Introduction

Stroke is the second leading cause of death worldwide and the third most common cause of disability, posing significant public health and socioeconomic challenges during treatment and post-stroke rehabilitation efforts1,2,3. The two primary types of stroke are ischemic stroke (IS) and hemorrhagic stroke (HS)4. Lifestyle modifications, particularly dietary changes and physical activity, are among the most effective strategies for preventing and reducing stroke risk5,6,7,8. Therefore, investigating the impact of specific nutrients on stroke risk may provide valuable insights into prevention and management strategies9,10. Recently, the role of dietary zinc intake in neurological diseases has drawn attention11,12,13.

Zinc is an essential trace mineral with anti-inflammatory and antioxidant properties, playing a critical role in maintaining membrane stability, cellular metabolism, cell proliferation and transformation, immune responses, and oxidative stress regulation14,15,16,17. Zinc is also necessary for maintaining the typical physiological processes in humans. It plays a critical role as a cofactor for antioxidant enzymes. Zinc deficiency has been linked to several chronic conditions, such as seizures, Alzheimer’s disease, diabetes, hypertension, depression, and cardiovascular diseases (CVD)18,19,20,21. Studies have reported that stroke patients tend to have lower serum zinc levels than healthy individuals and that zinc supplementation may aid in neurological recovery following a stroke22,23,24,25. However, these studies primarily focused on serum levels in stroke patients rather than examining dietary zinc intake in the general population.

In this study, we conducted a large cross-sectional analysis using National Health and Nutrition Examination Survey (NHANES) data to evaluate the potential relationship between dietary zinc consumption and stroke prevalence. We hypothesized that individuals with a history of stroke would have lower dietary zinc intake compared to the general population.

Methods

Study population

The NHANES dataset includes several cross-sectional continuous, stratified probability surveys of noninstitutionalized Americans26. This database has been maintained by the Centers for Disease Control and Prevention’s National Center for Health Statistics, which served as the source of data for the evaluation of the dietary and physical health of Americans in this study27,28. Through visits to homes and mobile examination centers (MEC), the NHANES collected relevant data for tests, meetings, and questionnaires for the assessment of the nutrition and health aspects of the study population28. Individuals participating in four 2-year NHANES study cycles from 2013 to 2020 provided the data for use in the present research. The study participants were adults of age > 18 years who had completed an interview earlier. Those with incomplete information about stroke, variables, or dietary zinc intake were excluded from the study.

Each protocol employed in the National Health and Nutrition Examination System research was authorized by the National Center for Health Statistics’ ethical review committee. In addition, before participating in the NHANES, each participant provided written informed consent.

Assessment of stroke

Self-reports were utilized as a part of the NHANES questionnaire to determine what constituted a stroke. As described elsewhere, the Medical Condition Questionnaire (MCQ) was used to detect stroke29. Previous research employing NHANES data employed self-reported stroke, and the findings of several studies demonstrated the validity of the self-reported evaluation method30,31,32. The participants were deemed to have experienced stroke if they responded with a “yes” to the question, “Has a doctor or other health professional ever told you that you had a stroke?”

Assessment of dietary zinc intake

Information on dietary zinc consumption was extracted from the NHANES nutrition investigation through 24-h dietary recalls. The NHANES dietary data includes zinc variables for the first and second 24-h recall. Therefore, we used the average zinc intake from the two 24-h recalls for the assessments. We also used the dietary supplement questionnaire to collect information about the intake of zinc from supplements. Although 24-h recalls are frequently used in dietary assessment, the intake on a single day is a poor estimator of long-term usual intake when compared with statistical modeling33. The zinc content of each daily consumable that was offered was added to determine each person’s daily zinc consumption. The comprehensive automated memory system meticulously created standardized questions and answer options specific to each food item. The Computer-Assisted Dietary Interview (CADI) System was used to accurately assess the nutritional content based on each person’s food and drink intake with reference to the Agriculture Department’s Automated Multiple Pass Method (AMPM)34,35. The national cancer institute (NCI) method was widely used for the estimation of routine dietary intake based on 24-h dietary intake data36. The zinc variables for the first and second 24-h recall, and the intake of zinc from supplements were used in this study. The dietary zinc intake levels were divided into quartiles (Q1–Q4) with values of ≤ 6.08 mg/day, 6.09–8.83 mg/day, 8.84–13.02 mg/day, and ≥ 13.03 mg/day.

Assessment of covariates

Age, gender, marital status, race/ethnicity, education level, family income, body mass index (BMI), smoking history, energy consumption, coronary heart disease, diabetes, hypertension, hypercholesteremia, antihypertensive medications, preventive aspirin, and lipid-lowering medications were among the many potential covariates that were evaluated, according to the literature32,37. Mexican Americans, non-Hispanic Whites, non-Hispanic Blacks, and other races were the categories for evaluated race and ethnicity. There were three categories for marital status: married, living with a partner, and living alone. Less than 9 years, 9–12 years, and > 12 years were the range for educational achievement. The poverty income ratio (PIR) divided the family income into three categories, according to the American government report, as follows: low (PIR < 1.3), medium (PIR > 1.3–3.5), and high (PIR > 3.5)38. To determine BMI, a standardized technique based on height and weight, was applied. The smoking status was classified as never smokers, current smokers, and former smokers based on the classifications from previous research39. Based on the questionnaire’s question about whether the doctor had previously been notified of the illness and drugs, the prior disease (i.e., hypertension, hypercholesteremia, diabetes, coronary heart disease, antihypertensive medications, preventative aspirin, and lipid-lowing medications) was determined.

