ORIGINAL CONTRIBUTION Neuropathological Correlates of Dysarthria in Progressive Supranuclear Palsy

Karen J. Kluin, MS, CCC, BC-NCD; Sid Gilman, MD; Norman L. Foster, MD; Anders A. F. Sima, MD, PhD; Constance J. D’Amato, BS; Leslie A. Bruch, MD; Laurie Bluemlein, RN, MS; Roderick Little, PhD; Jewel Johanns, PhD

Background: The dysarthria of progressive supra- gree of neuronal loss and gliosis in predetermined ana- nuclear palsy consists of prominent hypokinetic and tomical sites. spastic components with less prominent ataxic compo- nents. Results: All patients had hypokinetic and spastic dysar- thria, and 9 also had ataxic components. The severity of Objective: To correlate the types of dysarthria with the hypokinetic components was significantly correlated neuropathological changes in patients with progressive with the degree of neuronal loss and gliosis in the substan- supranuclear palsy. tia nigra pars compacta (r=0.61, P=.02) and pars reticu- lata (r=0.64, P=.01) but not in the subthalamic nucleus Design and Methods: In 14 patients with progres- (r=0.51, P=.07) or the striatum or (/r/ sive supranuclear palsy, we correlated the perceptual Ͻ0.34, PϾ.20). The severity of the spastic and ataxic com- findings with the neuropathological findings. A ponents was not significantly correlated with the neuro- dysarthria assessment was performed a mean±SD of pathological changes in the frontal cortex (r=0.20, P=.50) 31±15 months (range, 10-53 months) before death. The and (/r/Ͻ0.28, PϾ.33), respectively. deviant speech dimensions were rated on a scale of 0 (nor- mal) to 3 (severe). The neuropathological examination Conclusion: The hypokinetic dysarthria of progressive su- consisted of semiquantitative analysis of neuronal loss pranuclear palsy may result from degenerative changes in and gliosis by investigators (A.A.F.S., and L.A.B.) blinded the pars compacta and pars reticulata and to the clinical findings. Correlation and linear regres- not from changes in the striatum or globus pallidus. sion analysis were used to correlate the severity of the hypokinetic, spastic, and ataxic components with the de- Arch Neurol. 2001;58:265-269

YSARTHRIA is a cardinal volvement of the corticobulbar pathways feature of progressive su- is associated with spastic dysarthria, dis- pranuclear palsy (PSP).1 orders of the cerebellum and its connec- Dysarthria consists of a tions with ataxic dysarthria,8,9 and dis- combination of hypoki- eases of the extrapyramidal pathways with Dnetic, spastic, and ataxic components, usu- hypokinetic dysarthria.8 ally with prominent hypokinetic and spas- We examined the speech disorders of tic and less prominent ataxic features.2 14 patients with PSP who later underwent Neuropathological changes in patients with autopsy. We correlated the severity of the PSP involve neuronal loss and gliosis with specific components of the dysarthria with neurofibrillary tangles, argyrophilic and the neuropathological changes in struc- From the Departments of ␶ ␶ Speech-Language Pathology -positive threadlike structures, and -posi- tures that have been associated with dysar- (Ms Kluin), tive astrocytic tanglelike inclusions in many thria in other neurological diseases. Pre- (Mss Kluin and Bluemlein and subcortical regions, including the substan- liminary findings have been reported.10 Drs Gilman and Foster), tia nigra (SN), striatum, globus pallidus, Pathology (Dr Sima and subthalamic nuclei, periaqueductal gray, RESULTS Ms D’Amato), and Biostatistics pontine nuclei, inferior olives, cerebellar (Drs Little and Johanns), dentate nuclei, and certain cranial nerve nu- NEUROLOGICAL EXAMINATION University of Michigan Health clei.1,3-7 The anatomical locations of the neu- System, Ann Arbor; the ropathological changes responsible for the The 14 patients consisted of 7 men and 7 Department of Pathology, Wayne State University, dysarthria in patients with PSP have not women (mean±SD age at death, 69.0±5.8 Detroit, Mich (Dr Sima); and been determined. In other neurological dis- years; range, 59-80 years). The clinical diag- the Department of Pathology orders, the major types of dysarthria have nosis at the last clinic visit was PSP in all pa- and Microbiology, University of been linked to specific anatomical connec- tients. One patient initially was diagnosed as Nebraska, Omaha (Dr Bruch). tions in the . Bilateral in- havingAlzheimerdiseasebecauseofprogres-

