Dietary Antioxidants Associated with Slower Progression of Parkinsonian signs in Older Adults

Abstract

Background and Objective:

Progressive loss of motor function including parkinsonian signs is common in older adults. As diet may contribute to the motor decline, we tested the hypothesis that dietary intake of antioxidant nutrients (carotenoids, vitamin E and vitamin C) is related to the progression of parkinsonian signs in older adults.

Research Design and Methods:

A total of 682 participants without a clinical diagnosis of Parkinson’s Disease from the Rush Memory and Aging Project, were assessed annually over an average of 5.7 (±3.0) years using a 26-item modified version of the United Parkinson’s Disease Rating Scale. The scale assesses the severity of four parkinsonian signs (bradykinesia, gait, tremors, and rigidity) that were averaged to construct a global parkinsonian sign score. Nutrient intakes were assessed at baseline using a validated food frequency questionnaire. The associations between quintiles of antioxidant nutrient intakes and progression of parkinsonian signs were assessed using mixed effects models adjusted for age, sex, education, smoking.

Results:

In separate adjusted models, a slower rate of progressive parkinsonian signs was observed among those in the highest intake quintiles of total carotenoids (β= -0.06, 95%CI: -0.10 to -0.02,), beta-carotene from foods (β= -0.04, 95% CI:-0.08 to -0.0021), lutein-zeaxanthin (β= -0.05, 95%CI:-0.09 to -0.02), vitamin E from foods (β= -0.04, 95%CI:-0.08 to -0.01,) and vitamin C from foods (β= -0.06, 95%CI:-0.10 to -0.02), when compared to those in the lowest quintiles of intake.

Conclusion:

A higher level of dietary antioxidant nutrients may slow the rate of parkinsonian sign progression in older adults.

Introduction

Parkinsonian signs are common and progressive in older adults without a diagnosis of Parkinson’s disease (PD). It is estimated that by the age of 85 years, 50% or more will exhibit mild or severe parkinsonian signs^{1, 2}. Further, the presence of parkinsonian signs is associated with increased risk of disability, adverse health outcomes and mortality². Given its magnitude and the lack of effective treatments, reducing parkinsonian signs in the aging population is a health priority. Diet may be of potential benefit. For example, in a previous study, we observed that healthy dietary patterns, including the Mediterranean-DASH Intervention for Neurodegenerative Disease (MIND) and Mediterranean diets, are associated with reduced risk of parkinsonism and disability^{3, 4}. A characteristic feature of these diets is that they are rich in antioxidant nutrients that have been linked to the prevention of neurodegenerative conditions that involve oxidative stress and inflammation^{5 6}. The case-control studies have reported lower plasma antioxidant status⁷ and similar⁸ or lower⁹ antioxidant intake among people with Parkinsons’s diseaes (PD) when compared to the healthy controls. Prospective analysis on dietray antioxidants and PD risk indicated assocaition of dietary vitamin E with PD risk in the intial analysis¹⁰ but reported no association in longer follow-up⁶. Additionally, a randomized control trial reported flaxseed oil with vitamin E intake improved parkinsonian signs among PD pateints¹¹. However, it is unknown if dietary antioxidants is associated with parkinsonian signs of aging in people without PD diagnosis. In this study, we tested the hypothesis that the dietary antioxidants, including carotenoids, vitamin E and vitamin C are associated with progression of parkinsonian signs among community-dwelling older adults without a clinical diagnosis of PD.

