Essential tremor course and disability

Materials and methods.

Clinical diagnosis of ET was made when the patient had upper limb (UL) postural and/or kinetic tremor or head tremor for which no other neurologic or systemic cause could be identified.3,22-24⇓⇓⇓ Those patients who at initial assessment had isolated RT or RT and rigidity or bradykinesia were excluded, as were patients with prominent cerebellar dysfunction. Also excluded were patients who were receiving drugs that may cause postural/kinetic tremor, for example, lithium, and those with metabolic disorders, such as hyperthyroidism. Throughout the course, all patients were evaluated by the same neurologist at the MDCS. Site of tremor onset was based on history or early clinical observations. The spread of tremor to different anatomic sites was based on clinical follow-up. Tremor frequency and amplitude were evaluated visually. In some patients, handwriting sample, maze tracing, spiral drawings, and drinking from a cup were also assessed. Video recordings were made at some time on most but not all cases.

The severity of tremor was based on the visual assessment of tremor amplitude and the history of its impact on function and was recorded at each visit. To compare with previous status, additional objective assessments (handwriting, video, etc.) were made in some but not all cases. In patients assessed after 1987, Unified Parkinson’s Disease Rating Scale (UPDRS)25 for action or postural tremor was utilized. Tremor severity in the patients assessed prior to that was noted in the clinic records and was converted to UPDRS as follows: mild = 1, moderate = 2, marked = 3, severe = 4. For the purpose of this study, the anatomic site with the largest tremor amplitude ever observed was used as the tremor severity. Neurologic abnormalities detected during the follow-up were recorded prospectively and are included in this report. When tremor was evident in a limb fully supported against gravity with the patient resting in supine position, it was classified as RT. Additional diagnosis of Parkinson syndrome (PS) was made only when there was unequivocal evidence of bradykinesia, rigidity, and RT.16 In general, the ET patients were evaluated at the MDCS at annual intervals. Clinical information obtained between clinic visits by telephone or otherwise from the patient/family/physician, etc., was noted in the medical records and is included in this report. Response to drug therapy was classified primarily on the basis of patient/family observations, but in some patients, sequential objective assessments and video recordings were made. Presence of physical disability or psychological distress was based on the history provided by patients/family and on clinical assessment and was documented in the patient records prospectively.

Autopsy was performed after obtaining a written consent from the next of kin. Brain histologic examination including substantia nigra (SN) and cerebellum was performed by a qualified neuropathologist in every case. When autopsy was done within 24 hours of death, one-half of the brain was examined histologically and the other half was frozen for future studies.

Results.

All 20 ET patients (10 men and 10 women) clinically diagnosed with ET at the first MDCS visit who came to autopsy during 32 years (1970 to 2001) were included in the study.

Table E-1 in the supplementary material on the Neurology Web site (go to www.neurology.org) summarizes the major clinical features and autopsy findings. Onset of ET ranged from childhood to 68 years (median 46.5 years). The most common site of tremor onset was only UL in 15 of 20 (75%) cases. Four (20%) others had simultaneous cranial (head, lips, jaw, or vocal cord) and UL onset, and one (5%) had onset as head tremor alone.

Of the 15 UL-only onset cases, 10 (67%) subsequently developed cranial tremor. The most common sites of ET at final clinical assessment in the 14 cases (excluding 6 with clinical diagnosis of PS) was UL and cranial in 10 (71%) cases, including 2 cases with UL, cranial, and lower limb (LL) tremor. Of the four patients who had UL and cranial tremor onset, three (75%) had extension of tremor to LL; two of these also had PS with unequivocal bradykinesia, rigidity, and RT. Maximum tremor severity noted was UPDRS score of 2 in 9 of 14 (64%) cases without PS, and 1 patient had UPDRS 4 severity.25

RT without bradykinesia or rigidity was seen in 6 of 20 (30%) cases. None of those had SN or other basal ganglia pathology. Six (30%) other patients developed parkinsonian (PS) motor features: RT, bradykinesia, and rigidity. Two of those (Cases 18 and 19) at autopsy had progressive supranuclear palsy (PSP) and one (Case 5) had Lewy body (PD) pathology. One PS patient had basal ganglia status cribrosus (Case 2), and in two cases, parkinsonian features were attributed to neuroleptic usage (Cases 13 and 15).