(REPRINTED) ARCH NEUROL / VOL 58, FEB 2001 WWW.ARCHNEUROL.COM 265

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 PATIENTS AND METHODS The neuropathological diagnosis of PSP was based on the findings specified in the National Institute of Neuro- logical Disorders and criteria,3 including neuronal Patients with PSP have been studied in research protocols at loss, gliosis, and neurofibrillary tangles; the latter were de- the University of Michigan, Ann Arbor, since 1984, and many tected with a modified Bielschowsky silver stain in sub- have been followed up with longitudinal clinical assess- cortical nuclei.3,4,13 Coexistent Alzheimer disease was de- ments using the protocols of the Michigan Alzheimer’s Dis- termined using the Reagan criteria.3,14 Neuronal loss and ease Research Center since 1989. The evaluations included gliosis were graded semiquantitatively in 46 brain sec- neurological examinations, speech pathology assessments, tions stained with cresyl violet–Luxol fast blue–eosin and structural and functional imaging studies, and subsequent neu- phosphotungstic acid–hematoxylin using the following ropathological examinations at autopsy. We identified all pa- scale: absent is 0; mild, 1; moderate, 2; and severe, 3. In- tients with PSP examined postmortem who had been evalu- termediate changes were given half values. The pathologi- ated by a speech pathologist (K.J.K.) between 1984 and 1994. cal changes were scored by consensus of 2 neuropatholo- Evaluation of dysarthria included assessment of oral mo- gists (A.A.F.S., and L.A.B.), unaware of the clinical findings, tor and oral agility skills and perceptual speech analysis. Oral who viewed the sections concurrently in a multiheaded mi- motor examination consisted of assessment of muscular croscope. We formulated hypotheses concerning the ana- strength; coordination; accuracy; range of excursion; and tomical sites in the nervous system where neuropathologi- symmetry of head and neck, face, mandibular, tongue, pala- cal changes might be correlated with the components of topharyngeal, and respiratory muscles at rest and during re- dysarthria. Since the distribution of the data deviated from flex and voluntary movements. Oral agility was assessed by normality, we used Spearman rank correlation coeffi- oral diadochokinetic rates and the oral agility skills subtest cients to analyze the relation between hypokinetic, spas- of the Boston Diagnostic Examination.11 Perceptual tic, and ataxic dysarthria types and neuropathological ab- speech analysis included identification and rating of the se- normalities in these predetermined anatomical sites. The verity of the deviant speech dimensions during the exami- scores for neuronal loss and gliosis in each anatomical area nation and from videotaped or audiotaped samples of spon- were added to produce a composite neuropathological score. taneous speech, description of the Cookie Theft picture from We created a composite neuropathological score to limit the Boston Diagnostic Aphasia Examination,11 and oral read- problems with multiple comparisons, given the modest ing of the “Grandfather Passage.”8 We used the definitions sample size and the number of analyses conducted. Based of deviant speech dimensions of Darley et al8 and the Uni- on the association of hypokinetic dysarthria with several versity of Michigan classification of hypokinetic, ataxic, and nuclei of the , we correlated the hypokinetic spastic dysarthrias2,10,12 (Table 1). Each deviant speech di- dysarthria rating with the composite neuropathological mension identified was assigned a severity score extending scores in the SN pars compacta (SNc), SN pars reticulata from 0 (normal) to 3 (severe). Weighting factors were ap- (SNr), subthalamic nucleus, , , lat- plied to emphasize the deviant speech dimensions most char- eral globus pallidus, medial globus pallidus, and periaq- acteristic of each type of dysarthria.2,10,12 A total score was ob- ueductal gray. Based on the association of spastic dysar- tainedreflectingthedegreeofhypokinesia,,andataxia thria with corticobulbar projections, we correlated the spastic in speech. The possible scores ranged from 0 to 48 for each dysarthria rating with the composite neuropathological dysarthria type, with higher scores indicating more severe im- scores in the frontal cortex. Based on the association of ataxic pairment. To ensure consistency, all speech pathology exami- dysarthria with the cerebellum and related structures, we nations were performed by a single speech pathologist (K.J.K.). correlated the ataxic dysarthria rating and the composite Methods of measuring the severity of speech disorders have neuropathological scores in the cerebellar cortex, dentate been described in previous publications.2,9,10,12 nuclei, inferior olives, and red nuclei.

