The cognitive side of essential tremor: what are the therapeutic implications?

Abstract

While essential tremor (ET) has traditionally been categorized as a pure motor disease, cross-sectional and longitudinal studies of cognition in ET have demonstrated that these patients may have cognitive dysfunction. Recent epidemiological studies demonstrate an association between ET (particularly with onset after age 65) and increased risk for cognitive impairment and dementia. Although existing studies have generally conceptualized cognitive changes in ET as consistent with a ‘frontosubcortical’ or ‘corticocerebellar’ profile, results from these same studies suggest that cognitive impairment in ET may in fact be heterogeneous. Furthermore, the underlying mechanisms remain uncertain. Cognitive changes could be a byproduct of the cerebellar dysfunction of ET itself; alternately, they may be a feature of concomitant neurodegenerative diseases that have been associated in several studies with ET, including Alzheimer’s disease, Parkinson’s disease or progressive supranuclear palsy. While the study of cognitive dysfunction in ET has received research attention in recent years, the results of these studies have not been translated into the clinical domain and clinical practice. This review first summarizes the current literature on the potential relationships between ET and cognitive change. We then suggest areas of further clinical evaluation and treatment; these suggestions are directed at physicians caring for ET patients who may demonstrate or complain of cognitive impairment. As we discuss, clinicians should ideally screen ET patients for possible signs or symptoms of cognitive impairment in addition to assessing for psychiatric comorbidity and quality of life. These recommendations are in contrast to most current clinical practice, which does not routinely include such assessment among ET patients. To our knowledge, there have been no pharmacotherapeutic trials to date of any agent for cognitive change associated with ET. We believe that studies for this indication are now called for. Future efforts in this direction will also need to take into account the pathobiology of cognitive changes in ET, which itself is an area that is ripe for future investigations.

Introduction

Essential tremor (ET) traditionally has been categorized as a pure motor disease. However, cross-sectional and longitudinal studies have increasingly demonstrated that ET may be associated with cognitive impairment or dementia in a subsection of patients [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008; Gasparini et al. 2001; Troster et al. 2002; Lacritz et al. 2002; Benito-Leon et al. 2006a, 2006b, 2011; Louis et al. 2010a; Bermejo-Pareja et al. 2007; Thawani et al. 2009]. This association does not appear to be accounted for by medication side effects [Lombardi et al. 2001; Higginson et al. 2008] or psychiatric symptoms [Li et al. 2011], and may be related to cognitive changes associated with ET itself; alternately, it may be associated with concomitant neurodegenerative diseases that themselves have been reported to be associated with ET, including Alzheimer’s disease (AD), Parkinson’s disease (PD), Lewy body dementia (DLB) or progressive supranuclear palsy (PSP). This review summarizes the current literature on the potential relationships between ET and cognitive change, and suggests areas of further evaluation and treatment of ET patients with cognitive impairment.

Current literature on the relationships between ET and cognitive change

Association of ET with cognitive dysfunction

Studies on cognition in ET have predominantly assessed two areas: the integrity of various neuropsychological abilities, particularly executive functioning [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008; Gasparini et al. 2001; Sahin et al. 2006; Schwartz et al. 1999; Passamonti et al. 2011] and the extent to which patients with ET are at increased risk for mild cognitive impairment (MCI) or dementia [Benito-Leon et al. 2006b; Bermejo-Pareja et al. 2007; Thawani et al. 2009].

Evaluation of cognitive symptoms and neuropsychological deficits related to ET stem from early cognitive studies of cerebellar dysfunction. Beginning in the mid to late 1980s and early 1990s, observational studies described cognitive dysfunction related to cerebellar degeneration or stroke. Kish and colleagues described difficulties with concept formation, learning of paired associates and cognitive slowing in patients with olivopontocerebellar atrophy [Kish et al. 1988]. Patients with cerebellar cortical atrophy were found to have impaired executive function demonstrated by increased planning time on the Tower of Hanoi Test [Grafman et al. 1992], and by poor performance on tests of fluency and the initiation/ perseveration subtest of the Mattis Dementia Rating Scale [Appollonio et al. 1993]. Visual spatial deficits were described following excision of left cerebellar hemisphere tumors [Wallesch and Horn 1990], as well as in the setting of chronic phenytoin intoxication [Botez et al. 1985], and after left superior cerebellar artery territory infarction [Botez-Marquard et al. 1994]. Linguistic processing was impaired in patients with right cerebellar infarction, with symptoms including agrammatism [Silveri et al. 1994] and impaired linguistic error detection [Fiez et al. 1992].

