Association between positive history of essential tremor and disease progression in patients with Parkinson's disease

This study aimed to explore the effect of pre-existing essential tremor (ET) history on the disease progression of Parkinson’s disease (PD). We recruited and followed-up a group of PD patients from March 2009 to July 2020. The ET history of each patient was obtained by retrospective interviews or past medical records. Cox proportional hazards models with inverse probability of treatment weighting (IPTW) were used to estimate the hazard ratio (HR) with 95% confidence intervals (CIs). Of 785 patients who completed the followed-up visits, 61 patients (7.8%) reported a history of pre-existing ET. Cox regression models after IPTW indicated that the positive ET history in patients with PD was protective against time to United PD Rating Scale III 14-point increase (HR = 0.301, 95% CI = 0.134–0.678, P = 0.004), time to akinesia and rigidity 8-point increase (HR = 0.417, 95% CI = 0.218–0.796, P = 0.008), time to conversion to Hoehn and Yahr stage 3 (HR = 0.356, 95% CI = 0.131–0.969, P = 0.043), time to develop dyskinesia (HR = 0.160, 95% CI = 0.037–0.698, P = 0.015), and time to Montreal Cognitive Assessment 3-point decrease (HR = 0.389, 95% CI = 0.160–0.946, P = 0.037), but had no relationship with time to tremor 4-point increase (HR = 1.638, 95% CI = 0.822–3.266, P = 0.161) and time to death (HR = 0.713, 95% CI = 0.219–2.319, P = 0.574). Our study indicated that ET history in patients with PD is associated with a benign prognosis with slower motor and non-motor progression.


Scientific Reports
| (2020) 10:21749 | https://doi.org/10.1038/s41598-020-78794-1 www.nature.com/scientificreports/ To date, no data has systematically examined the impact of ET history on the progression of PD. Therefore, in the present study, we aimed to recruit and follow up a group of ET-PD patients and a group of PD patients without ET history with a duration of PD < 3 years at baseline to explore the positive ET history on the motor deterioration, onset of dyskinesia, cognitive decline, and survival of PD.
Comparison between patients with and without ET history before and after weighting. Before weighting, the standardized mean differences (SMD) values were greater than 0.1 in Body Mass Index (BMI), sex distribution, age, age of onset of PD, LEDD, levodopa use, dopamine agonist use, UPDRS part III score, motor subtypes, Hoehn and Yahr (H&Y) stage, Frontal Assessment Battery (FAB) score, Montreal Cognitive Assessment (MoCA) score, Hamilton Depression Rating Scale (HDRS) score, and Non-Motor Symptoms Scale (NMSS) score between patients with and without ET history (Table 1).
A total of 17 covariates at baseline including education, BMI, sex, age, age of onset of PD, disease duration of PD, LEDD, levodopa use, dopamine agonist use, BMI, motor subtypes, UPDRS III score, H&Y stage, MoCA score, FAB score, HDRS score, Hamilton Anxiety Rating Scale (HARS) score, and NMSS score were included for estimating the propensity score (PS). After weighting, the SMD values of each variable at baseline were reduced. Positive ET history in PD was protective against time to UPDRS III 14-point increase (unweighted P = 0.009; weighted P = 0.004), time to akinesia and rigidity 8-point increase (unweighted P = 0.014; weighted P = 0.008), time to conversion to H&Y stage 3 (unweighted P = 0.013; weighted P = 0.043), time to develop dyskinesia (unweighted P = 0.008; weighted P = 0.015), and time to MoCA 3-point decrease (unweighted P = 0.047; weighted P = 0.037), but had no relationship with time to tremor 4-point increase (unweighted P = 0.166; weighted P = 0.161), and time to death (unweighted P = 0.069; weighted P = 0.574) ( Table 2).

