Recent studies have debated the nutritional risks and benefits of the regular consumption of fruit juices related to the increase of overweight and obesity [6, 29–32]. There are some suggestions that calorie-free or low food energy content beverages (water, tea, coffee, low fat milk, skim milk, soy beverages) and non-calorically sweetened beverages, should be preferred to beverages with some nutritional benefits, such as fruit and vegetable juices, whole milk, alcoholic beverages and sports drinks, to prevent weight gain [29, 31]. However, others have shown that consumption of pure fruit juices, without added sweeteners, did not contribute to weight gain [6, 30, 32], but protected against chronic diseases such as coronary heart disease and some cancers  because of the contribution of bioactive compounds, including flavonoids and vitamins. The statement that the consumption of orange juice contributes to the development of weight gain and obesity was not confirmed in the present study nor by a recent epidemiological study with children , since orange juice consumption was not associated with body weight, body fat, waist circumference, obesity or physical activity in this study. The majority of the consumers and non-consumers of orange juice showed low physical activity with no statistical difference among groups. So, we suggested that physical activity had no influence on the singularities detected on the markers of lipid metabolism or on the intake of vitamin C and folate, which were influenced by the consumption of orange juice.
Together with excess weight, visceral fat can promote metabolic disorders, such as insulin resistance, dyslipidemias and high blood pressure [22, 35]. In our work, the percentage abdominal obesity among orange juice consumers did not differ from that of non-consumers. A meta-analysis showed that high consumption of fruits, juices and non-starchy vegetables may change body composition, reflected in BMI and abdominal obesity. Hence, consumption of these fresh foods may not help to reduce weight and waist circumference, but may help to delay or prevent their increase [36, 37]. Also, orange juice has high nutritional density with low food energy content  in comparison with other unfortified juices and beverages, and helps consumers to meet the nutritional requirements of vitamin C, folate and potassium . Orange juice can be considered a safe and protective energy source since it protects against oxidative or inflammatory stress .
Most participants of the present study followed a similar diet, since they had the two main meals of the day, breakfast and lunch, at the company’s restaurant from Monday through Friday. According to the dietary recalls, there was no difference among orange juice consumers and non-consumers regarding dietary energy, protein, carbohydrate, dietary fibers, total fruit, total fat, cholesterol, and potassium consumption. Moreover, the orange juice consumers showed significantly higher intake of vitamin C and folate, attributable to their average orange juice intake (480 mL/d). A previous experiment  verified a linear increase of vitamin C in the blood serum with increasing doses of orange juice, from 250 to 750 mL, while serum folate concentration only increased significantly when orange juice consumption reached 750 mL/d, suggesting that folate bioavailability is limited by its source and concentration in food.
Folate supplementation can decrease the homocysteine of individuals with originally normal and high levels, with dietary folate providing a beneficial impact on homocysteine levels . Orange juice is one of the most common sources of folate in the human diet: 250 mL of orange juice contains 47 μg of folate , which may help to control serum homocysteine levels . A mean orange juice intake of 480 mL/d provides an increase in dietary folate intake of 90 μg/d, and as much as 141 μg/d for those who consume 720 mL of orange juice. Most likely these increased folate intakes were responsible for the significant inverse correlation that was detected in the present study between orange juice and serum homocysteine levels, observed only for males.
The serum homocysteine levels of men with normal and high cholesterol levels who consumed orange juice daily did not differ significantly from those of their respective counterparts who did not consume orange juice daily. A previous study showed that consumption of orange juice below 750 mL/d did not affect serum homocysteine levels . However, another author observed that serum homocysteine levels only decreased when daily folate intake reached 250 μg . More recently, a study found that daily intake of 250 to 500 μg of folate reduced serum homocysteine levels by 11 to 20% . Dosages below 250 μg/d do not seem to affect serum homocysteine levels significantly; nevertheless, orange juice still contributes to the total daily folate intake.
