The n e w e ng l a n d j o u r na l of m e dic i n e

Review Article

Dan L. Longo, M.D., Editor

Multiple-System Atrophy
Alessandra Fanciulli, M.D., and Gregor K. Wenning, M.D., Ph.D.

M
ultiple-system atrophy is an adult-onset, fatal neurodegen- From the Department of Neurology, Divi-
erative disease characterized by progressive autonomic failure, parkinso- sion of Neurobiology, Innsbruck Medical
University, Innsbruck, Austria (A.F., G.K.W.);
nian features, and cerebellar and pyramidal features in various combina- and the Department of Neuroscience,
tions. It is classified as the parkinsonian subtype if parkinsonism is the predominant Mental Health and Sensory Organs, Sa-
feature and as the cerebellar subtype if cerebellar features predominate. pienza University of Rome, Rome (A.F.).
Address reprint requests to Prof. Wen-
With its variable clinical presentations, multiple-system atrophy presents a ma- ning at the Department of Neurology,
jor diagnostic challenge not only in neurology but also in other specialties, including Division of Neurobiology, Innsbruck Med-
cardiology, gastroenterology, urology, otolaryngology, and sleep medicine. Despite ical University, Anichstr. 35, A-6020 Inns-
bruck, Austria, or at g ­regor​ .­
wenning@​
having faster motor progression, multiple-system atrophy may masquerade as ­i-med​.­ac​.­at.
Parkinson’s disease or idiopathic late-onset cerebellar ataxia until advanced stages
N Engl J Med 2015;372:249-63.
of the disease. DOI: 10.1056/NEJMra1311488
The history of multiple-system atrophy reflects the varied clinical manifesta- Copyright © 2015 Massachusetts Medical Society.
tions of the disease. The term multiple-system atrophy was first coined in 1969 to
pool three previously described neurologic entities: olivopontocerebellar atrophy,
the Shy–Drager syndrome, and striatonigral degeneration.1 These entities correspond
to multiple-system atrophy with predominantly cerebellar, autonomic, or parkin-
sonian features, respectively.

Epidemiol o gic Fe at ur e s
Multiple-system atrophy is an orphan disease (Orpha number, ORPHA102 [www​
.­orpha​.­net]). The estimated mean incidence is 0.6 to 0.7 cases per 100,000 person-
years, with a range of 0.1 to 2.4 cases per 100,000 person-years.2 The estimated
point prevalence is 3.4 to 4.9 cases per 100,000 population, increasing to 7.8 per
100,000 among persons older than 40 years of age.3 Cases of the parkinsonian
subtype outnumber cases of the cerebellar subtype in most countries by 2:1 to 4:1,4-6
although the cerebellar subtype is more frequent in Japan,7 with genetic or epigen-
etic factors possibly exerting an influence. Disease onset is usually in the sixth de-
cade of life, with both sexes equally affected.8 The mean survival from the onset of
symptoms is 6 to 10 years,8-10 with few patients surviving more than 15 years.11

C ause s
No environmental factors are known to affect the risk of multiple-system atrophy.
As in Parkinson’s disease, nicotine use and alcohol consumption are less common
among patients with multiple-system atrophy than among healthy controls, which
indicates a possible pathophysiological link between Parkinson’s disease and multi-
ple-system atrophy.
Multiple-system atrophy is generally considered a sporadic disease. Nevertheless,
genetic factors play an etiologic role in some families. In a few European and Japa-
nese pedigrees, multiple-system atrophy has been transmitted in an autosomal domi-

n engl j med 372;3 nejm.org  January 15, 2015 249

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

nant or recessive inheritance pattern.12,13 Recently, Figure 1 (facing page). Pathophysiological Features of
a loss-of-function mutation in COQ2, encoding the Multiple-System Atrophy (MSA).
coenzyme Q10–synthesizing enzyme, was report- Genetic and environmental factors may contribute to
ed in Japanese familial and sporadic cases, but the initiation of the pathophysiological cascade of
the mutation was not detected in patients with MSA. Relocalization of p25α into the oligodendroglial
multiple-system atrophy who were from North soma is an early event, followed by cellular swelling
and abnormal uptake or overexpression of α-synuclein
America or Europe.14 Similarly, a discordant loss by oligodendroglia. α-Synuclein and p25α precipitate
of copy number of SHC2 was found in monozy- into glial cytoplasmic inclusions, which hinder neuro-
gotic twins and Japanese patients with sporadic nal trophic support and induce microglial activation.
multiple-system atrophy but not in patients in Misfolded α-synuclein released by dysfunctional oligo-
the United States.15,16 Mutations, duplications, and dendrocytes may be taken up by neighboring neurons
to form neuronal cytoplasmic inclusions. Neuroin-
triplications of SNCA, encoding α-synuclein, may flammation, loss of neurotrophic support, and neuro-
cause familial Parkinson’s disease with features nal dysfunction due to α-synuclein inclusions syner-
similar to multiple-system atrophy in some af- gistically contribute to neuronal death in the
fected persons.17 Two single-nucleotide polymor- striatonigral, olivopontocerebellar, and central auto-
phisms of the SNCA locus showed a significant nomic pathways, resulting in parkinsonism that is
poorly responsive to levodopa, cerebellar ataxia, and
association with multiple-system atrophy in a multidomain autonomic failure.
large series of European patients.18 This associa-
tion was confirmed in follow-up replication
studies19 but not in the preliminary analysis of ings.25 The main constituent of glial cytoplasmic
the first genomewide association study of multi- inclusions is misfolded α-synuclein, a protein nor-
ple-system atrophy.20 A G51D SNCA mutation was mally located in neuronal axons and synapses.
recently reported in a British pedigree with auto- Hence, multiple-system atrophy is classified as an
somal dominant juvenile parkinsonism and neu- oligodendroglial α-synucleinopathy, whereas Par-
ropathological findings compatible with both Par- kinson’s disease, dementia with Lewy bodies, and
kinson’s disease and multiple-system atrophy.21 pure autonomic failure26 are characterized by neu-
ronal α-synuclein aggregates (Lewy bodies).
Neuropathol o gic a l Fe at ur e s
Patho gene sis
Variable degrees of olivopontocerebellar atrophy
and striatonigral degeneration are typically found Although the pathogenic mechanisms underlying
at postmortem examination of patients with mul- multiple-system atrophy remain partially unclear,
tiple-system atrophy, which broadly reflect the converging evidence from preclinical models and
presence of ataxia and parkinsonism during life.22 postmortem studies suggests that it is a primary
In addition, neurodegenerative changes affect the oligodendrogliopathy (Fig. 1).22,27 Relocalization
central autonomic nervous system, including the of p25α, an important stabilizer of myelin in-
hypothalamus, noradrenergic and serotoninergic tegrity, into the oligodendroglial soma appears
brain-stem nuclei, dorsal nucleus of the vagus to precede α-synuclein aggregation.28 This is
nerve, nucleus ambiguus, intermediolateral col- followed by oligodendrocyte swelling and ab-
umns of the spinal cord, and Onuf nucleus.23 normal uptake or overexpression of α-synuclein
Frontal-lobe atrophy may be observed after a long by oligodendroglia.29,30 The interaction between
disease duration. p25α and α-synuclein promotes phosphorylation
Proteinaceous oligodendroglial cytoplasmic and aggregation of synuclein into insoluble
inclusions (also called Papp–Lantos bodies) are the oligomers first and glial cytoplasmic inclusions
histologic hallmark of multiple-system atrophy.24 later on. The formation of glial cytoplasmic in-
Less frequently, oligodendroglial nuclear, neuronal clusions, in turn, interferes with neuronal sup-
axonal, cytoplasmic, and nuclear inclusions can be port and activates quiescent microglial cells. As
observed as well. The density of glial cytoplasmic a result, progressively dysfunctional oligoden-
inclusions broadly reflects the distribution of neu- drocytes release misfolded α-synuclein into the
rodegenerative changes in the brains of patients extracellular space; this misfolded α-synuclein
with multiple-system atrophy.22 Activated microglia may be taken up by neighboring neurons to form
and reactive astrogliosis are other common find- neuronal cytoplasmic inclusions. At this point,

