Metabolic syndrome (MetS) is a pathophysiological state and cluster of ≥3 different cardiovascular risk factors, including abdominal obesity, insulin resistance, elevated blood pressure and dyslipidemia (1). Parallel to rapid economic development and changes in lifestyle, the prevalence of MetS is continuing to rise worldwide, affecting approximately a quarter of the adult population, and has become an important public health issue (2, 3). The Chinese National Nutrition and Health Surveillance (2010–2012) reported that the prevalence of MetS in Chinese adults aged 20 years or older was 18.7% and an estimated 189 million adults living with MetS in China (4). Likewise, the prevalence of MetS in U.S. adults reached 34.2% during 2007–2012, according to the National Health and Nutrition Examination Survey (NHNES) (5). Meanwhile, MetS has also been considered to be an important risk factor for many chronic non-communicable diseases (NCDs), such as type 2 diabetes, stroke, cardiovascular disease (6). Given the high morbidity and its strong link to some NCDs, early prevention of MetS is of almost importance. Although the precise etiology of MetS is not completely understood, known risk factors for MetS included genetic predisposition, smoking, alcohol consumption, sedentary lifestyle and high sugar or fat diets (6, 7).

Over the past several decades, abundant evidence has shown that dietary factors play the important role in the development of MetS (8). Previous studies have specially examined the associations between the consumption of specific foods or nutrients and risk of MetS (9–11). However, less attention has been paid for the association between different degree of food processing and MetS. Recently, the global consumption of ultra-processed foods (UPF) has been rising rapidly in some middle- or high-income countries, contributing to 25%~60% of daily energy intake (12–14). During the SARS-CoV-2 pandemic, many people were at risk of financial restrictions, which could easily translate into choosing UPFs, which are typically high in energy density, added sugars, salt, saturated and trans-fats, as well as low in dietary fiber, protein, vitamins and minerals (15). Apart of nutritional composition, UPFs are highly palatable, convenient, long shelf life and affordable (14). Thus, UPF intake has garnered considerable attention from scientific researchers. To date, some observational studies have explored the potential associations between UPF consumption and various adverse health outcomes, including obesity, type 2 diabetes, hypertension, and cardiovascular disease (16–19). Of note, several recent systematic review and meta-analyses have also been published to clarify the associations between consumption of UPF and type 2 diabetes, hypertension and all-cause mortality (20–22). Consequently, these studies provided fairly consistent support for the positive associations of UPF consumption with adverse health outcomes. After 2011, some epidemiological studies have also been carried out to explore the direct relationship between consumption of UPF and MetS risk (3, 23–30), but the results remain controversial. So far, five published studies have reported that higher intake of UPF was associated with an increased risk of MetS (23, 25–29), while others showed a null association (3, 24, 30). For instance, during a median follow-up time of 6 years (IQR: 3.0–9.0), an analysis of China Health and Nutrition Survey (CHNS) showed that higher long-term UPF consumption was associated with an increased risk of MetS in Chinese adults (RR: 1.17, 95%CI: 1.01–1.35) (23). Similarly, in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), Canhada et al., also found a positive association between consumption of UPF and the risk of MetS (RR: 1.19, 95%CI: 1.07–1.32) (25). In contrast, a recent prospective cohort study of 896 Brazil adults found no significant association between UPF consumption and risk for MetS (RR: 1.00, 95%CI: 0.99–1.01) (3). Indeed, previous meta-analyses have consistently shown a strong association between specific types of ultra-processed foods, such as soft drink, processed meat and risk of MetS (31, 32). Meanwhile, Lane et al. also published a systematic review and meta-analysis of observational studies reporting the association between consumption of ultraprocessed food and chronic non-communicable diseases (33). However, the above-mentioned meta-analysis regarding the relationship between UPF consumption and MetS risk was limited by a relatively limited number of studies available for inclusion at the time of publication (n = 4). Since then, several new epidemiological studies have also been published evaluating the association of UPF consumption with the risk of MetS (3, 23–25, 30). Furthermore, Lane et al.'s work has some limitations. For example, they only included four studies, which did not perform subgroup analyses to explore the potential sources of heterogeneity. Also, because the previous meta-analysis was based on four included studies, their results cannot be generalized to other populations. Therefore, we undertook a comprehensive systematic review and meta-analysis with the purpose to clarify the relationship between UPF consumption as defined by the NOVA framework and risk of MetS.

In the present study, the pooled results illustrated that high consumption of UPF was significantly associated with an increased risk of MetS. Nonetheless, the results of this meta-analysis must be interpreted with caution due to the high heterogeneity among the included studies. Moreover, subgroup analyses showed the positive association between high consumption of UPF and risk of MetS was more robust in cross-sectional studies and the subgroups of study quality <7. Likewise, the results of sensitivity analysis indicated that Magalhãesas et al.' study might be the source of heterogeneity. To the authors' knowledge, this is the first comprehensive systematic review and meta-analysis to assess the relationship between consumption of UPF and MetS risk. Our findings confirm the positive association results of previous studies and add to the growing evidence for the role of UPF consumption in diet-related chronic diseases, including MetS.

