Clinical Neurology and Neurosurgery 120 (2014) 14–19

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery

journal homepage: www.elsevier .com/ locate /c l ineuro

Hereditary spastic paraparesis in adults. A clinical and geneticperspective from Tuscany

Daniele Orsucci a, Loredana Petrucci a, Elena Caldarazzo Ienco a, Lucia Chico a,Paolo Simi b, Antonella Fogli b, Fulvia Baldinotti b, Costanza Simoncini a,Annalisa LoGerfo a, Cecilia Carlesi a, Alessia Arnoldi c, Maria Teresa Bassi c,Gabriele Siciliano a, Ubaldo Bonuccelli a, Michelangelo Mancuso a,*aNeurological Clinic, University of Pisa, ItalybU.O. Laboratorio Genetica Medica, Santa Chiara Hospital, Pisa, Italyc IRCCS E. Medea, Laboratory of Molecular Biology, Bosisio Parini, Lecco, Italy

A R T I C L E I N F O

Article history:

Received 20 April 2013Received in revised form 14 January 2014Accepted 8 February 2014Available online 17 February 2014

Keywords:4SMDHereditary spastic paraparesisHereditary spastic paraplegiasHSPSPGStrumpell-Lorrain

* Corresponding author. Tel.: +39 50 992E-mail address: mancusomichelangelo@

http://dx.doi.org/10.1016/j.clineuro.2014.020303-8467/ã 2014 Elsevier B.V. All rights r

443.gmail.com

.002eserved.

A B S T R A C T

Objective: Hereditary spastic paraparesis or paraplegias (HSPs) are a group of neurogenetic conditionswith prominent involvement of the pyramidal tracts. Aim of this study is the clinical and molecularcharacterization of a cohort of patients with HSP. Moreover, we aim to study the minimum prevalence ofHSP inourarea and toproposea schematic diagnostic approach toHSPpatients basedon the availabledatafrom the literature.Methods: Retrospective/perspective study on the subjects with clinical signs and symptoms indicative ofpure or complicated HSP, in whom other possible diagnosis were excluded by appropriateneuroradiological, neurophysiologic and laboratory studies, who have been evaluated by the Neuro-genetic Service of our clinic in last two years (2011–2012).Results: 45 patients were identified. The minimum prevalence of HSP in our areawas of about 2.17–3.43/100,000. The SF-36 (quality of life) and SPRS (disease progression) scoreswere inversely related; the time-saving, four-stage scale of motor disability could predict the SPRS scores with a high statisticalsignificance, andwe encourage its use in HSP. Our study confirms SPG4 as themajor cause of HSP. All SPG4patients had a pure HSP phenotype, and the dominant inheritance was evident in the great majority ofthese subjects. SPG7 was the second genetic cause. Other genotypes were rarer (SPG10, SPG11, SPG17).Conclusion: Exact molecular diagnosis will allow a more accurate patient counseling and, hopefully, willlead to specific, targeted, therapeutic options for these chronic, still incurable diseases.

ã 2014 Elsevier B.V. All rights reserved.

1. Introduction

Hereditary spastic paraparesis (HSPs), describes a heteroge-neous group of neurogenetic disorders caused by degeneration ofthe corticospinal tracts [1]. The key clinical findings are lower limbspasticity, with hyperreflexia and extensor plantar responses [1].Urinary urgency is also frequent. Age at onset is extremely variable,from childhood through to late adult life. Traditionally, theseconditions havebeendivided intopureHSPs and complicatedHSPs,depending on the presence of adjunctive neurological features(such as ataxia, thin corpus callosum, peripheral neuropathy, distal

(M. Mancuso).

amyotrophy, retinopathy, optic atrophy, extrapyramidal signs,cognitive dysfunction, deafness, epilepsy) [2].

HSP prevalence has been estimated 3–10/100,000 in Europe[3,4], but few epidemiological studies are available. A recentpopulation-based, cross-sectional study performed in SoutheastNorway showed a prevalence of 7.4/100,000 [5], and another studyin Estonia of 4.4/100,000 [6].

