Differentiation of Hereditary Spastic Paraparesis From Primary Lateral Sclerosis in Sporadic Adult-Onset Upper Motor Neuron Syndromes FREE

In patients with an apparently sporadic adult-onset upper motor neuron (UMN) syndrome, clinical differentiation between primary lateral sclerosis (PLS) and hereditary spastic paraparesis (HSP) can be problematic.¹ Primary lateral sclerosis is a rare sporadic disorder of progressive spinobulbar spasticity with a mostly spinal and occasionally bulbar region onset.^2- 12 Hereditary spastic paraparesis is a clinically and genetically heterogeneous group of disorders characterized by a slowly progressive spastic paraparesis.^1,13 To date, 15 genes and more than 20 additional loci have been identified for autosomal dominant, autosomal recessive, and X-linked forms of HSP.^14- 21 To exclude HSP, current diagnostic criteria for PLS require absence of a family history.⁵ However, researchers increasingly recognize that the apparently sporadic occurrence of HSP is not uncommon, and mutations of the spastin (SPG4) and paraplegin (SPG7) genes are a frequent cause.^1,22- 23

Differentiation between HSP and PLS is important for genetic counseling of family members and for the patients' prognosis because HSP generally has a more favorable prognosis than PLS.¹ Furthermore, progression to amyotrophic lateral sclerosis (ALS), as may occur in PLS,²⁴ is not expected in HSP. Symptomatic UMN involvement of the arms and particularly of the bulbar region is unusual in pure HSP and would favor a diagnosis of PLS.^1,25 However, PLS may present with a slowly progressive spastic paraparesis of the legs, similar to the typical HSP phenotype, for many years before the onset of arm or bulbar region symptoms.^1,9 To separate apparently sporadic HSP from PLS, previous studies used several clinical features, such as an age at onset of younger than 40 years,¹⁰ evidence of mild dorsal column impairment such as decreased vibratory sensation in the distal legs or abnormal leg somatosensory evoked potentials (SSEPs),¹ and symptoms of urinary urgency,^1,5 but how appropriate these criteria are has yet to be validated.

To study whether clinical characteristics can be used to differentiate between sporadic presentations of HSP and PLS in patients with apparently sporadic adult-onset UMN syndromes, we performed a nationwide search for patients in the Netherlands.

From November 1, 2002, through March 31, 2005, all Dutch neurologists were asked to enroll patients with an apparently sporadic adult-onset idiopathic UMN syndrome. In addition, 1 large referral center for patients with motor neuron disease in Leuven, Belgium, agreed to participate. The study protocol was approved by the medical ethics review board of the University Medical Center Utrecht, and written informed consent was obtained from each patient.

Inclusion criteria were a gradually progressive UMN syndrome with adult onset (age, ≥18 years) and a disease duration of at least 3 years. Exclusion criteria were a positive family history, clinical evidence of generalized lower motor neuron involvement fulfilling the revised El Escorial criteria for clinically probable or clinically definite ALS,²⁶ the presence of cerebellar or extrapyramidal signs, the presence of sensory signs other than decreased distal vibration sensation (absent at the ankles), and other causes of UMN loss that were investigated using a predefined series of laboratory tests. The presence of mild focal amyotrophy limited to 1 or 2 muscles in 1 region and the presence of fasciculation were allowed. Additional laboratory investigations were performed to exclude other causes, including serum biochemistry, plasma levels of vitamins B₁₂ and E, thyrotropin, and very-long-chain fatty acids; biliary alcohol levels in urine; serologic testing for syphilis, Lyme disease, human T-lymphotropic virus 1, and human immunodeficiency virus infection; and magnetic resonance imaging (MRI) of the brain and the spinal cord. Diagnostic MRI white matter changes, for example abnormalities suggestive of demyelinating disorders or leukodystrophy, and the presence of a thin corpus callosum were not allowed. We required that appropriate needle electromyography (EMG) had been performed at least 3 years after disease onset to exclude (laboratory-supported) ALS, according to the revised El Escorial criteria.²⁶ DNA analysis of the spastin gene (SPG4) (GenBank AJ246001, OMIM *604277), the most common cause of autosomal dominant pure HSP,²⁷ and the paraplegin gene (SPG7) (GenBank Y16610, OMIM *602783), which causes a form of autosomal recessive HSP with pure or complicated phenotypes,²⁸ was performed in all patients. Hereditary spastic paraparesis was genetically confirmed in 14 patients (7 with SPG4 and 7 with SPG7 mutations) as published previously.^22- 23 In approximately 75% of our patients, we performed additional genetic tests using multiplex ligation-dependent probe amplification to identify SPG4 exonic microdeletions,²⁹ but none were found.

