Nemaline Myopathies

Literature and regulatory information updated: 17 October 2026.

Congenital myopathies with nemaline rods caused by variants in genes of the sarcomeric thin filament (NEB, ACTA1, TPM2, TPM3, TNNT1, LMOD3, CFL2) and of the BTB-Kelch family (KBTBD13, KLHL40, KLHL41), together with MYPN and MYO18B; severe neonatal, intermediate, typical, childhood-onset, and adult-onset forms; histology, genetics, respiratory, nutritional, and orthopaedic management, and investigational therapies.

0. Reference Summary Table

Nemaline myopathy (NM) belongs to the congenital myopathies as a group. RYR1-related core-rod myopathies and malignant hyperthermia susceptibility are part of the differential diagnosis. Acquired sporadic late-onset nemaline myopathy (SLONM) is considered only in the differential diagnosis.

MANE Select (Matched Annotation from NCBI and EMBL-EBI) transcripts confirmed in VariantValidator [23].

Table sources: [2, 4, 9, 24, 25].

1. Introduction

1.1. Original description and early clinical series

Nemaline myopathy was described independently in 1963 by two North American groups. Shy, Engel, Somers, and Wanko published in Brain the article that coined the term, "Nemaline myopathy. A new congenital myopathy" [26]. In the same year, Conen, Murphy, and Donohue reported, among more than one hundred biopsies, the case of a 4-year-old boy with hypotonia and non-progressive weakness whose two gastrocnemius biopsies contained, on light and electron microscopy, small rod-shaped particles that they called "myogranules" [27]. The name derives from the Greek nema, thread, and refers to the threadlike structures seen in the muscle fibres; until then these patients had been labelled as having "amyotonia congenita" or benign congenital hypotonia [2]. Fox and colleagues date the first observation of these structures in hypotonic infants to 1958, five years before the syndrome was delineated [17].

The clinical heterogeneity led the 51st ENMC workshop, held in Naarden in 1997, to propose a classification into six forms defined by age at onset and by motor and respiratory severity: severe congenital, Amish, intermediate congenital, typical congenital, childhood onset, and adult onset [2, 28]. The application of this classification to 143 Australian and North American patients by Ryan and colleagues (2001) provided the first systematic description of the natural history, with the distribution of forms, mortality by form, and early predictors of survival that are still cited [24]. The authors concluded that "aggressive early management is warranted in most cases of congenital nemaline myopathy," because morbidity from respiratory infections and feeding difficulties decreased with age [24].

1.2. Milestones of molecular discovery

The first locus was mapped in 1992 to chromosome 1 in a large Australian family with autosomal dominant childhood-onset NM (NEM1) [29], and in 1995 the p.Met9Arg variant was identified in the slow tropomyosin encoded by TPM3 [30]. In 1999 recessive variants in nebulin (NEB), now the most frequent cause, were described [31], as were variants in skeletal α-actin (ACTA1), which account for half of severe forms [32]. In 2000 Johnston and colleagues mapped the Amish form to an interval of about 2 cM at 19q13.4 and found a stop codon in exon 11 of TNNT1 [16]; β-tropomyosin (TPM2) was added in 2002 [33], and the founder deletion of NEB exon 55 in Ashkenazi Jews in 2004 [34].

Between 2007 and 2017 the list grew to include cofilin-2 (CFL2) in 2007 [35], KBTBD13 in 2010 in a family with rods and cores [36], KLHL40 and KLHL41 in 2013 [37, 38], LMOD3 in 2014, identified by exome sequencing in 21 patients from 14 families [3, 21], MYO18B in 2015 in a Klippel-Feil syndrome with myopathy [39, 40], and MYPN in 2017 [22, 41]. The genes discovered after 2010 shifted understanding of the disease: NM was no longer solely a disease of thin-filament proteins but also included proteins of the ubiquitin-proteasome pathway that stabilise this filament [2]. More recently, the dominant form of TNNT1 (2017) [42] and dominant distal myopathy due to large heterozygous NEB deletions (2025) [7] have been described.

1.3. Nomenclature and reclassification

The GeneReviews chapter devoted to NM, published in 2002 and updated until 2015, was retired on 7 November 2019, on the grounds of "histological diagnosis without strong genetic correlation" [2]. The retirement reflects a paradigm shift: NM is a histological pattern shared by genetically distinct diseases, and management and counselling depend on the gene. Sewry and colleagues proposed in 2019 a classification that eliminates the intermediate and "other" classes and mixes clinical and genetic categories, revised in 2021; Amburgey and colleagues created a numerical score based on ambulation, respiratory support, and feeding [4, 43]. According to Hildebrandt and colleagues, none of these schemes yielded reliable predictors of long-term outcome [4].

OMIM retains the designation "nemaline myopathy" (NEM) for NEB (NEM2), TNNT1 (NEM5A, NEM5B, NEM5C), KBTBD13 (NEM6), CFL2 (NEM7), KLHL40 (NEM8), KLHL41 (NEM9), and LMOD3 (NEM10), but has reclassified as "congenital myopathy" (CMYO) the entities due to TPM3 (formerly NEM1, now CMYO4A and CMYO4B), ACTA1 (NEM3, now CMYO2A, CMYO2B, and CMYO2C), TPM2 (NEM4, now CMYO23), and MYPN (NEM11, now CMYO24) [8]. The correspondence between the old and new designations rests on matching OMIM numbers; the OMIM synonymy texts were not consulted because the site refused automated access.

The 277th ENMC workshop (Hoofddorp, June 2024; published in 2026) considered histological nomenclature outdated in a group with more than 40 genes and up to 40% of cases without a genetic diagnosis [1]. It defined congenital myopathies as genetic muscle diseases that typically present in the perinatal period or in infancy with hypotonia and weakness, usually non-progressive or slowly progressive, with distinctive structural fibre abnormalities and no overt dystrophic features; it adopted the acronym CMYO and proposed naming each entity in the order inheritance, gene, CMYO, and main histopathological finding, e.g., "AR NEB CMYO nemaline" for the former NEM2 and "AD ACTA1 CMYO nemaline" for CMYO2A [1]. Patient representatives rejected severity scales in the disease name, and OMIM had not adopted this format at the time [1, 8].

1.4. Definitions and scope

The abbreviation NM designates the inherited congenital myopathy in which nemaline rods are the dominant histological abnormality, or in which a pathogenic variant in an NM gene explains a compatible phenotype [1, 5]. The 2014 diagnostic consensus includes cap myopathy, zebra body myopathy, and core-rod myopathy within the NM family as pathological variants, and excludes SLONM from the congenital myopathies because of its late onset, rapidly progressive course, uncertain genetic basis, and association with monoclonal gammopathy [5]. The clinical forms follow the ENMC definitions as applied by Ryan (2001) and those of the modified Lehtokari classification used by de Feraudy (2026), which adds distal myopathy and the atypical form (ophthalmoplegia, cardiomyopathy) [24, 25].

RYR1-related myopathies with cores and rods ("core-rod") share histology with NM but carry their own risk of malignant hyperthermia. MYO18B, TNNT3, ADSS1, CAP2, RYR1 and RYR3 have been associated with myopathy with NM features without reaching moderate or definitive evidence for NM in ClinGen [4]; MYO18B is discussed alongside the NM genes because rods were present on biopsy in the reported patients [39, 40].

2. Molecular Genetics and Pathophysiology

2.1. The thin filament and the common pathogenic pathway

Skeletal muscle contraction depends on the interaction between thin filaments, anchored at the Z-disc, and thick myosin filaments. The thin filament is a double helix of polymerised globular actin monomers (F-actin) along which runs a nebulin molecule; at rest, tropomyosin dimers sterically block the myosin-binding sites, and the troponin complex, formed by troponin I (inhibitory), troponin T (which binds tropomyosin), and troponin C (which binds calcium), shifts tropomyosin into the actin groove when calcium binds to troponin C, exposing the myosin-binding sites [2]. Seven of the NM genes encode components of this filament or proteins that regulate its assembly (NEB, ACTA1, TPM2, TPM3, TNNT1, CFL2, LMOD3); three belong to the BTB-BACK-kelch (BBK) family and are involved in stabilisation or targeted degradation of sarcomeric proteins (KBTBD13, KLHL40, KLHL41) [2]. Myopalladin (MYPN) links nebulin to α-actinin at the Z-disc [3].

Nemaline rods are focal aggregates of Z-disc–derived proteins, in structural continuity with the Z-disc, composed mainly of α-actinin and also containing telethonin, filamin, myotilin, myozenin, and myopalladin [2, 4]. Rod formation is a common muscle response to diverse insults, which explains their presence in acquired diseases and in other genetic myopathies [2]. Hildebrandt and colleagues describe dysregulation of the thin filament, and thereby of the actin–myosin cross-bridge cycle, as the final common pathogenic pathway [4]. Binding of the KLHL40 protein to nebulin and leiomodin-3 places three of the genes in a single thin-filament stabilisation pathway [2, 44].

2.2. Relative distribution of genes

The proportion attributed to each gene depends on the era, recruitment method, and population. The estimates from GeneReviews (2015) and the 2014 diagnostic consensus came from research laboratory series; the figures from the 250th ENMC workshop (2019) pool genetically diagnosed patients from nine European and North American centres; the Boston cohort (Hildebrandt 2025) is a referral cohort biased towards severe forms; and the data from Cicala (2026) refer to all congenital myopathies diagnosed by panel in the United Kingdom, not only NM.

Sources: [2, 4–6, 45]. The ENMC 250 table contains summation inconsistencies in the original (Netherlands row with 66 patients for a total of 68; column of undiagnosed cases with 60 for a total of 69) [6].

The 250th workshop summarised that NEB (35%) and ACTA1 (24%) predominate, that TPM2, TPM3, and KLHL40 are frequent in paediatrics, that the dominant KBTBD13 founder variant makes NEM6 the most common form of NM in the Netherlands, and that KLHL41, CFL2, and MYPN are rare [6]. Most ACTA1-related NM is severe, but NEB remains a slightly more frequent cause of the severe form, according to Amburgey's natural history data (2017) cited by the workshop [6].

2.3. NEB (nebulin)

NEB has 183 exons spanning about 249 kb and is one of the largest genes in the human genome; it is expressed almost exclusively in skeletal muscle [2, 25]. Nebulin, 600 to more than 770 kDa depending on the isoform, accounts for 3–4% of myofibrillar protein, extends along the thin filament, and has 22–29 tandem super-repeats, each with seven simple repeats of 30–35 amino acids containing the SDxxYK motif, a probable actin-binding site; the C-terminal end is inserted into the Z-disc [2, 3, 25]. The protein stabilises F-actin, specifies thin-filament length, and is involved in Z-disc organisation and in regulation of contraction [3, 25]. The gene's architecture complicates analysis: 42 of the 183 exons undergo alternative splicing (exons 63–66 are skipped together, exons 143 and 144 are mutually exclusive, and exons 166–177 undergo extensive splicing that varies with developmental stage), and a block of eight exons between exons 82 and 105 is triplicated with nearly identical sequence (triplicate region, TRI) [6, 25].

The mechanism is deficiency of functional nebulin, with altered thin-filament length, sarcomere disorganisation, and reduced force generation [3, 25]. The classic disease is autosomal recessive, and most patients are compound heterozygous for private variants [6]. In the de Feraudy cohort (49 patients from 48 families, eight countries), 42 patients were compound heterozygous and 7 homozygous; of the 98 causal alleles, 58% predicted a premature stop codon (nonsense 17%, frameshift indel 41%), 34% affected splicing (including 4 synonymous variants), 5% were missense, and 3% were in-frame deletions or duplications; 51% of patients had at least one variant in exons 169–175, in the C-terminal region involved in Z-disc anchoring [25]. Nebulin tolerates considerable amino acid variation: gnomAD v4 lists more than 28,000 NEB variants, many of them missense in presumably healthy individuals, and only about 7% of the coding variants observed in a population exome database are pathogenic, which makes missense interpretation unusually difficult [3, 25].

Structural variants account for a substantial fraction of pathogenic alleles. Sagath and colleagues studied 35 families with custom comparative genomic hybridisation (CGH) arrays, exome, short-read genome, droplet digital PCR, and Sanger sequencing: in 23 recessive families there were intragenic deletions or duplications, or pathogenic gains of the triplicate region, in trans with a small variant, with no correlation between variant type and severity [7]. In the remaining 12 families, eight large heterozygous deletions of 52 to 97 exons, present in the heterozygous state (10 families) or in mosaic form (2), caused a predominantly distal, milder myopathy with dominant transmission; this was the first statistically significant genotype–phenotype association for NEB [7]. In six unrelated South African families, a heterozygous deletion of exons 14–77 caused dominant distal myopathy in 17 affected individuals, with median onset at 14 years (1–48), initial weakness of the ankle dorsiflexors and finger extensors, T1 hyperintensity of the tibialis anterior on magnetic resonance imaging (MRI), and non-specific myopathic changes on biopsy [46]. A gain of two or more additional copies of the triplicate region is interpreted as pathogenic [6].

The 2502-bp deletion that removes exon 55 and parts of introns 54 and 55 is the only recurrent variant with worldwide distribution. Anderson and colleagues found it in 5 patients from 5 Ashkenazi families; the resulting transcript encodes 35 fewer amino acids [34]. The Human Genome Variation Society (HGVS) notation on the MANE transcript is NM_001164508.2:c.7431+1919_7536+374del (GRCh38 NC_000002.12:g.151645758_151648259del; GRCh37 NC_000002.11:g.152502272_152504773del), identical in NM_001271208.2 and NM_004543.3; the historical notation NM_004543.3:c.7622-2025_7727+372del2502 is rejected by VariantValidator because of an exon boundary error [23]. Exon 55 is 105 nucleotides long, consistent with the in-frame loss of 35 amino acids described as p.(Arg2478_Asp2512del); this protein consequence is the authors' inference and not a VariantValidator prediction, which returns p.? for a deletion with intronic boundaries [23, 34]. In a worldwide series of 355 probands, the deletion was present in 14, including two families without known Ashkenazi ancestry but with the same haplotype; in 6 of 8 homozygotes the phenotype was more severe than the typical form [47]. In the Boston cohort, the deletion was associated with symptoms at birth more often than other NEB variants (odds ratio [OR] 7.15; 95% CI 1.2–78.9), and mortality was 3/13 versus 10/83, a difference that was not statistically significant (OR 2.2; 95% CI 0.56–8.99) [4]. Many patients with this deletion have cores on biopsy, according to de Feraudy [25].

Other recurrent variants occur in specific populations. In de Feraudy, the duplication c.24294_24297dup p.(Glu8100Serfs*5) in NM_001271208.1, which corresponds to NM_001164508.2:c.24189_24192dup p.(Glu8065SerfsTer5), was found in 8 unrelated families, 7 of them diagnosed in Argentina or Paraguay, with no shared haplotype in the three families studied, which the authors interpret as independent origin or an ancient founder effect [23, 25]. In the Brazilian cohort of Moreno (33 patients), the intronic variant NM_001164508.2:c.5343+5G>A, absent from gnomAD, was present in 5 patients, and the synonymous c.19944G>A p.(Ser6648=), with a splicing effect according to the authors, in 3 severely affected patients [23, 48].

Beyond NM, at the severe end NEB causes lethal arthrogryposis multiplex congenita 6 (AMC6, OMIM 619334), with contractures due to reduced fetal movement and death in utero or in infancy, and at the mild end recessive (ORPHA:399103) and dominant (ORPHA:708123) distal myopathies, core-rod myopathy, and adult-onset weakness [3, 8, 10, 48]. In nebulin-related distal myopathy, the biopsy may not show rods, so their absence does not exclude the gene [3, 10]. Hildebrandt described a patient with a dominant de novo NEB variant presenting as congenital fibre-type disproportion [4].

2.4. ACTA1 (skeletal α-actin)

ACTA1 has 7 exons and encodes skeletal α-actin, one of the six human actins, all highly conserved and all associated with disease [2, 3]. Because of this conservation, most non-synonymous variants are pathogenic, and no missense polymorphism was known as of 2014 [5, 6]. About 90% of pathogenic variants are dominant missense variants, mostly de novo; about 10% are recessive, usually null, with persistent expression of cardiac actin in all fibres; parental somatic and germline mosaicism has been documented [5, 6]. The mutant protein exerts a dominant-negative effect on thin-filament assembly and function, with misfolding, altered polymerisation, and aggregation; the combination of these effects produces cytoplasmic and intranuclear rods, thin-filament accumulation, and myofibrillar disorganisation, and the specificity of each variant explains part of the variation in severity [2]. Null mice upregulate cardiac and vascular actin in skeletal muscle [2].

The 2024 update of the variant database, with 607 reported variants, showed that the number of described phenotypes had increased from 5 to 20, that 74% of variants cause NM, and that a growing number cause cardiomyopathy and distal myopathy [3, 49]. OMIM distinguishes CMYO2A (AD, typical, from severe neonatal to mild childhood), CMYO2B (AR, severe infantile), and CMYO2C (AD, severe infantile, with reduced fetal movement, absence of antigravity movement at birth, and death from respiratory failure typically in the first year) [3, 8]. In the Boston cohort, the 2 patients with recessive loss-of-function variants had severe disease, and parental mosaicism was suspected in 2 families without confirmatory testing in the parents [4]. Allelic disorders include congenital fibre-type disproportion with selective hypotrophy of type 1 fibres, clinically indistinguishable from severe NM in 3 patients [2, 50], thin-filament excess myopathy, zebra body myopathy, cap myopathy, myopathy with cores (predominantly in type 2 fibres, unlike RYR1-related myopathies), rigid spine congenital muscular dystrophy due to a homozygous missense variant, and scapulohumeroperoneal myopathy (SHPM, OMIM 616852), a mild, late-onset form with scapular winging, foot drop, and distal weakness [2, 3, 8]. Cardiomyopathy is a rare but documented complication: Wang 2012 cites a fatal case of hypertrophic cardiomyopathy with p.Lys336Glu (historical numbering, = NM_001100.4:c.1012A>G p.(Lys338Glu)), and in 2026 a new likely pathogenic variant associated with dilated cardiomyopathy was published [18, 51].

2.5. TPM3 and TPM2 (tropomyosins)

TPM3 has 13 exons and several isoforms; slow α-tropomyosin 3 is expressed mainly in type 1 fibres [2, 3]. The founder variant from the original description, NM_152263.4:c.26T>G p.(Met9Arg), alters the N-terminal region that binds actin and troponin T and the head-to-tail interactions of the dimer, with a dominant-negative effect, altered calcium-activated force production, accelerated relaxation, and a shift in the force–frequency relationship [2, 23, 30]. The recessive variant NM_152263.4:c.94C>T p.(Gln32Ter), described in the homozygous state in an infant with extreme motor delay, appears in the older literature as Gln31Ter, a numbering that omits the initiator methionine [2, 23]. Recessive alleles reduce or abolish the protein [3]. OMIM separates CMYO4A (AD) from CMYO4B (AR, more severe, with congenital contractures, markedly delayed or absent walking, and respiratory insufficiency) [3, 8]. On biopsy, the most common TPM3 pattern is isolated fibre-type disproportion; rods, when present, are restricted to type 1 fibres and may coexist with caps immunoreactive for tropomyosin [3, 5]. Some TPM3 and TPM2 variants at tropomyosin–actin contact sites increase calcium sensitivity and cause hypercontractility (increased tone, trismus) with a non-specific biopsy lacking rods [3, 52]. TPM3 has numerous pseudogenes on other chromosomes, a known pitfall in short-read alignment [3, 53].

TPM2 has 10 exons and encodes β-tropomyosin, expressed mainly in type 1 fibres [2, 3]. By 2015, 30 distinct pathogenic variants were known, with phenotypes of NM, cap myopathy, core-rod myopathy, fibre-type disproportion, and distal arthrogryposis, with no clear relationship between variant type or position and the clinical or morphological phenotype [2, 54]. Two phenotypic groups can be distinguished: congenital myopathies (CMYO23, OMIM 609285), with usually moderate hypotonia and weakness, static or slowly progressive and without congenital contractures, in which cap structures are the most frequent morphological finding; and distal arthrogryposis types 1A and 2B4 (Sheldon-Hall syndrome), both under OMIM 108120, with camptodactyly, clubfoot, and mild or absent weakness [3, 8]. The disease is predominantly dominant (ClinGen: definitive AD association), and ClinGen dosage data (haploinsufficiency score 1, triplosensitivity score 0) indicate that dominant variants act through an altered protein and not through loss of dosage [3]. The rare recessive exception is complete absence of β-tropomyosin due to a homozygous null allele or intronic variant, which causes the non-lethal Escobar variant of multiple pterygium syndrome with congenital myopathy; the heterozygous parents are clinically normal [55, 56]. In the Swedish population cohort, the same TPM2 variant caused cap myopathy in one family member and NM in another [15].

