3'ournal ofNeurology, Neurosurgery, and Psychiatry 1996;61:615-620 615 Natural history of progressive supranuclear palsy J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.6.615 on 1 June 1996. Downloaded from (Steele-Richardson-Olszewski syndrome) and clinical predictors of survival: a clinicopathological study

I Litvan, C A Mangone, A McKee, M Veiny, A Parsa, K Jellinger, L D'Olhaberriague, K Ray Chaudhuri, R K B Pearce

Abstract Keywords: progressive supranuclear palsy; natural his- Objective-To analyse the natural history tory; survival ofprogressive supranuclear palsy (PSP or Steele-Richardson-Olszewski syndrome) Progressive supranuclear palsy (PSP or Steele- and clinical predictors of survival in 24 Richardson-Olszewski syndrome) causes patients with PSP confirmed by necropsy, postural instability, supranuclear vertical oph- who fulfilled the NINDS criteria for a thalmoplegia, parkinsonism unresponsive to neuropathological diagnosis of typical levodopa, pseudobulbar palsy, and mild PSP. dementia.' 2 However, patients with PSP with- Methods-Patients were selected from the out ophthalmoplegia or dementia, or present- research and clinical files of seven med- ing only with dementia or akinesia, have also ical centres involving tertiary centres of been reported.3-7 Austria, England, France, and the United In the absence of laboratory markers for the States. Clinical features were analysed in diagnosis of PSP, neuropathological examina- Neuroepidemiology detail. The patients' mean age at onset of tion remains the "gold standard" for its diag- Branch, National PSP was 63 (range 45-73) years. nosis. Neuropathologically confirmed cases of Institute of Results-The most frequent clinical fea- corticobasal degeneration, multiple system Neurological Disorders tures in at least of the and Stroke, National (occurring 75% atrophy, diffuse Lewy body disease, cere- Institutes ofHealth, patients) were early postural instability brovascular disease, subcortical gliosis, and Bethesda, Maryland, and falls, vertical supranuclear palsy, aki- prion disease have been clinically misdiag- USA netic-rigid predominant parkinsonian I Litvan nosed as PSP.8-'4 Thus PSP can be difficult to C A Mangone disorder characterised by symmetric diagnose because of its increasingly recognised L D'Olhaberriague bradykinesia and axial rigidity unrelieved clinical diversity and because the topographic A Parsa by levodopa, pseudobulbar palsy, and distribution of lesions in the basal ganglia and Department of frontal release signs. Occasionally, seg- Neuropathology, brainstem overlaps with what is found in other Massachusetts General mental dystonia or myoclonus were parkinsonian syndromes.'5 16 Moreover, PSP Hospital, Boston, described, but neither aphasia nor alien may be associated with more than one neu- Massachusetts, USA limb syndrome was reported. Fractures ropathological diagnosis, such as Alzheimer's http://jnnp.bmj.com/ A McKee occurred in 25% of the patients but were or Parkinson's disease.'7 Raymond Escourolle unrelated to the severity of the gait or to Neuropathology To develop a clinically useful description of Laboratory, Inserm U the presence of falls. Median survival the natural history of PSP which could 360, Hopital de la time was 5*6 (range 2-16-6) years. Onset improve the accuracy of its early diagnosis,'6 Salpetriere, Paris, offalls during the first year, early dyspha- France we reviewed the records of the first and last M Veiny gia, and incontinence predicted a shorter visits of 24 patients with neuropathologically

time. at on September 29, 2021 by guest. Protected copyright. Ludwig Boltzmann survival Age onset, sex, early typical PSP confirmed at necropsy,'8 and com- Institute of Clinical onset of dementia, vertical supranuclear pared our findings with those of other con- Neurobiology, Vienna, palsy, or axial rigidity had no effect on firmed series.'5 19-22 We also evaluated the Austria prognosis of survival. Pneumonia was the cause of death and clinical predictors of sur- K Jellinger most common immediate cause of death. Department of vival in these patients because both may pro- Neurology, Institute of PSP was most often clinically misdiag- vide useful insights into the management of Psychiatry, London, nosed as Parkinson's disease. Errors in patients with PSP. UK diagnosis suggest that PSP is underdiag- K Ray Chaudhuri nosed. Parkinson's