DESIGN and METHODS

Study Population

The study was conducted among participants of an ongoing longitudinal cohort study, the Rush Memory and Aging Project (MAP)¹². MAP recruits residents from more than 40 retirement communities and subsidized housing facilities in the Chicago area to participate in annual structured clinical evaluations. From 1997 to 2017, 1,876 participants were enrolled in the MAP cohort and assessed for parkinsonian signs. Dietary assessments were introduced to the study in 2004 and were completed by 1,062 participants without dementia, of which 905 had at least two annual assessments of parkinsonian signs for the study of parkinsonian sign progression. The first valid dietary assessment is used as the baseline for these analyses (analytic baseline). We also excluded participants who, at the analytic baseline had a clinical diagnosis of PD and were taking dopaminergic medication (n=10), had clinical parkinsonism (two or more parkinsonian signs; n=206) or had missing data on model covariates (n=7). After these exclusions, there were 682 participants included in these analyses (supplementary figure 1). The analytic sample was comparable to the entire MAP sample of 1,876 participants in age (mean of 80.3 versus 79.8 years); percentage female (75% versus 74%); body mass index (27.1 versus 27.3). However, due to the exclusion of those with dementia, PD and parkinsonism at baseline, the analytic sample had higher global cognitive scores (mean standardized scores of 0.23 vs. 0.04) and lower parkinsonian sign scores (mean scores of 4.9 versus 7.6) compared to the entire MAP sample.

Standard protocol approvals, registrations, and patient consents: The Institutional Review Board approved this study, and all the participants gave the written informed consent.

Any data not published within the article will be shared as de-identified upon request from any qualified investigator.

Assessment of Parkinsonian Signs

Parkinsonian signs were assessed annually via structured clinical evaluations conducted by trained nurse clinicians using a previously validated 26-item modified version of the United Parkinson’s Disease Rating Scale. The scale assesses four parkinsonian signs/domains: bradykinesia, parkinsonian gait, rigidity and tremors^{13, 14}. To examine the rate of progression of parkinsonian signs we used a global parkinsonian sign score (range from 0–100), computed as the average of all four parkinsonian sign scores. Higher scores indicate more severe parkinsonian signs.

Nutrient Assessment

Dietary Intakes of the carotenoids (alpha-carotene, beta-carotene, Lutein-zeaxanthin, Lycopene, β-cryptoxanthin), vitamin E and vitamin C were assessed at baseline using a modified Harvard semi-quantitative food frequency questionnaire (FFQ) that was validated for use in older Chicago community residents¹⁵. The FFQ ascertains the usual frequency of intake over the previous 12 months of >137 food items and dietary supplements. The reported frequency of consumption of foods was multiplied by the nutrient content of natural portion sizes (e.g., one banana) or sex-specific mean portion sizes reported by the oldest men and women of national surveys. The nutrient content of each food was based on United States Department of Agriculture (USDA) food composition tables and supplementary nutrient databases¹⁶. Daily nutrient intakes for each participant were computed by summing over the frequency based nutrient content of all the food items as well as the dietary supplements. In the present analysis, we also analyzed nutrient intake from food sources only as an independent variable. All nutrients were calorie-adjusted using the residual regression method within sex¹⁷. Lutein and zeaxanthin intakes are presented together because current laboratory methods do not provide individual quantification of these carotenoids in foods. Total carotenoid intake is the sum of intake levels of alpha-carotene, beta-carotene, lutein-zeaxanthin, lycopene, and β-cryptoxanthin.

Other Covariates

Non-dietary covariates were obtained from the participant’s clinical evaluation. Information on age (in years, calculated from birth date to clinical evaluation date), education (in years) and smoking history (never, past or current smoker) ad family history of PD (if any parent or sibling was diagnosed with PD) were self-reported at baseline. Body Mass index (BMI) was computed from the annually measured weight (kg) and height (meters) as kg/meters² and modeled as two indicator variables, BMI≤20 and BMI≥30 (based on existing literature on BMI categorization in aging cohort)¹⁸. Physical activity (hours per week) was also recorded in each clinical visit based on self-reported time spent over two weeks on five activities (walking for exercise, yard work, calisthenics, biking, and water exercise).

