Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jul 30.
Published in final edited form as: Neuroepidemiology. 2015 Jul 30;45(1):59–69. doi: 10.1159/000437228

Patterns of motor and non-motor features in medication-naïve Parkinsonism

Samay Jain 1,*, Seo-Young Park 2, Diane Comer 2
PMCID: PMC4543435  NIHMSID: NIHMS704731  PMID: 26227734

Abstract

Background

Parkinsonism is defined by motor features (tremor, bradykinesia, rigidity, and postural instability). Accompanying non-motor features (e.g., cognitive, autonomic, sleep disturbances) are under-recognized and under-treated. We hypothesized clinical patterns occurring in early, medication-naïve Parkinsonism are distinguished by features such as tremor, sleep, autonomic, and cognitive dysfunction.

Methods

Clinical and neuroimaging data were obtained in the Parkinson’s Progression Marker Initiative (PPMI). Group comparisons of Parkinsonism with dopaminergic deficits (PDD) (N=388)), controls (N=196), and Parkinsonism with scans without evidence of dopaminergic deficits (SWEDD’s) (N=64) were done with ANOVA, chi-square, and post-hoc pairwise tests. To examine clinical patterns within the PDD group, k-means clustering was performed with non-motor or motor features, or both.

Results

Among PDD, four non-motor patterns (% of PDD) (impulsive (14.9%), sleep-autonomic (22.9%), cognitive-olfactory (18.0%), and mild (44.1%)), four motor patterns (tremor plus bradykinesia (56.2%), tremor without bradykinesia (16.2%), postural instability (6.7%) and no tremor (20.9%)) and five combined motor/non-motor patterns (tremor with bradykinesia (42.3%), tremor without bradykinesia (15.5%), no tremor and mild non-motor features (17.0%), postural instability with sleep-autonomic disturbances (6.7%) and oldest onset cognitive-olfactory (18.6%)) were observed.

Conclusions

To our knowledge, this is the first description of non-motor clinical patterns in early, medication-naïve Parkinsonism, suggesting such features are intrinsic to Parkinsonian disorders.

Keywords: Parkinsonism, non-motor features, SWEDD, DaTScan

Introduction

Parkinsonism is common in neurological outpatient clinics, with an incidence ranging from 0.5/1000 person-years between the ages of 55 and 65 years, to 10.6/1000 for those over 85 years of age.[1] As there is currently no diagnostic test for Parkinsonism, identification relies solely on clinical signs comprised of characteristic motor features (e.g., tremor, bradykinesia, rigidity, gait instability) which are also the focus of treatment. Clinical and pathological evidence suggests that several idiopathic Parkinsonian disorders are multi-system, multi-organ diseases in which motor deficits are accompanied by non-motor features including cognitive, autonomic, psychiatric and sleep disturbances.[2] Non-motor features in Parkinson disease predate motor dysfunction by more than 20 years, and are linked to widespread neuropathological changes throughout the nervous system.[3] Although non-motor features often have greater impact on healthcare costs, quality of life, and institutionalization rates than motor features, they are under-recognized.[4, 5] Yet several non-motor features are treatable.

We used data collected by the Parkinson’s Progression Marker Initiative (PPMI) to test our hypothesis that specific patterns of non-motor features occur in patients with early idiopathic Parkinsonism who have never received anti-Parkinsonian medication. We sought to characterize clinical patterns of early Parkinsonism integrating non-motor and motor features because their recognition early in the course of Parkinsonism would facilitate development of more comprehensive early treatment strategies and inform concepts of pathogenesis. As deficits in dopamine neurotransmission underlie most idiopathic Parkinsonism, we also examined dopaminergic deficits in participants using data collected from neuroimaging which labeled dopamine transporters in the striatum. The status of striatal dopamine transporters led to two main groups of Parkinsonism: Parkinsonism with scans without evidence of dopaminergic deficits (SWEDD’s) and those with Parkinsonism with dopaminergic deficits (PDD).

Materials and Methods

Study Population

The PPMI is an ongoing observational multicenter cohort study designed to identify Parkinson Disease progression markers comprised of three groups: PDD (N=388), controls (N=196), and SWEDD’s (N=64). Details are published elsewhere and available at http://www.ppmi-info.org/.[6] The study was approved by the institutional review board of all participating sites. Written informed consent was obtained from all participants. Healthy control subjects had no significant neurological dysfunction, no first degree relative with Parkinson disease, a Montreal Cognitive Assessment (MOCA) > 26, and no detectable dopamine transporter deficit on neuroimaging (DaTscan, methods detailed below). Participants with Parkinsonism were diagnosed less than 2 years prior to the screening visit, untreated, and required to have an asymmetric resting tremor or asymmetric bradykinesia or two of bradykinesia, resting tremor and rigidity. Patients with clinical findings consistent with early stage Parkinsonism, no history of secondary causes of Parkinsonism and a dopamine transporter deficit on DaTscan imaging comprised the PDD group, and those without evidence of dopaminergic deficit comprised the SWEDD group. All participants underwent tests summarized in Table 1. We used data from the earliest time available (screening, baseline or first follow-up visit) downloaded from the PPMI database (download dates were 9 July 2013 for clinical assessments and 10 Sept 2013 for imaging).

Table 1.

Tests performed on all Participants in PPMI

Clinical Assessments
Motor
 (higher score is worse)		 Movement Disorder Society Unified Parkinson Disease Rating Scale (MDS-UPDRS)
Neurobehavior
 (higher scores are worse)		 Geriatric depression scale
State-trait anxiety inventory
Questionnaire for Impulsive-compulsive disorders (screening: any one item is considered positive)
Cognitive Testing
 (lower scores are worse)		 Montreal Cognitive Assessment (MOCA)
Hopkins verbal learning test – revised
Benton judgment of line orientation
Semantic fluency
Letter number sequencing
Symbol digit modalities test
Autonomic symptoms
 (higher score is worse)		 Scale for Outcome of Parkinson disease – Autonomic (SCOPA-AUT)
Sleep disorders
 (higher scores are worse)		 Epworth sleepiness scale
REM sleep disorder questionnaire
Olfactory testing
 (lower score is worse)		 University of Pennsylvania Smell Inventory (UPSIT)
Imaging & Biospecimen collection
Dopamine transporter imaging [123I]-FP-CITsingle photon emission computerized tomography (DaTscan)
Open in a new tab

Assessment of clinical features

Demographic and historical information were obtained on all participants, including age, sex, years of education, age of onset of first symptoms of PD, motor features present at diagnosis, and date of diagnosis. Parkinsonism was assessed by the Movement Disorder Society sponsored Unified Parkinson disease Rating Scale (MDS-UPDRS). Similar to other studies,[7, 8] motor features were summarized by taking the average score of MDS-UPDRS individual items for postural instability/gait (sum of items (3.10 through 3.14)/4), hypokinesia/rigidity (sum of items (3.1 through 3.9 + 3.14)/9), and tremor (sum of items (3.15a through 3.18)/10). Non-motor features were measured by the sum of MDS-UPDRS Part 1 and specific assessments of olfaction, cognition, sleep, autonomic, and psychiatric features summarized in Table 1.

