What predicts cognitive decline in de novo Parkinson's disease?

Neurobiology of Aging 33 (2012) 1127.e11–1127.e20 www.elsevier.com/locate/neuaging

What predicts cognitive decline in de novo Parkinson’s disease? Dario Arnaldia,*, Claudio Campusb, Michela Ferraraa, Francesco Famàa, Agnese Piccoa, Fabrizio De Carlic, Jennifer Accardoa, Andrea Brugnoloa, Gianmario Sambucetid, Silvia Morbellid, Flavio Nobilia,d a

Clinical Neurophysiology, Department of Neurosciences, Ophthalmology and Genetics, University of Genoa, Italy b Italian Institute of Technology (IIT), Genoa, Italy c Institute of Molecular Bioimaging and Physiology, National Research Council, Genoa, Italy d Nuclear Medicine, Department of Internal Medicine, University of Genoa, Italy Received 4 August 2011; received in revised form 8 November 2011; accepted 28 November 2011

Abstract Subtle cognitive impairment can be detected in early Parkinson’s disease (PD). In a consecutive series of de novo, drug-naive PD patients, we applied stepwise regression analysis to assess which clinical, neuropsychological, and functional neuroimaging (dopamine transporter [DAT] and perfusion single photon emission computed tomography [SPECT]) characteristics at baseline was predictive of cognitive decline during an average follow-up time of about 4 years. Decline both in executive (R2 ⫽ 0.54; p ⫽ 0.0001) and visuospatial (R2 ⫽ 0.56; p ⫽ 0.0001) functions was predicted by the couple of Unified Parkinson’s Disease Rating Scale (UPDRS)-III score and caudate dopamine transporter (DAT) uptake in the less affected hemisphere (LAH). Verbal memory and language decline was predicted instead by caudate DAT uptake and brain perfusion in a posterior parieto-temporal area of the less affected hemisphere (R2 ⫽ 0.42; p ⫽ 0.0005). No significant effect was shown for age, baseline neuropsychological scores, and levodopa equivalent dose at follow-up. The combined use of clinical structured examination and brain functional assessment by means of dual single photon emission computed tomography imaging appears as a powerful approach to predict cognitive decline in de novo PD patients. © 2012 Elsevier Inc. All rights reserved. Keywords: Parkinson’s disease; Dopamine transporter SPECT; Perfusion SPECT; Cognitive decline; Neuropsychological tests

1. Introduction Cognitive impairment is increasingly recognized in Parkinson’s disease (PD) since the early stages, when a subtle dysexecutive syndrome can be found in the majority of patients, mainly consisting of deficit in planning and cognitive flexibility (Kehagia et al., 2010). Moreover, deficit in learning, memory, and visuospatial function variably overlaps the dysexecutive syndrome and has contributed to the emerging concept of mild cognitive impairment (MCI) in PD (Caviness et al., 2007) as a condition possibly preceding * Corresponding author at: Clinical Neurophysiology, Department of Neurosciences, Ophthalmology and Genetics, University of Genoa, Via De Toni, 5, 16132 Genoa, Italy. Tel.: ⫹39 010 5554111; fax: ⫹39 010 5556893. E-mail address: [email protected] (D. Arnaldi). 0197-4580/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.neurobiolaging.2011.11.028

PD dementia (PDD), similarly to the conceptual construct adopted for Alzheimer’s disease (AD) (Aarsland et al., 2010). MCI in PD has been associated to subsequent decline in motor impairment, quality of life, and disability (Post et al., 2011) and has been shown to predict increased mortality risk (Lo et al., 2009). Thus, cognitive impairment could be an independent aspect of PD with a relevant role in determining functional outcome before the onset of PDD. Therefore, knowing the cognitive status and its evolution when PD is diagnosed for the first time could have a relevant prognostic meaning. Whereas advanced age, severity of motor disease, postural instability, and an akinetic-rigid syndrome are the main risk factors for the early onset of PDD (Aarsland et al., 2008), little is known on which signs or symptoms at disease onset are associated with mild but detectable, subsequent cognitive decline.

1127.e12

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

Few functional imaging studies have addressed this issue. An 18F-fluorodeoxyglucose-positron emission tomography (FDG-PET) study has identified a Parkinson’s disease-related metabolic cognitive pattern (PDCP) (Huang et al., 2008), significantly more expressed with the worsening of the disease (Huang et al., 2007b). This PDCP mainly includes precuneus, inferior parietal lobule, angular and lingual gyri, and superior and middle frontal gyri and significantly correlates with tests exploring executive functions, visuoperceptive abilities, memory, and language (Huang et al., 2007a). Hypometabolic PDCP expression has already been detected in the “presymptomatic” (i.e., ipsilateral) hemisphere of PD patients with hemi-parkinsonism (Tang and Eidelberg, 2010). In a single photon emission computed tomography (SPECT) study, at baseline perfusion levels in right medial frontal, left parietal, and left lenticular nucleus together with semantic and alternating word fluency, and Stroop interference index have been found to significantly predict cognitive decline after a 3-year follow-up in a group of drug-naive PD patients (Dujardin et al., 2004). Early acethylcholinesterase deficit has been shown in de novo PD patients by means of [11C]MP4A PET (Bohnen et al., 2003) but its role in subsequent decline needs to be further explored. Nigrostriatal degeneration, mainly at caudate level, and the consequent dopamine striato-frontal depletion syndrome has been repeatedly involved in cognitive impairment of PD (Cropley et al., 2006; Emre, 2003; Jokinen et al., 2009; Nobili et al., 2010; Rinne et al., 2000). Correlation between nigrocaudate dopaminergic impairment and tests exploring executive functions has been reported by means of [123I]␤-CIT or FP-CIT (N-␻-fluoro-propyl-2␤-carbomethoxy-3␤-(4-iodophenyl) nortropane) SPECT (Duchesne et al., 2002; Müller et al., 2000; Nobili et al., 2010) and PET radiopharmaceuticals, including presynaptic imaging with [18]F-DOPA (dihydroxyphenylalanine) (Rinne et al., 2000) and [11C]nomifensine (Marié et al., 1999), and postsynaptic D2 imaging with [11C]raclopride (Sawamoto et al., 2008). However, despite this evidence of a tight relationship between nigro-caudate dysfunction and executive function impairment, the role of nigrostriatal deafferentation in predicting subsequent cognitive decline at disease onset has not yet been investigated. In this study we collected baseline demographic, clinical, neuropsychological, and brain functional parameters, including dopamine transporter (DAT) activity and brain perfusion levels by means of SPECT, in a consecutive series of drug-naive patients with de novo PD. The patients were then treated with dopaminergic agents and a cognitive assessment was repeated after an average time of about 4 years. We were aimed at analyzing which parameters collected at baseline were significantly associated with cognitive decline 4 years later. 2. Methods 2.1. Patients Thirty consecutive patients with de novo PD, never treated with dopaminergic stimulation (drug-naive), were

