Cognitive Deficits in Essential Tremor

As mentioned above, early studies of essential tremor did not detect cognitive deficits in patients with this condition.^[1-6] Changes in cognition were, however, evident in patients with essential tremor undergoing deep brain stimulation (DBS).^[105,106] In one study, extensive psychometric evaluation of 40 essential tremor cases before and 3 months after DBS established that this technique was associated with improvements in visuoperceptual abilities, visual attention, delayed word list recognition, and prose recall in these patients.^[106] Unfortunately, this investigation did not include a control group against which improvements in cognitive functioning could be compared, although it did indicate that treatments designed to alleviate tremor can also alleviate cognitive deficits. Similar findings were observed in a case study conducted by Lucas et al. at the Mayo Clinic, Jackson, FL, USA. These authors showed that in addition to providing relief from tremor, bilateral thalamic DBS attenuated essential tremor-associated deficits in verbal fluency and verbal recall. In the study participant, deficits in verbal fluency and verbal recall were greater in the off-stimulation condition than when the stimulators were active.^[107] The studies described above provide evidence that patients with essential tremor can have cognitive deficits; however, the relationship between this disorder and cognitive dysfunction was not specifically addressed until early in the 21^st century.

Clinical Studies

In 2001, two papers explored the cognitive aspects of essential tremor. In the first of these studies, Gasparini et al.^[25] investigated frontal lobe dysfunction in 27 patients with essential tremor (15 with a familial history of essential tremor and 12 with a family history of PD), 15 patients with PD, and 15 healthy controls. All patients and controls were tested on the Wisconsin Card Sorting Test (WCST), Stroop Test, Verbal Fluency Task (FAS), Tower of Hanoi, and Trail Making Test (TMT). Both groups of essential tremor patients and the patients with PD had clear deficits compared with healthy controls on some parts of the Stroop Test and the WCST. Only the patients with essential tremor with a family history of PD and the patients with PD had deficits in attention (time to complete TMT series A) and verbal fluency compared with healthy controls. Both of the essential tremor groups had better overall scores on the test battery than the PD group, but their scores were worse than those obtained from the healthy control group. Nevertheless, the essential tremor group with a family history of PD were shown to have mild frontal lobe dysfunction.

In the second study, Lombardi et al.^[26] tested 18 patients with chronic essential tremor and 18 patients with PD on an extensive psychological battery that included tests of attention, executive functioning, visuospatial and memory functions, reasoning and conceptualization, and mood (Geriatric Depression Scale; see Supplementary Table 1). No control group was included in this study, but the data indicated that patients with essential tremor had cognitive impairments, including deficits in naming, short-term memory, working memory, verbal fluency and mental set-shifting. These patients were also demonstrated to experience significantly higher levels of depression than the patients with PD. In general, the deficits were mild in the essential tremor group, but ranged from unnoticeable to severe. In this study, the psychological deficits associated with essential tremor did not correlate with tremor severity or medications for tremor or depression.

Further studies have also shown that patients with essential tremor have several cognitive deficits (Supplementary Table 1).^[27-33] For example, Duane et al.^[27,28] showed in a retrospective evaluation of 55 patients with essential tremor that 28% of these patients had deficits in executive functioning (assessed on the WCST). Furthermore, substantial deficits in visual attention were seen in this group of patients, and the prevalence of depression (49%) and anxiety (55%) was high.

In a study of 13 chronic essential tremor cases who did not have dementia, Lacritz et al.^[29] detected mild impairment on one or more cognitive measures in 12 of the 13 patients. This group tended to have most difficulty on tasks associated with initiation, executive function and memory, and three patients included in the study had depression.

Tröster et al.^[30] have described the largest consecutive series of essential tremor cases to date. In their study, cognitive functioning was assessed in 101 essential tremor cases who were set to undergo either thalamotomy or thalamic stimulation. The group IQ was shown to be normal in this cohort; however, 25% of the group scored below the standard cut-off for dementia in the Mattis Dementia Rating Scale. Furthermore, after an extensive neuropsychological evaluation (see Supplementary Table 1), the group was shown to have significantly lower than average scores on measures of complex auditory and visual attention, response inhibition, executive functions, verbal fluency, and immediate recall of a word list. The deficits in these psychometric measures were generally small, suggesting that the cognitive changes associated with this disorder reflect mild clinical impairment. These deficits did not seem to be associated with comorbidities, and did not correlate with the incidence of anxiety or depression.

