC with severe neonatal hypotonia, as does SMN1/SMN2 analysis; Prader-Willi syndrome, with late prenatal hypotonia and transient, reducible limitations, requires methylation testing [4]. Standard exome sequencing does not capture the hypomorphic MYH3 allele NM_002470.4:c.-9+1G>A, located in the 5' untranslated region [9, 103]. The repetitive regions of NEB and TTN are refractory to exome mapping and sequencing; version 3 of the Australian panel improved coverage of these regions and CNV detection [34]. In DAIPT, a large homozygous CNV and an unidentified second allele illustrate that the exome may miss the second event [126]. WGS identified an intronic RYR1 variant as the second allele in early-onset FADS, and CRISPR activation revealed the splicing effect of an intronic NEB variant in fetuses with AMC, both described in 2026 [139, 140]. WES with CNV calling had not been implemented during the period of the Dutch cohort [33]. Mosaicism and oligogenicity, documented in the Turkish cohort, complicate interpretation [35]. Heterozygous loss-of-function MYH8 variants should not be interpreted as causal for DA7 [10].

The report should state the gene, the versioned transcript (preferably MANE Select), the variant in HGVS nomenclature, zygosity, the phase of biallelic variants, and the class according to the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria; in the Australian cohort, only class 4 and 5 variants were reported, except for two variants of uncertain significance (VUS) that were very likely causal [34]. Collaboration between diagnostics and research and functional genomics enabled VUS reclassification and phenotype expansion (CACNA1S, CHRNB1, GMPPB, STAC3) [34]. Parental segregation studies and Sanger confirmation are part of the process, and variants in unresolved families warrant reanalysis and sequencing of extended relatives [29, 35]. Rapid RNA analysis by amniocyte transdifferentiation was proposed in 2026 to resolve ambiguous variants prenatally [141].

5.7. Testing for maternal causes

In maternal myasthenia, whose incidence is 1–5 per million per year, the antibodies are mainly anti-AChR and rarely anti-MuSK or anti-LRP4; transplacental transfer causes transient neonatal myasthenia in about 10–15% of newborns and, rarely, a fetal AChR inactivation syndrome with persistent myopathy [4]. Antibodies specific to the fetal γ subunit are associated with fetal arthrogryposis, ranging from mild contractures to perinatal death; skeletal anomalies occur in up to 3.9% of children of mothers with myasthenia, and maternal anti-AChR antibodies have been found in up to 7.1% of FADS or AMC [4]. The test is inexpensive and was positive in 1–2% of mothers identified through a single affected infant; the mother may produce antibodies against the fetal receptor without having the disease [3, 142]. Maternal anti-AChR antibodies should be measured as early as possible and maternal serum stored for testing for other antibodies, including those directed against the fetal γ subunit when anti-AChR antibodies are positive [4].

5.8. Diagnostic algorithm

In lethal forms, autopsy with detailed neuropathology (brain, spinal cord, and nerves), muscle biopsy with electron microscopy, and banking of fetal and parental DNA give the best chance of a diagnosis; if autopsy is declined, a limited evaluation with muscle biopsy and skeletal radiographs is useful [3, 31]. A standardised autopsy protocol for arthrogryposis, with a registry and tissue bank, has been published [143]. In 30 consecutive cases of FADS with a normal karyotype in Leuven, extensive neuropathological study yielded a specific diagnosis in 16 (53%) — CNS in 9, spinal cord in 1, primary myopathy in 3, and syndromic in 3 —, neuromuscular findings without a diagnosis in 10 (33%), and normal results in 4 (13%) [144].

5.9. Prenatal diagnosis

Real-time ultrasound of fetal movement allows prenatal diagnosis of many forms; the Oxford Desk Reference recommends scans at 16, 20, 24, and 32 weeks in at-risk pregnancies, assessing the range of motion of each major joint, and notes that the limitations of ultrasound and the small chance of missing a significant contracture must be discussed [3]. Without a known genetic basis, prenatal diagnosis relies on serial ultrasound from 12 weeks [28, 31]; DA contractures are detectable at 18–24 weeks [7]; in one DA1 family the ultrasound diagnosis was made at 18 weeks from the extended wrist and clenched fingers throughout the examination, and in one DA3 family arthrogryposis was visible at 13 weeks [48, 55]. In LMPS, cystic hygroma and hydrops allow prenatal diagnosis in most pregnancies following an index case, and a hygroma interpreted as a cervical meningocele at 20 weeks delayed the diagnosis in one family, underscoring the importance of fetal examination after the loss [71]. Fetal clubfoot requires assessment of all joints and of movement: in an Oxford series of 149 cases of prenatally detected clubfoot (1991–2003), there were six false positives and 110 (74%) isolated anomalies, and in a Harvard series of 68 fetuses, 8 (11.8%) false positives and 4 (5.9%) abnormal karyotypes [135]. In the EUROCAT registries, the sensitivity of prenatal diagnosis was 69% (360/521 cases with information) [20]. Prenatal detection was 26% (28/107) in amyoplasia born after 1990, 53% (21/40) in infants in groups 1 and 2 of the pilot registry, 37% in a cohort of 301 infants in groups 1 and 2, and 100% in a high-risk population (n = 66) with serial fetal motor assessment, structural ultrasound, and a multidisciplinary team (indirect attributions from Arduç 2026) [33, 133].

The Netherlands has a seven-step care pathway, in an expert centre for FADS and AMC recognised since 2015, with a systematic 15-minute motor assessment before 24 weeks, the legal limit for termination of pregnancy in that country, which analyses movement patterns, quantity (frequency of general movements), and quality (variation in amplitude, speed, joint participation, and direction); movement is considered abnormal if at least the quality is abnormal [33]. Most fetuses with FADS do move: in half, quantity and differentiation were normal and quality was abnormal in all, and akinesia was observed in only 13 cases; in late-onset AMC, motility may not yet be reduced, and serial ultrasound is needed [33]. Amyoplasia accounted for 9% (6/64) of the prenatal cohort, compared with 25–30% of AMC in the postnatal population, because atrophy is difficult to recognise on second-trimester ultrasound; all cases of amyoplasia were diagnosed after birth [33]. Fetal brain MRI is complementary, by expert consensus [4].

In the absence of a lead, prenatal investigation includes CMA, analysis of SMN1, DMPK, and Prader-Willi region methylation, especially if there are clear signs of hypo/akinesia, maternal anti-AChR and anti-MuSK antibodies if an autoimmune cause is suspected, and, when anti-AChR antibodies are positive, antibodies against the fetal γ subunit; molecular diagnosis is by karyotype, CMA, a dedicated panel, or WES [4]. Arduç and colleagues recommend that each centre develop a care pathway for prenatally suspected contractures, code the phenotype with HPO terms, and offer WES when available; karyotype and rapid aneuploidy testing had no yield, microarray 2.5%, and the probability of a chromosomal cause is higher when there are multiple structural anomalies [33]. If termination of pregnancy for medical reasons (TOP) is requested by the parents and approved by the multidisciplinary prenatal diagnosis centre, fetal autopsy with neuropathology and muscle study should be offered systematically [4]. When the molecular basis is known, definitive prenatal diagnosis is by chorionic villus sampling or amniocentesis, and preimplantation genetic testing for monogenic disorders (PGT-M) is an option [4, 31]. No studies or guidelines define when and how to assess fetal movements and contractures, and failure to detect before 24 weeks deprives families of reproductive options [132, 133].

6. Differential Diagnosis

The differential diagnosis operates at two levels: between AMC and conditions that do not meet the definition (isolated contractures, hypokinesia without contractures, spina bifida, acquired limitations), and among the various causes of AMC, where the distinction between amyoplasia, DA, multiple pterygium syndromes, neuromuscular forms, and syndromic forms guides testing and counselling [1, 4].

In disorders of cellular trafficking, arthrogryposis is a prominent sign in vesicular transport defects due to SLC35A3 (a congenital disorder of glycosylation), ERGIC1 (neurogenic arthrogryposis multiplex), and KIAA1109 (Alkuraya-Kucinskas syndrome, with arthrogryposis and Dandy-Walker malformation), and in inter-organelle trafficking defects due to VPS33B and VIPAS39, which cause ARC1 and ARC2 syndromes [149].

DA are distinguished from congenital contractural arachnodactyly by the absence of marfanoid habitus, arachnodactyly, knee and elbow contractures, and "crumpled" ears, and often by pursed lips [39]. In congenital contractural arachnodactyly, physiotherapy leads to gradual postnatal improvement in most patients, progressive scoliosis or kyphosis in 64% is the most relevant health problem, and there is a severe form lethal in infancy with severe heart disease and gastrointestinal anomalies; FBN2 variants cluster in the exon 24–35 region, and skipping of exons 31 and 32 predisposes to aortic dilatation with a risk of dissection [83]. Prenatal ultrasound may reveal AMC or fetal akinesia in congenital contractural arachnodactyly [83].

Among the pterygium syndromes, Escobar syndrome is distinguished from LMPS by survival and the absence of hydrops, and from popliteal pterygium syndrome by the multiple distribution of the pterygia and the absence of lip pits and intraoral bands; the dominant MYH3-related form is accompanied by vertebral, carpal, and tarsal fusions, and the recessive TPM2-related form by nemaline myopathy [6, 30, 63, 145]. The presence of CNS anomalies argues against CHRNG [5, 111]. In mild forms, Escobar syndrome may be confused with Noonan syndrome [69]. In a term or near-term fetus without cystic hygroma or hydrops, the Pena-Shokeir phenotype is more likely than LMPS, because pterygia may be part of the former but hygroma and hydrops are not [71]. Marden-Walker syndrome is distinguished from the Pena-Shokeir phenotype by the absence of pulmonary hypoplasia and from Schwartz-Jampel syndrome by the absence of myotonia, according to a review of 22 cases [58]. Within the PIEZO2 group, cleft palate points to DA3, ocular abnormalities to DA5, and intellectual disability to Marden-Walker syndrome, although all three may represent variable expressivity of the same condition [55, 57]. In FADS, the distinction between intrinsic causes (brain, spinal cord, nerve, neuromuscular junction, muscle, bone, restrictive dermopathies, tendons, joints) and extrinsic causes (bicornuate or septate uterus, fibroids, multiple pregnancy) has practical consequences: secondary AMC responds well to physiotherapy and usually presents with isolated contractures, whereas primary AMC may be accompanied by akinesia, brain anomalies, reduced muscle mass, polyhydramnios, and noncephalic presentation [150]. The fetal akinesia sequence may include ear deformities and a small mandible, but usually without arachnodactyly; the pregnancy history and brain MRI help differentiate it from congenital contractural arachnodactyly [39].

Among the forms with hypotonia, MYBPC1-related myopathy with tremor mimics SMA, and normal reflexes and conduction velocities argue against SMA [12]. X-linked SMA should be tested for when the picture suggests SMA with normal SMN testing and there is evidence of X-linked inheritance; there are several other types of X-linked arthrogryposis, such as Wieacker-Wolff syndrome (ZC4H2) and occipital horn syndrome (ATP7A) [3, 16]. Diastrophic dysplasia has therapeutic relevance, because passive stretching, useful in most AMC, should be avoided [3].

7. Treatment and Risk Management Strategies

7.1. Principles and organisation of care

There is no disease-modifying therapy for AMC, DA, the multiple pterygium syndromes, or FADS; treatment varies with the specific diagnosis and rests on early and sustained rehabilitation, orthoses, serial casting, and orthopaedic surgery as an adjunct [3, 4, 36]. The exceptions with their own aetiological treatment are CMS and 5q SMA [4, 37]. The PNDS organises care around a medical-surgical team (PM&R, orthopaedics, paediatrics) and an allied health team (social worker, dietitian, occupational therapist, geneticist, physiotherapist, pain specialist, general practice, neonatology, neurology and paediatric neurology, obstetrics, speech therapy, orthotist-prosthetist, orthoptist depending on the aetiology, pulmonology, podiatry, psychology, psychomotor therapy, radiology, and rheumatology), usually coordinated by PM&R, with paediatric follow-up [4]. The international definition adds, as appropriate, cardiology, urology, ophthalmology, plastic surgery, dentistry and orthodontics, early transition planning, and genetic counselling [1]. Functional assessment is mandatory before any surgical decision, for example to preserve a tenodesis effect useful for grasp [4].