enrolled. The diagnosis of PD followed current criteria (Gelb et al., 1999). The patients underwent brain magnetic resonance imaging (MRI), or computed tomography (CT) in the case MRI was unfeasible, to rule out other brain diseases. Patients with brain infarcts on MRI/computed tomography or with a history of stroke or transient ischemic attacks were excluded, whereas the presence of small white matter hyperintensities on MRI was not an exclusion criterion. Dementia was excluded by means of clinical interview and questionnaires for activities of daily living (ADL) (Katz et al., 1970) and instrumental ADL (Lawton and Brody, 1969). The Clinical Dementia Rating (CDR) scale was 0 in 26 patients and 0.5 in 4 patients. The Mini Mental State Examination (MMSE) was used as a measure of global cognitive function. The 15-item geriatric depression scale was administered to assess depression. Motor severity of disease was assessed by the Unified Parkinson’s Disease Rating Scale, motor section (i.e., Unified Parkinson’s Disease Rating Scale [UPDRS]-III). Patients underwent a neuropsychological test battery, including: (1) 6-trial selective reminding test (SRT) for verbal episodic memory (immediate and delayed recall); (2) categorical and phonological verbal fluency; (3) figure copying of the mental deterioration battery (simple copy and copy with guiding landmarks) to assess visuoconstructional abilities; (4) Raven’s PM (Progressive Matrices) 47, investigating logical reasoning and visuospatial functions; (5) visual search test to study sustained attention and ideomotor speed; (6) Trailmaking test (A and B, with computation of B-A score) to explore visuomotor abilities, divided attention, and attention shifting; (7) Stroop color-word test for cognitive flexibility and executive functions; (8) symbol digit test to assess executive functions and working memory; (9) Corsi’s block design to investigate spatial memory; (10) digit span (forward) assessing auditory memory span; and (11) Clock Completion test as a mixed measure of executive functions, visuospatial abilities, and memory. References for tests and normative values are listed in a previous report (Nobili et al., 2010). The 30 patients underwent DAT-SPECT, while in 4 of them brain perfusion SPECT was not feasible because of logistic reasons. DAT-SPECT scans were visually reported by nuclear medicine physicians in our group (SM and GS). Moreover, semiquantitative uptake values (normalized on background activity) in each patient were compared with those from a reference control group ranging in age between 40 and 90, embedded within the BasGan V2 software (freely available on the Italian Association of Nuclear Medicine [AIMN] web site: www.aimn.it/struttura_index.php.), taking age into account (Guerra et al., 2009). DAT-SPECT scan was impaired in all patients by both visual and semiquantitative evaluation. All of them started treatment with dopamine agonists and/or L-DOPA, according to clinical judgment of the same neurologist and were followed-up with clinical and neuro-

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

psychological examinations (mean follow-up [F-U] time: 44.9 ⫾ 18.6 months, range: 14 – 82). One patient was lost at follow-up. L-DOPA equivalent dose was computed in each patient at follow-up, according to current criteria (Tomlinson et al., 2010). Thirteen patients were taking a dopamine agonist alone, 2 patients received L-DOPA alone, and 10 patients both L-DOPA and a dopamine agonist. Eight and 2 patients received MAO (monoamine oxidase)-B inhibitors or amantadine therapy also, respectively. The L-DOPA equivalent dose ranged from 60 to 1060 mg (mean: 402.9 ⫾ 210.6). Therefore, the final study group included 25 out of the original 30 patients. The diagnosis of PD was confirmed at follow-up in all patients, also based on the good motor response to dopaminergic therapy. Clinical Dementia Rating at F-U was 0 in 19 patients and 0.5 in 6 patients. As compared with the 25 patients of the study group, the 5 excluded patients were slightly younger and showed a prevalence of males, while no significant difference was found in education, MMSE score, 15-item Geriatric Depression Scale (GDS) score, and UPDRS-III score. These differences are likely the result of chance as no other selection criteria was adopted. The main clinical and demographic characteristics are listed in Table 1. The study protocol met the approval of the local Ethics Committee and an informed consent form was signed by all participants, in compliance with the Helsinki Declaration of 1975. 2.2. Brain [123I]FP-CIT SPECT and [99mTc]ethylcysteinate dimer (ECD) SPECT [123I]FP-CIT SPECT was acquired after intravenous administration of about 185 MBq of [123I]FP-CIT (DaTSCAN, GE Healthcare, Little Chalfont, Buckinghamshire, UK). Patients also underwent brain perfusion SPECT after intravenous administration of about 740 MBq of [99mTc]-ECD (Neurolite, Bristol-Myers-Squibb, New York, NY, USA) within 1 month. Both procedures were per-

1127.e13

formed according to the European Association of Nuclear Medicine (EANM) guidelines (Tatsch et al., 2002a, 2002b). SPECT was performed using a 2-headed Millennium VG camera (GE Healthcare) equipped with low-energy, highresolution, parallel-hole collimators. Acquisition started between 180 and 240 minutes after injection of [123I]-FP-CIT and lasted 40 minutes, while it started after 30 – 60 minutes after injection of [99mTc]-ECD and lasted 30 minutes A “‘step-and-shoot’” protocol was applied with a radius of rotation ⬍ 15 cm, and 120 projections evenly spaced over 360° were generated. Total counts ranged between 2.5 and 3 million ([123I]FP-CIT) or were higher than 5 million ([99mTc]-ECD). The pixel size of the acquisition matrix was 2.4 mm, thanks to an electronic zoom (zoom factor ⫽ 1.8) applied in the data collection phase. In the reconstruction phase also a digital zoom was used and the resulting images were sampled by isotropic voxels with 2.33 mm sides. Projections were processed by means of the ordered subsets expectation maximization (OSEM) algorithm (8 iterations, 10 subsets) (Hudson and Larkin, 1994) followed by post filtering (3D Gaussian filter with full width-half maximum ⫽ 8 mm). The ordered subsets expectation maximization algorithm included a proback pair accounting for collimator blur and photon attenuation. No compensation for scatter was performed. The 2D⫹1 approximation (Boccacci et al., 1999) was applied in the simulation of the space-variant collimator blur, whereas photon attenuation was modeled with the approximation of a linear coefficient uniform inside the skull and equal to 0.11 cm⫺1. The reconstructed [123I]-FP-CIT images were exported in analyze format and processed by the automatic BasGan algorithm (Calvini et al., 2007) based on a high definition, 3D striatal template, derived from Talairach’s atlas. An optimization protocol automatically performs fine adjustments in the positioning of blurred templates to best match the radioactive counts, and locates an occipital region of interest for background evaluation. Partial volume effect (PVE) correction is included in the process of uptake computation of caudate, putamen, and background. The partial

