Pediatric Nemaline Myopathy: A Systematic Review Using Individual Patient Data

Abstract

Nemaline myopathy is a skeletal muscle disease that affects 1 in 50 000 live births. The objective of this study was to develop a narrative synthesis of the findings of a systematic review of the latest case descriptions of patients with NM. A systematic search of MEDLINE, Embase, CINAHL, Web of Science, and Scopus was performed using Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines using the keywords pediatric, child, NM, nemaline rod, and rod myopathy. Case studies focused on pediatric NM and published in English between January 1, 2010, and December 31, 2020, in order to represent the most recent findings. Information was collected about the age of first signs, earliest presenting neuromuscular signs and symptoms, systems affected, progression, death, pathologic description, and genetic changes. Of a total of 385 records, 55 case reports or series were reviewed, covering 101 pediatric patients from 23 countries. We review varying presentations in children ranging in severity despite being caused by the same mutation, in addition to current and future clinical considerations relevant to the care of patients with NM. This review synthesizes genetic, histopathologic, and disease presentation findings from pediatric NM case reports. These data strengthen our understanding of the wide spectrum of disease seen in NM. Future studies are needed to identify the underlying molecular mechanism of pathology, to improve diagnostics, and to develop better methods to improve the quality of life for these patients.

Nemaline myopathy (NM) is a primary skeletal muscle disease and histopathologic diagnosis with variable clinical presentation and genetic causes. It has an estimated incidence of 1 in 50 000 live births.¹ The disorder was first described in 1963 in the case of a child with hypotonia and “microgranules” in a muscle biopsy.^2,3 Clinicohistopathologic diagnosis of several congenital myopathies includes the detection of such aggregates, distinguished as cores, electron-dense rods, or central nuclei.⁴ Rods detected in the sarcoplasm or inside the nucleus leads to a diagnosis of NM. These rods are often seen emanating from the muscle Z-disc, the anchor point of the sarcomere unit (Figure 1). Sarcomeric proteins have been shown to make up a portion of these aggregates (Figure 2).^5,6 Additionally, individual muscle fibers may be affected by either atrophy or hypertrophy.^7,8

Figure 1.

Pathologic findings in NM. Light microscopy of Gomori trichrome-stained muscle biopsy shows nemaline bodies within various fibers (left, cross-section). Electron-dense rods disrupt sarcomeric organization in the muscle (right, longitudinal section). Arrows highlight some of the present nemaline rods/bodies.

Figure 2.

Sarcomere-related proteins causative of NM. Created with BioRender.com.

To date, pathogenic variants in 12 genes have been identified as causative of NM (Table 1).⁹ Several of the implicated proteins localize to the thin filament of the sarcomere: skeletal alpha-actin 1 (ACTA1), cofilin (CFL2), leiomodin 3 (LMOD3), nebulin (NEB1), and kelch repeat and BTB domain containing protein 13 (KBTBD13).^10–15 Mutations have also been found in Kelch-like family member proteins (KLHL40, KLHL41) that regulate the protein turnover of thin filaments.¹⁶ Additional structural or regulatory components of the sarcomere may be affected as shown by mutations in genes encoding myosin 18B (MYO18B), myopalladin (MYPN), ryanodine receptor 1 (RYR1), slow skeletal muscle troponin T (TNNT1), fast skeletal muscle troponin T (TNNT3), and slow muscle alpha-tropomyosin (TPM2, TPM3).^17–21 As far as the functions of the proteins, skeletal alpha-actin is the main component of the actin filament, which is structurally supported by nebulin, troponin, tropomyosin, and myopalladin.^{10,13,17,19,21} Leiomodin is an actin nucleator, while cofilin is an actin-severing protein involved in maintaining the balance between filamentous and monomeric actin.^11,12 Meanwhile, the KBTBD13 adaptor protein works with KLHL40 and KLHL41 to regulate thin filament stability with the ubiquitin-proteasome system (Figure 2).^15,16

Table 1.

Genetics of Pediatric NM Case Reports (2010–2020): Genetic Changes Described in the Cases Reviewed Presently, Including Number of Cases by Gene Symbol, Genetic and Protein Changes, and a Description of Protein Function.