The present study found an inverse correlation between serum homocysteine levels and apo A-I, which is in agreement with a recent study of patients with coronary artery disease . A positive correlation was also observed between serum homocysteine levels and serum triglycerides (r = 0.17, p < 0.04). A study in animals found a direct relationship between high homocysteine levels and changes in lipid metabolism that increased the risk of coronary artery disease . Most participants, who consumed orange juice regularly, showed a slight reduction in homocysteine levels, that is, 8.5% in normolipidemic individuals and 11% in hypercholesterolemic individuals, although not significantly. High homocysteine levels in humans and animals are associated with endothelial dysfunction, accumulation of fats in the liver, low HDL- cholesterol and a slight change in total cholesterol [42, 44]. An experimental study using supra physiological dosages of homocysteine found that the underlying mechanism involves inhibition of apo A-I synthesis in the liver and consequent reduction of serum HDL- cholesterol and its functional and anti-inflammatory activities on the endothelium .
The dietary intake of saturated fat and cholesterol of the normolipidemic orange juice consumers or non-consumers was close to the dietary guidelines of American Heart Association . On the other hand, hypercholesterolemic subjects, both consumers and non-consumers, had higher intake of dietary cholesterol (> 300 mg/d), but the dietary saturated fat followed the recommended allowance (< 7% of total food energy intake). One of the strategies to reduce serum cholesterol according to the AHA is to consume foods containing stanol/sterol esters, such as orange juice fortified with stanol ester, or foods high in bioactive compounds which can improve the lipid profile .
The present study found that orange juice consumers had on average lower serum levels of total cholesterol and LDL-cholesterol in comparison with non-consumers. Presumably these effects are related to citrus flavonoid compounds found in orange juice [5, 43]. In vivo studies have shown that naringenin and hesperitin inhibited ACAT and microsomal transfer protein (MTP) activities, which are responsible for the synthesis and esterification of the cholesterol in the liver , resulting in lower Apo B secretion and consequently lower LDL-cholesterol levels . Also a significant negative correlation was found between orange juice consumption and total cholesterol, LDL-cholesterol and LDL/HDL ratio in humans [5, 10, 14, 46]. These study results elucidate the beneficial effect of orange juice on serum LDL-cholesterol levels in agreement with in vivo studies that demonstrate an effective cholesterol-lowering action of the bioactive components of orange juice, such as flavanones and vitamin C [5, 10, 14, 46].
Additionally a previous study showed a positive effect of orange juice on HDL-cholesterol levels with the consumption of 750 mL of orange juice per day , which was not verified in our study. Only 8% of our volunteers (data not shown) had a daily orange juice consumption that matched or exceeded the consumption of 750 mL per day, which may account for the difference.
The average of apo B and LDL-cholesterol of the orange juice consumers was lower in both normolipidemic and hyperlipidemic groups. Furthermore, an inverse correlation was found between regular orange juice consumption and apo B, reinforcing the LDL-cholesterol lowering effect of orange juice. An in vitro study found that hesperidin and naringin reduced the availability of lipids for assembly of apo B-containing lipoproteins: VLDL- and LDL-cholesterol, which is attributed to reduced ACAT 1 and ACAT 2 activities, selective decrease in ACAT 2 expression and reduced microsomal triglyceride transfer protein activity .
Serum triglyceride levels were statistically different between consumers and non-consumers of orange juice, despite the presence of sugars in the juice; however they were higher in the hypercholesterolemic group than in the normolipidemic. Animals given orange or grapefruit flavanones daily experienced a reduction in serum triglyceride levels  but studies with human beings have suggested that the magnitude of their effect on triglyceride levels depends primarily on other factors, such as age, gender, fasting glucose, insulin and triglyceride levels, insulin resistance and total fructose intake .
Although a smoking habit has been associated with a significant increase of stroke, high blood pressure, and damage to the arterial wall contributing to atherosclerosis, all of them considered as risk factors for cardiovascular disease , this behavior was not investigated in this study. The main focus of this investigation were the markers of lipid metabolism due to the benefit of the orange juice on total cholesterol and LDL-cholesterol in women and hyperlipidemic men, as previously detected from these authors [13, 14].
The first strength of this study was the volunteers have been already exposed to daily orange juice consumption and that they freely chose to consume or not. The second strength of this study was the exposure of individuals to the same main meals every day (breakfast and lunch). This environment was more homogeneous than individuals who had their meals at home or in different places. Nevertheless, be exposed to the same food environment, does not mean they ate the same foods. Therefore, the assessment of food consumption was conducted through the individual 24-hour recall. The limitation, however, was that the dietary information was obtained through a self-reported questionnaire, which may be subject to under or over reporting and interviewer bias, and because of the cross-sectional design, which is limited in verifying any relationship between cause and effect.