Table 1 Main demographic and clinical characteristics (mean ⫾ SD) in the de novo PD group n Age, y (range) Gender Education, y UPDRS-III score/baseline (range) MMSE score/baseline (range) GDS score/baseline (range) Follow up, mo (range) UPDRS-III score/F-U (range) MMSE score/F-U (range) GDS score/F-U (range)

Patients

Excluded patients

25 69.0 ⫾ 5.0 (60–79) 11 M/14 F 9.6 ⫾ 4.3 12.8 ⫾ 5.1 (5–26) 28.5 ⫾ 1.7 (24–30) 3.6 ⫾ 2.5 (0–9) 44.9 ⫾ 18.6 (14–82) 16.0 ⫾ 8.9 (6–45) 27.9 ⫾ 2.0 (23–30) 3.8 ⫾ 2.9 (0–10)

5 63.4 ⫾ 4.1 (59–67) 4 M/1 F 8.8 ⫾ 5.4 15.6 ⫾ 6.3 (10–26) 28.2 ⫾ 2.5 3.2 ⫾ 3.9 — — — —

p ⬍ 0.05 ⬍ 0.05 NS NS NS NS

Five PD patients were excluded from the original series of 30 either because perfusion SPECT was unfeasible (4 instances) or lost at follow-up (1 instance). Key: F, female; F-U, follow-up; GDS, 15-item Geriatric Depression Scale; M, male; MMSE, Mini Mental State Examination; NS, not significant; PD, Parkinson’s disease; SPECT, single photon emission computed tomography; UPDRS-III, Unified Parkinson’s Disease Rating Scale motor section.

1127.e14

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

volume effect correction performed by the method (detailed in Calvini et al., 2007) consists of an activity assignment in a Talairach-Tornoux atlas-based 3-compartment model of basal ganglia. Background uptake was subtracted by putamen and caudate uptake as follows (caudate or putamen uptake ⫺ background uptake)/background uptake. An example of BasGan V2 output with basal ganglia automatic tridimensional segmentation is shown in Fig. 1. In order to highlight correlation in the more and in the less affected hemispheres, respectively, the more affected hemisphere (MAH) was defined as the contralateral one to the side of the body with prevalence of motor symptoms, such as tremor, rigidity, or motor impairment, by the referring clinician. In 17 patients the MAH was the left one, while in the remaining 8 patients it was the right one. Thus, both DAT and perfusion SPECT were flipped in these 8 patients so as to have the MAH on the left side and the less affected hemisphere (LAH) in the right side, following the procedure adopted in previous reports (Nobili et al., 2010,

2011; Tang et al., 2010). In this way we favored to highlight findings caused by the disease process while missing the information of left/right hemisphere peculiarities. While the prevalence of left hemispheric impairment in PD at onset is a recognized, though still obscure, phenomenon (Djaldetti et al., 2006), we believe that 2 balanced groups with adequate number of patients with either right or left predominant impairment would be needed to further clarify this topic. 2.3. Statistics Because of the problem of multicollinearity among the 16 neuropsychological variables, preliminary analyses were performed on the original variables to minimize multicollinearity and to reduce the number of variables for further statistical analysis. As a first step, factor analysis with varimax rotation was applied to the 16 native neuropsychological measures at baseline, linearly detrended for education, to identify those scores expressing a similar part of

Fig. 1. Output of the BasGan software version 2 (freely available in the web at the Italian Association of Nuclear Medicine [AIMN] web site: www.aimn.it/struttura_index.php.), allowing automatic segmentation and semiquantification of [123I]FP-CIT uptake in the basal ganglia, with partial volume effect correction. The patient is a 73-year-old man with prevalent impairment of left side of the body; UPDRS-III: 16; Mini Mental State Examination (MMSE): 29.

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20 Table 2 Number of patients with abnormal neuropsychological test scores at baseline and at follow-up in each of the 16 tests

NPS-EX Visual search Stroop color Stroop color-word Corsi’s span Trailmaking A Symbol digit NPS-VS Raven’s PM47 matrices Constructional apraxia, simple copy Constructional apraxia, copy with guiding landmarks NPS-VM Categorical verbal fluency Phonological verbal fluency Episodic verbal memory, immediate recall Episodic verbal memory, delayed recall NPS-MIX Digit span Trailmaking B-A Clock completion

Baseline

Follow-up

2 2 0 4 6 2

2 0 4 8 11 5

0 0 3

1 4 11

0 0 0 2

1 1 2 2

1 5 7

2 9 10

1127.e15

UCL, London, UK, http://www.fil.ion.ucl.ac.uk/spm) and taking into account age, gender, and education as “nuisance” variables. The cluster correlating with NPS-EX includes bilateral posterior cingulate cortex and precuneus in the LAH (Fig. 2a), while the cluster correlating with NPS-VM includes precuneus, inferior parietal lobule, and

Key: NPS-EX, neuropsychological executive factor; NPS-MIX, neuropsychological “mixed” factor; NPS-VM, neuropsychological verbal memory factor; NPS-VS, neuropsychological visuospatial factor.

total variance. Factor analysis identified 4 cognitive factors, as detailed in a previous report (Nobili et al., 2010), including a dysexecutive (NPS-EX), a visuospatial (NPS-VS), a verbal memory (NPS-VM), and a “mixed” (NPS-MIX) factor. The composition of these factors can be seen in Table 2 as well. The percentage of variance explained by the 4 NPS factors was 32.8% for NPS-EX, 24.3% for NPS-VS, 20.6% for NPS-VM, and 13.5% for NPS-MIX (total variance explained ⫽ 91.2%). Because the NPS-EX factor included time trials that could be affected by hypokinesia and tremor, the test scores belonging to the factor were linearly detrended for the Trailmaking Test-A, as a paradigm of time trials. Factor analysis was run again but the factor composition was unchanged and the factor loadings even increased. A conventional threshold of 0.4 was applied to factor loadings (expressing the factor-variable correlation) to individuate the group of variables mainly represented by each factor. Factorial analysis was applied to follow-up values as well, confirming the same grouping established in the analysis of baseline values. Among the DAT-SPECT variables, the uptake values at caudate and putamen level in each side, normalized on background uptake, were considered. As for brain perfusion, in a previous work 2 brain perfusion cortical clusters correlating with the NPS-EX (volumetric region of interest [VROI] NPS-EX) and the NPS-VM (VROI NPS-VM) neuropsychological factors at baseline, respectively, have been identified (Nobili et al., 2011) using SPM8 (statistical parametric mapping; Welcome Trust Centre for Neuroimaging,

Fig. 2. (a) Cluster of significant correlation (SPM8; uncorrected height threshold p ⬍ 0.005 at voxel level) between brain perfusion SPECT values and NPS-EX factor scores at baseline. The cluster includes bilateral posterior cingulate cortex and precuneus in the LAH. (b) Cluster of significant correlation (SPM8; uncorrected height threshold p ⬍ 0.005 at voxel level) between brain perfusion SPECT values and NPS-VM factor scores at baseline. The cluster includes precuneus, inferior parietal lobule and superior temporal gyrus in the LAH. These clusters were saved as volumetric regions of interest and mean perfusion values used in stepwise regression analysis. The numerical details of the original analysis leading to identification of these clusters are reported in a previous report (Nobili et al., 2011).

