Associations of the Dietary Iron, Copper, and Selenium Level With Metabolic Syndrome: A Meta-Analysis of Observational Studies

Background Epidemiological studies have investigated the associations of dietary iron, copper, and selenium level with metabolic syndrome (MetS). However, their results are conflicting. This meta-analysis of observational study was, therefore, employed to investigate the associations above. Methods A comprehensive literature search was employed using PubMed, Web of Science, Embase, and Scopus database up to October 2021 (no restriction was set for the initiate time). The pooled relative risk (RR) of MetS for the highest vs. lowest dietary iron, copper, and selenium level was estimated, respectively. Results A total of 14 observational studies (55,131 participants) were identified as meeting the inclusion criteria. Specifically, 7 studies were related to the dietary iron level. The overall multivariable adjusted RR demonstrated that the dietary iron level was positively associated with MetS (RR = 1.27, 95% CI: 1.12–1.44; p < 0.001). With regard to the dietary copper level, 7 studies were included for meta-analysis. The overall multivariable adjusted RR showed that the dietary copper level was inversely associated with MetS (RR = 0.85, 95% CI: 0.78–0.93; p < 0.001). In addition, 4 studies were specified for the dietary selenium level. The overall multivariable adjusted RR indicated that the dietary selenium level was inversely associated with MetS (RR = 0.77, 95% CI: 0.63–0.95; p = 0.01) as well. Conclusion Our results suggest that the dietary iron level is positively associated with MetS, whereas a negative association between the dietary copper and selenium level and MetS is obtained. Further large well-designed prospective cohort studies are warranted to elaborate on the issues examined in this study.


INTRODUCTION
Elevated waist circumference, blood pressure, fasting blood glucose, triglycerides, and decreased high-density lipoprotein cholesterol (at least three of the five above metabolic abnormalities) are considered as the presence of metabolic syndrome (MetS) (1). MetS is a well-known attributable risk to diabetes, stroke, and coronary heart disease epidemic (2)(3)(4). Moreover, MetS increases the incidence of atherosclerotic cardiovascular disorder and complication that decreases longevity (5). However, the subjects suffering from MetS are progressively growing worldwide and the current global prevalence of MetS is between 11.6 and 62.5% (6). Although the etiology of MetS is not well understood yet, dietary factors are considered to be involved in MetS (7)(8)(9)(10)(11).
Micronutrients are important factors for cellular and biochemical functions (release of energy for synthesis and movement) (12). Iron, copper, and selenium are considered to be significant micronutrients and their dietary sources are meat, seeds, heme, tea, milk, nuts, cereals, eggs, fish, and so on (13)(14)(15)(16)(17)(18). Iron is one of the most abundant elements, which plays a significant role in various cellular processes, such as irondependent signaling, cellular respiration, DNA replication and synthesis, nucleic acid repair, and energy metabolism (19)(20)(21). Iron consumption, uptake, transfer, and storage are involved to maintain iron homeostasis (22). However, excess iron leads to inflammation and tissue damage, produces hydroxyl radicals (Haber-Weiss-Fenton reactions), which cause oxidative damage to cellular components (lipids, proteins, and DNA) (23,24). On the contrary, copper, a component of extracellular superoxide dismutase (25), is essential for iron uptake and signaling in eukaryotic organisms, energy metabolism, reactive oxygen species, and detoxification (26). In addition, copper plays an essential role in mitochondrial function and signaling involving mitophagy, bioenergetics, and dynamics, which affect cell fate by metabolic reprogramming (26). Selenium, also an essential micronutrient, is necessary to maintain the different cellular functions, such as signaling transduction pathways and immuneendocrine function (18). Moreover, selenium incorporates into selenoproteins and selenium-dependent enzymes (e.g., glutathione peroxidases), which are involved in intracellular redox regulation and modulation (27). Since oxidative stress and inflammation play a significant role in the pathophysiology of MetS (28), the dietary iron, copper, and selenium level is considered to be closely related to MetS.
A number of observational studies have been employed to investigate the associations of the dietary iron, copper, and selenium level with MetS (29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39)(40)(41)(42). However, their results are still conflicting. Thus, this meta-analysis of observational studies is employed to further investigate the above associations. It is hypothesized that the dietary iron level is positively associated with MetS, whereas the dietary copper and selenium level is inversely associated with MetS.

Search Strategy
Our meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (43). The PubMed, Web of Science, Embase, and Scopus electronic databases were searched up to October 2021 (no restriction was set for the initiate time) by using a combination of keywords that related to MetS ("metabolic syndrome"), iron ("iron, " "Fe"), copper ("copper, " "Cu"), and selenium ("selenium, " "Se"). No language restriction was set in the search strategy. We screened the titles and abstracts of all the articles and then read the full articles to identify the eligible studies.