2.6. TNNT1 (slow skeletal troponin T)

TNNT1 has 14 exons and encodes the slow skeletal isoform of troponin T, expressed only in type 1 fibres; it is not cardiac troponin (TNNT2), so cardiac disease protocols do not transfer to this entity [2, 3, 17]. Slow troponin T confers greater calcium sensitivity on the thin filament than the fast isoform [2]. The Amish founder variant is NM_003283.6:c.538G>T p.(Glu180Ter) on the MANE transcript (GRCh38 NC_000019.10:g.55137176C>A; ClinVar VCV 12440, pathogenic, multiple submitters with no conflicts, last evaluated on 10 July 2025) [23, 57]. The same variant is written c.505G>T p.(Glu169Ter) on the shorter isoforms NM_001126133.3 and NM_001291774.2; the notation "c.505G>T" used by Fox and colleagues and by GeneGuidance refers to this alternative transcript and not to a different variant, although these texts combine it with the protein numbering p.Glu180Ter of the long transcript [17, 23, 57]. Johnston (2000) described it as a stop codon in exon 11 that truncates the protein at amino acid 179, a numbering that predates current HGVS nomenclature [16].

The truncation removes the 83 C-terminal amino acids, eliminates the troponin C– and troponin I–binding sites, and greatly reduces affinity for tropomyosin; the fragment is translated in cell systems but is absent from patient muscle because of rapid proteolysis, so the variant behaves as a null allele [17, 58]. Strength is normal at birth because the fetal isoforms of fast and cardiac troponin T compensate for the absence of the slow isoform; postnatal onset and progression parallel the downregulation of these isoforms [2]. In the mouse, the p.Glu180* knock-in model has more severe myopathy than the gene knockout, and the mutant mRNA is detectable whereas the protein fragment is not; the authors propose that the fragment is cytotoxic and that gene replacement may not resolve this component [59]. A homozygous NM_003283.6:c.653C>G p.(Pro218Arg) variant caused lethal NM through altered splicing, with inclusion of exon 12' in 49.85% of transcripts, a mechanism distinct from classic loss of function [60].

OMIM distinguishes three entities: NEM5A (AR, severe infantile, "Amish"), NEM5B (AR, childhood onset, with proximal weakness, gait abnormality, and ambulation preserved until late adulthood, but with progressive restrictive respiratory insufficiency), and NEM5C (AD, milder, with onset in the first or second decade) [3, 8]. NEM5C rests on the variant NM_003283.6:c.311A>T p.(Glu104Val), proposed as dominant negative by Konersman (2017) [42]; ClinVar classifies it as of uncertain significance (VCV 940131, one submitter), and ClinGen rates the dominant association as "limited," versus "definitive" for biallelic loss of function, so an isolated heterozygous TNNT1 missense variant should not be reported as an established cause [3, 57]. Outside the Amish, families from Italy, Israel, Spain, Korea, Japan, and Palestine have been described; nine Palestinian children homozygous for c.574_577delinsTAGTGCTGT had a phenotype similar to the Amish form [3, 17].

2.7. KBTBD13

KBTBD13 has a single exon and encodes a 458-amino-acid (49-kDa) protein with a BTB/POZ domain and five Kelch repeats, expressed in skeletal and cardiac muscle [2]. The protein is a substrate adaptor of the cullin 3–based E3 ubiquitin ligase, binds actin, and modulates contraction and relaxation kinetics [3]. Heterozygous missense variants in the Kelch repeats disrupt the β-propeller and reduce the relaxation rate by about 35–40%, which explains the characteristic slowness of NEM6 [3, 61]. The three variants in the original description are NM_001101362.3:c.742C>A p.(Arg248Ser), c.1170G>C p.(Lys390Asn), and c.1222C>T p.(Arg408Cys); the last is the Dutch founder variant [23, 36]. In the p.Arg408Cys knock-in mouse, the variant displaces KBTBD13 from the sarcomere [62]. Because the mechanism is not dosage-related, a whole-gene deletion should not be reported as equivalent to the known missense variants [3]. KBTBD13 rods are frequently arranged in rings ("ring rods") and coexist with cores, which led the authors to propose the name rod-core myopathy [63].

2.8. KLHL40, KLHL41, and LMOD3

KLHL40 has 6 exons and encodes a 621-amino-acid BTB-BACK-kelch protein, specific to striated muscle and located in the A and I bands [2, 3]. Contrary to what its role as a ubiquitin ligase adaptor would predict, it stabilises its partners: it binds nebulin and blocks ubiquitination of leiomodin-3; in its absence, thin-filament proteins are lost, and the Klhl40-null mouse recapitulates the disease [44]. The protein is absent from patient muscle, and missense variants destabilise it [3, 37]. NEM8 is autosomal recessive; in southern China, the variant NM_152393.4:c.1516A>C p.(Thr506Pro) is a founder variant (ClinVar VCV 846771, pathogenic) [57, 64].

KLHL41 (sarcosin) has 6 exons and encodes a 606-amino-acid protein that regulates myoblast proliferation and differentiation, promotes lateral fusion of myofibrils, and stabilises the sarcomere through non-proteolytic ubiquitination; its loss dysregulates degradation of the NRAP protein [2, 3, 65]. The ClinGen curation of 28 October 2024, compiled 5 probands in 3 publications, with at least 8 variants (missense, in-frame indel, frameshift, and large deletion) and a homozygous loss-of-function mechanism [3]. Inheritance is recessive in OMIM and ClinGen; the Orphanet entry for the intermediate form (ORPHA:171433) lists dominant and recessive inheritance for the genes as a whole, which does not change the recessive inheritance of KLHL41 [3, 9]. In zebrafish, loss of Klhl41b reduces sarcomere size and causes defective assembly followed by progressive destabilisation and nemaline bodies [66].

LMOD3 has 3 exons and encodes leiomodin-3, a 560-amino-acid (65-kDa) protein with three actin-binding domains and a tropomyosin-binding helix; it localises near the pointed ends of thin filaments and is a potent actin nucleator, essential for myofibrillogenesis [2, 21]. In its absence, thin filaments are short and disorganised [21]. Truncating alleles cause the severe congenital form, usually lethal; the variant NM_198271.5:c.1648C>T p.(Leu550Phe) (ClinVar VCV 374498), found in the homozygous state or in trans with another missense variant in four unrelated patients from Tyrol, Upper Austria, and Bavaria, allows survival into adulthood, possibly through greater residual expression [3, 57, 67].

2.9. CFL2, MYPN, MYO18B, and genes with limited evidence

CFL2 has 5 exons and encodes cofilin-2, the muscle isoform of the cofilin/ADF family, which depolymerises actin in a pH-dependent manner and is required for actin isoform switching during postnatal sarcomere remodeling [2, 3]. The variant in the original consanguineous family of Middle Eastern origin is NM_138638.5:c.103G>A p.(Ala35Thr) (ClinVar VCV 8160); the notation "c.103C>A" that appears in secondary sources is incorrect, because the reference nucleotide is G [23, 35, 57]. The mutant protein is less soluble and depolymerises less, leading to actin accumulation in rods, minicores, and concentric laminated bodies [3, 35]. Subsequently, NM_138638.5:c.19G>A p.(Val7Met) was reported in an Iraqi Kurdish family with NM and myofibrillar myopathy, and a homozygous 4-bp deletion in patients ventilator-dependent from infancy [23, 68, 69]. On sequencing of 113 patients with NM of unknown genetic basis, only the original family had CFL2 variants [2].

MYPN encodes myopalladin, a Z-disc protein that links nebulin (in skeletal muscle) and nebulette (in the heart) to α-actinin and that also has a nuclear localisation [3]. Biallelic variants cause a mild, slowly progressive childhood-onset NM (CMYO24), with weakness and atrophy mainly of the lower limbs and neck; in Miyatake's four families, walking difficulties did not appear until around the fifth decade, and reduced respiratory function, cardiac involvement, and intranuclear rods were each observed in 2 patients [22]. They also cause cap myopathy with occasional rods and NM with caps in consanguineous South Asian families [41, 70]. ClinGen classifies the association of heterozygous variants with dilated cardiomyopathy (CMD1KK) as "limited" and with hypertrophic cardiomyopathy as "disputed," so an isolated heterozygous variant on a cardiomyopathy panel does not explain cardiac disease [3, 8].

MYO18B encodes an unconventional nucleocytoplasmic myosin required for sarcomere assembly in fast muscle [3]. Biallelic loss-of-function variants cause Klippel-Feil syndrome 4 (KFS4), with fusion of cervical vertebrae, congenital myopathy with rods, facial dysmorphism (ptosis, bulbous nose), and short stature; in Altuame's series of 7 patients, myopathy was universal and slowly progressive or non-progressive, and the Klippel-Feil anomaly and dysmorphism were present in 5 [39, 71]. Several Saudi patients share the founder variant NM_032608.7:c.6905C>A p.(Ser2302Ter) (ClinVar VCV 208842), and a homozygous nonsense variant in the last exon caused severe NM with cardiomyopathy [40, 57]. The somatic role of MYO18B as a tumour suppressor in lung cancer has no germline significance [3].

TNNT3, ADSS1, CAP2, RYR1, and RYR3 have been associated with myopathy with NM features without reaching moderate or definitive evidence for NM in ClinGen [4]. The 2023 UK congenital myopathy panel includes TNNT3 among the NM genes [45]. Muscle adenylosuccinate synthetase deficiency (ADSS1, formerly ADSSL1) is part of the differential diagnosis (section 6). In a series from a Japanese referral centre (repository from January 1978 to March 2019), Saito and colleagues identified 63 patients from 59 families with biallelic ADSS1 variants, the most frequent cause of genetically diagnosed NM at that centre; two variants, c.781G>A and c.919delA (published designations), accounted for 53.2% and 40.5% of alleles, suggesting a founder effect. Symptoms began in adolescence and involved the proximal and/or distal leg muscles, the tongue, the masseter, the diaphragm, and the paraspinal muscles; in the late stages, 26 of 63 patients had dysphagia with masticatory dysfunction, 12 of 48 hypertrophic cardiomyopathy, and 26 of 34 restrictive ventilatory insufficiency. MRI showed fatty infiltration of the periphery of the vastus lateralis, gastrocnemius, and soleus in all patients, and biopsy showed nemaline bodies with increased lipid droplets and myofibrillar disorganisation [72]. The frequency refers to the centre’s case series and is not an estimate of national prevalence.

2.10. Recurrent and founder variants with standardised nomenclature

General source of standardisation: VariantValidator (REST, GRCh38, MANE Select) and ClinVar, accessed on 17 October 2026 [23, 57].

2.11. Allelic disorders

2.12. Genotype–phenotype correlations, penetrance, and modifiers

Genotype–phenotype correlation is weak in recessive NEB disease: an update series of 159 families found no correlation, nor did the study of 35 families with structural variants in the recessive cases [7, 76]. The de Feraudy cohort suggests exploratory patterns: missense variants and in-frame indels occurred only in the typical, distal, and childhood-onset forms, all 7 intermediate forms had two truncating variants, canonical splice-site variants predominated in the severe forms, and 16 of the 18 patients with two truncating variants had at least one in an alternatively spliced exon; the authors infer that two truncations in constitutive exons would be incompatible with life, as in fetus P34, terminated at 17 weeks with akinesia, arthrogryposis, and hydrops [25]. Ophthalmoplegia, observed in only one adult patient with a heterozygous nonsense variant in alternative exon 143, led to the unvalidated hypothesis of a link with the higher expression of isoforms that include this exon in the extraocular muscles [25].

In ACTA1, variability is marked even within a family: three members with the same variant had very different severities, showing that genotype is not the only determinant [2, 77]. In NEB, two unrelated patients homozygous for the same variant required ventilation at 17 months and at 10 years, respectively, and only one had scoliosis; the authors invoke modifier genes and environmental factors [25]. In the Boston cohort, ACTA1 variants, compared with NEB variants, were associated with the severe congenital form (OR 5.4; 95% CI 2.31–12.11), invasive ventilation in the first year (OR 4.7; 1.9–11.0), and gastrostomy (OR 2.2; 1.2–3.9), with no difference after the first year [4]. In KLHL41, frameshift variants are associated with neonatal death and missense variants with survival into late childhood or young adulthood; in LMOD3 prognosis follows variant type [2, 3]. The number and distribution of rods do not correlate with severity or prognosis [3, 77].

There are insufficient data to estimate the penetrance of the dominant forms due to ACTA1, TPM3, TPM2, and KBTBD13 [2]. Parents mosaic for a dominant variant may be mildly or minimally affected [2], and in distal myopathy due to large NEB deletions two of the cases were mosaic [7]. In recessive inheritance, clinically healthy heterozygous parents may have NM changes on biopsy [2]. In muscle proteomics from 7 infants with NM (3 with dominant de novo ACTA1 variants and 4 compound heterozygous for NEB), 183 of 4846 proteins were dysregulated and none differed between the ACTA1 and NEB groups, which the authors interpret as an argument for gene-independent therapies (conference abstract) [78].

3. Clinical Spectrum and Manifestations by System

3.1. Common features

NM presents with weakness, hypotonia, and reduced or absent deep tendon reflexes, with weakness most pronounced in the face, neck flexors, and proximal limb muscles, and distal involvement in a subgroup [2]. In the congenital forms, the face is long and expressionless, with a "tented" upper lip, high-arched palate, and retrognathia; cognition is preserved, sensation is normal, and the extraocular muscles are usually spared [2, 5]. Joint hypermobility is common in early childhood and gives way over time to contractures and deformities, including scoliosis [2]. About 25% of children with congenital NM require tube feeding or gastrostomy in the first years [2]. Involvement of the diaphragm and intercostal muscles is common, and the degree of skeletal weakness does not necessarily reflect respiratory involvement, particularly in older children and adults [2, 24]. A cardiogenetics handbook describes disproportionate axial and respiratory involvement as common to all NM variants and as the usual determinant of long-term prognosis [79].

3.2. Clinical forms

The six forms of the ENMC classification overlap, and the distinction between the intermediate and typical forms is often made only retrospectively, because no single parameter in infancy separates them [2, 24]. Ryan considered patients typical if they crawled before 12 months and walked before 18 months [2]. The modified Lehtokari classification used by de Feraudy (2026) defines the severe form by absence of spontaneous movement or spontaneous respiration, or by severe contractures or fractures at birth; the intermediate form by early need for ventilatory support and inability to sit or walk, or wheelchair use before age 11 years; the typical form by delayed but achieved motor milestones, a slowly progressive or static course, and wheelchair use, if needed, only after age 11 years; and it adds distal myopathy and the atypical form [25].

Sources: [2, 4, 17, 24]. The Amburgey score assigns 6 points if the patient is non-ambulatory, 2 points for each level of respiratory support (none 0, nocturnal non-invasive only 1, daytime and nocturnal 2, invasive 3), and 1 point for a feeding tube, with a maximum of 13; the severe and intermediate classes did not differ from each other, nor did the typical from the mild [4]. The published interquartile range for age at diagnosis in the intermediate form is incompatible with the median and appears to be an editorial error in the article [4].

In the severe congenital form, hypotonia and weakness are evident at birth, with little spontaneous movement, sucking and swallowing difficulties, gastro-oesophageal reflux, and respiratory insufficiency; reduced fetal movement and polyhydramnios are frequent, and death in utero with fetal akinesia has been described [2]. Early mortality is due mainly to respiratory failure or aspiration pneumonia, but there are long-term survivors despite severe weakness and inadequate respiration at birth [2]. The intermediate form is distinguished by the early development of contractures [2]. In the typical form, respiratory involvement is less prominent but usual, often subclinical, with insidious nocturnal hypoventilation and frequent lower respiratory tract infections; weakness is proximal at presentation, and distal involvement appears late in a minority [2]. Respiratory compromise affected about 96% of patients with the intermediate form and about 37% of patients with the typical form in Ryan's series, according to the 2012 consensus [18].

The childhood-onset form was described in a large Australian family with dominant TPM3-related NM: normal early motor development, symmetric weakness of ankle dorsiflexion with foot drop at the end of the first or beginning of the second decade, mimicking a neuropathy, slow progression to all ankle movements and the proximal musculature, and two older relatives in wheelchairs by age 40 years [2, 29]. In a Dutch family with KBTBD13, childhood-onset proximal weakness was accompanied by slowness: patients moved "in slow motion" and could not jump or run, and physiological studies confirmed reduced contraction velocity and relaxation time [2, 61]. In the historical series, the adult-onset form grouped patients with generalised weakness between ages 20 and 50 years without prior symptoms or family history, sometimes with myalgia, rapid progression, cardiomyopathy, or "dropped head," now largely recognised as SLONM [2, 80, 81]. In the Boston cohort, the 12 adult-onset patients included met criteria for genetic or clinicopathological NM, and one patient with dominant TNNT1 disease had adult onset [4].

NEB-related distal myopathy presents with slowly progressive weakness of the leg and finger extensors and neck flexors, with onset between infancy and adulthood; in the dominant forms due to large deletions, weakness is milder, and 7 of the 17 South African patients reported sensory complaints without specific electrodiagnostic abnormalities [7, 10, 46]. In the de Feraudy cohort, 5 of 49 patients (10%) had the distal form [25].

3.3. Data from contemporary cohorts

Sources: [4, 6, 17, 24, 25, 48]. PEG, percutaneous endoscopic gastrostomy; FMUSP, Faculdade de Medicina da Universidade de São Paulo (University of São Paulo Medical School).

The cohorts measure different things and cannot be compared directly. The Boston cohort recruited through physician or self-referral, without independent examination, with data spanning three decades and a likely bias towards severe forms; of the 275 patients, 171 had a clinicopathological and molecular diagnosis, 70 a clinicopathological diagnosis only, 7 a molecular diagnosis only, and 32 had variants in an NM gene with another histological diagnosis (fibre-type disproportion in 20) [4]. Ancestry was European in 70%, Hispanic in 14%, and Ashkenazi Jewish in 8%, and 42% had a relative with NM [4]. The de Feraudy cohort comes 92% from Europe (mainly France) and South America and is not representative, but the proportions of typical (47%), intermediate or severe (27%), and mild (26%) forms are similar to those of earlier series [25]. The Brazilian cohort is single-centre, retrospective, and cross-sectional, with all patients examined by the first author, and does not allow population estimates [48].

In the de Feraudy cohort, weakness was present in all 49 patients, myopathic facies in 65%, ptosis in 12%, and joint hyperlaxity in 22%; the authors propose regarding hyperlaxity associated with weakness as a sign of NEB-related myopathy, after one patient had initially been thought to have Ehlers-Danlos syndrome [25]. Only 29% of those who walked did so within the normal World Health Organization window (17.5 months), and median creatine kinase (CK) was 44 U/L [25]. In the typical form (23 patients), all walked, at a median of 19 months, 3 lost ambulation, 61% required ventilation from a median age of 4 years, and 13% required nutritional support; in the intermediate form (7 patients), all were ventilated before age 3 years [25]. Ventilation was associated with nutritional support (p = 0.002); wheelchair use (10/27 versus 1/13; p = 0.07) and scoliosis (16/27 versus 4/13; p = 0.18) were more frequent in ventilated patients, without statistical significance [25].

In the Brazilian cohort, the last assessment showed independent ambulation in 55%, ambulation only with support in 12%, standing only or assistance with transfers in 9%, unsupported sitting in 12%, and inability to sit or lack of head control in 3 children aged 0.7 to 2.6 years; all had a high-arched palate, and only 30% ate all types of food [48]. The authors described tongue atrophy with a "triple furrow" pattern in 12 of 23 patients assessed, associated with dysphagia severity on the NdSSS (Neuromuscular Disease Swallowing Status Scale; p = 0.006), and proposed it as an accessible sign of bulbar involvement [48]. Support was needed in no domain in 36% of patients, in one domain in 30%, in two in 15%, and in all three (motor, respiratory, and bulbar) in 18% [48].

