The Relationship Between Flavonols Intake and Stroke in the Elderly: NHANES (2007-2010 and 2017-2018)

doi:10.21203/rs.3.rs-4989313/v1

Download PDF

Research Article

The Relationship Between Flavonols Intake and Stroke in the Elderly: NHANES (2007-2010 and 2017-2018)

https://doi.org/10.21203/rs.3.rs-4989313/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Stroke is a common fatal and disabling disease in the elderly. We investigated the correlation and potential benefits of dietary intake of flavonoid compounds and their subclasses in elderly stroke patients.

Methods

This study utilized data from the National Health and Nutrition Examination Survey (NHANES) from 2007–2010 and 2017–2018. Flavonoids intake was calculated based on food frequency questionnaires (FFQ) through a 24-hour dietary recall. Multivariable logistic regression analysis was employed to explore the relationship between flavonoids intake and their subclasses with stroke prevalence. Restricted cubic splines (RCS) were used to investigate the nonlinear relationship between flavonols subclasses and stroke. Multivariable logistic regression models were used to evaluate the association between flavonols intake and stroke among all participants and across different flavonols subgroups. Bayesian kernel machine regression (BKMR) was introduced to assess the overall effect of flavonols intake levels on the risk of stroke status. Considering the relatively high correlation among flavonols subclasses, we further implemented a hierarchical variable selection method, performing 50,000 iterations using the Markov Chain Monte Carlo algorithm. We then calculated the conditional posterior inclusion probability (condPIP).

Results

Including 3,806 elderly stroke patients, the study revealed an inverse relationship between dietary flavonoids and their subclasses and stroke prevalence. After adjusting for potential confounders, it was found that higher quartiles of flavonols intake were associated with lower stroke prevalence. Specifically, with each unit increase in flavonols (Q4) intake, the odds of stroke in the elderly decreased by 61% (OR = 0.390, 95% CI [0.209–0.728]; P = 0.005). Similar results were observed for the subclasses of flavonols. Subgroup analyses indicated that age and poverty index ratio (PIR) were effect modifiers in the relationship between flavonols intake and stroke. We further examined the intake levels of dietary flavonols subclasses such as isorhamnetin, kaempferol, myricetin, and quercetin and their association with stroke status stratified by population characteristics. In addition to age and PIR, hyperlipidemia and body mass index (BMI) were found to be the most common significant influencing factors in the relationship between flavonols subclasses and stroke prevalence. Furthermore, RCS revealed a “U”-shaped nonlinear relationship between flavonols, including their quercetin and kaempferol subclasses, and stroke, whereas the relationship between stroke and myricetin was linear. Our study also assessed the overall impact of dietary flavonols subclasses on stroke in the elderly and the interrelationships among these subclasses. The results consistently indicated a negative joint effect of flavonols subclass mixtures on the risk of stroke in the elderly. When evaluating the impact of individual flavonols subclasses on stroke outcomes, a potential dose-response relationship was observed, with increasing intake of myricetin being associated with a decreased risk of stroke.

Conclusion

These results emphasize that adhering to an increased dietary intake of flavonoid compounds, particularly flavonols and their subclasses such as myricetin, can significantly reduce the prevalence of stroke among the U.S. elderly population. This offers potential benefits for stroke patients, especially among elderly individuals aged 60–70 and those with higher incomes.

Flavonols

Stroke

Flavonoids

Flavonoid compounds

NHANES

Stroke is the second leading cause of death globally and is considered a major cause of disability among different populations worldwide L Zhang, H Lu and C Yang [1]. Epidemiological studies have found that the incidence of stroke increases with age. Evidence shows that [2] the incidence rate among people under the age of 55 is gradually rising, and it doubles after the age of 55. According to the recent Global Burden of Disease Study (GBD), the global cost of stroke is estimated to exceed $891 billion (accounting for 1.12% of the global GDP) [3], and the socioeconomic burden of stroke has increased over time, becoming a significant social and economic burden [4].

Scholars have relied on demographic data, regional characteristics, and the per capita GDP of countries to create predictive models indicating that from 2020 to 2050, the committee expects stroke mortality rates to increase by 50%, with disability-adjusted life-years (DALYs) growing over the same period from 144.8 million in 2020 to 189.3 million in 2050 [5]. Stroke not only affects individual brain health, such as causing cognitive impairment, but also leads to depression [6], dementia [7], and even death. Despite this, stroke is both preventable and treatable. Therefore, there is an urgent need to further explore potential dietary strategies for the prevention and management of stroke to alleviate the global burden and promote brain health.

In low- and middle-income countries, the incidence of stroke is high. Increasing research has demonstrated that dietary differences among individuals play an important role in the occurrence and development of stroke [8]. For example, adhering to an appropriate diet, particularly the Mediterranean diet, significantly reduces the risk of various brain diseases, especially ischemic stroke [9]. Plant-based diets are associated with a lower risk of stroke [10]. Flavonoids are a class of plant-specific secondary metabolites composed of benzopyran rings, widely found in fruits, herbs, stems, cereals, nuts, vegetables, flowers, and seeds [11]. These compounds are of great interest due to their various medicinal values, including anticancer, antioxidant, and anti-inflammatory [12]. Additionally, flavonoids exhibit significant neuroprotective and cardioprotective effects, showing great promise in the treatment of ischemic stroke. The brain has high oxygen consumption but low antioxidant capacity. Oxidative stress and neuroinflammation are two pathological events that occur immediately after a stroke [13]. Currently, an increasing number of pharmacological studies in vitro and in animal models have found that flavonoids can inhibit the expression of pro-inflammatory markers such as tumor necrosis factor-α (TNF-α), IL-6, and IL-1β. They alleviate ischemic brain injury through various neuroprotective targets and mechanisms, including promoting neurogenesis and exerting anti-neuroinflammatory effects [14]. For example, baicalin can resist immune cell infiltration [15] and exert neuroprotective effects by downregulating NF-kB and LOX-1 expression and the AMPK/Nrf2 pathway [16], thereby mitigating ischemic stroke. Additionally, as stroke is closely related to aging, some studies have found that flavonoids also inhibit ROS, oxidative stress, and frailty [17]. Therefore, flavonoids interventions as a potential strategy to reduce the occurrence and progression of stroke and related complications have attracted attention.

