html xmlns=”http://www.w3.org/1999/xhtml” xmlns:mml=”http://www.w3.org/1998/Math/MathML” xmlns:epub=”http://www.idpf.org/2007/ops”>
Background on cognitive impairment in Parkinson’s disease
Cognitive impairment is a very common problem in Parkinson’s disease (PD). Parkinson’s disease dementia (PDD) has a point prevalence estimated at about 40% [1], and prospective longitudinal studies find that up to 78% of PD patients can develop dementia over an 8-year period [2]. Parkinson’s disease dementia can have a significant impact on the quality of life of both patients and caregivers, and is associated with greater risk of nursing home placement and mortality [3, 4]. Parkinson’s disease patients also experience more subtle cognitive deficits not severe enough to warrant a diagnosis of dementia early in the course of the disease, with 25–30% of patients demonstrating some cognitive impairment even at the time of PD diagnosis [5]. Mild cognitive impairment in Parkinson’s disease (PD-MCI), defined as mild cognitive decline that is not normal for age and with essentially normal functional activities, has been increasingly recognized as a distinct clinical entity that may represent a transitional state between normal cognition and PDD [6]. Approximately 20–50% of nondemented PD patients meet the criteria for PD-MCI [7].
The profile of cognitive impairment in PD-MCI is heterogeneous and often characterized as either amnestic (i.e. memory deficit) or nonamnestic (impaired cognition other than memory) and as single domain or multiple domain (i.e. impairments in more than one cognitive area). Nonamnestic single-domain MCI is the most common subtype, representing approximately 42–62% of all individuals with PD-MCI [8–10]. Of PD-MCI patients with nonamnestic domain MCI, approximately 13–26% exhibit executive dysfunction and 18% were found to have visuospatial deficits [9, 10]. Overall, 16–24% of PD-MCI patients are characterized as amnestic single-domain MCI, while 24–40% of individuals are characterized as multiple-domain MCI (19% amnestic multiple domain; 5% nonamnestic multiple domain) [8–10].
Executive dysfunction is the most common cognitive disturbance and is usually evident early in the course of PD as well as sometime being an early sign of incident PDD [11]. Studies suggest that executive function deficits in PD result from a disruption in frontostriatal circuitry, particularly the dorsolateral prefrontal circuit [12–14]. Executive function deficits found in PD include impaired complex attention, verbal fluency, cognitive set-shifting, abstract thinking and planning [15–18]. Although there can be heterogeneity in the memory performance of PD patients [19, 20], the memory profile in general can be characterized by mild retrieval deficits with relatively spared retention of information. Episodic memory deficits in PD are thought to be a result of frontostriatal dysfunction that leads to impairments in higher-level strategy or organization of information during encoding and retrieval [21, 22]. Visuospatial deficits found in PD include impairments in perception of line orientation, visuomotor construction, facial recognition and memory for spatial location [23, 24].
The cognitive deficits in PDD are generally qualitatively similar to those with PD-MCI but are more severe and pervasive, affecting subjects’ functions [25]. The majority of individuals with PDD exhibit a subcortical profile, characterized by prominent executive dysfunction along with impaired attention, slowness of thought, and visuospatial and episodic memory deficits [26]. Compared with patients with Alzheimer’s disease (AD), PDD patients show more severe impairments in executive function and visuospatial processes, and less severe deficits in learning and memory [27–29]. However, some individuals with PDD may present with a “cortical” profile, which is characterized by memory (impaired retention of information) and language impairments that are found in cortical dementias such as AD [9]. The extent to which this represents a mixed pathology is unknown.
Given the frequency and progressive nature of cognitive deficits in PD and the impact that cognitive impairment has on patients and their families, it is critically important to identify treatments for cognitive impairment. Treatments that could improve cognitive functioning or delay or stop progression from PD-MCI to PDD are vitally important for this population. The following sections review the research conducted examining pharmacological and nonpharmacological treatments for cognitive impairment in PD.
