LAUSR

To the Editor, We would like to thank Cornelis et al. for their interest in our work, as well as for giving us the opportunity to put some more emphasis on aspects of our recently published article. The purpose of our study was to explore the effects of exercise training on the phenomenon of exercise oscillatory ventilation (EOV) based on describing and quantifying EOV pattern characteristics rather than approaching this phenomenon in a qualitative way. In absence of a universal EOV definition, we based our approach on the definition formulated by Corra et al. and currently adopted by the European Society of Cardiology and American Heart Association. To quantify the EOV phenomenon, we considered only oscillated cycles, finally determining and evaluating duration percentage, average amplitude and average length of oscillatory cycles during exercise as the EOV pattern characteristics. Percentage of EOV duration significantly reduced in EOV patients from 79.0 13.0% to 50.0 25.0% after the exercise programme, with only six out of the 20 patients remaining in the EOVdefinition (considering the cut-off duration value of 60%). As described, we employed a cohort of chronic heart failure (CHF) patients previously randomized in a study to investigate the effects of different exercise training regimes. Inclusion and exclusion criteria and the randomization process have been described in detail. Thirty-eight patients were finally included, a number somewhat lower than previously reported, due to technical difficulties to retrieve full raw data of cardiopulmonary exercise testing (CPET) before or after the rehabilitation programme. As thoroughly described, 20 patients out of the 38 exhibited EOV at baseline, while 18 patients (10 in the aerobic group and eight in the combined group, as obviously concluded by presented data) were classified as non-EOV patients. Furthermore, the exercise intervention was based on a high intensity interval regime. This has been shown to be a feasible, safe and effective modality in CHF patients. The effects of interval versus continuous aerobic regimes in terms of EOV characteristics remain to be decided, as we did not aim at comparing these training modalities. In any case, we undoubtedly applied a high intensity regime; 50% peak workload was defined based not on the CPET, as rather misunderstood, but on the so called ‘steep ramp test’, corresponding to intensities greater than 100% peak oxygen uptake. The main conclusion was that an exercise training programme improved periodic breathing during exercise by reducing the duration of EOV pattern. We provided an explanation based on efficacy of exercise training in general to increase exercise capacity (by improving several pathophysiological mechanisms) and possibly affect EOV appearance. The fact that both EOV and non-EOV patients similarly improved exercise capacity simply amplifies the role of exercise to improve EOV, beyond other benefits. We also commented on the role of combined training to enhance benefits of exercise rehabilitation and potentially further affect EOV characteristics. Finally, we have highlighted the necessity for development of appropriate software to study EOV. Recent work of Cornelis et al. contributes to this direction. Overall, although the prognostic importance of our findings need to be explored in larger studies, we are of the opinion that this work contributes not only to strengthening the importance of exercise training in CHF patients but also to the effort of quantifying