Statistical methods

We examined nutrient intake associations with change in global parkinsonian sign score using linear mixed models with random effects. The global parkinsonian score was square root-transformed to meet model assumptions of normality. Antioxidant nutrient intake was modeled in quintiles. Models were first adjusted for age, sex, education, smoking, a variable for time, and multiplicative terms of time with each model covariate. Further analyses added terms for other potential confounders/mediators to the basic model, including 1) Physical activity, and BMI 2) other anti-oxidant nutrients. Additionally, the basic model was also adjusted for family history of PD. BMI and physical activity were modeled as time-varying (for example, updated measures of BMI were modeled for each follow-up year analyzed). We also assessed potential modification of nutrient effects on parkinsonian signs by age (older or younger than 80 years), sex, and smoking status (ever smoked vs. non-smokers) by including three-way multiplicative terms between the nutrient of interest, the potential effect modifier and time in the basic model with significance level p≤0.05. All analyses were performed using ©SAS version 9.4.

RESULTS

The analytical sample was primarily white (93%) and female (75%), with an average age of 80.3 years (±7.1) at the time of the first dietary assessment. Many participants (42%) were either past or current smokers and only 8% reported a family history of PD. Baseline characteristics of the participants were similar by quintile of total carotenoid intake except that the highest quintile had fewer males, and higher physical activity scores compared to the lowest quintile (Table 1). The Spearman correlation coefficients between total carotenoids, vitamin E and vitamin C ranged from 0.20 to 0.60 whereas the carotenoids were moderately or highly associated with each other (r= 0.25 to 0.84), except lycopene (all r < 0.10) (Supplemental Table 1).

Table 1:

Baseline characteristics based on quintiles of total carotenoids intake in the study population

	N	Total carotenoids intake (energy adjusted, mcg/day)
		Q1	Q2	Q3	Q4	Q5
N	682	136	136	137	136	137
Median daily intake, mcg/day	682	6,541	10,388	13,169	16,411	24,400
Age, mean ± SD, years	682	81.4 ± 7.6	80.1 ± 7.3	80.3 ± 7.0	81.1 ± 6.5	78.88 ± 7.14
Male, %	682	28.7	25.7	25.5	22.8	21.9
Education, mean ± SD, years	682	14.1 ± 2.3	15.6 ± 2.9	15.1 ± 3.0	15.1 ± 2.9	15.8 ± 3.0
Smoking %, never	682	58.1	54.4	59.8	59.6	55.5
Physical activity, mean± SD, hours/week	681	2.8 ± 3.3	3.9 ± 4.3	3.6 ± 3.9	3.3 ± 2.9	4.7 ± 4.4
BMI, mean ± SD	665	26.8 ± 5.3	27.1 ± 4.9	26.7 ± 4.7	27.6 ± 5.2	27.3 ± 5.5
Total Energy, mean ± SD, kcal/d	682	1937 ± 681	1867 ± 448	1682 ± 522	1640 ± 470	1586 ± 416
Multi-vitamin use, %	682	64	72	78	72	71
Beta-carotene supplement use, %	682	4	4	9	10	8
Vitamin E supplement use, %	682	43	35	43	37	41
Vitamin C supplement use, %	682	49	31	36	48	44
Hypertension, %	676	74	75	80	77	74
Diabetes, %	682	24	19	23	22	24
Stroke, %	625	16	11	8	12	8
Myocardial Infarction	682	15	14	13	15	15

Carotenoids Intake and Progression of Parkinsonian signs

Participants were followed for an average of 5.7 (±3.0) years. In separate adjusted models, dietary intakes of total carotenoids, beta-carotene from food sources, and lutein-zeaxanthin were each inversely associated with the progression of parkinsonian signs. Intakes of total alpha-carotene, betacarotene, lycopene, and beta-cryptoxanthin, were not associated with slower progression of parkinsonian signs (Table 2). Figure 1.a provides a graphic characterization of the findings for total carotenoids, showing a significantly slower rate of progression for quintile 5 of carotenoid intake versus quintile 1. In subsequent analyses, we adjusted for physical activity and BMI in the basic model. There were no material differences in the results for total carotene (Q5 vs. Q1: β= -0.06 (95% CI: -0.10, -0.02; p for trend <0.0001) and lutein-zeaxanthin (Q5 vs. Q1: β= -0.06 (95% CI:-0.10, -0.02; p for trend= 0.004). However, beta-carotene intake from food sources (Q5 vs. Q1: β=-0.03 (95% CI: -0.07, 0.006; p for trend=0.15) was no longer statistically significant. Further adjusting for family history of PD in the basic model did not change the association total carotene (Q5 vs. Q1: β= -0.06 (95% CI: -0.10, -0.02; p for trend <0.0001) and lutein-zeaxanthin (Q5 vs. Q1: β= -0.05 (95% CI:-0.09, -0.01; p for trend= 0.005).

