INTRODUCTION The global lifestyle has been undergoing gradual changes that put the health of children and adolescents at risk and, in the future, the quality of life of adults and the elderly. Several factors contribute to the increase in this risk, but obesity is the trigger for many imbalances in daily life. Among them, we can highlight inadequate food consumption with the replacement of lunch and dinner with unhealthy snacks that are unfavorable to health ¹ . In addition, there is the low level of physical activity (LPA) and sedentary lifestyle resulting mainly from the increase in internet games and indiscriminate use of cell phones, which occupy adolescents for hours without any stimulus to energy expenditure ^2,3 . The Study of Cardiovascular Risks in Adolescents (ERICA) outlined the profile of Brazilian adolescents and showed that in Brazil, the prevalence of obesity and physical inactivity (<300 min/week) was 25.5% and 54.3%, respectively, with a higher percentage in females ³ . These two conditions in adults can induce structural changes, such as increased body and functional measurements and perimeters, and specifically in respiratory function, they can lead to restricted mobility of the thoracic cage, increased respiratory work, reduced volumes and capacities, and reduced tissue oxygenation ^4,5 . Obesity and low levels of physical activity in adolescents may not interfere with respiratory muscle strength and lung function ^6-8 . This is because in this age group, adolescents are able to adapt to the mechanical conditions imposed by excess weight and do not yet have the consolidation of the ribs and reduced lung compliance that occurs in adulthood ⁹ . Based on this information, the objective of the present study was to evaluate respiratory function and verify whether BMI and PAL influence muscle strength and lung function in obese and non-obese adolescents. METHODS This is a cross-sectional study carried out in the high school of Colégio SESI-Jundiaí. Adolescents between 13 and 18 years of age were recruited and those with heart disease, acute asthma attacks, and cognitive impairment were excluded, resulting in 123 students included in the research. The sample size calculation was performed using the GPower software, version 3.0, considering a power of 80%, effect size of 0.5, and significance level of <0.05. The research was approved by the Ethics and Research Committee of the Centro Universitário de Anápolis under No. 2,064,213/2017, following the guidelines of Resolution 466/12 of the National Health Council. All participants and guardians received information about the study and signed the Free and Informed Consent Form. Collections were made in a reserved room on the school premises between January and March 2018, always in the morning. The level of physical activity was measured using an adapted questionnaire, the Self-Administered Physical Activity Checklist, validated for Brazilian adolescents ¹⁰ . The instrument contains a list of 24 moderate to vigorous physical activities (≥ 3METs) and a blank space for any activity that is not described. The data were completed in the classroom, containing the frequency (days/week) and duration (hours/min/day) of the activities. Adolescents who practiced 300 min/week or more were considered active. Body mass was collected using a digital scale (Filizola, model 2096 PP), where the adolescents remained with as little clothing as possible and stood with their backs to the scale display. Height was measured in meters (m) using a stadiometer (Sanny). To perform this procedure, the adolescents were in an orthostatic position, with bare feet together. The BMI (body mass/height squared) and Z-score were then calculated , classified according to the World Health Organization ¹¹ . After the calculation, the adolescents were divided into two groups: eutrophic and overweight/obese. Waist circumference was measured with a metal anthropometric tape (Sanny) with the measurement point at the middle of the distance between the iliac crest and the lower costal margin (12th rib). Body composition was estimated by bioelectrical impedance analysis (BIA) using a tetrapolar device (Quantum II, RJL system). The adolescents were positioned supine on a non-conductive surface with their limbs approximately 30º apart. They had not exercised eight hours before, had not consumed alcohol in the 12 hours preceding the examination, and had not applied any type of body lotion on the day. Electrodes were positioned on the dorsal region of the hand (one between the head of the ulna and the radius, and the other on the proximal phalanx of the third toe) and on the foot (one electrode between the medial and lateral malleoli and the other in the region of the third metatarsal). Three measurements of R (resistance) and Xc (reactance) were performed, and the measurement with the highest value was used for analysis. The two bioelectrical measurements, R and Xc, in combination provided the phase angle (AF = arctangent Xc/R) ¹² . The values ​​of fat-free mass (FFM) and body fat (BF) were calculated using the formulas below ¹³ :

▪ FFM = 1.31 + (0.61 x height (cm) 2 / resistance) + 0.25 x body mass (kg) ▪ GC = body mass – FFM

Respiratory muscle strength was assessed using a digital vacuum manometer (MVD-300, Globalmed, Porto Alegre, Brazil). Maximum inspiratory pressure (Pi _max ) and maximum expiratory pressure (Pe _max ) were obtained from residual volume (RV) and total lung capacity (TLC), respectively ¹⁴ . The examination was performed with the adolescent in a sitting position, using a nose clip. Inspiratory and expiratory efforts were sustained for 1 second (s). Predicted values ​​were calculated according to Domènech-clar et al. ¹⁵ . Spirometry was performed using a portable device (Micro Quark, Cosmed) and disposable mouthpieces specific for adolescents aged 12–18 years. The acceptability and reproducibility criteria were those recommended by the American Thoracic Society/European Respiratory Society ¹⁶ . The measurements collected were forced expiratory volume in one second (FEV ₁ ), forced vital capacity (FVC) and the FEV ₁ /FVC ratio was calculated. The classification of disorders was performed using the lower limit of 70% of the predicted value. Data were expressed as mean, standard deviation, frequency and percentages. The Student’s t-test or Mann-Whitney test were used to compare groups according to data normality. Multiple regression analysis, stepwise method , was used to evaluate PAL and BMI as predictors of respiratory muscle strength and lung function. Models were adjusted for sex, age and sexual maturity. The value considered for p was <0.05. Data were analyzed using the Statistical Package for Social Sciences (SPSS). RESULTS The study had a total sample of 123 adolescents, 55.3% of whom were male and aged between 14 and 18 years. The prevalence of overweight/obesity was found in 26.8% of adolescents. There was no significant difference in PAL between the groups (p=0.69) and, of the eutrophic and overweight adolescents, 19 (57.6%) and 59 (65.6%) were active, respectively. Pi _max (p=0.007) and FVC (p=0.005) were higher in overweight adolescents, while the FEV ₁ /FVC ratio was lower (p=0.001) (Table 1). The phase angle was the only parameter that did not present a significant difference (p=0.26). Figure 1 shows a comparison of maximum respiratory pressures between overweight and normal-weight adolescents. Pimax _, which represents inspiratory muscle strength, was 16.3% higher in overweight adolescents. Only four adolescents in the overweight group had Pimax _within the expected range. None of the adolescents evaluated had Pemax _within the expected range for their age. Figure 2 shows the spirometry parameters. When comparing the groups, lung function showed FVC 11.6% higher in overweight adolescents, while the ratio was 5.5% lower. Two eutrophic adolescents and one overweight adolescent presented mixed ventilatory disorder. The PAL and BMI were tested in multiple regression as predictors of respiratory muscle strength and lung function (Table 2). The two parameters together, adjusted for sex, age and sexual maturity, were predictors of Pimax _, explaining the relationship by 30% and the FEV1/FVC ratio _was explained by 9%. BMI was a predictor of FEV1 ₍ adjusted R2 ₌ ⁴⁸ %, p=0.001) and FVC ( _adjusted ^R2 =51%, p<0.001), while PAL was a predictor of Pemax ₍ R2 ⁼ 41%, p=0.02) and %FEV1 ₍ R2 ⁼ 10%, p=0.008). DISCUSSION This study showed that overweight adolescents had higher _Pimax and FVC when compared to eutrophic adolescents, while the FEV ₁ /FVC ratio was lower. Most adolescents had respiratory muscle strength below the predicted level. Regarding lung function, only two eutrophic adolescents and one overweight adolescent had mixed respiratory disorder. PAL and BMI were shown to be direct predictors of respiratory muscle strength and lung function, and BMI was an inverse predictor of the FEV ₁ /FVC ratio. Pimax was higher in overweight adolescents. This result was also shown in a study that compared eutrophic and overweight individuals ₆ ^. However, there is evidence that Pimax _is similar in both groups ^8,17 . The higher Pimax _in overweight adolescents can be explained by biomechanical and physiological factors in which the distribution of body fat in the abdominal region would be the main mechanical factor ⁹ . However, in the present study, even with obese individuals presenting higher WC, the majority presented measurements within the predicted range for age and sex. It is worth noting that in this age group, the reduction in compliance and consolidation of the ribs do not yet have an influence on respiratory mechanics ⁹ . Regarding the assessment of lung function, this is essential to detect changes in airway conduction and the restrictive barrier imposed by obesity. FVC was higher, while the FEV ₁ /FVC ratio was lower in obese adolescents. There are similar results in the literature, in which overweight Thai children and adolescents tend to have a lower FEV ₁ /FVC ratio, which is consistent with airway obstruction ¹⁷ . In Canadian adolescents with and without excess weight, no differences in lung function were found ⁷ . And even though the obese adolescents in this study had higher FVC and lower FEV ₁ /FVC ratio, only three adolescents were diagnosed with mixed ventilatory disorder (two eutrophic and one overweight), showing that changes in lung volumes only occur in people with extreme BMI (>45 kg/m ² ), but who remain with normal partial oxygen pressure ¹⁸ . The level of physical activity assessed in min/week showed no significant difference among the adolescents, a fact also observed in another study carried out with children/adolescents ⁸. However, when the level of physical activity was used as a predictor of respiratory function, together with BMI, it influenced respiratory muscle strength, FEV ₁ and FEV ₁ /FVC ratio. BMI is the most widely used parameter to assess excess weight and is considered a general marker of health, showing a positive relationship with Pi _max , FVC and FEV ₁ and a negative relationship with the FEV ₁ /FVC ratio. These findings corroborate studies found in the literature ^7,8,19 that emphasize that at this stage of life the body is able to adapt to the condition of excess weight. However, with advancing age it is known that this relationship (excess weight and respiratory function) becomes inversely proportional and special attention is needed to food consumption and physical activity as strategies to prevent future problems. The strengths of this study are related to the study of respiratory function in adolescents, since the number of studies involving this population and the condition of excess weight is limited in the literature. Adolescents diagnosed with asthma were excluded, which made it possible to avoid selection bias due to the fact that this clinical condition already presents, to a greater or lesser degree, airway obstruction. Another point to be highlighted was the possibility of verifying the influence of BMI and PAL on respiratory muscle strength and lung function. It is important to emphasize that BMI is a general marker of obesity and it is possible that the distribution of body fat can influence respiratory function. Thus, other markers such as body fat percentage and blood levels of cholesterol and its fractions can be used to fill this gap. CONCLUSION Based on the study carried out, it is concluded that overweight adolescents had greater inspiratory force and FVC when compared to eutrophic ones, when the FEV ₁ /FVC ratio is lower. Most adolescents had respiratory muscle strength below that predicted for age, body mass and height. Regarding lung function, only two adolescents had mixed ventilatory disorder. It is believed that PAL and BMI are directly related to respiratory muscle strength and lung function in adolescents and that BMI was a negative predictor of the FEV ₁ /FVC ratio.