The best-characterized molecular mechanisms in HSPs areimpairment of transport of macromolecules and organelles,disturbance of mitochondrial function, or abnormalities of thedeveloping axon [1]. The genetics of HSP is complex and allmendelian modes of inheritance have been described [1]. Mostcases of autosomal dominant HSP are pure, whereas complicatedforms tend to be autosomal recessive; SPG4 (SPAST, spastin),SPG3A and SPG31 (REEP1) are the most common causes of

D. Orsucci et al. / Clinical Neurology and Neurosurgery 120 (2014) 14–19 15

autosomal dominant pure HSP [1]. Mutations in the SPG7 gene(paraplegin) are the most common cause of autosomal recessiveHSP, with both pure and complicated phenotypes [1]. Moreover, asignificant number of apparently sporadic cases has an SPGmutation, suggesting that these patients should not be excludedfrom genetic studies [7].

Aimof this study is the clinical andmolecular characterizationofa big cohort of patients with HSP evaluated at the NeurogeneticService of our clinic in last two years (2011–2012). Moreover, weaim to study the minimum prevalence of HSP in our area and topropose a schematic diagnostic approach to HSP patients based onthe available data from the literature.

2. Methods

2.1. Patients

We retrospectively reviewed the clinical and genetic data ofsubjects with clinical signs and symptoms indicative of pure orcomplicated HSP, in whom other possible diagnosis (i.e., multiple

Table 1Clinical features of HSP patients in our group.

No. Onset (years) Last eval. Follow up Pure/compl. Sex 4SMD

1 40 48 6 Pure F 12 44 46 2 Pure F 23 30 50 19 Pure F 34 25 34 7 Pure F 25 42 44 2 Pure M 16 43 48 3 Pure F 17 45 62 13 Pure M 38 10 61 5 Pure M 29 42 47 5 Pure M 1

10 50 76 10 Pure F 311 65 70 2 Pure F 112 30 53 10 Pure M 213 37 40 3 Pure M 214 27 38 10 Compl. M 215 35 57 5 Compl. F 216 46 51 5 Pure M 217 53 59 4 Compl. M 118 34 37 2 Pure M 219 26 38 5 Pure M 220 26 38 10 Pure F 421 27 69 13 Compl. M 322 24 33 9 Compl. F 323 55 60 5 Pure F 324 46 69 16 Pure F 225 43 64 21 Compl. F 226 12 41 9 Pure F 227 50 65 6 Pure M 328 49 50 1 Pure M 129 58 68 10 Compl. F 130 55 58 2 Compl. F 331 58 67 7 Pure M 232 32 34 2 Pure M 333 16 30 2 Compl. F 334 12 43 10 Pure F 235 62 64 1 Pure M 236 29 33 2 Compl. F 237 30 49 10 Pure M 238 21 42 20 Compl. F 239 53 58 5 Pure F 140 63 66 3 Pure F 441 24 28 4 Pure F 242 12 41 16 Pure F 243 10 51 2 Compl. F 244 50 62 4 Pure M 245 40 64 9 Pure F 2

a Apart from SPG4, 3A, 31 (analyzed in all patients). 4SMD, four-stage scale of motor dfeatures (compl.): cerebellar atrophy in patients 14, 15, 17 and 36; distal amyotrophy anddysarthria in patients 25 and 30; cognitive involvement in patients 29 and 30; thin cor

sclerosis, leucodystrophy, mitochondrial diseases, etc.) wereexcluded by appropriate neuroradiological, neurophysiologic andlaboratory studies. Those HSP cases (n=45) have been evaluated inlast two years (2011–2012).

A simple, four-stage functional scale ofmotordisability (1 =mildsymptoms,walkingwithout aid;2 =walkingwithout aidbutunableto run; 3 =walking with aid; 4 =wheelchair-dependent) [5] hasbeen utilized (Table 1). Moreover, in a subset of patientsperspectively evaluated in 2012, we could also perform standard-ized functional evaluations by means of the SF-36 scale of health-related quality of life [8] and the spastic paraplegia rating scale(SPRS) [9].