1127.e16

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

superior temporal gyrus in the LAH (Fig. 2b). These VROIs were saved, their mean counts were normalized on whole brain counts, and used in subsequent analyses. The R software (RDC-Team, 2008) was used for statistical analyses. As a first step, we evaluated the cognitive decline in each of the 4 NPS factors as the differences between baseline and follow-up factor scores (paired t test), corrected by the follow-up time. As a second step, stepwise forward multiple regression analysis (Faraway, 2004) (“best-fit” approach) was used to explore the relationships between cognitive decline in each of the 4 NPS factors and baseline parameters. Thus, 4 regression analyses were performed between the change in each NPS factor (i.e., ⌬NPSEX; ⌬NPS-VS; ⌬NPS-VM, and ⌬NPS-MIX) and the following baseline parameters: (1) age; (2) UPDRS-III score; (3) NPS baseline factor scores; (4) DAT-SPECT variables; and (5) 2 perfusion SPECT VROIs. Moreover, L-DOPA equivalent dose at follow-up time was included as a potential regressor of change in each NPS factor between baseline and follow-up. The “best fit” stepwise multiple regression finds the first variable as well as the best combination of 2 or more variables explaining the variance of decline in each NPS factor. At each step, the number of predictors is increased by 1. The maximum probability for a regressor to enter in the model was set to 0.05, while the minimum probability to be excluded was 0.1. The increase of the number of predictors finished when no significant (p ⬎ 0.05) improvement in the explained variance was observed. 3. Results 3.1. Neuropsychological factors All the 4 neuropsychological factor scores, corrected for the follow-up time, significantly decreased between baseline and follow-up evaluation (Table 3). Table 2 summarizes the number of patients scoring lower than ⫺1.5 standard deviation below the mean of normal reference values, according to the normative data for each test composing the factors, both at baseline and at follow-up examination. In detail, 14 (56%) and 18 (72%) patients were impaired on at least 1 neuropsychological test at baseline and at follow-up, respectively. One out of the 14 impaired patients at baseline recovered at follow-up, while 5 patients who were normal at baseline developed cognitive impairment at follow-up, all

of them showing impairment in more than 1 test (ranging from 2 to 5 tests). Among the remaining 13 patients with neuropsychological impairment at baseline, 6 were substantially unchanged (same number of impaired tests) but 7 further worsened at follow-up. Finally, in 6 instances neuropsychological tests were normal both at baseline and at follow-up. Overall, 34 (8.5%) test scores were impaired at baseline and 73 (18.2%) at follow-up. 3.2. Stepwise multiple regression analysis The best single predictor of ⌬NPS-EX was UPDRS-III score (Fig. 3a) with a slope of ⫺6.2 ⫾ 1.6 and an R2 ⫽ 0.38 (F(1,23) ⫽ 28.5; p ⫽ 0.0005). At the second step, the best couple of regressors was represented by the UPDRS-III score and DAT caudate uptake in the LAH, with a slope of ⫺5.5 ⫾ 1.6 (p ⫽ 0.002) and 35.0 ⫾ 19.0 (p ⫽ 0.04), respectively, which increased the explained variance of R2 ⫽ 0.16 (F(1,22) ⫽ 6.7; p ⫽ 0.04), leading to a total R2 ⫽ 0.54 (F(2,22) ⫽ 22.5; p ⫽ 0.0001). Adding a third variable did not lead to a significant increase in the variance explained by the model. The best single predictor of ⌬NPS-VS was DAT caudate uptake in the LAH (Fig. 3b) with a slope of 3.0 ⫾ 12.0 and an R2 ⫽ 0.26 (F(1,23) ⫽ 16.1; p ⫽ 0.005). The best couple of regressors was DAT caudate uptake in the LAH and the UPDRS-III score yielding a 0.30 increase of R2 (F(1,22) ⫽ 11.4; p ⫽ 0.02), with a regression coefficient of 26.0 ⫾ 11.0 (p ⫽ 0.02) and ⫺2.1 ⫾ 0.9 (p ⫽ 0.002), respectively, leading to a total R2 ⫽ 0.56 (F(2,22) ⫽ 16.2; p ⫽ 0.0001). No significant increase in the explained variance was found by adding further regressors. The best predictor of ⌬NPS-VM was the DAT caudate uptake in the LAH (Fig. 3c), with a regression coefficient of 35.0 ⫾ 14.0 and an R2 ⫽ 0.22 (F(1,23) ⫽ 6.4; p ⫽ 0.02). The best pair of regressors, corresponding to an R2 ⫽ 0.20 increase in the explained variance (F(1,22) ⫽ 6.2; p ⫽ 0.02) was composed by the caudate uptake in the MAH and the VROI NPS-VM, with a regression coefficient of 33.0 ⫾ 14.0 (p ⫽ 0.04) and 165.0 ⫾ 87.0 (p ⫽ 0.02), respectively, leading to a total R2 ⫽ 0.42 (F(2,22) ⫽ 5.6; p ⫽ 0.0005). Adding further regressors did not increase the variance explained. Finally, the best single predictor of ⌬NPS-MIX was the UPDRS-III score (Fig. 3d), with a slope of ⫺5.2 ⫾ 1.9 and

Table 3 Baseline neuropsychological factor scores, score decline (⌬) and follow-up time-normalized decline in 25 patients with de novo PD NPS factors

Baseline score (mean ⫾ DS)

⌬ NPS factors (mean ⫾ SE)

Follow-up time-normalized ⌬ (mean ⫾ SE)

t Test statistics

Confidence interval (95%)

P(t ⱖ 0)

NPS-EX NPS-VS NPS-VM NPS-MIX

21.3 ⫾ 37.2 15.6 ⫾ 31.8 20.2 ⫾ 46.4 26.4 ⫾ 64.2

⫺50.2 ⫾ 3.6 ⫺36.6 ⫾ 1.9 ⫺45.3 ⫾ 2.2 ⫺60.9 ⫾ 3.7

⫺1.34 ⫾ 0.1 ⫺0.87 ⫾ 0.05 ⫺1.01 ⫾ 0.06 ⫺1.49 ⫾ 0.09

⫺2.410 ⫺3.036 ⫺2.601 ⫺3.707

⫺1.916 to 0.148 ⫺1.062 to 0.202 ⫺0.129 to 1.119 ⫺1.761 to 0.301

0.012 0.003 0.0005 0.004

Key: ⌬, difference between baseline and follow-up neuropsychological factor scores; NPS-EX, neuropsychological executive factor; NPS-MIX, neuropsychological “mixed” factor; NPS-VM, neuropsychological verbal memory factor; NPS-VS, neuropsychological visuospatial factor.

