Cognitive and motor dysfunction in the early phase of Parkinson’s disease

Magdalena Eriksson Domellöf

Department of Pharmacology and Clinical Neuroscience

Department of Community Medicine and Rehabilitation

Umeå University, Sweden

Responsible publisher under swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7459-767-7

ISSN: 0346-6612

Elektronisk version tillgänglig på http://umu.diva-portal.org/

Tryck/Printed by: Print och Media, Umeå University, Umeå, Sweden 2013

Cover photo: Erik Domellöf

To Erik, Edith and Karin

I

Table of Contents Table of Contents I Abstract IV Abbreviations VI Svensk sammanfattning (Summary in Swedish) VII Original papers IX Introduction 1

Parkinson’s disease 1 Definition and diagnosis 1 Symptoms 2

Cardinal signs 2 Other motor symptoms 3 Non motor symptoms 3

Epidemiology 4 Histopathology and etiology 4 Pathology, biochemistry and physiology 5 Treatment of motor symptoms 6

Levodopa 6 Decarboxylase and COMT inhibitors 6 Dopamine agonist and MAO-B inhibitors 6 Treatment in later disease stages 7

Cognitive impairment 7 Cognitive impairment in PD 7

Executive functions 7 Memory 8 Attention 9 Visuospatial skills 9 Language function 9 PD-MCI 9 PDD 10

Epidemiology of cognitive impairment in PD 10 Risk factors for cognitive impairment in PD 12 Neuropathology of cognitive impairment in PDD 12 Genes and cognition in PD 13 Treatment of cognitive impairment and other non-motor features in PD 14 Cognitive motor relationship 15 Rationale of the thesis 15 Aims 17

Materials and methods 18 Study population 19 Non-participation 19 Assessment tools 23

II

Assessment of motor function 23 Assessment of cognitive function 23 Diagnosis 25 PD 26 Cognitive impairment 27 MCI 28 Dementia 28 Ethics 28 Statistics 29 Empirical studies 29 Paper I 29 Paper II 30 Paper III 31 Paper IV 33

Results 35 Paper I 35 Paper II 36 Paper III 36 Paper IV 38

Discussion 40 Main findings 40 Impaired cognitive functions 41 Proportions of MCI and Dementia 42 MCI 42 Dementia 42 Higher dementia and MCI rates 42 Predictive value of MDS-Task Force MCI criteria for PDD 43 Education, age, disease duration and gender aspects 44 Cognitive motor relationship 44 The association of bradykinesia with impaired working memory and executive

function 45 The association of posture and gait disturbances with impaired visuospatial

function and visuospatial memory 45 The effect of dopaminergic medication 47 Ethics 48 Methodological considerations 48 Study design 48 Loss to follow up and missing values 49 Statistics 49 Diagnosis of PD 50 UPDRS 50 SWEDDs 50 MCI and Dementia 50

III

Motor fluctuations 51 Neuropsychological testing 51 Control group 52 Future implications 52

Conclusions 54 Acknowledgements 55 References 57

IV

Abstract

Background: Parkinson’s disease (PD) is a chronic and progressive

neurodegenerative disease. The diagnosis is based on a combination of the

motor signs: tremor, bradykinesia, rigidity and postural abnormalities. Mild

Cognitive Impairment (MCI) is common early in the disease and a large

proportion of patients with PD develop dementia (PDD). Associations

between motor symptoms and cognitive decline have been suggested but the

results are inconclusive due to differences in the selection of participants and

variables tested. Large population based studies with comprehensive

neuropsychological investigation in newly diagnosed cases with PD followed

prospectively are rare. The aim of this thesis was to improve characterization

and understanding of cognition in PD, and to explore the relationship to

motor impairment in the early phase of PD. Methods: All new patients with

suspected idiopathic parkinsonism in the catchment area (142 ooo

inhabitants) were examined during a period of five years and four months.

Among other investigations, a comprehensive neuropsychological evaluation

was carried out in 119 of 148 patients with PD together with 30 age matched

healthy controls. Assessments were repeated after one three and five years.

Results: Patients performed worse than healthy controls in a majority of

neuropsychological tests. MCI at the time of diagnosis were found in 36%

according to recently published MCI criteria. Thirty% were cognitively

impaired using another definition. One fourth of the patients developed PDD

within five years after diagnosis and 25 % of those with MCI at baseline

reversed back to normal cognition. Age and MCI were significant predictors

of dementia. Education was an independent predictor for severe cognitive

dysfunction at diagnosis but did not predict PDD. Patients with MCI

converting to PDD had worse performance on visuospatial function,

semantic fluency, episodic memory, mental flexibility and conceptual

thinking. There were no differences in cognitive performance between

patients with predominant Postural and Gait Disturbances (PIGD) and the

tremor dominant subtype at the baseline investigation and belonging to the

PIGD subgroup at baseline did not predict PDD. Dementia converters

declined more rapidly than non-converters in posture/gait function.

Associations between bradykinesia and measures of executive functions and

working memory were found, and between posture and gait disturbances

and visuospatial function. Some of these associations were persistent after

one year. Patients receiving the dopamine agonist pramipexole performed

significantly worse on a measure of verbal fluency at the one year follow up.

Conclusions: The differences in proportions of cognitively impaired in the

different studies emphasize the value of joint criteria for PD-MCI. Even

when using such criteria, a substantial proportion of patients revert back to

V

normal function. The increase in motor disability in patients with PDD could

have several different causes that need to be further investigated. Associated

motor and cognitive dysfunctions could reflect common pathophysiological

processes in partly shared networks. Both dopaminergic and non-

dopaminergic motor and cognitive functions seems to be involved in PDD

which suggests that pharmacological treatment in PD needs to go beyond the

scope of dopaminergic deficiency in search for new therapies that would also

be effective for non-motor symptoms.

VI

Abbreviations

α-synuclein Alpha-synuclein

ß-amyloid Beta-amyloid

BBB Blood Brain Barrier

BVMT Brief Visuospatial Memory Test

BNT Boston Naming Test

COWAT Controlled Oral Word Assessment Tool

CBD Corticobasal Degeneration

DA Dopamine agonists

DBL Dementia with Levy Bodies

FCSRT Free and Cued Selective Reminding Test

fMRI Functional Magnetic Resonance Imaging

FP-CIT 123I-N (omega)-flouropropyl-2-ß-carbomethoxyl-3-ß-(4-

iodophenyl) nortropane

H&Y Hoehn & Yahr

ID Indeterminate

MADRS Montgomery-Åsberg Depression Rating Scale

MCI Mild Cognitive Impairment

MDS Movement Disorder Society

MMSE Mini-Mental State Examination

MSA Multiple System Atrophy

NYPUM Ny (New)-Parkinson Umeå

NPV Negative Predictive Value

PD Parkinson’s Disease

PDD Parkinson’s Disease Dementia

PIGD Postural Instability and Gait Disturbances

PPV Positive Predictive Value

PSP Progressive Supranuclear Palsy

UPDRS Unified Parkinson’s Disease Rating Scale

SD Standard Deviation

SE Standard Error

VII

Svensk sammanfattning (Summary in Swedish)

Parkinsons sjukdom (PS) är en kronisk och progressiv neurodegenerativ

sjukdom. Den kliniska diagnosen bygger på en kombination av motoriska

symptom: skakningar (tremor), rörelsehämning (bradykinesi), muskelstelhet

(rigiditet) samt balans och gångsvårigheter. Förekomsten av PS ökar med

ålder. Hos individer över 60 år är PS den vanligaste neurodegenerativa

sjukdomen efter Alzheimers sjukdom.

Kognitiva nedsättningar är vanliga redan vid tidig fas av PS och en stor del

utvecklar Parkinson demens. Kognitiva nedsättningar vid PS ger alvarliga

konsekvenser med ökad dödlighet och ökat beroende samt ökad stress för

vårdgivare Det är särskilt viktigt med en bättre beskrivning och förståelse för

kognitiva nedsättningar i tidig fas av sjukdomen eftersom det är då

interventioner med neuroprotektiva mediciner eller andra behandlingar

troligen ger bäst resultat. Att hitta specifika relationer mellan olika typer av

motoriska problem och kognitiva nedsättningar skulle kunna bidra till nya

idéer angående de underliggande patofysiologiska mekanismerna för

motoriska och kognitiva problem vilket kan ge en ökad förståelse för de

underliggande orsakerna.

De flesta tidigare studier som har undersökt kognitiv funktion vid PS har

varit tvärsnittsstudier och/eller inkluderat patienter i olika skeden av

sjukdomen. Associationer mellan motorproblem och kognitiva nedsättningar

har föreslagits men resultaten är även där ofullständiga på grund av

olikheter i de studerade patientgrupperna och vilka variabler som studerats.

Stora populationsbaserade studier av kognitiva problem vid PS med

omfattande neuropsykologisk utredning följda över tid är få. Därför var

målet med den aktuella avhandlingen att förbättra förståelsen av kognitiva

nedsättningar vid PS samt utforska relationen mellan kognition och motorik

i sjukdomens tidiga skede.

Detta gjordes genom en populationsbaserad cohort där målet var att

undersöka alla med misstänkt parkinsonism i upptagningsområdet för

Norrlands Universitets sjukhus. Alla inkluderade patienter genomgick

mängder av undersökningar mellan 1 januari 2004 och 31 april 2009, där

119 av 148 patienter med PS tillsammans med 30 friska kontroller

genomgick omfattande neuropsykologisk utredning vid studiens början samt

efter ett, tre och fem år.

I enlighet med tidigare studier presterade patienter med PS sämre än friska

kontroller på mängder av neuropsykologiska test. Andelen som

VIII

klassificerades som kognitivt nedsatta varierade mellan studierna mycket på

grund av att olika kriterier användes: 36 % när nyligen publicerade kriterier

anpassade till PS användes samt 30 % när en annan definition användes. En

fjärdedel av ursprungsgruppen utvecklade demens under uppföljningstiden

medan en fjärdedel av de med kognitiv nedsättning vid första mättillfället

återgick till normal kognition under uppföljningstiden. Ålder och MCI var

starka prediktorer för demens. Utbildningsnivå var starkt kopplat till

kognitiv nedsättning vid första mättillfället men predicerade inte senare

utveckling av demens. Patienter med kognitiv nedsättning som senare

utvecklade demens presterade sämre än de med kognitiv nedsättning som

inte utvecklade demens på flera olika kognitiva tester.

Det var inte någon skillnad i kognitiv funktion mellan patienter med

övervägande posturala problem och gång svårigheter (PIGD) jämfört med de

med övervägande Tremor. Att ha övervägande PIGD problem vid första

mättillfället predicerade inte utvecklingen av demens. Patienter som

utvecklade demens under uppföljningstiden fick å andra sidan mer posturala

problem och gångsvårigheter under uppföljningstiden än de som inte

utvecklade demens. Mönstret var liknande för bradykinesi och rigiditet, men

inte för tremor. Kopplingen mellan bradykinesi och sämre resultat på

arbetsminnestest, mental flexibilitet samt exekutiva funktioner samt mellan

posturala/gång problem och visuospatial funktion och visuospatialt minne

hittades. Vissa av dessa kopplingar visade sig också förändras parallellt efter

ett år. Det fanns inga signifikanta förändringar på de neuropsykologiska

testerna efter ett år men patienter som fick dopaminagonisten pramipexol

visade sig få sämre resultat på verbalt flöde vid ettårsuppföljningen.

Sammanfattningsvis så har denna studie visat olikheter i andelen kognitivt

nedsatta beroende på vilka kriterier som användes vilket betonar värdet av

en gemensamt kriterier för vad som utgör kognitiva nedsättningar vid PS.

Försämringen av motoriska problem i patienter med parkinson demens kan

bero på flera olika anledningar och behöver utredas ytterligare.

Kopplingarna mellan motoriska och kognitiva funktioner kan spegla

gemensamma patofysiologiska processer i delvist delade nätverk. Slutligen

så har vi visat att både traditionellt dopaminerga funktioner samt icke

dopaminerga funktioner på olika sätt är i kopplade till Parkinson demens

vilket ger en indikation till att interventioner vid PS behöver se bortom det

dopaminerga nätverken.

IX

Original papers

I. Elgh E, Domellöf M, Linder J, Edström M, Stenlund H, Forsgren L.

Cognitive functions in early Parkinson’s disease: a population based study.

European Journal of Neurology 2009:16:1278-1284

II. Domellöf ME, Elgh E, Forsgren L. The relation between cognition and

motor dysfunction in drug naïve newly diagnosed patients with Parkinson’s

disease. Movement Disorders. 2011:26:2183-2189

III. Domellöf ME, Forsgren L, Elgh E. Persistence of associations between

cognitive impairment and motor dysfunction in the early phase of

Parkinson’s disease. Journal of Neurology. 2013:9:2228-2236

IV. Domellöf ME, Forsgren L, Ekman U, Elgh E. Cognitive function in the

early phase of Parkinson’s disease, a longitudinal follow-up. Manuscript.

Papers are reprinted in this thesis with the kind permission of the

respective publisher.

X

1

Introduction

Parkinson’s disease (PD) is a chronic and progressive neurodegenerative

disease whose diagnosis is based on a combination of the motor signs:

tremor, bradykinesia, rigidity and postural abnormalities. The incidence of

PD increases with age and it is the most common neurodegenerative disease

next to Alzheimer’s disease (AD) in people over the age of 60 years. As life

expectancy increases and people over 60 years of age is the fastest growing

age group [1], the number of patients with PD will most likely increase.

PD was initially described by Dr James Parkinson in “An essay of the shaking

palsy” [2]. The description of the disorder was similar to how we describe it

today, as a progressive disease due to probable degenerative pathology in the

central nervous system. The main symptoms were characterized by resting

tremor, flexed posture and shuffling gait. The intellect and senses were

described as being intact. It was not until the beginning of the 20th century

that the first reports of mental deterioration in PD came (see Pollock and

Hornabrook for review) [3], and not until the 1970’s that dementia was

considered an important part of the clinical picture [4]. Today it is well

established that a substantial proportion of patients with PD develop

Parkinson’s Disease Dementia (PDD) [5] and that mild cognitive impairment

(MCI) is common at the time of diagnosis [6–8]. Despite this PD is still

primarily described as a movement disorder with treatment mostly focusing

on the reduction of motor symptoms.

Attempts have been made to find predictors for cognitive decline and

dementia in PD. Associations between motor symptoms and cognitive

decline have been suggested but the results are inconclusive due to

differences in the selection of participants and variables tested. Prospective

studies in large cohorts of well-defined newly diagnosed patients with PD

assessed with extensive neuropsychological protocols are rare [6–8]. This

thesis presents data from the NYPUM study where cognitive function in

early stages of PD and its association to motor function has been explored

both cross-sectionally and longitudinally.

Parkinson’s disease

Definition and diagnosis

PD is one of several disorders with parkinsonism. The most commonly used

diagnostic criteria for parkinsonism is the United Kingdom Parkinson’s

Disease Society Brain Bank (UK PDSBB) definition which is also used in this

2

thesis. It requires bradykinesia plus at least one of the following symptoms:

tremor, muscular rigidity or postural abnormalities for diagnosis [9].

Parkinsonism can be idiopathic, i.e. of unknown cause, and includes PD and

the atypical forms referred to as Parkinson plus syndromes or atypical

parkinsonism: Multiple System Atrophy (MSA), Progressive Supranuclear

Palsy (PSP), Cortico Basal Degeneration (CBD) and Dementia with Levy

Bodies (DLB). Parkinsonism that is secondary to known cause is classified as

secondary parkinsonism, e.g. treatment with neuroleptic drugs.

Diagnostic accuracy of PD is a big problem. Clinical-pathological studies

report incorrect diagnosis of PD in 10-24% of patients with advanced PD

[10]. This is due to the existence of several similar conditions with

parkinsonism but also due to the heterogeneity of PD both in presentation

and in response to medication. Clinical signs that may differentiate PD from

other parkinsonian disorders are: more often an asymmetrical onset of

motor symptoms, the presence of rest tremor and a positive clinical response

to dopaminergic medication [9].

No radiological methods can readily distinguish between the different

parkinsonian disorders. Nuclear imaging methods such as single photon

computed tomography (SPECT) or positron emission tomography (PET) can

reveal decreased dopaminergic nerve terminals in both PD and Parkinson

plus syndromes but do not distinguish between them [11]. A definite

diagnosis can only be confirmed with autopsy.

Symptoms

Cardinal signs

Bradykinesia is the main feature of parkinsonism and is characterized by

slowness or weakened and reduced amplitude of movements. Bradykinesia is

a collective term for hypokinesia and akineisa. Hypokinesia refers to reduced

spontaneous movement such as arm swing and reduced facial expression.

Akinesia refers to difficulty in initiating movement, which is more common

in the later stages of PD.

Tremor seen in PD is a slow (4-6 hertz) rest tremor, often initially unilateral.

It is common that the tremor increases with anxiety and affect. Other types

of tremor such as postural and action tremor can also be present.

Rigidity (stiffness) refers to an increased resistance to passive bending and

stretching of the patients arm. If the muscle tone varies it can give rise to

3

what is called the “cogwheel” phenomena with jerky movements during

bending of the limb.

