Reply from Glenn M. Stewart, Justin J. Kavanagh, Luke J. Haseler and Surendran Sabapathy

We thank the authors of the Letter to the Editor for their interest in our recent publication in The Journal of Physiology, where we examined cardiac responses – ventricular strain and cardiac troponin – to two acute bouts of prolonged constant‐load exercise of different intensities and duration but matched for total work. The exercise intensities we selected straddled the gas exchange threshold (GET), which we used as a non‐invasive estimate of the onset of blood lactate accumulation, or lactate threshold. The GET is an important exercise intensity descriptor as it demarcates the boundary between the moderate and heavy exercise intensity domains (Whipp, 1996). Our primary conclusions from this study were that the magnitude and time course of exercise‐induced cardiac perturbations are intensity dependent. Nicolò et al. (2016 b) question our methodological approach of matching the exercise bouts for total work completed and, consequently, our conclusions drawn from the results of the study. The authors propose matching exercise bouts based on perceived effort (‘isoeffort’), rather than total work, as a superior approach that will provide clearer outcomes.

Nicolò and colleagues’ greatest contention is directed toward our suggestion that the GET may demarcate an intensity threshold, above which perturbations in cardiac function, both during and in recovery from exercise, are exacerbated. We acknowledge that the purpose of our study was not to ascertain if the GET, per se, represented a boundary influencing the cardiac responses to exercise. Such a study aim would require, as the authors suggested, additional exercise bouts to better characterise the cardiac responses to exercise at and around the GET. However, given that there are distinct differences in a range of physiological responses to exercise performed below and above the GET (i.e. in the moderate and heavy intensity domains, respectively) (Whipp & Wasserman, 1972; Poole & Jones, 2012), and in light of our findings, we do not think it unreasonable to suggest that this intensity domain boundary may also be of relevance to the cardiac responses to acute exercise. Our conclusions, in this respect, also mirror those of Seiler et al. (2007), who examined cardiac autonomic modulation during recovery from exercise bouts performed at intensities below and above the GET.

As Nicolò et al. pointed out, the practice of matching total work when comparing constant‐load exercise bouts comprising different intensities and durations is well established (Pollock, 1973). In proposing an ‘isoeffort’ approach, Nicolò et al. have relied on their own observational study that compared intermittent exercise bouts performed at a given subjective measure of ‘maximal session effort’. Prescribing ‘isoeffort’ exercise may be useful for comparing multiple bouts of intermittent exercise when exercise duration is insufficient to achieve steady state and the physiological response is profoundly influenced by the work‐to‐rest ratio (Karlsson & Saltin, 1971). However, the authors have failed to contextualise their approach within the framework of studying the cardiac responses to exercise. While we do not dispute the utility of ratings of perceived exertion (RPE) when setting or regulating exercise intensity, ‘isoeffort’ constant‐load exercise may not adequately separate the moderate and heavy intensity domains, and total mechanical, metabolic and cardiac work would differ between trials. Importantly, our study design accounted for several mediators of cardiac stress such as exercise intensity, duration and total mechanical work. Even more pertinent to our study, accumulative metabolic demand and rate–pressure product (an index of myocardial O₂ demand, Gobel et al. 1978) were comparable between exercise trials.

We also believe the authors’ assertion that an effort ‘matching approach can be achieved … by means of the highly predictable curvilinear relationship between exercise intensity and duration’ is not appropriate for the non‐fatiguing exercise bouts we utilised. The hyperbolic power–duration model proposed by Monod & Scherrer (1965), and currently embodied in the critical power concept (Morton & Hodgson, 1996), suggests that the tolerable exercise duration is predictable across a range of exercise intensities above critical power. Thus, the hyperbolic power–duration model does not apply to the exercise intensities we investigated (well below critical power), given that tolerable exercise duration is theoretically infinite for intensities below critical power (Jones et al. 2010).

Finally, the authors propose respiratory frequency (f _R) as a physiological correlate of effort (Nicolò et al. 2016 a), and invoke the concept of ‘central command’ in explaining the link between these variables. Whilst a bivariate relationship between f _R and effort may certainly be present during exercise, hyperpnoea is mediated by several neural, mechanical and humoral mechanisms, of which none are solely responsible for controlling respiration during exercise (Forster et al. 2012). Indeed, studies by colleagues in our laboratory have recently shown that the experimental modulation of mood alters the perceived intensity of breathlessness (an important contributor to global perceived effort) during exercise, whilst f _R remains unchanged (Sharma et al. 2016). Thus, the notion proposed by the authors that f _R is a surrogate measure of ‘central command’ is, at present, unsubstantiated.

In summary, our original paper (Stewart et al. 2016) described the influence of manipulating exercise intensity and duration, while controlling total mechanical work, on functional and biochemical perturbations in the human heart. We accounted for several mediators of cardiac stress, including exercise intensity, duration and total mechanical/metabolic/cardiac work; we do not believe that measures of perceived effort would provide additional mechanistic insight. Accordingly, our original conclusion which stated that the magnitude and time course of exercise‐induced functional and biochemical cardiac perturbations are influenced by exercise intensity is justified.

Additional information

Competing interests

None declared.