Reply from M. J. Gibala, J. P. Little, M. J. MadDonald and J. A. Hawley

Abstract

Thank you for the opportunity to respond to the letter submitted by Gayda and colleagues in response to our recent review published in The Journal of Physiology (Gibala et al. 2012). With regards to their first comment regarding our new ‘practical’ high-intensity interval exercise (HIIE) protocol, we disagree with the assertion that ‘exercise intensity at 60% of peak power cannot be considered high intensity.’ In our efforts to develop a low-volume HIIE protocol that can be applied across different cohorts including clinical populations, we devised a model comprising 10 × 60 s work bouts at an intensity eliciting ∼85–90% of maximal heart rate (HRmax; averaged over the 10 intervals), interspersed by 60 s of recovery. We have found that the percentage of peak power output (PPO; determined using a standard ramp test to volitional fatigue which does not always elicit peak O2 uptake) that approximates the desired target heart rate (i.e. the % of HRmax) varies considerably between subjects and is exercise-mode specific. For example, in the study by Hood et al. (2011) which was conducted on sedentary healthy adults, a workload equivalent to 60% of PPO during upright cycling was sufficient to elicit a training intensity of ∼90% HRmax. However, in our recent study conducted on patients with type 2 diabetes, the intensity required to elicit ∼90% HRmax was ∼95% of PPO determined during recumbent cycling (Little et al. 2011). We agree with the assertion by Gayda and colleagues that ‘acute physiological responses during different HIIE protocols as well as patient's safety, tolerance and comfort should be tested before their implementation into training programs’. Ongoing protocol optimization work in our laboratory reveal that when interval exercise was prescribed as 80% of PPO in coronary artery disease (CAD) patients – most of whom were taking beta-blocker medication – the 10 × 60 s protocol resulted in peak heart rates during the exercise that averaged ∼85% of age-predicted HRmax. Further, the 10 × 60 s protocol was best tolerated and rated as most preferred by CAD patients in comparison with a modified Wingate protocol (repeated 30 s efforts at 100% PPO with 4 min unloaded cycling for recovery), the standard aerobic interval training protocol used by Wisloff and colleagues (2007), or a moderate-intensity continuous exercise (MICE) protocol. It is likely that high-intensity interval training (HIT) does not conform to a ‘one size fits all’ approach and the interval training stimulus needs to be tailored to individuals depending on their initial level of fitness, exercise tolerance, use of prescription medications and other factors. We also concur with the other main comment by Gayda and colleagues that ‘the superiority of this HIIE protocol [our 10 × 60 s ‘hard’/60 s ‘easy’ model]… needs to be demonstrated.’ Indeed, our review concluded ‘One aspect that is unclear from the present literature is the precise intensity and minimal volume of training that is needed to potentiate the effect of the stimulus-adaptation on outcomes such as mitochondrial biogenesis and relevant health markers. To answer such questions, a complex series of studies needs to be undertaken that systematically ‘titrate’ levels of the ‘training impulse’ and determine subsequent cellular, performance and clinical responses after divergent training interventions.’ Specifically with respect to the use of HIIE in patients with cardiovascular risk or cardiovascular disease, the letter by Gayda and colleagues highlights four references from their laboratory that were not cited in our review. Given the relatively broad scope of our review and the fact that Journal guidelines restricted the number of references to 50, it was obviously not possible to cite all relevant work. Moreover, two of the citations listed by Gayda et al. were acute exercise studies (whereas the focus of our review was training adaptations) and the other two citations were a journal abstract and a recent paper published in February 2012 (neither of which we had access to at the time of submission of our original manuscript). We are also aware of the pioneering research conducted by Meyer and colleagues (e.g. Meyer et al. 1998) and have acknowledged this work in a previous commentary (MacDonald & Currie, 2009). We apologize to all authors whose work on interval training we could not cite due to the broad focus of our review and referencing limitations imposed by The Journal.