Abstract
Background:
Some epidemiology studies suggest that atherosclerotic cardiovascular disease (ASCVD) risk factors increase the risk of developing Parkinson’s disease (PD). However, conflicting data suggest lower rates of ASCVD in PD.
Objective:
The objective of this study is to determine, with data from a longitudinal clinicopathological study, whether ASCVD risk factors are associated with a PD diagnosis and/or increased brain or peripheral load of Lewy-type synucleinopathy (LTS).
Methods:
All subjects were followed to autopsy and neuropathological examination in the Arizona Study of Aging and Neurodegenerative Disorders (AZSAND). Multivariable regression models, including age, gender, and smoking history, were used to investigate the association of a PD diagnosis or brain or submandibular gland LTS load with ASCVD risk factors.
Results:
150 subjects were included (PD n = 60, controls n = 90). Univariable comparisons and regression models showed a general trend to inverse associations. The multivariable odds ratio (OR) of brain LTS load for carotid artery disease was 0.93 (95% CI: 0.86 to 0.98; p = 0.02), for anticoagulant use 0.95 (95% CI: 0.90 to 0.99; p = 0.04) and for abnormal heart weight 0.96 (95% CI: 0.92 to 0.99; p = 0.01). Composite clinical and overall (clinical + pathology composite risk scores) composite risk scores were also significantly lower in the PD subjects (p = 0.0164 and 0.0187, respectively). Submandibular gland LTS load was not significantly related to ASCVD conditions.
Conclusions:
This study shows associations of higher brain LTS with lower prevalence of both clinical and pathological indices of ASCVD in PD subjects versus age-similar controls. We suggest that this is due to α-synuclein pathology-induced sympathetic denervation in PD.
INTRODUCTION
Parkinson’s disease (PD) is the second most common neurodegenerative disease, following Alzheimer’s disease (AD). While many studies have linked atherosclerotic cardiovascular disease (ASCVD) risk with the risk of developing AD, associations with PD have been inconsistent. ASCVD risk factors include gender, age, race, total cholesterol, HDL-cholesterol, systolic blood pressure, treatment for high blood pressure, diabetes and smoking. Two prospective and case-controlled studies demonstrated no association between hypertension (HTN) [1], high blood cholesterol [1], diabetes (DM) [1], or vascular risk factors [2] and PD. Indeed, it was hypothesized that perhaps high cholesterol could lower PD risk [1]. In contrast, other prospective studies identified an association between high cholesterol [3], HTN [4], DM [5], and obesity [6] with an increased risk for developing PD. These large prospective epidemiologic studies have been well-powered, but while identifying incident PD, they lack neuropathologic correlation or confirmation of diagnosis. Three autopsy studies directly assessing cerebrovascular disease did not find significant differences between neuropathologically-confirmed PD and control subjects [7–9], while a fourth found that PD subjects had significantly less small-vessel disease, as well as a significantly decreased prevalence of clinical ASCVD risk factors [10].
There have been no reported studies that directly correlate PD Lewy-type synucleinopathy (LTS) density with ASCVD risk factors. It is well known from our previous work and that of others, that LTS in PD is extensive, both peripherally and centrally [11, 12]. In particular, LTS is found in preganglionic autonomic neurons, sympathetic ganglia, the vagus nerve, the heart [13, 14] and the adrenal gland [11, 15]. This extensive pathology in the autonomic nervous system may explain the dysautonomia experienced by PD patients, which can manifest as pre-motor and non-motor symptoms. Both parasympathetic and sympathetic activity play an important role in homeostasis including vasomotor control and hormonal release, which affect heart rate, blood pressure, digestion and metabolism [16]. We hypothesize that the LTS load in vasomotor areas of the brain, adrenal gland and the peripheral autonomic nervous system leads to a decrease in sympathetic tone and subsequent reduction of blood pressure and other vascular risk factors.
We report the findings of our pathology-confirmed case-control cohort identifying reductions of some ASCVD risk factors and postmortem cardiac pathology measures in PD subjects, as well as correlations of lower ASCVD risk factors with increased brain LTS load.
