Journal ofNeurology, Neurosurgery, and Psychiatry. Special Supplement 1989:78-89

Multiple system atrophy the nature of the beastNIALL QUINN

From the University Department ofClinical Neurology, Institute ofNeurology, The National HospitalforNervous Diseases, London, UK

SUMMARY Multiple system atrophy (MSA) is generally considered a rare disease, but may accountfor up to 10% of patients with Parkinsonism. The profusion of names for this disease, which maypresent to general physicians, psychiatrists, cardiologists, autonomic specialists, general neurologistsand those with a special interest in Parkinsonism (this author's own perspective) or cerebellardisorders, together with ignorance of its protean manifestations, may account for its under-recognition and misdiagnosis. In this article, the history and nosology of the condition areconsidered, and provisional diagnostic criteria are advanced. The usefulness (or otherwise) ofancillary investigations is addressed, and the shortcomings of current methods of treatment arestressed. As with idiopathic Parkinson's disease, the ultimate goal of eradicating the disease entailsbetter diagnosis in order to establish the cause, and thence to develop a radical treatment capable ofpreventing or arresting the disease process.

Multiple system atrophy (MSA)' is a degenerativedisease of the central nervous system. Clinically it maypresent with any combination of extrapyramidal,pyramidal, cerebellar and autonomic signs and symp-toms, and may change its clinical emphasis as itevolves. Pathologically, it is characterised by cell lossand gliosis occurring in a selection of the following "atrisk" structures: substantia nigra, caudate, putamen(especially posterior part), globus pallidus (especiallypars externa), inferior olives, pontine nuclei, cerebellarPurkinje cells, intermediolateral cell columns of thespinal cord and Onuf's nucleus. Locus coeruleus,dorsal motor nucleus of vagus, vestibular nuclei,pyramidal tracts and anterior horns may also beinvolved. This pattern of involvement may hold thekey to the mechanism of cell loss, explaining whycertain cells are susceptible to whatever causes thedisease whilst adjacent cells and structures are spared.Together with this distribution of cell loss, the absenceof Lewy bodies distinguishes it from idiopathic Park-inson's disease and the lack of neurofibrillary tanglesseparates it from Steele-Richardson-Olszewski diseaseand post-encephalitic Parkinsonism.MSA has always been considered a rare condition,

yet it has also attracted a number of different andcomplex names related to its varied clinical and

Address for reprint requests: Dr Niall Quinn, University Departmentof Neurology, Institute of Neurology, Queen Square, London WCI3BG, UK.

Accepted January 1989

pathological features (see fig). This has resulted inconsiderable confusion among neurologists. The mainaim of this review is hopefully to bring together thestrands to weave a coherent picture of the condition.

The history ofMSAOnce a disease has been delineated, it is alwaysinstructive to return to earlier descriptions which, withhindsight, probably dealt with the same condition.Although other French and German workers mayhave described cases of this condition in the latenineteenth century, Dejerine and Thomas,2 in 1900, in

Fig Simple is beautiful.

78

Jdwi/o4 JtoshA9<,

S1S

She-)t (4).

018flP6dA

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

Multiple system atrophy-the nature ofthe beast

describing, two sporadic cases, were the first tointroduce the term olivopontocerebellar atrophy(OPCA). Their first case was a woman of 52 years, thesecond a man of 42. Both developed ataxia, dysarth-ria, akinesia, rigidity and brisk leg reflexes. Thewoman developed incontinence of urine, and diedafter 3 years. The man developed tremor, incontinenceof urine, and probable symptomatic postural hypoten-sion ("etourdissements"). Pathological examinationof the first case revealed atrophy of the olives, ponsand cerebellum. The substantia nigra and striatumwere not examined, or at least not commented upon.The next piece of the jigsaw was a paper by

Bradbury and Eggleston in the American Heart Jour-nalin 1925 entitled "Postural hypotension. A report ofthree cases".3 Here they described three male subjectswhose symptoms began in middle life (aged 36, 47, 60years) and who were seen after from 3 to 7 years ofillness. Common to all three were symptomatic pos-tural hypotension and anhidrosis. Patient 1 hadadditional impotence and constipation, cases 2 and 3had unequal pupils and brisk leg reflexes, and case 3had bilateral extensor plantar responses. Two patientsdied suddenly, and two years later4 they reported post-mortem findings in one of these cases, but unfortun-ately the central nervous system was not examined.Over subsequent years, a number of authors began torecognise in some patients a clinical association be-tween orthostatic hypotension and disease of thecentral nervous system, especially Parkinsonism.56The case of Langston5 was a man of 56 with a 7 yearhistory of autonomic failure followed by Parkinson-ism with antecollis, husky and uncontrollable speech,"peculiar involuntary movements resembling laugh-ter" and emotionalism. Young's patient,6 a man of43,also had a 5 year history comprising autonomicfailure, Parkinsonism, an unsteady, staggering gaitand a husky voice. A left vocal cord paresis was noted.The next developments took place in 1960. It is

important to remember that this was before thelevodopa era, and to keep in mind that today thedegree of clinical response to levodopa is the singlemost useful diagnostic pointer in the differentialdiagnosis of Parkinsonism. Shy and Drager in 19607described "A neurological syndrome associated withorthostatic hypotension". They had seen four cases,and reported two in detail, with pathology in one.Both were male. Case 1 began at age 39, and was seenat age 46 after 7 years of disease. Case 2 began at age49, and died at age 551 after 6j years of disease. Bothhad marked autonomic failure, slurred speech,impaired co-ordination and distal muscle wasting.Both had tremor at rest affecting the hands, and case 1also had tremor on movement. Case 1 had pyramidalsigns, case 2 equivocal plantar responses and fascicula-tions. Both are said to have had external ocular muscle

79paresis, but in case 1 this was a slight exophoria and incase 2 "inability to converge and weakness of themedial rectus muscle bilaterally". Both had reducedfacial expression, case 2 noted slowness ofmovements,and rigidity was present in both.Post-mortem examination ofcase 2 showed cell loss

and/or gliosis in many areas, but most marked in thecaudate, substantia nigra, olives, locus coeruleus,cerebellum and intermediolateral cell columns of thespinal cord; the putamen, globus pallidus and ponswere also affected. In the introduction, the authorsdefined the full syndrome as comprising the followingfeatures: orthostatic hypotension, urinary and rectalincontinence, loss of sweating, iris atrophy, externalocular palsies, rigidity, tremor, loss of associatedmovements, impotence, the findings of an atonicbladder and loss of the rectal sphincter tone, fascicula-tions*, wasting of distal muscles, evidence of aneuropathic lesion in the electromyogram that sug-gests involvement of the anterior horn cells*, and thefinding of a neuropathic lesion in the muscle biopsy.Items with an asterisk* were found only in case 2. It isof interest to note that neither cerebellar norpyramidal signs appear in this definition of the fullsyndrome.

In the previous year Van der Eecken, Adams andvan Bogaert had made a brief presentation to theAmerican Association of Neuropathologists entitled"Striopallidal-nigral degeneration. An hitherto unde-scribed lesion in paralysis agitans" which was pub-lished in 196O.' They called attention to the finding ofpronounced shrinkage and brownish discoloration ofparts of the putamen and globus pallidus as well as theusual depigmentation of the substantia nigra. Micro-scopically, one of the most striking lesions was thevirtual disappearance ofthe small cells ofboth caudateand putamen.The same authors (Adams et a) in 1961 published a

longer paper entitled: "Degenerescences nigro-strieeset cerebello-nigro-striees".9 In it, they gave clinical andpathological details of 3 patients (2 M, 1 F) withsporadic disease. All had an akinetic-rigid syndromewith tremor (slight action, slight rest, and 4-5 Hzrespectively). Reflexes were brisk in all three, plantarsextensor in case 2, and there was ataxia and intentiontremor in case 3. Cases 1 and 3 had severe speechdisturbance, case 1 had blackouts and impotence, andcase 2 double incontinence. The pathology involvednot only striatum and nigra, but also olives andcerebellum in all three cases, with additional pontinelesions in cases 1 and 3. Case 2 also had Lewy bodies insubstantia nigra. Neither clinically nor pathologically,therefore, was striatonigral degeneration a pure andrestricted entity resulting from pathology of thesestructures alone.

