Clinical manifestations

Presentation and course

Typical manifestations. Symptoms in most patients become evident in late childhood or adolescence. Protruding scapulae (winging) may be noted by the caregivers of the child, especially if they had previously seen other affected family members. The next manifestation is likely to be difficulty raising the arms, often asymmetrically. Difficulty lifting objects may follow, but the arms are likely to maintain function for many years. Sometimes, the distortion of shoulder girdle structures makes it difficult for the person to be fitted for shirts, blouses, or jackets. The disorder progresses slowly to include truncal muscles as well as lower extremity muscles. If there is leg weakness, it is apt to be initially more distal than proximal, so there is a tendency for a foot-drop gait. This gives rise to a scapuloperoneal distribution of limb weakness. Facial weakness may be subjectively asymptomatic, with many individuals retrospectively stating that they have never been able to whistle or suck from a straw. Most dramatically and rarely, some patients may sleep with their eyes open.

With the exception of infantile-onset facioscapulohumeral muscular dystrophy, facial weakness is not usually severe early on in children, but many affected individuals have never been able to whistle. Most dramatically, but uncommonly, some patients tend to sleep with their eyes open.

Signs on examination parallel the distribution of weakness. Weakness of eye closure is sometimes manifested by a wide-eyed appearance or inability to close the eyes against resistance. The lips tend to be full or even slightly everted. There is also loss of wrinkles overlying the affected facial muscles. The most dramatic set of signs is caused by inability to fix the scapulae. At rest, the scapulae deviate laterally. There may be winging of the scapulae with the arms dependent, in attempts to raise the arms to shoulder level by abducting the arms, or in pushing against a wall with the hands at shoulder level and elbows straight. Viewed from in front of the patient, the tips of the scapulae rise above the clavicle, presenting a typical facioscapulohumeral appearance. The distinctive picture may be augmented by poorly developed pectoral muscles or frank pectus excavatum, so that the chest seems to be caved-in. If the pectorals are severely wasted, the anterior axillary fold, which is formed by the pectorals and normally runs at a diagonal from the chest to the head of the humerus, is lost, and the fold is either vertical or even reversed.

As a result of the scapular disorder, the arms cannot be raised to shoulder level even before there is detectable weakness of the supraspinati, infraspinati, or deltoids on manual muscle tests. The unusual pattern of bumps and depressions from protruding bones and preserved muscle alternating with small muscles has been called the “poly-hill” sign (64).

Weakness of muscles in the anterior compartment of the legs may be detected even before there is an abnormality of gait. Tendon reflexes may be either preserved or absent in weaker individuals.

The abdominal muscles are often affected, resulting in a protuberant abdomen and difficulty sitting up from a supine position. The Beevor sign, which is upward movement of the umbilicus on flexion of the neck, is attributed to selective weakness of the lower part of the rectus abdominus so that the upper fibers are stronger and pull the umbilicus upward (16). The sign was present in 15 of 28 patients with FSHD, as compared to two of 65 patients with other myopathies. Given its specificity in FSHD, a positive Beevor sign can be a useful early diagnostic sign in patients without overt abdominal muscle weakness.

Spectrum of disease manifestations. Early-onset FSHD was reported to occur in up to 21% of patients in one review (25). Early-onset was defined as age of onset of facial weakness (before 5) and shoulder girdle weakness (before 10). These patients can present with more severe muscle weakness and more rapid progression. Approximately half the patients will have systemic features (including hearing loss, retinal abnormalities, respiratory insufficiency, and developmental delay). De novo/sporadic cases are more likely in this population (25). However, the classification of early onset FSHD was, in many cases, based on recall in adults with FSHD, which likely introduced a selection bias overestimating the true incidence of early onset FSHD. In a more unbiased nationwide, prospective, cross-sectional study, patients less than 17 years old resembled classic FSHD in presentation and genetics (23). Furthermore, long-term follow-up of these patients also suggest a heterogeneous disease course (24), with mild and slow progression of facial weakness, pain, fatigue, and lumbar hyperlordosis after two years (13).

In addition to age of onset, disease severity is broadly influenced by the number of the residual D4Z4 repeats on chromosome 4 (see genetics).