Postural instability and gait disturbances are often present in PD. They are

characterized by slow pace while walking, small shuffling steps, balance

problems and gives an increased risk of falling. Postural instability is more

common in the later stages of PD and is believed to be due to impaired

postural reflexes.

Other motor symptoms

Apart from the motor symptoms that are used in the diagnostic procedure

there are several other common motor features such as impaired articulatory

ability (dysarthria), soft speech resulting from lack of coordination in the

vocal musculature (hypophonia), swallowing difficulties (dysphagia) and

drooling (sialorrhoea). All combined these are referred to as bulbar

functions.

In later stages of the disease motor complications and freezing of gate are

common phenomena. Motor complications are usually a result of long term

treatment and consist of motor fluctuations and dyskinesias. Motor

fluctuations refer to decline in motor performance usually in the wearing off

phase of the medication cycle. Sometimes with disease progression there can

also be off-periods that seem to appear more or less at random [12].

Dyskinesias are involuntary movements that are presented as stereotypic,

choreatic or dystonic movements. They usually appear at peak dose (when

the L-dopa has reach the plateau) and more rarely when the drug levels rise

or fall [12].

Non motor symptoms

Apart from the motor features in PD there are a range of non-motor features

and some of them are common. Non-motor symptoms in PD include

neuropsychiatric symptoms with depression, apathy, hallucinations,

cognitive impairment and dementia, sleep disorders with restless legs, REM-

sleep behavior disorder, excessive daytime somnolence, vivid dreaming and

insomnia, autonomic symptoms with bladder disturbances, orthostatic

hypotension and sweating, sensory symptoms with olfactory disturbance,

pain and visual dysfunction and gastrointestinal symptoms with

constipation. Most of the non-motor features develop after motor onset but

there are a few that can be present years before diagnosis: olfactory

dysfunction, mild dysautonomia as well as mood and sleep disturbances [13].

4

Epidemiology

The prevalence of PD in people over 65 years of age is around 1-2% and

increases from 0.6% in the ages 65-69 to 2.8% in the ages 85-89 [14]. The

cumulative incidence (which can be regarded as the lifetime risk) of PD up to

89 years of age is close to three% [15]. Early onset PD, i.e. clinical signs

developing before the age of 50 constitutes only three-four% of the PD

population [15, 16].

The incidence of PD in high quality studies ranges from 8.4 to 20.0 per 100

000 with a mean of 14.5 per 100 000 (95 % CI, 12.2-17.3) [17]. The mean age

at symptom onset is in the late 60’s and population based studies report age

at diagnosis to around 70 years of age [8, 15]. Most studies have reported

slightly higher prevalence for PD in men, but there are also studies reporting

no differences between men and women (see de Lau et al for review) [18].

Histopathology and etiology

Most cases with PD have an unknown etiology. Both genetic and

environmental factors have been implicated and PD is most likely caused by

an interaction of genetic, aging and environmental factors. Environmental

factors that have been linked to PD are lifestyle, dietary and occupational

exposures, such as increased risk with pesticide exposure or protective effect

of substances such as caffeine and smoking [18].

There are subsets of patients of about 10% that report a positive family

history [18]. Some families show an autosomal dominant inheritance

pattern, others a recessive inheritance pattern. Through the genetic forms of

PD, the hope is to unravel biological processes that are also of importance for

the vast majority with sporadic PD.

5

FIGURE 1. Sixty times magnification of Lewy-bodies and Lewy-neuritis in substantia nigra in a patient with Parkinson’s disease. (photo: Suraj Rajan)

Pathology, biochemistry and physiology

A histopathological diagnosis of PD requires loss of dopaminergic cells in

substantia nigra pars compacta and Lewy-bodies and Lewy-neurites in some

of the surviving dopaminergic neurons. Lewy-bodies and Lewy-neuritis are

mainly composed of the protein alpha synuclein (α-synuclein) which is

encoded by the SNCA gene [19]. Lewy bodies are not exclusively found in the

nigrostriatal network, and they are believed to spread throughout the brain

in a sequential manner during different stages of the disease with onset in

the olfactory structures and the dorsal motor nucleus of the vagus nerve [20,

21].

The deterioration of dopaminergic neurons in the substantia nigra projecting

to the basal ganglia is believed to cause two of the cardinal signs seen in PD,

rigidity and bradykinesia. Bradykinesia has been described as the best

clinical correlate of the nigrostriatal lesion [22]. Tremor has been suggested

to be both a result of the disruption in the striatopallidal circuit and a result

of a disruption in the cerebello-thalamo-cortical circuit [23], as well as

involving serotonergic dysfunction [24].

Postural abnormalities that often appear at later disease stages are believed

to have origins other than dopaminergic failure. They have been connected

to cholinergic denervation of the pedunculopeptine nucleus (PPN) [25, 26].

Increased neocortical ß-amyloid deposition which is common in AD has also

been associated with more severe postural and gait disturbances in PD [27].

6

Treatment of motor symptoms

Available treatments are mainly focused on reducing motor symptoms. They

have no effect on underlying pathophysiological processes and consequently

do not have curative or disease modifying effects. The most commonly used

treatments are pharmacological interventions with drugs that affect the

dopamine system, so called dopaminergic drugs.

Levodopa

The first drug available for treatment of PD was levodopa; it was developed

in the early 1960s based on the findings of the Swedish Nobel prize winner,

professor Arvid Carlson and early clinical trials by Ehringer and

Hornykiewicz [28, 29]. Dopamine cannot pass through the blood brain

barrier (BBB) but levodopa (L-dopa), a precursor of dopamine, can. L-dopa

is taken up by dopaminergic neurons and decarboxylates into dopamine

presynaptically. When released it binds to both D1 class receptors (including

D1 and D5) and D2 class receptors (including D2, D3 and D4)

postsynaptically. Dopamine that does not bind to postsynaptic receptors is

taken up into the presynaptic cell by the dopamine transporter.

Decarboxylase and COMT inhibitors

To prevent levodopa from being metabolized to dopamine in the periphery

before passing the BBB and entering the brain, levodopa is combined with

decarboxylase inhibitors (carbidopa or benserazide) and sometimes also

with catechol-O-methyl transferase (COMT) inhibitors (entacapone or

tolcapone). The result is that more of the drug enters the brain, and

peripheral side effects are reduced (e.g. nausea, hypotension).

Dopamine agonist and MAO-B inhibitors

Other dopaminergic drugs are dopamine agonists and monoamine oxidase

(MAO) B inhibitors. MAO-B metabolizes dopamine in the dopaminergic

neuron and inhibition of the enzyme increases the availability of dopamine.

Dopamine agonists (DA) act directly on the postsynaptic system. The various

DA used have different receptor profiles. The commonly used non ergot

dopamine agonists, such as pramipexole and ropinirole have a high affinity

for D2 class receptors [30]. DA is today often used as monotherapy in early

phase of PD, mostly in younger patients, in order to reduce/delay motor

complications. Although use of DA early in the disease gives the benefits of

delaying the start of motor fluctuations accompanied with the use of L-dopa

there are other side-effects of DA such as higher risk of hallucinations,

orthostatic hypotension, excess daytime sleepiness, sleep attacks and

impulse control disorder [31].

7

Treatment in later disease stages

In later stages of the disease a portable pump delivering apomorphine

subcutaneously or levodopa delivered intestinally can be used for continuous

dopaminergic stimulation to reduce variation in motor performance (motor

fluctuations and dyskinesias). An antagonist on the glutamate N-methyl-D-

aspartate (NMDA) receptor (amantadine) can also be tried for treatment of

dyskinesias and fluctuations. Neurosurgical interventions with targeted

lesions in the brain in advanced cases with PD have largely been replaced by

Deep Brain Stimulation (DBS). Most DBS operations for PD target the

subthalamic nucleus which has a central role in the motor circuits that are

disturbed in PD.

Cognitive impairment

Cognition can be defined as higher order brain functions that help the

individual to interact with the environment in a successful way. Sometimes

the cognitive functions do not work as intended, due to functional or

structural impairment. This can lead to dysfunction for the individual. These

disturbances can be reversible if caused by stress, medication, depression, or

sleep deprivation. They can also be due to degenerative processes which can

only be treated symptomatically. There are different ways to describe and

measure human cognition. No cognitive tests measures only one cognitive

domain and tests usually tap into different cognitive functions. A poor result

in a single test can be due to damage in various parts of the brain.

Cognitive impairment in PD

The profile of cognitive impairment in PD varies between individuals and is

as heterogenic as the rest of the clinical picture, probably because of the

diverse nature of the underlying pathology. Already in the early stages of the

disease a wide range of cognitive impairments have been demonstrated [6–

8], e.g. in memory, visuospatial function and executive functions. Some

suggest that executive problems are related to the early phase of the disease

due to an altered dopaminergic tone in the frontal cortex. On the other hand

impairment in visuospatial dysfunction and semantic verbal production

display an involvement of temporal and posterior structures. The

visuospatial dysfunction has been suggested to be related to later disease

stages and subsequent development of dementia [32].

Executive functions

Executive functions include a set of abilities that control and regulate other

cognitive functions. It can be described as our ability to plan, perform

abstract reasoning, solve problems, focus despite disturbances and shift

8

focus when appropriate. Executive functions have been linked to distributed

networks with an interaction between prefrontal and subcortical regions.

Cognitive decline in PD has sometimes been described as resembling the

pattern seen in frontal lobe patients with mainly frontally mediated attention

and executive problems [33]. Executive functions are affected both in PD

and PDD and have been suggested to be more affected in PDD than in AD

(the most common dementia disorder)[34].

Memory

Human memory consists of multiple systems. A basic distinction can be

made between short-term or working memory and long-term memory

(LTM). Working memory temporarily hold information while LTM refers to

the ability to store information [35]. Working memory has been shown to be

distinctly different from LTM. Connections between dopamine, aging and

cognition [36], especially working memory processes have been suggested

[37].

LTM can be subdivided into declarative and non-declarative memory. In

turn declarative LTM can be separated into semantic and episodic memory.

Semantic memory refers to a network of associations and concepts of basic

knowledge about the world. Episodic memory refers to information about

personally experience of past events. Different memory processes involved

are encoding, consolidation and retrieval of information (Figure 2.).

Disturbances of any of the three memory components, i.e. encoding,

consolidation or retrieval, can result in memory failure.

The medial temporal lobes are involved in acquisition and retrieval of new

episodic memories [38] . Consolidation of new memories has been linked to

hippocampus and surrounding structures [39].

FIGURE 2. Overview of memory processes

9

Frontally mediated cognitive processes that are engaged in working memory

and executive function are believed to be collaborating in long term memory

as well, especially in free recall [40]. In a test situation episodic memory is

often assessed by asking a person to recall or recognize information learned

at the time. Episodic memory impairment is present in PDD although some

studies claim that it is less severe than in AD [41]. Some suggest that

memory impairment in PD is more related to a frontally mediated retrieval

deficit than to an encoding problem. In PDD there is evidence of recognition

deficiencies as well (see Emre et al for review) [34].

Attention

Attention is a multidimensional function that involves processes that focus,

select, divide, sustain and inhibit behavior. Attention is important for all

cognitive skills and, in tests, especially difficult to separate from working

memory and executive functions. The term attention has been used

interchangeably with executive functions in some prior PD studies [34].

Visuospatial skills

Visuospatial function includes mental imagery and navigation, distance and

depth perception and visuospatial construction. It is the ability to

understand visual representations and their spatial relationships and is

governed by several different pathways originating from parietal cortex

(Occipito-parietal, parieto prefrontal, parieto pre-motor and parieto-medial

temporal) (see Kravitz et al for review) [42]. Both constructional abilities and

visuospatial function without the demands of fine motor control have been

shown more affected in PDD than in AD [34].

Language function

Language can be described as the capacity for acquiring and using complex

systems of communication. Language functions include abilities such as

reading, arithmetic, oral and written word production and comprehension.

Common test for language function are verbal fluency and word

comprehension test. Some studies use tests of verbal fluency as a measure of

language function whereas other includes it to measure executive functions.

Patients with PDD are considered to have less language impairment than

patients with AD [34].

PD-MCI

The concept of Mild Cognitive Impairment (MCI) was developed by Petersen

et al. [43] to detect individuals with an increased risk of developing AD. It is

defined as a transition stage between normal aging and dementia. The

original criterion was focused on memory impairment but it was later

suggested that also patients with impairment in other cognitive domains

10

could be at risk for developing dementia [44]. Studies have applied

Petersen’s MCI criteria on patients with PD with various results. To create

conformity between studies and clinicians within the field of PD a task force

commissioned by the Movement Disorder Society adapted the MCI criteria

to fit the specific cognitive profile seen in PD [45] and constructed guidelines

to assist in the MCI classification. The PD-MCI criteria are based on a

combination of literature review and expert consensus.

PDD

Dementia is defined as a progressive, irreversible deterioration of cognitive

function. The pathologies behind dementia disorders can be different types

of neurodegeneration and vascular lesions. The definition of dementia in the

Diagnostic and Statistical Manual of Mental Disorders IV is an overall

decline in intellectual function including difficulties with language, simple

calculations, planning and judgment, and motor skills. Furthermore a loss of

memory that is severe enough to interfere with activities of daily living that is

not due to physical decline. PDD require onset of motor symptoms at least

one year before the onset of dementia [34]. The one year rule is to

differentiate between PDD and DLB.

Epidemiology of cognitive impairment in PD

Five to seven percent of adults over the age of 60 are demented [46]. Of the

dementia population three-four% have PDD [47]. The proportion of patients

with PDD in PD populations varies between 28% and 90% [48] depending

on selection of participants and criteria used.

In community based studies of PDD the annual incidence rate has been

around 100 per 1000 person years [49–51] which corresponds to around

10% of the PD population developing dementia each year. The annual

incidence rate of dementia in population based studies with newly diagnosed

patients with PD has been 38.7 per 1000 person-years (95% CI, 23.9-59.3)]

[32]. See table 1 for longitudinal studies of newly diagnosed patients with

PD.

The proportion of patients with MCI in studies of newly diagnosed patients

with PD has ranged between 19 and 36% [6–8, 52, 53] with a higher

proportion in community-based samples. In adults older than 65 years in the

general population the proportion of MCI has ranged between three and 19%

[54].

11

TABLE 1. Overview of community based studies of prospectively followed cases with newly diagnosed PD.

Study, inclusion Follow up years

Cases/ controls

Mean age

Started treatment for PD

Cognitive domains: tests included

Sidney multicenter [55, 56]a

1985-1988

20 91/50* 63 13.6% Vocabulary, Block Design, Ravens Colored Progressive Matrices, RAVLT, Benton visual retention test, simple and choice reaction time test, COWAT, Austin Maze, Western aphasia battery

CamPaIGN [8, 32]b

2000-2002 10 159/na 69.9 47.2% Phonemic fluency (FAS), Category fluency

(animals), Pattern and Spatial Recognition Memory, ToL

CARPA [6, 57]b 2002-2005

5 115/70

66.2 67.8% Psychomotor speed: Digit symbol test, TMT A, Stroop A and B Attention: Digit span, TMT B, Stroop C, Language: BNT Memory: RAVLT, RBMT, Logical Memory Test, face recognition, Visual Association Test Executive functions: MWCST, category fluency, similarities, ToL

Visuospatial/Construction skills: JOLOT, GIT, Clock drawing test

ParkWest [7, 58]c

2004-2006 3 196/201 67.3 0% Memory: California verbal learning test

Visuospatial ability: Silhouettes and Cube test Attention/Executive function: category fluency, serial seven from MMSE, Stroop test

areferences for three and five year follow up, breferences for baseline and five year follow up, creferences for baseline and three year follow up.

*Also included cases with dementia before the onset of motor signs.ToL= Tower of London, TMT=Trail Making Test, BNT=Boston Naming Test,

RAVLT= Ray Auditory Verbal Learning Test, RBMT=Rivermead Behavioural Memory Test, MWCST= Modified Wisconsin Card Sorting Test,

JOLOT=Judgement of Line Orientation, GIT=Groeningen Inteligence Test, COWAT=Controlled Word Association Test.

12

Risk factors for cognitive impairment in PD

Identification of clinical factors that predict development of dementia are

important for clinical practice and disease management [59]. Age [55],

depressive symptoms, specific neuropsychological impairments [60], specific

motor impairment, male sex, fewer years of education, visual hallucination,

REM sleep disorder and orthostatic hypotension have all been associated

with an increased risk of PDD. For MCI older age, motor disease severity,

non-tremor dominant motor phenotype and fewer years of education have

been reported as associated factors.

Older age is one of the primary clinical features predicting PDD [61, 62].

Early presence of frontal executive problems has been pointed out to be a

predictor for development of PDD [63]. Recent research has started to

evaluate this and connects the development of PDD also to posterior

cognitive problems such as visuospatial, verbal function and episodic

memory (see Kehagia, et al. for review) [64]. Preliminary results suggest that

PD-MCI with posterior cognitive deficits predicts a shorter time to PDD [45].