sive dementia, but later developed rigidity and supranuclear tation of vertical gaze was greater than limitation of hori- gaze palsy. Several others were diagnosed as having Parkin- zontal gaze in all patients. At their last clinical examination, son disease (PD) elsewhere, but had clear signs of PSP by nearly all patients had completely lost voluntary vertical eye the time of our initial examination. Twelve patients had re- movements. Gait was a common complaint early in ceived levodopa during their illness. No benefit was noted the disease, but frequently became less apparent as the pa- in 11, and 1 had a “poor” response. Motor speech was as- tients became increasingly immobile. One patient had rest- sessed at the time of the initial examination. At this time, ingdistaltremor,and2othershadextensorplantarresponses. all 14 patients had cognitive impairments, limb rigidity, su- pranuclear gaze palsy, and frequent falls. Although we regu- SPEECH PATHOLOGY EVALUATION larlyseeourpatientsat6-monthintervals,somesubjectswere unable to return for examination as they became more im- In most patients, the speech pathology evaluation was paired or were institutionalized. Consequently, the last clini- performed only at the time of the patient’s initial exami- cal neurological examination ratings were performed on av- nation, a mean±SD of 31±15 months (range, 10-53 erageamean±SDof8±8monthsbeforedeath.Themean±SD months) before death. All patients had mixed dysar- duration of neurological symptoms before death was 7±2 thria with hypokinetic and spastic components, and 9 also years (range, 3-10 years). Most of the patients had moder- had ataxic components (Table 2). Hypokinetic dysar- ate to severe and gaze limitations at their last thria scores ranged from 4 to 39 (mean, 15), spastic dys- examination. Bradykinesia and axial rigidity were the most arthria scores from 6 to 34 (mean, 18), and ataxic dys- severemotorsigns,and7patientswereunabletostand.Limi- arthria scores from 0 to 20 (mean, 5). All patients had