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

1127.e17

Fig. 3. Linear regression of (a) ⌬NPS-EX and the Unified Parkinson’s Disease Rating Scale (UPDRS)-III score (R2 ⫽ 0.38; F(1,23) ⫽ 28.5, p ⫽ 0.0005); (b) ⌬NPS-VS and dopamine transporter (DAT) caudate uptake in the less affected hemisphere (LAH) (R2 ⫽ 0.26; F(1,23) ⫽ 16.1; p ⫽ 0.005); (c) ⌬NPS-VM and DAT caudate uptake in the LAH (R2 ⫽ 0.22; F(1,23) ⫽ 6.4; p ⫽ 0.02); (d) ⌬NPS-MIX and the UPDRS-III score (R2 ⫽ 0.25; F(1,23) ⫽ 7.6, p ⫽ 0.01). ⌬ is normalized to the follow-up time (⌬/TFU). The dark line represents the best fit.

an R2 ⫽ 0.25 (F(1,23) ⫽ 7.6; p ⫽ 0.01), while the best pair of regressors was the UPDRS-III score and the DAT caudate uptake in the LAH, providing an R2 increase of 0.26 (F(1,22) ⫽ 9.2, p ⫽ 0.02) with coefficients of ⫺4.2 ⫾ 1.8 (p ⫽ 0.02) and 46.0 ⫾ 22.0 (p ⫽ 0.02), respectively, leading to a total R2 ⫽ 0.51 (F(2,22) ⫽ 12.7; p ⫽ 0.0001). Adding further regressors did not increase the variance explained. 4. Discussion In this study we observed the cognitive decline in 25 drug naive, de novo PD patients over a mean follow-up period of 4 years. All of the neuropsychological factors showed a significant decline from baseline values. Verbal tasks, including verbal memory and verbal fluency tests showed the most significant decline, followed by visuospatial functions and by sparse tests including shifting attention and working memory tests, and finally by executive functions. Decline in executive functions survived detrend correction for motor speed, as derived from the Trailmaking A test score. All of the patients underwent chronic dopaminergic stimulation but this did not result in a significant effect on cognitive decline between baseline and follow-up, in keeping with the view that memory, language, and visuoconstruction are less likely to be improved by dopaminergic stimulation (Kehagia et al., 2010) and with longitudinal studies evidence of cognitive worsening after 3–5 years of follow-up, despite dopaminergic therapy (Muslimovic´ et al., 2009; Williams-Gray et al., 2007). Fourteen patients showed impairment in at least 1 cognitive test at baseline. This number grew to 18 at follow-up examination, in keeping with the notion that de novo PD patients are twice as

likely to develop MCI than are healthy elderly individuals (Kehagia et al., 2010). The motor severity of the disease and caudate DAT uptake in the LAH were the best couple of regressors predicting cognitive decline in all neuropsychological factors, but verbal memory/language. These data show that assessing the motor severity of the disease by means of a simple clinical structured scale is a valuable tool to predict not just worsening of motor disease (Ferguson et al., 2008; Roos et al., 1996; Zhao et al., 2010) but also subsequent decline in executive and visuospatial functions. This finding is in line with those studies showing that in patients at different stages of the disease the UPDRS-III score is related to the progression to dementia (Hughes et al., 2000; Levy et al., 2002). Here we have shown that the UPDRS-III score is predictive of cognitive decline prior of dementia onset (no patient in this group developed dementia at follow-up) and since the time of diagnosis. On the contrary, while neuropsychological test scores were reported to predict the onset of dementia (Levy et al., 2002), we did not find any significant predictive ability of baseline neuropsychological factor scores toward subsequent cognitive decline and prior of dementia onset. This negative and somewhat unexpected finding is more difficult to be explained and might be interpreted by taking into account that the worsening of the disease is not linear and not homogeneous among patients (Alves et al., 2005). More intriguing from a pathophysiological standpoint is the predicting role of DAT depletion at the level of caudate nucleus in the LAH. First, this finding confirms previous data pointing to the nigro-caudate rather than to nigro-putaminal do-

1127.e18

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

paminergic deafferentation as having a role in cognitive dysfunction (Grahn et al., 2008; Nobili et al., 2010; Redgrave et al., 2010; Rinne et al., 2000). Second, the DAT caudate uptake in the LAH (much more than in the MAH) predicted the decline. SPECT and PET studies have shown that caudate nucleus in the LAH is the last basal ganglia involved by nigrostriatal degeneration, thus meaning that it is the most preserved in de novo PD, while the caudate of the MAH is already more affected (Dauer and Przedborski, 2003; Filippi et al., 2005; Grahn et al., 2008). Thus, DAT uptake in the LAH may express better the sample variability and explain a larger part of variance. Indeed, the caudate uptake values showed a larger distribution in the LAH (range: 1.64 – 4.37) than in the MAH (range: 1.52–3.84). Both nigrostriatal dopaminergic activity and cortical function have been increasingly evaluated in de novo PD, showing correlation with cognitive status in cross-sectional studies. Nigrostriatal degeneration, mainly at the caudate level, has been correlated with several tests assessing cognitive—mainly executive-functions, such as the Stroop colorword test (Rinne et al., 2000) and the object alternation test (Marié et al., 1999). The PDCP has been recently evaluated in the LAH with the aim to assess cortical metabolism in a presymptomatic hemisphere. It has been shown that PDCP correlates with nigrostriatal denervation, is already expressed in the LAH at the time of clinical diagnosis and is detectable at least 2 years before the onset of symptoms in the unaffected side of the body (Tang et al., 2010). Finally, decline in verbal memory and verbal fluency (here grouped together into the same NPS-VM factor) was not predicted by the motor severity of the disease but by DAT uptake (in the LAH or in the MAH, according to subsequent steps of stepwise regression analysis) and by brain perfusion in posterior parieto-temporal cortex of the LAH. This VROI derives from a previous cross-sectional study where we showed virtually in the same patient group the significant correlation between NPS-VM factor and brain perfusion level in this region of the LAH, including the precuneus, the inferior parietal lobule, and the superior temporal gyrus (Nobili et al., 2011). Moreover, posterior associative cortical hypoperfusion has been shown in PD patients with the amnestic form of MCI (Nobili et al., 2009). This finding is especially intriguing because on one hand it shows that nigro-caudate dopaminergic impairment is involved in memory and language decline, and on the other hand it highlights the role of early posterior cortical dysfunction, probably not based on dopaminergic but rather on cholinergic deficit (Bohnen and Albin, 2011; Kehagia et al., 2010). Effectively, cholinergic deficit has been increasingly reported in PD since the early stages and is suggested to produce the posterior association cortical dysfunction similarly, and even more severely, to Alzheimer’s disease (Bohnen et al., 2003). The meaning of our findings could be that both neurotransmitter systems are involved in worsen-