250 n engl j med 372;3 nejm.org  January 15, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Multiple-System Atrophy

Possible genetic factors Possible environmental factors

COQ2 mutation Agricultural employment
SHC2 copy number loss Organic solvents
SNCA multiplications or single-nucleotide polymorphisms Plastic monomers
Unidentified autosomal recessive or dominant mutations Pesticides
Metal dusts

Oligodendrocyte α-Synuclein–p25α
Relocalization of swelling glial cytoplasmic
p25α into the inclusions
oligodendrocyte
soma
Neuron
Ectopic α-synuclein
localization

Microglial
activation

Inflammatory
mediators

Astroglial
activation

Neuronal
Neuronal
Loss of cytoplasmic
degeneration
oligodendroglial inclusions
neurotrophic support

Possible prion-like
propagation to
functionally connected
areas

Striatonigral degeneration Olivopontocerebellar atrophy Degeneration of brain stem and

medullary autonomic nuclei

Poorly levodopa-responsive parkinsonism Cerebellar syndrome Multidomain autonomic failure

n engl j med 372;3 nejm.org January 15, 2015 251

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

neuroinflammation, loss of oligodendroglial neu- disproportionate antecollis (severe forward neck

rotrophic support, and neuronal dysfunction due flexion that interferes with eating, speaking, and
to α-synuclein inclusions may synergistically pro- vision), as well as hand or foot dystonia, are as-
mote neuronal death and subsequent reactive sociated with the motor presentation of multi-
astrogliosis. Toxic α-synuclein species may then ple-system atrophy in 16 to 42% of patients.36
spread in a prion-like fashion to other function- Recurrent falls, dysphonia (voice-tone changes),
ally connected brain areas,31 leading to the multi- dysarthria (difficulty in articulating words), drool-
system neuronal involvement that is typical of ing, and dysphagia are defining features of ad-
multiple-system atrophy. vanced disease.

Nonmotor Features
Cl inic a l Pr e sen tat ion
Early and severe autonomic failure is a key feature
Like Parkinson’s disease, multiple-system atrophy of multiple-system atrophy, and the most frequent-
has a prodromal premotor phase in 20 to 75% of ly affected domains are urogenital and cardio-
cases, including sexual dysfunction, urinary urge vascular. Erectile dysfunction typically occurs at
incontinence or retention, orthostatic hypotension, disease onset in male patients. Genital hyposen-
inspiratory stridor, and rapid-eye-movement sleep sitivity during intercourse characterizes the sexual
behavior disorder months to years before the first dysfunction in women. Urinary dysfunction in-
motor symptoms appear.32 cludes urinary urgency and frequency, urge incon-
tinence, nocturia, and, less commonly, incomplete
Motor Features bladder emptying.37 Urinary failure may be masked
Parkinsonism, with slowness of movements, ri- by concomitant prostatic hypertrophy in men or
gidity, and a tendency to fall, characterizes the perineal laxity due to multiple labors or uterine
motor presentation of the parkinsonian subtype prolapse in women. Remarkably, as many as 50%
of multiple-system atrophy.5 The motor findings of these patients undergo futile genitourinary sur-
are sometimes asymmetrical and may be mark- gery before the correct diagnosis is determined.38
edly so. Parkinson-like “pill-rolling” rest tremor is Severe orthostatic hypotension, defined as a
unusual, whereas irregular postural and action blood pressure decrease of 30 mm Hg systolic or
tremor with superimposed jerks is seen in as many 15 mm Hg diastolic within 3 minutes after a pas-
as 50% of patients with multiple-system atro- sive head-up tilt or standing from the recumbent
phy.5 Progressive degeneration of the striatum ac- position,39 is the main feature of cardiovascular
counts for the poor response or lack of response autonomic failure in clinically established multi-
to levodopa, which is a mandatory diagnostic cri- ple-system atrophy. It is manifested as recurrent
terion for probable multiple-system atrophy of the syncope, light-headedness (dizziness), weakness,
parkinsonian subtype. Nevertheless, a transient nausea, tremulousness, headache, or “coat-hanger
response to levodopa may be observed in approxi- pain” (pain in the neck and shoulder region) on
mately 40% of patients during early disease stag- standing, but it may also be asymptomatic.40 Post-
es33; this is sometimes accompanied by drug-in- prandial hypotension and supine and nocturnal
duced involuntary movements, such as head–neck hypertension accompany orthostatic hypotension
dystonia, an involuntary muscle contraction re- in half of patients with multiple-system atrophy.41
sulting in abnormal twisting postures.34 Respiratory disturbances are characteristic of
Cerebellar ataxia predominates in the motor multiple-system atrophy. Diurnal or nocturnal in-
presentation of the cerebellar subtype of multiple- spiratory stridor develops in as many as 50% of
system atrophy.5,35 Cerebellar features consist of patients at some time36 but is more frequent in
a wide-based gait, uncoordinated limb movements, advanced disease than in earlier stages, and sleep
action tremor, and spontaneous, gaze-evoked, or apneas affect about 40% of patients. Episodic noc-
positional downbeat nystagmus. Spasticity or py- turnal inspiratory stridor and sleep apnea may
ramidal weakness should cast doubts on a diag- occur together.42 Other features of autonomic
nosis of multiple-system atrophy, but generalized failure in multiple-system atrophy include consti-
hyperreflexia, as well as a Babinski sign, may pation,43 pupillomotor abnormalities, and vaso-
occur in 30 to 50% of cases.5 motor and thermoregulatory failure with sweat-
Abnormal postures, including bent spine and ing that is diminished and ultimately absent.44,45

252 n engl j med 372;3 nejm.org  January 15, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Multiple-System Atrophy

Neurology
Psychiatry
Parkinsonism
Depression
Cerebellar features
Anxiety
Pyramidal signs
Frontal executive dysfunction
Sleep Medicine
REM sleep behavior disorder
Ear, Nose, and Throat
Sleep apneas
Stridor
Excessive daytime sleepiness
Dysarthria
Restless legs syndrome
Dysphonia
Dysphagia
Cardiology
Pneumology Syncope
Aspiration pneumonia Orthostatic hypotension
Postprandial hypotension
Nocturnal hypertension
Leg edema

Gastroenterology
Dysphagia
Constipation
Diarrhea

Dermatology
Hypohidrosis or anhidrosis
Vasomotor abnormalities

Urology
Sexual dysfunction
Urinary urgency and frequency
Nocturia
Urinary urge incontinence
Urinary retention
Recurrent urinary tract infections

Figure 2. Multidisciplinary Presentation of MSA.

Dementia or visual hallucinations are not priate emotional context) can occur,36 as well as
consistent with a diagnosis of multiple-system behavioral changes, including depression, anxi-
atrophy; these symptoms in the presence of par- ety, panic attacks, and suicidal ideation. Finally,
kinsonism and autonomic failure should prompt as many as 50% of patients report disabling
consideration of dementia with Lewy bodies.39 pain; advanced disease, dystonia, and female sex
However, frontal-lobe dysfunction with attention are possible risk factors for this symptom.47 A
deficits has been reported in one third of cas- visual summary of the varied clinical manifesta-
es.46 Emotional incontinence (i.e., inappropriate tions of multiple-system atrophy is provided in
laughing or crying in the absence of the appro- Figure 2.

n engl j med 372;3 nejm.org January 15, 2015 253

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Recurrent falls
Sexual dysfunction
Indwelling catheter
Urinary dysfunction Frontal executive dysfunction
REM sleep behavior disorder
Unintelligible speech
Orthostatic hypotension
Gastrostomy
Stridor
Tracheostomy
Parkinsonism
Bronchopneumonia
Cerebellar features
Uroseptic fever
Multidomain autonomic failure
Sudden death
Pyramidal signs

Stage at which death can occur

Probable MSA
Premotor MSA Possible MSA
0 3 6 9
Years

Figure 3. Natural History of MSA.

The premotor phase of MSA can last for months to years. Year 0 denotes the time of onset of motor symptoms. A diagnosis of definite
MSA is not possible until the postmortem examination is performed.

