Essential tremor (ET) is perhaps the most common adult-onset movement disorder.1,2 It is a sporadic or familial disorder characterized by postural-action tremor, with a frequency of 5—8 Hz. ET predominantly involves the hands but may also spread to the head, the vocal cords, the lips, and, rarely, the lower extremities.3,4 Clinically, ET progresses slowly. In some cases in the advanced stage, daily living activities, such as eating and writing, may become impaired.5,6 Although postural-action tremor is the most common symptom of ET, other extrapyramidal symptoms (e.g., rigidity, bradykinesia, rest tremor, gait abnormalities) and cerebellar dysfunction (e.g., intentional tremor) have been reported in patients with ET.7—13
The pathophysiological basis of ET is still controversial.14 Neuropathological studies have not identified any evidence of morphological and biochemical abnormalities in the brains of ET patients.15 However, some clinical and brain metabolic imaging studies suggest that cerebello-thalamo-cortical loops may have a role in the pathophysiology of ET. For example, lesions of cerebellum, pons and thalamus have been reported to abolish or reduce ET.16—19 Positron emission tomography studies have demonstrated bilateral cerebellar hyperactivity in ET. Furthermore, magnetic resonance spectroscopy studies have revealed cerebellar neuronal damage or loss in patients with ET.20—24
Although ET is traditionally classified of as a pure motor system disease, in recent years, it was reported that the impairments in cognitive functions can be associated with ET.25—28 The results of these studies conducted among relatively elderly population have shown that a mild attentional impairment and frontal executive dysfunction are among the clinical features of ET.
Unlike the studies mentioned above, we examined cognitive functions in a newly diagnosed, nonmedicated, relatively young patient group with ET and comparison subjects using a comprehensive neuropsychological test battery in order to eliminate confounders.
Previous studies reported inverse correlation between the regional brain blood flow by single photon emission computed tomography (SPECT) and the poor performance on the neuropsychological tests.
Additionally, we investigated the regional cerebral blood flow (rCBF) of both groups by technetium-99m-HexaMethyl Propylene Amine Oxime-Single photon emission computed tomography (technetium-99m-HMPAO SPECT).
We assessed 16 patients (nine women and seven men) with ET, who met the diagnostic criteria of ET as proposed by Bain et al.,29 by a comprehensive neuropsychological test battery and compared them with comparison subjects (9 women and 7 men). The patient group was comprised of consecutive referrals to our movement disorders outpatient clinic who had never been diagnosed and treated for tremor before. The patients who were on specific medication for ET were not included in this study. Fifteen of the patients had bilateral tremor in the upper extremities and one had only head tremor. In four of the 15 patients who had bilateral upper extremity tremor, there was also both head tremor and vocal tremor, only head in one, vocal in another. Eight of the patients had positive family history for ET. The other possible causes of tremor and reversible causes of cognitive dysfunction were ruled out by normal serum thyroid hormone, vitamin B12 levels and negative VDRL. The patients were not taking alcohol, except one whose tremor was responsive to alcohol. None of the subjects had depression or other psychiatric, and systemic illness or head trauma history. Psychiatric comorbidity was excluded with the clinical interview and Hamilton Depression Rating Scale (HDRS)30 was applied specifically to quantify the probable depressive symptomatology in both the patient and the comparison groups. Patients and comparison subjects were matched for age, gender, education and handedness. For patients with ET, the severity of tremor was quantified using the Clinical Rating Scale for Tremor (CRST).31 Before the study, informed consent was obtained from patients and comparison subjects.
A single-headed gamma camera (GE, Starcam 4000i, U.S.) fitted with a low energy high resolution collimator was used. Sixty-four 30-second frames were collected during a 360o rotation in a 64x64 matrix. Acquisition began 15 to 20 minutes following the injection of 740 MBq technetium-99m HMPAO (Ceretec, Amersham, UK) in a quiet room and eyes closed.