Gene Symbol	Cases	Protein Function	Genetic and Protein Changes	References
ACTA1	22	Principal actin isoform of skeletal muscle thin filament	c.283C>A, p.Asn94Lys	Saito 2011
			c.350A>G, p.Asn117Ser	Yang 2016
			c.356G>A, p.Glu85Lys	Lehtokari 2018
			c.407T>C, p.Val136Ala	Ennis 2015
			c.430C > T, p.Leu144Phe	Saito 2011
			c.448A>G, p.Thr150Ala	Miyatake 2014
			c.455G>C, p.Gly152Ala	Ravenscroft 2011
			c.478G>A, p.Gly160Ser	Yokoi 2019
			c.487C>G, p.His163Asp	Moreno 2017
			c.557A>G, p.Asp186Gly	Levesque 2013
			c.611C>T, p.Thr204Ile	Moreno 2017
			c.760A>C, p.Asn254His	Pula 2020
			c801G>C, p.Gly270Arg	Saito 2011
			c.868G>A, p.Asp290Asn	Seidahmed 2016
			c.871A>T, p.Tyr5927HisfsX17	Moureau-Le Lan 2018
			c.911delG, p.Gly304AlafsX24	Friedman 2014
			c.1049C>T, p.Ser350Leu	Moreno 2017
			c.1074G>T, p.Trp358Cys	Gatayama 2013
			c.1075A>C, p.Ile359Leu	Yeşilbaş 2019
			c.1127G > C, p.Cys376Ser	Waisayarat 2015
CFL2	6	Actin-severing and actin-binding protein	c.19G>A, p.Val7Met	Ockeloen 2012
			c.100_103delAAAG, p.Lys34Glnfs*6	Ong 2014
			c.235G>T, p.Asp79Tyr	Fattori 2018
			c.256G>C, p.Asp86His	Fattori 2018
			c.281delC, p.Ser94LeufsTer6	Fattori 2018
KLHL40	10	Binding partner of E3 ligase cullin 3	c.604delG, p.Ala202Argfs*56	Natera-de Benito 2016
			c.1405G>T, p.Gly469Cys	Kawase 2015
			c.1516A>C p.Thr506Pro	Yeung 2020
			c.1327G>A, p.Gly433Ser	Yeung 2020
			c.1498C>T, p.Arg500Cys	Seferian 2016
			c.1513G>C, p.Ala505Pro	Natera-de Benito 2016
			chr3:42727712G>A, p.Trp201Ter	Avasthi 2019
LMOD3	5	Actin nucleator	c.366delG, p.Lys122AsnFs*6	Marguet 2020
			c.882dupA, p.Asp295Argfs*2	Michael 2018
			c.1069G>T, p.Glu357*	Michael 2018
			c.1628G>T, p.Arg543Leu	Marguet 2020
			p.Glu121ArgfsTer5	Berkenstadt 2018
			p.L245del	Berkenstadt 2018
MYO18B	2	Unconventional myosin, function not well characterized	c.6496G>T; p.Glu2166*	Malfatti 2015
NEB1	24	Structural component of sarcomere, binds actin	c.300dup, p.Tyr101fs*5 int49	Malfatti 2014
			c.3458 + 1G>A	Kapoor 2012
			c.1825dup, p.Pro541Profs*2	Lehtokari 2011
			c.2414 + 5G>A	Lehtokari 2011
			c.5060G > A, p.W1687X	Kiiski 2015
			c.5343 + 5G>A, p.Arg1747_Thr1778del	Malfatti 2014
			c.5574C > G, p.Tyr1858Stop	Malfatti 2014
			c.6496-G > A, p.2166_2234del	Malfatti 2014
			c.7432 + 1916_7535 + 372del, p.Arg2478_ Asp2512del	Malfatti 2014
			c.11086A>C, p.Thr3696Pro	Moureau-Le Lan 2018
			c.13066delT, p.Tyr4356Thrfs*8 ex110	Malfatti 2014
			c.17535G > A, p.Glu5845Glu	Malfatti 2014
			c.17779_17780delTA, p.Tyr5927HisfsX17	Moureau-Le Lan 2018
			c.18676C>T, p.Gln6226*	Malfatti 2014
			c.19101 + 5G > A, p.Leu6333_Glu6367del	Malfatti 2014
			c.20928G>T; p.Gly6976Gly	Malfatti 2014
			c.21076CC>T, p.Arg7026Ter	Moureau-Le Lan 2018
			c.21796_21810delinsT, p.Pro7266fs*30	Moureau-Le Lan 2018
			c.22273del, p.Val7425Serfs49*	Malfatti 2014
			c.22591–3C>G, p.7531Val_ Ser7564del	Malfatti 2014
			c.2310 + 5G>A, c.17779_17780delTA, p.His738_Asp770del	Moureau-Le Lan 2018
			c.23420_23421del, p.Arg7807Serfs*16	Malfatti 2014
			c.24250_24253dupGTCA, p.T8085fsX8100	Gajda 2015
			c.24269del, p.Arg8090fs*54	Malfatti 2014
			c.24372_24375dupAAGA, p.Val8126fs	Scoto 2013
			c.24440_24441insGTCA, p.Pro8148Serfs*15	Malfatti 2014
			c.24527_24528delCT, p.P8176fsX8179	Gajda 2015
			c.24579G>A, p.Ser8193Ser ex119	Malfatti 2014
			c.24686_ 24687del, p.Glu8229Glufs*18	Malfatti 2014
			c.24735_ 24736del(AG), p.Arg8245fs*1	Malfatti 2014
			c.163689G>T, GAG>UGA	Kapoor 2012
			hg19 chr1:g.(154,156,325_154,156,028) _(154,173,059_154,177,712)	Kiiski 2015
RYR1	2	Intracellular calcium channels at the sarcoplasmic reticulum	c.4455–4G>A c.4718C>T, p.Pro1573Leu c.7585G>A, p.Asp2529Asn	Tiberi 2020 Kondo 2012 Kondo 2012
TNNT1	15	Muscle contraction regulator	c.309 + 1G>A	van der Pol 2014
			c.323C>G, p.Ser108X	Marra 2015
			c.574_577delinsTAGTGCTGT	Abdulhaq 2015
			c.661G>T, p.Glu221X	D’Amico 2019
			arr[GRCh37] 19q13.42(55652193_55663445)x0	Streff 2019
TPM2	2	Muscle contraction regulator	c.415_417delGAG, p.E139del	Citirak 2014
Unknown	13