Disease Progression and Prognosis Di agnosis

Multiple-system atrophy is characterized by a
relentless worsening of motor and nonmotor Because of its protean manifestations, multiple-
symptoms during an average time frame of 10 system atrophy may be misdiagnosed, especially
years, with more rapid progression at the onset at disease onset. An autonomic presentation can
(Fig. 3).9 Approximately 50% of patients require be indistinguishable from pure autonomic fail-
walking aids within 3 years after the onset of ure, and parkinsonism with autonomic involve-
motor symptoms,7 60% require a wheelchair after ment may be misdiagnosed as Parkinson’s disease.
5 years,10 and the median time before the patient Patients with the cerebellar subtype generally pres-
is bedridden is 6 to 8 years.7 The causes of death ent with late-onset cerebellar ataxia and addi-
in multiple-system atrophy commonly include tional features of autonomic failure. The cerebel-
bronchopneumonia, urosepsis, or sudden death. lar phenotype can mimic ataxias induced by toxins
Sudden death often occurs at night as a result of (e.g., alcohol, chemotherapeutic agents, lead,
either acute bilateral vocal-cord paralysis or acute lithium, and toluene) or by vitamin B1 deficiency,
disruption of the brain-stem cardiorespiratory as well as immune-mediated or genetic ataxias,
drive. Counseling patients on life expectancy can such as fragile X–associated tremor ataxia syn-
be challenging for the treating physician, because drome, spinocerebellar ataxia (especially type 6),
there are reports of both aggressive variants with or late-onset Friedreich’s ataxia.54
a disease duration of less than 3 years48 and more
benign cases with prolonged survival.11 Older age Clinical Diagnostic Criteria
at onset,7,8,10,49-51 a parkinsonian phenotype,9,50 and The consensus guidelines define three degrees of
early development of severe autonomic failure7,50,52,53 certainty for the diagnosis of multiple-system at-
are negative prognostic factors, whereas a cere- rophy: definite, probable, and possible.39 A diag-
bellar phenotype8 and later onset of autonomic nosis of definite multiple-system atrophy requires
failure11 predict slower disease progression. postmortem evidence of widespread α-synuclein–

254 n engl j med 372;3 nejm.org January 15, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Multiple-System Atrophy

Table 1. Diagnostic Criteria for Multiple-System Atrophy (MSA).*

Definite MSA
Neuropathological findings during postmortem examination must include the following:
Widespread and abundant cerebral α-synuclein–positive GCIs
Neurodegenerative changes in striatonigral or olivopontocerebellar region
Probable MSA
Sporadic, progressive disease in adults (onset after 30 yr of age) characterized by autonomic failure, including urinary incontinence
(with erectile dysfunction in men), or an orthostatic decrease in blood pressure by at least 30 mm Hg systolic or 15 mm Hg
diastolic within 3 min of standing, plus one of the following:
Parkinsonism (slowness of movements, rigidity, and tendency to fall) with poor response to levodopa (parkinsonian subtype
[MSA-P])
A cerebellar syndrome (wide-based gait, uncoordinated limb movements, action tremor, and nystagmus) (cerebellar subtype
[MSA-C])
Possible MSA
A sporadic, progressive, adult-onset disease characterized by the following:
Parkinsonism (slowness of movements, rigidity, and tendency to fall) or a cerebellar syndrome (wide-based gait, uncoordinated limb
movements, action tremor, and nystagmus)
At least one feature suggesting autonomic dysfunction (otherwise unexplained urinary urgency or frequency, incomplete bladder
emptying, erectile dysfunction in men, or a substantial orthostatic blood-pressure decline that does not meet the level re-
quired for probable MSA)
At least one of the following additional features:
Possible MSA-P or MSA-C: Babinski sign with hyperreflexia, stridor
Possible MSA-P: rapidly progressive parkinsonism; poor response to levodopa; recurrent falls within 3 yr after the onset of motor
symptoms; cerebellar features (wide-based gait; cerebellar dysarthria; uncoordinated limb movements; or spontaneous,
gaze-evoked, or positional downbeat nystagmus); recurrent choking within 5 yr after the onset of motor symptoms; atrophy
on MRI of the putamen, middle cerebellar peduncle, pons, or cerebellum; hypometabolism on FDG-PET in the putamen,
brain stem, or cerebellum
Possible MSA-C: parkinsonism; atrophy on MRI of the putamen, middle cerebellar peduncle, or pons; hypometabolism on FDG-
PET in the putamen; presynaptic nigrostriatal dopaminergic denervation on SPECT or PET
Features supporting the diagnosis of MSA (red flags) and features not supporting the diagnosis
Supporting: head–neck dystonia; disproportionate antecollis; bent spine (forward, lateral, or both); contractures of the hands or feet; in-
spiratory sighs; severe dysphonia; severe dysarthria; new or increased snoring; cold hands and feet; emotional incontinence
(pathologic laughter or crying); jerky, irregular, or postural or action tremor
Not supporting: classic “pill-rolling” rest tremor, clinically significant neuropathy, hallucinations not induced by drugs, onset after 75 yr
of age, family history of ataxia or parkinsonism, dementia (in accordance with DSM-IV criteria), white-matter lesions suggest-
ing multiple sclerosis

* The information is adapted from Gilman et al.,39 with permission. DSM-IV denotes Diagnostic and Statistical Manual of Mental Disorders, 4th
edition; FDG-PET [18F]-fluorodeoxyglucose–positron-emission tomography; GCI glial cytoplasmic inclusion; and SPECT single-photon-emis-
sion computed tomography.

positive glial cytoplasmic inclusions with con- sonism or cerebellar ataxia is accompanied by at
comitant olivopontocerebellar atrophy or striato- least one feature suggestive of autonomic failure
nigral degeneration. Probable multiple-system and at least one of the additional features listed
atrophy is defined as a sporadic, progressive in Table 1. The additional presence of one or more
disorder in adults (onset after the age of 30 “red flags” further supports a diagnosis of multi-
years), characterized by severe autonomic failure ple-system atrophy. Exclusion criteria need to be
plus predominantly levodopa-refractory parkin- considered as well (Table 1).
sonism (in the parkinsonian subtype) or cerebellar
ataxia (in the cerebellar subtype). Ancillary Investigations
Finally, a diagnosis of possible multiple-system The diagnosis of multiple-system atrophy is based
atrophy can be made if a sporadic, progressive, on the medical history and neurologic findings.
adult-onset disorder with predominant parkin- Nevertheless, ancillary investigations may help

n engl j med 372;3 nejm.org  January 15, 2015 255

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 256
Table 2. Ancillary Investigations in MSA.*

Investigation Characteristic Findings Notes

Neuroimaging
CT Ruling out of MSA-mimicking brain lesions CT is recommended for patients with contraindications to MRI.
1.5-Tesla MRI In MSA-P: atrophy of putamen, middle cerebellar peduncle, pons, or Putaminal rim sign and hot-cross-bun sign have high specificity but
cerebellum; “putaminal rim sign” (hyperintense signal of the dor- low sensitivity for the diagnosis of MSA.55
solateral border of putamen) plus putaminal hypointensity in T2-
weighted sequences; in MSA-C: atrophy of putamen, middle cere-
bellar peduncle, or pons; “hot-cross-bun sign” (cruciform hyperin-
tensity in the pons) in T2-weighted sequences
Diffusion-weighted MRI Increased putaminal diffusivity, differentiating MSA-P from PD; in- Diffusion-weighted MRI has high sensitivity and specificity.55
creased diffusivity in the middle cerebellar peduncle, differentiating Increased diffusivity in the middle cerebellar peduncle has been
MSA-P from PD and PSP also observed in fragile X–associated tremor ataxia syndrome.56
FDG-PET Hypometabolism in putamen, brain stem, or cerebellum, which indi- FDG-PET is recommended for patients with contraindications for
cates possible MSA-P in patients with atypical parkinsonism39; hy- MRI.
pometabolism in putamen, which indicates possible MSA-C in pa-
tients with idiopathic late-onset cerebellar ataxia39
The