Pharmacological interventions
There are no standard pharmacological treatments for the cognitive impairment of patients with PD-MCI; however, there are good clinical practice guidelines that advocate the discontinuation of medications that have the potential of deteriorating cognition when the symptoms of cognitive disturbances start to manifest. Medications that can worsen cognition include anticholinergic agents (commonly used to treat tremor or urinary disturbances), amantadine; dopamine agonists and benzodiazepines (reviewed in [30]). Good clinical practice guidelines include ruling out metabolic disorders (e.g. thyroid or hepatic dysfunction), infections (e.g. usually urinary) and vitamin deficits (e.g. hypovitaminosis of vitamin D, hypovitaminosis D, B or folic acid B or folic acid).
Moreover, there have been no therapeutic trials to improve the cognitive aspects of patients with PD-MCI except for a pilot study with rasagiline, a monoamine oxidase B (MAO-B) inhibitor, which showed improvements in working memory and verbal fluency [31] [Table 11.1]. The results of this study have led to an ongoing 24-week, multicenter, double-blind, placebo-controlled study to determine whether 1mg/day of rasagaline is more effective than placebo in improving the cognitive dysfunction of PD-MCI patients.
Pharmacologic and nonpharmacological interventions for cognitive impairment in Parkinson’s disease
Study | Sample/study design | Method | Intervention | Cognitive outcome | Results |
---|---|---|---|---|---|
Pharmacological interventions | |||||
Emre et al. (2004) [32]: multicenter involving multiple countries | 541 PDD patients (rivastigmine=362, placebo=179), 410 completed the study | 24-weeks double-blind, placebo RCT | Rivastigmine 9–12mg vs placebo | ADAS-Cog, ADCS-CGIC | ADAS-Cog, ADCS-CGIC improved significantly vs placebo |
Dubois et al. (2012) [33]: multicenter involving multiple countries | 550 PDD patients (5mg=195 patients, 10mg= 182 patients, placebo=173 patients), intention to treat analysis | 24-weeks, double-blind, placebo RCT | Donepezil 5 or 10mg vs placebo | ADAS-Cog, CGIC+ | ADAS-Cog: not significant in intent to treat, but removing country interaction analysis improved significantly vs placebo at both doses; the CGIC improved significantly in the 10mg group |
Hanagasi et al. (2011) [31]: Turkey | 55 PD-MCI: 48 patients completed the study (rasagaline=23, placebo=25), seven dropouts not included in the analyses | 12-week double-blind, placebo RCT | Rasagaline 1mg/day vs placebo | No primary endpoint (exploratory study) | WM (DS backward), VF and composite attention domain z-score improved significantly with rasagaline vs placebo |
Emre et al. (2010) [34]: multicenter involving multiple countries | 199 patients: 34 with DLB and 62 with PDD on memantine 20mg/day and 41 with DLB and 58 with PDD on placebo | 24-week double-blind, placebo RCT | Memantine 20mg/day vs placebo | No primary endpoint | No significant differences were noted between the two treatments in patients with PDD |
Aarsland et al. (2009) [35]: Norway, Sweden and UK | 72 out of 75 patients (40 with PDD and 32 with DLB): 34 with memantine and 38 with placebo; 56 (78%) completed the study | 24-week double-blind, placebo RCT | Memantine 20mg/day vs placebo | Clinical global impression of change (CGIC) | Better CGIC scores |
Cognitive rehabilitation | |||||
Restorative | |||||