Similarly, adult-onset forms may also be severely affected; in some series, about 20% of adults over the age of 50 lose the ability to walk and are in wheelchairs (60).

Many affected individuals in a family are asymptomatic, but show unequivocal signs of the dystrophy on examination (97). About 30% of all familial cases are asymptomatic. If the parents of an affected person are clinically normal and do not show the typical molecular changes associated with FSHD, and if more than one sibling is affected, a de novo mutation in the germline is suspected (37). Alternatively, one of the parents is a somatic mosaic for the FSHD mutation with minimal or no symptoms of FSHD. At least 12% of familial cases have normal neurologic examination, have no symptoms, and are nonpenetrant FSHD1 carriers (105).

Hearing loss and retinal vascular disease are associated with facioscapulohumeral muscular dystrophy, but they are rarely symptomatic except in the more severely affected individuals, almost exclusively most commonly in childhood-onset facioscapulohumeral muscular dystrophy (05; 58). Studies confirm that symptomatic retinal vasculopathy and hearing loss happen almost exclusively and occur in individuals with large deletions, leaving only one to three residual repeats (83). Similarly, hearing loss is also largely restricted to patients with one to three repeats (49). To that extent, FSHD is pleiotropic; that is, the genetic abnormality affects multiple organ systems in ways not understood. In adult-onset disease, however, hearing loss is not more common in people with FSHD than in the general population (69; 93; 94). Mental retardation and epilepsy are exceedingly rare (93; 28; 13).

Calf hypertrophy is distinctly unusual but has been recorded (57; 34). Respiratory failure is also infrequent but should be considered in more severely affected individuals (54; 07; 109; 79; 89). More commonly, patients have a restrictive ventilatory pattern, with up to 45% with that pattern on pulmonary function studies (89). Cardiac involvement is uncommon and consists of mainly conduction abnormalities that are usually benign (40; 19; 92; 39; 14). Dysphagia is rare, even in advanced disease (109). As discussed below, carriers of the mutation may diverge from the consensus criteria for diagnosis. Also, homozygosity for the 4q35 deletion does not result in a more severe phenotype (90).

Clinical diagnostic criteria. An international workshop developed detailed diagnostic criteria for FSHD (59), which remain largely applicable (88). The major criteria that define FSHD are onset in facial or periscapular muscles sparing ocular, pharyngeal, and lingual muscles. Supporting clinical evidence of FSHD include:

The presence of severe contracture, cardiomyopathy, or symptomatic respiratory failure at the time of initial evaluation should suggest alternative diagnoses. With the availability of highly sensitive and specific genetic testing, a muscle biopsy should be performed in individuals to look for an alternative diagnosis only when genetic testing is negative and an alternative diagnosis amenable to pathologic diagnosis is being considered (56). In facioscapulohumeral muscular dystrophy, EMG and muscle biopsy usually show nonspecific myopathic patterns, and serum creatine kinase values are normal or modestly elevated (not more than five times the normal limit).

Despite the fairly typical clinical picture in FSHD, with onset in the face and shoulders followed by humeral and then lower extremity involvement later, the presenting symptoms and signs can vary greatly.

Genetic diagnosis. FSHD, in its most common form, results from a partial deletion of a large repetitive DNA element called a macrosatellite repeat. This form of FSHD, now referred to as FSHD1, represents about 95% of patients with FSHD (68). The FSHD-associated repetitive element is known as D4Z4 and is present in the subtelomeric region of the long arm of chromosome (4q35). Each D4Z4 repeat is 3.3 kb in size, and normal individuals typically have 11 to 100 D4Z4 repeats on each copy of chromosome 4q35. In FSHD1, one copy of 4q35 will have one to 10 repeats. The genetic test uses a probe that recognizes a sequence just proximal to the repeats. Subsequent digestion with the EcoRI restriction enzyme yields a DNA fragment that spans the D4Z4 array. Consequently, the fragment size reported on commercial FSHD genetic testing will vary with the number of D4Z4 repeats. Thus, normal alleles having more than 10 repeats are greater than 38 kb in size, whereas an FSHD-associated allele with one to 10 residual repeats is between 10 and 38 kb in size (56).