The physiotherapy and occupational therapy assessment includes pain (EDIN or NFCS scales in infants and a visual analogue scale from age 6; pain is prominent in carriers of PIEZO2 variants), the skin (creases, commissures, dimples, bands, pterygia, syndactylies), passive and active goniometry of each joint of the four limbs and hands, muscle testing (difficult until age 4–5 years, relying on observation against and without gravity, the Active Movement Scale, which is not validated, and video), the spine (rib hump, hyperlordosis, sagittal plane), gait (quantified gait analysis with disappointing results; serial video useful), respiratory function (thorax frequently cylindrical with rolled-in shoulders, especially in amyoplasia), transfers, grasp, and ADL [4]. An AMC-specific global motor functional classification (GMFC-AMC), developed through an international Delphi with French and Spanish adaptation, has a protocol published in December 2025, with no results as of the update date [151].

7.2. International rehabilitation recommendations (2025)

The recommendations of Dahan-Oliel and colleagues resulted from a five-phase integrated knowledge translation process: a survey of 65 therapists from 9 countries (upper and lower limb stretching was the most used intervention, followed by strengthening, orthoses, positioning, activity-based training, and assistive devices); three panels in 2021 with professionals, people with lived experience, and researchers from 10 countries, using the GRADE method adapted to solely observational, low-quality evidence (levels III and IV); a modified Delphi between April and October 2022 (41 participants in the first round, 37 in the second, retention threshold ≥80% agreement); external AGREE II appraisal by five experts (two from Brazil, one from Canada, two from the United States), with an overall quality of 96.6%; and interviews with 15 clinicians and 4 managers [36]. All recommendations are weak ("we suggest"): the desirable effects probably outweigh the undesirable ones, but with appreciable uncertainty [36].

Goals vary with life stage: in infancy, high-intensity intervention across multiple settings, focused on range of motion, activation of movement, orthoses and casts, floor mobility (rolling, sitting), exploratory fine motor skills, and caregiver well-being; in the young child, on transitions, standing and walking, wheeled mobility, play, and early ADL; at school age, on episodic care, school accessibility, transfers, assistive technology, and independence; in adolescents and adults, on needs-based services, a healthy lifestyle (cardiovascular exercise, avoiding weight gain), pain, participation, and transition to adulthood [36, 152]. The stated limitations are the inclusion of only two low- or middle-income countries (Peru and Poland), applicability conditioned by regional, cultural, and socioeconomic factors, and the absence of recommendations on prenatal counselling, transition, intimacy, and overall multidisciplinary medical management [36]. The Swedish cohort of 131 patients supports early multidisciplinary assessment, early stimulation of active movement, and minimisation of immobilisation [125].

7.3. Rehabilitation from birth to age 2–3 years according to the PNDS

The upper limb goals are maximal function, with hand-to-mouth (feeding) and reaching the various parts of the body (hygiene) with grasp; in the lower limbs, ranges of motion that allow stable standing and a stable sitting position without harm to the spine [4]. In the first months, physiotherapy is most often daily and mobilises all joints through all movements, without losing knee extension while working on flexion, below the pain threshold, respecting fatigability and bone fragility (mobilisation under slight traction or joint decompression), with global motor stimulation and prone and side-lying positioning [4]. Positioning orthoses are used from the first days, especially during sleep; because the usual 6-month replacement interval is insufficient, hand splints in the first year should be renewed every 15 days to every 3 weeks if necessary, using lightweight, breathable materials, because hyperhidrosis is common [4].

Table source: PNDS appendices 7 and 8 [4]. In particular cases, with two clubfeet and two irreducible hips, the feet are treated and nonintervention is discussed for the hips; with one irreducible hip, the feet are treated and the hip is then operated on at about 1 year; with one reducible hip, casts and harness, with surgery if these fail; with simultaneous foot and knee involvement, an above-knee cast with progressive correction [4].

After the first months, walking is possible if hip flexion deformity is <30° and knee flexion deformity <20°, with favourable muscular and neurological factors, regardless of the severity of the deformities at birth [4]. In the upper limbs, the minimum goal is "hand-to-mouth" on one side and "hand-to-perineum" on the other, with gross grasp and thumb opposition. In amyoplasia, emphasis is placed on shoulder external rotation; the elbow is usually extended (biceps absent, triceps active), and mobilisation into flexion on at least one side is paramount; supination, wrist realignment, first web-space opening, and finger curling are worked on [4]. The lower limb joint goals are, at the hip, extension to 0°, flexion to 90°, and abduction ≥30° (for sitting on the toilet); at the knee, flexion to 90° and full extension; at the ankle, dorsiflexion to 90° and maximal extension of the first metatarsophalangeal joint [4]. Targeted muscle stimulation is useful in amyoplasia, in which the existing fibres are normal (quadriceps, gluteals), as are hydrotherapy, exoskeletons, and resistance training through play; compensations (mouth, feet) should be respected, moderating those that will cause wear in adulthood, and assistive devices offered in the face of movements potentially harmful at 15–20 years [4]. Stable sitting at 8–9 months is an important milestone; orthoses are used mainly during sleep, except for functional hand orthoses; speech therapy assesses oral motor function, swallowing, chewing, articulation, and praxis [4].

7.4. Hand orthoses and exoskeletons

Hand orthoses are worn continuously in the first months, except during hygiene and affectionate contact, and thereafter only during sleep; night orthoses are always forearm-based (wrist-hand-finger), made of thin (1.6 mm at birth and then 2.5 mm), breathable low-temperature thermoplastic, with the wrist in maximal extension if there is spontaneous flexion, neutral deviation, first web-space opening, preserved palmar arches, fingers in extension or in the intrinsic-plus position, and a counter-support under the first metacarpal to prevent Z-thumb; fastening is with Velcro or cohesive bandage, never adhesive [4]. Imported exoskeletons are generally effective only from age 6–8 years, for reasons of size and assistance, and custom-made systems are useful before then [4]. The Playskin Lift exoskeletal garment was associated with improved object play in children with arthrogryposis, and the WREX upper limb exoskeleton was the subject of a completed outcomes study [153, 154]. From age 3 until the end of growth, night orthoses are maintained throughout growth, alternating one night in two per side if bilateral, with therapeutic breaks only with before-and-after assessment and attention to growth spurts; rehabilitation takes place several times a week [4].

7.5. Orthopaedics by segment

Foot. For rigid equinovarus foot, the PNDS recommends the Ponseti method with systematic, repeatable Achilles tenotomy, adapted so as not to induce knee flexion deformity through casting with the knee flexed; casting may be prolonged, and maintenance with above-knee foot splints, if possible with the knee in extension, is essential; Denis-Browne–type splints are considered illogical in AMC, because they rely on self-correction through movement [4]. In a systematic review with meta-analysis of 5 case series (53 patients, 102 feet), initial Ponseti success in arthrogrypotic clubfoot was 91% (95% CI 0.79–0.96), final success 68% at 5.8 years of follow-up, and recurrence 30% (95% CI 0.14–0.52); the authors, from Brazil, recommend Ponseti as initial treatment, with early detection and treatment of recurrences [155]. In a series of 10 children (17 feet) followed for a mean of 5.8 years, initial correction was achieved in all with a mean of 8 casts (4–10) and Achilles tenotomy in 94.1%, and two-thirds had a plantigrade, functional, painless foot; more casts were needed and the risk of recurrence was higher than in idiopathic clubfoot [156]. At ages 4.0–6.9 years, children with arthrogryposis treated with Ponseti had residual equinovarus and limitations in transfers, mobility, sports, and overall function, but the method produced a braceable foot that postpones invasive surgery [157]. Convex foot (vertical talus) is difficult to treat orthopaedically, with uncertain results, and foot surgery is frequent because of recurrent equinovarus or convex foot [4]. After age 3, extensive posterior releases include tenectomies and wide capsulotomy, with resection of 2–3 cm of the Achilles tendon and also of the flexor hallucis longus, flexor digitorum longus, or peroneal tendons; at the end of growth, tarsal bone surgery (talectomy or triple arthrodesis) remains, with prolonged immobilisation and monitoring for pressure sores [4].

Hip. Well-tolerated bilateral dislocation does not warrant reduction, and it is prudent not to operate, because surgery after failure or recurrence has potentially serious complications (femoral head necrosis, early osteoarthritis); unilateral dislocation should be considered for surgical reduction from 9 months [4]. To sit without harm to the spine, hip flexion ≥100° is required [4]. Lower limb management in amyoplasia requires a very early and intensive approach, surgical techniques adapted to the hip, knee, foot, and ankle, and patient-centred outcome measures; the authors stress the need for specific guidelines [158].

Knee. Knees in flexion deformity or recurvatum are mobilised to maximal extension (walking) and flexion (sitting) and maintained with positioning splints; quadriceps tenotomy is not performed in hyperextended knees [4]. From age 2–3 years to the end of growth, extension osteotomies are indicated if flexion deformity is >40° in ambulatory patients, combined with hamstring tenectomy or release (the hamstrings are usually fibrotic and of no functional value, as assessed by muscle MRI) rather than lengthening, with postoperative splints for at least 6 months; they can be performed early, but there is a risk of loss of correction before age 3 and of a second peripubertal procedure; gradual correction with a circular external fixator is another option [4]. Correction of hip and knee contractures improves walking potential [159].

Upper limb. The role of surgery is limited and individualised; initial treatment with stretching and splints yields significant gains in the first years, and despite lower function, quality of life is comparable to that of age-matched peers [160]. The first option is posterior elbow release (posterior capsulotomy and V-Y triceps tendon plasty) to improve flexion, performed early, before age 2; transfer for active flexion is considered only in light of triceps quality; hemitransfer of the long head of the triceps to the bicipital tuberosity is possible if passive flexion is sufficient for the hand to reach the mouth; transfer of the whole triceps or of the latissimus dorsi is contraindicated, because of progressive elbow flexion deformity with consequences for feeding and hygiene [4]. Later, external rotation osteotomy of the humerus places the hand in a more functional plane, carpal osteotomy for wrist flexion has good long-term results, and wrist arthrodesis is proscribed because of poor results [4]. The 2020 review describes advances in elbow function, wrist repositioning, and thumb position [159].

Surgical timing. In a 2026 systematic review (42 studies, 729 patients, 1,088 joints), mean age at intervention was lowest for the foot and hip and highest for the upper limb and spine; soft tissue releases and tendon procedures predominated early and bony reconstructions later; recurrence was highest after isolated soft tissue procedures of the foot and knee (30–100%) and lowest after osteotomies, muscle transfers, and definitive arthrodeses of the wrist, shoulder, and spine (<20%); heterogeneity and the absence of standardised protocols prompted the preparation of a consensus on the timing and sequence of surgery, not yet published in the sources consulted [161]. Useful secondary bone deformities, especially rotational ones, should not be corrected [4]. In orthopaedic cohorts, more than two-thirds of children with amyoplasia undergo surgery (mean of 5.7 procedures), and Freeman-Sheldon syndrome involves about 10 procedures per individual [44, 50].