Cerasa et al. (2014 ) [36]: Italy | Eight CR vs seven placebo (visuomotor task), ND PD with an attention, PS and/or EF deficit | Blind RCT | 6 weeks of two 1h computerized attention (RehaCom) trainings per week | Attention, PS, EF, VF, VS, and verbal and visual memory | PS (SDMT) and attention (DS forward) improved compared with controls; increased RS activity in parieto-prefrontal network |
Disbrow et al. (2012) [37]: USA | 14 pre-training task impaired vs 15 unimpaired ND PD | Non-RCT | 10 days (h/day unclear) of computerized EF monitoring training (number sequencing) | Attention (DS), PS/EF (TMT), VF | Impaired group improved on EF motor training task; TMT B/TMT A improved in both groups |
Mohlman et al. (2011 ) [38]: USA | 14 ND PD with cognitive complaints | Non-RCT, within subject design | 4 weeks of one 90 min attention process training (APT-II) per week | Attention (DS), PS/EF (TMT, CWIT), VF (COWAT) | VF (COWAT), WM (DS backward) EF (TMT B) and PS (CWIT color word trial) improved |
París et al. (2011 ) [39]: Spain | 16 CR vs 12 speech therapy ND PD | Double-blinded RCT | 4 weeks of three 45 min computer + homework sessions per week | Attention, PS, EF, VS/VC, VF, visual and verbal learning and memory | Attention (DS), PS, EF (TMT B, TOL), VS/VC, category fluency, visual learning and memory improved |
Sammer et al. (2006 ) [40]: Germany | 14 CR vs 12 stc ND PD | RCT | Ten 30 min WM/EF training sessions | Attention, EF (BADS, TMT equivalent), face–name memory | EF (BADS) improved |
Restorative+ | |||||
Sinforini et al. (2004 ) [41]: Italy | 20 PD-MCI | Non-RCT, within subject design | 6 weeks of two 1h CT (TNP software) + motor rehab sessions | Attention (DS), VF (FAS), memory (Babcock’s story), EF (Raven’s Matrices; WCST, Stroop), MMSE | EF (Raven’s matrices), VF (FAS) and story memory improved |
Reuter et al. (2012 ) [42]: Germany | 222 PD-MCI split into three CT groups | Double-blinded RCT | At least 14 1h sessions: 71 CT (A) vs 75 CT+transfer (compensatory) (B) vs 76 CT+transfer+motor (C) training | ADAS-Cog, SCOPA-Cog, BADS, PASAT | All groups improved on all cognitive measures; group C did the best |
Naismith et al. (2013) [43]: Australia | 35 CR vs 15 waiting-list ND PD (62% PD-MCI) | Single-blinded RCT | 7 weeks of 2h sessions (1h psychoeducation + 1h computerized CT | EF/PS (TMT), VF (COWAT), learning/ memory (LMT) | Episodic learning and memory (LMT) improved |
Cognitive stimulation | |||||
Nombela et al. (2012) [44]: Spain | Five puzzle vs five stc ND PD (MMSE >24) | Non-RCT | 6 months of daily Sudoku exercises | MMSE, MDRS, modified Stroop | EF (Stroop) improved; reduced cortical activation (fMRI) |
Pompeu et al. (2012) [45]: Brazil | 16 Wii vs 16 balance exercise ND PD | Parallel, single-blinded RCT | 7 weeks of two 1h sessions | MoCA | Overall cognition (MoCA) improved in both groups pre- to post-intervention; no difference between groups |
Physical exercise | |||||
Cruise et al. (2011) [46]: Australia | 15 PD vs 13 stc ND PD | Non-RCT | 12 weeks of two 1h aerobic and resistance training | MMSE, FAS and animal VF, Stockings of Cambridge, CANTAB-eclipse pattern recognition and spatial recognition memory, spatial working memory | Spatial WM, VF (FAS and animal) improved |
Nocera et al. (2013) [47]: USA | 15 PD vs 6 stc ND PD | Single-blinded RCT | 16 weeks of three 1h tai chi sessions | WM (DS backward test), FAS, Category Fluency, Stroop, TMT | No statistical improvement, but a trend (P = 0.08) trend toward improvement in WM (DS backward), associated with a large effect size of 0.89) |
