Sensitization to common foods and early vascular aging: associations and the mitigating effects of health behaviors

Abstract

Background

Sensitization to common foods is typically considered clinically irrelevant in individuals without symptomatic food allergies. However, recent studies found an association between IgE specific to the mammalian oligosaccharide galactose-α-1,3-galactose and cardiovascular disease (CVD). The aims of this study are to determine whether common food sensitization is associated with early vascular aging (EVA) and to examine whether healthier lifestyle behaviors modifies the association in individuals without CVD.

Methods

This was a cross-sectional, population-based study of 2788 American participants aged 30 years or older without cardiovascular disease. Total and specific IgE levels for common foods were measured. EVA was defined based on the 10th percentile of the difference between chronological age (CA) and vascular age (VA). Logistic regression models were employed to assess the associations between food sensitization and EVA, and whether healthy lifestyle modified the association. Poisson regression models, ordinal logistic regression models, and linear regressions were performed as sensitivity analysis.

Results

Sensitization to at least one food allergen associated with an increased risk of EVA (odds ratio [OR] 1.91 [95% confidence interval (CI), 1.1 to 3.3]). Milk sensitization demonstrated the most robust association (OR 7.18, [95% CI, 2.5 to 20.62]). Additionally, moderate to vigorous activities (MVA) (OR 0.33 [95% CI, 0.11 to 0.97]) and sufficient sleep duration (OR, 0.21 [95% CI, 0.07 to 0.65]) mitigate the association between food sensitization and EVA. Similar results were presented in Poisson regression models, ordinal logistic regression models, and linear regressions.

Conclusions

The findings that common foods sensitization is independently associated with EVA, and that MVA and adequate sleep duration mitigate the association, have significant public health implications. Further research is needed to elucidate the mechanisms.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-21951-w.

Background

Food-specific immunoglobulin E (IgE) was traditionally considered clinically irrelevant in individuals without symptomatic food allergies [1]. However, recent studies have found that IgE specific to mammalian oligosaccharide galactose-α-1,3-galactose (α-Gal) is independently associated with coronary artery disease (CAD) [2, 3]. These findings suggest that immunologic responses to food have implications beyond food allergies. Notably, food-specific IgE is prevalent in approximately 15% of the US populations, and most individuals sensitized to food without clinical allergic reactions do not avoid consuming the foods to which they are sensitized [4]. Therefore, sensitization to common foods may be an important risk factor for cardiovascular disease (CVD). However, the relationship between sensitization to common foods and CVD risk has not been studied.

CVD risk can vary considerably among population with the same chronological age (CA), with some exhibiting an abnormally higher risk. Vascular age (VA) is a means used to express CVD risk in terms of age. Early Vascular Aging (EVA) refers to vascular structures and functions that deteriorate more pronouncedly than what would be expected for CA [5]. Research indicates that EVA offers a more precise evaluation of CVD risk than CA and enhances the prediction of cardiovascular events beyond traditional CVD risk factors [6, 7].

The aim of this study is to investigate whether sensitization to common food allergens is associated with EVA in a population without CVD. Additionally, we examined whether a healthier lifestyle modifies the association between food sensitization and EVA.

Methods

Population

The National Health and Nutrition Examination Survey (NHANES) is a population-based survey conducted by the US Centers for Disease Control and Prevention. Informed consent was obtained from all participants, and approved by the National Center for Health Statistics Ethics Review Board. This cross-sectional population-based study included 2788 participants aged 30 years or older without CVD (including coronary heart disease, congestive heart failure, heart attack, stroke, or angina) derived from NHANES 2005–2006, as IgE levels were only measured in this cycle. Participants with missing data on IgE levels and variables required for vascular age calculation were excluded from the analysis (Additional file Fig. S1).

Exposure definition

In NHANES, total IgE and specific IgE levels to common food allergens (cow’s milk, hen’s egg, peanut, and shrimp) and aeroallergens were measured as previously reported [8]. Sensitization was defined as IgE ≥ 2 kU/L [4, 9]. Aeroallergen and food sensitizations were categorized as either present or absent [9]. Total IgE levels were log10-transformed to conform to a normal distribution [9].