7.6. Spine

Spinal surveillance is systematic [4, 119]. The international definition proposes spinal casting for infantile curves, bracing for curves of 20°–50°, and surgery for curves ≥50° with rapid progression (>5° per year), with a growth-friendly implant if remaining growth is significant or fusion if the patient is near maturity, and recommends MRI of the entire neuraxis before surgery or in the presence of neurological signs, to exclude Chiari malformation, tethered cord, or diastematomyelia [1]. The PNDS considers orthopaedic treatment of scoliosis of little efficacy and recommends surgery as soon as remaining growth allows (closure of the triradiate cartilage), with instrumentation to the pelvis if the patient is nonambulatory or has pelvic obliquity; surgery is difficult in a rigid spine and requires multidisciplinary teams [4]. In a North American database (PearlDiver, 2010–2023), AMC accounted for 115 (2.5%) of 4,600 paediatric posterior spinal fusions for deformity; after 1:4 matching, the 84 patients with AMC had a higher likelihood of 90-day adverse events and of 5-year reoperation (21.4% vs 5.8%) than patients with idiopathic scoliosis [162]. In 9 Turkish patients with Escobar syndrome, scoliosis surgery achieved a mean correction of 65%, without major complications [163]. Preoperative halo traction has been reported in one case of CPSFS1B [9]. In the series of 105 children with tethered cord in 40%, the authors recommend spinal MRI for all patients with AMC and describe potential benefits of surgical release (level IV evidence), a position that diverges from the selective MRI recommendations of the international definition and the PNDS [1, 4, 120].

7.7. Fractures

Fractures require immobilisation that does not add stiffness or displace useful mobility, and a perfect reduction, preferably surgical, because a 10° angular change can compromise what little useful movement there is [4]. In the newborn, handling should support the entire long bone and avoid rotational forces, and the sudden release of a fixed contracture should raise suspicion of an occult fracture [136].

7.8. From the end of growth to adulthood

In the transition to adulthood, the night orthosis is maintained as long as possible, with joint assessment every 6 months, and driving, independent living, workplace ergonomics, and intimacy are addressed [4]. In adults, the goals are pain, stiffness, and independence, with weekly or cyclical maintenance physiotherapy, analgesic physiotherapy, hydrotherapy and spa therapy, adapted sports, joint and spinal protection, adaptation of the home, workplace, and vehicle, a wheelchair for long distances, exoskeletons, and analgesic night orthoses [4]. The PNDS mentions corticosteroid or hyaluronic acid injections without specifying drug, dose, or route, so there is no validated dosing in the sources consulted [4]. A manual wheelchair that wears out the shoulders should not be offered: powered aids or human assistance are preferred when the upper limbs are affected, and obesity reduces independence [4]. Surgery in adults is rare and targets pain and fractures; arthrodesis is preferred to achieve a painless joint, because mobility never improves after surgery; total hip replacement, and more rarely knee replacement, is very difficult, requires wide capsuloligamentous resection, and has a slow recovery; osteoarthritis and lumbar and cervical disc degeneration occur earlier [4]. Bone quality, pain (osteoporosis, early osteoarthritis, low back pain, lower limb arthralgia), anxiety, fatigue, difficulties in sexual life, self-esteem, and loneliness should be assessed and supported [4].

7.9. Anaesthesia and the perioperative period

The number of surgeries ranges from none to several dozen per patient [4]. Venous access is difficult, because of reduced venous capital and atypical vessel courses (odds ratio up to 8.1 compared with people without AMC), requiring additional preparation time; the airway and intubation are difficult when mouth opening and/or neck extension are limited, and a face mask or laryngeal mask serves for short procedures or to prepare for intubation; positioning is difficult [4]. The trismus of DA7 may not be evident before surgery and laryngoscopy may fail; paediatric anaesthesia with a mask and spontaneous ventilation without muscle relaxants, and a successful blind nasotracheal intubation, have been reported in a series of three patients [10]. In Freeman-Sheldon syndrome, anaesthesia is a major risk (difficult airway, difficult venous access, unpredictable response to muscle relaxants), with anaesthetic or surgical complications in half of a cohort of 32 patients, and muscle rigidity after halothane anaesthesia has been reported [9, 50]. In DA5D, micrognathia heralds a difficult airway [13].

The risk of malignant hyperthermia is theoretical and appears very low across AMC as a whole, with scant data and a few reported hypermetabolic episodes; when a muscular cause is confirmed (congenital myopathy) or the form is clinically myogenic, the PNDS recommends preferring halogenated agent–free or regional anaesthesia [4]. Recessive RYR1 variants in FADS and multiple pterygium syndrome with malignant hyperthermia require the same precautions in the patient, in carrier parents, and in siblings [3, 31, 116]. The anaesthetic guidance presented is that of the PNDS; the European Malignant Hyperthermia Group guidelines were not among the sources consulted.

7.10. Management by entity

Rehabilitation in the form of passive stretching should be avoided in diastrophic dysplasia [3]. Schooling takes place in mainstream settings, with adaptations of seating, transfers, accessibility, and pencil grip, a duplicate set of textbooks, aide support, and a schedule compatible with rehabilitation; the PNDS proposes a therapeutic education programme with age-specific modules (0–5, 6–12, and 13–18 years, and in adults genetics, pain, fatigue, and ageing) and psychological support before genetic testing, at disclosure of the diagnosis, and throughout the life of the child, parents, siblings, and adult [4]. Patient organisations should be offered systematically [4].

7.11. Drugs and regulatory status

No medicinal product indicated for AMC, DA, the multiple pterygium syndromes, or FADS was identified in the sources consulted (Europe PMC and ClinicalTrials.gov), nor any regulatory submissions to the Food and Drug Administration (FDA) or the European Medicines Agency (EMA) [154]. The only drugs mentioned in the core sources are joint injections of corticosteroids or hyaluronic acid in adults (drug, dose, and route not specified), immunoglobulins during pregnancy in mothers with anti-AChR antibodies (no dose), and anaesthetic agents, for which the recommendation is to avoid halogenated agents in myogenic forms [4].

7.12. Registered trials and studies

Table source: ClinicalTrials.gov, accessed 18 October 2026 [154]. None of the registered studies is a pharmacological trial, and none has a primary outcome with results published in the sources consulted; the absence of trials in the registry does not prove that no studies exist in other registries. The available therapeutic evidence is observational: case series and retrospective cohorts from orthopaedic centres, with selection bias towards non-lethal forms, heterogeneity of diagnoses and techniques, and no randomised control groups [36, 155, 161]. Success and recurrence figures from the series cannot be compared with one another, and recurrence proportions of 30% to 100% after soft tissue procedures reflect small series and differing definitions [161]. The rehabilitation recommendations envisage a cluster-randomised trial of online training modules, not yet published [36].

7.13. Developments in the past 12 months

The Chinese consensus on the diagnosis and clinical management of AMC (2025 edition), published in July 2025, classifies AMC into amyoplasia, DA, forms with CNS involvement, and neuromuscular diseases and defines self-care as the goal of treatment [38]. In utero therapy to increase fetal movement remains a proposal, with no trial [132].

7.14. Pregnancy, transition, and divergences between sources

In women with myasthenia and anti-AChR antibodies, the risk of AMC recurrence is very high, and maternal thymectomy and immunoglobulins during pregnancy could prevent recurrences, according to the PNDS, with limited evidence; the absence of maternal symptoms does not exclude this cause [4]. Pregnancy in a woman with AMC is not addressed in the sources consulted beyond anaesthetic and venous access issues [4]. The transition to adult care should be prepared in late adolescence, with adult rheumatology and orthopaedics, prolonged rehabilitation after surgery, and a pain clinic [4].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks by scenario

The recurrence risk depends on the specific diagnosis, and the strategy for a new pregnancy should be discussed at a genetics clinic before conception, with genetic analysis of both parents [3, 4].

In the EUROCAT registries, the same or a similar anomaly was reported in 13% of siblings and in 12% of the maternal family of AMC cases [20]. Most siblings with fetal akinesia are similarly affected, but intrafamilial variability increases among distant relatives [31, 111]. In families with Escobar syndrome, the increased frequency of miscarriages and stillbirths and the coexistence, within the same family, of siblings who died in infancy and a surviving patient suggest that severity may vary even with the same variant [29].

8.2. Cascade testing, carriers, and mosaicism

In recessive inheritance, parental testing confirms the phase of the variants and enables carrier testing in relatives; in X-linked inheritance, the mother of an affected boy with more than one affected relative is an obligate heterozygote, and the father of an affected boy is neither affected nor hemizygous [16]. A woman with more than one affected child, no other affected relatives, and the variant not detected in leucocytes probably has germline mosaicism [16]. In trismus-pseudocamptodactyly syndrome, two affected children of unaffected parents indicate probable gonadal mosaicism, and in Sheldon-Hall syndrome transmission may occur from a mildly affected or somatically mosaic parent, so examination of the parents is always necessary [7, 82]. In congenital contractural arachnodactyly, somatic mosaicism for the FBN2 variant was demonstrated in the father of two affected sisters, and in Escobar syndrome the father of an affected child had minimal signs (difficulty opening the mouth fully, scoliosis, pectus excavatum, mild cutaneous syndactyly, carpal anomalies), raising the hypothesis of dominant inheritance or a manifesting heterozygote; analysis of the CHRNG coding regions in the child identified no missense or frameshift variant, and OMIM also describes the family in the entry for the dominant form [64, 66, 69]. In DAIPT, the heterozygous parents of one patient had distal joint laxity [59]. When the cause is RYR1, carrier parents and siblings should be informed of the potential risk of malignant hyperthermia [31, 116]. In France, since 2013, there has been a legal obligation to inform relatives when prevention or counselling is possible, including potentially carrier women in X-linked forms [4].

8.3. Prenatal and preimplantation diagnosis

When the variant is known and the risk is above that of the population, prenatal diagnosis is by chorionic villus sampling or amniocentesis, and PGT-M with in vitro fertilisation is an option, after approval by a multidisciplinary prenatal diagnosis centre; there are five PGT centres in France (Strasbourg, Paris, Montpellier, Nantes, and Grenoble) [4]. In the Australian cohort, several couples with a molecular diagnosis used in vitro fertilisation with PGT-M or prenatal diagnosis in subsequent pregnancies [34]. For CHRNG, with known biallelic variants, prenatal or preimplantation testing is available, and the value of the molecular diagnosis is diagnostic and reproductive, not therapeutic [5]. Without a molecular diagnosis, prenatal diagnosis relies on serial ultrasound from 12 weeks, with the sensitivity limitations of ultrasound, and the strategy should be reviewed in each pregnancy, because panels and sequencing allow more couples to reach a molecular diagnosis [31]. In the Dutch cohort, eight women had more than one affected pregnancy (25 pregnancies), all in group 3 [33].

8.4. Counselling, minors, and psychological support

Psychological support before genetic testing and at disclosure of results addresses motivations, anticipation of consequences, the family dimension, guilt, and coping resources, and is directed at the child, parents, siblings, and adult (plans for parenthood, pain, ageing) [4]. Genetic counselling is recommended in all forms [1]. The sources consulted describe no specific recommendations on carrier testing of minors in the recessive forms within scope. In adults with AMC, psychological support covers plans for parenthood, and the PNDS therapeutic education programme includes a genetics module [4].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

The figure of 1 in 3000 live births, repeated in reviews, in the Oxford Desk Reference, and in the 2006 Finnish study itself, probably comes from a small Finnish hospital series of 11 cases (1961–1965), with no population basis [1, 3, 150]. Population estimates range from 1/3300 to 1/12,000, and the differences result from the case definition (coded AMC vs multiple congenital contractures in the broad sense), the inclusion of TOP and fetal deaths, the source (registries with ICD codes or clinical review), the period, and the difficulty of coding in ICD-10 [1, 21]. The European and Western Australian registries converge at about 1/12,000, and the Alberta and Swedish studies, with a broad definition, at 1/4300–1/5100 [20–22, 167]. The PNDS gives a range of 1/3000 to 1/12,000, and the rehabilitation recommendations 1/3000 to 1/56,000 depending on region, classification, and coding [4, 36]. Prenatal incidence is higher than birth incidence, because of high intrauterine mortality [150]. For comparison, isolated congenital clubfoot occurs in about 1/1000 births [4].

