Vegetarian Diets and Cardiovascular Risk in Women

Background: Vegetarian diets have favorable effects on cardiovascular risk, provided that they do not contain ultra-processed foods (UPF). Objective: To compare the metabolic profile, cardiovascular risk, body composition, and food consumption in vegan (VEG), lacto-ovo vegetarian (LOV), and omnivorous (OMNI) women. To verify the association between UPF consumption and cardiovascular risk. Methods: Cross-sectional study with 119 VEG (n = 43), LOV (n = 38), and OMNI (n = 38) women. Anthropometric and biochemical parameters and the Framingham risk score were assessed. Food consumption was assessed by means of a 3-day food register, and intake of macronutrients, micronutrients, and UPF was estimated. The correlation between UPF consumption and cardiovascular risk was assessed using Spearman’s coefficient, with a significance level of 5%. Results: The groups showed low cardiovascular risk, without significant difference between them. The VEG and LOV groups had lower body mass index, neck circumference, body shape index, and systolic blood pressure (p < 0.05) than the OMNI group; greater consumption of carbohydrates, sugars, dietary fibers, micronutrients, beta-carotene, and carotenoids; and lower consumption of total fat, saturated fatty acids, and cholesterol (p < 0.05). Consumption of UPF was lower in the LOV group (5.7 [0.0– 19.8]) than in the OMNI group (14.9 [5.1 – 22.3]; p < 0.05). UPF consumption was associated with SBP (ρ = 0.439; p = 0.007) and blood sugar (ρ = 0.422; p = 0.010) in the VEG group, and in the LOV group it was inversely associated with LDL-c (ρ = −0.456; p = 0.010). Conclusion: Vegetarian women showed better body composition and dietary quality than OMNI women. It is important to take consumption of UPF in vegetarians into consideration, in order to improve cardiovascular risk in women.

literature have investigated the consumption of UPF in vegetarians, especially in women. 13,14 It is important to comprehend the nutritional profile of women and to compare different DPs in order to verify their associations with diseases, such as CVD, thus establishing nutritional behavior protocols for prevention. In this manner, the objective of this study was to compare the metabolic profile, cardiovascular risk, body composition, and food consumption in women who adhered to vegan (VEG), lacto-ovo-vegetarian (LOV), and omnivorous (OMNI) diets, as well as to verify consumption of UPF.

Study Groups
This cross-sectional study was conducted between January and July 2019, with a convenience sample of 119 women selected at a clinical nutrition outpatient clinic, in Rio de Janeiro. Women ages 20 to 59 years old were selected; they had adhered to the DP for at least 6 months, and they were divided into the following groups: VEG (no consumption of any products of animal origin), LOV (consumption of eggs, milk, and dairy products), and OMNI (consumption of red meat, fish, chicken, eggs, milk, and dairy products. Pregnant and breastfeeding women were excluded.
This study received approval from the Ethics Committee of the Clementino Fraga Filho University Hospital of the Federal University of Rio de Janeiro, and it was registered under number 89033118.1.0000.5257. All participants signed a free and informed consent form.