Statistical analyses

All analyses were conducted using R version 4.3.3 and Free Statistics version 1.92. Categorical data were presented as counts and frequencies, while continuous variables were expressed as means and standard deviations (SDs). The relationships between dietary zinc intake and stroke were assessed using odds ratios (ORs) and 95% confidence intervals (CIs), calculated through multivariable logistic regression models. Zinc consumption in food was categorized into quartiles for comparison. Three multivariable logistic regression models were developed, as shown in Table 3: models 1, 2, and 3. Model 1 was adjusted for sociodemographic variables (age, sex, marital status, race/ethnicity, education level, and family income). Model 2 was adjusted for the same sociodemographic variables as Model 1, in addition to smoking status, BMI, and energy consumption. Model 3 was adjusted for all variables in Model 2, in addition to coronary heart disease, diabetes, hypertension, hypercholesteremia, antihypertensive drug use, preventive aspirin use, and lipid-lowing drug use. A restricted cubic spline (RCS) analysis was performed to assess potential non-linear associations, adjusted based on Model 3 variables. Additionally, sensitivity and stratified analyses were conducted across different subgroups to verify the robustness of the findings. A P value of < 0.05 was considered statistically significant.

Results

Study population

A total of 35,706 adults participated in the NHANES 2013–2020 surveys. After the inclusion criteria were applied, 13,908 participants under the age of 18 years old, 12,663 participants were excluded due to missing data on calorie intake, smoking, BMI, marital status, education, and family income, and 6853 participants were excluded due to incomplete data on hypertension, hypercholesteremia, diabetes, coronary heart disease, medication use (antihypertensive drugs, lipid-lowering drugs, preventive aspirin), dietary zinc intake, and stroke history. As a result, 2642 participants met the inclusion criteria for this cross-sectional study. A flowchart summarizing the inclusion process is presented in Fig. 1.

Fig. 1
figure 1

Inclusion and exclusion flowchart of the study participants.

Full size image

Participant characteristics

Table 1 summarizes the characteristics of the study participants, categorized by quartiles of dietary zinc intake. The study sample included 1438 women (53.7%), with a mean age of 62.8 years. Higher dietary zinc intake was associated with younger age groups, a larger proportion of men, non-Hispanic white ethnicity, married or cohabiting, higher educational attainment, middle-class family incomes, lower smoking rates, higher energy consumption, and higher BMI. Supplementary Table 1 provides a comparison of key characteristics between included and excluded participants.

Table 1 Participants characteristics by categories of dietary zinc intake.
Full size table

Relationship between dietary zinc intake and stroke

Univariate analysis identified significant associations between stroke and several factors, including age, non-Hispanic black ethnicity, marital status, higher family income, smoking history, BMI, hypercholesteremia, coronary heart disease, diabetes, medication use (antihypertensive drugs, preventive aspirin, and lipid-lowering drugs). Detailed results are presented in Table 2.

Table 2 Association of covariates and stroke risk.
Full size table

Multivariable analysis observed an inverse association between dietary zinc intake and stroke. After adjusting for potential confounders, participants in Q2 (6.09–8.83 mg/day) had a significantly lower odds of stroke (adjusted OR: 0.64, 95% CI: 0.41–0.99, p = 0.044) compared to those in Q1 (≤ 6.08 mg/day) (Table 3). Furthermore, RCS analysis revealed an L-shaped relationship between dietary zinc intake and stroke odds (p = 0.041), as illustrated in Fig. 2.

Table 3 Association between dietary zinc intake and stroke.
Full size table
Fig. 2
figure 2

Association between dietary zinc intake and stroke.

Full size image

The analytical threshold analysis revealed that the OR for stroke prevalence was 0.858 (95% CI: 0.743–0.99, p = 0.037) for individuals consuming less than 8.82 mg by zinc per day. This suggests that for every additional 1 mg of dietary zinc intake per day, the odds of stroke are lower by 14.2% than before. However, when daily zinc intake reached ≥ 8.82 mg/day, no further correlation was observed between dietary zinc consumption and stroke odds (Table 4). This finding indicates that beyond a certain threshold, higher zinc intake through food is no longer associated with lower odds of stroke.

Table 4 Threshold effect analysis of the relationship of dietary zinc intake with stroke.
Full size table

Subgroup analyses

Subgroup analyses were conducted to examine the impact of various factors (sex, age, marital status, education level, family income, and BMI) on the relationship between dietary zinc intake and stroke. As shown in Table 5, no significant associations or interaction effects were found among these subgroups.