Image reconstruction was performed using ramp filtered back projection with a Butterworth filter without attenuation correction. Transverse slices, reoriented parallel to the orbito-meatal plane were obtained using anatomically-defined 47 cortical regions of interest in a total of eight levels (F1). The slice that displayed the occipital cortex in Level 3 without any cerebellar activity was chosen as reference slice. Two regions of interest (ROIs) were placed on this reference slice. Same slices were used across all patients. Other ROIs were sampled from all cortical and cerebellar regions as indicated in F1. For each ROI, a cortex/occipital ratio was derived by dividing mean counts in the cortical ROI by the mean counts in the reference ipsilateral occipital region. The ratio of these counts was used as a measure of rCBF.
The neuropsychological test battery was designed to assess different domains of cognitive functions; global attention, language, memory, visuospatial functions, and executive functions. All tests are presented in t1.
The age, education, duration of disease, gender, handedness, means for SPECT measures and neuropsychological test parameters of patients with ET and comparison subjects were compared using student-t test. The Pearson correlation coefficient was used to determine the correlation between CRST scores and the results of neuropsychological tests, and SPECT measures.
The demographic features of two groups and the CRST score and disease duration of the patients with ET are presented in t2. Fifteen of the patients were right-handed, and one was left handed whereas 13 of the comparison subjects were right-handed, and three were left-handed. In all patients, except one patient who had head tremor, tremor involved the dominant-hand. The mean and SD of the HDRS scores in the patient and comparison groups were 2.93 (SD = 1.48) (range;2—8) and 4.06 (SD =1.8) (range;1—5), respectively. There were no statistically significant difference between age, gender, education levels, handedness of the patients with ET and comparison subjects. The patient who has head tremor was among the patients without the SPECT evaluation.
In 11 of 16 patients and in nine of 16 comparison subjects, the rCBF could be measured by technetium-99m-HMPAO SPECT. All missing subjects could not be investigated before the study was completed due to a temporary technical problem of the gamma camera system.
The means and standard deviations of all calculated rCBF index values are presented in t3. There was no statistically significant difference between SPECT measures of patients and comparison subjects. However, in SPECT evaluation of 11 patients with ET, there was a statistically significant inverse correlation between bilateral frontal blood flow and CRST scores (F2).
Bilateral frontal blood flow in Levels 3 and 4 and the right frontal blood flow in Level 5 decreased while CRST scores increased. No correlation was found between cerebellar, parietal, temporal lobe blood flow and CRST scores as well as the blood flow of frontal lobe in Levels 7 and 8.
Neuropsychological Test Results
Test results are summarized in t4. In the California Verbal Learning Test (CVLT), patients scored significantly worse on the fifth learning trial, free recall intrusions, delayed recognition and "discriminability." Patients obtained significantly lower scores on Benton Facial Recognition (BFR), Benton Line Orientation (BLO) tests and made significantly more perseverations in verbal fluency tasks. The differences in other tests scores did not reach statistical significance. Similarly, there was no significant difference between the SPECT measures of patients with ET and comparison subjects.
However, there were statistically significant correlations between CRST scores and some Wisconsin Card Sorting Test (WCST) measures, namely, the total number of responses, completed categories, perseverative responses and errors, and CRST scores and Hooper Visual Organization Test (HVOT) scores (F4).
Recently, it has been reported that the cognitive functions subserved by prefrontal cortex and frontocerebellar network were impaired in patients with ET.
Gasparini et al.25 found that the group with ET and a family history of ET showed poorer performences on frontal lobe tasks involving conceptualization and set-shifting, than those of Parkinson’s disease (PD) and comparison subjects. Lombardi et al.26 reported that impairments in working memory-attention, verbal fluency, visual confrontation naming and verbal learning and memory. Also, they demonstrated that the levels of depression of patients with ET were higher. Lacritz et al.27 found mild cognitive impairment characterized with executive dysfunction in the subject with severe ET.