Neuroimaging

All subjects underwent dopamine transporter imaging by DaTscan (an intravenous injection of [123I]-FP-CIT containing activity in the range of 111–185 MBq over 15–20 seconds followed by single photon emission computerized tomography (SPECT) imaging done 3–6 hours later).[9] Details are available on the PPMI website at http://www.ppmi-info.org/. Regions of interest were placed on the left and right caudate, the left and right putamen, and the occipital cortex (reference tissue). Count densities for each region were used to calculate striatal binding ratios (SBR’s) for each of the four striatal regions (SBR = (target region/reference region) −1).

Statistical Analyses

Descriptive statistics, ANOVA F-tests, t tests, and chi-square tests summarized baseline characteristics and group comparisons. Three k-means cluster analyses were performed among PDD participants: one based on 14 non-motor variables, one based on 7 motor variables, and one based on both non-motor and motor variables. We included 388 of the 423 PDD participants without any missing values for clustering. Variables were standardized before clustering so that each had a mean zero and standard deviation one. For binary variables, we assigned values zero or 1 and treated them in the same way as continuous variables. As we standardized all variables before clustering so that each variable has mean zero and standard deviation 1, these values (0 and 1) might have changed. To empirically determine the number of clusters, we compared the sum of squared error (SSE) for a number of cluster solutions.[10] SSE is the sum of the squared distance between each member of a cluster and its cluster centroid. We looked for a point with a sudden drop of SSE to find the number of clusters. Also, we produced 250 randomized versions of the original input data by randomly scrambling all entries of the data matrix, and calculated SSE against cluster solutions for the randomized data. If a data set has strong clusters, the SSE of the actual data should decrease more quickly than the random data as the number of clusters increase. We also looked at the Gap statistic as another measure for estimating the number of clusters.[11] In this way, we chose 4, 4, and 5 clusters for clustering based on non-motor, motor, and combined variables, respectively. We then compared the resulting clusters with ANOVA F-test and chi-square test for continuous and binary variables, respectively. For variables that were significantly different across clusters (p≤0.05), we performed post-hoc pairwise analysis using ANOVA with a Tukey adjustment. We labeled clinical patterns using descriptors based on variables that were significantly different among clusters. For example, if cognitive and olfactory testing were significantly worse in a particular cluster, that cluster was labeled cognitive-olfactory. (SAS version 9.3 (2012)[12] was used to prepare downloaded datasets then analyzed by R version 3.0.1 (2013)[13]).

Results

Group comparisons are in Table 2. Compared to controls, PDD dopamine transporter imaging SBR’s and olfactory function (University of Pennsylvania Smell Inventory (UPSIT) scores) were lower and scores for posture/gait, scores for hypokinesia/rigidity and tremor abnormalities were higher. SWEDD’s scored highest in severity of non-motor features for the MDS-UPDRS Part 1, Scale for Outcome of Parkinson disease – Autonomic (SCOPA-AUT), and Epworth Sleepiness Scale and had the highest proportion of individuals with impulsive/compulsive behaviors. Controls performed best in most cognitive tests. Clustering using non-motor features yielded four patterns in the PDD group (Table 3): (1) Impulsive: presence of impulsive/compulsive behaviors,; (2) Sleep–autonomic: most severe non-motor (MDS-UPDRS Part 1), autonomic (SCOPA-AUT) and REM sleep disorder symptoms; (3) Cognitive-olfactory: performed worst on all cognitive tests and had low UPSIT scores; and (4) Mild: no impulsive/compulsive behaviors and the best UPSIT performance. This four cluster solution accounted for 24.7% of the variance.

Table 2.

Group Comparisons

Group Controls
N=196		 PDD
N=388		 SWEDD
N=64
Age at baseline(years) 60.8 (11.2) 61.5 (9.79) 61.0 (10.0)
Age of onset 59.6 (10.0) 58.8 (10.5)
Sex (N (%women)) 70 (35.7%) 132 (34.0%) 24 (37.5%)
Time from first symptom to baseline visit (months) 22.7 (23.7) 23.5 (27.5)
Time from first symptom to diagnosis (months) 17.1 (22.2) 18.6 (26.3)
Total Years of Education 16.0 (2.89) 15.6 (3.00) 15.1 (3.87)
Non-motor measures
MDS Non Motor Pt 1 1,2 2.92 (2.97) 5.55 (4.05) 8.25 (6.47)
UPSIT1,2 34.0 (4.85) 22.3 (8.28) 31.4 (6.23)
SCOPA-AUTO1,2 8.88 (7.43) 13.4 (9.55) 17.2 (12.2)
Cognitive Measures
 Benton Line Judgement 13.1 (1.98) 12.8 (2.12) 12.8 (2.38)
 Hopkins verbal learning 2 15.6 (2.29) 14.7 (2.59) 14.4 (2.55)
 Letter number sequence 3 10.9 (2.57) 10.6 (2.66) 9.88 (2.66)
 Montreal Cognitive Assessment Battery2 28.2 (1.11) 27.2 (2.33) 27.1 (2.44)
 Semantic Fluency 2 51.8 (11.2) 48.9 (11.7) 45.2 (12.4)
 Symbol Digit2 46.8 (10.5) 41.3 (9.87) 41.2 (11.9)
Sleep-related Measures
 REM2 2.85 (2.26) 4.13 (2.68) 4.55 (2.86)
 Epworth Sleepiness1 5.64 (3.42) 5.83 (3.46) 8.08 (4.80)
Psychiatric Measures
 Presence of Impulsive/Compulsive behaviors1 36 (18.7%) 77 (19.8%) 21 (32.8%)
 Depression 5.18 (1.38) 5.27 (1.45) 5.64 (1.71)
 Anxiety State and Trait 47.0 (3.50) 46.6 (3.88) 46.6 (3.83)
Motor Measures
Posture/Gait score1,2 0.04 (0.10) 0.34 (0.28) 0.21 (0.31)
Hypokinesia/Rigidity score1,2 0.04 (0.08) 0.80 (0.41) 0.48 (0.43)
Tremor score2 0.03 (0.08) 0.43 (0.31) 0.44 (0.29)
Motor Feature present at dx (N (% of group)):
 Tremor 0 301 (77.6%) 53 (84.1%)
 Rigidity4 0 298 (76.8%) 37 (58.7%)
 Bradykinesia 0 321 (82.7%) 51 (81.0%)
 Postural Instability 0 26 (6.70%) 8 (12.9%)
Asymmetry of motor features4
 Left side more affected 0 166 (42.8%) 15 (23.4%)
 Right side more affected 0 212 (54.6%) 44 (68.8%)
 Both sides equally affected 0 10 (2.58%) 5 (7.81%)
NHY (Hoehn and Yahr Stage)2 0 1.55 (0.51) 1.42 (0.50)
Neuroimaging
DATScan
 R Caudate4,5 2.98 (0.62) 1.99 (0.58) 2.81 (0.60)
 L Caudate4,5 3.03 (0.64) 1.99 (0.57) 2.82 (0.58)
 R Putamen4,5 2.17 (0.61) 0.86 (0.37) 2.07 (0.52)
 L Putamen4,5 2.16 (0.59) 0.83 (0.36) 2.03 (0.52)
Open in a new tab