ing of memory and language, in keeping with the identification of a PDCP hypometabolic pattern, including both frontal and parietal associative regions (Eidelberg, 2009; Tang and Eidelberg, 2010). Despite NPS-VM factor disclosed the highest statistically significant decline between baseline and follow-up, just a minority of patients scored lower than normal threshold (see Table 2). This was due to the relatively high scores at baseline, because then about half of the patients worsened in each of the 3 tests composing the factor, while the rest were mostly unchanged and only a few improved. No significant effect of L-DOPA equivalent dose on cognitive decline between baseline and follow-up was found. Dopaminergic chronic stimulation may ameliorate some aspects of cognition, such as cognitive flexibility and working memory (Cools et al., 2003; Costa et al., 2009; Lewis et al., 2005), likely by improving functioning within the frontostriatal dopamine pathways, but might have deleterious effects on other cognitive domains, such as some aspects of learning (Cools et al., 2006; Frank and Claus, 2006) and attention (Swainson et al., 2006). Other aspects of cognition, mainly including visuospatial function (Lee et al., 1998), visual recognition memory, conditional associative learning, and verbal memory seem to be dopamineindependent and unaffected by medication status (Kehagia et al., 2010), especially with increasing duration of disease (Owen et al., 1993; Sahakian et al., 1988). In some clinicalbased neuropsychological studies, dopaminergic stimulation had no significant effect on routine neuropsychological tests after 6 months of chronic treatment (Relja and Klepac, 2006). Moreover, our patients were reassessed after a mean time of about 4 years and hence it is likely that even an expected positive effect on some tests has been overcome by worsening of disease, which is in keeping with previous longitudinal studies with similar follow-up time (Muslimovic´ et al., 2009; Williams-Gray et al., 2007). A limitation of this study is that the mean age at onset of PD (69.0; see age in Table 1) was a bit higher than in the general PD population which may have influenced the results to some extent (for instance, the percentage of patients with at least 1 test below the normal limit was rather high, i.e., 56% at baseline and 72% after 4 years). This bias could be the result of chance, because no criteria other than consecutive enrollment were adopted. Also, living in a very old urban area might have affected recruitment. Although age has been shown to have no significant effect in the stepwise regression model in this sample, the present data might not be applicable to the younger general PD population. Another limitation is that the follow-up time was not homogeneous, ranging from little more than 1 year to almost 7 years. Such a source of inhomogeneity may have affected the results, although the analysis was fully normalized for the follow-up time. In conclusion, our results show that cognitive decline after an average time of 4 years can be previewed in de novo

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

PD patients, mainly on the basis of the UPDRS-III score and on the DAT uptake in the caudate nucleus of the less affected hemisphere. Exceptions were verbal memory and verbal fluency that were predicted by DAT caudate uptake and by cortical perfusion level in a posterior parieto-temporal cortical area, but not by the UPDRS-III score. On the contrary, neither baseline neuropsychological factor values nor age were useful to this purpose. Automatic semiquantitative measures of DAT caudate uptake and brain perfusion by means of SPECT/PET could be an objective tool to support clinicians in previewing cognitive decline and in planning intervention. Disclosure statement The authors disclose no conflicts of interest. The study protocol met the approval of the local Ethics Committee and an informed consent form was signed by all participants, in compliance with the Helsinki Declaration of 1975. References RDC-Team, 2008. R: A Language and Environment for Statistical Computing R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org. Aarsland, D., Beyer, M.K., Kurz, M.W., 2008. Dementia in Parkinson’s disease. Curr. Opin. Neurol. 21, 676 – 682. Aarsland, D., Bronnick, K., Williams-Gray, C., Weintraub, D., Marder, K., Kulisevsky, J., Burn, D., Barone, P., Pagonabarraga, J., Allcock, L., Santangelo, G., Foltynie, T., Janvin, C., Larsen, J.P., Barker, R.A., Emre, M., 2010. Mild cognitive impairment in Parkinson disease: a multicenter pooled analysis. Neurology 75, 1062–1069. Alves, G., Wentzel-Larsen, T., Aarsland, D., Larsen, J.P., 2005. Progression of motor impairment and disability in Parkinson disease: a population-based study. Neurology 65, 1436 –1441. Boccacci, P., Bonetto, P., Calvini, P., Formiconi, A.R., 1999. A simple model for the efficient correction of collimator blur in 3D SPECT imaging. Inverse Probl. 15, 907–930. Bohnen, N.I., Albin, R.L., 2011. The cholinergic system and Parkinson disease. Behav. Brain Res. 221, 564 –573. Bohnen, N.I., Kaufer, D.I., Ivanco, L.S., Lopresti, B., Koeppe, R.A., Davis, J.G., Mathis, C.A., Moore, R.Y., DeKosky, S.T., 2003. Cortical cholinergic function is more severely affected in parkinsonian dementia than in Alzheimer disease: an in vivo positron emission tomographic study. Arch. Neurol. 60, 1745–1748. Calvini, P., Rodriguez, G., Inguglia, F., Mignone, A., Guerra, U.P., Nobili, F., 2007. The basal ganglia matching tools package for striatal uptake semi-quantification: description and validation. Eur. J. Nucl. Med. Mol. Imaging 34, 1240 –1253. Caviness, J.N., Driver-Dunckley, E., Connor, D.J., Sabbagh, M.N., Hentz, J.G., Noble, B., Evidente, V.G., Shill, H.A., Adler, C.H., 2007. Defining mild cognitive impairment in Parkinson’s disease. Mov. Disord. 22, 1272–1277. Cools, R., Altamirano, L., D’Esposito, M., 2006. Reversal learning in Parkinson’s disease depends on medication status and outcome valence. Neuropsychologia 44, 1663–1673. Cools, R., Barker, R.A., Sahakian, B.J., Robbins, T.W., 2003. L-Dopa medication remediates cognitive inflexibility, but increases impulsivity in patients with Parkinson’s disease. Neuropsychologia 41, 1431–1441. Costa, A., Peppe, A., Dell’Agnello, G., Caltagirone, C., Carlesimo, G.A., 2009. Dopamine and cognitive functioning in de novo subjects with

