THE EFFECT OF HIGH-INTENSITY INTERVAL TRAINING ON POST-EXERCISE OXYGEN CONSUMPTION: A META-ANALYSIS

ABSTRACT Introduction: The objective of this study was to present a systematic review and meta-analysis to compare total excess post-exercise oxygen consumption (EPOC) for two training intervention models in healthy individuals, and the secondary objective was to understand whether oxygen consumption after exercise could really promote a meaningful help. Design: To design a meta-analysis review to compare two training intervention models (experimental: high-intensity interval training; and control: continuous moderate-intensity) and their effects on total EPOC in healthy individuals. Participants: Seventeen studies were considered to be of good methodological quality and with a low risk of bias. Methods: Literature searches were performed using the electronic databases with no restriction on year of publication. The keywords used were obtained by consulting Mesh Terms (PubMed) and DeCS (BIREME Health Science Descriptors). Results: The present study findings showed a tendency (random-effects model: 0.87, 95%-CI [0.35,1.38], I2=73%, p<0.01) to increase EPOC when measured following high-intensity interval training. Conclusions: Our study focused on the analysis of high- and moderate-intensity oxygen uptake results following exercise. Despite the growing popularity of high-intensity interval training, we found that the acute and chronic benefits remain limited. We understand that the lack of a standard protocol and standard training variables provides limited consensus to determine the magnitude of the EPOC. We suggest that longitudinal experimental studies may provide more robust conclusions. Another confounding factor in the studies investigated was the magnitude (time in minutes) of VO2 measurements when assessing EPOC. Measurement times ranged from 60 min to 720 min. Longitudinal studies and controlled experimental designs would facilitate more precise measurements and correct subject numbers would provide accurate effect sizes. Systematic reviewb of Level II studies.


INTRODUCTION
Due to the increase of obesity in recent years, there has been an increase in the search for strategies to help reduce fat mass. One non--pharmacological strategy is exercise. Several training designs, models of exercise, and different intensities and durations have been used to increase energy expenditure during and after exercise. Energy expenditure post exercise is normally quantified by measuring excess post exercise oxygen consumption (EPOC). [1][2][3][4][5] However, finding the exercise mode that increases energy expenditure after exercise is difficult. In addition, an intensity that can be used to control, maintain and decrease body weight and control diseases associated with obesity is also desirable. Energy expenditure during and post-exercise is measured by oxygen uptake (VO 2 ) using a gas analyzer. 6,7 During exercise, there is an increase in VO 2 to support increased energy needs. Post exercise, VO 2 does not return to resting levels immediately and may remain elevated for some time. Exercise intensity is an important factor in the determination of excess post-exercise oxygen consumption (EPOC). 3 There are several models of training (resistance training, high-intensity training, and continuous training) that can increase EPOC between 1 to 48 hours above resting levels. 1,3,[8][9][10][11][12][13] . In this context, it has been suggested that there is a curvilinear relationship between EPOC magnitude (total O 2 consumed during recovery) and exercise intensity.
High-intensity interval training (HIIT) has been recommended because of the relatively rapid, increased amount of energy expenditure during and after exercise when compared to continuous aerobic training. However, aerobic training has been reported as an effective method to control or lose weight. 14 On the other hand, resistance training (RT) has been described as intermittent in nature and might induce a prolonged EPOC during recovery. 2 Tucker et al. 15 have suggested that it is unlikely that the greater fat loss observed after interval exercise training reported in some studies is due to greater EPOC after interval exercise. In this context, Binzen et al. 2 investigated the acute effects of 45 min of RT on EPOC and substrate oxidation 120 min following exercise in moderately trained women. The overall 2h EPOC was 6.2 L (RT: 33.4 ± 5.1 L vs. control: 27.2 ± 0.3 L), corresponding to an 18.6% elevation over the measurement period.
The literature seems inconclusive about the magnitude effect of EPOC and the relationship with the intensity of exercise during training. Nowadays, professionals have been recommending high-intensity interval training models related to the relative and absolute increases in energy expenditure following exercise. However, high-intensity interval training models may not be effective for all individuals, especially sedentary, elderly and overweight/obese individuals. [16][17][18] Objectives Therefore, the aim of this study was to present a systematic review and meta-analysis to compare two training intervention models (experimental: high-intensity interval training; and control: continuous moderate-intensity) in total oxygen consumption during recovery (EPOC) in healthy individuals in training, and the secondary objective was to understand whether oxygen consumption after exercise really could promoter meaningfully help.

