Dietary Intake and Arterial Stiffness in Children and Adolescents: A Systematic Review

Arterial stiffness is a risk factor for cardiovascular disease that is affected by diet. However, research understanding how these dietary risk factors are related to arterial stiffness during childhood is limited. The purpose of this review was to determine whether various dietary factors were associated with arterial stiffness in the pediatric population. Five databases were systematically searched. Intervention studies, cross-sectional and cohort studies were included that investigated nutrient or food intake and outcomes of arterial stiffness, primarily measured by pulse wave velocity (PWV) and augmentation index (AIx), in the pediatric population (aged 0–18 years). A final 19 studies (six intervention and 13 observational) were included. Only two intervention studies, including a vitamin D and omega-3 supplementation trial, found protective effects on PWV and AIx in adolescents. Findings from observational studies were overall inconsistent and varied. There was limited evidence to indicate a protective effect of a healthy dietary pattern on arterial stiffness and an adverse effect of total fat intake, sodium intake and fast-food consumption. Overall, results indicated that some dietary factors may be associated with arterial stiffness in pediatric populations; however, inconsistencies were observed across all study designs. Further longitudinal and intervention studies are warranted to confirm the potential associations found in this review.


Introduction
Globally, cardiovascular disease (CVD) is the largest contributor to non-communicable diseases, with the total number of deaths from CVD steadily increasing from 12.1 million in 1990 to 18.6 million in 2019 [1]. CVD is considered a largely preventable disease, as a plethora of the associated risk factors are considered modifiable, including smoking, hypertension and poor nutrition [2]. Importantly, early prevention is fundamental to reducing CVD prevalence over the lifespan, as subclinical changes in cardiovascular structure and function can commence during childhood [3,4].
Arterial stiffness is increasingly used as a pre-clinical measure of CVD. Decreased elasticity of arterial walls represents one of the earliest markers of damage to the vascular system [5]. Among adults, arterial stiffness is an independent predictor of cardiovascular events and mortality and all-cause mortality [6]; among adolescents and young adults, it has been associated with target organ damage [7]. While various methods are available to determine the degree of arterial stiffness, the gold standard is pulse wave velocity (PWV) [8,9]. Stiffening of the arteries naturally progresses with age over the lifespan; however, the process can be accelerated with certain conditions, such as obesity [10,11] and via environmental factors [9]. A few systematic reviews have investigated whether a relationship is evident between dietary intake and arterial stiffness, with the majority focusing on single nutrients and all in adult populations [12][13][14][15]. For instance, an earlier systematic review of randomized control trials (RCTs) conducted in 2012 explored whether nutrient and dietary interventions reduced arterial stiffness in adults [12]. Findings from this review showed some evidence of an effect via supplementation with omega-3 fatty acids and soy isoflavones [12]. A more recent systematic review and meta-analysis of RCTs published in 2018 investigated the effect of dietary sodium restriction in adults on arterial stiffness and found that decreased consumption was linked to reduced arterial stiffness [14]. Further, an umbrella review of systematic reviews and meta-analyses were conducted in 2019, which investigated the effect of vitamin C supplementation towards cardiovascular markers, including arterial stiffness [13]. Ultimately, no effects were exhibited between vitamin C supplementation and arterial stiffness [13].
While there is emerging literature examining the relationship between nutrient and dietary intake [16][17][18], including nutrient supplementation [19], and measures of arterial stiffness among children, to date, no systematic reviews have been conducted within the pediatric population. Understanding how modifiable lifestyle factors, such as dietary intake, are related to arterial stiffness during childhood is important for the development and implementation of dietary strategies to protect cardiovascular health. Therefore, the aim of this systematic review was to determine whether dietary factors, including patterns, nutrients and food intake, were associated with arterial stiffness within the pediatric population.

Materials and Methods
This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [20] and was registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration number CRD42022315443).