The most common treatment for ischemic stroke is recombinant tissue plasminogen activator (rt-PA) therapy. However, due to the limited therapeutic window and secondary brain injury after reperfusion (I/R), most patients are still not eligible for this treatment. There are few clinical studies exploring the relationship between dietary flavonoid intake, including specific subtypes, and stroke, making it urgent to find new drugs for stroke treatment. Our investigation is based on data from the National Health and Nutrition Examination Survey (NHANES) in the United States. By analyzing subgroups of the stroke population and introducing Bayesian Kernel Machine Regression (BKMR), we aim to assess the combined effects of flavonols subgroups on stroke in the elderly. This study seeks to reveal the correlation and potential benefits of flavonoid compounds and their subtypes for elderly stroke patients. Our goal is to reduce the disease burden in this population and improve outcomes.

2.1 Study participants

NHANES is a series of cross-sectional surveys designed to assess the health status of the entire US population. Conducted by the National Center for Health Statistics (NCHS), NHANES includes demographic information, dietary data, various physical examination metrics, and health-related data. The study protocol was authorized by the NCHS Research Ethics Review Board, and all participants provided written informed consent.

Currently, only three NHANES cycles (2007–2008, 2009–2010, and 2017–2018) have released data on dietary flavonoids intake. The total number of participants in these surveys was 29,940. This study primarily included individuals aged 60 and older, excluding 23,563 participants under the age of 60, resulting in a sample of 6,377 individuals. Further exclusions were made for missing data: 1,366 participants with missing flavonoid compounds intake data (including flavonoids, isoflavones, anthocyanidins, flavan-3-ols, flavanones, flavones, and flavonols such as isorhamnetin, kaempferol, myricetin, and quercetin), 15 participants with missing stroke data, and 1,190 participants with missing covariate data (including age, gender, race, education level, poverty level, BMI, smoking, alcohol consumption, physical activity, HEI-2015, diabetes, hypertension, hyperlipidemia, and sleep disorders). The final study population consisted of 3,806 individuals, representing a weighted population of 44,947,393 (Fig. 1).

2.2 Assessment of dietary flavonoids intakes

In this study, the intake of flavonoids was based on the consumption of foods and beverages (including water), without considering the intake of flavonoids supplements or medications. The Flavonoid Database created by the Food Surveys Research Group (FSRG) was used to estimate flavonoids intake in the US. This database provides flavonoids values for all foods and beverages in the Food and Nutrient Database for Dietary Studies (FNDDS) versions 4.1 and 5.0, which were used to code dietary intake data and calculate nutrient intake for NHANES 2007–2008 and 2009–2010, respectively.

The FNDDS is used to code and analyze dietary intake data from the National Health and Nutrition Examination Survey, providing data files for flavonoids intake that correspond to the NHANES dietary data releases. These files contain the content (mg/100g) of 29 individual flavonoids within six classes of flavonoids in all foods and beverages. For each reported food/beverage item, the daily total intake of the 29 individual flavonoids, six classes of flavonoids (flavonoids, flavan-3-ols, flavanones, flavones, flavonols, and isoflavones), and total flavonoids consumed on Day 1 and Day 2 were included. Flavonoids intake was calculated by qualified nutritionists through interviews using food frequency questionnaires (FFQ). The intake of different flavonoids was determined by multiplying the 24-hour retrospective dietary intake type by the flavonoids content of each food. The data was then coded according to the FNDDS manual, and the flavonoid content of the coded foods was calculated. Qualified nutritionists collected dietary data, including the intake of flavonoids and other nutrients, through 24-hour food recall interviews.

2.3 Assessment of Stroke

As in previously published articles based on NHANES data, we used the medical conditions questionnaire to determine outcomes [18, 19]. Participants were asked in the health questionnaire, ‘MCQ160F: Has a doctor or other health professional ever told {you/SP} that {you/s/he}. . .had a stroke?’. Those who answered ‘yes’ were considered to have had a stroke, while those who chose other options were considered not to have had a stroke.

2.4 Study Covariates

The covariates in this study included age, gender (female or male), and race, categorized into four groups (non-Hispanic White, non-Hispanic Black, Mexican American, and Others). Education level was divided into three categories (less than 9th grade, including uneducated participants, high school graduate/GED or equivalent, some college or AA degree, and college graduate or above). The poverty-to-income ratio [20] was calculated by dividing family (or individual) income by the poverty guidelines specific to the survey year. If respondents reported an income of < $20,000 or ≥ $20,000 only, this value was not calculated. BMI, Healthy Eating Index (HEI-2015), and physical activity (include work activity, recreational activity, and walking or bicycling; if any one of these three activities is marked as ‘yes’, then physical activity is marked as ‘yes’.) Smoking status was categorized into three groups: never (smoked less than 100 cigarettes in life), former (smoked more than 100 cigarettes in life and smoke not at all now), and now (smoked more than 100 cigarettes in life and smoke some days or every day). Alcohol intake was categorized into five groups (never, former, mild, moderate, and heavy) based on self-reporting. Current drinking status was defined as ‘heavy’ if subjects reported drinking five or more drinks on ten or more days in the past 12 months. If they reported drinking five or more drinks on fewer than ten days in the past 12 months, their current drinking status was defined as ‘moderate’. If they reported drinking fewer than 12 times in their lifetime, they were defined as ‘never.’ Others were defined as ‘mild.’ Additionally, recognized stroke-related chronic disease risk factors, such as sleep disorders and hyperlipidemia diagnosed by a doctor, as well as diabetes mellitus (DM) (doctor-diagnosed, taking glucose-lowering medication, excluding pregnant women) and hypertension (including doctor-diagnosed, taking antihypertensive medication, or having abnormal blood pressure readings on three consecutive measurements with systolic ≥ 140 mmHg or diastolic ≥ 90 mmHg), are also included as covariates. Detailed information on the methods and quality of covariate determination can be obtained from the NHANES website.