The most commonly used drug for tremor was propranolol in 10 (50%) cases. Six (60%) of those reported symptomatic benefit. The remaining four either did not tolerate the drug or had no benefit. Levodopa was used in nine (45%) cases, three of whom reported symptomatic benefit. These three included one case with PD (Case 5) and two cases (Cases 18 and 19) with PSP. The third most commonly used drug was primidone in six (30%) cases. Some benefit was reported by half (3/6) of the cases. No patient with ET + RT or ET + PS had complete resolution of RT on drug therapy.

Six of the 14 (43%) with ET and ET + RT (but no rigidity and bradykinesia) reported psychological distress or had a psychiatric diagnosis, including 1 case with alcoholism. One of these cases (Case 8) had only head tremor. Physical disability related to tremor was reported by 10 (71%) of these 14 cases. In 3 of the 14 (21%), the maximum recorded tremor severity was UPDRS 1.25 Of those three, only one (33%) reported physical disability. Of the nine cases with tremor score of 2,25 seven (78%) reported physical disability. Tremor score was 3 in one and 4 in another case; both had physical disability due to tremor.

There was no systemic illness pattern that was unexpected in the elderly.

Neuropathologic findings of PSP were seen in two and PD26-28⇓⇓ in one case. All three had clinical evidence of PS. Aside from those three, no patient revealed SN pathology on routine neuropathologic examination. Cerebral arteriosclerosis with ischemic changes was noted in 4 of 20 (20%) cases. Mild cerebellar Purkinje cell loss was noted in two; one of those also had ischemic cerebral stroke, and the other had diabetes mellitus and coronary artery disease. We did not perform detailed cell count on any brain area.

Discussion.

The most common site of ET onset in our patients was the UL alone or cranial and UL in 19 of 20 (95%) cases. This figure is comparable with that reported by others.4 Isolated leg tremor is regarded as an indication against ET diagnosis,4 and none of our patients had leg tremor as the initial manifestation. RT is a common manifestation of PD at onset29 and during the course of illness,30 but its significance in ET remains to be established.18 Fluorodopa PET study was performed in 31 tremor patients.17 Eleven of those had predominant RT. Four of these 11 had onset as resting leg tremor, and in 2, RT was restricted to LL at last assessment; both had reduced ipsilateral arm swing, thus suggesting a diagnosis of PS. All 11 of the predominant RT cases had PET abnormality consistent with PS.17 The data provided indicate that 7 of these 11 patients had other clinical features of PS.17 Several other patients in that study were reported to have leg tremor in conjunction with tremor elsewhere, but the site of tremor onset was not identified.17 It was concluded that patients with onset as leg tremor had reduced striatal fluorodopa uptake comparable with PD.17 Five of our patients developed LL RT later in the course of illness. Three patients also had bradykinesia and rigidity (PS). The remaining two (Cases 17 and 20) had leg RT but no rigidity or bradykinesia. Neither of those two had SN pathology. Comparison of our pathologic observations to PET observations17 in LL RT onset cases is not possible as no LL-onset case was diagnosed as ET in our clinic. In a separate clinicopathologic study,29 we observed that all leg RT-onset cases had PD.

With time, ET may spread to other sites,31 but the pattern of evolution has not been documented in longitudinally followed cases. Our study confirms that extension of tremor to other anatomic sites is common in this disorder. Six (30%) of our patients had additional clinical features of PS, which could confound the ET extension pattern. Therefore, we did not include those six cases in the evolution of anatomic site of ET. Of the remaining 14 ET or ET + RT-only cases, 11 (79%) had only UL onset. Of those 11 cases, 8 (73%) had subsequent cranial extension of tremor. At final assessment, 9 of 14 (64%) in this group had both the UL and the cranial-only tremor.

