Acute effects of high intensity interval training on blood pressure in overweight/obese adolescents

The aim of this study was to analyze the acute responses of blood pressure (BP) to a high intensity interval training (HIIT) session on a treadmill in adolescents with excess body weight. This is a study with crossover randomized design the sample consisted of 11 male adolescents aged 15 to 18 years. The volunteers performed two experimental protocols: HIIT and control. The HIIT protocol consisted of fi ve series at 85 to 95% of VO2peak for one minute, interspersed by three minutes of recovery at 40 to 50% of VO2peak. Before and after the exercise protocols, hemodynamic parameters were obtained using an automatic ambulatory monitoring equipment (SpaceLabs model 90207), programmed to perform measurements every 20 minutes from 12:00 to 22:00 hours. A signifi cant difference in systolic blood pressure (SBP) was only observed between the control and HIIT protocols (131.90 ± 7.93 vs 124.18 ± 7.56 mmHg, respectively) for the fi rst hour after exercising. Comparisons of preand post-session (10 hours) hemodynamic measures between conditions (HIIT vs. control) demonstrated that HIIT promoted a signifi cant reduction in mean blood pressure compared to pre-session values. In conclusion the HIIT protocol resulted in a low magnitude hypotensive effect on post-exercise hemodynamic measures with respect to SBP.


Introduction
Acute effects of high intensity interval training on blood pressure in overweight/obese adolescents Th e high prevalence of obesity in the pediatric population is considered a worldwide public health concern, as this morbidity has been associated with diverse health risks 1,2 . A condition that is typically associated with excess body weight is hypertension 3,4 epidemiologic studies in obese children and adolescents reported a prevalence of hypertension ranging from 47% to 62%.
Studies have also observed associations between lower levels of physical activity and elevated blood pressure (BP) in epidemiological investigations 5,6 . In this context, regular engagement in physical exercise, with special emphasis on aerobic continuous exercise, has been recommended for individuals with arterial hypertension 7 . In children and adolescents, evidence indicates that aerobic exercise promotes long-term 8 as well as acute hypotensive effects 9 , the latter being termed "post-exercise hypotension". Research suggest that hypotensive eff ects after aerobic exercise sessions are able to predict the chronic hypotensive eff ect 10 .
Post-exercise hypotension and its possible mechanisms have been described in the literature 7 , however, few studies have investigated the eff ects of manipulating exercise prescription components, including training methods, on the magnitude and duration of the hypotensive response. Th e most commonly prescribed training method is continuous exercise, which is based on dynamic or cyclic exercises (e.g., walking, running, cycling) of low to moderate intensity lasting for prolonged periods. Th e research by Carpio-Rivera et al. 11 observed that incremental exercise protocols produce the largest reductions in BP. Conversely, the high intensity interval training (HIIT) method is characterized by repeated series of short or medium duration exercises (10s to 5 min) performed at an intensity higher than the threshold speed of lactate 12 .
Studies examining the eff ects of exercise intensity, duration, and methods of aerobic training on the magnitude and duration of post-exercise hypotensive responses have reported contradictory results 13,14 . Th us, it is still necessary to elucidate the eff ects of high-intensity interval exercise on post-Methods exercise hemodynamic responses in adolescents with excess weight. A better understanding of this subject is of major clinical relevance, since the magnitude and duration of post-exercise hypotension induces better modulation of BP in the long term 10 .
Th erefore, the objective of the present study was to examine the acute responses of BP to a high intensity interval exercise session in adolescents with excess body weight.
Th is was a randomized crossover design study. Participants consisted of male adolescents who met the following inclusion criteria: age between 15 and 18 years; no diagnosis of cardiovascular disease; classified as overweight or obesity (percentile > 85 th ); no diagnosis of diabetes mellitus and/or systemic arterial hypertension; insuffi ciently active for at least three months; not using medication; and not presenting conditions that could limit engagement in physical exercises (e.g., labyrinthitis and osteoarticular dysfunctions, among others). Th e exclusion criteria adopted were: not presenting intermediate chronotype; presenting any type of sleep disorder; having a reported stage of sexual maturation < G4; and not meeting the pre-participation recommendations in physical tests and experimental sessions.
Participants were duly informed about all procedures, potential risks, and benefi ts of the investigation. Written assent was obtained from adolescents interested in participating in the study. Their responsible guardians provided written informed consent for participation. All participants were instructed to maintain their normal routine during the period of participation in the study.
Sample size analysis was performed using Gpower 3.1 software. Th e parameters adopted for sample size calculation were: a power of 0.80;  = 0.05; correlation coeffi cient of 0.5; correction of sphericity of 1; and an eff ect size of 0.50, according to the procedures suggested by Beck 15 and results of Carpio-Rivera et al. 11 . Th e calculation resulted in a required sample of 12 individuals, providing a power of 0.87.
Data collection was carried out in the Laboratory of Biodynamics of Human Movement, at the Center of Health Sciences, at the University. Initially, all volunteers interested in participating in the study underwent screening to ensure their eligibility according to the inclusion criteria. In their second visit, participants were submitted to a treadmill cardiopulmonary exercise test. After a minimum interval of two days and a maximum of seven days, the volunteers performed two experimental sessions in a simple random order: control and HIIT.

