Abstract
Background:
Mild parkinsonian signs (MPS) are common in the older adult and associated with a wide range of adverse health outcomes. There is limited data on the prevalence of MPS and its significance.
Objective:
To determine the prevalence of MPS in the community ambulant population and to evaluate the relationship of MPS with prodromal features of Parkinson’s disease (PD) and cognition.
Methods:
This cross-sectional community-based study involved participants aged ≥50 years. Parkinsonian signs were assessed using the modified Unified Parkinson’s Disease Rating Scale (mUPDRS) and cognition using the Montreal Cognitive Assessment (MoCA). Premotor symptoms of PD were screened using a self-reported questionnaire. Linear regression was used to assess the association of MPS with premotor symptoms of PD and cognitive impairment.
Results:
Of 392 eligible participants, MPS was present in 105 (26.8%). Mean age of participants with MPS was 68.8±6.9 years and without MPS was 66.1±5.9 years (p < 0.001). Multivariate analysis revealed that MoCA scores were significantly lower in the MPS group (β= –0.152, 95% CI = –0.009, –0.138, p < 0.05). A significant correlation between the presence of REM sleep behavior disorder (RBD) and total MPS scores (β= 0.107, 95% CI = 0.053, 1.490, p < 0.05) was also found. Neither vascular risk factors nor other premotor symptoms were significantly associated with MPS.
Conclusion:
MPS is common and closely related to cognitive impairment and increasing age. Presence of RBD is predictive of higher MPS scores. This study highlights the necessity of other investigations or sensitive risk markers to identify subjects at future risk of PD.
INTRODUCTION
Mild parkinsonian signs (MPS), which include mild bradykinesia, rigidity, resting tremor, gait and balance disturbances, are common in the elderly population with a prevalence ranging from 15% to 52% [1–3]. The presence of MPS may represent the mild end of a disease spectrum that spans from normal aging to neurodegenerative disease, including Parkinson’s disease (PD) and Alzheimer’s disease (AD) [4]. It remains unclear whether individuals with MPS progress to a distinct neurological disease entity and if so, whether they develop primarily PD, AD, vascular parkinsonian syndromes or vascular dementia [4]. MPS has been shown to be associated with dementia [5–7], vascular risk factors or disease [8]and adverse health outcomes, such as, increased functional disability [1, 9] and mortality [3].
Prodromal PD refers to the stage wherein early symptoms or signs of neurodegeneration are present, but clinical diagnosis based on fully evolved motor features is not yet possible [10]. In prodromal PD, transient or subtle motor dysfunction, in parallel with non-motor features, may be present prior to diagnosis [11]. A few studies have reported the association between MPS and the presence of PD risk markers and substantia nigra hyperechogenicity [12, 13]. However, the clinical significance of MPS remains to be fully understood.
This study was designed to assess the prevalence of MPS in a community ambulant elderly population, and to evaluate for a possible association of MPS with cognition, prodromal markers of PD, cardiovascular risk factors and disease.
METHODS
The study sample consisted of subjects aged 50 years and above who participated in a community health screen conducted in May 2017 and May 2019 in Singapore. Participants were recruited using advertisements and completed a standardized questionnaire which included demographic information, cardiovascular (CVS) risk factors, coexisting medical illness, and medication use. Participants with history of stroke, dementia, intracranial malignancy, current neuroleptic medication use, significant orthopedic abnormalities that would limit mobility or the performance of the Movement Disorders Society – Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) and known and newly diagnosed PD or Parkinson plus syndromes were excluded from the study.
The presence of prodromal markers of PD was also assessed with a self-reported questionnaire which included questions derived from the MDS prodromal criteria pertaining to the presence or absence of constipation, impaired smell, excessive day time sleepiness, REM sleep behavior disorder (RBD), depression, anxiety, urinary disturbance, and low blood pressure [10].
All subjects underwent neurological assessment using an abbreviated version of the MDS-UPDRS (mUPDRS) [14]. This is a validated scale that had been used in previous epidemiological studies to determine MPS prevalence [5, 6] and was performed by trained assessors familiar with the MDS-UPDRS. The MPS score was derived from ten items from the MDS-UPDRS including facial expression, speech, tremor at rest (in any body region), rigidity (rated separately at the neck, right arm, left arm, right leg and left leg), posture and body bradykinesia; with a maximum possible score of 40 [15]. In the present study, the presence of MPS was defined when any of the following conditions was met: 1) two or more UPDRS ratings = 1, or 2) one UPDRS rating ≥2, or 3) UPDRS rest tremor rating ≥1 [5].
