With an estimated prevalence of 25 to 45%, non-alcoholic fatty liver disease (NAFLD) is one of the most common chronic liver diseases globally, putting an enormous burden on the health and economic systems (1, 2). Metabolic Associated Fatty Liver Disease (MAFLD) is considered to be a more appropriate concept than NAFLD given the growing prevalence of metabolic dysfunction (3). Patients with MAFLD were diagnosed with a definite criterion instead of exclusion conditions, which is distinguished from NAFLD. Several recent studies have pointed out that MAFLD has a higher prevalence and a more severe liver injury compared with NAFLD (4, 5). There is no doubt that MAFLD is a major health problem worldwide and will likely become a serious public health issue if the government does not act. There is currently no clinically approved medication with proven effectiveness for MAFLD. Lifestyle modifications such as increasing exercise and changing diet may be useful in the early stage of MAFLD (6). Therefore, investigating the potential dietary strategies for MAFLD could contribute to the prevention and progression of this disease.

The pathogenesis of NAFLD results in a series of ‘multiple hits’ hypothesis (7). Although the mechanism of NAFLD is complex, containing abdominal adiposity, insulin resistance, inflammation, and other factors. Notably, oxidative stress, which generates excessive reactive oxygen species (ROS), is involved in the entire development of NAFLD (8). Flavonoids are natural compounds found in many fruits, vegetables, red wine, and tea. There are several subclasses of flavonoids, such as anthocyanins, isoflavones, flavones, flavanones, flavan-3-ols, and isoflavones. Due to the effect of attenuating oxidative stress and inflammation, flavonoids present an attractive therapy for hepatic steatosis (9). Previous studies showed that the prevalence of metabolic-related diseases was lower with higher flavonoid intake (10–12).

Several studies of in vitro and animal model experiments confirmed the potential effect of flavonoids in the prevention and progression of NAFLD (13). Ying et al. found that high-fat diet-induced rats treated with flavonoids could alleviate insulin resistance and inflammation, and significantly enhance the levels of antioxidants (14). A clinical trial reported that patients who received supplementation of anthocyanin, a subclass of flavonoid, could markedly decrease plasma levels of ALT and glucose in NAFLD individuals (15). The NHANES 2005-2010 result showed an inverse relationship between flavonoid intake and the risk of NAFLD (16). Moreover, a prospective study of the Chinese population pointed out that higher flavonoid intake reduced the risk of NAFLD progression in elderly obese or overweight individuals (17). These results suggest that flavonoids could be beneficial to NAFLD patients. Nevertheless, previous NAFLD studies could not accurately depict the association between flavonoid intake and MAFLD because MAFLD is more heterogeneous than NAFLD. To our knowledge, there was no clinical evidence linking flavonoids and subclasses intake to MAFLD. Thus, we aim to investigate the association between flavonoids and subclasses intake with MAFLD prevalence in this cross-section study.

The overall weighted prevalence of MAFLD was 41.93%, suggesting that the soaring epidemic of this disease has caused a serious burden on public health. Hence, the primary prevention of MAFLD through dietary and behavioral modification is an urgent need. This is the first study reporting the association between flavonoid and subclasses intake and the risk of MAFLD in U.S. adults based on NHANES. Although the higher total flavonoid intake was not associated with a lower prevalence of MAFLD, the anthocyanidin intake had an inverse association with the risk of MAFLD. Moreover, the protective effect of higher anthocyanidin intake was more significant in male, Non-Hispanic White, Non-Hispanic Asia, overweight or obesity, and individuals combined with hyperlipidemia. Higher isoflavone intake may have a benefit to prevent MAFLD in individuals of younger, Non-Hispanic Black, Non-Hispanic Asia and higher HEI-2015 scores. The racial groups had an interaction effect with anthocyanin intake on the risk of MAFLD. There was a positive correlation between HDL and anthocyanidin intake, and a negative correlation between FPG and isoflavone intake. These results promise that dietary modification of increasing anthocyanidin and isoflavone intake provides a possible prevention approach for MAFLD, especially in particular groups.