Vascular age and early vascular aging

In 2008, D’Agostino et al. [10] proposed the Framingham Risk Score (FRS) to assess general cardiovascular risk, incorporating factors such as age, systolic blood pressure, hypertension treatment, high-density lipoprotein, total cholesterol, diabetes, and smoking. To facilitate a better understanding of risk, they developed a sex-specific ‘vascular age’ concept. An individual’s vascular age is calculated as the age of a person with the same FRS-predicted risk score but with all other risk factors at normal levels. The difference between CA and VA (CA-VA), termed Δ-age, was used to define EVA, with the 10th percentile of Δ-age serving as the cutoff [11].

Consumption of foods

Dietary information was collected using standardized questionnaires. Participants were asked about their liquid milk consumption, including the quantity and type of milk consumed in the past 30 days, as well as the frequency of milk used in cereal over the past 12 months. Milk consumption was defined as cow’s milk only (whole milk, 2% milk, or nonfat/skim milk). Egg and peanut consumption were quantified using 12-month food frequency questionnaires. However, the question about peanut consumption included nuts and seeds, which excluded participants who do not eat peanut but peanut consumer is not definitively identified. Shrimp consumption in the past 30 days was also inquired [12].

Covariates

Demographic information, including age, sex, ethnicity, marital status, income level, and education, was self-reported. Lifestyle factors included alcohol consumption, physical activity, sleep duration, and the healthy eating index (HEI). Comorbidities included asthma, chronic kidney disease (CKD), hypertension, and depression. Body mass index (BMI) was calculated as the ratio of weight (kg) to the square of height (m). Detailed definitions of covariates are presented in Additional file Table S1.

Statistical analysis

Prior to data analysis, we employed the ‘random forest’ method to impute missing values including all 15 covariates. Continuous variables were expressed as mean ± standard deviation (SD) or median with interquartile range (IQR), as appropriate. Categorical variables were reported as counts and percentages. Wilcoxon rank sum test and Pearson’s Chi-squared test or Fisher’s exact test were adopted to compare the group difference.

To study the association between sensitization to individual and overall food and EVA, logistic regression models were created. Model 1 is unadjusted, Model 2 adjusted for age, ethnicity, marital, income level, education, and asthma, Model 3 further adjusted for alcohol drinking, physical activity, sleep duration, HEI, hypertension, CKD, depression, BMI and C-reactive protein. The relationship between sensitization to individual and overall aeroallergen, and total IgE and EVA were similarly examined. To explore whether allergen consumption modified the observed relationship, we stratified the Model 3 by allergen consumption.

To evaluate the modification of health behaviors included in Life’s Essential 8 (moderate to vigorous activities [MVA], healthy diet, sufficient sleep, and reduced BMI) on the association [13], we classified participants based on IgE levels and health behavior status. Smoking was excluded from this analysis as it is a significant component of the FRS. Food sensitization was categorized into higher and lower IgE levels based on the presence or absence of sensitization. Total IgE levels were classified as higher or lower using the mean log10-transformed total IgE as the cutoff. The HEI was dichotomized at the mean HEI value. Sleep duration was categorized as less than 7 h per day or ≥ 7 h per day. BMI was classified using a cutoff value of 25 kg/m².

Several sensitivity analyses were performed. First, Poisson regression was used to estimate the prevalence ratios (PR) to assess the relationship between sensitization, total IgE, and EVA [14]. Second, EVA and supernormal VA were defined by the 10th and 90th percentiles of Δ-age, respectively, with values between these percentiles classified as normal VA [11]. Ordinal logistic regressions were conducted to estimate the association between sensitization, total IgE, and vascular age status. Third, linear regressions using the ordinary least squares (OLS) method were performed to evaluate the association between sensitization, total IgE, and Δ-age. Fourth, complete-case analyses were conducted to avoid imputation bias due to missing data. Additionally, association between sensitization and health behaviors and EVA were examined in sensitivity analyses.

All statistical analyses were performed in R (version 4.3.2). P < 0.05 was considered significant (2-tailed).

Results

Participants characteristics

A total of 2,788 participants were included in this study (282 with EVA). Among them, 112 (4.01%) were sensitized to at least 1 common food allergen, with specific sensitizations as follows: 18 (0.65%) to milk, 54 (1.94%) to peanut, 7 (0.25%) to egg, 47 (1.69%) to shrimp, and 695 (24.93%) to aeroallergens. The mean log10-transformed total IgE level was 1.67 ± 0.65 kU/L.