These clinical observations were reinforced by functional neuroimaging experiments showing cerebellar activation by nonmotor tasks [Allen et al. 1997; Fiez and Raichle, 1997] and by behavioral studies [Schmahmann, 1996] that indicated involvement of the cerebellum in cognitive processing and emotion. Together, clinical and imaging findings led to a designation of the ‘cerebellar cognitive affective syndrome’, which included impairments in executive function (including planning, set-shifting, abstract reasoning, verbal fluency and working memory), often with perseveration or inattention, visual spatial disorganization and impaired visual spatial memory, and difficulties with language production including dysprosodia, agrammatism and mild anomia [Schmahmann and Sherman, 1998; Akshoomoff et al. 1997; Hallet and Grafman, 1997; Molinari et al. 1997; Desmond and Fiez, 1998; Rapoport et al. 2000].

As the cerebellothalamocortical basis of tremor in ET began to be defined in the late 1990s and early 2000s [Middleton and Strick, 2000a, 2000b, 2001; Montgomery et al. 2000; Deuschl et al. 2000; Singer et al. 1994], the possibility of cerebellum-influenced cognitive impairment in ET also began to be considered. Given the deficits outlined in the cerebellar cognitive affective syndrome, it was expected that cognitive changes in ET might include changes in executive function, abstract reasoning, verbal fluency, working memory, attention, visual spatial skills, and language. Two initial studies evaluated the effect of thalamic deep brain stimulation (DBS) on verbal fluency, verbal recall and quality of life in ET [Troster et al. 1999; Lucas et al. 2000] and demonstrated improvement in recall and quality of life and variable results in verbal fluency following DBS. However, Gasparini and colleagues were the first to publish findings concerning neuropsychological function in ET. Utilizing cognitive measures to evaluate frontal lobe function, the researchers compared the average neuropsychological test scores of small groups of patients with ET and a family history of ET (n = 15), patients with ET and a family history of PD (n = 12), patients with PD (n = 15), and healthy controls (n = 15) [Gasparini et al. 2001]. Relative to the healthy control group, the group with ET and a family history of ET demonstrated poorer performance on a task involving conceptualization and set-shifting [Wisconsin Card Sorting Test (WCST)] and on an inhibition task (Stroop task interference condition), but not on another executive function task assessing planning (Tower of London). The group with ET but a family history of PD demonstrated impairments on the same tasks, as well as on a letter fluency task (FAS). The ET group with a family history of PD generally performed more like PD than did the ET with family history of ET group. The authors hypothesized that these findings might occur if ET and PD represented points on a spectrum of motor disorders with executive cognitive dysfunction. Into such a continuum, cognitive impairment among ET patients with a family history of ET versus ET patients with a family history of PD might occur by intermediate degrees, similar to each other but different from either individuals without neurological disease or PD-only patients. They concluded that further studies on larger sample groups were warranted to investigate the cognitive profile of ET to provide a better understanding of its relationship with PD. Utilizing a broader range of neuropsychological tests, Lombardi and colleagues found that, relative to normative test values, 18 ET patients undergoing evaluation for surgical treatment of tremor demonstrated impaired semantic fluency [Lombardi et al. 2001]. Furthermore, when patients’ actual scores were compared with their own expected scores given their estimated intelligence quotient (IQ), the patients also showed impairments in letter fluency, visual confrontation naming, verbal recall, and working memory and attention. While ET patients in this study also demonstrated difficulty with WCST, they demonstrated strengths on other executive measures of conceptualization (including matrix reasoning and similarities) and on visuospatial tasks including facial matching and the Hooper visual organization task. In summary, these initial findings provided preliminary evidence that ET might compromise performance on some, but not all, measures of attention and executive function and may also incorporate nonexecutive abilities.

Over the past decade, numerous studies have continued the neuropsychological evaluation of cognition in patients with ET. For the sake of conciseness, these study findings, including numbers of cases versus controls as well as the specific cognitive domains evaluated and neuropsychological batteries utilized, are summarized in Table 1. All studies used either self-established [Lombardi et al. 2001; Duane and Vermilion, 2002] or previously published standardized criteria for diagnosis of ET [Fahn et al. 1993; Louis et al. 1998; Salemi et al. 1994; Bain et al. 2000]. In addition, other neurological conditions that could mimic ET, such as PD, were specifically evaluated and potential participants with these symptoms were excluded [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008; Gasparini et al. 2001; Lacritz et al. 2002; Benito-Leon et al. 2006a; Sahin et al. 2006; Schwartz et al. 1999; Passamonti et al. 2011; Duane and Vermilion, 2002]. There were certain limitations to some studies. While numbers of cases and controls were limited in most individual studies (the exception being the study by Benito-Leon and colleagues, with 232 cases and 696 controls [Benito-Leon et al. 2006a]), when taken as a whole, overall participant numbers are robust. In addition, while most studies either excluded potential participants with psychiatric comorbidities or medications that could impact cognition [Higginson et al. 2008; Gasparini et al. 2001; Passamonti et al. 2011], or controlled for these conditions [Benito-Leon et al. 2006a], other studies did not [Duane and Vermilion, 2002].