Participants
Study Design

Screening
The adolescents were screened before participating in the study, which consisted of obtaining the following information: sociodemographic data, clinical data (history of cardiovascular, metabolic, or osteoarticular diseases, symptoms suggestive of heart disease, and use of medications), chronotype, presence of sleep disorders, habitual physical activity level, biological maturation, anthropometric measures (body mass, height, waist circumference, and skinfolds), and BP measurement at rest.
Twenty-four volunteers agreed to participate in the study. At the initial screening, 10 did not accept or could not be included because they did not meet the established criteria. Th us, 14 adolescents and their parents signed the consent form and initiated the preliminary procedures. Of these, three adolescents left the study for personal reasons. Th us, 11 volunteers completed the experimental protocols, resulting in a power of 0.82.

Identifi cation of chronotype and sleep quality index
To assess participants' chronotype, the Morningness-Eveningness Questionnaire (MEQ) was used to determine if the volunteer was a morning, evening, or intermediate person 16 . For the analysis of sleep quality, the Pittsburgh Sleep Quality Index (PSQI) 17 was used. Volunteers who presented indices greater than 10 points, indicating the presence of a sleep disorder, were not included because poor sleep quality could interfere with cardiovascular functions, including blood pressure.

Habitual physical activity level
Habitual physical activity level was analyzed using the International Physical Activity Questionnaire (IPAQ-8, short version). Those who did not reach the World Health Organization recommendations for physical activity for children and adolescents 18 and had not engaged in exercise programs in the previous three months were included in the study.

Biological Maturation
Biological maturation was assessed by selfevaluation of secondary sexual characteristics (genitals) 19 . Volunteers who reported a sexual maturation stage of G4 or G5 were included in the study.

Anthropometric Measurements
Body mass was evaluated using a Welmy digital scale, with a precision of 100 grams. Height was measured using a portable stadiometer (Welmy®), with a precision of 0.1 centimeter. Body mass index (BMI) was calculated by dividing body mass in kilograms by height in meters squared. The waist and hip circumferences were measured using an inextensible anthropometric tape (Sanny®). Volunteers presenting BMI above the 85 th percentile according to the criteria for age and gender of the World Health Organization 20 were included in the study. Triceps, subscapular, and medial leg skinfolds were measured to the nearest 1.0 mm using a Lange skinfold caliper. To estimate the relative fat, the equation proposed by Slaughter et al. 21 for male children and adolescents aged seven to 18 years was used.

Resting blood pressure
To evaluated the values of systolic (SBP), diastolic blood pressure (DBP), and mean blood pressure (MBP) at rest, a SpaceLabs automatic oscillometric monitor model 90207 was used in the non-dominant arm with dimensions appropriate for the circumference of the arm 22 . Next, three measurements were taken in the sitting position with a two minute interval between them.

Cardiopulmonary exercise test on a treadmill
The cardiolpulmonary test was conducted in a laboratory environment under controlled temperature (20 to 22°C). Participants performed an incremental test on a motorized treadmill (INBRAMED Super ATL) to measure the peak oxygen consumption (VO 2peak ) and determine the ventilatory threshold. A modified Balke treadmill protocol was used according to the recommendations of the American Heart Association 23 . During the protocol, respiratory and metabolic variables were measured every 10 seconds by a computerized metabolic system (VO2000 metabolic gas analyzer). A silicone facial mask set with mouth and nose seals was worn by the participants. The Polar FT1 was used to monitor heart rate during the test. Subjective perception of effort (CR-20) was obtained at the final 15 seconds of each stage. The test was discontinued when one or more of the following criteria were observed: voluntary fatigue or inability to maintain the pre-determined speed of the stage, R > 1.1, subjective perceived exertion (CR-20) above 18, and achievement of maximal heart rate (HR max ) estimated by the equation HR max = 208 -0.7 x age.