It seems surprising that accounts of striatal patho-

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

80logy in some cases of Parkinsonism should only havebegun to appear in the 1960s. However, some suchcases were in fact described earlier, but taken to be partof the spectrum ofParkinson's disease. Thus Messing,in 1930, described the case of a woman aged 62 years'0who presented "the classical picture of Parkinson'sdisease". Post-mortem examination of the brainrevealed "total atrophy of the ponto-cerebellar sys-tem, partial atrophy of the inferior olives and theirconnections, marked atrophy of the supero-lateralportions of the cerebellum and of the vermis ... andfibrosis of the globus pallidus ... together withdysmyelination of the putamen and the anteriorportion of the caudate nucleus ... This was certainly acase of Parkinson's disease, with an unusualadditional lesion: atrophy ofthe cerebellar pathways".Malamud's 1957 atlas of neuropathology" devoted

one page of text to "Parkinson's disease (paralysisagitans) ... Besides the postencephalitic and othersymptomatic forms of Parkinsonism, there is aprimary degeneration form that generally begins inlate middle or advanced age ... The predominantpathology is either in the striopallidum or in thesubstantia nigra, although some cases show morediffuse cerebral involvement. Case 1: a 73 year old manwith a two year history ... gradually became bedfast... He exhibited a pill-rolling tremor. . ., rigidity andweakness ..., a slurred, unintelligible speech, yet arelatively normal mental status. His course wasrapidly downhill. The brain showed symmetrical atro-phy of the putamen and globus pallidus, with grayishdiscoloration of the former". Neither of theseaccounts document nigral pathology, nor is there anymention of the presence or absence of Lewy bodies.Thus, despite the major discoveries of both Lewy andTretiakoff in the early 1900s, there was still noneuropathological consensus on the diagnosis ofidiopathic Parkinson's disease until almost half acentury had elapsed.Thus far, we have encountered four "conditions"

which at first sight appeared separate, but may in factall include subjects with the same disease: OPCA,idiopathic orthostatic hypotension, the Shy-Dragersyndrome and striatonigral degeneration. Grahamand Oppenheimer' deserve our thanks for writing in1969: ". . . unnecessary confusion is caused by invent-ing new names, of the type "pallido-subthalamo-vestibular atrophy" for unusual syndromes. What wewish to avoid is the multiplication of names for"disease entities" which, in fact, are merely theexpression ofneuronal atrophy in a variety of overlap-ping combinations. We therefore propose to use theterm "multiple system atrophy" ".

During the early 1970s it became apparent thatpatients with neurogenic autonomic failure sometimeshad Lewy body pathology rather than the changes of

Quinn

MSA. A key clinicopathological review by Bannisterand Oppenheimer in 1972 entitled: "Degenerativediseases of the nervous system associated with auto-nomic failure"'2 considered this division. They des-cribed 16 pathologically studied cases of autonomicfailure: four personal cases (three MSA, one Lewybody) and 12 cases from the literature (eight MSA,four Lewy body). In the five Lewy body cases (4 M,1 F) mean age of disease onset was 65 years (range56-73), mean survival 5 years (range 1-5-13) and meanage at death 70 years. In the 11 MSA cases (8 M,. 3 F),mean age at onset was 49 years (range 36-58), meansurvival 6 years (range 2-11) and mean age at death 55years. The striking difference in age of disease onsetdisguises the much worse prognosis of MSA cases.Among the five Lewy body cases, three had pureautonomic failure and two Parkinsonism with auto-nomic failure; pyramidal and cerebellar signs wereabsent. Among the 11 MSA patients, all had auto-nomic failure with pyramidal and/or cerebellar signs,and eight had additional Parkinsonism. Although inthis series no MSA patients had isolated Parkinsonismwith autonomic failure, this combination can undoub-tedly occur. No patients in either group had receivedlevodopa.

Pathologically, all 11 cases ofMSA with progressiveautonomic failure had involvement of pigmentedbrainstem nuclei. Intermediolateral cell columns wereinvolved in 10, putamen in 10 and caudate in four,whilst the olives were affected in 10, pons in six,cerebellum in 10, pyramidal tracts in five and anteriorhorns in four.

Takei and Mirra, in 1973, described seven cases ofstriatonigral degeneration (with pathology).'3 Six ofthe seven had been diagnosed in life as Parkinson'sdisease, four had pyramidal signs, two ataxia and oneorthostatic hypotension. Pathologically, in only onecase was involvement limited to the striatum andnigra. They concluded: "Although we treated all ourseven cases as examples of SND, which implies adisorder involving two sites, other structures wereinvolved in some cases as well. The term is onlyjustified with the understanding that these two sites arethe most dominant ofmultiple sites ofinvolvement....It would be more meaningful to group these condi-tions under a simple name, such as multiple systematrophy". Unfortunately, they got carried away withthe attractiveness of this notion, proposing thatHuntington's disease, Pick's disease and Friedreich'sataxia, among others, be included under this label.Whilst multisystem degeneration might be a usefulshorthand term for these different conditions, the termmultiple system atrophy refers only to the specificdisease (or possibly group of diseases) addressed here.

Returning to the cerebellar presentation of MSA,Miyazaki'4 in 1978 described 45 patients labelled

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

Multiple system atrophy-the nature ofthe beastclinically as suffering from sporadic adult onsetOPCA. Twenty had positive autonomic function testsfor subclinical orthostatic hypotension; 17 of these(85%) subsequently developed urinary incontinence,seven (35%) had extrapyramidal symptoms, and eight(40%) had pyramidal signs.

In their review of striatonigral degeneration,'5Adams and Salam-Adams excellently summarised theclinical heterogeneity of MSA: "SND, OPCD andSDS symptomatically can occur singly or combined inall possible ways. Each of the clinical syndromes maymark the beginning of the disease, but if the patientsurvives for many years the appearance of the othertwo syndromes can be predicted". Unfortunately, thereview ends by proposing the term "striatonigral-cerebellar-autonomic degeneration"! An additionalproblem in clinical assessment of these patients is thatit can be well nigh impossible to state whether a patientwith severe Parkinsonism incorporating markeddysarthria and postural instability has cerebellar signs,and vice-versa. Moreover, the increase in tendon jerksin a rigid limb and the occasional presence of a"striatal toe" or "dystonic foot response" makes itdifficult to recognise pyramidal signs in extra-pyramidal disease.One of the particular features of multiple system

atrophy is that it sits astride a number of disciplinesand sub-specialities. Part of the difficulty experiencedby many physicians in conceptually getting to gripswith the condition stems from precisely this character-istic. For me, the analogy of the blindfolded menexamining different parts of an elephant and comingaway with different impressions of the nature of thebeast is an excellent one. Much of what has beenwritten about MSA has been written by doctors with aspecial interest in the autonomic nervous system, forobvious reasons since the disease offers such anexcellent model for its understanding. However,prominent autonomic symptoms may be absent invery many patients early in the disease, and indeedthroughout its course in some subjects. My ownperspective of MSA, viewing from a Parkinson'sdisease clinic, is of an irregular lumpy iceberg. Abovethe surface is one mini-iceberg of pure autonomicfailure, and another of OPCA. Beneath the waves liethe bulk ofMSA cases, these "striatonigral" variantswho, unrecognised, swell the ranks of Parkinson'sdisease clinics, participating in drug trials and trans-plantation programmes. The gradual recognition oftheir existence owes itself to three factors. Firstly,exposure to many Parkinsonian patients makes iteasier to recognise what is typical and hence what isatypical. Secondly, the pathological proof in thosecases clinically recognised as MSA who have had apost-mortem examination helps to confirm and hencevalidate clinical diagnostic criteria. And thirdly the

81picture emerging from large post-mortem series ofParkinsonian brains is that cases of MSA are morecommon than previously thought, and also frequentlymis-diagnosed in life.