Discussion
In a sample of patients with early-stage PD, we firstly assessed the association between ET history and disease progression and found that ET history was a protective factor for motor and non-motor progression, especially for akinesia and rigidity deterioration, cognitive decline, and dyskinesia development. However, we found that the presence of ET had no relationship with tremor deterioration and mortality.
Heterogeneity of motor manifestations of PD patients suggested that PD had distinct clinical patterns implying distinct anatomic, biochemical, and pathologic changes 11 . In patients with PD, tremor was often less responsive to dopamine replacement therapy than rigidity and bradykinesia, and had a quite variable response to levodopa 12 . Therefore, tremor was regarded as the motor manifestation which is mostly independent of the other motor signs. Both rest and action tremors can frequently occur in PD, which were associated with multifaceted phenomenology and, possibly, pathophysiology 13 . However, the exact origin of tremor in PD remains unclear. The deficit in the basal ganglia-thalamocortical circuit is one potential tremor generator 14 .
The role of tremor in PD has been disclosed. In patients with PD, tremor is often combined with a family history of parkinsonism, early age of onset, and slower progression 15 . Patients with tremor at onset also had a slower disease progression than those without 7,16 . Furthermore, patients with TD phenotype were not only associated with more severe rest tremor but also had more severe action tremor, suggesting that both rest and action tremor can contribute to the classification of motor phenotype. Therefore, ET-PD patients would be more likely to belong to the TD phenotype of PD 7 . Previous analyses reported that patients with TD phenotype usually had a more benign course compared to those with PIGD phenotype 7 .These discoveries supported our finding that PD patients with a positive history of ET were associated with a slower motor and non-motor progression of the disease. Furthermore, our findings suggested that the role of ET should be considered when defining the PD subtype.
The variable rate of motor sign progression between patients with and without ET history suggested different pathological, genetic, and biochemical mechanisms as well as possible distinct causes for phenotypically different disorders. A pathological link between ET and PD has been established by the finding of Lewy body in a proportion of ET cases 2 . In addition, a 123-I ioflupane SPECT study found evidence of minimal dopaminergic deficits in the caudate nucleus of ET patients 17 , which suggested that some ET patients had a subtle dopaminergic deficit. However, it was still unknown whether ET-PD patients were associated with a slower progression of Lewy body aggregation than PD patients without ET history, which required to be verified by further longitudinal pathologic studies. Moreover, it was also unclear whether genetic factors played a potential role in the disease progression of PD. For example, both Leucine-rich repeat and Ig domain containing 1 gene (LINGO1) and its www.nature.com/scientificreports/ paralog LINGO2 mutations have been reported to be associated with ET and PD 18 . The significance of hereditary factors on the motor progression of PD remained to be determined.
A new viewpoint suggested that ET-PD syndrome was a distinct clinical entity rather than a combination of the classical phenotype of ET and PD. There was evidence indicating that ET-PD patients had peculiar clinical, neurophysiological, and imaging features 19 . A previous study found that patients with ET-PD had extrapyramidal signs like PD patients with TD subtype, but with more symmetric rest tremor and lower severity of rigidity 19 . Furthermore, the electrophysiological findings in that study 19 indicated that rest tremor in patients with ET-PD had a synchronous pattern, as in ET with rest tremor, instead of the classical alternating pattern of the rest tremor in PD. If we regarded PD patients with a history of ET as a distinct clinical entity, our findings supported the benign course of this syndrome.
The association between ET history and a slower decline in cognition in our study was consistent with a previous study, which demonstrated that ET was associated with mild deficits in attention, executive functions, memory, and, possibly, other cognitive processes, and most of the clinical series and neurological tests regarded ET as a mildly progressive movement disorder characterized by monosymptomatic tremor 20 . However, the cardinal motor symptoms in PD including tremor, rigidity, bradykinesia, and PIGD often progressed at different rates and showed variable responsiveness to levodopa therapy, challenging the view that baseline motor subtype classification was a significant predictor of death and cognitive decline 21 . The DATATOP study found that there was no difference in performance on neuropsychological tests between patients with TD and PIGD subtypes, suggesting relative independence of motor phenotype and cognitive changes in PD 7 . In addition, a communitybased study indicated that patients with TD subtype at baseline did not become demented until they developed PIGD subtype, and dementia did not occur among patients with persistent TD subtype of Parkinsonism 8 . The same study also found that PD patients with PIGD subtype were associated with accelerated cognitive decline and highly increased risk for subsequent dementia, suggesting that PIGD rather than tremor shared common or paralleled neuropathology with dementia 8 .
Some limitations should be recognized. First, some patients (39/61) diagnosed as ET-PD was based on the self-reported history of ET, which might contribute to recall bias. Second, although the robust methods and statistical methods were used, some potential unmeasured confounders might still bias our results. Third, the prevalence of ET history in PD was relatively small and therefore a selection bias cannot be ruled out. Forth, some patients with ET history reported an indefinable onset age of PD, which might contribute to a controversial disease duration of PD. Fifth, the relatively short observation of disease progression of some patients was not sufficient to conclude the influence of ET history on the long-term outcomes.

Conclusions
Our study indicated that ET history in PD patients was associated with a benign prognosis with slower motor and non-motor progression and had no impact on survival.