Study Selection
The titles, abstracts, and full texts of all the retrieved studies were reviewed by two researchers independently. Disagreements were resolved by discussions. The included studies were required to meet the following criteria: (1) the study design is an observational study; (2) the outcomes include the associations of the dietary iron, copper, and selenium level with MetS; and (3) the relative risk (RR) or odds ratio (OR) with 95% CI was reported. The exclusion criteria were listed as follows: (1) duplicated or irrelevant articles; (2) reviews, letters, or case reports; (3) randomized controlled trials; and (4) non-human studies.

Data Extraction
The data were extracted by two researchers independently and disagreements were resolved by discussions. The information about the first author and year of publication, location, age, gender, sample size, study design, adjustments, exposure, category of exposure, effect estimates, and diagnostic criteria of MetS was collected. The corresponding effect estimates with 95% CIs for the highest vs. lowest dietary iron, copper, and selenium level and MetS was extracted (adjusted for the maximum number of confounding variables).

Quality Assessment
The Newcastle-Ottawa Scale (NOS) criteria for non-randomized studies were employed to assess the quality of each included study. The NOS is based on three broad perspectives: (1) the selection process of the study cohorts; (2) the comparability among the different cohorts; and (3) the identification of exposure or outcome of the study cohorts. Disagreements with respect to the methodological quality were resolved by mutual consultation.

Statistical Analyses
The RR for MetS was the outcome measure in this meta-analysis. The I 2 statistic, which measures the percentage of total variation across studies due to heterogeneity, was examined (I 2 > 50% was considered as heterogeneity). If significant heterogeneity was observed among the studies, a random-effects model was used; otherwise, a fixed-effects model was accepted. Begg's test was employed to assess the publication bias (44). A p-value of < 0.05 was considered statistically significant. Moreover, subgroup analysis for study design, diagnostic criteria of MetS, sample size, exposure assessment, type of iron, and the population was employed. After eliminating 749 duplicated articles, 1,044 articles were screened according to the titles and abstracts and 751 irrelevant studies were removed. Then, 182 reviews, case reports, or letters; 68 non-human studies; and 29 randomized controlled trial studies were excluded, respectively. Eventually, a total of 14 studies were identified for this meta-analysis. Table 1 presents the main characteristics of the included studies. These studies were published between 2010 and 2021. Eight studies were performed in Asian countries [Korea (31), China (32,35,37,39,41), and Iran (33,40)]. The other 6 studies were conducted in Brazil (29,38,42), USA (30), Spain (36), and Columbia (34), respectively. Except for the study by Bruscato (only female) (29), both the male and female participants were considered. The sample size ranged from 284 to 15,051 for a total of 55,131. The dietary iron, copper, and selenium level were assessed by a food-frequency questionnaire (FFQ) in 3 studies (35,37,40) and 24 h or 3-day recall method in 11 studies (29-34, 36, 38, 39, 41, 42). The criteria for MetS were the National Cholesterol Education Program-Adult Treatment Panel III (NCEP ATP III) (31,32,(38)(39)(40)(41), the International Diabetes Federation (IDF) (29,33,37), and the American Heart Association (AHA) (30,35,36) in 6, 3, and 3 studies, respectively. Moreover, some other criteria (34,42) were also employed for adolescents.

The Relative Risk of MetS for the Highest vs. Lowest Dietary Iron Level
The overall multivariable adjusted RR showed that the dietary iron level was positively associated with MetS (RR = 1.27, 95% CI: 1.12-1.44; p < 0.001) (Figure 2). No substantial level of heterogeneity was obtained among various studies (p = 0.097, I 2 = 44.1%). No evidence of publication bias existed according to the Begg's rank correlation test (p = 1.000).

The Relative Risk of MetS for the Highest vs. Lowest Dietary Copper Level
The overall multivariable adjusted RR showed that the dietary copper level was negatively associated with MetS (RR = 0.85, 95% CI: 0.78-0.93; p < 0.001) (Figure 3). No substantial level of heterogeneity was obtained among various studies (p = 0.391, I 2 = 5.3%). No evidence of publication bias existed according to the Begg's rank correlation test (p = 0.754).