Ridgel et al. (2011) [48]: USA | 19 PD (cognitive status unknown) | Non-RCT, within subject design | Three sessions, over 3 weeks, of 30 min passive cycling | TMT A and B | EF (TMT B) improved |
Tabak et al. (2013) [49]: USA | Two PDD patients (MoCA <22) | Non-RCT | 24 1h sessions cycling (three times a week for 8 weeks) | EF and processing speed, MoCA, Color Trails 1 and 2, Parkinson’s Disease Cognitive Rating Scale | EF (all measures) improved |
Tanaka et al. (2009) [50]: Brazil | 10 PE vs 10 stc* ND PD | Non-RCT | Group exercise, mainly aerobic, 1h sessions, three times a week for 6 months | EF (WCST) and PS (symbol search) | EF improved |
Neural stimulation | |||||
Boggio et al. (2005) [51]: Brazil | 13 active rTMS and placebo drug treatment vs 12 sham rTMS and active drug treatment (fluoxetine, 20mg/day) PD with major or minor depression | Double-blinded RCT | rTMS to left DLPFC for 10 sessions over 2-week period | EF (WCST, Stroop Color Word Interference test, TMT-B, FAS), visuospatial (HVOT), reasoning (Raven’s Matrices), and WM (DS) | No difference between treatment groups; both groups improved in EF (WCST, Stroop) and VS (HVOT) at 2 weeks post-intervention; both groups improved on Hooper and WCST at 8 weeks post-intervention |
Boggio et al. (2006) [52]: Brazil | 18 ND PD | Non-RCT, within subject design | Single session tDCS to left DLPFC with two different intensities (1 mA and 2 mA) versus tDCS to primary motor cortex with two different intensities (1 mA and 2 mA) versus sham tDCS 20 min, each session 48h apart | WM | WM improved with 2 mA to the left DLPFC |
Epstein et al. (2007) [53]: USA | 12 ND PD with moderate to severe depression | Non-RCT, open study | 10 sessions over 2 weeks of rTMS to the left DLPFC | MMSE, RBANS, Brief Test of Attention, DRS | Overall cognition (DRS total score), EF (DRS Conceptualization) and memory (DRS Memory) improved |
Pal et al. (2010) [54]: Hungary | 12 rTMS vs 10 sham rTMS ND PD with mild to moderate depression | Double-blinded RCT | rTMS to the left DLPFC for 10 days | MMSE, EF (TMT, Stroop Color Word Test) | EF (Stroop) at 1 day and 30 days post-intervention improved |
Sedlackova et al. (2009) [55]: Czech Republic | 10 PD (cognitive status unknown) | Non-RCT, within subject design | One 30 min session rTMS to left dorsal premotor cortex and DLPFC and occipital cortex (control condition) | EF (WM, choice reaction time, VF, TMT) | No improvement |
Srovnalova et al. (2011) [56]: Czech Republic | 10 ND PD | Non-RCT, within subject design | One session of rTMS to left and right inferior frontal gyri in sequence vs sham rTMS | EF (Stroop Color Word test, Frontal Assessment Battery) | PS and EF (all conditions of the Stroop) improved |
Srovnalova et al. (2012) [57]: Czech Republic | 10 ND PD | Non-RCT, within subject design | Two sessions of rTMS to right and two sessions to left DLPFC vs sham rTMS | EF (Tower of London) | EF (Tower of London) after rTMS to right DLPFC improved |
Furukawa et al. (2009) [58]: Japan | Six PD with impaired performance on WCST (≤4 categories) and MMSE≤26 | Non-RCT | Under EEG monitoring, rTMS to the frontal region (Fz) for 3 months, 1200 stimulations | TMT-B, WCST, WAIS-R | EF (WCST and TMT-B) improved |
