INTRODUCTION

Cardiometabolic risk (CMR) represents a predictive factor for cardiovascular and metabolic disorders, and is characterized by the presence of abdominal obesity, hyperglycemia, hypertriglyceridemia, reduced HDL-c, or hypertension (¹). According to the International Diabetes Federation (IDF), a quarter of the world's population has CMR (¹). Studies have shown its development in healthy populations, and increasingly in younger age groups, in contrast to the classic profile of its incidence (²,³).

Diet is an important modifiable risk factor for CMR and non-communicable diseases (NCD). Since the beginning of the early history of modern nutrition science, dietary guidelines and food policies have been based on nutrient-focused approaches to address health promotion and disease prevention in the population (e.g., focusing on total fat, saturated fat, or sugar rather than overall food and diet quality) (⁴). Recent advances in nutrition science support that foods and diet patterns, instead of nutrient-focused metrics, explain many of the effects of diet on NCD (⁴,⁵).

There has been a worldwide increase in the intake of ready-made or pre-made products for consumption, the so-called ultraprocessed foods (UPF) (⁶,⁷). In Brazil, this scenario has motivated the elaboration of dietary guidelines for the Brazilian population, based on the purpose and extent of food processing, encouraging the consumption of unprocessed and minimally processed food (MPF) to the detriment of UPF (⁵). Therefore, the investigation of adverse health outcomes according to a food-processing classification seems promising (⁸).

Although available in the literature, the relationship of UPF with dyslipidemia, CMR, and NCD is still incipient (2,8-10). Only three studies to date have explored the association between UPF and CMR (²,⁸,¹¹) using the NOVA food classification system, from which derived the new dietary guidelines for the Brazilian population (⁴). Despite some limitations, these studies found an association between high consumption of ultraprocessed food and CMR (²,⁸,¹¹).

Further studies discussing the relationship between the degree of food processing and CMR are needed to encourage government agencies to draw up public policies and contribute to the prevention of chronic disease (¹²). Thus, our study aimed to evaluate the association between food consumption, based on the degree of processing, and CMR in young adults.

METHODS

STUDY DESIGN AND POPULATION

This is a comparative cross-sectional study in a simple random sample. The sample consisted of young adults of both sexes, aged between 18 and 25 years, who were students of health majors at public and private universities in the State of Sergipe, Brazil. The sample was calculated according to Miot (¹³), with a prevalence among those exposed of 29.6 % (¹⁴), a level of significance of 5 %, and a test power of 80 % to estimate a minimum size of 30 individuals for each group. The final sample consisted of 60 individuals in each group, as shown in figure 1. For the exclusion criteria the following was considered: evidence of diseases related to oxidative stress, chronic inflammation, use of medications or nutritional treatments capable of changing the energy balance, inappropriate food consumption, unhealthy lipid profile, impaired plasma insulin levels and glucose metabolism, individuals with unstable body weight in the past six months (allowing 10 % variation above or below), individuals on special diets for at least three months, the conditions of gestation or lactation, and being an elite athlete. Volunteers with missing data were also excluded at the end of the study.

Figure 1.  Flow diagram of the study sample selection. 

RESULTS

A total of 89 women (74.2 %) and 31 men (25.8 %) were included in the study sample. The most frequent CMR component was low HDL-c concentration (21.2 %), followed by abdominal obesity (15.0 %), hypertriglyceridemia (9.2 %), hypertension (7.6 %), and hyperglycemia (5.9 %). There were significant differences in CMR components among participants, indicating the categorization of the groups as proposed (Table I).

Table I.  Anthropometric, body composition, clinical, biochemical, and lifestyle characteristics categorized by presence of cardiometabolic risk in young adults (n = 120) 

Med: median; IQR: interquartile range; CMR: cardiometabolic risk; HOMA-IR: insulin resistance index. ^*Statistical level of 5 %, Mann-Whitney test.

There was a high caloric contribution of UPF to the diet of this population, although no intergroup statistical differences were found for the CMR group (70.6 %) and the group with no CMR (67.4 %), as shown in table II.

Table II.  Daily intake of energy and macronutrients according to food processing level and presence of cardiometabolic risk in young adults (n = 120) 

Med: median; IQR: interquartile range; CMR: cardiometabolic risk; MPF: in natura and minimally processed foods; UPF: ultraprocessed foods. Percentages derived from the total energy value (TEV) of the diet. *Statistical level of 5 %, Mann-Whitney test.