18F-dopa PET, 123Iβ-CIT SPECT, and Presynaptic nigrostriatal dopaminergic denervation This finding is suggestive of MSA-C in patients with idiopathic, late-
123I-FP SPECT onset cerebellar ataxia.39
11C-raclopride PET and 123I-IBZM Postsynaptic dopamine-receptor loss in the striatum, differentiating These tests have suboptimal diagnostic accuracy.
SPECT MSA-P from PD
18F-fluorodopamine PET and Normal cardiac radiotracer uptake, differentiating MSA-P from PD57 These tests have suboptimal diagnostic accuracy. Normal uptake in-
123I-MIBG scintigraphy dicates the presence of spared cardiac postganglionic sympathet-
ic fibers in MSA.
Transcranial ultrasonography† Hyperechogenicity of the lentiform nucleus (putamen plus globus pall- The test cannot be performed in about 10% of patients because of
idus) combined with normal echogenicity of the substantia nigra, the absence of an acoustic temporal bone window.
differentiating MSA-P from PD58
Autonomic testing
n e w e ng l a n d j o u r na l

Urologic domain

The New England Journal of Medicine

of

Urodynamic examination Vesical detrusor overactivity, absence of detrusor–sphincter coordina- Masses, relevant surgical scars, prolapses, and urinary tract infec-
tion, bladder atony tions need to be ruled out in advance; videourodynamics can pro-
vide additional anatomical information in particular cases.
Ultrasonography Large postvoiding residual urine volumes (>100 ml) The test is recommended for patients reporting voiding difficulties.

n engl j med 372;3 nejm.org  January 15, 2015

Cardiovascular domain

Copyright © 2015 Massachusetts Medical Society. All rights reserved.

m e dic i n e

Standing test OH with blunted heart-rate increase on standing39 Perform the test after the patient has been in the supine position for
10 min. OH can be symptomatic or asymptomatic.
Passive head-up tilt test OH with blunted heart-rate increase on tilting39; decreased heart-rate The test requires devices for continuous heart rate and blood-pressure
variability at rest and during orthostatic challenge59; normal supine monitoring. OH can be symptomatic or asymptomatic.
plasma noradrenaline levels, differentiating MSA from PAF60†; lack Assessment of noradrenaline levels provides suboptimal diagnos-
of plasma vasopressin rise during head-up tilt, differentiating MSA tic accuracy. Consider dietary and environmental factors as well as
from PAF61† drugs modifying noradrenaline levels. A lack of plasma vasopressin
rise during head-up tilt indicates disruption of baroreceptive path-
ways in MSA.

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
 Investigation Characteristic Findings Notes
Valsalva maneuver Impaired cardiovascular autonomic reflexes with an absence of blood- The test requires devices for continuous heart rate and blood-pres-
pressure overshoot during late phase II and phase IV of the maneu- sure monitoring.
ver62,63
24-hr ambulatory blood-pressure Nocturnal hypertension
monitoring
Respiratory domain
Sleep laryngoscopy† Nocturnal stridor, subclinical paralysis of vocal-cord abductors
Polysomnography Sleep apnea The test is recommended for patients with snoring or excessive day-
time sleepiness; it is also indicated for the diagnosis of RBD.
Gastrointestinal domain
Videofluoroscopic swallow test Silent aspiration The test is recommended for patients with swallowing difficulties, ex-
cessive cough, or drooling (high suspicion of incipient dysphagia).
Measurement of colon-transit Loss of anal-sphincter relaxation, decreased anal tone
time43 and rectal manometry†
Anal-sphincter electromyography† Anal-sphincter abnormalities, differentiating MSA-P from PD and The test has suboptimal diagnostic accuracy; it does not distinguish
MSA-C from idiopathic late-onset cerebellar ataxia64 MSA-P from PSP. A history of multiple or traumatic deliveries,
lower abdominal surgery, prostatectomy, hemorrhoidectomy,
and chronic constipation may produce abnormal results.
Thermoregulatory domain
Thermoregulatory sweat test† Rapidly progressive failure of whole-body sweating, differentiating
MSA-P from PD44
Quantitative sudomotor axon re- Normal findings, differentiating MSA-P from PD65 Normal findings indicate sparing of skin postganglionic sympathetic
flex test† innervation in MSA.
Laboratory investigations
Multiple-System Atrophy

Clonidine growth hormone stimula- Absence of rise in plasma growth hormone level, differentiating The test has suboptimal diagnostic accuracy. The absence of an in-
tion test† MSA-P from PD and PSP65,66 crease in plasma growth hormone indicates α2 adrenoreceptor
loss in the hypothalamus in MSA.

n engl j med 372;3 nejm.org  January 15, 2015

The New England Journal of Medicine

Arginine growth hormone stimulation Absence of rise in plasma growth hormone level, differentiating The absence of an increase in plasma growth hormone indicates dis-
test† MSA-P from PD67 ruption of the hypothalamopituitary axis in MSA. The test does not
distinguish between MSA-C and idiopathic, late-onset cerebellar
ataxia.
Genetic testing COQ2 mutations (in Japanese patients)14; absence of mutations asso-

Copyright © 2015 Massachusetts Medical Society. All rights reserved.

ciated with fragile X–associated tremor ataxia syndrome,
Friedreich’s ataxia, and SCA types 1, 2, 3, 6, and 1754 in cases of a
cerebellar presentation
Other testing Normal results of tests for vitamin B1 deficiency, paraneoplastic ataxia, an-
ti-GAD ataxia, gluten ataxia, post–Epstein–Barr virus cerebellitis, and
Hashimoto’s thyroiditis54 for patients with a cerebellar presentation
Neuropsychological testing Frontal executive dysfunction,46 ruling out overt dementia39

* There are no established guidelines for ancillary testing in cases of suspected MSA. Test results that may help rule out conditions that mimic MSA or that support the clinical diagnosis
are listed. CT denotes computed tomography, 123Iβ-CIT 2β-carbomethoxy-3β-(4-iodophenyl)tropane, 123I-FP ioflupane, 123I-IBZM iodobenzamide, 123I-MIBG metaiodobenzylguanidine,

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
MRI magnetic resonance imaging, OH orthostatic hypotension, PAF pure autonomic failure, PD Parkinson’s disease, PSP progressive supranuclear palsy, RBD rapid-eye-movement
sleep behavior disorder, and SCA spinocerebellar ataxia.
† This test is not routinely used and therefore should be considered investigational.