Patients with predominantly postural instability and gait difficulties (PIGD)

have been suggested to have a faster rate of cognitive decline [65]. However,

following patients with early PD for up to 10 years have not rendered in the

same results [66]. Some mean that it is change from tremor dominant

subtype to PIGD dominant subtype that is a predictor for developing PDD

[67].

Neuropathology of cognitive impairment in PDD

The neuropathology underlying cognitive decline and PDD is heterogeneous

and studies exploring pathological correlates of cognitive impairment in PD

have rendered conflicting results [68].

Cortical Lewy bodies and Lewy neuritis have been shown to be the most

significant correlate of dementia in PD [69]. α-synuclein pathology in the

parahippocampal gyrus and anterior cingulate gyrus is more pronounced in

PDD than in PD [70, 71]. Some individuals have pronounced α-synuclein

burden without being demented, which suggests that α-synuclein burden by

itself is not sufficient for dementia. Other factors such as cognitive reserve or

brain plasticity might determine the cognitive performance in relation to the

α-synuclein burden (See Irvin et al for review) [59]. Furthermore, some

patients with PDD express only small amounts of α-synuclein. Cholinergic

deficits have been seen in patients with PDD that have less pronounced

cortical and limbic Lewy bodies and Lewy neuritis.

13

Some studies claim that less than 10% of PDD cases have coexisting AD [72,

73], whereas other have reported proportions as high as 30-40% [69, 74, 75].

For example low CSF ß amyloid levels, have been linked to development of

PDD [76]. Differences between demented and non-demented patients have

also been found in the distribution of neurofibrillary tau pathology (also

common in Alzheimer disease). The spread for non-demented patients was

restricted to the entorhinal areas whereas neurofibrillary tangles had spread

to the rest of the limbic system, lateral temporal areas and beyond in

patients with PDD [21].

The contribution of vascular lesions to PDD is not well studied but some

indications of higher degrees of vascular lesions in PDD than PD without

dementia have been found [77]. A cross-sectional study combining the

cortical Lewy, ß-amyloid and tau stages have shown that this combination

perfectly discriminates between demented and none demented PD [76].

Genes and cognition in PD

Familial associations to the development of PDD have been reported [78].

However, not much is known about how genes contribute to cognitive

impairment and PDD and only a few of the findings have been linked to

biological processes likely to be involved in cognitive impairment and/or

PDD

The genetic polymorphism Catechol-O-methyl-transferase (COMT) gene

(Val158Met) has been suggested to have an impact on executive functions

through its link to the dopamine system but has not been connected to PDD

[32]. Monoamine oxidase (MAO) has also been suggested to be a candidate

gene linked to executive heterogeneity in PD.

Some of the familial forms of PD are linked to PDD. The gene SNCA coding

for α-synuclein on chromosome four can be multiplied and cause increases

in the concentration of α-synuclein. Triplicated cases are more often than

duplicated cases associated with cognitive decline and dementia with a more

rapid progression. A polymorphism of the DYRK1A gene has been connected

to the PDD and LBD [79] through effects on a kinase that phosphorylates α-

synuclein and amyloid precursor protein. Mutations in the ATP13A2 gene

results in a rare genetic variant that can cause young onset parkinsonism

(Kutor-Rakeb syndrome: PARK 9). These patients become demented and the

predominant pathology is atrophy with iron accumulation in basal ganglia

[80]. The MAPT H1/H1 genotype seems to have a strong influence on

cognitive functions, especially posterior cortical cognitive impairments and

give a higher risk of developing PDD [81].

14

Treatment of cognitive impairment and other non-motor

features in PD

Treatment for non-motor and non-dopaminergic symptoms in PD has

started to be evaluated because of the recent focus on the disabling effect of

the non-dopaminergic and non-motor features of the disease.

Cholinesterase inhibitors have been shown to have positive effect for

cognitive dysfunction and PDD, but also for behavioural disturbances and

activities of daily living in patients with PDD [82]. Rivastigmine is the

cholinesterase inhibitor with strongest evidence as an effective treatment of

PDD [83]. Special characteristics of responders are hard to find but those

with visual hallucinations [84] and those with elevated levels of

homocysteine [85] seem to respond especially well.

For psychotic symptoms in PD a gradual reduction of antiparkinson

medication is recommended [86]. If neuroleptic treatment is needed,

atypical forms (e.g. clozapine) shows the best effect/side effect profile for PD

patients [87].

Treating depression in patients with PD is complicated. A recent meta-

analysis concluded that there is insufficient evidence to suggest selective

serotonin reuptake inhibitors (SSRI), serotonin and norepinephrine

reuptake inhibitors (SNRIs), pramipexole or pergolide for treatment of

depression in patients with PD [88]. According to the same meta-analysis

the treatment with the best effect on reducing depressive symptoms in PD

was tricyclic antidepressants (TCAs).

Dopaminergic medication has been suggested to have both detrimental and

positive effects on cognitive functions. One explanation for this is that the

relationship between dopamine levels and cognitive performance is believed

to follow an inverted U-shaped curve, where both low and high levels of

dopamine cause impaired cognitive performance [89]. Individuals with

initially low levels of dopamine are believed to improve performance with

intake of dopaminergic drugs while patients with higher levels decline in

performance due to excessive levels of dopamine. Different functions are

mediated by different brain networks and affected differently by

dopaminergic depletion. In early PD for example, the loss of dopaminergic

neurons in striatum is most prominent in putamen and dorsal caudate, both

involved in the motor and dorsolateral circuit. The ventral striatum, involved

in limbic and orbitofrontal circuits is mostly intact.

15

High levels of dopamine agonists and levodopa are believed to have a

negative effect on reversal learning, decision-making and impulse control

[64]. The dopamine agonist pramipexole, has been suggested to have a more

harmful effect than pergolide [64]. Some studies have suggested that short

term use of the dopamine agonist pramipexole might cause decline in short

term verbal memory, attentional-executive functions and verbal fluency

[90]. A slight cognitive decline in semantic verbal fluency, executive

functions, verbal learning and memory has been shown shortly after

subthalamic deep brain stimulation [91].

Cognitive motor relationship

The association of motor control and cognitive function has been studied in

children [92], patients with brain injury [93] and patients with

neurodegenerative disorders [94]. One goal in linking motor and cognitive

function is to find which aspects of cognitive and motor functions are

processed by a given area or brain network [95]. Associations between

cognitive and motor function have been found in the elderly and in patients

with neurological disease. These associations can be unspecific reflecting co-

existence of common age-related syndromes or reflecting a more widespread

pathology affecting both motor and cognitive structures. They can also be

specific as in the association of specific cognitive and motor functions being

governed by the same neural networks.

Apart from the relation between PIGD subtype and PDD [67, 96], other

cognitive motor relationships have been suggested in PD. Early findings

reported bradykinesia and rigidity [4] to be associated with PDD and the

severity of bradykinesia has been connected with visuospatial reasoning and

psychomotor speed [97]. Others found no motor cognitive relationships [98,

99]. The results are inconclusive due to differences in the selection of

participants and variables tested.

Rationale of the thesis

Cognitive impairment in PD has severe consequences with increased

mortality, nursing home placement and caregiver stress [100]. A better

characterization and understanding of cognition in the early phase of the

disease is particularly important as this is the phase when early intervention

with potential neuroprotective drugs or other therapies is likely to be most

effective. Studying motor and cognitive relationships in early stages of the

disease is important to be able to estimate the risk of early development of

dementia or other type of cognitive problems with regard to motor function.

Also finding relationships between specific motor and cognitive functions

16

may provide new ideas about the underlying pathophysiological processes

for the motor and non-motor functions in PD.

Most studies on cognitive function in PD have been retrospective, cross

sectional, including a mixture of incident and prevalent cases or been biased

towards younger cases. Good descriptions of cognition in the early phase of

PD in unselected study populations followed prospectively were almost

completely lacking [8] when this study was initiated [15]. Prospective studies

in large cohorts of well-defined newly diagnosed patients with PD assessed

with extensive neuropsychological protocols can help connect cognitive

motor associations to specific functions rather than global cognitive and

motor decline and perhaps dissociate predictive clinical factors from

associated factors.

17

Aims

The aim of this thesis was to the improve characterization and

understanding of cognition in early phases of PD, to investigate clinical

determinants of cognitive decline and dementia and to explore what aspects

of cognitive function are connected to different motor functions. Further

aims were to investigate the variability of cognitive performance in PD in

patients followed prospectively from time of diagnosis and during the

following five years.

1. To describe the character and predictors of cognitive dysfunction in a

population based cohort with drug naive newly diagnosed PD. (Paper I)

2. To explore the relationship between cognition and motor dysfunction in a

population based cohort with drug naive newly diagnosed PD. (Paper II)

3. To explore if motor and cognitive variables associated at baseline change

in parallel after one year. (Paper III)

4. To report the magnitude of cognitive change one year after diagnosis and

investigate if different types of dopaminergic medication have an impact on

the results. (Paper III)

5. To explore the five-year course of cognitive functions in a prospectively

followed cohort of patients with PD and search for predictors for PDD (Paper

IV).

5. To explore the difference in the evolution of motor function in patients

that develop PDD compared to those who do not develop PDD (Paper IV).

18

Materials and methods

This thesis is based on data from the NYPUM-project (new parkinsonism in

Umeå), a population-based study of idiopathic parkinsonism addressing

etiological, diagnostic and prognostic factors. The patients were recruited

from southern part of Västerbotten County in northern Sweden with a

catchment area of 142 000 inhabitants.

Patients were included from January 1st 2004 to April 30th 2009 and

classified in to different forms of parkinsonism (PSP, MSA, CBG and LBD)

according to established clinical criteria. Cases were investigated extensively

at the time of presentation and repeatedly during follow up which was

between 4.5-9 years except for those who died. To avoid selection bias cases

were identified prospectively during the inclusion period and through many

sources to make case identification as complete as possible. A total of 185

patients with idiopathic parkinsonism were identified.

FIGURE 3. Map of the investigation area (Pantzare Information AB, Luleå,

Sweden)

19

Study population

The study populations participating in the studies of this thesis are presented

below. The differences in participants between the studies are due to sample

selection and change in diagnosis during follow up. Paper I is based on the

first four years of inclusion whereas paper II, III and IV include the whole

sample. Figure 4 (a, b, c, d) describes the flow chart of participants in each

paper.

Thirty age and sex matched controls based on the 50 first patients included

in the study were recruited. The controls were recruited by advertisements in

the local newspaper or among friends and family of the PD participants.

Requirements for controls were that they needed to be healthy with no

neurological disorders and normal neurological examination, and have a

normal FP-CIT scan.

Non-participation

About 20% (Paper I 21%, Paper II-IV 19%) of the patients declined to

participate in the neuropsychological assessment at baseline. They were

significantly older (79 vs. 69 years, p<0.001), had various medical

conditions, such as blindness, deafness, severe cardiac disease and had

higher scores on the UPDRS part III (35 vs. 26, p<0.001) and scored worse

on the MMSE (27.5 vs. 28.7; p=0.04).

There were also a substantial amount of patients that did not participate at

follow up. This loss was bigger for the neuropsychological evaluation than for

the study as a whole. Patients that did not participate in neuropsychological

testing at follow-up were older, had fewer years of education and higher

scores on the UPDRS. Furthermore, they performed worse on all

neuropsychological tests at baseline except a test of language.

20

FIGURE 4A. Flow chart showing the participants in Paper I FIGURE 4B. Flow chart showing the participants in Paper II

21

FIGURE 4C. Flow chart showing the participants in Paper III

22

FIGURE 4D. Flow chart showing the participants in paper IV

23

Assessment tools

Global cognitive ability was measured with the Mini-mental State

Examination (MMSE) [101]. Depression was assessed with the Montgomery

and Åsberg Depression Rating Scale (MADRS); participants with scores over

eight were considered to have a mild depression and severe depression if

scores were 18 or over [102]. Dopamine transporter (DAT) Single-photon

emission computed tomography (SPECT)-imaging using 123I-Ioflupane ([123I]

FP-CIT) was performed within 1-2 months following inclusion. Levodopa

Equivalent Dose (LED) was calculated based on the conversion factors

suggested by Tomlinson et al [103].

Assessment of motor function

The severity of parkinsonism was measured by the Unified Parkinson’s

Disease Rating Scale (UPDRS) [9] and the Hoehn and Yahr staging.

Variables of the different motor-scores were calculated from the UPDRS III

[104]: the sum of UPDRS III items 20 and 21 for tremor, item 22 for rigidity,

the sum of items 24, 25, 26, and 31 for bradykinesia, the sum of items 27, 28,

29, and 30 for the posture and gait score (arising from chair, posture, gait

and postural stability), and the sum of items 18 and 19 for the bulbar score

(speech and facial expression) [105].

Previously published divisions of motor subtype were used to divide patients

into groups based on predominant motor feature: postural instability and

gait disturbances (PIGD), tremor or indeterminate phenotypes [106]. A

patient was classified as having tremor dominant PD if mean tremor score

divided by mean PIGD score was >=1.5, PIGD subtype if mean tremor score

divided by mean PIGD score was <=1.0 and indeterminate subtype if mean

tremor score divided by mean PIGD score was between >1.0 and <1.5.

Assessment of cognitive function

Cognitive functions were thoroughly investigated at the time of inclusion

(baseline) and after one, three, and five years. The patients were tested at

their optimal motor stage. The tools used were neuropsychological tests of

verbal and non-verbal episodic memory, working memory, attention,

psychomotor function, visuospatial function and executive function. Table 2

shows the total battery of neuropsychological tests with an explanation of

what cognitive domain each test was intended to measure and how the test

was executed. The tests are listed in the order of presentation.

24

The tests were chosen to assess important cognitive functions within a

reasonable amount of time. Since motor function is affected in PD, tests that

are not affected by motor function were chosen when possible. All tests

included in the battery are frequently used both in research and in clinical

practice and have good validity and reliability. Most of the tests have age and

education adjusted norms.

Instruments were chosen to minimise the effect of repeated testing, e.g.

parallel versions. Tests that did not have alternate forms or could be affected

by close repeated measurements were excluded from the 12-month follow-up

[107].

The assessments were performed by trained research assistants with

backgrounds in psychology and neuroscience supervised by a consultant in

clinical neuropsychology. To assure that the assessments would be coherent

within and between testers they were instructed to perform the assessment

in accordance with the standardized protocol of the different tests. They

practiced before they performed the first testing accompanied by a senior

tester during one or more testing and performed an assessment on a patient

during supervision with feedback. The test protocols were checked for errors.

Patients were encouraged to participate in all neuropsychological tests. For

some participants this was impossible due to tiredness or technical issues.

The tests with most missing cases were the WCST (n=15), logical memory

(n=9) and logical memory delayed (n=11). Missing data varied between one

and six for the other neuropsychological variables at baseline. Cases were

omitted if they had missing values on most tests. Otherwise we conducted

pairwise deletion of the data.

25

Neuropsychological test Cognitive domain Execution

1. Finger tapping test Psychomotor speed Tap left and right index

finger during 10 s

2. Free and Cued Selective

Reminding test

(FCSRT)[108]

Episodic memory Cued and free recall

after encoding 16 words

x3

3. Mental Control from

WMS[109]

Conceptual tracking Counting back and forth

and reciting alphabet

4. Trail Making Test A

(TMT A)[110]

Attention/psychomotor

speed

Drawing lines between

numbers

5. Trail Making Test B

(TMT B)[110]

Attention/psychomotor

speed

Drawing lines

alternating between

numbers & letters

6. Benton Judgement of Line

Orientation (JOLOT)[111]

Visuospatial function Judging the slope of two

lines

7. Digit span forward and

backward[109]

Working memory/

attention

Repeating numbers

forward and backward

8. FCSRT delayed recall Episodic memory Cued and free recall

after a delay

9. Boston Naming Test (BNT) Verbal naming Picture naming

10. Brief Visuospatial Memory

Test [112]

Visuospatial memory Drawing 6 figures after

10 min encoding x3

11. Associative Learning Test

(WMS)

Episodic memory Remembering word

pairs x3

12. Logical memory, immediate

recall (WMS)

Episodic memory Repeating short story

directly after encoding

13. Controlled Oral Word

Association Test

(COWAT)[113], phonemic

Verbal function/executive

function

Naming as many words

beginning with the letter

F,A or S, during one

minute

14. COWAT, categories Verbal function/executive

function

Naming as many words

from a given category:

animals, colours or

fruits

15. BVMT delayed recall Visuospatial memory Drawing encoded figures

after a 25 minute delay

16. Logical memory, delayed

recall

Episodic memory Reciting short story

after delay

17. Wisconsin Card Sorting

Test-computer version

Executive function Matching cards based on

unknown rules

TABLE 2. Total battery of neuropsychological tests in order of presentation.

26

Diagnosis

PD

PD was diagnosed according to the United Kingdom Parkinson’s Disease

Society Brain Bank (UK PDSBB) criteria for definite PD, requiring at least

three supportive criteria. Cases with 1-2 supportive criteria were classified as

probable PD. All cases were checked against diagnostic criteria for MSA

[114], PSP [115],CBD [116]and DLB [117]. In cases that fulfilled the

diagnostic criteria for multiple syndromes we classified the patient with the

atypical diagnosis. Definite PD and probable PD are included in all studies.