2.5 Statistical analysis

All statistical analyses were performed using the R programming language (version 4.2.3). Weighted means (odds ratios (OR) and 95% confidence intervals (CI)) described continuous variables as mean ± standard deviation, and weighted percentages (OR and 95% CI) described categorical variables as percentages. A value of P < 0.05 was considered statistically significant. To assess differences between groups, weighted chi-square tests were used for categorical variables, and weighted regression models were used for continuous variables. Due to the varying selection probabilities of NHANES participants and to prevent oversampling and reduce error and bias, we adjusted for weights in the analysis. Multivariate logistic regression models were used to evaluate the relationship between flavonols intake levels and stroke in all participants, as well as the relationship between different flavonols subgroups and stroke. The log-normal distribution of flavonoids intake was considered to better conform to normality. Given the potential non-linear and non-additive effects among flavonols subgroups, we introduced BKMR to assess the joint effects of flavonols subgroups on stroke in the elderly, the influence of each flavonols subclass as part of flavonols intake, and potential interactions between different flavonols subclasses. Considering the binary outcome (stroke or no stroke), we used the probability link function for flavonol subclasses. Pearson correlation coefficients were used to group these flavonol subclasses. Due to the relatively high correlation among flavonol subclasses, we further implemented a hierarchical variable selection method, using the Markov Chain Monte Carlo (MCMC) algorithm with 50,000 iterations. We then calculated the conditional posterior inclusion probability (condPIP).

3.1 Baseline characteristics of the study population

Table 1 provides an overview of the demographic and baseline clinical characteristics of the participants. This study included 3,806 participants, among whom 331 (8.70%) were stroke patients. Among all participants, the average age was 69.39 ± 0.15 years. There were 1,918 females (50.39%), and the largest racial group was non-Hispanic white (2,138 participants, 56.17%). Notably, significant differences in demographic and baseline clinical characteristics were observed between stroke and non-stroke participants (P < 0.0001). The proportion of individuals aged over 70 was higher among stroke patients [208 (68.00%) vs. 123 (32.00%), P < 0.0001]. Stroke patients also had lower education levels (P = 0.02), lower HEI scores (P = 0.02), less physical activity (P < 0.0001), and higher prevalence rates of diabetes, hypertension, and sleep disorders (P < 0.0001). Additionally, alcohol consumption seemed to influence stroke occurrence. We also compared the dietary intake of flavonoids between stroke and non-stroke subjects. Surprisingly, the total flavonoids intake, as well as intakes of anthocyanidins, flavan-3-ols, flavones, and flavonols, were significantly higher in the non-stroke group, with the difference in flavonols being the most significant (P < 0.0001) (Table 1).

Table 1

Weighted characteristics of participants grouped by stroke status included in NHANES
Variables	Total (n = 3806)	Non-stroke (n = 3475)	Stroke (n = 331)	P-value
Age, n(%), year				< 0.0001
60–70	1909(50.158)	1786(56.265)	123(32.001)
>=70	1897(49.842)	1689(43.735)	208(67.999)
Sex, n(%)				0.563
Male	1888(49.606)	1714(43.962)	174(41.361)
Female	1918(50.394)	1761(56.038)	157(58.639)
Race, n(%)				0.784
Non-Hispanic White	2138(56.174)	1943(80.822)	195(80.883)
Non-Hispanic Black	725(19.049)	651(7.871)	74(9.143)
Mexican American	432(11.35)	403(3.978)	29(3.664)
Others	511(13.426)	478(7.329)	33(6.309)
Education, n(%)				0.016
Below high school	1067(28.035)	950(16.609)	117(24.664)
High school	926(24.33)	842(25.991)	84(29.354)
Above high school	1813(47.635)	1683(57.400)	130(45.982)
PIR, n(%)				0.234
<1	528(13.873)	463(7.513)	65(9.968)
>=1	3278(86.127)	3012(92.487)	266(90.032)
BMI, n(%), kg/m2				0.710
<25	889(23.358)	812(23.268)	77(20.719)
25–30	1403(36.863)	1279(33.987)	124(35.582)
>=30	1514(39.779)	1384(42.745)	130(43.700)
Healthy eating index, n(%)				0.018
Inadequate	1186(31.161)	1049(28.432)	137(39.877)
Average	1952(51.287)	1805(54.025)	147(46.575)
Optimal	668(17.551)	621(17.543)	47(13.549)
Physical activity, n(%)				< 0.0001
No	1425(37.441)	1243(30.509)	182(52.494)
Yes	2381(62.559)	2232(69.491)	149(47.506)
Smoking status, n(%)				0.576
Never	1837(48.266)	1705(50.769)	132(46.015)
Former	1526(40.095)	1375(40.042)	151(43.585)
Now	443(11.64)	395( 9.190)	48(10.400)
Alcohol consumption, n(%)				0.002
Never	629(16.527)	570(13.686)	59(15.293)
Former	832(21.86)	723(16.409)	109(29.574)
Mild	1636(42.985)	1520(48.762)	116(41.041)
Moderate	416(10.93)	389(13.509)	27( 7.058)
Heavy	293(7.698)	273(7.633)	20(7.033)
DM, n(%)				< 0.001
No	2871(75.434)	2664(80.669)	207(63.944)
Yes	935(24.566)	811(19.331)	124(36.056)
Hypertension, n(%)				< 0.001
No	1179(30.977)	1138(36.273)	41(16.035)
Yes	2627(69.023)	2337(63.727)	290(83.965)
Sleep disorder, n(%)				< 0.001
No	2717(71.387)	2516(69.263)	201(53.627)
Yes	1089(28.613)	959(30.737)	130(46.373)
Hyperlipidemia, n(%)				0.382
No	630(16.553)	588(14.653)	42(12.080)
Yes	3176(83.447)	2887(85.347)	289(87.920)
Total flavonoids, median(IQR), mg/d	88.430( 31.355,288.705)	92.605(32.305,298.465)	58.840(21.900,172.155)	0.002
Isoflavones, median(IQR), mg/d	0.010(0.000,0.050)	0.010(0.000,0.050)	0.010(0.000,0.035)	0.125
Anthocyanidins, median(IQR), mg/d	4.340( 0.570,21.440)	4.555(0.615,21.865)	3.210(0.275,17.430)	0.082
Flavan-3-ols, median(IQR), mg/d	20.470( 6.910,220.440)	21.215(7.075,238.275)	12.745(4.955,119.895)	0.004
Flavanones, median(IQR), mg/d	1.430( 0.125,22.295)	1.455(0.135,22.100)	0.865(0.005,24.310)	0.183
Flavones, median(IQR), mg/d	0.605(0.235,1.270)	0.615(0.235,1.290)	0.475(0.180,1.065)	0.045
Flavonols, median(IQR), mg/d	14.210( 7.725,23.595)	14.510(7.915,24.250)	10.370(6.365,16.930)	< 0.0001

Variables are presented as median (IQR) or numbers (percentages). P values are derived from the Wilcoxon test for continuous variables without normal distribution and chi-square tests for categorical variables. NHANES, National Health and Nutrition Examination Survey; BMI, body mass index; PIR, poverty income ratio; DM, diabetes mellitus; IQR, interquartile range.

