The aim of this study was to establish the effects of a HIT intervention upon a number of CVD risk factors in adolescent youth. Results revealed that body mass and waist circumference was maintained in the HIT group although significant increases in these measures were evident in the control group. This suggests that the intervention may have had an impact on body mass and waist circumference maintenance by limiting their increase during the intervention period as observed within the control group. Although previous investigations involving youth have noted significant reductions in body composition through a HIT intervention [13, 14], both of these studies were of a longer duration involving overweight and obese participants. Given the short duration of this study and that only 2 individuals had at risk waist circumference levels (data not shown), our findings are unsurprising.
A significant reduction in systolic BP and increase in CRF post-intervention confirms our findings from an earlier investigation . Reductions in systolic BP through aerobic exercise training are well established with approximate reductions of 6 to 10 mm Hg seen in previously sedentary men and women of all ages . This decrease in systematic vascular resistance through aerobic exercise training has also been linked to the reduction in systolic BP  and appears to suggest that central adaptations may be responsible for the adaptations noted in the HIT group. Nonetheless, previous HIT investigations involving healthy adults appear to contradict the suggestion that central adaptations are linked to an increase in CRF [40, 41].
For instance, participants involved in a 6 week HIT intervention undertaken 3 times per week experienced significant peripheral (a-vO2 difference) but not central (maximal cardiac output) adaptations  whereas in the study by Rakobowchuk and colleagues, an intervention undertaken 5 times per week for 6 weeks improved peripheral, but not central, vascular structure . Whether peripheral adaptations can explain the increase in CRF evident in this cohort is unclear. Given the nature of our school based intervention it was not possible to undertake similar measures in this cohort given the number of participants and the time restrictions of the school curriculum. Nonetheless, previous investigations that have involved HIT interventions have consistently demonstrated peripheral adaptation through improved oxygen delivery to working muscles , enzymatic  and mitochondrial adaptations [12, 42]. Furthermore, some have also suggested that reductions in systolic BP from aerobic exercise training are due to an increased stroke volume through improved systemic vascular resistance (peripheral adaptation) rather than an increased cardiac output . This is plausible since resting HR typically decreases after a period of exercise. From these studies it is evident that HIT interventions can stimulate peripheral adaptations which can enhance CRF, and perhaps systolic BP, and may explain the adaptations noted in this cohort.
Although these adaptations may be surprising given the initial level of CRF evident in this cohort (mean baseline 20MSFT levels of the cohort was in the top 90th percentile ), the design of our protocol may have ensured a large aerobic contribution towards energy production. Whilst our understanding of exercise metabolism is limited by ethical and methodological constraints in youth populations, evidence from adults may offer a plausible explanation for the increase in CRF evident in the intervention group. For instance, at the onset of the first repetition maximal power generation will partly be reliant upon muscle PCr stores to contribute towards ATP regeneration. As each exercise repetition is of 30s duration interspersed with 30s recovery, the decrease in PCr availability during subsequent repetitions  means individuals will rely more heavily upon both glycolytic and in particular, oxidative phosphorylation for continued energy production . It was evident in this study that participants were unable to cover the same distance during subsequent bouts of exercise, especially towards the end of the intervention period (data not shown), which could indicate a greater reliance upon oxidative phosphorylation as exercise continues. This reliance may have become more pronounced in the latter repetitions of each session and towards the end of the intervention period given the inclusion of more repetitions and reduced recovery period, and is supported by our observations of the distances covered. It may be that the 1:1 exercise and recovery ratio provides enough stimuli to induce aerobic adaptations within the mitochondria (peripheral adaptation) since it is required to replenish ATP at a high rate with a decreasing anaerobic contribution .