Since the diagnosis changed at follow up for a small number of patients, the

latest diagnosis available at the time of statistical analysis for each paper was

used.

BOX 1. UK PDSBB clinical diagnostic criteria for PD

Step 1. Diagnosis of parkinsonism Bradykinesia and at least one of the following:

a. Muscular rigidity

b. 4-6 hz tremor

c. Postural instability

Step 2. Exclusion criteria for PD

History of repeated strokes and head

injuries

History of definite encephalitis

Oculogyric crises

Neuroleptic treatment at onset of

symptoms

More than one affected relative

Sustained remission

Strictly unilateral features after three

years

Supranuclear gaze palsy

Cerebellar signs

Early severe autonomic

involvement**

Early severe dementia with

disturbances of memory and

language

Babinski sign

Presence of cerebral tumor or

communicating hydrocephalus

Negative response to large dose of

levodopa

MTPT exposure

Step 3. Supportive prospective

criteria for PDD. Three or more

required for PD definite

Unilateral onset

Rest tremor

Progressive disorder

Persistent asymmetry affecting the

side of onset most

Excellent response to levodopa

Severe levodopa induced chorea

Levodopa response for 5 years or

more

Clinical course of 10 years or more

*We excluded the criterion ”more than one affected relative” since it is now known that PD can have a genetic background **Defined as symptomatic orthostatic blood pressure fall or presence of urinary incontinence within 12 months of baseline visit.

27

Cognitive impairment

In Paper I a cognitive domain was considered to be impaired if more than

half of single test results within that domain were below cutoff level. Lezak et

al have presented a classification system for ability levels based on a

statistically defined range of scores [118]. Average performance is according

to this classification: mean +/-0.6 SD low averages, -0.6 to -1.3 SD

borderline, -1.3 to – 2.0 SD impaired. We chose -1.5 SD as a cut-off level

score for lowered test performance, which is commonly accepted in clinical

practice.

Paper I Paper IV

Cognitive domain Neuropsychological test Neuropsychological test

Episodic memory FCSRT free and cued recall

Logical Memory and Paired

Associative Learning from WMS

(healthy controls)

BVMT

FCSRT free recall

BVMT total

BVMT delayed

Working memory Digit span from WAIS III

Digit Span backwards

TMT B

Attention TMT A and B

Executive functioning WCST

Mental control from WMS

WCST total errors

WCST preservative errors

Animal fluency

Visuospatial

functioning

The Benton Judgment of Line

orientation

The Benton Judgment of Line

orientation

Verbal function Boston naming test

Controlled Oral Word Association

(FAS and categories)

Boston naming test

TABLE 3. Neuropsychological measures used for classification of cognitive

impairment (Paper I) and MCI according to MDS Task force criteria (Paper IV).

FCST=Free and cued selective reminding test, WMS=Weschler memory scale,

BVMT=Brief Visuospatial Memory Test,TMT Trail Making Test, WCST=Wisconsin Card

Sorting Test, BNT=Boston Naming Test

28

MCI

To classify patients as MCI in Paper IV the MDS Task Force criteria were

used [45]. The task force guidelines require a diagnosis of PD based on the

UK PD Brain Bank criteria [119], gradual decline in cognitive ability reported

by either patient, informant or observed by clinician, cognitive deficits on

either formal neuropsychological testing or a scale of global cognitive

abilities, and cognitive deficits that is not sufficient to interfere significantly

with functional independence, no dementia and a minimum of two

neuropsychological tests with 1-2 SD below the mean value of established

norms or a control group, and subjective cognitive complaints. As in Paper I,

1.5 SD below established norms were used to detect impaired domains in

Paper IV.

To capture subjective cognitive complaints from decline from premorbid

level, information on self-perceived cognitive decline or information from

family or friends were gathered through a semi structured short

questionnaire given to the patient when enrolling in the study. The

questionnaire addressed the patient and/or family members experience of

cognitive decline, a question of how the patient function cognitively was

asked before the neuropsychological assessment and information regarding

self-perceived memory and concentration from the Parkinson’s Disease

Questionnaire 39 (PDQ-39) [120].

Dementia

A diagnosis of PDD was made according to published criteria through a

consensus between three of the authors in paper IV (MED, LF and UE) [34].

The PDD diagnosis requires a diagnosis of PD a minimum of one year prior

to the onset of dementia and cognitive deficiency severe enough to affect

Activities of Daily Living (ADL). All available information were used

including medical files, neuropsychological testing, temporal cognitive

decline measured by repeated neuropsychological testing or MMSE, and

interview by the study nurse and information from family members.

Ethics

Written, informed consent was obtained from all participants and healthy

controls. The study was approved by the Regional Ethics Committee at Umeå

University.

29

Statistics

All statistical analyzes were two tailed and p-values under 0.05 were

considered significant for most analyzes. The Statistical Package for the

Social Sciences (SPSS) versions 15, 19 and 21 and the Predictive Analytics

Software (PASW) Statistics 17.0 were used for statistical analysis. Depending

on the distribution of scores parametric and non-parametric tests was used

where appropriate. The residuals of the regression models were explored to

see if they fulfilled the assumptions of normality, linearity and

homoscedasticity.

Empirical studies

Paper I

Patients with newly diagnosed PD were compared to healthy controls on the

total battery of cognitive tests. Patients and controls were also classified as

being impaired in different cognitive domains. Neuropsychological

investigations were performed in 88 of 111 classified as PD. Only two had

received pharmacological treatment for PD before the neuropsychological

testing (low doses for a few weeks). Two individuals did not complete enough

tests to make it possible to judge impairment in different domains. Table 2

shows the tests that were used for each domain. Those impaired in one or

more domains were compared to those with normal function. Furthermore,

those with two or more impaired domains were compared to the rest.

Characteristics of this sample are presented in table 4.

Patients

n=88

mean (SD, range)

Healthy controls

n=30

mean (SD, range)

P-value

Sex (female/male) 39/49 14/16 0.825

Age, years 68.1 (9.3) 68.2 (6.6) 0.979

Education, years 9.9 (4.1) 11.5 (3.5) 0.055

MMSE 28.7 (1.4,24-30) 29.1 (0.8,28-30) 0.074

Mild depression a n (%) 14 (16%) 0 <0.001

UPDRS III 23.8 (9.7,5-48) 0

Dopaminergic medication 2 0

FP-CIT scan normal 5 30

TABLE 4. Characteristic of the population in Paper I

MMSE= Mini Mental State Examination, MADRS=Montgomery and Åsbergs Rating

Scale;UPDRS= Unified Parkinson’s Disease Rating Scale, DAT= dopamine transporter. aMADSR 9-17.Reprinted from Paper I with kind permission from the copyright holders.

30

TABLE 5. Characteristic of the sample in Paper II.

PIGD=Postural impairment and gait disturbances, MMSE= Mini Mental State Examination,

MADRS=Montgomery and Åsberg Rating Scale; UPDRS= Unified Parkinson’s Disease Rating

Scale. Reprinted from Paper II with kind permission from the copyright holders.

Independent two-tailed t-tests were used to analyze differences in

demographics, clinical characteristics and cognitive tests raw scores between

patients and controls. For adjusted p-value linear regressions were

performed with age, gender, years of education and psychomotor function as

covariates for each cognitive variable. Patients being impaired in one or

more cognitive domain were compared to those with intact cognition on

demographic and clinical variables analyzed with Student’s t-test or Mann-

Whitney U-test. Binary logistic regression was performed between patients

with two or more cognitive domains impaired for the purpose of exploring

predictors for cognitive impairments. Demographic and clinical variables

that significantly differed between the groups in independent analyses were

included as covariates in the model: years of education, disease duration,

bradykinesia, facial expression, rigidity and total UPDRS III subscore.

Paper II

The associations between cognitive and motor functions were assessed.

Furthermore, a comparison between PIGD/Tremor/Indeterminate

phenotypes was made for baseline cognitive variables. Neuropsychological

assessments were performed in 122 of 150 patients that fulfilled the

diagnostic criteria for PD. Seventeen patients that started their dopaminergic

treatment before performing the neuropsychological investigation and two

with a normal FP-CIT scan were not included in the study. Thus 103 patients

were included (Table 5).

Total

n=103

mean (SD)

PIGD

n=55

mean (SD)

Tremor n=35

mean (SD)

Indeterminate

n=13

mean (SD)

P-

value

Sex (f/m) 41/62 20/35 15/20 6/7 0.731

Age (years) 68.4 (9.2) 69.5 (8.7) 63.0 (9.9) 69.3 (9.4) 0.492

Years of

education

9.9 (4.1) 10.4 (4.9) 9.4 (2.9) 9.3 (3.2) 0.894

Disease

duration

(months)

22 (23) 20 (16) 27 (33) 19 (11) 0.866

MMSE 28.8 (1.3) 28.6 (1.4) 29.0 (1.1) 25.5 (1.1) 0.496

UPDRS III 24.8 (10.6) 27.5 (10.5) 21.1 (9.9) 23.5 (10.3) 0.020

31

Differences in demographic, clinical and cognitive characteristics between

the PIGD, tremor and indeterminate group were analyzed with two-tailed

ANOVA, Kruskal-Wallis, chi-square or Fisher’s exact test when appropriate.

The nonparametric Spearman’s rho was used to explore correlations

between demographic, motor and neuropsychological variables. Multiple

linear regression analysis was performed to explore if the relationships

found in the correlation analysis between motor and cognitive variables were

unique or affected by other variables. To meet the assumptions of normality

and reduce skewing and outlier influence, some variables were transformed

with square root transformation or logarithm transformation. Most variables

were approximately normally distributed after transformation.

Paper III

The persistence of associations between motor score and cognitive functions

were investigated together with the magnitude of cognitive change between

baseline and follow-up for the total group and for patients receiving different

dopaminergic medication (pramipexole ± levodopa and without dopamine

agonists). Data from baseline and 12 month assessment were used.

Neuropsychological assessments were performed in 122 of 150 patients that

fulfilled the diagnostic criteria for PD. Six patients were excluded due to: not

enough tests performed (n=1), having dementia at baseline (n=1), normal

SPECT FP CIT scan (n=3) or severe depression according to MADRS (n=1).

Of the remaining 115 patients 88 (76%) participated in the 12 month

neuropsychological follow up. Twelve patients had started their medication

for parkinsonism before the baseline evaluation and were therefore excluded

from the analysis of the different effects of dopaminergic medication. Five

patients did not receive either levodopa or dopamine agonist at follow-up.

Two patients that received ropinirole instead of pramipexole were excluded

from the medication analysis (Table 6).

32

Spearman’s rho’s was used to assess correlation in change between baseline

and 12 month follow-up in cognitive and motor variables for the total group.

The analysis was based on the 12-month follow-up score subtracted from the

baseline score. For all cognitive variables, except for TMT A and TMT B, a

negative score meant improvement. For all motor variables a negative score

meant decline in function. Thus, a negative correlation between motor and

cognitive scores meant that an improvement in motor function corresponded

to an improvement in cognitive function, except for TMT A and B. The

cognitive variables that showed a cognitive-motor relationship in study II

and the cognitive and motor variables that correlated in their differences

between baseline and the 12 month follow-up were used as dependent

variables in separate linear regressions. The motor variable that correlated

with the cognitive variable of interest was used as the independent variable.

Age at baseline, years of education, LED and baseline score of the dependent

variable were used as covariates.

To assess the differences between baseline and the 12 month assessment

paired sample t-tests, Mann-Whitney U-tests, chi-square or Fisher’s exact

test were applied when appropriate. Change in cognitive performance

depending on the dopaminergic medication (dopamine agonist ±levodopa

vs. no agonist) was explored. Cognitive variables that had shown to be

affected by the intake of pramipexole in a previous study (FCSRT

recognition, verbal fluency, category fluency, TMT A and TMT B) [90] was

analyzed with univariate analysis of covariance (ANCOVA). The follow up

score of the variable of interest was the dominant variable; medication was

Total,

mean±SD

n=88

Total,

mean±SD

n=74

Pramipexole

mean±SD

n=34

No agonist

mean±SD

n=40

p-value

Sex

(men/women)

57/31 48/26 21/13 27/13 0.607

Age 67.5 (9.3) 67.5 (9.2) 62.0 (7.5) 72.2 (7.8) <0.001

Years of education 10.2 (3.8) 10.3 (3.8) 11.8 (4.0) 8.95 (3.1) 0.001

MMSE 28.8 (1.2) 28.8 (1.3) 28.9 (1.2) 28.7 (1.3) 0.464

UPDRS III 24.8 (10.2) 24.3 (10.3) 23.7 (10.2) 24.9 (10.5) 0.620

MADRS 4.2 (3.3) 4.4 (3.2) 4.1 (3.7) 4.7 (2.8) 0.408

LED (12 month) 331 (174) 317 (173) 312 (172) 320 (175) 0.486

Medication

L-dopa 62 (70%) 50 (68%) 15 (44%) 35(88%)

Pramipexole 39 (44%) 34 (46%) 34(100%) 0

Selegeline 6 (7%) 4 (5%) 2 (8.8%) 2(5%)

TABLE 6. Demographics and baseline measures of the total group (n=88) and the

74 patients drug naïve at baseline, by type of dopaminergic treatment in Paper III.

MMSE= Mini Mental State Examination, MADRS=Montgomery and Åsbergs Rating Scale;

UPDRS= Unified Parkinson’s Disease Rating Scale. Reprinted from Paper III with kind

permission from the copyright holders.

33

TABLE 7. Baseline characteristics for the total group and for MCI+ and MCI- in

Paper IV.

UPDRS=Unified Parkinson’s Disease Rating Scale, MADRS=Montgomery and Asberg

Depression Rating Scale (MADRS), MMSE= Mini Mental State Examination

entered as between subject factors and the baseline score of the variable of

investigation, age, years of education and LED were entered as covariates.

Interaction effects between medication and the covariates were checked

before the full factorial models were executed. Further, the proportion that

declined in variables with significant differences in the different medication

groups were calculated and a Chi square test was performed to evaluate if the

groups differed (two or more words less at follow-up vs. those with

improvement or smaller change). U-shaped associations between motor and

cognitive variables were checked separately for each medication group in a

regression model with cognitive variables as dependent variable, squared

change in motor score (motor²) as independent variable and change in

motor score as covariate.

Paper IV

Neuropsychological investigations were performed in 119 of 150 patients that

fulfilled the diagnostic criteria for PD. Eight were excluded, four had normal

presynaptic dopamine uptake, two could not perform tests and two had

severe depression. Thus 111 patients with PD were included. There were 17

(15 %) participants that had started their dopaminergic treatment before the

baseline assessment and these had been on dopaminergic medication for an

average of three months. Baseline, one year, three and five year follow up

data for the patients were included. (Table 7)

Total MCI-

(n=71)

MCI+

(n=40)

p-value

Age 68.7 (9.4) 67.4 (9.4) 71.0 (9.1) 0.048

Years of education 10.1 (4.1) 10.6 (4.4) 9.3 (3.5) 0.088

Sex (m/f) 68/43 38/33 30/10 0.026

Duration (months) 22.2 (22) 22 (22) 21 (15) 0.673

MMSE 28.7 (1.3) 29.1 (0.9) 28.0 (1.6) <0.001

MADRS 4.3 (3.4) 3.8 (3.5) 5.2 (3.1) 0.053

Started dopaminergic

treatment

15% 13% 20% 0.304

UPDRS III 25.4 (10.9) 23.4 (10.3) 28.9 (11.0) 0.012

34

Proportions and the annual incidence of dementia with a 95% confidence

interval (CI) were calculated. The time of dementia were estimated as the

midpoint between assessments that dementia was diagnosed. Person-years

at risk were calculated as the total follow up time until dementia or study end

for those not developing dementia. Incidence was calculated as the number

of cases with dementia divided by the total number of person years at risk

during follow up. To study the association of age, education, PIGD sub-type

and MCI to the development of dementia a Cox-proportional hazard model

was performed.

Patients with MCI (MCI+) that converted to dementia were compared to

patients with MCI who did not convert on baseline-demographics or

cognitive and -motor measurements. Student’s t-test, Mann-Whitney U-tests

and chi-square tests were used when appropriate.

Population average regression models for correlated data i.e. Generalized

Estimation Equations (GEE) were applied to investigate the association of

dementia to the evolvement of motor scores. GEE was chosen because it can

model non-normal distributed data and take non-independent

measurements such as repeated measurements or clustered data into

consideration. It is also flexible with unbalanced and missing data [121]. An

autoregressive model with Tweedie log link function was used. The

dependent variable was the motor score (posture/gait, bradykinesia, tremor,

rigidity or bulbar) in five separate models. The factors were set to dementia

status (dementia converters and non-converters) and follow-up (baseline,

one year, three years and five years). No dementia and baseline testing were

set to be the redundant parameters to which the other parameters were

compared. Covariates were age at baseline and LED. The significance level

was set to p<0.01 because of five dependent measures being tested.

35

TABLE 8. Differences between patients with impaired cognition and intact cognition.(Paper I)

MMSE= Mini Mental State Examination, MADRS=Montgomery and Åsbergs Rating Scale;

UPDRS= Unified Parkinson’s Disease Rating Scale. Reprinted from Paper I with kind

permission from the copyright holders.

