The First Case Studies

Written documentation of the natural history of SMA dates to 1891, when Guido Werdnig, an Austrian neurologist at the University of Graz, wrote case studies on 3 year old Wilhelm Bauer and his 1 year old brother. Wilhelm developed weakness in his proximal limbs at about 1 year of age, and over the subsequent 2 years progressively lost muscle tone, voluntary movements, and the ability to swallow and hold up his head. Just before Wilhelm’s 3rd birthday, Werdnig noted that the boy exhibited weakened muscles, large fat deposits, flexed legs, tremor in hands and arms, and very limited, labored movement.

Less than a month after this clinical examination, Wilhelm developed rales and exhibited retraction of intercostal muscles. Two days later, he developed dyspnea and fever; several days after the onset of dyspnea, he passed away. His younger brother lived until 6 years of age and died of similar respiratory complications. Werdnig completed a microscopic examination of the older brother’s spinal cord cross sections and gastrocnemius muscle. Within the second and third cervical nerves, he noted several abnormal findings: regions of the lateral tracts lacking in myelinated fibers, degenerated anterior funiculi, indistinct processes in the ganglion cells of the anterior horn of the spinal cord, and many empty cell-beds. Posterior horns and roots were observed to be normal ( Fig. 5.1 ). These findings were consistent across many of the cervical, thoracic, and lumbar sections.

Histological analysis of Werdnig’s first case study. Cross section of seventh cervical level showing honeycombed anterior horn (I); cross section of second lumbar level (II); anterior horn at L2–3 level, demonstrating empty cell beds (b, c) and some shriveled ganglion cells (III); posterior columns at sixth cervical nerve level (IV); posterior columns at sixth cervical level, higher magnification (V); longitudinal section of gastrocnemius muscle displaying fiber atrophy (VI).

Upon examination of the gastrocnemius muscle, he noted groups of muscle fibers separated from one another by large masses of fatty tissue. These fibers included single fibers with disrupted cross-striations and degraded contractile tissue that resembled flattened tubes with nuclei ( Fig. 5.1 ).

Based on the clinical history and autopsy, Werdnig concluded that Wilhelm, along with his younger brother who similarly lost muscle function in his legs yet maintained normal sensation, had an infantile, familial muscular atrophy. This illness, Werdnig posited, resembled a neurological illness due to its swift course of atrophy. Thus, it was distinct from muscular dystrophy, or a slow degradation of muscle mass due to a lack of functional protein in the muscle.

Over the next decade, German neurologist Johann Hoffmann published five papers in accord with Werdnig’s conclusions. He documented a total of seven cases across four families, all with onset around 1 year of age. Age of death varied between 14 months and 5 years. Hoffmann’s autopsies echoed two of Werdnig’s key observations: atrophy of muscle in the extremities (with proximal onset) and damaged cells in the anterior horn of the spinal cord. Hoffmann agreed with the majority of Werdnig’s evidence, disputing only his finding that affected children exhibited hand tremor. Through these case studies, Hoffmann reinforced evidence for a familial, early-onset motor neuron disease, which he called spinale muskelatrophie , or “spinal muscular atrophy.”

The Emergence of Subtypes

Over the next century, additional case studies helped solidify the medical definition of SMA and divide the illness into broad severity categories (severe, intermediate, mild).

Following Werdnig and Hoffmann’s influential findings, two papers published at the turn of the 20th century shed light on the most severe form of infantile SMA. In 1899, Sylvestre presented “Paralysie flasque de quatre membres et des muscles du tronc (sauf le diaphragme) chez un nouveau-né” (translation: Flaccid paralysis of the four limbs and intercostal muscles (except the diaphragm) in a newborn) at the Pediatric Society of Paris. The case, which chronicles a 2 month old with spinal muscle atrophy since birth, was later published in the Bulletins de la Société de Pédiatrie de Paris . Two other children in the infant’s family exhibited similar symptoms and passed away before 6 months of age.

Four years later, British neurologist Charles Edward Beevor published a comparative study of one case of severe infantile SMA and one case of spinal cord hemorrhage at birth. Two infants were admitted to the National Hospital, Queen Square, in London, with similar patterns of paralysis in the legs and torso. Beevor included detailed notes on the family history, case history, and postmortem examination of both infants with accompanying illustrations by Dr. F Batten.

