Dietary habits have been involved in the pathophysiology of diseases. Lifestyle interventions have confirmed that a healthy diet combined with physical activity are able to improve underlying mechanisms of metabolic and cardiovascular diseases, such as oxidative stress, inflammation and insulin resistance [1, 2]. Several benefits from a fiber-rich diet including fruits & vegetables (FV) on health have been described. International recommendations of 5 servings a day of this food group are based also on their high micronutrient content, which can induce favorable cardiometabolic outcomes [3, 4]. High FV consumption has been associated with decreased incidence and mortality from a variety of obesity-related diseases including type 2 diabetes and cardiovascular diseases [2, 5–7], which are conditions of oxidative stress and chronic inflammation [2, 4, 7, 8]. These associations have been attributed either as to an interaction between micro and macronutrients, or to nutrients in isolation.

Pro-vitamin A carotenoids and vitamins C and E, present in FV, are essential for proper physiological functioning. The importance of vitamin E for maintaining oxidative-antioxidant balance is widely recognized [8, 9], but this must be accompanied by vitamin C in order to enhance antioxidant protection [8–10]. Pro-vitamin A carotenoids are present in brightly colored FV; such micronutrients modulate immune system and exert a protective action by reducing LDL-cholesterol oxidation via induction of antioxidant enzymes [10, 11]. Although the synergism of different nutrients consumed in a daily diet could limit the ability to separate out specific effects, reported in observational studies [12], it was of our interest investigating how the intake of those particular vitamins or their main souces (FV) could be associated with circulating markers of oxidative stress and inflammation, then contributing to increase cardiometabolic risk.

The reported benefits of FV on health are in part dependent on their content of natural antioxidant as vitamins A, C, and E, since oxidative stress is an underlying mechanism for several chronic diseases [9, 13, 14]. Studies in animal models and humans have provided evidence about the pathways by which micronutrients in FV may induce protective cardiometabolic effects [4, 7, 15]. Unbalance between free radicals production and antioxidant enzymes causes damage of multiple cellular components and even death [13, 16], and it would be expected that restoring redox equilibrium could attenuate the risk of diseases such as type 2 diabetes and atherosclerosis [15–17]. However, multivitamins supplementation in randomized trials have not proven to reduce cardiovascular morbidity and mortality [9, 13, 16]. Considering that high doses of certain vitamins may adversely affect lipid concentrations, natural intake through a healthy diet has been recommended [13, 16].

Several circulating biomarkers have been used to indicate in vivo oxidative stress (urinary F2-isoprostanes and oxidized LDL – oxLDL) or antioxidant defense (glutathione peroxidase, superoxide dismutase – SOD or total antioxidant capacity), although none is considered the gold standard [4, 8, 10, 15]. It is known that oxidative stress alters intracellular signaling pathways inducing insulin resistance, a process which is aggravated by oxidative stress-triggered inflammatory mechanisms [15, 17]. A pro-inflammatory state has been detected by elevated C-reactive protein (CRP) concentrations. Tumor necrosis factor alpha (TNF-α) is a major cytokine that affects phosphorylation of IRS-1 into serine and GLUT4 translocation, reducing cellular glucose uptake [18]; TNF-α mediates the production of proteins either with pro- or with anti-inflammatory properties, such as adiponectin [19, 20]. Disturbances in these mechanisms are not routinely evaluated in clinical practice. Considering that dietary habits influence these mechanisms, assessment of food consumption could provide indirect way to evaluate oxidative-antioxidant balance and inflammatory status.

In humans, beneficial impact of FV intake on the incidence of type 2 diabetes was reported [5, 6]. However, a study conducted in Brazil suggested that high intake of fruits and juices could favor glucose metabolism disturbances [21], calling attention that analysis of different contents in food groups could result in contrasting effects. In addition, different populations may also respond distinctly due to the cardiometabolic risk level and/or genetic background.

We raised the hypothesis that dietary assessment by the FV group and/or vitamins could provide valuable information to help identify relationships between consumption and certain biological disturbances in a sample of Brazilian population, known to be highly miscegenated. This study aimed at investigating whether the intake of the FV group as well as pro-vitamin A carotenoids and vitamins C and E were associated with circulating markers of oxidative stress (oxidized LDL and superoxide dismutase), inflammation (CRP, TNF-α and adiponectin) and insulin resistance (HOMA-IR) in Brazilians individuals at cardiometabolic risk.

Participants consumed a mean of 1,800 ± 672 kcal/day (50.4 ± 6.8% carbohydrate, 18.1 ± 4.2% protein and 31.2 ± 5.8% fat). Mean energy intake, percentage macronutrients, vitamins and FV were similar between genders. A high proportion of individuals were below recommended levels for micronutrients (98% for pro-vitamin A carotenoids, 89% for vitamin E and 70% for vitamin C). On average, subjects consumed 3.8 ± 2.8 servings/day of FV. Mean time spent on physical activities was 181 min/week, ranging from zero to 1,260 minutes/week.

The oxLDL concentration was correlated to HOMA-IR (r = 0.436, p = 0.03), while SOD concentration was correlated to both adiponectin (r = 0.184, p < 0.01) and CRP (r = −0.198, p < 0.01) concentrations.

In an at-risk sample of the Brazilian population, this study reinforced a beneficial role of FV, as well as antioxidant vitamins, in the pathophysiological process of cardiometabolic diseases, major causes of mortality worldwide. This assertion is based on findings of inverse associations of oxidative stress and inflammation biomarkers with the intake of these nutrients. It is noteworthy that our results remained significant even after adjusting for variables which reflect a healthy lifestyle, such as physical activity, saturated fat intake and smoking.