Fregni et al. (2004) [59]: USA and Brazil | 21 rTMS and placebo drug treatment vs 21 sham rTMS and active drug treatment (fluoxetine 20mg/day) | Double-blinded RCT | 10 days of rTMS to left DLPFC | MMSE | Both groups improved on the MMSE, but the active rTMS group improved faster (showed more improvement at 2 weeks, but at 8 weeks no significant difference between groups) |
PDD, Parkinson’s disease dementia; DLB, dementia with Lewy bodies; ADAS-Cog, Alzheimer’s Disease Assessment Scale – cognitive subscale; CIBIC+ (global function), Clinician’s Interview-based Impression of Change plus caregiver input; PD-MCI, Individuals with Parkinson’s disease and Mild Cognitive Impairment; WM, working memory; DS, digit span; VF, verbal fluency; CR, cognitive rehabilitation; ND, nondemented; PS, processing speed; EF, executive function; RCT, randomized controlled trial; VS, visuospatial; SDMT, Symbol Digit Modality Test; RS, resting-state fMRI; TMT, Trail Making Test; COWAT, Controlled Oral Word Association Test; CWIT, Color Word Interference Test; VC, visuoconstruction; TOL, Tower of London; stc, standard of care; BADS, Behavioural Assessment of the Dysexecutive Syndrome; Restorative+, restorative CR+ motor training and/or compensatory CR; CT, cognitive training; WCST, Wisconsin Card Sorting Test; MMSE, Mini Mental Status Examination; MDRS, Mattis Dementia Rating Scale; SCOPA-Cog, Scales for Outcome of Parkinson’s Disease – Cognition; PASAT, Paced Auditory Serial Attention Task; LMT, Logical Memory Test; MoCA, Montreal Cognitive Assessment; PE, physical exercise intervention; CANTAB, Cambridge Neuropsychological Test Automated Battery; RBANS, Repeatable Battery for Assessment of Neuropsychological Status; HVOT, Hooper Visual Organization Test; WAIS-R, Wechsler Adult Intelligence Scale – Revised; DLPFC, dorsolateral prefrontal cortex; tDCS, transcranial direct current stimulation; rTMS, repetitive transcranial magnetic stimulation.
* Controls from a prior study.
To improve the cognitive dysfunction of patients with PDD on the other hand, in addition to the above good clinical practice guidelines, the use of cholinesterase inhibitors such as rivastigmine, donepezil and galantamine provide modest benefit [34, 31]). There is good evidence (level A evidence) from a double-blind, randomized, placebo-controlled study that rivastigmine has moderate beneficial effects on cognition in PDD [32] (Table 11.1). One large and two small donepezil randomized, double-blind, placebo-controlled studies [33, 60, 61] also showed cognitive benefits (level A evidence) (Table 11.1). Despite these promising findings, cholinesterase inhibitors have side effects that include worsening of motor symptoms (tremor), nausea, vomiting, diarrhea and urinary disturbances.
Memantine, a N-methyl-d-aspartate (NMDA) receptor antagonist, which is also used in moderately advanced AD, does not or minimally benefits patients with PDD [34, 35] (Table 11.1), and it is at present not recommended. Neuroleptics are used to treat the neuropsychiatric disturbances (i.e. hallucinations, agitation, delusions), but do not influence cognition and may worsen the parkinsonism as well as leading to a neuroleptic malignant syndrome. Similarly, antidepressants useful for depression or anxiety do not appear to improve cognition in this population. Moreover, the use of serotoninergic antidepressants and selegiline in high doses may lead to a serotoninergic syndrome, which manifests with cognitive, somatic and autonomic disturbances.