257
 The n e w e ng l a n d j o u r na l of m e dic i n e

confirm the diagnosis, rule out alternative causes, Nonmotor Features

and tailor management strategies (Table 2). Patients with multiple-system atrophy who have
neurogenic bladder symptoms should be screened
regularly for urinary tract infections. Urinary urge
Pr incipl e s of Ther a py
incontinence due to detrusor overactivity can be
Only symptomatic therapy is available at present, treated with antimuscarinic agents. However, anti-
including pharmacologic and nonpharmacologic cholinergic side effects, such as confusion and
approaches (see Table S1 in the Supplementary cognitive worsening, must be monitored. Botuli-
Appendix, available with the full text of this ar- num toxin injections in the detrusor muscle may
ticle at NEJM.org). Unfortunately, the evidence be tried for patients whose condition does not
level is low for these approaches, with few ex- respond to treatment with antimuscarinic agents,
ceptions. Most of the commonly used drugs are and a 10° to 20° head-up tilt during sleep and bed-
prescribed off-label.68 We recommend a multi- time administration of desmopressin can amelio-
disciplinary management approach to optimally rate nocturia. Clean intermittent self-catheteriza-
address the patients’ needs.69 tion is the first-line therapy for urinary retention
with postvoid residual volumes above 100 ml.
Symptom Management Unfortunately, this approach can cause urethral
Motor Features ulceration in the long term, and suprapubic in-
In the parkinsonian subtype of multiple-system dwelling catheterization may become necessary.
atrophy, a slow increase in the levodopa dose is Preliminary evidence suggests the use of a blad-
advisable to minimize exacerbation of orthostatic der stimulator as a possible alternative to self-
hypotension, edema, and nausea. A transient ben- catheterization for patients with multiple-system
eficial response to levodopa is seen in as many atrophy. Add-on pharmacologic therapies for uri-
as 40% of patients.5 Levodopa withdrawal in pa- nary retention are designed either to enhance
tients with no apparent response occasionally vesical detrusor contractility (cholinergic agents)
causes abrupt and sometimes irreversible wors- or to promote urethral smooth sphincter relax-
ening of motor abnormalities. Complete discon- ation (α1-adrenoreceptor antagonists).74 Sildenafil,
tinuation of levodopa treatment is therefore not a phosphodiesterase type 5 inhibitor, can reverse
recommended for patients who do not have major erectile dysfunction in men with multiple-system
side effects. Dopamine agonists are less likely to atrophy, but worsening of orthostatic hypotension
provide a motor benefit in multiple-system atro- is a recognized side effect.75 Intracavernous injec-
phy but may be tried in cases of levodopa-induced tion of vasodilatory prostaglandins (e.g., alprosta-
dystonia; evidence for the effectiveness of aman- dil) can be used as an alternative approach. No
tadine, a selective antagonist of NMDA receptors, data are available on the management of sexual
is controversial.70 Treatment with amantadine can dysfunction in women with multiple-system at-
be attempted, but if no amelioration is achieved, rophy.
it should be discontinued. Local botulinum toxin Patients with orthostatic hypotension should
injections can be helpful for disabling hand, be trained to recognize and avoid triggers such
foot, or axial dystonia. as rapid postural change, heavy meals, straining
There is no specific therapy available for cer- while coughing or passing stool, and exposure
ebellar symptoms. Clonazepam may ameliorate to hot temperatures. If patients feel light-headed
myoclonus or action tremor in multiple-system (dizzy), physical countermaneuvers such as cross-
atrophy. Open-label gabapentin71 and buspirone72 ing the legs, squatting, or tensing muscles may
have also been effective in single cases. prevent syncope. Other nonpharmacologic mea-
Complementary neurorehabilitation programs, sures to ameliorate orthostatic hypotension include
including occupational, physical, and speech ther- increasing water and salt intake, keeping the head
apy, can be helpful for patients with the parkinso- of the bed raised during sleep, and using com-
nian or cerebellar subtype in preventing choking pression stockings or abdominal binders.40
and falls and in augmenting general coping and In patients with severe orthostatic hypoten-
communication abilities.73 sion, pharmacologic measures are advisable to

258 n engl j med 372;3 nejm.org  January 15, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Multiple-System Atrophy

minimize the risk of injurious falls. Drugs with ing that results from impaired swallowing in
hypotensive side effects (e.g., long-acting calcium advanced multiple-system atrophy. Liquid thick-
antagonists and antianginal agents) should be eners (e.g., honey) and a chin-down posture while
avoided or at least administered in the evening. swallowing can prevent choking in dysphagic
Midodrine and droxidopa, sympathomimetic patients. In advanced stages, a percutaneous en-
agents that increase arteriolar tone, are specifi- doscopic gastrostomy allows for enteral feeding
cally licensed by the Food and Drug Administra- and lowers the risk of aspiration pneumonia, but
tion for the symptomatic treatment of neurogenic it needs to be discussed with the patient well in
orthostatic hypotension.76-79 Off-label administra- advance in order to obtain informed consent.
tion of fludrocortisone can be helpful, in addition Constipation can be very difficult to treat in
to increasing the intravascular volume.80 Beneficial patients with multiple-system atrophy. Satisfying
effects have also been reported for a range of other results are likely to be achieved if alimentary
drugs used off-label to treat orthostatic hypoten- measures are combined with regular adminis-
sion associated with autonomic failure.81-85 Exac- tration of osmotic bulking agents.93
erbation of supine hypertension is a frequent side Low-dose clonazepam at bedtime can be con-
effect of therapies for orthostatic hypotension, and sidered as treatment for severe rapid-eye-move-
patients receiving such therapies should therefore ment sleep behavior disorder, but it may aggra-
be monitored. In the case of supine hypertension, vate nocturnal stridor or sleep apnea. Preliminary
recumbence should be avoided in the daytime, evidence suggests that melatonin may be an al-
and a snack before bedtime as well as a head-up ternative treatment in this situation.94
tilt position during sleep may control milder forms Cognitive impairment in patients with multi-
of nocturnal hypertension. If a reverse-dipping ple-system atrophy does not usually require treat-
profile (i.e., a nocturnal increase in blood pres- ment, but pharmacologic intervention may be
sure with respect to daytime) is documented dur- needed for those with severe depression, anxiety,
ing 24-hour ambulatory blood-pressure monitor- or emotional incontinence. Selective serotonin-
ing and does not respond to nonpharmacologic reuptake inhibitors are preferable to tricyclic anti-
measures, bedtime administration of short-act- depressants because they are more effective and
ing antihypertensive agents may be considered. less likely to worsen orthostatic hypotension and
Postprandial hypotension can be minimized by urinary retention.
avoiding both alcohol and excessive single-meal
caloric intake or by increasing water and coffee
T owa r d Dise a se Modific at ion
consumption. In severe cases, caffeine,86 octreo-
tide,87 or acarbose88 administration before eating Multiple-system atrophy–specific biomarkers are
may be helpful. needed to facilitate the early identification and
Continuous positive airway pressure — or, in enrollment of patients in disease-modifying clini-
resistant cases, biphasic positive airway pressure cal trials, at early or even presymptomatic disease
— is the therapy of choice for patients with stages. Perhaps surprisingly, blood and cerebro-
multiple-system atrophy who have isolated noc- spinal fluid α-synuclein levels appear to be neither
turnal inspiratory stridor or sleep apnea.89 Uni- sensitive nor specific for multiple-system atrophy.
lateral botulinum toxin injection in the vocal-cord Phosphorylated oligomeric α-synuclein or other
adductors90 may be considered for more severely markers, such as neurofilament light chain, FLT3
affected patients. Tracheostomy89 may effectively ligand, total tau, or amyloid β42, which may also
relieve airway obstruction at the laryngeal level be combined with α-synuclein, need to be explored
and prevent respiratory crisis due to acute bilateral further as candidate biomarkers in body fluids.95
paralysis of the vocal-cord abductors, but it can- Preliminary evidence suggests that positron-emis-
not eliminate the risk of sudden death, because sion tomography (PET) with 11C-2-(2-[2-dimethyl-
fatal sleep apnea can still occur. aminothiazol-5-yl]ethenyl)-6-(2-[fluoro]ethoxy)
Oral glycopyrrolate91 or botulinum toxin in- benzoxazole, an α-synuclein ligand, may allow
jections in the salivary glands92 (both of which visualization of the density of glial cytoplasmic
decrease saliva production) can relieve the drool- inclusions in regions of the brain in vivo.96 If