As mentioned above, individuals with large deletions resulting in a contracted allele have the most severe clinical manifestations. Individuals with one to two repeats tend to include patients with early onset FSHD. Patients with three to six D4Z4 repeat units are more affected than patients with seven to nine D4Z4 repeat units and are more likely to have facial involvement (53). Standard genetic testing for FSHD1 is both highly sensitive (93%) and specific (98%) (86).

About 5% of patients with clinically typical facioscapulohumeral muscular dystrophy do not have D4Z4 contractions but have severe DNA hypomethylation of both copies of 4q35 (10). This hypomethylation of 4q35 is now known to be the result of mutations in the SMCHD1 gene, a gene involved in X chromosome inactivation (42) and also associated with arhinia or Bosma arhinia microphthalmia syndrome (51) or DNA methyltransferase 3B (DNMT3B). Monoallelic DNMT3B mutation can cause facioscapulohumeral muscular dystrophy phenotype, whereas biallelic mutations cause autosomal recessive immunodeficiency, centromeric instability, and facial anomalies (ICF syndrome) (100). Ligand‐dependent nuclear receptor interacting factor 1 (LRIF1; also known as HBiX1) is associated with D4Z4 chromatin relaxation, and a carrier of homozygous mutation of this gene has been reported to have facioscapulohumeral muscular dystrophy phenotype (29). This form of facioscapulohumeral muscular dystrophy (FSHD2) is clinically identical to FSHD1. Genetic testing for FSHD1 and FSHD2 can be done in a hierarchical manner commercially.

Prognosis and complications

The disease is slowly progressive and does not shorten mortality. In a survey, 61 of 126 affected adults were currently working, only nine needed adjustments for disease-related handicaps, and 85% considered their jobs satisfying (106).

Medical complications exist. Twenty percent of patients above the age of 50 will eventually use a wheelchair. Several studies have examined pulmonary function in patients with facioscapulohumeral muscular dystrophy. Up to approximately one-third of patients or nonambulant patients may suffer from respiratory insufficiency (52; 110). The pattern on pulmonary function testing is mainly restrictive (79; 52; 110; 89). Patients with restrictive lung disease can be asymptomatic to requiring the use of noninvasive ventilation. In a single center study, the frequency of use of noninvasive ventilation was reported to be 14% (52). However, this estimate may be too high as another nationwide Dutch study found that only 1% of patients with FSHD had clinically significant respiratory failure requiring noninvasive or invasive ventilator support (112). Predisposing factors for restrictive lung disease include age of onset, D4Z4 fragment length, wheelchair dependence, pelvic girdle weakness, (kypho-)scoliosis or lumbar lordosis, and pectus excavatum (112; 79; 52; 110). Patients with predisposing factors should have periodic pulmonary function testing. Sleep-disordered breathing including hypoventilation syndrome and obstructive sleep apnea is frequent and may be subclinical (52; 73). Spinal deformities are also common, in approximately 20% to 30% of patients; these are predominantly lumbar hyperlordosis or, less commonly, scoliosis, with childhood-onset cases being the most severe (57). Additionally, camptocormia due to weakness of paraspinal muscles can be seen in older patients. Patients can also complain of pain and difficulty sleeping with poor sleep quality (31; 36).

Severe childhood-onset FSHD typically have large deletions (ie, residual D4Z4 allele size of 10 to 18 kb) and are at higher risk of developing a potentially preventable exudative retinopathy also known as Coats disease; such patients need periodic surveillance with dilated indirect ophthalmoscopy (83). These same patients are also at risk for symptomatic hearing loss and should have routine audiometric testing. Moreover, hearing loss can be progressive, and children with large deletions should continue to be monitored even if they passed newborn hearing screening (49).

Pregnancy is usually tolerated well, but some women later have more difficulty caring for children (72; 08).

