INTRODUCTION — Cognitive dysfunction is common in Parkinson disease (PD) and exists on a continuum of severity; prevalence increases with the duration of the movement disorder. When severe, dementia often surpasses the motor features of PD as a major cause of disability and mortality.
While PD can coexist with other common causes of dementia, such as Alzheimer disease (AD) and vascular dementia, mild cognitive impairment (MCI) and dementia are increasingly recognized as common features of PD itself. The clinical characteristics and course of cognitive impairment, its pathologic features, and the most appropriate treatment are areas of current investigation. Clinical features can generally distinguish between PD and other movement disorders associated with dementia, such as dementia with Lewy bodies (DLB) or progressive supranuclear palsy (PSP). However, whether PD dementia (PDD) and DLB are distinct disorders, or whether they represent different presentations of the same disease, is an area of debate and investigation [1,2].
This topic will discuss cognitive impairment and dementia associated with PD. Other aspects of PD and DLB are discussed separately. (See "Clinical manifestations of Parkinson disease" and "Management of nonmotor symptoms in Parkinson disease" and "Initial pharmacologic treatment of Parkinson disease" and "Epidemiology, pathology, and pathogenesis of dementia with Lewy bodies" and "Clinical features and diagnosis of dementia with Lewy bodies" and "Prognosis and treatment of dementia with Lewy bodies".)
Incidence and prevalence — In one study, PD dementia (PDD) was estimated to account for 3.6 percent of all cases of dementia in the population .
Dementia is a common feature of PD . Cross-sectional studies suggest that the mean prevalence of dementia is 30 to 40 percent . In prospective cohort studies, the incidence rate of dementia in patients with PD is consistently estimated at approximately 100 per 1000 patient-years, a rate almost five- to sixfold higher than controls without PD [3,5-7]. These studies suggest that dementia is nearly inevitable in patients with PD; cumulative incidence rates of over 80 percent have been reported in patients followed for more than 20 years after onset of PD [8-10].
Milder degrees of cognitive impairment are also common in patients with PD, particularly early in the disease course [11-14]. In studies of patients with a new diagnosis of PD, the prevalence of mild cognitive impairment (MCI) ranges from approximately 10 to 35 percent [10,12,13,15].
Risk factors — Older age, age at onset of PD ≥60 years, duration of PD, and severity of parkinsonism impact the incidence of dementia in PD [6-8,16-19]:
●In a community-based study, the prevalence of PDD in the general population was 787 per 100,000 for those ≥80 years, while the prevalence for those <50 years was zero .
●In a prospective, follow-up study evaluating the effect of disease duration, a 26 percent prevalence of dementia was noted at baseline in 224 patients with a mean disease duration of 11.5 years . The prevalence of dementia was 52 percent four years later, and 78 percent eight years later. The mean PD duration at the onset of dementia in this study was 14 years.
●In a population-based study, 37 percent of patients whose PD began after the age 70 years were demented, compared with 9 percent of patients whose symptoms had begun before 70 years . After five years of follow-up, the prevalences had risen to 62 and 17 percent, respectively.
●In a population-based study, the relative risk (RR) for dementia was 9.7 in older (≥72 years), high-severity (Unified Parkinson Disease Rating Scale [UPDRS] >24) PD patients, compared with young, low-severity patients . Young, high-severity patients and older, low-severity patients were not at higher risk of dementia compared with the reference population, suggesting that age and PD severity interact to increase the risk of dementia.
Additional risk factors for the development of more severe or early cognitive impairment in patients with PD include the presence of rapid eye movement sleep behavior disorder (RBD), autonomic dysfunction (especially hypotension), hyposmia, abnormal visual color discrimination, slow resting electroencephalography (EEG) frequencies, and gait dysfunction [21-24].
Poor cognitive functioning at baseline has also been associated with an increased risk of parkinsonism .
Genetic factors — Although the majority of cases of PD appear to be sporadic, monogenetic forms of PD are increasingly recognized, some of which often involve early or severe cognitive changes (table 1). Examples include PD associated with mutations in the glucocerebrosidase (GBA) gene [26-29] and some forms of alpha-synuclein (SNCA) associated PD [30-32]. By contrast, dementia is an uncommon feature of PD associated with parkin (PARK2) mutations . (See "Etiology and pathogenesis of Parkinson disease", section on 'Genetics'.)
There may also be genetic risk factors for the development of dementia in patients with sporadic PD .