MATERIALS AND METHODS
All subjects completed informed consent and were enrolled in the Arizona Study of Aging and Neurodegenerative Disorders (AZSAND) [17]. The AZSAND database was queried for all cases with a clinicopathological diagnosis. Exclusion criteria are outlined in the results section. All subjects received standardized movement disorder examinations and a clinical diagnosis by Movement Disorders fellowship trained Neurologists (EDD, CHA, SHM, HAS). Autopsies were performed by the Banner Sun Health Research Institute Brain and Body Donation Program (BBDP). Details regarding the clinical and neuropathological methods of AZSAND and the BBDP have previously been published [18, 19]. For this study the AZSAND database was queried for subjects’ past or current medical history, smoking history and medication use.
Pathology assessments
All subjects underwent autopsy and a standardized neuropathological assessment [17, 18]. This post mortem assessment included Lewy body staging using the Unified Lewy Body Staging System [19]. Lewy-type synucleinopathy densities in brain and submandibular gland were determined semi-quantitatively with reference to a standardized template published by the Dementia with Lewy Bodies Consortium [20]. Postmortem cardiac measurements included: heart weight, degree of coronary artery stenosis, presence of myocardial infarction (MI), circumference, and calcification of heart valves. The estimation of coronary artery stenosis was done in a standardized manner. The anterior and posterior descending arteries were transected at 0.5 cm intervals along their lengths and the degree of stenosis by yellowish-white atheromatous plaque was graded as mild, moderate or severe according to example gross photographs. For this study, the highest score at any transect of either measured artery was used. Abnormal heart weight was defined as two standard deviations above the mean from reference normative data [21].
Statistical analysis
Demographics, clinical characteristics, and the prevalence of ASCVD medical history conditions as well as postmortem pathology conditions, were compared between PD, controls and controls with incidental Lewy body disease (ILBD; defined as non-demented and without parkinsonism) using the two-sample t-test, Wilcoxon rank sum test, Chi-square test or Fisher’s Exact test as applicable. Individual ASCVD and pathology conditions were compared as well as composite risk scores derived from summations of the number of conditions possessed by each individual. Multivariable linear and logistic regression models adjusting for age, gender, and smoking history were used to investigate the association of a PD diagnosis, or of brain or submandibular gland LTS load, with individual and composite ASCVD risk factors. Results were adjusted for multiple comparisons using the false discovery rate method.
RESULTS
We identified 732 subjects with a final clinicopathological diagnosis of control, control with ILBD that had whole-body autopsy including the heart. The following subjects were excluded due to other major disease conditions: 308 dementia without PD, 15 progressive supranuclear palsy (PSP), 5 multiple system atrophy (MSA), 10 parkinsonism NOS, 5 incidental PSP, 1 Huntington’s disease, 15 Other miscellaneous conditions. From the remaining 373 cases, 223 subjects without heart pathology data available were excluded, leaving 150 subjects (PD n = 60, controls n = 73; ILBD n = 18). As controls with and without ILBD did not differ in neurological or ASCVD measures (Table 1), the two groups were combined for the remainder of the analyses presented here (Tables 2– 6). The average age for PD (n = 60) was 79.9 (6.7) years vs 85.1 (11.0) years for controls (n = 90; p = 0.0005). PD subjects were 66.7% male vs 52.2% for the controls (p = 0.08).
Demographics and clinicopathological comparisons: controls and controls with incidental Lewy body disease (ILBD)
1Wilcoxon rank sum p-value; 2Chi-Square p-value.
Demographics and clinicopathological comparisons between controls and PD subjects
1Wilcoxon rank sum p-value; 2Chi-Square p-value.
On univariable analysis, specific individual ASCVD conditions (Table 3) that significantly differed between PD and control were carotid artery disease (1.7% vs 16.70%, p = 0.004), anti-coagulant usage (7.7% vs 25.3%, p = 0.013), anti-hypertensive usage (34.6% vs 64.6%, p = 0.0012) and abnormally high heart weight (13.3% vs 37.8%, p = 0.0014). Others were not significantly different, although diabetes was close to the significance level (18.3% vs 31.3%, p = 0.09). The group differences in the composite clinical risk score and the overall (clinical + pathology factors) risk score were highly significant (p < 0.0001), with lower scores in the PD subjects.
Prevalence rates for clinical and pathology cardiovascular risk factors in the two groups
Results shown are number of subjects with each condition, percentage of group with each condition, number of subjects for whom data was available, and p-value for the comparison.