Methods
Subjects. All procedures of the study were supported by the Ethics Committee of West China Hospital, Sichuan University (No. 2015236). Initially, 4486 PD patients were seen and registered in the Department of Neurology, West China Hospital of Sichuan University, between March 2009 and July 2019 (Fig. 1). All participants provided written informed consent and met the clinical diagnostic criteria for PD based on the Unified Kingdom PD Society Brain Bank 22 and also verified by the MDS clinical diagnostic criteria 23 . All the methods were carried out in accordance with the relevant guidelines and regulations.
To explore the influence of ET history on the disease prognosis, participants who met the following inclusion criteria were invited to finish at least one (range 1-10) face-to-face follow-up examination (n = 807): (1) H&Y stage < 3; (2) disease duration < 3 years; (3) assessment at "off " medication; (4) absence of motor complications including motor fluctuation and dyskinesia; (5) absence of dementia. The mean interview intervals for each adjacent visit was set for more than one year (Supplementary Table 1). During the follow-up visit, 22 patients withdraw informed consent, 31 lost contact and 94 died (Fig. 1). Among the 94 patients died, 81 had no face-toface follow-up data on the motor and cognitive assessments, which was not included as censors in the analysis on the motor and cognitive progression. This is because these patients who had died were probably caused by motor and cognitive progression, which may have a potential effect on the clinical outcomes. Finally, 785 patients who provided information on survival outcomes and 704 patients who had the motor and cognitive outcomes were incorporated into the data analysis.
Definition of ET-PD. The diagnosis of ET-PD was based on previously defined criteria 24 and required that: (1) the ET diagnosis was present for > 5 years before the PD diagnosis, (2) the initial ET was characterized by moderate or greater amplitude action tremor without any motor signs of PD, and (3) the initial ET diagnosis occurred in absence of any red flags for possible emerging PD. These subjects with a long-lasting history of ET who developed PD phenotype were classified as ET-PD patients. Past clinical history data on ET history were obtained from patients and their relatives or extracted by their medical records, when available.
Baseline assessments. At baseline, trained neurologists in our movement disorder center completed a standardized assessment for all patients. Demographic and clinical data including sex, age, height, weight, age of onset of PD, PD duration, years of schooling, and therapeutic schedule were collected. BMI was calculated as body weight (kg) divided by height squared (m 2 ). Each patient was classified as TD, Intermediate, or PIGD phenotype based on a previous method 25 . The LEDD for each subject was calculated based on their therapeutic schedule 26 .
Scientific Reports | (2020) 10:21749 | https://doi.org/10.1038/s41598-020-78794-1 www.nature.com/scientificreports/ The UPDRS part III 27 and H&Y stage (range 1-5) 28 were both applied to evaluate the severity of motor symptoms. If possible, all patients were asked to withdraw medications > 12 h at the follow-up visit (484/673, 71.9%). For those patients who did not provide an "off " score, we estimated an "off " medication score by adding the difference value of the study population's mean "off "-score and mean "on"-score to the patient's "on" medication score 29 .
The Chinese version of the NMSS 30 was used to assess the severity of the global NMS burden. In addition, cognitive function was evaluated using the MoCA (range 0-30) 31 and the FAB (range 0-18) 32 , with lower scores indicating poor cognition. The severity of depression was assessed by the HDRS (24 items) 33 , and the severity of anxiety was evaluated with the HARS 34 .

Definition of clinical outcomes
Death. Continuous mortality surveillance was performed mainly throughout the follow-up of patients and their relatives mainly using telephone visits. It lasted until July 1, 2020, which was approximately 11 years after our study began (2009), with as many as 14 years of follow-up for mortality after the patients first diagnosed in Motor decline. It is reported that a change of 2.5-5.2 points on the UPDRS III score is a clinically significant difference 35 , so we defined a fast motor progression as a 14-point increased in the UPDRS-III score (mean of 4-point per year) based on the mean follow-up period from the baseline to the last face-to-face visit or censoring (3.6 ± 2.2 years). Time to the event was defined as the time from the baseline to follow-up visit in which a 14-point increase was first reached. To separately explore the influence of ET history on the tremor and akinesia and rigidity progression of PD, time to tremor 4-point increase and time to akinesia and rigidity 8-point increase were set as another clinical outcomes. The increased score for the tremor or akinesia and rigidity was determined based on the proportion of each symptom in the UPDRS III multiplied by 14. Time to conversion to H&Y stage 3 was set as an additional motor deterioration event, which was defined as the time from the onset of PD to first follow up examinations in which the patient firstly reached a score of H&Y stage 3. Statistical analyses. Frequencies and descriptive statistics were utilized to summarize the baseline data for the unweighted and weighted samples. Baseline data were reported percentages for categorical variables, and both mean ± standard deviation (SD) and median (quartiles) for continuous variables. The Shapiro-Wilk test was used to check the normality for each variable (Supplementary Table 2).

Dyskinesia.
To balance the differences in various baseline variables between patients with and without ET history, a PS weighting method was selected. The PS model was constructed by conducting a multivariable logistic regression model in which patients with and without ET history was regressed on all baseline variables possibly related to the clinical outcomes, including education, BMI, sex, age, age of onset of PD, disease duration of PD, LEDD, levodopa use, dopamine agonist use, BMI, motor subtypes, UPDRS-III score, H&Y stage, MoCA score, FAB score, HDRS score, HARS score, and NMSS score. The estimated PS was made as to the predicted probability of presenting ET history in each subject. The inverse probability of treatment weighting (IPTW) 23 was then calculated as the inverse of the PS for the patients with ET history and as the inverse of (1 − PS) for the patients without ET history. This approach created a pseudo-population in which the exposure of ET was independent of measured confounders.
To rate bias reduction after the PS weighting, SMD was compared between patients with and without ET history before and after weighting, with a threshold of < 10% designated to indicate between-group balance. Cox proportional hazards models were used to estimate the HR and examine 95% CIs of the unweighting and weighting samples. The plotted Schoenfeld and time was used to test the proportional hypothesis, and P > 0.05 suggested the data met the assumption of equal proportional risk. To visually assess the study groups on the risk of developing a disease milestone, the Kaplan-Meier survival curves were plotted and the log-rank tests were applied .

Scientific Reports
| (2020) 10:21749 | https://doi.org/10.1038/s41598-020-78794-1 www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.