Literature Search
A literature search was performed across five databases, including MEDLINE Complete and CINAHL Complete (via EBSCOhost platform), Scopus, Cochrane Library and Embase (via Elsevier), identifying studies published from inception to 4 March 2022. The Population; Intervention/Exposure; Comparison; Outcome (PICO) framework was utilized to develop the research question and search strategy. Participants: Children and adolescents aged 0-18 years; Intervention/Exposure: Food, nutrient intake or dietary pattern; Comparison: for intervention studies, children with less exposure will be compared to the control group, e.g., an intervention group receiving a dietary supplement compared to control group not receiving supplementation; Outcomes: Indexes of arterial stiffness. These include pulse wave velocity (PWV), which, when measured at the carotid-femoral site, is considered the reference method for arterial stiffness [8,9], augmentation index (AIx) and arterial distensibility. PWV measures the speed (m/s) for the arterial pulse to travel between the carotid and femoral arteries, with higher values representing increased stiffness in the arteries [9]. AIx is an indirect measure that calculates the ratio between increased pressure of reflected wave and pulse pressure or simply the difference between systolic and diastolic blood pressure [21]. An increased AIx infers greater arterial stiffness.
The search strategy included three concepts related to the research aim: (1) dietary factors; (2) indexes of arterial stiffness; (3) population of interest. Search terms were supplemented with appropriate Medical Subject Headings (MeSH). Examples of search terms included: "dietary intake" OR "nutrient intake" OR Food* OR "dietary pattern" OR vitamin OR mineral (concept 1) AND "arterial stiff*" OR "pulse wave velocity" OR "vascular ageing" (concept 2) AND child* OR adolesce* OR pediatric (concept 3). The full search strategy across all databases can be found in supplementary file S1 (Supplementary Tables S1-S5). Reference lists of included studies were searched to identify any additional studies.

Eligibility Criteria
To be eligible for inclusion, studies were required to meet the following criteria: have used a cross-sectional, cohort or intervention study design; have investigated nutrient or food intake; and measured outcomes of arterial stiffness in children and/or adolescents.
Eligible studies were sourced from the aforementioned databases, being limited to English only and human studies, but with no limits on the date of publication. Conference abstracts were excluded. Studies that included participants older than 18 years of age were included if the mean age of the sample was <18 years. Any studies that included participants with a chronic or genetic disease (e.g., type 1 diabetes) were excluded, as the focus of the review was to determine the effects of dietary factors on generally healthy children and adolescents (i.e., not targeting any specific health condition). However, given the high prevalence of overweight and obesity in the general pediatric population, no restriction was placed on children with these conditions. Furthermore, understanding the potential effect of diet on arterial stiffness among children with obesity is important, given the increased risk of arterial stiffening within this group [11]. Studies that did not assess the exposure or have a dietary intervention, such as those monitoring effects of physical exercise, were excluded. Included indexes of arterial stiffness were PWV, AIx and carotid or brachial artery distensibility.

Data Management and Extraction
Articles obtained through the literature search were imported to EndNote, and then exported into Covidence, where duplicate records were removed. Screening at title and abstract level, followed by full-text review, was completed in duplicate by two independent reviewers. Any conflicts for article inclusion were resolved via discussion, and if necessary, a third reviewer was consulted. A data extraction table was created and piloted with two studies and refined to ensure all relevant data were collected. Data were then extracted from eligible studies independently by two reviewers. Extracted data included: aim, study design, sample characteristics, measures of exposure and outcome, intervention details (where relevant), statistical analysis (including covariate adjustment) and study findings. In all instances, the most adjusted model was preferentially extracted.

Quality Assessment
The Academy of Nutrition and Dietetics Quality Criteria Checklist was used to assess overall quality and risk of bias of included studies [22] independently by two reviewers. Any disagreement in assessment was resolved via discussion. The tool consists of four relevance questions that assessed significance related to the topic of interest and dietetic practice, followed by a further ten questions (each with sub-questions) that scrutinized the validity of the study, including participant selection, blinding, methods of measurement and statistical analysis [22]. Each question was answered with yes (Y), no (N), unclear (U) or not applicable (NA). Each study received an overall positive, neutral or negative quality rating based on validity question answers. A positive rating indicated a low risk of bias, a neutral score indicated moderate risk of bias and a negative score indicated high risk of bias. Studies allocated an 'N' for six or more validity questions were rated negative. Those with answers for validity questions 2, 3, 6 and 7 that were less than exceptionally strong were rated neutral. For studies with the majority of validity questions assigned 'Y' (including questions 2, 3, 6, 7), a positive rating was allocated [22].

