Vitamin C and Metabolic Syndrome: A Meta-Analysis of Observational Studies

Background: The association between vitamin C and metabolic syndrome (MetS) has been evaluated in several epidemiological studies with conflicting results. This meta-analysis was therefore employed to further investigate the above issue. Methods: The observational studies on the associations of dietary and circulating (serum and plasma) vitamin C levels with MetS were searched in the PubMed, Web of Science, and Embase database up to April 2021. The pooled relative risk (RR) of MetS for the highest vs. lowest dietary and circulating vitamin C levels and the standard mean difference (SMD) of dietary and circulating vitamin C levels for MetS vs. control subjects were calculated, respectively. Results: A total of 28 observational studies were identified in this meta-analysis. Specifically, 23 studies were related to the dietary vitamin C level. The overall multivariable-adjusted RR demonstrated that the dietary vitamin C level was inversely associated with MetS (RR = 0.93, 95% CI: 0.88–0.97; P = 0.003). Moreover, the overall combined SMD showed that the dietary vitamin C level in MetS was lower than that in control subjects (SMD = −0.04, 95% CI: −0.08 to −0.01; P = 0.024). With regard to the circulating vitamin C level, 11 studies were included. The overall multivariable-adjusted RR demonstrated that the circulating vitamin C level was inversely associated with MetS (RR = 0.60, 95% CI: 0.49–0.74; P < 0.001). In addition, the overall combined SMD showed that the circulating vitamin C level in MetS was lower than that in control subjects (SMD=-0.82, 95%CI: −1.24 to −0.40; P < 0.001). Conclusions: Current evidence suggests that both dietary and circulating vitamin C level is inversely associated with MetS. However, due to the limitation of the available evidence, more well-designed prospective studies are still needed.


INTRODUCTION
Metabolic syndrome (MetS) is defined as the presence of the following five metabolic abnormalities (at least three): elevated waist circumference, blood pressure, fasting blood glucose, triglycerides, and decreased high-density lipoprotein cholesterol (1). Affecting 25% of the population in the developed world, MetS has been considered an important public health issue in parallel to obesity and diabetes (2). Although the etiology of MetS is not well-understood yet, dietary factors are thought to be involved in MetS (3).
Vitamin C, an essential water-soluble micronutrient traditionally utilized to prevent and treat scurvy (also known as ascorbic acid), is one of the most common antioxidants. Fruit and vegetable consumption, which are equipped with abundant vitamin C, have been demonstrated to be inversely associated with MetS in our previous study (4). Moreover, vitamin C consumption is associated with a lower risk of type 2 diabetes (5) and hypertension (6). Fundamentally, the induction of vitamin C deficiency could lead to a phenotype characterized by insulin resistance, weight gain, dyslipidemia, and hepatic steatosis (7). Above all, it is speculated that vitamin C levels may be negatively associated with MetS.
To the best of our knowledge, a number of observational studies have examined the associations of dietary and circulating vitamin C level with MetS . However, no final conclusion can be obtained. Thus, the present meta-analysis of observational studies was employed to investigate the issue further. It was hypothesized that the dietary and circulating vitamin C was 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 (36). The PubMed, Web of Science, and Embase electronic database were searched during April 2021 by using a combination of keywords and in-text words related to MetS ("metabolic syndrome") and vitamin C ("vitamin C" and "ascorbic acid"). No language and MetS diagnostic criteria restrictions were set in the search strategy. The titles and abstracts of all articles were first screened. Then, the full articles were read to identify the eligible studies. Moreover, the reference lists for the retrieved articles were reviewed to include additional studies.

Study Selection
Two researchers (YZ and JD) reviewed the titles, abstracts, and full texts of all retrieved studies independently. Disagreements were resolved by discussions and mutual consultations. The included studies were required to meet the following criteria: (1) the study design was an observational study; (2) the outcomes included the associations of dietary and circulating vitamin C level with MetS; and (3) the relative risk (RR), odds ratio (OR), or standard mean difference (SMD) with 95% confidence interval (CI) were 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
Two researchers extracted the data (YZ and JD) independently, and disagreements were resolved by consensus. The information about the first author, year of publication, location, age, gender, sample size, study design, adjustments, exposure, category of exposure, effect estimates, adjustments, and diagnostic criteria of MetS, were collected. The corresponding effect estimates with 95% CIs for the highest vs. lowest dietary levels and the circulating vitamin C level were extracted (adjusted for the maximum number of confounding variables). Moreover, the dietary and the circulating vitamin C levels (mean ± SD) were also extracted to calculate the SMD (MetS vs. control).

Quality Assessment
Quality assessment was conducted according to the Newcastle-Ottawa (NOS) criteria for non-randomized studies, which is based on three broad perspectives: the selection process of study cohorts, the comparability among different cohorts, and the identification of exposure or outcome of study cohorts. Disagreements with respect to the methodological quality were resolved by discussion and mutual consultation. Studies in other languages were translated into English for evaluation. A study awarded seven or more stars were considered high-quality (37), which was the basis for subgroup analysis for study quality.