Clinical vignette

A man with no known family history of neuromuscular disease was a varsity swimmer and speed skater in college, and he had been actively athletic throughout his adult years. He had several episodes of acute low back pain, and, at 70 years of age, had paresthesias in both legs on prolonged standing and then had difficulty standing erect, especially after walking. When he shaved, he bent forward and then had difficulty standing up. Lumbar laminectomy gave no benefit, and right hip replacement helped the leg discomfort but not the difficulty standing erect, so he began using a cane. In airports he had to use a wheelchair. His social life was constricted because he could not stand. He had difficulty raising his right arm and stopped playing tennis at 72 years of age. By the age of 75 he could not raise his right arm to fix his collar or comb his hair; similar problems on the left were milder. He had difficulty rising from low seats but could go up or down stairs. He had always had difficulty blowing up balloons and whistling but did close his eyes in sleep. Past history and system review were unremarkable. There was no consanguinity in the family, and his four sons, ages 41 to 49, were all asymptomatic.

Eye closure was weak against resistance. His lips were slightly everted, and his whistle was weak. Speech was clear. Neck flexion was weak against resistance. In sitting he supported his head with his right arm, supporting the arm by leaning on the elbow. There was not an overtly typical facioscapulohumeral appearance when he was viewed from the front, but the tips of the scapulae did protrude above the clavicle. Similarly, there was no typical scapular winging, but the left scapula protruded more than the right in testing the supraspinati or in pushing against the wall. He could not raise his arms to shoulder level, but they could be stretched overhead passively without any unusual resistance. The following shoulder muscles were all strong: supraspinati, infraspinati, trapezius, and rhomboids. Similarly, there was no weakness of the deltoids, biceps, triceps, or hand muscles. He could walk on heels or toes, and could rise from a standard chair without using his hands. There was no weakness of leg muscles on manual testing. As soon as he stood to walk, he bent forward from the trunk and his head began to drop. Tendon reflexes were absent. There was no focal wasting, no Hoffmann or Babinski signs, and no clonus. Sensation was normal.

EMG was a mixed pattern, with a predominantly myopathic pattern in proximal limb muscles but some neurogenic features distally. Nerve conduction studies gave normal results except for findings compatible with asymptomatic carpal tunnel median neuropathies. Serum creatine kinase level was normal.

DNA diagnosis (Athena). After EcoRI and EcoRI-BlnI restriction endonuclease digestion, the patient had a 34 kb allele 1 and greater than 48 kb allele 2. In normal people, allele size is greater than 42 kb; borderline values are 38 to 41 kb; FSHD deletions give result in values of 10 to 38 kb. The patient’s values repeat size were clearly in the FSHD range, but there was also evidence of a translocation with 10q26, which raised some uncertainty of interpretation. However, given the clinical picture compatible with the diagnosis (despite the late onset), the DNA evidence also favored the diagnosis of FSHD. Whether this was a new mutation could not be ascertained without clinical and DNA studies of his parents.

Comment. This case history is atypical in several respects, including late onset, lack of other symptomatic relatives, and symptomatic predominance as the bent spine syndrome. All, however, have been reported, and DNA analysis proved the diagnosis.

Biological basis

Etiology and pathogenesis

Genetics. FSHD1 is inherited as an autosomal dominant disorder; it is caused by a deletion of a critical number of D4Z4 repeats on one copy of 4q35. However, the contraction is necessary but not sufficient to cause FSHD1. The contraction has to occur on a 4q35 that is of the “A” sequence variant distal to the last repeat.

Most of the remaining patients (about 5%) with typical clinical features of facioscapulohumeral muscular dystrophy have FSHD2. Patients with FSHD2 have normal number of D4Z4 repeats on both copies of 4q35 but demonstrate profound DNA hypomethylation at the D4Z4 locus (10). DNA hypomethylation indicates a more open chromatin structure, and therefore, containing genes that are more likely to be expressed. This is caused by a gene locus different than the FSHD1 4q locus. In about 80% of cases, FSHD2 is associated with mutations in the SMCHD1 gene on chromosome 18p (42) and DMNT3B gene on chromosome 20 (100). Inheritance of FSHD2 is considered digenic. FSHD2 results from the inheritance of two independent genetic predispositions: a) mutation of the SMCHD1 or DMNT3B gene, and b) the presence of at least one “A” variant distal to the D4Z4 repeats. Although genetically distinct, both FSHD1 and FSHD2 trigger the same molecular mechanism.