Magnetic resonance imaging (MRI) of the brain and the spinalcord, as well as appropriate neurophysiological and laboratoryinvestigations (including in some instances genetic studies forspino-cerebellar ataxias, Friedreich’s ataxia, ARSACS, mitochon-drial DNA mutations, etc.), have been performed in all patients inorder to exclude secondary cases of the paraparesis and to findeventual “complicating” features (e.g., neuropathy, thinning ofthe corpus callosum, etc.). Patients without genetically confirmed

SF36 SPRS Family history Screened genes (SPG)a Genetic diagnosis

618 3 AD SPG4387 11 AD 7 SPG4516 22 AD SPG4374 16 AD SPG4446 6 AD SPG4629 3 AD SPG4394 23 AD SPG4604.5 8 No SPG4498 5 AD SPG4

AD SPG4230 11 AD SPG4661 17 AD SPG4362 16 No SPG4323 14 ar 7 SPG7532 9 ar 7,17 SPG7175 21 No 7 SPG7638 1 No 7 SPG7

No 7 SPG7393 10 AD 7,10,17 SPG10302 29 No 7,10,11 SPG11244 23 AD SPG17

No 7,11 Unknown164 24 No 7,10,11 Unknown233 14 No 7,10,11 Unknown256 20 AD 7,10 Unknown200 4 No 7,10,11 Unknown409 23 AD 7,10 Unknown

No 7 UnknownNo 6,7,13 Unknown

397 13 No 7,11,15 Unknown450 14 No 7,10,11 Unknown310 22 No 7 Unknown467 23 No 7,10,11 Unknown

No 5,7,10,11 Unknown223 11 No 7,10,11 Unknown441 14 No 7 Unknown

No 7 UnknownNo 6,7,10,11,13,15 UnknownNo 7,10,11 Unknown

410 30 AD 7,10 Unknown613 15 No 7,10,11 Unknown527 17 No 7 Unknown600.5 10 No 7, 10,11 Unknown587 10 No 7,10,11 Unknown501 9 No 7,10,11 Unknown

isability (see text); AD, autosomal dominant; ar, autosomal recessive. Complicatinghearing loss in patient 21; white matter involvement in patients 21, 22, 38 and 43;

pus callosum in patient 30; ataxia (without cerebellar atrophy) in patient 33.

16 D. Orsucci et al. / Clinical Neurology and Neurosurgery 120 (2014) 14–19

diagnosis or clear family history were included only after carefulexclusion of other known causes of spastic paraplegia. Moreover,the majority of our patients have undergone a neurologicalfollow-up of several years (see Table 1), further reinforcing thediagnosis of HSP (e.g., differential diagnosis with primary lateralsclerosis) [10].

Genetic analysis (automatic sequencing and deletion researchby multiplex ligation-dependent probe amplification assay) havebeen requested for SPG4, SPG3A, SPG31 in all patients with clearautosomal dominant family history, and for SPG4, SPG3A, SPG31,SPG7 in all patients without clear dominant inheritance. Furthermolecular studies have been performed in most undiagnosedpatients (i.e., SPG10, SPG11, etc.) on the basis of the clinical picture[11] (see Table 1).

2.2. Statistical analysis

Datawere expressed asmeans� standard deviation. Analysis ofdatawas carried out usingMedCalc1 Version 7.3.0.1. The datawerecorrelated by Kendall’s coefficient of rank correlation (t), orcompared by unpaired two-tailed Student’s t-test. A P value <0.05was considered as significant.

3. Results

3.1. Patient features

A total of 45 patients were identified. Male subjects were 19/45(42.2%). 6/45 patients (13.3%) had childhood onset (�16 years). 32/45 patients had a pure HSP phenotype (71.1%). The molecularly-confirmed patients (Table 2) were 21/45 (46.7%): 13 were SPG4-positive (28.9%), five SPG7 (11.1%), one SPG10, one SPG11, one SPG17-mutated (2.2% each). Despite extensive search for SPG3A and SPG31mutations in all patients, no subjects with these genotypes werepresent in our series. The mean age at onset was 37.4�15.4 years,the mean age at last evaluation 51.2�12.8 years, and the meanfollow-up period in our center 7.0�5.3 years.