The pathogenic driver of MAFLD is characterized by the liver’s inability to handle the energy substrates leading to toxic lipid accumulation (25). The dysfunction in lipid metabolism leads to the activation of oxidative stress, stimulation of ROS, the release of pro-inflammatory cytokines and apoptotic cell death, and subsequent stimulation of inflammation and fibrogenesis (26). Based on previous evidence from in vitro and in vivo experiments, flavonoids could be beneficial in inhibiting the occurrence and progression of NAFLD, and the mechanism is mainly related to their anti-inflammatory and antioxidant effect (16, 17). In our study, a higher level of flavonoid intake was benefit to improve the MAFLD-related indicators, such as HDL, ALT, TG, and Hs-CRP, however, there was no difference in total flavonoid intake between non-MAFLD and MAFLD individuals. The findings implied that total flavonoid intake is not an independent risk factor for MAFLD and that various subclasses of components may have different effects. More emphasis should be placed on the association between flavonoid subclasses and MAFLD.

According to the FNDDS database, there are six subclasses of flavonoids as follows have been evaluated: anthocyanidins, flavanones, flavones, flavan-3-ols, flavanols, and isoflavones. Our study showed that two subclasses of flavonoids, anthocyanidins and isoflavones, could be potential protective factors in the risk of MAFLD, and this difference was significant in anthocyanidins. This reason might be the higher consumed of anthocyanin intake in the daily diet of the U.S. population (27). The most common source of anthocyanidins are berries, appealing vegetables, and cereals. Previous studies reported that anthocyanidin treatment could increase insulin secretion and decrease the level of plasma LDL and hepatic TG contents as antioxidants in obese mice or rats (28, 29). Blueberry polyphenols improved obesity and metabolic alterations in high-fat diet induced mice through modulation of the gut microbiota (30). Further studies illustrated that cherry anthocyanidins are protected against NAFLD via activating the autophagy pathway (31). In summary, anthocyanins could prevent and improve NAFLD, which has been confirmed in vitro and animal experiments. However, few clinical studies reported the effects of anthocyanidins on NAFLD/MAFLD. A randomized double-blind clinical study showed that a supplement of anthocyanin decreased liver injury biomarkers and improved insulin resistance and clinical evolution of NAFLD patients (32). Our results show that anthocyanin intake is significantly higher in patients with non-fatty liver and negatively correlated with the risk of MAFLD. This result is an important supplement to the correlation between anthocyanins and MAFLD.

We discovered that the higher anthocyanidin intake with a lower risk of MAFLD was obvious in men, Non-Hispanic White, Non-Hispanic Asia, overweight/obesity and hyperlipidemia participants. A large number of studies have proved that man, age, and obesity are independent risk factors associated with hepatic steatosis (33–35). In animal and human studies, young and female individuals are protected in dysmetabolism resulting from the ability to convert excess fatty acids into ketone body production and enhanced sex-specific browning (36). It’s possible that estrogen deficiency exacerbated hepatic steatosis and inflammatory changes in mice and zebrafish models (37, 38). Obesity was closely associated with the prevalence and severity of NAFLD and advanced disease. When the capacity of adipose tissue to store energy was overloaded, hepatocytes store the extra lipids, leading to ectopic fat accumulation, insulin resistance, and hepatic steatosis (39). The result of the generalized linear regression model analysis was in line with the result of multiple regression analysis. Race as an independent risk factor for MAFLD enhanced the effect of anthocyanin intake on MAFLD in our results. The positive correlation between HDL and anthocyanidin intake could explain that higher anthocyanidin intake is protective against MAFLD in participants combined with hyperlipemia. Although dietary intervention alone was unlikely to prevent the incidence and progression of MAFLD, the findings that anthocyanins had a protective role against MAFLD, especially in these subgroups, provides a practical preventive measure to minimize the risk of MAFLD. Moreover, many studies have confirmed that dietary polyphenol is a new target to prevent NAFLD. Daily intake of 1,500 mg curcumin could reduce serum cholesterol, glucose, and ALT in patients with NAFLD (40). A study of systematic review and meta-analysis showed that curcumin, naringenin, hesperidin, catechin, and silymarin may have therapeutic efficacy in patients with NAFLD (41). The result of our study on anthocyanins is complementary to the research on the correlation between polyphenols and MAFLD.

