Physical activity and organized exercise regimens are widely acknowledged as fundamental approaches to managing cardiometabolic health, irrespective of additional lifestyle considerations (Edwards et al., 2023). An extensive body of empirical research spanning the last decade has consistently documented notable enhancements in diverse cardiometabolic fitness parameters and health indicators after exercise interventions, including glycemic markers (Boer et al., 2014; Martin et al., 2015; Dias et al., 2018; Kranen et al., 2023), blood lipid profile (Ahmadi et al., 2020; Paahoo et al., 2021; Khaliltahmasebi et al., 2022), and blood pressure (Espinoza-Silva et al., 2019; Ketelhut et al., 2020; Oliveira et al., 2021; Salus et al., 2022a; 2022b). Ultimately, these cumulative effects diminish the overall risk of cardiovascular disease within the broader population (Kranen et al., 2023; Bond et al., 2015).

Cardiovascular disease stands as the primary etiology of noncommunicable mortality globally and is projected to endure as an enduring and ubiquitous threat internationally (Kranen et al., 2023). A seemingly inevitable escalation in obesity and diabetes presently constitutes the foremost obstacle in the ongoing endeavor to diminish the burden of cardiovascular disease worldwide, and cholesterol, especially low-density lipoprotein cholesterol, is considered the primary determinant of cardiovascular disease (Timmis et al., 2020; Sayevand et al., 2022). Blood pressure also has a linear correlation with the incidence of stroke and myocardial infarction (Prospective Studies Collaboration, 2002), and treatment aimed at reducing blood pressure levels protects patients against cardiovascular events (Xie et al., 2016).

Different exercise interventions have traditionally been prescribed to diminish the abovementioned factors, with the majority focusing on moderate-intensity training, generally completed by running (Edwards et al., 2023). However, despite the indisputable advantages of global exercise training guidelines (Bull et al., 2020), adopting and complying with the existing guidelines remain inadequate (Edwards et al., 2023), with the “insufficient time” as a primary hindrance to their consistent involvement regularly (Gillen and Gibala, 2014). Hence, it seems sensible to investigate innovative exercise training methods that could enhance the acceptance and commitment of the general public.

High-intensity interval training (HIIT), characterized by repeated bouts of intensive work interspersed by periods of low-intensity exercise or complete rest (Gillen and Gibala, 2014), has been introduced as an alternative to moderate-intensity training. Despite a lower time commitment, HIIT induces numerous adaptations resembling moderate-intensity training (Little et al., 2010). HIIT may also have superior effects on modification of cardiometabolic risk in youth than moderate-intensity training (Hay et al., 2012; Carson et al., 2014). Adolescents also indicated higher enjoyment during HIIT than moderate-intensity training suggesting HIIT as a useful approach for preventing cardiovascular disease (Kranen et al., 2023). Several methods for prescribing HIIT have been created to assist athletes in reaching targeted exercise intensities during their training sessions in a structured and individualized way. These methods may include using a rating of perceived exertion (RPE) to guide intensity, utilizing maximal aerobic speed and power metrics, which are thought to be critical prescription components for many categories; anaerobic speed/power reserve measures; and the upper limits of high-intensity performance above the velocity or power linked to maximal oxygen uptake [V̇O_2max (v/pV̇O_2max). Repeated all-out sprinting efforts or sprint interval training (SIT) is another form of HIIT performed at all-out or maximal intensities over a given distance or duration. Since such exercise is consistently performed all-out, it can be prescribed without pretesting individual reference intensities (i.e., v/pV̇O_2max) (Laursen and Buchheit, 2019).