Data Synthesis
Data were grouped according to the study design, and then further stratified based on the assessed dietary component (e.g., omega-3, vitamin D, dietary patterns, sodium). Due to the limited number of studies retrieved and heterogeneity in dietary exposures, indexes of arterial stiffness, covariate adjustment and methodologies used across studies, no meta-analysis was performed.

Study Selection
In total, 8033 articles were identified across five databases, 3248 of which were duplicates and removed ( Figure 1). Following title and abstract screening, 111 articles were retrieved for full-text assessment, 92 of which were excluded, leaving 19 articles for inclusion. No further studies were identified following a screening of the reference list of the included studies.

Data Synthesis
Data were grouped according to the study design, and then further stratified based on the assessed dietary component (e.g., omega-3, vitamin D, dietary patterns, sodium). Due to the limited number of studies retrieved and heterogeneity in dietary exposures, indexes of arterial stiffness, covariate adjustment and methodologies used across studies, no meta-analysis was performed.

Findings from Intervention Studies
Across the six intervention studies, sample sizes ranged from 25 to 225 participants, and intervention duration ranged from 12 weeks to 3 years. With the exception of one study conducted in 8-year-old children [27], the remaining were conducted in adolescents aged between 10 and 18 years. Four of the studies were completed in participants who were overweight or obese [19,[24][25][26] (Table 4).

Vitamin D Supplementation
Three RCTs examined the effect of vitamin D supplementation on PWV [19,23,25]. All studies used a lower dose of vitamin D in the active control, and the dosage in the intervention group ranged from 2000 IU/d of vitamin D3 to 120,000 IU/d. All studies showed increases in plasma 25(OH)D within the intervention group vs. control at the end of the intervention period. Two of these studies, both completed with adolescents who were overweight or obese, found no effect of vitamin D supplementation on cf-PWV following a 6-or 12-month intervention [19,25]. Conversely, the third study completed in apparently healthy African-American adolescents found an improvement in cf-PWV following vitamin D supplementation for a 16-week period [23]; however, there was no change in carotid-radial PWV or carotid-distal PWV.

Omega-3 Fatty Acid Supplementation
Two intervention studies examined the effect of omega-3 supplementation on arterial stiffness [26,27]. The first randomized cross-over trial, completed in obese adolescents, showed no effect on carotid-radial PWV following 3 months of 1.2 g of omega-3 supplementation; however, there was an improvement in AIx [26]. Similarly, a 3-year follow-up of a previous 5-year parallel RCT showed no difference in brachial-PWV measured at 8 years of age when comparing children who had consumed a diet supplemented with omega-3 fatty acids and a control diet for the first 5 years of life [27]. In addition, this study found no effect on AIx or carotid artery distensibility.

Milk Protein Supplementation
The final parallel RCT, which examined the effect of different types of milk proteins (e.g., supplementing diets with skimmed milk and casein and whey-based protein drinks) found the consumption of these drinks over a 12-week period resulted in no change in cf-PWV or AIx among overweight and obese adolescents when compared to a pretest control group or water group [24].

Prospective Cohort Studies
Across the four prospective cohort studies, the follow-up period ranged from 6 to 10 years and included between 93 and 4024 participants [18,[28][29][30]. All were based on population birth cohorts.

Cross-Sectional Studies
The majority of cross-sectional studies (6/9) were based on community samples of children recruited from schools, including elementary [17,[33][34][35] and secondary schools [32,36] or a combination of both [34]. One of these school-based studies specifically targeted the recruitment of obese children [34]. Another cross-sectional study was completed on adolescents who were overweight or obese and recruited for an intervention study (reported above) [24,31]. The sample size of cross-sectional studies ranged from 75 to 1780 participants (Table 5).