(REPRINTED) ARCH NEUROL / VOL 58, FEB 2001 WWW.ARCHNEUROL.COM 266

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 masked faces. Thirteen patients had impaired lingual rapid nonverbal agility skills scores ranged from 0 to 11 of a alternating movements ranging in severity from mild to total of 12 (mean, 4.5). The verbal agility skills scores severe, with a mean rating of moderate. Ten patients had ranged from 0 to 14 of a total of 15 (mean, 9). a brisk jaw jerk, and 9 had a hyperactive gag reflex. The NEUROPATHOLOGICAL FINDINGS Table 1. Deviant Speech Dimensions for Hypokinetic, Spastic, and Ataxic Dysarthrias: Whole brain weight ranged from 951 to 1390 g (mean, 1197 University of Michigan Classification* g). Gross inspection demonstrated no abnormal cerebral atrophy. Pallor of the SN was seen in all patients. Exten- Hypokinetic Dysarthria Spastic Dysarthria Ataxic Dysarthria sive subcortical neuropathological changes were found in Low volume (able to Strained-strangled Excess and equal a distribution characteristic of PSP, with neurofibrillary 3,13 increase on sound (3) stress (3) tangles and extensive neuronal loss and gliosis. Neuro- command) (3) fibrillary tangles were found in all patients in the subtha- Monopitch (2) Reduced stress (2) Irregular articulatory lamic nucleus, SNc, third nuclear complex, periaqueduc- breakdown (2) tal gray, and pontine nuclei. In 13 patients, neurofibrillary Loudness decay (2) Harsh voice Alternating loudness (continuous) (2) variation (2) tangles were found in the lateral globus pallidus, locus ce- Short rushes of Slow rate (2) Fluctuating pitch ruleus, and inferior olivary nucleus. Tangles were also found speech (2) levels (2) in the putamen of 6 patients, in the claustrum of 5, and in Increased speaking rate Low pitch (2) Variable rate (2) the caudate nucleus of 3. Neuronal loss and gliosis accom- over time (2) panied these changes, and were most severe in the SNc, Imprecise phonemes Imprecise Harsh voice periaqueductal gray, subthalamic nucleus, and medial glo- over time (1) phonemes (1) (transient) (1) Decreased stress (1) Monoloudness (1) Breathy voice bus pallidus. Neuronal loss, gliosis, and neurofibrillary (transient) (1) tangles were apparent in the regions chosen for correla- Repetition of sounds, Hypernasality Altered nasality tional analysis (Table 3). Two patients had neuropatho- words, and (continuous) (1) (transient) (1) logical changes of Alzheimer disease and PSP, and the other phrases (1) 12 had only pure PSP. Inappropriate silences Monopitch (1) Voice (1) or difficulty initiating phonation (1) CORRELATION OF WITH Breathy voice Prolonged phonemes, Audible inspiration NEUROPATHOLOGICAL CHANGES (continuous) (1) intervals, or both (1) (1) The severity of the hypokinetic component of dysar- *Data from Kluin et al.12 The numbers in parentheses indicate weighing factors used in obtaining numerical scores for the dysarthria types. Higher thria was significantly correlated with the neuropatho- factors are used for the deviant speech dimensions considered most logical score in the SNc and SNr. A sizeable correlation characteristic of each type of dysarthria. was also found with the degree of neuronal loss and

Table 2. Speech Pathology Evaluation

Severity* Total No. Dysarthria Type None Mild Moderate Severe of Patients Hypokinetic 0 6 5 3 14 Spastic 0 3 7 4 14 Ataxic 5 6 3 0 14

*Data are given as the number of patients with each type of dysarthia. The severity of dysarthia was based on a composite score of the deviant speech dimensions.

Table 3. Neuropathological Findings*

Frequency of Severity of Neuronal Loss and Gliosis

Region 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Substantia nigra pars compacta 0000000010715 Substantia nigra pars reticulata 0000105020114 Lateral globus pallidus 0000112010603 Medial globus pallidus 0010012020404 Subthalamic nucleus 0000002040413 Periaqueductal gray 0000001020614 Caudate nucleus 1221404000000 Putamen 5330101100000

*Data are given as the number of patients with neuronal loss and gliosis in each region. The scores for neuronal loss and gliosis in each anatomical area were added together for a composite neuropathological score.