The identification of FSHD2 has also exposed another modifier of FSHD disease severity beyond the number of residual repeats. The co-occurrence of both FSHD1 and FSHD2 was identified in a family (75). In this kindred, individuals with either FSHD1 or FSHD2 associated with SMCHD1 mutations had mild disease, whereas individuals with a short (FSHD1-sized) allele and an SMCHD1 mutation had much more severe disease. Other, as yet to be uncovered, gene modifiers could account for intrafamilial variability in disease severity.

There may be a 1% to 2% prevalence of a contracted D4Z4 repeat on a permissive 4qA haplotype without an FSHD family history (11; 78). These de novo genetic changes may not be symptomatic because of the large epigenetic contribution to this disease. Penetrance (17% to 53%) appears to be correlated with smaller repeat size, lower methylation status, and age (26; 91; 76; 67; 77; 111). The smaller the repeat size, the more penetrant the disease. However, nonpenetrant mutation carriers must be distinguished from asymptomatic mutation carriers; the latter complained of no symptoms, yet on exam have mild weakness.

Pathophysiology. The genetic lesion in FSHD is a loss of a critical number of a repetitive DNA element; therefore, there is, as such, no disruption or mutation of the sequence of an expressed gene. How a contraction of D4Z4 repeats results in FSHD remained unclear for a number of years. One critical piece of evidence was that haploinsufficiency for 4q35, deletion of one copy of the distal 4q35, does not result in FSHD. This suggested that at least one copy of D4Z4 repeat is necessary for FSHD and that the FSHD genetic defect results in a deleterious gain of function. This implied that part of the genetic sequence within repeat was essential to cause trigger disease (95). An open reading frame called DUX4, present in each D4Z4 copy, was suggested as a putative FSHD gene. However, DUX4 expression in either normal or FSHD tissue was difficult to demonstrate. Moreover, the DUX4 sequence lacked some of the components of an expressed gene, suggesting that it may be a pseudogene (30; 50; 21). Subsequently analysis of the chromatin structure also indicated that the contraction of the D4Z4 repeats changes the chromatin structure from a closed heterochromatic conformation to a more permissive euchromatic state, the latter more typically seen in expressed DNA sequences (101; 113). Finally, it was demonstrated that the subtelomeric region distal to the repeats comes in two variants or haplotypes: A or B. Contraction of a critical number of repeats on only the “A” background results in FSHD (46).

Evidence for a unifying hypothesis for the pathophysiology of FSHD emerged in 2010 from these apparently disparate findings. The hypothesis states that contraction of the repeats results in relaxation of the chromatin structure, allowing the expression of the DUX4 gene within the repeats. However, as the DUX4 sequence lacks the mRNA-stabilizing polyadenylation tail (polyA) sequence, no DUX4 protein is produced. The one exception is the DUX4 in the most distal repeat unit, which can splice the polyA sequence that is present in the “A” but not the “B” 4q35 sequence variant distal to the last repeat (44). Thus, FSHD results only if loss of a critical number of D4Z4 repeats occurs on a 4q35 allele with a distal “A” variant. This event results in the derepression of a gene, DUX4, which is normally switched off differentiated cells (82).

Although genetically different, FSHD2 appears to also be the result of reactivation of the DUX4 gene. In FSHD2, there is no contraction of the number of D4Z4 repeats, but both copies of the 4q35 D4Z4 sequences are profoundly hypomethylated, resulting in a chromatin structure that is conducive to gene expression similar to contracted D4Z4 repeat in FSHD1 (11). This DNA hypomethylation was shown to be the result of mutations in the SMCHD1 gene in most cases of FSHD2 (42) as well as DNMT3B (100). If such hypomethylation occurs in an individual with at least one “A” haplotype, DUX4 can be expressed.

The emerging evidence is that DUX4 is highly toxic to cells inducing apoptotic cell death, rendering cells more susceptible to oxidative stress and interfering with normal myogenesis (38; 04; 03; 81).