Table 2:

Estimated effects (beta and 95% confidence interval) for association of carotenoids, vitamin E and C with progression of parkinsonian signs among 682 MAP participants:

Model	Quintile of total vitamin intake					P for trend
	Q1	Q2	Q3	Q4	Q5
Total carotenoids
Median intake, mcg/day	6,541	10,388	13,169	16,411	24,400
beta 95% CI	Ref	−0.02 (−0.06,0.01)	−0.004 (−0.04,0.03)	−0.002 (−0.04,0.04)	−0.06 (−0.10,−0.02)	<0.001
Alpha carotene
Median intake, mcg/day	260	517	745	1117	1889
beta 95% CI	Ref	−0.04 (−0.08, 0.003)	−0.01 (−0.05, 0.03)	−0.04 (−0.08, 0.001)	−0.04 (−0.08, 0.002)	0.13
Beta carotene from food sources
Median intake, mcg/day	1548	2641	3501	5439	8114
beta 95% CI	Ref	−0.02 (−0.06,0.01)	−0.01 (−0.05,0.03)	−0.03 (−0.08,0.002)	−0.04 (−0.08,−0.002)	0.04
Beta-carotene (food + supplement)
Median intake, mcg/day	1848	2967	4250	6022	9440
beta 95% CI	Ref	−0.01 (−0.05,0.02)	−0.03 (−0.07,0.006)	−0.04 (−0.08,0.002)	−0.04 (−0.08,0.003)	0.08
Lutein-zeaxanthin
Median intake, mcg/day	1144	2060	2885	4110	8197
beta 95% CI	Ref	−0.02 (−0.06,0.02)	−0.02 (−0.06,0.02)	−0.03 (−0.07,0.004)	−0.05 (−0.09,−0.02)	0.004
Lycopene
Median intake, mcg/day	1084	3051	4644	6554	9562
beta 95% CI	Ref	−0.03 (−0.08,0.003)	−0.04 (−0.08,−0.01)	−0.03 (−0.07,0.01)	−0.04 (−0.08,0.003)	0.14
β-cryptoxanthin
Median intake, mcg/day	53	105	160	224	311
beta 95% CI	Ref	−0.02 (−0.06,0.02)	−0.03 (−0.07,0.007)	0.001 (−0.04,0.04)	−0.04 (−0.07, 0.001)	0.18
Total Vitamin E (food+supplements)
Median intake, mg/day	5.33	16.42	25.07	47.32	225.62
beta 95% CI	Ref	−0.0006 (−0.04,0.04)	0.01 (−0.03,0.05)	−0.02 (−0.03,0.06)	−0.004 (−0.04, 0.03)	0.53
Vitamin E (from food sources only)
Median intake, mg/day	4.35	5.18	5.78	6.53	8.15
beta 95% CI	Ref	−0.03 (−0.07,0.01)	−0.05 (−0.09, −0.008)	−0.04 (−0.08,0.004)	−0.04 (−0.08, −0.01)	0.05
Total Vitamin C (food+supplements)
Median intake, mg/day	93	148	206	361	990
beta 95% CI	Ref	−0.03 (−0.07,0.01)	−0.05 (−0.08,−0.004)	−0.004 (−0.04, 0.04)	−0.01 (−0.06, 0.03)	0.53
Vitamin C (from food sources only)
Median intake, mg/day	54	87	115	146	201
beta 95% CI	Ref	−0.03 (−0.07,0.007)	−0.05 (−0.09,−0.01)	−0.03 (−0.08,0.006)	−0.06 (−0.10,−0.02)	0.02

Linear mixed models with random effects were used. Basic model was adjusted for age, sex, education, smoking, lag and interaction term between lag and other covariates

Figure 1:

Change in the global parkinsonian score (square root transformed) for the person in quintile 1, 3 and 5 of (a) total carotenoids, (b) vitamin E from food sources and (c) vitamin C from food sources, based on mixed models adjusted for age, sex, education, and smoking.