3.2 Association of flavonoids and subclasses of flavonols intake levels with stroke status

We converted the intake of the six classes of flavonoids (flavonoids, isoflavones, anthocyanidins, flavan-3-ols, flavanones, flavones, and flavonols) and total flavonoid intake into categorical variables (Q1-4). Based on dietary flavonoids quartiles, what stands out in Table 2 is that participants with the highest dietary flavonols (Q4) intake showed a lower presence of stroke in the adjusted model. We found that with each unit increase in flavonols (Q4) intake, the likelihood of stroke in the elderly decreased by 61.0% (OR = 0.390, 95% CI [0.209–0.728]; P = 0.005). Both unadjusted (P < 0.0001) and adjusted (P = 0.007) models for flavonols (Q4) intake showed significant trends, consistent with the trends observed in Table 1. Therefore, we further studied the intake of flavonols subclasses (isorhamnetin, kaempferol, myricetin, and quercetin) by converting them into categorical variables (Q1-4) for model adjustment to explore their correlation with stroke (Table 3). We found that both unadjusted and adjusted models for quercetin (Q4) intake showed significant trends (P = 0.026). With each unit increase in quercetin (Q4) intake, the likelihood of stroke decreased by 52.8% (OR = 0.472, 95% CI [0.245–0.907]).

Table 2

Weighted logistic regression results of flavonoids intake levels with stroke status
Variables	Model1 OR(95%CI)	P-trend/ P-value	Model 2 OR(95%CI)	P-trend/ P-value	Model 3 OR(95%CI)	P-trend/ P-value
Total flavonoids categories		0.003		0.076		0.101
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.934(0.557,1.563)	0.789	1.093(0.603,1.981)	0.761	1.101(0.592,2.047)	0.750
Q3	0.577(0.336,0.991)	0.046	0.775(0.402,1.493)	0.431	0.764(0.385,1.516)	0.423
Q4	0.468(0.263,0.832)	0.011	0.600(0.306,1.176)	0.130	0.628(0.317,1.245)	0.172
Log-transformed total flavonoids continuous	0.804(0.705,0.916)	0.002	0.870(0.747,1.013)	0.071	0.876(0.750,1.023)	0.091
Isoflavones categories		0.061		0.518		0.707
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	1.125(0.663,1.908)	0.656	1.306(0.730,2.337)	0.353	1.446(0.825,2.535)	0.187
Q3	0.682(0.443,1.049)	0.080	0.783(0.488,1.257)	0.298	0.795(0.490,1.289)	0.334
Q4	0.735(0.455,1.187)	0.203	0.953(0.570,1.595)	0.850	1.031(0.618,1.719)	0.903
Log-transformed isoflavones continuous	0.963(0.919,1.010)	0.115	0.992(0.942,1.045)	0.762	0.998(0.948,1.050)	0.926
Anthocyanidins categories		0.128		0.855		0.904
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.747(0.381,1.465)	0.388	0.806(0.422,1.539)	0.498	0.784(0.397,1.547)	0.465
Q3	0.798(0.485,1.315)	0.368	0.925(0.528,1.622)	0.778	0.919(0.510,1.655)	0.768
Q4	0.657(0.399,1.080)	0.096	0.906(0.503,1.634)	0.734	0.917(0.488,1.724)	0.778
Log-transformed anthocyanidins continuous	0.984(0.953,1.016)	0.304	1.010(0.966,1.055)	0.651	1.010(0.963,1.058)	0.677
Flavan-3-ols categories		0.004		0.081		0.095
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.995(0.614,1.613)	0.983	1.147(0.669,1.966)	0.606	1.149(0.665,1.985)	0.603
Q3	0.710(0.443,1.138)	0.150	1.003(0.592,1.699)	0.991	0.995(0.555,1.785)	0.986
Q4	0.474(0.255,0.883)	0.020	0.595(0.302,1.173)	0.128	0.609(0.312,1.190)	0.139
Log-transformed flavan-3-ols continuous	0.918(0.860,0.980)	0.011	0.953(0.886,1.025)	0.184	0.956(0.886,1.031)	0.227
Flavanones categories		0.235		0.795		0.969
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.488(0.298,0.798)	0.005	0.573(0.343,0.957)	0.035	0.588(0.349,0.992)	0.047
Q3	0.395(0.229,0.682)	0.001	0.497(0.289,0.857)	0.014	0.522(0.298,0.913)	0.025
Q4	0.766(0.483,1.217)	0.252	0.991(0.590,1.663)	0.970	1.068(0.643,1.772)	0.791
Log-transformed flavanones continuous	0.951(0.910,0.994)	0.026	0.972(0.928,1.019)	0.229	0.977(0.932,1.023)	0.303
Flavones categories		0.120		0.993		0.970
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.871(0.518,1.466)	0.596	0.941(0.521,1.702)	0.835	0.942(0.505,1.756)	0.843
Q3	0.805(0.462,1.400)	0.434	0.972(0.523,1.807)	0.925	0.987(0.533,1.829)	0.966
Q4	0.666(0.387,1.144)	0.137	0.994(0.521,1.896)	0.986	0.974(0.501,1.893)	0.935
Log-transformed flavones continuous	0.937(0.869,1.010)	0.089	1.004(0.910,1.108)	0.933	1.004(0.905,1.114)	0.937
Flavonols categories		< 0.0001		0.007		0.007
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.794(0.505,1.249)	0.311	0.973(0.614,1.540)	0.902	0.881(0.545,1.423)	0.588
Q3	0.545(0.322,0.920)	0.024	0.742(0.409,1.346)	0.312	0.723(0.401,1.303)	0.265
Q4	0.286(0.166,0.490)	< 0.0001	0.395(0.212,0.736)	0.005	0.390(0.209,0.728)	0.005
Log-transformed flavonols continuous	0.664(0.561,0.784)	< 0.0001	0.764(0.620,0.940)	0.013	0.772(0.629,0.947)	0.015

Model 1: No covariate was adjusted. Model 2: age, sex, race, education, poverty income ratio, body mass index, smoking status, alcohol consumption, physical activity, and healthy eating index were adjusted. Model 3: age, sex, race, education, poverty income ratio, body mass index, smoking status, alcohol consumption, physical activity, healthy eating index, diabetes mellitus, hypertension, sleep disorder, and hyperlipidemia were adjusted.