n engl j med 372;3 nejm.org  January 15, 2015 259

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

replicated and standardized, this approach may with the cerebellar subtype attenuated clinical
become an effective diagnostic and prognostic progression as well as the loss of cerebral glucose
molecular imaging marker of multiple-system metabolism and gray-matter density at 1 year of
atrophy. follow-up. However, cerebral ischemic lesions
Thanks to research consortia such as the (asymptomatic in all but one patient) were seen
European Multiple System Atrophy Study Group, in both the placebo group and the treatment
the French Multiple System Atrophy Reference group.100 These may have been related to the
Center, and the U.S. Autonomic Disorders Consor- administration procedure itself, but stem-cell–
tium, several intervention trials of orally active induced asymptomatic strokes in some patients
candidate neuroprotective agents have been com- cannot be ruled out. The exact mechanism of ac-
pleted in the past decade, using rates of motor tion of stem-cell therapy in multiple-system atrophy
decline as primary outcome measures and find- is also unclear at present. If they do not replace
ings from magnetic resonance imaging or PET lost neurons, stem cells might prevent neurode-
imaging as secondary outcome measures (see generation by releasing neurotrophic factors or
Table S2 in the Supplementary Appendix). Despite attenuating neuroinflammation.100,101 In our opin-
preclinical evidence in favor of neuroprotection, ion, the outcome of mesenchymal stem-cell thera-
thus far, riluzole,35 minocycline,97 rifampin (also py for multiple-system atrophy will depend on
called rifampicin),98 and rasagiline99 have shown carefully designed trial protocols that are informed
no benefit. Several factors may account for the by preclinical studies.
therapeutic failure seen in these trials. Currently Future trials will benefit from accelerated
available multiple-system atrophy preclinical mod- interventional target discovery, biomarker devel-
els may not reflect the pathologic complexity of opment, and early patient recruitment. To this
multiple-system atrophy, preclinical interventional purpose, a global multiple-system atrophy regis-
protocols may be substantially different from hu- try (GLOMSAR) has been established by the In-
man ones, and clinical end points may be insuffi- ternational Parkinson and Movement Disorder
cient to detect disease-modifying effects in the Society Study Group on Multiple System Atrophy
short term. Moreover, a putative neuroprotective with the support of the U.S. Multiple System Atro-
compound might be effective in early or even phy Coalition (www​.­multiple-system-atrophy​.­org).
preclinical stages of the disease but be futile in Dr. Wenning reports receiving consulting fees from Astra-
later stages, when much of the neuronal reserve Zeneca and lecture fees from Lundbeck. No other potential
conflict of interest relevant to this article was reported.
has already been lost. Disclosure forms provided by the authors are available with
A randomized, placebo-controlled trial showed the full text of this article at NEJM.org.
that intraarterial infusion of autologous mesen- We dedicate this review to Kerry Simon and all patients with
multiple-system atrophy under our care, whose fight against
chymal stem cells, followed by three intravenous this challenging disease we admire. We thank Sid Gilman and
stem-cell infusions, 1 month apart, in patients Ryan Walsh for providing their expert opinion on the review.

References
1. Quinn N. Multiple system atrophy — JP, et al. Presentation, diagnosis, and 9. Wenning GK, Geser F, Krismer F, et
the nature of the beast. J Neurol Neuro- management of multiple system atrophy al. The natural history of multiple system
surg Psychiatry 1989;​Suppl:​78-89. in Europe: final analysis of the European atrophy: a prospective European cohort
2. Bower JH, Maraganore DM, McDonnell multiple system atrophy registry. Mov study. Lancet Neurol 2013;​12:​264-74.
SK, Rocca WA. Incidence of progressive su- Disord 2010;​25:​2604-12. 10. Klockgether T, Lüdtke R, Kramer B, et
pranuclear palsy and multiple system atro- 6. Kim HJ, Jeon BS, Lee JY, Yun JY. Sur- al. The natural history of degenerative
phy in Olmsted County, Minnesota, 1976 to vival of Korean patients with multiple sys- ataxia: a retrospective study in 466 pa-
1990. Neurology 1997;​49:​1284-8. tem atrophy. Mov Disord 2011;​26:​909-12. tients. Brain 1998;​121:​589-600.
3. Schrag A, Ben-Shlomo Y, Quinn NP. 7. Watanabe H, Saito Y, Terao S, et al. 11. Petrovic IN, Ling H, Asi Y, et al. Mul-
Prevalence of progressive supranuclear Progression and prognosis in multiple tiple system atrophy-parkinsonism with
palsy and multiple system atrophy: a cross- system atrophy: an analysis of 230 Japa- slow progression and prolonged survival:
sectional study. Lancet 1999;​354:​1771-5. nese patients. Brain 2002;​125:​1070-83. a diagnostic catch. Mov Disord 2012;​27:​
4. Gilman S, May SJ, Shults CW, et al. 8. Ben-Shlomo Y, Wenning GK, Tison F, 1186-90.
The North American Multiple System At- Quinn NP. Survival of patients with 12. Itoh K, Kasai T, Tsuji Y, et al. Definite
rophy Study Group. J Neural Transm pathologically proven multiple system at- familial multiple system atrophy with un-
2005;​112:​1687-94. rophy: a meta-analysis. Neurology 1997;​ known genetics. Neuropathology 2014;​
5. Köllensperger M, Geser F, Ndayisaba 48:​384-93. 34:​309-13.

260 n engl j med 372;3 nejm.org  January 15, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Multiple-System Atrophy