SWEDD=Scans without evidence of dopaminergic deficits

Please refer to Table 1 for Legend. Unless noted, all values in parentheses are standard deviations of raw scores.

1

SWEDD group is significantly different from Control and PDD groups

2

Control group is significantly different from SWEDD and PDD groups

3

SWEDD group is significantly different from Control group but not from PDD group

4

PDD group is significantly different from SWEDD group but not controls

5

Control group is significantly different from PDD group but not SWEDD group

Table 3.

Results of clustering using only non-motor features

Cluster Description Impulsive Cognitive-Olfactory Mild Sleep-autonomic
N (% of all PDD participants) 58 (14.9%) 70 (18.0%) 171 (44.1%) 89 (22.9%)
Non-motor measures used for clustering
MDS Non Motor Pt 11,2 6.60 (3.44) 4.20 (2.25) 3.52 (2.46) 9.82 (4.51)
UPSIT3,4 22.0 (7.81) 17.4 (7.18) 25.7 (7.42) 19.9 (8.24)
SCOPA-AUTO (high is worse)1,2 15.3 (8.91) 11.4 (7.78) 9.3 (6.02) 21.5 (11.30)
Cognitive Measures
 Benton Line Judgement5,6 13.3 (1.62) 10.6 (2.34) 13.6 (1.54) 12.7 (2.05)
 Hopkins verbal learning2,5 15.5 (2.23) 12.0 (2.36) 15.7 (2.01) 14.2 (2.42)
 Letter number sequence2,5 11.5 (2.70) 7.93 (2.11) 11.6 (2.38) 10.3 (1.94)
 Montreal Cognitive Assessment Battery5,6 27.4 (2.11) 25.4 (2.90) 28.0 (1.64) 26.8 (2.27)
 Semantic Fluency2,5 52.3 (10.68) 38.9 (8.01) 53.6 (7.55) 45.6 (8.20)
 Symbol Digit2,5 46.0 (10.37) 31.3 (8.01) 45.2 (7.55) 38.7 (8.20)
Sleep-related Measures
 REM2,7 3.84 (2.49) 4.26 (2.80) 3.01 (1.79) 6.38 (2.77)
 Epworth Sleepiness6 6.28 (3.36) 5.67 (3.37) 5.17 (3.18) 6.93 (3.84)
Psychiatric Measures
 Presence of Impulsive/Compulsive behaviors* 58 (100%) 9 (12.9%) 0 (0%) 10 (11.2%)
 Depression8 5.02 (1.61) 5.64 (1.57) 5.04 (1.30) 5.60 (1.43)
 Anxiety State and Trait2 47.2 (3.70) 47.9 (3.93) 46.9 (3.60) 44.7 (3.87)
Motor Measures
Posture/Gait score9 0.28 (0.23) 0.37 (0.28) 0.28 (0.28) 0.45 (0.28)
Hypokinesia/Rigidity score9 0.70 (0.30) 0.86 (0.49) 0.75 (0.39) 0.92 (0.41)
Tremor score 0.42 (0.27) 0.51 (0.33) 0.39 (0.32) 0.44 (0.30)
Motor Feature present at dx (N (% of cluster)):
Tremor 46 (79.3%) 56 (80.0%) 131 (76.6%) 68 (22.9%)
Rigidity 43 (74.1%) 54 (77.1%) 129 (75.4%) 72 (18.6%)
Bradykinesia 51 (87.9%) 58 (82.9%) 140 (81.8%) 72 (18.6%)
Postural Instability* 2 (3.4%) 1 (1.4%) 10 (5.8%) 13 (14.6%)
Left side affected 23 (39.7%) 31 (44.3%) 77 (45.0%) 35 (39.3%)
Right side affected 35 (60.3%) 39 (55.7%) 89 (52.0%) 49 (55.1%)
Both sides affected 0 (0%) 0 (0%) 5 (2.9%) 5 (5.6%)
NHY (Hoehn and Yahr Stage) 1.53 (0.50) 1.60 (0.49) 1.49 (0.52) 1.65 (0.48)
Other characteristics
Age onset (years)9,10 56.5 (10.6) 66.2 (7.62) 56.5 (9.65) 62.4 (8.84)
Sex (N (%women)) 24 (41.4%) 17 (24.3%) 64 (37.4%) 27 (30.3%)
Time from 1st symptom to baseline visit (months) 27.4 (43.7) 17.9 (14.9) 21.8 (17.3) 25.2 (21.2)
Time from 1st symptom to diagnosis (months) 21.4 (39.8) 13.2 (14.6) 16.5 (16.6) 18.5 (20.5)
Total Years of Education5 15.7 (2.58) 14.0 (3.33) 16.1 (2.89) 15.7 (2.83)
Neuro-imaging
DaTScan
 R Caudate11 2.18 (0.63) 1.89 (0.61) 2.04 (0.51) 1.85 (0.59)
 L Caudate 2.11 (0.59) 1.89 (0.55) 2.03 (0.53) 1.94 (0.63)
 R Putamen11 0.98 (0.43) 0.80 (0.34) 0.88 (0.35) 0.79 (0.38)
 L Putamen 0.86 (0.37) 0.79 (0.33) 0.84 (0.34) 0.81 (0.40)
Open in a new tab

Please refer to Table 1 for Legend. Unless noted, all values in parentheses are standard deviations of raw scores, and significant differences are p<0.05 after Tukey adjustment.