Table 1.

Results of neuropsychological evaluation studies in patients with ET.

	Schwartz et al. (1999)	Gasparini et al. (2001)	Lombardi et al. (2001)	Duane and Vermilion, (2002)	Lacritz et al. (2002)	Troster et al. (2002)	Benito-Leon et al. (2006a)	Sahin et al. (2006)	Higginson et al. (2008)	Frisina et al. (2009)	Kim et al. (2009)	Passamonti et al. (2011)
(Cases) [Controls]	(23) [25]	(27) [15]	(18) [–]	(55) [–]	(13) [–]	(101) [–]	(232) [696]	(16) [16]	(24) [21]	(34) [–]	(34) [33]	(15) [15]
Global cognitive function
MMSE (or similar test)			=		=	=	++			+	++	=
Calculated IQ							=
Clock drawing test								=
Attention and working memory
Digit span forward (WAIS)			++	=				=	++		++	=
Digit span backward			++					=	++		++	=
Spatial span forward (WMS)			=
Spatial span backward			=
Symbol search (WAIS)									+
Letter–number sequencing (WAIS)			++
Picture completion (WAIS)			=						++
Trail Making Test Part A		=					++	=
Brief test of attention (working memory)						++
Visual attention (several tests)				++
Executive functions
Stroop test		++			++	++		+
WCST		++	++	=	++	=		=
Tower of London		=
Tower of Hanoi								=
Go-No-Go test											=
Ruff total design					++
Frontal assessment battery												+
Modified card sorting test												=
Matrix reasoning (WAIS)			=						=
Similarities (WAIS)			=						+
Memory
CVLT relayed recall			+		=	++		=
CVLT Delayed recognition			=		=	++		+
Logical memory recall (WMS)						=	+		+
Hopkins Verbal Learning Free Recall											++
Hopkins Verbal Learning Delayed Recall											++
Hopkins Verbal Learning Recognition											=
Figural memory (WMS)						=
Faces (WMS)									+
Rey Complex Figure Immediate Recall											=
Rey Complex Figure Delayed Recall				=							=
Rey Complex Figure Recognition											++
WMS: visual reproduction, % retention					++
RAVLT (immediate recall)				=			+					=
RAVLT (delayed recall)				=			+					+
Language
Letter fluency		++	++		+	++	++	++	=		++	=
Semantic fluency (animal naming)			++		=	++	=		++		++
Vocabulary (WAIS)			=
BNT			+		=	++		=	+		++
Language comprehension (Token Test)												+
Visuospatial
Benton facial recognition test			=			++		+
Benton line orientation test								+
Rey Complex Figure Copy				=				=			=
Block designs (WAIS)					=			=	++
Hooper visual organization test			=			++		=
Judgment of Lines Orientation												=
Visual recognition deficits					+	+		+			=
Mood
Depression			+	++	+				+			=
BDI												=
HAM-A (anxiety)												=
Motor
Grooved Pegboard						++
Self-designed visuomotor tracking test	+

Key

⁼p value not significant (p > 0.05)

⁺p ≤ 0.05

⁺⁺p ≤ 0.001

not assessed in given study

BDI, Beck Depression Inventory; BNT, Boston Naming Test; CVLT, California Verbal Learning Test; ET, essential tremor; HAM-A, Hamilton Anxiety Rating Scale; IQ, intelligence quotient; MMSE, Mini-Mental State Examination; RAVLT, Rey Auditory Verbal Learning Test; WAIS, Wechsler Adult Intelligence Scales; WCST, Wisconsin Card Sorting Test WMS; Wechsler Memory Scale.