In another study, Sahin et al.^[31] assessed the clinical characteristics of a group of young essential tremor cases (n = 16), and measured their brain activity using single-photon emission CT (SPECT). The study showed that these patients had deficits in global attention, language, memory, visuospatial functions, and executive functions compared with the comparison group (healthy controls). However, no clear relationship between the SPECT data and neuropsychological findings was observed.

In two further studies, Higginson et al.^[32] and Kim et al.^[33] aimed to identify which cognitive deficits were associated with essential tremor by use of extended psychometric batteries. These studies both included meticulous selection of cases and controls. Results from the Higginson et al. study indicated that the essential tremor group performed significantly less well than normal controls across multiple domains, including attention, processing speed, memory and visuoperception, and performed similarly to patients with PD. No relationship between the psychological deficits and the incidence of depression or use of medications was found in the essential tremor group. Of note, the significantly younger age of the control group is a limitation of this study. The results of the Kim et al.^[33] study showed that the mean Mini-Mental State Examination (MMSE) score was significantly lower in the essential tremor group than in controls. Moreover, the essential tremor cases were also shown to have mild deficits in attention, working memory, verbal memory and executive functions. Age and educational status significantly correlated with psychological deficits in this patient cohort.

Fields et al. have reported psychometric findings in 40 patients with essential tremor 12 months after thalamic DBS.^[108] After this period of time, no substantial deleterious effects on cognition were observed, and DBS was associated with a significant reduction in the patients' tremor score and anxiety levels. The quality of life of the patients was also shown to improve. However, many of the essential tremor cases showed declines rather than improvements on language and visual memory tests. Verbal fluency declined significantly in four (10%) of the patients, all of whom showed diminished verbal fluency at baseline, indicating that reduced preoperative verbal fluency might predispose the individual to further fluency declines after DBS. This interesting prospective study clearly indicates that the cognitive disturbances associated with essential tremor are not dependent on the tremor itself, as the tremor was ameliorated with DBS treatment; however, cognitive disturbances might have been progressive in this cohort.

A review of the above evidence indicates that deficits in attention, executive functions, verbal fluency and several memory functions including working memory— phonemic and delayed memory—are consistently present in patients with essential tremor.^[26,29,32] By contrast, visuospatial functions, except for visual spatial attention, do not seem to be consistently affected in patients with this disorder. The above literature does not indicate that deficits in general intelligence or global cognitive performance are characteristic features of essential tremor, although the Kim et al.^[33] study did show a small but statistically significant (P <0.001) decrease in MMSE score in patients with this condition compared with controls. Of note, however, global cognitive function was evaluated only in a minority of the studies mentioned above. In one study, information processing speed (symbol search) was slower in patients with essential tremor than in healthy controls,^[32] whereas simple calculation (assessed on the MMSE) and reasoning were not affected in the majority of patients in the above clinical series.^[18,30,32] Some of the studies indicated that patients with essential tremor perform at a similar standard to patients with PD on psychometric tests. Nevertheless, in general, patients with essential tremor performed less well on tests of word fluency and attention and better on tests of reasoning and calculation than patients with PD.^[25,26,32]

The nature of the essential tremor-related cognitive deficits highlighted above are mild, but their prevalence indicates that dysfunction in the dorsolateral prefrontal cortex and/or its connections to other brain areas, including the cerebral-cerebellar loop, might be contributing factors to the pathophysiology of this tremorogenic disorder.^{[25,30-33,40,65]}

The studies reviewed above have a number of limitations. For example, some of the studies performed psychometric analyses with the sole intention of monitoring the possible adverse effects of DBS in patients with essential tremor, rather than directly exploring the relationship between cognitive deficits and this condition. As a result, the lack of psychometric objectives in some of the studies might have had a negative impact on their capacity to identify cognitive dysfunction in essential tremor cases. Most of the clinical data, with the exception of the Tröster et al. study,^[30] were derived from small numbers of study participants, were based on limited psychometric evaluations that focused mainly on executive functions, or did not employ a control group. In fact, only four of the above studies had a control group,^[25,31-33] and in two of these studies the mean ages of the control groups were noticeably younger than those of the essential tremor groups.^[25,32] Furthermore, the psychometric data obtained in the studies were not always corrected for confounding factors such as depression, medication or comorbidities.^[12] In the clinical series in which these corrections were made, no differences in cognitive performance were observed between patients with essential tremor alone and patients with this disorder who also had depression or were taking medications, such as sedatives.^[26,29,32] In general, patients with essential tremor who had severe comorbidities were not included in the above studies,^[30] indicating that these clinical series might have been subject to selection bias.