Infant Feeding Practices Prospective Cohort Studies
Two prospective cohort studies examined infant feeding practices (breastfeeding or introduction to solids) [28,29] with mixed findings reported. For breastfeeding, one study reported a protective effect [29], while the other reported a negative effect [28]. In a large study, (n = 4024) de Jonge et al. [29] reported that at 6 years of age, cf-PWV was significantly lower in children who had been breastfed vs. never breastfed during the first 12 months of life. This study adjusted for a comprehensive list of maternal-and child-related covariates ( Table 5). In contrast, in a much smaller study (n = 93), which included adjustment for child-related covariates only, Schack-Nielsen reported that longer duration of breastfeeding during the first 9 months of life was associated with significantly higher carotid-femoral-PWV measured at age 10 years; however, no association was observed for carotid-radial PWV [28]. The cohort study by de Jonge [29] also assessed the association between the timing of the introduction of solids and cf-PWV at 6 years of age and reported no association between these two variables.

Cross-Sectional Studies
One cross-sectional study examined the relationship between breastfeeding and arterial stiffness [33]. In this study of 520 children, no association was found between the duration of exclusive breastfeeding (retrospectively collected) and cf-PWV measured at age 9-10 years (adjusted for some child-related covariates) [33].

Dietary Patterns Prospective Cohort Studies
One prospective cohort study examined the association between dietary patterns and arterial stiffness [18]. In this cohort of children followed from age 4 to 15 years (n = 188), it was reported that among four empirically derived dietary pattern trajectories ('unhealthy', 'moderately unhealthy', 'moderately healthy' and 'healthy'), none were associated with cf-PWV measured at 15 years of age (adjusted for maternal and child-related covariates) [18].

Cross-Sectional Studies
Five cross-sectional studies examined the association between dietary patterns and arterial stiffness [17,32,[34][35][36]. Three of these studies reported null findings [17,34,35], and two reported inverse associations (e.g., protective) [32,36]. Empirically derived dietary patterns were identified and assessed in two of these studies [17,35]. Firstly, among Italian primary schoolchildren (n = 300), there were no associations identified between 'healthy' or 'unhealthy' dietary patterns and cf-PWV [17]. Secondly, among New Zealand primary schoolchildren (n = 389), there were no associations between identified dietary patterns characterized by 'snacks' or 'fruit and vegetables' and either cf-PWV or Aix [35]. Both of these cross-sectional studies adjusted for a range of child-related covariates, including physical activity ( Table 5). The remaining three studies used scoring systems to assess adherence to either a Mediterranean dietary pattern (n = 2) [32,34] or a heart-healthy dietary pattern (n = 1) [36]. One study assessing adherence to the Mediterranean dietary pattern completed in elementary school students from Greece (n = 277) reported that higher adherence to this dietary pattern was significantly inversely associated with Aix (e.g., protective) (adjusted for child-related covariates) [32]. Conversely, in a small sample of Italian schoolchildren (n = 75), all with obesity, there was no association with adherence to a Mediterranean dietary pattern and cf-PWV (adjusted for child-related covariates) [34]. Finally, in a crude unadjusted analysis of Italian elementary schoolchildren (n = 387), cf-PWV was significantly lower (e.g., protective) among children with an 'ideal' heart-healthy diet score compared to children with a 'not ideal' score [36].

Consumption of Specific Food Groups Cross-Sectional Studies
Three cross-sectional studies assessed the association between consumption of specific food groups and arterial stiffness with mixed findings [17,31,37]. In a sample of Italian primary schoolchildren (n = 300), Giontella et al. [17] examined the association between 14 different foods/food groups and PWV. The only food found to be positively associated with cf-PWV was fast food. No associations were reported for cereals and tubers, vegetables, fruit, eggs, meat, dairy products, sweets, legumes, fish, nuts, extra virgin olive oil, animalderived fat or seed oil [17]. Conversely, in a large population-based sample (n = 1780) of Australian children aged 11-12 years, it was reported that neither fast food nor sugarsweetened beverage consumption was associated with cf-PWV [37]. Finally, in a sample of adolescents who were overweight or obese (n = 183), there was a trend whereby greater consumption of milk and yoghurt per day was associated with lower cf-PWV (p = 0.05) (e.g., protective), but no association was reported with Aix [31] All of these studies were adjusted for a range of child-related covariates ( Table 5).