The bulbar region was considered affected if pseudobulbar dysarthria was present. The arms and legs were considered affected if there was at least 1 of the following findings on the neurological examination results: spasticity (defined as a modified Ashworth Scale³⁰ score of ≥2), obvious loss of dexterity due to spasticity (evaluated at the examination), pathological hyperreflexia (defined as an Institute of Neurological Diseases and Stroke Myotatic Reflex Scale³¹ score of 4 or 5 or extensor plantar response), or a UMN paresis (Medical Research Council grade of no greater than 4).³² We classified patients as having asymmetry if, for any affected body region, spasticity (modified Ashworth Scale score of ≥2), paresis (Medical Research Council paresis grade of ≤4), or dexterity loss were only present unilaterally. Functional impairment was assessed using the Amyotrophic Lateral Sclerosis Functional Rating Scale.³³

At inclusion, patients underwent standardized needle EMG of 3 muscles per limb and 2 thoracic and 3 bulbar muscles, with the findings interpreted according to the revised El Escorial criteria.²⁶ Diagnosis of probable laboratory-supported ALS according to the revised El Escorial criteria requires that at least 2 regions fulfill the EMG criteria. Results of leg SSEP studies were available for 33 patients, and the results were classified as normal or abnormal.

Differences between groups were tested using the Pearson χ² and Fisher exact tests. We used the Kruskal-Wallis and Mann-Whitney tests to evaluate differences in the case of nonnormal distribution of values. Statistical analysis was performed by one of us (F.B.).

The medical files of 182 eligible patients were screened for this study. Of these, 23 patients were unable or unwilling to participate, and 55 were excluded on the basis of inclusion and exclusion criteria. In 14 of the remaining 104 patients, a pathogenic mutation of the spastin (SPG4) or paraplegin (SPG7) gene was found (7 with SPG4 and 7 with SPG7 mutations).^22- 23 In the 14 patients with the SPG4 or SPG7 mutation, 13 had a typical pure HSP phenotype with only leg involvement, and 1 patient with the SPG7 mutation also developed UMN symptoms of the arms (arm hyperreflexia associated with loss of dexterity) after 1 year. Of the other 90 patients, 39 had a phenotype similar to typical pure HSP (involvement of the legs only), 36 had bulbar region involvement suggestive of PLS, and 15 patients had a phenotype that was difficult to classify as being similar to HSP or suggestive of PLS (UMN involvement of arms and legs). Of the 36 patients with bulbar region involvement, 28 also had symptoms in the arms and legs, 2 patients in the legs, and 2 in the arms. The patients' characteristics are presented in Table 1.