Within the comprehensive description of Case 1, the child with SMA, several details stand out that link the case to previous documentation of SMA. Of the 5 week old infant’s seven siblings, three had exhibited similar paralysis and died before 8 months of age, demonstrating familial inheritance. In terms of clinical history, Beevor’s description of muscle degradation paralleled that of Sylvestre: flaccid limbs, paralysis of limb and intercostal muscles, and normal diaphragm function. Finally, the postmortem examination revealed atrophy of anterior horn cells at the cervical, thoracic, and lumbar levels that echoed Werdnig and Hoffmann’s findings a decade earlier. A comparison of Case 1 and normal tissue revealed that Case 1’s anterior horn cells at the level of the second lumbar were significantly smaller, measuring less than 0.03 mm, while those of the normal tissue measured between 0.04 and 0.05 mm. All peripheral nerves appeared normal in Case 1 with the exception of degenerate nerves in the brachial plexuses. Additional spinal cord degeneration was found in lumbar to cervical posterior columns and in the 5th right cervical root. Staining of right bicep muscle tissue showed extremely atrophied muscle fibers (0.008 mm in diameter compared to a normal diameter of 0.02 mm).

Although Beevor’s case paralleled that of Werdnig and Hoffmann in many of its clinical details, it was distinct in its age of onset and clinical progression. The infant’s mother reported that she felt no movement during pregnancy, suggesting that the paralysis developed in utero. The newborn was also fully paralyzed at birth. The infant passed away at 8 weeks, a considerably more rapid progression than that of Werdnig’s and Hoffmann’s cases. Thus, Beevor established comprehensive evidence for the existence of a congenital form of SMA. In our modern classification system, this case is representative of type 0 or type 1 SMA.

The third major subtype of SMA, a milder variety with age of onset in childhood and adolescence, emerged in the literature in the 1950s. Two research groups initially described this variant of SMA as a condition simulating muscular dystrophy. Swedish neurologists Wohlfart, Eliason, and Fex published a study of three families with hereditary muscle weakness and atrophy caused by nerve damage, and concluded that it could possibly be a “benign variant” of severe SMA. One year later, Kugelberg and Welander presented 12 cases of progressive limb weakness originally misdiagnosed as muscular dystrophy but later found to be neurogenic through muscle biopsy and electromyographic data. The study found a pattern of onset of muscle weakness in thigh and pelvic muscles and loss of knee reflexes as the first neurological positive finding. Muscle weakness gradually progressed in these cases over 20–40 years. Kugelberg and Welander also argued that this disease was distinct from Werdnig–Hoffmann syndrome due to its delayed onset and milder symptoms, a theory that genetic analysis has since overturned. Today, type III SMA is interchangeably referred to as Kugelberg–Welander disease, or juvenile SMA.

Modern Clinical Presentation and Diagnostic Criteria

The first classification system was created by Byers and Banker in the 1960s and included Group I, Group II, and Group III designations, in order from most to least severe muscle degradation and earliest to latest onset, based on 52 cases of muscular atrophy at Boston Children’s Hospital. Group I cases were characterized by onset before 2 months of age, widespread muscle weakness, and early death. Group II cases were defined by an onset between 2 and 12 months of age, initially localized muscle weakness, and longer survival. Finally, Group III included cases with onset after 1 year of age and survival measured in years as opposed to months.

In the early 1990s, soon after the discovery of gene linkage to chromosome 5q, the International SMA Collaboration convened to establish specific, standard diagnostic criteria for SMA to assist clinicians in recognition and diagnosis of disease. Inclusion criteria were broken up into two categories: weakness, further characterized as symmetrical, proximal > distal, legs >arms, and trunk involved; and denervation, confirmed by electromyographic diagnosis, muscle biopsy, and clinical fasciculations. Exclusions included central nervous system dysfunction, arthrogryposis, involvement of other neurologic system of other organs, sensory loss, eye muscle weakness, and marked facial weakness.

In 1998, the 59th International Workshop of the European Neuromus-cular Center revised the above criteria to include electrophysiological, histopathological, and genetic inclusion and exclusion criteria. Notably, when present alongside clinical symptoms, a mutation in or absence of the SMN gene is diagnostic. Additional modified inclusion criteria include abnormal spontaneous electromyographic activity, increased mean duration and amplitude of motor unit action potentials, and atrophic and type I hypertrophic fibers. The revised criteria also accounted for observed exceptions to the 1990 criteria; for instance, severe congenital cases often exhibit some sensory loss and arthrogryposis.

These workshops distilled the volume of SMA studies into the three broad categories of severe, intermediate, and mild (I–III) that echoed Byers and Banker’s prior organization ( Table 5.2 ).

Patients with type I SMA develop symptoms by 6 months of age and face a 38.1% survival probability at 2 years of life. These SMA patients account for approximately 50% of cases, making it the most common form of the disease. Clinical characteristics of type I patients include generalized hypotonia, impaired bulbar function, and respiratory insufficiency. These children are defined by their inability to sit unsupported. An increase in survival of type I patients has been shown with the use of noninvasive, assisted ventilation. This type of SMA is also referred to as Werdnig–Hoffmann disease, despite the fact that the clinical criteria do not match Werdnig or Hoffmann’s case studies.