Before examining larger and more recent series, it issalutary to consider a paper from Tygstrup andN0rholm'6 in 1963, prior to the introduction oflevodopa. They reported the necropsy results in 12patients from a larger series who had died from 1 weekto 4 years after a stereotactic operation for Parkinson-ism. All showed degeneration of substantia nigra.However, only five had Lewy bodies: two of theremaining seven had multiple system atrophy. Takeiand Mirra'3 found MSA in 7 9% of 89 brains ofpatients with Parkinsonism. This led them to feel "thatstriatonigral degeneration is not an extremely raredisease, but that this group constitutes a significantproportion of Parkinsonian cases." Duvoisin, largelyon clinical grounds, estimated that in his own consult-ing practice OPCA accounts for 5 to 6% of allParkinsonism.'7 In my own experience in the King'sCollege Hospital Parkinson's disease clinic, almostevery study on Parkinson's disease has includedpatients who later turned out to have clinical orpathologically proven MSA. The difficulty in diag-nosis is greatest in untreated patients: four of20 (20%)patients in a neuropsychological study of de novopatients with Parkinson's disease'8 turned out to havepathologically proven (n = 1) or clinical (n = 3)MSA. Jellinger (personal communication) found it in11% of 110 brains in his 1957-70, 3-6% of 490 brainsin his 1971-87, and 5-1% of 600 brains in his total1957-87 series ofbrains from patients with Parkinson-ism dying in three hospitals in and around Vienna. Inthe United Kingdom, the Parkinson's Disease Societyestablished a Brain Bank in 1984. Patients volunteerfor the register, and are assessed at yearly intervals by aconsultant neurologist with a special interest in Park-inson's disease. Disregarding those cases especiallysent to the Brain Bank by neurologists because ofatypical disease, 11% of the first 83 brains receivedhad MSA, and a further 7% did not have IPD. It ispossible that this figure for MSA is an over-estimate ofthe prevalence of the condition, since clearly in thefirst years of operation of a brain bank more regis-tered individuals with MSA than with IPD will diebecause of differing prognosis. It should, however,reflect the incidence of the disease. MSA is thereforenot a rare disease. If, say, 10% of the Parkinsonianpopulation of the UK were affected, this would makeprevalence 16-4 per 100,000, that is, more than twicethat of Huntington's disease.An idea of the age of onset and prognosis can be

obtained from tables in the large reviews of patho-logically proven MSA of Oppenheimer'9 (n = 41, in achapter on the neuropathology of progressive auto-

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

82nomic failure) and of Gosset et al0 (n = 35, in anarticle on SND and OPCA, reproduced in Adams andSalam-Adams' chapter in the Handbook of ClinicalNeurology on SND'5). Between them, these tablesdocument 59 different cases of pathologically provenMSA with age of onset in 58 and disease duration in57. Median age ofdisease onset was 53 years (range 36to 74), with 62% starting between 45 and 59 years, and90% between 40 and 64 years. Older patients withParkinsonism and autonomic failure are thus morelikely to have Lewy body disease. Median diseaseduration to death was 5 years (range 1 to 11), and 88%were dead within 7 years of disease onset. Someauthors have stressed a male preponderance of cases.Thus, in considering 41 cases of "autonomic failurewith MSA", Oppenheimer found 26 males and 15females. However, Gosset et al, considering 35 cases of"SND and OPCD" found 14 males, 20 females andone sex unknown. Taken together, these seriesdocument MSA in 58 cases of known sex. Males wereaffected in 31 cases, females in 27. It is possible thatbecause impotence and urinary symptoms tend to benoted more in males than anorgasmia and urinarydisturbance in (especially parous) females, anartifically high rate of autonomic disturbance may berecorded among the former. However, MSA seems toaffect the sexes equally.

Nosology and diagnostic criteriaIt can reasonably be asserted that Shy and Dragerwere responsible for first placing the combination ofautonomic failure, neurological signs and theassociated CNS pathology firmly on the neurologicalmap. However, confusion reigns over how the termShy-Drager syndrome should be used. The twopatients they described had clinical features incom-patible with idiopathic Parkinson's disease (IPD). Theterm cannot therefore be used as a form of shorthandto refer simply to Parkinsonism with autonomicfailure (which can be due either to Lewy body diseaseor MSA), though many physicians use it that way.Also, should the autonomic failure always precede theneurological features? This uncertainty is exemplifiedby Chokroverty,2' who wrote: "Although there areoccasional reports of patients with Shy-Drager syn-drome presenting initially with somatic neurologicalmanifestations before dysautonomic symptoms, it isdifficult to decide whether these patients do in facthave the Shy-Drager syndrome". Moreover, one ofhispreviously published cases (case 1 in an article inBrain' entitled "The syndrome ofprimary orthostatichypotension") developed somatic neurologicalfeatures 31 years after the onset ofpure dysautonomiaand, at necropsy, had classic pathological findings of"the Shy-Drager syndrome"." Does Shy-Drager syn-drome refer only to the full clinical syndrome they

Quinn

described (which I have never seen), or to the path-ology which revealed atrophy ofmultiple systems? Myown view is that, whilst acknowledging their majorcontribution in recognising the somatic neurologicalfeatures associated with some cases of autonomicfailure and in describing the underlying pathology, weshould cease to use the term Shy-Drager syndromesince its meaning has become so imprecise. If the termis to be retained at all, then I think it should refer to theclinical picture of Parkinsonism with autonomicfailureplus features incompatible with IPD, viz lack ofresponse to levodopa, or the presence of pyramidal orcerebellar signs.

Progressive autonomic failure (PAF) is a term thatgradually supplanted idiopathic orthostatic hypoten-sion. More recently, terminology has again changed,so that PAF now stands for pure autonomic failure.23But how is autonomic failure defined? All

physicians recognise "barn-door" autonomic failurewhen confronted with it, but where does mild auto-nomic dysfunction (as seen in many patients withParkinson's disease and also some normal subject) endand autonomic failure begin? The autonomic researchcommunity have not yet agreed universal criteria for adiagnosis of autonomic failure.2' There are certainly anumber of tests of autonomic function which havebeen validated in control populations, and for which itis possible to state that an individual lies outside thenormal range, or one or two standard deviations fromthe mean. But how many of these single tests need tobe how abnormal before autonomic failure is diag-nosed, and how can the diagnosis be made in lesssevere cases without recourse to an autonomic func-tion laboratory? McLeod and Tuck' require abnor-mal results in two or more of the following tests: theresponse of blood pressure to change in posture andisometric contraction, heart rate response to standing,variation in heart rate with respiration, Valsalva ratio,sweat tests and plasma noradrenaline measurements.Moreover, none of the usual tests of autonomicfunction can distinguish between autonomic failure aspart of MSA and as part of Lewy body disease,although dynamic tests of circulating noradrenalineconcentrations may help (see below).Even the definition of orthostatic hypotension

varies greatly from one group to another (table 1).132Beyond the question of what numerical drop insystolic/diastolic/mean blood pressure on assumingthe erect posture is significant, other confoundingfactors abound. How long should the subject be supinebefore standing? Should the patient's blood pressurebe recorded after active standing (some MSA patientsmay be unable to do this because of their motordisability) or using a tilt-table (the latter will give adifferent pulse and blood pressure profile),33 and forhow long? Should Korotkow phase IV or V be used?

