Menu
Home
Log in / Register
 
Home arrow Health arrow Cognitive impairment and dementia in Parkinson disease
Source

Deficits in individual cognitive domains in PD-D

Attention and executive functions

The term ‘executive functions’ refers to a set of cognitive functions that are responsible for the planning, initiation, sequencing, and monitoring of complex goal-directed behaviour [6]. Historically, ‘executive functions’ has been used in at least two different ways. The older use of the term referred to ‘higher’/‘frontal’ functions such as insight, will, abstraction, and judgement. In the recent literature, the term usually means cybernetic ‘control’ functions [6] and is used more or less synonymously with concepts such as ‘top-down’ cognitive control or attentional control functions [11].

Thus, attentional control refers to the same phenomena as the cybernetic executive control functions. However, the term attention may also refer to ‘bottom-up’ phenomena such as orienting to stimulus novelty and startle, as well as to basic alertness or vigilance. Alertness and vigilance may be defined as readiness to detect and respond to stimuli, related to cortical arousal. Further, the term ‘selective attention’ refers to the ability to filter out task-irrelevant stimuli and to facilitate the processing of task-relevant stimuli [11]. Alertness and vigilance are basic requisites for behaviour involving the processing of external events and have a global impact on behaviour and cognition [8]. Focus on the importance of alertness and vigilance has increased as the concept of ‘fluctuating attention’ has become a ‘core feature’ in the diagnostic criteria for DLB [12], since such fluctuations could be viewed as a form of alertness deficit.

A problem in measuring executive functions and attention is that there are no pure tests of such functions. These functions are cognitive component-processes of both goal-directed behaviour and bottom-up driven processing of external stimuli. This may pose a problem if, for instance, one only uses visual test-paradigms for measuring attention in a patient group with severe visuospatial impairment. However, different tasks have different requirements for attention and executive functions, and executive and attentional deficits may be inferred if patients tend to perform worse on such tasks than on less demanding ones.

An overview of studies in which PD-D was diagnosed according to explicit criteria is shown in Table 4.1. In the columns of the table, effect sizes, measured with Cohen’s d [13], defined as the difference of the means divided by the pooled standard deviations, are given for the comparisons of PD-D with different patients groups (DLB, AD, and PD patients without dementia) and healthy controls (HC). The pooled standard deviation of the compared groups was calculated as recommended by Hedges and Olkin [14]. An effect size of 0.2 to 0.3 is considered small, 0.5 medium, and >0.8 a ‘large’ effect [13]. A positive number indicates that the group represented in the column performed better than the PD-D group, while a negative number indicates worse performance than the PD-D group.

Attention and executive functions are somewhat more severely affected in PD-D than in AD, with a medium average effect size of 0.14. It appears that attention and executive functions are

Table 4.1 Executive functions and attention

Test

HC

DLB

AD

PD

CAMCOG: attention [15]

1.71

-0.04

0.49

1.46

verbal fluency:

Letter [16]

-0.13

0.52

Letter [17]

0

Letter [18]

1.91

0.43

Letter [19]

0.23

Letter [20]

-0.70

-0.22

Letter [21]

1.94

-0.01

Category [16]

0.26

0.38

Category [17]

-0.33

Category [18]

1.65

-0.19

Category (supermarket) [20]

-0.18

-0.61

Category (animals) [20]

-0.71

-0.65

Category [21]

1.21

-0.05

Category and letter [22]

0.65

-0.14

0.33

Trail making test:

Part A [17]

-1.36a

Part A [22]

1.48

0.05a

0.66

Part B [17]

-0.13a

Part A (ZvT version) [21]

1.46

-0.34a

Dementia Rating Scale:

Initiation and perseveration [18]

2.04

0

Attention [23]

0.03

Initiation and perseveration [23]

0.30

Wisconsin card sorting test:

Criteria [17]

-0.29

Categories [18]

1.66

0.54

Categories [19]

-0.04

Cancellation

Shape time [16]

-0.04a

1.06a

TMX time [16]

-0.41

0.83

Stroop test:

Interference [17]

-0.41

Interference [20]

-0.24

0.55

Frontal assessment battery [17]

0

Test

HC

DLB

AD

PD

Digit span:

WAIS total [17]

0

Forward [19]

-0.09

Forward [20]

0.18

0.06

Forward [21]

1.67

-0.33

Forward [22]

0

-0.40

0.32

Backward [19]

-0.10

Backward [20]

-0.66

0.08

Backward [21]

2.72

0.17

Backward [22]

0.40

-0.23

0.44

RBANS: attention [24]

0.72

1.46

Mental control (WMS) [22]

0.38

-0.43

0.41

Digit symbol (WAIS) [22]

1.72

0

0.92

Choice reaction time [25]b

1.60a

0.04a

0.86a

1.20a

Simple reaction time [25]b

1.39

0.341

0.74

1.21

Serial 7s [26]

0.72

Average effect size

1.42

-0.21, -0.11c

0.14

0.84

Numbers represent Cohen's d relative to PD-D.