Evaluating the results as presented in Table 1, it is apparent that over the past decade many studies have evaluated multiple cognitive domains for evidence of cognitive change in patients with ET. Over time, studies have become increasingly complex and sophisticated in the cognitive batteries that they include for cognitive assessment. As a result, many areas of cognition have been comprehensively assessed in ET patients. Most studies demonstrated that patients with ET have clear evidence of cognitive impairment. The majority of studies evaluating executive functioning (seven out of nine) have found deficits in this area, as manifested by low scores on tests such as the WCST [Passamonti et al. 2011], Stroop Test [Gasparini et al. 2001; Troster et al. 2002; Lacritz et al. 2002] Frontal Assessment Battery [Passamonti et al. 2011], verbal fluency tests [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008; Gasparini et al. 2001; Troster et al. 2002; Lacritz et al. 2002; Sahin et al. 2006], and the Similarities [Lombardi et al. 2001] subtest of the Wechsler Adult Intelligence Scales (WAIS). Deficits in working memory (e.g. WAIS Digit Span forward [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008] and backward, and the WAIS symbol search [Higginson et al. 2008] and letter–number sequencing tests [Lombardi et al. 2001]), verbal fluency (letter fluency [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Gasparini et al. 2001; Troster et al. 2002; Lacritz et al. 2002; Benito-Leon et al. 2006a; Sahin et al 2006] and semantic fluency (animal naming) [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008; Troster et al. 2002] and verbal recall (e.g. California Verbal Learning Test (CVLT) [Lombardi et al. 2001; Troster et al. 2002], Wechsler Memory Scale (WMS) Logical Memory Test [Higginson et al. 2008; Benito-Leon et al. 2006a] and the Hopkins Verbal Learning Test Free and Delayed Recall [Kim et al. 2009]) might also be best conceptualized as executive deficits. However, impairment has also been reported on tasks that generally have low executive demands in areas including global cognition: Mini-Mental State Examination (MMSE) [Kim et al. 2009; Frisina et al. 2009]); language (e.g. Boston Naming Test [Kim et al. 2009; Frisina et al. 2009; Lombardi et al. 2001; Higginson et al. 2008; Troster et al. 2002]); verbal recognition (e.g. CVLT [Troster et al. 2002; Sahin et al. 2006], WMS Logical Memory Test [Higginson et al. 2008; Benito-Leon et al. 2006a] and the Hopkins Verbal Learning Test [Kim et al. 2009]); and visual spatial function (e.g. Benton facial recognition test [Troster et al. 2002; Sahin et al. 2006], Benton line orientation test [Sahin et al. 2006], WAIS Block Design [Higginson et al. 2008] and the Hooper Visual Organization Test [Troster et al. 2002; Duane and Vermilion, 2002]).

The biological basis for the cognitive dysfunction associated with ET is unclear, although a number of mechanisms may be responsible. It has been hypothesized that cognitive deficits are mediated through cerebellar, thalamic and frontal connections that are compromised in ET [Kim et al. 2009; Troster et al. 2002; Sahin et al. 2006]. Much of the literature supports that ET is a disorder of cerebellar dysregulation, including the presence of intention tremor [Leegwater-Kim et al. 2006; Louis et al. 2009b], gait ataxia [Singer et al. 1994; Stolze et al. 2001], oculomotor abnormalities [Helmchen et al. 2003] and problems with dysrhythmia and motor learning [Trillenberg et al. 2006; Avanzino et al. 2009; Bares et al. 2010; Farkas et al. 2006; Shill et al. 2009; Kronenbuerger et al. 2007]. In addition, postmortem studies in ET have indicated the presence of a variety of structural and degenerative changes in the cerebellum, including increased number of Purkinje cell axonal swellings (‘torpedoes’) [Louis et al. 2007b; Axelrad et al. 2008], increased number of displaced or heterotopic Purkinje cells [Kuo et al. 2011], reduction in number of Purkinje cells in some studies [Axelrad et al. 2008] and hypertrophy of basket cell processes (‘hairy baskets’) [Erickson-Davis et al 2010]. Moreover, DBS surgery in ET targets the specific thalamic nucleus (i.e. ventral intermediate) that is the cerebellar receiving area, providing further support for pathway dysregulation in ET. More recently, the presence of dysregulated cerebellar pathways in ET has been supported by neuroimaging findings, including functional magnetic resonance imaging (MRI) [Bucher et al. 1997], positron emission tomography [Jenkins et al. 1993; Wills et al. 1994; Colebatch et al. 1990], ¹H magnetic resonance spectroscopic imaging [Louis et al. 2002; Pagan et al. 2003], diffusion tensor imaging [Shin et al. 2008; Klein et al. 2011; Nicoletti et al. 2010], voxel-based morphometry [Benito-Leon et al. 2009; Quattrone et al. 2008] and studies using other automated volumetric methods [Cerasa et al. 2009] which demonstrate the presence of functional, metabolic and structural abnormalities in the cerebellum of ET patients.