1127.e19

Parkinson’s disease: effects of pramipexole and pergolide on working memory. Neuropsychologia 47, 1374 –1381. Cropley, V.L., Fujita, M., Innis, R.B., Nathan, P.J., 2006. Molecular imaging of the dopaminergic system and its association with human cognitive function. Biol. Psychiatry 59, 898 –907. Dauer, W., Przedborski, S., 2003. Parkinson’s disease: mechanisms and models. Neuron 39, 889 –909. Djaldetti, R., Ziv, I., Melamed, E., 2006. The mystery of motor asymmetry in Parkinson’s disease. Lancet Neurol. 5, 796 – 802. Duchesne, N., Soucy, J.P., Masson, H., Chouinard, S., Bédard, M.A., 2002. Cognitive deficits and striatal dopaminergic denervation in Parkinson’s disease: a single photon emission computed tomography study using 123iodine-beta-CIT in patients on and off levodopa. Clin. Neuropharmacol. 25, 216 –224. Dujardin, K., Defebvre, L., Duhamel, A., Lecouffe, P., Rogelet, P., Steinling, M., Destée, A., 2004. Cognitive and SPECT characteristics predict progression of Parkinson’s disease in newly diagnosed patients. J. Neurol. 251, 1383–1392. Eidelberg, D., 2009. Metabolic brain networks in neurodegenerative disorders: a functional imaging approach. Trends Neurosci. 32, 548 –557. Emre, M., 2003. What causes mental dysfunction in Parkinson’s disease? Mov. Disord. 18(suppl 6), S63–S71. Faraway, J., 2004. Linear models with R. Boca Raton, FL, USA: CRC Press. Ferguson, L.W., Rajput, M.L., Muhajarine, N., Shah, S.M., Rajput, A., 2008. Clinical features at first visit and rapid disease progression in Parkinson’s disease. Parkinsonism Relat. Disord. 14, 431– 435. Filippi, L., Manni, C., Pierantozzi, M., Brusa, L., Danieli, R., Stanzione, P., Schillaci, O., 2005. 123I-FP-CIT semi-quantitative SPECT detects preclinical bilateral dopaminergic deficit in early Parkinson’s disease with unilateral symptoms. Nucl. Med. Commun. 26, 421– 426. Frank, M.J., Claus, E.D., 2006. Anatomy of a decision: striato-orbitofrontal interactions in reinforcement learning, decision making, and reversal. Psychol. Rev. 113, 300 –326. Gelb, D.J., Oliver, E., Gilman, S., 1999. Diagnostic criteria for Parkinson disease. Arch. Neurol. 56, 33–39. Grahn, J.A., Parkinson, J.A., Owen, A.M., 2008. The cognitive functions of the caudate nucleus. Prog. Neurobiol. 86, 141–155. Guerra, U.P., Poli, G.L., Andreoletti, M., Nobili, F., Inguglia, F., Morbelli, S., Sestini, S., Mazzeo, C., Castagnoli, A., Calcagni, M.L., Varrone, A., 2009. Creation of a normal database for striatal uptake semi-quantification of 123I–FP-CIT SPECT. Q. J. Nucl. Med. Mol. Imaging 53 (suppl 1), 103. Huang, C., Mattis, P., Perrine, K., Brown, N., Dhawan, V., Eidelberg, D., 2008. Metabolic abnormalities associated with mild cognitive impairment in Parkinson disease. Neurology 70, 1470 –1477. Huang, C., Mattis, P., Tang, C., Perrine, K., Carbon, M., Eidelberg, D., 2007a. Metabolic brain networks associated with cognitive function in Parkinson’s disease. Neuroimage 34, 714 –723. Huang, C., Tang, C., Feigin, A., Lesser, M., Ma, Y., Pourfar, M., Dhawan, V., Eidelberg, D., 2007b. Changes in network activity with the progression of Parkinson’s disease. Brain 130, 1834 –1846. Hudson, H.M., Larkin, R.S., 1994. Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans. Med. Imaging 13, 601– 609. Hughes, T.A., Ross, H.F., Musa, S., Bhattacherjee, S., Nathan, R.N., Mindham, R.H., Spokes, E.G., 2000. A 10-year study of the incidence of and factors predicting dementia in Parkinson’s disease. Neurology 54, 1596 –1602. Jokinen, P., Brück, A., Aalto, S., Forsback, S., Parkkola, R., Rinne, J.O., 2009. Impaired cognitive performance in Parkinson’s disease is related to caudate dopaminergic hypofunction and hippocampal atrophy. Parkinsonism Relat. Disord. 15, 88 –93. Katz, S., Downs, T.D., Cash, H.R., Grotz, R.C., 1970. Progress in development of the index of ADL. Gerontologist 10, 20 –30.