OR, odds ratio; CI, confidence intervals. Q1, 1st quartile; Q2, 2nd quartile; Q3, 3rd quartile; Q4, 4th quartile. * denotes P < 0.05, ** denotes P < 0.001.

Table 3

Weighted logistic regression results of subclasses of flavonols intake levels with stroke status
Variables	Model1 OR(95%CI)	P-trend/ P-value	Model 2 OR(95%CI)	P-trend/ P-value	Model 3 OR(95%CI)	P-trend/ P-value
Isorhamnetin categories		< 0.001		0.077		0.058
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	1.251(0.733,2.135)	0.403	1.404(0.786,2.510)	0.240	1.463(0.806,2.654)	0.199
Q3	0.876(0.568,1.351)	0.541	1.154(0.723,1.842)	0.535	1.115(0.692,1.797)	0.640
Q4	0.447(0.256,0.782)	0.006	0.604(0.324,1.126)	0.108	0.611(0.336,1.112)	0.102
Log-transformed isorhamnetin continuous	0.939(0.903,0.976)	0.002	0.971(0.930,1.015)	0.183	0.969(0.931,1.010)	0.128
Kaempferol categories		< 0.001		0.016		0.028
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	1.252(0.860,1.823)	0.235	1.394(0.932,2.086)	0.102	1.482(0.985,2.231)	0.058
Q3	0.739(0.448,1.220)	0.231	0.909(0.547,1.511)	0.703	0.937(0.570,1.541)	0.789
Q4	0.410(0.236,0.712)	0.002	0.526(0.280,0.987)	0.046	0.576(0.309,1.072)	0.079
Log-transformed kaempferol continuous	0.831(0.754,0.916)	< 0.001	0.894(0.804,0.993)	0.038	0.913(0.820,1.017)	0.094
Myricetin categories		0.002		0.067		0.076
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.850(0.536,1.347)	0.480	0.995(0.643,1.539)	0.982	0.972(0.616,1.535)	0.899
Q3	0.617(0.359,1.062)	0.080	0.783(0.438,1.399)	0.394	0.730(0.396,1.347)	0.298
Q4	0.469(0.278,0.790)	0.005	0.614(0.353,1.068)	0.082	0.623(0.354,1.096)	0.096
Log-transformed myricetin continuous	0.789(0.717,0.869)	< 0.0001	0.831(0.735,0.940)	0.005	0.835(0.739,0.945)	0.006
Quercetin categories		< 0.001		0.027		0.022
Q1	1.00[Reference]		1.00[Reference]		1.00[Reference]
Q2	0.801(0.519,1.237)	0.310	1.041(0.685,1.581)	0.846	0.980(0.644,1.490)	0.920
Q3	0.632(0.384,1.041)	0.070	0.851(0.491,1.474)	0.551	0.797(0.466,1.362)	0.388
Q4	0.332(0.185,0.594)	< 0.001	0.480(0.250,0.920)	0.029	0.472(0.245,0.907)	0.026
Log-transformed quercetin continuous	0.683(0.577,0.809)	< 0.0001	0.781(0.631,0.968)	0.025	0.784(0.637,0.966)	0.024

Model 1: no covariate was adjusted. Model 2: age, sex, race, education, poverty income ratio, body mass index, smoking status, alcohol consumption, physical activity, and healthy eating index were adjusted. Model 3: age, sex, race, education, poverty income ratio, body mass index, smoking status, alcohol consumption, physical activity, healthy eating index, diabetes mellitus, hypertension, sleep disorder, and hyperlipidemia were adjusted.

OR, odds ratio; CI, confidence intervals. Q1, 1st quartile; Q2, 2nd quartile; Q3, 3rd quartile; Q4, 4th quartile. * denotes P < 0.05, ** denotes P < 0.001.

3.3 Subgroup analyses

To further observe whether the conclusions hold true across different subgroups, we conducted a subgroup analysis to study the correlation between dietary flavonols intake and stroke in various populations (Fig. 2). The variables registered in Model 3 were retained in this analysis, with the population stratified by age, gender, race, education, poverty, BMI, HEI, physical activity, alcohol consumption, smoking, and different disease conditions. Considering the potential differences among these populations and specific environments, we also tested their interactions.

Notably, the subgroup analysis showed that the correlation between flavonols intake and the occurrence of stroke was stable and not significantly affected by gender, age, education level, race/ethnicity, hypertension, smoking status, or diabetes status. However, age and poverty level were effect modifiers in the relationship between flavonols intake and stroke. For age, the OR was 0.672 (95% CI: [0.498–0.907]) for elderly individuals aged 60–70, and 0.815 (95% CI: [0.651–1.021]) for those over 70 years old (P for interaction = 0.038). For poverty level, the OR was 1.355 (95% CI: [0.888–2.068]) for individuals with a PIR < 1, and 0.708 (95% CI: [0.562–0.891]) for those with a PIR ≥ 1 (P for interaction = 0.002) (Fig. 2).Additionally, we found that in individuals aged ≥ 70, apart from non-Hispanic Whites, those with less than a college education, PIR < 1, abnormal BMI, abnormal HEI, never smokers, those without physical exercise, those with a smoking history, non-heavy drinkers, individuals with DM, without hypertension, with sleep disorders, and without hyperlipidemia, the association between flavonols intake and stroke remained stable.

We further explored the association between dietary flavonols subclasses intake levels and stroke status, stratified by group characteristics. We observed similar results. BMI and hyperlipidemia were two significant influencing factors in the relationship between the flavonols subclass isorhamnetin and stroke prevalence (P for interaction = 0.040 and 0.014, respectively) (Supplementary Figure S1). PIR was a significant influencing factor in the relationship between the flavonols subclass kaempferol and stroke prevalence (P for interaction = 0.01) (Supplementary Figure S2). Age and hypertension were two significant influencing factors in the relationship between the flavonols subclass myricetin and stroke prevalence (P for interaction = 0.005 and 0.008, respectively) (Supplementary Figure S3). Lastly, we found that age, BMI, and PIR were three significant influencing factors in the relationship between the flavonols subclass quercetin and stroke prevalence (P for interaction = 0.024, 0.008, and 0.001, respectively) (Supplementary Figure S4).