The idea that cognitive impairment in ET may reflect the distal effects of cerebellar changes via cerebellothalamocortical or cerebellothalamofrontal networks [Kim et al. 2009; Troster et al. 2002; Sahin et al. 2006] is consistent with a growing literature documenting the role of the cerebellum in facilitating psychological, cognitive and learning activities [Lombardi et al. 2001; Schmahmann and Sherman, 1998; Middleton and Strick, 2000a; Watson, 1978; Peterburs et al. 2010]. In addition, many reports of neuropsychological performance in patients with ET have highlighted a pattern of attentional and executive dysfunction that is analogous to that observed in neurodegenerative diseases affecting subcortical white matter [Schmidtke and Hull, 2002; Traykov et al. 2002; Tierney et al. 2001; Doddy et al. 1998; Lafosse et al. 1997] and basal ganglia structures [Troster et al. 2002; Lundervold et al. 1994; Pillon et al. 1995; Cummings, 1998].

While cognitive impairment in ET may be mediated by the processes or pathways specifically involved in the disease, an alternate hypothesis is that some cases of ET may also manifest a spectrum of neurodegenerative diseases, including AD, DLB, PD or Parkinson Plus Disorders. As a result, associated cognitive impairment and dementia actually may occur as a result of these concomitant neurodegenerative diseases (i.e. ‘ET plus’) [Louis, 2005b]. Although existing studies have generally conceptualized cognitive changes in ET as consistent with a ‘frontosubcortical’ or ‘corticocerebellar’ profile, results from these same studies suggest that cognitive impairment in ET may in fact be heterogeneous, with a subset of changes uncharacteristic of cerebellar pathology. For example, three out of five studies that have evaluated verbal recognition memory (a measure of memory storage rather than retrieval) document impairment in patients with ET [Kim et al. 2009; Lombardi et al. 2001; Troster et al. 2002; Lacritz et al. 2002; Sahin et al. 2006]. This cognitive deficit is generally not a component of cerebellar cognitive syndromes [Neau et al. 2000; Daum and Ackermann, 1997] or of frontosubcortical impairment in dementia [Schmidtke and Hull. 2002; Traykov et al. 2002; Tierney et al. 2001; Doddy et al. 1998; Lafosse et al. 1997; Lundervold et al. 1994; Pillon et al. 1995; Cummings, 1998]. Rather, recognition deficits reflect a primary amnestic syndrome that is a classic sign of hippocampal dysfunction [Beyer et al. 2013; Deweer et al. 1995; Manns et al. 2003] and a cardinal feature of AD [Remy et al. 2005; Lekeu et al. 2003; Hamilton et al. 2004; Pillon et al. 1993]. Moreover, while language and visual spatial deficits in ET may be accounted for by disruptions in cerebellocortical networks, it may also be the case that such deficits reflect local changes in the temporal and parietal cortex consistent with early AD.

Association of late-onset ET with risk for MCI and dementia

This latter hypothesis is further supported by recent epidemiological studies demonstrating an association between ET (especially with onset after age 65) and risk for dementia. A series of papers from the Neurological Disorders in Central Spain (NEDICES) study examined the associations between ET and cognitive dysfunction [Benito-Leon et al. 2006a; Louis et al. 2010b], prevalent MCI [Benito-Leon et al. 2011], and both prevalent and incident dementia [Benito-Leon et al. 2006b; Bermejo-Pareja et al. 2007]. ET cases were gathered from three communities in central Spain and all participants were 65 years of age or older. The investigators used a population-based case–control study design matching patients with ET and controls for age, gender, education, depressive symptoms and medications that could potentially affect cognition. The neuropsychological battery tested global cognitive performance, executive function, memory and pre-morbid intelligence. Subjects were also asked whether they suffered from forgetfulness or depression. The investigators demonstrated that ET cases performed less well than controls on neuropsychological testing, particularly on tests of memory and executive abilities, and a larger proportion of cases reported forgetfulness [Benito-Leon et al. 2006a; Louis et al. 2010b] Moreover, cross-sectional analyses revealed that ET with tremor onset after age 65 was associated with an increased odds of MCI [adjusted odds ratio (OR) = 1.57, 95% confidence interval (CI) 1.03–2.38, p = 0.03], whereas cases with tremor onset prior to age 65 years and controls were equally likely to have MCI (adjusted OR = 0.73, 95% CI 0.34–1.57, p = 0.43) [Benito-Leon et al. 2011]. These study subjects were then followed longitudinally for a mean of 3.2 years, and the rate of cognitive decline appeared to be faster in ET cases with tremor onset after 65 years of age than in controls [Louis et al. 2010a]. The same study demonstrated an association between ET with onset after 65 years of age and prevalent dementia [Benito-Leon et al. 2006b] (adjusted OR = 1.70, 95% CI 1.04–2.76, p = 0.03), as well as incident dementia [Bermejo-Pareja et al. 2007] over a mean follow-up duration of 3.2 years [relative risk (RR) = 1.98, 95% CI 1.14–3.45, p = 0.01).