NM symptoms may present at any point during fetal development to adulthood. Typically, NM presents similarly to other congenital myopathies: hypotonia, muscle weakness, and decreased or absent myostatic stretch reflexes. Defects are observed in the musculoskeletal system, such as fractures, kyphoscoliosis, joint contractures, skull enlargement, pectum excavatum, and pes planus. Creatine kinase levels are normal, although this is also the case in other myopathies. Many patients are born with characteristic myopathic facies affecting mainly the mouth, resulting in a high-arched palate and drooling; occasionally patients present with ophthalmoplegia, which is rarely seen in other musculoskeletal conditions. Complications in the respiratory system are common.²² In rare cases, cardiac and renal systems may also be affected secondary to the musculoskeletal defects.^23–25 The differential diagnosis for patients with severe hypotonia and bulbar weakness is broad, but at birth includes myotonic dystrophy type 1, congenital myasthenic syndromes, spinal muscular atrophy type 0/1, mitochondrial myopathies, and Pompe disease.^26,27

The goal of this systematic review was to synthesize latest published findings in pediatric patients with NM as an initial step toward describing genotype-phenotype descriptions. Our examination of case reports published between 2010 and 2020 details novel genetic findings, newly described clinical presentations, and clinical considerations for the care of pediatric NM patients.

Methods

A systematic search of Pubmed, Embase, CINAHL, Web of Science, and Scopus was performed using Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.²⁸ The search strategy used the keywords (pediatric* OR child*) AND (NM OR nemaline rod OR rod myopathy). This protocol was not registered. Included case series or studies were published between January 1, 2010, and December 31, 2020; originally published in English; had pediatric patients (<21 years old); and focused on NM as the main topic. The time period selected limited our search to publications with the most recent peer-reviewed findings on pediatric NM. Reviews, meta-analyses, book chapters, and abstracts were excluded. Selected references were exported to Sciwheel, at which point all duplicates were removed.²⁹ Titles were read for concordance with inclusion criteria as a screen, after which abstracts were read to select for eligible articles. Full text of all eligible articles was read, and any that did not fulfill the inclusion criteria were removed from consideration. These full-text articles fulfilled the CARE guidelines for case reports.³⁰ Data were extracted and tabulated about age of first signs, earliest presenting neuromuscular signs and symptoms, accompanying body systems affected, progression, death, pathologic description, genetic changes, and consanguinity. These data were synthesized with attention to the recommendations of “Methodological Quality and Synthesis of Case Series and Case Reports.”³¹ Extracted data are available on request.

Results

The database search yielded a total of 385 records (Figure 3). Of the 69 eligible articles, 55 case reports or series of 101 pediatric patients from 23 countries were selected for this review (Supplementary Material). By genetic mutation, reports implicated ACTA1, KLHL40, NEB, and TNNT1, in addition to other genes with a minor contribution (Table 1).

Figure 3.

Study selection process using PRISMA flow diagram.