Macronutrient Intake Prospective Cohort Studies
One prospective cohort study assessed the association between macronutrient intake and arterial stiffness, with mixed findings across different macronutrients [30]. The prospective cohort study was based on a population sample of children (n = 2427) with measures of dietary intake at 14 months and cf-PWV at age 6 years [30]. There was evidence that higher total fat intakes (highest vs. lowest tertile) at 14 months were associated with increased cf-PWV at 6 years (e.g., adverse effect), and total carbohydrate and mono-and disaccharides (middle vs. lowest tertile) were inversely associated with cf-PWV (e.g., protective effect). However, there were no other associations with total protein or saturated fat intake, nor other sub-components of these macronutrients (e.g., vegetable and animal protein, mono and polyunsaturated fat or polysaccharides) (models adjusted for maternal and child-related covariates, including energy intake) [30].

Cross-Sectional Studies
One cross-sectional study assessed the relationship between macronutrient intake and arterial stiffness [31]. In this study of adolescents with overweight or obesity (n = 183), there was no association between the percentage of total energy from fat intake and cf-PWV and a positive association between the percentage of energy from protein intake and cf-PWV; however, consumption of neither macronutrient was associated with AIx (adjusted for child-related covariates) [31]. In addition, one prospective cohort study reported a positive cross-sectional association between the percentage of energy from fat intake and PWV at 10 years of age [28].

Sodium Intake Cross-Sectional Studies
One cross-sectional study examined the association between sodium intake and arterial stiffness [16]. Brady et al. reported associations between sodium density (mg/kcal/d) and three indexes of arterial stiffness (cf-PWV, AIx, brachial artery distensibility) in a sample of USA participants (n = 614), which included children and young adults (10-24 years, mean age 17.9 years) [16]. With a higher sodium diet, a significant inverse association was reported for brachial artery distensibility (adverse effect) and positive associations for carotid-femoral-PWV and AIx (both adverse effects) (adjusted for child-related covariates, including body fat and systolic blood pressure).

Discussion
This systematic review is the first to examine the association between diet and arterial stiffness in children and adolescents. Overall, the evidence across the range of dietary factors examined in this review was scarce. There was limited evidence to suggest a protective effect of vitamin D and omega-3 supplementation and a healthy dietary pattern on arterial stiffness in the pediatric population versus an adverse effect from total fat intake, sodium intake and fast-food consumption. A lack of significant findings across intervention and longitudinal studies may be reflective of the longer time period required for meaningful vascular remodelling to occur and induce changes to arterial stiffness [38]. Overall, the clinical implications of the observed effects on arterial stiffness are unclear, as the overall effect estimates were generally small and may not be clinically relevant. Nonetheless, a previous systematic review found that a relatively small 1 m/s increase in PWV was associated with a 15% increase in cardiovascular mortality and a 14% increase in overall CVD events in adults [6]. Although the implications of increased arterial stiffness in the pediatric population on future cardiovascular outcomes are largely unexplored, relatively small changes in PWV during childhood may still hold relevance for future cardiovascular health.

Intervention Studies
Overall, there were very few intervention studies examining the effect of dietary intervention on arterial stiffness in children, and all of these were based on single-nutrient dietary supplementation. The effect of vitamin D supplementation on arterial stiffness was inconsistent, with two studies showing no effect in an overweight/obese population and a single study in African-American youth finding a significant improvement in cf-PWV, the gold standard measure for arterial stiffness, following supplementation [8,9]. The mechanism for these vasculoprotective effects may be attributed to an array of direct or indirect actions on vascular cells, such as suppression of the renin-angiotensin system, impacts on calcium metabolism and reducing inflammation and oxidative stress on cells [23]. However, the magnitude of this decrease in cf-PWV observed post-supplementation was small (~0.08 m/s) and may not be clinically significant when considering implications

Discussion
This systematic review is the first to examine the association between diet and arterial stiffness in children and adolescents. Overall, the evidence across the range of dietary factors examined in this review was scarce. There was limited evidence to suggest a protective effect of vitamin D and omega-3 supplementation and a healthy dietary pattern on arterial stiffness in the pediatric population versus an adverse effect from total fat intake, sodium intake and fast-food consumption. A lack of significant findings across intervention and longitudinal studies may be reflective of the longer time period required for meaningful vascular remodelling to occur and induce changes to arterial stiffness [38]. Overall, the clinical implications of the observed effects on arterial stiffness are unclear, as the overall effect estimates were generally small and may not be clinically relevant. Nonetheless, a previous systematic review found that a relatively small 1 m/s increase in PWV was associated with a 15% increase in cardiovascular mortality and a 14% increase in overall CVD events in adults [6]. Although the implications of increased arterial stiffness in the pediatric population on future cardiovascular outcomes are largely unexplored, relatively small changes in PWV during childhood may still hold relevance for future cardiovascular health.