3.4 RCS analysis

We used RCS to explore the relationship between flavonols and their four subclasses of stroke. There was a ‘U’-shaped non-linear relationship between log-transformed flavonols and stroke (P nonlinear = 0.0025) (Fig. 3a), with a turning point at 1.775. This suggests that, beyond the saturation point, higher flavonols intake provides additional stroke protection. Similarly, the flavonol subclasses quercetin and kaempferol also exhibited a ‘U’-shaped non-linear relationship with stroke. For quercetin, as shown in Fig. 3e, P nonlinear = 0.0048, with a turning point at 1.301. For kaempferol, as shown in Fig. 3c, P nonlinear = 0.0003, with a turning point at -0.015. In contrast, the relationship between stroke and myricetin was linear (P for overall = 0.0003, P nonlinear > 0.05), indicating that higher intake of myricetin consistently provided more protection against stroke (Fig. 3d). The relationship between stroke and isorhamnetin was not significant (Fig. 3b).

3.5 Overall impact of dietary flavonols subclass intake on stroke

Previous studies did not consider the relationship between the intake of flavonols subclasses and stroke. Figure 4 shows the Pearson correlation analysis of the relationships among different flavonols subclasses.

BKMR was used to estimate the overall effect of flavonols intake levels on the risk of stroke status. We evaluated changes in the stroke outcome by comparing the four flavonols subclasses fixed at different percentiles to their median values, ultimately finding that flavonols intake was inversely correlated with stroke outcomes in the elderly (Fig. 5), which further supports our previous findings. When the other flavonols subclasses were fixed at the 25th, 50th, and 75th percentiles, the assessment of the impact of individual flavonols subclasses on stroke outcomes revealed that as the intake of myricetin increased, the risk of stroke decreased (Supplementary Figure S5).

We further explored the nonlinear relationships between the exposure of each flavonol subclass and stroke outcomes using univariate exposure-response curves (Fig. 6). When the levels of the other three flavonol subclass variables were fixed at their median, the inhibitory effect of myricetin on stroke outcomes in the elderly gradually strengthened, showing a significant inverse relationship until reaching a turning point. The other three flavonol subclasses exhibited a “U”-shaped relationship, consistent with the results of RCS.

Our investigation covered 3,806 eligible participants diagnosed with stroke. By comparing the dietary intake of flavonoids and their subclasses between stroke and non-stroke subjects, we found, surprisingly, that the total flavonoids intake, as well as intakes of anthocyanidins, flavan-3-ols, flavones, and flavonols, were significantly higher in the non-stroke group, with flavonols showing the most significant difference (P < 0.0001). Our study demonstrated a negative correlation between the subclasses of dietary flavonoids intake and stroke. In fact, some pharmacological studies have shown that dietary flavonols subclasses such as fisetin [21], quercetin [22, 23], chalcones [24], and rutin [25] have neuroprotective effects, which is consistent with our findings. Additionally, an increasing number of prospective cohort studies have shown that consuming more flavonoid-rich fruits, particularly citrus fruits, strawberries, lemons, and grapes, is associated with a lower risk of stroke [26, 27]. Therefore, combining our research findings, flavonoids may have great potential in addressing the subsequent inflammatory response and oxidative stress during reperfusion, which can cause secondary tissue damage [28]. Additionally, it is necessary to further identify which flavonoid in the subclass play a crucial role in clinically preventing stroke.

After adjusting for potential confounding factors, higher quartiles of flavonols intake were associated with lower stroke prevalence. For every unit increase in flavonols intake (Q4), the likelihood of stroke in the elderly decreased by 61% (OR = 0.390, 95% CI [0.209–0.728]; P = 0.005). Additionally, this association remained stable among individuals aged 70 and older, except for non-Hispanic Whites, those with less than a college education, PIR < 1, abnormal BMI, abnormal HEI, never smokers, those without physical exercise, those with a smoking history, non-heavy drinkers, individuals with DM, without hypertension, with sleep disorders, and without hyperlipidemia. Similar results were observed for flavonols subclasses. Participants with the highest quartile of quercetin intake (Q4) showed a lower presence of stroke in the adjusted model. For every unit increase in quercetin intake (Q4), the likelihood of stroke in the elderly decreased by 52.8% (OR = 0.472, 95% CI [0.245–0.907]). This is consistent with previous studies, which have shown that quercetin can inhibit the formation of lipid peroxides through the Nrf2/HO-1 signaling pathway and has neuroprotective potential by participating in the regulation of iron concentration [29]. Additionally, animal I/R models have found that flavonols subclasses can improve synaptic plasticity by preventing Aβ aggregation, inhibiting tau hyperphosphorylation, and reducing neuroinflammation and oxidative stress [30].

Notably, to further examine whether the conclusions are consistent across different subgroups, our subgroup analysis revealed that age and poverty level are effect modifiers in the relationship between flavonols intake and stroke. The protective effect of flavonols was most pronounced in elderly individuals aged 60–70. Additionally, the stroke incidence increased by 35.5% (95% CI: [0.888–2.068]) in individuals with a PIR < 1, whereas it decreased by 29.2% (95% CI: [0.562–0.891]) in those with a PIR ≥ 1, indicating a negative correlation between income and stroke incidence. Cognitive function post-stroke varies significantly among individuals, potentially because low income is perceived as a stressful event, which affects the microstructural integrity of the entire brain. This stress impacts brain plasticity through neurobiological pathways, leading to unhealthy behaviors such as smoking and a poor diet. In contrast, high-income individuals typically have healthier lifestyles, including more physical exercise, which can enhance brain health by increasing neurotrophic factors and brain plasticity [31]. Although this hypothesis remains controversial, our findings support the idea that higher income can enable elderly individuals aged 60–70 to benefit from the stroke-preventive effects of dietary flavonols.

We further explored the association between dietary intake levels of flavonols subclasses (isorhamnetin, kaempferol, myricetin, and quercetin) and stroke status, stratified by population characteristics. We observed that, in addition to age and PIR, hyperlipidemia and BMI are the most common significant influencing factors in the relationship between flavonols subclasses and stroke prevalence. This suggests that flavonols may reduce the incidence of stroke in the elderly by modulating lipid metabolism levels. Current studies indicate that citrus flavonoids are a good source for treating hyperlipidemia [32], which have shown beneficial effects on obesity [33]. Specifically, quercetin can regulate the PI3K/Akt/NF-κB signaling pathway to promote microglia/macrophage M2 polarization, thereby treating cerebral ischemia-reperfusion injury [34]. Although the specific mechanisms require further in vivo and in vitro experimental validation.