Genetics

Improved clinical availability of next-generation sequencing has increased identification of novel mutations associated with pediatric NM, especially exome sequencing.³² The 88 cases covered by this review with available genetic information implicate several variants as causative of NM: 24% deletions and 6% duplications. These resulted in 25% missense mutations and 9% frameshift mutations. These changes were predicted to cause protein truncations in 19% of the cases. NM mutations were found to be de novo in 11% of cases. Further examination of the sequencing data revealed that 30% of the patients were heterozygous for the mutated gene, 23% were homozygous, and 7% were compound heterozygous.

NEB and ACTA1 have been identified as the most commonly mutated genes in NM. Of the pediatric cases examined for this review, 25% had mutations in NEB whereas 22% in ACTA1. Consistent with previous literature, many of the NEB mutations were frameshifts and present in compound heterozygous combinations.³³ Fifty percent of the ACTA1 mutations were missense, and 45% were reported as de novo in the proband. For 2 patients, the ACTA1 mutations were predicted to produce a truncated protein or affect the binding site for actin’s interactors. Three cases with ACTA1 mutations had parents with gonadal mosaicism, which had not been previously reported, whereas one case showed somatic mosaicism.^34–36 Identifying mosaic mutations required additional in-depth sequencing because of being very-low-grade.

New mutations were found in less common causative genes, such as CFL2, KLHL40, RYR, TNNT1, and TPM3 (Table 1). New variants in TNNT1 were documented in non-Amish populations.^37–39 van der Pol et al predict that the combination of splice site mutations and a deletion in TNNT1 would produce short in-frame transcripts that would be targeted by nonsense-mediated decay.³⁹ One family harbored a novel 4-basepair deletion in CFL2 likely causing a premature stop codon in the protein.⁴⁰ Three patients from 2 unrelated families described by Fattori et al had mutations that may cause misfolding, lack of actin-binding, and protein degradation of cofilin 2.^40,41 In southern China, researchers identified a KLHL40 founder mutation that inherited both homozygous and heterozygous, which often leads to a truncation of the protein.⁴² Several new mutations in tropomyosin genes, including TPM3, have also been identified.^43,44 RYR mutations have been associated with various myopathies, but rarely with NM. Massively parallel sequencing in one patient in this review brought to light a new mutation in RYR1, causing NM with ophthalmoplegia.⁴⁵

Histopathologic Features

Pathology information was available for 64% of the patients reviewed (Table 2). Of these, 75% confirmed presence of cytoplasmic nemaline rods or bodies. Curiously, one case report about a patient with an LMOD3 mutation noted that the rods were surrounded by a filamentous halo.⁴⁶ Eight percent of patients had intranuclear rods; 4 of the 5 patients with intranuclear rods had a confirmed mutation in ACTA1, with the fifth not having any information about genetic causes. Internalized nuclei (found in 9%) were seen in patients with different mutated genes, including ACTA1, CFL2, and TNNT1.^38,41,44,47

Table 2.

Main Clinicopathologic Features Reported.

