While regular exercise training represents one of the most powerful strategies to reduce the development of metabolic disease in healthy adults , most adults fail to meet current guidelines for participation . These guidelines largely focus on the time spent carrying out moderately intense activity (or total energy expenditure), typically require many hours of exercise each week and can fail to modify risk factors relevant for disease prevention . In addition, as time commitment is perceived as a major barrier, driving or contributing to low compliance, then these guidelines may not be the most logical approach for improving public health. In the present study we demonstrate for the first time that only a few minutes of high intensity interval exercise performed over two weeks is required to substantially improve both insulin action and glucose homeostasis in sedentary young males. This is both a physiologically important observation and potentially useful as it highlights a preventative intervention that could logically be implemented as an early strategy to prevent age related development of cardiovascular disease.
Interestingly, despite employing long-term training interventions (2–16 months) the majority of studies investigating classic aerobic [10, 20–25] or strength training programs [26–28] have observed only a reduction in insulin area under the curve (AUC) in response to a glucose load following training, without a concomitant reduction in glucose AUC, indicating only a partial improvement in insulin action. Furthermore, walking based interventions may not reduce risk factors in the target population where prevention is the key objective . Some longitudinal exercise studies have shown reductions in glucose AUC [29–31], but post-training OGTTs were performed within 24 hours of the last exercise bout and therefore reflect the combined impact of acute and chronic exercise . In contrast, Hughes et al demonstrated reduced glucose AUC in elderly subjects without a concomitant change in insulin AUC .
The low volume, high intensity training utilized in the current study significantly reduced both glucose AUC (-12%) and insulin AUC (-37%), with a sustained improved insulin action until at least day three after the last exercise session. This was achieved without changes in body weight, and with a weekly energy-'cost' of training of ~225 kcal during the first training week and ~275 kcal during the second training week. This very modest increase in calorie consumption is in stark contrast to the ~2000–3000 kcal·week-1 consumed during a typical aerobic training program [25, 34]. This implies, but does not prove, that the mechanism underpinning the benefits we observed with HIT, may be distinct from those responsible for the more modest improvements in insulin action with classic time-consuming aerobic training. While much focus is being given to increasing calorie consumption to ward off weight gain, it is clear that improving metabolic fitness may be just as important as limiting gains body mass index.
Failure for insulin to adequately control blood glucose following a meal is known as 'insulin resistance'. Skeletal muscle is considered the major tissue responsible for the uptake of glucose following a meal, or a glucose or insulin challenge  such that it is entirely reasonable to assume that the improvement in glucose and insulin AUC observed in the present study reflected improved muscle glucose uptake. The limiting step in glucose disposal is considered to be its transport into the skeletal muscle  and GLUT4 is the most abundant glucose transporter in skeletal muscle. Increased GLUT4 concentration with endurance training has been suggested to be an important factor regulating insulin sensitivity [6, 33]. Burgomaster et al. reported that skeletal muscle GLUT4 levels increase by ~20% after one week of HIT, and remarkably remained elevated over 6 weeks of training and a subsequent 6-week period of detraining . Given the similarity between our study and the aerobic performance improvement, protocols and subjects in the Burgomaster et al studies, these studies should be comparable and thus an increase in GLUT4 may partly explain our findings. However, increased GLUT4 concentration does not always fully explain training-induced improvements in insulin sensitivity and key regulatory proteins down-stream in the insulin signalling pathway are more activated in response to insulin following aerobic training . HIT produces similar changes in skeletal muscle markers of carbohydrate and lipid metabolism to aerobic training , so it should be investigated whether HIT also produces similar adaptations of the insulin signalling pathway as seen following traditional aerobic training .
Improved whole body glucose disposal following training has been associated with an increase in insulin stimulated glycogen synthesis . HIT has at least two novel features, firstly unlike walking or moderate intensity aerobic training, it involves the activation of a large muscle mass and secondly this is associated with a very high glycogen breakdown-turnover. The combination of these two factors means that a greater proportion of muscle fibres will need to replenish their carbohydrate stores, compared with what would be encountered following moderate intensity aerobic training. Muscle contraction under conditions of metabolic stress (such as incurred during HIT) results in very rapid glycogen degradation  and this would almost certainly alter the binding of a variety of glycogen associated proteins [40, 41]. Thus we suspect that remodelling of the glycogen pool, altering the molecules' branching architecture , may well be important in regulating skeletal muscle insulin sensitivity following HIT. Currently, high intensity muscle contraction is the only feasible strategy for remodelling of the entire muscle glycogen pool in humans.
Insulin sensitivity may also be regulated by plasma NEFA concentration. Pharmacological lowering of plasma NEFA levels has been shown to positively regulate insulin sensitivity during an OGTT , whereas raising plasma NEFA concentration, through lipid infusion, lowers glucose infusion rate during peripheral insulinemia-euglycemia in young men . In contrast, exercise training has been shown to have little or no effect on fasting plasma NEFA concentration, insulin mediated lipolysis or NEFA release during exercise [41–44]. In the present study, HIT was associated with a 17% decrease in fasting plasma NEFA concentration without a concomitant change in fasting insulin. Furthermore, there was a 26% reduction in NEFA AUC during OGTT following HIT despite a 37% reduction in the plasma insulin AUC. This suggests that insulin was able to inhibit lipolysis to a greater extent following HIT and while these changes are more modest than those observed with lipid powering drugs, the long terms benefits may still be of significance.