Statistical Analyses
The RR for MetS and SMD for dietary and circulating vitamin C levels were the outcome measures in the present study. The I 2 statistic, which measures the percentage of total variation across studies due to heterogeneity, was examined (I 2 > 50% was considered heterogeneity). If significant heterogeneity was observed among the studies, the random-effects model was used; otherwise, the fixed effects model was accepted. Begg's test was employed to assess the publication bias (38). A p < 0.05 was considered statistically significant. Moreover, subgroup analysis for study design, diagnostic criteria of MetS, geographical region, sample size, exposure assessment, study quality, adjustment of BMI, and physical activity was conducted for RR analysis. In addition, subgroup analysis for diagnostic criteria of MetS, geographical region, sample size, study quality, and exposure assessment was employed for SMD analysis. Of note, Kim et al. separated their effect estimates by high and low physical activity (PA), which served as two independent datasets (22). 485) were retrieved during the initial literature search. After eliminating 202 duplicated articles, 603 articles were screened according to the titles and abstracts and 451 irrelevant studies were excluded. Then, 47 reviews, case reports, or letters; 49 nonhuman studies; and 28 randomized control trials studies were removed, respectively. Thereafter, three additional studies were acquired from the reference lists for the retrieved articles (33)(34)(35). Moreover, three studies were excluded for duplicated data (39)(40)(41). Eventually, a total of 28 studies were selected for this meta-analysis.

RR of MetS for the Highest vs. Lowest Dietary Vitamin C Category
The overall multivariable-adjusted RR demonstrated that the dietary vitamin C level was negatively associated with MetS (RR = 0.93, 95% CI: 0.88-0.97; P = 0.003) (Figure 2). A substantial level of heterogeneity was obtained among various studies (P = 0.003, I 2 = 54.5%

SMD of the Dietary Vitamin C Level for MetS vs. Control Subjects
The overall combined SMD showed that the dietary vitamin C level in MetS was lower than that in control subjects (SMD = −0.04, 95% CI: −0.08 to −0.01; P = 0.024) (Figure 3). A substantial level of heterogeneity was obtained among the various studies (P < 0.001, I 2 = 69.0%). No evidence of publication bias existed according to Begg's rank-correlation test (P = 0.314).
The results of the subgroup analysis are presented in Table 2.

RR of MetS for the Highest vs. Lowest Circulating Vitamin C Category
The overall multivariable-adjusted RR demonstrated that the circulating vitamin C level was negatively associated with MetS (RR = 0.60, 95% CI: 0.49 to 0.74; P < 0.001) (Figure 4).
No substantial level of heterogeneity was obtained among various studies (P = 0.249, I 2 = 22.7%). No evidence of publication bias existed according to Begg's rank-correlation test (P = 0.536). The results of subgroup analysis were presented in Table 3. The above findings were confirmed in

SMD of the Circulating Vitamin C Level for MetS vs. Control Subjects
The overall combined SMD showed that the circulating vitamin C level in MetS was lower than that in control subjects (SMD = −0.82, 95% CI: −1.24 to −0.40; P < 0.001) (Figure 5). A substantial level of heterogeneity was obtained among the various studies (P < 0.001, I 2 = 98.3%). No evidence of publication bias existed according to Begg's rank-correlation test (P = 0.076). The results of subgroup analysis are presented in Table 4.