Of the 757 EUROCAT cases, 504 (67%) were live births, 199 (26%) TOP, and 54 (7%) fetal deaths; of the 381 live births with known survival, 87 (23%) died in the first week, and perinatal mortality was 32%; AMC was isolated in 282 (37%), associated with a syndrome or chromosomal anomaly in 90 (12%), and with other major malformations in 385 (51%); maternal illness or first-trimester drugs did not appear to be associated, but data were available for only about one-third of mothers [20]. In Alberta, a specific diagnosis was reached in 43% of group II and 65% of group III, with high mortality mainly in the first month (45% of total losses) [21]. In Sweden, the groups by level of the motor system were cerebral (23), spinal (16), mechanical restriction (3), neuromuscular (12), and connective tissue (9); the first three were mostly sporadic, and the cerebral group had the highest mortality, more contractures at birth, more feeding difficulties, and less independent walking; in 8 cases of myopathy, the contractures normalised during follow-up [22]. In Western Australia, 9/30 had significant CNS anomalies, 7/30 anterior horn and/or peripheral nerve involvement, 3/30 primary muscle disease, and 11/30 syndromes without a neuropathological lesion (mainly DA or amyoplasia); mortality was 37% and clubfoot occurred in 23/30 [167].

By entity, Orphanet estimates Escobar syndrome and LMPS at less than 1/1,000,000 (64 and 28 cases or families, respectively), FADS at 0.6/1,000,000 in Europe, DA5D at less than 1/1,000,000 (33 cases), DAIPT at about 16 cases, and Freeman-Sheldon syndrome at about 100 cases or families [11, 13, 23–25, 27]. Neurogenic AMC has a birth prevalence of 4.3/100,000 in Europe, a figure Orphanet classifies as not validated [77]. The prevalence of congenital contractural arachnodactyly is unknown, with about 70 probands described [39]. The prevalence of DA as a whole is unknown [1].

9.2. Proportion of the major groups

Amyoplasia accounts for up to one-third of AMC in the PNDS and the Oxford Desk Reference, 20–30% in Smith's, and 25–30% in the postnatal population according to Hall, and amyoplasia and DA together for more than half [1, 3, 4, 33, 44]. In orthopaedic centre cohorts, with selection bias towards non-lethal forms, amyoplasia accounted for 122/256 (47.7%) and DA for 62/256 (24.2%) in eight North American hospitals, and for 47/114 (41.2%) and 49/114 (43.0%) in a Canadian centre (1979–2016); these figures are not prevalences [164, 168]. In the Dutch prenatal cohort, amyoplasia accounted for only 9% (6/64), because it is rarely recognised on ultrasound [33].

9.3. Founder populations and consanguinity

In Finland, LCCS is a characteristic disease and the main cause of lethal arthrogryposis: in the 1987–2002 national study, there were 39 cases of LCCS (1/25,250) among 141 lethal arthrogryposes, with total fetal immobility from 12 weeks, hydrops, and death before 32 weeks; the study predates the identification of GLE1, and not all cases were molecularly confirmed [14, 26]. OMIM summarises the same study as 39 LCCS cases among 115 cases of lethal arthrogryposis; the abstract of the original article gives 141, and that is the denominator used [26, 76]. An earlier, morphological national study from 1979–1992 had estimated an incidence of 1/19,000 births, with the grandparents' birthplaces concentrated in the sparsely populated north-east of the country [76]. The GLE1 Fin-major variant has an allele frequency of 0.012 in gnomAD Finns [103]; in a 2008 study, the carrier frequency was 2% in north-eastern Finland and 1% in the general Finnish population, and CAAHD, unlike LCCS1, shows no clustering in the north-east [17, 76]. In a Dutch genetic isolate, 14 fetuses with FADS1 were homozygous for the same MUSK variant [73]. In Turkey, in mostly consanguineous families, CHRNG was the most frequent gene, in 12 of 116 Turkish families (10.3%), followed by ECEL1 in six families [35]; in an earlier series, CHRNG and ECEL1 were the most frequent known genes [117]. The recurrent CHRNG variant c.753_754del has been described as of Turkish origin, in lethal and non-lethal phenotypes [29]. In the Australian cohort, the CHRNG c.459dup duplication was the most frequent pathogenic variant, with a cohort allele frequency of 0.0132 vs 0.000343 in gnomAD; this difference reflects the selection of a patient cohort, not population frequency [34]. MYBPC1-related LCCS4 is due to a founder variant in Bedouin tribes, and the ECEL1 carrier frequency was 1 in 24 (4.2%) among healthy residents of the probands' village in one study, with a probable founder variant also in an Indian cohort; these figures cannot be generalised [12, 13]. Trismus-pseudocamptodactyly syndrome, with a single recurrent variant, arose independently in North American and European families [10]; one of the first large American pedigrees was attributed to a Dutch girl who emigrated to Tennessee in the late 18th century, but haplotype analysis showed that the US families and the Dutch family do not share a founder [62]. A relatively high frequency of Escobar syndrome has been noted in Kuwait, with no population estimate in the sources consulted [69]. Bartsocas-Papas syndrome has a birth prevalence of 1–9/1,000,000 in Spain [145]. In congenital contractural arachnodactyly, most reported patients are of European ancestry, probably owing to ascertainment bias [39].

9.4. Portugal, Brazil, and gaps

The sources consulted contain no epidemiological data on AMC from Portugal or Brazil. The Zika virus epidemic, from 2015 onwards and mainly in Brazil, became the most frequent infectious cause of AMC, with contractures in up to 30% of newborns with congenital Zika, always with microcephaly [1, 4]. Two of the five external reviewers of the international rehabilitation recommendations were from Brazil, and the meta-analysis of the Ponseti method in AMC was conducted by Brazilian authors [36, 155]. The absence of data in the sources consulted does not demonstrate the absence of published studies in these countries.

9.5. Special populations

Among adults, most walk, but fewer than half are fully independent, and pain is common, especially in the feet and back [159]. Amyoplasia has an increased incidence in monozygotic twins, with only one twin affected [44, 124]. In women with myasthenia, pregnancy assessment includes the risk of fetal AMC [4]. In France, AMC qualifies for exemption from patient co-payments under the long-term illness scheme [4].

9.6. European networks, consortia, and patient organisations

Portuguese and Brazilian organisations were not among the sources consulted.

10. Illustrative Clinical Case

In 2016, Kariminejad and colleagues described three Iranian families with Escobar syndrome, identified among nine Iranian cases of arthrogryposis (two fetuses and seven children) studied by exome sequencing; CHRNG was suspected in four and confirmed in three, with the homozygous variants confirmed by Sanger sequencing in the patients, heterozygous in the parents, and absent in 120 Iranian donors [29]. The pedigrees of the three families are shown in Figure 1 (https://pmc.ncbi.nlm.nih.gov/articles/PMC4886457/figure/Fig1/), the clinical features in Figure 2 (https://pmc.ncbi.nlm.nih.gov/articles/PMC4886457/figure/Fig2/; panels a–c of family 1 at 1 month, d–f of family 2, and g–j of family 3 at 7 years), and the variants in Figure 3 (https://pmc.ncbi.nlm.nih.gov/articles/PMC4886457/figure/Fig3/) [29]. The Figure 2 legend designates the family 1 case as IV:1, whereas the text and table designate it IV:3.

In family 3, the male proband (IV:3) was born to first-cousin parents. The couple's first child (IV:1), with multiple contractures and pterygia, died at 6 months; the proband was born of a twin pregnancy, and the co-twin (IV:2), also affected, died at 1.5 years; a subsequent pregnancy ended in fetal death at 20 weeks (IV:4) [29]. The proband weighed 2500 g at birth (5th percentile) and had pterygia of the knees and elbows from birth. He achieved head control at 1 year, stood at 2 years, and walked at 2.5 years, with normal cognition and speech; he underwent surgery for bilateral inguinal hernias at age 3, and epistaxis led to a diagnosis of type 3 von Willebrand disease [29]. At age 7 he measured 117 cm (below the 25th percentile), weighed 16 kg (below the 3rd percentile), and had a head circumference of 49.5 cm; he had low anterior and posterior hairlines, downslanting palpebral fissures, epicanthus, a broad nose, long philtrum, high palate, micrognathia, small and malpositioned teeth, pterygia of the neck, axillae, and antecubital and intercrural regions, elbow and wrist contractures, camptodactyly of all fingers, cutaneous syndactyly, rocker-bottom feet, an overlapping second toe, micropenis, and cryptorchidism [29]. He was homozygous for NM_005199.5:c.401_402del p.(Pro134ArgfsTer43), a recurrent variant that forms a hotspot with c.753_754del [29, 103].

In family 1, the female proband (IV:3), daughter of first cousins with two spontaneous miscarriages at 6 and 7 weeks, underwent amniocentesis for a high risk of trisomy 21 on first-trimester screening, with a normal karyotype, and was born by elective caesarean section at term weighing 2800 g (10th percentile), measuring 44 cm (below the 3rd percentile), with a head circumference of 35 cm (50th percentile). At 1 month she had a very short neck, mild axillary, elbow, and knee pterygia, contractures of the elbows, wrists, fingers, knees, and ankles, clenched hands with the thumb across the palm, clubfeet and rocker-bottom feet with almost no ankle movement, a haemangioma of the forehead and nose, strabismus, a flat nasal root, and downturned corners of the mouth; whole-body radiographs showed no major abnormalities [29]. She was homozygous for a novel variant, described by the authors as c.1009_1010delCA and normalised as NM_005199.5:c.1010_1011del p.(His337LeufsTer60) [29, 103]. In family 2, the female proband, of non-consanguineous parents, had reduced fetal movements perceived by the mother and was born at 39 weeks by caesarean section for breech presentation (3150 g, 43 cm); at 2.5 months she had torticollis, contractures of the shoulders, elbows, wrists, fingers, knees, and great toes, clenched hands, sloping shoulders with reduced muscle mass, elbow dimples, an axillary pterygium, a capillary haemangioma of the nasal tip and forehead, micrognathia, and radiological scoliosis, with normal echocardiography and brain, abdominal, renal, and hip ultrasound; she was homozygous for NM_005199.5:c.753_754del p.(Val253AlafsTer44) [29, 103].

The authors found no correlation between variant position and phenotype and emphasised intra- and interfamilial variability, including the coexistence of lethal and non-lethal forms with the same variant [29]. The phenotype changed from infancy to school age: the pterygia became evident only later, the clenched hands improved until only camptodactyly remained at age 7, the typical facial appearance (downslanting fissures, ptosis, long philtrum, expressionless face) was absent in infancy, and the capillary haemangioma faded [29]. Because the γ→ε switch of the AChR is complete at about 33 weeks of gestation, patients do not have typical postnatal myasthenia, and the authors considered postnatal muscle biopsy of little value; they recommended following infants with AMC, because many develop the characteristic Escobar picture in the first years [29]. The report does not describe the patients' orthopaedic or rehabilitative treatment, which follows the general rehabilitation and orthopaedic recommendations; the coexistence of type 3 von Willebrand disease in the family 3 proband is an individual finding with no established relationship to CHRNG in the sources consulted.

11. Prognosis and Long-Term Follow-Up

11.1. Vital prognosis

Prognosis depends mainly on the cause and on the presence of CNS involvement. In the EUROCAT registries, 23% of live births with known survival died in the first week and perinatal mortality was 32%, figures that include the lethal and syndromic forms [20]; in Western Australia, mortality was 37% [167]; in Alberta, 45% of losses occurred in the first month [21]. In the Dutch prenatal cohort, group 1+2 (13 cases) had 9 (69%) live births and 4 (31%) TOP, and group 3 (51 cases, 68 pregnancies) had 3 live births (4%), 46 TOP (68%), 12 fetal deaths (18%), and 7 neonatal deaths (10%) [33]. In the 2020–2024 Chinese series, 95.16% (59/62) of pregnancies ended in TOP, and of the three newborns, two died in the neonatal period or in infancy [134]. These figures reflect parental decisions and each country's policies, not the natural history without intervention.