Vitamin E and Vitamin C Intake and Progression of Parkinsonian signs

Vitamin E from food sources (p for trend=0.05) and vitamin C from food sources (p for trend= 0.03) each had a significant inverse association with the progression of parkinsonian signs (Table 2). Figures 1.b and 1.c provide graphic representations of the rates of parkinsonian signs progression over time for participants in the lowest quintiles of vitamin E and vitamin C intakes from food sources compared with intake quintiles 3 and 5. In further analyses that included adjustment for physical activity, and BMI to the basic model, the association of vitamin C from food sources (Q5 vs. Q1: β= -0.05 (95% CI:-0.09, -0.01; p for trend= 0.05) as well as vitamin E from food sources (Q5 vs. Q1: β= -0.04 (95% CI:-0.08, -0.003; p for trend= 0.09) with a slower rate of parkinsonian signs progression remained . These associations were also retained in the models additionally adjusted for family history of PD (vitamin C from food sources (Q5 vs. Q1: β= -0.06 (95% CI:-0.10, -0.02; p for trend= 0.02); vitamin E from food sources (Q5 vs. Q1: β= -0.04 (95% CI:-0.08, -0.004; p for trend= 0.06)). Total vitamin E and total vitamin C intake (intake from food and supplements) was not associated with parkinsonian signs progression.

To understand the independent effects of total carotenoids, lutein-zeaxanthin, vitamin E from food sources, and vitamin C from food sources, we re-analyzed the models with adjustment for other antioxidant nutrients. The associations of delayed parkinsonian signs progression with total carotenoids (Q5 vs. Q1: β= -0.05, p=0.03), lutein-zeaxanthin (Q5 vs. Q1: β= -0.04, p=0.05), and vitamin C (Q5 vs. Q1: β= -0.04, p value= 0.04) were retained in these analyses. However, the vitamin E association was no longer statistically significant (Q5 vs. Q1: β= -0.02, p value= 0.28).

We further investigated whether the observed associations were modified by age, sex, and smoking. The only statistically significant modification of effects was that for beta-carotene from food sources by age (p= 0.01) and vitamin C from food sources by sex (p=0.02). Higher beta-carotene intake from food sources was associated with the slower parkinsonian signs progression among adults less than 80 years of age (Q5 vs Q1: β=-0.07, p=0.01) but not for older adults > 80 years old (Q5 vs Q1: β=0.01, p=0.51). Vitamin C intake from foods was associated with slower progression of parkinsonian signs in men (Q5 vs Q1: β=-0.15, p=0.004) but not in women (Q5 vs Q1: β=-0.04, p=0.09).

DISCUSSION

To our knowledge, this is the first longitudinal study examining the association between dietary intake of antioxidant nutrients and progression of parkinsonian signs among community-dwelling older adults. In this prospective analysis, dietary intakes of antioxidant nutrients including carotenoids, vitamin E and vitamin C were associated with slower progression of parkinsonian signs. These data suggest a possible role of antioxidant nutrients in the diet as a modifiable risk factor in slowing the progression of a common heterogeneous disorder that is known to affect motor functions in older adults.