1127.e20

D. Arnaldi et al. / Neurobiology of Aging 33 (2012) 1127.e11–1127.e20

Kehagia, A.A., Barker, R.A., Robbins, T.W., 2010. Neuropsychological and clinical heterogeneity of cognitive impairment and dementia in patients with Parkinson’s disease. Lancet Neurol. 9, 1200 –1213. Lawton, M.P., Brody, E.M., 1969. Assessment of older people: selfmaintaining and instrumental activities of daily living. Gerontologist 9, 179 –186. Lee, A.C., Harris, J.P., Calvert, J.E., 1998. Impairments of mental rotation in Parkinson’s disease. Neuropsychologia 36, 109 –114. Levy, G., Jacobs, D.M., Tang, M.X., Côté, L.J., Louis, E.D., Alfaro, B., Mejia, H., Stern, Y., Marder, K., 2002. Memory and executive function impairment predict dementia in Parkinson’s disease. Mov. Disord. 17, 1221–1226. Lewis, S.J., Slabosz, A., Robbins, T.W., Barker, R.A., Owen, A.M., 2005. Dopaminergic basis for deficits in working memory but not attentional set-shifting in Parkinson’s disease. Neuropsychologia 43, 823– 832. Lo, R.Y., Tanner, C.M., Albers, K.B., Leimpeter, A.D., Fross, R.D., Bernstein, A.L., McGuire, V., Quesenberry, C.P., Nelson, L.M., Van Den Eeden, S.K., 2009. Clinical features in early Parkinson disease and survival. Arch. Neurol. 66, 1353–1358. Marié, R.M., Barré, L., Dupuy, B., Viader, F., Defer, G., Baron, J.C., 1999. Relationships between striatal dopamine denervation and frontal executive tests in Parkinson’s disease. Neurosci. Lett. 260, 77– 80. Müller, U., Wächter, T., Barthel, H., Reuter, M., von Cramon, D.Y., 2000. Striatal [123I]beta-CIT SPECT and prefrontal cognitive functions in Parkinson’s disease. J. Neural Transm. 107, 303–319. Muslimovic´, D., Post, B., Speelman, J.D., De Haan, R.J., Schmand, B., 2009. Cognitive decline in Parkinson’s disease: a prospective longitudinal study. J. Int. Neuropsychol. Soc. 15, 426 – 437. Nobili, F., Campus, C., Arnaldi, D., De Carli, F., Cabassi, G., Brugnolo, A., Dessi, B., Morbelli, S., Sambuceti, G., Abbruzzese, G., Rodriguez, G., 2010. Cognitive-nigrostriatal relationships in de novo, drug-naive Parkinson’s disease patients: a [I-123]FP-CIT SPECT study. Mov. Disord. 25, 35– 43. Nobili, F., Abbruzzese, G., Morbelli, S., Marchese, R., Girtler, N., Dessi, B., Brugnolo, A., Canepa, C., Drosos, G.C., Sambuceti, G., Rodriguez, G., 2009. Amnestic mild cognitive impairment in Parkinson’s disease: a brain perfusion SPECT study. Mov. Disord. 24, 414 – 421. Nobili, F., Arnaldi, D., Campus, C., Ferrara, M., De Carli, F., Brugnolo, A., Dessi, B., Girtler, N., Morbelli, S., Abruzzese, G., Sambuceti, G., Rodriguez, G., 2011. Brain perfusion correlates of cognitive and nigrostriatal functions in de novo Parkinson’s disease. Eur. J. Nucl. Med. Mol. Imaging 38, 2209 –2218. Owen, A.M., Beksinska, M., James, M., Leigh, P.N., Summers, B.A., Marsden, C.D., Quinn, N.P., Sahakian, B.J., Robbins, T.W., 1993. Visuospatial memory deficits at different stages of Parkinson’s disease. Neuropsychologia 31, 627– 644. Post, B., Muslimovic, D., van Geloven, N., Speelman, J.D., Schmand, B., de Haan, R.J., CARPA-study group, 2011. Progression and prognostic factors of motor impairment, disability and quality of life in newly diagnosed Parkinson’s disease. Mov. Disord. 26, 449 – 456.

Redgrave, P., Rodriguez, M., Smith, Y., Rodriguez-Oroz, M.C., Lehericy, S., Bergman, H., Agid, Y., DeLong, M.R., Obeso, J.A., 2010. Goaldirected and habitual control in the basal ganglia: implications for Parkinson’s disease. Nat. Rev. Neurosci. 11, 760 –772. Relja, M., Klepac, N., 2006. A dopamine agonist, pramipexole, and cognitive functions in Parkinson’s disease. J. Neurol. Sci. 248, 251–254. Rinne, J.O., Portin, R., Ruottinen, H., Nurmi, E., Bergman, J., Haaparanta, M., Solin, O., 2000. Cognitive impairment and the brain dopaminergic system in Parkinson disease: [18F]fluorodopa positron emission tomographic study. Arch. Neurol. 57, 470 – 475. Roos, R.A., Jongen, J.C., van der Velde, E.A., 1996. Clinical course of patients with idiopathic Parkinson’s disease. Mov. Disord. 11, 236 – 242. Sahakian, B.J., Morris, R.G., Evenden, J.L., Heald, A., Levy, R., Philpot, M., Robbins, T.W., 1988. A comparative study of visuospatial memory and learning in Alzheimer-type dementia and Parkinson’s disease. Brain J. Neurol. 111, 695–718. Sawamoto, N., Piccini, P., Hotton, G., Pavese, N., Thielemans, K., Brooks, D.J., 2008. Cognitive deficits and striato-frontal dopamine release in Parkinson’s disease. Brain 131, 1294 –1302. Swainson, R., SenGupta, D., Shetty, T., Watkins, L.H., Summers, B.A., Sahakian, B.J., Polkey, C.E., Barker, R.A., Robbins, T.W., 2006. Impaired dimensional selection but intact use of reward feedback during visual discrimination learning in Parkinson’s disease. Neuropsychologia 44, 1290 –1304. Tang, C.C., Eidelberg, D., 2010. Abnormal metabolic brain networks in Parkinson’s disease from blackboard to bedside. Prog. Brain Res. 184, 161–176. Tang, C.C., Poston, K.L., Dhawan, V., Eidelberg, D., 2010. Abnormalities in metabolic network activity precede the onset of motor symptoms in Parkinson’s disease. J. Neurosci. 30, 1049 –1056. Tatsch, K., Asenbaum, S., Bartenstein, P., Catafau, A., Halldin, C., Pilowsky, L.S., Pupi, A., 2002a. European Association of Nuclear Medicine procedure guidelines for brain neurotransmission SPET using (123)I-labelled dopamine D(2) transporter ligands. Eur. J. Nucl. Med. Mol. Imaging 29, BP30 –BP35. Tatsch, K., Asenbaum, S., Bartenstein, P., Catafau, A., Halldin, C., Pilowsky, L.S., Pupi, A., 2002b. European Association of Nuclear Medicine procedure guidelines for brain perfusion SPET using (99m)Tclabelled radiopharmaceuticals. Eur. J. Nucl. Med. Mol. Imaging 29, BP36-42. Tomlinson, C.L., Stowe, R., Patel, S., Rick, C., Gray, R., Clarke, C.E., 2010. Systematic review of levodopa dose equivalency reporting in Parkinson’s disease. Mov. Disord. 25, 2649 –2653. Williams-Gray, C.H., Foltynie, T., Brayne, C.E., Robbins, T.W., Barker, R.A., 2007. Evolution of cognitive dysfunction in an incident Parkinson’s disease cohort. Brain J. Neurol. 130, 1787–1798. Zhao, Y.J., Wee, H.L., Chan, Y.H., Seah, S.H., Au, W.L., Lau, P.N., Pica, E.C., Li, S.C., Luo, N., Tan, L.C., 2010. Progression of Parkinson’s disease as evaluated by Hoehn and Yahr stage transition times. Mov. Disord. 25, 702–708.