*

significantly different by chi-square

1

Impulsive group significantly different from all other groups

2

Sleep/Autonomic group significantly different from all other groups

3

Mild significantly different from all other groups

4

Impulsive group significantly different from Olfactory/Cognition group

5

Cognitive-olfactory group significantly different from all other groups

6

Sleep/Autonomic group significantly different from Mild group

7

Mild group significantly different from Cognitive-olfactory group

8

Mild group significantly different from Sleep/Autonomic and Cognitive-olfactory group

9

Sleep/Autonomic group significantly different from Impulsive and Mild groups

10

Cognitive-olfactory group significantly different from Impulsive and Mild groups

11

Impulsive group significantly different from Sleep/Autonomic and Cognitive-olfactory group

Clustering using motor features also yielded four patterns in the PDD group (Table 4): (1) Tremor plus bradykinesia: tremor and bradykinesia at the time of diagnosis; (2) Tremor without bradykinesia: tremor and no bradykinesia at the time of diagnosis; (3) Postural instability: postural instability at the time of diagnosis and the highest posture/gait scores at baseline; and (4) No tremor: no tremor at the time of diagnosis. This four cluster solution accounted for 47.0% of the variance.

Table 4.

Results of clustering using only motor features

Cluster Description Tremor with Brady Tremor no Brady Postural Instability No tremor
N=388 (% of all PDD participants) 218 (56.2%) 63 (16.2%) 26 (6.7%) 81 (20.9%)
Motor characteristics used for clustering
Motor Feature present at dx (N (% of cluster)):
 Tremor* 218 (100%) 62 (98.4%) 21 (80.8%) 0 (0%)
 Rigidity* 177 (82.2%) 26 (41.3%) 24 (92.3%) 71 (87.7%)
 Bradykinesia* 218 (100%) 0 (0%) 24 (92.3%) 79 (97.5%)
 Postural Instability* 0 (0%) 0 (0%) 26 (100%) 0 (0%)
Posture/Gait score1 0.32 (0.27) 0.29 (0.27) 0.55 (0.34) 0.34 (0.29)
Hypokinesia/Rigidity score2 0.80 (0.39) 0.72 (0.43) 0.98 (0.47) 0.82 (0.40)
Tremor score3 0.49 (0.31) 0.58 (0.28) 0.45 (0.22) 0.15 (0.17)
Non-motor characteristics
MDS Non Motor Pt 11 5.53 (3.91) 5.14 (4.43) 8.12 (4.76) 5.09 (3.64)
UPSIT 22.7 (7.84) 22.0 (8.29) 21.9 (8.73) 21.8 (9.34)
SCOPA-AUTO4 13.9 (10.3) 11.8 (8.85) 18.0 (9.05) 11.7 (7.44)
Cognitive Measures
 Benton Line Judgement 12.8 (2.11) 12.4 (2.32) 13.0 (1.80) 12.9 (2.11)
 Hopkins verbal learning 14.7 (2.60) 14.6 (2.58) 14.3 (2.59) 14.7 (2.63)
 Letter number sequence 10.6 (2.64) 10.5 (2.68) 10.4 (2.63) 10.8 (2.75)
 Montreal Cognitive Assessment Battery2 27.2 (2.42) 26.6 (2.39) 28.0 (1.62) 27.4 (2.13)
 Semantic Fluency 49.0 (11.3) 48.2 (11.9) 49.2 (10.5) 49.0 (13.2)
 Symbol Digit 40.6 (10.26) 43.0 (8.51) 41.2 (8.57) 42.1 (10.1)
Sleep-related measures
 REM1 4.12 (2.61) 3.76 (2.35) 5.57 (3.37) 3.99 (2.76)
 Epworth Sleepiness 5.70 (3.38) 5.51 (3.61) 7.54 (3.57) 5.89 (3.44)
Psychiatric Measures
 Presence of Impulsive/Compulsive behaviors 48 (22.0%) 9 (14.3%) 7 (26.9%) 13 (16.0%)
 Depression 5.25 (1.51) 5.41 (1.12) 5.69 (1.44) 5.09 (1.52)
 Anxiety State and Trait5 46.9 (3.79) 47.0 (3.83) 46.6 (3.35) 45.5 (4.16)
Other Characteristics
NHY (Hoehn and Yahr Stage) 1.57 (0.50) 1.48 (0.53) 1.69 (0.47) 1.52 (0.50)
Age onset (years) 59.6 (9.93) 61.3 (9.95) 60.0 (9.83) 58.2 (10.5)
Sex (N (%)) 72 (33.0%) 25 (39.7%) 7 (26.9%) 26 (32.1%)
Time from 1st symptom to baseline visit (months) 23.9 (27.6) 22.6 (17.8) 28.7 (23.1) 17.9 (14.3)
Time from 1st symptom to diagnosis (months) 18.3 (25.8) 15.6 (16.9) 21.5 (23.2) 13.8 (13.3)
Side affected at time of diagnosis
 Left side affected at diagnosis 97 (44.5%) 19 (30.2%) 12 (46.2%) 38 (46.9%)
 Right side affected at diagnosis 117 (53.7%) 43 (68.3%) 12 (46.2%) 40 (49.4%)
 Both sides affected at diagnosis 4 (1.8%) 1 (1.6%) 2 (7.7%) 3 (3.8%)
Total Years of Education 15.6 (3.02) 15.7 (3.00) 16.1 (2.32) 15.0 (3.13)
Neuro-imaging
DaTScan
 R Caudate 1.99 (0.54) 2.11 (0.58) 1.87 (0.65) 1.93 (0.61)
 L Caudate 2.01 (0.55) 2.14 (0.59) 1.94 (0.60) 1.88 (0.57)
 R Putamen 0.86 (0.39) 0.96 (0.36) 0.76 (0.34) 0.82 (0.31)
 L Putamen 0.85 (0.38) 0.85 (0.32) 0.74 (0.22) 0.78 (0.35)
Open in a new tab

Please refer to Table 1 for Legend. Unless noted, all values in parentheses are standard deviations of raw scores, and significant differences are p<0.05 after Tukey adjustment.