Intervention Studies
Overall, there were very few intervention studies examining the effect of dietary intervention on arterial stiffness in children, and all of these were based on single-nutrient dietary supplementation. The effect of vitamin D supplementation on arterial stiffness was inconsistent, with two studies showing no effect in an overweight/obese population and a single study in African-American youth finding a significant improvement in cf-PWV, the gold standard measure for arterial stiffness, following supplementation [8,9]. The mechanism for these vasculoprotective effects may be attributed to an array of direct or indirect actions on vascular cells, such as suppression of the renin-angiotensin system, impacts on calcium metabolism and reducing inflammation and oxidative stress on cells [23]. However, the magnitude of this decrease in cf-PWV observed post-supplementation was small (~0.08 m/s) and may not be clinically significant when considering implications on future CVD risk [6]. Furthermore, black youth are at increased risk of vitamin D deficiency in comparison to the white population; Dong et al. [39] previously found that compared to white subjects, Black youth had significantly lower plasma 25-hydroxyvitamin D levels in every season of the year. Thus, the favourable outcomes seen in the Black youth cannot be extrapolated to the broader pediatric population, as these underlying physiological differences may impact outcomes.
The evidence for omega-3 supplementation is inconclusive, with several limitations observed in these studies. Based on evidence from two lower-quality studies, one detected no changes in brachial-PWV, arterial structure or function in a 3-year follow-up of an intervention in young children [27]. A major limitation of this study was that it did not consider potential confounding diet and lifestyle factors between the cessation of interventions and the 3-year follow-up measurements. Another RCT found no significant effects on cr-PWV post-supplementation; however, a reduction in AIx was observed [26]. Comparability and validity of these studies are limited due to variations in sites measured for PWV and the overall lack of use of the gold-standard outcome measure for PWV (e.g., cf-PWV). A previous systematic review in adults found that chronic omega-3 supplementation between six weeks and two years in duration was effective in reducing PWV and improving arterial compliance [12]. Nevertheless, such differences may be indicative of the physiological differences based on age; thus, further studies of long-term omega-3 supplementation in children that utilize the gold standard outcome measure of cf-PWV are warranted to determine if a beneficial effect may be seen in the pediatric population.