Table 1 presents the characteristics of the participants. Those sensitized to at least 1 common food allergen were more likely to be younger, male, non-Hispanic white, and had higher levels of C-reactive protein and a higher prevalence of asthma. However, there were no significant differences in marital status, income level, education, or health behaviors.

Table 1.

Participants characteristics

Characteristic	Overall, N = 2,788a	Food sensitization overall		P valueb
		None, N = 2,676a	Any, N = 112a
Age, year	51.00 (40.00, 64.00)	51.00 (40.00, 65.00)	47.00 (37.75, 60.25)	0.033
Sex				0.009
Female	1,383 (49.61%)	1,341 (50.11%)	42 (37.50%)
Male	1,405 (50.39%)	1,335 (49.89%)	70 (62.50%)
Ethnicity				< 0.001
Non-Hispanic White	1,462 (52.44%)	1,417 (52.95%)	45 (40.18%)
Non-Hispanic Black	598 (21.45%)	555 (20.74%)	43 (38.39%)
Mexican American	531 (19.05%)	512 (19.13%)	19 (16.96%)
Other	197 (7.07%)	192 (7.17%)	5 (4.46%)
Marital				0.3
Married or living with a partner	1,856 (66.57%)	1,787 (66.78%)	69 (61.61%)
Not married nor living with a partner	932 (33.43%)	889 (33.22%)	43 (38.39%)
Income Levels				0.11
Low	1,084 (38.88%)	1,051 (39.28%)	33 (29.46%)
Median	1,056 (37.88%)	1,006 (37.59%)	50 (44.64%)
High	648 (23.24%)	619 (23.13%)	29 (25.89%)
Education				0.9
Less than high school	732 (26.26%)	702 (26.23%)	30 (26.79%)
High school graduate or higher	2,056 (73.74%)	1,974 (73.77%)	82 (73.21%)
Smoking Status				0.2
Never	2,179 (78.16%)	2,098 (78.40%)	81 (72.32%)
Current	609 (21.84%)	578 (21.60%)	31 (27.68%)
Alcohol Drinking				0.078
Never	331 (11.87%)	315 (11.77%)	16 (14.29%)
Former	582 (20.88%)	568 (21.23%)	14 (12.50%)
Current	1,875 (67.25%)	1,793 (67.00%)	82 (73.21%)
HEI Score	53.60 ± 12.83	53.70 ± 12.84	51.19 ± 12.54	0.067
Sleep Duration				0.3
< 7 h/d	1,065 (38.20%)	1,018 (38.04%)	47 (41.96%)
7–8 h/d	1,544 (55.38%)	1,489 (55.64%)	55 (49.11%)
> 8 h/d	179 (6.42%)	169 (6.32%)	10 (8.93%)
MVA				0.6
No	1,080 (38.74%)	1,039 (38.83%)	41 (36.61%)
Yes	1,708 (61.26%)	1,637 (61.17%)	71 (63.39%)
Cholesterol, mmol/L	204.28 ± 40.90	204.22 ± 40.76	205.60 ± 44.19	> 0.9
HDL-C, mmol/L	54.81 ± 16.58	54.91 ± 16.70	52.36 ± 13.23	0.2
CRP, mg/dl	0.20 (0.09, 0.49)	0.20 (0.09, 0.48)	0.26 (0.11, 0.68)	0.016
BMI, kg.m2	28.81 ± 6.50	28.79 ± 6.53	29.49 ± 5.74	0.13
Hypertension				0.3
No	1,597 (57.28%)	1,538 (57.47%)	59 (52.68%)
Yes	1,191 (42.72%)	1,138 (42.53%)	53 (47.32%)
CKD				0.8
No	2,522 (90.46%)	2,420 (90.43%)	102 (91.07%)
Yes	266 (9.54%)	256 (9.57%)	10 (8.93%)
Diabetes				0.3
No	2,361 (84.68%)	2,270 (84.83%)	91 (81.25%)
Yes	427 (15.32%)	406 (15.17%)	21 (18.75%)
Asthma				< 0.001
No	2,466 (88.45%)	2,382 (89.01%)	84 (75.00%)
Yes	322 (11.55%)	294 (10.99%)	28 (25.00%)
Depression				0.6
No	2,670 (95.77%)	2,561 (95.70%)	109 (97.32%)
Yes	118 (4.23%)	115 (4.30%)	3 (2.68%)