Results

Paper I

Patients with PD performed significantly worse than healthy controls in tests

measuring episodic memory (free recall), attention, psychomotor function

and category fluency. Thirty% of the patients were impaired in one or more

cognitive domain and 16% were impaired in two or more cognitive domains.

Impairment was evenly distributed between episodic, executive and verbal

functions. Speech, facial expression, rigidity and bradykinesia were

significantly more affected amongst the cognitively impaired patients.

Compared to patients with intact cognition, patients with impaired cognition

had lower scores on the MMSE, shorter disease duration and more motor

problems (facial expression, speech, bradykinesia and rigidity) (Table 8).

The strongest predictor for impairment in two or more cognitive domains

was years of education (10.3 years for <2 vs. 7.8 years for >2 cognitive

domains impaired, p=0.01).

1≥ Impaired domains

n=26

Intact cognition

n=60

P-value

Sex (female/male) 8/18 29/31 0.131

Age, years 65.9 (8.4) 68.9 (9.7) 0.084

Years of education 9.8 (3.4) 9.9 (4.5) 0.659

MMSE 28.0 (1.8) 29.0 (1.1) 0.018*

MADRS 5.4 (4.0) 3.8 (3.7) 0.078

UPDRS III 28.3 (10.4) 21.8 (8.9) 0.004**

Speech 0.8 (0.7) 0.3 (0.5) <0.001**

Facial expression 1.5 (0.8) 0.8 (0.8) 0.001**

Tremor 2.8 (2.7) 2.8 (2.4) 0.775

Rigidity 7.4 (4.2) 4.8 (3.6) 0.004**

Bradykinesia 9.8 (4.4) 7.8 (3.8) 0.029*

Axial impairment 2.6 (2.1) 2.3 (1.5) 0.507

36

TABLE 9.Results for linear regression analysis to assess the association between motor and cognitive performance with age, sex education and other motor signs controlled for. (Paper II)

1log transformation, 2square root stransformation, β=beta, R2=how much of the variance in

the dependent variable is explained by the independent variables. WCST Wisconsin Card

Sorting Task, TMT Trail Making Test, BVMT Brief Visuospatial Memory Test, FCSRT Free and

Cued Selective Reminding Test, BNT Boston Naming Test. Reprinted from Paper II with kind

permission from the copyright holders.

Paper II

No significant differences were found between PIGD, tremor or

indeterminate subtypes in their cognitive test performance at baseline.

Instead there was an association between different cognitive test scores and

different motor functions measured by the UPDRS III. Correlation analysis

showed that bradykinesia, posture/gait score and bulbar function were

correlated to a range of cognitive functions. Tremor did not correlate to any

cognitive measures. After controlling for age, sex, education and the other

motor signs bradykinesia was associated with WCST, digit span and TMT B.

Postural instability and gait function were associated with visuospatial

function and visuospatial episodic memory while bulbar dysfunction were

associated with episodic verbal memory (Table 9).

R² Standardized Β p-value

Bradykinesia

WCST category completed 0.232 -0.246 0.022*

Digit span 0.214 -0.288 0.002**

TMTB² 0.444 0.197 0.038*

PIGD²

Line orientation 0.187 -0.251 0.016*

BVMT total 0.364 -0.210 0.022*

Bulbar dysfunction²

FSCRT recognition 0.330 -0.329 0.001**

FSCRT total¹ 0.248 -0.373 0.001**

Mental control 0.162 -0.299 0.003**

37

Paper III

The associations found in Paper II partly remained when looking at change

between baseline and 12 months. Change in postural instability and gait

problems correlated with change in visuospatial memory and change in

bradykinesia correlated with change in working memory after controlling for

age at baseline, years of education, and LED at the 12-month follow-up

(Table 10).

No differences in the total group were found between baseline and follow-up

in any of the cognitive variables, except for a small improvement in MMSE

(baseline mean 28.8, follow-up mean 29.2, p=0.025) and MADRS (baseline

mean 4.37, follow-up mean 3.34), p=0.017). Significant changes in the motor

variables were seen in UPDRS III total (baseline mean 24.4, follow-up mean

22.0), p=0.014) as well as for tremor (baseline mean 2.5, follow-up mean

1.49, p=<0.001) and posture/gait sub-score (baseline mean 2.36, follow up

1.88, p=0.008). The only cognitive test that significantly differed between

medication groups was phonemic fluency [F (1.67) =5.244 p=0.025, partial

eta square=0.073, ANCOVA]. Patients receiving pramipexole± levodopa

performed significantly worse at the one-year follow-up (baseline mean: 43.7

words, follow-up mean: 38.5). The patients that did not receive pramipexole

showed a non-significant improvement between assessments (baseline

mean: 36.2 words, follow-up mean: 37.5). In the pramipexole± levodopa

group 74% produced two or more words less at follow-up compared to only

31% of those not receiving pramipexole. For patients receiving pramipexole

the correlation between bradykinesia and digit span forward was U-shaped.

Standardized B 95% CI for B p-value

PIGD

BVMT total -0.272 -1.601- -0.190 0.014*

Line orientation -0.099 -0.658- 0.254 0.381

Bradykinesia

Digit span -0.300 -0.311- -0.050 0.007**

TMT B 0.243 0.212-7.315 0.038*

BVMT total -0.160 -0.496-0.082 0.159

Digit span forward

(n=34)

Bradykinesia2 -0.473 -0.047—0.006 0.014*

Bradykinesia -0.042 -0.159-0.126 0.818

TABLE 10.Results for linear regression analysis to assess the association between change in motor and cognitive performance in 88 patients with PD. after controlling for baseline age, years of education and LED.

PIGD=postural imbalance and gait disturbance. BVMT=Brief visuospatial memory test.

FCSRT=Free and Cued Selective Reminding Test. TMT B=Trail Making Test part B. *p-

value<0.05, **p-value<0.01, Bradykinesia2 statistics based on only patients receiving

pramipexole. Reprinted from Paper III with kind permission from the copyright holders.

38

That is, patients with less bradykinesia at follow-up either improved or

deteriorated in their working memory performance.

Paper IV

Twenty-seven patients with PD (24%) developed dementia within five years,

corresponding to an annual incidence of 56.0 per 1000 person-years (95%

CI, 37.2-80.3). Forty patients (36%) had MCI at baseline of which 22 (55%)

transitioned to dementia, corresponding to an annual incidence of 148 per

1000 person-years (95% CI, 94.9-215.0) in patients with MCI. Five (7%)

patients with normal cognition at diagnosis transitioned to dementia

corresponding to an annual incidence of 15 per 1000 person-years (95% CI,

4.9-34.6). Seven patients with MCI at baseline reversed back to normal

cognition and three patients shifted between MCI and normal cognition

between assessments. The sensitivity for MCI to predict dementia in early

stages of PD was 81%, the specificity 79%, the PPV 55% and the NPV 92%.

MCI converters performed significantly poorer than non-converters on

several cognitive measures at baseline: MMSE, FCSRT free recall, BVMT,

mental control, line orientation, animal fluency, and TMT B.

MCI (B=-2.486, Exp(B)=0.083, CI for Exp(B) 0.031-0.225, p<0.001) and

baseline age (B=1.077, Exp(B)=2.936, CI for Exp(B) 1.23-6.99, p=0.003)

were both significant predictors for dementia. Education and motor subtype

at baseline were not significant predictors of dementia. None of the patients

with the tremor subtype at the latest visit had developed dementia

There were significant negative changes for posture and gait impairment

(Wald Chi-Square 23.862 p<0.001), bulbar dysfunction (Wald Chi-Square

20.627, p<0.001), and bradykinesia (Wald Chi-Square 54.963, p<0.001)

during the follow-up period. Dementia converters had more bradykinesia

(Wald Chi-Square 25.035, p<0.001), rigidity (Wald Chi-Square 20.376,

p<0.001), bulbar dysfunction (Wald Chi-Square 12.557, p<0.001) as well as

posture and gait impairment (Wald Chi-Square 16.231, p<0.001). There was

an interaction effect between follow-up and dementia for posture and gait

(Wald Chi-Square 14.523, p=0.002); i.e. patients developing PDD differed

from PD patients that did not develop dementia in their evolvement of

posture and gait impairment (Figure 6a). The interaction effect for follow up

and dementia were close to significant for bradykinesia (figure 6b) between

baseline and three years (p=0.02), and for rigidity (figure 6c) between

baseline and five years (p=0.01) but not significant for the total follow up

period (rigidity: Wald Chi-Square 8.214, p=0.042, bradykinesia: Wald Chi-

39

Square 6.449, p=0.092). The evolvement of tremor and bulbar dysfunction

did not significantly differ between patients.

FIGURE 6. The evolvement of a)

postural instability/gait

disturbances, b) bradykinesia and

c) rigidity in patients converting

to PDD compared to those who

did not. Significant p-values

indicate interaction between time

and dementia status compared to

baseline.

40

Discussion

Main findings

The main focus of this thesis is cognitive function in early PD and its

association to motor disturbances. In line with previous studies [6–8, 98] we

found, in comparison with healthy controls, that cognitive impairment is

common in newly diagnosed patients with PD and that a range of functions

are affected already at diagnosis. Patients with cognitive impairment at

baseline had significantly more motor problems according to the UPDRS III,

than those with intact cognition, and education was an independent

predictor for impairment in two or more cognitive domains at baseline but

did not predict dementia within five years.

Associations were found between bradykinesia and worse performance on

tasks measuring working memory and executive functions as well as between

posture/gait dysfunction and lower scores on tests measuring visuospatial

function and visuospatial memory. Some of these associations persisted also

after one year.

MCI at the time of diagnosis was found in 36% when diagnosed by recently

published MDS Task Force criteria (Paper IV). Thirty% were classified as

having cognitive impairment when using slightly different criteria (Paper I).

One fourth of the patients developed PDD within five years after diagnosis.

Age at diagnosis and MCI according to MDS Task Force criteria were

significant predictors of dementia. There were 25 % of those classified as

having MCI at baseline that converted to normal cognition or reversed back

and forth over the 1.5 SD limit between assessments.

Patients with MCI converting to PDD within five years had worse

performance on several cognitive tests at baseline: visuospatial function,

semantic fluency, episodic memory, conceptual tracking and mental

flexibility; but not in tests measuring executive function, working memory,

verbal phonemic fluency and language function.

There were no differences in baseline measures of cognitive function

between PIGD, tremor and indeterminate phenotype. PIGD phenotype at

baseline did not predict dementia within five years. None developed

dementia among those having a tremor dominant phenotype. Worsening of

posture/gait was more pronounced in patients developing dementia than in

patients that did not. A similar pattern was seen in bradykinesia and rigidity

but not for tremor and bulbar dysfunction.

41

Patients treated with the dopamine agonist pramipexole performed

significantly worse on a measure of verbal fluency at the one year follow up

compared with patients not treated with the drug.

Impaired cognitive functions

Compared with healthy controls, patients with PD performed significantly

worse already at the time of diagnosis on tests measuring attention, semantic

fluency and episodic memory (free recall). Attention/working memory and

episodic memory were the domains mostly affected followed by executive,

visuospatial and verbal function. This shows a diversity of cognitive decline

within patients with PD, in line with previous studies [6].This could be

expected considering the heterogeneity of the disease.

A diverse pattern of differences in baseline test performance between

patients with MCI converting to dementia compared with those not

converting to dementia was shown in Paper IV. WCST, a measure of

executive function, together with working memory, phonemic fluency and

verbal naming did not significantly differ between MCI converters compared

with non-converters at baseline. Interestingly many of the same tests did not

differ between patients and controls in Paper I.

Previous research has connected executive impairment [122, 123], and

memory [123] as predictors of PDD. Neuropsychological measures with a

more posterior cortical basis such as pentagon copying and semantic fluency

have been suggested to be predictors of dementia [124]. A recent study

comparing MCI converting to dementia within two years with those that did

not convert proposed that frontal lobe associated performance together with

atrophy in both frontostriatal areas and cholinergic structures, and not

visuospatial function, were predictors of dementia in PD-MCI [125].

Tests that significantly differed between MCI converters and non-converters

in Paper IV were tests measuring visuospatial function and semantic fluency

in line with Williams-Gray et al. [66], but also tests measuring episodic

memory free and delayed recall, conceptual tracking, and mental flexibility

where the latter tests are believed to be more frontally mediated.

Together this shows a diverse picture of cognitive function associated with

PDD with a range of different cognitive functions that are different between

MCI converters and non-converters. We need to look at both the striatal

network in combination with a more widespread cortical network in the

search for predictors for PDD.

42

Proportions of MCI and Dementia

MCI

The proportion of patients that were classified as being cognitively impaired

differed somewhat between Paper I (30%) and Paper IV (36%). Different

classification procedures with stricter criteria and removal of TMT A and

TMT B from the classification battery in Paper I and the use of MDS Task

Force criteria in paper IV were the reasons for this. The MDS Task Force

criteria had not yet been published when paper I was assembled. The

differences of results within the same study population depending on which

classification is used show that MCI classification relies on the tests included

and the cut-off values used and emphasizes the value of joint criteria for PD-

MCI, in research and in clinical practice.

Proportions of MCI in newly diagnosed cases with PD published prior to the

MDS Task Force criteria have ranged between 19 to 36% [6–8, 52, 53, 126].

The proportions of patients classified as MCI according to the MDS Task

Force criteria in newly diagnosed PD have been 25 to 35% [57, 58, 127]. This

makes the proportion of 36% in Paper IV at the higher range.

Dementia

In cross-sectional studies the prevalence of dementia has been around 30%

and increases over 20 years to 76% [128]. The majority of studies have been

based on prevalent PD cases with different disease duration. Following

patients from disease onset gives a more accurate estimate of dementia rates

and give a better base for determining the dementia risk. Studies on

prospectively studied newly diagnosed cases with PD are rare, and the

proportion of patients developing PDD in such studies within five years has

been 17% [32, 129] [incidence 38.7 per 1000 person-years (95% CI, 23.9-

59.3)] [32]. This makes the proportion of patients developing dementia

(24.3%) and the incidence rate (56 per 1000 person-years) slightly higher

than reported in previous studies on newly diagnosed cases, but longer

follow up times have rendered incidence rates similar to our study [54.7 per

1000 person-years (95% CI, 35.4-74.1)] [66].

Higher dementia and MCI rates

There could be several reasons for the high dementia and MCI rates in this

cohort. Cognition was mapped extensively and was more thoroughly

investigated than in most other longitudinal studies [58, 60]. Our patients

were older [57] and they had less education [57, 58]. Furthermore, the

43

annual incidence of PD reported from the NYPUM cohort is among the

highest reported in the literature [15]. This could be due to meticulous case

finding strategies that might also influence the proportion of patients with

cognitive impairment.

Predictive value of MDS-Task Force MCI criteria for PDD

The MDS Task Force MCI criteria were found to have a relatively good

sensitivity and specificity where 22 of 40 (55%) patients with MCI developed

dementia within 5 years. There are two longitudinal studies on newly

diagnosed patients with PD using MDS Task Force guidelines for MCI [57,

58]. A three year follow up where ten out of 37 (27%) patients with MCI

converted to dementia [58] and a five-year follow-up where 11 out of 43

(26%) converted [57]. The three-year study had an annual incidence rate of

dementia of 98.9 per 1000 person-years in patients with MCI at diagnosis.

After three years there were 11 out 40 patients with MCI at baseline that had

converted to dementia in our study and only one of 71 of those with normal

cognition.

Half of the patients that were classified as having MCI at baseline did not

develop dementia within five years. With prolonged follow-up the PPV might

increase as more patients with MCI may develop dementia, the NPV will

most likely decrease as more patients classified as having normal cognition

at baseline will probably also convert to dementia.

During follow-up 25% that were initially classified as MCI at baseline

reversed to normal cognitive function or shifted between normal cognition

and MCI between follow-up. In those cases the MCI at baseline might be due

to reversible causes, such as nervousness, depression or sleep deprivation,

rather than neurodegenerative. In some cases the cognitive improvement

could also be due to the intake of dopaminergic medication [130].

Previous studies in the general population have shown a reversion rate from

MCI to normal cognition between 13-41% [131–133]. A recently published

study using MDS Task Force criteria reported a reversion rate from MCI to

normal cognition to around 20% [58]. A study from Norway suggests that

patients with PD classified as MCI both at baseline and after 12 months have

a higher proportion of MCI converting to dementia as well as a lower

proportion of MCI reversing back to normal cognition than for those having

MCI at baseline or at the 12-month follow-up [58].

44

Education, age, disease duration and gender aspects

In Paper I shorter length of education was a predictor for having more than

two cognitive domains impaired. Less education did not predict the

development of dementia. Previous longitudinal studies on non PD

participants have found that short education does not increase the risk of

developing dementia [134, 135] but Kryscio et al. [134] found that education

significantly predicted transition from normal cognition to MCI.