*

Significantly different by chi-square

1

Postural Instability significantly different from all other groups

2

Postural Instability significantly different from Tremor-no-bradykinesia

3

No Tremor significantly different from all other groups

4

Postural Instability significantly different from Tremor-no-bradykinesia and No-Tremor groups

5

No Tremor significantly different from Tremor-no-bradykinesia

When we used both non-motor and motor features for clustering, five patterns emerged in the PDD group (Table 5): (1) Tremor plus bradykinesia and (2) Tremor without bradykinesia were characterized as described above; (3) No tremor and mild non-motor symptoms (No Tremor-mild): no tremor at diagnosis and lower severity for several non-motor features; (4) Postural instability with sleep and autonomic features: postural instability at the time of diagnosis and the most severe sleep and autonomic symptoms; and (5) Oldest onset cognitive-olfactory: the oldest age of onset of PDD with the worst cognitive and olfactory performance. This five cluster solution accounted for 22.7% of the variance.

Table 5.

Result of clustering using both non-motor and motor features

Cluster Description Tremor with Bradykinesia Tremor no Bradykinesia No Tremor and mild PISA OCO
N=388 (% of entire sample) 164 (42.3%) 60 (15.5%) 66 (17.0%) 26 (6.7%) 72 (18.6%)
Non-motor and Motor measures used for clustering
MDS Non Motor Pt 11,3 5.37 (3.94) 5.25 (4.52) 4.88 (3.73) 8.12 (4.76) 5.89 (3.62)
UPSIT5 24.0 (7.42) 22.5 (8.14) 23.9 (8.84) 21.9 (8.73) 17.0 (7.50)
SCOPA-AUTO 13.4 (10.1) 12.1 (8.96) 10.7 (6.90) 18.0 (9.05) 14.9 (10.2)
Cognitive Measures
 Benton Line Judgement 13.4 (1.64) 12.5 (2.33) 13.4 (1.73) 13.0 (1.80) 11.0 (2.34)
 Hopkins verbal learning2 15.1 (2.10) 14.8 (2.41) 15.3 (2.17) 14.3 (2.59) 12.0 (2.35)
 Letter number sequence2 11.4 (2.31) 10.8 (2.51) 11.5 (2.35) 10.4 (2.63) 7.9 (2.03)
 Montreal Cognitive Assessment Battery2,4 27.7 (1.88) 26.8 (2.27) 27.8 (1.77) 28.0 (1.62) 25.5 (2.98)
 Semantic Fluency2 51.6 (10.6) 49.1 (11.4) 52.4 (11.9) 49.2 (10.5) 39.4 (9.55)
 Symbol Digit2 44.0 (8.63) 44.0 (7.04) 44.9 (8.50) 41.2 (8.57) 29.9 (7.68)
Sleep-related Measures
 REM1,5 3.73 (2.40) 3.75 (2.40) 3.62 (2.44) 5.57 (3.37) 5.34 (2.97)
 Epworth Sleepiness6 5.64 (3.27) 5.48 (3.65) 5.42 (3.00) 7.54 (3.57) 6.31 (3.93)
Psychiatric Measures
 Presence of Impulsive/Compulsive behaviors 36 (21.9%) 9 (15.0%) 10 (15.2%) 7 (26.9%) 15 (20.8%)
 Depression 5.14 (1.42) 5.47 (1.11) 5.14 (1.57) 5.69 (1.44) 5.38 (1.64)
 Anxiety State and Trait 46.8 (3.84) 46.9 (3.90) 45.6 (3.74) 46.6 (3.35) 46.9 (4.23)
Motor Measures
Posture/Gait score1,5 0.29 (0.25) 0.28 (0.27) 0.29 (0.27) 0.55 (0.34) 0.47 (0.28)
Hypokinesia/Rigidity score5,7 0.76 (0.36) 0.67 (0.38) 0.78 (0.37) 0.98 (0.47) 0.98 (0.48)
Tremor score8 0.47 (0.30) 0.57 (0.28) 0.13 (0.17) 0.45 (0.22) 0.49 (0.31)
Motor Feature at diagnosis (N (% of cluster)):
 Tremor* 164 (100%) 59 (98.3%) 0 (0%) 21 (80.7%) 57 (79.2%)
 Rigidity* 133 (81.1%) 25 (41.7%) 59 (89.4%) 24 (92.3%) 57 (79.2%)
 Bradykinesia* 164 (100%) 0 (0%) 65 (98.5%) 24 (92.3%) 68 (94.4%)
 Postural Instability* 0 (0%) 0 (0%) 0 (0%) 26 (100%) 0 (0%)
Other characteristics
Age onset (years)2 57.5 (9.83) 60.7 (9.83) 56.8 (10.7) 60.0 (9.84) 66.1 (6.87)
NHY (Hoehn and Yahr Stage) 1.54 (0.51) 1.45 (0.53) 1.50 (0.50) 1.69 (0.47) 1.67 (0.47)
Sex (N (%)) 56 (34.1%) 25 (41.7%) 23 (38.4%) 9 (34.6%) 19 (26.4%)
Time from 1st symptom to baseline visit (months) 24.9 (30.5) 22.8 (18.0) 18.6 (14.2) 28.7 (23.1) 19.3 (15.7)
Time from 1st symptom to diagnosis (months) 19.0 (28.5) 15.7 (17.1) 14.3 (13.4) 21.5 (23.2) 14.8 (14.4)
Side affected at time of diagnosis
 Left side affected 72 (43.9%) 19 (31.7%) 31 (47.0%) 12 (46.2%) 32 (44.4%)
 Right side affected 89 (54.3%) 40 (66.7%) 33 (50.0%) 12 (46.2%) 38 (52.8%)
 Both sides affected 3 (1.8%) 1 (1.7%) 2 (3.0%) 2 (7.7%) 2 (2.8%)
Total Years of Education9 16.1 (2.62) 15.9 (2.93) 15.3 (2.92) 16.1 (2.32) 14.2 (3.72)
Neuro-imaging
DaTScan
 R Caudate10 2.04 (0.55) 2.10 (0.60) 1.97 (0.57) 1.87 (0.65) 1.82 (0.57)
 L Caudate 2.03 (0.55) 2.15 (0.59) 1.94 (0.53) 1.94 (0.60) 1.87 (0.60)
 R Putamen10 0.88 (0.40) 0.97 (0.37) 0.83 (0.30) 0.76 (0.34) 0.78 (0.36)
 L Putamen 0.84 (0.35) 0.86 (0.32) 0.80 (0.36) 0.74 (0.22) 0.82 (0.41)
Open in a new tab

PISA= Postural Instability, Sleep and Autonomic features;

OCO: Oldest onset cognitive and olfactory features

Please refer to Table 1 for Legend._Unless noted, all values in parentheses are standard deviations of raw scores, and significant differences are p<0.05 after Tukey adjustment.