Disease Conclusion-Progressive onset of early Society Brain Tissue Methods Bank, Institute of postural instability with falls or supranu- Patients were selected from the research and Neurology, London, clear vertical palsy in the fifth decade, clinical neuropathological files of seven.med- UK should suggest the diagnosis of PSP. ical centres of four countries (Austria, R K B Pearce Onset of falls during the first year are Correspondence to: England, France, and the United States) by Dr Irene Iitvan, Federal emphasised, as they could lead to an early neuropathologists who based their selection on Building, Room 714, diagnosis and influence the prognosis of National Institute of the recently published National Institute of Neurological Disorders and patients with PSP. Whether appropriate Neurological Disorders and Stroke (NINDS) Stroke, National Institutes of treatment of the dysphagia could prolong Health, Bethesda, Maryland neuropathological criteria for the diagnosis of 20892, USA. the survival of PSP patients needs to be PSP.18 Only neuropathologically typical cases Received 23 November 1995 explored. of PSP, which were shown to have substantial and in revised form 5 February 1996 reliability,23 were included in the study. Accepted 8 February 1996 (7 Neurol Neurosurg Psychiatry 1996;61:615-620) Briefly, neuropathologically typical PSP 616 Litvan, Mangone, McKee, Vemny, Parsa, _Jellinger, et al J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.6.615 on 1 June 1996. Downloaded from includes a high density of neurofibrillary tan- women and 15 men, a typical sex distribution gles and neuropil threads in at least three of for patients with PSP. The mean age at onset of the following areas: pallidum, subthalamic PSP was 63 (SD 7) years; mean age at the first nucleus, substantia nigra, and pons, and a low visit was 66-4 (SD 7) years; and mean age at to high density of neurofibrillary tangles or death was 69-6 (SD 6 6) years. None of the neuropil threads in at least three of the follow- patients had a familial parkinsonian disorder. ing areas: striatum, oculomotor complex, Of the 24 patients, 21 were right handed, one medulla, and dentate nucleus. It also requires was left-handed, one was ambidextrous, and the exclusion of other disorders including large one had no record of handedness. or numerous infarcts; profound diffuse or focal atrophy; Lewy bodies; changes diagnos- INITIAL SYMPTOMS OF PSP tic of Alzheimer's disease; oligodendroglial The onset of PSP symptoms was insidious and argyrophilic inclusions; Pick bodies; diffuse disease progression was steady. The initial spongiosis; or prion P positive amyloid symptom was most often (63%, n = 15) pos- plaques. All cases met the criteria for inclusion tural instability. Both postural instability and in the study-that is, neuropathological diag- falls occurred during the first year in 58% of nosis of PSP made with at least a 75% degree the patients. Dysarthria was the second most of certainty by the neuropathologists who pro- common symptom (33%, n = 8), followed by vided the cases, and detailed neurological bradykinesia (13%, n = 3). Visual distur- examinations, including neurooculomotor bances (diplopia, blurred vision, burning eyes, examination, on the first and last visits. light sensitivity) were the first symptoms in The patients' records were abstracted on 13% (n = 3) of the patients. Cognitive or standardised forms by eight of us (AM, MV, behavioural changes generally followed these KJ, KRC, RKBP, LD, IL, or CAM), who fol- initial symptoms, but were the first symptoms lowed strict instructions. Signs or symptoms in 8% (n = 2) of the patients. were recorded as missing data if they were not mentioned in the records. However, as the SYMPTOMS AT FIRST VISIT data were retrospectively collected, we The first clinical visit occurred a mean of 3-7 assumed that neurologists performed com- years (range 1-11 years) after onset of disease. plete examinations and considered that a fea- At the first visit, most of the patients with PSP ture (for example, supranuclear palsy) was had gait disorder and postural instability, a absent when they reported that the examina- history of falls, bilateral bradykinesia, a pre- tion was "within normal limits" (for example, dominant akinetic-rigid course, axial rigidity, cranial nerves). For the purpose of this study, vertical supranuclear