Three patients lived in Pisa town (population 87,215), nine inPisa province (population 412,729), 29 in “Area Vasta Nord-Ovest”in Tuscany (provinces of Massa Carrara, Livorno, Lucca and Pisa –

population 1,338,450) (Italian Institute of Statistics – ISTAT, “2011

Table 2Molecular findings in our genetically-confirmed patients.

No. Gene Mutation(s)

1 SPG4 c.1210–1212delTTT deletion (p.Phe42 SPG4 c.343del10 frameshift ten-nucleotid3 SPG4 c.1684C>T stop mutation (p.Arg5624 SPG4 c.806C>G stop mutation (p.Tyr2695 SPG4 c.532C>T stop mutation (p.Gln178s6 SPG4 c.1243_1244insA insertion (framesh7 SPG4 c.1281dupT frameshift insertion (p.8 SPG4 c.870+3A>G mutation (with deletio9 SPG4 c.1450G>C substitution (p.Gly484A

10 SPG4 c.1243_1244insA insertion (framesh11 SPG4 c.1243_1244insA insertion (framesh12 SPG4 c.1243_1244insA insertion (framesh13 SPG4 macrodeletion of exons 1–1714 SPG7 macrodeletion of exons 13–17 and15 SPG7 p.Ala510Val and c.2213_2214insA (f16 SPG7 homozygous c.233T>A stop mutati17 SPG7 homozygous c.1529C>T mutation (18 SPG7 homozygous c.1617delC frameshift19 SPG10 c.611G>A substitution (p.Arg204Gl20 SPG11 c.6883_6885delACC (p.Thr2295_2221 SPG17 c.263A>G transition (p.Asn88Ser)

Population and housing census”, http://dati.istat.it/Index.aspx?lang=en), with a minimum prevalence of at least 3.43/100,000;2.18/100,000 and 2.17/100,000, respectively.

3.2. Neuroimaging findings

Cerebellar atrophywas present in four patients: 14,15,17 (SPG7-mutated) and 36 (see Table 1). White matter involvement was alsoobserved in four subjects: 21 (SPG17mutation), 22, 38 and 43. Thincorpus callosumwas present in patient 30. Neuroimaging findingswere unremarkable in the other patients.

3.3. Quality of life and clinical scales

These scales were available for the 36 patients. SF-36 and SPRSscore were inversely related between them (P = 0.012) (Fig. 1A). Ofnote, the four-stage scale ofmotordisability [5], not yet validatedorcompared with other well-known quality of life or HSP diseaseprogression scales, could predict the SPRS scores with a highlysignificant P<0.0001 (Fig. 1B). Moreover, this simplified scale wasinversely relatedwith the SF-36 scale (n=36, t =�0.248, P = 0.032).

The clinical scales were not related with age at onset or age atlast evaluation.

Urinary urgency were reported in 24/45 patients (53.3%). Theproportion of SPG4 and/or SPG7 mutations was not different inpatientswith urinary symptoms comparedwith the other patients.Moreover, no significant differences in quality of life scores norother clinical parameters (i.e., actual age and age at onset) wereobserved between the groupwith urinary symptoms and the othersubjects.

3.4. SPG4 patients

Age at onset (38.7�13.2 years versus 36.8�16.3 years) andage at last evaluation (52.2�11.9 years versus 50.8�13.3 years)did not differ between the 13 SPG4 patients and the other 32subjects. The clinical and quality of life scores were notsignificantly different between the two groups. All the SPG4subjects had a pure HSP phenotype. 13/32 patients with pure HSP,in our series, had SPG4 mutations (40.6%). Only two patients withSPG4 did not refer a clearly dominant family history; of the 16patients with a dominant family history, 11 had SPG4 mutations

Reference

04del) [7]e deletion (with stop codon p.Arg115stop) unreportedstop) [7]stop) [7]top) unreportedift with p.Tyr415Stop) [13]Ala428CysfsX442) [12]n of exon 5 in the transcript) [15]rg) unreportedift with p.Tyr415stop) [13]ift with p.Tyr415stop) [13]ift with p.Tyr415stop) [13]

[14]c.1529C>T mutation (p.Ala510Val) [19]rameshift) with absent paraplegin reactivity [19], unreportedon (p.Leu78stop) [18]p.Ala510Val) [19]mutation (leading to p.Ser539stop) [18]n) [17]96del) and c.7000G>C (p.Ala2334Pro) both unreported

[16]

[(Fig._1)TD$FIG]

Fig. 1. Quality of life in patients with hereditary spastic paraparesis. (A) The SF-36 scores were inversely related with the SPRS scores (n =36, t =�0.293, P =0.012). (B) Thesimplified scale of motor disability correlated with the SPRS score with a high statistical significance (n =36, t = 0.722, P<0.0001).