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

83Multiple system atrophy-the nature of the beast

Table I Some definitions oforthostatic hypotension

Schatz et al, 19635: The simultaneous occurrence of dizziness and syncope with a decline in standing blood pressure of 30/20 mm Hg.Thomas & Schirger, 1963a: The simultaneous occurrence of giddiness, syncope or both, in association with a persistent drop of at least 50mm Hg in systolic and a corresponding drop in diastolic pressure upon standing.

Johnson et al, 1965f: A fall in systolic pressure of more than 20 mm Hg.Nick et al, 19672: An orthostatic drop of > 40 mm Hg systolic and > 30 mm Hg diastolic.Thomas & Schirger, 197029: A minimal orthostatic drop of 25 mm Hg in diastolic blood pressure from supine to standing position on at least

one ofmany blood pressure readings.Thomas et al, 19813°: A consistent and persistent fall of 30 mm Hg or more in systolic and 15 mm Hg or more in diastolic pressure,accompanied by clinical symptoms.

Cohen et al, 1987': The presence of orthostatic syncope or presyncope and sphygmomanometric confirmation (orthostatic reduction insystolic BP by > 30 mm Hg, or mean arterial pressure by > 20 mm Hg).

Bannister & Mathias, 198832: When a fall of more than 20 mm Hg systoiic pressure on standing is found in a patient with symptoms, furtherinvestigation is justified.

Should symptoms be produced? Does failure to recordan abnormal drop on one occasion mean the patientdoes not have orthostatic hypotension? Does a singlerecording ofabnormal postural drop mean the subjecthas orthostatic hypotension? What importance shouldone attach to post-prandial orthostatic hypotension,or when the patient is being treated with levodopa or adopamine agonist?

Quite apart from these uncertainties about thediagnosis of autonomic failure, the diagnosis ofMSAalso rests on shifting sands. The most recent criteriaused at the Mayo Clinic3' require: "The presence ofautonomic failure (see table 1). When striatonigral orolivopontocerebellar involvement was confirmed onneurological examination, the diagnosis of MSA wasmade". Unless levodopa unresponsiveness is specifiedthese criteria could potentially include as MSApatients with Lewy body Parkinson's disease withautonomic failure!The term "Parkinson plus syndrome" has recently

enjoyed favour among neurologists attempting toseparate out cases with atypical Parkinsonism whosefeatures are incompatible with the idiopathic disease.Whilst this expression is useful, particularly in avoid-ing definitive alternative diagnosis where none can bemade with certainty, it neglects the single mostsuspicious feature, viz the lack of response tolevodopa. Perhaps "Parkinson minus syndrome"would be just as clinically useful. Thus, a dramaticclinical motor response to treatment with levodopa atan adequate dose for an adequate period of time is forpractical purposes always seen in patients with patho-logically proven uncomplicated Lewy body Parkin-son's disease.4 A poor or absent response is the rule inMSA patients, although transient significant benefithas occasionally been reported.35 In the absence ofclear pyramidal or cerebellar signs, this differenceconstitutes a major clinical tool for differentiatingbetween IPD and MSA. It therefore seemsappropriate that it should be included among diagnos-tic criteria for MSA with Parkinsonian features. Onedrawback, however, concerns patients who cannottolerate an adequate trial of levodopa (with peripheral

decarboxylase inhibitor) because of faintness or sick-ness. Nevertheless, whilst some such patients seem tohave IPD, a higher proportion of pathologicallyproven MSA subjects have manifested suchintolerance to levodopa preparations. Curiously,some of these are able to tolerate bromocriptine (withdomperidone), but usually without dramatic resultantclinical benefit.

Although a few patients may display inappropriatelaughter or crying, numerous clinical accounts ofpathologically confirmed MSA have called attentionto the striking preservation of intellect in the face ofsevere motor disability. Dementia seems therefore notto be an integral feature of MSA. CoincidentalAlzheimer's disease can certainly occur in MSApatients,3s6- but at a rate no higher than among thegeneral age-matched population. In addition, a num-ber of cases showing Parkinsonism, dementia andautonomic dysfunction, with poor or absent responseto levodopa, in life have been shown at post-mortem tobe suffering from "cortical" or "diffuse" Lewy bodydisease.39 For practical purposes, dementia rules outuncomplicated MSA.Lewy bodies have been found restricted largely to

pigmented brainstem nuclei in some patients alsoshowing the striatal pathology of MSA.404 We nowappreciate that incidental Lewy bodies, with somedegree of cell loss, may be seen in the brains of manyindividuals who did not manifest clinical Parkinson-ism during life.49 It does not now seem necessary topropose that cases showing MSA and Lewy bodieshave a "transitional variant" of the disease," merelythat this is a chance superimposition of a commonpathological finding. However, the finding of inciden-tal Lewy bodies does raise the possibility of anerroneous pathological diagnosis of IPD if thestriatum is not also examined. IPD must have Lewybodies in substantia nigra, but the presence of Lewybodies and cell loss in substantia nigra is not synony-mous with clinical IPD, and hence does not excludeMSA.The question of inheritance of MSA is best ap-

proached through those cases presenting with Parkin-

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

84sonism and autonomic failure. There has been only ahandful of cases with pathologically proven MSAwhere one other family member is affected by Parkin-sonism,12 1340 a rate corresponding to the chanceoccurrence of IPD. The family reported by Lewis in1964' is always cited as evidence that MSA can beinherited, but there was no pathology and the clinicalfeatures were atypical. Thus, the four cases examinedhad had the disease for 7, 34 and 19 years, and 6months respectively. The clinical features includedorthostatic hypotension, cerebellar, pyramidal andextrapyramidal signs, but there was also amyotrophy,fasciculation, footdrop and pes cavus. Given thisunusual clinical picture and given that no patho-logically proven cases of truly familial typical MSAhave been recorded, it seems likely that this family didnot have MSA. Could they perhaps have had theadult-onset familial leukodystrophy with autonomicfailure described by Eldridge et al?5' The othercelebrated instance ofinherited so-called striato-nigraldegeneration is the Joseph family.52 In this largedominant pedigree, disease onset usually occurs in thetwenties or thirties, and the clinical features varyconsiderably. Spasticity, ophthalmoplegia, peripheralneuropathy and cerebellar signs frequently occur, withonly mild extrapyramidal signs and without promi-nent autonomic failure. The pathology showsinvariant symmetrical degeneration of the cerebellumand thoracic cord, variable degeneration of the basispontis, oculomotor nuclei, peripheral nerves, nigra,striatum, but the inferior olives are always spared.MSA of SND predominance seems therefore to be asporadic disease.When one turns to OPCA, the genetic picture is

more complex. Among adult-onset cases of OPCA,rather less than half seem to be inherited, as adominant trait; the remainder have apparently arisensporadically. Neither clinical features alone nor path-ology may suffice to differentiate between the inheritedand the sporadic form in an individual case, althoughgroup differences exist. Thus, in Berciano's 1982review53 of 117 cases of pathologically confirmedOPCA, 39% of 54 familial cases showed Parkinsonism(55% of 63 sporadic cases), 50% pyramidal signs(46%), 96% cerebellar signs (87%) and 39% sphincterdisturbance (48%). Pathologically, in addition toinvolvement of olives, pons and cerebellum, striatumwas affected in 22% of familial cases (38% ofsporadic) and substantia nigra in 46% (48%). Thus,neither Parkinsonian features nor sphincter distur-bance, and neither striatal nor nigral pathology,clearly differentiates between familial and sporadiccases of OPCA. However, mean age of onset inhereditary cases is earlier than in sporadic cases (28years versus 49 years in the clinicopathological studyof Berciano53; 39 years versus 49 years in the clinical