A wide range of HIIT studies have shown significant impacts on different cardiometabolic risk factors (Fereshtian et al., 2017; Costa et al., 2018; Sheykhlouvand et al., 2018; 2022; 2024; Campbell et al., 2019; Reljic et al., 2020; Edwards et al., 2022; 2023; Gharaat et al., 2024; Tao et al., 2024; Song and Sheykhlouvand, 2024). Although these studies provide invaluable information regarding the effects of HIIT on different cardiometabolic risk factors, the studies’ variables are limited to single risk factors and exercise modalities, which may restrict their ability to capture a complete picture of the factors that could be affected by different type of interval interventions. Moreover, studies regarding children and adolescents are limited. Over the past two decades, childhood obesity has escalated to epidemic levels globally (Grossman et al., 2017). An epidemiological study conducted in 2017 revealed that there are approximately 107.7 million obese children in 195 countries, and the prevalence of obesity among children is higher compared to adults (GBD 2015 Obesity Collaborators, 2017). A recent update by the World Health Organization (WHO, www.who.int) indicates that over 390 million children and adolescents aged 5–19 years were overweight in 2022, including 160 million who were living with obesity. Childhood obesity not only raises the risk of cardiovascular disease but also leads to conditions such as adult coronary heart disease, hypertension, metabolic syndrome, and type II diabetes mellitus (Cole and Lobstein, 2012). Studies showed that disorders of glycolipid metabolism may originate from childhood; obesity can accelerate this situation (Cao et al., 2021). Hence, identifying the effective interventions in modulating the mentioned risk factors would be of value. Although some studies have been dedicated to these age categories, no previous study has performed a systematic review and meta-analysis to investigate the effect of HIIT and SIT on cardiometabolic health markers in children and adolescents. Hence, the present study aimed to conduct a novel meta-analysis on the effects of different intensive interval interventions (HIIT and SIT) on multiple cardiometabolic health indicators in children and adolescents.

This systematic review and meta-analysis are reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guideline.

Three electronic databases, including PubMed, Web of Science, and Scopus, were searched. The search strategy was performed on July 2024 using the following keywords: "high intensity interval training" OR "high-intensity interval training" OR "high intensity interval exercise" OR "high-intensity interval exercise" OR "high-intensity intermittent training" OR "high intensity intermittent training" OR "high intensity intermittent exercise" OR "high-intensity intermittent exercise" OR "aerobic interval training" OR "aerobic-interval training" OR "aerobic interval exercise" OR "aerobic-interval exercise" OR "interval training" OR "interval exercise" OR "sprint interval training" OR "sprint interval exercise" OR "sprint training" AND children OR childhood OR child* OR adolescent OR youth OR pediatrics (Table 1). In addition, the search was updated up to July 2024. Also, the references list of applicable articles and Google Scholar were manually searched to find other relevant articles. The search terms and strategies were performed in consultation with information specialists to ensure comprehensiveness and rigor in conducting the literature search.

Articles were included if they met the following criteria: (1) peer-reviewed and English languages published articles, (2) randomized and non-randomized control trials comparing HIIT versus non-exercise control groups, (3) studies involving children and adolescents with mean age ranges from 6 to 18 years, (4) studies involving exercise training with intervention duration ≥ two weeks, (5) outcomes measures including glycemia markers including fasting glucose and insulin and insulin resistance, lipid profiles including triglyceride, total cholesterol, low-density lipoprotein cholesterol and high-density lipoprotein cholesterol, and blood pressure including systolic blood pressure and diastolic blood pressures. Studies were excluded if participants were≥ 18 years old, articles were published in non-English-language journals, studies included a single-arm trial (without a control group), and studies included exercise training combined with other interventions, such as caloric restriction. In addition, non-original studies, such as reviews, were excluded. Two reviewers independently selected and identified studies according to inclusion and exclusion criteria, and disagreements were resolved by discussion. All articles were included in EndNote software, and after removing duplicates, a screen was conducted based on the title and abstract. Subsequently, the remaining articles were screened based on their full text.

Two reviewers independently conducted data extractions on a pre-designed template. Data retrieved from each study included study design, participants’ characteristics, including sample size, body mass index, age, sex, and health status, HIIT characteristics, including program description, intensity, duration, frequency, and numbers of sets, and outcomes data. To calculate the effect size, mean and standard deviation (SD) or mean changes and their SDs in the training and control groups were extracted. When required, these data were extracted from the figure using Getdata or were calculated from other data such as median and data range or standard errors (Reeves et al., 2008; Wan et al., 2014). In addition, if a study had more than one intervention arm, all were included as a separated trial.

Study quality was assessed using the Physiotherapy Evidence Database (PEDro) which contains 11 items. Two items, including blinding of participants and intervention, were excluded from the scale (De Morton, 2009). Finally, the quality of the included studies was assessed using 9 items that were performed by two independent reviewers. Details are provided in Table 2.