METHODS
The meta-analysis review was carried out in accordance with the recommendations of Khan et al. 19 considering: 1) framing of the questions for a literature review; 2) identification of the relevant research; 3) evaluation of the quality of the studies; 4) summary of the evidence; 5) and interpretation of the results. In addition, we adhered to the 27 items by checklists of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA). To ensure transparency and complete communication this systematic review and meta-analysis complied with suggestions outlined previously. 20 The research questions were defined by the PICOS model in accordance with PRISMA guidelines, as follows: 1. Population: males and females with experience in training. 2. Intervention: an acute session which incorporated a high-intensity training design. 3. Comparator: oxygen uptake compared to other interventions (moderate-intensity training). 4. Outcomes: amount of oxygen uptake after exercise. 5. Study design: randomized controlled designs, counterbalanced crossover or repeated measure designs that investigated the acute oxygen uptake responses from high-intensity training.
The review was approved and registered at National Institute for Health Research -International prospective register of systematic reviews (PROSPERO) CRD42020170195 last 04/28/2020.

Type of Studies
We included randomized clinical trials (with parallel-group design, within-person design, cluster design, or the first phase of cross-over trials) evaluating the mindfulness strategies and programs on the resistance training systems, compared to each resistance training system. We excluded non-randomized clinical trials, such as cohorts, case-control, and case reports studies. We did not impose language, publication date, or status restrictions for potentially retrieved records.

Types of participants
We included studies of adult both genre (aged 18 years or over) adults with or without experience in resistance training, without diagnostic diseases.

Type of interventions
We included studies that assessed the effects of training using high-intensity compared to low-or moderate-intensity exercise.

Type of outcome measures
Primary outcomes 1. Oxygen consumption (liter) and calorie (kcal) 2. Adverse effects (e. g., worsening of the parameters mentioned above after treatment) Secondary outcomes 1. Metabolic changes 2. Change in level of cardiorespiratory fitness 3. Effect of exercise on heart rate

Literature search
For this review, literature searches were performed using the (Virtual Library for Health -BVS, PubMed, Embase, Ebsco SPORTDiscus and Science Direct) electronic databases without any year restriction. Manual reference searching was performed to identify other relevant studies. The keywords used were obtained through consultation of Mesh Terms (PubMed) and DeCS (keywords of subjects in BIREME health science). The combination of excess post-exercise oxygen consumption (epoc; oxygen consumption; oxygen; metabolic equivalent) and high intensity interval training (High-Intensity; Sprint Interval Training; High-Intensity Intermittent; exercise) with "AND" and "OR" combination: epoc and exercise (pubmed) ((epoc[All Fields] AND ("exercise"[MeSH Terms] OR "exercise"[All Fields])) AND Search((epoc[Title/Abstract]) AND (("Oxygen Consumption"[Mesh] OR "Consumption, Oxygen" OR "Consumptions, Oxygen" OR "Oxygen Consumption" OR "Metabolic Equivalent"))) AND (("High-Intensity Interval Training"[Mesh]High Intensity Interval Training OR "High-Intensity Interval Trainings" OR " Interval Training, High-Intensity" OR "Interval Trainings, High-Intensity" OR "Training, High-Intensity Interval" OR "Trainings, High-Intensity Interval" OR "High-Intensity Intermittent Exercise" OR "Exercise, High-Intensity Intermittent" OR "Exercises, High-Intensity Intermittent" OR "High-Intensity Intermittent Exercises" OR "Sprint Interval Training" OR "Sprint Interval Trainings")). After the removal of duplicates, the title and abstract of each article were initially screened for suitability. Full-text articles were retrieved in order to determine inclusion or exclusion. Two authors (BCL and EFR) performed the search independently. In the case of any selection bias, a third assessor (GAJ) was included. The search was conducted throughout January 2018 and updated in December of 2019.