Quinnstudy ofHarding4). Moreover, mean disease durationto death in familial cases was 14-9 years, and insporadic cases was 6-3 years.53 In the clinical study ofHarding' on 36 living sporadic cases of late onsetcerebellar ataxia, mean duration of symptoms was12-5 years. Only three (8%) patients had impassivefacies, but no other extrapyramidal features. Ninepatients (25%) had urinary symptoms. This seriesclearly differs from what we recognise as MSA, partlybecause only cerebellar ataxia was needed forinclusion, and perhaps partly because, in an effort toexclude cases with paraneoplastic cerebellar degenera-tion, patients with a subacute course or a diseaseduration of less than two years were excluded. Ophth-almoplegia was more frequent, and pigmentary retinaldegeneration and optic atrophy only found, in familialcases.5' Since the predominantly striatonigral variantofMSA is sporadic, it seems appropriate to include inthe rubric of MSA only sporadic cases of OPCA(although this restriction would be at variance with thewider usage of the term by Oppenheimer himself toinclude familial OPCA). Moreover, to be consideredas clinically probable cases of MSA, they should alsodisplay either Parkinsonian features or autonomicfailure, or both.

Neuropathological examination of the brain andspinal cord cannot at present distinguish betweenmany cases of familial OPCA and sporadic cases ofMSA, although the value of intermediolateral columncell counts in this respect has not been fully assessed.Should this deter one from separating such cases? Ithink not. A myocardial infarct due to coronaryatherosclerosis may look the same in someone withfamilial hypercholesterolaemia as in a heavy eater andsmoker, yet the ultimate cause is different: on the onehand genetic, on the other acquired. It should notsurprise us that a similar or even an identical patho-logical picture should develop via a common mechan-ism triggered by different root causes, one genetic andone perhaps acquired.A number of MSA subjects may display abnormal

eye movements. Most commonly, this consists ofimpaired or absent convergence, hypometric volun-tary saccades and/or saccadic pursuit, and sometimesa supranuclear gaze palsy for up (and rarely down)gaze.55 Indeed, MSA is one of a number of causes ofthe clinical picture of progressive supranuclear palsy,which is often erroneously taken to be synonymouswith Steele-Richardson-Olszewski disease. Thus, anumber of cases of MSA would fulfil the clinicaldiagnostic criteria for Steele-Richardson diseaseproposed by both Golbe et al56 and Lees.5" Because ofthe potential for confusion between the two diseases, Iwould propose that a predominant downgaze paresisshould constitute an exclusion criterion for the clinicaldiagnosis of MSA.

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

Multiple system atrophy-the nature of the beastTable 2 Multiple system atrophy: proposed diagnosticcriteria

SND type OPCA type(predominant (predominantlyparkinsonism) cerebellar)

Sporadic adult-onset Possiblenon/poorlylevodopa responsiveparkinsonism*

Above, plus severe Probable Sporadic adult-onsetsymptomatic cerebellar ± pyramidalautonomic failuret syndrome*+ /or cerebellar signs with severe+ /or pyramidal signs symptomatic

autonomic failuret+ /or parkinsonism

p/m confinned Definite p/m confirmed

*Without DSM III dementia, predominant downgaze PSNP orother identifiable cause.tPostural syncope or presyncope and/or marked urinaryincontinence or retention not due to other causes.Sporadic: one other case of typical clinical IPD among 1st or 2nddegree relatives allowable.Adult onset: onset age 30 years or above.

On the basis of the above considerations, andfollowing the proposals of groups formulating diag-nostic criteria for multiple sclerosis' and for Alz-heimer's disease,57 I propose the diagnostic classifica-tion set out in table 2. Undoubtedly this will needfurther modification in the future, particularly inregard to the definition of autonomic failure, and alsoin the light of pathological verification of casesfulfilling these diagnostic criteria. Moreover, thisclinical diagnostic schema mirrors the drawbacksconsidered above, in that it fails to exclude cases ofSteele-Richardson disease not yet displaying apredominant downgaze paresis.

"Redflags" (table 3)Besides the poor response to levodopa, and theadditional presence ofpyramidal or cerebellar signs orautonomic failure as major diagnostic criteria, certainother clinical features may either raise a suspicion ofMSA, or at least suggest that one might not be dealingwith IPD.

Symmetrical onset and absence ofa classical restingtremor have often been proposed as features distin-guishing MSA from IPD. However, marked andpersistent asymmetry and the presence of a classicalresting tremor may both be seen in cases of path-ologically proven MSA. Early instability and fallssuggest a cause other than IPD. A consequence of thisis that among patients whose Parkinsonism is revealedby, and persists after, treatment with neurolepticagents (eg prochlorperazine) given to combat vaguefeelings of unsteadiness, those with alternative causes

85Table 3 "Redflags": clinicalfeatures raising doubts abouta diagnosis ofideopathic Parkinson's disease

Early instability and falls ?Absent/atypical levodopaAIMs

Rapid progression ?Disproportionate antecollisIrregular jerky tremor ?PH of hypertensionMyoclonus ?ContracturesPoor or absent response to an ?Raynaud's or ergotism

adequate trial of levodopaAutonomic failureAbnormal eye movements beyondIPD range

Severe dysarthria/dysphoniaPain unrelieved by levodopaSevere levodopa intolerance Peripheral neuropathy

*Pyramidal signs*Cerebellar signstDSM III dementia

Never part of IPD.tUncommon in both IPD and MSA.

of Parkinsonism may be over-represented.Rapid clinical deterioration despite dopaminergic

treatment is also highly suspicious. Indeed, manyphysicians recognise this, and only consider MSA inParkinsonian patients with cerebellar features orpyramidal signs together with profound autonomicfailure who deteriorate towards death within 5 or 6years. Yet a number of cases ofMSA survive 10 yearsor longer after the onset of clinical symptoms or signs,and may never develop cerebellar or pyramidal signs.Failure to recognise this leads to a self-fulfillingprophecy whereby MSA cases with longer diseaseprogression are not diagnosed as such in life.Similarly, failure to recognise the unusual nature ofthe case makes a post-mortem examination less likelyto be sought.

Dating the onset ofthe disease is also problematical.Should one take the first sign of relatively non-specificimpotence or sphincter disturbance, or date diseaseonset only from the onset of symptomatic posturalhypotension or of the first motor symptoms and signs?Indeed, do any of these features in fact correspond tothe actual onset of the disease process? There is nowevidence to suggest that the onset of IPD mayconsiderably antedate the first clinical signs, perhapsby 20 years or even more. Given the redundancymargin of central neuronal populations, it seemsprobable that in MSA, too, disease onset may longantedate clinical features. In this respect, it is ofinterest to examine the past history and previousrecords of cases of clinical MSA. A number have hadarterial hypertension diagnosed, and sometimestreated, several years before the first symptoms, onlyto later become normo- or hypotensive. Could thisrelate to early denervation hypersensitivity causingsupine hypertension? It is tempting also to speculateon the significance of Raynaud's phenomenon occur-ring before, or during, treatment with dopaminergic

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

86drugs. I vividly recall a patient with apparentuntreated IPD who, on pergolide monotherapy,developed severe vasospasm of all fingers of bothhands. His Parkinsonism did not improve and he wenton to die of pathologically proven MSA. Was theRaynaud's a manifestation of an ergot drug acting onan already abnormal autonomic nervous system?Another man had attended 27 years earlier with a

complaint of dizziness of unknown cause, whichsubsequently settled. Could this have been, in retro-spect, the first sign of his illness?