The pattern of disease progression in the 14 patients with the SPG4 or SPG7 mutation and the other 90 patients with a sporadic UMN syndrome is shown in Table 2. All 14 patients with the SPG4 or SPG7 mutation and 78 of the other patients with sporadic UMN disease (87%) had symptom onset in the legs. Of the 36 patients with bulbar region symptoms, 25 had symptom onset in the legs. The Figure shows the time from onset of leg symptoms (HSP phenotype) to development of bulbar region symptoms (PLS phenotype) in these patients. More than half of these patients developed bulbar region symptoms within 5 years, which may be an underestimation because disease duration for some patients was only 3 years. However, the time range was broad, and 1 patient even developed a PLS phenotype after 18 years. It is therefore possible that the patients in our study with a sporadic UMN disease and only leg involvement may still develop bulbar region symptoms, even after a relatively long disease duration. Because previous studies used clinical characteristics such as onset before age 40 years,¹⁰ evidence of mild dorsal column impairment,¹ and symptoms of urinary urgency^1,5 to separate HSP from PLS, we compared the frequency of these and other clinical characteristics in the 14 patients with the SPG4 or SPG7 mutation with those of the other 90 patients with UMN disease and with those of the 36 patients with at least bulbar region symptoms (Table 1). A significant difference was found only for age at onset (younger in patients with the SPG4 or SPG7 mutation compared with all patients without the SPG4 or SPG7 mutation [P = .01] and with the patients with bulbar region involvement [PLS phenotype] [P = .002]). There were no other significant differences in clinical features between patients with the SPG4 or SPG7 mutation and the patients with the PLS phenotype. The range of age at onset, however, was similar in patients with the SPG4 or SPG7 mutation (29-69 years), in all other patients with UMN disease (18-76 years), and in patients with at least bulbar region involvement (32-76 years). Six of the 14 patients with the SPG4 or SPG7 mutation had symptoms at 40 years or older, and 3 of the 36 patients with at least bulbar region symptoms had onset before age 40 years. Evidence of mild dorsal column impairment (decreased vibratory sense or abnormal leg SSEP), symptoms of urinary urgency, and mild EMG abnormalities were present in the patients with the SPG4 or SPG7 mutation and in the patients with the PLS phenotype.

Four patients had symptom onset in the arms; of these, 3 developed bulbar region symptoms (Table 2). Eight patients had onset in the bulbar region that progressed in various patterns to the arms or legs in 6.

Four of the 25 patients with symptom onset in the legs who developed bulbar region symptoms showed an asymmetrical pattern of progression, with only 1 leg being affected before symptoms progressed to the arms or bulbar region. This pattern of asymmetrical progression was also seen in 4 of the 16 patients with involvement of only the arms and legs, but not in the 14 patients with the SPG4 or SPG7 mutation.

Differentiation between HSP and PLS is important for genetic counseling of family members and for the patients' prognosis because HSP generally has a more favorable prognosis than PLS.¹ Furthermore, progression to ALS, as may occur in PLS,²⁴ is not to be expected in HSP. Previous studies used several clinical and laboratory features, such as onset before age 40 years,¹⁰ evidence of mild dorsal column impairment (eg, decreased vibratory sensation in the distal legs or abnormal leg SSEPs),¹ and symptoms of urinary urgency^1,5; however, the appropriateness of these criteria has yet to be validated. For that reason, we compared the presence of these features in a large cohort of patients with an apparently sporadic adult-onset UMN syndrome that consists of patients who had genetically proved (the SPG4 or SPG7 mutation) HSP (subtype 1), a phenotype similar to typical pure HSP (involvement of the legs only) (subtype 2), a phenotype (involvement of the arms and legs) that was difficult to classify as being similar to HSP or suggestive of PLS (subtype 3), or bulbar region involvement suggestive of PLS (subtype 4). The main objective of our study was to investigate whether clinical or laboratory features are useful to differentiate between patients in subtypes 1 and 4, who have the most definite diagnosis of HSP or PLS. Our results demonstrate that these features are not useful in clinical practice for distinguishing HSP from PLS in the individual patient. There was substantial overlap in age at onset between phenotypes: the youngest onset in our study occurred in a patient with a PLS phenotype at 32 years, and it was even younger (23 and 26 years) in other PLS studies,^8,34 whereas 6 patients with the SPG4 or SPG7 mutation had onset at 40 years or older; the oldest onset was 69 years. Symptoms of urinary urgency, previously considered atypical for PLS,⁵ were even more frequent in our patients with bulbar region involvement (53%) compared with patients with the SPG4 and SPG7 mutations (43%) and were also reported in more recent PLS series.^8- 9 Signs of mild dorsal column impairment (decreased vibratory sensation in the distal legs or abnormal leg SSEPs), considered indicative of a diagnosis of HSP instead of PLS in a previous study,¹ were also found in patients with a phenotype suggestive of PLS (bulbar region involvement) in our study, which confirms several previous PLS studies.^8,34 Because 3 of our 7 patients with the SPG7 mutation (and none of the other patients) developed cerebellar involvement during follow-up, its appearance could be an important clinical clue for SPG7 mutation as a cause in apparently sporadic UMN syndromes.²³