●Both the epsilon 2 (ε2) and epsilon 4 (ε4) alleles of the apolipoprotein E (APOE) gene have been implicated as possible risk factors for PDD [34-37]. As an example, in a prospective cohort study of 212 patients with PD, the presence of the ε4 allele was associated with more rapid cognitive decline . However, other studies and meta-analyses have found no or only modest associations between PDD and alleles of the APOE gene [39,40]. In AD, APOE ε4 is well established as a risk factor, whereas APOE ε2 has a protective effect. APOE ε4 was found to be associated with the development of PDD in one study; the authors posited that the gene may have a greater effect on the severity of dementia than on its onset . (See "Genetics of Alzheimer disease", section on 'Apolipoprotein E'.)
●Allelic variation in the gene encoding microtubule-associated protein tau (MAPT) appears to affect susceptibility as well. The genotype H1/H2 has been implicated in the development of PDD [42,43]. In one cohort of 126 patients with PD, the MAPT H1/H1 genotype was strongly associated with incident dementia (OR = 12.1) . In a subsequent study, this haplotype was noted to have a specific association with declining memory, but not the overall rate of cognitive decline . Similarly, a high catechol-O-methyl transferase (COMT) activity haplotype has been found to be associated with a higher risk of developing cognitive decline in PD .
Neuropathology — The anatomic and pathologic basis of PD dementia (PDD) is not fully understood and is likely heterogeneous. Historically, the dementia in PD was often attributed to coexisting Alzheimer or vascular pathology. However, neuropathologic studies have found that the degree of Lewy body pathology (Lewy bodies and Lewy neurites) in the cortex and other areas correlates better with cognitive decline and dementia than does the degree of Alzheimer pathology, which may be relatively modest in these patients [45-51].
Lewy bodies constitute the pathologic hallmark of PD, and to some extent there is a relatively predictable neuroanatomic spread of Lewy body pathology in PD that correlates with clinical dementia. Lewy body pathology is initially most prominent in the olfactory system and lower brainstem, then progresses to involve the midbrain structures. In later stages, other brain structures are involved, including the cortex (figure 1). The neuropathologic stage of PD correlates with the severity of both motor parkinsonism and dementia . However, not all patients with PD and dementia have cortical Lewy body pathology . (See "Etiology and pathogenesis of Parkinson disease", section on 'Pathology'.)
The density of cortical Lewy body pathology has also been implicated in the degree of cognitive impairment [45-51]. In one study, Lewy body density in the entorhinal cortex and anterior cingulate accounted for up to one-third of the variability in clinical dementia severity . In another study, the severity of Lewy body pathology in the basal forebrain and hippocampus distinguished PD patients with and without dementia .
Important exceptions are noted; some individuals have significant cognitive decline in the absence of cortical Lewy pathology, while others have normal cognitive function in the presence of widespread cortical pathology [52,55]. A few studies have found that reduced amyloid beta (1 to 42) in the cerebrospinal fluid is associated with cognitive decline in patients with PD, as it is in other patient groups [56-58]. Given the substantial prevalence of both Alzheimer disease (AD) and PD, some patients with PD have cognitive decline that is attributable to or exacerbated by pathologic features of AD such as neurofibrillary tangles and beta-amyloid neuritic plaques . (See "Mild cognitive impairment: Prognosis and treatment", section on 'CSF biomarkers' and "Clinical features and diagnosis of Alzheimer disease", section on 'Other laboratory testing'.)
Magnetic resonance imaging (MRI) and neuropathologic studies demonstrate more prominent global brain atrophy in PD patients with dementia than those without [59-62]. By contrast, patients with PD without dementia have rates of brain atrophy similar to controls. MRI studies have found that atrophy in patients with PDD is most prominent in limbic and paralimbic structures, including the hippocampus and the anterior cingulate gyrus [60,63]. Another neuroimaging study found that patients with PDD have more white matter hyperintensities in the cholinergic pathways compared with cognitively intact patients with PD .
Whether the neuropathology of PDD differs from that seen in dementia with Lewy bodies (DLB), or whether both conditions exist on a continuum is not clear; there are few comparative neuropathologic studies [65,66]. Cortical Lewy bodies and Lewy neurites appear to substantially contribute to the pathologic substrate for dementia in both conditions [65,67,68]. One small study did find that neuronal loss in the substantia nigra was more significant in PDD than in DLB; other features including cortical Lewy pathology and Alzheimer pathology did not differ in PDD and DLB . By contrast, an evaluation of brain pathology in 57 patients with DLB or PDD found that a longer duration of parkinsonism prior to dementia was associated with more pronounced cortical cholinergic deficits and less severe cortical Lewy body pathology .
Biochemical correlates — There is mounting evidence that cholinergic deficit is a consistent biochemical correlate of cognitive dysfunction in PD:
●Loss of cholinergic neurons in the nucleus basalis of Meynert and decreased cholinergic activity in the cortex appear to be at least as significant in PDD as in AD, if not more so [71-75].