Our finding of shorter disease duration among those with impaired cognitive

function in Paper I suggests an aggressive PD phenotype with cognitive

decline early in the disease. However in Paper IV we found no significant

differences in disease duration between those with MCI at baseline

compared to the rest. Disease duration did not differ between those

developing dementia compared with those who did not. This is in line with a

prospective population-based study that through separating general age

from age at motor onset showed that patients with PD have a similar

dementia age regardless of when they first are presented with motor

symptoms [136].

The only significant gender difference found were the difference between

how many were classified as MCI according to the MDS Task Force criteria

where 75% of those with MCI were male. The reason for this is hard to grasp

since this has not been shown in previous studies. There could be gender

differences in whether to agree to participate in the neuropsychological

testing if one experiences cognitive decline. One explanation could be that

female patients with PD have been suggested to more often be presented

with the tremor dominant phenotype [137]. However, there were no

significant differences in motor subtypes between male and female

participants in the present study.

Cognitive motor relationship

A pattern of motor cognitive associations already in early stages of PD could

be depicted from the connection found between motor and cognitive

measures, and the different motor and cognitive variables associated with

MCI and dementia. Bradykinesia was associated with worse working

memory performance and executive functions whereas postural instability

and gait disturbances were associated with visuospatial memory and

visuospatial function. Tremor showed no association with any cognitive

measures either at baseline or in the change over time. Bradykinesia, rigidity

and postural and gait disturbances were all more severe in those with MCI

45

and those developing PDD. Furthermore, there was a more pronounced

worsening of motor functioning over time in those developing dementia.

Some cross-sectional studies have investigated the cognitive motor

relationship in PD [4, 97, 98, 138]. Most of the earlier studies have been

rather small and included prevalent cases with various disease duration that

have received dopaminergic treatment of various duration. It can be

problematic to demonstrate motor cognitive relationships in medicated

patients since dopaminergic medication has shown to affect motor and

cognitive abilities differently. Correlation between the UPDRS motor score

and a range of neuropsychological measures, but not visuospatial function

has been reported [139], although a recent study using objective measures of

motor dysfunction connected visuospatial function to postural instability in

PD [94].

The association of bradykinesia with impaired working memory

and executive function

Both bradykinesia and decline in executive functions and attention/working

memory are common in PD. Bradykinesia [22], working memory [37] and

executive functions have been linked to the frontostriatal brain network and

further linked to nigrostriatal dysfunction. The association of bradykinesia

with executive dysfunction and attention/working memory dysfunction

found in Paper II supports the possibility of joint underlying networks.

Patients classified as having cognitive impairment (Paper I and Paper IV)

had more severe bradykinesia at baseline which suggests dopaminergic

involvement in MCI in early stages of PD. A recent cross-sectional functional

magnetic resonance imaging (fMRI) study of early PD linked cognitive

impairment in PD to frontostriatal dysfunction by showing under

recruitment in dorsal caudate nucleus and bilateral anterior cingulate while

performing an updating task in patients with PD-MCI [140].

Patients that developed PDD had more bradykinesia and rigidity than those

not developing dementia. This also indicates early frontostriatal

dopaminergic involvement in PDD.

The association of posture and gait disturbances with impaired

visuospatial function and visuospatial memory

Significant associations between postural and gait scores and visuospatial

memory and visuospatial function were found and further confirmed in the

46

association between change in posture and gait score and visuospatial

memory in Paper III.

PIGD/tremor and indeterminate subtype did not significantly differ in any

cognitive tests at diagnosis. The evolvement of postural instability and gait

scores differed between those developing dementia compared with those

who did not, with a significantly more pronounced decline in those

developing dementia. We showed in line with previous results that PIGD

subtype at the time of diagnosis did not predict the development of dementia

[66]. Together this shows that PIGD phenotype does not seem to be a

predictor of cognitive impairment in early stages of PD or a predictor of

developing dementia. It rather seems to be a parallel process with decline in

visuospatial function and memory and the development of PDD. In the

general population poor gait and cognition have been connected [141].

Dementia and falls often coexist and gait impairments have been related to

the severity of cognitive impairments [142].

Previous studies have shown that PIGD phenotype at diagnosis predicts

worse outcome in PD [66, 143]. PIGD has also been shown more common

among patients with worse cognitive performance [144] and has been

identified as a risk factor for dementia [96]. Both previous studies [96, 144]

included cases in various disease stages that had started their dopaminergic

medication before the first assessment. In line with our results is a study by

Alves et al (2006), which suggested that changing from tremor to PIGD

subtype is a risk factor of dementia and that having a persistent tremor

phenotype was protective of dementia [67].

A recent study found a connection between increased neocortical ß-amyloid

deposit and higher PIGD features in PDD which could possibly be one

underlying factor explaining the association of postural instability and gait

disturbances with impairment in visuospatial function and memory and later

development of dementia [27]. A specific association of gait dysfunction,

cognitive impairment and reduced cholinergic activity has been described

with several lines of evidence (see Ambioni et al. for review) [145].

Other explanations could be the association of degeneration of the

pedunculopontine nucleus (PPN) and the lateral pontine tegmentum area

(rich in acetylcholine) in gait disturbances [146]. Impaired cholinergic

integrity of the PPN has been associated with frequent falls in PD [147]

47

The effect of dopaminergic medication

Study III showed that there were some differences in change in cognitive

functions over one year (phonemic fluency) and the association of the change

between motor and cognitive functions (digit span forward and

bradykinesia) in patients receiving the dopamine agonist pramipexole

compared to patients who did not. Since we did not compare different

agonist we cannot tell whether our finding is a specific effect for pramipexole

or an effect that could also be found in other DA. Pramipexole is believed to

have a high affinity for D3 receptors [148] which could be connected to both

findings. A possible explanation for the decline in verbal fluency for patients

receiving DA could be that D3 auto receptors have been shown to contribute

to the presynaptic regulation of tonically released dopamine which leads to a

moderate inhibitory action on locomotion [148]. If this is also the case for

cognitive function, this could result in an inhibitory effect on word

production.

The U-shaped association for change in bradykinesia and working memory

could possibly be explained by the U-shaped association between dopamine

levels and cognitive functions. Bradykinesia has been described as the best

measure of the nigrostriatal lesion [22] and is related to the dorsal striatum

that is severely dopamine depleted in the early stages of the disease. There is

a high density of D3 auto receptors in the ventral striatum that is less

dopamine depleted early on in the disease. Therefore the U-shaped

association for patients receiving DA perhaps indicates that pramipexole is

more likely to impair working memory by overdosing than other types of

medication.

These findings have to be interpreted with caution. A lot of comparisons

were made and the findings could because of this be spurious. Still the

finding could be indicative of a possible influence of DA on certain aspects of

cognition that needs to be further investigated. In line with our result Brusa

et al. showed in 20 right handed patients with PD that pramipexole

produced significant impairment of short term verbal memory, attentional

executive functions and verbal fluency [90]. None of the previous studies

that have investigated the more prolonged effect of DA (6-24 months) has

included patients that received pramipexole as monotherapy. Patients

receiving pramipexole differed from the other patients in age and baseline

performance which could have influenced the results. This was because the

decision on the type of dopaminergic medication used was made by the

treating physician based on personal experience and the common practice of

using dopamine agonist in younger onset patients (before 65-70 years of

age).

48

Ethics

Studying elderly clinical populations in which a large proportion of the

patients eventually will develop dementia renders certain demands of ethical

considerations. Both cognitively impaired and borderline cognitively

impaired patients with PD have impairments in their decisional capacity

measured by the MacArthur Competence Assessment Tool for Clinical

Research [149]. This might be especially problematic when enrolling patients

with MCI whose reduced decisional capacity might be less obvious. It is also

important to note that many of the patients will change in their decisional

capacity during the course of the study. When enrolling patients in a

prospective study with repeated investigations it can be hard for the patient

to grasp how the extensive assessments will interfere with their daily life at

present, but even more so in the future, especially for a patient that is not

functioning at an optimal cognitive level.

Since the NYPUM study is implemented at a hospital there is close contact

between clinic and science. A clinical study where many health care persons

are involved can make the participants feel pressure to give consent and

continue in the research project. It is important to make clear that the

decision whether to remain in the study or not should under no

circumstances influence the future care of the patient.

Studies have shown that incidental findings of significant clinical importance

in fMRI research are around one-two % [150]. There is probably an even

higher proportion in elderly populations. This highlights the importance of

having a plan for taking care of possible findings and for bringing back the

results to the patient. One example of an ethical challenge could be whether

the participant should be told that they are a part of a risk group.

Methodological considerations

Study design

The main strength of the study is the population-based approach with a close

to complete identification of cases (making generalization of results possible)

and an extensive assessment of neuropsychological functions within a few

weeks of diagnosis. This allows testing of almost all patients prior to the

introduction of pharmacological treatment with dopaminergic agents. The

neuropsychological evaluation is repeated after one, three and five years and

can provide knowledge of the prognosis of cognitive function in PD. It also

makes it possible to identify factors at diagnosis that are predictive of

different outcomes and thereby can imply causality. This is in contrast to

49

cross-sectional investigations that only can confirm associations. One

limitation is the lack of histopathologically confirmed diagnosis.

Loss to follow up and missing values

A methodological problem seen in longitudinal studies of the elderly is a

cohort bias, i.e. selective loss to follow-up. There are many drop-outs, people

are old and die, they have a chronic progressive disease that affects them to

such a degree that they are unable to participate at follow-up. In the present

study already at the baseline testing, the dropouts differed from those

participating in the neuropsychological investigation. This is a problem when

describing the true progression for the entire population with PD. The

picture provided here is probably a little bit too optimistic. In the group that

developed dementia there were only five patients performing the five year

neuropsychological follow up. Therefore describing the longitudinal change

in the cognitive measures would be misguiding. Fortunately there were more

evaluations of these participants and most of them remained in the study

allowing evaluation of cognitive function/dementia even when they did not

perform the extensive neuropsychological test battery. This particular loss to

follow-up is hard to avoid but one way of minimizing the skewed loss to

follow-up is to perform home visits for the patients who are too tired to come

to the hospital.

There is also loss of individual measures due to tiredness and technical

issues. WCST was the test with most missing values which could be one

explanation why the result of WCST did not differ between the groups. We

did not omit patients who had not performed isolated tests. Since the

missing values were rather few for most variables, and distributed in

different cases, we used pairwise deletion of the data (except for a few cases).

The drawback with this method is that the parameters of the model will be

based on different set of data. The positive side is that we do not lose as

much power as we would have if all cases would have been omitted.

Statistics

A lot of variables and comparisons were used in the studies which increases

the risk for type I errors. We did not want to exclude possible true

relationships by adjusting for alpha error inflation by applying the strict

Bonferroni adjustment [151]. The risk for Dementia in PD seems to be

exponential, i.e. there is an exponential increase of new dementia cases with

age. This is problematic when using the Cox proportional Hazard model.

50

Diagnosis of PD

The number of patients with PD at baseline is different in all the four papers.

This is due to differences in the number of patients that had been included at

the time of the particular paper and to changes in diagnosis in some patients

as the disease progressed. Since the diagnosis of PD is more accurate as time

progresses and diagnostic criteria that are based on what happens

prospectively, e.g. the response to dopaminergic treatment, the diagnosis is

destined to change in some individuals.

UPDRS

The total score of UPDRS III is a validated and reliable tool for measuring

the progression of motor dysfunction in PD. The division into the different

motor components (e.g. bradykinesia, rigidity etc.), even though used in

several studies [106], is not validated as used in the thesis. UPDRS is a

clinical scale that relies solely on the investigators judgment. An alternative

to the clinical measures could be to use quantitative measures of the

different motor functions which could also enable description of other

aspects of motor functions. This could give a more precise picture into what

part of movement that is correlated with different cognitive functions.

SWEDDs

In patients clinically diagnosed with PD four-15% have normal imaging of

dopaminergic function (Scans Without Evidence of Dopaminergic Deficit

[SWEDDs]) [152]. SWEDDs patients are not likely to progress in severity or

develop motor and non-motor complications and it is highly likely that they

do not have PD at all [153]. We have treated the SWEDDs differently in the

different papers. In Paper I SWEDDs were included in the analysis. In Paper

II-IV the SWEDDs were excluded. The reason for this is that we did not have

as many prospective scans in Paper I and therefore we could not tell if the

normal datscan was an expression of early disease or if they were actually

true SWEDDs. Patients in Papers II –IV had been repeatedly scanned for

three to five years when the studies were assembled and therefore the

SWEDD classification was considered more stable.

MCI and Dementia

We tried to find variables that differed between MCI that developed

dementia and those who did not as a way of dissociating patients developing

early dementia from the rest of the group at diagnosis. The question that we

hoped to answer was how patients with MCI that converted to dementia in

51

the early phase differed from those who did not. It is likely that additional

patients from the MCI group will develop dementia therefore the specificity

and sensitivity for MCI is only for the early phase of PD, i.e. up to five years

after diagnosis.

Motor fluctuations

The occurrence of dyskinesias and motor fluctuations is common, and in the

long term management most patients will be affected by those treatment-

related side effects [154]. There is some evidence that cognitive performance

together with arousal state might be affected during motor fluctuations in

PD [155]. This is important when studying patients with PD for a longer

period of time. In the present study the ambition was to avoid testing if

substantial dyskinesias or other wearing-of symptoms were present. In the

cases where this was impossible or if the testing for some reason was

performed anyway, this was noted in the test protocol. Since the study

population was in the early phase of PD, dyskinesias and motor fluctuations

were minor problems.

Neuropsychological testing

Different neuropsychological tests aim to capture certain aspects of cognition

and manage to do so to a certain degree. However, it is complicated to apply

a reductionist view when looking at cognition since different cognitive

functions does not operate separately from each other. This is important to

have in mind when using standardized neuropsychological tests. It is obvious

in the PD literature that researchers are using different cognitive tests to

measure the same function. There are also examples were different studies

claim that a certain task measures different abilities. This is one contributing

factor to the various results between studies.

Some of the tests used in the studies require motor function such as both

TMT A and B. TMT B but not TMT A is correlated with bradykinesia which

indicates that it is likely that it is mental flexibility that contributes to the

association. BVMT is a test that requires drawing which can indeed be

influenced by especially tremor. According to the instructions for the test, for

example shaky handwriting should be overlooked in the judgment of

correctness of the figures.

The aim for the present study was to assemble a test battery that captured

different functional levels and could be performed even when functions start

to deteriorate. This is important to be able to measure change and at the

same time avoid ceiling as well as floor effects. It is also important for

52

minimizing the loss of study patients over time. If tests get too complicated

when functions deteriorate it can be hard to assess patients with severe

cognitive decline.

Control group

The small control group is a weakness of the study. The control group was

assessed with the UPDRS III, but associations between motor and cognitive

scores could not be calculated in the control group, as the controls had scores

of zero, at baseline and later. Thus it was not possible to see if the pattern of

association differed between the control group and the PD group. It has been

discussed whether to use norm values based on a control group or normative

values from other cohorts. Well-developed normative values taken from big

cohorts strictly following the standardization of the tests in similar age

groups that are being studied provide adequate data for comparison with the

study group of interest. It may be difficult to find such norm values,

especially with the tests in an assessment battery taken from different

sources. The optimal solution would be a large sample of controls tested

under the same conditions as the clinical subjects under investigation.

Future implications

The papers in the thesis discuss cognitive dysfunction in early stages of PD

and its relation to motor function for baseline and follow-up data up to five

years after diagnosis. We have so far completed the eight-year follow-up for

patients recruited during the first two years. This will allow investigations of

baseline measures that predict early onset and late onset dementia. Later we

will also receive autopsy results that will confirm the clinical diagnosis of PD

and enable identification of cellular signatures associated with the

development of dementia.

With longer follow-up we can study patients maintaining cognitive function.

This will make it possible to investigate what distinguishes patients with

preserved cognitive function from the large proportion of patients that

decline.

Combining different motor and cognitive measures as predictors for

dementia might display subtypes that go beyond the scope of studying

cognition and motor functions separately. Studying patients with postural

and gait impairment together with impairment in visuospatial memory and

function might give a better prediction for future outcome. Furthermore,

investigating objective measures of movement might provide a more specific

pattern of deficit connected to specific cognitive functions.

53

Blood samples and cerebrospinal fluid (CSF) from patients and controls will

enable genotyping and metabolomic profiling which could assist in the

interpretation of the motor and cognitive associations discussed in the

present thesis as well as finding predictors for cognitive decline and

dementia.

54

Conclusions

The findings in this thesis confirm results from previous studies, namely that

a large proportion of patients are cognitively impaired already at the time of

diagnosis with a range of cognitive functions being affected. The present

study is unique in that it combines being truly community-based with

employing an extensive neuropsychological test battery with repeated

investigation. All patients within the study are referred to the same

neurological department. This resulted in almost a complete recruitment of

affected cases all in early disease stages.

The findings show a diverse picture of cognitive and motor functions with

both dopaminergic and non-dopaminergic motor and cognitive functions

associated with cognitive impairment and PDD. This indicates that we need

to look at both the striatal network in combination with a more widespread

cortical network in the search for mechanisms behind cognitive impairment

in PD. The associations found between motor and cognitive dysfunction

could reflect shared underlying pathology in motor and cognitive

disturbances.