^a Median (IQR) or Mean (SD); population (percentages, %)

^b Wilcoxon rank sum test; Pearson’s Chi-squared test or Fisher’s exact test

HEI, healthy eating index; MVA, moderate to vigorous activity; HDL-C, high density lipoprotein cholesterol; CRP, C-reactive protein; BMI, body mass index; CKD, chronic kidney disease

Additional file Fig. S2 presents the pattern of missing data, with depression having the highest proportion of missing values (5.42%, n = 151). After excluding participants with missing data, 2,477 participants were included in the complete-case analyses. Among these participants, 100 (4.04%) were sensitized to at least 1 common food allergen, with specific sensitizations as follows: 17 (0.69%) to milk, 46 (1.86%) to peanut, 5 (0.20%) to egg, 45 (1.82%) to shrimp, and 627 (25.31%) to aeroallergens. The mean log10-transformed total IgE level was 1.67 ± 0.65 kU/L.

Association between sensitization and early vascular aging

Sensitization to 1 or more common food allergens was significantly associated with EVA (odds ratio [OR], 1.91 [95% confidence interval (CI), 1.1 to 3.3]) in the fully adjusted model (Table 2). Regarding individual food sensitizations, sensitization to cow’s milk (OR, 7.18 [95% CI, 2.5 to 20.62]) and egg (OR, 5.62 [95% CI, 1.15 to 27.52]) were associated with an increased risk of EVA.

Table 2.

Association between sensitization and early vascular aging

Sensitization	Model1a	P Values	Model2a	P Values	Model3a	P Values
Milk	5.77[2.22,15.02]	< 0.001	6.5[2.43,17.45]	< 0.001	7.18[2.5,20.62]	< 0.001
Egg	6.73[1.5,30.21]	0.013	6.38[1.35,30.19]	0.019	5.62[1.15,27.52]	0.033
Peanut	2.06[1.02,4.13]	0.043	1.56[0.76,3.19]	0.223	1.93[0.91,4.1]	0.088
Shrimp	1.57[0.7,3.54]	0.277	1.45[0.63,3.33]	0.385	1.45[0.59,3.56]	0.419
Food sensitization overall	2[1.21,3.3]	0.006	1.77[1.05,2.95]	0.031	1.91[1.1,3.3]	0.021
Aeroallergen	1.04[0.78,1.37]	0.805	0.88[0.65,1.18]	0.378	0.91[0.67,1.25]	0.563
Total IgE (per Log10 increase)	1.46[1.21,1.76]	< 0.001	1.42[1.16,1.73]	0.001	1.3[1.06,1.61]	0.013

^a Data are presented as odds ratios [95% confidence intervals]. Model 1 was unadjusted; Model 2 adjusted for age, ethnicity, marital, income level, education, and asthma; Model 3 adjusted for Model 2 + alcohol drinking, moderate to vigorous activity, sleep duration, healthy eating index, hypertension, chronic kidney disease, depression body mass index and C-reactive protein. IgE, immunoglobulin E

The association between food sensitization and EVA remained robust in sensitivity analyses. Poisson regression yielded PR of 1.64 [95% CI, 1.09 to 2.46] for any food sensitization, 4.01 [95% CI, 2.28 to 7.06] for sensitization to cow’s milk, and 3.53 [95% CI, 1.81 to 6.89] for sensitization to egg (Additional file Table S2). The association between sensitization to milk and egg and VA status, as estimated by ordinal logistic regression, persisted with common odds ratios (cOR) of 5.96 [95% CI, 2.11 to 16.29] and 5.43 [95% CI, 1.1 to 26.2], respectively (Additional file Table S2). Sensitization to any food allergen and specifically to milk was significantly associated with Δ-age based on OLS regression (Additional file Table S3), with beta coefficients of -5.47 [95% CI, -10.13 to -0.82] and − 3.46 [95% CI, -5.37 to -1.55], respectively. However, the association between sensitization to egg and Δ-age was not significant (beta coefficients, -4.91[95%CI, -12.37 to 2.55], P = 0.197). The association between any food sensitization, sensitization to milk and egg and vascular aging status persisted in complete-case analyses with OR of 1.59 [95%CI, 1 to 2.52], 8.97 [95% CI, 3.06 to 26.35] and 8.79 [95% CI, 1.34 to 57.54], respectively (Additional file Table S4).