Participants who were suspected to have a previously undiagnosed Parkinsonian syndrome were further assessed by a movement disorders specialist. Participants were diagnosed to have PD if they had met the UK Brain Bank Criteria for PD [16]. A cardinal sign was considered to be present when the UPDRS rating was >2.
Cognition was evaluated using the Montreal Cognitive Assessment (MoCA) [17] in participants who were involved in 2017 health screen. The subjects were classified as cognitively normal if the adjusted MoCA was ≥27 and cognitively impaired if the adjusted MoCA score was <27 [18].
The study was approved by the Singhealth CIRB (Centralised Instituitional Review Board).
Statistical analysis
Statistical analysis was performed using SPSS Version 24 for Windows (Armonk, NY: IBM Corp). Baseline demographics and clinical characteristics were compared between subjects with and without MPS using Chi-square or Fisher’s exact test for categorical variables and two-sample t test and Mann-Whitney U test for continuous variables. Descriptive statistics are given either as median (range) and mean±standard deviation for non-categorical data or as percentages of total for categorical data. Linear regression was used to assess the association between cognitive impairment and MPS in the group who had the MoCA performed. A similar model was used to analyze the relationship of prodromal markers of Parkinson’s disease with MPS. Multivariate linear regression was undertaken to examine the influence of potential confounding factors such as age, sex and CVS risk factors. The significance level was set at p < 0.05. Multivariate logistic regression model was performed to investigate the relation of specific cognitive domain to MPS among individual with cognitive impairment. These comparisons are exploratory hence no corrections for multiple comparisons was made.
RESULTS
A total of 422 subjects participated in the study: 19 were excluded because of known PD (n = 5), significant orthopedic comorbidities (n = 11), cerebellar ataxia (n = 1) and known dementia (n = 1). One additional subject was excluded due to non-completion of assessments. A further 11 (2.7%) subjects were excluded as they had fulfilled the clinical diagnostic criteria for PD, and thus represented cases of newly diagnosed PD.
Thus, a total of 392 participants were available for analysis. Of these, only 249 subjects who were screened in 2017 had MoCA assessments performed as part of their assessment. The mean age of the study population was 66.8±6.3 years with 120 (30.6%) males and 272 (69.4%) females. Of these 392 participants, 105 (26.8%) were found to have MPS. MPS were observed in 20.6% of subjects less than 65 years, 28.9% of subjects between the age of 65–74 years and 32.9% in those above 75 years.
The mean age of subjects with MPS (68.8±6.9 years) was significantly higher than those without MPS (66.1±5.9 years) (p value <0.001). Male subjects tended to have higher MPS scores than female subjects (1.34±2.1 vs 0.89±1.5) (p value = 0.01). The adjusted MoCA score was lower in the MPS group (25±4) compared to those without MPS (27±3) (p value <0.001). The demographic and clinical characteristics of the subjects are summarized in Table 1.
Demographic and clinical characteristics of subjects
MPS, mild parkinsonian signs; MoCA, Montreal Cognitive Assessment. Continuous variables are reported as mean/median±standard deviation, categorical variables are reported as frequency (percent). All the values are χ2 except *which is derived from two sample t test (t stat) and Mann-Whitney U test (z stat). **Cognitive screening was done in 249 participants. Cognitive impairment = 124, Normal cognition = 125. Statistically significant differences (p < 0.05) are highlighted in
The proportion of subjects who reported a history of smoking, hypertension, diabetes mellitus, heart disease and hyperlipidemia did not differ between the two groups (Table 1). The proportion of subjects reporting individual prodromal PD symptoms (anxiety, depression, constipation, impaired smell, erectile dysfunction, excessive day time sleepiness, RBD, low blood pressure and urinary dysfunction) were not statistically significant between the two groups (Table 2).
Premotor symptoms in subjects without MPSvs with MPS
MPS, mild parkinsonian signs; RBD, REM sleep behaviour disorder.