The differences in proportions of cognitively impaired in Paper I and Paper

IV were mostly due to the use of different criteria. This shows that MCI

classification is largely determined by the tests included in diagnosis and the

cut-off values used. This emphasizes the value of joint criteria for PD-MCI,

both in research and in clinic. Even when using joint criteria there is a

substantial amount of patients that reverts back to normal cognitive function

at follow-up.

There could be several explanations for increased motor severity in patients

with PDD. Patients with more posture and gait difficulties could be more

advanced in their disease progression, i.e. dementia and postural and gait

difficulties may represent a more advanced disease stage. Alternative

explanations could be that patients with dementia for unknown reasons

respond less well to medication compared with patients without dementia

and therefore have worse and declining motor performance. The parallel

decline in motor and cognitive functions might also affect each other (e.g.

less physical activity due to motor dysfunction resulting in a more inactive

life which leads to accelerating cognitive decline which in turn results in

worsening of motor performance). To answer these questions further studies

that are specifically designed to answer these questions are needed.

55

Acknowledgements

There are several people I would like to thank that in different ways have

contributed to this thesis.

First of all, I would like to express my gratitude to my supervisors Lars

Forsgren and Eva Elgh. Lars for giving me the opportunity of being a

doctoral student within the NYPUM-project and for always replying fast to

my queries, small or large! Eva for recruiting me to the NYPUM-project and

for great support along the way!

I would also like to thank all participants of the NYPUM project that with

great effort endured the neuropsychological examination on top of all the

other investigations, year after year.

A big thank you goes out to those helping us collecting the

neuropsychological data: Theres Åberg, Urban Ekman, Sofia Nording,

Carina Berg, John Wallert Holmqvist, Ida Andersson, Anton Andersson

and Neda Dimova Bränström. You all did a tremendous job!

Apart from great help with the data collection I would like to give special

thanks to Sofia and Urban. Sofia for great help with the project while I was

on maternity leave and for being a wonderful friend! Urban for rewarding

discussions on MCI, dementia, manuscripts, and everything in between.

Being a doctoral student within the NYPUM project became more fun when

you joined the crew!

Also, I would like to thank everyone else involved in the NYPUM project. A

special thanks to Jörgen Andersson for extracting data from the NYPUM

database whenever I needed it and filling in for me when writing the thesis

did not allow me to fulfill my project duties, and Mona Edström for doing a

great job with coordinating the project.

Thank you all great “brains” at the UFBI for inspiring lab meetings, with a

special thanks to Mikael Anderson for being tremendously patient when

helping me with SPM (even though I unfortunately did not include fMRI-

data in my thesis)

I’m further grateful to my colleagues at the Department of Geriatrics, past

and present! Even though I might have come across as an odd bird in your

“fika”-room, you have all showed tremendous hospitality. On top of learning

56

a lot about geriatric research when attending your seminars, I’ve had

wonderful discussions and laughs in the “fika”-room.

Loads of love to all of my friends who with walks, long conversations on the

telephone, trips to the beach, yoga sessions, coffee, laughs and food have

been tremendous support throughout this process, you know who you are!

A loving thank you to my family: Rose Eriksson that apart from being my

wonderful mother also did the proofreading of this thesis. Nils Klingberg

and Johan Eriksson for your brotherly support. My father Sigvard Eriksson,

I wish you were still with us. Karin and Edith for simply being the joy of my

life! Last but not least I want to thank Erik for taking a tremendous

responsibility for family duties during the last couple of months and on top

of that being patient with all my worries! I truly love you!

57

References

1. World population aging, Economic and social affairs, UN [http://www.un.org/en/development/desa/population/theme/ageing/index.shtml]

2. Parkinson J. An Essay on the Shaking Palsy. London: Sherwood, Neely, and Jones; 1817:66.

3. Pollock M, Hornabrook RW. The prevalence, natural history and dementia of Parkinson’s disease. Brain 1966, 89:429–448.

4. Marttila RJ, Rinne UK. Dementia in Parkinson’s disease. Acta Neurol Scand 1976, 54:431–441.

5. Aarsland D, Beyer MK, Kurz MW. Dementia in Parkinson’s disease. Curr Opin Neurol 2008, 21:676–82.

6. Muslimovic D, Post B, Speelman JD, Schmand B. Cognitive profile of patients with newly diagnosed Parkinson disease. Neurology 2005, 65:1239–45.

7. Aarsland D, Brønnick K, Larsen JP, Tysnes OB, Alves G. Cognitive impairment in incident, untreated Parkinson disease: the Norwegian ParkWest study. Neurology 2009, 72:1121–6.

8. Foltynie T, Brayne CEG, Robbins TW, Barker RA. The cognitive ability of an incident cohort of Parkinson’s patients in the UK. The CamPaIGN study. Brain 2004, 127:550–60.

9. Fahn S, Elton R. Recent Development in Parkinson’s Disease. New York: Mcmillan Health Care Information; 1987:153–164.

10. Hughes AJ, Ben-shlomo Y, Daniel SE, Lees AJ. What features improve the accuracy of clinical diagnosis in Parkinson’s disease: A clinicopathologic study. Neurology 1992, 57:34–38.

11. Mo SJ, Linder J, Forsgren L, Larsson A, Johansson L, Riklund K. Pre- and postsynaptic dopamine SPECT in the early phase of idiopathic parkinsonism: a population-based study. Eur J Nucl Med Mol Imaging 2010, 37:2154–64.

12. Fox SH, Lang AE. Levodopa-related motor complications--phenomenology. Mov Disord 2008, 23:509–14.

58

13. Tolosa E, Gaig C, Santamaría J, Compta Y. Diagnosis and the premotor phase of Parkinson disease. Neurology 2009, 72:12–20.

14. De Rijk MC, Launer LJ, Berger K, Breteler MM, Dartigues JF, Baldereschi M, et al. Prevalence of Parkinson’s disease in Europe: A collaborative study of population-based cohorts. Neurology 2000, 54:21–23.

15. Linder J, Stenlund H, Forsgren L. Incidence of Parkinson’s disease and parkinsonism in northern Sweden: a population-based study. Mov Disord 2010, 25:341–8.

16. Van Den Eeden SK. Incidence of Parkinson’s Disease: Variation by Age, Gender, and Race/Ethnicity. Am J Epidemiol 2003, 157:1015–1022.

17. Caslake R, Taylor K, Scott N, Gordon J, Harris C, Wilde K, et al. Age-, gender-, and socioeconomic status-specific incidence of Parkinson’s disease and parkinsonism in North East Scotland: The PINE study. Parkinsonism Relat Disord 2013, 19:515–521.

18. De Lau LM, Breteler MM. Epidemiology of Parkinson’s disease. Lancet Neurol 2006, 5:525–535.

19. Singleton AB, Farrer M, Johnson J, Singleton A, Hague S, Kachergus J, et al. alpha-Synuclein locus triplication causes Parkinson’s disease. Science 2003, 302:841.

20. Braak H, Ghebremedhin E, Rüb U, Bratzke H, Del Tredici K. Stages in the development of Parkinson’s disease-related pathology. Cell Tissue Res 2004, 318:121–34.

21. Braak H, Bohl JR, Müller CM, Rüb U, de Vos RAI, Del Tredici K. Stanley Fahn Lecture 2005: The staging procedure for the inclusion body pathology associated with sporadic Parkinson’s disease reconsidered. Mov Disord 2006, 21:2042–51.

22. Vingerhoets FJ, Schulzer M, Calne DB, Snow BJ. Which clinical sign of Parkinson’s disease best reflects the nigrostriatal lesion? Ann Neurol 1997, 41:58–64.

23. Helmich RC, Hallett M, Deuschl G, Toni I, Bloem BR. Cerebral causes and consequences of parkinsonian resting tremor: a tale of two circuits? Brain 2012, 135:3206–3226.

59

24. Doder M, Rabiner EA, Turjanski N, Lees AJ, Brooks DJ. Tremor in Parkinson’s disease and serotonergic dysfunction: An 11C-WAY 100635 PET study. Neurology 2003, 60:601–605.

25. Bohnen NI, Müller MLTM, Koeppe R a, Studenski S a, Kilbourn M a, Frey K a, et al. History of falls in Parkinson disease is associated with reduced cholinergic activity. Neurology 2009, 73:1670–6.

26. Yarnall A, Rochester L, Burn DJ. The interplay of cholinergic function, attention, and falls in Parkinson’s disease. Mov Disord 2011, 26:2496–503.

27. Müller MLTM, Frey KA, Petrou M, Kotagal V, Koeppe RA, Albin RL, et al. β-Amyloid and postural instability and gait difficulty in Parkinson’s disease at risk for dementia. Mov Disord 2013, 28:296–301.

28. Carlsson A. The occurence, distribution and physiological role of catecholamines in the nervous system. Pharmacol Rev 1959:490–493.

29. Ehringer H, Hornykiewicz O. Distribution of noradrenaline and dopamine (3-hydroxytyramine) in the human brain and their behaviour in diseases of the extrapyramidal system. Klin Wochenschr 1960:1236–1239.

30. Moustafa AA, Herzallah MM, Gluck MA. Dissociating the Cognitive Effects of Levodopa versus Dopamine Agonists in a Neurocomputational Model of Learning in Parkinson’s Disease. Neurodegener Diseas 2013, 11:102–11.

31. Lang AE. When and how should treatment be started in Parkinson disease? Neurology 2009, 72:39–43.

32. Williams-Gray CH, Evans JR, Goris A, Foltynie T, Ban M, Robbins TW, et al. The distinct cognitive syndromes of Parkinson’s disease: 5 year follow-up of the CamPaIGN cohort. Brain 2009, 132:2958–69.

33. Taylor AE, Saint-Cyr JA, Lang AE. Frontal lobe dysfunction in Parkinson’s disease. Brain 1986, 109:845–883.

34. Emre M, Aarsland D, Brown R, Burn DJ, Duyckaerts C, Mizuno Y, et al. Clinical diagnostic criteria for dementia associated with Parkinson’s disease. Mov Disord 2007, 22:1689–707.

60

35. Baddeley A, Hitch G. Working Memory. In Psychol Learn Motiv. Edited by Bower G. New York: Academic Press; 1974:47–89.

36. Bäckman L, Nyberg L, Lindenberger U, Li S-C, Farde L. The correlative triad among aging, dopamine, and cognition: current status and future prospects. Neurosci Biobehav Rev 2006, 30:791–807.

37. Bäckman L, Nyberg L, Soveri A, Johansson J, Andersson M, Dahlin E, et al. Effects of working-memory training on striatal dopamine release. Science 2011, 333:718.

38. Squire LR, Stark CEL, Clark RE. The medial temporal lobe. Annu Rev Neurosci 2004, 27:279–306.

39. Squire LR. Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychol Rev 1992, 99:195–231.

40. Fletcher PC. Frontal lobes and human memory: Insights from functional neuroimaging. Brain 2001, 124:849–881.

41. Aarsland D. Performance on the dementia rating scale in Parkinson’s disease with dementia and dementia with Lewy bodies: comparison with progressive supranuclear palsy and Alzheimer's disease. J Neurol Neurosurg Psychiatry 2003, 74:1215–1220.

42. Kravitz DJ, Saleem KS, Baker CI, Mishkin M. A new neural framework for visuospatial processing. Nat Rev Neurosci 2011, 12:217–30.

43. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Kokmen E, Tangelos EG. Aging, memory, and mild cognitive impairment. Int psychogeriatrics IPA 1997, 9:65–69.

44. Winblad B, Palmer K, Kivipelto M, Jelic V, Fratiglioni L, Wahlund L-O, et al. Mild cognitive impairment--beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med 2004, 256:240–6.

45. Litvan I, Goldman JG, Tröster AI, Schmand BA, Weintraub D, Petersen RC, et al. Diagnostic criteria for mild cognitive impairment in Parkinson’s disease: Movement Disorder Society Task Force guidelines. Mov Disord 2012, 27:349–56.

61

46. Prince M, Bryce R, Albanese E, Wimo A, Ribeiro W, Ferri CP. The global prevalence of dementia: A systematic review and metaanalysis. Alzheimer’s Dement 2013, 9:63–75.

47. Aarsland D, Zaccai J, Brayne C. A systematic review of prevalence studies of dementia in Parkinson’s disease. Mov Disord 2005, 20:1255–63.

48. Janvin CC, Aarsland D, Larsen JP. Cognitive predictors of dementia in Parkinson’s disease: a community-based, 4-year longitudinal study. J Geriatr Psych Neur 2005, 18:149–54.

49. Aarsland D, Andersen K, Larsen JP, Lolk A, Nielsen H, Kragh-Sørensen P. Risk of dementia in Parkinson’s disease: a community-based, prospective study. Neurology 2001, 56:730–6.

50. Marder K, Tang MX, Cote L, Stern Y, Mayeux R. The frequency and associated risk factors for dementia in patients with Parkinson’s disease. Arch Neurol 1995, 52:695–701.

51. Hobson P, Gallacher J, Meara J. Cross-sectional survey of Parkinson’s disease and parkinsonism in a rural area of the United Kingdom. Mov Disord 2005, 20:995–8.

52. Dalrymple-Alford JC, Livingston L, MacAskill MR, Graham C, Melzer TR, Porter RJ, et al. Characterizing mild cognitive impairment in Parkinson’s disease. Mov Disord 2011, 26:629–636.

53. Hoops S, Nazem S, Siderowf AD, Duda JE, Xie SX, Stern MB, et al. Validity of the MoCA and MMSE in the detection of MCI and dementia in Parkinson disease. Neurology 2009, 73:1738–1745.

54. Gauthier S, Reisberg B, Zaudig M, Petersen RC, Ritchie K, Broich K, et al. Mild cognitive impairment. Lancet 2006, 367:1262–1270.

55. Reid WG, Hely MA, Morris JG, Broe GA, Adena M, Sullivan DJ, et al. A longitudinal of Parkinson’s disease: clinical and neuropsychological correlates of dementia. J Clin Neurosci 1996, 3:327–33.

56. Hely MA, Morris JG, Rail D, Reid WG, O’Sullivan DJ, Williamson PM, et al. The Sydney Multicentre Study of Parkinson’s disease: a report on the first 3 years. J Neurol Neurosurg Psychiatry 1989, 52:324–328.

62

57. Broeders M, de Bie RMA, Velseboer DC, Speelman JD, Muslimovic D, Schmand B. Evolution of mild cognitive impairment in Parkinson disease. Neurology 2013, 81:346–52.

58. Pedersen KF, Larsen JP, Tysnes O-B, Alves G. Prognosis of mild cognitive impairment in early Parkinson disease: the Norwegian ParkWest study. JAMA Neurol 2013, 70:580–6.

59. Irwin DJ, Lee VM-Y, Trojanowski JQ. Parkinson’s disease dementia: convergence of α-synuclein, tau and amyloid-β pathologies. Nat Rev Neurosci 2013, 14:626–36.

60. Williams-Gray CH, Foltynie T, Brayne CEG, Robbins TW, Barker R a. Evolution of cognitive dysfunction in an incident Parkinson’s disease cohort. Brain 2007, 130:1787–98.

61. Kempster PA, O’Sullivan SS, Holton JL, Revesz T, Lees AJ. Relationships between age and late progression of Parkinson’s disease: a clinico-pathological study. Brain 2010, 133:1755–62.

62. Levy G, Schupf N, Tang M-X, Cote LJ, Louis ED, Mejia H, et al. Combined effect of age and severity on the risk of dementia in Parkinson’s disease. Ann Neurol 2002, 51:722–9.

63. Woods SP, Tröster AI. Prodromal frontal/executive dysfunction predicts incident dementia in Parkinson’s disease. JINS 2003, 9:17–24.

64. Kehagia AA, Barker RA, Robbins TW. Neuropsychological and clinical heterogeneity of cognitive impairment and dementia in patients with Parkinson’s disease. Lancet Neurol 2010, 9:1200–13.

65. Burn DJ, Rowan EN, Allan LM, Molloy S, O’Brien JT, McKeith IG. Motor subtype and cognitive decline in Parkinson’s disease, Parkinson's disease with dementia, and dementia with Lewy bodies. J Neurol Neurosurg Psychiatry 2006, 77:585–9.

66. Williams-Gray CH, Mason SL, Evans JR, Foltynie T, Brayne C, Robbins TW, et al. The CamPaIGN study of Parkinson’s disease: 10-year outlook in an incident population-based cohort. J Neurol Neurosurg Psychiatry 2013, 84:1258–64.

67. Alves G, Larsen JP, Emre M, Wentzel-Larsen T, Aarsland D. Changes in motor subtype and risk for incident dementia in Parkinson’s disease. Mov Disord 2006, 21:1123–30.

63

68. Svenningsson P, Westman E, Ballard C, Aarsland D. Cognitive impairment in patients with Parkinson’s disease: diagnosis, biomarkers, and treatment. Lancet Neurol 2012, 11:697–707.

69. Irwin DJ, White MT, Toledo JB, Xie SX, Robinson JL, Van Deerlin V, et al. Neuropathologic substrates of Parkinson disease dementia. Ann Neurol 2012, 72:587–98.