Experimental Protocols
Upon arriving at the laboratory, participants rested for 10 minutes before resting hemodynamic measures were taken (SBP, DBP, MBP, and HR). After that, the adolescents performed the control or interval protocol randomly. After the end of the session, participants hemodynamic measures were taken again after a resting period of 10 minutes in order to minimize the infl uence of exercise-induced cardiovascular stress. Considering that dehydration Results may be an intervening variable, each adolescent received a 500 ml bottle of mineral water. Bladder emptying occurred at the criteria of the volunteers during the protocols.
Th e HIIT protocol used in this study was adopted considering the tolerance of adolescents to highintensity exercise and evidence from investigations that reported signifi cant eff ects of HIIT on health parameters in obese populations 14,24 . Th us, the interval exercise protocol consisted of three minutes of warmup, with a velocity of 4.0 km/h. Subsequently, the volunteers were submitted to fi ve sets of 1:3, that was, one minute in the active phase (above the ventilatory threshold, 85 to 95% of VO 2peak ) followed by three minutes in the active recovery phase (40 to 50% of VO 2peak ). At the end of the fi ve sets, the adolescents performed a cool-down period for three minutes, gradually reducing the speed before stopping the treadmill belt. Th e total duration of the session was approximately 30 minutes.
For the control protocol, participants received the same instructions as the HIIT protocol, but participants rested in the seated position for 30 minutes and were instructed not to perform physical activities of moderate to vigorous intensity on the day of monitoring.

Ambulatory BP measurement
The analysis of ambulatory blood pressure monitoring (ABPM) was based on the position stand from the American Heart Association 22 . Th us, after the end of the experimental protocol (10 minutes), the ambulatory BP measurement was performed on the non-dominant arm using an oscillometric automatic monitor (SpaceLabs, model

Statistical treatment
Th e Shapiro-Wilk test was used to verify the normality of the data. Participant characteristics are presented as mean and standard deviations, and median and interquartile range for the numerical variables, and frequency distribution for the categorical variables. Th e eff ects of the experimental protocols on hemodynamic responses (SBP, DBP, MBP, and HR) were tested using repeated measures analysis of variance, considering experimental protocol (control and HIIT) and time (presession and post-session). When the assumption of sphericity (Mauchly's test) was violated, the Greenhouse-Geisser correction was applied. If an eff ect and/or interaction was identifi ed by the F test, the Bonferroni post hoc test was applied to locate the diff erences between the means. Eff ect sizes (ƞ²) are presented for signifi cant diff erences. Data analyses were conducted in the statistical package SPSS version 20.0 (SPSS, USA, 2012), adopting a level of signifi cance of p < 0.05. 90207). Th e monitor was programmed to perform measurements every 20 minutes from 12:00 to 22:00 hours. Th is period was suffi cient to obtain daily representative measurements of blood pressure, as there is indication that six hours of measurement are associated with the results obtained from 24 hours in the pediatric population 25 . Th e volunteers were instructed not to talk, sleep, or perform physical activities when the device was triggered and recording the measurements. In addition, for each protocol, participants received a diary to record stressful situations, meal times, transit times, and means of locomotion.
The general characteristics of the volunteers are described in FIGURE 1A demonstrates the SBP responses over the 10-hour period following the control and HIIT protocols. A significant interaction of time x experimental protocol was observed (F=2.488; p=0.008; ²=0.110). Th e post hoc test indicated a signifi cant diff erence in SBP means between the control and HIIT protocols only in the period from 12:00 to 13:00 hours, indicating post-exercise hypotension induced by HIIT in relation to the control protocol. DBP and MBP responses over the 10hour period following the control and HIIT protocols are presented in FIGURES 1B and 1C, respectively. No effects of the protocols on these hemodynamic measures were found.
Regarding HR responses over the 10-hour period after the control and HIIT protocols (   Data are expressed as mean and standard deviations. *p < 0.05 compared to the control protocol. #p < 0.05 compared to pre-session.