palsy, and dysarthria upward gaze limitation was considered abnor- (table 2). Gait was unstable (n = 23), small mal when either a restriction, in pursuit, or stepped (n = 12), or broad based (n = 5). voluntary gaze, or both, of at least 50% of the Falls were backwards in eight of 11 patients normal range was described, or the upward for whom the direction of the falls was supranuclear palsy was rated as moderate to described. severe. The severity of gait disturbance was Supranuclear gaze deficits involved initially classified as follows: 0 = not affected; 1 = min- either downward or upward gaze and, later, imal (impaired but no assistance is needed); 2 horizontal gaze. The patients rarely presented http://jnnp.bmj.com/ = mild (needs the use of a stick or walker); 3 = with vertical supranuclear palsy at onset; it moderate (needs the assistance of one or more usually occurred three years after onset of dis- persons); and 4 = severe (unable to walk even ease, and in three patients it never developed. with assistance). Moderate to severe upward gaze palsy was Survival was calculated by the Kaplan- more frequent than downward gaze abnormal- Meier and proportional hazards (Cox) life ities (n = 19 v n = 16). Pursuit and voluntary

table analysis.24 Analysis of variance saccades were affected early in the course of on September 29, 2021 by guest. Protected copyright. (ANOVA), the x2 test, and the Spearman r the disease, and often patients had abnormali- test were used for statistical analysis, as ties in convergence (although such data were appropriate.24 Results are expressed as mean often missing). (SD); statistical significance was defined as Speech was slurred (n = 15), dysphonic P < 0-05. (n = 12), slow (n = 9), palilalic (n = 2), ataxic (n = 1), or unintelligible (n = 1). Frontal lobe type symptomatology (mainly decreased flu- Results ency, concrete thought, and difficulty with DEMOGRAPHIC analogies, less often perseveration and imita- Table 1 shows the main demographic charac- tion behaviour), and personality changes teristics of the 24 patients. There were nine (mostly apathy and depression) was present in

Table 1 Characteristics ofpatients with PSP in three series confirmed by necropsy Present series Collins et al"' De Bruin et alP Sex 9F/15M 3F/9M 34F/51M* Age at onset (y) (mean (range)) 63 (45-73) 66 (48-77) 62 (47-78) Disease duration (y) (mean (range)) 6e6 (2-16o6) 563 (2-10) 5 (1-13) Time to first visit (mean (SD)) 3s7 (2) 1-5 (039) NA Time between visits (y) (mean (SD)) 2(2 (1) NA NA *Sex was not specified in five cases. No patients had history of encephalitis or familial disease; NA = data not available. Steele-Richardson-Olszewski syndrome 617 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.6.615 on 1 June 1996. Downloaded from Table 2 Features ofpatients with PSP in three series confirmed by necropsy Present series Present series Collins et aP' De Bruin et aP' first visit last visit last visit last visit Features (n = 24) (%) (n = 24) (%) (n = 12) (%) (n = 90) (%) Disease course: Akinetic-rigid 15 (63) 19 (79) NA NA Mixed disease 4 (17) 4 (17) NA NA Tremor dominant 1 (4) 1 (4) NA NA Extrapyramidal features: Gait disorders 23 (96) 24 (100) NA 63 (71) Postural instability 22a (92) 24 (100) NA NA Falls 20 (83) 24 (100) 12 (100) 54 (60) Falls in first year 14 (58) - 9 (75) NA Bilateral bradykinesia 21' (88) 23a (96) 11' (100) 60 (67) Axial rigidity 15b (63) 20' (83) NA 52 (58) Tremor, rigidity, or bradykinesia (at least two) 18 (75) 24 (100) 1 la (100) NA Tremor, rigidity, and bradykinesia 3 (13) 4' (17) 2' (18) NA Neck dystonia 5' (21) 11 (46) 4 (33) 43 (48) Rest tremor 5 (21) 3 (13) 2a (18) 15 (17) Unilateral dystonia 1 (4) 1 (4) 2' (18) 18 (20) Treatment: Levodopa None or unknown 12 (52) 9 (35) 3 (25) NA Pronounced response 2 (8) 4 (27) 2' (18) NA No response 9 (39) 11 (73) 7 (64) NA Other features: Supranuclear vertical palsy 19 (79) 20' (19 downward) 11 (10 downward) 61 (69) Dysarthria 18b (75) 24 (100) 118 (100) 60 (67) Frontal release signs 13 (54) 19' (79) NA NA Frontal lobe type symptomatology 1ld (46) 4e (58) NA NA Extensor plantar 5 (21) 9 (38) 8' (66) 30 (34) Dysphagia 4 (16) 20 (83) 6' (55) 51 (58) Anomia 1 (4) 4 (17) NA NA aData missing in one patient; bdata missing in two patients; cdata missing in three patients; ddata missing in four patients; 'data missing in seven patients; NA = data not available.