●Anticholinergic drugs often exacerbate cognitive deficits in PD patients . In addition, in one observational study, prolonged use of these agents in early-stage PD increased the risk of subsequent dementia .
●Cholinesterase inhibitors improve cognitive function modestly in PDD. (See 'Cholinesterase inhibitors' below.)
By contrast, dopaminergic loss, while central to the pathogenesis of motor symptoms, is not consistently linked with the degree of cognitive dysfunction in patients with PD . While some neuroimaging studies have implicated frontostriatal dopaminergic pathways , dopaminergic medication has mild, variable effects on cognitive performance; it may improve short-term memory early in the disease but not in more advanced patients [80,81]. While primary dopaminergic influence on cognition may be minimal, one study has shown that dopaminergic and cholinergic degenerations exhibit both independent and, possibly more importantly, interactive contributions to cognitive impairment in PD . (See 'Treatment' below.)
Other neurotransmitter systems may be involved in producing visual hallucinations. A positron emission tomography study found increased serotoninergic type 2A receptor binding in nondemented PD patients with visual hallucinations compared with those without hallucinations . A subsequent study found an increase in the number of serotonergic type 2A receptors in the inferolateral temporal cortex of PD patients with hallucinations .
Cognitive impairment — Cognitive impairment in PD has a heterogeneous cognitive profile that is different from that of Alzheimer disease (AD) [4,85,86]. The general pattern is one of executive dysfunction and impaired visuospatial function, with less prominent memory deficits and relatively preserved language function (table 2) [12,85,87,88].
Executive dysfunction is often present early in the disease course and is manifested by deficiencies in set shifting, attention, and planning . Tests of face recognition, one measure of visuospatial function, are impaired early in the course of PD . Other measures of visuospatial function become impaired in more severe PD and are more impaired in those with, rather than without, dementia. All of these deficits may impair higher-level functioning well before patients meet criteria for dementia [90,91].
Memory deficits, while less prominent than in AD, do occur in PD dementia (PDD) but appear to be related to retrieval of learned information, which is improved by cuing. Conspicuously absent in PDD are aphasia, apraxia, and severe memory loss, features that are more common in AD [6,7,92-96]. (See "Clinical features and diagnosis of Alzheimer disease", section on 'Cardinal symptoms'.)
Comparison of neuropsychological test profiles among patients with dementia with Lewy bodies (DLB), AD, and PDD reveals no differences between DLB and PDD [97,98]. AD patients are more likely to have significant impairment on memory testing, while patients with PDD and DLB are more likely to have visuoperceptive, visuoconstructive, and attentional deficits [97,99].
Neuropsychiatric symptoms — A variety of neuropsychiatric and behavioral symptoms may occur in patients with PD; while many such symptoms can occur in PD patients without dementia, they are increasingly common in patients with more severe cognitive impairment.
Visual hallucinations are common in PD with and without dementia and can be identified in as many as 50 percent of all PD patients [100,101]. Complex visual hallucinations (eg, animals, people) are the most common type, but other visual disturbances can occur, including sensations of presence, sensations of movement in the periphery, and illusionary experiences . Auditory, tactile, and olfactory hallucinations can also occur but are less common.
The presence of visual hallucinations is associated strongly with cognitive dysfunction and to a lesser extent with rigidity and autonomic dysfunction . A variety of antiparkinson drugs, most notably anticholinergic agents, dopaminergic agents, and amantadine, can exacerbate visual hallucinations. In PD patients with normal cognition, those with hallucinations are more likely to develop dementia than are patients without this symptom [6,8,17,104].
Other manifestations of psychosis in PD include delusions, often paranoid and related to spousal infidelity and persecution. These also may be exacerbated by antiparkinson drugs. The neuropsychiatric features of PD are discussed in more detail separately. (See "Management of nonmotor symptoms in Parkinson disease", section on 'Psychosis'.)
PD patients with and without dementia are also subject to depression, anxiety, and sleep disorders [101,102]. Depression occurs in 30 to 40 percent and correlates poorly with degree of motor impairment, but may be predictive of cognitive decline [66,105]. (See "Clinical manifestations of Parkinson disease", section on 'Mood disorders'.)
Sleep disorders include sleep fragmentation, nightmares, and rapid eye movement (REM) sleep behavior disorder (RBD). The last is a parasomnia characterized by vivid dreams in REM sleep without the usual accompanying muscle atonia. This causes individuals to "act out" their dreams, especially when they are vivid or frightening. RBD is generally accepted to be a predictor for development of an alpha-synucleinopathy such as PD or DLB, often appearing years before the onset of overt neurologic or cognitive symptoms . Rates of mild cognitive impairment (MCI) are three to six times higher in PD patients with RBD compared with those without RBD, and RBD is a marker for earlier onset of dementia [102,107-109]. (See "Rapid eye movement sleep behavior disorder".)