RT, which is characteristic of PS, has also been reported in some ET cases during the course of illness.12,17,31,32⇓⇓⇓ The significance of this evolution remains controversial.12,16,33⇓⇓ Most reports on this topic are based on clinical assessment alone. Two autopsy studies found no pathologic evidence of PD16,34⇓ in ET + RT cases. Twenty predominant postural tremor (ET) cases were studied with ¹⁸F-dopa PET scanning.17 Six (30%) of those had additional RT (but not bradykinesia and rigidity). None of these six manifested PET scan abnormality. They concluded that low-amplitude RT was not indicative of nigrostriatal system dysfunction, but the predominant RT was indicative of nigrostriatal dysfunction.17 As the tremor amplitude and frequency35 evolve with time and PET methodology has certain sensitivity thresholds, these studies do not provide the level of evidence comparable to autopsy findings. Based on the available literature and our observations, we conclude that a proportion of ET patients develop RT in the UL, LL, or both as natural progression of this disorder.

The relationship between ET and PD remains controversial. Some observers have reported significantly increased risk of PD12,13⇓ in ET cases, whereas others could not substantiate that.2,15,16,36⇓⇓⇓ PS consists of a heterogeneous group of disorders, and definitive diagnosis of PD26,27,37⇓⇓ is based on histologic examination. Even after prolonged follow-up, errors in identifying the underlying PS pathology are well known.27,37⇓ Neither of our two PSP cases manifested ophthalmoplegia,38 and both improved on levodopa. Pathologic studies revealed that only one of the patients with PS had Lewy body disease. Had we not performed autopsy studies, we could have erroneously assumed that all six (30%) of the ET cases clinically diagnosed as PS had PD pathology.

Autopsy studies indicate that 3.8% of individuals in the sixth decade and 12.8% in the ninth decade have incidental Lewy body inclusions.39,40⇓ Such cases are regarded as having preclinical PD39,41⇓ and may easily manifest parkinsonian features on neuroleptics.42 If the prevalence of PD in ET cases were 24 times greater than in the general population, as suggested in one study,12 we would have expected Lewy body inclusions in at least (3.8% × 24) 91% of our patients. However, only 1 of 20 (5%) of our ET patients had Lewy body pathology. Thus, the risk of PD in our ET cases was similar to that in the general population.

It has been suggested33 that ET patients may have slowly progressive PD, where the first manifestation is postural tremor. Our data do not support that hypothesis. Even after several decades of onset, the ET and ET + RT cases had no pathologic evidence of PD.

One of our PSP cases survived 29 years and the other nearly 80 years after the tremor onset. Given the duration of the symptoms, it is highly unlikely that tremor in these two cases was consequent to PSP pathology.38 We are not aware of any report indicating higher risk of PSP in the ET. We conclude that the association between ET and PSP and ET and PD each was incidental in our patients.

Functional disability in ET is now well recognized8 and is related to the tremor site and amplitude. Our study confirms the observations in cross-sectional studies indicating that physical disability increases with the duration of illness.8 The most common problem is the UL functional impairment. Some community and chronic care ET patients with identifiable tremor-related disability may not spontaneously report functional disability,8,43,44⇓⇓ indicating a need for careful assessment of disability in these patients. Specialty movement disorder clinics, which are most likely to pursue detailed assessment of disability, have limitations as only a small proportion of ET cases attend such clinics.8 A recent report indicates that disability among the specialty clinic cases is more pronounced than in the community-based ET patients.8

There is no drug that is specific for ET. The most commonly used drugs are propranolol and primidone.18,32⇓ In our patients, propranolol was used in 10 cases, with 60% reporting some benefit, and primidone was prescribed to 6, with 50% reporting tremor improvement. Levodopa was used in nine cases. Table E-1 in the supplementary material on the Neurology Web site shows that all except one levodopa-treated patient had RT. Only those three cases that had clinical features of PS reported some improvement on levodopa.