The meta-analysis of the included studies was conducted using comprehensive meta-analysis software version 3 (CMA3). To calculate effect size, standardized mean differences (SMD), weighted mean differences (WMD), and confidence intervals were calculated using a random effect model. The effect size was expressed as WMD when all studies reported the outcome in the same unit and was expressed as SMD when all studies reported the outcomes in the same units. For SMD, SMD of 0.2 to 0.49, 0.5 to 0.79, and > 0.8 were considered to represent small, medium, and large effect sizes, respectively (Cohen, 2013). Heterogeneity was assessed using Q statistics and I² values, where I² values of 25%, 50% and 70% were considered to represent low, moderate and high heterogeneity, respectively (Reeves et al., 2008). Publication bias was assessed using visual interpretation of a funnel plot and the Egger test (Sterne and Egger, 2001). In addition, several subgroup analyses were performed based on age (children: age < 12 years and adolescents: age ≥ 12 years), obesity (obese and non-obese), and mode of interval training (SIT and HIIT). The sensitivity of the analysis was determined by removing individual studies to ensure that individual results do not influence the results. In addition, analysis sensitivity was performed by removing non-randomized trials and studies with a high risk of bias.

The database searches yielded 948 articles from Scopus, 829 articles from PubMed, and 662 articles from Web of Science, of which 1529 articles remained after removing duplicate records. Then, 1398 articles were eliminated based on titles and abstracts and subsequently 131 articles were removed based on the full-text screen as presented in Figure 1. In addition, one study was added after search updates (R). Finally, 31 articles met all inclusion criteria and were included in the meta-analysis (R)(Tjønna et al., 2009; Rosenkranz et al., 2012; Racil et al., 2013; 2016a; 2016b; Boer et al., 2014; Martin et al., 2015; Chuensiri et al., 2018; Cvetković et al., 2018; van Biljon et al., 2018; Delgado-Floody et al., 2018; Dias et al., 2018; Espinoza-Silva et al., 2019; 2023; Taylor et al., 2019; Ahmadi et al., 2020; Ketelhut et al., 2020; Plavsic et al., 2020; McNarry et al., 2021; Oliveira et al., 2021; Paahoo et al., 2021; Khaliltahmasebi et al., 2022; Abassi et al., 2022; Meng et al., 2022; Martínez-Vizcaíno et al., 2022; Williams et al., 2022; Salus et al., 2022a; 2022b; Kranen et al., 2023; Tadiotto et al., 2023). Among the included studies, 23 were randomized control trials, and the remaining eight were non-randomized control trials (Espinoza-Silva et al., 2019; 2023; Salus et al., 2022a; 2022b; Delgado-Floody et al., 2018; Tadiotto et al., 2023; Taylor et al., 2019; van Biljon et al., 2018). Thirty-one studies included 2,496 children and adolescents with sample sizes ranging from16 (Rosenkranz et al., 2012) to 562 (Martínez-Vizcaíno et al., 2022) and mean ages ranging from 6.4 (Espinoza-Silva et al., 2023) to 17.7 (Boer et al., 2014) years old. Table 3 presents more details of participants' characteristics. Of the included studies, 17 used SIT (Tjønna et al., 2009; Rosenkranz et al., 2012; Racil et al., 2013; 2016a, 2016b; Boer et al., 2014; Martin et al., 2015; Cvetković et al., 2018; van Biljon et al., 2018; Taylor et al., 2019; Ketelhut et al., 2020; Paahoo et al., 2021; Meng et al., 2022; Williams et al., 2022; Abassi et al., 2022; Khaliltahmasebi et al., 2022; Tadiotto et al., 2023), and 14 studies used HIIT (R) (Dias et al., 2018; Ahmadi et al., 2020; Kranen et al., 2023; Espinoza-Silva et al., 2019; 2023; Oliveira et al., 2021; Chuensiri et al., 2018; Delgado-Floody et al., 2018; Martínez-Vizcaíno et al., 2022; McNarry et al., 2021; Plavsic et al., 2020; Tjønna et al., 2009; van Biljon et al., 2018), of which intervention duration ranged from two weeks (Williams et al., 2022) to 12 months (Meng et al., 2022; Tjønna et al., 2009). Twelve weeks were most relevant. More details of HIIT characteristics are presented in Table 3.