Parkinsonian signs are common motor impairments in older adults and are associated with diverse adverse health outcomes². Our findings for total dietary carotenoids and lutein are supported by several studies that reported positive associations of carotenoids with other motor outcomes including walking ability¹⁹, muscle strength²⁰ and Parkinson’s disease risk^21–23. However, some observational studies of middle-aged populations reported null associations between dietary carotenoids and physical decline²⁴ and PD risk¹⁰. Findings on lutein-zeaxanthin intake and parkinsonian signs are also supported by the randomized trial of lutein-zeaxanthin supplementation that improved visual-motor reaction time²⁵ and motor abnormalities²⁶ in young adults.

Other important antioxidant nutrients assessed in this study were vitamin E, and vitamin C. We found that both vitamin E and vitamin C from food sources were associated with slower progression of parkinsonian signs. The association of dietary vitamin E with parkinsonian signs is consistent with previous findings of vitamin E associations with improved physical performance²⁴ and decreased PD risk^{10, 23}. The population-based studies on dietary vitamin C indicate mixed findings, some report null results with physical decline²⁴ and PD risk¹⁰, whereas another reported positive associations with skeletal muscular strength²⁷. We observed that the association of vitamin C on parkinsonian signs was stronger in males. Our results are similar to findings from the Boston Area Community Health Cohort in which vitamin C was associated with physical decline only in men²⁴. However, in a Swedish cohort, vitamin C intake was associated with lower risk of PD only in women²³. We speculate the restriction of the vitamin C association to males in our study may be attributed to higher fat-free mass and thus higher distribution volume of vitamin C in older men compared to older women²⁸ which may lead to more antioxidant effect at similar intake levels.

The mechanism underlying the associations between antioxidant nutrients and parkinsonian signs presented in this paper is unclear. Carotenoids, vitamin E, and vitamin C are key dietary antioxidant nutrients that may reduce oxidation and the production of reactive oxygen species. High plasma levels of these nutrients are also associated with reduced inflammatory markers²⁹. Oxidative stress and inflammation are two main mechanisms for neurodegenerative disorders in older adults³⁰. Various in-vitro studies have indicated that carotenoids may decrease oxidative stress and inflammation via different direct and indirect pathways³¹. Animal models suggest vitamin E reduces oxidative stress in the brain³², restores synaptic plasticity⁹ and has a neuroprotective effect in aging rats³³. Similarly, vitamin C is reported to scavenge free radicals, prevent membrane lipid peroxidation in the brain,³⁴ and modulate glutamatergic neurotransmission^{35, 36} in various experimental models. Recent work suggests that subclinical mixed brain pathologies such as Alzheimer’s disease, Lewy body dementia, and cerebrovascular disease pathologies may contribute to the presence of severe parkinsonian signs without the overt neurological disease in older adults³⁷. Thus, a healthy diet rich in antioxidant nutrients with free radical scavenging properties may play an important role in preventing parkinsonian signs and its progression in old age.

Some of the important strengths of our study include the longitudinal design with a repeated annual assessment of parkinsonian signs with mean follow-up time of 5.7 (±3) years, nutrient intake assessments with a comprehensive, validated questionnaire for older adults, comprehensive assessment of multiple potential confounders, and the community-dwelling study cohort. Limitations of the study include the non-diverse study population of the cohort which limits the generalizability of these results to overall diverse US population of older adults and its observational design that prevents establishing causality between nutrient intake and parkinsonian signs. Additionally, the analytical sample used in the cohort is small to tease out the multiple independent effects of these correlated nutrients on parkinsonism.

CONCLUSION

Parkinsonian signs in older adults is a prevalent motor impairment that progresses with time and is associated with disability, adverse health outcomes, poor quality of life, and mortality. Currently, there is no known cure. Thus, the potential for prevention or delay in the progression of late-life motor decline through dietary intervention is important to public health. We found a higher level of dietary antioxidant nutrients intake may slow the rate of parkinsonian signs progression in older adults. Previous studies have reported the association of antioxidant nutrients on cognitive decline. Thus, focusing on these nutrients in older adults may help with late-life cognition as well as motor decline. Further studies are required to establish these findings through replication in more diverse populations and, ultimately, a randomized dietary intervention trial to confirm causality between antioxidant nutrient intake and parkinsonian signs.