	n (%)	Associated genes
Age at first reported signs
Fetal	31 (31)	ACTA1, KLHL40, LMOD3, NEB1, RYR1, TNNT3, TNNT1
Birth	38 (38)	ACTA1, CFL2, KLHL40, MYO18B, NEB1, TNNT1, TPM2
Infancy (<1 y)	13 (13)	CFL2, LMOD3, NEB1, TNNT1
1–13 y	13 (13)	ACTA1, CFL2, NEB
Early signs
Decreased fetal movements	9 (9)	ACTA1, KLHL40
Locked-in state	1 (1)	KLHL40
Polyhydramnios	14 (14)	ACTA1, KLHL40, LMOD3, NEB1, RYR1
Neonatal hypotonia	64 (64)	ACTA1, CFL2, KLHL40, LMOD3, MYO18B, NEB1, RYR1, TNNT1, TNNT3
Delayed motor development	29 (29)	ACTA1, CFL2, KLHL40, LMOD3, NEB1, RYR1, TNNT1, TPM2
Abnormal gait/frequent falls	10 (10)	ACTA1, CFL2, KLHL40, LMOD3, NEB1, TPM2
Muscle weakness
Unspecified	22 (22)	ACTA1, CFL2, KLHL40, LMOD3, NEB1, TNNT1
Axial and proximal	31 (31)	ACTA1, CFL2, KLHL40, LMOD3, MYO18B, NEB1, TNNT1, TNNT3, TPM2
Distal	23 (23)	ACTA1, CFL2, KLHL40, LMOD3, MYO18B, NEB1, TNNT1
Tremor	7 (7)	TNNT1
Feeding difficulty	16 (16)	ACTA1, CFL2, KLHL40, MYO18B, NEB1, TNNT3
Early respiratory difficulty	36 (36)	ACTA1, CFL2, KLHL40, LMOD3, NEB1, RYR1
Spinal curvature	37 (37)	ACTA1, CFL2, LMOD3, NEB1, RYR1, TNNT1, TNNT3
Scoliosis	15 (15)	ACTA1, CFL2, LMOD3, NEB1, RYR1, TNNT1, TNNT3
Kyphosis	11 (11)	ACTA1, CFL2, LMOD3, TNNT1
Rigid spine	5 (5)	ACTA1, LMOD3, NEB1, TNNT1
Lordosis	5 (5)	ACTA1, CFL2, LMOD3, NEB1
Facial involvement	55 (55)
Facial weakness	29 (53)	ACTA1, CFL2, NEB1, RYR1, TNNT3, TPM2
Ptosis	6 (11)	KLHL40, LMOD3, NEB1, TPM2
Ophthalmoplegia	1 (2)	RYR1
Facial dysmorphias	29 (53)
High-arched palate	23 (42)	ACTA1, CFL2, LMOD3, MYO18B, NEB1, RYR1, TNNT1, TNNT3, TPM2
Micrognathia	4 (7)	ACTA1, NEB1
Cleft palate/lip	2 (4)	KLHL40
Myopathic facies	9 (16)	ACTA1, KLHL40, LMOD3, TNNT1
Elongated face	7 (13)	ACTA1, NEB, RYR1
Macrocephaly	2 (4)	NEB
Dysmorphic features	3 (5)	NEB
Thoracic deformities
Pectus excavatum	2 (2)	ACTA1, MYO18B
Pectus carinatum	5 (5)	TNNT1
Respiratory	57 (56)	ACTA1, CFL2, KLHL40, MYO18B, NEB1, RYR1, TNNT1
Tracheotomy	15 (27)	ACTA1, NEB1, RYR1, TNNT1
Mechanical ventilation	30 (54)	ACTA1, LMOD3, NEB1, TNNT1, RYR1
Respiratory insufficiency	6 (11)	ACTA1, CFL2, NEB1
Sleep apnea	2 (4)	NEB1
Pleural effusion, chylothorax	3 (5)	ACTA1, KLHL40
Cardiac	13 (13)
Hypertrophy	2 (15)	MYO18B
Ventricular dilatation	4 (31)	ACTA1, MYO18B, TNNT1
Sudden cardiac arrest	2 (15)	NEB1
Atrioseptal defect	2 (15)	LMOD3
Cardiomegaly	1 (8)
Bradycardia	1 (8)	RYR1
Transient supraventricular tachycardia	1 (8)	MYO18B
Joints/skeletal	39 (39)
Contractures	21 (54)	ACTA1, CFL2, KLHL40, LMOD3, NEB1, RYR1, TNNT1
Arthrogryposis	10 (26)	ACTA1, LMOD3, NEB1
Club feet	7 (18)	KLHL40, LMOD3, NEB1, TNNT1
Fractures	3 (8)	KLHL40
Hip hyperlaxity	3 (8)	KLHL40, NEB1, TNNT3
Gastrointestinal involvement	3 (3)	ACTA1, TNNT1
Neurologic
Intellectual disability	1 (1)	NEB1
Decreased white matter	3 (3)	ACTA1
Wheelchair bound	7 (7)	ACTA1, CFL2, KLHL40, NEB1
Death	36 (36)
0–1 mo	5 (14)	ACTA1, NEB1
1–6 mo	12 (33)	ACTA1, KLHL40, LMOD3, MYO18B, NEB1, RYR1
6 mo-1 y	3 (8)	ACTA1, KLHL40, NEB1, TNNT3,
1–5 y	7 (19)	KLHL40, NEB1, TNNT1
5–10 y	5 (14)	ACTA1, NEB1, TNNT1
>10 y	5 (14)	NEB1, TNNT1
Cause of death
Sepsis	3 (8)	ACTA1, KLHL40, MYO18B
Respiratory insufficiency	3 (8)	ACTA1, TNNT3
Cardiopulmonary arrest	3 (8)	ACTA1, KLHL40, RYR1
Infection	2 (6)	KLHL40, LMOD3,
Hypoxic-ischemic brain injury	1 (3)
Terminated pregnancy	4 (4)	KLHL40, LMOD3
Pathology features	65 (64)
Cytoplasmic nemaline rods	49 (75)
Intranuclear rods	5 (8)	ACTA1
Fiber size variation	26 (40)	ACTA1, CFL2, LMOD3, MYO18B, NEB1, TNNT1
Internalized nuclei	6 (9)	ACTA1, CFL2, MYO18B, TNNT1
Increased fibrous connective tissue	18 (28)	ACTA1, CFL2, TNNT1, TNNT3
Inflammatory cell infiltration	3 (5)	ACTA1, TNNT1
Atrophic fibers	8 (12)	ACTA1, MYO18B, TNNT3
Fingerprint bodies	1 (2)	LMOD3
Rods surrounded by halos	1 (2)	LMOD3
Mitochondrial aggregates	1 (2)	NEB1
Myofibrillar degradation	3 (5)	ACTA1, CFL2, KLHL40