11 of 20 patients. Dysphagia and neck dysto- attention earlier (33-8 (7) months) than those nia were infrequent. who did not (59 (9) months; ANOVA, P < Details of treatment and initial response 0 04). were known for 15 patients (table 2). Seventy three per cent of them did not benefit from SYMPTOMS AT LAST VISIT levodopa treatment, and 27% (n = 2) had a The last clinical visit occurred a mean of 5 9 good response to levodopa (50-70% benefit), (range, 1 6-16) years after onset of disease. In but the benefit lasted less than one year in one general, symptoms and signs that were appar- patient, and was unknown in another. ent early in the course of PSP progressed Fractures occurred in 25% of the patients, steadily up to the last visit. The gait was mostly early in the course of the disease abnormal with postural instability and (mean, 2-9 years after onset; range 0-5-6 bradykinesia in all of our patients (table 2);

years). Gait severity or falls during the first 67% (n = 16) of the patients used a wheel- http://jnnp.bmj.com/ year of the disease were not correlated with the chair. Axial rigidity was more common (83%, n number of fractures. = 20) than retrocollis (46%, n = 11). Speech Age at onset of symptoms was significantly was always impaired: it was slurred (n = 22), correlated with the onset of falls during the dysphonic (n = 19), slow (n = 15), unintelligi- first year of disease (Spearman r = 0-82; P < ble (n = 10), palilalic (n = 3), mute (n = 3), 0-00- 1). Patients who had recurrent falls dur- echolalic (n = 2), ataxic (n = 2), or tachy- ing the first year of disease were significantly phemic (n = 1). Few patients had what could older (67 (1 1) years) than those who did not be considered unusual motor features: one on September 29, 2021 by guest. Protected copyright. (56-4 (1-3) years; ANOVA, P < 0-0001). patient had unilateral dystonia, one Patients with PSP who had begun falling dur- myoclonus, three asymmetric parkinsonian ing the first year of disease sought medical signs, one motor neuron disease in the lower limbs, and one both tremor dominant disorder and resting tremor at onset of disease. A few Kaplan-Meier life table 1-0 patients had anomia, but aphasia or alien limb analysis ofpatients with PSP after onset of 0.9 syndrome was never reported. symptoms. 0.8 SURVIVAL (, 0-7 The median survival from onset of PSP was in 0.6 5-6 (95% confidence interval 4'8-7 1 years; a 05 figure), and after the first clinical visit it was gO.5 2'7 years. The onset of falls during the first 0 o. 0.4 - year together with dysphagia and incontinence 20 CL 0.3 at the first visit predicted a shorter survival (X2 = 13-5, P < 0-003). The median survival of 0.2 patients who had begun to fall during the first 0-1 year of disease was shorter (5.2 years) than < 00 1I 1I 1I those who did not (6-8 years; P 0 05). 1 2 3 4 5 6 7 8 9 10 12 14 16 Survival of two patients with PSP with early Years from onset incontinence (3-2 and 3-3 years), and of one 618 Litvan, Mangone, McKee, Vemny, Parsa, _Jellinger, et al J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.6.615 on 1 June 1996. Downloaded from with early dystonia (two years) was brief. The 45, with rapidly progressive postural instability only patient with tremor at onset and tremor and falls, followed by dysarthria and supranu- dominant disease had the longest survival clear vertical palsy, affecting either upward or (1 6-6 years). Although patients with early dys- downward gaze. Similar symptoms often phagia had a shorter survival (3 9 years) than occurred in other series confirmed by those without (5-8 years), early dysphagia did necropsy. 19-22 Although falls often occur in not independently predict survival. Age at elderly people secondary to various causes,25 26 onset, sex, early onset of dementia, vertical the diagnosis of PSP should be considered supranuclear palsy, or axial rigidity had no when falls are associated with postural insta- effect on prognosis of survival. bility. Vertical supranuclear palsy, considered to be the hallmark of the disease, was present CLINICAL DIAGNOSIS AND CAUSE OF DEATH at disease onset in only a few patients. Much The neurologists who clinically followed up more commonly, vertical supranuclear palsy these patients made a diagnosis of PSP in 58% occurred several years after disease onset, or of the patients at the first visit (11 typical, rarely, never developed, as previously three atypical); Parkinson's disease in 21 % (n = reported.3 15 22 27 28 Downward gaze palsy is con- 5); corticobasal degeneration in 4% (n = 1); sidered to be more specific for the diagnosis of multiple system atrophy in 4% (n = 1); PSP than is upward gaze palsy primarily Alzheimer's disease in 4% (n = 1); and other because of age related