Discussion
Exercise is an efficient non-pharmacological approach for prevention, treatment, and control of various morbidities 18 . However, the type and dose of exercise required to induce health benefi ts are still controversial. Th e high-intensity post-exercise hemodynamic responses assessed by means of ABPM in adolescents with excess body weight have not been investigated thoroughly in the literature.
Our results indicate that a single session of HIIT promoted a signifi cant reduction in SBP in the fi rst 60 minutes post-exercise when compared to the control condition, characterizing a hypotensive eff ect after the session.
According to the study by Brito et al. 13 , the characteristics of the population, such as age, BMI, and blood pressure status influence post-exercise hemodynamic responses. Therefore, in order to minimize bias in the study results, we only recruited overweight or obese male adolescents not diagnosed with hypertension. Th e HIIT protocol used in this study was based on the adolescents' tolerance of highintensity exercise and on the evidence of investigations that reported signifi cant eff ects of HIIT on health parameters in morbid populations, such as: obese, hypertensive, diabetic, and metabolic syndrome 14,24 .
Regarding post-exercise hemodynamic responses in adolescents with excess body weight, the results from the present study demonstrated that SBP was signifi cantly reduced in the fi rst hour (12:00 to 13:00 hours) following the HIIT protocol (approximately -4 mmHg) when compared to the control protocol. Th ese results are close to the magnitude of eff ect observed after HIIT in normotensive individuals, in which a post-exercise BP reduction between 3 and 8 mmHg was observed [26][27][28] . Th e literature suggests that post-exercise hypotension is greater in hypertensive individuals compared to normotensive individuals 7 , yet this could not be verifi ed in this study as we did not include a control group. An important fi nding from the present study was that post-exercise hypotension was of low magnitude and short duration, suggesting that frequent engagement in exercise is necessary for systematic reductions of BP.
Some methodological aspects in exercise prescription may have contributed to the similarity of the fi ndings of the present study with the results observed in the literature regarding the magnitude and duration of the hypotensive eff ect induced by HIIT in normotensive young adults. For example, the fi nal exercise volume (27 min. at ~ 70% VO 2peak ) from this study amounted to similar volumes used in the studies by Jones et al. 26 , Angadi et al. 28 , and Miyashita et al. 27 . Jones et al. 26 and Angadi et al. 28 employed a protocol of three 10-minutes series of 70% VO 2peak intensity exercise, whereas Miyashita et al. 27 . prescribed 10 series of 70% VO 2peak intensity exercise lasting for three minutes each.
Th e DBP and MBP over the 10-hour follow-up did not demonstrate signifi cant alterations either for the control or HIIT protocols. However, Figure 1C indicates that the HIIT protocol attenuated the increase in MBP between 12:00 and 1:00 PM, and promoted reductions in MBP values of approximately 2 mm Hg in subsequent hours. After the HIIT protocol, HR remained elevated from 12:00 to 3:00 p.m in relation to the pre-session and control conditions. An increase in ambulatory HR has been reported after performing aerobic exercises and may be attributed to increased sympathetic modulation and decreased vagal activity of the heart 13 .
In the present study, the comparison of pre-and post-session (10 hours) hemodynamic measurements between the experimental protocols indicated that HIIT promoted a reduction of 2 mmHg in MBP compared to pre-session (p < 0.05). Cohort studies indicate that outcomes of this magnitude can potentially reduce cerebral vascular accident mortality by up to 6%, coronary heart disease mortality by up to 4%, and all-cause mortality by as much as 3% 29 .
Reductions in ambulatory BP have a greater clinical signifi cance in relation to causal and selfassessed BP because such reductions are more valid in the prognosis for cardiovascular disease 22 . ABPM includes multiple measurements that more accurately refl ect an individual's BP values throughout daily activities. In addition, this technology eliminates many problems associated with clinical assessments such as observer bias and the white coat syndrome 22 .
A high intensity physical exercise session induces considerable cardiovascular stress during its performance and, consequently, promotes homeostatic imbalance in the body of children and adolescents 9 . Considering the post-exercise hypotension of low magnitude and short duration and the concomitant increase in HR, it is postulated that in the present study the physiological mechanisms of BP control acted in order to minimize pressure stress generated during the HIIT protocol and reach homeostatic balance. Among the principal mechanisms involved in this adjustment is the reduction in peripheral vascular resistance mediated by vasodilator agents produced during the session, such as nitric oxide, histamines, and prostaglandins 7 .
The hypotensive mechanisms of exercise are multifactorial and may be different between individuals 13 . This diversity may explain the contradictory results in this area of study in recent decades. Th e literature points out that the chronic eff ects of exercise on BP result from the accumulation of the acute physiological responses of each session 10 . Liu et al. 10 demonstrated a strong and signifi cant positive association (r = 0.89) between the reduction in BP after a single aerobic exercise session and the reduction in BP at rest after a certain period of aerobic training.
Th is research presents limitations such as lack of control of participants' eating habits, stress, and physical activity, which could alter BP values. However, to minimize bias, volunteers received guidelines to maintain similar routines of physical activity, eating, and sleep time in the 24 hours prior to the intervention and during the monitoring period 22 . In addition, the experimental design of this study included a control session (without exercise) to allow for withinsubject comparison of the exercise eff ect on BP.
Additional studies should be conducted using the HIIT physical training method in different populations and larger samples to assess BP responses to different exercise components, as well as to further evaluate the potential applications of the HIIT method in the prevention and treatment of hypertension in young individuals.
In conclusion the results of the present investigation indicate that, in adolescents with excess body weight, a HIIT protocol induced signifi cant SBP reductions in the fi rst 60 minutes post-exercise in relation to the control protocol. Th e results further indicate that the post-exercise hypotensive eff ect of the HIIT was of low magnitude and short duration.

Confl ict of interest
Th ey have no confl icts of interest.