Although the number of patients with genetically confirmed HSP included in our study was relatively low, we chose to use only those HSP patients with a sporadic presentation because patients with familial HSP may present with different clinical features. Hereditary spastic paraparesis was genetically confirmed in 14 of 104 patients (7 with the SPG4 mutation and 7 with the SPG7 mutation). We tested only the SPG4 and SPG7 genes to identify patients with genetically proved HSP, but more extensive testing would have identified only a relatively small number of additional patients from subgroups 2 and 3, described in the following paragraphs. More importantly, a more extensive screen for HSP genes would not have influenced our main conclusion that features that have been considered indicative of HSP (onset at <40 years, mild dorsal column involvement, and urinary urgency) appear not to exclude PLS and that onset at 40 years or older is not uncommon in apparently sporadic HSP.

Genetic testing is the only reliable way to differentiate HSP from PLS in patients with apparently sporadic adult-onset UMN syndromes, at least for those who have symptom onset in the legs. More than 30 genetic loci and 15 of the responsible genes have been identified for autosomal dominant, autosomal recessive, and X-linked forms of HSP.^14- 21 Mutation of SPG4, the most common cause of autosomal dominant HSP (in approximately 40%), is a frequent cause of apparently sporadic spastic paraparesis (12%-13%).^22,35 Mutations of SPG7, previously reported in 1.5% to 6% of autosomal recessive cases of HSP, are an additional frequent cause of sporadic HSP (13% in this series).²³ The role of the other known HSP genes in patients with apparently sporadic disease is largely unknown. The atlastin gene (SPG3A) mutation is the second most common cause of autosomal dominant HSP (approximately 10%) but is mostly associated with infantile or childhood onset, although patients with adult onset have been reported.³⁶ Mutations in the novel mitochondrial protein REEP1 (REEP1) accounted for 6.5% of an unselected HSP population in a study,¹⁷ whereas another study³⁷ detected REEP1 mutations in 3% of a clinically mixed sample of patients with pure and complicated HSP, including 2 of 119 patients with sporadic disease (1.7%; only 1 patient with adult-onset HSP). The ZFYVE27 gene (SPG33) was mutated in 1 of 43 patients with autosomal dominant HSP in 1 study.¹⁸ Mutations in the NIPA1 gene (SPG6) are a rare cause of autosomal dominant HSP in Europe (<1%).³⁸ The remaining known HSP genes are associated with complicated forms of HSP or probably each account for less than 1% of HSP cases. Male patients with proteolipid protein mutations (SPG2) are excluded, in part on the basis of the MRI criteria because many SPG2 patients have MRI changes, but in addition on the basis of our age-at-onset criterion because the SPG2 mutation is associated with early onset.³⁹ We believe that the SPG11 mutation is practically excluded in our patients because it seems to be rare in sporadic disease and is associated with a thin corpus callosum, which was not seen in our patients.⁴⁰