No significant differences were observed for daily consumption in grams of macronutrients, or for MPF and UPF groups categorized by the presence of CMR. However, there was a lower percentage of total energy from total carbohydrates and a greater percentage from total lipids in all foods, and in UPFs for the group with CMR. This explains why UPF contributed to the disproportion in fat intake between groups. It was not possible to identify the influence of type of processing on the difference of total carbohydrate percentage between the groups.

It is important to note that despite consuming more total carbohydrates, the non-CMR group did not exceed the maximum value of the Reference Daily Intake as established by the National Research Council's Food and Nutrition Board and the WHO recommendation. Significant differences were observed for the consumption in grams of macro- and micro-nutrients when comparing UPF and MPF (p < 0.01), regardless of the presence of CMR (data not shown in the table).

The increased consumption of MPF grams reflected a greater consumption of fruits, natural juices, eggs, and fish. A greater intake of UPF grams was associated with a greater intake of simple pasta, breads, processed meats, cheese, brined vegetables, sweetened and artificial juices, soft drinks, oils and fats, roasted snacks, chips, sandwiches and snacks, sweets and desserts, and traditional dishes (Table III).

Table III.  Daily food consumption according to food processing level in the diet of young adults (n = 120) 

MPF: in natura and minimally processed foods; UPF: ultra-processed foods; Med: median; IQR: interquartile range. ^*Noodles and cereal bar; ^**Tapioca, fritada, feijoada. ^†Statistical level of 1 %; ^‡Statistical level of 4 %; Kruskal-Wallis test. Values on the same line with the same letter are not significantly different.

In the logistic regression models on the associations of MPF and UPF consumption with cardiometabolic risk components (Table IV), the consumption of UPF was shown to be a risk factor for abdominal obesity in the crude model (OR = 1.09; 95 % CI = 1.00-1.18). The consumption of MPF provided protection against increases LDL-c independently of sex, physical exercise, and alcohol consumption (OR = 0.70; 95 % CI = 0.50-0.98).

Table IV.  Crude and adjusted associations of MPF and UPF consumption with cardiometabolic risk and lipid profile in young adults (n = 120) 

OR: odds ratio; 95 % CI: 95 % confidence interval; MPF: in natura and minimally processed foods; UPF: ultraprocessed foods; LDL-c: low density lipoprotein cholesterol; HDL-c: high density lipoprotein cholesterol; SBP: systolic blood pressure; DBP: diastolic blood pressure. ^*Adjusted for sex, physical activity, and alcoholic beverages; ^†p < 0.05.

DISCUSSION

The results found by this study reveal an important contribution of processed and ultra-processed foods to the diet of Brazilian young adults, regardless of the presence of CMR, which corroborates data already found for a young population in Brazil regarding a daily caloric contribution of 51.2 % from UPF (⁵). Meanwhile, a report of the Pan American Health Organization (PAHO) pointed out a 26.7 % increase in sales for beverages and UPFs in Latin America while these sales decreased by 9.8 % in North America (⁶).

The relationship between UPF and non-communicable chronic diseases is due to the poor nutritional quality of these foods and their great content in fats and sugars (⁶). A study evaluating the relationship between total macronutrients and clinical outcomes (²⁶), not considering the degree of processing of the consumed foods, did not observe an influence of lipids or carbohydrates from UPF on their total intake or on the manifestation of CMR as we found.

A high consumption of fat is associated with dyslipidemia (³,²⁷) and other non-communicable chronic diseases, especially cardiovascular diseases, which are the main cause of mortality in the world's population (²⁸). UPFs are rich in fats, especially in saturated and trans fats, which make food more attractive and palatable (⁶). At the same time fats attract the population to the consumption of UPF, they contribute to vascular endothelial dysfunction prior to dyslipidemias, atherosclerosis (²⁷) and other diseases, including those related to cognitive disorders such as Alzheimer's disease and dementia (²⁹).

Processed meats, as well as fatty and processed snacks, are sources of saturated fats and are associated with an increased risk of cardiovascular disease (³⁰). In 2015 the WHO pointed out the consumption of processed meats as a determinant of colorectal carcinogenesis in humans (³¹). Although traditional dishes represent a cultural value of the region, their greater consumption may be related to an increase in UPF intake since feijoada and fritada have ingredients with high fat and energy contents.