*

significantly different by chi-square

1

PISA significantly different from Tremor with Bradykinesia, Tremor-no-bradykinesia and No-Tremor groups

2

Cognitive-olfactory group significantly different from all other groups

3

Tremor plus Bradykinesia significantly different from Tremor-no-bradykinesia

4

Tremor-no-bradykinesia significantly different from No-Tremor

5

Cognitive-olfactory group significantly different from Tremor plus Bradykinesia, Tremor-no-bradykinesia and No-Tremor groups

6

PISA significantly different from No-Tremor group

7

PISA significantly different from Tremor-no-bradykinesia group

8

No-Tremor significantly different from all other groups

9

Cognitive-olfactory group significantly different from Tremor plus Bradykinesia, Tremor-no-bradykinesia and PISA

10

Tremor-no-bradykinesia significantly different from OCO group

11

Cognitive-olfactory group significantly different from PISA and No Tremor

12

Cognitive-olfactory group significantly different from No Tremor

Discussion

To our knowledge, this is the first report characterizing non-motor clinical patterns in early medication naïve Parkinsonism. We found non-motor symptoms, particularly sleep and autonomic features, to be worse in SWEDD’s than controls or PDD. We also found SWEDD’s to have highest prevalence of impulsive/compulsive behaviors, and this may result in them reporting more non-motor symptoms. Although SWEDD’s could be very early Parkinsonism which later demonstrate dopamine transport deficits, in other studies follow up of SWEDD’s over 4 years does not demonstrate decreasing striatal dopamine in most cases.[14–16] SWEDD’s may represent other conditions such as secondary Parkinsonism, Huntington disease, adult-onset dystonic tremor, essential tremor, psychogenic tremor or Fragile X permutation.[17] As in another study, we found olfactory function in SWEDD’s was better than PD.[18] Unlike our results, a separate study reported SWEDD’s have less severe non-motor issues of urinary symptoms, sleep disturbances, and behavior as reported by the Non-Motor Symptoms Scale.[19] This study adds to prior reports because we focus on clinical patterns in early medication naïve Parkinsonism with confirmed status of striatal dopamine transporter binding. Unlike our study, others who have employed clustering using non-motor features did so in Parkinson disease after a substantial proportion of participants were exposed to dopaminergic treatment, controls were not included, and the means to distinguish SWEDD’s was not available.[20, 21] Others have investigated clusters later in the disease course,[22, 23] when a major concern about non-motor features is the degree to which they are intrinsic to Parkinsonism, as some may be secondary to medication side effects. For example, dopaminergic medication may contribute to impulse control disorders (ICD’s), psychosis or orthostatic hypotension.[24] Our results in participants never exposed to anti-Parkinsonian medication provide strong evidence that such non-motor features are also intrinsic to Parkinsonism.

Regarding non-motor patterns, the impulsive pattern includes ICD’s such as gambling, shopping, sexual behavior, and eating.[25] Younger age of onset of Parkinson disease is associated with ICD’s,[26] and this cluster was younger than the sleep-autonomic and cognitive-olfactory patterns. Dopamine is involved in regulating motivation, drive and learning stimulus-reinforcement behaviors.[27]The impulsive pattern had significantly higher dopamine transporter binding in the right caudate and putamen than the sleep-autonomic and cognitive-olfactory patterns. SWEDD’s, which were excluded from clustering analyses, had higher dopamine transporter binding and also had higher impulsivity scores.

The sleep-autonomic pattern is consistent with studies finding autonomic dysfunction among those with REM sleep behavior disorder (RBD).[28],[29] RBD is a parasomnia in which patients “act-out” dreams with motor movements while in REM sleep. Neurodegenerative disease eventually develops in up to 80% of RBD cases, with the most common being Parkinson disease.[30] The cognitive-olfactory pattern is consistent with olfactory test scores showing correlations with verbal memory and executive performance.[31] The duration of education for the cognitive-olfactory pattern was lowest among all clusters. This supports the concept of cognitive reserve, which posits that lifelong experiences, including education, can increase cognitive tolerance of age or disease related changes.[32] The mild non-motor pattern may reflect non-motor features in their nascent phase that may become apparent as Parkinsonism progresses, or a milder non-motor variant of Parkinsonism.

Clustering patterns we observed with motor features of tremor-predominant and postural-instability/gait subtypes[7, 33] are similar to other reports in Parkinson disease. The postural instability pattern had the most severe non-motor features. To our knowledge this is the first report of two motor patterns: tremor with bradykinesia and tremor without bradykinesia.

Although the numeric differences in individual variables among the clusters are small, the usefulness of this study lies in the combination of variables resulting in distinct clinical patterns. These results suggest specific patterns of non-motor features manifest early in the course of Parkinsonism. However, this is a single analysis based on data at an early point in disease. Before on the clinical significance of these patterns can be established, the evolution of these clinical patterns with longitudinal follow up is necessary. Within PDD, DaTScan results were similar despite the clinically heterogeneous patterns observed. It is possible either non-dopaminergic pathways or areas outside of the striatum underlie nonmotor or motor features.