Observatoinal Studies
The effect of a range of dietary patterns on arterial stiffness has been investigated in children. Most of these have been cross-sectional studies, with very limited evidence from higher-quality, prospective study designs. The most frequently studied dietary exposure across cross-sectional studies was dietary patterns (Figure 1). The overall evidence for associations between dietary patterns and arterial stiffness was mixed, with the majority reporting null effects [17,34,35]. Furthermore, evidence from the one well-designed prospective cohort study assessing dietary pattern trajectories during early to middle childhood found no association with changes in cf-PWV measured in adolescence [18]. Among adults, a Mediterranean diet has been associated with improved arterial stiffness in adult populations [40]. It is thought that the high fruit, vegetable, nut, fish and olive oil content of the Mediterranean diet improves vascular measures due to their antioxidant effect, thus reducing oxidative stress [41,42]. Further studies in children are required to determine the potential effect of healthy dietary patterns, such as the Mediterranean diet, on arterial stiffness in the pediatric population.
Limited evidence exists for associations between various food groups and arterial stiffness. One study found that from 14 different food groups, the only association with cf-PWV was with fast food intake, suggesting a detrimental effect [17]. On the contrary, Saraf et al. [37] reported no association between the consumption of takeaway foods or sugar-sweetened beverages and cf-PWV.
Two studies examined the relationship between dietary macronutrient composition and arterial stiffness, with some evidence to suggest a relationship between these factors, although the magnitude of these associations was small, raising questions of clinical significance [30,31]. Higher total fat intakes were associated with increased arterial stiffness in one well-designed cohort study with a long six-year follow-up period [30]. A possible mechanism for this relationship is higher intakes of fat, particularly saturated fatty acids, being associated with increased plasma LDL-cholesterol concentrations [43], which result in vascular remodelling and may initiate the underlying processes of atherosclerotic plaque formation, suspected to begin during childhood and progress later into life [44]. However, this same study [30] found no association between different types of fat (including saturated fat) and PWV. Further studies are required to clarify the relationship between different types of fat intake and arterial stiffness within the pediatric population.
A relationship was also evident between arterial stiffness and carbohydrate intake, with significant improvements in arterial stiffness seen at moderate-high intake, compared to low intake [30]. However, the effect size of these findings was small and does not provide information on carbohydrate sources or overall diet quality. Additionally, this study is limited as habitual dietary intake may not be represented due to the single-timepoint dietary assessment and potential confounding dietary changes over the 6-year follow-up period were not considered.
One small, cross-sectional study reported a positive relationship between percentage energy intake from protein and PWV in a cohort of children with habitual low milk intake [31]. The study attributed the detrimental effect of protein to high meat consumption; however, it did not report on the type of protein consumed. Given the variability in health effects of different protein sources, the quality of protein is important to consider [45]. The study also reported a trend for an inverse association between milk intake and PWV (i.e., protective) [31]; however, an RCT in the same cohort found no effect of milk proteins on arterial stiffness [24]. The 12-week intervention period of the RCT may not have been long enough to produce a meaningful result. Milk is expected to play a protective role in vascular health due to its high essential amino acid, vitamin and mineral content and the role of these in protein synthesis and fatty acid oxidization; however, further research is warranted to examine the potential effect on markers of arterial stiffness among children [46].
There was limited evidence examining the relationship between breastfeeding and arterial stiffness in childhood, with two cohort studies of varying quality showing inconsistent results. Overall, the evidence presented in this review is not strong enough to infer that a relationship exists between breastfeeding and arterial stiffness due to both a small sample size in one study [28] and the potential for residual childhood dietary confounding factors that were not adjusted for in analysis [28,29]. Furthermore, breastfeeding is an integral part of dietary guidelines and has been well established to have a range of health benefits, including reduced risk of developing CVD risk factors, such as obesity, high blood pressure and type 2 diabetes [47,48].
Only one study has investigated the association between salt intake and arterial stiffness in children with findings consistent with those in the adult population, with lower sodium intake associated with protective effects on a range of arterial stiffness markers [16]. Salt intake has previously been shown to be detrimental to endothelial function and arterial stiffness in normotensive and hypertensive individuals [14,49]. Salt has been found to have these vascular effects independent of blood pressure, making it highly relevant for a young, healthy population [12,50]. Its effect is potentially due to high salt intake increasing the generation of reactive oxygen species, which thereby reduces the stimulation of nitric oxide, a potent vasodilator [51]. Further studies that include long follow-up periods into adolescence and young adulthood are required to confirm these findings.

