In this nationally representative cross-sectional study, we discussed the association between dietary fructose intake and the risk of MetS among Chinese residents aged 45 and above. The consumption of dietary fructose for urban residents was 11.6 g/day and for rural residents was 7.6 g/day. Under the current dietary fructose intake status, we did not find an association between dietary fructose intake and the risk of MetS in both urban and rural residents aged 45 and above. However, there was a significant inverse association between dietary fructose intake and MetS for urban residents who participating in physical activity.
A large number of researches suggested that fructose was a culprit in the occurrence of MetS through several metabolic pathways, such as increasing hepatic de novo lipogenesis in the liver(25), depleting ATP stores which resulting in increasing generation of uric acid via purine pathway(26, 27), affecting on plasma lipids, lipoprotein, and apolipoproteins (28, 29), and host-gastrointestinal microbe interactions (30, 31). However, there were still disputes between mechanism studies and population epidemiological studies. According to the systematic reviews and meta-analysis, high doses of fructose (≥ 100 g/day) increases serum TG concentration (10, 32), low to middle doses of fructose (0 ~ 90 g/day) have a benefit effect in HbA1c (13, 33). But fructose did not increase the risk of hypertension and type 2 diabetes (12, 34), also, it did not affect serum HDL-C concentration (32) and cause weight gain when it was substituted for other carbohydrate in diets providing similar calories (35).
In present study, we did not find an association between dietary fructose intake and the risk of MetS among Chinese residents aged 45 and above. The results of this study were similar to the study from the NHANES 1999–2006 which showed fructose ordinary consumption (approximately 37% of total sugars and 9% of daily energy in the US population) had no association with the risk of MetS (17). Both two studies were population-based cross-sectional studies. However, a systematic review and meta-analysis discussing the association of fructose consumption and components of metabolic syndrome reported that fructose consumption was positively associated with FPG, TG and SBP, and negatively associated with HDL-C (9). We supposed several reasons for the difference. On the one hand, the fructose sources were different. Food sources of fructose in this meta-analysis were from industrialized foods. In our study, however, fruits and products, vegetables and vegetable products were the most dominant food sources, accounting for more than 50% of dietary fructose. One study reported that most food sources of dietary fructose (especially fruits) did not have a harmful effect on indicators of health (HbA1c, fasting insulin), but several food sources of fructose (especially sugars-sweetened beverages) adding excess energy to diets showed negative effects (36). On the other hand, the fructose intake was different. Fructose provided at least 15% of daily energy requirements in the 15 studies included in this meta-analysis. In our study, however, the average dietary fructose intakes for urban and rural residents were 11.6 g/day and 7.6 g/day, respectively. They contributed less than 3% of energy requirements. Several systematic reviews reported that a continuous exposure to high fructose intake may have adverse health effects (37, 38). Previous study has shown that the percentage of total calories from added sugar containing food of Chinese residents in 2010–2010 was 9.09%, which was under the recommended limits (10%) of WHO (39, 40). In addition, some researchers argued that the before-after design used by the authors, the lack of adjustment for energy as an important confounding variable, and unclear statistical methods render their results uninterpretable. Under calorie-matched conditions, this systematic review and meta-analysis cannot infer that fructose uniquely affects most components of MetS (41). In this study, we not only adjusted the confounding factors, including energy, but also stratified analyze the variables (gender, physical activity, smoking, and alcohol use) that might influence the risk of MetS.
Interestingly, we found that the risk of MetS decreased with the increase of the quartile levels of dietary fructose intake for urban residents who participating in physical activity. In recent years, a growing number of researches supported the idea that physical activity might play a role of modulator for fructose’s health effects (37, 38, 42–44). Fructose was generally processed in splanchnic organs (small bowel, liver, kidneys) to glucose, lactate, and fatty acids, which serve as metabolic energy substrates in extra-splanchnic organs and tissues (37). As fructose uptake and fructolysis were unregulated processes, the amount of metabolic energy substrates was proportionate to fructose intake (42). For sedentary subjects, high fructose intake caused an overflow of metabolic energy substrates which resulted in increased gluconeogenesis, de novo lipogenesis, and triglyceride-rich lipoprotein secretion in the liver (42). In contrast, for physically active subjects with high fructose intake and high energy expenditure, fructose was mainly metabolized into glucose and lactate that can be readily oxidized to support ATP synthesis, resulting in a net lactate release from splanchnic organs (mostly the liver) to the working muscle (42). This ‘reverse Cori cycle’ may be advantageous to improve performance by acting on central fatigue and/or alter metabolic regulation (43, 44). An animal study showed that the naked mole-rat can resist hypoxia and acidosis by increasing fructolysis (45). In our study, dietary fructose intake in the fourth quartile of urban and rural residents was 25.6 g/day and 16.8 g/day, respectively, both of which were relatively low dosage. A series of systematic reviews and meta-analyses have reported that small doses of fructose, or fructose in substitution for glucose or sucrose, may have beneficial effects or not any adverse effects on the components of the MetS (12–15, 33, 46, 47). Based on the above points, we suggested that physical activity and relatively low fructose intake may have a beneficial synergistic effect on MetS.
Several limitations should be considered in the present study. First, this cross-sectional study has a natural disadvantage to address causal relationship between dietary fructose intake and MetS. Second, added fructose was not distinguished in this study. In previous studies, the intake status of added fructose and its relationship with metabolic disease were the focus of attention. However, the consumption of added fructose was very low in our study population. Third, the accuracy of dietary information was limited by the accuracy of recall of the participants and the specificity with which the reported foods were mapped in the dietary recall records. To minimize this situation, all interviewers completed a strict training program with detailed methodologies on administration of the dietary questionnaire. Forth, three consecutive 24-h dietary records may not reflect long-term dietary habits. More high-quality cohort studies and randomized controlled trials were needed to evaluate the association between dietary fructose intake and the risk of MetS.