Anthropometric, body composition, blood pressure, and cardiovascular risk assessment
Anthropometric assessment was carried out, including measurements of body mass in kg, height in m, waist circumference (WC) in cm, and neck circumference (NC) in cm. Body mass index (BMI) was calculated (weight/height²) and classified in accordance with the parameters established by the World Health Organization. 15 The waist-to-height ratio (WHR), 16 visceral adiposity index (VAI), 17 LAP, 18 BRI, 19 and ABSI were also calculated and classified. 20 Body composition was assessed by tetrapolar bioimpedance (Biodynamics 450,biodynamics corporation, Washington), and body fat percentage data from the Informatics Department of the Unified Health System (DATASUS, acronym in Portuguese), in 2019, 17.2% of deaths that occurred in women of childbearing age were due to circulatory system diseases, thus representing the second leading cause of mortality in this group. 3 Considering that the majority of cardiometabolic risk factors do not occur with clinical manifestations, early identification can be important in order to modify prognosis of CVD. Accordingly, some tools have been suggested to predict cardiovascular risk, such as the Framingham Risk Score (FRS), which assesses short-term risk of coronary artery disease (CAD), and traditional anthropometric indices, such as body mass index (BMI), waist circumference (WC), and waist-to-height ratio (WHR), as well as new indices, such as lipid accumulation product (LAP), body roundness index (BRI), and body shape index (ABSI). 4,5 In relation to strategies for modifying the course of CVD, diet plays a crucial role. Different dietary patterns (DP) such as the Dietary Approaches to Stop Hypertension (DASH) eating plan, the Mediterranean DP, and the vegetarian DP have been proposed by the American Heart Association/American College of Cardiology (AHA/ACC) both for prevention and treatment of CVD and in general, emphasizing the increased consumption of vegetables, fruits, whole grains, and legumes and limited intake of red meat, sweets, sugar-sweetened beverages, and salty or highly processed foods. 6 In recent years, the number of vegetarians has increased, and this DP has been associated with health benefits, given that it involves reduced cardiometabolic risk factors, and it may contribute to a lower prevalence of CVD. [7][8][9] The benefits of the vegetarian DP result from increased consumption of vegetables, sources of fiber, and phytonutrients, which reduce inflammation and oxidative stress, providing cardiovascular protection. 10 Although plant-based diets have been associated with lower risk of CVD, they may not always have beneficial health effects, in the event that they are rich in unhealthy plant foods (sugar-sweetened juices/beverages, refined grains, fried potatoes, sweets). 11 It is known that consumption of ultra-processed foods (UPF) is associated with the main cardiovascular risk factors, such as obesity, hypertension, dyslipidemia, hyperglycemia, and hyperinsulinemia; reduced consumption of these foods is, therefore, recommended. 12 Few studies in the was classified according to Lohman et al. 21 . Blood pressure was measured by the auscultatory method. 22 Cardiovascular risk was calculated by means of the FRS, using the standardized calculator (https://qxmd.com/ calculate/calculator_253/framingham-risk-score-atp-iii), which takes the following variables into account: sex, age, total cholesterol, high-density lipoprotein (HDL), systolic blood pressure (SBP), use of medications for systemic arterial hypertension, and smoking. 4

Biochemical Assessment
Blood samples were collected in the morning, after a 12hour fasting period, in tubes with gel to obtain serum; 30 minutes after collection, blood samples were centrifuged for 15 minutes at 4,000 rpm. Serum concentrations of glucose, triglycerides, HDL, and total cholesterol were determined by the enzymatic method, using an automated biochemical analyzer (Labtest Diagnostica SA, Vista Alegre, Lagoa Santa -Mg. Brazil). Low-density lipoprotein (LDL) concentrations were calculated using the Friedewald formula. 23 Serum insulin concentration was obtained by chemiluminescence, and insulin resistance was estimated using the homeostasis model assessment of insulin resistance (HOMA-IR) index. 24

Assessment of Food Consumption and Degree of Physical Activity
Consumption of energy, macronutrients, and micronutrients was assessed using a 3-day food register (2 typical and 1 atypical days), and data were analyzed using Food Processor software, version 7.2 (EshaResearch, Salem, MA, USA). Assessment of the adequacy of nutritional composition of micronutrients was based on Dietary Reference Intakes. 25 Adequate consumption of fiber, saturated fat, and sodium was established based on the guidelines of the Brazilian Society of Cardiology. 26 Consumption of processed foods and UPF was classified according to the NOVA classification, by Monteiro et al. 27 Degree of physical activity was assessed by the short version of the International Physical Activity Questionnaire. 28

Statistical Analysis
Statistical analyses were conducted using IBM® SPSS® Statistics software, version 25 (Armonk, NY, USA). Categorical variables were shown as percentages and analyzed using the chi-square (χ²) test. Normality of variables was evaluated using the Kolmogorov-Smirnov method. Continuous variables were represented as median and interquartile range, because they did not show normal distribution. For comparison between groups, the Kruskal-Wallis test with post hoc Bonferroni was used. Spearman's correlation coefficients were used to evaluate correlations between percentage of total energy value from UPF and cardiovascular risk factors. Values were considered significant when p < 0.05.