Searching other resources
Additionally, we checked the reference list and citations of eligible studies, grey literature (Open Grey, www.opengrey.eu), and related systematic reviews. Where required, we attempted to contact the authors of the original reports for clarification or to request missing data.

Inclusion and exclusion criteria
Studies were included in this review if they met the following criteria: (a) implemented high-intensity in comparison to moderate-intensity; (b) results reported in oxygen consumption (liter) and calorie (kcal); (c) the study had an acute design or part thereof; and (d) was published in an English-language peer-reviewed journal.

Selection of studies and reviewing process
To increase reliability, two researchers (GAJ and DR) performed the analyses independently during all stages of the study, and in the case of a discrepancy, a third assessor (GAJ) was used as a moderator. For all included articles, the following data were extracted: (1) study characteristics (author, year, sample size and study design); (2) participant demographics (age, sex and training experience); (3) protocols of the training (high-intensity, and moderate-intensity structure [i.e. rest period, number of sets and repetitions, duration the session, exercise selection and intensity according to the previous studies]); 21,22 and (4) outcome measures (VO 2 [L], and showed calorie value [kcal]) post-intervention and reported an average change and standard deviation using a validated measure. The reference lists of articles retrieved were then screened for any additional articles that had relevance to the topic, according to previous publications 23 (Figure 1).

Data extraction and management
Data extraction forms were used to extract data from each study. Data extracted included the size and characteristics of the sample (i.e., age, gender, body weight, height, mass fat, free-fat mass, experience of resistance training), characteristics of the interventions (study design, number of sessions, duration of each session of treatment, intensity of training, a model of training), instruments used to evaluate the outcomes (oxygen consumption), and results of the included studies. Two independent reviewers performed the data extraction. Any disagreements were resolved by a third reviewer. When data were not available in the manuscripts or in the case of uncertainty, the authors were contacted where possible for clarification.

Risk of bias and quality assessment
The risk of bias in the studies was assessed by three authors (GAJ, DR and AF) the according to The Joanna Briggs Institute (JBI) Critical Appraisal tools for use in JBI Systematic Reviews. 24 The JBI critical appraisal checklist for analytical cross-sectional studies for analyzing by the risk of bias was assessed by considering the following questions: were the criteria for inclusion in the sample clearly defined? were the study subjects and the setting described in detail? was the exposure measured in a valid and reliable way? were objective, standard criteria used for the measurement of the conditions? were confounding factors identified? were strategies to deal with confounding factors stated? were the outcomes measured in a valid and reliable way? was appropriate statistical analysis used?
These systematic reviews incorporated a process of critique and appraisal of the research evidence. Therefore, the purpose of this appraisal was to assess the methodological quality of a study and to determine the extent to which a study has addressed the possibility of bias in its design. Conduct, and analysis according to previous models were employed in this meta-analysis 25,26 (Figures 2 and 3).

Measure of treatment effect
The random-effects meta-analysis was conducted for the performance variable oxygen consumption. The performance variable outcome was presented as standardized mean differences SMD ± standard deviation (SD), and 95% confidence interval (CI) values. For each study, SMD was computed such that positive values indicate that the intervention group (i.e. high-intensity training) was superior to the control group (i.e. moderate-intensity training). 27 Dealing with missing data Missing data was dealt with as outlined in Chapter 10 of the Cochrane Handbook of Systematic Reviews; hence, where possible, we performed intention-to-treat analysis for primary and secondary outcomes (randomized studies). Irrespective of the study design, we tried to contact the trial investigators or sponsors to obtain missing outcome data. Where these data remain unavailable, we rated the relevant domains of the Cochrane tool for assessing the risk of bias accordingly.