The relationship between total IgE levels and EVA persisted in logistic regression (Table 2) and several sensitivity analyses (Additional file Table S2-S4). Association between aeroallergen sensitization and EVA were not statistically significant (Table 2, Additional file Table S2-S4 and Fig. S3).

Food consumption modified the association between sensitization and early vascular aging

We stratified participants by their food consumption status (Table 3). The association between milk sensitization and EVA persisted among milk consumers (OR, 8.54 [95%CI, 2.07 to 35.22]). Peanut sensitization emerged as a statistically significant risk factor for EVA (OR, 2.79 [95% CI, 1.11 to 7]). Shrimp sensitization become borderline significantly associated with EVA (P = 0.051) [15, 16]. However, the association between egg sensitization and EVA disappeared among egg consumers (OR, 5.75 [95%CI, 0.75 to 44.15], P = 0.093). The associations between milk sensitization, peanut sensitization, and shrimp sensitization and EVA were strengthened in specific food consumers. Similar results were observed in Poisson regression, ordinal logistic regression, and complete-case analyses (Additional file Table S5-S6).

Table 3.

Association between food sensitization and early vascular aging stratified by consumption of allergen

Sensitization	n/N	Consumersa	P Value	n/N	Non-consumersa	P Value
Milk	144/1476	8.54[2.07,35.22]	0.003	56/567	7.7[0.39,152.34]	0.18
Egg	182/1892	5.75[0.75,44.15]	0.093	18/151	NA b	NA b
Peanut	177/1815	2.79[1.11,7]	0.029	23/228	NA b	NA b
Shrimp	148/1315	3.6[1,13.03]	0.051	23/209	NA b	NA b

^a Data are presented as odds ratios [95% confidence intervals]. ^b There were too few participants in these categories to calculate ORs

n/N, participants with early vascular aging/all subgroup participants. Adjusted for age, ethnicity, marital, income level, education, asthma, alcohol drinking, moderate to vigorous activity, sleep duration, healthy eating index, hypertension, chronic kidney disease, depression body mass index and C-reactive protein

Health behaviors mitigate the association between sensitization and early vascular aging

Compared with participants at the highest risk of EVA (those with food sensitization and unhealthy behaviors), the OR for participants sensitized to any food and engaging in MVA was 0.33 [95% CI, 0.11 to 0.97]. The OR for participants with food sensitization who slept ≥ 7 h per day was 0.21 [95% CI, 0.07 to 0.65]. No significant risk mitigation was observed in participants with food sensitization and a high HEI or those with a BMI < 25 kg/m² (Fig. 1). Similar results were presented in Poisson regression and ordinal logistic regression (Additional file Fig. S4). However, healthy eating and reduced BMI significantly mitigated the association between food sensitization and EVA in complete-case analyses (Additional file Fig. S5).

Fig. 1.

The combined association of health behaviors and food sensitization and total IgE with early vascular aging (EVA). (A) the combined association of moderate to vigorous activity (MVA) and IgE with EVA; (B) the combined association of healthy eating index and IgE with EVA; (C) the combined association of sleep duration and IgE with EVA; (D) the combined association of body mass index and IgE with EVA. The red squares and error bars indicate a statistical difference. Adjusted for age, ethnicity, marital, income level, education, asthma, alcohol drinking, moderate to vigorous activity, sleep duration, healthy eating index, hypertension, chronic kidney disease, depression body mass index and C-reactive protein

Discussion

This large cross-sectional study, involving 2,788 participants without CVD, provides robust evidence supporting the significant association between common food sensitization and EVA. In addition, we also demonstrated that MVA and sufficient sleep duration mitigate the association of common food sensitization with EVA.