Myoclonic jerks, usually affecting the fingers, ofsmall amplitude, and often stretch-sensitive, occur in anumber of patients with MSA, but are otherwise rarein non-demented Parkinsonians. An irregular, jerkytremor of the hands is also a feature more commonlyseen in MSA than IPD.

Bizarre contractures of the hands may be morecommon in non-idiopathic Parkinsonism. Well recog-nised in post-encephalitic cases, and in Parkinsonism-dementia, they can also occur in MSA.363858 A morefrequent and characteristic feature suggesting MSA isthe development ofa relatively fixed, disproportionateantecollis53859 hampering feeding, communication andvision.

Speech is almost always more severely affected inMSA than in IPD. As well as the low volumemonotony of Parkinsonism, a quivering, irregular,severely hypophonic or slurring dysarthria is "often so

characteristic that the diagnosis can be suggested bylistening to the patient on the telephone".59 Res-piratory stridor, especially at night, is also a helpfuldiagnostic pointer, since vocal cord paralysis com-monly occurs.'Marked focal muscular atrophy and fasciculations,

whilst reported in some cases with MSA,7 are in myexperience most unusual.

Pain, characteristically in, or maximal in, the mostseverely affected limbs, is not uncommon in IPD. Insuch cases, it is usually dramatically improved orabolished when levodopa turns the patient "on".6'Some cases of MSA may experience similar deepaching limb pains59 which, in contrast, persist through-out the day and are not relieved by levodopa.

In a patient with Parkinsonism, the significance ofthe presence and degree of autonomic symp-tomatology may be very difficult to assess. I recentlyasked 34 patients (23 male, 11 female, mean age 49-6,SD 8-9 years, mean disease duration 11, SD 4 9 years)to complete an autonomic questionnaire. All had beendiagnosed as suffering from Parkinson's disease, andwere attending a weekend meeting for youngersufferers organised by the Parkinson's DiseaseSociety. The group could certainly have included oneor more cases of MSA, but overall I had no reason todoubt the diagnosis of IPD. Nocturia at least once

Quinnnightly was experienced by 85%, daytime frequency ofsix times or more by 74%, and 71% had been "caughtshort" with a dribble at least occasionally. Twenty fiveper cent said their hands and feet were bone dry in hotor humid conditions, 15% experienced increasingfaintness between the onset ofParkinson's disease anddopaminergic treatment, and 37-5% subsequent totreatment (corresponding figures for postural black-outs were 6% and 9% respectively). Normal erectionswere achieved only occasionally or never by 23% ofmales, with occasional erectile problems in as many as73%. Our judgement of the diagnostic significance ofautonomic symptoms in Parkinsonism as a pointer toautonomic failure must be weighed against these rates,and also those for an age-matched control population(work in progress).

InvestigationsAutonomic function tests can demonstrate variousaspects of autonomic failure, but not its cause.However, resting supine plasma noradrenaline (NA)levels, and their response to standing, may help todifferentiate between groups of subjects with MSAand those with pure autonomic failure (which mostcommonly shows Lewy body pathology'"). Thus,MSA patients usually have normal or slightly elevatedplasma NA levels, whereas those with PAF usuallyhave low plasma NA levels.62 In contrast to normalsubjects, neither group increases the plasma NA levelon standing.62 However, the clinical usefulness ofplasma NA levels as a diagnostic tool in individualcases has not been demonstrated.62Computed tomographic (CT) scanning appears to

be of only limited usefulness in the diagnosis ofMSA.In subjects with a predominantly cerebellar presenta-tion, cerebellar and/or brainstem atrophymay be seen,but since this is also seen in many familial cases ofadult onset cerebellar atrophy this finding may nothelp in differential diagnosis. Among cases presentinga predominantly, or a pure, Parkinsonian syndromethe CT scan is usually normal. In only a minority ofcases will the scan demonstrate clear atrophy ofposterior fossa structures. Striatal CT images arenormal in all forms of MSA.

Magnetic resonance imaging, however, may wellprove useful in demonstrating striatal pathology inMSA, and its absence in IPD. Early studies using 1 5tesla T2 weighted images suggested that cases of"Parkinson's disease" with rapid progression andpoor response to treatment showed altered signalparticularly in putamen, and it was proposed thismight be due to abnormal deposition of iron in thisstructure. The debate over iron content of the puta-minal pigments63 and its relationship to the MRIimages continues unresolved at present. However,whatever the ultimate cause of the altered signal, it is

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

Multiple system atrophy-the nature ofthe beastbeginning to look as if MRI scanning can detect theputaminal pathology so characteristic of striatonigralcases ofMSA, thus differentiating them from IPD.""

Positron emission tomography (PET) scans ofMSApatients have so far been few in number. It comes as nosurprise that '8F-fluorodopa scans show low concen-tration of activity in striatum, as in IPD.6" The mostmodem scanners with sufficient resolution alsodemonstrate a lower caudate uptake in MSA subjectsin contrast to relatively preserved caudate uptake inIPD.' However, it is likely that measures of post-synaptic function will prove more useful in differen-tiating between MSA and IPD, so that the results of"C-raclopride studies are awaited with interest. Ifthese scans can demonstrate clear and unequivocaldifferences, and if a dopamine receptor ligand can bedeveloped for SPET studies, then SPET scanning mayemerge as a useful diagnostic tool.

Electrophysiological investigations have beenapplied to a number of patients with clinical MSA.Nerve conduction studies may sometimes reveal asubclinical polyneuropathy,3' and EMG studies maysuggest neurogenic atrophy in some individuals.7Thermal threshold, visual and somatosensory evokedresponses, and electrical and magnetic central motorconduction times have not yet been adequatelyexplored. The EEG is not useful. One group hasdemonstrated an abnormal latency and an abnor-mality of the amplitude ratio of waves V/I of thebrainstem auditory evoked responses (BSAEPs) in 11and 13 patients out of 14 respectively with a clinicaldiagnosis of MSA.' However, our own unpublishedinvestigations showing normal BSAEPs in a numberof, mostly less severely affected, subjects (includingone with pathologically proven MSA) lead us to doubtthe specificity of this abnormality in patients earlier inthe course of the disease.One particular electrophysiological investigation

appears to be extremely useful in supporting a clinicaldiagnosis of MSA. In an initial study in 14 patientswith clinical MSA (Parkinsonism, autonomic failureand pyramidal signs in all 14, additional cerebellarsigns in two), individual motor units recorded from thestriated component of the urethral sphincter wereconsistently abnormal, showing polyphasia and longduration.' Subsequent studies in a further 26 subjectswith probable, 15 with possible MSA, and 13 withprobable IPD, have shown that this investigation hashigh specificity (0 92) but less sensitivity (0.62) indifferentiating between probable MSA and probableIPD.' Moreover, one probable MSA subject withabnormal results has since died, with pathologicalconfirmation of the diagnosis.

It is therefore apparent that no single clinical featureor investigational abnormality is diagnostic of MSA.Rather, the complete clinical picture together with the

87results of a number of investigations must be takeninto account in making the diagnosis. For the timebeing, only clinical criteria have been advanced intable 2. However, when more clinicopathologicalcorrelation has permitted validation of diagnostictests, used singly or, more likely, in concert, we mayarrive at a schema similar to that for multiple sclerosis,where categories of laboratory supported diagnosis'have appeared.