Excessive daytime sleepiness (EDS) is also common in patients with PD, both with and without dementia, although the latter has been identified as a risk factor. Some patients have sudden sleep attacks. Both symptoms pose risks for patients who drive. (See "Clinical manifestations of Parkinson disease", section on 'Excessive daytime sleepiness' and "Management of nonmotor symptoms in Parkinson disease", section on 'Daytime sleepiness'.)
Cardinal motor features — Patients with PDD have core motor features of PD that include (table 3):
●Bradykinesia and akinesia
Resting tremor is more specific to PD than are other parkinsonian features; however, PD patients with tremor as their predominant motor symptom are less likely to develop cognitive impairment than are those with more prominent postural instability and gait impairment [110-112]. (See "Clinical manifestations of Parkinson disease", section on 'Cardinal features'.)
Other features — Other signs of neurologic impairment in patients with and without cognitive impairment include autonomic dysfunction, olfactory dysfunction, pain and sensory disturbances, and mood disorders. These are summarized in the tables (table 3 and table 4) and discussed in more detail elsewhere. (See "Clinical manifestations of Parkinson disease".)
Clinical course — The severity of cognitive impairment tends to progress in most patients over the course of the disease, although reported rates of conversion from MCI to dementia are quite variable [92,93].
While most studies suggest that early evidence of cognitive impairment in PD identifies a higher risk of future dementia [6,7,94,113], the prognosis is somewhat uncertain. In a study of PD patients diagnosed within the past 3.5 years, 41 percent had MCI based on Montreal Cognitive Assessment (MoCA) testing, and an additional 21 percent converted from normal cognition to MCI after an additional 18 months of follow-up; 5 percent of patients progressed from MCI to dementia over the same time period . In another study of patients with a longer disease history (mean disease duration five years) and normal baseline cognition, all patients who developed incident MCI over the course of the study period converted to dementia within the next five years . However, in a more recent five-year study, only 29 percent of initially cognitively normal patients developing MCI went on to dementia; 28 percent of patients with baseline MCI and 24 percent of those with incident MCI reverted to normal cognition at study end .
Frank dementia typically occurs late in the clinical course of PD, whereas it is often one of the presenting features of DLB. Similarly, hallucinations are a late-appearing consequence of PD or PDD but are seen early in the course of DLB and may be among the presenting features of the illness. (See 'Dementia with Lewy bodies' below.)
DIFFERENTIAL DIAGNOSIS — PD dementia (PDD) is one of a number of conditions that include dementia and parkinsonism as prominent features.
Dementia with Lewy bodies — The differentiation of PDD and dementia with Lewy bodies (DLB) is somewhat arbitrary. In PDD, dementia occurs in the setting of well-established parkinsonism, while in DLB, dementia usually occurs before, at the same time as, or very shortly after the development of parkinsonian signs. If parkinsonism is present for more than one year before the onset of dementia, it is officially classified as PDD. This "one-year rule" is somewhat arbitrary; the length of time that parkinsonism precedes other symptoms in otherwise similar patients does not correlate with pathologic differences [116,117].
Other features that may help distinguish between PDD and DLB are a faster clinical decline, earlier onset of hallucinations and delusions, and decreased levodopa responsivity for DLB compared with PDD . Parkinsonian features are more likely to be bilaterally asymmetric and more severe in PDD according to most, although not all, studies . Tremor is more common in PDD than DLB . Significant fluctuations in cognition are common in DLB, but rarely occur in PDD.
In one study, volumetric analysis of magnetic resonance imaging (MRI) scans found that patients with DLB had more pronounced cortical atrophy than did patients with PDD despite having similar dementia severity . Another study also found that patients with PDD and DLB with similar dementia severity could be distinguished by patterns of fractional anisotropy on diffuse tensor imaging MRI . However, it seems unlikely that these observations are sufficiently sensitive and specific to aid in diagnosis of DLB versus PDD in individual patients .
Despite these observed clinical trends, no individual symptom characteristic or test finding reliably distinguishes the motor parkinsonism of PD versus DLB. It remains uncertain even whether these are distinct disorders (versus the onset of clinical symptoms of the same underlying pathology at different times). Both disorders are characterized pathologically as an alpha-synucleinopathy. (See "Clinical features and diagnosis of dementia with Lewy bodies".)
Progressive supranuclear palsy — Progressive supranuclear palsy (PSP) is a rare syndrome that can mimic PDD early on. Distinctive early features of this disorder include vertical supranuclear gaze palsy and prominent postural instability with falls [124,125]. The response to levodopa is typically poor, but approximately 20 percent of patients may have some improvement with L-dopa in early stages of the disease . Bradykinesia and rigidity are typically symmetric in onset . Behavioral changes including apathy, disinhibition, dysphoria, and anxiety are common . (See "Diagnosis and differential diagnosis of Parkinson disease", section on 'Progressive supranuclear palsy'.)