Correlation between cognitive function and MPS
249 participants underwent cognitive testing using the MoCA. To determine the association between cognitive function and MPS, participants were divided into 2 groups; cognitively normal and impaired cognition according to adjusted MoCA scores. 125 subjects had normal cognition and 124 subjects were cognitively impaired. Among 249 subjects who underwent cognitive testing, 56 participants had MPS and 193 participants had no MPS. Cognitive impairment was thus present in 42/56 (75%) subjects in the MPS group and 82/193 subjects (42.5%) in those without MPS. We found that cognitive impairment was strongly associated with presence of mild parkinsonian signs (MPS) (p < 0.001). We further conducted multivariate linear regression analysis to examine the relationship between severity of cognitive impairment and after adjusting for age, gender, hypertension, diabetes mellitus, heart disease, smoking and hyperlipidemia. The results showed that lower adjusted MoCA scores predicted higher total MPS scores (β= –0.152, 95% CI = –0.009 to –0.138, p < 0.05) (Table 3).
Relation of MoCA score to total MPS score (n = 249)
Dependent variable: total MPS score. *MoCA score adjusted by educational level. Estimated from Linear regression model adjusted for age, gender, diabetes mellitus, high blood pressure, hypercholesterolemia, heart disease and smoking. Statistically significant differences (p < 0.05) are highlighted in
In addition, we performed multivariable logistic regression analysis to investigate the relationship between specific cognitive functions (Executive function, Visuospatial function, Memory and Language) and MPS among individuals with cognitive impairment. There was no significant difference in relation of specific cognitive domains to MPS (p value all >0.05).
Correlations between prodromal/premotor PD symptoms and MPS
A series of linear regression models were performed to determine potential relationships between prodromal symptoms (depression, anxiety, constipation, impaired smell, RBD, excessive day time sleepiness, erectile dysfunction, urinary dysfunction and low BP) and total MPS score.
In multivariate regression analysis after adjusting for age and gender, weak correlation between RBD and total MPS score was found (β= 0.107, 95% CI = 0.053, 1.490, p < 0.05). Participants with MPS who reported RBD had a higher total MPS score compared to those who had not reported RBD (1.7±2.3 vs 0.9±1.6), (p value <0.05). There were no significant associations identified between the other premotor symptoms and the total MPS score (see Table 4).
Premotor symptoms in relation to total MPS score (n = 391)
RBD, REM sleep behaviour disorder; MPS, mild parkinsonian signs. Dependent variable: total MPS score. *Estimated from Linear regression model adjusted for age and gender. Statistically significant differences (p < 0.05) are highlighted in
DISCUSSION
In this community based cross-sectional study, the overall prevalence of MPS was 27%. MPS was more prevalent amongst the elderly with its prevalence increasing with age. Subjects with MPS had lower adjusted MoCA scores compared to participants without MPS. Total MPS score was also found to be higher in those who reported RBD, but this observation was not present with the other individual prodromal symptoms.
In prior cross-sectional studies, estimates of MPS prevalence have ranged from 15% to 50% of older adults [2–4, 8]. This variation in reported prevalence may be partly explained by differences in study methodology and definitions of MPS adopted by the different studies. Some studies had defined MPS as the presence of any one of the UPDRS components rated 1 or higher [1, 21]. Others have defined this more rigorously as the presence of ≥2 parkinsonian signs or a score of ≥2 for ≥1 item [2, 22]. In order to separate MPS from the signs of normal aging and avoid the influence of other co-morbid disease, we used the stricter criteria of requiring at least two signs with a rating ≥1 (or one sign with a rating ≥2) to define MPS and excluded subjects with arthritis, and other significant medical comorbidities which could affect motor UPDRS scoring thus providing a conservative estimate of the prevalence of MPS. However, as our study had been carried out as part of a community health screening event which could have selected for healthier subjects who were ambulant and thus underestimate the true prevalence of MPS in the community.