A second prospective, population-based study of elders (study enrollment age ≥65 years but ET age of onset unknown) was conducted in northern Manhattan, New York [Thawani et al. 2009] and demonstrated similar results. In cross-sectional analyses, 31 of 124 (25.0%) ET cases had prevalent dementia versus 198 of 2161 (9.2%) controls (adjusted OR = 1.84, 95% CI 1.13–2.98, p = 0.01). In prospective analyses, 17 of 93 (18.3%) ET cases versus 171 of 1963 (8.7%) controls developed incident dementia [adjusted hazard ratio (HR) = 1.64, 95% CI 0.99–2.72, p = 0.055). Thus, in a second population-based study of elders, ET was associated with both increased odds of prevalent dementia and increased risk of incident dementia.

In both the New York [Thawani et al. 2009] and Spanish studies [Benito-Leon et al. 2006b; Bermejo-Pareja et al. 2007], the majority of ET cases who developed incident dementia carried clinical diagnoses of probable AD, making this a likely etiology of their cognitive symptoms. Postmortem studies in different study groups also suggest that there may be an increase in Alzheimer-type changes in the ET brain (i.e. neurofibrillary tangles and neuritic plaques) [Louis et al. 2007b], providing further support for this etiologic hypothesis.

Cortical Lewy bodies have not been a prominent feature of ET pathology in most studies [Louis et al. 2007b; Shill et al. 2008]. However, one group recently described a subset of ET patients whose postmortem examination revealed an anatomically restricted manifestation and distinctive pattern of DLB (with many Lewy bodies in the locus ceruleus, rare Lewy bodies in the substantial innominata and dorsal vagal nuclei, and rare or absent Lewy bodies in the substantia nigra), which has not been observed in patients with subclinical or clinically manifest PD [Louis et al. 2005]. This raises the question as to whether a proportion of ET cases might have forms of DLB; however, the role that Lewy body pathology may play in cognitive change, MCI, or dementia in patients with ET has yet to be investigated prospectively or in large epidemiologic studies.

Neurodegenerative diseases traditionally categorized as Parkinson Plus Disorders may also play a role in cognitive impairment in ET. Our group recently reported an evaluation of 89 ET patients prospectively collected at the ET Centralized Brain Repository over the course of its first 9 years [Louis et al. 2013]. Of this group, 11 (12.4%) ET patients were found to have postmortem changes consistent with PSP. PSP is a syndrome characterized by supranuclear palsy, postural instability and some degree of Parkinsonism in many patients; it may also be characterized by dementia. Five patients in our group had cognitive complaints later in life and four of these individuals were diagnosed by their treating physician as having had dementia. The latency from onset of ET to dementia ranged from 5 to 47 years. The prevalence of PSP in this ET sample was larger than the population prevalence of PSP (0.001–0.00655%), and was also 2–5 times the proportion of normal cases with incidental PSP reported in two prior autopsy series [Shill et al. 2008; Rajput et al. 2004]. In a Canadian ET postmortem series, two out of 20 (10%) ET patients were also reported to have had PSP, which is a value that is similar to the 11 out of 89 (12.4%) reported above [Rajput et al. 2004], but given the small numbers (n = 2 with ET + PSP), the authors did not draw attention to this possible connection. In an Arizona ET postmortem series, one out of 24 (4.2%) had PSP [Shill et al. 2008]; however, this study specifically excluded subjects with ET and PD as well as ET with dementia. These findings raise the question of what proportion of ET patients who develop what is presumed during life to be concomitant PD or AD actually have PSP. This is particularly important when speaking about cognitive change in ET, as executive impairment is a notable feature of PSP [Williams and Lees, 2009; Litvan et al. 1996] and thus may account for some instances of ET patients with cognitive impairment.