With economic development and changes in lifestyle, the prevalence of MetS is increasing around the world (2). It is reported that the standardized prevalence of MetS was 31.1% in 2015–2017, and approximately a third of adults have MetS in China (7). Given the high prevalence and burden on public health, more attention is now needed to prevent the occurrence of MetS. As we all know, dietary factors, as a component of lifestyle, play a key role in the prevention of MetS (8). Over the past decades, the food supply industries have increased the commercialization of UPF, and usual diets have also shifted toward the consumption of UPF, characterized by high in energy density, added sugar, saturated and trans-fats, as well as lower in fiber (41). In recent decade, the global consumption of UPF has increased rapidly, contributing about 25%~60% of total daily energy intake in some high- and middle-income countries (19). However, there are considerable differences in UPF consumption between developing and developed countries (42, 43). For example, UPF consumption has already accounted for more than 50% of total energy intake in countries such as United States, Canada and United kingdom (44–46). By contrast, the median contribution of UPF intake to the total daily energy was 10.5% in China (47). Although the overall consumption of UPF in China is currently lower than that observed in some high-income countries, the increased trend was dramatic, especially in highly urbanized areas (48). According to China's most recent census in 2020, the proportion of individuals aged 65 years and above in China has reached nearly 14%, indicating that China is becoming an elderly society (49). UPF consumption is associated with poorer diet quality (i.e. low dietary fiber, fruits and vegetables intake), which can lead to frailty. A recent cross-sectional study conducted by Zupo et al., offered the evidence of food processing contribution to poor nutrition in the aging population (50). Hence, the influence of UPF consumption on chronic diseases has garnered considerable attention from researchers. Until now, some previous observational studies have shown that higher consumption of UPF is significantly associated with adverse health outcomes, such as the increased risks for obesity, hypertension, diabetes, metabolic syndrome and cancers (16, 20–22). However, epidemiological evidence regarding the effect of consumption of UPF on the risk of MetS is limited and inconclusive (13, 23–30). In this study, we found a significant positive relationship between UPF intake and MetS risk, although there was evidence of high heterogeneity across studies (I² = 85.0%; P < 0.001). This is in agreement with findings from some previous studies reporting that high UPF consumption is associated with an increased risk of MetS (23, 25, 27–29). In the China Health and Nutrition Survey (CHNS), Pan et al. reported that higher long-term UPF consumption was associated with an increased risk of MetS in Chinese adults (RR = 1.17; 95% CI: 1.01–1.35) over a follow-up period of 29 years (23). Similarly, a recent study from the Brazilian longitudinal study of adult health also showed that higher consumption of UPFs was associated with an increased risk of MetS (RR = 1.19; 95%CI: 1.07- 1.32) (25). However, contrary to our finding, a recent cross-sectional study conducted by Barbosa et al., showed no significant relationship between UPF consumption and MetS risk (RR = 1.09, 95%CI: 0.89–1.32) (24). The differences in assessment of UPF, amounts and types of UPF of different populations, and definitions of UPF consumption levels, at least in part, explain the discrepant results between the different studies (23). Moreover, the inconsistency may also result from the significant difference in the statistical power. Previous studies undertaken in Lebanon and Brazil have documented a small number of MetS, which results in the limited power (26, 30). Furthermore, it is worth noticing that these inconsistent findings may be attributed to the differences in sociodemographic factors, such as age, race, and income, which have been found to be associated with UPF consumption (12). Researchers have proposed several potential plausible mechanisms that may explain the observed positive association between UPF consumption and risk of MetS, although conflicting results were reported. First, UPFs tend to be energy dense and often have high amounts of saturated fat, trans-fat and added sugars. Previous studies have demonstrated that excessive consumption of saturated fat and added sugars are associated with higher risk of MetS (51). Second, the detrimental effect of UPF consumption on MetS may also partly be attributed to lower intakes of minimally processed foods such as whole grains, vegetables, fruits, which are shown to be inversely associated with risk of MetS (52). Third, beyond the poor nutritional composition, other components common present in UPF, such as emulsifiers and artificial sweeteners have been implicated in changes in gut microbiota, glucose intolerance and insulin resistance, which could lead to the progression of MetS (53–55). Fourth, UPF may be contaminated with packaging contact materials, such as phthalates and bisphenol A, are involved in endocrine disruption and insulin resistance (56). A previous meta-analysis of 33 epidemiological studies on Bisphenol A and risk of cardiometabolic disorders showed that higher concentration of Bisphenol A was associated with an increased risk of cardiometablic outcomes (57). Finally, food processing (particularly heat treatments) is largely associated with the loss of physical and structural characteristics of the food substrate, and has been reported to be associated with lower satiety potential and higher glycemic responses (58). Given the above, these plausible mechanisms could explain the positive association between UPF consumption and risk of MetS.