Although the relationship between sensitization to common food and EVA is a novel finding, numerous studies have highlighted the important role of allergic immune pathways in CVD. Mast cells express the high-affinity IgE receptor (FcεR1) and release various inflammatory mediators upon allergen-induced cross-linking of IgE bound to FcεR1 [17]. These cells infiltrate the myocardium and are present within vascular, where they contribute to cardiac development and remodeling [18, 19]. Moreover, mast cells are closely associated with the formation of atherosclerotic plaques, cardiac fibrosis, and the progression to heart failure [19, 20]. IgE is a key component of the protein network involved in allergen/antigen signaling responses [21]. In an observational study, Guo et al. found that serum total IgE levels were higher in patients with CAD, and elevated serum total IgE was identified as an independent risk factor for multivessel disease. A significant linear relationship between serum total IgE and Gensini score was also observed [22]. Additionally, two cross-sectional studies indicated that IgE sensitization to an unusual food, α-Gal, is independently associated with coronary plaque instability, increased non-calcified plaque burden, and the risk of ST-segment elevation myocardial infarction (STEMI) [2, 3]. More recently, a study based on the NHANES and Multi-Ethnic Study of Atherosclerosis (MESA) cohorts further revealed that IgE sensitization to common foods is associated with an increased risk of cardiovascular mortality [9]. In our study, the association between food sensitization and EVA remained consistent across multiple analytical models, with milk sensitization demonstrating a particularly robust relationship. While peanut sensitization was not significantly associated with EVA in the overall participant population, a significant and stronger association emerged among individuals who consumed peanuts. Among participants who consumed milk, eggs, and shrimp, the associations with EVA were strengthened, although the associations between egg sensitization and shrimp sensitization and EVA did not reach statistical significance.

Conversely, sensitization to aeroallergens associated with a reduced risk of EVA, though this relationship was not statistically significant. Sensitization is regarded as a risk factor for allergic or inflammatory diseases, with evidence suggesting that polysensitization is associated with more severe allergic symptoms [23]. However, it may also result in a more comprehensive immune response that provides protection against allergens over time [24]. In our study, polysensitization to aeroallergens was observed in 464 participants, markedly more frequent than polysensitization to common foods, which was present in only 11 participants. This discrepancy may account for the associated reduced risk of EVA with aeroallergen sensitization. Additionally, the relationship between food-specific IgE and CVD is thought to be modulated by gastrointestinal exposure [9]. Ingested food allergens enter circulation and interact with IgE bound to cardiac mast cells, potentially inducing localized chronic inflammation and contributing to the development of CVD [25]. However, exposure from aeroallergens at much lower levels compared to food allergens (nanograms versus grams), which may explain why aeroallergen sensitization does not appear to elevate the risk of EVA [9].

Our study has substantial public health implications. Despite substantial efforts, CVD remains the leading cause of mortality worldwide [26]. An epidemiological analysis based on Global Burden of Disease (GBD) data indicates that from 1990 to 2019, the burden of CVD has consistently increased in nearly all middle- to high-income countries, with prevalence cases rising from 271 million to 523 million [27]. Alarmingly, some high-income countries have experienced a reversal in the trend, transitioning from declining to increasing CVD incidence rates [27]. Research indicates that most individuals with elevated food-specific IgE levels consume the foods to which they are sensitized, despite being unlikely to exhibit clinical allergies [4]. This behavior is associated with an increased risk of CVD, particularly among individuals who consume the foods to which they are sensitized. Although the proportion of individuals with food sensitization is relatively small within our population, milk and egg remain major sources of protein for humans [28]. Considering the large population size, even a small percentage of sensitized individuals can translate to a substantial number of affected people.

We further found that both MVA and sufficient sleep duration significantly alleviate the association between common food sensitization and EVA. The potential protective association of reducing BMI to below 25 kg/m² should be interpreted with caution. Although many models did not demonstrate a statistically significant reduction in EVA risk for participants with high levels of food-specific IgE and BMI below 25 kg/m² compared to those with high food-specific IgE and BMI of 25 kg/m² or higher, a significant reduction in EVA risk was observed among participants with low levels of food-specific IgE and BMI below 25 kg/m². This reduction was more substantial than that seen in participants with high levels of food-specific IgE and BMI below 25 kg/m². When comparing participants with high levels of total IgE and BMI of 25 kg/m² or higher to those with high total IgE and BMI below 25 kg/m², a significant reduction in EVA risk was detected. An even more pronounced reduction was observed in participants with low levels of total IgE and BMI below 25 kg/m². These findings suggest that reducing BMI below 25 kg/m² may also decrease the elevated EVA risk associated with increased IgE levels. Interestingly, participants with high levels of food-specific IgE and high HEI exhibited a higher risk of EVA compared to those with high levels of food-specific IgE and low HEI. We hypothesize that this is attributable to protein intake being a significant component of the HEI [29]. Individuals with higher HEI scores consume more protein daily, resulting in more pronounced sensitization.