Subjects with PD had varying brain loads of LTS. The regression models for individual ASCVD risk factors (Table 4) against LTS brain load generally showed inverse associations. Thus, as the LTS load increases, the likelihood of having several cardiovascular disease indicators or risk factors decreases. Three of these models reached significance. The multivariable odds ratio (OR) of brain LTS load for carotid artery disease was 0.93 (95% CI: 0.86 to 0.98; p = 0.02), for anticoagulant use 0.95 (95% CI: 0.90 to 0.99; p = 0.04) and for enlarged heart 0.96 (95% CI: 0.92 to 0.99; p = 0.012). After adjustments for multiple comparisons, however, none of these individual ASCVD regressions remained significant. Of composite risk scores (Table 5) linear regression analysis found both the clinical score and overall composite scores were significantly lower in PD subjects (p = 0.016 and 0.019, respectively) after adjustment for multiple comparisons.
Multivariable logistic regression model results for brain summary LTS density vs individual ASCVD conditions
Fifteen subjects without LTS density score or smoking status were excluded (n = 135). Results are adjusted for age, gender and smoking history.
Forty of 60 PD cases had peripheral alpha-synuclein pathology load assessment in the submandibular gland and 35 of the 40 PD cases assessed (85.4%) were positive for submandibular gland LTS while none of 73 assessed controls (with or without ILBD) were positive. Regressions of submandibular gland LTS load against ASCVD measures (Table 6) showed only one of the 26 regressions was significant (in the opposite direction as expected, with a higher rate of left ventricular myocardial infarction in PD subjects).
Multivariable linear models results for composite cardiovascular risk scores. Results are adjusted for age, gender and smoking history
1ASCVD clinical risk score = Diabetes + Hyperlipidemia + Hypertension + AF + Stroke + PVD + Transient ischemic attack + Heart Disease + CAD + Obesity + Statin Use + Anti-Cog Use + BP Med Use + Anti-Platelet Use. 2ASCVD pathology risk score = abnormal heart weight + abnormal left coronary artery stenosis + abnormal right coronary artery stenosis + abnormal left ventricular myocardial infarct + abnormal mitral calcification of heart valves + abnormal triscupid calcification of heart valves + abnormal pulmonary calcification of heart valves + abnormal aortic calcification of heart valves + abnormal triscupid circumference of heart valves + abnormal pulmonary circumference of heart valves + abnormal mitral circumference of heart valves + abnormal aortic circumference of heart valves. 3ASCVD overall risk score = ASCVD clinical risk score + ASCVD path risk score.
Multivariable logistic regression model results for ASCVD using submandibular gland LTS density as primary predictor
Thirty-seven subjects were excluded due to absence of a submandibular gland LTS density score (n = 113). Results are adjusted for age, gender and smoking history.
\enlargethispage 3pt
As the control group was slightly higher in age than the PD group, we did a sensitivity analysis to explore whether this was a critical factor. We used the propensity score matching method to match the PD cases and controls based on their age at death, gender and smoking history. There were 74 subjects (37 PD cases and 37 controls) who could be matched. We compared the 74 matched patients with the 76 “unmatched” patients to make sure the 74 matched subjects representative of the total 150 subjects, and found that these two groups are similar regarding their demographics and clinical characteristics. We then repeated the logistic regression models, excluding age, gender and smoking history because these had been rendered non-contributory because of the matching process, and obtained regression results that were essentially the same as in our original analysis.
DISCUSSION
This study showed that subjects with a PD diagnosis have a lower prevalence of ASCVD risk factors and heart disease, and this same relationship is accentuated for PD subjects with higher brain LTS loads. Our case-control analysis of individual ASCVD factors revealed a consistent general trend toward higher rates of ASCVD risk factors in our control group vs PD, although only reaching statistical significance for a history of carotid artery disease, anti-coagulant usage, anti-hypertensive usage and abnormally high heart weight. Cigarette smoking has been repeatedly found to be associated with decreased PD risk, for reasons that are still unclear [22].
With regard to LTS brain load, the multivariable regression models showed a statistically significant difference for carotid artery disease, anticoagulant use, and heart weight, all of which were reduced in the PD group. It is surprising that the regression models utilizing submandibular gland LTS did not support those done with brain LTS. This may be due to greater variability in the SMG measures, as these are taken from a single anatomical site while the brain LTS load is a summary density from 10 sites.