The pathological hallmark of the most common form of PDD, Parkinson disease, is Lewy-related pathology. Lewy formations are aggregates of the protein α-synuclein, and their distribution throughout the nervous system is thought to underlie both motor and non-motor features. Recently, Lewy-related pathology has been recognized to occur throughout the brain, spinal cord, and peripheral autonomic nervous system in Parkinson disease, and these regions underlie non-motor features. Neuropathological variability in extranigral regions may account for clinical heterogeneity we observed. For example, the density of α -synculein pathology in the olfactory bulb corresponds with olfactory deficits and correlates significantly with cognitive testing (the mini-mental state exam), supporting our clinical pattern of OCO. Multiple pathological studies demonstrate that non-tremor Parkinson disease cases have more severe cortical Lewy pathology, and clinically these cases are more likely to have some cognitive impairment.[34]

Although medication naïve Parkinsonism allows one to appreciate clinical features that are not secondary to medication effects, a major limitation is that we are limited in our ability to diagnose specific Parkinsonian syndromes. This is due to the fact that diagnostic criteria for Parkinson disease, the most common PDD, includes excellent response to dopaminergic medication, and lack of response may lead one to suspect another Parkinsonian disorder. While we were able to determine the status of striatal dopaminergic transporters with DaTScan, in the absence of a trial of dopaminergic medication, it is possible a clinical pattern in PDD could, in part, represent another disorder, such as multiple system atrophy in which autonomic dysfunction is more prominent. Results of cluster analysis are dependent on the variables selected for clustering, such that if different researchers used different features, different clinical patterns may emerge. Replication in other cohorts is necessary to determine validity and generalizability. Given that the PPMI includes those with very early Parkinsonism, it is possible that long-term clinical patterns have not yet emerged, particularly among mild non-motor patterns.

Conclusions

Although the temporal sequence of non-motor and motor features cannot be determined in our cross-sectional analyses, this study suggests heterogeneity in Parkinsonism exists very early in the course of disease. The presence of non-motor features in medication naïve participants suggests that non-motor features are intrinsic to Parkinsonian disorders. Even within the first 3 years of diagnosis among untreated Parkinson disease patients, non-motor symptoms make a larger contribution to diminished quality of life compared to motor features.[5] Several non-motor symptoms are treatable, including sleep disturbances, autonomic dysfunction, cognitive impairment, and psychiatric disorders. The presence of non-motor patterns in early Parkinsonism demonstrate the need for comprehensive treatment strategies which encompass both motor and non-motor features to begin near the time of diagnosis.

Acknowledgments

Funding

Data used in the preparation of this article were obtained from the Parkinson’s Progression Markers Initiative (PPMI) database (www.ppmi-info.org/data). For up-to-date information on the study, visit www.ppmi-info.org. This project was funded by the Michael J. Fox Foundation for Parkinson’s Research and the NIH/NINDS 1K23 NS070867. The PPMI – a public-private partnership – is funded by the Michael J. Fox Foundation for Parkinson’s Research and funding partners, including Biogen Idec, F. Hoffman-La Roche Ltd., GE Healthcare, Genentech, Abbvie, Avid Radiopharmaceuticals, Bristol-Myers Squibb, Covance, GlaxoSmithKline, Lilly, Lundbeck, Merk, Meso Scale Discovery, Piramal, UCB and Pfizer Inc.

Footnotes

Disclosures:

Dr. Jain reports no disclosures.

Dr. Park reports no disclosures.

Ms. Comer reports no disclosures

Conflicts of Interest: None

Financial Disclosures

None

Contributor Information

Seo-Young Park, Email: parksy@upmc.edu.

Diane Comer, Email: comerdm@upmc.edu.