Results
A total of 119 women were included in the study. Their characteristics are shown in Table 1. Of these participants, 43 (36%) were VEG, with median age of 29 (24 to 35) years; 38 (32%) were LOV, with median age of 27.5 (22 to 36) years; and 38 (32%) were OMNI, with median age of 33.5 (27 to 40) years. The VEG and LOV group had higher level of schooling than the OMNI group.
The prevalence of excess body mass (overweight or obesity) in the study population was 25.6% in VEG, 26.3% in LOV, and 57.8% in OMNI. Even though the cutoff values of the anthropometric indices were within adequate values for all groups, VEG and LOV had lower BMI, NC, and ABSI compared to OMNI. Furthermore, SBP was significantly lower in the groups that adhered to vegetarian DPs (VEG and LOV). Moreover, the LOV group had significantly lower WHR, WC, BRI, and DBP values than the OMNI group, and LDL-c concentrations were lower in the VEG group than in the OMNI group (Table 2).
With the aim of assessing short-term risk of CAD, the FRS was calculated, and it showed that all study groups had low (1%) risk of CAD in the short term (≤ 10 years), without any statistically significant difference (p = 0.340).
When assessing food consumption, it was observed that VEG and LOV had higher consumption of carbohydrates; sugars; dietary fiber; vitamins A, C, and E; potassium; beta-carotene; and carotenoids, as well as lower consumption of total fat, saturated fatty acids, cholesterol, sodium, and Na/Kcal ratio than the OMNI group. Moreover, UPF consumption was significantly lower in LOV (5.7 [0.0 -19.8]) than in OMNI (14.9 [5.1 -22.3]; p < 0.05), as can be observed in Table 3.
The correlation test between percentage of TEV from UPF and cardiovascular risk factors demonstrated that consumption of UPF was positively associated with SBP (ρ = 0.439; p = 0.007) and blood sugar (ρ = 0.422; p = 0.010) in the VEG group and negatively associated with  Table 4. No associations were observed between UPF consumption and the FRS and 10-year risk of CVD in any of the groups.
The main results of this article can be seen in Figure 1.