Assessment of heterogeneity
We assessed statistical heterogeneity employing the Cochran Q test to determine the strength of evidence that any heterogeneity was genuine. We considered a threshold of P-value < 0.1 as an indicator of whether heterogeneity (genuine variation in effect sizes) is present. In addition, we examined and interpreted the I² statistic as follows: < 25% (no heterogeneity); 25% to 49% (low heterogeneity); 50% to 74% (moderate heterogeneity); ≥ 75% (high heterogeneity). 28  Therefore, the effect of training type was determined by standardized SMD values post-intervention after calculating the inverse of the variance. 29,30 The amount of heterogeneity was estimated (with the DerSimonian-Laird estimator) and incorporated into the standard error of the estimated average effect and the corresponding confidence interval.

Assessment of reporting biases
Funnel plots and Trim and fill were used to assess publication bias using Egger's regression tests where non-significant asymmetry indicated no bias. 31

Data synthesis
The meta-analyses We used the metafor package version 1.2-1 and rmeta version 3.0 implemented in R-3.6.2 software for Mac to perform and synthesize the direct and indirect evidence of the oxygen consumption post exercise effect. Therefore, all analyses were performed using package meta in R version 1.0.4.4 -© 2009-2016 RStudio, Inc (The R Foundation for Statistical Computing, Vienna, Austria). An α level of p < 0.05 was used to determine statistical significance.

Subgroup analysis and investigation of heterogeneity
We performed subgroup analysis in case of heterogeneity considering the following variables: VO 2 or kcal post-exercise: RT = studies that included only resistance exercise, and used instrument equipment, free-weights; RU = studies that included only running training and used a treadmill; and CY = studies that included only cycling training, and used a cycle-ergometer).

Sensitivity analysis
We performed a sensitivity analysis that included: Effects of risk of bias by excluding trials with high or unclear risk of bias; Influence of unpublished studies excluding trials with abstracts only; and Influence of sponsorship by excluding industry-funded studies.

Characteristics of included trials and participants
The 17 studies were judged to be of good methodological quality and at low risk of bias. Full details of the risk of bias are presented in (supplementary material). The presented high-intensity interval training n = 152 experimental group and moderate-intensity training n = 150 control groups (total of 302 adults from both genders were included and randomized respectively). The mean age ranged from 26 The intervention characteristics are outlined in Table 2. Interventions were conducted to compare high-intensity interval training and continuous moderate-intensity in all studies. Ten studies 1,4,5,32-38 reported by VO 2 in liters; three studies 2,39,40 reported outcomes only in energy expenditure in calories (kcal); and five studies 15,41-44 reported outcomes in both (VO 2 and kcal) respectively.
The cycle ergometer was the most common modality of exercise (eight studies), 4,5,15,34,36,37,41,43 followed by the Treadmill (seven studies), 4,32,33,35,37,38,44 and resistance exercise was the least common modality of exercise (two studies). 1,42 Table 2 shows a large effect in favor of high-intensity training for VO 2 post-exercise using a cycle ergometer or treadmill. There were different significances in favor of high-intensity exercise for energy expenditure (calorie) when the intervention was resistance exercise.
Thus, there is evidence that the results of the meta-analysis were influenced by a publication bias. After analysis the asymmetry in funnel (t = 1.09; = -0.61 (13), p-value = 0.30), we used the Trim and Fill method for the adjusted effect size.
The adjusting for publication bias showed studies used of high-intensity training seems a positive influence in increasing energy expenditure and/ or uptake oxygen post-exercise. The full details are summarized in Figure 4.