In LMPS, all patients are stillborn or die in the immediate neonatal period [70]; in LCCS1, death occurs before 32 weeks [14]; in the Pena-Shokeir phenotype, about 30% are stillborn and most liveborn infants die of pulmonary hypoplasia in the first month [72]; in a 1988 review of about 60 cases, only 5 had survived beyond 28 days, and one child lived to 20 months [73]. In GLE1-related CAAHD, most Finnish cases were lethal in the perinatal period, but there are survivors to school age with ventilation and tube feeding [17]. X-linked infantile SMA has a greatly shortened life expectancy due to ventilatory failure [16]. In Escobar syndrome, about 6% die in the first year from respiratory complications, and the subsequent course is stable [30, 110]. In Freeman-Sheldon syndrome, early mortality from aspiration may occur [50]. In sarcomeric DA (Sheldon-Hall) and in DA5D, life expectancy and cognition are normal [7, 13].

11.2. Functional prognosis

In amyoplasia, 85% of children walk by age 5 with appropriate physiotherapy, and most attend mainstream schooling and are independent in ADL [44]. In a Canadian centre, 98% of children with DA walked independently and most were independent in self-care and transfers [168]. In Freeman-Sheldon syndrome, the mean age at walking is 18 months (10–48) and 83% use a walking aid [50]. In DA1, 20% of adults have straight, functional fingers [47]. In Escobar syndrome, most become ambulatory and have normal intelligence [30]. Muscle strength, contractures, cognition, and response to therapy determine independence in walking, and antigravity strength and passive range of motion of the upper limbs determine independence in ADL; when cognition is preserved, high educational attainment and good self-reported quality of life are observed [1].

In a registry of 143 children under 66 months of age, gross motor function was the most limited domain (more than 90% near or below the cut-off of the ASQ-3, Ages and Stages Questionnaire, at 2 standard deviations), fine motor function was limited in 53.1%, and communication and problem-solving were relatively preserved; a longer neonatal intensive care stay was associated with worse communication and fine motor function [165]. Among 23 young adults (aged 18–36 years) after paediatric care, 83% reported good or excellent health, 30% lived independently, 69% were community ambulators, 57% were employed, mean pain was 2.6/10 (legs and feet), and 56% were satisfied with life [172]. Most adults walk, but fewer than half are fully independent [159]. The PNDS describes a classification into five types of walking prognosis in adults, which it does not detail [4].

11.3. Follow-up schedule

The PNDS sets out intensive follow-up from the neonatal period to the end of growth and then throughout life, with a reference PM&R team and a local physiotherapist at the forefront, plus an occupational therapist, a speech and language therapist (lingual and oral motor function, swallowing, chewing, language), and a psychomotor therapist; the frequency, type, and content of visits depend on severity and were defined by expert opinion, with no fixed intervals in the main text [4]. In adults, follow-up aims to maintain gains and monitor regression, pain, epilepsy, and associated neurological lesions, with adult rheumatology and orthopaedics [4]. Respiratory follow-up from childhood is recommended in certain forms [4]. An emergency card for AMC was in preparation at the Grenoble reference centre [4].

The intervals in the table come from the PNDS, based on expert opinion; the sources consulted provide no validated surveillance intervals for hearing, respiratory function, or the spine, and when the interval is not defined, frequency is decided by the team according to severity [4, 36].

12. Summary and Concluding Remarks

Arthrogryposis multiplex congenita is a clinical sign defined by congenital joint contractures in two or more body areas that do not progress to unaffected joints and whose common denominator is decreased fetal movement. More than 400 genes and various non-genetic, maternal, and mechanical causes produce this sign, and amyoplasia and the distal arthrogryposes together account for more than half of cases [1, 15]. The birth prevalence of coded AMC in European registries is about 1/12,000, and the widely cited figure of 1/3000 probably originates from a hospital series of 11 cases [1, 20].

Amyoplasia is sporadic, symmetrical, without CNS involvement, and carries a recurrence risk equal to that of the population, provided that typical criteria are met and differentials have been excluded; it has a good motor prognosis with early rehabilitation [4, 44]. The distal arthrogryposes are mostly dominant and result from gain of function of fast sarcomeric proteins or of the PIEZO2 channel, with recessive exceptions such as ECEL1-related DA5D, loss-of-function PIEZO2 DAIPT, ADAMTS15-related DA12, and some families with MYLPF-related DA1C; the OMIM series now runs from DA1 to DA12, with the former DA8 reclassified as CPSFS1A and DA4, DA6, and DA10 still without a gene [4, 11, 48, 96]. Escobar syndrome and LMPS form a spectrum of the fetal γ subunit of the AChR: the same CHRNG variant can cause the lethal or non-lethal form in different families, with about 95% concordance between siblings; the absence of postnatal myasthenia is explained by the γ→ε switch in the third trimester, and the course of the non-lethal form is stable, dominated by pterygia, scoliosis, and early respiratory complications [30, 90, 110, 111].

Diagnosis rests on careful clinical examination, examination of the parents, and first-line exome or genome sequencing, supplemented by tests for what sequencing does not detect (DMPK expansion, SMN1, 15q11–q13 methylation) and by maternal anti-AChR antibodies; the reported yields of 28% to 73.5% come from selected cohorts, and a negative result does not exclude a genetic cause [4, 33, 35]. Prenatal detection is low in the general population, about 25% before 24 weeks, and improves with systematic fetal motor assessment in dedicated centres [33, 132].

There is no specific pharmacological treatment. Management is based on rehabilitation from birth, orthoses, serial casting, and individualised orthopaedic surgery, with 2025 international recommendations of low strength and the 2021 French PNDS as the only European national guideline identified in the sources consulted [4, 36]. Anaesthesia, the airway, venous access, and the risk of malignant hyperthermia in myogenic forms require planning [4]. Identification of CMS and SMA, which have their own treatments, is the main therapeutic consequence of the aetiological diagnosis, and the molecular diagnosis defines the recurrence risk and enables prenatal and preimplantation diagnosis [4, 37].