restrictions in upward disease in 8% (n = 2). At the last visit, a diag- gaze.2930 In our patients, however, moderate to nosis of PSP was made in 88% of the patients severe upward gaze palsy was more frequent (14 typical, seven atypical). False negative than downward gaze abnormalities. Our find- misdiagnosis occurred with Parkinson's dis- ings suggest that downward gaze palsy might ease (n = 1), corticobasal degeneration (n = not be a requirement for the diagnosis of early 1), and other disorders (n = 1). PSP, and that upward gaze palsy in association Thirteen of the 20 patients in whom the with early gait instability and postural falls cause of death was known died of pneumonia may be a better predictor of the disease. (two with aspiration pneumonia), four of car- Our patients with PSP also showed (as in diovascular disorders (pulmonary emboli, other series; table 2) impaired behaviour con- myocardial infarct, congestive heart failure), sistent with frontal lobe dysfunction. In addi- and three of renal infections. tion, they developed non-levodopa responsive parkinsonism that was typically manifested as LABORATORY TESTS a predominantly axial, symmetric akinetic- Eighteen of the 24 patients had CT, MRI, or rigid syndrome, and occasionally with exten- both. Eight patients had diffuse atrophy; one sor neck dystonia. Despite its emphasis in the had atrophy of the tectum of the brainstem at a medical literature,29 neck dystonia was an later stage; one had ventricular dilation com- infrequent early feature in our patients. Rest patible with normal pressure hydrocephalus; tremor and segmental dystonia were rarely one had right temporal opercular infarct with- seen, as in other series.2' A study22 which eval- out a clinical history of stroke, which was not uated 16 pathologically confirmed PSP cases

confirmed at necropsy; one had multiple foci within three years of onset found that the http://jnnp.bmj.com/ of prolonged T2 in the white matter and puta- symptoms progressed rapidly in 94%; in 81% men; and six were reported to be normal. onset was symmetric; 50% had vertical Six of the 24 patients had a single-photon supranuclear palsy; 56% had axial dystonia; emission computed tomography (SPECT). and 50% cognitive impairment. Likewise, The iodoamphetamine SPECT showed varied symptoms in our series progressed rapidly and results: one patient had bilateral frontal and steadily for an average of six years. basal ganglia hypoperfusion; one had bilateral Although clinical series based on cases of orbitofrontal and left thalamic hypoperfusion; PSP confirmed by necropsy may have more on September 29, 2021 by guest. Protected copyright. one had left basal ganglia, and left temporo- atypical cases, the demographics from our 24 parietal hypoperfusion; one had left parieto- patients with PSP confirmed by necropsy are occipital and left frontal hypoperfusion; and similar to those of other clinicopathological one had right frontal, right basal ganglia and and clinical series.'5 19 22 31 The heterogeneity of left cerebellum hypoperfusion; one scan was patients with PSP is also increasingly recog- read as within normal limits. One patient had a nised.'5 Patients with PSP without ophthalmo- PET with '8fluorodeoxyglucose, which showed plegia may have fewer falls and less axial bilateral frontal hypometabolism. Only one dystonia, dysarthria, dysphagia, and rigidity.'5 patient had a barium swallowing study, which In our series, patients with PSP without early showed difficulty initiating the swallow, lin- downward gaze palsy also had less axial rigidity. gual propulsion difficulties, delayed pharyn- Interestingly, survival was longest in one of geal transit, and slow oral transit. our three patients with PSP who never devel- Fourteen patients with PSP had an EEG; oped ophthalmoplegia and had other features 11 had diffuse, frontal or bitemporal slowing suggestive of Parkinson's disease, such as and three were normal. asymmetric resting tremor at onset, tremor dominant disease, and late onset of dysarthria, dysphagia, and postural instability. Discussion Differences in the occurrence of clinical fea- NATURAL HISTORY OF PSP tures among the published series 9 20 may be The patients with PSP usually presented in the partially explained by incomplete records. seventh decade, and never before the age of Even though we chose only records that were Steele-Richardson-Olszewski syndrome 619