Epidemiology

The literature suggests widely variable regional differences for prevalence. Lunt and Harper noted reports of one in 435,000 people in Wisconsin and figures for Europe from one in 17,000 people to one in 250,000 people (48). They noted that both Padberg and Becker estimated the rate in Europe to be about one in 20,000 people. The figure in Wales was 4.4 in 100,000 people. A re-estimation of the prevalence of FSHD in the Netherlands now puts the figure at 2.4 in 20,000 people, more than double prior estimates (09). In the northern Spanish region, the prevalence was found to be approximately 5.2 out of 100,000 people (61). In China, the prevalence of FSHD1 is estimated to be 0.75 per 1,000,000 people (105).

Prevention

FSHD is an autosomal dominant disease; thus, genetic counseling has long been available (102), and testing is now made more precise by DNA analysis. Probes are available to make the diagnosis in an individual case. Prenatal diagnosis is possible, but because the disease is often so mild, diagnosis of the fetus seems to be requested less often than in some other conditions. Preimplantation genetic diagnosis is possible as well but is complicated in FSHD. FSHD1 genetic testing is not PCR-based and, therefore, requires more DNA than can be obtained from a single cell. Preimplantation genetic diagnosis can still be performed indirectly by using linkage analysis to polymorphic markers proximal to the D4Z4 repeats. This latter method requires the availability of several affected and unaffected family members for testing, and its accuracy in predicting the status of the embryo is not clear (88).

Differential diagnosis

Neurogenic facioscapulohumeral muscular atrophy (or spinal muscular atrophy of facioscapulohumeral distribution). This condition has been described (20), but DNA analysis indicates that it is really FSHD with inexplicably erroneous neurogenic patterns in EMG or muscle biopsy (99; 48). Misleading EMG patterns have been reported less often as recordings have improved technically and quantitative analysis of motor unit potentials has become more widely available. Since the advent of the DNA test, there have been no such reports of a neurogenic form.

Polymyositis. Two problems have caused FSHD to be confused with polymyositis. First, for reasons not known, collections of inflammatory cells may be found in the muscle biopsies of patients with typically autosomal dominant FSHD. When this is encountered in a sporadic, nonfamilial case, there may be diagnostic uncertainty. Treatment with prednisone or other immunosuppressive drugs has been used in these patients without benefit.

The other problem arises when patients with an acquired myopathy have weakness in facioscapulohumeral distribution. However, these patients do not have the typical appearance of the abnormal chest and shoulder girdle that characterizes FSHD.

Scapuloperoneal syndrome. Clinicians have been impressed by a syndrome that seems identical to FSHD, except that the face is not involved. This seems a slender reed to differentiate genetic diseases. The shoulder girdle weakness and appearance are identical to those of FSHD. It seemed that the condition could be allelic with FSHD because some individuals in typical FSHD families lack evident facial weakness. Moreover, like FSHD, the scapuloperoneal syndrome is usually myopathic and autosomal dominant; and also like FSHD, there have seemed to be neurogenic forms, and some cases did not meet criteria for autosomal dominant transmission. (X-linked families were probably instances of Emery-Dreifuss syndrome.) This issue seems to be on the road to resolution as a result of DNA analysis. Linkage to 4q35 has been excluded in several scapuloperoneal families (48; 87). Wilhelmsen and colleagues not only excluded linkage to 4q35 in one family but also found linkage of scapuloperoneal muscular dystrophy to chromosome 12, proving that this is a disorder distinct from the facioscapulohumeral disorder (108). The difference can be suspected clinically because the face is spared in all members of the family (not just a few individuals), but proof of diagnosis requires DNA analysis for linkage to 4q35 or chromosome 12. Proof for misdiagnosing the chromosome 12 scapuloperoneal muscular dystrophy led to the discovery of an X-linked scapuloperoneal dystrophy in a large Italian-American family, identifying mutations in a gene encoding four-and-a-half LIM protein 1 (FHL1) (65).

Limb-girdle muscular dystrophies. Calpainopathies (LGMD2A) have been reported to mimic the FSHD clinical phenotype as have patients with mutations in the valosin-containing protein (VCP) gene (74). The latter is often associated with Paget’s disease and frontotemporal dementia and has distinctive features on muscle biopsy.