Additionally, there is a ‘U’-shaped non-linear relationship between flavonols intake and stroke, as well as between its subclasses quercetin and kaempferol and stroke. This may indicate that reaching a certain threshold of flavonols intake is necessary to confer additional protective effects against stroke. This finding further emphasizes and validates the role of flavonols in the pathogenesis of stroke. Moreover, our study assessed the overall impact of dietary flavonols subclasses on stroke in the elderly and the interrelationship among these subclasses. Our results consistently demonstrate a negative combined effect of flavonols subclasses intake on the risk of stroke in the elderly. Specifically, we observed a potential dose-response relationship when evaluating the impact of individual flavonols subclasses on stroke outcomes. Unlike the other three flavonols subclasses, increasing myricetin intake is significantly linearly and inversely correlated with reduced stroke risk. This suggests that myricetin in the diet might be strongly associated with a substantial reduction in stroke prevalence. Brain damage following ischemia is a neurodegenerative condition. Myricetin has been shown to possess strong antioxidant and anti-inflammatory properties [35], which can mitigate apoptosis following ischemic stroke by reducing excitotoxicity, oxidative stress, and inflammation-induced apoptosis [36] This indicates its potential value in treating neurodegenerative diseases involving protein misfolding, including post-ischemic neurodegenerative disorders.

Our findings not only emphasize the importance of increasing dietary flavonoids intake in stroke patients but also explore the protective effects of flavonols and their subclasses. The study’s significant advantage lies in employing multiple statistical methods to examine flavonols and their subclasses, suggesting potential intervention strategies, and identifying differences among flavonoid subclasses. The use of machine learning on large-scale data further enhances the robustness of the statistical results. However, there are limitations. Long-term follow-up data attrition in the database may introduce selection bias. Diagnosis based on questionnaire responses cannot distinguish between ischemic and hemorrhagic stroke, nor exclude related causes such as cerebral aneurysms, genetic factors, and trauma, affecting diagnostic accuracy. Additionally, the elderly patient sample results in an insufficient sample size, limiting the ability to conduct sensitivity analyses to evaluate the impact of flavonoids on stroke mortality.

In conclusion, maintaining increased dietary flavonoids intake can significantly reduce the prevalence of stroke in the US population, offering potential benefits to stroke patients. In this study, flavonols and their subclasses played a crucial role, showing independent or combined protective effects against disease prevalence, particularly among elderly individuals aged 60–70 and those with higher income levels. We strongly advocate for personalized management and treatment of stroke patients. Additionally, we recommend that future research adopt more rigorous experimental designs and utilize comprehensive data sources to further validate and expand these findings.

In conclusion, dietary intake of flavonoids is negatively correlated with stroke incidence, particularly flavonols and their subclasses such as myricetin, can significantly reduce the prevalence of stroke among the U.S. elderly population. To establish a definitive relationship, further research involving cell, animal, and human studies is necessary.

Ethics statement

Research involving human participants was reviewed and approved by NHANES, and participants provided written informed consent for their involvement.

Funding

This work was funded by the TCM Scientific Research Project of Shanghai National Health Commission (No. 2022QN017); Shanghai Municipal Commission of Health (the "Flagship" Department Construction Project of Integrated Traditional Chinese and Western Medicine for Geriatrics 2024); The Science and Technology Commission of Shanghai Municipality (No.17dz2307500).

Author Contribution

XC conceived and designed the study. XC, TP, and JL were responsible for data analysis and retrieval, contributing to the initial data analysis and interpretation. XC drafted the initial manuscript. TP and LH revised the manuscript. All authors contributed to the article and approved the submitted version.

Acknowledgement

We thank the respondents in the NHANES database for providing data for this study and the Medical Science Data Center of Fudan University for their support. However, the content of this paper does not represent any official position of NHANES. The authors had full access to the data in the study and take responsibility for the accuracy of the data analysis.

Data Availability

Data is provided within the manuscript or supplementary information files. All data from this study can be provided by the first author upon request.