Nonpharmacological interventions
Cognitive rehabilitation
One of the most promising nonpharmacological treatments for cognitive impairment in PD is cognitive rehabilitation. Cognitive rehabilitation is a structured intervention that aims to improve, maintain or delay the decline of cognitive skills with the ultimate goal of improving the person’s ability to function in everyday life [62]. There are two main cognitive rehabilitation approaches: (i) restorative/remediative, which focuses on retraining with the goal of regaining specific cognitive skills lost due to trauma or disease; and (ii) compensatory, which provides alternative approaches to adapt to and “work around” cognitive weaknesses. Restorative methods frequently consist of computerized drills and/or practice worksheets aimed at remediating targeted cognitive deficit(s). Compensatory rehabilitation focuses on the teaching of specific strategies, such as mnemonics, time management and systematic problem solving, in order to improve functional cognitive skills in everyday life. Cognitive rehabilitation has been documented to be efficacious in nonprogressive conditions, such as traumatic brain injury and stroke [63], and there is a growing body of evidence of its efficacy with progressive conditions, such as mild AD [64] and amnestic MCI (reviewed in [65]).
Published data on the efficacy of cognitive rehabilitation in individuals with PD is sparse. To the best of our knowledge, only eight studies have evaluated cognition as an outcome following restorative and/or compensatory cognitive rehabilitation in PD, and only two of these studies were specific to PD-MCI (Table 11.1). Two additional studies [44, 45] used video or puzzle game practice as the primary treatment modality, which we have detailed in a separate section below (see ‘Cognitive stimulation’). All of the studies detailed in Table 11.1 evaluated a restorative-based cognitive rehabilitation approach, which entailed 5–14h of computer- and/or worksheet-based cognitive training based on analog versions of standard neuropsychological tests [37–43]. Two studies used aspects of compensatory strategy training and psychoeducation in addition to restorative training [42, 43], while one study conducted parallel motor training in addition to cognitive training [41]. Each study detailed in Table 11.1 revealed improvements in the respective cognitive domain in which the patients were trained, including improvements in executive function, memory and attention. With regard to the best approach, only one study to date has attempted to delineate which treatment strategy may be most efficacious. This study, conducted by Reuter et al. [42], compared a cognitive computer-based training program alone, with a compensatory (transfer) training program, and with both a compensatory and a motor training program in 222 nondemented PD patients. While all three interventions were deemed effective for improving cognition, the multimodal cognitive-psychomotor approach was found to be the most effective [42].
Taken together, there is evidence that cognitive rehabilitation improves cognition in nondemented PD. However, various treatment strategies and methodologies render interstudy comparisons and conclusions difficult. All studies to date have been limited by small sample sizes (n=6–35), minimal use of control groups, mixed cognitively normal and MCI samples, and lack of use of a standard MCI diagnosis. Based on their review of studies on nonpharmacological interventions for PD conducted between 1985 and 2012, Hindle et al. [66] concluded that only one study, conducted by Paris et al. [39], was acceptable with regard to bias. In addition, there is limited information on the transfer of improvements to the everyday lives of individuals with PD and an impact on quality of life. The compensatory approach may be more beneficial to everyday, “real world” function and performance [67], but it has yet to be validated independently in PD. Finally, maintenance and long-term effects of treatment maintenance are virtually unknown, as only two studies conducted follow-up testing. These studies indicated that the effects of cognitive rehabilitation combined with motor training were maintained at the 6 months follow-up, yet a lack of control group comparisons in both studies limit drawing conclusions of efficacy [41, 42]. Future studies are critically needed to validate these intervention strategies and their long-term effects in PD-MCI.
To the best of our knowledge, no published literature exists on the efficacy of cognitive rehabilitation in individuals with PDD. However, there are several published studies on the efficacy of cognitive rehabilitation in AD. The findings in this population have been controversial, but there is evidence that cognitive training can improve and maintain cognition in mild to moderate AD, as well as improving their quality of life, mood and functional abilities [68]. Evidence further supports the use of a compensatory strategy approach (e.g. use of a memory notebook) compared with interactive computer games involving memory, concentration and problem-solving skills as superior in improving cognitive performance in AD [64]. Still, it is difficult to generalize these findings to individuals with PDD, as the neuropathological mechanisms and neuropsychological profiles of PDD are quite different from AD, especially in the early stages [69]. Efficacy studies are critically needed to determine the best cognitive intervention strategies for PDD.

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree

