Introduction

The prevalence of metabolic syndrome (MetS) in Mexican adolescents is higher (6.5-19.2%)^1,2than in other ethnic groups (4.5%).³Diet and physical activity play a role in the development of MetS.⁴

Dietary patterns (DP) are defined as “nutritional variables grouped according to some nutritional criteria, in which variables are reduced into a smaller number of variables through statistical manipulation”.⁵DP have the advantage of evaluating the potential synergistic effects of foods and nutrients and of reducing epidemiological limitations in comparison with single-food or nutrient approaches. Dietary guidelines might be focused on a food-based approach and not only on a nutrient-based approach that is unclear and favors the consumption of industrialized products designed to meet individual nutrient goals rather than achieving a healthy diet.⁶

In adults, DP predict obesity, MetS, and other chronic diseases; nevertheless, some inconsistencies exist.^5,7,8In adolescents, some reports suggest that a “Western” or “obesogenic” DP is positively associated with overweight.⁹Others failed to detect such association¹⁰or with MetS.¹¹Moreover, an inverse association between the “fast food and sweet” DP and obesity was reported.¹²More studies are needed in adolescents because they have unique nutritional needs and cultural particularities and the still existing inconsistencies require further evaluation. We evaluated the association of DP with MetS and metabolic markers (insulin resistance [IR] and lipids) in adolescents.

Materials and methods

We conducted a cross-sectional study in public high schools in Guadalajara, Jalisco, Mexico. Adolescents who were willing to participate were included in the study (n= 681). Participation rate was 89.3%. Participants with incomplete data (n= 9), or with implausible total energy intake (<800 kcal/day or >6000 kcal/day) (n= 18) were excluded from analysis. Finally, 654 participants were included. Previous studies including some of these participants have been reported.¹³The Institutional Review Board of the Mexican Institute of Social Security approved the protocol. Written informed consent was obtained from all the participants and their parents.

Questionnaires and anthropometric measurements were performed by nutritionists. After five minutes of rest, two systolic blood pressure (SBP) and diastolic blood pressure (DBP) readings were taken with a digital baumanometer (Omron HEM-751; Vernon Hills, IL, USA).

A venipuncture blood sample was collected after a 12-h fast. Serum samples were centrifuged and stored at -80^oC until analysis. Glucose levels were determined with the hexokinase method in an automated system (Synchron CX4; Beckman Coulter, Inc., Brea, CA, USA) and insulin with an immunometric method utilizing an Immulite 2000 analyzer (Diagnostic Products Co., Los Angeles CA, USA). IR was estimated with HOMA-IR= fasting insulin (μU/ml) × fasting glucose (mmol/l)/22.5. Total cholesterol and triglycerides were estimated by conventional enzymatic procedures. High-density lipoprotein-cholesterol (HDL-C) and low-density lipoprotein-cholesterol (LDL-C) were determined directly by immunochemical methods utilizing an ILab 300 Plus analyzer (Instrumentation Laboratory, Ltd., Birchwood, Warrington, UK).

Results

Mean age of the participants was 15.8 ± 1.0 years, 51.7% were women, 28.8% were overweight or obese. MetS prevalence according to different definitions was: 5.1% (IDF); 7.2% (Cook and collaborators); 8.1% (Ford and collaborators), and 16.4% (Ferranti and collaborators).

Three DP were identified (table II): 1) “DP1”, characterized by lower energy intake and lower consumption of cholesterol; 2) “DP2”, characterized by higher intake of protein, cholesterol, saturated fats, sodium, dietary fiber, vitamins, and minerals, and 3) “DP3”, characterized by higher energy, carbohydrate, sucrose, fructose, and alcohol intake and lower protein intake (table III). Regarding food composition, the “DP1” was characterized bytortilla, the “DP2” by whole fat dairy products, meat, refined grains, fruits, and milk beverages, and, the “DP3”, by Mexican food, sweetened beverages, sweet baked goods, sweets with fat, snacks, sweets, and alcohol. It is noteworthy that the energy intake of unhealthy foods (fast foods, sweetened beverages, sweet baked goods, sweets with fat, snacks, sweets, and alcohol) was high in the three DP, being highest in the “DP3” (49.3%), then the “DP1” (40.2%), and finally, the “DP2” (36.9%). Contrariwise, the energy intake of healthy foods (legumes and nuts, fruits, vegetables, wholegrains,tortilla, eggs, fish, poultry, low-fat dairy products, and avocado) was low in the three DP: 17.1% in the “DP3”, 21.2% in the “DP1”, and finally, 20.1% in the “DP2”.

Adolescents consuming mainly the “DP3” were older, mainly female, smokers and physically active (p<0.05 for all, data not shown).

Unhealthy eating habits were different according to DP; more than one half of adolescents who skipped breakfast were in the “DP1” or in the “DP3”. Moreover, lunch away-from-home, fast food consumption away-from-home, eating chicken skin and adding ≥3 teaspoons of sugar to beverages was higher in the “DP3” (p<0.05 for all) (table IV). The remaining eating habits did not differ among dietary groups.

Finally, interactions between sex and DP were found (p<0.05); therefore, multiple logistic regression analyses were performed after stratifying by sex. Among males, the “DP3” was associated with MetS (Cook and colleagues OR, 12.14; 95%CI, 1.66-89.05; de Ferranti and colleagues OR, 5.10; 95%CI, 1.20-21.72, and Ford and colleagues OR, 9.29; 95%CI, 1.44-59.73), high triglycerides (OR, 3.89; 95%CI, 1.01-15.07), and HOMA-IR (OR, 6.66; 95%CI, 1.12-39.70). Also, the “DP2” was associated with AO (OR, 5.11; 95%CI, 1.57-16.66). Females showed no statistically significant association (table V).

Discussion

Our results suggest that DP are associated in different ways with obesity and MetS, and the “DP3” has the greatest risk of MetS, hypertriglyceridemia, and IR, while the “DP2” exhibits a greater risk of AO among adolescent males, but not among females.

The “DP3” was associated with MetS (defined by Cook and colleagues, de Ferranti and colleagues, and Ford and colleagues), hypertriglyceridemia, and IR in this study. Sucrose consumption was higher in the 75th percentile (11.9% of total energy). In this DP, compared with the other two, sucrose consumption exceeds the World Health Organization recommendation (≤10% of total energy intake).²⁶In the Mexican adolescents studied in Ensanut 2006, high-energy beverages (soft drinks, sweetened juices, aguas frescas) accounted for 12.7% of the total kcal/day,²⁷a figure lower than in our findings (13.2%) in the “DP3”. In Mexican adolescents studied in the Health Workers Cohort Study, a Western DP (characterized by soft drinks, snacks, and corn tortillas) was found to be associated with IR.²⁸Some studies in adult populations suggest that unfavorable diets (rich in sugar-sweetened beverages, fried potatoes, and red and processed meats) are associated with glucose and insulin;²⁹the consumption of fructose-sweetened beverages decreases insulin sensitivity and increases postprandial hypertriglyceridemia.³⁰In addition, one or two servings/day of sugar-sweetened beverages increase the risk for diabetes and MetS.³¹We do not know the mechanism of the relationship between the “DP3” with IR and with triglycerides; however, fructose might play a key role in hepatic IR through activation of the carbohydrate-responsive element-binding protein which prevents insulin from suppressing glucose production and stimulates de novo lipogenesis.³²

Furthermore, in our study, only 21.3% of adolescents consumed the dietary fiber recommendation of at least 14 g /1 000 kcal³³and no differences were found between dietary patterns (“DP1”= 20.7%, “DP2” = 26.2% and “DP3”= 16.7%;p= 0.320). These results are similar to a study in Mexican adolescents in which almost 80% did not consume the recommended dietary fiber.²⁸

Fast food accounted for highest energy consumption in the three DP in our study (13.7-15.5%). This reached 16.8% in males from “DP2”, which may explain, in part, the association of this DP with AO. Similar to our study, fast food energy intake in adolescents of the NHANES 2011-2012 study was 16.9% and increased to 18.6% in obese.³⁴A study in Iranian adolescents found that fast food consumption in the highest vs. the lowest quartile increased the incidence of AO.³⁵

On the other hand, we found that sodium and sucrose intakes in the “DP3” were higher than in the “DP1”. Previously, a three-times higher risk of developing MetS in the highest vs. the lowest quartile of sweet and salty snacks was found in children and adolescents.³⁶Additionally, Mexican school-children, consuming a Western DP (high in sweetened beverages, salty snacks, cakes, and sweets) had more overweight.³⁷Moreover, in USA adolescents, sweetened beverage consumption increased 74 g/day per each additional 1g/salt/day.³⁸In a study in children from Mexico City, salty-food consumption was mentioned as one of the reasons for drinking soft sweetened drinks.³⁹ Unfortunately energy-dense food consumption has increased in children and adolescents; not only because these foods are inexpensive, good-tasting, and available,⁴⁰but because of their ability to exert an effect on hedonic and motivational processes.⁴¹

Additionally, unhealthy eating habits (skipping breakfast, eating away-from-home, fast food consumption away-from-home, eating chicken skin, and sugar added to beverages) were higher in the “DP3” in our study. An increase in the consumption of calories in USA children and adolescents was found between 1977 and 2006: consumption away-from-home increased 255% during this period and fast foods contributed to the largest energy intake from foods prepared away-from-home.⁴²In Mexican children, the availability of unhealthy foods (snacks, chocolates, sweets, sugary drinks, and antojitos) on the way to school ranged from 22-31%, and foods and beverages eaten away-from-home contributed to obesity increase.⁴³On the other hand, Lebanese adolescents consuming a Western DP were more likely to eat away-from-home and to skip breakfast than the traditional DP⁴⁴In our study, such behaviors were more frequent in the “DP3”.

Sample size was higher in females than in males, nevertheless, only significant associations were found in males. We cannot establish a biological reason for this, but in our adolescents MetS prevalence according to Cook and colleagues, was higher in males than females (12.1 vs. 7.6%;p= 0.090), as well as according to Ford and colleagues (11.1 vs. 5.3%;p= 0.007) and de Ferranti and colleagues (19.7 vs. 14.2%;p= 0.061). Energy intake was also higher among males than females (2 795 vs. 2 464 kcal/day;p<0.001). Others have found that, only among males, an increase in the percentage of energy from fat was associated with AO, and a Western DP was associated with a higher risk of overweight and hypertriglyceridemia.⁹To the contrary, among Korean prepubertal girls, a balanced DP was negatively associated with triglycerides and a Western DP was positively associated with MetS.⁴⁵More studies evaluating these associations according to sex are needed.

We must be careful in the interpretation of our results, because the nature of our cross-sectional analysis cannot establish causal relationships. Thus, future studies are required to answer this question. The FFQ is widely used in epidemiological studies due to its advantages; however, it overestimates consumption; therefore, interpretation of results should be conducted with caution. Although food groups were formed according to their nutritional value, it is possible that complete objectivity might not been achieved. Despite such limitations, we found associations between obesity (and MetS) and DP that are similar to those of previous studies. On the other hand, the strengths of the study are that confounders were controlled with multiple logistic regressions.

In conclusion, we found that the “DP3” increased the risk of MetS, hypertriglyceridemia, and IR, and that the “DP2” increased the risk of AO in male adolescents. Moreover, unhealthy eating habits were higher among “DP3” consumers. Promotion of a healthy DP is needed in order to reduce obesity and MetS in adolescents.