UPF generally include a great amount of added sugar. In the US diet, the content of added sugars in UPF was eightfold higher than in processed foods, and fivefold higher than in MPF and processed culinary ingredients grouped together (³²). There is already evidence of a relationship of added and refined sugar with coronary artery disease. This alerts to the importance of excessive sugar consumption as a potential risk factor for cardiovascular diseases besides saturated fat.

A study evaluating the association between dietary patterns and metabolic syndrome in healthy adults found a positive association of sugar and fast-food consumption with LDL-c levels (³). A Brazilian cohort of 3- to 4-year-old children observed that UPFs were predictors of increases in total cholesterol and LDL-c from pre-school to school age. Foods with the greatest energy contribution in the above-mentioned study included breads, crackers, salty snacks, and candies (⁹). Meanwhile, another cohort study in Brazil observed that the caloric contribution of MPF had been reduced by 5.1 percentage points between 1987 and 2009 (³³).

Recommendations to prevent cardiovascular disease advise the consumption of fish, fruits and vegetables due to their independent protective effect, which becomes greater when combined with or making up eating patterns like the Mediterranean diet (³⁴,³⁵). In a recent systematic review and meta-analysis, fish and fruits were among the ten foods indicated as cardiometabolic protectors. This same study also highlighted processed meats and sugar-sweetened beverages as harmful to cardiometabolic health (³⁵). According to some explanations the protective role of eggs is promoted by unsaturated phospholipids that contribute to lowering cholesterol and blood pressure levels (³⁶).

The number, size, and density of LDL-c particles in serum total cholesterol may influence its relationship with cardiovascular disease. The origin and amount of saturated fat and simple sugars in foods affect via different paths LDL-c levels and cardiovascular health. Because processed and ultraprocessed foods offer a greater nutritional imbalance and are rich sources of saturated fats and simple sugars, their consumption may promote a reduction of LDL-c particle size, with LDL-c thus becoming more susceptible to oxidation and atherogenic, thrombotic, and inflammatory processes (³⁷).

A randomized controlled trial tested beverages with different sugar concentrations in healthy adults, and observed that even low and moderate concentrations of sugar reduced the size of LDL-c particles and consequently increased their atherogenic power (³⁸). Furthermore, controlling or reducing LDL-c levels is a well-established strategy to prevent cardiovascular outcomes such as the already known effect of LDL-c reduction on the onset of mini stroke (²⁷).

In addition to characteristics such as practicality and palatability, strong media influence is another factor that motivates a greater consumption of UPF. In Brazil, beverage advertising is the third largest category of television commercials, where UPF represent 60.7 % of advertisements versus only 7.0 % related to MPF (³⁹).

The increased consumption of fruits and unsweetened juices observed in the MPF diet may have substantially contributed to the protection conferred against LDL-c in the population of this study. Likewise, we believe that the higher consumption of fish has also protected against LDL-c alteration in the MPF group. According to the degree of food processing, the study reinforces the notion that healthy diets should be based on a low consumption of UPF and, especially, on a high daily consumption of MPF for the prevention of unfavorable clinical outcomes.

This study had some limitations that must be considered. Because this was an observational cross-sectional study, a causal association of MPF and UPF consumption with cardiometabolic risk cannot be inferred. Also, any generalization of the findings beyond the overall and male populations must be cautious since the majority of the study sample was made up of females (74.2 %). The FFQ method employed in this study to investigate food intake may also represent another limitation. Although FFQs are relatively inexpensive and easily applied, their success depends on the memory of the respondent, and requires well-trained investigators to obtain accurate estimates of the portions and frequencies consumed. However, this method provided reliable estimates of the average dietary intake of the study population (⁴⁰).

To minimize the risk of bias in the study, the sample was randomly selected, protocols and training were used to optimize data collection, and only one researcher performed all anthropometric measures in triplicate for better accuracy of the measured values. To minimize the risk of bias in the collection of food consumption, a food photography gallery was used to help participants report the portions and quantities consumed.

CONCLUSIONS

Fats from UPF contributed to a greater total energy value in the group with CMR. Consumption of UPF was shown to be a risk factor for abdominal obesity whereas consumption of MPF was an independent protective factor against changes in LDL-c levels. Our results support the need for more studies on this perspective, as well as public policies targeting diets high in unprocessed and MPF and low in UPF, in order to contribute to the prevention and control of cardiometabolic diseases in young populations.