Association of active behavior, body mass index, blood pressure and inflammatory cytokine in adolescents

Introduction: Physical inactivity is associated with obesity, chronic inflammation and the occurrence of chronic non-communicable diseases. Objective: To investigate the association of active behavior, body mass index (BMI), diastolic blood pressure (DBP) with serum levels of inflammatory interleukins in adolescents. Methods: Cross-sectional population-based study of adolescents from public schools in São Luís, Maranhão. A theoretical model was built, based on Structural Equation Modeling, to explore the effects of the latent variable “active behavior” adjusted for socioeconomic status and gender, with effects on BMI, DBP and inflammatory load. Results: Active behavior was associated with lower DBP values (standardized coefficient (SC) = -0.146; p = 0.029), while higher BMI was associated with higher DBP values (SC = 0.177; p < 0.001). For the highest values of active behavior there was a higher inflammatory load (SC = 0.442; p < 0.001); and a higher BMI was associated with a higher inflammatory burden (SC = 0.145; p = 0.025). Female gender had a higher inflammatory burden (SC = 0.282; p < 0.001). There was an indirect effect for females on lower inflammatory load values via reduction in active behavior (SC = -0.155; p < 0.001). Conclusion: Active behavior reduces blood pressure levels in adolescents; while the increase in inflammatory cytokines induced by active behavior may be involved in their anti-inflammatory response for disease prevention.


Introduction
The substantial increase in the prevalence of physical inactivity is associated with obesity, chronic inflammation and global mortality from chronic non-communicable diseases. 1,2 Considering that physical inactivity is a modifiable risk factor, it is recommended to use active behaviors and/or physical activity (PA) with a view to reducing the risk of heart disease, stroke, diabetes, breast and colon cancer, 2 since there is evidence of the anti-inflammatory and antioxidant effects of physical exercise (PE), involving the immune system, muscle and adipose tissue. 3,4 Excess adipose tissue is associated with the secretion and release of pro-inflammatory cytokines, 5,6 with the increase in their serum levels and the development of low-intensity chronic inflammation. On the other hand, adopting active behavior contributes to the reduction of adipose mass with or without weight loss, 4 modulates inflammatory and anti-inflammatory cytokines (myokines) 3,5,7 and, consequently, prevents diseases.
Among the cytokines there are pro-inflammatory interleukins 1β, IL-6, IL-8 and TNFα, which are involved with metabolic alterations such as insulin resistance, atheogenesis, dyslipidemia, favoring increased cardiovascular risk. 7 The benefits of active behavior for the prevention or adjuvant treatment of diseases depend on its type, intensity and duration. 8,9 Both PA and PE are associated with muscle contraction and energy expenditure. However, by definition, PA refers to any body movement involving skeletal muscles in different activities, whether domestic, occupational, active displacements, physical education, among others, while PE, the resulting muscle work, aims to optimize the individual's performance to meet their daily activity demands, 1 both being elements of active behavior.
On PA in adolescents, there is evidence of gains in cardiorespiratory conditioning associated with decreased inflammatory and adiposity markers. 4,10 A study (n = 1089) demonstrated that overweight and obese adolescents have higher inflammatory scores than eutrophic ones, and increased levels of PA decreased inflammation and visceral adiposity. 11 Another study found that regardless of diet, active behavior was associated with reduced plasma levels of inflammatory markers in obese adolescents. 12 Measuring the active behavior of adolescents is a challenge in choosing the most appropriate method. 13 In the complexity of understanding its effects, it is relevant to explore the direct and indirect effects of the adoption of a non-sedentary life in inflammatory markers, because studies on this theme in adolescents have emerged. [10][11][12] Therefore, this study aimed to investigate the association of active behavior, body mass index (BMI) and blood pressure with serum levels of inflammatory cytokines.  other variables of the model: active behavior, BMI, DBP. As a hypothesis, it is thought that active behavior is inversely associated with BMI and the variable sex, explaining the "active behavior" and the "inflammatory load" (Figure 1).

Modeling with structural equations
Structural equation modeling (SEM) was used as a statistical procedure to test the hypotheses about the relationships between latent and observable variables, as well as to reduce measurement errors (bias-free) in statistical estimation 23 eand to present the maximum likelihood estimate to deal with absent data. 27 For the analysis of latent variables, the standardized and significant factor loadings were adopted as assumptions.  The estimator "root of weighted mean quadratic residue and adjusted variance" (weighted least squares means and variance adjusted) and theta parameterization were used to control residual variances. 28 The change indexes (modindex) command was adopted for suggestions for changes in the latent variable. 29

Results
According to the sociodemographic characteristics of the mother and adolescents, it was observed that 18.5% studied four years or less and 40.9% had a family income of 1 to < 3 minimum wages. Of the 405 adolescents, 44.4% were male, 13.6% were obese and/ or overweight, and 5.6% had DBP > 80 mmHg (Table 1).
For the latent variable "active behavior", the mean of the indicators was 962.

Discussion
An association of active behavior with inflammatory load was found, suggesting that the highest Mets of PA in adolescents contribute to inflammation. These findings corroborate other studies. [30][31][32][33] Forti et al., 9 in a study with healthy young people, found that acute PA increased levels of inflammatory cytokines, while chronic PA decreased IL-6 levels and increased the production of anti-inflammatory cytokines.
Other studies have shown that cardiorespiratory fitness was indirectly associated with inflammatory cytokines in adolescents, 10,11 suggesting that maintaining active behavior (through PA) increases cardiorespiratory fitness and decreases the level of inflammatory markers.  production and release of pro-inflammatory cytokines. 6,7 This result corroborates a longitudinal study with 843 adolescents, 26 that demonstrated the association of BMI with inflammatory markers. Agostinis-Sobrinho et al. 10 showed the association of higher BMI values, low PA levels and high score of inflammatory markers.
In this study, high BMI was associated with higher DBP values. The increase in BMI in adolescents was associated with low rates of active behavior and, consequently, lower stimulation of nitric oxide production and higher risk of dyslipidemia and arterial hypertension. 26 Family SES was associated with higher BMI values. This is one of the aspects related to the process of nutritional transition experienced by Brazilian society in the last three decades, due to the reduction in the prevalence of malnutrition and increased obesity, which involves social factors such as the growth of multi-processed food industries and, consequently, higher consumption of fast foods, regardless of social strata. 33 A study conducted with adolescents from urban and rural areas also found a direct association between SES and BMI. 25 The adoption of active behavior by adolescents, however, is encountered with some obstacles. In public schools, physical education as a compulsory discipline in high school is not practiced regularly, either due to the lack of teacher, appropriate place, among others. In addition, leisure sports and active outdoor commuting are often discouraged by public insecurity in urban centers.
As strengths of this study, we highlight the use of statistical analysis based on a theoretical model Regarding the limitations of the study, one can attribute: the design of the study (cross-sectional), which prevents the establishing temporality in the associations found, as well as the cause-effect relationship, although the statistic adopted aims at an analysis of the direct and indirect effects between the variables. Another limitation was the lack of criteria to control biological collection (abstinence from PA).