Anxiety and depression are known causes of distress in ET.8 Eleven (55%) of our 20 cases had one of the following diagnoses: depression, alcoholism, obsessive–compulsive personality, or social embarrassment and anxiety. The cause-and-effect relationship was not systematically explored. The ET and psychological manifestations may have common pathophysiology, the tremor may lead to psychiatric manifestations, or it may represent incidental comorbidity.

Comorbid disorders may contribute to physical disability in ET patients. Six (30%) of our cases had motor manifestations of PS: RT, bradykinesia, and rigidity. We excluded those six patients to ascertain the physical disability attributable to ET. Of the 14 ET or ET + RT only, 10 (71%) reported physical disability in activities of daily living, which could be attributed to ET. As reported in another large ET study,8 the major disability in our cases was related to UL function: drinking, feeding, writing, etc.

ET remains a clinical diagnosis as there is no electrophysiologic or biologic marker. There is ongoing debate whether both postural and kinetic tremor are essential for the diagnosis of ET.7,45⇓ With time, both postural and kinetic UL tremor was evident in most of our cases.

We did not observe any other illness pattern that would be unexpected in the elderly.

Observations based on ET patients and in animal models of ET suggest dysfunction of the cerebellum or its afferent or efferent connections.4,20,21,46,47⇓⇓⇓⇓ The tremor in ET has some similarity to intention tremor seen in cerebellar disorders,48 and some ET patients have mild ataxia.20,31,49⇓⇓ However, unlike other common cerebellar disorders, most ET patients do not manifest clinical features of cerebellar dysfunction such as ataxia and dysmetria.4 PET studies21,50⇓ show excessive increase in cerebellar blood flow with activity in the ET cases. One ET patient was reported having improvement of ipsilateral tremor following cerebellar infarction.46 These observations all point to dysfunction of the cerebellum or its afferent or efferent connections. Routine histologic studies in our patients revealed some cerebellar pathology in two cases; each had mild Purkinje cell loss. One (Case 4) was a 95-year-old woman who had a cerebral infarct. The second (Case 11) had coronary artery disease and diabetes mellitus and died at age 80. In the absence of autopsy brain studies on age-matched non-ET subjects with comparable agonal status, we cannot comment on the significance of our cerebellar findings. Two patients (Cases 2 and 9, parent and child) each had mild ataxia on clinical assessment. On routine histology, neither revealed cerebellar pathology.

As yet, there is no reliable animal model of ET.20 More focused morphometric studies of ET and control brains are therefore needed to determine the role of the cerebellum in the pathophysiology of this disorder.

Most experts regard ET as a genetic disorder, but some environmental factor(s) may play a role.51,52⇓ We did not collect detailed family history of tremor in all cases. In another unrelated study of families that had an ET proband and a secondary case with any type of movement disorder (each evaluated at the MDCS), we found that 91% of the secondary cases also had ET.53

We recognize that this study includes selected ET cases. Our health care system and the MDCS policies provide equal access to all residents. There is no cost to ET patients to attend the MDCS, as all provincial residents are covered by the Provincial Health Plan (without premiums). Brain autopsy studies are routinely offered to all movement disorder patients seen at the MDCS, and there is no cost to the family for transport of the body or the autopsy. The postmortem study can only be performed after a written consent from the next of kin. Typically, autopsy study is initiated by the family, as most patients die away from the MDCS. Reasons for the family interest and consent for autopsy are complex and variable and could constitute a significant source of bias. Referral of complicated cases to the MDCS is another source of bias. As such, our observations cannot be regarded as representative of ET cases in the general population.