Association of ET with psychiatric disorders

Psychiatric issues such as depression and anxiety, personality changes, social phobias and low measures of subjective health status or health-related quality of life are all frequently observed in patients with ET [Li et al. 2011; Miller et al. 2007; Siddiqui et al. 2009; Louis, 2005a, 2010a; Poewe et al. 1990; Lorenz et al. 2006; Chatterjee et al. 2004; Louis et al. 2007a, 2008; Ozel-Kizil et al. 2008; Findley, 2004; Galvin, 2009; Woods et al. 2008; Dogu et al. 2005]. Some studies have estimated that as many as 30% of patients with ET have mild depressive symptoms [Lorenz et al. 2006]. Depression in ET may have different clinical manifestations than in patients with primary affective disorders, in that patients with ET are more likely to report concentration difficulties and fatigue [Li et al. 2011]. Although the association between psychiatric issues and ET is documented in many studies, more information is needed regarding their initiation and evolution. Traditionally, psychiatric symptoms have been regarded as occurring in response to debilitating neurological symptoms in ET [Chatterjee et al. 2004]. However, other evidence suggests that depression might occur before the motor symptoms of ET manifest; in such situations this psychiatric disorder cannot be a secondary response to disability [Louis, 2010a; Louis et al. 2007a; Dogu et al. 2005].The pathophysiological basis of depression and anxiety in patients with ET is currently unknown and awaits prospective studies. Importantly, however, depression does not appear to account for cognitive deficits in ET [Li et al. 2011].

Therapeutic implications

Implications for patient care: assessment and counseling

A variety of studies demonstrate that patients with ET are at increased risk for cognitive deficits. Specifically, those with tremor onset at age 65 or older are at increased risk for MCI (OR = 1.57, 95% CI 1.03–2.38, versus controls) [Benito-Leon et al. 2011] with a faster rate of cognitive decline once impairment has occurred [Louis et al. 2010a]. Similarly, individuals with ET are also at increased risk for prevalent dementia (OR = 1.70–1.84 versus controls) [Bermejo-Pareja et al. 2007; Thawani et al. 2009], and incident dementia (RR = 1.64–1.98 versus controls) [Bermejo-Pareja et al. 2007; Thawani et al. 2009], with the most frequent presumed etiology being concomitant AD [Louis et al. 2009b].

What are the therapeutic implications of these findings? To begin with, clinicians should consider routinely assessing ET patients (particularly those with older age of onset) for possible signs or symptoms of cognitive impairment. Briefly, such an interview could include an assessment of the patient’s ability to recall recent events and appointments, and their ability to independently conduct somewhat complex tasks such as management of finances, driving and housework. Ancillary interviews of close friends or family members for their opinions of the patient’s cognition should also be strongly considered. Clinician initiation of this discussion may assist many patients with ET and their families who are reluctant to volunteer such information spontaneously. Patients who are experiencing cognitive difficulties may have uncertainty about the meaning of their problems, and an informed and open discussion with an educated practitioner can allay some of their anxieties. These suggestions are in contrast to current clinical practice, which does not routinely include such assessment of ET patients and which may further judge evidence of such cognitive decline, if detected, to be inconsequential or related to normal aging or medication side effects [Tolosa and Molineuvo, 2010].

If concerns are raised during the clinician interview with the patient or family members, additional follow-up steps may be taken. For example, psychometric screening tests such as the Montreal Cognitive Assessment (MoCA) [Nasreddine et al. 2005] take approximately 10 minutes to administer in an office environment and can help the clinician to assess a patient’s cognitive performance and to track the clinical evolution of these cognitive deficits over time. The MoCA has been shown to have excellent specificity and sensitivity for detecting MCI and dementia, and for tracking cognitive change over time, outperforming the MMSE in these regards [Dong et al. 2012; Freitas et al. 2013]. The Addenbrooke’s Cognitive Examination Revised (ACE-R) [Mioshi et al. 2006] has also been shown to be particularly useful in the detection of cognitive impairment in individuals with movement disorders [Bak et al. 2005]. The ACE-R, designed to be a more thorough cognitive screening tool than the MMSE, includes six subscores evaluating orientation, attention, memory, verbal fluency, language and visuospatial functioning. The ACE-R has high sensitivity and specificity for the diagnosis of dementia in PD using a cutoff score of 83 [Reyes et al. 2009], and more recent work has shown it to be a highly useful screen for the diagnosis of MCI in PD using a cutoff score of 89 [McColgan et al. 2012]. Moreover, performance on aspects of the ACE-R, particularly the verbal fluency subscore, has been shown to inform the differential diagnosis of Parkinsonian syndromes including PD, PSP and corticobasal degeneration (CBD) [Rittman et al. 2013]. Examination of performance across ACE-R subscores may also provide preliminary information regarding the profile of cognitive impairment and the extent to which patients present with primary deficits in executive function versus memory. However, individuals who demonstrate cognitive deficits on screening should be referred for more extensive neuropsychological evaluations through a neuropsychologist to determine the extent of cognitive limitations and to more comprehensively examine whether the pattern of deficits is consistent with a known neurodegenerative etiology such as AD, PD, DLB or PSP.