Figure 2, Figure 3 and Figure 4 depict the change in fasting glucose, insulin, and insulin resistance following HIIT compared to the control group. Overall, HIIT significantly decreased insulin [SMD: –0.78 (CI: –1.41 to –0.15), p = 0.01, 13 trials] and insulin resistance [SMD: –1.01 (CI: –1.64 to –0.39), p = 0.002, 12 trials], but not glucose [WMD: –0.07 mmol/l (CI: –0.18 to 0.03), p = 0.18, 18 trials]. There were significant heterogeneities between studies for insulin (I² = 87.81, p = 0.001), glucose (I² = 54.44, p = 0.005), and insulin resistance (I² = 86.46, p = 0.001). Visual interpretation of funnel plot and Egger’s test suggested publication bias for insulin resistance (p = 0.04) and only Egger’s test showed publication bias for insulin (p = 0.07). Both Visual interpretation of the funnel plot, and Egger’s test did not show publication bias for glucose (p = 0.10). The sensitivity of analyses shows that excluding non-randomized studies on overall effect sizes was insignificant. In addition, the sensitivity of analyses by removing studies with a high risk of bias did not change the significance.

After analyzing the different subgroup analyses, insulin was decreased by SIT [SMD: –1.21 (CI: –1.94 to –0.48), p = 0.001, 9 trials] and in those with obesity [SMD: –1.12 (CI: –1.81 to –0.42), p = 0.002, 9 trials] and adolescents [SMD: –1.00 (CI: –1.70 to –0.29), p = 0.005, 10 trials]. Subgroup analyses for insulin resistance indicate that SIT decreased insulin resistance [SMD: –1.30 (CI: –2.12 to –0.48), p = 0.002, 9 trials] and in those with obesity [SMD: –0.84 (CI: –1.35 to –0.32), p = 0.001, 10 trials] and adolescents [SMD: –1.25 (CI: –2.01 to –0.48), p = 0.001, 9 trials]. Glucose was decreased by SIT [WMD: –0.11 mmol/l (CI: -0.22 to –0.00), p = 0.03, 10 trials]

Figure 5, Figure 6, Figure 7 and Figure 8 depict the changes in triglyceride, total cholesterol, low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol following HIIT compared to the control group. Overall, HIIT significantly decreased triglyceride [WMD: –0.12 mmol/l (CI: –0.21 to –0.04), p = 0.004, 18 trials], total cholesterol [WMD: –0.23 mmol/l (CI: –0.37 to –0.08), p = 0.002, 17 trials] and low-density lipoprotein cholesterol [WMD: –0.23 mmol/l (CI: –0.35 to –0.11), p = 0.001, 17 trials] and increased high-density lipoprotein cholesterol [WMD: 0.07 mmol/l (CI: 0.02 to 0.12), p = 0.004, 18 trials]. There were significant heterogeneities between studies for triglyceride (I² = 44.82, p = 0.02), total cholesterol (I² = 57.72, p = 0.002), low-density lipoprotein cholesterol (I² = 57.29, p = 0.002) and high-density lipoprotein cholesterol (I² = 56.40, p = 0.001). Visual interpretation of the funnel plot suggested publication bias for triglyceride, total cholesterol, and high-density lipoprotein cholesterol, but Egger’s test did not show publication bias for triglyceride (p = 0.13), total cholesterol (p = 0.70), and high-density lipoprotein cholesterol (p = 0.96). In addition, both visual interpretation of funnel plot and Egger’s test did not suggest publication bias for low-density lipoprotein cholesterol (p = 0.33). Sensitivity of analyses shows that the either exclusion of studies and non-randomized studies on overall effect sizes were not significant. In addition, the visual interpretation of the funnel plot and Egger’s test did not suggest publication bias for low-density lipoprotein cholesterol (p = 0.33).