13. References

1. Dahan-Oliel N, Cachecho S, Barnes D, Bedard T, Davison AM, Dieterich K, et al. International multidisciplinary collaboration toward an annotated definition of arthrogryposis multiplex congenita. Am J Med Genet C Semin Med Genet. 2019;181(3):288-299. doi:10.1002/ajmg.c.31721. PMID: 31282072. PMC6771513.
2. Hall JG. Arthrogryposis (multiple congenital contractures): diagnostic approach to etiology, classification, genetics, and general principles. Eur J Med Genet. 2014;57(8):464-72. doi:10.1016/j.ejmg.2014.03.008. PMID: 24704792.
3. Firth HV, Hurst JA. Arthrogryposis. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 66–69.
4. Centre de Référence Anomalies du Développement et Syndromes Malformatifs; Centre de Référence Maladies Neuromusculaires; Filières AnDDI-Rares et FILNEMUS. Protocole National de Diagnostic et de Soins (PNDS): Arthrogryposes multiples congénitales. Saint-Denis: Haute Autorité de Santé; September 2021 [accessed 2026-10-18]. Available from: https://www.has-sante.fr/upload/docs/application/pdf/2021-10/texte_pnds_arthrogryposesmultiplescongenitales.pdf
5. GeneGuidance. Clinical curation of the CHRNG gene. Local database (geneguidance.db) [accessed 2026-10-18].
6. GeneGuidance. Clinical curation of the TPM2 gene. Local database (geneguidance.db) [accessed 2026-10-18].
7. GeneGuidance. Clinical curation of the TNNI2 gene. Local database (geneguidance.db) [accessed 2026-10-18].
8. GeneGuidance. Clinical curation of the TNNT3 gene. Local database (geneguidance.db) [accessed 2026-10-18].
9. GeneGuidance. Clinical curation of the MYH3 gene. Local database (geneguidance.db) [accessed 2026-10-18].
10. GeneGuidance. Clinical curation of the MYH8 gene. Local database (geneguidance.db) [accessed 2026-10-18].
11. GeneGuidance. Clinical curation of the PIEZO2 gene. Local database (geneguidance.db) [accessed 2026-10-18].
12. GeneGuidance. Clinical curation of the MYBPC1 gene. Local database (geneguidance.db) [accessed 2026-10-18].
13. GeneGuidance. Clinical curation of the ECEL1 gene. Local database (geneguidance.db) [accessed 2026-10-18].
14. GeneGuidance. Clinical curation of the GLE1 gene. Local database (geneguidance.db) [accessed 2026-10-18].
15. Kiefer J, Hall JG. Gene ontology analysis of arthrogryposis (multiple congenital contractures). Am J Med Genet C Semin Med Genet. 2019;181(3):310-326. doi:10.1002/ajmg.c.31733. PMID: 31369690.
16. Baumbach-Reardon L, Hunter JM, Ahearn ME, Pfautsch M. Spinal muscular atrophy, X-linked infantile. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Created 2008-10-30; updated 2021-07-29; stated review 2024-05-09. NBK2594. https://www.ncbi.nlm.nih.gov/books/NBK2594/
17. Online Mendelian Inheritance in Man (OMIM). #611890 CONGENITAL ARTHROGRYPOSIS WITH ANTERIOR HORN CELL DISEASE; CAAHD. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/611890. Accessed 2026 Oct 18.
18. Online Mendelian Inheritance in Man (OMIM). #619110 ARTHROGRYPOSIS, DISTAL, TYPE 1C; DA1C. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/619110. Accessed 2026 Oct 18.
19. Online Mendelian Inheritance in Man (OMIM). #620545 ARTHROGRYPOSIS, DISTAL, TYPE 12; DA12. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/620545. Accessed 2026 Oct 18.
20. Hoff JM, Loane M, Gilhus NE, Rasmussen S, Daltveit AK. Arthrogryposis multiplexa congenita: an epidemiologic study of nearly 9 million births in 24 EUROCAT registers. Eur J Obstet Gynecol Reprod Biol. 2011;159(2):347-50. doi:10.1016/j.ejogrb.2011.09.027. PMID: 22005589.
21. Lowry RB, Sibbald B, Bedard T, Hall JG. Prevalence of multiple congenital contractures including arthrogryposis multiplex congenita in Alberta, Canada, and a strategy for classification and coding. Birth Defects Res A Clin Mol Teratol. 2010;88(12):1057-61. doi:10.1002/bdra.20738. PMID: 21157886.
22. Darin N, Kimber E, Kroksmark AK, Tulinius M. Multiple congenital contractures: birth prevalence, etiology, and outcome. J Pediatr. 2002;140(1):61-7. doi:10.1067/mpd.2002.121148. PMID: 11815765.
23. Orphanet. Autosomal recessive multiple pterygium syndrome. ORPHA:2990. https://www.orpha.net/en/disease/detail/2990
24. Orphanet. Lethal multiple pterygium syndrome. ORPHA:33108. https://www.orpha.net/en/disease/detail/33108
25. Orphanet. Fetal akinesia deformation sequence. ORPHA:994. https://www.orpha.net/en/disease/detail/994
26. Pakkasjärvi N, Ritvanen A, Herva R, Peltonen L, Kestilä M, Ignatius J. Lethal congenital contracture syndrome (LCCS) and other lethal arthrogryposes in Finland--an epidemiological study. Am J Med Genet A. 2006;140A(17):1834-9. doi:10.1002/ajmg.a.31381. PMID: 16892327.
27. Orphanet. Freeman-Sheldon syndrome. ORPHA:2053. https://www.orpha.net/en/disease/detail/2053
28. Hall JG. Pena-Shokeir phenotype (fetal akinesia deformation sequence) revisited. Birth Defects Res A Clin Mol Teratol. 2009;85(8):677-94. doi:10.1002/bdra.20611. PMID: 19645055.
29. Kariminejad A, Almadani N, Khoshaeen A, Olsson B, Moslemi AR, Tajsharghi H. Truncating CHRNG mutations associated with interfamilial variability of the severity of the Escobar variant of multiple pterygium syndrome. BMC Genet. 2016;17(1):71. doi:10.1186/s12863-016-0382-5. PMID: 27245440. PMC4886457.
30. Jones KL, Jones MC, del Campo M. Escobar syndrome (multiple pterygium syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 434–435.
31. Firth HV, Hurst JA. Fetal akinesia. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 724–726.
32. Dieterich K, Le Tanno P, Kimber E, Jouk PS, Hall J, Giampietro P. The diagnostic workup in a patient with AMC: Overview of the clinical evaluation and paraclinical analyses with review of the literature. Am J Med Genet C Semin Med Genet. 2019;181(3):337-344. doi:10.1002/ajmg.c.31730. PMID: 31368648.
33. Arduç A, De Vries JIP, B Tan-Sindhunata M, Waisfisz Q, Pajkrt E, Linskens IH. Perinatal genetic diagnostic yield in a population of fetuses with the phenotype arthrogryposis multiplex congenita: a cohort study 2007-2021. Eur J Hum Genet. 2026;34(2):216-226. doi:10.1038/s41431-025-01848-3. PMID: 40195522. PMC12859011.
34. Ravenscroft G, Clayton JS, Faiz F, Sivadorai P, Milnes D, Cincotta R, et al. Neurogenetic fetal akinesia and arthrogryposis: genetics, expanding genotype-phenotypes and functional genomics. J Med Genet. 2021;58(9):609-618. doi:10.1136/jmedgenet-2020-106901. PMID: 33060286. PMC8328565.
35. Pehlivan D, Bayram Y, Gunes N, Coban Akdemir Z, Shukla A, Bierhals T, et al. The Genomics of Arthrogryposis, a Complex Trait: Candidate Genes and Further Evidence for Oligogenic Inheritance. Am J Hum Genet. 2019;105(1):132-150. doi:10.1016/j.ajhg.2019.05.015. PMID: 31230720. PMC6612529.
36. Dahan-Oliel N, Cachecho S, Araujo C, Fąfara A, Lacombe F, Samargian A, et al. Consensus-based recommendations for the rehabilitation of children with arthrogryposis multiplex congenita: an integrated knowledge translation approach. Orphanet J Rare Dis. 2025;20(1):168. doi:10.1186/s13023-025-03671-x. PMID: 40205440. PMC11983950.
37. Abicht A, Müller JS, Lochmüller H. Congenital myasthenic syndromes overview. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Created 2003-05-09; updated 2021-12-23. NBK1168. https://www.ncbi.nlm.nih.gov/books/NBK1168/
38. Obstetrics and Gynecology Ultrasound Group, Ultrasound Medical Branch of Chinese Medical Association; Clinical Genetics Group, Medical Geneticist Branch of Chinese Medical Doctor Association; Group of Genetic Disease Prevention and Control, Birth Defect Prevention and Control Committee of Chinese Society of Preventive Medicine. [Expert consensus on the clinical diagnosis and management of arthrogryposis multiplex congenita (2025 edition)]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi. 2025;42(7):820-829. doi:10.3760/cma.j.cn511374-20250518-00306. PMID: 40947408.
39. Callewaert B. Congenital contractural arachnodactyly. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–. Created 2001-01-23; updated 2019-10-21; stated review 2022-07-14. NBK1386. https://www.ncbi.nlm.nih.gov/books/NBK1386/
40. ERN ITHACA — European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders [Internet]. Available from: https://ern-ithaca.eu/ [accessed 2026-10-18].
41. ERN EURO-NMD — European Reference Network for Rare Neuromuscular Diseases [Internet]. Available from: https://ern-euro-nmd.eu/ [accessed 2026-10-18].
42. ERN BOND — European Reference Network on Rare Bone Disorders [Internet]. Available from: https://ernbond.eu/ [accessed 2026-10-18].
43. Orphanet. Arthrogryposis multiplex congenita. ORPHA:1037. https://www.orpha.net/en/disease/detail/1037
44. Jones KL, Jones MC, del Campo M. Amyoplasia congenita disruptive sequence. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 226–229.
45. Human Phenotype Ontology Consortium. HPO annotations (phenotype.hpoa, genes_to_disease.txt), release 2026-09-02 [accessed 2026-10-18]. Available from: https://github.com/obophenotype/human-phenotype-ontology/releases
46. Orphanet. ORPHAcodes API (preferred names) [accessed 2026-10-18]. Available from: https://api.orphacode.org
47. Jones KL, Jones MC, del Campo M. Distal arthrogryposis syndrome, type 1. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 230–233.
48. Online Mendelian Inheritance in Man (OMIM). #108120 ARTHROGRYPOSIS, DISTAL, TYPE 1A; DA1A. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/108120. Accessed 2026 Oct 18.
49. Online Mendelian Inheritance in Man (OMIM). #614335 ARTHROGRYPOSIS, DISTAL, TYPE 1B; DA1B. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/614335. Accessed 2026 Oct 18.
50. Jones KL, Jones MC, del Campo M. Freeman-Sheldon syndrome (whistling face syndrome, distal arthrogryposis type 2A). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 300–301.
51. Online Mendelian Inheritance in Man (OMIM). #193700 ARTHROGRYPOSIS, DISTAL, TYPE 2A; DA2A. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/193700. Accessed 2026 Oct 18.
52. Online Mendelian Inheritance in Man (OMIM). #601680 ARTHROGRYPOSIS, DISTAL, TYPE 2B1; DA2B1. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/601680. Accessed 2026 Oct 18.
53. Online Mendelian Inheritance in Man (OMIM). #618435 ARTHROGRYPOSIS, DISTAL, TYPE 2B2; DA2B2. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/618435. Accessed 2026 Oct 18.
54. Online Mendelian Inheritance in Man (OMIM). #618436 ARTHROGRYPOSIS, DISTAL, TYPE 2B3; DA2B3. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/618436. Accessed 2026 Oct 18.
55. Online Mendelian Inheritance in Man (OMIM). #114300 ARTHROGRYPOSIS, DISTAL, TYPE 3; DA3. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/114300. Accessed 2026 Oct 18.
56. Online Mendelian Inheritance in Man (OMIM). %609128 ARTHROGRYPOSIS, DISTAL, TYPE 4; DA4. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/609128. Accessed 2026 Oct 18.
57. Online Mendelian Inheritance in Man (OMIM). #108145 ARTHROGRYPOSIS, DISTAL, TYPE 5; DA5. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/108145. Accessed 2026 Oct 18.
58. Online Mendelian Inheritance in Man (OMIM). #248700 MARDEN-WALKER SYNDROME; MWKS. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/248700. Accessed 2026 Oct 18.
59. Online Mendelian Inheritance in Man (OMIM). #617146 ARTHROGRYPOSIS, DISTAL, WITH IMPAIRED PROPRIOCEPTION AND TOUCH; DAIPT. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/617146. Accessed 2026 Oct 18.
60. Online Mendelian Inheritance in Man (OMIM). #615065 ARTHROGRYPOSIS, DISTAL, TYPE 5D; DA5D. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/615065. Accessed 2026 Oct 18.
61. Online Mendelian Inheritance in Man (OMIM). 108200 ARTHROGRYPOSIS, DISTAL, TYPE 6; DA6. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/108200. Accessed 2026 Oct 18.
62. Online Mendelian Inheritance in Man (OMIM). #158300 ARTHROGRYPOSIS, DISTAL, TYPE 7; DA7. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/158300. Accessed 2026 Oct 18.