13. Stemberger S, Scholz SW, Singleton multiple system atrophy with Parkinson’s ic hypotension in autonomic failure. Hy-
AB, Wenning GK. Genetic players in mul- disease and dementia with Lewy bodies. pertension 2003;​42:​136-42.
tiple system atrophy: unfolding the nature Neurosci Lett 1998;​251:​205-8. 42. Ghorayeb I, Bioulac B, Tison F. Rela-
of the beast. Neurobiol Aging 2011;​ 27. Wenning GK, Stefanova N, Jellinger tionship between stridor and sleep ap-
32(10):​1924.e5-1924.e14. KA, Poewe W, Schlossmacher MG. Multi- noea syndrome: is it as simple as that? J
14. Multiple-System Atrophy Research ple system atrophy: a primary oligoden- Neurol Neurosurg Psychiatry 2004;​ 75:​
Collaboration. Mutations in COQ2 in fa- drogliopathy. Ann Neurol 2008;​ 64:​239- 512-3.
milial and sporadic multiple-system atro- 46. 43. Sakakibara R, Odaka T, Uchiyama T,
phy. N Engl J Med 2014;​369:​233-44. 28. Song YJ, Lundvig DM, Huang Y, et al. et al. Colonic transit time, sphincter
15. Ferguson MC, Garland EM, Hedges L, p25alpha Relocalizes in oligodendroglia EMG, and rectoanal videomanometry in
et al. SHC2 gene copy number in multiple from myelin to cytoplasmic inclusions in multiple system atrophy. Mov Disord
system atrophy (MSA). Clin Auton Res multiple system atrophy. Am J Pathol 2004;​19:​924-9.
2014;​24:​25-30. 2007;​171:​1291-303. 44. Lipp A, Sandroni P, Ahlskog JE, et al.
16. Sasaki H, Emi M, Iijima H, et al. Copy 29. Asi YT, Simpson JE, Heath PR, et al. Prospective differentiation of multiple
number loss of (src homology 2 domain Alpha-synuclein mRNA expression in oli- system atrophy from Parkinson disease,
containing)-transforming protein 2 (SHC2) godendrocytes in MSA. Glia 2014;​62:​964- with and without autonomic failure. Arch
gene: discordant loss in monozygotic 70. Neurol 2009;​66:​742-50.
twins and frequent loss in patients with 30. Reyes JF, Rey NL, Bousset L, Melki R, 45. Klein C, Brown R, Wenning G, Quinn
multiple system atrophy. Mol Brain 2011;​ Brundin P, Angot E. Alpha-synuclein N. The “cold hands sign” in multiple sys-
4:​24. transfers from neurons to oligodendro- tem atrophy. Mov Disord 1997;​12:​514-8.
17. Gwinn K, Devine MJ, Jin LW, et al. cytes. Glia 2014;​62:​387-98. 46. Stankovic I, Krismer F, Jesic A, et al.
Clinical features, with video documenta- 31. Watts JC, Giles K, Oehler A, et al. Cognitive impairment in multiple system
tion, of the original familial Lewy body Transmission of multiple system atrophy atrophy: a position statement by the Neu-
parkinsonism caused by α-synuclein trip- prions to transgenic mice. Proc Natl Acad ropsychology Task Force of the MDS Mul-
lication (Iowa kindred). Mov Disord 2011;​ Sci U S A 2013;​110:​19555-60. tiple System Atrophy (MODIMSA) study
26:​2134-6. 32. Jecmenica-Lukic M, Poewe W, Tolosa group. Mov Disord 2014;​29:​857-67.
18. Scholz SW, Houlden H, Schulte C, et E, Wenning GK. Premotor signs and 47. Tison F, Wenning GK, Volonte MA,
al. SNCA variants are associated with in- symptoms of multiple system atrophy. Poewe WR, Henry P, Quinn NP. Pain in
creased risk for multiple system atrophy. Lancet Neurol 2012;​11:​361-8. multiple system atrophy. J Neurol 1996;​
Ann Neurol 2009;​65:​610-4. 33. Wenning GK, Tison F, Ben Shlomo Y, 243:​153-6.
19. Al-Chalabi A, Dürr A, Wood NW, et Daniel SE, Quinn NP. Multiple system 48. Wakabayashi K, Mori F, Nishie M, et
al. Genetic variants of the alpha-synuclein atrophy: a review of 203 pathologically al. An autopsy case of early (“minimal
gene SNCA are associated with multiple proven cases. Mov Disord 1997;​12:​133- change”) olivopontocerebellar atrophy
system atrophy. PLoS One 2009;​ 4(9):​ 47. (multiple system atrophy-cerebellar). Acta
e7114. 34. Boesch SM, Wenning GK, Ransmayr Neuropathol 2005;​110:​185-90.
20. Sailer A. First genome-wide associa- G, Poewe W. Dystonia in multiple system 49. Wenning GK, Ben Shlomo Y, Magal-
tion study in multiple system atrophy. atrophy. J Neurol Neurosurg Psychiatry hães M, Daniel SE, Quinn NP. Clinical
Mov Disord 2012;​27:​Suppl 1:​S467. 2002;​72:​300-3. features and natural history of multiple
21. Kiely AP, Asi YT, Kara E, et al. 35. Bensimon G, Ludolph A, Agid Y, Vid- system atrophy: an analysis of 100 cases.
α-Synucleinopathy associated with G51D ailhet M, Payan C, Leigh PN. Riluzole Brain 1994;​117:​835-45.
SNCA mutation: a link between Parkin- treatment, survival and diagnostic criteria 50. O’Sullivan SS, Massey LA, Williams
son’s disease and multiple system atro- in Parkinson plus disorders: the NNIPPS DR, et al. Clinical outcomes of progres-
phy? Acta Neuropathol 2013;​125:​753-69. study. Brain 2009;​132:​156-71. sive supranuclear palsy and multiple sys-
22. Ahmed Z, Asi YT, Sailer A, et al. The 36. Köllensperger M, Geser F, Seppi K, et tem atrophy. Brain 2008;​131:​1362-72.
neuropathology, pathophysiology and ge- al. Red flags for multiple system atrophy. 51. Roncevic D, Palma JA, Martinez J,
netics of multiple system atrophy. Neuro- Mov Disord 2008;​23:​1093-9. Goulding N, Norcliffe-Kaufmann L,
pathol Appl Neurobiol 2012;​38:​4-24. 37. Sakakibara R, Hattori T, Uchiyama T, Kaufmann H. Cerebellar and parkinso-
23. Benarroch EE. The central autonomic et al. Urinary dysfunction and orthostatic nian phenotypes in multiple system atro-
network: functional organization, dys- hypotension in multiple system atrophy: phy: similarities, differences and survival.
function, and perspective. Mayo Clin Proc which is the more common and earlier J Neural Transm 2014;​121:​507-12.
1993;​68:​988-1001. manifestation? J Neurol Neurosurg Psy- 52. Fanciulli A, Strano S, Colosimo C,
24. Papp MI, Kahn JE, Lantos PL. Glial chiatry 2000;​68:​65-9. Caltagirone C, Spalletta G, Pontieri FE.
cytoplasmic inclusions in the CNS of pa- 38. Beck RO, Betts CD, Fowler CJ. Genito- The potential prognostic role of cardio-
tients with multiple system atrophy (stria- urinary dysfunction in multiple system vascular autonomic failure in α-sy­ n­
u­
tonigral degeneration, olivopontocerebel- atrophy: clinical features and treatment clein­­opathies. Eur J Neurol 2013;​20:​231-
lar atrophy and Shy-Drager syndrome). J in 62 cases. J Urol 1994;​151:​1336-41. 5.
Neurol Sci 1989;​94:​79-100. 39. Gilman S, Wenning GK, Low PA, et al. 53. Figueroa JJ, Singer W, Parsaik A, et al.
25. Ozawa T, Paviour D, Quinn NP, et al. Second consensus statement on the diag- Multiple system atrophy: prognostic indi-
The spectrum of pathological involve- nosis of multiple system atrophy. Neurol- cators of survival. Mov Disord 2014;​29:​
ment of the striatonigral and olivoponto- ogy 2008;​71:​670-6. 1151-7.
cerebellar systems in multiple system at- 40. Freeman R. Clinical practice. Neuro- 54. Lin DJ, Hermann KL, Schmahmann
rophy: clinicopathological correlations. genic orthostatic hypotension. N Engl J JD. Multiple system atrophy of the cere-
Brain 2004;​127:​2657-71. Med 2008;​358:​615-24. bellar type: clinical state of the art. Mov
26. Spillantini MG, Crowther RA, Jakes 41. Goldstein DS, Pechnik S, Holmes C, Disord 2014;​29:​294-304.
R, Cairns NJ, Lantos PL, Goedert M. Fila- Eldadah B, Sharabi Y. Association be- 55. Brooks DJ, Seppi K. Proposed neuro-
mentous alpha-synuclein inclusions link tween supine hypertension and orthostat- imaging criteria for the diagnosis of mul-