Clinical Implications and Future Research
From the current systematic review, limited evidence suggested that some pediatric dietary factors may modulate arterial stiffness; these tended to be in line with general healthy eating patterns and dietary guidelines. In particular, higher total fat and salt intake and less healthy dietary patterns overall were linked to increased arterial stiffness. These relationships are in line with general dietary recommendations, which emphasize the importance of limiting total fat intake and minimizing intake of discretionary foods that are high in fat and salt in order to reduce the risk of various adverse health outcomes, such as overweight/obesity and CVD [48,52,53].
However, given the thin spread of evidence across all study designs for associations between diet and arterial stiffness in children, further research is required to confirm any potential associations found in the current review. Previous studies outside the scope of this review have examined the relationship between childhood dietary factors and arterial stiffness outcomes later into adulthood. A recent large cohort study found no correlation between childhood calcium intakes and PWV when measured in later adulthood [54].
However, another large cohort study by Aatola et al. [38] that examined the relationship between lifetime risk factors and its implications on PWV in adulthood found that persistently higher consumption of fruit and vegetables in childhood and adulthood was associated with lower adulthood PWV, thereby displaying a protective relationship between arterial stiffness and fruit and vegetable consumption. Although the implications of increased arterial stiffness in the pediatric population on future cardiovascular outcomes are largely unexplored, evidence of an association with target organ damage in children [7] alongside the potential impact of childhood dietary risk factors on arterial stiffening and CVD risk later in life [6,7,55] provides a strong rationale for further research in this area.
Given the strong findings in adult populations for the roles of omega-3 supplementation and salt intake on modulating arterial stiffness, it is suggested that future research focuses on these dietary interventions [12,56]. High-quality randomized controlled trials for single nutrient interactions such as omega-3 and salt intake are recommended to ascertain if a causal effect on arterial stiffening exists in the paediatric population. Furthermore, well-designed cohort studies, such as those assessing a range of dietary patterns during the formative childhood years with longer follow-up periods into adulthood, such as that seen in the Young Finns study [38,54], are recommended to allow for meaningful changes in vascular structure and function to occur. Additionally, a standardized measure of arterial stiffness, cf-PWV, should be used as this has been recommended in the paediatric population [9,55].

Strength and Limitations
Research on this topic is not without limitations. Firstly, heterogeneity amongst dietary variables, study designs and measures of arterial stiffness made it difficult to synthesise results. This, combined with a low number of included studies, meant a meta-analysis could not be performed. Not all studies employed the reference method for arterial stiffness measurement (cf-PWV) [9,57]. Furthermore, this review only included research with participants with a population mean age of <18 years. Although results are specifically relevant to the paediatric population, significant findings have been reported in studies of greater duration beyond childhood and adolescent years [38], implying that a longer follow-up period may be required to observe the impact of dietary intake changes during childhood on arterial stiffness in adulthood. As the review was limited to the inclusion of studies published in English only, studies published in languages other than English may have been overlooked.
In terms of the study methodology of included research, several limitations were identified. Significant variables remained unadjusted for in multiple studies, including physical activity [34] and cardiorespiratory markers [18], which have been shown to impact arterial stiffness [58]. Three studies assessing the effect of breastfeeding on arterial stiffness in later childhood had a follow-up period between six and ten years, with dietary measures unaccounted for between infancy and follow-up [28,29,33]. Inconsistency in outcome measurement occurred in three studies, with arterial stiffness not measured in all participants included in final analyses, reducing the validity of the data presented [24,27,28]. Multiple observational studies reported a small sample size [18,28,31,34]. The overarching tools used for dietary analysis were food frequency questionnaires and dietary recall, leading to the possibility of recall bias or social desirability bias, in which participants will report data thought to be more 'desirable' by society [59]. However, objective measurement of diet is near impossible for cohort studies analysing usual dietary habits. Overall, the quality of research included in this review is not exceptional, with a significant number of studies reporting insubstantial information on withdrawals, poor participant accountability, biased participant selection and/or unblinded approaches to variable measurement, increasing the risk of bias. However, the majority of included studies employed valid and reliable outcome measures and all studies reported no conflict of interest.
This systematic review has several strengths. The PRISMA guidelines were followed throughout the review process. This review is strengthened by a comprehensive search strategy and data extraction process. A search strategy encompassing a broad range of search terms was developed and used across five databases to obtain relevant articles. Data extraction and quality assessment were completed in duplicate to minimise the risk of bias. Data extraction was piloted, and amendments were made to ensure that all relevant information was captured from the included research.

Conclusions
This systematic review found that various dietary exposures may play a role in mediating arterial stiffness in a paediatric population. Although evidence of single nutrient interactions from intervention trials was limited, evidence from observational studies indicated that higher total fat and salt intakes may be associated with increased arterial stiffening in children. However, the strength of this evidence is inconsistent, methodologies are varied and only a small number of studies have been conducted in this area. Given the evidence for meaningful impacts of various dietary exposures on modulating arterial stiffness in adults, more robust studies in the paediatric population that implement longer follow-up periods are warranted to ascertain any potential associations and ensure that sufficient time is provided for vascular remodelling to induce meaningful changes in arterial stiffness.