(REPRINTED) ARCH NEUROL / VOL 58, FEB 2001 WWW.ARCHNEUROL.COM 267

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 gliosis in the subthalamic nucleus, but this did not r=−0.43, PϾ.12; SNr: r=−0.15, PϾ.60). Furthermore, the reach statistical significance (Table 4 and Figure). correlations between hypokinetic dysarthria and neuro- Correlations with other predetermined anatomical sites pathological scores in the SN persisted when the sum of in the striatum and globus pallidus were smaller and spastic and ataxic dysarthria was controlled by linear statistically insignificant (/r/Ͻ0.34, PϾ.20). We regression. Specifically, the P values for these partial cor- adjusted for the duration from the speech pathology relations were .05 for the SNc and .03 for the SNr. evaluation to death by multiple regression with dysar- Spastic dysarthria components were not closely cor- thria as outcome, and the adjusted correlations were related with neuropathological changes in the frontal cor- similar. tex (r=0.20, P=.50), and ataxic dysarthria components did To assess whether the relation between neuropatho- not correlate significantly with the neuropathological logical changes in the SN and hypokinetic dysarthria might changes in the dentate nucleus, inferior olives, or red nu- be attributable to symptoms of dysarthria in general, we clei (/r/Ͻ0.28, PϾ.33). All patients except 1 had neuro- correlated the sum of 2 nonhypokinetic types of dysar- pathological scores of 0 in the cerebellar cortex, prevent- thria, spastic and ataxic, with the severity of the neuro- ing correlations of the severity of ataxic dysarthria with pathological changes. We found no significant correla- pathological features in this site. The dysarthria scores of tions in any of the regions (PϾ.12 for all); estimated the 2 patients with Alzheimer disease and PSP were within correlations with the SN were in fact slightly negative (SNc: the same range as those of the patients with pure PSP.

Table 4. Correlations of Hypokinetic Dysarthia COMMENT With Neuropathological Changes This study revealed significant correlations between the intensity of the neuropathological changes in the SNc and Hypokinetic Dysarthria SNr and the severity of the hypokinetic components of Region Correlation P dysarthria in patients with PSP. We found no signifi- Substantia nigra pars compacta 0.61 .02 cant correlations with neuropathological changes in the Substantia nigra pars reticulata 0.64 .01 striatum, globus pallidus, or periaqueductal gray. We se- Lateral globus pallidus −0.13 .66 lected the SN for correlation in relation to the hypoki- Medial globus pallidus 0.34 .24 netic components of dysarthria because of the similarity Subthalamic nucleus 0.51 .07 of the hypokinetic dysarthria in patients with PSP to that Periaqueductal gray −0.13 .66 Caudate nucleus −0.18 .53 in patients with PD and the known intense neuropatho- Putamen 0.10 .73 logical changes in this site in those with both PD and PSP. Darley et al8 proposed that hypokinetic dysarthria re-

40 1 40 A B 1

30 30

1 1 1 20 1 20 1 1 1 1 1 1 1 1 2 1 10 10 1 1 1 1 1 1 1 1 1 1

0 1462 3 5 0 142 3 5 6 Substantia Nigra Pars Compacta Substantia Nigra Pars Reticulata 40 40 C 1D 1 Neuropathological Score 30 30

1 1 1 20 1 20 1 1 1 1 2 1 1 1 1

10 1 10 1 1 1 1 1 1 1 11 1 1

0 142 3 5 60 1 2 3 4 5 6 Subthalamic Nucleus Striatum

Severity of hypokinetic dysarthria and neuropathological score in predetermined regions. A significant relation was found between the severity of hypokinetic dysarthria and the neuropathological score in the substantia nigra pars compacta (A) and the substantia nigra pars reticulata (B). There was a relation that did not reach statistical significance between the severity of hypokinetic dysarthria and the neuropathological score in the subthalamic nucleus (C) but no correlation for the striatum (caudate nucleus and putamen) (D). The number 1 signifies data from 1 patient; 2, data from 2 patients at overlapping sites.