63. Chong JX, Burrage LC, Beck AE, Marvin CT, McMillin MJ, Shively KM, et al. Autosomal-Dominant Multiple Pterygium Syndrome Is Caused by Mutations in MYH3. Am J Hum Genet. 2015;96(5):841-9. doi:10.1016/j.ajhg.2015.04.004. PMID: 25957469. PMC4570285.
64. Online Mendelian Inheritance in Man (OMIM). #178110 CONTRACTURES, PTERYGIA, AND SPONDYLOCARPOTARSAL FUSION SYNDROME 1A; CPSFS1A. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/178110. Accessed 2026 Oct 18.
65. Online Mendelian Inheritance in Man (OMIM). #618469 CONTRACTURES, PTERYGIA, AND SPONDYLOCARPOTARSAL FUSION SYNDROME 1B; CPSFS1B. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/618469. Accessed 2026 Oct 18.
66. Online Mendelian Inheritance in Man (OMIM). #121050 CONTRACTURAL ARACHNODACTYLY, CONGENITAL; CCA. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/121050. Accessed 2026 Oct 18.
67. Online Mendelian Inheritance in Man (OMIM). %187370 ARTHROGRYPOSIS, DISTAL, TYPE 10; DA10. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/187370. Accessed 2026 Oct 18.
68. Online Mendelian Inheritance in Man (OMIM). #620019 ARTHROGRYPOSIS, DISTAL, TYPE 11; DA11. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/620019. Accessed 2026 Oct 18.
69. Online Mendelian Inheritance in Man (OMIM). #265000 MULTIPLE PTERYGIUM SYNDROME, ESCOBAR VARIANT; EVMPS. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/265000. Accessed 2026 Oct 18.
70. Jones KL, Jones MC, del Campo M. Lethal multiple pterygium syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 240–241.
71. Online Mendelian Inheritance in Man (OMIM). #253290 MULTIPLE PTERYGIUM SYNDROME, LETHAL TYPE; LMPS. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/253290. Accessed 2026 Oct 18.
72. Jones KL, Jones MC, del Campo M. Pena-Shokeir phenotype (fetal akinesia/hypokinesia sequence). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 234–235.
73. Online Mendelian Inheritance in Man (OMIM). #208150 FETAL AKINESIA DEFORMATION SEQUENCE 1; FADS1. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/208150. Accessed 2026 Oct 18.
74. Online Mendelian Inheritance in Man (OMIM). #618388 FETAL AKINESIA DEFORMATION SEQUENCE 2; FADS2. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/618388. Accessed 2026 Oct 18.
75. Online Mendelian Inheritance in Man (OMIM). #618389 FETAL AKINESIA DEFORMATION SEQUENCE 3; FADS3. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/618389. Accessed 2026 Oct 18.
76. Online Mendelian Inheritance in Man (OMIM). #253310 LETHAL CONGENITAL CONTRACTURE SYNDROME 1; LCCS1. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/253310. Accessed 2026 Oct 18.
77. Orphanet. Neurogenic arthrogryposis multiplex congenita. ORPHA:1143. https://www.orpha.net/en/disease/detail/1143
78. Orphanet. Multiple pterygium-malignant hyperthermia syndrome. ORPHA:2215. https://www.orpha.net/en/disease/detail/2215
79. Cachecho S, Elfassy C, Hamdy R, Rosenbaum P, Dahan-Oliel N. Arthrogryposis multiplex congenita definition: Update using an international consensus-based approach. Am J Med Genet C Semin Med Genet. 2019;181(3):280-287. doi:10.1002/ajmg.c.31739. PMID: 31452331.
80. Bedard T, Lowry RB. Disease coding systems for arthrogryposis multiplex congenita. Am J Med Genet C Semin Med Genet. 2019;181(3):304-309. doi:10.1002/ajmg.c.31718. PMID: 31232506.
81. Escobar V, Bixler D, Gleiser S, Weaver DD, Gibbs T. Multiple pterygium syndrome. Am J Dis Child. 1978;132(6):609-611. doi:10.1001/archpedi.1978.02120310073016. PMID: 655146.
82. Jones KL, Jones MC, del Campo M. Hecht syndrome (trismus pseudocamptodactyly syndrome, distal arthrogryposis type 7). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 312–313.
83. Jones KL, Jones MC, del Campo M. Beals syndrome (congenital contractural arachnodactyly syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 666–669.
84. Hall JG, Reed SD, Rosenbaum KN, Gershanik J, Chen H, Wilson KM. Limb pterygium syndromes: a review and report of eleven patients. Am J Med Genet. 1982;12(4):377-409. doi:10.1002/ajmg.1320120404. PMID: 7124793.
85. Hall JG, Reed SD, Greene G. The distal arthrogryposes: delineation of new entities — review and nosologic discussion. Am J Med Genet. 1982;11(2):185-239. doi:10.1002/ajmg.1320110208. PMID: 7039311.
86. Bamshad M, Jorde LB, Carey JC. A revised and extended classification of the distal arthrogryposes. Am J Med Genet. 1996;65(4):277-81. doi:10.1002/(SICI)1096-8628(19961111)65:4<277::AID-AJMG6>3.0.CO;2-M. PMID: 8923935.
87. Dahan-Oliel N, Hall J, Samargian A, Sawatzky B, van Bosse H. Summary of the 3rd international symposium on arthrogryposis. Am J Med Genet C Semin Med Genet. 2019;181(3):277-279. doi:10.1002/ajmg.c.31705. PMID: 31120641.
88. Nematollahi S, Bouzaglo N, Sanchez VC, de-Vries JIP, Dieterich K, Fafara A, et al. The International Consortium for Arthrogryposis: A Collaborative Framework for Early Detection, Care, Research, and Education. Am J Med Genet C Semin Med Genet. 2026. doi:10.1002/ajmg.c.70013. PMID: 42362964.
89. Toydemir RM, Rutherford A, Whitby FG, Jorde LB, Carey JC, Bamshad MJ. Mutations in embryonic myosin heavy chain (MYH3) cause Freeman-Sheldon syndrome and Sheldon-Hall syndrome. Nat Genet. 2006;38(5):561-5. doi:10.1038/ng1775. PMID: 16642020.
90. Hoffmann K, Muller JS, Stricker S, Megarbane A, Rajab A, Lindner TH, et al. Escobar syndrome is a prenatal myasthenia caused by disruption of the acetylcholine receptor fetal gamma subunit. Am J Hum Genet. 2006;79(2):303-12. doi:10.1086/506257. PMID: 16826520. PMC1559482.
91. Morgan NV, Brueton LA, Cox P, Greally MT, Tolmie J, Pasha S, et al. Mutations in the embryonal subunit of the acetylcholine receptor (CHRNG) cause lethal and Escobar variants of multiple pterygium syndrome. Am J Hum Genet. 2006;79(2):390-5. doi:10.1086/506256. PMID: 16826531. PMC1559492.
92. Vogt J, Harrison BJ, Spearman H, Cossins J, Vermeer S, ten Cate LN, et al. Mutation analysis of CHRNA1, CHRNB1, CHRND, and RAPSN genes in multiple pterygium syndrome/fetal akinesia patients. Am J Hum Genet. 2008;82(1):222-7. doi:10.1016/j.ajhg.2007.09.016. PMID: 18179903. PMC2253973.
93. Michalk A, Stricker S, Becker J, Rupps R, Pantzar T, Miertus J, et al. Acetylcholine receptor pathway mutations explain various fetal akinesia deformation sequence disorders. Am J Hum Genet. 2008;82(2):464-76. doi:10.1016/j.ajhg.2007.11.006. PMID: 18252226. PMC2427255.
94. Nousiainen HO, Kestilä M, Pakkasjärvi N, Honkala H, Kuure S, Tallila J, et al. Mutations in mRNA export mediator GLE1 result in a fetal motoneuron disease. Nat Genet. 2008;40(2):155-7. doi:10.1038/ng.2007.65. PMID: 18204449. PMC2684619.
95. Vogt J, Morgan NV, Marton T, Maxwell S, Harrison BJ, Beeson D, et al. Germline mutation in DOK7 associated with fetal akinesia deformation sequence. J Med Genet. 2009;46(5):338-40. doi:10.1136/jmg.2008.065425. PMID: 19261599.
96. McMillin MJ, Below JE, Shively KM, Beck AE, Gildersleeve HI, Pinner J, et al. Mutations in ECEL1 cause distal arthrogryposis type 5D. Am J Hum Genet. 2013;92(1):150-6. doi:10.1016/j.ajhg.2012.11.014. PMID: 23261301. PMC3542461.
97. McMillin MJ, Beck AE, Chong JX, Shively KM, Buckingham KJ, Gildersleeve HI, et al. Mutations in PIEZO2 cause Gordon syndrome, Marden-Walker syndrome, and distal arthrogryposis type 5. Am J Hum Genet. 2014;94(5):734-44. doi:10.1016/j.ajhg.2014.03.015. PMID: 24726473. PMC4067551.
98. Hall JG, Kimber E, Dieterich K. Classification of arthrogryposis. Am J Med Genet C Semin Med Genet. 2019;181(3):300-303. doi:10.1002/ajmg.c.31716. PMID: 31271512.
99. Bamshad M, Van Heest AE, Pleasure D. Arthrogryposis: a review and update. J Bone Joint Surg Am. 2009;91 Suppl 4(Suppl 4):40-6. doi:10.2106/JBJS.I.00281. PMID: 19571066. PMC2698792.
100. Hall JG, Kiefer J. Arthrogryposis as a Syndrome: Gene Ontology Analysis. Mol Syndromol. 2016;7(3):101-9. doi:10.1159/000446617. PMID: 27587986. PMC4988256.
101. Melki J. Genetic Bases of Arthrogryposis Multiplex Congenita. Annu Rev Genomics Hum Genet. 2026;27(1):239-245. doi:10.1146/annurev-genom-120324-031410. PMID: 41729706.
102. Laquerriere A, Jaber D, Abiusi E, Maluenda J, Mejlachowicz D, Vivanti A, et al. Phenotypic spectrum and genomics of undiagnosed arthrogryposis multiplex congenita. J Med Genet. 2022;59(6):559-567. doi:10.1136/jmedgenet-2020-107595. PMID: 33820833. PMC9132874.
103. National Center for Biotechnology Information. ClinVar [accessed 2026-10-18]. Available from: https://www.ncbi.nlm.nih.gov/clinvar/
104. Zgheib O, Rio-Frio T, Guipponi M, Nouspikel T, Fokstuen S, Abramowicz M, et al. Biallelic MYH3 Variants Cause Distal Arthrogryposis in Compound Heterozygosity and a Subclinical Phenotype in Simple Heterozygosity. Codominance or Recessive Inheritance?. Clin Genet. 2026;109(6):1059-1063. doi:10.1111/cge.70142. PMID: 41562327. PMC13167627.
105. Altin N, Mamchaoui K, Ohana J, Bigot A, Corradi B, Maragliano L, et al. The Emerging TNNT3 Spectrum: From Distal Arthrogryposis to Congenital Myopathy. Hum Mutat. 2025;2025:1785045. doi:10.1155/humu/1785045. PMID: 41473596. PMC12745331.
106. Online Mendelian Inheritance in Man (OMIM). #609285 CONGENITAL MYOPATHY 23; CMYO23. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/609285. Accessed 2026 Oct 18.
107. Online Mendelian Inheritance in Man (OMIM). #618524 CONGENITAL MYOPATHY 16; CMYO16. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/618524. Accessed 2026 Oct 18.
108. Online Mendelian Inheritance in Man (OMIM). #614915 LETHAL CONGENITAL CONTRACTURE SYNDROME 4; LCCS4. Baltimore (MD): Johns Hopkins University. Available from: https://www.omim.org/entry/614915. Accessed 2026 Oct 18.
109. Capece G, Di Feo MF, Melnik E, Dadali E, Markova T, Verloes A, et al. Beyond distal arthrogryposis: refining the phenotypic landscape of PIEZO2-related disorders. Brain. 2026. doi:10.1093/brain/awag249. PMID: 42496149.
110. Carrera-García L, Natera-de Benito D, Dieterich K, de la Banda MGG, Felter A, Inarejos E, et al. CHRNG-related nonlethal multiple pterygium syndrome: Muscle imaging pattern and clinical, histopathological, and molecular genetic findings. Am J Med Genet A. 2019;179(6):915-926. doi:10.1002/ajmg.a.61122. PMID: 30868735.
111. Vogt J, Morgan NV, Rehal P, Faivre L, Brueton LA, Becker K, et al. CHRNG genotype-phenotype correlations in the multiple pterygium syndromes. J Med Genet. 2012;49(1):21-6. doi:10.1136/jmedgenet-2011-100378. PMID: 22167768.
112. GeneGuidance. Clinical curation of the CHRNA1 gene. Local database (geneguidance.db) [accessed 2026-10-18].
113. GeneGuidance. Clinical curation of the CHRND gene. Local database (geneguidance.db) [accessed 2026-10-18].
114. GeneGuidance. Clinical curation of the DOK7 gene. Local database (geneguidance.db) [accessed 2026-10-18].
115. Firth HV, Hurst JA. Spinal muscular atrophy (SMA). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 526–527.
116. McKie AB, Alsaedi A, Vogt J, Stuurman KE, Weiss MM, Shakeel H, et al. Germline mutations in RYR1 are associated with foetal akinesia deformation sequence/lethal multiple pterygium syndrome. Acta Neuropathol Commun. 2014;2:148. doi:10.1186/s40478-014-0148-0. PMID: 25476234. PMC4271450.
117. Bayram Y, Karaca E, Coban Akdemir Z, Yilmaz EO, Tayfun GA, Aydin H, et al. Molecular etiology of arthrogryposis in multiple families of mostly Turkish origin. J Clin Invest. 2016;126(2):762-78. doi:10.1172/JCI84457. PMID: 26752647. PMC4731160.
118. Dieterich K, Kimber E, Hall JG. Central nervous system involvement in arthrogryposis multiplex congenita: Overview of causes, diagnosis, and care. Am J Med Genet C Semin Med Genet. 2019;181(3):345-353. doi:10.1002/ajmg.c.31732. PMID: 31410997.
119. Mom A, Bourg V, Dai S, Milot AM, Perennou D, Dieterich K. Spinal Involvement in a Pediatric and Adult Cohort of Patients With Arthrogryposis Multiplex Congenita. Am J Med Genet C Semin Med Genet. 2026. doi:10.1002/ajmg.c.70009. PMID: 42161899.