DISCUSSION
In the present meta-analysis, a total of 28 observational studies were identified, and the pooled analysis showed that both the dietary and the circulating vitamin C level were inversely associated with MetS. Since oxidative stress and inflammation play a significant role in the pathophysiology of MetS (42), the underlying mechanism behind the negative association between vitamin C and MetS can be listed as follows: on the one hand, vitamin C serves as a strong antioxidant, which prevents other compounds from being oxidized (42). Vitamin C can donate electrons, scavenge harmful free radicals, and leave the ascorbyl radical (relatively stable and unreactive) (43,44). Indeed, antioxidants could lower blood pressure by oxidation of cGMP-dependent protein kinase (45). Several meta-analysis studies have also indicated a potential beneficial effect of vitamins antioxidant on type 2 diabetes (46, 47). A higher intake of antioxidant vitamins and serum total antioxidant status from antioxidant supplementation were associated with a decreased waist circumference and low-density lipoprotein to high-density lipoprotein-cholesterol ratio (40). On the other hand, the neutrophil is an important regulatory marker for acute and chronic inflammation. The hyperplasia and hypertrophy of adipose tissue, the main source of various inflammatory mediators, is closely associated with the MetS-associated inflammation (48). The neutrophil chemotaxis to adipose tissue (with subsequent phagocytosis of microbes and clearance by macrophages) could therefore reduce inflammation. Indeed, vitamin C has the potential to improve neutrophil chemotaxis in both human and animal studies (49)(50)(51). In addition, patients with impaired neutrophil bacterial phagocytosis can be significantly improved by vitamin C supplementation (longlasting clinical improvement) (52). Importantly, vitamin Cdeficient neutrophils were not phagocytosed by macrophages in vitro and persisted at inflammatory loci in vivo (53). Taken together, vitamin C is able to alleviate the inflammatory response by influencing neutrophil chemotaxis in response to inflammatory mediators, enhancing phagocytosis of microbes by neutrophils and neutrophil clearance by macrophages (52).
The effect of vitamin C on MetS has been extensively investigated in experimental animal and clinical studies. Bilbis et al. found that 4-week supplementation of vitamin C could decrease the weight gain, blood pressure, glucose, insulin, insulin resistance, total cholesterol, triglycerides, low-density lipoprotein  cholesterol, and very low-density lipoprotein cholesterol in a salt-loaded animal model (54), indicating a strong biological effect of vitamin C on the circulating profile in MetS. Moreover, unlike animals, humans are dependent upon dietary vitamin C due to mutations in L-gulono-γ-lactone oxidase (Gulo, the final enzyme for vitamin C biosynthesis). The induction of vitamin C deficiency in Gulo-/-mice was associated with a phenotype characterized by insulin resistance, weight gain, dyslipidemia, and hepatic steatosis (7). Indeed, this animal model was humanized and necessarily reflected the issue. Importantly, Abd indicated that the combination of vitamin C and Lmethionine had the most beneficial effect on hyperglycemia, dyslipidemia, abnormal coagulation indices, and oxidative stress in the alloxan-induced diabetes model. Asynergistic biological effects may exist between vitamin C and others, which indicated that the combination therapy strategy for vitamin C is promising (55). With regard to the clinical evidence, Farag et al. found that vitamin C supplementation could result in a significant reduction in BMI, and the combination of physical activities [exercise alone was reported to be beneficial to MetS (56, 57)] and vitamin C supplements may further improve systolic blood pressure and serum levels of total cholesterol in MetS patients (58,59).
Taken together, the current experimental and clinical evidence strongly supports the potential beneficial effect of vitamin C on MetS. Moreover, the vitamin C combination therapy (with Lmethionine or exercise) seems to be a promising strategy for the management of MetS, which might be recommended clinically. Interestingly, the inverse relationship between dietary vitamin C level and MetS was only obtained in crosssectional studies. However, the number of cohort studies is rather small (only two), which may inevitably reduce the reliability. Although vitamin C may be beneficial to the components or complications of MetS (e.g., decreased blood pressure, glucose, insulin, insulin resistance, etc.), it is not necessarily a reflection of MetS prevention. One cohort study combined vitamin C and B 1 as a whole, and the effect of vitamin B 1 could not be specified (27). Interestingly, the inconsistent result with regard to diagnostic criteria of MetS and exposure assessment was acquired. It was speculated that NCEP ATP III criteria and recall method seems to be more precise and suitable for vitamin C evaluation. With regard to the SMD analysis for the dietary vitamin C levels, the findings disappeared in small samplesized (<1,000) studies. Indeed, large sample-sized studies may be more reliable to address the issues. Importantly, no study has specified the dietary season variation, and our results might be influenced by the seasonal variability of vitamin C levels. Taken together, more well-designed prospective cohort studies are still needed.
Our study has several strengthens. To begin with, this is the first meta-analysis of observational studies on the associations of dietary and circulating vitamin C levels with MetS. In addition, the included studies are analyzed based on the adjusted results and large samples. Finally, our findings are consistent with the current corresponding experimental and clinical studies. Above all, our study might provide helpful information to better consider the effect of vitamin C on MetS.
On the other hand, we should also acknowledge the limitations of this study. (1) The substantial level of heterogeneity might have distorted the reliability of our results. (2) Due to the limited relevant literature, only three prospective cohort studies were identified totally (causal relationships could not be obtained). (3) The classification of exposure may vary greatly among individuals. (4) The selection of adjusted factors and definition of MetS was not uniform. (5) One included study has combined the effect estimates for vitamin C and vitamin B 1 as a whole (27), some issues could not be addressed. (6) The seasonal variation in vitamin C level cannot be considered in this study. These limitations may weaken the significance of our study.

CONCLUSIONS
Current evidence suggests that both dietary and circulating vitamin C levels are inversely associated with MetS. However, due to the limitation of the available evidence, more well-designed prospective studies are still needed.

DATA AVAILABILITY STATEMENT
Publicly available datasets were analyzed in this study. This data can be found at: PubMed, Web of Science, Embase.

AUTHOR CONTRIBUTIONS
YZ, HG, and JD conceived the idea, performed the statistical analysis, and drafted this meta-analysis. YZ and JD selected retrieved relevant papers. QL and YL assessed each study. YZ was the guarantors of the overall content. All authors revised and approved the final manuscript.