After analyzing the different subgroups, total cholesterol was decreased by SIT [WMD: –0.26 mmol/l (CI: –0.47 to –0.04), p = 0.01, 10 trials] and tended to decrease by HIIT [WMD: –0.17 mmol/l (CI: –0.34 to –0.00), p = 0.05, 7 trials], in those with obesity [WMD: –0.28 mmol/l (CI: –0.41 to –0.16), p = 0.001, 14 trials], adolescent [WMD: –0.19 mmol/l (CI: –0.31 to –0.06), p = 0.003, 9 trials] and children [WMD: –0.30 mmol/l (CI: –0.60 to –0.01), p = 0.04, 8 trials]. HIIT [WMD: –0.15 mmol/l (CI: –0.29 to –0.01), p = 0.03, 8 trials] and SIT [WMD: –0.11 mmol/l (CI: –0.22 to –0.002), p = 0.04, 10 trials] decreased triglyceride, in those with obesity [WMD: –0.15 mmol/l (CI: –0.35 to –0.05), p = 0.003, 15 trials], adolescents [WMD: –0.06 mmol/l (CI: –0.12 to –0.00), p = 0.03, 10 trials], and children [WMD: –0.27 mmol/l (CI: –0.47 to –0.08), p = 0.005, 8 trials]. SIT decreased low-density lipoprotein cholesterol [WMD: –0.38 mmol/l (CI: –0.49 to –0.26), p=0.001, 9 trials] in those with obesity [WMD: –0.28 mmol/l (CI: –0.39 to –0.16), p = 0.001, 13 trials], adolescent [WMD: –0.15 mmol/l (CI: –0.26 to –0.04), p = 0.005, 10 trials] and children [WMD: –0.37 mmol/l (CI: –0.59 to –0.15), p = 0.001, 7 trials]. High-density lipoprotein cholesterol was increased by SIT [WMD: 0.06 mmol/l (CI: 0.04 to 0.09), p = 0.001, 9 trials], in those with obesity [WMD: 0.06 mmol/l (CI: 0.04 to 0.08), p = 0.001, 14 trials], adolescent [WMD: 0.09 mmol/l (CI: 0.008 to 0.18), p = 0.03, 11 trials] and children [WMD: 0.07 mmol/l (CI: 0.05 to 0.10), p = 0.001, 7 trials].

Figure 9 and Figure 10 depict the changes in systolic blood pressure and diastolic blood pressures following HIIT compared to the control group. Overall, HIIT significantly decreased systolic blood pressure [WMD: –2.35mm Hg (CI: –3.79 to –0.91), p = 0.001, 24 trials], but not diastolic blood pressures [WMD: –0.92 mm Hg (CI: –2.31 to 0.47), p = 0.19, 24 trials]. Significant heterogeneities existed between systolic blood pressure studies (I2=40.68, p = 0.02) and diastolic blood pressures (I² = 60.11, p = 0.001). Visual interpretation of the funnel plot and Egger’s test did not suggest publication bias for systolic blood pressure (p=0.74), and only Visual interpretation of the funnel plot showed publication bias for diastolic blood pressures (Egger’s test p = 0.90). The sensitivity of analyses shows that the exclusion of studies on overall effect sizes was not significant. However, removing non-randomized studies led to a significant effect size for diastolic blood pressures [WMD: –1.90 mm Hg (CI: –3.71 to –0.10), p=0.03, 14 trials]. The sensitivity of analyses shows that excluding non-randomized studies on overall effect sizes was insignificant. In addition, the sensitivity of analyses by removing studies with a high risk of bias did not change the significance.

After analyzing the different subgroup analyses, systolic blood pressure was decreased by SIT [WMD: –3.97 mm Hg (CI: –6.33 to –1.61), p = 0.001, 11 trials] and in those with obesity [WMD: –3.30 mm Hg (CI: –4.79 to –1.81), p = 0.001, 18 trials] and children [WMD: -1.79 mm Hg (CI: –3.13 to –0.45), p = 0.008, 13 trials] and tended to significant in adolescents [WMD: –2.69 mm Hg (CI: –5.53 to 0.14), p = 0.06, 11 trials]. Subgroup analyses show that diastolic blood pressure only was decreased by SIT [WMD: –2.22 mm Hg (CI: –3.97 to –0.47), p = 0.01, 11 trials].

The present meta-analysis analyzed glycemic markers, lipid profiles, and blood pressure to determine the impact of HIIT on cardiometabolic health in children and adolescents by analyzing glycemic markers, lipid profiles, and blood pressure. Our results demonstrated the broad efficacy of HIIT, with significant improvements in insulin and insulin resistance, triglyceride, total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, systolic blood pressure, and diastolic blood pressure in the mentioned categories. Importantly, subgroup analysis revealed SIT as a highly effective modality of HIIT in modulating the mentioned variables. These findings reinforce the significance of HIIT in regulating cardiometabolic health.