Mitochondrial myopathy. Hudgson and colleagues were the first to record ragged-red fibers in a myopathy that clinically simulated FSHD; Worsfold and colleagues found evidence of impaired oxidative metabolism in that family. Rowland and colleagues described a man with a facioscapulohumeral distribution of weakness and prominent ragged-red fibers in muscle biopsy (71). There was no deletion of mitochondrial DNA, but a point mutation was not found. He had a prominent cardiomyopathy with congestive heart failure and could not stand erect because of trunk weakness; these are all features unlike FSHD. Slipetz and colleagues found deficiency of mitochondrial complex 3 in a 44-year-old woman with apparent FSHD and atrial re-entrant tachycardia (80). The shoulder girdle deformity made it difficult to keep the leads of a pacemaker in place. There were no ragged-red fibers. In another family with autosomal dominant scapulohumeral myopathy, there were no ragged red fibers, and the finding of a point mutation in an affected person may not have been decisive (62).

Emery-Dreifuss muscular dystrophy. This condition is characterized and defined by distinctive clinical features: cardiac arrhythmia (especially atrial paralysis requiring insertion of a pacemaker); contractures at elbows, knees, ankles, and spine; lack of facial weakness; and X-linked inheritance. None of these features is found in FSHD, and there should be no problem in this distinction. Occasionally, Emery-Dreifuss muscular dystrophy appears in successive generations to suggest autosomal dominant inheritance. Even then, however, the clinical features that define this disorder differ from those of FSHD.

Other myopathies. Individual cases have been reported of other myopathies simulating the facioscapulohumeral distribution of weakness. Among those diseases are inclusion-body myositis, desmin-storage myopathy, centronuclear myopathy, and acid maltase deficiency. Each of these conditions is defined by specific morphologic changes on muscle biopsy.

Diagnostic workup

The diagnosis of FSHD is usually evident clinically and confirmed by genetic testing. If there is doubt as to whether the condition is myopathic or neurogenic, a serum creatine kinase and EMG are helpful. Serum creatine kinase value should not be more than 10 times normal unless there is some complicating feature. In patients with infantile or early childhood–onset FSHD, the presence of hearing loss or retinal vascular disease is supportive of the diagnosis of FSHD. A muscle biopsy is not needed for the diagnosis of FSHD. A biopsy is necessary if genetic testing is negative to rule out an alternative diagnosis.

A diagnostic algorithm of genetic analysis has been described (68). Blood cells from patients are analyzed for the D4Z4 repeat size and the A/B haplotypes on chromosomes 4 and 10. If the D4Z4 repeat size is not contracted and the haplotype is the nonpermissive B haplotype, then it is unlikely to be FSHD. If one permissive haplotype is present, then methylation assay of the 4q chromosome should be pursued, with abnormal results on methylation assay prompting a search for FSHD2-associated gene mutations. There are diagnostic exceptions with translocations between 4q35 and 10q26 (43), somatic mosaicism, and duplication of the D4Z4 region, with a noncontracted D4Z4 region obscuring a contracted D4Z4 region on the same chromosome (45). Other testing alternatives have been proposed to use molecular combing, single-molecule optical mapping (85), nanopore sequencing, and bisulfite sequencing to measure methylation levels (27).

Management

With education, training, and emotional support, many patients are active in diverse occupations. AAN evidence-based guidelines now provide a basis for the uniform management of FSHD and its potential complications (86). The guidelines address several relevant issues, including exercise, scapular fixation, pain management, and respiratory compromise.

Clinicians may worry about exercising weak muscles. Trials have shown improvement in aerobic capacity with aerobic exercise training, though little to no improvement in dynamic strength (103).