D. Orsucci et al. / Clinical Neurology and Neurosurgery 120 (2014) 14–19 17

(68.8%), one SPG10, one SPG17 and three are still underinvestigation (see Table 1).

3.5. SPG7 patients

Age at onset (39.0�10.4 years versus 37.2�16.0 years) and ageat last evaluation (48.4�10.4 years versus 51.6�13.1 years) did notdiffer between the five SPG7 patients and the other 40 subjects. Theclinical and quality of life scores were not significantly differentbetween the two groups. Two SPG7 patients had a pure HSPphenotype, the other three had cerebellar atrophy and mildcerebellar signs at the neurological examination. Two patient had afamily history indicative of a recessive disease, the other three hadno family history.

4. Discussion

We reviewed the clinical and genetic data of 45 subjects withclinical signs and symptoms indicative of pure or complicated HSP.

The extensive genetic analysis allowed us to obtain a definitivegeneticdiagnosis in abouthalf of patients.Wehavealsodetected sixnovel mutations, not previously associated with HSP (see Table 2).Theminimumprevalence ofHSP in our areawas of about 2.17–3.43/100,000.

The three most common autosomal dominant HSPs (SPG3A,SPG4, and SPG31), as well as the recessive SPG11, result frommutations in proteins which are in concert implicated in theformation of the tubular endoplasmic reticulum network, withimpairment of the relationship between endoplasmic reticulumtubules and the microtubule cytoskeleton [2]. SPG17 proteinlocalize to the endoplasmic reticulum also [2]. In SPG10 there ismotor-based transport impairments (due to mutation in an ATP-dependent motor that move cargoes in the anterograde directionalong axons) [2]. In SPG7, the most common recessive HSP, bothmitochondrial function and axonal transport are impaired [2].

Consensus guidelines [11] state that patients with pure HSP anda dominant family history of spastic paraparesis should be tested

[(Fig._2)TD$FIG]

Fig. 2. A possible approach to the genetic diagnosis of hereditary spastic paraparesis. X-linked forms (SPG1 and 2) have not been considered here because they cause complexneurological syndromesof the childhood. Recessive formshavebeen considered togetherwith the apparently sporadic ones, because only rarely it is possible to establish that agiven patient probably has a recessive disease (e.g., if the parents are consanguineous). However, when there is strong evidence of recessive inheritance, SPG7 is first analyzed,whereas SPG11 (and SPG15 in complicated cases) are studied in second instance.This table is conceived as a flow chart, but the real diagnostic process depends on the specific clinical presentation of the patient, and is not a rigid process. For example, a given“complicating” features will make a specific genetic diagnosis more probable (e.g., cerebellar involvement for SPG7 and 15; thin corpus callosum for SPG11 and 15; cognitiveimpairment forSPG11and15aswell;distal amyotrophyforSPG15and17;neuropathy forSPG10and15), andthis shouldbeconsideredat thebeginningof thediagnosticwork-up.

18 D. Orsucci et al. / Clinical Neurology and Neurosurgery 120 (2014) 14–19

for SPG4 (direct sequencing and subsequently deletion search). As athird step, sequencing of atlastin (SPG3A) in subjects with a pureform and onset under 20 years is recommended; sequencing ofREEP1 (SPG31) and KIF5A (SPG10) can be considered in remainingmutation-negative dominant families, the latter particularly whena neuropathy is present [11]. Molecular testing first for SPG11 andsecond SPG15 is recommended in recessive HSP and thin corpuscallosum [11]. SPG7 may be tested especially when cerebellarfeatures are present [11]. For other recessive forms, no generalrecommendation can be given. Sporadic subjects with progressivespastic paraparesis where other causes of spasticity have beencarefully excluded should be tested for SPG4 mutations includingdeletion search. In negative cases, sequencing of SPG7 may beproposed [11]. These points are reasonable. However, a moreextensive approach may be warranted; a possible diagnostic flow-chart (based on the available data from the literature [11]) is shownin Fig. 2.