Elevated fasting blood glucose and insulin levels are widely recognized as established factors that increase the risk of developing metabolic disorders (Martin et al., 2015). Following analysis of randomized-controlled trials, our findings indicate a statistically significant decrease in insulin levels (12 studies) and insulin resistance (11 studies) in children and adolescents following HIIT. More importantly, the majority of these studies have employed SIT as an interval intervention, highlighting its strong influence in modulating these metabolic indicators. However, glucose levels only decreased in response to SIT and exhibited an unresponsiveness to HIIT, demonstrating the ineffectiveness of HIIT in targeting this metabolic marker. In a recent meta-analysis investigating the effects of HIIT on cardiometabolic risk factors in childhood obesity, Zhu and colleagues 2021) have reported that HIIT is effective in fasting Insulin and Insulin sensitivity. In another meta-analysis of randomized controlled trials, Cao and colleagues (2021) found no statistically significant difference between HIIT and moderate-intensity continuous training on in fasting Insulin and Insulin sensitivity in overweight and obese children. However, the mentioned studies haven’t specified interval intervention modality (i.e., HIIT vs. SIT). Our results indicated that studies employing SIT interventions observed suitable modulation in metabolic biomarkers mentioned above. A growing body of studies exploring the effects of high-intensity interval interventions on improving metabolic health has argued the potential mechanism affecting insulin sensitivity following such interventions (Gillen and Gibala, 2014; Gibala, 2018; Ryan et al., 2020). In line with early studies (Fell et al., 1982; Bogardus et al., 1983), Ryan and colleagues (2020) recently indicated that the post-exercise decrease in muscle glycogen content could be considered the main contributor to the short-lived improvement in insulin sensitivity which could also be known as the acute effects of the most recent exercise training. This effect may persist for at least a few days after exercise. Long-term metabolic adaptations in skeletal muscles could result from enhanced oxidative capacity and mitochondrial respiratory protein abundance (Ryan et al., 2020; Kelley et al., 1999; Goodpaster et al., 2003; 2013; Holloszy, 2009; 2013; Muoio and Neufer, 2012). Research has indicated that insulin-sensitive populations possess higher mitochondrial density than insulin-resistant individuals, and a training-induced increase in this factor and oxidative capacity may influence insulin sensitivity (Goodpaster, 2013). According to proponents of this theory, high mitochondrial density may be associated with an increased fat-burning ratio at rest (Kelley et al., 1999; Goodpaster et al., 2003).

Bioactive fatty acid metabolites can induce insulin resistance by activating proinflammatory pathways (Schenk and Horowitz, 2007). Enhanced fat oxidation could limit the accumulation of bioactive lipid species and prevent a direct hindrance to insulin signaling (Goodpaster, 2013). However, despite the positive effects of exercise on mitochondrial density, research indicates unaltered resting fat oxidation after the training (Kanaley et al., 2001; Van Aggel-Leijssen et al., 2002), making this hypothesis controversial (Ryan et al., 2020). Overall, the exact mechanisms underlying these effects still need to be fully understood, and further research is warranted to elucidate the relationship between HIIT and insulin sensitivity.

Modified lipid profile is another positive effect of HIIT, which our findings support. Statistically significant improvements in triglycerides, total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol in children and adolescents reinforce previous systematic review and meta-analysis of randomized controlled trials demonstrating HIIT's effects on the maintenance of optimal health (Edwards et al., 2023). However, our findings do not comply with Batacan and colleagues' study (2017), which could result from including recently published studies in our work and excluding non-randomized trials. The most remarkable effects of SIT on lipid profile highlight the effectiveness of this training modality, which hasn’t been specified with the previous investigations. Previous meta-analyses have confirmed a dose-dependent association between elevated triglyceride (Liu et al., 2013), low-density lipoprotein cholesterol, and declined high-density lipoprotein cholesterol levels (Jung et al., 2022) with higher risks of cardiovascular disease and all-cause mortality. By enhancing catecholamines and growth hormone levels (Langin, 2006; Trapp et al., 2007), HIIT stimulates lipolysis of adipose and increases free fatty acid availability to be utilized by active muscles (Dias et al., 2018; Burguera et al., 2000). Enhanced the capacity of the skeletal muscles in utilizing lipids, reduces plasma lipid levels (Earnest et al., 2013) and improves lipid profile (Mann et al., 2014). The mechanism may also include an elevation in lecithin cholesterol acyltransferase (LCAT) (an enzyme transferring ester to high-density lipoprotein cholesterol [Calabresi and Franceschini, 2010]), which is increased in response to HIIT (Rahmati-Ahmadabad et al., 2018; Lira et al., 2019). “The cholesterol removal process is known as reverse cholesterol transport” (Mann et al., 2014). Cholesterol is removed from circulation through this process for disposal, which, in turn, is a result of an increase in LCAT and a decrease in cholesterol ester transferring protein (a responsible enzyme for transferring high-density lipoprotein cholesterol to other lipoproteins) in response to HIIT (Lira et al., 2019). The increased enzyme activity mentioned above following HIIT elevates the ability of active muscles in fatty acids oxidization originating from plasma, triglyceride, and low-density lipoprotein cholesterol (Mann et al., 2014; Shaw et al., 2009) and improves lipid profile.