complete and participating centres are dedi- longer survival or whether early dystonia may J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.6.615 on 1 June 1996. Downloaded from cated to the study of parkinsonian disorders, shorten survival time. this is a retrospective study. Thus there may Patients are not often questioned about have been inconsistent recording of symptoms symptoms suggestive of dysphagia (coughing and signs. Moreover, there are always prob- or choking during meals) or silent aspiration lems of reliability of recollection of data, par- (fever of an unknown origin),4' 42 and barium ticularly as the first visit occurred a mean of swallowing studies are rarely requested, even 3-7 years after the onset of symptoms. Patients though patients with PSP are often aware of may have not recalled the events and physi- their difficulty in swallowing.43 In our series, a cians may not have recorded patients' com- barium swallow test was requested in only one plaints or may not have probed them patient, although the chance of detecting aspi- sufficiently. However, as these are tertiary cen- ration on clinical examinations is poor.44 tres, it is possible that patients would have Indeed, this is similar to what we find in our come to their first visit with records from other own clinical practices. Evaluation of patients primary centres. Only prospective population with PSP by speech therapists specialising in studies in which patient information is swallowing disturbances may prevent compli- recorded in a standardised manner can pro- cations such as silent aspiration pneumonia- vide us with the information for a definite nat- probably one of the most frequent causes of ural history of PSP. death in this disorder. Patients may benefit from changes in the consistency of the diet or MISDIAGNOSIS from the low morbidity and mortality of cur- Errors in the diagnosis of our cases of PSP by rent techniques such as percutaneous endo- the primary neurologists suggest that PSP is scopic gastrostomy. underdiagnosed-especially early in the course The single highest cause of death in our of the disease-as previously suggested."5 16 32 33 patients with PSP was pneumonia, whereas it The diagnosis of PSP is often confounded is ischaemic heart disease in normal controls because several of its classic symptoms also and patients with Parkinson's disease.45 In our occur in related disorders. For instance, study, the cause of death cannot be explained supranuclear vertical gaze palsy occurs in by a certification artifact,46 because the events other parkinsonian disorders, such as corti- were recorded from findings at necropsy. cobasal degeneration, diffuse Lewy body dis- Case-control studies with larger samples are ease, Creutzfeldt-Jakob disease, subcortical needed to determine whether early manage- gliosis, or less commonly, multiple system ment of dysphagia or other common compli- atrophy.'012 16223437 In addition, supranuclear cations prolong the survival of patients with vertical palsy occurs in other treatable condi- PSP. tions, such as Whipple's disease38 or arte- riosclerotic pseudoparkinsonism. In the early LABORATORY TESTS stages of the disease, PSP was most often con- The diagnosis of PSP was often not supported fused with Parkinson's disease by the primary by MRI, principally because brainstem atro- neurologists, whereas experienced neurologists phy was not routinely evaluated,47A9 although (specialists in movement disorders) were most the MRIs did exclude other diagnoses. The likely to confuse PSP with corticobasal degen- SPECT scans were not useful because they http://jnnp.bmj.com/ eration. 16 did not show the typical pattern of bilateral Features such as supranuclear vertical gaze frontal hypometabolism disclosed by '8fluo- palsy, rapid disease progression, early postural rodeoxyglucose PET.5053 Although iodoam- instability with falls, symmetric onset of aki- phetamine SPECT54 and imaging of dopamine netic-rigid parkinsonism that is unrelieved by D2 receptors with 123I-IBZM SPECT55 may be levodopa, or pyramidal signs, help differenti- preferable, studies are needed to determine

ate PSP from Parkinson's disease.39 Alien limb the role and cost effectiveness of SPECT in on September 29, 2021 by guest. Protected copyright. syndrome, severe asymmetric parkinsonism, diagnosing PSP. An EEG was only helpful in cortical sensory deficits, severe ideomotor excluding other rare disorders such as apraxia, myoclonus, or late onset of postural Creutzfeldt-Jakob disease. instability may help to differentiate corti- In summary, better knowledge of the pre- cobasal degeneration from PSP.36 39 40 The senting symptoms, natural history of the dis- confounding symptomatology points to the ease, and features that predict survival will be necessity of finding biological markers that will helpful for making an earlier and more accu- help make an earlier and definitive diagnosis of rate diagnosis and prognosis of patients with PSP. PSP.