120. Nugraha HK, Hariharan AR, Huser AJ, Feldman DS, Asadi-Moghaddam K. Do Not Forget the Spine MRI in Children With Arthrogryposis Multiplex Congenita: High Prevalence of Tethered Spinal Cord and Preliminary Clinical Findings Following Detethering. J Pediatr Orthop. 2025;45(2):e186-e191. doi:10.1097/BPO.0000000000002839. PMID: 39380182.
121. Hermansen MV, Ørstavik K, Steen U, Toft M, Brunborg C, Wekre LL. The EXPLAIN Study: Exploring Arthrogryposis Multiplex Congenita in Adults in Norway - A Description of Demographic, Medical, and Neurological Findings. Am J Med Genet C Semin Med Genet. 2026. doi:10.1002/ajmg.c.70017. PMID: 42271604.
122. Cirillo A, Collins J, Sawatzky B, Hamdy R, Dahan-Oliel N. Pain among children and adults living with arthrogryposis multiplex congenita: A scoping review. Am J Med Genet C Semin Med Genet. 2019;181(3):436-453. doi:10.1002/ajmg.c.31725. PMID: 31347265.
123. Sions JM, Donohoe M, Vallery CE, Stauffer SJ, Pohlig RT. Pain and Health: Sex-Specific Differences From a Survey Study of Adults With Arthrogryposis Multiplex Congenita. Am J Med Genet C Semin Med Genet. 2026. doi:10.1002/ajmg.c.70021. PMID: 42552986.
124. Hall JG, Aldinger KA, Tanaka KI. Amyoplasia revisited. Am J Med Genet A. 2014;164A(3):700-30. doi:10.1002/ajmg.a.36395. PMID: 24459070.
125. Kimber E. AMC: amyoplasia and distal arthrogryposis. J Child Orthop. 2015;9(6):427-32. doi:10.1007/s11832-015-0689-1. PMID: 26537820. PMC4661149.
126. Diodato D, Bosco L, Catteruccia M, Sancesario A, Jofrè J, Suarez B, et al. Distal arthrogryposis with impaired proprioception and touch: description of 9 additional cases harbouring novel PIEZO2 variants and literature review. Eur J Paediatr Neurol. 2026;61:24-35. doi:10.1016/j.ejpn.2026.01.003. PMID: 41780226.
127. Jones KL, Jones MC, del Campo M. Spondylocarpotarsal synostosis syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 602–603.
128. Firth HV, Hurst JA. Floppy infant. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 154–157.
129. Fountain MD, Aten E, Cho MT, Juusola J, et al. The phenotypic spectrum of Schaaf-Yang syndrome: 18 new affected individuals from 14 families. Genet Med. 2017;19:45-52. doi:10.1038/gim.2016.53. PMID: 27195816.
130. Firth HV, Hurst JA. Prolonged neonatal jaundice and jaundice in infants below 6 months. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 288–291.
131. Gissen P, Johnson CA, Morgan NV, Stapelbroek JM, et al. Mutations in VPS33B, encoding a regulator of SNARE-dependent membrane fusion, cause arthrogryposis-renal dysfunction-cholestasis (ARC) syndrome. Nat Genet. 2004;36:400-404. doi:10.1038/ng1325. PMID: 15052268.
132. Filges I, Tercanli S, Hall JG. Fetal arthrogryposis: Challenges and perspectives for prenatal detection and management. Am J Med Genet C Semin Med Genet. 2019;181(3):327-336. doi:10.1002/ajmg.c.31723. PMID: 31318155.
133. Filges I, Hall JG. Failure to identify antenatal multiple congenital contractures and fetal akinesia--proposal of guidelines to improve diagnosis. Prenat Diagn. 2013;33(1):61-74. doi:10.1002/pd.4011. PMID: 23296716.
134. Yao T, Yuan L, Dong X, Wang Y, Zhang C, Liu L, et al. Prenatal Diagnosis of Arthrogryposis Multiplex Congenita (AMC): Ultrasound and Genetic Findings in 69 Fetuses From 67 Unrelated Families. Prenat Diagn. 2026;46(5-6):881-891. doi:10.1002/pd.70122. PMID: 41936055.
135. Firth HV, Hurst JA. Talipes (club-foot). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 788–789.
136. Watanabe D, Hasebe Y, Maebayashi Y, Katsumata N, Nemoto A, Naitoh A. Sudden Release of a Fixed Contracture as a Clue to Occult Humeral Fracture in a Neonate With Arthrogryposis Multiplex Congenita. Cureus. 2026;18(7):e112944. doi:10.7759/cureus.112944. PMID: 42614726. PMC13483570.
137. Pergande M, Motameny S, Özdemir Ö, Kreutzer M, Wang H, Daimagüler HS, et al. The genomic and clinical landscape of fetal akinesia. Genet Med. 2020;22(3):511-523. doi:10.1038/s41436-019-0680-1. PMID: 31680123.
138. Todd EJ, Yau KS, Ong R, Slee J, McGillivray G, Barnett CP, et al. Next generation sequencing in a large cohort of patients presenting with neuromuscular disease before or at birth. Orphanet J Rare Dis. 2015;10:148. doi:10.1186/s13023-015-0364-0. PMID: 26578207. PMC4650299.
139. Wang M, Hong J, Han S, Jin P, Xu C, Qian Y, et al. Whole Genome Sequence Identifies the Second Allele: An Intronic Variant in RYR1 Contributes to Early-Onset Fetal Akinesia Deformation Sequence. Mol Syndromol. 2026. doi:10.1159/000550111. PMID: 41878156. PMC13008415.
140. Misceo D, Terkelsen T, Bøen Keim SM, Bjørnstad PM, Strand ME, Orszagh VC, et al. CRISPR Activation Reveals the Spliceogenicity of an Intronic NEB Variant in Fetuses With Arthrogryposis Multiplex Congenita 6. Clin Genet. 2026;109(4):772-777. doi:10.1111/cge.70097. PMID: 41186962. PMC12958004.
141. Qi Y, Sun J, He W, Wu J, Yin A. Rapid functional RNA Analysis via amniocyte transdifferentiation resolves prenatal variant ambiguity in fetal akinesia syndrome. Genet Med Open. 2026;4:104393. doi:10.1016/j.gimo.2026.104393. PMID: 42317288. PMC13272537.
142. Polizzi A, Huson SM, Vincent A. Teratogen update: maternal myasthenia gravis as a cause of congenital arthrogryposis. Teratology. 2000;62(5):332-341. doi:10.1002/1096-9926(200011)62:5<332::AID-TERA7>3.0.CO;2-E. PMID: 11029151.
143. Oberg KC, Magaki S, Hall JG. A standardized autopsy protocol for arthrogryposis (multiple congenital contractures). Am J Med Genet C Semin Med Genet. 2019;181(3):474-478. doi:10.1002/ajmg.c.31731. PMID: 31373772.
144. Witters I, Moerman P, Fryns JP. Fetal akinesia deformation sequence: a study of 30 consecutive in utero diagnoses. Am J Med Genet. 2002;113(1):23-8. doi:10.1002/ajmg.10698. PMID: 12400062.
145. Jones KL, Jones MC, del Campo M. Popliteal pterygium syndrome (facio-genito-popliteal syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 432–433.
146. Campos IFS, Ribeiro EM. Miopatias congênitas. In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 285–288.
147. Jones KL, Jones MC, del Campo M. Baraitser-Winter syndrome. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 314–321.
148. Firth HV, Hurst JA. Ptosis, blepharophimosis, and other eyelid anomalies. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 292–295.
149. García-Cazorla Á, Dionisi-Vici C, Saudubray JM. Disorders of Cellular Trafficking. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. doi:10.1007/978-3-662-63123-2_44
150. Niles KM, Blaser S, Shannon P, Chitayat D. Fetal arthrogryposis multiplex congenita/fetal akinesia deformation sequence (FADS)-Aetiology, diagnosis, and management. Prenat Diagn. 2019;39(9):720-731. doi:10.1002/pd.5505. PMID: 31218730.
151. Araujo CRS, Hyer L, Sienko SE, Buckon C, Costa C, Natera-de Benito D, et al. Gross motor functional classification for arthrogryposis multiplex congenita: protocol for co-development involving public with lived and professional experience. Res Involv Engagem. 2025;12(1):19. doi:10.1186/s40900-025-00827-8. PMID: 41402879. PMC12884608.
152. Wagner LV, Cherry JS, Sawatzky BJ, Fąfara A, Elfassy C, Eriksson M, et al. Rehabilitation across the lifespan for individuals with arthrogryposis. Am J Med Genet C Semin Med Genet. 2019;181(3):385-392. doi:10.1002/ajmg.c.31729. PMID: 31397084.
153. Babik I, Cunha AB, Lobo MA. Play with objects in children with arthrogryposis: Effects of intervention with the Playskin Lift™ exoskeletal garment. Am J Med Genet C Semin Med Genet. 2019;181(3):393-403. doi:10.1002/ajmg.c.31719. PMID: 31232529. PMC10740315.
154. ClinicalTrials.gov API v2 [accessed 2026-10-18]. Available from: https://clinicaltrials.gov/api/v2/studies
155. Bravin AC, Ferraz Ferreira G, Nogueira MP. Effectiveness of the Ponseti method for congenital clubfoot in patients with arthrogryposis: A systematic review and meta-analysis of observational studies. J Child Orthop. 2024;18(1):72-78. doi:10.1177/18632521231214778. PMID: 38348434. PMC10859111.
156. Matar HE, Beirne P, Garg N. The effectiveness of the Ponseti method for treating clubfoot associated with arthrogryposis: up to 8 years follow-up. J Child Orthop. 2016;10(1):15-8. doi:10.1007/s11832-016-0712-1. PMID: 26833334. PMC4763152.
157. Church C, McGowan A, Henley J, Donohoe M, Niiler T, Shrader MW, et al. The 5-Year Outcome of the Ponseti Method in Children With Idiopathic Clubfoot and Arthrogryposis. J Pediatr Orthop. 2020;40(7):e641-e646. doi:10.1097/BPO.0000000000001524. PMID: 32032217.
158. Hamdy RC, van Bosse H, Altiok H, Abu-Dalu K, Kotlarsky P, Fafara A, et al. Treatment and outcomes of arthrogryposis in the lower extremity. Am J Med Genet C Semin Med Genet. 2019;181(3):372-384. doi:10.1002/ajmg.c.31734. PMID: 31479584.
159. van Bosse HJP. Orthopaedic care of the child with arthrogryposis: a 2020 overview. Curr Opin Pediatr. 2020;32(1):76-85. doi:10.1097/MOP.0000000000000847. PMID: 31743218.
160. Oishi S, Agranovich O, Zlotolow D, Wall L, Stutz C, Pajardi G, et al. Treatment and outcomes of arthrogryposis in the upper extremity. Am J Med Genet C Semin Med Genet. 2019;181(3):363-371. doi:10.1002/ajmg.c.31722. PMID: 31268234.
161. Mounsef PJ, Laroche M, Ben Letaifa R, Legler J, Gagnon M, Dahan-Oliel N, et al. Surgical timing and management patterns across joints in arthrogryposis: a systematic review. J Pediatr Orthop B. 2026. doi:10.1097/BPB.0000000000001332. PMID: 41691531.
162. Smith-Voudouris J, Yang A, Gouzoulis MJ, Tuason DA, Grauer JN, Frumberg DB. Pediatric patients with arthrogryposis have increased early complications and long-term reoperation risk following posterior spinal fusion. Spine Deform. 2026;14(3):967-975. doi:10.1007/s43390-026-01277-2. PMID: 41520309.
163. Aslan M, Korkmaz MF. A Clinical and Demographic Investigation on Cases With Multiple Pterygium Syndrome-Escobar Type. Congenit Anom (Kyoto). 2026;66(1):e70084. doi:10.1002/cga.70084. PMID: 42855427.
164. Zidan A, Snider L, Rampakakis E, Hamdy R, Rauch F, Hyer LC, et al. Factors Associated With Functional Mobility in 256 Children With Arthrogryposis Multiplex Congenita: A Multicentric Cross-Sectional Study. Arch Phys Med Rehabil. 2026;107(8):1915-1924. doi:10.1016/j.apmr.2025.12.014. PMID: 41475428.
165. Zidan A, Pasia SC, Rampakakis E, Hamdy R, Rauch F, Hyer LC, et al. Milestone Attainment in Young Children With Arthrogryposis Multiplex Congenita: Developmental Profile and Associated Factors. Am J Med Genet C Semin Med Genet. 2026. doi:10.1002/ajmg.c.70006. PMID: 41834695.
166. Panebianco CJ, Huang Y, Khatib N, Gottlieb DC, Essaidi M, Ahmed S, et al. Maternal Exercise Rescues Fetal Akinesia-Impaired Joint and Bone Development. FASEB J. 2025;39(24):e71341. doi:10.1096/fj.202503192R. PMID: 41388826. PMC12701528.
167. Silberstein EP, Kakulas BA. Arthrogryposis multiplex congenita in Western Australia. J Paediatr Child Health. 1998;34(6):518-23. doi:10.1046/j.1440-1754.1998.00296.x. PMID: 9928642.
168. Hansen-Jaumard D, Elfassy C, Montpetit K, Ghalimah B, Hamdy R, Dahan-Oliel N. A review of the orthopedic interventions and functional outcomes among a cohort of 114 children with arthrogryposis multiplex congenita. J Pediatr Rehabil Med. 2020;13(3):263-271. doi:10.3233/PRM-190657. PMID: 33104047.
169. Arthrogryposis Multiplex Congenita Support, Inc. (AMCSI) [Internet]. Available from: https://amcsupport.org/ [accessed 2026-10-18].
170. The Arthrogryposis Group (TAG), United Kingdom [Internet]. Available from: https://arthrogryposis.co.uk/ [accessed 2026-10-18].
171. Alliance Arthrogrypose (France) [Internet]. Available from: https://www.arthrogrypose.fr/ [accessed 2026-10-18].
172. Altiok H, Flanagan A, Krzak JJ, Hassani S. Quality of life, satisfaction with life, and functional mobility of young adults with arthrogryposis after leaving pediatric care. Am J Med Genet C Semin Med Genet. 2019;181(3):461-468. doi:10.1002/ajmg.c.31717. PMID: 31260186.