Significant improvements were also observed in blood pressure after HIIT. Since hypertension remains the primary modifiable risk factor for cardiovascular disease and all-cause mortality (Lim et al., 2012), the significant decreases observed due to HIIT hold essential clinical significance (Edwards et al., 2023). Analyses of 24 randomized-controlled trials indicated that various HIIT forms significantly decrease systolic blood pressure. However, diastolic blood pressure only decreases with SIT, indicating the importance of this training modality on blood pressure. In line with our findings, other meta-analyses have shown the positive effects of interval interventions on lowering blood pressure (Cao et al., 2021; Zhu et al., 2021). However, our study specifically focused on the effects of different interval interventions (HIIT vs. SIT) on blood pressure in this category, and indicated the effects of SIT on diastolic blood pressure. Studies have indicated that the change in blood pressure in healthy individuals with normal blood pressure is short-lived and an acute effect of the most recent exercise training (Whyte et al., 2010; Rossow et al., 2010; Burns et al., 2012). By contrast, improved blood pressure following HIIT could be more long-lasting in hypertensive subjects (Chuensiri et al., 2018). Improved endothelial function by HIIT is one of the main contributing factors in chronically reducing blood pressure (Ciolac et al., 2010). Several HIIT studies have reported reduced arterial stiffness (AS) indicated by measuring brachial-ankle pulse wave velocity (Chuensiri et al., 2018; Ciolac et al., 2010; Guimarães et al., 2010). More intensive interventions may result in a greater effect on AS reduction in people already exhibiting some alterations in vascular elasticity (Ciolac et al., 2010). Mechanisms underpinning the reduction in AS by HIIT are not fully elucidated. However, such a change could be mediated by “the removal of chronic restraint on the arterial smooth muscle cells provided by the sympathetic adrenergic vasoconstrictor tone (Sugawara et al., 2009) as well as endothelin-1” (Maeda et al., 2009). Accumulative evidence indicated vascular reactivity through exercise-induced blood flow and shear stress (Ciolac et al., 2010; Niebauer and Cooke, 1996). Endothelium-dependent vasodilation is mediated by upregulated expression of mRNA for nitric oxide (NO) synthesis in endothelial cells exposed to laminar shear stress through long-term change in blood flow by HIIT (Sessa et al., 1992; Noris et al., 1995). Long-term change in flow could also mediate vascular remodeling by enlarging vessels, changing vascular structure (Niebauer and Cooke, 1996), and decreasing blood pressure.

The clinical implications of our study underscore the transformative potential of high-intensity interval training in enhancing cardiometabolic health among children and adolescents. Through a comprehensive meta-analysis, we observed substantial improvements in insulin levels, insulin resistance, lipid profiles, and blood pressure, with sprint interval training emerging as particularly efficacious. These findings align with and bolster current guidelines advocating for vigorous physical activity to mitigate cardiometabolic risks in youth. Integrating HIIT, especially SIT, into pediatric exercise regimens can significantly improve metabolic markers, offering a promising strategy to reduce the long-term risk of cardiovascular diseases in this vulnerable population.

Our study has several limitations that should be considered when interpreting results. There were significant heterogeneities for several outcomes. The HIIT type, age, and BMI of participants may be the source of heterogeneity. In addition, the present meta-analysis was not limited to non-randomized trials where these types of studies have less clinical value. However, we performed subgroup analysis based on the type of studies, and our result remained significant.

Our findings indicate the wide-ranging effectiveness of high-intensity interval training, resulting in significant enhancements in insulin levels, insulin resistance, triglycerides, total cholesterol, low-density lipoprotein, high-density lipoprotein, systolic blood pressure, and diastolic blood pressure within the specified categories. Sprint interval training emerged as an exceptionally potent form of HIIT for influencing these parameters when we analyzed subgroups. These results underscore the importance of HIIT in managing cardiovascular and metabolic health, highlighting its apparent potential for incorporation into guidelines and clinical practice.