In the present study, we found significant differences between performances of the patients with ET and comparison subjects in neuropsychological tests assessing visuospatial functions and some measures of verbal memory. There were no significant differences in other tests assessing attention, visual memory, language and executive functions, except the the number of perseverations in verbal fluency, between two groups. However, we found that CRST scores of patients with ET were correlated with poorer performance on some WCST measures and HVOT.
It has been shown by clinical and experimental studies that WCST is a sensitive test of the frontal network functions such as working memory, set shifting, mental flexibility, problem solving, and conceptualization.36,37 In our study, the total number of responses, completed categories, perseverative responses and errors on WCST in ET group were inversely correlated with tremor severity. All the studies that investigated neuropsychological functions in patients with ET demonstrated that the performance on WCST of the patients with ET is poorer than normal group or normative data. Our results are similar to those of the other studies.
According to our results, the impairments in visuospatial functions were revealed by tests requiring visual processing such as BLO, BFR, HVOT, but not requiring manipulation and construction such as block design (BD) and Hanoi tower. These findings may suggest that the poor performance in patients with ET arises from pure visuoperceptual deficits but not from a motor problem caused by tremor. Tröster et al.28 showed that visuoperceptual abilities as measured by BFR and HVOT were significantly less than average. Lacritz et al.27 found that the constructional abilities of patients with ET are within normal limits by BD. The posterior parietal cortex receives afferent connections from cerebellum via pons and thalamus. Visuospatial dysfunction following cerebellar damage may reflect the involvement of cerebello-ponto-thalamo-parietal pathways.38,39 It has been known that the cerebellum plays a role in pathophysiology of ET.4,8,9 Visuospatial disturbances in patients with ET may be due to the involvement of afferent pathways from cerebellum to posterior parietal lobe.
Although the performance of the patient group was poorer in almost all the measures of the CVLT than that of the comparison group, the differences between groups reached statistical significance on fifth learning trial, free recall intrusions, recognition hits and discriminability.
In contrast to Gasparani et al., Lombardi et al., and Lacritz et al., we did not find significant differences between ET and comparison groups in the verbal fluency.25,26,27 However, we observed significantly higher number of perseverations on the verbal fluency in ET group. This finding is also in accordance with the hypothesis of an executive dysfunction.
Furthermore, bilateral frontal hypoperfusion shown by SPECT scanning in the patient group correlates with both CRST scores and some of the neuropsychological test scores that are sensitive to frontal lobe functions (i.e., WCST). Higher CRST scores were associated with lower bilateral frontal blood flow. This result supports the frontal dysfunction in the patient with ET.
The results of our study, which was conducted in a relatively young sample of patients with ET, confirmed the results of the aforementioned studies that suggested a frontal network dysfunction.
The most important difference of the present study from the other three studies was the age of patients included to the study. The mean age of our cases was 29.6 years (range=18—52), whereas the mean ages of the cases in the studies of Gasparini et al. , Lombardi et al. , and Lacritz et al. was 68.8 years, 66.1(range=36—83), and 70 years (range=43—86), respectively.25,26,27 Age effects on the frontal lobe functions are well known. Normal aging affects first and foremost the frontal lobes.40,41 Thus, the confirmation of a frontal network involvement in a young patient group with ET may clarify the probable confounding age effect in the previous studies. Second, our group was comprised of nonmedicated patients that might be considered an additional advantage of the present study in terms of the confounders. Yet, relatively small sample size, especially in the SPECT component of the study could be seen as the weakness of our study. This might explain why the comparison of the rCBF’s did not reveal any differences between the two groups despite the significant inverse correlation between the ET severity and frontal rCBF in the patient group.
The present study shows that subclinical cognitive deficits, including frontal executive and visuospatial functions and verbal memory impairments, may be classified as clinical features of ET. These results support that the dysfunction of fronto-cerebellar circuits may have a role in ET pathophysiology.
The authors thank Dr. Hakan Gurvit for his contributions and Dr. Yuksel Bek and Dr. Mustafa Ertas for statistical analysis.