Other common features included fiber size variation (40%) and increased endomysial fibrous connective tissue (28%). Five percent of cases described some form of inflammatory cell infiltration. One study found that 2 patients had necrotic muscle fibers with T cells that stained positive for CD4 and negative to CD8.⁴⁸ Atrophic fibers were found in 12% of patients with no correlation to age of first symptoms or death at the time of case report. Detailed analysis of patients with NEB mutations categorized them into 3 groups by time of symptom onset, which correlated with differences on histology such as differences in the amount of sarcomeric dissociation, rod pattern, and fiber type.⁴⁹

Spectrum of Disease

The selected case reports described a variety of clinical presentations, some with symptoms in organ systems beyond the typical muscle weakness of NM (Table 2). The first reported signs often came as early as fetal development for 9% of cases, with a decrease in fetal movements or polyhydramnios on antenatal ultrasonography, typically during the second or third trimesters. The most common early sign of disease was neonatal hypotonia, seen in 64% of children, followed by early respiratory difficulty (36%) and spinal curvature (37%). More than half of patients had respiratory complications throughout childhood, including respiratory failure as evidenced by placement of tracheostomy. Mutations in MYO18B, TNNT1, and ACTA1 were seen in some cases that involved cardiomyopathy.^17,50–52 One patient presented early with left-sided hypertrophy causing pulmonary hypertension and right ventricular dilatation.¹⁷ Dilated cardiomyopathy was identified in 2 patients with ACTA1 mutations; this was paired with dyskinesia of the left ventricle in one child.^34,51

Muscle weakness presented differently across patients. Pattern of weakness was axial and proximal in 31% of cases, whereas distal involvement was reported for 23%. Facial weakness was noted for just under one-third of children, although a broad spectrum of facial dysmorphic patterns was described. Some patients are born with significant craniofacial deformities, including cleft lip, atrophy of facial and masticator muscles, and jaw deformity, which can impact the patient’s ability to close their mouth.⁵³ Joint contractures and arthrogryposis were common, described in almost 40% of patients. Another possible presenting symptom is hyperextension of the neck, for which Tiberi et al recommend neurologic examination after birth to check for possible muscle disorder such as NM.⁵⁴ One patient with decreased movement in utero was found to have very fragile ribs on postnatal chest radiograph.²⁶

The majority of patients exhibiting early lethality (36% of patients were deceased) were diseased within the first year of life; causes of death included sepsis, respiratory insufficiency, cardiopulmonary arrest, infection, or hypoxic-ischemic brain injury.

Patients with NM typically have normal intelligence, but Saito et al documented patients with ACTA1 dominant missense mutations with delays on word comprehension testing that “may result from abnormal development of the central nervous system, not only from hypoxic events or limited social experiences.”⁵⁵ All 3 patients had decreased white matter volume, enlargement of the lateral ventricles, and frontal lobe hypoplasia.

Clinical severity varied significantly, even among patients with mutations in the same gene. Some with ACTA1 mutations were reported to have mild phenotypes, whereas one patient had dysautonomia in addition to myopathic symptoms.^56–58 Anticipation was seen in a family harboring an ACTA1 missense mutation, where the mother had some weakness while her child had myopathic facies, high-arched palate, lumbar hyperlordosis, and delayed motor milestones.⁵⁹ One child with an ACTA1 mutation of Thai ancestry had primary pulmonary lymphangiectasia on histology, in addition to pleural adhesions and a choroid plexus papilloma causing communicating hydrocephalus.⁴⁷ For KLHL40 mutations, some patients had moderate presentation whereas others experienced fetal akinesia or a total locked-in state because of the severity of their disease.^60–62 LMOD3 mutations also could present prenatally as decreased fetal movements, polyhydramnios, and arthrogryposis, or later in childhood, with a milder phenotype or even disease progression.^12,46,63 Mutations in TNNT1 were reported in patients of varied ethnic backgrounds, including a case series of Amish patients presenting with the prototypical progressive muscle weakness, contractures, and tremors; of Italian siblings with severe failure to thrive and rigid spine; and of a Palestinian cohort that showed signs of transient tremors, progressive spinal rigidity, and limb contractures.^64–66