n engl j med 372;3 nejm.org  January 15, 2015 261

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

tiple system atrophy. Mov Disord 2009;​24:​ 70. Wenning GK. Placebo-controlled tri- orthostatic hypotension. J Neurol Neuro-
949-64. al of amantadine in multiple-system at- surg Psychiatry 2003;​74:​1294-8.
56. Apartis E, Blancher A, Meissner WG, rophy. Clin Neuropharmacol 2005;​ 28:​ 84. Perera R, Isola L, Kaufmann H. Effect
et al. FXTAS: new insights and the need 225-7. of recombinant erythropoietin on anemia
for revised diagnostic criteria. Neurology 71. Gazulla J, Benavente MI. Improve- and orthostatic hypotension in primary
2012;​79:​1898-907. ments in the symptoms of olivopontocer- autonomic failure. Clin Auton Res 1995;​5:​
57. Goldstein DS. Sympathetic neuroim- ebellar atrophy with gabapentin. Rev 211-3.
aging. Handb Clin Neurol 2013;​117:​365- Neurol 2005;​40:​285-8. 85. Fouad-Tarazi FM, Okabe M, Goren H.
70. 72. Heo JH, Lee ST, Chu K, Kim M. The Alpha sympathomimetic treatment of au-
58. Behnke S, Berg D, Naumann M, Beck- efficacy of combined estrogen and buspi- tonomic insufficiency with orthostatic
er G. Differentiation of Parkinson’s dis- rone treatment in olivopontocerebellar hypotension. Am J Med 1995;​99:​604-10.
ease and atypical parkinsonian syndromes atrophy. J Neurol Sci 2008;​271:​87-90. 86. Onrot J, Goldberg MR, Biaggioni I,
by transcranial ultrasound. J Neurol Neu- 73. Jain S, Dawson J, Quinn NP, Playford Hollister AS, Kingaid D, Robertson D. He-
rosurg Psychiatry 2005;​76:​423-5. ED. Occupational therapy in multiple sys- modynamic and humoral effects of caf-
59. Bordet R, Benhadjali J, Destee A, tem atrophy: a pilot randomized con- feine in autonomic failure: therapeutic
Hurtevent JF, Bourriez JL, Guieu JD. Sym- trolled trial. Mov Disord 2004;​19:​1360-4. implications for postprandial hypoten-
pathetic skin response and R-R interval 74. Winge K, Fowler CJ. Bladder dysfunc- sion. N Engl J Med 1985;​313:​549-54.
variability in multiple system atrophy and tion in Parkinsonism: mechanisms, prev- 87. Alam M, Smith G, Bleasdale-Barr K,
idiopathic Parkinson’s disease. Mov Dis- alence, symptoms, and management. Pavitt DV, Mathias CJ. Effects of the pep-
ord 1996;​11:​268-72. Mov Disord 2006;​21:​737-45. tide release inhibitor, octreotide, on day-
60. Goldstein DS, Holmes C, Sharabi Y, 75. Hussain IF, Brady CM, Swinn MJ, time hypotension and on nocturnal hy-
Brentzel S, Eisenhofer G. Plasma levels of Mathias CJ, Fowler CJ. Treatment of erec- pertension in primary autonomic failure.
catechols and metanephrines in neuro- tile dysfunction with sildenafil citrate (Vi- J Hypertens 1995;​13:​1664-9.
genic orthostatic hypotension. Neurology agra) in parkinsonism due to Parkinson’s 88. Fukushima T, Asahina M, Fujinuma
2003;​60:​1327-32. disease or multiple system atrophy with Y, et al. Role of intestinal peptides and the
61. Kaufmann H, Oribe E, Miller M, observations on orthostatic hypotension. autonomic nervous system in postpran-
Knott P, Wiltshire-Clement M, Yahr MD. J Neurol Neurosurg Psychiatry 2001;​ 71:​ dial hypotension in patients with multiple
Hypotension-induced vasopressin release 371-4. system atrophy. J Neurol 2013;​260:​475-83.
distinguishes between pure autonomic 76. Jankovic J, Gilden JL, Hiner BC, et al. 89. Iranzo A. Management of sleep-disor-
failure and multiple system atrophy with Neurogenic orthostatic hypotension: a dered breathing in multiple system atro-
autonomic failure. Neurology 1992;​ 42:​ double-blind, placebo-controlled study phy. Sleep Med 2005;​6:​297-300.
590-3. with midodrine. Am J Med 1993;​95:​38-48. 90. Merlo IM, Occhini A, Pacchetti C, Al-
62. Fanciulli A, Strano S, Ndayisaba JP, et 77. Wright RA, Kaufmann HC, Perera R, fonsi E. Not paralysis, but dystonia causes
al. Detecting nocturnal hypertension in et al. A double-blind, dose-response study stridor in multiple system atrophy. Neu-
Parkinson’s disease and multiple system of midodrine in neurogenic orthostatic rology 2002;​58:​649-52.
atrophy: proposal of a decision-support hypotension. Neurology 1998;​51:​120-4. 91. Seppi K, Weintraub D, Coelho M, et
algorithm. J Neurol 2014;​261:​1291-9. 78. Low PA, Gilden JL, Freeman R, Sheng al. The Movement Disorder Society evi-
63. Low PA, Tomalia VA, Park KJ. Auto- KN, McElligott MA. Efficacy of midodrine dence-based medicine review update:
nomic function tests: some clinical appli- vs placebo in neurogenic orthostatic hy- treatments for the non-motor symptoms
cations. J Clin Neurol 2013;​9:​1-8. potension: a randomized, double-blind of Parkinson’s disease. Mov Disord 2011;​
64. Vodusek DB. How to diagnose MSA multicenter study. JAMA 1997;​ 277:​
1046- 26:​Suppl 3:​S42-S80.
early: the role of sphincter EMG. J Neural 51. 92. Mancini F, Zangaglia R, Cristina S, et
Transm 2005;​112:​1657-68. 79. Freeman R, Landsberg L, Young J. The al. Double-blind, placebo-controlled study
65. Kimber J, Mathias CJ, Lees AJ, et al. treatment of neurogenic orthostatic hy- to evaluate the efficacy and safety of bot-
Physiological, pharmacological and neu- potension with 3,4-DL-threo-dihydroxy- ulinum toxin type A in the treatment of
rohormonal assessment of autonomic phenylserine: a randomized, placebo-con- drooling in parkinsonism. Mov Disord
function in progressive supranuclear pal- trolled, crossover trial. Neurology 1999;​ 2003;​18:​685-8.
sy. Brain 2000;​123:​1422-30. 53:​2151-7. 93. Eichhorn TE, Oertel WH. Macrogol
66. Kimber JR, Watson L, Mathias CJ. 80. Chobanian AV, Volicer L, Tifft CP, 3350/electrolyte improves constipation in
Distinction of idiopathic Parkinson’s dis- Gavras H, Liang CS, Faxon D. Mineralo- Parkinson’s disease and multiple system
ease from multiple-system atrophy by corticoid-induced hypertension in pa- atrophy. Mov Disord 2001;​16:​1176-7.
stimulation of growth-hormone release tients with orthostatic hypotension. N 94. Boeve BF, Silber MH, Ferman TJ. Mel-
with clonidine. Lancet 1997;​349:​1877-81. Engl J Med 1979;​301:​68-73. atonin for treatment of REM sleep behav-
67. Pellecchia MT, Longo K, Pivonello R, 81. Jordan J, Shannon JR, Biaggioni I, ior disorder in neurologic disorders: results
et al. Multiple system atrophy is distin- Norman R, Black BK, Robertson D. Con- in 14 patients. Sleep Med 2003;​4:​281-4.
guished from idiopathic Parkinson’s dis- trasting actions of pressor agents in se- 95. Zhang F, Chen J, Zhao L, Dong C.
ease by the arginine growth hormone vere autonomic failure. Am J Med 1998;​ Candidate biomarkers of multiple system
stimulation test. Ann Neurol 2006;​ 60:​ 105:​116-24. atrophy in cerebrospinal fluid. Rev Neu-
611-5. 82. Sakakibara R, Matsuda S, Uchiyama rosci 2014;​25:​653-62.
68. Rey MV, Perez-Lloret S, Pavy-Le Traon T, Yoshiyama M, Yamanishi T, Hattori T. 96. Kikuchi A, Takeda A, Okamura N, et
A, Meissner WG, Tison F, Rascol O. A The effect of intranasal desmopressin on al. In vivo visualization of alpha-synuclein
cross-sectional study on drug use in mul- nocturnal waking in urination in multiple deposition by carbon-11-labelled 2-[2-(2-
tiple system atrophy. CNS Drugs 2014;​28:​ system atrophy patients with nocturnal dimethylaminothiazol-5-yl)ethenyl]-6-
483-90. polyuria. Clin Auton Res 2003;​13:​106-8. [2-(fluoro)ethoxy]benzoxazole positron
69. Fanciulli A, Wenning GK. Treatment. 83. Singer W, Opfer-Gehrking TL, McPhee emission tomography in multiple system
In:​Wenning GK, Fanciulli A, ed. Multiple BR, Hilz MJ, Bharucha AE, Low PA. Ace- atrophy. Brain 2010;​133:​1772-8.
system atrophy. Vienna:​Springer-Verlag, tylcholinesterase inhibition: a novel ap- 97. Dodel R, Spottke A, Gerhard A, et al.
2014:​169-94. proach in the treatment of neurogenic Minocycline 1-year therapy in multiple-

262 n engl j med 372;3 nejm.org  January 15, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Multiple-System Atrophy

system-atrophy: effect on clinical symp- A randomized, placebo-controlled clinical 101. Stemberger S, Jamnig A, Stefanova
toms and [(11)C] (R)-PK11195 PET trial to assess the effects of rasagiline in N, Lepperdinger G, Reindl M, Wenning
(MEMSA-trial). Mov Disord 2010;​25:​97-107. patients with multiple system atrophy of GK. Mesenchymal stem cells in a trans-
98. Low PA, Robertson D, Gilman S, et parkinsonian subtype. Mov Disord 2012;​ genic mouse model of multiple system
al. Efficacy and safety of rifampicin for 27:​Suppl 1:​S390. atrophy: immunomodulation and neuro-
multiple system atrophy: a randomised, 100. Lee PH, Lee JE, Kim HS, et al. A ran- protection. PLoS One 2011;​6(5):​e19808.
double-blind, placebo-controlled trial. domized trial of mesenchymal stem cells Copyright © 2015 Massachusetts Medical Society.
Lancet Neurol 2014;​13:​268-75. in multiple system atrophy. Ann Neurol
99. Poewe W, Barone P, Gliadi N, et al. 2012;​72:​32-40.

n engl j med 372;3 nejm.org  January 15, 2015 263

The New England Journal of Medicine
Downloaded from nejm.org at UNIV OF PENN LIBRARY on January 14, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.