CAMCOG, Cambridge Cognitive Examination [27]; ZvT, Zahlen-verbindungs-Test; RBANS, Repeatable Battery for the Assessment of Neuropsychological Status [28]; TMX, Consonant Trigram Cancellation; WAIS, Wechsler Adult Intelligence Scale; WMS, Wechsler Memory Scale.

a Reaction time data reversed. b This study had large MMSE and age differences.

c Weighted meta-analytic average effect size, according to sample size; p = 0.106.

more severely affected in DLB than in PD-D, with a small average effect size of -0.21, but a fixed- effects meta-analysis revealed a weighted average mean difference of -0.11 (p = 0.106). Hence, there is no proven statistical difference between DLB and PD-D with regard to attention and executive functions. Fig. 4.1 gives a graphical depiction.

This meta-analytic review should be interpreted with caution as there are several results that come from the same sample, just with different tests, and hence sample characteristics specific to these studies may exert too much weight in the results.

There were a few studies that could not be included in Table 4.1 as the required statistics were not reported. In a study by Aarsland et al. [29], 60 patients with DLB, 35 with PD-D, 49 with progressive supranuclear palsy (PSP), and 29 with AD were compared using subscores from the Mattis Dementia Rating Scale (DRS) [30]. The groups were not matched for age, education, and severity of dementia, but statistical correction was used. Further, results were presented separately for mild to moderate versus severe dementia. The main findings showed that patients with DLB and PD-D suffering from severe dementia had a similar cognitive profile. For mild to moderate dementia, the PD-D group had a higher conceptualization score than the DLB group. This subscale measures semantic language abilities and abstraction. No other differences were found.

Fixed-effects meta-analysis

Fig. 4.1 Fixed-effects meta-analysis: DLB versus PD-D, attention and executive functions. Studies cited in the figure are Park et al. [20], Mosimann et al. [15], Noe et al. [16], Mondon et al. [17], and Ballard et al. [25].

PD-D patients with mild to moderate dementia performed worse than AD patients on ‘initiation and perseveration, a measure of executive functions. However, this subscale contains several complex motor tasks, possibly confounding the results.

There are two studies analysing components of attentional processes in PD-D using event- related potentials (ERPs). Perriol and colleagues [73] investigated a sensory filtering mechanism, pre-pulse inhibition (PPI), in PD-D, DLB, AD, and HC (10 subjects in each group). PPI is calculated as the amplitude of N1/P2 ERPs to a startle stimulus after a preceding (about 120 ms in this case) non-startling stimulus. The authors found reduced PPI in DLB and PD-D compared with the HC, and the DLB group also showed reduced PPI when compared with the AD group. Thus, the authors concluded that sensory filtering was most severely affected in DLB, followed by PD-D, while the AD patients did not differ significantly from the HC.

Brannick and colleagues [31] investigated an auditory automatic stimulus change-detection mechanism by measuring the mismatch negativity (MMN) ERP [32] in PD-D, DLB, AD, PD, and HC subjects. The amplitude of the MMN for the PD-D group was significantly attenuated compared with DLB, PD, and HC. Further, as compared with PD, DLB, AD, and HC, the PD-D group significantly regularly missed target stimuli in an auditory ‘oddball-distracter’ task which significantly correlated with MMN amplitude. Thus, the authors concluded that the PD-D patients had a more severe auditory attention deficit than DLB and AD, related to automatic (bottom-up driven) detection of stimulus deviance. A notable finding was that the MMN of the PD group was equal to that of the HC group, there was not even a tendency towards amplitude-attenuation. Hence, there appears to be a qualitative difference in automatic stimulus detection in PD-D versus PD that may be associated with the development of dementia in PD. The more severe deficit in PD-D compared with DLB was also surprising. A possible explanation could be that the attentional deficit in DLB is more pronounced in visual modality, as most of the research on attentional deficits in DLB and PD-D has used visual tests. Clearly, it should not be concluded that attention in general is impaired based on tests in a single sensory modality.

The finding that executive impairment is marked in PD-D is not surprising, given that cognitive impairment in PD without dementia is also frequently characterized by a dysexecutive syndrome [3, 33, 34]. However, it is not clear how the cognitive deficits in PD progress to PD-D,

i.e. if this process is one of gradual change or of qualitative differences. While in earlier studies progression of dopaminergic deficits was held responsible for cognitive impairment in PD-D, the role of other neural systems and neurotransmitters, including noradrenergic, serotonergic, and cholinergic systems, was subsequently emphasized [35]. This view has been most clearly put forward by Kehagia et al. [2], who argue that PD-D is first and foremost associated with a severe cholinergic deficit that sets this condition apart from PD with dysexecutive features. Kehagia calls this ‘the dual-syndrome hypothesis’.

The deficits of attention and executive functions in PD-D have a special significance, as it has been shown that attentional deficits in PD-D are the most important cognitive predictor of ability to perform activities of daily living (ADL) in a very large sample of PD-D patients [36]. Further, it has been shown that attention [37] and executive functions [38] appear to be the cognitive functions that are most closely associated with visual hallucinations in PD. There is a possibility that such attentional-executive deficits may partly underlie impairments in other cognitive domains, especially visuospatial dysfunction [39] (later in this chapter we shall see that this may not be the whole story, as DLB appears to be associated with more severe visuospatial deficits).

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Mathematics
Political science
Philosophy
Psychology
Religion
Sociology
Travel