The PD subjects were less likely to have an abnormally enlarged heart. This is a new finding as our study is the first to include heart weight in an analysis of PD and control subjects. Increased ASCVD generally is accompanied by increased heart weight due mostly to the effects of prolonged hypertension. We propose that the decreased heart weight, and of other ASCVD factors in PD is likely secondary to the decrease in peripheral sympathetic tone due to LTS burden in sympathetic ganglia and nerves. The sympathetic nervous system plays a crucial role in blood pressure, heart rate, and blood vessel contractility, with an additional impact on cholesterol and glucose levels [16]. Sympathetic drive raises blood pressure, increases heart rate, constricts blood vessels, elevates blood glucose levels, and leads to dyslipidemia [16]. All of these effects of sympathetic over-activity could lead to the development of vascular risk factors and disease. The extent of LTS both centrally and peripherally in PD likely reduces sympathetic function, thus disrupting the parasympathetic and sympathetic homeostasis. This could explain the decrease in vascular risk factors and disease noted in autopsied PD subjects.
Supporting our hypothesis, the prevalence of history of diabetes, hypertension, usage of anti-hypertensive drugs, hyperlipidemia and heart disease were all lower in the PD subjects, however, the differences were not pronounced and none of these reached the significance level after adjustment for age, gender and smoking history. This is therefore relatively weak evidence for our sympathetic drive hypothesis. However, a previously-published autopsy study [10] also found a decrease in prevalence of hypertension, diabetes and heart disease, reaching the significance level for heart disease and for a summation of vascular risk factors. An alternative explanation for the lower heart weights in PD subjects might be their decreased level of mobility and exercise. Another possibility is that our hypertension prevalence for PD was overestimated, because they were ascertained as simply present or absent over multiple years of clinical observation, with any single record of hypertension being sufficient. It is possible that PD subjects have equivalent rates of hypertension in prodromal or early-stage disease but decreased rates in later-stage disease, as sympathetic denervation becomes more pronounced. Supporting a later-stage effect of PD on these, we did not find significant ASCVD differences between ILBD cases and non-ILBD controls, suggesting that cardiovascular effects are not evident in prodromal PD.
This study has several limitations including a relatively small sample size and an inability to control for possible medication effects. A strength of this study is its primacy in the investigation of how a PD clinicopathological diagnosis and LTS brain load are related to clinically-documented ASCVD risk factors and postmortem heart pathology in neuropathologically confirmed PD cases. It is difficult to compare our results to prior studies due to the differences in study design. Prior studies were either case controlled or prospective epidemiological studies looking at the risk for incident PD, and the great majority lacked neuropathological confirmation of PD diagnosis. A similar neuropathology-confirmed study has previously reported similarly decreased ASCVD indices in PD [10]. We suggest that an important factor leading to reduced ASCVD risk factors and manifest disease may be the sympathetic denervation that results from the progressive development of LTS in the peripheral sympathetic nervous system [13, 14].
COMPLIANCE WITH ETHICAL PUBLICATION GUIDELINES
The Banner Sun Health Research Institute IRB and Mayo Clinic IRB approved this study. All subjects signed informed consent and were enrolled in the Arizona Study of Aging and Neurodegenerative Disorders (AZSAND), an ongoing longitudinal clinicopathologic study in the Banner Sun Health Research Institute Brain and Body Donation Program (BBDP). All procedures involving experiments on human subjects are done in accord with the ethical standards of the Committee on Human Experimentation of the institution in which the experiments were done or in accord with the Helsinki Declaration of 1975.
CONFLICT OF INTEREST
EDD
NZ
HAS
CMB
SHM
TGB
EYZ
CHA
Footnotes
ACKNOWLEDGMENTS
The Brain and Body Donation Program has been supported by the National Institute of Neurological Disorders and Stroke (U24 NS072026 National Brain and Tissue Resource for Parkinson’s Disease and Related Disorders), the National Institute on Aging (P30 AG19610 Arizona Alzheimer’s Disease Core Center), the Arizona Department of Health Services (contract 211002, Arizona Alzheimer’s Research Center), the Arizona Biomedical Research Commission (contracts 4001, 0011, 05-901 and 1001 to the Arizona Parkinson’s Disease Consortium) and the Michael J. Fox Foundation for Parkinson’s Research.