Multiple system atrophy — An umbrella term for olivopontocerebellar atrophy, striatonigral degeneration, and Shy-Drager syndrome, multiple system atrophy (MSA) commonly presents with parkinsonism. Other features can include dysautonomia, cerebellar ataxia, and corticospinal tract deficits.
The prominence of these manifestations and the symmetry of onset, absence of tremor, and poor response to levodopa suggest this diagnosis rather than PDD . Traditionally, the dementia of MSA has been thought to be milder than that seen in PSP or corticobasal degeneration (CBD) ; however, a large prospective cohort of patients with PSP and MSA found that dementia, while slightly less common in MSA, was no less severe . (See "Multiple system atrophy: Clinical features and diagnosis".)
Corticobasal degeneration — Patients with this condition (CBD; also referred to as corticobasal syndrome [CBS] when neuropathologic confirmation is lacking) can manifest asymmetric parkinsonism, including bradykinesia, rigidity, and postural instability. More distinctive features can include ideomotor apraxia, alien limb phenomenon, aphasia, and loss of cortical sensory function . Absence of tremor and lack of levodopa response are typical for CBD and help to distinguish it from PDD. (See "Corticobasal degeneration".)
Alzheimer disease — Although subtle extrapyramidal signs may be present in early stages of Alzheimer disease (AD), clinically evident parkinsonism does not develop until late stage AD . In most cases, the relative timing of the appearance of dementia and parkinsonism is obvious, such that this feature, in itself, does not confuse AD and PDD.
Alternatively, AD may develop in a patient with PD, as the two disorders are not rare. This might be hard to distinguish from PDD early on, but the ultimate appearance of cortical dysfunction, such as aphasia or apraxia, or a more profound amnesic syndrome usually suggests the presence of the second dementing condition. Prominent neuropsychiatric symptoms, however, suggest that the dementia is due to PDD rather than AD .
Cerebrovascular disease — Infarctions in multiple vascular territories affecting periventricular and subcortical white matter, basal ganglia, and brainstem can produce dementia and a vascular parkinsonism. An abrupt onset of symptoms, predominant involvement of the lower extremities, a stepwise course, vascular risk factors, focal neurologic signs, pseudobulbar palsy, and evidence of vascular disease on neuroimaging suggest this diagnosis. (See "Etiology, clinical manifestations, and diagnosis of vascular dementia".)
Other considerations — When cognitive impairment occurs in the setting of PD, clinicians should consider possibilities other than dementia. Depression and other neuropsychiatric symptoms are common in PD and may contribute to cognitive impairment. Motor impairments of PD, particularly physical inactivity and bradyphrenia, may lead to the appearance of dementia. Similarly, patients with PD should be screened for causes of delirium that can lead to superimposed cognitive impairment, usually presenting more acutely and with altered sensorium and disrupted attention.
Patients with AD or other dementias may be treated with neuroleptic drugs associated with extrapyramidal side effects. Valproate also has been associated with a syndrome of reversible parkinsonism and cognitive decline (see "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Valproate'). Because medication effects can be prolonged, even for a year or more, after medication discontinuation, a complete history of medication use should be elicited .
EVALUATION — The evaluation of a patient with suspected cognitive impairment first establishes the spectrum of cognitive and behavioral changes and provides a measure of its severity. The cornerstone of the evaluation is the clinical interview, which ideally should be supplemented by interview of an informant. The latter can be particularly helpful in determining premorbid level of functioning and pattern of decline, which the patient may not be able to describe accurately.
There are several basic tests of global cognitive function that can be performed easily at the bedside to screen for cognitive dysfunction and follow changes over time. Of the available tools, we prefer the Montreal Cognitive Assessment (MoCA), which has been found to be an accurate brief bedside test for cognitive impairment in patients with PD . The MoCA is accessible online and in several languages. The Mini-Mental State Examination (MMSE) cannot be solely relied upon to detect disabling cognitive impairment in PD dementia (PDD) because it is not very sensitive to executive dysfunction, a key feature of PDD. (See 'Cognitive impairment' above.)
Neuropsychological testing is more sensitive for defining cognitive impairments than global cognitive scales and provides a more detailed assessment of the pattern of cognitive dysfunction . While not required in all patients for diagnostic purposes, formal neuropsychological testing can be particularly useful when there is diagnostic uncertainty after clinical assessment, when it is important to assess for risk of future cognitive decline, or to help identify patients with patterns of cognitive dysfunction that pose risk for future loss of functional independence. (See "Mild cognitive impairment: Epidemiology, pathology, and clinical assessment", section on 'Neuropsychological testing'.)