In our analyses, it is important to point out that there was evidence for high between-study heterogeneity among all included studies (I² = 85.0%, P < 0.0001). Although between-study heterogeneity is common in meta-analysis (59), exploring the potential sources of heterogeneity is the essential. Thus, we performed subgroup analyses based on study design (cohort or cross-sectional studies), exposure assessment (FFQ or 24 h dietary recall), study quality (≥7 or <7), mean age (≥55 or <55), study area (developed countries or developing countries) and sample size (<5,000 or ≥5,000) to address sources of heterogeneity. The results of subgroup analyses showed that heterogeneity might be mainly due to the difference in study design and exposure assessment. When the results were stratified by exposure assessment, the heterogeneity among included studies decreased from 85.0 to 25.7%. There are several possible explanations for the high heterogeneity. First, different levels of UPF consumption in included studies may explain, to some extent, the high between-study heterogeneity. Second, six of the included studies were cross-sectional in design. Given the observational nature of included studies, we cannot assume the causality of the observed association. Likewise, the results may be susceptible to recall bias, resulting from dietary survey methods (i.e. FFQs and 24-h dietary recalls) in the observational studies. Third, considering that UPF consumption varied across different populations, and that definitions of UPF consumption levels varied in different studies, despite the RRs or ORs were all from the highest category (taking the lowest category as the reference), different populations and studies may have different definitions of UPF consumption levels, thereby causing the substantial heterogeneity. Fourth, different models used to control potential confounding variables in included studies may explain the heterogeneity observed in our analyses. There was an inconsistent adjustment for potential confounding variables in the included studies. As a consequence, it is inevitable that we have high levels of heterogeneity when combining studies. Finally, the considerable heterogeneity persisted in subgroup analyses, indicating the presence of other unknown confounding factors.

This study has several strengths and limitations. First, to our knowledge, this is the first comprehensive systematic review and meta-analysis to evaluate the relationship between consumption of UPF and the risk of MetS. Our findings add to the growing evidence for the impact of UPF consumption on MetS and help inform public policy for the prevention and management of MetS. Second, a rigorous selection of articles was conducted based on the predetermined inclusion criteria, with the inclusion only of studies in which the classification of UPFs faithfully followed the characteristics proposed by the NOVA system. Third, MetS cases were ascertained through medical records, reducing the risk of misclassification. Fourth, the quality of the included studies was moderate to high, and the reported RRs were multivariate and adjusted for some known confounders, such as sex, age, physical activity and total energy intake. Meanwhile, we also performed subgroup and sensitivity analyses to explore the potential sources of heterogeneity. Fifth, no signs of publication bias were evident in the funnel plot, and the Begg's test for publication bias was non-significant. Despite the above-mentioned strengths, this study also has some limitations that should be acknowledged. First, although our findings show a positive relationship between UPF consumption and MetS risk, the majority of included studies are cross-sectional in design, which are limited by the potential for reverse causality. Second, five of the included studies used the FFQs that were not specifically designed to assess the NOVA classification groups, which might have led to an under- or over-estimation of the size of the observed associations. Likewise, information bias, as a consequence of self-reported data on dietary intake, might have occurred. Third, even though all of included studies in this meta-analysis have adjusted for a wide range of important confounders, residual confounding from unmeasured factors cannot be completely ruled out. Also, there was also an inconsistent adjustment for potential confounders in the all included studies. Consequently, the data included in this meta-analysis might suffer from differing degrees of completeness and accuracy. Fourth, high heterogeneity was found in this meta-analysis, which might have distorted the reliability of our results. Although we performed subgroup and sensitivity analyses to explore the potential sources of heterogeneity, we were unable to adequately ascertain and explain the sources of inter-study heterogeneity. Finally, given the limited number of included studies and the fact that most of them were performed in developing countries, caution is advised in the interpretation and extrapolation of our findings.

In summary, findings from this study suggest that higher consumption of UPF is significantly related to an increased risk of developing MetS. Our findings add valuable evidence to the existing literature showing a positive relationship between UPF consumption and risk of MetS, and highlight the importance of limiting UPF consumption in decreasing the modifiable burden of MetS. Thus, active discouragement of UPF consumption should be considered as part of MetS prevention strategies. Considering the high level of evidence provided in the included studies, more well-designed prospective studies, particularly in different geographic regions and settings, are warranted to further confirm these findings.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

CS conceived and designed the systematic review, meta-analysis, and interpreted the results. XZ and JZ acquired the data. LS and QZ performed the statistical analysis and obtained the funding. LS conceived the idea and drafted this manuscript. All authors critically revised the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.