We thank Drs Jean-Jacques Hauw, Susan Daniel, Dennis CLINICAL PREDICTORS OF SURVIVAL AND Dickson, Dikran S Horoupian, Peter L Lantos, Kurt Jellinger, MANAGEMENT ISSUES and Ann McKee for providing the cases for the NINDS data- Onset of falls during the first year and early base. We also thank Devera G Schoenberg for skilful editing. dysphagia and incontinence were useful pre- dictors of shorter survival of patients with PSP. It is unclear whether these features indi- 1 Steele JC, Richardson JC, Olszewski J. Progressive cate an unusually aggressive course of the dis- supranuclear palsy. Arch Neurol 1964;10:333-59. 2 Duvoisin RC. Clinical diagnosis. In: Litvan I, Agid Y, eds. ease or alternatively, promote secondary, life Progressive supranuclear palsy: clinical and research threatening complications. Other studies are approaches. New York: Oxford University Press, 1992: 15-33. also needed to determine whether tremor 3 Nuwer M. Progressive supranuclear palsy despite normal dominant disease could be a predictor of eye movements. Arch Neurol 1981;38:784. 620 Litvan, Mangone, McKee, Vemny, Parsa, _Jellinger, et al

4 Dubas F, Gray F, Escourolle R. Maladie de Steele- 1992:44-88. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.6.615 on 1 June 1996. Downloaded from Richardson-Olszewski sans ophtalmoplegie. Rev Neurol 31 Maher ER, Lees AJ. The clinical features and natural his- 1983;139:407-16. tory of the Steele-Richardson-Olszewski syndrome (pro- 5 Masliah E, Hansen L, Quijada S, et al. Late onset dementia gressive supranuclear palsy). Neurology 1986;36: 1005-8. with argyrophilic grains and subcortical tangles or atypical 32 Golbe LI, Davis PH, Schoenberg BS, Duvoisin RC. progressive supranuclear palsy? Ann Neurol 1991;29: Prevalence and natural history of progressive supranu- 389-96. clear palsy. Neurology 1988;38: 1031-4. 6 Matsuo H, Takashima H, Kishikawa M, et al. Pure akinesia: 33 Golbe LI. Epidemiology. In: Litvan I, Agid Y, eds. an atypical manifestation of progressive supranuclear Progressive supranuclear palsy: clinical and research palsy. J Neurol Neurosurg Psychiatry 1991 ;54:397-400. approaches. New York: Oxford University Press, 1992: 7 Mizusawa H. Pure akinesia and progressive supranuclear 33-43. palsy. No To Shinkei 1993;45:113-8. 34 Feamley JM, Revesz T, Brooks DJ, Frackowiak RS, Lees 8 Dubinsky RM, Jankovic J. Progressive supranuclear palsy AJ. Diffuse Lewy body disease presenting with a and a multi-infarct state. Neurology 1987;37:570-6. supranuclear gaze palsy. J7 Neurol Neurosurg Psychiatry 9 Will RG, Lees AJ, Gibb W, Barnard RO. A case of progres- 199 1;54: 159-6 1. sive subcortical gliosis presenting clinically as Steele- 35 De Bruin VM, Lees AJ, Daniel SE. Diffuse Lewy body disease Richardson-Olszewski syndrome. Jf Neurol Neurosurg presenting with supranuclear gaze palsy, parkinsonism, Psychiatry 1988;51:1224-7. and dementia: a case report. Mov Disord 1992;7:355-8. 10 Gibb WR, Luthert PJ, Marsden CD. Corticobasal degener- 36 Rinne JO, Lee MS, Thompson PD, Marsden CD. ation. Brain 1989;112:1171-92. Corticobasal degeneration. A clinical study of 36 cases. 11 Foster NL, Gilman S, Berent S, et al. Progressive subcorti- Brain 1994;117:1183-96. cal gliosis and progressive supranuclear palsy can have 37 Wenning G, Ben Shlomo Y, Magalhaes M, Daniel S, similar clinical and PET abnormalities. J Neurol Quinn N. Clinical features and natural history of multi- Neurosurg Psychiatry 1992;55:707-13. ple system atrophy. An analysis of 100 cases. Brain 1994; 12 Jankovic J, Rajput AH, Golbe LI, Goodman JC. What is it? 117:835-45. Case 1, 1993: parkinsonism, dysautonomia, and ophthal- 38 Amarenco P, Roullet E, Hannoun L, Marteau R. Pro- moparesis. Mov Disord 1993;8:525-32. gressive supranuclear palsy as the sole manifestation of 13 Winikates J, Jankovic J. Vascular progressive supranuclear systemic Whipple's disease treated with pefloxacine [let- palsy. Neural Transm Suppl 1994;42:189-201. ter]. J7 Neurol Neurosurg Psychiatry 1991 ;54: 1121-2. 14 Revesz T, Daniel SE, Lees AJ, Will RG. A case of progres- 39 Quinn N. Parkinsonism-recognition and differential diag- sive subcortical gliosis associated with deposition of nosis. BMJ 1995;310:447-52. abnormal prion protein. J Neurol Neurosurg Psychiatry 40 Lang AE, Bergeron C, Pollanen MS, Ashby P. Parietal 1996;58:759-60. Pick's disease mimicking cortical-basal ganglionic degen- 15 Daniel SE, De Bruin VMS, Lees AJ. The clinical and eration. Neurology 1994;44: 1436-40. pathological spectrum of Steele-Richardson-Olszewski 41 Feinberg MJ, Knebl J, Tully J, Segall L. Aspiration and the syndrome (progressive supranuclear palsy): a reappraisal. elderly. Dysphagia 1990;5:61-71. Brain 1995;118:759-70. 