Treatable causes of cognitive impairment and dementia should be excluded (see 'Other considerations' above). All medications should be reviewed, particularly those that have been introduced recently. Anticholinergics, tricyclic antidepressants, benzodiazepines, and dopamine agonists are among the more common medications to cause or aggravate cognitive dysfunction in patients with PD. The evaluation also generally includes a neuroimaging study (usually magnetic resonance imaging [MRI]) and basic laboratory evaluations (eg, vitamin B12 level and thyroid function tests). This topic is discussed in more detail elsewhere. (See "Evaluation of cognitive impairment and dementia".)
Because of the frequent comorbidity of depression in PD, screening for depression as an alternative cause or a contributor to cognitive impairment is recommended. The Beck Depression Inventory, the Hamilton Depression Rating Scale, and the Montgomery-Asberg Depression Rating Scale have all demonstrated utility in this setting [135-138] but are available only by license requiring a fee. The Patient Health Questionnaire (PHQ-9) is another questionnaire for depression screening that has been validated in the general population and is available in the public domain (table 5). (See "Screening for depression in adults".)
DIAGNOSIS — Cognitive impairment in PD exists on a continuum of severity. Based on the extent to which it interferes with daily activities, it is generally divided into two major categories: mild cognitive impairment (MCI) and PD dementia (PDD). MCI and dementia in patients with PD are clinical diagnoses that should be suspected in patients with a decline in cognitive abilities that is interfering with daily function (for PDD).
Both diagnoses are contingent upon established parkinsonism (of at least one year) and the exclusion of other diagnoses. (See 'Differential diagnosis' above.)
There are no specific findings on neuroimaging, neuropsychological testing, or laboratory results that are useful in the positive diagnosis of PDD.
Mild cognitive impairment — MCI appears to be prevalent in patients with PD. In one meta-analysis of 41 studies (7053 patients with PD), the prevalence of MCI was estimated to be 40 percent .
Clinical criteria for the diagnosis MCI in patients with PD were first proposed by an international consensus panel in 2012  and have since been validated. A diagnosis of PD-MCI requires all of the following :
●A firmly established diagnosis of PD (see "Diagnosis and differential diagnosis of Parkinson disease", section on 'Diagnosis')
●Gradual decline, in the context of established PD, in cognitive ability reported by either the patient or informant, or observed by the clinician
●Cognitive deficits on either formal neuropsychological testing or a scale of global cognitive abilities
●Cognitive deficits are not sufficient to interfere with functional independence, although subtle difficulties on complex functional tasks may be present
●Other primary explanations for cognitive impairment (eg, delirium, stroke, major depression, metabolic abnormalities, adverse effects of medication, or head trauma) have been excluded
●Other PD-associated comorbid conditions (eg, motor impairment or severe anxiety, depression, excessive daytime sleepiness [EDS], or psychosis) have not significantly influenced cognitive testing, in the opinion of the clinician
Parkinson disease dementia — The core features of PDD include the following :
●A firmly established diagnosis of PD (see "Diagnosis and differential diagnosis of Parkinson disease", section on 'Diagnosis')
●A dementia syndrome with insidious onset and slow progression, developing in the context of established PD and diagnosed by history, clinical, and mental examination, defined as:
•Impairment in more than one cognitive domain (attention, executive function, visuospatial function, memory) (see 'Cognitive impairment' above)
•Decline from a premorbid level
•Deficits severe enough to impair daily life (social, occupational, or personal care), independent of the impairment ascribable to motor or autonomic symptoms
Behavioral features such as apathy, changes in personality or mood, hallucinations, delusions, and EDS are supportive of the diagnosis but not required .
TREATMENT — The treatment of PD dementia (PDD) is symptomatic. No therapies have been shown to modify the course of the disease or influence prognosis.
Cognitive impairment — We initiate treatment trials of cholinesterase inhibitors and/or memantine in a stepwise fashion in patients with PDD, monitoring for side effects and tapering if no improvement or side effects develop.
Cholinesterase inhibitors — We suggest the use of cholinesterase inhibitors in patients with PDD. Most, but not all, studies of cholinesterase inhibitors in PDD have noted a mild to moderate benefit but an increased risk of side effects, including worsened tremor and nausea [140-150]:
●Rivastigmine was evaluated in a 24-week, double-blind, placebo-controlled study of 501 patients with mild to moderate PDD and was found to result in moderate improvement in dementia, mean improvement of 2.1 points on the Alzheimer Disease Assessment Scale-Cognitive Subscale (ADAS-cog) score compared with 0.7-point decline in the placebo-treated group . Clinically meaningful improvements were seen in 20 and 14.5 percent in the treatment and placebo groups, respectively, while clinically meaningful worsening was observed in 13 and 23 percent. This suggests that overall, 15 percent of patients benefited from treatment .