While neuroimaging methods have been used to support the diagnosis of several cognitive neurodegenerative diseases, no such studies to date have evaluated the use of specific neuroimaging tests to investigate the nature of cognitive changes in ET. It may be possible to use functional imaging such as fluorodeoxyglucose positron emission tomography (FDG-PET) and single photon emission computed tomography (SPECT) to determine whether some ET patients also have AD pathology, particularly if the patient’s cognitive profile is supportive of this possibility; however, specific studies have not yet been performed to evaluate this potential association. While ¹²³I-2β-carbometoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl) nortropane (¹²³I-FP-CIT) scans for the dopamine transporter have demonstrated the capacity to distinguish ET and healthy controls from patients with PD, DLB and PSP based on reduced levels of dopamine active transporter (DAT) in the latter conditions [Roselli et al. 2010], they cannot distinguish ET from healthy controls. As a result, future studies are needed to specifically examine these different imaging indications.

There are many potential functional aspects to cognitive impairment and the evaluation of cognitive performance would allow clinicians to proactively identify areas of potential cognitive and functional difficulty [Kronenbuerger et al. 2009; Louis et al. 2010a]. In a recent population-based study, lower cognitive test scores in ET were associated with more self-reported functional difficulty, indicating that cognitive deficits in ET seem to have functional consequences [Louis, 2010c]. This was supported by a second study which demonstrated that Minimum Data Set Activities of Daily Living Section (MDS-ADL) scores were significantly correlated with overall cognitive level (as assessed by the MMSE) but not with motor disability among patients with ET. Overall, these findings highlight that the cognitive aspects of ET may be more functionally disabling than the hallmark motor features. If areas of cognitive functional difficulty are identified, modifications for daily cognitive tasks can be discussed with the patient and his or her family to identify means of assistance or reassignment of difficult tasks to others.

The clinician should keep in mind that functional difficulties that seem out of proportion to any underlying cognitive impairment may be a sign of a co-occurring mood disturbance. Thus, clinicians should consider assessing these patients for quality of life and depression [Louis, 2010c; Louis et al. 2010a], as standard treatment with selective serotonin reuptake inhibitors (SSRIs) or other antidepressant regimens is available and effective in patients with ET.

Potential pharmacotherapies for cognitive complaints in ET

To our knowledge, there have been no clinical trials to date of any agent for cognitive change associated with ET. As a result, studies for this indication are needed. Given that many of these patients with ET demonstrate impairment in attention and memory, it may make sense for trials of cholinesterase inhibitors such as donepezil, galantamine and rivastigmine to be conducted. Memantine, an N-methyl-d-aspartic acid (NMDA) receptor antagonist used for treatment of moderate to severe AD has been investigated in studies for its anti-tremor effects in animals [Iseri et al. 2011] and in humans [Handforth et al. 2010]. While animal studies showed that memantine had a neuroprotective effect on cerebellar and inferior olivary neurons [Iseri et al. 2011], human studies indicate that the average effect of memantine on tremor was mild or not significant except for a small subset of patients for whom memantine conferred meaningful tremor benefit [Handforth et al. 2010]. It should be noted that while the beneficial effect of memantine on tremor was primarily noted at higher dosages (i.e. 30–40 mg/day), adverse events were also more common at higher doses and included dizziness, somnolence and fatigue [Handforth et al. 2010]. Cognition was not studied as an endpoint in either of these studies.

Further study recommendations

Future studies are needed to better characterize the cardinal neuropsychological features of ET to aid in accurate clinical diagnoses and appropriately categorize patients for inclusion in clinical treatment trials. Additional case-control and longitudinal studies are needed to examine the potential relationships between ET and the mechanisms contributing to cognitive, personality and psychological change. Autopsy studies and genetic susceptibility studies are also critical for exploration of potential common biologic and genetic pathways between ET and neurodegenerative diseases associated with cognitive impairment such as AD, PD, DLB and PSP. Neuroimaging studies present another modality for examining these potential relationships. Finally, once these patterns and relationships are better characterized, clinical trials could examine the potential utility of cholinesterase inhibitors and memantine for treatment of cognitive impairment in ET.