Our study confirms that the most common form of autosomaldominant HSP is caused by mutations in the SPG4 gene, encodingspastin. While epidemiological available data reveal that SPG4mutations account for up to 45% of autosomal dominant forms [12]and �20% of sporadic cases [13], in our series the frequency isdifferent: 68.8% and 7.4%, respectively. The higher proportion ofSPG4mutations in our patient is likely due to a founder effect (e.g.,patients 6, 10, 11, 12 probably come from the same ancestor – seeTable 2).

SPG4 mutations more commonly lead to pure forms of thedisease. Age at onset is highly variable, ranging from infancy up tothe eighth decade. All types of DNA alterations are observed,including missense, non-sense, insertions or deletions [14], andrare splice-site mutations [15]; the pathogenic mechanism ishaploinsufficiency (the disease occurs once the level of functionalspastin falls below a critical level) [14]. SPG3A has been reported as

the second most frequent form of dominant HSP (more commonlyas a pureHSP) [7]. However, despite extensive screening,wedidnotidentify any SPG3Amutations, and this form seems very rare in theItalian population. Among the dominant forms, we have alsoidentified a SPG17patientwith a “Silver syndrome”phenotype (HSPwith distal amyotrophy) [16], and a SPG10 patient with pure HSP.SPG10 may lead to both pure and complicated phenotypes (e.g.,Silver syndrome-like) [17].

Themost frequent of the recessive forms in our population is theSPG7, as expected [18,19]. We have also identified a SPG11-positivepatient, who had a pure HSP phenotype.

Age at onset, clinical and quality of life scores were not differentbetween SPG4 or SPG7 patients and the other patients. In our group,the SF-36 (quality of life) and SPRS (disease progression) scoreswere inversely related, confirming that health-relatedqualityof lifeis a valid parameter in HSP that should be considered in upcomingtherapeutic trials [20]. This data was expected and confirmedprevious studies [20]. No effect of the HSP genotype on quality oflife was observed. Notably, the time-saving, four-stage scale ofmotor disability [5] (suggested acronym 4SMD), not validated orcompared with other scales to date, could predict the SPRS scoreswith a very high statistical significance (see Fig. 1B), and wasinversely related with the SF-36 scale. Therefore, we encourage itsroutinaryuse inHSPpatients; further studies areneeded to confirm(or not) its possible usefulness in other neurological andneuromuscular diseases as well.

5. Conclusion

Our study confirms SPG4 as the major cause of HSP; of note, allSPG4 patients had a pure HSP phenotype, and the dominantinheritancewasevident in thegreatmajorityof these subjects. SPG7was the second genetic cause; cerebellar atrophy was a frequent

D. Orsucci et al. / Clinical Neurology and Neurosurgery 120 (2014) 14–19 19

(60%) “complicating feature” in SPG7 patients. Other genotypeswere rarer (SPG10, SPG11, SPG17) or absent (SPG3A, SPG31) in ourseries.

An exactmolecular diagnosis will allowamore accurate patientcounseling and, hopefully, will lead to specific, targeted, therapeu-tic options for these chronic, still incurable diseases.

References

[1]

Salinas S, Proukakis C, Crosby A, Warner TT. Hereditary spastic paraplegia:clinical features and pathogeneticmechanisms. Lancet Neurol 2008;7:1127–38.

[2]

Blackstone C. Cellular pathways of hereditary spastic paraplegia. Annu RevNeurosci 2012;35:25–47.

[3]

McMonagle P, Webb S, Hutchinson M. The prevalence of “pure’’ autosomaldominant hereditary spastic paraparesis in the island of Ireland. J NeurolNeurosurg Psychiatry 2002;72:43–6.

[4]

Silva MC, Coutinho P, Pinheiro CD, Neves JM, Serrano P. Hereditary ataxias andspastic paraplegias:methodological aspects of a prevalence study in Portugal. JClin Epidemiol 1997;50:1377–84.