●Donepezil therapy was evaluated in a randomized study of 550 patients with PDD . After 24 weeks, active therapy was associated with significant benefit on the ADAS-cog score but not when the analysis was performed using the predefined statistical model that included a treatment-by-country interaction term. Donepezil also appeared to be associated with benefit by other, although not all, primary and secondary outcome measures, including measures of executive function and attention.
Another potential benefit of cholinesterase inhibitors in PDD is improvement in neuropsychiatric symptoms, such as hallucinations. While this benefit has been inconsistently observed in studies, increased dropouts due to worsened tremor, nausea, and vomiting were also reported [141,146,147]. (See "Management of neuropsychiatric symptoms of dementia", section on 'Antidementia drugs'.)
Based on the available data showing a benefit for cholinesterase inhibitors, albeit modest, we suggest their use in PDD if the patient does not experience intolerable side effects. We generally administer these drugs for a trial period of two to three months and then review the patient's response with family. Treatment is continued if improvement is noted either on bedside testing or by the family. We taper patients off treatment if there has been no improvement or if there are intolerable side effects. When cholinesterase inhibitors are discontinued, they should not be abruptly terminated, if at all possible, but rather tapered to avoid sudden cognitive and behavioral worsening . A more detailed discussion of the use of cholinesterase inhibitors in dementia is found separately. (See "Cholinesterase inhibitors in the treatment of dementia".)
Memantine — Memantine has reported efficacy in moderate to severe Alzheimer disease (AD) and in vascular dementia. One 24-week randomized controlled study of 72 patients with either dementia with Lewy bodies (DLB) or PDD found that patients treated with memantine performed better on the primary outcome assessment measure, the clinical global impression of change, but not on other secondary outcome measures . In a more recent, 24-week randomized controlled study, DLB patients, but not PDD patients, were improved on the same outcome measure . Memantine was well tolerated in these trials and in another shorter study of patients with PDD . However, hallucinations and worsened neuropsychiatric symptoms have occasionally been reported with the use of memantine, suggesting some caution with its use in PDD [155-157].
Psychosis — Visual hallucinations and delusions are a frequent manifestation of PD, particularly in advanced stages of the disease, and patients with PDD may be more sensitive to the neuropsychiatric side effects of antiparkinsonian medications.
Although stopping all potentially offending antiparkinsonian drugs is usually not an option for patients, dose reduction may be possible. Anticholinergic drugs are poorly tolerated in patients with PDD and should be the first medications to be tapered.
When pharmacotherapy is felt to be necessary based on patient distress or risk of harm to self or others, quetiapine and clozapine in low doses are preferred over other antipsychotics. No drug is uniformly effective, however, and patients and families should be made aware of the risk of motor side effects as well as the increased risk of stroke, myocardial infarction, and death that has been associated with antipsychotics as a class . Treatment of psychosis and hallucinations in patients with PD is discussed in more detail separately [135,136,159,160]. (See "Management of nonmotor symptoms in Parkinson disease", section on 'Psychosis'.)
Motor parkinsonism — Treatment of motor parkinsonian symptoms is similar in PD with and without dementia. However, patients with PDD may be particularly susceptible to the neuropsychiatric side effects of dopaminergic medications; these may be dose limiting and/or require concomitant use of antipsychotic agents. Observational studies suggest that L-dopa does not adversely affect cognition overall ; however, in other studies, dopaminergic therapy has been shown to have beneficial effects on some specific cognitive measures and detrimental effects on others [162,163]. Individual baseline cognition and learning style may modulate the effect of dopamine as well .
Anticholinergic agents frequently exacerbate cognitive impairment and are generally avoided in patients with PDD. (See "Initial pharmacologic treatment of Parkinson disease", section on 'Role of anticholinergics'.)
In a study of 29 nondemented PD patients, a program of anabolic and aerobic exercise resulted in an improvement in executive function . The efficacy of exercise in improving cognition in PD is yet to be proven.
Patients with PDD are generally excluded from consideration for deep brain stimulation (DBS) therapy because of poor outcomes. As an example, in a study of 41 patients evaluated after suboptimal response to DBS, 20 percent of these patients had preoperative dementia . On the other hand, in PD patients with mild or no dementia, accumulated evidence suggests that bilateral subthalamic DBS can result in small declines in measures of executive function and a moderate decline in verbal fluency . A review of nine studies presenting head-to-head comparisons of subthalamic DBS with globus pallidus DBS noted that in two of the nine reports, patients undergoing subthalamic DBS fared slightly better than those receiving pallidal surgery on measures of attention and memory, yet quality of life was similar in the two groups .