The Relative Risk of MetS for the Highest vs. Lowest Dietary Selenium Level
The overall multivariable adjusted RR showed that the dietary selenium level was negatively associated with MetS (RR = 0.77, 95% CI: 0.63-0.95; p = 0.01) (Figure 4). No substantial level of heterogeneity was obtained among various studies (p = 0.985, I 2 = 0.0%). No evidence of publication bias existed according to the Begg's rank correlation test (p = 1.000). Table 4

DISCUSSION
A total of 14 observational studies were identified in the present meta-analysis. The pooled results showed that the dietary iron level was positively associated with MetS, whereas a negative association between the dietary copper and selenium level and MetS was obtained. The pathophysiology of MetS is involved in oxidative stress and inflammation. As a strong pro-oxidant, iron causes oxidative stress and damage to pancreatic beta cells, which decreases the synthesis and secretion of insulin, impairs insulin signaling, and then alters the glucose metabolism (45,46). Recently, ironmediated cell death (ferroptosis) has also been reported to induce cardiomyocyte damage and plays an important role in cardiovascular disorders-related pathology (47). On the contrary, copper and selenium are important antioxidants that act against oxidative stress (25,27). Copper is served as a cofactor of the copper/zinc superoxide dismutase, a protein located in both the cytosol and the mitochondrial inner membrane space to relieve the electron transport chain-generated reactive oxygen species (26). Differently from other metals, selenium works by incorporation into proteins by a cotranslational mechanism (as part of the amino acid selenocysteine) (48). Most selenium proteins participate in antioxidant defense and redox state regulation, particularly the families of glutathione peroxidases      and thioredoxin reductases (48). Copper deficiency is associated with the increased high-density lipoprotein (HDL) cholesterol level in rats (49) and blood cholesterol levels in humans (50).
Similarly, selenium is also considered to prevent high-fat dietinduced hyperglycemia and liver damage in rats (51) and type 2 diabetes mellitus and cardiovascular disease in humans (52).
The above may significantly account for the major findings of this study. Ferritin, a ubiquitous intracellular protein, is important in the regulation of iron homeostasis (53). Similarly, a meta-analysis of the observational study suggested that the increased ferritin level is positively associated with MetS (54). Moreover, iron chelation therapy (reduce serum ferritin level) was associated with improved serum glucose and HDL levels (55). The phlebotomy with a reduction of body iron could significantly improve cardiovascular risk and glycemic control in patients with MetS (56). The above findings are decent supplements to ours.
Heme iron exists in most animal foods, whereas the rest in animal or plant food is non-heme iron (57). Non-heme iron is absorbed less efficiently than heme iron (∼5% non-heme iron and 25% heme iron are absorbed from diet) (41). The epidemiological data indicate that the effect of heme and nonheme iron on MetS may vary from the regional variety (39,40) and heme iron alone cannot reflect the iron status (58,59). Indeed, our results with regard to heme and non-heme iron are quite different. The specification of the iron subtype should be considered in further study.
Interestingly, our findings are only confirmed in the NCEP ATP III/IDF diagnostic criteria and > 1,000 sample-sized studies. It is speculated that MetS diagnostic criteria or sample size may influence the results of this study. Moreover, the negative relationship between the dietary iron and selenium and MetS is only specified in the 24 h or 3-day recall method and adult population. Although the number of studies for FFQ and adolescents is limited, the recall method might be precise to reflect the issues and some age-related differences with the dietary pathology of MetS cannot be fully excluded. In addition, the issue of race should also be noted. The corresponding findings only exist in Asians, but not in Americans (with regard to the dietary iron and selenium). Our results suggest that the potential effect of racial variation should not be ignored either.
Our findings can be incorporated into the daily lives of subjects suffering from MetS. The programs to build awareness with collaboration between physicians and nutritionists should be encouraged in the future. For example, reinforce the dietary education in MetS subjects: avoid the dietary iron overdose or copper/selenium deficiency. Nevertheless, the toxicity of excess copper intake should also be emphasized. Excess copper intake is reported to induce oxidative stress, damage to the mitochondrial, contributes to apoptosis, DNA damage, and inflammatory responses (60,61). Therefore, careful validation by randomized controlled trial/prospective cohort study is still needed before its clinical application.
This study has several strengthens. To begin with, this is the first meta-analysis of observational studies on the associations of the dietary iron, copper, and selenium level with MetS. In addition, the included studies are analyzed based on the adjusted results and large samples. Moreover, the limited heterogeneity level may reflect the decent reliability of our results. Finally, our findings might provide significant information to better consider the dietary effects on MetS.
The limitations of this study should also be acknowledged. First, only 2 prospective cohort studies were identified totally due to the limited relevant literature (causal relationships could not be obtained). Second, the classification of exposure varies greatly among individuals. Third, the adjusted factors and definition of MetS were not uniform. Forth, one included study has combined the data for the dietary iron and red meat as a whole (40). These limitations may weaken the significance of this study.

CONCLUSION
Our results suggest that the dietary iron level is positively associated with MetS, whereas a negative association between the dietary copper and selenium level and MetS is obtained. Further large well-designed prospective cohort studies are warranted to elaborate on the issues examined in this study.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
YZ and JD conceived the idea and drafted this manuscript and selected and retrieved relevant papers. ZL and QL performed the statistical analysis. JL and HG assessed each study. YZ was the guarantor of the overall content. All the authors revised and approved the final version of the manuscript.