Significant improvements in CMJ and 10 m sprint performance were noted in the HIT group whereas the control group experienced a significant decrease in CMJ performance only. It appears that the intervention which involved maximal acceleration and deceleration phases as well as explosive turns had a positive effect on muscular power, and confirms the findings of others that have shown an increase in muscular power through HIT interventions [19, 47]. A logical explanation is that neurological adaptations (motor-unit recruitment) may explain the increase noted in explosive muscular power but evidence of this in adolescents is scarce. Although changes in agility favoured the HIT group, this did not reach significance while the control group experienced a significant decrease in agility post-intervention. It is well established that an increase in agility performance and maximal sprinting speed are reliant upon adaptations in a number of factors including muscular power, acceleration speed, balance and coordination . While it appears that the HIT intervention was appropriate to induce significant increases in acceleration speed and muscular power (Table 3), it may be that balance and coordination did not improve at a sufficient level to induce a significant increase in agility performance.
A significant decrease in LDL concentration was evident in both the HIT and control groups while total cholesterol also decreased significantly in the control group. The large decrease in LDL evident in both groups post-intervention (Table 4) would have a subsequent effect upon total cholesterol levels, and can explain the significant reduction noted within the control group. Others have found similar positive effects of exercise on LDL in adolescent cohorts [49, 50] while others have not . Although participants were encouraged to maintain normal physical activity and dietary behaviours throughout the intervention period, the decrease in LDL evident in the control group could have been a result of increased physical activity, improved diet or a combination of both despite our recommendations. This however doesn’t appear plausible given that both body mass and waist circumference significantly increased post-intervention in the control group whereas CMJ and 10m sprint performance significantly decreased, as did CRF and 10-m sprint performance (albeit not significantly). It’s more likely that the dietary and physical activity behaviour of those participants in the control group was poor throughout the intervention period.
No significant differences were evident in the seven other metabolic CVD risk factors measured post-intervention but is unsurprising given the apparent healthy nature of this adolescent cohort. For instance, mean baseline 20 MSFT levels for both groups were in the top 90th percentile  while only 2 individuals presented with at risk waist circumference (data not shown). It is plausible that individuals who have poor anthropometrical and CRF profiles would have unfavourable metabolic profiles. Perhaps these individuals may be more susceptible to positive metabolic changes through the HIT intervention.
The findings from this study indicate that it is feasible to implement HIT interventions for adolescents within the school setting and for them to adhere to such a rigorous protocol. Findings from the focus groups and informal discussions revealed that adolescents are able to undertake strenuous intensive exercise and enjoy the nature of the activity. Whilst enjoyment was not specifically measured, we are confident that participant responses are accurate given that the mean adherence rate for the 21 exercise sessions was 16.6 ± 1.5 (~80%).
An additional important result from this study was the role of the PE teachers. Despite their initial trepidation and concerns regarding the intervention, unbeknown to us, the PE teachers had coordinated their lunch break to perform the HIT intervention prior to the start of the study. From informal discussions, the teachers themselves wanted to undertake the intervention to experience the challenges the pupils would soon face. Once the teachers began to provide positive feedback of their own experience of the intervention, we agreed that they would take an active role in emphasizing the importance and benefits of completing the intervention to the participants. As our relationship with the teachers strengthened, we focused our attention towards building relationships with the participants. As we were present during each exercise session, we were able to engage in informal discussions with participants and understand their reasons for participating. At the very least, this gave us a starting point for discussion during the subsequent sessions and weeks and was viewed positively by the participants during the focus group sessions post-intervention. From our experience, the strong relationships built with both teachers and participants allowed us to solve any issue that arose together, and appears vital for maintaining adherence and facilitating research outcomes.
The findings of this study, as well as our previous investigations [16, 19] suggest that HIT may be an effective strategy to enhance measures of physical fitness in adolescent cohorts. As our studies have used apparently healthy adolescent cohorts, it is unsurprising that limited changes have occurred in the metabolic profiles of cohorts post-intervention. Also, the observations of this study are limited as it was not possible for allocation concealment or blinding since this study was conducted in only one school. Whilst attempts to reduce this confounding bias were undertaken by controlling for confounding variables in our analysis, it is plausible that the experimental treatment effect of the HIT intervention may have been exaggerated in this study. So whilst our findings suggest that HIT is a feasible method of physical activity in adolescent cohorts, further work is now required to establish the effects of HIT interventions on measures of health and well-being in a larger cohort of school children, taken from different schools through the use of a clustered randomized controlled trial.