Clinical Care Considerations

Care for patients with NM should follow standard multidisciplinary supportive care as reviewed in the Consensus Statement on Standard of Care for Congenital Myopathies.⁶⁷ Biopsy is critical to the diagnosis of NM; however, selection of the biopsy location may introduce sampling error. A potential advance was discussed by one paper that showed the potential of using imaging—magnetic resonance imaging, in particular—to guide the decision of what muscle to biopsy.²⁷ Better understanding of the genetic underpinnings of NM may allow for making the diagnosis of NM without the need for biopsy.³²

NM patients require multifaceted care that addresses specific needs and prevention involving wide expertise from specialists and health care professions. The varied presentations of NM in some patients in this review bring to light several considerations for clinical practice.

Respiratory function is commonly affected in NM patients, as 36% of the patients described had early respiratory difficulty and 56% had neuromuscular respiratory failure or weakness. Some patients experienced recurrent respiratory infection, which is cause for concern given that respiratory insufficiency, cardiopulmonary arrest, infection, and sepsis were the majority of the reported causes of death across case reports. NM patients may also experience complications postprocedure such as pleurodesis, where pneumothorax can be treated using biphasic cuirass ventilation.⁶⁸ Monitoring sleep and pulmonary function over time is important to assess for nocturnal hypoventilation, obstructive sleep apnea, and restrictive lung disease. One patient underwent inspiratory muscle strength training for 2 weeks after major surgery. This short-term intervention increased her mean inspiratory pressure and allowed her to breathe unassisted for 11–13 hoursper day, compared with her medical historyof restrictive lung disease and recurrent ventilatory failure postoperation.⁶⁹

NM has variable impacts on the motor function of each patient and, therefore, care involves a combination of orthopedic, physical therapy, occupational/speech therapy, and neuropharmacologic agents. Speech therapy involving oral motor exercises, tongue strengthening exercises, and diadochokinesia exercises were used by one study team to help a patient with severe dysarthria improve the intelligibility of her speech.⁷⁰ Two case reports present rare instances where there is a role for neuropharmacologic agents in improving the life of patients with NM. The report by Sahin et al documented decreased drooling and spontaneous extremity movement from a child that was previously immobile after treatment with _L-tyrosine.⁷¹ For a child with a KLHL40 mutation presenting with myasthenic symptoms without antibodies directed against the acetylcholine receptor, use of the acetylcholinesterase inhibitor pyridostigmine greatly improved endurance and strength, which regressed when the medication was no longer administered.⁷²

Given that the histologic appearance of NM resembles that of core-rod myopathy, it is recommended that anesthesia be approached with caution in NM patients so as to mitigate any theoretical risk for malignant hyperthermia particularly if a mutation in RYR1 is present.⁷³ In a case of cleft palate repair, the team was able to avoid the use of succinylcholine by using propofol and fentanyl for induction, intubation, and adjustment of IV access, with rocuronium for facial muscle relaxation.⁷⁴ Intubation itself can pose challenges because of the facial dysmorphias of some patients.⁷⁵

For children, follow-up care should occur at regular intervals, which should be more frequent (3–4 months) for infants as opposed to older children (6–12 months). The majority of the deceased patients reviewed passed away during the first year of life, suggesting that the timing of visits for children with early-onset NM symptoms may need to occur more frequently.

Discussion

NM is a myopathy that can range from severe to mild with weakness manifesting at birth. Defects of sarcomere components or its regulators result in actin aggregates within the muscle, seen easily on electron microscopy of muscle biopsy. Several important knowledge gaps remain, including how the disease affects children across all ages. The articles surveyed in this review deepen the understanding of pediatric NM patients and provide clues that may aid in identifying subtle cases after careful evaluation.

The continued improvement in sequencing and its increased accessibility makes the possibility of identifying causative genes more feasible, including for those that produce larger proteins such as nebulin. Sequencing results confirm the heterogeneity of genes that cause NM, and several new variants were identified in pediatric NM patients, which expand our understanding of how it is inherited. The majority of patients reviewed had mutations in NEB and ACTA1 as has been previously reported. New examples of gonadal mosaicism leading to inheritance were identified in ACTA1 patients by using more sophisticated sequencing modalities. Patients from non-Amish populations were found to harbor TNNT1 mutations different from the Amish founder mutation, which resulted in a different phenotypic picture with failure to thrive and spinal rigidity rather than only progressive muscle weakness, contractures, and tremors. Many of the case reports implicated nonsense-mediated decay as the mechanism by which translation of faulty transcript of NM genes is reduced, in particular for recessive form of the disease.