We detected a positive association of FV consumption with SOD and an inverse association with oxLDL, although the latter had disappeared in the adjusted model. SOD is an essential enzyme that eliminate superoxide radical (O2–) protecting cells from damage induced by free radicals; on the other hand, it is known that oxidation of the LDL particle increases its atherogenic potential [29]. Our findings are in agreement with a recent cross-sectional study, as well as the Women’s Health Across the Nation in which participants with higher intake of vegetables had lower urinary F2-isoprostanes levels, another biomarker of oxidation [30, 31]. Also in intervention studies, individuals who consumed a diet rich in FV had significantly lower concentrations of oxidative stress biomarkers [32, 33].

Beneficial properties of FV include an interaction of nutrient and non-nutritive components in whole foods [1–7]. Despite knowing that the combination of nutrients in foods have greater health benefits when combined than when consumed alone [2], each nutrient per se might have its own effect on the risk of cardiometabolic diseases. Therefore, our study also focused on the associations of pro-vitamin A carotenoids and vitamins C and E intakes with the biomarkers of oxidative stress. Actually, associations of these vitamins with oxidative stress were previously demonstrated [29, 34, 35]. In participants of present study, intakes of vitamins C and E were inversely associated with oxLDL concentrations. Lower levels of this atherogenic particle would be particularly desirable taking into consideration the prediabetic condition of the participants, which increases cardiovascular risk [1, 2, 15, 17]. Oxidative stress triggers inflammation stimulating the NF-кB pathway [36–38] and both processes were shown to be strongly associated with insulin resistance [15, 17]. Along the same lines, we found associations of oxLDL concentrations with HOMA-IR and of SOD with adiponectin and CRP, supporting a role for these parameters in assessing oxidative and inflammatory states. These findings are in agreement with the results of the Women’s Health Study [39], indicating that pro-inflammatory pathways contribute to the onset and/or aggravation of insulin resistance, a major pathophysiologic basis of cardiometabolic diseases.

Considering that low intake of FV is a major modifiable risk factor contributing the risk for several diseases in epidemiological studies [39–41], for clinical purposes, the number of servings of this food group may be a useful indicator of oxidative status and low grade inflammation. International recommendations have established a minimum of 5 servings of FV per day (equivalent to 400 grams) as part of a healthy diet [1, 3]. This was the rationale for the categorization of FV intake in the present study. In our sample, number of servings (as well as pro-vitamin A carotenoids and vitamins E and C intakes) fell below this recommendation, since the participants consumed a mean of 3.8 servings/day of FV. This finding was somehow expected since previous studies consistently reported that people living in both developed and developing countries consume low amounts of this food group [3, 42]. A minimal intake of 5 servings per day of FV should guarantee the amount necessary to reduce oxidative stress and inflammation, involved in the genesis of cardiometabolic diseases. In fact, in our sample, those individuals who consumed the recommended amount of FV had significantly lower waist circumference and higher adiponectin concentrations. Concordantly, other investigators found that diets with high antioxidant capacity were related to increased adiponectin concentrations [43]. Considering that abdominal adiposity is a central feature for the development of insulin resistance, coupled with the favorable effects of adiponectin on these processes [18–20], this dietary habit should be highly desirable for reducing the cardiometabolic risk of these individuals. HOMA-IR was not able to identify differences in insulin sensitivity across categories of FV intake. Based on our body of findings, we speculate that habit of FV consumption may be not only indicative of a healthy lifestyle which protects against body fat accumulation. FV contents may favor per se antioxidant and anti-inflammatory mechanisms, resulting in attenuation of insulin resistance and atherogenesis.

The present study has limitations and strengths. Circulating levels of vitamins were not available whereas the cross-sectional design precluded determination of causality. Since our sample derived from a longitudinal study, sample size was not calculated to test the current associations. Despite a relatively small sample, this proved sufficient to allow detection of important associations of FV intake with oxidative stress, inflammation status and insulin resistance in a population known to be genetically heterogeneous. The application of epidemiologic methods to nutrition is fraught with complications because of the highly interrelated nature of dietary exposures. For this reason, it is difficult to separate out the specific effects of nutrients or foods despite the common practice of examining the role of single nutrients or foods in relation to disease risk. The combination of these nutrients in foods has greater health benefits than the individual nutrients alone. Another limitation of the present study is related to the intrinsic weakness of the available techniques to quantify food intake, since they are dependent of individual report [44]. This context may result in the detection of associations of small magnitudes between nutrients and diseases. Precautions were taken to improve the accuracy of the dietary data collected. Strength of our study was the detailed assessment of dietary intake in this at-risk sample. Individuals provided more reliable diet record data by giving a three-day food record, including a weekend day. It is also noteworthy that data on two markers of antioxidant-oxidative status were available, namely SOD and oxLDL concentrations, given that no single gold-standard is defined in literature.

In summary, the assessment of FV or selected vitamins intakes in clinical settings may be useful to identify oxidative stress and inflammation involved in the genesis of cardiometabolic diseases and for motivating at-risk patients for changing dietary habits. This study supports that international guidelines of ≥ 5 servings per day of FV should be emphasized in an attempt to attenuate underlying mechanisms of chronic metabolic and cardiovascular diseases.