It remains unclear whether the occurrence of MPS is largely attributable to vascular risk factors related to aging which in turn lead to age related decline in nigrostriatal dopaminergic activity [4]. Several epidemiological studies of healthy elderly persons with MPS have shown a positive correlation between vascular risk factors or vascular disease and MPS, and hypothesized that vascular risk factors and vascular events might result in pathological changes in the basal ganglia and hence mild parkinsonian signs [8, 24]. In contrast, we did not find any association between cardiovascular risk factors and MPS. The presence of MPS in elderly individuals might reflect, in part, the accumulation of vascular pathological changes in the basal ganglia or white matter regions caused by preventable vascular diseases [8]. However, MPS is likely to be caused by neurodegenerative process in younger individuals around 65 years [13]. This might explain the lack of association between vascular risk factors or diseases and MPS in our study as participants in our study were much younger (mean age 67 years) than the study by Louis et al. (mean age 78 years). Gender disparity in our study may be one of the contributing factors which can explain the lack of an association between CVS risk factors and MPS as the incidence of cardiovascular disease is known to be higher in men than in women of similar age [25]. Previous imaging studies have also shown that small vessel disease burden is associated with the presence of MPS [23]. However, our pilot study did not include brain imaging and thus we could not detect silent infarcts or small vessel disease in the brain. However, some studies have suggested that MPS may be more predictive of future PD development, rather than reflective of underlying vascular disease or mixed pathology [11]. This includes an autopsy study that demonstrated that significant reductions in dopaminergic terminals and neurons can be seen in subjects with MPS [26]. Our lack of association between self-reported CVS risk factors and vascular disease is more supportive of the latter hypothesis.
Several cross-sectional studies have now demonstrated an association between MPS and mild cognitive impairment [5, 27]. MPS and cognitive impairment may share a similar pathogeneses; with the presence of one increasing the odds of having the other [5]. The prevalence and severity of MPS has been shown to increase as the severity of cognitive impairment increases [27]. Results from the current study extend the findings from previous studies and suggest that lower cognitive function was associated with increased MPS. Lerche et al. postulated that MPS in individuals aged 75 or above are more likely to be caused by vascular diseases rather than neurodegeneration [13]. However, our findings remained significant and still unchanged after accounting for age, vascular risk factors and disease, suggesting that MPS in individual with cognitive impairment are not solely explained by vascular factors.
In this study, the presence of RBD was associated with higher MPS scores (β= 0.107, p < 0.05). However, the prevalence of other prodromal PD symptoms was similar between the MPS and non-MPS groups. There have been conflicting reports in literature: Uemera et al. used both validated screening questionnaires and also single question screens to determine the presence of prodromal symptoms of PD and found an increased prevalence of depression but not probable RBD, constipation, hyposmia or orthostatic hypotension in those with MPS [2]. However, in a much larger cross-sectional study, Lerche et al. reported that individuals with MPS, in addition to probable RBD, had a greater frequency of depression, hyposmia and autonomic dysfunction than controls [12] but had used more rigorous screening questionnaires or objective measures to assess prodromal symptoms. In contrast, we used a single-question screening for probable RBD similar to what has been used in other studies [28] which can reliably detect RBD, and it may have allowed us to pick up differences in RBD prevalence between MPS patients and controls; but not for the other prodromal symptoms. In this study, we also showed that subjects who reported probable RBD had a higher total MPS scores compared to those who had not reported probable RBD. This association remained between RBD and MPS even after controlling for age and gender.
The strengths of this study are that this is a community-based study using strict criteria to detect MPS in community ambulant population, which excluded possible confounding factors that may result in the over-diagnosis of MPS, and we had assessed both CVS risk factors and prodromal PD symptoms in this cohort. However, as this was a cross-sectional study, the long-term outcomes of the associations found cannot be assessed in our study. Moreover, we did not perform cognitive screening for all participants and had utilized self-reported questionnaires for the detection of CVS risk factors and premotor symptoms.
Conclusions
In summary, MPS is strongly correlated with increasing age and cognitive impairment. We found that the presence of RBD, which is a specific marker of future neurodegenerative disease, is predictive of higher MPS scores. However, other prodromal non-motor symptoms of PD were not found to be associated with MPS. Observations from this study highlights the need of future longitudinal studies that involved additional investigations and/or sensitive risk markers to determine natural course of MPS and its relation to PD.
CONFLICT OF INTEREST
Authors have no conflict of interest to report.
Footnotes
ACKNOWLEDGMENTS
The authors will like to thank all participants for their contribution towards this study. We also acknowledge those who helped during community health screening conducted in 2017 and 2019. The research is supported by the Singapore National Research Foundation under its Translational and Clinical Flagship Programme (TCR12dec010) and administered by the Singapore Ministry of Health’s National Medical Research Council.