(REPRINTED) ARCH NEUROL / VOL 58, FEB 2001 WWW.ARCHNEUROL.COM 268

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 flected extrapyramidal dysfunction based on examina- pathological changes doubtless advanced in the SN and tion of patients with PD. In patients with both PSP and the other structures affected from the time of diagnosis PD, there is severe neurodegeneration within the SNc and and dysarthria assessment to the time of death. Despite decreased inhibitory input to the SNr from the globus these concerns, the correlations of hypokinetic dysar- pallidus, but only in those with PSP is there substantial thria with neuropathological changes in the SNc and SNr loss of SNr neurons.15 In patients with PSP, the striatum were significant, perhaps reflecting the importance of these and the globus pallidus are also involved, but in those sites in the pathogenesis of hypokinetic dysarthria. with PD these regions are spared. The severity of neu- ropathological changes was considerably less in the stria- Accepted for publication August 1, 2000. tum than in the SN, which may be a reason for the non- This study was supported in part by grant P50AG08671 significant correlations for this structure. The medial from the National Institute on Aging, National Institutes of globus pallidus, subthalamic nucleus, and periaqueduc- Health, Bethesda, Md (Michigan Alzheimer’s Disease Re- tal gray had severe neuronal loss and gliosis, but the neu- search Center). ropathological changes in these structures were not sig- Corresponding author and reprints: Karen J. Kluin, MS, nificantly correlated with the severity of hypokinetic CCC, BC-NCD, Department of Speech-Language Pathol- dysarthria. Hence the findings in the present study sug- ogy, University of Michigan Health System, 1D203 Uni- gest that the most important structures in the pathophysi- versity Hospital, 1500 E Medical Center Dr, Ann Arbor, MI ology of hypokinetic dysarthria in patients with PSP may 48109-0043. be the SNc and the SNr. The SNr is the output station of the basal ganglia and projects to anterior/ventral lateral, REFERENCES mediodorsal, and midline thalamic nuclei, superior col- liculus, and pedunculopontine nucleus in the brain- 16 1. Steele JC. Progressive supranuclear palsy. Brain. 1972;95:693-704. stem. The role of the SNr in patients with dysarthria may 2. Kluin KJ, Foster NL, Berent S, Gilman S. Perceptual analysis of speech disor- be related to the projections into the . ders in progressive supranuclear palsy. Neurology. 1993;43:563-566. 3. Hauw JJ, Daniel SE, Dickson D, et al. Preliminary NINDS neuropathological cri- We found no significant correlations between the spas- teria for Steele-Richardson-Olszewski syndrome (progressive supranuclear palsy). tic and ataxic components of dysarthria and the severity Neurology. 1994;44:2015-2019. of neuropathological abnormalities in sites known to be 4. Daniel SE, de Bruin VM, Lees AJ. The clinical and pathological spectrum of Steele- Richardson-Olszewski syndrome (progressive supranuclear palsy): a reappraisal. affected in those with PSP and associated with these types Brain. 1995;118:759-770. of dysarthria. We also correlated with neuropathological 5. Lantos PL. The neuropathology of progressive supranuclear palsy. J Neural Transm scores the sum of the 2 nonhypokinetic types of dysar- Suppl. 1994;42:137-152. 6. Probst A, Langui D, Lautenschlager C, Ulrich J, Brion JP, Anderton BH. Progres- thria, spastic and ataxic. We found no correlations in any sive supranuclear palsy. Acta Neuropathol (Berl). 1988;77:61-68. of the regions studied, indicating that symptoms of dys- 7. Yamada T, McGeer PL, McGeer EG. Appearance of paired, nucleated, tau- positive glia in patients with progressive supranuclear palsy brain tissue. Neu- arthria in general do not account for the correlation be- rosci Lett. 1992;135:99-102. tween hypokinetic dysarthria and neuropathological 8. Darley FL, Aronson AE, Brown JR. Motor Speech Disorders. Philadelphia, Pa: changes in the SN. While these results should be inter- WB Saunders Co; 1975. 