To the best of our knowledge, this study represents the first investigation into the relationship between sensitization to common foods and EVA, as well as the impact of healthier behaviors in moderating this association. The findings provide significant public health implications. However, several limitations must be considered. First, the sample size for individuals sensitized to specific foods is relatively small, which may influence the stability and reliability of the results. To address this, we employed multiple analytical methods in this study. Additionally, sensitization to common food was more strongly associated with EVA in food consumers. All those results tend to support our interpretations. Second, the limited number of participants sensitized to specific foods may affect the representativeness of the sample. Applying sample weights may amplify the influence of these few cases and increase standard errors and reduces the precision of estimates [30]. Consequently, as previous studies did, we did not adopt weights in our analysis, resulting in findings that are not nationally representative [31–33]. Third, the study employs an observational design. Although we adjusted for numerous potential confounders, there may still be unidentified variables that were not controlled for, and the observed associations do not establish causality. Fourth, allergy is a clinical diagnosis based on reproducible acute reactions and the inability to tolerate specific foods [4, 34]. Since NHANES does not provide data on food allergy symptoms, we are unable to determine whether participants experienced food allergy symptoms. However, consistent with previous studies, we consider that individuals who report consumption of certain foods in dietary questionnaires are not clinically allergic [9]. Additionally, our subgroup analysis revealed that among individuals consuming specific foods, the association between food sensitization and EVA was strengthened, indicating that the relationship between food sensitization and EVA is most pronounced in individuals who continue to consume specific foods and not exhibiting food allergy symptoms.

Conclusions

Our study demonstrates that sensitization to common foods is independently associated with EVA in individuals without CVD. Additionally, MVA and adequate sleep duration were found to mitigate the elevated risk of EVA associated to common food sensitization. These findings highlight the necessity for further research to elucidate the underlying biological mechanisms.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

IgE

Immunoglobulin E

α-Gal

Mammalian oligosaccharide galactose-α-1,3-galactose

CAD

Coronary artery disease

CVD

Cardiovascular disease

CA

Chronological age

VA

Vascular age

EVA

Early vascular aging

NHANES

National Health and Nutrition Examination Survey

HEI

Healthy eating index

CKD

Chronic kidney disease

BMI

Body mass index

SD

Standard deviation

IQR

Interquartile range

MVA

Moderate to vigorous activities

FRS

Framingham Risk Score

PR

Prevalence ratio

OLS

Ordinary least squares

OR

Odds ratio

CI

Confidence interval

cOR

Common odds ratio

FcεR1

High-affinity IgE receptor

STEMI

ST-segment elevation myocardial infraction

MESA

Multi-Ethnic Study of Atherosclerosis

GBD

Global Burden of Disease

Author contributions

Dr DC had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: RS, XL, DC; Statistical analysis: RS, SC, QS, HW; Acquisition, or interpretation of data: RS, ZZ, HW; Drafting of the manuscript: RS, SC, ZZ; Critical review of the manuscript for important intellectual content: DC, XL, FP, JL; Obtained funding: DC; Administrative, technical, or material support: DC, JL, FP.

Funding

This work was supported by Fujian Provincial Health and Family Planning Commission (grant numbers: 2021ZQNZD005); Joint Funds for the Innovation of Science and Technology of Fujian Province (grant numbers 2023Y9107); and National Natural Science Foundation of China (grant numbers 82271591).

Data availability

The datasets generated and/or analysed during the current study are available in the NHANES website, https://wwwn.cdc.gov/nchs/nhanes/default.aspx.

Declarations

Ethics approval and consent to participate

The National Health and Nutrition Examination Survey (NHANES), conducted by the US Centers for Disease Control and Prevention and was approved by the National Center for Health Statistics Ethics Review Board. Informed consents have been obtained from all participants. As NHANES data are deidentified and anonymous during analysis, secondary analyses require no additional ethical approval or informed consent. This study was in compliance with the ethical standards of the 1964 Declaration of Helsinki and its amendments.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.