PROGNOSIS AND COUNSELING — Dementia in PD is associated with reduced patient and caregiver quality of life, reduced survival, increased risks of nursing home admission, and other neuropsychiatric symptoms [6,19,34,135,169-172]. Among a cohort of 180 patients with PD not demented at baseline, incident dementia was associated with a twofold increase in mortality over a mean four-year follow-up, even after controlling for severity of motor symptoms .
Cognitive impairment in patients with PD can affect decision-making capacity. While brief tests of cognitive function such as the Montreal Cognitive Assessment (MoCA) can be useful in predicting the likelihood of impaired capacity, they do not substitute for a formal assessment of capacity, particularly when high-risk decisions are being contemplated. (See "Assessment of decision-making capacity in adults".)
As dementia progresses, patients require greater oversight. A variety of safety concerns arise in a relatively predictable manner, and addressing these issues proactively with patients and families can prevent serious incidents. Driving safety is often one of the earlier issues to be confronted. In contrast to what many patients and families may believe, driving impairment in patients with mild to moderate PD is correlated mostly with cognitive and visual deficits, rather than motor impairment . These and other management issues are discussed in detail separately. (See "Management of the patient with dementia" and "Care of patients with advanced dementia", section on 'Decision-making support'.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Parkinson disease".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Parkinson disease (The Basics)")
SUMMARY AND RECOMMENDATIONS — Cognitive dysfunction and dementia are common in Parkinson disease (PD).
●The risk of dementia in patients with PD increases with age of onset, age of the patient, duration, and severity of the illness. Additional risk factors include the presence of rapid eye movement (REM) sleep behavior disorder (RBD), autonomic dysfunction, and abnormal visual color discrimination. (See 'Epidemiology' above.)
●Neuropathologic studies suggest that Lewy body pathology rather than coincident Alzheimer or cerebrovascular disease is responsible for dementia in most patients with PD. (See 'Pathophysiology' above.)
●The cognitive dysfunction of PD dementia (PDD) is distinct from Alzheimer disease (AD). With PDD, memory impairment is less prominent early on, while executive dysfunction and visual spatial impairments are features that may be apparent and functionally limiting before the patient meets criteria for dementia. (See 'Cognitive impairment' above.)
●Other disorders that produce dementia and parkinsonism include dementia with Lewy bodies (DLB), progressive supranuclear palsy (PSP), multiple system atrophy (MSA), and corticobasal degeneration (CBD). PDD is distinguished from DLB by the onset of dementia in the setting of well-established parkinsonism of at least one year's duration. (See 'Differential diagnosis' above.)
●The evaluation of a patient with suspected cognitive impairment first establishes the spectrum of cognitive and behavioral changes and provides a measure of its severity. The cornerstone of the evaluation is the clinical interview, which ideally should be supplemented by interview of an informant. Neuropsychological testing may aid in the diagnosis of dementia; laboratory studies and neuroimaging are performed to exclude other conditions. The Mini-Mental State Examination (MMSE) cannot be solely relied upon to detect disabling cognitive impairment in PDD because it is not very sensitive to executive dysfunction. (See 'Evaluation' above.)
●Cognitive impairment in PD exists on a continuum of severity. Based on the extent to which it interferes with daily activities, it is generally divided into two major categories: mild cognitive impairment (MCI) and PDD. Both MCI and dementia are clinical diagnoses in patients with firmly established PD of at least one year's duration that should be suspected in patients with a decline in cognitive abilities that is interfering with daily function (for PDD). (See 'Diagnosis' above.)
●We suggest the use of cholinesterase inhibitors in patients with PDD (Grade 2C). These appear to confer a modest benefit for cognition in patients with PDD; evidence for improved neuropsychiatric symptoms is less convincing. Treatment response should be monitored regularly, and therapy should be tapered off if there are no observed benefits or if side effects are intolerable. (See 'Cholinesterase inhibitors' above.)
●Visual hallucinations and delusions are a frequent manifestation of PD, particularly in advanced stages of the disease, and patients with PDD may be more sensitive to the neuropsychiatric side effects of antiparkinsonian medications. Although stopping all potentially offending antiparkinsonian drugs is usually not an option for patients, dose reduction may be possible, starting with anticholinergic medications. When pharmacotherapy is felt to be necessary based on patient distress or risk of harm to self or others, trials of selected medications in low doses are reasonable. (See "Management of nonmotor symptoms in Parkinson disease", section on 'Psychosis'.)
●The treatment of motor parkinsonism is similar to that for the patient with PD without dementia. An exception is that dementia is a relative contraindication for deep brain stimulation (DBS) therapy. (See 'Treatment' above.)