3.4. Manifestations by system

Respiratory. Weakness of the diaphragm and intercostal muscles causes a restrictive ventilatory defect with hypoventilation that begins during sleep and precedes daytime abnormalities; this disorder cannot be predicted from limb strength [18]. Respiratory involvement disproportionate to limb weakness is common in NEB- or ACTA1-related NM and in selected patients with TPM3, and respiratory failure may progress in adolescence or adulthood even without prior compromise [5, 18]. In an NM cohort cited by van Kleef, inspiratory muscle strength was below the reference range in 96% of patients [19]. In the Boston cohort, 13 of 131 ambulatory patients required invasive ventilation and 17 positive-pressure ventilation during follow-up, and 3 of the 13 discontinued invasive ventilation in childhood [4]. In TNNT1-related NM, respiratory weakness combines with chest wall rigidity to cause severe early hypoventilation, which requires high pressures on bilevel positive airway pressure (BiPAP) non-invasive ventilation and responds poorly to escalation; in one of the three non-Amish patients described by Zambon, severe respiratory compromise coexisted with relatively mild motor involvement [3, 82]. Chest deformity and respiratory weakness can cause pulmonary hypertension and cor pulmonale through chronic hypoxia and hypercapnia [79].

Bulbar, feeding, and speech. Dysphagia and feeding difficulties result from facial and bulbar weakness and, according to the Brazilian authors, from orofacial deformities (high-arched palate, malocclusion, inability to close the mouth) [2, 48]. In a review cited by the 2012 consensus, 17 of 19 children with NM had feeding difficulties, most with weak suck in infancy, and gastrostomy was associated with better growth and fewer respiratory infections; feeding difficulties tend to improve with age [18]. Disproportionate bulbar involvement is common in NM, particularly NEB-related [18], and in a natural history study cited by the 250th workshop it affected 52% of patients with ACTA1 and 57% of patients with NEB [6]. Hypernasal voice, dysarthria, and sialorrhea are frequent [2, 18]. In the Brazilian cohort, a patient with recurrent pulmonary infections, tracheostomised at 18 months, stopped having infections and reduced ventilation to sleep-time use only after gastrostomy, which the authors present as an example of unrecognised aspiration [48].

Musculoskeletal. Scoliosis, with or without rigid spine, is particularly common in forms with marked axial weakness, including NEB [18]. In the Brazilian cohort, skeletal deformities affected 63% of patients older than 3 years and all adults, and the mean age of patients with vertebral deformity was 24 years versus 12 years in the others [48]. Clubfoot, distal arthrogryposis, and early fixed kyphoscoliosis point to NM in the newborn; foot drop and pes cavus occur in forms with distal involvement (NEB, TPM3, TPM2) [5]. Fractures at birth occur in the severe forms, notably due to KLHL40 [5], and in the Boston cohort a patient with mild symptoms had a clavicle fracture at birth [4]. Congenital hip dislocation is possible in any congenital myopathy, although more common in the RYR1-related forms [18].

Heart. Cardiac contractility is usually normal [2]. In the Boston cohort, 12 of 226 patients with available information had cardiomyopathy (2 with NEB, 5 with ACTA1, 1 with TPM2, and 4 without a molecular diagnosis), and the authors consider heart disease rare outside critical neonatal illness [4]; in the de Feraudy cohort none of the 49 patients with NEB had cardiomyopathy [25]. Cases of dilated or hypertrophic cardiomyopathy with heart failure and sudden death have been described, mainly with ACTA1 [18, 51, 79]. KBTBD13 carries its own cardiac risk: among 65 p.(Arg408Cys) carriers from three families, 12% had left ventricular dilatation, 29% an ejection fraction below 50%, 8% atrial fibrillation, 9% ventricular tachycardia, and 20% repolarisation abnormalities; 5 had an implantable cardioverter-defibrillator, and there were 3 sudden cardiac deaths [75]. MYPN and MYO18B are also associated with cardiac involvement [22, 40]. In an international survey of patients and caregivers in 33 countries, self-reported cardiac complications affected 7% of congenital-onset forms and 16% of childhood- or adult-onset forms [83].

Central nervous system, eyes, and other. Central nervous system involvement is rare; significant cognitive impairment should prompt reconsideration of the diagnosis, except when there is hypoxic-ischaemic encephalopathy after severe respiratory failure at birth [18]. Ocular abnormalities are very rare in NM [18], but ophthalmoplegia occurs in the severe LMOD3-related forms [3], and in the Boston cohort 6 of 236 patients had ophthalmoplegia [4]. Fatigue, pain, and sleep disturbances affect quality of life in adults [6, 84]. In the international survey, all participants reported daily difficulties with mobility, fatigue, breathing, and mental health, regardless of severity [83].

Pregnancy. Pregnancy and delivery were relatively well tolerated by women with NM in Ryan's series [2]. The 2012 consensus notes that pregnancy has not been studied systematically in the congenital myopathies, that motor and respiratory worsening has been reported, and that the risk is greater in subtypes with diaphragmatic weakness [18]. When the fetus is affected, obstetric complications are frequent: polyhydramnios, reduced fetal movement, malpresentation, and fetal distress [2]; in Ryan's series there were obstetric complications in 49 of the 143 cases [24].

3.5. Gene-specific phenotypes

NEM5A (TNNT1, Amish). Fox and colleagues compiled 106 patients of Amish ancestry born between October 1923 and June 2017, from 54 families in Pennsylvania (97), Maryland (5), New York (3), and Indiana (1) [17]. Birth weight was normal (3.4 ± 0.1 kg), but 94% had failure to thrive by 9 months; neonatal signs were axial hypotonia, mild rigidity of the pelvic and shoulder girdles, and tremors of the limbs and chin with sustained clonus on provocation, which disappeared within a few months as proximal contractures developed [17]. By 12 months, hip abduction was limited to less than 10° in most; atrophy and weakness progressed in parallel with the contractures, and all children developed severe pectus carinatum with a rigid chest [17]. Of the 18 children born in the last two decades with detailed data, 61% rolled to prone at 7.6 ± 1.1 months and later lost this ability, only 2 (11%) sat unsupported, and receptive language and social development were preserved [17]. Median survival was 18 months (0.2–66 months), 76% died of respiratory failure by age 2 years, and only one child lived past age 4 years [17]. Biopsy in two children aged 14 and 16 months showed marked variation in fibre size due to type 1 fibre hypotrophy and type 2 fibre hypertrophy, rods, myofibrillar disorganisation, and vacuoles, with complete absence of slow troponin T on western blot [17].

In the international WiTNNess natural history study, 16 children with TNNT1-related infantile NM (median age 2.3 years) had a median ventilation-free survival of 20.2 months and a 100% probability of death or permanent mechanical ventilation by 60 months; only 2 of 16 sat unsupported and none stood independently, and laboratory and imaging data suggest that the disease is restricted to skeletal muscle [85]. Childhood-onset NEM5B presents with proximal weakness and gait abnormality, with most patients ambulatory until late adulthood but with restrictive respiratory insufficiency and declining forced vital capacity (FVC); a good motor prognosis does not guarantee a good respiratory prognosis [3]. Dominant NEM5C manifests in the first or second decade with difficulty walking on the heels, waddling gait, proximal weakness, and Gowers sign [3, 42].

NEM6 (KBTBD13). In the largest systematic study, of 24 Dutch patients (19 women aged 19 to 76 years and 5 men aged 25 to 57 years), 96% reported weakness and 96% slowness of movement since childhood, and 83% difficulty running [84]. Weakness was usually mild, axial, proximal, and distal, and objective function was little affected (median Motor Function Measure [MFM] score 91%; median Mini-BESTest 24 of 28; 6-minute walk distance below the lower limit of normal in only 17%), but quality of life was reduced, fatigue increased, and one third of patients fell at least once during 100 days of prospective follow-up [84]. All had a reduced muscle relaxation rate measured by transcranial magnetic stimulation (median 6.5/s; lower limit of normal 10.1/s), and diagnosis was often delayed until adulthood [84]. Biopsy in 18 patients showed rods in 85% on light microscopy and 89% on electron microscopy, and cores in 79% and 83%, respectively [63].

NEM8 (KLHL40). This is the severe end of NM, frequently prenatal, with fetal akinesia or hypokinesia, contractures, fractures, respiratory failure, and swallowing difficulty at birth; in the original description, almost half the patients had no spontaneous antigravity movement, and the variant accounted for up to 28% of tested cases in a Japanese cohort with severe NM [5, 37]. In the six patients from southern China with the c.1516A>C founder variant, there was reduced fetal movement, polyhydramnios, breech presentation, and clubfoot; two pregnancies were terminated, and the liveborn infants survived between 49 days and 17 months [64]. The disease is not uniformly lethal: an 8-year-old homozygous girl, with absent protein in muscle, had only hypotonia and weak suck at birth, motor delay, mild walking difficulty, and fatigability [86].

NEM10 (LMOD3). The severe phenotype includes a hypotonic, immobile, ventilator-dependent newborn, congenital contractures, ophthalmoplegia, prematurity, polyhydramnios, and, in the most severe cases, fetal hydrops [3, 21]. A child followed for 8 years had bilateral ophthalmoplegia, visual impairment, multiple contractures, scoliosis, home mechanical ventilation, and gastrostomy, with no voluntary movement beyond a slight nod of the head [87]. The four patients with p.(Leu550Phe) survived into adulthood, three were ambulatory, and all had marked facial weakness [67].

NEM9 (KLHL41) and NEM7 (CFL2). The KLHL41 phenotype ranges from death in infancy to slowly progressive distal weakness with preserved ambulation [3, 38]. In the original CFL2 family, two sisters had congenital hypotonia, delayed motor milestones, frequent falls, and inability to run; biopsy of one at age 2 years showed rods, occasional minicores, and actin accumulations, and that of the older sister at age 4 years only non-specific changes [35]. In the Boston cohort, the two patients with CFL2 and the two with KLHL41 were alive at last follow-up [4].

ACTA1, TPM3, and TPM2. ACTA1 spans the whole spectrum, from the severe congenital form with death in the first year to childhood myopathy with survival into adulthood, and in some patients weakness progresses severely in late childhood [2, 79]. Dominant TPM3-related NM (CMYO4A) presents with mild motor delay, proximal and neck flexor weakness, difficulty running, and fatigability; many have reduced vital capacity without symptoms, which calls for active screening, and in disease with caps MRI shows a pattern of fatty infiltration with masseter hypertrophy [3]. TPM2-related NM has a typical congenital phenotype, sometimes asymmetric weakness, and, with some variants, distal arthrogryposis or large-joint contractures with pterygia [5]; in the Boston cohort there were no deaths among the 9 patients with TPM2 (ages 7 to 58 years), and the median age at diagnosis was 13 years [4].

3.6. Natural history and treated disease

The Amish form is the only setting in which the natural history of NM has been observed without life-support technologies: Amish parents uniformly declined hospitalisation, gastrostomy, and mechanical ventilation, care was palliative and home-based, and over almost a century there was no correlation between year of birth and age at death [17]. Nine Palestinian children homozygous for another TNNT1 variant had a similar phenotype, but 78% were fed by gastrostomy and all started ventilatory support at a median of 17 months; when ventilator dependence is taken as the equivalent of death, "ventilator-free" survival falls from 70 to 17 months, almost identical to that of the Amish form [17].

In the Boston cohort, survival appeared bimodal, with declines in the first year of life and after age 60 years, and the group without a molecular diagnosis had shorter survival, which the authors attribute to an older cohort with less access to molecular testing and life-prolonging interventions [4]. In 2001 Ryan described a stormy initial course, with frequent respiratory infections, often followed by stabilisation [2, 24]. Limb weakness is generally non-progressive, and those who walk rarely lose ambulation in childhood; in adulthood loss is possible, and contractures, scoliosis, fractures, and weight gain contribute to it [18]. In adults the course is not always static, with slow worsening possibly accelerated by ageing [1].

4. Clinical Recognition and Red Flags

4.1. Prenatal period

Prenatal suspicion arises with polyhydramnios, reduced fetal movement, breech presentation, clubfoot, multiple contractures, pterygia, or hydrops [1, 2]. In cohorts of congenital myopathies with onset in utero, recessive TTN is the most frequent cause (about 10% of diagnoses), followed, among others, by recessive loss-of-function NEB and RYR1 variants and dominant de novo ACTA1 variants; about half remain without a molecular diagnosis, and maternal antibodies against the fetal acetylcholine receptor (FARAD, fetal acetylcholine receptor inactivation syndrome) mimic these forms without a genetic cause [1]. KLHL40, LMOD3, and NEB are the causes of NM most associated with fetal akinesia and arthrogryposis, and severe forms of NM are among the muscular causes of the fetal akinesia sequence [5, 88, 89]. In a Chinese family with successive pregnancies with clubfoot, polyhydramnios, and neonatal death, with normal karyotype and chromosomal microarray, fetal exome sequencing identified two compound heterozygous KLHL40 variants, and fetal tissue showed markedly reduced protein and nemaline bodies [90].

4.2. Newborn and infant

Congenital myopathies account for less than 10% of neonatal hypotonia [1]. In the hypotonic newborn, hypotonia of central origin relatively spares the face and limbs, whereas facial weakness, sucking and swallowing difficulty, and early respiratory distress point to a congenital myopathy, such as severe NM or myotubular myopathy, or to congenital myotonic dystrophy; in spinal muscular atrophy (SMA) type 1 the face is spared and early respiratory distress is rare [91]. The 2014 consensus associates severe congenital NM with marked lower facial weakness with an open mouth and "tented" upper lip, long face, dolichocephaly and high-arched palate, prominent bulbar weakness, and respiratory failure at birth; very severe neonatal hypotonia without antigravity movement suggests NM, most likely due to ACTA1, NEB, or KLHL40 [5]. Early fixed kyphoscoliosis, clubfoot, and distal arthrogryposis are compatible with NM, and severe arthrogryposis with fetal akinesia sequence suggests ACTA1, NEB, or KLHL40 [5].

Tremors of the limbs and chin, with mild rigidity of the hips and shoulders, are the neonatal clue to TNNT1-related NM and disappear in the first months [2, 17]. The overlap between severe ACTA1-related NM and perinatal asphyxia with hypoxic-ischaemic encephalopathy is a source of diagnostic delay: when profound weakness persists after electroencephalographic recovery, the picture is no longer consistent with the typical course of encephalopathy and warrants genetic testing (section 10) [74].

4.3. Child

In the older child, motor delay, waddling gait, difficulty running, Gowers sign, hypernasal voice, and recurrent respiratory infections are the most common presentations of typical NM [2]. Early scoliosis, foot drop, and pes cavus in forms with distal involvement (NEB, TPM3, TPM2) require exclusion of a neuropathy with electromyography (EMG) and nerve conduction studies [5]. Nocturnal hypoventilation in an ambulatory patient is a sign of NEB-, TPM3-, or ACTA1-related NM, with a differential diagnosis that includes juvenile Pompe disease, some congenital myasthenic syndromes, and collagen VI–related disorders [5]. Slowness of movement, inability to run or jump, and fatigue point to KBTBD13 [2, 84]. Joint hyperlaxity associated with weakness should prompt consideration of NEB [25]. Pseudohypertrophy suggests another diagnosis [5].

4.4. Adolescent and adult

In a study of 44 adults with congenital myopathy presented at the 250th workshop, 26 had onset in adulthood (median 47 years) and 23 had previously received other diagnoses, mainly non-neurological [6]. The 277th workshop reports that neonatal or childhood onset can be identified in more than 80% of adults, that mild forms may reach a clinician only in adolescence or adulthood, with long delays and misdiagnoses, and that NM is the single histopathological subtype most often encountered in adults [1]. In the Boston cohort, median age at diagnosis was 11 years in the mild or juvenile form and 56 years in the adult-onset form [4]. In an adult with no family history and subacute proximal or axial weakness after age 40 years, "dropped head," dysphagia, weight loss, or monoclonal gammopathy, the leading hypothesis is SLONM (section 6) [81, 92].

4.5. Gene-specific clinical clues

4.6. Red flags and emergency signs

Signs that argue against a diagnosis of congenital myopathy are hyperreflexia or other central nervous system abnormalities outside hypoxic encephalopathy, tongue fasciculations (SMA), dysmorphism beyond "myopathic facies," acidosis or elevated lactate, rapid deterioration in the first year, and extreme hyperlaxity (collagen VI) [5]. A CK more than 5 times the upper limit of normal suggests muscular dystrophy [1, 5].

The emergency signs are respiratory and bulbar. Recurrent respiratory infections, chronic atelectasis, failure to thrive, and aspiration pneumonia indicate hypoventilation or aspiration, which may be silent, and prompt a sleep study, swallowing assessment, and discussion of non-invasive ventilation [18]. Because nocturnal hypoventilation precedes daytime abnormalities and cannot be predicted from limb strength, the absence of complaints does not exclude it [18]. In Ryan's series, very severe neonatal respiratory disease and arthrogryposis multiplex congenita were associated with death in the first year in all patients but one, and independent walking before 18 months predicted survival [2, 24]. Any surgery or sedation, even minor, carries a risk of prolonged respiratory depression [18]. Prolonged immobilisation after illness or surgery can markedly worsen weakness [2].

4.7. Newborn screening and carrier screening

No population-based newborn screening for NM exists in the sources consulted. In Amish communities, the Clinic for Special Children laboratory tests high-risk newborns for the TNNT1 founder variant by high-resolution melt analysis in the first 1–2 days of life, allowing diagnosis before the onset of weakness [17]. Several recessive and X-linked congenital myopathies are included in reproductive carrier screening programmes [1], and the NEB exon 55 deletion is part of carrier screening targeted to the Ashkenazi Jewish population [11, 93].

5. Diagnosis: Criteria, Testing, and Algorithms

5.1. Diagnostic criteria

The diagnostic criteria for NM were established by the 1997 ENMC workshop (published in 1998) and rest on three elements: nemaline rods on muscle biopsy, a clinical myopathy, and molecular testing [4, 28]. The 2015 GeneReviews accepted the diagnosis in the presence of rods on biopsy and/or pathogenic variants in one of the known genes [2]. The 2014 consensus specifies that there is no minimum number of rods per biopsy, but that they must be the most prominent histological abnormality, because RYR1-related myopathies may have rods in a few fibres [5]. The 277th ENMC workshop defines congenital myopathy as the combination of compatible symptoms with a genetic diagnosis that explains the phenotype or with a compatible biopsy, allows the transitional term "probable CMYO" when there is neither a gene nor a biopsy, and reserves the term CMYO for cases with a confirmed genetic diagnosis or compatible histology [1]. In practice, a diagnosis of NM is established in a patient with a compatible phenotype in the presence of pathogenic or likely pathogenic variants in an NM gene with the appropriate inheritance, or in the presence of rods as the dominant finding on biopsy when genetic testing is inconclusive.

5.2. Non-genetic investigations

CK is usually normal or mildly elevated [2, 5]. In the Boston cohort it was elevated in only 16 of 116 cases with available information [4]; in the de Feraudy cohort the median was 44 U/L (interquartile range 27–176; reference 50–200 U/L), with mild elevation in two patients (473 and 600 U/L) [25]; and in the Swedish cohort, which includes all congenital myopathies, it was normal (below 240 IU/L) in 55 of 69 patients and elevated to between 264 and 978 IU/L in the remaining 14 [15]. In the first week of life CK may be non-specifically elevated, and values more than 5 times normal suggest dystrophy, with rare exceptions in RYR1 and TTN [1, 5].

EMG may be normal in young and mildly affected patients and is usually myopathic in older ones, with small, polyphasic motor unit potentials (MUPs) and an early, full recruitment pattern; in the distal forms and in the severely affected newborn it may appear neurogenic [1, 2, 5]. In the Boston cohort, EMG was abnormal in 78 of 132 patients [4]. Nerve conduction is normal, with low motor amplitudes when there is marked loss of muscle mass [2]. Repetitive stimulation at 3 Hz and single-fibre EMG serve to exclude a congenital myasthenic syndrome, although they are less well tolerated and harder to interpret in the newborn [1, 5]; in the Swedish cohort, none of the 33 patients studied with repetitive stimulation had a decrement [15].

Muscle imaging distinguishes a myopathic from a neuropathic pattern, guides the choice of muscle to biopsy, and helps prioritise genes [2, 5]. Ultrasound is performed at the bedside without sedation and is preferred in young children [1]. MRI should use T1 turbo spin echo or Dixon sequences, preferably whole-body, to include the pterygoids, tongue, neck, and trunk [1]. In NEB-related NM, mild forms show relative sparing of the thigh and early involvement of the lower leg, especially the tibialis anterior and, to a lesser extent, the soleus, and severe forms diffuse involvement of the lower limbs with sparing of the gastrocnemii; whole-body MRI shows selective involvement of the lateral pterygoid, sparing the other masticatory muscles and the tongue [5, 94]. In ACTA1-related NM, involvement is more proximal (sartorius, gluteus maximus) and the gastrocnemii are consistently spared; the tibialis anterior, tongue, and lateral pterygoid are involved with both genes [6]. TPM2 involves the temporalis and the distal leg (soleus, extensors, and flexors), without specific findings in the thigh, and TPM3-related cap myopathy has a pattern of fatty infiltration with masseter hypertrophy [3, 5]. The 277th workshop considers, however, that the NEB pattern is underexplored or heterogeneous, and that the well-characterised patterns are those of SELENON, dominant RYR1, DNM2, MYH7, and ACTA1 [1].

5.3. Muscle biopsy

Biopsy is no longer mandatory as a first-line test when genetic testing is rapid and unequivocal, but it remains essential for inconclusive results, novel genes, and the need for functional validation [1]. A clinically affected but not end-stage muscle should be biopsied, and sampling more than one muscle should be considered, because findings vary among muscle groups [2, 77]. Rods are rarely seen on haematoxylin and eosin and appear as red or purple structures against a blue-green background on modified Gomori trichrome, tending to cluster beneath the sarcolemma and around nuclei; in very small fibres they are hard to see, and electron microscopy is useful [2, 5]. On electron microscopy, nemaline bodies are electron-dense structures 1 to 7 µm long and 0.3 to 2 µm wide, in continuity with the Z-discs and with a lattice ultrastructure similar to theirs; areas of total myofibrillar disorganisation coexist with relatively normal sarcomeres [2]. Immunohistochemistry for α-actinin labels the rods [5]. Common accompanying findings are type 1 fibre predominance or uniformity, type 1 fibre hypotrophy, and very small fibres (less than about 5 µm) expressing fetal myosin [5]; predominance and atrophy increase with age [2].

According to Sewry (2019), rod shape is distinctive enough to suggest the gene only for KLHL40 and LMOD3, and even for LMOD3 the pattern is characteristic without being specific [3, 43, 87].

Biopsy has known limitations. The proportion of fibres with rods varies among patients and among muscles, and the first biopsy may not show them because of sampling error [2]. Histology may differ between samples from the same patient and change with age [1, 4]. In the de Feraudy cohort, 28 of 40 biopsies showed isolated rods and 11 a core-rod pattern; in one severely affected newborn rods were seen only on electron microscopy, in one patient with adult onset only the third biopsy, at age 40 years, showed them, and in one intermediate form neither rods nor cores were found [25]. Neonatal biopsies are technically difficult because of small fibre size and overlap with congenital dystrophy [1]. Rods are not pathognomonic and occur in mitochondrial myopathies, dermatomyositis, myotonic dystrophy type 1, human immunodeficiency virus (HIV)–associated myopathy, Hodgkin disease, and normal extraocular muscle [2].

5.4. Genetic strategy

The 277th ENMC workshop recommends a "genetics-first" approach: massively parallel sequencing panels, exome, or genome as first line in congenital hypotonia or suspected congenital myopathy, often before biopsy and after exclusion of appropriate alternatives [1]. In the critically ill newborn, in atypical presentations, and in neonatal hypotonia, rapid exome or genome sequencing is the first option, with a yield of 20–60% in neonatal intensive care units (NICUs), where genetic diseases account for 10–20% of admissions; rapid microarray and targeted tests for Prader-Willi syndrome, myotonic dystrophy type 1 (DM1), and SMA, which sequencing does not detect, should also be considered [1]. In older children and adults, multigene panels are appropriate [1]. If the result is inconclusive, biopsy, muscle imaging, long-read sequencing, RNA sequencing of muscle or fibroblasts, and other omics follow [1].

The 2014 consensus already provided for direct genetic testing, without biopsy, in characteristic phenotypes and in the very severely affected newborn for whom withdrawal of care is being considered, with ACTA1 and KLHL40 as priority genes for severe congenital NM; in this setting it recommends collecting, immediately after death, biopsies of more than one muscle (quadriceps and deltoid), deoxyribonucleic acid (DNA), and a fibroblast line [5]. Rapid molecular diagnosis in the critically ill newborn has a high yield, and parents value early prognostic information [4].

Yield depends on patient selection. At the UK reference service of the Dubowitz Neuromuscular Centre, 2352 panel analyses in 1927 patients with congenital myopathy or dystrophy (2014–2023) yielded a diagnosis in 28.7%, with a maximum of 47.2% in 2016 and a minimum of 11.4% in 2021; the mean yield fell from 29.41% in 2014–2019 to 15.97% in 2020–2023, after referral criteria were broadened, whereas the same group's earlier series, with phenotype-directed Sanger sequencing, had achieved 63% in the congenital myopathies [45]. In the Beggs laboratory, 70% of 918 families with congenital myopathy had a genetic diagnosis, and 84% of those solved had variants in 9 genes (ACTA1, DNM2, MTM1, NEB, RYR1, SELENON, TPM2, TPM3, TTN) [1]. The UK congenital myopathy panel grew from 22 genes in 2014 to 61 in 2023 and includes ACTA1, CFL2, KBTBD13, KLHL40, KLHL41, LMOD3, MYPN, NEB, TNNT1, TNNT3, TPM2, and TPM3 [45].

5.5. Test limitations

A negative result does not exclude NM. NEB is the main technical obstacle: its 183 exons, the triplicate region poorly aligned by short reads, and structural variants (intragenic deletions and duplications and copy-number changes of the triplicate region), which make up a substantial minority of pathogenic alleles, require targeted CGH array, droplet digital PCR, or genome sequencing, not just exome [3, 7]. In the UK series, 79 of the 502 undiagnosed patients with at least one variant had a single heterozygous pathogenic variant in a recessive gene, NEB being the second most frequent gene (17.7%), and NEB was also the gene with the second-highest number of variants of uncertain significance (VUS, 14.2%) [45]. In the Brazilian cohort, 20% of patients had only one identified variant and a "probable" diagnosis confirmed by biopsy; the authors presume an undetected second variant (CNV, gain or loss in the triplicate region, intronic variant) [48]. In an Australian panel with mean coverage above 300-fold and CNV detection down to the single-exon level, 10% of solved cases involved CNVs, including homozygous SMN1 deletions and DMPK expansions in patients referred as having myopathy [1]. In the de Feraudy cohort, five non-canonical splice variants remained VUS for lack of muscle RNA, and no large dominant deletions were detected, which may reflect the limitations of routine methods [25].

Other genes have their own pitfalls. TPM3 has numerous pseudogenes that hamper alignment [3]. In KBTBD13 the mechanism is missense, and a whole-gene deletion is not equivalent to the known causal variants [3]. In ACTA1 and in large NEB deletions there is mosaicism in the parent or in the patient, which low-depth sequencing may not detect [5, 7]. Synonymous or deep intronic variants that alter splicing can be confirmed only with RNA: in the de Feraudy cohort, the synonymous NM_001164508.2:c.20928G>T p.(Gly6976=) (identical in NM_001271208.2) caused partial, out-of-frame skipping of exon 139 [25]. Variant databases have a European ancestry bias (86% of published genetic data), which hampers interpretation in other populations [1]. A biopsy without rods does not exclude NEB, and the proportion of variants detectable by a method is not equivalent to the clinical sensitivity of the test [3, 25].

5.6. Protein and functional tests

When genetic testing is inconclusive, protein analysis in muscle guides or confirms the gene. Western blot or immunostaining for slow troponin T is the non-molecular test most specific for TNNT1 [3, 17]. The KLHL40 protein is absent from patient muscle, including in mildly affected patients [3, 86]. Immunolabeling of myopalladin at the Z-disc (MYPN) and demonstration of absence of the skeletal β-tropomyosin isoform in recessive TPM2 forms are specialised-centre assays [3]. Nebulin immunohistochemistry is normal in most patients with NEB variants and shows abnormalities that guide testing in only a small proportion [2]. Muscle RNA transcript analysis confirms the effect of splice variants [25, 48].

5.7. Reporting and variant interpretation

The report should state the gene, the reference transcript with version (MANE Select, e.g., NM_001164508.2 for NEB and NM_003283.6 for TNNT1), the HGVS notation at the c. and p. levels, the genomic coordinate with the genome build, zygosity, the phase of the variants when known, and the class according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) [4, 23]. In a disease with so many alternative transcripts, the same variant may carry different numbers, as with the TNNT1 founder variant (c.538G>T on NM_003283.6, c.505G>T on NM_001126133.3), and conversion between transcripts should be done in VariantValidator or ClinVar, not by hand [23, 57]. The research rule used in the Boston cohort, which classified as NEB-related NM patients with a clinicopathological diagnosis and a single pathogenic or likely pathogenic NEB variant with no other cause, is not a clinical diagnostic criterion [4].

VUS are frequent in large genes (NEB, RYR1). Family studies help classify them: a healthy sibling with the same pair of alleles excludes them as the cause, and a dominant variant present in healthy relatives is unlikely to be causal or has reduced penetrance [5]. In the de Feraudy cohort, a missense VUS in trans with different frameshift variants in a symptomatic mother and her daughter constituted pseudodominant inheritance of a recessive disease [25]. Periodic reanalysis of sequencing data and multidisciplinary discussion of VUS are part of recommended practice [45].

5.8. Diagnostic algorithm

5.9. Prenatal diagnosis and molecular autopsy

Once the familial variants have been identified, prenatal diagnosis and preimplantation genetic testing are possible [2]. Without a prior diagnosis, prenatal ultrasound detection of akinesia and fetal exome sequencing increase the diagnostic yield, as in the KLHL40 case in which the fetal exome and fetal tissue histology confirmed NEM8 after several affected pregnancies [90]. In fetus P34 of the de Feraudy cohort, termination at 17 weeks for akinesia, arthrogryposis, and hydrops allowed identification of two truncating NEB variants in constitutive exons [25]. When a severely affected newborn dies without a diagnosis, collection of DNA, muscle tissue from more than one site, and fibroblasts preserves the possibility of a diagnosis for the family [5]. Storage of DNA from affected individuals is recommended [2].

6. Differential Diagnosis

SLONM is the main differential diagnosis in adults, and its treatment differs radically. The 2014 consensus excludes it from the congenital myopathies, and Hildebrandt considers non-familial adult-onset NM a distinct entity, with a pathophysiology that appears to involve immune or inflammatory dysregulation and that may not be monogenic [4, 5]. The 294th ENMC workshop (March 2026) defined red flags for paraproteinaemic myopathies (onset at age 40 years or older, subacute proximal and/or axial weakness, normal or mildly elevated CK, dysphagia, weight loss, systemic manifestations and poor response to conventional immunotherapy), proposed replacing the designation "SLONM-MGUS" with "SLONM-MP" (paraproteinaemic myopathy) and recommended routine Congo red staining of biopsies from patients with monoclonal protein and plasma cell–directed therapy [92]. In the review of 144 patients, those with monoclonal gammopathy had more respiratory involvement, more prominent rods, less complete recovery and more loss of ambulation, and the best outcomes were achieved with two or more regimens including intravenous immunoglobulin and/or autologous stem cell transplantation [95]. Without treatment, the 2012 consensus describes death from respiratory disease within 1 to 5 years [18].

In the newborn, the central differential diagnosis is with congenital DM1, myotubular myopathy, congenital myasthenic syndromes, SMA, and hypoxic-ischaemic encephalopathy [5, 91]. DM1 and SMA require dedicated tests, because the DMPK expansion and the SMN1 deletion escape routine sequencing [1]. The distinction from myasthenic syndromes has therapeutic consequences and can be misleading: the patient with KLHL40-related NM treated with pyridostigmine was initially classified as having DOK7 congenital myasthenic syndrome, even though repetitive stimulation at 3 and 50 Hz showed no decrement, and only review of the biopsy with Gomori trichrome and electron microscopy revealed rods [73]. Hypoxic-ischaemic encephalopathy may coexist with severe NM when respiratory failure at birth causes asphyxia, which delays suspicion of myopathy (section 10) [18, 74].

In children and adults, RYR1-related core-rod myopathy is distinguished by well-demarcated cores in type 1 fibres, whereas the cores in ACTA1-related NM predominate in type 2 fibres and those in KBTBD13 are poorly structured [2]. Sparse rods in a biopsy with cores point to RYR1, and the distinction has anaesthetic implications [5, 18]. Distal NEB-related NM and dominant TPM3-related NM with foot drop mimic a neuropathy, and EMG and nerve conduction studies settle the question [5]. The presence of rods is not enough for a diagnosis of primary NM: there must be an appropriate clinical picture, because rod formation is a common muscle response [2].

7. Treatment and Risk Management Strategies

7.1. Principles and reference documents

There is no disease-modifying therapy and no drug approved by the Food and Drug Administration (FDA) or the European Medicines Agency (EMA) for any form of NM: the EMA medicines list contains no product with an indication in nemaline myopathy, and ClinicalTrials.gov had, as of 17 October 2026, no ongoing NM-specific drug or gene therapy trial [3, 98, 99]. Management is supportive, multidisciplinary, and anticipatory, and follows the 2012 international consensus statement on the standard of care for congenital myopathies (Wang and colleagues), developed by a committee of 59 members from 10 disciplines, with two online surveys and an in-person meeting at Stanford in May 2010; the recommendations are expert consensus, without formal grading of evidence, and distinguish "highly recommended" from "recommended" [18]. A multidisciplinary approach improves quality of life and may influence survival [2]. The 277th ENMC workshop announced a revision of these standards of care by working groups, with no document published to date [1]. The intensity of intervention is defined by the individual course and not only by the gene name [3].

The 2012 congenital myopathy care manual predates the discovery of several NM genes and the recognition of the cardiac risk of KBTBD13; the general recommendations apply to all genes, and genotype-specific considerations (subsection 7.8) come from later studies [18, 75]. The general congenital myopathy care plan applies to NM.

7.2. Communicating the diagnosis and follow-up

The diagnostic disclosure meeting should take place as soon as there is a clinical diagnosis, even without a genetic diagnosis, and should address the diagnosis, prognosis, recurrence risk, treatment plan, and available support, including patient organisations and research [18]. Recommended follow-up is every 3–4 months for infants younger than 12 months and every 6–12 months for older children, ideally in a multidisciplinary clinic; at each visit, breathing, speech, swallowing, cardiac signs, scoliosis, contractures, and growth are assessed [18]. Each visit reinforces healthy weight, exercise, nutrition with vitamin D supplementation, immunisations, preventive motor and respiratory physical therapy, and a reminder of anaesthetic risk [18]. Without a specific diagnosis, the workup should be reviewed at each visit, considering repeat biopsy and muscle imaging [18]. After diagnosis, initial evaluations include pulmonary function tests and screening for nocturnal hypoxia, assessment of feeding and growth, examination for contractures and scoliosis, physical therapy, occupational therapy, speech therapy, orthodontics if there are palatal anomalies, screening for cardiomyopathy, and a genetics consultation [2].

Myalgia and fatigue are managed with activity modification, mobility aids (walker, scooter, wheelchair), and rehabilitation; the consensus mentions non-steroidal anti-inflammatory drugs, gabapentin, and massage for pain, without doses [18]. By extrapolation from adult peripheral neuropathic pain, an indication in the Portuguese summary of product characteristics (SmPC) for Neurontin®, gabapentin is started at 300 mg on day 1, 300 mg twice on day 2, and 300 mg three times on day 3 (or 900 mg/day in three doses), with increases of 300 mg/day every 2–3 days up to a maximum of 3600 mg/day; the US label, for adult postherpetic neuralgia, titrates up to 1800 mg/day [100, 101]. Below age 18 years, gabapentin is authorised only for epilepsy (in the SmPC, 25–35 mg/kg/day in three doses from age 6 years), so its use for pain in NM is off-label at any age [100]. In NEM6, fatigue and quality of life should be asked about directly, and falls warrant risk assessment, balance training, and home safety measures [3, 84].

7.3. Respiratory care

Surveillance. All patients able to cooperate should undergo spirometry with flow–volume loop and FVC, maximal inspiratory and expiratory pressures (MIP and MEP), peak cough flow, and spot oximetry, at least annually if the risk is low and more often, e.g., every 6 months, if the risk is high or there are recurrent respiratory symptoms [18]. Vital capacity should also be measured supine: a fall of more than 20% relative to the sitting position suggests significant diaphragmatic weakness and risk of nocturnal hypoventilation, with or without daytime symptoms [18]. Because height does not reflect lung growth when there is scoliosis, arm span, ulnar length, or knee–hip distance is used, and patients are compared with themselves over time [18]. Pressure tests and spirometry are unreliable below age 6 years; sniff nasal inspiratory pressure (SNIP) is useful from age 2 years, and diaphragm ultrasound assesses diaphragm thickness, thickening, and excursion [6].

Source: Table 2 of the 2012 consensus [18].

Sleep-disordered breathing (obstructive apnoea, hypoventilation, hypoxaemia, hypercapnia) precedes daytime abnormalities and cannot be predicted from limb strength, so it should be assessed early and routinely; attended polysomnography is the most complete test, and the acceptable minimum is overnight oximetry with capnography or, if unavailable, venous blood gas with bicarbonate [18]. In TNNT1-related NM, all patients are at high risk and should be assessed every 6 months [3].

Airway clearance and prevention. Highly recommended for all are pneumococcal and influenza vaccines in addition to the routine schedule, mechanical or manual assisted cough (mechanical insufflation-exsufflation, air stacking, manual assisted cough) when cough is ineffective, taught during a stable phase, and scheduled secretion mobilisation once or twice daily in patients with severe weakness, even when stable (percussion, high-frequency chest wall oscillation, intrapulmonary percussive ventilation, as tolerated) [18]. Preventive chest expansion may be beneficial in early respiratory weakness, with limited evidence; bronchodilators and mucolytics are not routinely recommended [18].

Ventilation. Bilevel non-invasive ventilation with a backup rate is recommended for hypoventilation; continuous positive airway pressure (CPAP) is not recommended, because it does not treat hypoventilation [18]. The indications are chronic respiratory failure, sleep-disordered breathing documented on polysomnography, recurrent respiratory infections, and chronic atelectasis; use is nocturnal and extends to daytime if there is acute or progressive worsening [18]. In TNNT1-related NM, chest wall rigidity requires unusually high pressures from early on, and escalation has limited benefit, which should be anticipated [3, 82]. Intubation is indicated for hypoxaemia or hypercapnia refractory to non-invasive ventilation [18]. Tracheostomy is considered when non-invasive ventilation is needed 24 hours a day or there are repeated extubation failures; the decision is made with the family and is not an emergency [18]. Chronic 24-hour invasive ventilation outside the neonatal period is rare in the congenital myopathies, except in myotubular myopathy and severe congenital NM, and in this context palliative care should be discussed [18].

Acute respiratory infection. In acute infection, chest radiography, oximetry, venous or capillary blood gas, spirometry, and peak cough flow are performed when possible; assisted cough, secretion mobilisation, and non-invasive ventilation are intensified, including during the day, and the consensus notes that antibiotics are often used early [18]. In acute respiratory failure, immediate support is provided with non-invasive ventilation and intensive airway clearance when possible, and invasive ventilation if these fail, extubating to non-invasive ventilation as soon as possible [18]. Aggressive treatment of lower respiratory tract infections is among the measures that influence survival [2].

Inspiratory muscle training. In the Dutch pilot study, 9 patients with NM and MIP below 70% of predicted (7 women and 2 men, 36.6 ± 20.5 years) completed 8 weeks of sham training followed by 8 weeks of active training with a threshold pressure device, two 15-minute sessions a day, 5 days a week, with a load increasing from 15% to 30% of MIP [19]. MIP rose from 43 ± 15.9 to 47 ± 16.6 cmH2O after active training (difference of 4 ± 3.4 cmH2O, 11.4 ± 9.8%; p = 0.045 on analysis of variance), with no improvement in MEP, SNIP, FVC, peak cough flow, or diaphragm thickness; adherence was 96.6%, and there were no serious adverse events [19]. The authors consider the training feasible, but the clinical relevance of a 4-cmH2O gain is uncertain, because the minimal clinically important difference is unknown; patients with MIP too low for the minimum load of the device (9 cmH2O) could not train, and the design was not randomised [19]. Inspiratory muscle training is not part of the guidelines, and mechanical ventilatory support and assisted cough techniques remain the mainstay of treatment [19].

7.4. Feeding, swallowing, speech, and oral health

Weight and height are measured at every visit, at least every 3 months in infants, with surrogate measures (arm span, ulnar length) when necessary; dietary assessment is done at every visit, and calcium and vitamin D annually [18]. Swallowing is assessed clinically by a therapist, with videofluoroscopy when aspiration is suspected, which may be silent, without cough; chest radiography helps detect chronic aspiration [18]. In infants, positioning, thickening, caloric enrichment, adapted nipples, and nipple shields are used; a nasogastric tube is a short-term solution, and gastrostomy should be strongly considered in the newborn with very severe myopathy, with a nasojejunal or gastrojejunal tube if there is reflux, while maintaining oral stimulation [18]. In older children, gastrostomy is indicated when caloric needs are not met, while avoiding excess weight [18]. In the Boston cohort, gastrostomy was removed in 21 of the 105 patients who had one, and neither the signs at birth nor the gene predicted this removal; the authors acknowledge that some families keep it for periods of intercurrent illness [4].

Gastroesophageal reflux disease is treated with proton pump inhibitors, H2 receptor antagonists, or antacids, thickening, and positioning, with antireflux surgery in selected cases; constipation and slow gastric emptying are treated with fluids, mobilisation, and laxatives, and some centres use erythromycin as a prokinetic [18]. There is no consensus on sialorrhea: the 2012 consensus does not recommend botulinum toxin or salivary gland ligation, accepts systemic anticholinergics with caution because of secretion thickening and constipation, and mentions a small study in which L-tyrosine reduced sialorrhea in children with NM [18, 102]. Speech therapy includes augmentative communication, a speech prosthesis for hypernasality, pharyngoplasty in selected cases, and a palatal lift [18]. Oral health requires a paediatric dentist by the first year, toothbrushing twice a day, review every 6 months, and an orthodontist at 6–8 years; extensive orthodontics should be avoided when the face is very weak, and surgery for severe malocclusion should not be considered because of the anaesthetic risk [18].

7.5. Orthopaedics, rehabilitation, exercise, and bone health

Exercise and mobility. Regular symmetric resistance exercise, ideally concentric, at least 2–3 times a week, has no evidence of harm; exercise-induced fatigue and pain and high-impact sports should be avoided [18]. In a mouse model, resistance exercise counteracted worsening after immobilisation, and the GeneReviews authors observed clinical improvement with regular low-impact exercise (cycling, swimming) in the typical form, without controlled data in humans [2]. Standing is strongly recommended, even in very weak children (standing frames, parapodium, orthoses), with care to avoid metaphyseal fractures from overtightened straps [18]. A power wheelchair can be used from age 2 years when appropriate, with seating systems, a seat belt, and adapted driving controls [18].

Contractures. Passive and active-assisted stretching, static and progressive splints, serial casting, and spinal bracing are used [18]. Botulinum toxin is contraindicated in the skeletal muscle of children with primary muscle disease, for any indication, because it causes disuse atrophy for 3–6 months and may delay rehabilitation or render the patient non-ambulatory [3, 18]. Knee contractures greater than 15–25° that prevent standing or walking warrant consideration of medial hamstring lengthening, and neonatal clubfoot is treated as in healthy children (serial casting, percutaneous tenotomy, night splints) [18]. After fractures or tendon surgery, ambulation should be maintained or resumed within 1–2 days, with internal fixation, lightweight casts, or hinged orthoses, because prolonged immobilisation markedly worsens weakness [2, 18].

Scoliosis. The spine is examined at every visit [18]. In non-ambulatory patients, sitting anteroposterior and lateral radiographs are obtained at the first clinical sign and every 6 months if the curve is worsening; a soft orthosis for sitting balance at 18–24 months is temporary; bracing for curves of 20–40° in the sitting position is not needed when lying down or at night; for progressive curves greater than 50°, surgery is considered, with growing rods in some children younger than 10 years and definitive fusion from the upper thoracic spine to the pelvis near skeletal maturity, often at 11–13 years in severe neuromuscular diseases; delaying surgery for rigid curves compromises correction [18]. In ambulatory patients, a standing radiograph is obtained at clinical diagnosis and annually; bracing is used for curves of 20–40° if it does not interfere with walking, growing rods before age 10 years for curves greater than 40°, and posterior fusion from age 12 years for curves greater than 50°, without fusion to the pelvis, to preserve ambulation [18]. Neuromuscular scoliosis may worsen after maturity [18]. In Ryan's series, 4 of the 5 patients with unexpected postoperative respiratory failure had undergone scoliosis surgery (subsection 7.7) [2].

Hip and upper limbs. Hip ultrasound is performed in the early postnatal period; dislocation is treated with a Pavlik harness, with immobilisation limited to 10–12 weeks, and persistent dislocation after age 1 year with open surgery depending on severity [18]. In non-ambulatory patients there is no consensus on keeping painless hips reduced, and in ambulatory patients surgery is rarely indicated because of the risk of losing ambulation [18]. In the upper limbs, wrist, hand, and elbow splints are used, and elbow surgery is rarely indicated [18].

Bone health. The consensus recommends vitamin D 400 IU daily in all children, annual measurement of 25-hydroxyvitamin D, calcium at the recommended daily allowance, Tanner staging from age 11 years, and recording of fractures at every visit [18]. Intensive evaluation (dual-energy X-ray absorptiometry or paediatric quantitative computed tomography and lateral thoracolumbar radiography) is indicated for a low-impact pathological long-bone or vertebral fracture, two or more traumatic fractures, bone pain, severe osteopenia, or instrumentation failure, with exclusion of secondary causes (pubertal delay, coeliac disease, thyroid disease); bisphosphonates are considered after dental assessment and by an experienced physician [18].

7.6. Cardiac surveillance and treatment

Primary cardiomyopathy is rare in the congenital myopathies; it has been described with ACTA1, DNM2, and TPM2, and in NM there are also reports of transient heart failure and electrocardiographic abnormalities of uncertain significance [18]. The 2012 consensus proposes evaluation by a cardiologist about every 2 years in asymptomatic patients, particularly while the genetic diagnosis is uncertain, and more often if there are symptoms or abnormalities on the electrocardiogram (ECG) or echocardiogram [18]. The most frequent cardiac risk across NM is cor pulmonale secondary to chronic respiratory failure and chest deformity [18, 79]. In KBTBD13-related NM, the authors of the study of 65 carriers recommend referring patients to cardiology, with ECG and echocardiography, and Holter monitoring when indicated, and propose including the gene in cardiomyopathy panels; this gene–cardiac disease relationship comes from a single group and has not been curated by ClinGen or the Gene Curation Coalition (GenCC) [3, 75]. For MYO18B, baseline and periodic cardiac evaluation is recommended [3, 40]. For TNNT1, primary cardiomyopathy has not been described [3, 85]. Heart failure and arrhythmias are treated according to general cardiology guidelines, which the sources consulted do not adapt to NM.

7.7. Anaesthesia, malignant hyperthermia, and dantrolene

Anaesthesia is generally well tolerated: in Ryan's series, none of the 130 patients who underwent at least one surgery had malignant hyperthermia, but 5 had unexpected postoperative respiratory failure (4 after scoliosis correction and 1 after fundoplication), 3 required prolonged ventilation, and 1 died [2, 24]. Preoperative assessment of pulmonary function is essential to choose the timing of surgery and reduce anaesthetic risk [2]. There is a risk of prolonged respiratory depression even after minor procedures; non-invasive ventilation serves as a postoperative bridge, and in spinal fusion it should be started before surgery or the patient extubated directly onto it [18]. In MYO18B-related NM, cervical instability should be assessed before neck manipulation or intubation [3].

NM is not definitively associated with malignant hyperthermia, although bradycardia and mild hyperthermia during cardiac surgery have been reported; susceptibility is greatest in dominant or de novo RYR1-related myopathies, King-Denborough syndrome and Native American myopathy, and there is little evidence of susceptibility in the other genetically defined congenital myopathies [2, 18]. The 2012 consensus recommends regarding any neuromuscular patient as potentially susceptible, particularly if the genotype is unknown or the histology non-specific: a "clean" technique with a purged machine, avoidance of suxamethonium and inhalational anaesthetics (except nitrous oxide) when possible, extensive cardiorespiratory monitoring with capnography, intensive care availability and dantrolene on the anaesthetic trolley [18]. GeneReviews advised avoiding neuromuscular blocking agents when possible, because of the link between core-rod forms and RYR1 [2]. Wang 2012 states that "prophylactic dantrolene is not recommended prior to anesthesia, even in cases where malignant hyperthermia susceptibility has been established" [18].

Dantrolene sodium is a direct-acting muscle relaxant that treats skeletal muscle hypermetabolism in a malignant hyperthermia crisis; it has no indication in NM outside this setting [103, 104]. Doses differ between the US labels and the European summary of product characteristics (SmPC):

Sources: [103–105]. Revonto® and Hikma's intravenous dantrolene are also authorised in the USA [103]. The 2012 consensus dose ("1 mg/kg to a maximum level of 10 mg/kg") corresponds to the US labels [18, 103]. The practical consequence is twofold: the treatment dose depends on the formulation available, because the European SmPC starts at 2.5 mg/kg; and the prophylaxis provided for in the US labels for susceptible patients is not recommended by the congenital myopathy consensus and is not included in the Agilus® SmPC [18, 103, 104]. The sources consulted do not include the INFARMED or Brazilian Health Regulatory Agency (ANVISA) registers, so the availability of the formulations in Portugal and Brazil was not documented.

7.8. Genotype-specific management considerations

7.9. Off-label drugs and supplements

None of the following drugs has an approved indication in NM; their use is off-label and rests on case series, isolated reports, or trial protocols without published results [3]. The doses given are those in the sources and do not constitute dosing recommendations.

The authors of the pyridostigmine report emphasise that myasthenic features and response to acetylcholinesterase inhibitors had not been described in NM or in KLHL40-related myopathy, and merely ask whether a therapeutic trial should be considered in the congenital myopathies [73]. The authors of the second report used a fivefold lower dose and observed no benefit [108]. Wang 2012 does not mention pyridostigmine, and no guideline gives a dose for NM [18]. In the mouse model of severe ACTA1-related NM, L-tyrosine improved strength [2].

None of the labels gives a weight-based dose for pyridostigmine. For myasthenia gravis, the closest approved indication, the Portuguese SmPC for Mestinon® (60-mg tablets) specifies 1–3 tablets 2–4 times a day in adults and, in children, an initial dose of 30 mg below age 6 years and 60 mg from 6 to 12 years, with increments of 15–30 mg per day until maximal improvement and usual total requirements of 30–360 mg/day; the US label gives an average of 600 mg/day (60–1500 mg/day) and states that paediatric safety and efficacy have not been established [110, 111]. The 90 mg/day in report 1 at 12 months falls within the SmPC daily total but above the initial paediatric dose, and the 360 mg/day at 9 years coincides with the upper limit of usual requirements; both are off-label by indication [73, 110]. Oral ephedrine has no SmPC in Portugal, where only injectable forms exist; in the USA, over-the-counter tablets (Primatene®, 12.5 mg of ephedrine hydrochloride) are intended for mild intermittent asthma from age 12 years, 1–2 tablets every 4 hours up to 150 mg in 24 hours, and should not be used below age 12 years [112]. The ephedrine in report 1, started at 5.5 years, is off-label by age and indication. The oral salbutamol in the COMPIS protocol (2 mg three times a day) is within the dosing of the US syrup label for ages 6–14 years (2 mg three or four times a day, up to 24 mg/day), although for a different indication (bronchospasm) [113].

7.10. Regulatory status

Sources: [3, 18, 98, 103–105].

7.11. Clinical trials and interventional studies

Also registered are a study of aerobic training in 16 patients with congenital myopathies (NCT02020187, Copenhagen, completed in 2014) and a study of whole-body electrical muscle stimulation in neuromuscular diseases (NCT07478172, recruiting), whose results are not part of the sources reviewed [99]. There is no randomised controlled trial with published results in any form of NM. The only interventional study with results is a non-randomised pilot study of 9 patients, with a physiological outcome and no functional measure; the salbutamol trial, completed in 2025, has not released results, and the L-tyrosine trial has not been updated since 2014 [19, 107, 109]. Absence of released results is not equivalent to a negative result, but it precludes any recommendation. Relative gains, such as the 11.4% in MIP, must be read alongside the absolute value (4 cmH2O) and the lack of a validated minimal clinically important difference [19]. No negative drug trial in patients has been published; the known negative results are preclinical (levosimendan in human fibres, the myostatin inhibitor ActRIIB-mFc in the nebulin-deficient mouse) [115, 116]. The barriers to translation identified are rarity, heterogeneity of genotypes, phenotypes, and trajectories, lack of longitudinal natural history data, and absence of validated outcomes and biomarkers [117].

Trial infrastructure is under construction. The 250th ENMC workshop (2019) proposed parallel natural history studies in Europe and in the USA and Canada, with a common dataset, inclusion of all genotypes, and age-specific outcomes every 6 months: CHOP-INTEND and HINE-2 from 0 to 2 years, MFM20 from 2 to 5 years, and MFM32, timed tests, grip and pinch strength (MyoGrip, MyoPinch), and sitting and supine respiratory tests from 5 to 18 years [6]. In the tables, CHOP-INTEND denotes the Children's Hospital of Philadelphia Infant Test of Neuromuscular Disorders, HINE-2 section 2 of the Hammersmith Infant Neurological Examination, AIMS the Alberta Infant Motor Scale, PDMS-3 the Peabody Developmental Motor Scales (third edition), 6MWT the 6-minute walk test, NSAA the North Star Ambulatory Assessment, and PUL the Performance of the Upper Limb. The registered studies are as follows:

7.12. Investigational therapies and developments in the past 12 months

A 2022 review identified at least eleven experimental approaches, directed at the causal gene, contraction mechanisms (calcium and myosin), myogenesis, or muscle function in general [117]. In 2019, the 250th workshop considered cell therapy unfeasible because of body-wide distribution, myostatin inhibition capable of increasing mass without correcting the defect, gene or protein replacement best suited to recessive loss-of-function forms, and genome editing an option for inactivating dominant alleles [6]. Literature and registry searches performed in October 2026 found no studies of base or prime editing, NEB exon skipping, or allele-specific silencing of ACTA1, and no clinical trials of gene therapy, troponin activators, or NAD+ precursors in NM; a negative search does not demonstrate the absence of unpublished programmes [99].

In the past 12 months (October 2025 to October 2026), developments were as follows, by level of evidence. In preclinical research, Kbtbd13 silencing, nicotinamide riboside in NEM6, and the role of Klhl41b were published, and the KLHL40 gene replacement preprint appeared [62, 66, 124, 125]. In human studies, the following were published: the cohort of 49 patients with NEB variants [25], population epidemiology from western Sweden [15], ten-year diagnostic yield in the United Kingdom [45], dominant distal myopathy due to NEB deletion in South Africa [46], prenatal diagnosis of NEM8 [90], an ACTA1 variant with dilated cardiomyopathy [51], an international survey of patients and caregivers [83], and a qualitative study with 17 Spanish parents [133]. Among consensus documents, the 277th ENMC workshop on nomenclature and diagnosis [1] and the 294th on paraproteinaemic myopathies and SLONM [92] were published. Among registries, the Spanish and Belgian natural history studies started and the READYCOM protocol was published [121–123]. There was no CHMP opinion, regulatory authorisation, sponsor announcement about a trial, or product withdrawal relating to NM in the sources consulted.

7.13. Agents and practices to avoid

7.14. Special populations, transition, and palliative care

Pregnancy and lactation. All pregnancies in women with congenital myopathy are high risk and should be followed by high-risk obstetrics together with pulmonology and cardiology, particularly in subtypes with diaphragmatic weakness [18]. Respiratory assessment before and during pregnancy follows the parameters in subsection 7.3. In the case reported by Pei, the symptoms of a woman with an ACTA1 variant began during pregnancy (section 10) [134]. The sources consulted do not address lactation or the safety of supportive medications during breastfeeding; the Dantrium® label places dantrolene in pregnancy category C [103].

Transition and adulthood. The consensus recommends a coordinated plan for transition from paediatric to adult care [18]. In adults, fatigue, pain, and sleep should be assessed with dedicated instruments, and patients with reduced respiratory function may be offered respiratory muscle training [6, 19]. The international survey showed mental health difficulties in all severity groups [83], and the Spanish qualitative study with 17 parents described unremitting caregiving demands, essential but unstable formal support, and reorganisation of family life [133].

Palliative care. In the severe neonatal forms (ACTA1, KLHL40, LMOD3, NEM5A), early and explicit discussion of prognosis, ventilatory support options, and palliative care is part of management, and for chronic 24-hour invasive ventilation the consensus calls for discussion of palliative care with the family [3, 18, 74]. A decision to withdraw life support, when made, should be accompanied by collection of samples for diagnosis (subsection 5.9) [5].

8. Reproductive Options and Family Counselling

8.1. Inheritance and recurrence risk by gene

8.2. Dominant inheritance, de novo cases, and mosaicism

In dominant NM, some probands have an affected parent and others a de novo variant [2]. In ACTA1, most cases are simplex (no other affected relative), and two families with parental mosaicism for dominant variants have been described [2, 77, 135]; in the Boston cohort, parental mosaicism was suspected in two further families, without confirmatory testing [4]. An apparently negative family history can be confirmed only after evaluating the parents, because the disease may not have been recognised, the parent may have died early, or the form may be late-onset; the parent in whom the variant arose may have somatic mosaicism and be mildly or minimally affected [2]. When a variant is identified in ACTA1, TPM3, TPM2, or KBTBD13, testing is offered to the parents, and parents should not undergo biopsy before testing for the familial variants has been exhausted, because its interpretation is difficult [2]. The risk to siblings is 50% if a parent is affected or carries the variant; if the parents are clinically normal and negative for the variant, the risk is low but not zero, because germline mosaicism has been reported in ACTA1 and is possible in other dominant genes [2]. The offspring of an individual with dominant disease have a 50% risk [2].

8.3. Recessive inheritance, consanguinity, and pseudodominance

In recessive NM, the parents are asymptomatic obligate carriers; each sibling has a 25% chance of being affected, a 50% chance of being an asymptomatic carrier, and a 25% chance of being neither affected nor a carrier, and an unaffected sibling has a 2/3 probability of being a carrier [2]. The children of an affected individual are obligate carriers, and the parents' siblings have a 50% probability of being carriers [2]. In Ryan's series of 143 patients from 110 families, inheritance was recessive in 29 patients from 15 families (20%), dominant in 41 patients from 22 families (29%), and undetermined in the remainder [2, 24]. In the Boston cohort, 19 patients were born to consanguineous unions, but information was missing for 141 [4]; in the de Feraudy cohort, 3 of the 6 consanguineous families had homozygous variants and 3 had two different variants [25].

Apparently vertical transmission does not exclude recessive inheritance: in one family in the de Feraudy cohort, an affected mother and daughter shared a NEB missense VUS and each had a different frameshift variant in trans, a case of pseudodominance [25]. Some healthy heterozygous parents have NM changes on biopsy, so biopsy of a single parent does not distinguish dominant from recessive inheritance; one parent with clinical disease and typical histology and the other normal suggests dominant inheritance, and two normal parents suggest a de novo variant, germline mosaicism, or recessive inheritance [2]. In apparently de novo cases, non-medical explanations (alternative paternity or maternity, adoption) may also be involved [2]. In the recessive form of TPM2, unlike the dominant rule for this gene, the recurrence risk is 25% [3].

8.4. Carriers and screening in founder populations

The NEB exon 55 deletion had a carrier frequency of 1 in 108 in a random sample of 4090 Ashkenazi Jews (Anderson 2004) and 1 in 168 (0.006) in carrier screening of individuals of 100% Ashkenazi ancestry in the New York metropolitan area (Scott 2010); the two estimates come from different samples and eras [34, 93]. The GeneReviews resource on Ashkenazi founder variants attributes more than 95% of pathogenic NEB variants in this population to this deletion [11]. Because the deletion has been found in families without known Ashkenazi ancestry with the same haplotype, ancestry should not preclude testing [47].

In Old Order Amish communities, the carrier frequency of the TNNT1 founder variant is 6.5% (1 in 16) according to Fox (2018), corresponding to an incidence of 1 in 960 births; Johnston (2000) estimated an incidence of about 1 in 500, and the GeneReviews resource on Amish founder variants gives a carrier frequency of 1 in 16 to 1 in 19 in Lancaster (Pennsylvania), calculated from exomes of 301 controls, with this variant accounting for about 100% of TNNT1 variants in this population [12, 16, 17]. The three figures use different designs and samples and apply only to these communities. Recessive congenital myopathies are part of several reproductive carrier screening programmes [1]. When one member of a couple is a carrier, testing of the partner should include methods capable of detecting NEB structural variants (subsection 5.5).

8.5. Prenatal and preimplantation diagnosis

Risk and reproductive options should be discussed before pregnancy, and genetic counselling should be offered to affected, carrier, or at-risk young adults [2]. Once the familial variants have been identified, prenatal diagnosis and preimplantation genetic testing are possible [2]. In the severe recessive forms due to KLHL40 and LMOD3, recurrence in successive pregnancies has been described, and in the southern Chinese series two pregnancies affected by KLHL40 variants were terminated [3, 64]. When the familial variant is not known, prenatal ultrasound may show polyhydramnios, reduced movement, clubfoot, or contractures, and fetal exome sequencing may establish the diagnosis [90]. Storage of DNA from affected individuals is recommended, particularly in the lethal forms [2, 5].

8.6. At-risk relatives and minors

Older and younger siblings of a proband should be evaluated for early surveillance: with a known variant, by molecular testing; without a known variant, by targeted clinical examination, with neurophysiological studies to decide on biopsy [2]. In recessive NM, testing of asymptomatic siblings is considered for early diagnosis, because it allows respiratory surveillance to begin before symptoms [2, 18]. In NEM6, cardiology assessment of carrier relatives is warranted by the risk of cardiomyopathy and arrhythmias [75]. In Amish communities, testing of high-risk newborns in the first 1–2 days of life allows diagnosis before the onset of weakness [17]. In NEM5A, counselling should be explicit about prognosis, because all patients in the cohort of 106 died by age 6 years [3, 17]. The sources consulted do not specifically address carrier testing in minors.

9. Special Considerations: Population Diversity and Epidemiology

9.1. Global frequency

NM accounts for about 17% of all congenital myopathies, according to a study cited by GeneReviews [2]. Michael and colleagues emphasise that, in a disease with significant mortality, point prevalence counts only survivors and underestimates occurrence, which is better captured by incidence or birth prevalence; they consider their own results an underestimate, because diagnostic methods changed over the period [15]. In the same Swedish cohort, TPM2 was found in 4 patients with variable histology, and NEB and TPM3 in one patient each [15]. The 250th workshop was unable to obtain prevalence or incidence data; because until 2019 the Dubowitz Neuromuscular Centre was the only UK centre testing NM genes, the experts considered the 116 UK patients a good estimate of the affected population, which is an opinion and not a population rate [6].

9.2. Distribution of genes by country

Source: Table 1 of the 250th ENMC workshop, compiled by Anna Sarkozy (Dubowitz Neuromuscular Centre) [6]. The United Kingdom also had 1 patient with two recessive MYO18B variants without pathological findings of NM; KLHL41 and MYPN had no patients. The numbers are counts of genetically diagnosed patients at each centre and reflect the composition of the centres, not the relative prevalence in the countries. The predominance of NEB in Finland (29 of 33) and of KBTBD13 in the Netherlands (38 of 68) reflects local research and founder effects [6].

9.3. Founder populations and recurrent variants

9.4. USA, Europe, Brazil, and Portugal

North American data come mainly from the Boston cohort (275 patients; European ancestry in 70%, Hispanic in 14%, Ashkenazi in 8%, and African American, Asian, and Native American in 5–6% each), the Amish communities, and Ashkenazi screening in New York [4, 17, 93]. The Boston cohort is a referral cohort and does not allow prevalence to be estimated [4]. In Europe, there are population data from western Sweden, diagnostic yield data from the United Kingdom, cohorts from France, the Netherlands, Finland, Germany, Italy, and Spain, and a Spanish qualitative study on home care [6, 15, 25, 45, 133]. In the United Kingdom, NEB accounted for 8.7% and ACTA1 for 6.7% of the 345 panel diagnoses of congenital myopathy between 2014 and 2023 [45].

Brazil has a cohort of 33 patients with NEB-related NM from the Hospital das Clínicas of the University of São Paulo Medical School (2011–2022), with 45 variants, 23 of them novel, the description of the "triple furrow" tongue atrophy sign, and two recurrent variants (c.5343+5G>A in 5 patients and c.19944G>A in 3); it is a single-centre cohort and not a population estimate [48]. Brazilian investigators took part in the 2012 standard-of-care consensus and the 250th ENMC workshop [6, 18], and the de Feraudy cohort included patients from Argentina, Chile, and Paraguay [25]. The sources consulted include no epidemiological data or cohorts from Portugal; their absence from these sources does not demonstrate the absence of published data. Genetic databases contain 86% data of European ancestry, which hampers variant classification in other populations [1].

9.5. Special populations

Adults with mild forms reach a diagnosis late, often after other non-neurological diagnoses [1, 6]. Women with NM who become pregnant should be followed as high-risk pregnancies [18]. Amish communities have cultural values that, in Fox's series, led families uniformly to decline hospitalisation, gastrostomy, and ventilation, which calls for tailored counselling and respect for end-of-life decisions [17]. In resource-limited settings, the 277th workshop describes local tests targeting founder variants, exemplified by RYR1 in South Africa [1].

9.6. European reference networks and patient organisations

The European Reference Network for rare neuromuscular diseases is ERN EURO-NMD (https://ern-euro-nmd.eu/), to which several authors of the Dutch inspiratory training study belong [19]. The European Neuromuscular Centre (ENMC, https://www.enmc.org/) organised the 1997, 2019, 2024, and 2026 workshops on NM, congenital myopathies, and SLONM, with lay summaries in several languages [1, 6, 92]. TREAT-NMD (https://treat-nmd.org/) cofunded the 2012 consensus and the 277th workshop [1, 18]. In France, the FILNEMUS network supported the de Feraudy cohort [25].

A Foundation Building Strength (https://www.buildingstrength.org/), in the USA, is an organisation dedicated exclusively to NM, founded in 2008, that funded the 2012 consensus and the inspiratory training study and collaborates in the NM-CTRN network; according to the organisation itself, it has funded 42 studies and has more than 375 people with NM enrolled in the CMDIR [18, 118, 138]. Muscular Dystrophy UK maintains a page on NM reviewed in February 2025, with the six-type classification and recommendations on respiratory surveillance, gastrostomy, and moderate aerobic exercise, and collaborates in the UK natural history study [120, 139]. Information and support are also provided by the Muscular Dystrophy Association (https://www.mda.org/disease/congenital-myopathies), Cure CMD (https://www.curecmd.org/), and the CMDIR registry (https://www.cmdir.org/) [2, 20]. The neuromuscular gene table (https://www.musclegenetable.fr/) and Orphanet (https://www.orpha.net/en/disease/detail/607) are resources for professionals [1, 9]. The sources consulted do not identify NM-specific patient organisations in Portugal or Brazil.

10. Illustrative Clinical Case

10.1. Severe congenital de novo ACTA1-related myopathy in a newborn with initially suspected perinatal asphyxia

Zhao, Deng, and Cai published in January 2026 the case of a Chinese newborn girl with neonatal-onset ACTA1-related myopathy [74]. The mother was 29 years old, G3P2, with a previous uterine scar, and the parents were not consanguineous; there was no family history of neuromuscular or metabolic disease, and a 7-year-old brother had normal motor development [74]. Prenatal ultrasound showed polyhydramnios, with 2500 mL of amniotic fluid, without other structural anomalies [74]. The infant was delivered by caesarean section at 39 weeks and 1 day, with Apgar scores of 3 at 1 minute and 5 at 5 minutes, and required immediate resuscitation [74].

Examination showed lethargy, expressionless facies, profound generalised hypotonia, absence of spontaneous movement (only minimal finger movements), respiratory failure requiring mechanical ventilation, bulbar dysfunction with weak sucking and swallowing and excessive oral secretions, absent Moro and grasp reflexes, and preserved eye movements; in the discussion, the authors mention arthrogryposis and hypertonic contractures against a background of global hypotonia [74]. Blood gas analysis revealed metabolic acidosis (pH 7.27; pCO2 46.2 mmHg; bicarbonate 20 mmol/L; base excess −5.3), and lactate was 2.7 mmol/L; CK peaked at 578 U/L and normalised by day 9 [74]. The ECG showed sinus tachycardia at 177 beats per minute with ventricular ectopy, and the echocardiogram a haemodynamically insignificant 1.2-mm ductus arteriosus, with an ejection fraction of 53% [74]. The electroencephalogram (EEG) on day 1 showed multifocal epileptiform discharges and subclinical seizures; EMG on day 10 showed reduced motor amplitudes (right median nerve 0.64 mV, left 1.94 mV) and fibrillations; brain MRI showed punctate periventricular lesions with T1 and T2 hyperintensity and restricted diffusion, without malformations [74].

Exome sequencing, requested on day 3, identified the heterozygous ACTA1 variant NM_001100.4:c.227G>A p.(Gly76Asp) (GRCh38 NC_000001.11:g.229432783C>T), absent in both parents by Sanger sequencing and therefore de novo, classified as pathogenic by criteria PS2, PM1, PM2, and PP3; the article does not state the transcript, and the HGVS notation was confirmed in VariantValidator [23, 74]. The diagnosis was confirmed on day 16 [74]. No muscle biopsy was performed, which the authors present as a limitation.

The infant received 7 days of conventional mechanical ventilation, followed by unsuccessful attempts to wean to non-invasive positive-pressure ventilation (NIPPV) [74]. As neuroprotective measures for suspected hypoxic-ischaemic encephalopathy, she received therapeutic hypothermia and vitamin B1 and B12 supplements, and swallowing rehabilitation was started, with minimal improvement [74]. The EEG improved progressively through day 7, but profound neuromuscular weakness persisted, which the authors considered inconsistent with the typical recovery from hypoxic-ischaemic encephalopathy [74]. Given the prognosis, the family decided to withdraw life support on day 18, and the infant died shortly afterwards [74].

The article's figures were reviewed. Figure 1 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12864113/figure/F1/) shows the newborn in supine and lateral positions, with a nasal non-invasive ventilation interface, monitoring sensors, and oximetry, upper limbs flexed against the trunk with closed fists, lower limbs abducted and externally rotated, and an open mouth with little expression. Figure 2 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12864113/figure/F2/) shows axial T1, T2, and diffusion brain MRI slices with the periventricular foci marked. Figure 3 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12864113/figure/F3/) shows Sanger electropherograms with a double G/A peak in the infant and a single G peak in the parents. Figure 4 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12864113/figure/F4/) is the timeline: conventional mechanical ventilation on day 1, NIPPV on day 7, genetic testing from day 3 to day 16, MRI on day 9, diagnosis on day 16, and withdrawal of treatment on day 18; the abbreviation "CMV" in the figure means conventional mechanical ventilation and not cytomegalovirus, although the publisher's alt text reads it as a diagnosis [74].

The case illustrates three points from the consensus statements. First, the severe ACTA1-related form overlaps with perinatal asphyxia: respiratory failure at birth can cause true hypoxic-ischaemic encephalopathy, as the EEG and MRI show, and the encephalopathy does not explain the persistent weakness [18, 74]. Second, the 2014 consensus already identified ACTA1 and KLHL40 as targets for direct genetic testing in the very severely affected newborn, and the 277th workshop recommends rapid exome or genome sequencing as first line in the NICU [1, 5]. Third, the severity warrants early palliative care consultation, as the authors propose [74]. Therapeutic hypothermia and vitamins were directed at the presumed encephalopathy and are not treatments for NM. The absence of biopsy or collection of muscle tissue or fibroblasts after the decision to withdraw support departs from the 2014 recommendation to preserve samples for diagnosis and research, although in this case the genetic diagnosis had already been established [5]. The family, with a de novo variant, has a low but not zero recurrence risk, because of the germline mosaicism described in ACTA1 (section 8) [2, 5].

10.2. Adult-onset ACTA1-related myopathy with first symptoms during pregnancy

Pei and colleagues described a 33-year-old Chinese woman whose symptoms began in 2018, during pregnancy, with a sensation of heaviness and swelling in the lower limbs and elevated CK and CK-MB, not investigated at the time; weakness worsened from 2021, with difficulty climbing stairs, squatting, and rising [134]. In 2023 gait was normal, upper limb strength normal, and lower limb strength grade 4 of 5 on the Medical Research Council scale, with mildly reduced tone, diminished patellar reflexes, and mild atrophy of the thighs and lower legs, without scoliosis and with normal sensation [134]. CK was 1011 U/L, nerve conduction was normal, EMG showed fibrillations, positive sharp waves, and myopathic MUPs, and thigh MRI showed oedema and fatty infiltration, mainly of the vastus medialis [134]. Genetic testing identified the heterozygous ACTA1 variant NM_001100.4:c.417G>C p.(Gln139His) in exon 3, classified as pathogenic by the laboratory, as well as two heterozygous VUS in MTM1 and TNNT1 [23, 134]. Biopsy was offered and declined by the patient because of its invasiveness, cost, and the genetic diagnosis already established; several relatives across four generations had muscle weakness, without genetic confirmation [134].

Figure 1 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC12599729/figure/F1/), which was reviewed, shows two coronal thigh MRI slices with heterogeneous hyperintensity on fat-suppressed T2 and bilateral heterogeneous intramuscular signal more evident in the lower half of the thighs; Figure 2 (https://pmc.ncbi.nlm.nih.gov/articles/PMC12599729/figure/F2/) is the five-generation pedigree with vertical transmission, including transmission from the patient to a son, consistent with dominant inheritance; the legend uses "myasthenic manifestations" to denote weakness, without neuromuscular junction data [134]. The patient received supportive care and remained stable on telephone follow-up [134].

The case serves to illustrate genetic diagnosis without biopsy and the possibility of onset or worsening during pregnancy, but it has limitations that set it apart from the usual NM profile: a CK more than 5 times normal and spontaneous activity on EMG are atypical and, according to the 2014 consensus, suggest other diagnoses, and there is no histology confirming rods [5, 134]. Segregation analysis in affected relatives, recommended for classifying dominant variants, was not performed [5]. For a future pregnancy, the 2012 consensus would recommend follow-up by high-risk obstetrics with pulmonology and cardiology [18].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis by clinical form

In Ryan's series, cumulative survival differed significantly among the severe, intermediate, and typical forms; respiratory failure caused the death of 17 of the 23 patients with the severe form, 8 of the 29 with the intermediate form, 4 of the 66 with the typical form, 1 of the 19 with childhood onset, and none of the 6 with adult onset [2, 24]. Hypotonia and severe weakness in infancy did not predict early mortality, unlike very severe neonatal respiratory disease and arthrogryposis, which were associated with death in the first year in all patients but one; independent walking before 18 months predicted survival, and most patients with the typical form eventually walked [2, 24]. An earlier review of 14 patients from London and 85 from the literature had identified neonatal hypotonia as the main prognostic sign, with a retrospective classification lacking detail [2, 140]. The 2012 consensus considers survival into adulthood likely in all congenital myopathies, except when there is severe respiratory insufficiency and profound weakness in infancy, and notes that patients with a severe neonatal presentation due to ACTA1 often die in the first year [18]. The mild non-progressive or slowly progressive form is compatible with a normal life expectancy [96].

In the Boston cohort, 61 of 263 live-born patients died, with a mean age at death of 6.1 years and a median of 0.2 years; median cohort survival was 66 years, with a bimodal curve, and of the 8 patients with information after age 65 years, 2 died at 66 years and 6 lived longer [4]. Invasive ventilation before age 1 year (51 patients, 31 deaths, median survival 0.75 years) and respiratory failure at birth (54 patients, median 0.68 years) separated an early-mortality group from patients without support before age 1 year (134 patients, 9 deaths) and those without respiratory failure at birth (209 patients, 26 deaths), whose medians were not reached [4]. Of the 153 patients who had walked, 25 lost ambulation, totally or partially, during follow-up, and the median time to loss could not be estimated [4].

Respiratory morbidity accumulates with age in the forms that survive. In the Brazilian NEB cohort, half the patients older than 3 years had started ventilatory support before age 20 years, and 85% required it by age 40 years; skeletal deformities affected all adults [48]. In the de Feraudy cohort, the median age at initiation of ventilation, estimated by Kaplan-Meier analysis, was 16 years [25]. Progressive respiratory failure may appear in adolescence or adulthood even without previous compromise [18].

11.2. Prognosis by gene

11.3. Early predictors

In the Boston cohort, respiratory distress at birth was associated with gastrostomy (OR 4.8; 95% CI 2.6–8.6), invasive ventilation in the first year (OR 2.8; 1.5–5.5), and invasive ventilation after the first year (OR 3.3; 1.8–6.1); hypotonia at birth was associated with gastrostomy (OR 2.2; 1.3–3.7) but not with invasive ventilation [4]. The authors made 18 comparisons with α = 0.05, which would be expected to yield about 0.9 significant results by chance, and 7 had CIs that excluded 1 [4]. The Amburgey score is useful in prospective follow-up, but it does not distinguish the severe from the intermediate form or the typical from the mild form and, because it depends on age at walking, does not predict outcome in the first years; none of the available classifications yielded reliable predictors of long-term outcome [4]. The number and distribution of rods have no prognostic value [3, 77].

11.4. Surveillance plan

11.5. Follow-up in adulthood

In adults the course is not always static, with slow worsening possibly accelerated by ageing [1]. Loss of ambulation in adulthood is possible, and contractures, scoliosis, fractures, and weight gain contribute to it [18]. Neuromuscular scoliosis may worsen after skeletal maturity [18]. Fatigue, pain, sleep, and mental health should be part of follow-up, because the international survey showed daily difficulties in all severity groups [6, 83]. Patients with respiratory weakness may be candidates for inspiratory muscle training, with functional benefit yet to be demonstrated [19]. Participation in registries and natural history studies has been recommended since 2012 and is now possible in several national and international studies (subsection 7.11) [6, 18].

12. Summary and Concluding Remarks

Nemaline myopathy is a histological pattern, defined by Z-disc–derived rods, shared by a group of genetic diseases of the sarcomeric thin filament and of the BTB-Kelch proteins that stabilise it [2, 5]. Eleven genes have moderate or definitive evidence in ClinGen, and NEB and ACTA1 together account for about 60% of genetically diagnosed patients in European and North American cohorts; ACTA1 accounts for about half of severe congenital forms, almost always through dominant de novo variants [4–6]. The retirement of the GeneReviews chapter in 2019 and the nomenclature proposed by the 277th ENMC workshop, which orders inheritance, gene, CMYO, and histological finding, reflect the shift to a gene-centred diagnosis; OMIM retains the NEM designation for NEB, TNNT1, KBTBD13, CFL2, KLHL40, KLHL41, and LMOD3 and uses CMYO for ACTA1, TPM2, TPM3, and MYPN [1, 2, 8].

Diagnosis now begins with genetics: panel, exome, or genome with CNV analysis, and rapid exome or genome sequencing in the severely affected newborn, with muscle biopsy reserved for inconclusive cases [1]. NEB requires methods capable of detecting structural variants and changes in the triplicate region, and a single variant in a recessive gene, a biopsy without rods, or a negative panel does not exclude the diagnosis [7, 25, 45]. Standardising variants on MANE transcripts avoids frequent errors, such as the dual numbering of the Amish TNNT1 founder variant (c.538G>T on NM_003283.6 and c.505G>T on NM_001126133.3) or the incorrect notation "c.103C>A" for CFL2 [23, 57].

The clinical spectrum ranges from fetal akinesia and neonatal death to mild weakness with a normal life expectancy. The feature that cuts across the forms is respiratory involvement disproportionate to limb weakness, with nocturnal hypoventilation preceding daytime complaints, together with bulbar dysfunction, especially with NEB, and scoliosis [5, 18, 48]. Some genes carry their own risks: the chest wall rigidity and infantile lethality of TNNT1 in the Amish form, the muscle slowness and cardiomyopathy with arrhythmias of KBTBD13, the occasional cardiac involvement of ACTA1, MYPN, and MYO18B, and the cervical instability of MYO18B [3, 17, 51, 75]. Recent cohorts from Boston, France and South America, Brazil, and Sweden have quantified the need for ventilation (42% in Boston, combining invasive and non-invasive; 55% in São Paulo; 67% in the de Feraudy NEB cohort), for tube or gastrostomy feeding (25–38%), and survival, but they are referral or single-centre cohorts and do not measure prevalence, which is around 0.2 per 100,000 inhabitants [4, 13, 15, 25, 48].

There is no approved treatment. Management follows the 2012 consensus: respiratory surveillance every 6–12 months with sitting and supine testing and polysomnography every 1–2 years, bilevel non-invasive ventilation and assisted cough, nutrition with early gastrostomy in severe forms, orthopaedic management of scoliosis and contractures without botulinum toxin, periodic cardiology assessment, and anaesthetic planning avoiding suxamethonium and volatile agents when possible, with dantrolene available but without prophylaxis [18]. The dantrolene dose for a malignant hyperthermia crisis differs between the US labels (1 mg/kg up to 10 mg/kg) and the European SmPC for Agilus® (2.5 mg/kg every 10 minutes) [103, 104]. L-tyrosine, pyridostigmine, and salbutamol are supported only by case reports, small series, or trials without released results, and inspiratory muscle training produced a small gain in MIP of uncertain clinical relevance [19, 73, 102, 108, 109]. Gene-directed therapies (KLHL40 replacement, KBTBD13 silencing) and thin-filament modulators are at the preclinical stage, and the natural history study infrastructure under way in the USA, the United Kingdom, Spain, Belgium, and the Netherlands is laying the groundwork for future trials [62, 118, 120, 121, 124, 126].

13. References

1. Villar-Quiles RN, Hayes LH, Raga S, Bönnemann C, Oates E, Dowling J, et al. 277th ENMC international workshop: Congenital myopathies: revising and revisiting nomenclature and diagnostic guidelines, 21-23 June 2024, Hoofddorp, The Netherlands. Neuromuscul Disord. 2026;60:106328. doi:10.1016/j.nmd.2025.106328.
2. North KN, Ryan MM. Nemaline Myopathy – RETIRED CHAPTER, FOR HISTORICAL REFERENCE ONLY. 2002 Jun 19 [updated 2014 Sep 18; revised 2015 Jun 11; retired 2019 Nov 7]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1288/
3. GeneGuidance. Curated clinical records for the genes NEB, ACTA1, TPM2, TPM3, TNNT1, KBTBD13, KLHL40, KLHL41, LMOD3, MYPN, CFL2, and MYO18B [database]. Version of 26 September 2026.
4. Hildebrandt C, Genetti CA, Logvinenko T, Win W, Barraza-Flores P, Hayes LH, et al. Retrospective Cohort Analysis of Clinical, Molecular, and Histopathologic Characteristics of 275 Patients With Nemaline Myopathy. Neurol Genet. 2025;11(4):e200277. doi:10.1212/NXG.0000000000200277.
5. North KN, Wang CH, Clarke N, Jungbluth H, Vainzof M, Dowling JJ, et al. Approach to the diagnosis of congenital myopathies. Neuromuscul Disord. 2014;24(2):97-116. doi:10.1016/j.nmd.2013.11.003.
6. Neuhaus SB, Wallgren-Pettersson C, Bönnemann CG, Schara U, Servais L. 250th ENMC International Workshop: Clinical trial readiness in nemaline myopathy 6-8 September 2019, Hoofdorp, the Netherlands. Neuromuscul Disord. 2020;30(10):866-875. doi:10.1016/j.nmd.2020.08.356.
7. Sagath L, Kiiski K, Naidu K, Patel K, Jonson PH, Laarne M, et al. Structural variation in nebulin and its impact on phenotype and inheritance: establishing a dominant distal phenotype caused by large deletions. Eur J Hum Genet. 2025;33(9):1153-1162. doi:10.1038/s41431-025-01891-0.
8. Online Mendelian Inheritance in Man (OMIM). Entries 108120, 161800, 255310, 256030, 605355, 609273, 609284, 609285, 610687, 615248, 615348, 615731, 616165, 616549, 616852, 617336, 619334, 620265, 620278, 620386, and 620389 (titles retrieved via NCBI E-utilities on 17 October 2026). Baltimore: Johns Hopkins University. Available from: https://omim.org/
9. Orphanet. Nemaline myopathy (ORPHA:607) and entities ORPHA:171430, 171433, 171436, 171439, 171442, 457074, 98902, 708123, and 447974 [Internet]. Orphadata data of 23 June 2026, accessed 17 October 2026. Available from: https://www.orpha.net/en/disease/detail/607.
10. Orphanet. Autosomal recessive distal nebulin myopathy (ORPHA:399103) [Internet]. Accessed 17 October 2026. Available from: https://www.orpha.net/en/disease/detail/399103.
11. Wallace SE, Bean LJH. Resources for Genetics Professionals — Genetic Disorders Associated with Founder Variants Common in the Ashkenazi Jewish Population. 2018 Dec 13 [revised 2022 Oct 20]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/n/gene/founder-ashkenazi/
12. Wallace SE, Puffenberger EG, Bean LJH. Resources for Genetics Professionals — Genetic Disorders Associated with Founder Variants Common in the Amish Population. 2020 Jun 18 [revised 2023 Dec 7]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/n/gene/founder-amish/
13. Huang K, Bi FF, Yang H. A Systematic Review and Meta-Analysis of the Prevalence of Congenital Myopathy. Front Neurol. 2021;12:761636. doi:10.3389/fneur.2021.761636.
14. Norwood FL, Harling C, Chinnery PF, Eagle M, Bushby K, Straub V. Prevalence of genetic muscle disease in Northern England: in-depth analysis of a muscle clinic population. Brain. 2009;132(Pt 11):3175-86. doi:10.1093/brain/awp236.
15. Michael E, Hedberg-Oldfors C, Tulinius M, Nordström S, Oldfors A, Darin N. Incidence and Prevalence of Congenital Myopathies - A Population-Based Study From Western Sweden. Ann Neurol. 2026;99(2):382-392. doi:10.1002/ana.78078.
16. Johnston JJ, Kelley RI, Crawford TO, Morton DH, Agarwala R, Koch T, et al. A novel nemaline myopathy in the Amish caused by a mutation in troponin T1. Am J Hum Genet. 2000;67(4):814-21. doi:10.1086/303089.
17. Fox MD, Carson VJ, Feng HZ, Lawlor MW, Gray JT, Brigatti KW, et al. TNNT1 nemaline myopathy: natural history and therapeutic frontier. Hum Mol Genet. 2018;27(18):3272-3282. doi:10.1093/hmg/ddy233.
18. Wang CH, Dowling JJ, North K, Schroth MK, Sejersen T, Shapiro F, et al. Consensus statement on standard of care for congenital myopathies. J Child Neurol. 2012;27(3):363-82. doi:10.1177/0883073812436605.
19. van Kleef ESB, Langer D, van Engelen BGM, Ottenheijm CAC, Voermans NC, Doorduin J. Inspiratory Muscle Training in Nemaline Myopathy. J Neuromuscul Dis. 2023;10(5):825-834. doi:10.3233/JND-221665.
20. Cure CMD. Congenital Muscle Disease International Registry (CMDIR). ClinicalTrials.gov NCT01403402. Available from: https://clinicaltrials.gov/study/NCT01403402.
21. Yuen M, Sandaradura SA, Dowling JJ, Kostyukova AS, Moroz N, Quinlan KG, et al. Leiomodin-3 dysfunction results in thin filament disorganization and nemaline myopathy. J Clin Invest. 2014;124(11):4693-708. doi:10.1172/JCI75199.
22. Miyatake S, Mitsuhashi S, Hayashi YK, Purevjav E, Nishikawa A, Koshimizu E, et al. Biallelic Mutations in MYPN, Encoding Myopalladin, Are Associated with Childhood-Onset, Slowly Progressive Nemaline Myopathy. Am J Hum Genet. 2017;100(1):169-178. doi:10.1016/j.ajhg.2016.11.017.
23. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. Available from: https://variantvalidator.org.
24. Ryan MM, Schnell C, Strickland CD, Shield LK, Morgan G, Iannaccone ST, et al. Nemaline myopathy: a clinical study of 143 cases. Ann Neurol. 2001;50(3):312-20. doi:10.1002/ana.1080.
25. de Feraudy Y, Maino A, Boedec M, Lornage X, Romero NB, Monges S, et al. Clinical, Histological, and Genetic Characterization of a Large Cohort of 49 Patients With Nebulin-Related Congenital Myopathy. Hum Mutat. 2026;2026:4385685. doi:10.1155/humu/4385685.
26. SHY GM, ENGEL WK, SOMERS JE, WANKO T. NEMALINE MYOPATHY. A NEW CONGENITAL MYOPATHY. Brain. 1963;86:793-810. doi:10.1093/brain/86.4.793.
27. CONEN PE, MURPHY EG, DONOHUE WL. LIGHT AND ELECTRON MICROSCOPIC STUDIES OF "MYOGRANULES" IN A CHILD WITH HYPOTONIA AND MUSCLE WEAKNESS. Can Med Assoc J. 1963;89(19):983-6.
28. Wallgren-Pettersson C, Beggs AH, Laing NG. 51st ENMC International Workshop: Nemaline Myopathy. 13-15 June 1997, Naarden, The Netherlands. Neuromuscul Disord. 1998;8(1):53-6.
29. Laing NG, Majda BT, Akkari PA, Layton MG, Mulley JC, Phillips H, et al. Assignment of a gene (NEMI) for autosomal dominant nemaline myopathy to chromosome I. Am J Hum Genet. 1992;50(3):576-83.
30. Laing NG, Wilton SD, Akkari PA, Dorosz S, Boundy K, Kneebone C, et al. A mutation in the alpha tropomyosin gene TPM3 associated with autosomal dominant nemaline myopathy. Nat Genet. 1995;9(1):75-79. doi:10.1038/ng0195-75.
31. Pelin K, Hilpelä P, Donner K, Sewry C, Akkari PA, Wilton SD, et al. Mutations in the nebulin gene associated with autosomal recessive nemaline myopathy. Proc Natl Acad Sci U S A. 1999;96(5):2305-10. doi:10.1073/pnas.96.5.2305.
32. Nowak KJ, Wattanasirichaigoon D, Goebel HH, Wilce M, Pelin K, Donner K, et al. Mutations in the skeletal muscle alpha-actin gene in patients with actin myopathy and nemaline myopathy. Nat Genet. 1999;23(2):208-12. doi:10.1038/13837.
33. Donner K, Ollikainen M, Ridanpää M, Christen HJ, Goebel HH, de Visser M, et al. Mutations in the beta-tropomyosin (TPM2) gene--a rare cause of nemaline myopathy. Neuromuscul Disord. 2002;12(2):151-8. doi:10.1016/s0960-8966(01)00252-8.
34. Anderson SL, Ekstein J, Donnelly MC, Keefe EM, Toto NR, LeVoci LA, et al. Nemaline myopathy in the Ashkenazi Jewish population is caused by a deletion in the nebulin gene. Hum Genet. 2004;115(3):185-90. doi:10.1007/s00439-004-1140-8.
35. Agrawal PB, Greenleaf RS, Tomczak KK, Lehtokari VL, Wallgren-Pettersson C, Wallefeld W, et al. Nemaline myopathy with minicores caused by mutation of the CFL2 gene encoding the skeletal muscle actin-binding protein, cofilin-2. Am J Hum Genet. 2007;80(1):162-7. doi:10.1086/510402.
36. Sambuughin N, Yau KS, Olivé M, Duff RM, Bayarsaikhan M, Lu S, et al. Dominant mutations in KBTBD13, a member of the BTB/Kelch family, cause nemaline myopathy with cores. Am J Hum Genet. 2010;87(6):842-7. doi:10.1016/j.ajhg.2010.10.020.
37. Ravenscroft G, Miyatake S, Lehtokari VL, Todd EJ, Vornanen P, Yau KS, et al. Mutations in KLHL40 are a frequent cause of severe autosomal-recessive nemaline myopathy. Am J Hum Genet. 2013;93(1):6-18. doi:10.1016/j.ajhg.2013.05.004.
38. Gupta VA, Ravenscroft G, Shaheen R, Todd EJ, Swanson LC, Shiina M, et al. Identification of KLHL41 Mutations Implicates BTB-Kelch-Mediated Ubiquitination as an Alternate Pathway to Myofibrillar Disruption in Nemaline Myopathy. Am J Hum Genet. 2013;93(6):1108-17. doi:10.1016/j.ajhg.2013.10.020.
39. Alazami AM, Kentab AY, Faqeih E, Mohamed JY, Alkhalidi H, Hijazi H, et al. A novel syndrome of Klippel-Feil anomaly, myopathy, and characteristic facies is linked to a null mutation in MYO18B. J Med Genet. 2015;52(6):400-4. doi:10.1136/jmedgenet-2014-102964.
40. Malfatti E, Böhm J, Lacène E, Beuvin M, Romero NB, Laporte J. A Premature Stop Codon in MYO18B is Associated with Severe Nemaline Myopathy with Cardiomyopathy. J Neuromuscul Dis. 2015;2(3):219-227. doi:10.3233/JND-150085.
41. Lornage X, Malfatti E, Chéraud C, Schneider R, Biancalana V, Cuisset JM, et al. Recessive MYPN mutations cause cap myopathy with occasional nemaline rods. Ann Neurol. 2017;81(3):467-473. doi:10.1002/ana.24900.
42. Konersman CG, Freyermuth F, Winder TL, Lawlor MW, Lagier-Tourenne C, Patel SB. Novel autosomal dominant TNNT1 mutation causing nemaline myopathy. Mol Genet Genomic Med. 2017;5(6):678-691. doi:10.1002/mgg3.325.
43. Sewry CA, Laitila JM, Wallgren-Pettersson C. Nemaline myopathies: a current view. J Muscle Res Cell Motil. 2019;40(2):111-126. doi:10.1007/s10974-019-09519-9.
44. Garg A, O'Rourke J, Long C, Doering J, Ravenscroft G, Bezprozvannaya S, et al. KLHL40 deficiency destabilizes thin filament proteins and promotes nemaline myopathy. J Clin Invest. 2014;124(8):3529-39. doi:10.1172/JCI74994.
45. Cicala G, Mccauley J, Phadke R, Mueller J, Robb SA, Manzur AY, et al. Genetic Landscape and Diagnostic Outcomes of UK Patients With Congenital Myopathies and Muscular Dystrophies Over a 10-Year Period. Neurol Genet. 2026;12(2):e200354. doi:10.1212/NXG.0000000000200354.
46. Henning F, Naidu K, Thomas P, Schoeman M, Zaharie D, Patel K, et al. Phenotypic characterization of dominantly inherited distal nebulin myopathy. Neuromuscul Disord. 2026;62:106415. doi:10.1016/j.nmd.2026.106415.
47. Lehtokari VL, Greenleaf RS, DeChene ET, Kellinsalmi M, Pelin K, Laing NG, et al. The exon 55 deletion in the nebulin gene--one single founder mutation with world-wide occurrence. Neuromuscul Disord. 2009;19(3):179-81. doi:10.1016/j.nmd.2008.12.001.
48. Moreno CAM, Artilheiro MC, Fonseca ATQSM, Camelo CG, de Medeiros GC, Sassi FC, et al. Clinical Manifestation of Nebulin-Associated Nemaline Myopathy. Neurol Genet. 2023;9(1):e200056. doi:10.1212/NXG.0000000000200056.
49. Clayton JS, Johari M, Taylor RL, Dofash L, Allan G, Monahan G, et al. An Update on Reported Variants in the Skeletal Muscle α-Actin (ACTA1) Gene. Hum Mutat. 2024;2024:6496088. doi:10.1155/2024/6496088.
50. Laing NG, Clarke NF, Dye DE, Liyanage K, Walker KR, Kobayashi Y, et al. Actin mutations are one cause of congenital fibre type disproportion. Ann Neurol. 2004;56(5):689-94. doi:10.1002/ana.20260.
51. Lerchner T, Kaiser FJ, Chodyla MK, Schlosser T, Surowy H, Kurt M, et al. A novel likely pathogenic ACTA1 variant is associated with dilated cardiomyopathy in nemaline myopathy. ESC Heart Fail. 2026;. doi:10.1093/eschf/xvag227.
52. Wallgren-Pettersson C, Jokela M, Lehtokari VL, Tyynismaa H, Sainio MT, Ylikallio E, et al. Variants in tropomyosins TPM2 and TPM3 causing muscle hypertonia. Neuromuscul Disord. 2024;35:29-32. doi:10.1016/j.nmd.2023.12.006.
53. Lambert MR, Gussoni E. Tropomyosin 3 (TPM3) function in skeletal muscle and in myopathy. Skelet Muscle. 2023;13(1):18. doi:10.1186/s13395-023-00327-x.
54. Tajsharghi H, Ohlsson M, Palm L, Oldfors A. Myopathies associated with β-tropomyosin mutations. Neuromuscul Disord. 2012;22(11):923-33. doi:10.1016/j.nmd.2012.05.018.
55. Monnier N, Lunardi J, Marty I, Mezin P, Labarre-Vila A, Dieterich K, et al. Absence of beta-tropomyosin is a new cause of Escobar syndrome associated with nemaline myopathy. Neuromuscul Disord. 2009;19(2):118-23. doi:10.1016/j.nmd.2008.11.009.
56. Schirwani S, Sarkozy A, Phadke R, Childs AM, Mein R, Ismail A, et al. Homozygous intronic variants in TPM2 cause recessively inherited Escobar variant of multiple pterygium syndrome and congenital myopathy. Neuromuscul Disord. 2021;31(4):359-366. doi:10.1016/j.nmd.2020.09.033.
57. Landrum MJ, Lee JM, Benson M, et al. ClinVar [Internet]. Bethesda (MD): National Center for Biotechnology Information. Records VCV000012440, VCV000940131, VCV000008160, VCV000846771, VCV000374498, and VCV000208842, accessed 17 October 2026. Available from: https://www.ncbi.nlm.nih.gov/clinvar/
58. Jin JP, Brotto MA, Hossain MM, Huang QQ, Brotto LS, Nosek TM, et al. Truncation by Glu180 nonsense mutation results in complete loss of slow skeletal muscle troponin T in a lethal nemaline myopathy. J Biol Chem. 2003;278(28):26159-65. doi:10.1074/jbc.M303469200.
59. Feng HZ, Strauss KA, Jin JP. Potential cytotoxicity of truncated slow skeletal muscle troponin T (ssTnT) in a loss of function TNNT1 myopathy mouse model. FEBS J. 2025;292(20):5525-5539. doi:10.1111/febs.70165.
60. Laarne M, Oghabian A, Laitila J, Isohanni P, Tynninen O, Zhao F, et al. A homozygous single-nucleotide variant in TNNT1 causes abnormal troponin T isoform expression in a patient with severe nemaline myopathy: a case report. J Neuromuscul Dis. 2025;12(5):689-698. doi:10.1177/22143602251339569.
61. Pauw-Gommans IM, Gerrits KH, de Haan A, van Engelen BG. Muscle slowness in a family with nemaline myopathy. Neuromuscul Disord. 2006;16(8):477-80. doi:10.1016/j.nmd.2006.05.003.
62. Galli RA, Vonk LA, Baelde RJ, van den Heuvel A, Bogaards S, Conijn S, et al. Kbtbd13 knockdown restores muscle function in a clinically relevant mouse model of nemaline myopathy type 6. Sci Transl Med. 2026;18(855):eadx2116. doi:10.1126/scitranslmed.adx2116.
63. Bouman K, Küsters B, De Winter JM, Gillet C, Van Kleef ESB, Eshuis L, et al. NEM6, KBTBD13-Related Congenital Myopathy: Myopathological Analysis in 18 Dutch Patients Reveals Ring Rods Fibers, Cores, Nuclear Clumps, and Granulo-Filamentous Protein Material. J Neuropathol Exp Neurol. 2021;80(4):366-376. doi:10.1093/jnen/nlab012.
64. Yeung KS, Yu FNY, Fung CW, Wong S, Lee HHC, Fung STH, et al. The KLHL40 c.1516A>C is a Chinese-specific founder mutation causing nemaline myopathy 8: Report of six patients with pre- and postnatal phenotypes. Mol Genet Genomic Med. 2020;8(7):e1229. doi:10.1002/mgg3.1229.
65. Ramirez-Martinez A, Cenik BK, Bezprozvannaya S, Chen B, Bassel-Duby R, Liu N, et al. KLHL41 stabilizes skeletal muscle sarcomeres by nonproteolytic ubiquitination. Elife. 2017;6. doi:10.7554/eLife.26439.
66. Han H, Tate GR, Shi J, Bongiorno AH, Mansur A, Widrick J, et al. Klhl41b orchestrates sarcomere assembly and growth to drive skeletal muscle hypertrophy in vivo. Skelet Muscle. 2026;16:38. doi:10.1186/s13395-026-00453-2.
67. Schatz UA, Weiss S, Wenninger S, Schoser B, Muss WH, Bittner RE, et al. Evidence of mild founder LMOD3 mutations causing nemaline myopathy 10 in Germany and Austria. Neurology. 2018;91(18):e1690-e1694. doi:10.1212/WNL.0000000000006428.