For those with severe limitation of arm functions, surgical scapular fixation can improve shoulder range of motion. In one series, 12 patients were followed for 3 to 21 years after surgery (06; 33; 96; 01; 66; 22). All but one were pleased with the improved abduction and better performance in lifting and carrying objects. Two patients had adverse effects: transient brachial plexus palsy in the first and a frozen shoulder in the second. Other adverse effects include pleural effusion, atelectasis, scapular fractures, and rib fractures (02). Others have reported favorable results (33; 12; 66; 18). In other series, patients expressed satisfaction with the cosmetic effects as well as improved function (33; 96; 01; 22). Erşen and colleagues found that the beneficial effect on shoulder range of motion persists with time. There appears to be a minor but significant complication rate that requires surgical revision (18). Orrell and colleagues concluded that a controlled trial of scapular fixation could not be done (55). Based on the fairly weak level of evidence, the AAN guidelines recommend a cautious approach to scapular fixation (86). Before recommending scapular fixation, the potential improvement in shoulder range of motion can be assessed by bedside manual fixation of the scapula to see what gain in range of motion fixation might achieve. Additionally, the overall rate of disease progression in the individual and the muscle strength in the targeted arm need to be considered.

Because symptomatic retinal vascular disease and hearing loss are now known to be limited to patients with large deletions (ie, residual D4Z4 allele size of 10 to 18 kb), screening for these complications should be limited to this subgroup of patients with FSHD. Severe retinal vascular disease can lead to a potentially preventable exudative retinopathy also known as Coats disease. Such patients need periodic surveillance with dilated indirect ophthalmoscopy (83; 86). Similarly, patients with large deletions are predisposed to hearing loss. Audiometric testing is especially important in infants and young children as, if not detected and treated early, hearing loss can lead to delayed language development (49; 86).

Although significant respiratory compromise is relatively uncommon, clinicians should be vigilant in patients with the predisposing factors listed above who should have routine measurement of their FVC and asked about symptoms of nocturnal hypoventilation, such as orthopnea, non-restful sleep, frequent nightmares, and early-morning headaches. All patients should have baseline pulmonary function testing done, and those with signs of restrictive lung disease (FVC < 60% predicted) and with predisposing factors should have regular pulmonary function testing (86).

Pain is a commonly reported symptom by patients, both children and adults (13; 36). Clinicians should ask about and monitor patient’s pain and discuss risks and benefits of physical therapy or pharmacological interventions. Fatigue is also a commonly reported symptom and may be also intertwined with pain.

A systematic literature review and retrospective study demonstrated no evidence of cardiac structural involvement (86; 39; 14). Additionally, the evidence for the predilection of patients with FSHD to atrial arrhythmias is insufficient to reliably determine its frequency, but is likely very low. As such, no routine cardiac testing is needed unless a patient is symptomatic.

Drug therapy aimed at nonspecific enhancement of muscle function has been largely ineffective (70; 17; 107). Based on the lack of evidence of efficacy, the AAN guidelines do not recommend the use of any of the drugs studied (prednisone, albuterol, diltiazem) for the treatment of muscle weakness in facioscapulohumeral muscular dystrophy (86). The results of a 17-week randomized placebo-controlled trial of an antioxidants cocktail have been published (63). There was no significant change in the primary outcome—the 2-minute walk, but there were modest improvements in some secondary outcome measures. A phase 2 randomized placebo-controlled, double-blind 48-week study of losmapimod, a p38 protein kinase inhibitor hypothesized to reduce DUX4 expression, has been completed (32). The study failed to meet the primary endpoint of reducing DUX4 expression in muscle, but it showed changes in reachable working space and MRI fat infiltration that were promising enough to warrant a phase 3 clinical trial.

However, the identification of DUX4 misexpression as the primary cause of FSHD has, for the first time, provided a targeted approach to the treatment of FSHD and is actively being pursued in multiple exciting approaches (47).

Special considerations

Pregnancy

There are conflicting reports about increased operative deliveries and preterm births in women with FSHD (72; 08). Additionally, about 25% of women with FSHD report subjective, persistent worsening of muscle function following pregnancy.

Anesthesia

No special risks of anesthesia have been recognized. However, as respiratory compromise in indolent neuromuscular diseases can be asymptomatic, a preoperative measurement of pulmonary functions is recommended. The presence of restrictive lung disease may require more prudent extubation and more careful use of medications that suppress respiratory drive.