Zhang L, Lu H, Yang C. Global, regional, and national burden of stroke from 1990 to 2019: A temporal trend analysis based on the Global Burden of Disease Study 2019. Int j stroke. 2024. https://doi.org/10.1177/17474930241246955.
Scott CA, Li L, Rothwell PM. Diverging Temporal Trends in Stroke Incidence in Younger vs Older People: A Systematic Review and Meta-analysis. Jama neurol. 2022;79:1036–48.
Feigin VL, Brainin M, Norrving B, et al. World Stroke Organization (WSO): Global Stroke Fact Sheet 2022. Int j stroke. 2022;17:18–29.
Kuriakose D, Xiao Z. Pathophysiology and Treatment of Stroke: Present Status and Future Perspectives. Int J Mol Sci. 2020;21:7609.
Feigin VL, Owolabi MO. Pragmatic solutions to reduce the global burden of stroke: a World Stroke Organization-Lancet Neurology Commission. Lancet neurol. 2023;22:1160–206.
Jeong S, Chokkalla AK, Davis CK, et al. Post-stroke depression: epigenetic and epitranscriptomic modifications and their interplay with gut microbiota. Mol psychiatr. 2023;28:4044–55.
Rost NS, Brodtmann A, Pase MP, et al. Post-Stroke Cognitive Impairment and Dementia. Circ res. 2022;130:1252–71.
Maalouf E, Hallit S, Salameh P, et al. Eating Behaviors, Lifestyle, and Ischemic Stroke: A Lebanese Case-Control Study. Int J Environ Res Public Health. 2023;20:1487.
Cherian L, Wang Y, Fakuda K, et al. Mediterranean-Dash Intervention for Neurodegenerative Delay (MIND) Diet Slows Cognitive Decline After Stroke. J Prev Alzheimers Dis. 2019;6:267–73.
Okekunle AP, Asowata O, Akpa OM, et al. Dietary patterns associated with stroke among West Africans: A case-control study. Int j stroke. 2023;18:193–200.
Chen S, Wang X, Cheng Y, et al. A Review of Classification, Biosynthesis, Biological Activities and Potential Applications of Flavonoids. Molecules. 2023;28:4982.
Ullah A, Munir S, Badshah SL, et al. Important Flavonoids and Their Role as a Therapeutic Agent. Molecules. 2020;25:5243.
Teleanu DM, Niculescu AG, Lungu II, et al. An Overview of Oxidative Stress, Neuroinflammation, and Neurodegenerative Diseases. Int J Mol Sci. 2022;23:5938.
Lu W, Chen Z, Wen J. Flavonoids and ischemic stroke-induced neuroinflammation: Focus on the glial cells. Biomed pharmacother. 2024. https://doi.org/10.1016/j.biopha.2023.115847.
Hao Z, Zhang Z, Zhao Y, et al. Baicalin Reduces Immune Cell Infiltration by Inhibiting Inflammation and Protecting Tight Junctions in Ischemic Stroke Injury. Am j Chin med. 2023;51:355–72.
Yuan Y, Men W, Shan X, et al. Baicalein exerts neuroprotective effect against ischaemic/reperfusion injury via alteration of NF-kB and LOX and AMPK/Nrf2 pathway. Inflammopharmacology. 2020;28:1327–41.
Oei S, Millar CL, Nguyen Lily TN, et al. Higher intake of dietary flavonols, specifically dietary quercetin, is associated with lower odds of frailty onset over 12 years of follow-up among adults in the Framingham Heart Study. Am j clin nutr. 2023;118:27–33.
Yan S, Sha S, Li S, et al. Association between hypertension and stroke in US adults in the National Health and Nutrition Examination Surveys (NHANES) 2007 to 2018. Postgrad med. 2023;135:187–94.
Dong W, Yang Z. Association of Dietary Fiber Intake With Myocardial Infarction and Stroke Events in US Adults: A Cross-Sectional Study of NHANES 2011–2018. Front Nutr. 2022;9:936926.
Balestrino R, Schapira AHV. Parkinson disease. Eur j neurol. 2020;27:27–42.
Maher P. Preventing and Treating Neurological Disorders with the Flavonol Fisetin. Brain Plast. 2021;6:155–66.
Guo C, Wang WJ, Liao YC et al. (2022) Effect and Mechanisms of Quercetin for Experimental Focal Cerebral Ischemia: A Systematic Review and Meta-Analysis. Oxid med cell longev. https://doi.org/10.1155/2022/9749461
Zhang L, Ma J, Yang F, et al. Neuroprotective Effects of Quercetin on Ischemic Stroke: A Literature Review. Front Pharmacol. 2022. https://doi.org/10.3389/fphar.2022.854249.
Adelusi TI, Akinbolaji GR, Yin X, et al. Neurotrophic, anti-neuroinflammatory, and redox balance mechanisms of chalcones. Eur j pharmacol. 2021. https://doi.org/10.1016/j.ejphar.2020.173695.
Rana AK, Kumar R, Shukla DN, et al. Lithium co-administration with rutin improves post-stroke neurological outcomes via suppressing Gsk-3β activity in a rat model. Free radical bio med. 2023;207:107–19.
Gao Q, Dong JY, Cui R, et al. Consumption of flavonoid-rich fruits, flavonoids from fruits and stroke risk: a prospective cohort study. Brit j nutr. 2021;126:1717–24.
Li XQ, Wang C, Yang T, et al. A meta-analysis of prospective cohort studies of flavonoid subclasses and stroke risk. Phytother res. 2022;36:1103–14.
Ucar BI, Ucar G, Saha S, et al. Pharmacological Protection against Ischemia-Reperfusion Injury by Regulating the Nrf2-Keap1-ARE Signaling Pathway. Antioxid (Basel). 2021;10:823.
Peng C, Ai Q, Zhao F, et al. Quercetin attenuates cerebral ischemic injury by inhibiting ferroptosis via Nrf2/HO-1 signaling pathway. Eur j pharmacol. 2024. https://doi.org/10.1016/j.ejphar.2023.176264.
Ren Y, Qu S. Constituent isoflavones of Puerariae radix as a potential neuroprotector in cognitive impairment: Evidence from preclinical studies. Ageing res rev. 2023. https://doi.org/10.1016/j.arr.2023.102040.
Tian J, Wang Y, Guo L, et al. Association of Income with Post-Stroke Cognition and the Underlying Neuroanatomical Mechanism. Brain Sci. 2023;13:363.
Ling Y, Shi Z, Yang X et al. (2020) Hypolipidemic effect of pure total flavonoids from peel of Citrus (PTFC) on hamsters of hyperlipidemia and its potential mechanism. Exp gerontol. https://doi.org/10.1016/j.exger.2019.110786
Xiao PT, Liu SY, Kuang YJ, et al. Network pharmacology analysis and experimental validation to explore the mechanism of sea buckthorn flavonoids on hyperlipidemia. J ethnopharmacol. 2021. https://doi.org/10.1016/j.jep.2020.113380.
Li L, Jiang W, Yu B et al. (2023) Quercetin improves cerebral ischemia/reperfusion injury by promoting microglia/macrophages M2 polarization via regulating PI3K/Akt/NF-κB signaling pathway. Biomed pharmacother. https://doi.org/10.1016/j.biopha.2023.115653
Pluta R, Januszewski S, Czuczwar SJ. Myricetin as a Promising Molecule for the Treatment of Post-Ischemic Brain Neurodegeneration. Nutrients. 2021;13:343.
Zhang L, Zhou T, Ji Q, et al. Myricetin improves apoptosis after ischemic stroke via inhibiting MAPK-ERK pathway. Mol biol rep. 2023;50:2545–57.

No competing interests reported.

Supplementarymaterial.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

The Relationship Between Flavonols Intake and Stroke in the Elderly: NHANES (2007-2010 and 2017-2018)

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1 Introduction

2 Materials and methods

2.1 Study participants

2.2 Assessment of dietary flavonoids intakes

2.3 Assessment of Stroke

2.4 Study Covariates

2.5 Statistical analysis

3 Results

3.1 Baseline characteristics of the study population

3.2 Association of flavonoids and subclasses of flavonols intake levels with stroke status

3.3 Subgroup analyses

3.4 RCS analysis

3.5 Overall impact of dietary flavonols subclass intake on stroke

4 Discussion

Conclusion

Declarations

Ethics statement

Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1