68. Ockeloen CW, Gilhuis HJ, Pfundt R, Kamsteeg EJ, Agrawal PB, Beggs AH, et al. Congenital myopathy caused by a novel missense mutation in the CFL2 gene. Neuromuscul Disord. 2012;22(7):632-9. doi:10.1016/j.nmd.2012.03.008.
69. Ong RW, AlSaman A, Selcen D, Arabshahi A, Yau KS, Ravenscroft G, et al. Novel cofilin-2 (CFL2) four base pair deletion causing nemaline myopathy. J Neurol Neurosurg Psychiatry. 2014;85(9):1058-60. doi:10.1136/jnnp-2014-307608.
70. Polavarapu K, Bardhan M, Anjanappa RM, Vengalil S, Preethish-Kumar V, Shingavi L, et al. Nemaline Rod/Cap Myopathy Due to Novel Homozygous MYPN Mutations: The First Report from South Asia and Comprehensive Literature Review. J Clin Neurol. 2021;17(3):409-418. doi:10.3988/jcn.2021.17.3.409.
71. Altuame FD, Haldeman-Englert C, Cupler E, Al Muhaizea MA, Al-Zaidan HI, Hashem M, et al. Further delineation of MYO18B-related autosomal recessive Klippel-Feil syndrome with myopathy and facial dysmorphism. Am J Med Genet A. 2021;185(2):370-376. doi:10.1002/ajmg.a.61957.
72. Saito Y, Nishikawa A, Iida A, Mori-Yoshimura M, Oya Y, Ishiyama A, et al. ADSSL1 myopathy is the most common nemaline myopathy in Japan with variable clinical features. Neurology. 2020;95(11):e1500-e1511. doi:10.1212/WNL.0000000000010237. PMID: 32646962.
73. Natera-de Benito D, Nascimento A, Abicht A, Ortez C, Jou C, Müller JS, et al. KLHL40-related nemaline myopathy with a sustained, positive response to treatment with acetylcholinesterase inhibitors. J Neurol. 2016;263(3):517-23. doi:10.1007/s00415-015-8015-x.
74. Zhao L, Deng F, Cai B. ACTA1-related congenital myopathy in a neonate: a case report and literature review. Front Pediatr. 2025;13:1706982 (published 20 January 2026). doi:10.3389/fped.2025.1706982.
75. de Winter JM, Bouman K, Strom J, Methawasin M, Jongbloed JDH, van der Roest W, et al. KBTBD13 is a novel cardiomyopathy gene. Hum Mutat. 2022;43(12):1860-1865. doi:10.1002/humu.24499.
76. Lehtokari VL, Kiiski K, Sandaradura SA, Laporte J, Repo P, Frey JA, et al. Mutation update: the spectra of nebulin variants and associated myopathies. Hum Mutat. 2014;35(12):1418-26. doi:10.1002/humu.22693.
77. Ryan MM, Ilkovski B, Strickland CD, Schnell C, Sanoudou D, Midgett C, et al. Clinical course correlates poorly with muscle pathology in nemaline myopathy. Neurology. 2003;60(4):665-73. doi:10.1212/01.wnl.0000046585.81304.bc.
78. Hedberg-Oldfors C, Bedir AZ, Visuttijai K, Michael E, Oldfors A. Proteomic insights into infantile nemaline myopathy reveal dysregulated proteins and cellular pathways [abstract PP01.08]. In: Abstracts of the 19th International Congress on Neuromuscular Diseases, 7–11 July 2026. J Neuromuscul Dis. 2026. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC13434919/
79. Schade van Westrum SM, Wahbi K, Bonne G, van der Kooi AJ. Hereditary Neuromuscular Diseases and Cardiac Involvement. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 371–385.
80. Lomen-Hoerth C, Simmons ML, Dearmond SJ, Layzer RB. Adult-onset nemaline myopathy: Another cause of dropped head. Muscle Nerve. 1999;22(8):1146-50. doi:10.1002/(sici)1097-4598(199908)22:8<1146::aid-mus23>3.0.co;2-u.
81. Chahin N, Selcen D, Engel AG. Sporadic late onset nemaline myopathy. Neurology. 2005;65(8):1158-64. doi:10.1212/01.wnl.0000180362.90078.dc.
82. Zambon AA, Abel F, Linnane B, O'Rourke D, Phadke R, Sewry CA, et al. Troponin-T type 1 (TNNT1)-related nemaline myopathy: unique respiratory phenotype and muscle pathology findings. Neuromuscul Disord. 2022;32(3):245-254. doi:10.1016/j.nmd.2022.01.014.
83. Cossette SA, van de Camp SAJH, Lee LM, Lambert MR, Servais L, Voermans NC. Nemaline myopathy patient and caregiver perspectives: a multinational cross-sectional study. Neuromuscul Disord. 2026;68:107427. doi:10.1016/j.nmd.2026.107427.
84. van Kleef ESB, Bouman K, Molenaar JPF, Küsters B, Groothuis JT, Olivé M, et al. Nemaline Myopathy Type 6 Caused by Variants in the KBTBD13 Gene: A Cross-Sectional Study of 24 Patients. Neurol Genet. 2024;10(6):e200214. doi:10.1212/NXG.0000000000200214.
85. Strauss KA, Carson VJ, Bolettieri E, Everett M, Bollinger A, Bowser LE, et al. WiTNNess: an international natural history study of infantile-onset TNNT1 myopathy. Ann Clin Transl Neurol. 2023;10(11):1972-1984. doi:10.1002/acn3.51884.
86. Seferian AM, Malfatti E, Bosson C, Pelletier L, Taytard J, Forin V, et al. Mild clinical presentation in KLHL40-related nemaline myopathy (NEM 8). Neuromuscul Disord. 2016;26(10):712-716. doi:10.1016/j.nmd.2016.07.011.
87. Michael E, Hedberg-Oldfors C, Wilmar P, Visuttijai K, Oldfors A, Darin N. Long-term follow-up and characteristic pathological findings in severe nemaline myopathy due to LMOD3 mutations. Neuromuscul Disord. 2019;29(2):108-113. doi:10.1016/j.nmd.2018.12.009.
88. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 724–727 (Fetal akinesia).
89. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 234–235 (Pena-Shokeir phenotype).
90. Zhuang J, Hong M, Huang Q, Jiang Y, Wang J. Prenatal Diagnosis and Functional Analysis of Two Compound Heterozygous Variants in the KLHL40 Gene Causing Nemaline Myopathy 8. Mol Genet Genomic Med. 2026;14(7):e70270. doi:10.1002/mgg3.70270.
91. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 154–157 (Floppy infant).
92. Lauletta A, Torchia E, Schwarz JM, Garibaldi M, Kleefeld F, Laforet P, et al. 294th ENMC international workshop: Diagnosis and management of paraproteinemic myopathies focusing on sporadic late-onset nemaline myopathy (SLONM) and light-chain (AL) amyloid myopathy. 27th-29th March 2026, Hoofddorp, The Netherlands. Neuromuscul Disord. 2026;67:107394. doi:10.1016/j.nmd.2026.107394.
93. Scott SA, Edelmann L, Liu L, Luo M, Desnick RJ, Kornreich R. Experience with carrier screening and prenatal diagnosis for 16 Ashkenazi Jewish genetic diseases. Hum Mutat. 2010;31(11):1240-50. doi:10.1002/humu.21327.
94. Jungbluth H, Sewry CA, Counsell S, Allsop J, Chattopadhyay A, Mercuri E, et al. Magnetic resonance imaging of muscle in nemaline myopathy. Neuromuscul Disord. 2004;14(12):779-84. doi:10.1016/j.nmd.2004.08.005.
95. Lauletta A, Forcina F, Merlonghi G, Fionda L, Leonardi L, Costanzo R, et al. Sporadic late-onset nemaline myopathy (SLONM): Data from a case series and literature review of 144 patients. Autoimmun Rev. 2026;25(2):103960. doi:10.1016/j.autrev.2025.103960.
96. Campos IFS, Ribeiro EM. Miopatias congênitas [Congenital myopathies]. In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar [Genetic Syndromes: Recognize to Treat]. Fortaleza: EdUnichristus; 2021. p. 285–288.
97. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 32, section 32.4.2: Muscle-Specific Adenylosuccinate Synthase Deficiency.
98. European Medicines Agency. Download medicine data: list of medicines for human use [Internet]. Accessed 17 October 2026. Available from: https://www.ema.europa.eu/en/medicines/download-medicine-data.
99. National Library of Medicine. ClinicalTrials.gov [Internet]. Search for studies on nemaline myopathy and congenital myopathies, accessed 17 October 2026. Available from: https://clinicaltrials.gov/
100. Viatris Healthcare. Neurontin 100 mg, 300 mg e 400 mg cápsulas — resumo das características do medicamento [Neurontin 100 mg, 300 mg, and 400 mg capsules — summary of product characteristics]. INFARMED, Infomed (med_id 6057); first authorisation 27 Sep 1996, last renewal 6 Feb 2009. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=6057&tipo_doc=rcm (accessed 26 October 2026).
101. Viatris Specialty LLC. Neurontin® (gabapentin) [prescribing information]. DailyMed set id 97935fd9-1d4a-43b6-a5d9-de994591187b; version 2026-05-12. NDA020235. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=97935fd9-1d4a-43b6-a5d9-de994591187b (accessed 26 October 2026).
102. Ryan MM, Sy C, Rudge S, Ellaway C, Ketteridge D, Roddick LG, et al. Dietary L-tyrosine supplementation in nemaline myopathy. J Child Neurol. 2008;23(6):609-13. doi:10.1177/0883073807309794.
103. Par Health USA. Dantrium® Intravenous (dantrolene sodium for injection) [prescribing information]. DailyMed set id 4df35098-8702-46be-ac67-30cfdf1aa570; version 2026-03-10. NDA018264 (original approval 1979-09-19). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=4df35098-8702-46be-ac67-30cfdf1aa570.
104. European Medicines Agency. Agilus (dantrolene sodium hemiheptahydrate) — EPAR, Summary of Product Characteristics. EMEA/H/C/006009; holder Norgine B.V.; CHMP opinion 21 Mar 2024; authorisation 29 May 2024. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/agilus.
105. Eagle Pharmaceuticals. Ryanodex® (dantrolene sodium) for injectable suspension [prescribing information]. DailyMed set id 8f7b3ac0-604d-4c78-b545-5e0f8ea3d698; version 2024-10-30. NDA205579 (original approval 2014-07-22). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=8f7b3ac0-604d-4c78-b545-5e0f8ea3d698.
106. Dalal N, Naranje K, Moriangthem A, Singh A. L-tyrosine for treatment of an infant with nemaline rod myopathy. BMJ Case Rep. 2024;17(6):e259303. doi:10.1136/bcr-2023-259303.
107. Hadassah Medical Organization. Treatment of TNNT1-Myopathy With L-Tyrosine. ClinicalTrials.gov NCT02035501. https://clinicaltrials.gov/study/NCT02035501.
108. Buchignani B, Marinella G, Pasquariello R, et al. KLHL40-Related Myopathy: A Systematic Review and Insight into a Follow-up Biomarker via a New Case Report. Genes (Basel). 2024;15(2):208. doi:10.3390/genes15020208.
109. Västra Götaland Region. COMPIS – Congenital Myopathy Intervention Study (oral salbutamol). ClinicalTrials.gov NCT05099107. https://clinicaltrials.gov/study/NCT05099107.
110. Mestinon 60 mg comprimidos revestidos (brometo de piridostigmina) — resumo das características do medicamento [Mestinon 60 mg coated tablets (pyridostigmine bromide) — summary of product characteristics]. INFARMED, Infomed; text approved 8 Jul 2020. Available from: https://extranet.infarmed.pt/INFOMED-fo/ (accessed 26 October 2026).
111. Amneal Pharmaceuticals of New York. Pyridostigmine bromide tablets [prescribing information]. DailyMed set id 0f56facf-1f0a-4ba5-876d-757b114dfa45; version 2025-10-16. ANDA040502. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=0f56facf-1f0a-4ba5-876d-757b114dfa45 (accessed 26 October 2026).
112. Foundation Consumer Brands. Primatene® Asthma Tablets (ephedrine hydrochloride 12.5 mg) [OTC drug facts]. DailyMed set id 2170f081-c03f-47e8-bc63-4052c5e653fa; version 2024-11-07. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=2170f081-c03f-47e8-bc63-4052c5e653fa (accessed 26 October 2026).
113. Chartwell RX. Albuterol sulfate syrup [prescribing information]. DailyMed set id 560df50c-e2c4-459f-9d89-bf7e6588949e; version 2025-03-28. ANDA077788. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=560df50c-e2c4-459f-9d89-bf7e6588949e (accessed 26 October 2026).
114. Radboud University Medical Center. Inspiratory Muscle Training in Patients With Nemaline Myopathy. ClinicalTrials.gov NCT03728803. https://clinicaltrials.gov/study/NCT03728803.
115. de Winter JM, Joureau B, Sequeira V, Clarke NF, van der Velden J, Stienen GJ, et al. Effect of levosimendan on the contractility of muscle fibers from nemaline myopathy patients with mutations in the nebulin gene. Skelet Muscle. 2015;5:12. doi:10.1186/s13395-015-0037-7.
116. Tinklenberg JA, Siebers EM, Beatka MJ, Fickau BA, Ayres S, Meng H, et al. Myostatin inhibition using ActRIIB-mFc does not produce weight gain or strength in the nebulin conditional KO mouse. J Neuropathol Exp Neurol. 2019;78(2):130-139. doi:10.1093/jnen/nly120.
117. Fisher G, Mackels L, Markati T, Sarkozy A, Ochala J, Jungbluth H, et al. Early clinical and pre-clinical therapy development in nemaline myopathy. Expert Opin Ther Targets. 2022;26(10):853-867. doi:10.1080/14728222.2022.2157258.
118. Stanford University. Nemaline Myopathy Clinical Research Network (NM-CTRN). ClinicalTrials.gov NCT06774703. https://clinicaltrials.gov/study/NCT06774703.
119. Clinic for Special Children. WiTNNess: An International Natural History Study of Autosomal Recessive TNNT1 Myopathy. ClinicalTrials.gov NCT06374719. https://clinicaltrials.gov/study/NCT06374719.
120. University of Oxford. NatHis-NM-MDUK. ClinicalTrials.gov NCT06670378. https://clinicaltrials.gov/study/NCT06670378.
121. Vall d'Hebron Research Institute. Natural History Study for Patients With Nemaline Myopathy in Spain. ClinicalTrials.gov NCT07488806. https://clinicaltrials.gov/study/NCT07488806.
122. CHU de Liège. Natural History Study for Patients With Nemaline Myopathy in Belgium. ClinicalTrials.gov NCT07201636. https://clinicaltrials.gov/study/NCT07201636.
123. van de Camp SAJH, Brenninkmeijer R, van Doorn JLM, de Laat ECM, Pomp L, Stinissen L, et al. READYCOM: protocol for a 2-year prospective natural history and cross-sectional muscle-fatigability study for improving trial readiness in congenital myopathies. BMJ Open. 2026;16(8):e119529. doi:10.1136/bmjopen-2026-119529.
124. Tate GR, McNulty J, Widrick J, Gupta VA. KLHL40 gene replacement therapy in severe nemaline myopathy improves survival and skeletal muscle function in a preclinical mouse model. bioRxiv [preprint]. 2026 Sep 23. doi:10.64898/2026.09.23.753846.
125. Baelde RJ, Vonk LA, Nollet EE, Galli RA, Monteiro AF, Bastu S, et al. Nicotinamide riboside prevents mitochondrial dysfunction in nemaline myopathy type 6. Hum Mol Genet. 2026;35(12). doi:10.1093/hmg/ddag023.
126. Galli RA, Borsboom TC, Gineste C, Brocca L, Rossi M, Hwee DT, et al. Tirasemtiv enhances submaximal muscle tension in an Acta1:p.Asp286Gly mouse model of nemaline myopathy. J Gen Physiol. 2024;156(4):e202313471. doi:10.1085/jgp.202313471.
127. de Winter JM, Gineste C, Minardi E, Brocca L, Rossi M, Borsboom T, et al. Acute and chronic tirasemtiv treatment improves in vivo and in vitro muscle performance in actin-based nemaline myopathy mice. Hum Mol Genet. 2021;30(14):1305-1320. doi:10.1093/hmg/ddab112.
128. Lee EJ, Kolb J, Hwee DT, Malik FI, Granzier HL. Functional characterization of the intact diaphragm in a nebulin-based nemaline myopathy (NM) model — effects of the fast skeletal muscle troponin activator tirasemtiv. Int J Mol Sci. 2019;20(20):5008. doi:10.3390/ijms20205008.
129. Karimi E, Gohlke J, van der Borgh M, Lindqvist J, Hourani Z, Kolb J, et al. Characterization of NEB pathogenic variants in patients reveals novel nemaline myopathy disease mechanisms and omecamtiv mecarbil force effects. Acta Neuropathol. 2024;147(1):72. doi:10.1007/s00401-024-02726-w.
130. Lindqvist J, Granzier H. Pharmacological inhibition of myostatin in a mouse model of typical nemaline myopathy increases muscle size and force. Int J Mol Sci. 2023;24(20):15124. doi:10.3390/ijms242015124.
131. Casey JG, Kim ES, Joseph R, Li F, Granzier H, Gupta VA. NRAP reduction rescues sarcomere defects in nebulin-related nemaline myopathy. Hum Mol Genet. 2023;32(10):1711-1721. doi:10.1093/hmg/ddad011.
132. Coulson Z, Kolb J, Sabha N, Karimi E, Hourani Z, Ottenheijm C, et al. Generation of a novel mouse model of nemaline myopathy due to recurrent NEB exon 55 deletion. Skelet Muscle. 2025;15(1):8. doi:10.1186/s13395-025-00378-2.
133. Arjona RM, Velarde-García JF, Del Cerro EP, Monroy AM. Parents' Perspectives on Home Care for Children With Nemaline Myopathy: A Qualitative Study. Child Care Health Dev. 2026;52(4):e70301. doi:10.1111/cch.70301.
134. Pei X, Zhai Y, Yang F, Lu W. CARE-compliant case report: Nemaline myopathy caused by the ACTA1 p.Q139H missense mutation. Medicine (Baltimore). 2025;104(45):e45459. doi:10.1097/MD.0000000000045459.
135. Wallgren-Pettersson C, Pelin K, Nowak KJ, Muntoni F, Romero NB, Goebel HH, et al. Genotype-phenotype correlations in nemaline myopathy caused by mutations in the genes for nebulin and skeletal muscle alpha-actin. Neuromuscul Disord. 2004;14(8-9):461-70. doi:10.1016/j.nmd.2004.03.006.
136. Deenen JC, Verbeek AL, Verschuuren JJ, van Engelen BG, Voermans NC. Prevalence and incidence rates of 17 neuromuscular disorders: An updated review of the literature. J Neuromuscul Dis. 2025;12(6):713-722. doi:10.1177/22143602241313118.
137. Amburgey K, McNamara N, Bennett LR, McCormick ME, Acsadi G, Dowling JJ. Prevalence of congenital myopathies in a representative pediatric united states population. Ann Neurol. 2011;70(4):662-5. doi:10.1002/ana.22510.
138. A Foundation Building Strength. Institutional website [Internet]. Accessed 17 October 2026. Available from: https://www.buildingstrength.org/
139. Muscular Dystrophy UK. Nemaline myopathy (NM) [Internet]. Reviewed February 2025. Accessed 17 October 2026. Available from: https://www.musculardystrophyuk.org/conditions/a-z/nemaline-myopathy/
140. Martinez BA, Lake BD. Childhood nemaline myopathy: a review of clinical presentation in relation to prognosis. Dev Med Child Neurol. 1987;29(6):815-20. doi:10.1111/j.1469-8749.1987.tb08832.x.