9. Kluin KJ, Gilman S, Markel DS, Koeppe RA, Rosenthal G, Junck L. Speech dis- preted with caution given the limited sample size, they are orders in olivopontocerebellar atrophy correlate with positron emission tomog- consistent with a link with the SN that is specific to the raphy findings. Ann Neurol. 1988;23:547-554. hypokinetic dysarthria components. 10. Kluin KJ, Gilman S, Foster NL, et al. Degeneration of substantia nigra pars compacta contributes to the dysarthria of PSP [abstract]. Neurology. 1997;48(suppl 2):A97. Dysarthria has been reported as the second most 11. Goodglass H, Kaplan E. The Assessment of Aphasia and Related Disorders. 2nd common clinical manifestation of PSP.17 The SN, globus ed. Philadelphia, Pa: Lea & Febiger; 1983. 12. Kluin KJ, Gilman S, Lohman M, Junck L. Characteristics of the dysarthria of mul- pallidus, and subthalamic nucleus are the areas with the tiple system atrophy. Arch Neurol. 1996;53:545-548. most severe neuropathological abnormalities in pa- 13. Litvan I, Hauw JJ, Bartko JJ, et al. Validity and reliability of the preliminary NINDS tients with PSP, raising the possibility that the patho- neuropathologic criteria for progressive supranuclear palsy and related disor- 18 ders. J Neuropathol Exp Neurol. 1996;55:97-105. logical process starts there. 14. The National Institute on Aging, and the Reagan Institute Working Group on The patients described in this study appear to repre- Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Dis- sent an appropriate sample of patients with PSP. All met ease. Consensus recommendations for the postmortem diagnosis of Alzhei- 1,3-5,13 mer’s disease. Neurobiol Aging. 1997;18(suppl):S1-S2. clinical and neuropathological criteria for PSP. All had 15. Hardman CD, Halliday GM, McRitchie DA, Cartwright HR, Morris JGL. Progressive mixed dysarthria with hypokinetic and spastic compo- supranuclear palsy affects both the substantia nigra pars compacta and reticulata. Exp Neurol. 1997;144:183-192. nents, and more than half had ataxic components. The spec- 16. Albin RL. Pathophysiology of parkinsonism and dyskinesias. In: Krauss JK, Gross- trum of clinical findings in these patients was similar to man RG, Jankovic J, eds. Pallidal Surgery for the Treatment of Parkinson’s Disease those in previous reports.4,17 In keeping with an earlier re- and Movement Disorders. Philadelphia, Pa: Lippincott-Raven Publishers; 1998: 19 41-54. port, our series included an equivalent proportion of men 17. Litvan I, Agid Y, Jankovic J, et al. Accuracy of clinical criteria for the diagnosis of and women; however, a recent report20 stated that PSP af- progressive supranuclear palsy (Steele-Richardson-Olszewski syndrome). Neu- fects men more frequently than women. As in our sample, rology. 1996;46:922-930. 18. Verny M, Duyckaerts C, Agid Y, Hauw JJ. The significance of cortical pathology in PSP, resting is unusual, dysarthria and gaze limi- in progressive supranuclear palsy clinico-pathological data in 10 cases. Brain. tation are consistently present late in the illness, and ver- 1996;119:1123-1136. 19. Golbe LI. Epidemiology. In: Litvan I, Agid Y, eds. Progressive Supranuclear Palsy: tical eye movements are more severely affected than hori- Clinical and Research Approaches. New York, NY: Oxford University Press Inc; 4,17,20,21 zontal eye movements. 1992:34-43. In this study, the time from diagnosis and speech 20. Santacruz P, Uttl B, Litvan I, Grafman J. Progressive supranuclear palsy: a sur- vey of the disease course. Neurology. 1998;50:1637-1647. pathology evaluation to death was long, and many of our 21. Duvoisin RC. Differential diagnosis of PSP. J Neural Transm Suppl. 1994;42: patients became anarthric during that time. The neuro- 51-67.

(REPRINTED) ARCH NEUROL / VOL 58, FEB 2001 WWW.ARCHNEUROL.COM 269

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021