Discussion
In this study, women who adhered to a vegetarian DP had better body composition and dietary quality, in comparison with the OMNI DP. In the VEG group, UPF consumption was associated with higher blood sugar and SBP.
Our sample was characterized by young women, who were apparently healthy, and the majority were active.
Consequently, the FRS indicated low short-term risk of CAD for all groups, and the biochemical parameters were within the limits of normality. Navarro et al. 29 carried out a cross-sectional study, with 88 apparently healthy men (44 vegetarians and 44 omnivores), age ≥ 35 years, and they observed FRS < 10; however, they found that risk of CAD, assessed by the FRS, was lower in vegetarians, as were some cardiovascular risk factors, suggesting that a plant-based diet could be considered protective for cardiovascular health.
The identification of individuals who are susceptible to developing CVD is extremely important and, at the same time, challenging, especially in those who are asymptomatic. The performance of scores for predicting cardiovascular risk varies considerably between populations, and evidence that supports the use of cardiovascular risk scores for primary  31 Accordingly, it is necessary to create new models in order to assess cardiovascular risk in this specific group and to determine the role of these scores in predicting cardiovascular risk in primary prevention.
When comparing DPs, studies have shown that vegetarians have lower BMI and WC than omnivores. 9,32 In a cohort of 49,098 adults in Taiwan, a lower prevalence of overweight was observed in vegetarians than in nonvegetarians; the authors also found that, for each year on a vegan diet, the risk of obesity decreased by 7%. 33 We also found that omnivores had more overweight and greater WC, NC, WHR, ABSI, and BRI than vegetarians. New anthropometric indices have been used to assess the risk of CVD; the BRI, for example, is an index based on WC and height, and it has demonstrated a good capacity for identifying risk of CAD in women; 5 there are, however, no studies in the literature that compare these new indices in vegetarians and non-vegetarians. Different DPs, and food choices may contribute to the development of diseases, and food choices may contribute to the development of diseases. In this study, we observed that the VEG and LOV groups had more balanced diets that were rich in fibers and adequate in terms of nutrients, such as vitamins A, C, and E and carotenoids, and low in saturated fatty acids, cholesterol, and sodium. Bowman et al. 34 also observed differences between the vegetarian and omnivorous DPs; the first was characterized by greater consumption of micronutrients and lower consumption of saturated fat and sodium.
Greater intake of fruits, vegetables, and legumes is associated with lower risk of CVD, acute myocardial infarction, cardiovascular mortality, non-cardiovascular mortality, and total mortality. This is because these foods contain antioxidants and polyphenols, such as vitamin C, vitamin E, and carotenoids, which can prevent lipid   oxidation in the arterial vessel walls, reduce blood pressure, and improve endothelial function, in addition to the fibers that are associated with reduced insulin, total cholesterol, and LDL. 35 In contrast, saturated fatty acids influence the production of inflammatory cytokines and insulin resistance, and sodium increases blood pressure; consequently, increased consumption of these nutrients is associated with increased cardiovascular mortality. [36][37][38] Although vegetarian diets are favorable to health, 8,39 they may often not bring health benefits in the event that they are composed of processed foods. 11 We observed greater consumption of UPF in the OMNI group, in comparison with LOV. The VEG group, even though they had low consumption of UPF, had elevated sugar intake (> 10% of TEV). Silveira et al. 13 studied 503 vegetarians (83.7% women), and they observed that 60% consumed UPF and sugar-sweetened beverages; the frequency of excess daily intake of UPF (≥ 3 times daily) and sugarsweetened beverages (≥ 3 times daily) were 16% and 20%, respectively. Furthermore, excessive consumption of UPF (≥ 3 times daily) was independently associated with overweight. A recent cross-sectional study of the NutriNet-Santé cohort, conducted in France with 21,212 participants with different DP (omnivorous, pescovegetarian, vegetarian, and vegan), found that increased avoidance of foods of animal origin was associated with increased consumption of UPF, demonstrating that not all vegetarian diets necessarily bring health benefits, due to the potential effects of UPF.
We observed an association of UPF consumption with SBP and blood sugar in the VEG group. UPF are energy-dense, and they usually have higher total fat, saturated fat, sugar, and salt contents, as well as a lower amount of fiber and vitamins. High consumption of UPF is associated with increased prevalence of obesity, dyslipidemia, metabolic syndrome, and CVD. 12,40 It is, accordingly, necessary to reduce consumption of these foods, regardless of the DP adopted.
Study limitations include the small number of volunteers, which can make it difficult to generalize the results. Participants were young, apparently healthy, and active; it is likely due to this that they did not show alterations in biochemical parameters and had low FRS scores. The lack of a tool to assess cardiovascular risk in young women is also a limitation in the current literature. Finally, the cross-sectional study design limits the capacity to establish a causal association, making it necessary to conduct prospective longitudinal studies in the future in order to confirm these results. Nonetheless, this was the first Brazilian study to characterize UPF consumption and associate it with cardiovascular risk factors in vegetarian women, comparing them with other DPs.

Conclusion
Women who adhered to vegetarian DP had better body composition and dietary quality than those with OMNI DP, suggesting that the former DP can confer benefits with respect to cardiovascular protection in young women. Nevertheless, future studies should consider consumption of UPF in vegetarians as a modifiable risk factor for CVD.