TreatmentSadly, the treatment of MSA is most disappointing.Levodopa preparations may transiently give usefulimprovement, but often either cause no improvement,or give rise to intolerable nausea and vomiting and/orsymptomatic postural hypotension." However, itshould be realised that some patients who seem not tobe responding to levodopa can still deteriorate sig-nificantly when the drug is withdrawn. Head-up tilt ofthe bed at night, elastic support stockings, fluorocor-tisone and/or indomethacin, caffeine, DDAVP oratrial pacing may help cardiovascular problems.24"Anticholinergics can reduce urinary frequency (butmay precipitate retention). Retention with overflow,or incomplete bladder emptying with infections, canbe managed with intermittent self-catheterisation (ifhand mobility permits) or indwelling suprapubic orurethral catheterisation. High fibre diet, bulk lax-atives, suppositories or enemas may be needed forconstipation. Tracheostomy for intermittent res-piratory stridor (especially at night) must be con-sidered with the utmost care. It may only cruellyprolong a precarious existence before inevitable death.Similarly, cricopharyngeal myotomy or gastrostomyin patients with severe dysphagia is seldom, if ever,indicated. The paramedical specialities of physio-therapy, occupational therapy, speech therapy (oftendirected at swallowing problems and communicationaids rather than improving speech per se) and socialwork often are of much more practical use thananything the physician can offer. With their graveprognosis and their lack ofresponse to levodopa, thesepatients are, to my mind, more in need of a break-through in neuronal implant treatment than thosewith idiopathic Parkinson's disease. If embryonicnigral implants can be shown to work in IPD, thenlogically the next patient groups to be offered (thistime striatal) implant treatment should be those withMSA.We desperately need better diagnosis ofMSA if we

are to seek the cause, and hence a preventive orretardant treatment, for this terrible affliction. In themeantime, we need to draw on all our skills andcompassion to help and comfort those who are struckby this disease.

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

88I gratefully acknowledge the inspiration and friend-ship of David Marsden who kindled my interest inmovement disorders, and to whom this article isdedicated. Thanks also to Mel Calman for drawing thefig, and to Pamela Green for typing the manuscript.

References

1 Graham JG, Oppenheimer DR. Orthostatic hypotension andnicotine sensitivity in a case of multiple system atrophy. JNeurol Neurosurg Psychiatry 1969;32:28-34.

2 Dejerine J, Thomas AA. L'atrophie olivo-ponto-cerebelleuse.Nouv Iconog de le Salpetriere 1900;13:330-70.

3 Bradbury S, Eggleston C. Postural hypotension. A report of threecases. Am Heart J 1925;1:73-86.

4 Bradbury S, Eggleston C. Postural hypotension: Autopsy on a

case. Am Heart J 1927;3:105-6.5 Langston W. Orthostatic hypotension; report of a case. Ann Int

Med 1936;10:688-95.6 Young RH. Association of postural hypotension with sympathetic

nervous system dysfunction; case report with review ofneurologic features associated with postural hypotension. AnnInt Med 1944;15:910-6.

7 Shy GM, Drager GA. A neurologic syndrome associated withorthostatic hypotension. Arch Neurol 1960;2:511-27.

8 van der Eecken H, Adams RD, van Bogaert L. Striopallidal-nigraldegeneration. An hitherto undescribed lesion in paralysisagitans. J Neuropathol Exp Neurol 1960;19:159-61.

9 Adams RD, van Bogaert L, van der Eecken H. Degenerescencesnigro-stri6es et cerebello-nigro-stri6es. Psychiat Neurol1961;142:219-59.

10 Messing Z. Atrophic olivo-ponto-c6rebelleuse dans un cas demaladie de Parkinson. Rev Neurol (Paris) 1930;60:498-9.

11 Malamud N. Atlas of Neuropathology. Berkeley: University ofCalifornia Press, 1957:256.

12 Bannister R, Oppenheimer DR. Degenerative diseases of thenervous system associated with autonomic failure. Brain1972;95:457-74.

13 Takei Y, Mirra SS. Striatonigral degeneration: A form of multiplesystem atrophy with clinical parkinsonism. In: ZimmernanHM, ed. Progress in Neuropathology, vol 2. New York: Gruneand Stratton, 1973:217-51.

14 Miyazaki M. Shy-Drager syndrome: A nosological entity? Theproblem of orthostatic hypotension. In: Sobue I, ed. Spin-ocerebellar Degenerations. Baltimore: University Park Press,1980:35-44.

15 Adams RD, Salam-Adams M. Striatonigral degeneration. In:Vinken PJ, BruynCW, Klawans HL, eds. Handbook ofClinicalNeurology vol 5 (49): Extrapyramidal disorders. Amsterdam:Elsevier Science Publishers BV, 1986:205-12.

16 Tygstrup I, N0rholm T. Neuropathological findings in 12 patientsoperated for parkinsonism. Acta Neurol Scand 1963:39(suppl4):188-95.

17 Duvoisin RC. An apology and an introduction to the olivopon-tocerebellar atrophies. Adv Neurol 1984;41:5-12.

18 Canavan AGM, Passingham RE, Marsden CD, Quinn N, WykeM, Polkey CE. The performance of patients with Parkinson'sdisease on problem solving tasks. Neuropsychologia (in press).

19 Oppenheimer D. Neuropathology of progressive autonomicfailure. In: Bannister R, ed. Autonomic Failure. A Textbook ofClinical Disorders of the Autonomic Nervous System. Oxford:Oxford University Press. 1st ed, 1982;267-83.

20 Gosset A, Pellissier JF, Delpuech F, Khalil R. Degenerescencestriato-nigrique associee a une atrophie olivo-ponto-cerebelleuse. Etude anatomo-clinique de trois cas. Discussionnosologique. Rev Neurol. (Paris) 1983;139:125-39.

21 Chokroverty S. Autonomic dysfunction in olivopontocerebellaratrophy. Adv Neurol 1984;41:105-41.

Quinn

22 Chokroverty S, Barron KD, Katz FH, del Greco F, Sharp JT.The syndrome of primary orthostatic hypotension. Brain1969;92:743-86.

23 Bannister R. Introduction and classification. In: Bannister R, ed.Autonomic Failure. A Textbook of Clinical Disorders of theAutonomic Nervous System. 2nd ed. Oxford: Oxford UniversityPress, 1988;1-20.

24 McLeod JG, Tuck RR. Disorders of the autonomic nervoussystem: Part 2. Investigation and treatment. Ann Neurol1987;21:519-29.

25 Schatz IJ, Podoisky S, Frame B. Idiopathic orthostatic hypoten-sion. JAMA 1963;186:537-40.

26 Thomas JE, Schirger A. Neurologic manifestations in idiopathicorthostatic hypotension. Arch Neurol 1963;8:204-8.

27 Johnson RH, Smith AC, Spalding JME, Wollner L. Effect ofposture on blood pressure in elderly patients. Lancet1965;i:731-3.

28 Nick J, Contamin F, Escourolle R, Guillard A, Marcantoni J-P.Hypotension orthostatique idiopathique avec syndromeneurologique complexe a predominance extra-pyramidale. RevNeurol (Paris) 1967;116:213-27.

29 Thomas JE, Schirger A. Idiopathic orthostatic hypotension. Astudy of its natural history in 57 neurologically affectedpatients. Arch Neurol 1970;22:289-93.

30 Thomas JE, Schirger A, Fealey RD, Sheps SG. Orthostatichypotension. Mayo Clin Proc 198 1;56:117-25.

31 Cohen J, Low P, Fealey R, Sheps S, Jiang N-S. Somatic andautonomic function in progressive autonomic failure andmultiple system atrophy. Ann Neurol 1987;22:692-9.

32 Bannister R, Mathias C. Testing autonomic reflexes. In: BannisterR, ed. Autonomic Failure. A Textbook of Clinical Disorders ofthe Autonomic Nervous System. Oxford: Oxford UniversityPress, 2nd ed, 1988;289-307.