42 Longemann JA. Evaluation and treatment ofswallowing disor- 16 Litvan I, Agid Y, Jankovic J, et al. Accuracy of clinical crite- ders. San Diego: College-Hill Press, 1983. ria for the diagnosis of progressive supranuclear palsy 43 Sonies BC. Swallowing and speech disturbances. In: Litvan (Steele-Richardson-Olszewski syndrome). Neurology I, Agid Y, eds. Progressive supranuclear palsy: clinical and 1996 (in press). research approaches. New York: Oxford University Press, 17 Gearing M, Olson DA, Watts RL, Mirra SS. Progressive 1992:240-53. supranuclear palsy: neuropathologic and clinical hetero- 44 Splaingard ML, Hutchins B, Sulton LD, Chaudhuri G. geneity. Neurology 1994;44:1015-24. Aspiration in rehabilitation patients: videofluoroscopy vs 18 Hauw JJ, Daniel SE, Dickson D, et al. Preliminary NINDS bedside clinical assessment. Arch Phys Med Rehab 1988; neuropathologic criteria for Steele-Richardson-Olszewski 69:637-40. syndrome (progressive supranuclear palsy). Neurology 45 Ben-Shlomo Y, Marmot MG. Survival and cause of death in 1994;44:2015-9. a cohort of patients with parkinsonism: possible clues to 19 Brusa A, Mancardi GL, Bugiani 0. Progressive supranu- aetiology? Jf Neurol Neurosurg Psychiatry 1995;58:293-9. clear palsy 1979: an overview. Ital J Neurol Sci 1980;4: 46 Kircher T, Nelson J, Burdo H. The autopsy as a measure of 205-22. accuracy of the death certificate. N Engl _J Med 1985; 20 De Bruin VM, Lees AJ. Subcortical neurofibrillary degen- 313:1263. eration presenting as Steele-Richardson-Olszewski and 47 Schonfeld SM, Golbe LI, Sage JI, Safer JN, Duvoisin RC. other related syndromes: a review of 90 pathologically Computed tomographic findings in progressive supranu- verified cases. Mov Disord 1994;9:381-9. clear palsy: correlation with clinical grade. Mov Disord 21 Collins SJ, Ahlskog JE, Parisi JE, Maraganore DM. 1987;2:263-78. Progressive supranuclear palsy: neuropathologically 48 Savoiardo M, Strada L, Girotti F, et al. MR imaging in pro- based diagnostic criteria. JT Neurol Neurosurg Psychiatry gressive supranuclear palsy and Shy-Drager syndrome. J 1995;58: 167-73. ComputAssist Tomogr 1989;13:555-60. 22 Colosimo C, Albanese A, Hughes AJ, de Bruin VM, Lees 49 Savoiardo M, Girotti F, Strada L, Ciceri E. Magnetic reso- AJ. Some specific clinical features differentiate multiple nance imaging in progressive supranuclear palsy and system atrophy (striatonigral variety) from Parkinson's other parkinsonian disorders. J Neural Transm Suppl http://jnnp.bmj.com/ disease. Arch Neurol 1995;52:294-8. 1994;42:93-1 10. 23 Litvan I, Hauw JJ, Bartko JJ, et al. Validity and reliability of 50 Bum DJ, Sawle GV, Brooks DJ. Differential diagnosis of the preliminary NINDS neuropathological criteria for Parkinson's disease, multiple system atrophy, and Steele- progressive supranuclear palsy and related disorders. J Richardson-Olszewski syndrome: discriminant analysis of Neuropathol Exp Neurol 1996;55:97-105. striatal "8F-dopa PET data. J7 Neurol Neurosurg Psychiatry 24 SAS Institute. JMP statistics for the Apple Macintosh. Cary, 1994;57:278-84. NC: SAS Institute Inc, 1994. 51 Brooks DJ. PET studies in progressive supranuclear palsy. _J 25 Tinetti ME, Speechley M. Prevention of falls among the Neural Transm Suppl 1994;42:119-34. elderly. N EnglJ Med 1989;320: 1055-9. 52 Blin J, Horwitz B, Baron J, Agid Y. Does frontal cortex 26 Tinetti ME, Inouye SK, Gill TM, Doucette JT. Shared risk hypometabolism in progressive supranuclear palsy result

factors for falls, incontinence, and functional depen- from subcortical dysfunction? Eur J Neurol 1995;1: on September 29, 2021 by guest. Protected copyright. dence. Unifying the approach to geriatric syndromes. 221-8. J7AMA 1995;273:1348-53. 53 Foster NL, Gilman S, Berent S, et al. Progressive subcorti- 27 Pfaffenbach DD, Layton DDJ, Kearns TP. Ocular mani- cal gliosis and progressive supranuclear palsy can have festations in progressive supranuclear palsy. Am J similar clinical and PET abnormalities. _J Neurol Ophthalmol 1972;74: 1179-84. Neurosurg Psychiatry 1992;55:707-13. 28 Davis PH, Bergeron C, McLachlan DR. Atypical presenta- 54 Ghika J, Tennis M, Growdon J, Hoffman E, Johnson K. tion of progressive supranuclear palsy. Ann Neurol Environment-driven responses in progressive supranu- 1985;17: clear palsy. J Neurol Sci 1995;130:104-1 1. 337-43. 55 VanRoyen E, Verhoeff NF, Speelman JD, Woletrs EC, 29 Lees AJ. The Steele-Richardson-Olszewski syndrome (pro- Kuiper MA, Janssen AG. Multiple system atrophy and gressive supranuclear palsy). Mov Disord 1987;2:272-87. progressive supranuclear palsy. Diminished striatal D2 30 Troost B. Neuro-ophthalmological aspects. In: Litvan I, dopamine receptor activity demonstrated by '23I-IBZM Agid Y, eds. Progressive supranuclear palsy: clinical and single-photon emission computed tomography. Arch research approaches. New York: Oxford University Press, Neurol 1993;50:513-6.