Despite these limitations, the studies consistently showed that patients with essential tremor had cognitive deficits. Moreover, the cognitive deficits and mood disorders commonly experienced by patients with this tremor are similar to the cognitive and mood disorders described in cerebellar cognitive affective syndrome (CCAS).^{[25,30,32,109-111]} CCAS has been observed in patients with acute and chronic cerebellar disorders, in which cerebellar damage is considered to underlie the changes in cognition and mood associated with the disorder.^[109,110] This syndrome, which is produced by diverse cerebellar damage and/or damage to cerebellar connections to other brain areas, would explain the neuropsychological and emotional findings in patients with essential tremor.^{[25,30,32,111]} Of note, however, cerebellar lesions frequently produce inconsistencies in cognitive impairments, and patients with diseases of the cerebellum often perform within the normal range on cognitive tests.^[112]

Population Studies

The limitations observed in the clinical series described above have been partially overcome in a population-based study performed in Spain. The NEDICES study assessed the prevalence and incidence of the main neurological disorders (including essential tremor) in 5,278 elderly individuals.^[54,55,113] The whole study population was invited to complete a brief psychometric examination,^[34] the results of which are summarized in Table 1 . The study identified 232 individuals who were judged to have essential tremor, and analysis of their psychometric tests revealed that this group of individuals had mostly mild deficits in attention, executive function, information-processing speed, immediate and delayed memory and verbal fluency when compared with controls (n = 696).^[34] Furthermore, the cognitive performance of the essential tremor group (as assessed on the 37-point MMSE) was very mildly decreased compared with the control group,^[34] and depressive symptoms were increased in the former group.^[70,71] Mild cognitive impairment was most prominent in patients with senile essential tremor.^[114]

In this population-based study, cognition was measured at baseline and 3 years later. During this time period, cognitive decline was significantly greater in patients with essential tremor than in controls (Figure 1).^[65] This finding is in agreement with the observation that in this cohort, elderly patients with essential tremor had an increased risk of dementia.^[36,115] An association between essential tremor and dementia has also been demonstrated in a population-based cohort study conducted by Thawani et al. in New York, USA.^[37] The main cause of dementia in these two population-based surveys was AD.^[37,115] The relationship between essential tremor and AD is not yet fully understood, although some possible mechanisms have been put forward.^[100,101] The NEDICES cohort and other studies^[35,74,116] indicate that the functional incapacity experienced by patients with essential tremor is related more to cognitive performance than to tremor; a similar finding was made in a study of patients with essential tremor in nursing homes.^[74]

Figure 1.

Enlarge

37-MMSE Scores in Patients With Essential Tremor and Controls in the Neurological Disorders in Central Spain (NEDICES) Cohort.⁶⁵ Baseline and follow-up 37-MMSE scores for controls are shown on the left; baseline and follow-up 37-MMSE scores for patients with essential tremor are shown on the right. At baseline, the mean 37-MMSE score in patients with essential tremor was 28.8 ± 5.8 points compared with 30.2 ± 4.8 points for controls (Mann-Whitney, P = 0.02). After the 3 year follow-up period, the 37-MMSE score declined by 0.70 ± 3.21 points in essential tremor cases compared with 0.11 ± 3.81 points in controls (Mann-Whitney, P = 0.03).⁶⁵ Abbreviation: 37-MMSE, 37-point Mini-Mental State Examination.

In summary, the clinical studies and the population-based NEDICES study reviewed above have produced concordant findings. These studies have all shown that deficits in attention, verbal fluency, memory (immediate and delayed) and possibly other subtle psychological deficits such as visuoperceptual deficits are common in patients with essential tremor. These patients are also more likely to have mild global cognitive impairment or depression than are healthy controls ( Box 3 ). The cognitive deficits and mood disorders that patients with essential tremor can experience might represent a sub-clinical form of CCAS. An alternative explanation is that essential tremor might be caused by a dynamic oscillatory disturbance of the motor system and that this disturbance might have a detrimental effect on the nervous system and underlie the cognitive deficits associated with this condition;^[12] however, this hypothesis seems speculative. Clinical, pathological^[44] and epidemiological data,^[34,36,37] as well as results from studies of thalamic DBS,^[108] in patients with essential tremor, indicate that this condition is not a simple monosymptomatic tremor disorder as was once thought.