The patients with the SPG4 or SPG7 mutation all presented with UMN symptoms in the legs and that progressed to the arms in only 1 patient. Most patients (87%) without the SPG4 or SPG7 mutation initially presented with symptoms similar to the patients with the SPG4 or SPG7 mutation. We showed that bulbar region UMN symptoms may start many years after symptom onset in the legs (≤18 years), so there can be a long period during which PLS and (apparently sporadic) HSP are clinically similar. We considered the presence of bulbar region symptoms to be part of the PLS phenotype. Although we cannot fully rule out the possibility that genes will be discovered in patients with an HSP phenotype who develop bulbar region symptoms, the absence of bulbar region symptoms in adult-onset familial HSP makes this unlikely. Only 1 family was reported to have an autosomal dominant adult-onset UMN disease resembling typical PLS associated with a locus on chromosome 4p (PLS1).^41- 42 The autosomal recessive UMN diseases associated with alsin (ALS2) mutations, which typically progress to the arms and the bulbar region, seem to be limited to the early-onset forms, described as juvenile PLS, infantile ascending HSP, and, if associated with lower motor neuron involvement, juvenile ALS.⁴³

Some features may support a diagnosis of PLS, such as an onset of disease in the arms or the bulbar region and development of bulbar region symptoms or marked asymmetry during the disease course. The presence of mild focal atrophy or fasciculation may also support a diagnosis of PLS, but these were not frequently found in our study, so this requires confirmation in future larger studies. Development of UMN symptoms in the arms after disease onset in the legs usually supports a diagnosis of PLS but does not exclude HSP, because arm involvement was seen in 1 of our patients with the SPG7 mutation and was also reported previously in other patients with genetically confirmed HSP.^25,44 Despite our finding that unilateral symptom onset was common in patients with (36%) and without (53%) the SPG4 or SPG7 mutation, marked asymmetry at clinical evaluation was not seen in our patients with the SPG4 or SPG7 mutation. Although current diagnostic criteria for PLS require “symmetrical distribution of symptoms,”⁵ we and others found that such asymmetry may indeed be indicative of PLS.^8- 9 This should be verified in large prospective studies.

In patients with an apparently sporadic adult-onset UMN syndrome and symptom onset in the legs, differentiation of sporadic presentations of HSP from PLS based on clinical characteristics is unreliable and therefore depends on genetic test results. Disease onset in the arms or bulbar region, development of bulbar region involvement, or marked asymmetry during the disease course may support a diagnosis of PLS. New and more widely available genetic testing possibilities for HSP would be beneficial for patients with apparently sporadic UMN syndromes because HSP generally has a more favorable prognosis than PLS.

Correspondence: Leonard H. van den Berg, MD, PhD, Department of Neurology, Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, the Netherlands (l.h.vandenberg@umcutrecht.nl).

Accepted for Publication: October 3, 2008.

Author Contributions: Dr van den Berg had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Brugman, Wokke, and van den Berg. Acquisition of data: Brugman, Franssen, de Visser, de Jong, Faber, Kremer, Schelhaas, Bruyn, Kuks, Robberecht, and Wokke. Analysis and interpretation of data: Brugman, Veldink, Franssen, de Jong, Kremer, van Doorn, Verschuuren, Robberecht, Wokke, and van den Berg. Drafting of the manuscript: Brugman, Veldink, de Jong, Kuks, and van den Berg. Critical revision of the manuscript for important intellectual content: Brugman, Veldink, Franssen, de Visser, de Jong, Faber, Kremer, Schelhaas, van Doorn, Verschuuren, Bruyn, Kuks, Robberecht, Wokke, and van den Berg. Statistical analysis: Brugman and Veldink. Obtained funding: van den Berg. Administrative, technical, and material support: Brugman, Franssen, and Schelhaas. Study supervision: de Jong, Wokke, and van den Berg.

Financial Disclosure: None reported.

Funding/Support: This study was supported by the Netherlands Organization for Scientific Research (Dr van den Berg) and the Interuniversity Attraction Poles program P6/43 of the Belgian Federal Science Policy Office (Dr Robberecht).

Role of the Sponsor: The funding organizations had no role in the design and conduct of the study; the collection, management, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.

Additional Contributions: We are grateful for the cooperation of all the patients, the referring neurologists, and the Dutch patient support organization for neuromuscular diseases (Vereniging Spierziekten Nederland). Gerda Valk and Nienke de Goeijen, MS, the research nurses, provided assistance in the clinical evaluation of patients.