[5]

Erichsen AK, Koht J, Stray-Pedersen A, Abdelnoor M, Tallaksen CM. Prevalenceof hereditary ataxia and spastic paraplegia in southeast Norway: a population-based study. Brain 2009;132:1577–88.

[6]

Braschinsky M, Luus SM, Gross-Paju K, Haldre S. The prevalence of hereditaryspastic paraplegia and the occurrence of SPG4 mutations in Estonia. Neuro-epidemiology 2009;32:89–93.

[7]

Alvarez V, Sanchez-Ferrero E, Beetz C, Diaz M, Alonso B, Corao AI, et al.Mutational spectrum of the SPG4 (SPAST) and SPG3A (ATL1) genes in Spanishpatients with hereditary spastic paraplegia. BMC Neurol 2010;10:89.

[8]

Apolone G, Mosconi P. The Italian SF-36 health survey: translation, validationand norming. J Clin Epidemiol 1998;51:1025–36.

[9]

Schule R, Holland-Letz T, Klimpe S, Kassubek J, Klopstock T, Mall V, et al. Thespastic paraplegia rating scale (SPRS): a reliable and valid measure of diseaseseverity. Neurology 2006;67:430–4.

[10]

Brugman F, Veldink JH, Franssen H, de Visser M, de Jong JM, Faber CG, et al.Differentiation of hereditary spastic paraparesis from primary lateral sclerosisin sporadic adult-onset upper motor neuron syndromes. Arch Neurol2009;66:509–14.

[11]

Gasser T, Finsterer J, Baets J, Van Broeckhoven C, Di Donato S, Fontaine B, et al.EFNS guidelines on the molecular diagnosis of ataxias and spastic paraplegias.Eur J Neurol 2010;17:179–88.

[12]

Magariello A, Muglia M, Patitucci A, Mazzei R, Conforti FL, Gabriele AL, et al.Novel spastin (SPG4) mutations in Italian patients with hereditary spasticparaplegia. Neuromuscul Disord 2006;16:387–90.

[13]

Magariello A, Muglia M, Patitucci A, Ungaro C, Mazzei R, Gabriele AL, et al.Mutation analysis of the SPG4 gene in Italian patients with pure andcomplicated forms of spastic paraplegia. J Neurol Sci 2010;288:96–100.

[14]

DepienneC, FedirkoE, Forlani S, CazeneuveC,Ribai P, Feki I, et al. Exondeletionsof SPG4 are a frequent cause of hereditary spastic paraplegia. J Med Genet2007;44:281–4.

[15]

Lim JS, Sung JJ, Hong YH, Park SS, Park KS, Cha JI, et al. A novel splicingmutation(c.870+3A>G) in SPG4 in a Korean family with hereditary spastic paraplegia.J Neurol Sci 2010;290:186–9.

[16]

Windpassinger C, Auer-Grumbach M, Irobi J, Patel H, Petek E, Horl G, et al.Heterozygous missense mutations in BSCL2 are associated with distalhereditary motor neuropathy and Silver syndrome. Nat Genet2004;36:271–6.

[17]

Goizet C, Boukhris A, Mundwiller E, Tallaksen C, Forlani S, Toutain A, et al.Complicated forms of autosomal dominant hereditary spastic paraplegia arefrequent in SPG10. Hum Mutat 2009;30:E376–85.

[18]

Arnoldi A, Tonelli A, Crippa F, Villani G, Pacelli C, Sironi M, et al. A clinical,genetic, and biochemical characterization of SPG7 mutations in a largecohort of patients with hereditary spastic paraplegia. Hum Mutat2008;29:522–31.

[19]

Sanchez-Ferrero E, Coto E, Beetz C, Gamez J, Corao A, Diaz M, et al. SPG7mutational screening in spastic paraplegia patients supports a dominant effectfor some mutations and a pathogenic role for p.A510V. Clin Genet 2012;.

[20]

Klimpe S, Schule R, Kassubek J, Otto S, Kohl Z, Klebe S, et al. Disease severityaffects quality of life of hereditary spastic paraplegia patients. Eur J Neurol2012;19:168–71.