DISCUSSION
Our study focused specifically on evaluating high-intensity interval training and moderate-intensity results for oxygen uptake following exercise. We observed that the studies are not conclusive in relation to EPOC. Our study demonstrated a tendency (random-effects model: 0.87, 95%-IC [0.35; 1.38], I 2 = 73%, p < 0.01) to increase EPOC when the exercise performed was high-intensity interval training (Figure 4).
After adjusting for publication bias, studies that used high-intensity interval training seemed to positively influence the increase in energy expenditure and/or oxygen uptake after exercise. However, there was a large heterogeneity observed (I 2 = 73%) between the studies (Figure 4). Were the criteria for inclusion in the sample clearly defined?
Were the study subjects and the setting described in delaiI?
Was the exposure measured in a valid and reliable way?
Were objective, standard criteria used for measurement of the condition?
Were confounding factors identified?
Were strategies to deal with confounding factors stated?
Were the outcomes measured in a valid and reliable way?
Was appropriate statistical analysis used?

0%
10% 20% 30% 40% 50% 60% 70% 80% 90% 100%   The heterogeneity among the studies can be explained by examination of different variables including, age, sex, physical condition, oxygen collection instrument, type of protocol, VO 2 intensity, effort strength time, modality mode, weekly frequency, and the magnitude (time-min) of the analysis of oxygen consumption after exercise. In particular, the EPOC magnitude effect presented ranges from 10 to 90 minutes. 4,5,15,32,34,38,39,43,44 Other studies were analyzed for over 90 min. 1,2,[35][36][37]41 Despite the limitations by heterogeneity, our data demonstrate that the intensity of effort can be considered as a determining factor for increasing EPOC during exercises using both treadmills and bicycles. As for caloric parameters, we observed significant changes only in RT (Table 2).
Although our results point to a significant trend towards high intensity exercise, studies are inconclusive when analyzed individually.
For example, studies by Turker et al. 15 compared EPOC after high--intensity interval exercise (HIE), and sprint interval exercise (SIE), and steady-state exercise (SSE). Ten recreationally active males participated in a randomized crossover trial. Although 3h EPOC and total net EE after exercise were higher (p=0.01) for SIE (22.0 ± 9.3 L; 110 ± 47 kcal) compared to SSE (12.8 ± 8.5 L; 64 ± 43 kcal), total (exercise + post exercise) net O 2 consumed and net EE were greater (p=0.03) for SSE (69.5 ± 18.4 L; 348 ± 92 kcal) than for SIE (54.2 ± 12.0 L; 271 ± 60 kcal). On the other hand, Schaun et al. 44 compared the energy expenditure during and after two treadmill protocols, high-intensity interval training (HIIT) and moderate continuous training (CONT), in young adult men. The protocols HIIT (8 bouts, 20s at 130% of the velocity associated with the VO 2 max. on a treadmill with 10s of RI) versus CONT (30min on a treadmill at a submaximal velocity equivalent to 90-95% of HR associated with the anaerobic threshold). No difference was found between the groups for VO 2 , EE and EPOC post-exercise and were higher than HIIT (69.31 ± 10.88; 26.27 ± 2.28 kcal, respectively).

CONCLUSIONS
Our study focused on the analysis of high and moderate-intensity exercise results and effects on oxygen uptake following exercise. Despite the growing popularity of high-intensity interval training, we found that acute and chronic benefits remain limited. We understand that lack a of similar protocols and the control of the variables that influence training outcomes, will affect the measures that are used to determine EPOC magnitude. We also suggest that controlled longitudinal studies would reveal additional perspectives in relation to the measurement of EPOC. A further confounding factor is the magnitude (time in minutes) of VO 2 measurement during EPOC assessment as it ranged from 60 min to 720 min. Longitudinal studies and controlled experimental design would permit a higher combination of effect size which would be a desirable outcome. The findings of the present study showed a tendency (random-effects model: 0.87, 95%-IC [0.35; 1.38], I 2 = 73%, p < 0.01) for increases in EPOC post exercise when the exercise performed prior to EPOC measurement was high-intensity interval training.