Assessing the Relationship Between ET and Dementia

Specific Aims:

Essential tremor (ET) is a common neurological condition which increases with age and contributes to significant disability by interfering with activities of daily living (ADLs) in most affected patients. Dementia is another common neurological condition of varying etiologies – most commonly Alzheimer’s disease, vascular dementia and Lewy body disorders to name a few. A population study of ET demonstrated that more mildly affected, largely untreated ET individuals may be more likely to complain of memory problems and have deficits at testing (Benito-Leon J et al, 2011). In the past, our group has compared the risk of developing dementia in subjects with ET versus controls without tremor in a large, well-categorized cohort of individuals involved in a longitudinal aging study, the Arizona Study of Aging and Neurodegenerative Disorders (AZSAND) (Shill HA et al, 2014). Based on the size of our cohort at the time, we concluded that ET was not associated with an increased risk of dementia but that a subset of subjects with older age onset/shorter duration tremor may be at higher risk. We would like to better assess and further study this more conclusively given the substantial increase in the size of our ET cohort, thanks to funding from the IETF, over the years as well as better characterize the tremors in patients with dementia related to both Lewy body and non-Lewy body disorders.

Aim 1: To evaluate the risk of developing dementia in patients with ET

Here we intend to compare the risk of developing dementia in subjects with ET versus controls without tremor in our large, well-categorized cohort of individuals involved in a longitudinal aging study (AZSAND). All subjects are followed annually with a comprehensive standardized motor and neuropsychological test battery to assess cognitive function as well as clinical evaluations for tremors by a trained movement disorders specialist.

Aim 2: To determine the prevalence and characteristics of tremors in patients with dementia related Lewy body disorders vs. non-Lewy body disorders

Relatively little is known regarding the prevalence and characteristics of tremors in patients with dementia. We will assess the overall presence of postural and action tremors in patients with dementia and further compare the tremors in dementia patients with Lewy body disorders (Parkinson’s disease, dementia with Lewy bodies and multiple system atrophy) to non-Lewy body disorders (Alzheimer’s disease, vascular dementia and progressive supranuclear palsy).

Aim 1 Analysis

Figure 1. Participant Flow Chart

Summary:

There were 545 controls and 105 prevalent ET and 104 Incident ET patients included for Aim 1 analysis. Prevalent ET subjects entered the study with a diagnosis of ET at baseline visit and Incident ET subjects developed ET during follow-up in the study.

The longitudinal cognitive testing data were extracted to get the follow up cognitive diagnosis for the analysis population. Everyone included in the analysis population had to have follow up cognitive diagnosis to enter the study. Therefore, there are no competing risk events before cognitive diagnosis (i.e, death preventing dementia happening).  Due to the sparse follow up after 10 years, everyone’s follow up date were censored at follow up 10 years. Reverse Kaplan-Meier curve for the developing of dementia were generated and log-rank test were used to test if the developing dementia distribution were different between Prevalent ET, Incident ET and controls (Table 1). Further, multivariable Cox regression adjusting for age at baseline cognitive test or ET diagnosis (for Incident ET), gender, cognitive status and ET symptom onset duration were used to see if there is any group difference in terms of developing dementia after adjusting for potential confounders (Table 2).

The results show that in the multivariable analysis, both ET groups (Incident and Prevalent ET) didn’t have significant difference in terms of developing dementia compared to controls.

Table 2.  Multivariable Cox regression for developing dementia

Aim 2 Analysis

In the current frozen database (as of 07/29/21), there are 924 subjects had final clinicopathologic diagnosis. Out of them, 521 had final dementia diagnosis and the further 497 had final movement diagnosis (so their tremor can be defined by postural and kinetic tremor) were included in the Analysis Aim 2.

Table 3 shows the autopsy demographics and the final clinicopathologic diagnosis comparison between dementia without lewy body and dementia with lewy body groups. We can see that dementia without lewy body group had significant older age (87.3 vs. 83.7 years old, p<.001), more females (51.7 % vs. 34.6%, p<.001) compared to dementia with lewy body group.

In the univariate analysis, we can see that the dementia without lewy body group has similar tremor prevalence (defined by postural tremor or kinetic tremor, 40.0% vs. 34.6%, p=0.218) compared to dementia with lewy body group.

Multivariable logistic regression adjusted for death age and gender also showed that there is no difference between dementia groups (OR for dementia with lewy body 0.80, p=0.260) in terms of having a tremor (postural or kinetic tremor) at the last movement exam (Table 4).