Isoflavones have been shown in animal models and in vitro studies to protect against NAFLD via multiple mechanisms that modulate oxidative stress, lipid synthesis, fatty acid β-oxidation, peroxisome proliferator-activated receptor α (PPARα) activity, and aldose reductase/polyol production (42). Dietary soy isoflavones could prevent the accumulation of hepatic lipid droplets and delay the progression of NAFLD due to the regulation of lipogenesis and lipolysis, suppression of hepatic PPARγ2, and promotion of fatty acid oxidation (43, 44). There were limited epidemiological studies reporting the association of isoflavone intake with NAFLD. A Chinese cohort epidemiological study showed that higher soy food intake ≥ 4 times/week reduced the 25% prevalence of NAFLD in Chinese adults (45). Another study using the data from Chinese adults in the Nutrition Health Atlas Project reported that total isoflavones intake was inversely associated with NAFLD, hyperlipidemia, and hypertension (46). The protective effect of higher isoflavones intake in Asia individuals for MAFLD prevalence was in line with the Chinese cohort studies. Several studies have found that higher diet quality (higher scores of HEI-2015) was associated with a lower risk of NAFLD (47, 48). Our study discovered that higher isoflavone intake was associated with a lower risk of MAFLD among participants with higher HEI-2015 scores. This finding supported prior research and suggested that higher isoflavone intake might benefit from a higher diet quality. Previous studies have shown that young and Black individuals have lower risk of NAFLD (49). We found that higher isoflavones intake was negatively associated with MALFD prevalence in these subgroups, indicating that isoflavones might be a marginally effective intervention in MAFLD prevention. The mechanism of higher isoflavone intake was protective against MAFLD in the context of diabetes since isoflavones activated PPAR, which played a central role in the regulation of blood glucose homeostasis and insulin sensitivity (50). This could be the reason for negative relationship between isoflavone intake and plasma level of FPG. Besides, there was a negative association between isoflavone intake with plasma level of ALT but a positive association with AST, both ALT and AST are the markers of liver injury, so isoflavone intake could not beneficial in liver injury. Moreover, we found that the Q2 of isoflavone intake increased the risk of MAFLD. A previous study in line with our result, the supplement of daidzein in Zucker fatty rats did not decrease the body weight and serum leptin, liver steatosis scores, and energy intake did not show significant differences either (51). Another study investigated the effect of 2-heptyl-formononetin (C7F) and formononetin, an O-methylated isoflavone, on lipid metabolism in C57BL/6J mice. The diet-supplemented formononetin or C7F increased the level of triglycerides and induced hepatic steatosis compared with a cholesterol-enriched diet (52). However, a randomized double-blind controlled trial showed that patients of NAFLD daily supplemented with 250 mg genistein for 8-weeks could improve fat metabolism and reduce insulin resistance (53). As a result, the effect of isoflavone intake on fatty liver disease was controversial. More research is required to establish the isoflavone impact on MAFLD and the underlying mechanism.

To our knowledge, this is the first national population-based study to evaluate the relationship between flavonoid and subclasses (especially anthocyanins and isoflavones) intake and the risk of MAFLD. With the large-scale representative survey data, our analysis results are effective and reliable for adjusting the potential confounding variables. The relationship between anthocyanin or isoflavone intake and the risk of MAFLD was evaluated in diverse individual groups. This may give a diet suggestion for the prevention of MAFLD and metabolic diseases. However, the study’s limitations should be acknowledged. Firstly, this study is a cross-sectional investigation that could only identify relationships rather than causality. Secondly, a person’s long-term diet intake was not well represented by the 24-h recall, and MAFLD may have occurred before they provided the interview data, implying that the bias of flavonoid intake was unavoidable. Thirdly, our results only represent the U.S. population. Although this survey included results from multiple racial groups, flavonoid intake may vary substantially depending on the diet habits and dietary components of different regions. Furthermore, despite multiple confounding variables have been adjusted, residual confounding unknown to us may affect the results. Finally, further research is needed to corroborate our findings and the underlying mechanisms.

In summary, this study provided evidence that higher anthocyanin intake is associated with a lower risk of MAFLD for the U.S. adult population. There was a negative relationship between isoflavone intake and MAFLD prevalence in participants of younger (age < 50), Non-Hispanic Black, Non-Hispanic Asia, and higher scores of HEI-2015. These results imply that higher anthocyanin and isoflavone intake could prevent MAFLD in the dietary modification. Further studies are warranted to elucidate our findings and confirm the mechanism.

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

The studies involving human participants were reviewed and approved by Institutional Review Board of the National Center for Health Statistics (NCHS). The patients/participants provided their written informed consent to participate in this study.

JT designed the research, collected the data, performed the statistical analysis, and drafted the manuscript. YZ investigated and curated data. JX and KX contributed to the conception and literature search. LC and ML supervised the statistical analysis, reviewed and edited the manuscript. All authors have approved the final manuscript.