Type II SMA is categorized today by an onset of proximal limb weakness at 6–18 months of age with progressive weakness, scoliosis, joint contractures, and substantial respiratory morbidity developing into childhood. These children achieve the ability to sit independently but never stand or walk. It is usually not fatal (70% of those affected reach 25 years of age) but necessitates wheelchair assistance. It is also referred to as Dubowitz disease, after the English neurologist Victor Dubowitz, who cataloged many additional intermediate SMA cases in 1960.

Patients with type III SMA develop onset of proximal weakness at >1 year of age and experience milder symptoms and a normal lifespan. Clinical characteristics include heel cord tightness and contractures, tremor, lumbar lordosis, and an increased risk for fractures and scoliosis. Though weakness continues to progress throughout childhood and adolescence, type III SMA patients will achieve independent ambulation.

Additional Diagnostic Categories and Challenges

It is worth noting that several different categorizations of SMA subtypes persist in the literature. Certain clinicians advocate for three solely functional categories: nonsitters, sitters, and walkers. Others utilize a four-category system, with SMA subtypes I, II, III, and IV. These are assigned on the basis of clinical factors such as age of onset, motor milestones, and prognosis. Types I through IV are, respectively, infantile (Werdnig–Hoffmann), intermediate, mild, and adult-onset. Type IV SMA, or adult-onset SMA, is characterized by mild proximal limb weakness and an onset after 18 years of age. Onset of limb weakness usually occurs after 30 years of age and occurs gradually, with proximal lower limbs affected first. Quadriceps, ileo-psoas, glutei, triceps brachialis, and deltoid are the muscles that most commonly show signs of atrophy. Additional symptoms vary widely but may include hand tremors and tongue fasciculations. There are no major motor impairments and the natural lifespan is normal. SMN1 mutations are implicated in some cases of Type IV SMA, but in many cases there are no clear genetic causes.

Type I can be further divided into IA (congenital onset), IB (0–3 months of age onset), and IC (3–6 months of age onset). IB and IC are referred to as “nonsitters” due to a lack of ability to sit unsupported. In a five-subtype system, such as that proposed by MacLeod in 1999, antenatal onset is distinctly referred to as Type 0, while early onset (0–6 months) is named Type 1. Type 0, known also as arthrogryposis multiplex congenital type, is at the most severe end of the SMA spectrum with age of onset in utero and death within weeks of birth. The average lifespan of these infants is less than 6 months. Patients exhibit congenital hypotonia, weakness, proximal joint contractures, and respiratory failure. These patients require respiratory support measures to maintain life.

A growing body of evidence suggests the substantial variability within diagnostic categories and calls into question the reliability of certain diagnostic criteria. For instance, a 2002 study by Borkowska and colleagues followed 349 SMA Type 1 patients and observed that 36, or approximately 10%, survived until their 5th birthdays. Classic diagnostic criteria maintain that Type I patients will die before their 2nd birthdays. This finding therefore disputes the notion that data on age of death should be used as the basis for creation of subtypes.

Victor Dubrowitz proposed a new numerical system of classification that further breaks down types I, II, and III into intervals of 0.1. Therefore, a child with early onset and severe paralysis would not necessarily fall under the same umbrella type as a child with early onset who is barely unable to sit without aid. This system could account for the variability of survival, motor, and respiratory function within broad types.

Our current knowledge of SMN protein function supports Dubrowitz’s notion of an SMA severity spectrum. The SMN protein is likely multifunctional, and its effects are thus likely dependent on specific protein levels. Select inhibitions of function may correlate to specific degrees of severity of disease progression. Our growing understanding of SMN protein function will help to further elucidate the complex variation of SMA presentation in the future. A more precise classification of the many forms of this illness will assist in accurate diagnosis and tailored treatment for each individual afflicted.

Our modern understanding of SMN protein and SMA subtypes also suggests that the disease may be neurodevelopmental in addition to neurodegenerative. SMN protein levels are particularly high in motor neurons during prenatal nervous system development in healthy individuals, suggesting a critical role for SMN in normal motor function. Furthermore, SMA type 0 and type I patients exhibit pathological features consistent with incomplete formation of motor neuron pathways: immature motor neurons and myofibers with ongoing apoptosis. In a mouse model of SMA, neuromuscular junction synapses were impaired due to decreased vesicle density and conservation of the fetal achetylcholine receptor subunit. It is posited that Type II and III SMA, associated with higher SMN2 copy number and thus higher levels of SMN protein, are likely more neurodegenerative in nature: motor neurons and muscle tissue develop normally during the gestational period but are vulnerable to degradation over time.