Table 5 Subgroup analysis by dietary zinc intake.
Full size table

Discussion

This large-scale cross-sectional study of adult Americans revealed an L-shaped relationship between dietary zinc intake and stroke prevalence. This inflection point analysis showed that when dietary zinc intake was less than 8.82 mg/day, stroke prevalence was lower with higher zinc intake. However, for individuals consuming ≥ 8.82 mg/day, the odds of stroke remained unchanged, even with higher zinc consumption. Subgroup analyses confirmed the robustness of this association. To our knowledge, this is the first cross-sectional study to examine the relationship between dietary zinc intake and stroke in the NHANES database.

Research has suggested a link between zinc levels and stroke prevalence. In stroke patients with low zinc intake, restoring zinc levels to normal may improve neurological recovery23. A nested case-control study in hypertensive individuals found a significant inverse relationship between plasma zinc levels and the prevalence of first-time HS22. Similarly, patients with IS were found to have significantly lower blood zinc concentrations than healthy controls40. However, a meta-analysis of 12 case-control studies reported that elevated blood zinc levels might be associated with an increased prevalence of IS, while no significant association was observed between zinc levels and HS risk41. Interestingly, none of these studies assessed the relationship between dietary zinc intake and stroke prevalence in the general population. Instead, most studies have focused on smaller sample sizes or shorter study durations. Our study, leveraging the NHANES dataset, provides a unique opportunity to examine the dose-response relationship between dietary zinc intake and stroke, while adjusting for multiple confounding variables and conducting stratified analyses.

Our study found an L-shaped connection between the odds of stroke and between dietary zinc intake. Specifically, within a certain range, higher dietary zinc intake is associated with lower stroke prevalence. Individuals with lower zinc intake (< 8.82 mg/day) appear to benefit the most from increasing their dietary zinc consumption. However, for individuals consuming ≥ 8.82 mg/day, the prevalence of stroke does not decrease further with additional zinc intake. Currently, the recommended daily zinc intake is 11 mg for men and 8 mg for women, with higher requirements for pregnant women and children42. However, our study indicates that many adult Americans, particularly women, fail to meet their daily zinc intake requirements. Excessive zinc intake can lead to adverse effects, including headaches, loss of appetite, nausea, vomiting, and diarrhea. Prolonged high zinc consumption may interfere with the absorption and metabolism of essential trace elements, such as iron, copper, and calcium, potentially leading to iron-deficiency anemia, copper deficiency, and calcium deficiency, weakened immune function, reduced high-density lipoprotein cholesterol levels, and neurological complications.

According to the data, zinc plays a crucial role in both the pathophysiology and normal functioning of the brain43. However, the fundamental mechanisms linking zinc consumption to stroke remain unclear. Research suggests that zinc is associated with both IS and HS22,23. Zinc has anti-inflammatory and antioxidant properties that protect vascular cells from free radical damage and inflammatory injury. Additionally, it plays a role in regulating cytokine production44. Atherosclerosis, the primary cause of CVD, including peripheral artery disease and stroke, is influenced by zinc through its interactions with immune cells, vascular smooth muscle cells, and endothelial cells45. Zinc is essential for the activation and transformation of monocytes into macrophages, the formation of foam cells via oxidized low-density lipoprotein (LDL) uptake in atherosclerosis, monocyte-endothelial adhesion, and diapedesis16. Clinical trials have shown that zinc supplementation can lower serum LDL cholesterol levels, regulate blood glucose levels, and reduce blood pressure44. Furthermore, zinc plays both beneficial and detrimental roles in blood pressure regulation, glucose metabolism, lipid metabolism, and other atherosclerosis-related risk factors20.

This study has several limitations. First, the dietary assessment relied on a 24-h recall method, which may have introduced recall bias, potentially affecting the accuracy of nutritional intake measurements. However, the 24-h recall method is considered more precise in capturing detailed information about food types and portion sizes compared to food frequency questionnaires46,47. Additionally, the CIs at the extreme ends in Fig. 2 are wide, indicating uncertainty in these regions. This is primarily due to challenges in study design and data collection, such as the small number of participants in these ranges, as well as external factors beyond our control. Furthermore, stroke diagnosis was based on self-reported data of participants, which may have led to misclassification or reporting bias. The survey also did not differentiate between HS and IS, limiting the specificity of our findings. Another limitation is the study population, which consisted solely of American adults, restricting the generalizability of the results to other ethnic or demographic groups. Additionally, we were unable to analyze specific populations due to the exclusion of individuals with incomplete data, potentially introducing selection bias. Despite using regression models and stratified analyses to control for confounding variables, residual unmeasured confounders may still be present. Finally, given the cross-sectional nature of this study, we cannot establish a causal relationship between dietary zinc intake and stroke prevalence. Further long-term studies, including randomized controlled trials, are necessary to confirm these findings.

Conclusions

Our study suggests an L-shaped relationship between dietary zinc intake and stroke presence in American adults. Within a specific range, higher dietary zinc consumption was associated with a lower odds of stroke. However, beyond a certain intake level, the protective effect plateaued. Further randomized controlled trials are needed to determine whether increasing dietary zinc intake can effectively reduce stroke prevalence.