70. Kövari E, Gold G, Herrmann FR, Canuto A, Hof PR, Bouras C, et al. Lewy body densities in the entorhinal and anterior cingulate cortex predict cognitive deficits in Parkinson’s disease. Acta Neuropathol 2003, 106:83–8.

71. Mattila PM, Rinne JO, Helenius H, Dickson DW, Röyttä M. Alpha-synuclein-immunoreactive cortical Lewy bodies are associated with cognitive impairment in Parkinson’s disease. Acta Neuropathol 2000, 100:285–290.

72. Braak H, Rüb U, Jansen Steur ENH, Del Tredici K, de Vos RAI. Cognitive status correlates with neuropathologic stage in Parkinson disease. Neurology 2005, 64:1404–10.

73. Apaydin H, Ahlskog JE, Parisi JE, Boeve BF, Dickson DW. Parkinson Disease NeuropathologyLater-Developing Dementia and Loss of the Levodopa Response. Arch Neurol 2002, 59:102–112.

74. Hurtig HI, Trojanowski JQ, Galvin J, Ewbank D, Schmidt ML, Lee VM-Y, et al. Alpha-synuclein cortical Lewy bodies correlate with dementia in Parkinson’s disease. Neurology 2000, 54:1916–1921.

75. Galvin JE, Pollack J, Morris JC. Clinical phenotype of Parkinson disease dementia. Neurology 2006, 67:1605–11.

76. Compta Y, Pereira JB, Ríos J, Ibarretxe-Bilbao N, Junqué C, Bargalló N, et al. Combined dementia-risk biomarkers in Parkinson’s disease: A prospective longitudinal study. Parkinsonism Relat Disord 2013, 19:717–724.

77. Jellinger KA, Attems J. Prevalence and impact of vascular and Alzheimer pathologies in Lewy body disease. Acta Neuropathol 2008, 115:427–36.

78. Kurz MW, Schlitter AM, Larsen JP, Ballard C, Aarsland D. Familial occurrence of dementia and parkinsonism: a systematic review. Dement Geriatr Cogn Disord 2006, 22:288–95.

64

79. Jones EL, Aarsland D, Londos E, Ballard C. A pilot study examining associations between DYRK1A and α-synuclein dementias. Neurodegener Dis 2012, 10:229–31.

80. Ramirez A, Heimbach A, Gründemann J, Stiller B, Hampshire D, Cid LP, et al. Hereditary parkinsonism with dementia is caused by mutations in ATP13A2, encoding a lysosomal type 5 P-type ATPase. Nat Genet 2006, 38:1184–91.

81. Setó-Salvia N, Clarimón J, Pagonabarraga J, Pascual-Sedano B, Campolongo A, Combarros O, et al. Dementia risk in Parkinson disease: disentangling the role of MAPT haplotypes. Arch Neurol 2011, 68:359–64.

82. Rolinski M, Fox C, Maidment I, McShane R. Cholinesterase inhibitors for dementia with Lewy bodies, Parkinson’s disease dementia and cognitive impairment in Parkinson's disease. Cochrane database Syst Rev 2012, 3:CD006504.

83. Seppi K, Weintraub D, Coelho M, Perez-Lloret S, Fox SH, Katzenschlager R, et al. The Movement Disorder Society Evidence-Based Medicine Review Update: Treatments for the non-motor symptoms of Parkinson’s disease. Mov Disord 2011, 26:S42–80.

84. Burn D, Emre M, McKeith I, De Deyn PP, Aarsland D, Hsu C, et al. Effects of rivastigmine in patients with and without visual hallucinations in dementia associated with Parkinson’s disease. Mov Disord 2006, 21:1899–907.

85. Barone P, Burn DJ, van Laar T, Hsu C, Poewe W, Lane RM. Rivastigmine versus placebo in hyperhomocysteinemic Parkinson’s disease dementia patients. Mov Disord 2008, 23:1532–40.

86. Zahodne LB, Fernandez HH. Pathophysiology and Treatment of Psychosis in Parkinson’s Disease. Drugs Aging 2008, 25:665–682.

87. Pollak P. Clozapine in drug induced psychosis in Parkinson’s disease: a randomised, placebo controlled study with open follow up. J Neurol Neurosurg Psychiatry 2004, 75:689–695.

88. Liu J, Dong J, Wang L, Su Y, Yan P, Sun S. Comparative Efficacy and Acceptability of Antidepressants in Parkinson’s Disease: A Network Meta-Analysis. PLoS One 2013, 8:e76651.

89. Cools R, D’Esposito M. Inverted-U-shaped dopamine actions on human working memory and cognitive control. Biol Psychiatry 2011, 69:113–25.

65

90. Brusa L, Bassi A, Stefani A, Pierantozzi M, Peppe A, Caramia MD, et al. Pramipexole in comparison to l-dopa: a neuropsychological study. J Neural Transm 2003, 110:373–80.

91. Parsons TD, Rogers SA, Braaten AJ, Woods SP, Tröster AI. Cognitive sequelae of subthalamic nucleus deep brain stimulation in Parkinson’s disease: a meta-analysis. Lancet Neurol 2006, 5:578–588.

92. Diamond A. Close Interrelation of Motor Development and Cognitive Development and of the Cerebellum and Prefrontal Cortex. Child Dev 2000, 71:44–56.

93. Sosnoff JJ, Broglio SP, Ferrara MS. Cognitive and motor function are associated following mild traumatic brain injury. Exp Brain Res 2008, 187:563–71.

94. Amboni M, Barone P, Iuppariello L, Lista I, Tranfaglia R, Fasano A, et al. Gait patterns in parkinsonian patients with or without mild cognitive impairment. Mov Disord 2012, 27:1536–43.

95. Georgopoulos AP. Neural aspects of cognitive motor control. Curr Opin Neurobiol 2000, 10:238–241.

96. Burn DJ, Rowan EN, Allan LM, Molloy S, O’Brien JT, McKeith IG. Motor subtype and cognitive decline in Parkinson’s disease, Parkinson's disease with dementia, and dementia with Lewy bodies. J Neurol Neurosurg Psychiatry 2006, 77:585–9.

97. Mortimer JA, Pirozzolo FJ, Hansch EC, Webster DD. Relationship of motor symptoms to intellectual deficits in Parkinson disease. Neurology 1982, 32:133–133.

98. Cooper JA, Sagar HJ, Jordan N, Harvey NS, Sullivan E V. Cognitive impairment in early, untreated parkinson’s disease and its relationship to motor disability. Brain 1991, 114:2095–2122.

99. Hietanen M, Teräväinen H. Cognitive performance in early Parkinson’s disease. Acta Neurol Scand 2009, 73:151–159.

100. Aarsland D, Larsen JP, Tandberg E, Laake K. Predictors of nursing home placement in Parkinson’s disease: A population-based, prospective study. .

66

101. Folstein M, Folstein S, Mchugh P. Mini-mental state- practical method for grading cognitive state of patients for clinician. J Psychiatr Res 1975, 12:189–198.

102. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Brit J Psychiatry 1979, 134:382–389.

103. Tomlinson CL, Stowe R, Patel S, Rick C, Gray R, Clarke CE. Systematic review of levodopa dose equivalency reporting in Parkinson’s disease. Mov Disord Off J Mov Disord Soc 2010, 25:2649–2653.

104. Louis ED Cote L, Alfaro B, Mejia H, Marder K TMX. PRogression of parkinsonian signs in parkinson disease. Arch Neurol 1999, 56:334–337.

105. Uc EY, McDermott MP, Marder KS, Anderson SW, Litvan I, Como PG, et al. Incidence of and risk factors for cognitive impairment in an early Parkinson disease clinical trial cohort. Neurology 2009, 73:1469–1477.

106. 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, 40:1529–34.

107. Benedict RH, Zgaljardic DJ. Practice Effects During Repeated Administrations of Memory Tests with and Without Alternate Forms. Volume 20. Psychology Press; 1998:339–352.

108. Buschke H. Selective reminding for analysis of memory and learning. J Verb Learn Verb Behav 1973, 12:543–550.

109. Wechsler D. Wechsler Adult Intelligence Scale, 3rd edition(WAIS III): Test Manual, 3rd Edn. Psychology. New York; 1997.

110. Reitan RM. Validity of the Trail Making Test as an indicator of organic brain damage. Percept Mot Ski 1958, 8:271–276.

111. Benton AL. Contributions to Neuropsychological Assessment: A Clinical Manual. Oxford University Press; 1994:159.

112. Benedict RHB, Schretlen D, Groninger L, Dobraski M, Shpritz B. Revision of the Brief Visuospatial Memory Test: Studies of Normal Performance, Reliability and Validity. Psychol Assess 1996, 8:145–153.

67

113. Spreen O, Strauss E. A Compendium of Neuropsychological Tests: Administration , Norms , and Commentary , Third Edition. Volume 2nd. Oxford University Press; 1991:1240.

114. Gilman S, Low P., Quinn N, Albanese A, Ben-Shlomo Y, Fowler C., et al. Consensus statement on the diagnosis of multiple system atrophy. J Neurol Sci 1999, 163:94–98.

115. Litvan I, Agid Y, Calne D, Campbell G, Dubois B, Duvoisin RC, et al. Clinical research criteria for the diagnosis of progressive supranuclear palsy (Steele-Richardson-Olszewski syndrome): Report of the NINDS-SPSP International Workshop. Neurology 1996, 47:1–9.

116. Litvan I, Bhatia KP, Burn DJ, Goetz CG, Lang AE, McKeith I, et al. Movement Disorders Society Scientific Issues Committee report: SIC Task Force appraisal of clinical diagnostic criteria for Parkinsonian disorders. Mov Disord 2003, 18:467–86.

117. McKeith IG, Dickson DW, Lowe J, Emre M, O’Brien JT, Feldman H, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology 2005, 65:1863–72.

118. Lezak M, Howiesson D, Loring D. Neuropsychological Assessment. New York: Oxford University Press; 2004:158–160.

119. Gibb W, Lees A. The relevence of the Lewy body to the pathogenesis of ideopathic Parkinson’s disease. J Neurosurg psychiatry 1988, 51:745–752.

120. Peto V, Jenkinson C, Fitzpatrick R, Greenhall R. The development and validation of a short measure of functioning and well being for individuals with Parkinson’s disease. Qual Life Res 1995, 4:241–248.

121. Hardin JW, Hilbe JM. Generalized Estimating Equations. CRC Press; 2013:261.

122. Janvin CC, Larsen JP, Aarsland D, Hugdahl K. Subtypes of mild cognitive impairment in Parkinson’s disease: progression to dementia. Mov Disord Off J Mov Disord Soc 2006, 21:1343–9.

123. Levy G, Jacobs DM, Tang M-X, Côté LJ, Louis ED, Alfaro B, et al. Memory and executive function impairment predict dementia in Parkinson’s disease. Mov Disord 2002, 17:1221–6.

68

124. Williams-Gray CH, Mason SL, Evans JR, Foltynie T, Brayne C, Robbins TW, et al. The CamPaIGN study of Parkinson’s disease: 10-year outlook in an incident population-based cohort. J Neurol Neurosurg Psychiatry 2013, 84:1258–64.

125. Lee JE, Cho KH, Song SK, Kim HJ, Lee HS, Sohn YH, et al. Exploratory analysis of neuropsychological and neuroanatomical correlates of progressive mild cognitive impairment in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2013:jnnp–2013–305062–.

126. Caviness JN, Driver-Dunckley E, Connor DJ, Sabbagh MN, Hentz JG, Noble B, et al. Defining mild cognitive impairment in Parkinson’s disease. Mov Disord 2007, 22:1272–1277.

127. Marras C, Armstrong MJ, Meaney CA, Fox S, Rothberg B, Reginold W, et al. Measuring mild cognitive impairment in patients with Parkinson’s disease. Mov Disord 2013.

128. Aarsland D, Kurz MW. The epidemiology of dementia associated with Parkinson’s disease. Brain Pathol 2010, 20:633–9.

129. Broeders M, Velseboer DC, De Bie R, Speelman JD, Muslimovic D, Post B, et al. Cognitive Change in Newly-Diagnosed Patients with Parkinson’s Disease: A 5-Year Follow-up Study. J Int Neuropsychol Soc JINS 2013:1–14.

130. Cools R. Dopaminergic modulation of cognitive function-implications for L-DOPA treatment in Parkinson’s disease. Neurosci Biobehav R 2006, 30:1–23.

131. Boyle PA, Wilson RS, Aggarwal NT, Tang Y, Bennett DA. Mild cognitive impairment: risk of Alzheimer disease and rate of cognitive decline. Neurology 2006, 67:441–5.

132. Manly JJ, Tang M-X, Schupf N, Stern Y, Vonsattel J-PG, Mayeux R. Frequency and course of mild cognitive impairment in a multiethnic community. Ann Neurol 2008, 63:494–506.

133. Larrieu S, Letenneur L, Orgogozo JM, Fabrigoule C, Amieva H, Le Carret N, et al. Incidence and outcome of mild cognitive impairment in a population-based prospective cohort. Neurology 2002, 59:1594–1599.

134. Kryscio RJ, Schmitt FA, Salazar JC, Mendiondo MS, Markesbery WR. Risk factors for transitions from normal to mild cognitive impairment and dementia. Neurology 2006, 66:828–32.

69

135. Cobb JL, Wolf PA, Au R, White R, D’Agostino RB. The effect of education on the incidence of dementia and Alzheimer’s disease in the Framingham Study. Neurology 1995, 45:1707–1712.

136. Aarsland D, Kvaløy JT, Andersen K, Larsen JP, Tang MX, Lolk A, et al. The effect of age of onset of PD on risk of dementia. J Neurol 2007, 254:38–45.

137. Haaxma CA, Bloem BR, Borm GF, Oyen WJG, Leenders KL, Eshuis S, et al. Gender differences in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2007, 78:819–24.

138. Mayeux R, Stern Y. Intellectual dysfunction and dementia in Parkinson disease. Adv Neurol 1983, 38:211–27.

139. Riggeal BD, Crucian GP, Seignourel P, Jacobson CE, Okun MS, Rodriguez R l, et al. Cognitive decline tracks motor progression and not disease duration in Parkinson patients. Neuropsychiatr Dis Treat 2007, 3:955–8.

140. Ekman U, Eriksson J, Forsgren L, Mo SJ, Riklund K, Nyberg L. Functional brain activity and presynaptic dopamine uptake in patients with Parkinson’s disease and mild cognitive impairment: a cross-sectional study. Lancet Neurol 2012, 11:679–687.

141. Montero-Odasso M, Verghese J, Beauchet O, Hausdorff JM. Gait and cognition: a complementary approach to understanding brain function and the risk of falling. J Am Geriatr Soc 2012, 60:2127–36.

142. Van Iersel MB, Hoefsloot W, Munneke M, Bloem BR, Olde Rikkert MGM. Systematic review of quantitative clinical gait analysis in patients with dementia. Z Gerontol Geriatr 2004, 37:27–32.

143. Post B, Merkus MP, de Haan RJ, Speelman JD. Prognostic factors for the progression of Parkinson’s disease: a systematic review. Mov Disord 2007, 22:1839–51.

144. Verbaan D, Marinus J, Visser M, van Rooden SM, Stiggelbout a M, Middelkoop H a M, et al. Cognitive impairment in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2007, 78:1182–7.

145. Amboni M, Barone P, Hausdorff JM. Cognitive contributions to gait and falls: Evidence and implications. Mov Disord 2013, 28:1520–33.

70

146. Devos D, Defebvre L, Bordet R. Dopaminergic and non-dopaminergic pharmacological hypotheses for gait disorders in Parkinson’s disease. Fundam Clin Pharmacol 2010, 24:407–21.

147. Müller MLTM, Bohnen NI. Cholinergic dysfunction in Parkinson’s disease. Curr Neurol Neurosci Rep 2013, 13:377.

148. Beaulieu J-M, Gainetdinov RR. The physiology, signaling, and pharmacology of dopamine receptors. Pharmacol Rev 2011, 63:182–217.

149. Karlawish J, Cary M, Moelter ST, Siderowf A, Sullo E, Xie S, et al. Cognitive impairment and PD patients’ capacity to consent to research. Neurology 2013, 81:801–7.

150. Mamourian A. Incidental Findings on Research Functional MR Images: Should We Look? AJNR Am J Neuroradiol 2004, 25:520–522.

151. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990, 1:43–46.

152. Marek K, Jennings D, Seibyl J. Dopamine agonists and Parkinson’s disease progression: what can we learn from neuroimaging studies. Ann Neurol 2003, 53:160–166.

153. Schwingenschuh P, Ruge D, Edwards MJ, Terranova C, Katschnig P, Carrillo F, et al. Distinguishing SWEDDs patients with asymmetric resting tremor from Parkinson’s disease: a clinical and electrophysiological study. Mov Disord 2010, 25:560–9.

154. Marsden CD, Parkes JD. Success and problems of long-term levodopa therapy in Parkinson’s disease. Lancet 1977, 309:345–349.

155. Brown RG, Marsden CD, Quinn N, Wyke MA. Alterations in cognitive performance and affect-arousal state during fluctuations in motor function in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1984, 47:454–465.

1