References

  • 1.Aerts MB, Esselink RA, Post B, van de Warrenburg BP, Bloem BR. Improving the diagnostic accuracy in parkinsonism: a three-pronged approach. Practical neurology. 2012;12(2):77–87. doi: 10.1136/practneurol-2011-000132. Epub 2012/03/28. [DOI] [PubMed] [Google Scholar]
  • 2.Jellinger KA. Formation and development of Lewy pathology: a critical update. Journal of neurology. 2009;256(Suppl 3):270–9. doi: 10.1007/s00415-009-5243-y. [DOI] [PubMed] [Google Scholar]
  • 3.Jellinger KA. Journal of neural transmission. Vienna, Austria: 1996. Neuropathobiology of non-motor symptoms in Parkinson disease; p. 2015. Epub 2015/05/16. [DOI] [PubMed] [Google Scholar]
  • 4.Chaudhuri KR, Healy DG, Schapira AH. Non-motor symptoms of Parkinson’s disease: diagnosis and management. Lancet neurology. 2006;5(3):235–45. doi: 10.1016/S1474-4422(06)70373-8. [DOI] [PubMed] [Google Scholar]
  • 5.Muller B, Assmus J, Herlofson K, Larsen JP, Tysnes OB. Importance of motor vs. non-motor symptoms for health-related quality of life in early Parkinson’s disease. Parkinsonism & related disorders. 2013;19(11):1027–32. doi: 10.1016/j.parkreldis.2013.07.010. Epub 2013/08/07. [DOI] [PubMed] [Google Scholar]
  • 6.The Parkinson Progression Marker Initiative (PPMI) Progress in neurobiology. 2011;95(4):629–35. doi: 10.1016/j.pneurobio.2011.09.005. Epub 2011/09/21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Lewis SJ, Foltynie T, Blackwell AD, Robbins TW, Owen AM, Barker RA. Heterogeneity of Parkinson’s disease in the early clinical stages using a data driven approach. Journal of neurology, neurosurgery, and psychiatry. 2005;76(3):343–8. doi: 10.1136/jnnp.2003.033530. Epub 2005/02/18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Zetusky WJ, Jankovic J, Pirozzolo FJ. The heterogeneity of Parkinson’s disease: clinical and prognostic implications. Neurology. 1985;35(4):522–6. doi: 10.1212/wnl.35.4.522. Epub 1985/04/01. [DOI] [PubMed] [Google Scholar]
  • 9.Benamer TS, Patterson J, Grosset DG, Booij J, de Bruin K, van Royen E, et al. Accurate differentiation of parkinsonism and essential tremor using visual assessment of [123I]-FP-CIT SPECT imaging: the [123I]-FP-CIT study group. Mov Disord. 2000;15(3):503–10. Epub 2000/06/01. [PubMed] [Google Scholar]
  • 10.Peeples M. R Script for K-Means Cluster Analysis. 2011 http://www.mattpeeplesnet/kmeanshtml (January 7, 2015)
  • 11.Tibshirani R, Walther G, T H. Estimating the Number of Clusters in a Dataset via the Gap Statistic. Journal of the Royal Statistical Society, Series B. 2000;63:411–23. [Google Scholar]
  • 12.What’s New in SAS 9.3. Cary, NC: SAS Institute, Inc; 2012. [Google Scholar]
  • 13.Team RC. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. [Google Scholar]
  • 14.Marek K, Jennings D, Seibyl J. Long-term follow-up of patients with scans without evidence of dopaminergic deficit (SWEDD) in the ELLDOPA study. Neurology. 2005;64(suppl 1):A274. [Google Scholar]
  • 15.Peall KMH, Bajaj N. SWEDD-UK Meeting Report. Advances in Clinical Neuroscience and Rehabilitation. 2010;10(4):32–7. [Google Scholar]
  • 16.Batla A, Erro R, Stamelou M, Schneider SA, Schwingenschuh P, Ganos C, et al. Patients with scans without evidence of dopaminergic deficit: A long-term follow-up study. Mov Disord. 2014;29(14):1820–5. doi: 10.1002/mds.26018. Epub 2014/10/29. [DOI] [PubMed] [Google Scholar]
  • 17.Bajaj N. SWEDD for the General Neurologist. Advances in Clinical Neuroscience and Rehabilitation. 2010;10(4):30–1. [Google Scholar]
  • 18.Silveira-Moriyama L, Schwingenschuh P, O’Donnell A, Schneider SA, Mir P, Carrillo F, et al. Olfaction in patients with suspected parkinsonism and scans without evidence of dopaminergic deficit (SWEDDs) Journal of neurology, neurosurgery, and psychiatry. 2009;80(7):744–8. doi: 10.1136/jnnp.2009.172825. Epub 2009/03/12. [DOI] [PubMed] [Google Scholar]
  • 19.Jang WAJ, Kim HT. Non motor symptoms in PD patients with SWEDDs [abstract] Mov Disord. 2013;28(Suppl 1):299. [Google Scholar]
  • 20.Erro R, Vitale C, Amboni M, Picillo M, Moccia M, Longo K, et al. The heterogeneity of early Parkinson’s disease: a cluster analysis on newly diagnosed untreated patients. PloS one. 2013;8(8):e70244. doi: 10.1371/journal.pone.0070244. Epub 2013/08/13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yang HJ, Kim YE, Yun JY, Kim HJ, Jeon BS. Identifying the clusters within nonmotor manifestations in early Parkinson’s disease by using unsupervised cluster analysis. PloS one. 2014;9(3):e91906. doi: 10.1371/journal.pone.0091906. Epub 2014/03/20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.van Rooden SM, Colas F, Martinez-Martin P, Visser M, Verbaan D, Marinus J, et al. Clinical subtypes of Parkinson’s disease. Mov Disord. 2011;26(1):51–8. doi: 10.1002/mds.23346. Epub 2011/02/16. [DOI] [PubMed] [Google Scholar]
  • 23.Thenganatt MA, Jankovic J. Parkinson Disease Subtypes. JAMA neurology. 2014 doi: 10.1001/jamaneurol.2013.6233. Epub 2014/02/12. [DOI] [PubMed] [Google Scholar]
  • 24.Connolly BS, Lang AE. Pharmacological treatment of Parkinson disease: a review. Jama. 2014;311(16):1670–83. doi: 10.1001/jama.2014.3654. Epub 2014/04/24. [DOI] [PubMed] [Google Scholar]
  • 25.Weintraub D. Impulse control disorders in Parkinson’s disease: prevalence and possible risk factors. Parkinsonism & related disorders. 2009;15(Suppl 3):S110–3. doi: 10.1016/S1353-8020(09)70794-1. Epub 2010/01/30. [DOI] [PubMed] [Google Scholar]
  • 26.Weintraub D, Nirenberg MJ. Impulse control and related disorders in Parkinson’s disease. Neurodegenerative diseases. 2013;11(2):63–71. doi: 10.1159/000341996. Epub 2012/10/06. [DOI] [PubMed] [Google Scholar]
  • 27.Jain S, Waters C. Controversies with Dopamine Agonists: somnolence, valvulopathy and repetitive behaviors. Current Drug Therapy. 2007;2(1):17–20. [Google Scholar]
  • 28.Postuma RB, Gagnon JF, Montplaisir JY. REM sleep behavior disorder: From dreams to neurodegeneration. Neurobiology of disease. 2012 doi: 10.1016/j.nbd.2011.10.003. [DOI] [PubMed] [Google Scholar]
  • 29.Postuma RB, Lang AE, Massicotte-Marquez J, Montplaisir J. Potential early markers of Parkinson disease in idiopathic REM sleep behavior disorder. Neurology. 2006;66(6):845–51. doi: 10.1212/01.wnl.0000203648.80727.5b. [DOI] [PubMed] [Google Scholar]
  • 30.Schenck CH, Boeve BF, Mahowald MW. Delayed emergence of a parkinsonian disorder or dementia in 81% of older men initially diagnosed with idiopathic rapid eye movement sleep behavior disorder: a 16-year update on a previously reported series. Sleep medicine. 2013;14(8):744–8. doi: 10.1016/j.sleep.2012.10.009. Epub 2013/01/26. [DOI] [PubMed] [Google Scholar]
  • 31.Doty RL. Olfaction in Parkinson’s disease and related disorders. Neurobiology of disease. 2012;46(3):527–52. doi: 10.1016/j.nbd.2011.10.026. Epub 2011/12/24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Stern Y. Cognitive reserve in ageing and Alzheimer’s disease. Lancet neurology. 2012;11(11):1006–12. doi: 10.1016/S1474-4422(12)70191-6. Epub 2012/10/20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Jankovic J, McDermott M, Carter J, Gauthier S, Goetz C, Golbe L, et al. Variable expression of Parkinson’s disease: a base-line analysis of the DATATOP cohort. The Parkinson Study Group. Neurology. 1990;40(10):1529–34. doi: 10.1212/wnl.40.10.1529. Epub 1990/10/01. [DOI] [PubMed] [Google Scholar]
  • 34.van de Berg WD, Hepp DH, Dijkstra AA, Rozemuller JA, Berendse HW, Foncke E. Patterns of alpha-synuclein pathology in incidental cases and clinical subtypes of Parkinson’s disease. Parkinsonism & related disorders. 2012;18(Suppl 1):S28–30. doi: 10.1016/S1353-8020(11)70011-6. Epub 2011/12/23. [DOI] [PubMed] [Google Scholar]

RESOURCES