Increased information about causative genes and mutations provides the opportunity to incorporate genetics into the diagnostic process. Looking forward, as the genetic profile of NM further develops, it may be possible to shift away from a reliance on muscle biopsy given that this procedure is invasive and can cause discomfort for a child. However, much remains to be discovered since there are cases where a mutated gene has not yet been identified.^{26,48,52,53,58,68,69,70,71} Sequencing is not as accessible across the globe and may pose financial hardship in some cases; thus, it is unlikely to become the sole source of diagnostic information in the near future.

The cases described not only the expected presence of nemaline rods on histology but also details about fiber size variation, filaments surrounding the rods, and inflammatory cell infiltration. These subtleties require further study to identify if these findings correlate with genotype or clinical severity.

The clinical features presented mirror many of the existing descriptions of NM, yet they highlight additional opportunities for clinical intervention to improve the lives of NM patients. A number of cases listed decreased fetal movements or antenatal ultrasonographic findings such as polyhydramnios as early signs; although not specific enough to be diagnostic, these may serve to document the need for a neuromuscular examination at birth. Subtle findings at birth such as hyperextension of the neck should also lead the clinician to consider a more thorough neurologic examination.⁵⁴ Given that early respiratory difficulty was cited in more than one-third of patients, more needs to be done to identify ways to provide respiratory support for neonates with suspected NM with special attention to the various craniofacial deformities found in some patients. Respiratory health is critical, considering the number of patients described as experiencing recurrent infection and dying from pulmonary complications particularly in the first year of life.

Little was mentioned about management of spinal curvature despite almost 40% of patients experiencing this complication. Among these cases, there were examples of both dilated and hypertrophic cardiomyopathy in patients with mutations in ACTA1, TNNT1, and MYO18B. Cardiomyopathy had been reported in ACTA1 and Amish TNNT1 patients previously; however, this report of a pediatric patient with a mutation MYO18B was one of the first that showed an association with hypertrophy.¹⁷ The case of MYO18B also highlight that cardiac phenotypes seen in model organisms may serve to identify genes that may later be found to be important in skeletal muscle and be linked to NM pathology. The article by Saito et al⁵⁵ introduces the point that normal intelligence may not be common to all patients with NM and therefore emphasizes the need for more detailed analysis of the pediatric brain. Lastly, these cases collectively underscore how presentation is not consistent across patients with mutations in the same gene; future studies examining patients with different mutations in the same gene should carefully track the precise symptoms that may distinguish severity.

To compile the most recent clinical reports of NM, this review is limited by the findings shared in the literature. Therefore, the trends identified in the results are based on the information explicitly described in the case reports, which varied in their detail (Table 2). The lack of histologic or genetic results for some patients reduces the correlations that can be made between genotype and phenotype. Studies that include a large cohort of patients with NM with different genetic mutations are needed to identify such patterns, especially if sequencing and testing are done in a comparable manner. These studies will allow for more robust correlation between genotype and phenotype that may be useful in predicting prognosis.

Greater interplay between clinical and basic science findings will be critical in developing clinical prevention and treatment strategies. Future studies may consider tests, procedures, and treatments for NM. For example, further study should examine the value of imaging, as suggested by Ennis et al.²⁷ The value of L-tyrosine as treatment should also be considered, although there have been mixed results in human and animal studies. Ryan et al describe improvement in bulbar function, activity level, and exercise tolerance in 4 patients who received L-tyrosine supplementation; however, the underlying genetic change was identified in only 1 patient in TPM3.⁷⁶ A study in mice with a mutation in ACTA1 showed similar improvements in mobility in addition to showing a decrease in rods and degenerating fibers on muscle biopsy.⁷⁷ However, work in zebrafish models of nebulin NM have not shown these changes after administration of _L-tyrosine or other supplements.⁷⁸ These findings are similar to a study using 2 established mouse models and one zebrafish model with ACTA1 mutations.⁷⁹ Continued genetic studies are needed to characterize the responsible genes that cause other forms of NM. More reports sharing best practices in improving quality of life for NM patients would benefit the field, particularly related to respiratory and orthopedic care that would benefit many of the patients reviewed.

Overall, this review demonstrates the progress that has been, and is yet to be, made in the diagnosis and mechanistic understanding of NM in children. More research, particularly into the importance of sarcomeric structures and regulatory proteins in maintaining healthy muscle, will be needed. These concepts will be critical to establishing best practices and standards for diagnosis and treatment in pediatric NM.