33 Wieling W. Standing, orthostatic stress, and autonomic function.In: Bannister R, ed. Autonomic Failure. A Textbook of ClinicalDisorders of the Autonomic Nervous System. Oxford: OxfordUniversity Press, 2nd ed, 1988;308-20.

34 Gibb WRG. Accuracy in the clinical diagnosis of parkinsoniansyndromes. Postgrad Med J 1988;64:345-5 1.

35 Lees AJ. The treatment of multiple system atrophy: striatonigraldegeneration and olivopontocerebellar degeneration. In: Bann-ister R, ed. Autonomic Failure. A Textbook ofClinical Disordersof the Autonomic Nervous System. Oxford: Oxford UniversityPress, 2nd ed, 1988;596-604.

36 Trotter JL. Striato-nigral degeneration. Alzheimer's disease andinflammatory changes. Neurology 1973;23:1211-6.

35 Kosaka K, lizuka R, Mizutani Y, Kondo T, Nagatsu T. Striato-nigral degeneration combined with Alzheimer's disease. ActaNeuropathol (Berl) 198 1;54:253-6.

38 Neumann MA. Pontocerebellar atrophy combined with vesti-bular-reticular degeneration. J Neuropathol Exp Neurol 1977;36:321-37.

39 Burkhardt CR, Filley CM, Kleinschmidt-DeMasters BK, de laMonte S, Norenberg MD, Schneck SA. Diffuse Lewy bodydisease and progressive dementia. Neurology 1988;38:1520-8.

40 Guard 0, Sindou M, Carrier H. Syndrome de Shy-Drager; unenouvelle observation anatomo-clinique. Lyon Med 1974;231:1075-84.

41 Adams RD, van Bogaert L, van der Eecken H. Striato-nigraldegeneration. J Neuropathol Exp Neurol 1964;23:584-608.

42 Jellinger K, Danielczyk W. Striato-nigrale degeneration. ActaNeuropathol 1968;10:242-57.

43 Thapedi IM, Ashenhurst EM, Rozdilsky B. Shy-Drager syn-drome: report of an autopsied case. Neurology 1971 ;21:26-32.

44 Kaiya H. Spino-olivo-ponto-cerebello-nigral atrophy with Lewybodies and binucleated nerve cells: a case report. ActaNeuropathol 1974;30:263-9.

45 Yamamura Y, Ohama E, Yoshimura N, Atsumi T, Oyanagi S,Ikuta F. Striato-nigral degeneration, a form of multiple system

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from

Multiple system atrophy-the nature of the beastatrophy allied to olivopontocerebellar atrophy. Adv Neurol Sci1974;18:89-105.

46 Kobayashi H, Mukouyama M, Muroga T, et al. A case ofolivopontocerebellar atrophy, Parkinson syndrome and Lewybodies in the substantia nigra and locus coeruleus. Adv NeurolSci 1979;23:599.

47 Yoshimura M, Shimada H, Nakura H, Tomonaga M. Twoautopsy cases of Lewy type Parkinson's disease with Shy-Drager syndrome. Trans Soc Pathol Jap 1980;69:432.

48 Sima AAF, Hoag G, Rozdilsky B. Shy-Drager syndrome-thetransitional variant. Clin Neuropathology 1987;6:49-54.

49 Gibb WRG, Lees AJ. The relevance of the Lewy body to thepathogenesis of idiopathic Parkinson's disease. J NeurolNeurosurg Psychiatry 1988;51:745-52.

50 Lewis P. Familial orthostatic hypotension. Brain 1964;87:719-28.

51 Eldridge R, Anayiotos CP, Schlesinger S, et al. Hereditary adult-onset leukodystrophy simulating chronic progressive multiplesclerosis. N EnglJ Med 1984;311:948-53.

52 Rosenberg RN. Joseph disease: An autosomal dominant motorsystem degeneration. Adv Neurol 1984;41:179-93.

53 Berciano J. Olivopontocerebellar atrophy. A review of 117 cases.J Neurol Sci 1982;53:253-72.

54 Harding AE. Idiopathic late onset cerebellar ataxia. A clinical andgenetic study of 36 cases. J Neurol Sci 1981;51:259-71.

55 Lepore FE. Disorders of ocular motility in the olivoponto-cerebellar atrophies. Adv Neurol 1984;41:97-103.

56 Golbe LI, Davis PH, Schoenberg BS, Duvoisin RC. Prevalenceand natural history of progressive supranuclear palsy.Neurology, 1988;38:1031-4.

57 Lees AJ. The Steele-Richardson-Olszewski syndrome (progressivesupranuclear palsy). In: Marsden CD, Fahn S, eds. MovementDisorders 2, London: Butterworths, 1987;272-87.

58 Quinn NP, Ring H, Honavar M, Marsden CD. Contractures oftheextremities in parkinsonian subjects. A report of three caseswith a possible association with bromocriptine treatment. ClinNeuropharmacol 1988;11:268-77.

59 Caplan LR. Clinical features of sporadic (Dejerine-Thomas)olivopontocerebellar atrophy. Adv Neurol 1984;41:217-24.

8960 Williams A, Hanson D, Calne DB. Vocal cord paralysis in the Shy-

Drager syndrome. J Neurol Neurosurg Psychiatry 1979;42:151-3.

61 Quinn NP, Koller WC, Lang AE, Marsden CD. Painful Parkin-son's disease. Lancet 1986;i: 1366-9.

62 Polinsky RJ. Neurotransmitter and neuropeptide function inautonomic failure. In: Bannister R, ed. Autonomic Failure. ATextbook of Clinical Disorders of the Autonomic NervousSystem. Oxford: Oxford University Press, 2nd ed. 1988;321-47.

63 Borit A, Rubinstein LJ, Urich H. The striatonigral degenerations.Putaminal pigments and nosology. Brain 1975;98:101-12.

64 Drayer BP, Olanow W, Burger P, Johnson GA, Herfkens R,Riederer S. Parkinson plus syndrome: Diagnosis using highfield MR imaging of brain iron. Radiology 1986;159:493-8.

65 Pastakia B, Polinsky R, DiChiro G, Simmons JT, Brown R, WenerL. Multiple system atrophy (Shy-Drager syndrome): MRimaging. Radiology 1986;159:499-502.

66 Rutledge JN, Hilal SK, Silver AJ, Defendini R, Fahn S. Study ofmovement disorders and brain iron by MR. Am J Neuroradiol1987;8:397-41 1.

67 Brooks DJ, Salmon EP, Bannister R, Mathias C, FrackowiakRSJ. The integrity of the dopaminergic system in multiplesystem atrophy and pure autonomic failure studied with PET.J Neurol Neurosurg Psychiatry 1989 (in press).

68 Prasher D, Bannister R. Brain stem auditory evoked potentials inpatients with multiple system atrophy with progressive auton-omic failure (Shy-Drager syndrome). J Neurol NeurosurgPsychiatry 1986;49:278-89.

69 Kirby R, Fowler C, Gosling J, Bannister R. Urethro-vesicaldysfunction in progressive autonomic failure with multiplesystem atrophy. J Neurol Neurosurg Psychiatry 1986;49:554-62.

70 Eardley I, Quinn NP, Fowler CJ, Kirby RS, Parkhouse HM,Marsden CD, Bannister R. The value of urethral sphincterelectroyography in the differential diagnosis of parkinsonism.Br J Urol (in press).

71 Editorial. Management of orthostatic hypotension. Lancet1987;i: 197-8.

Protected by copyright.

on April 9, 2022 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.52.Suppl.78 on 1 June 1989. D

ownloaded from