From the authors
We would like to thank L. Bianchi for his letter to the editors and Ambrosino 1 for his editorial regarding our paper on the effects of training with inspiratory pressure support in patients with chronic obstructive pulmonary disease (COPD) 2. Both reactions focus on the important issue of whether noninvasive ventilatory support (NIVS) during training could now be recommended as an aid in the practical setting of pulmonary rehabilitation.
Is the complex game of assisted ventilation or “mechanical doping” during exercise training in COPD worth the candle? We agree with L. Bianchi that the literature on the effects of training with NIVS in COPD yielded some conflicting results. It must be noted, however, that the study lacking an additional effect of training with NIVS included patients with a relatively mild airway obstruction, well-preserved exercise performance and an absence of inspiratory muscle weakness 3. In such patients, the ventilatory system is not the primary factor limiting exercise performance. In addition, relatively high levels of ventilation have to be supported, putting a significant demand on the capacity of the ventilator. In the absence of a clear indication for NIVS during exercise or, alternatively, insufficient ventilatory assistance, it is likely that NIVS may work as a deterrent rather than a support to exercise with poor compliance and high drop-out rate as a result.
Pulmonary rehabilitation is by definition a multi-component intervention based on the individual needs of a patient 4. For methodological reasons, we deliberately chose to limit the experimental intervention to cycle exercise training, comparing the effects of two different intensities of NIVS on excercise performance. It is likely that this mono-component approach contributed to the relatively modest results with respect to walking distance and quality of life, especially in the control group. The outcome of the whole may be greater than the sum of its parts. For this reason, we recommend combining NIVS with other components if it is considered as aid during exercise training.
Undoubtedly, it is possible to achieve adequate exercise training intensities and physiological benefit from general exercise training in many patients with COPD without the use of any assistance at all. However, it is also documented that a considerable number of patients may be unable to reach a satisfactory training stimulus because of intolerable exertional dyspnoea 5. As L. Bianchi suggested, it is likely that we can find the best candidates for ventilatory-assisted exercise training among such patients. Furthermore, in our study, the gain in training intensity in patients training with mechanical doping could be partly explained by the magnitude of the acute effect of ventilatory assistance on exercise tolerance at the start of training. That is, the more exercise performance improved acutely due to the application of NIVS, the more patients were able to tolerate higher intensities during training. Previously we have shown that acute effects of NIVS were inversely related to maximal inspiratory pressure 6. Thus, the weaker the inspiratory muscles, the greater the gain in exercise tolerance. These findings may provide clues to the selection of patients for exercise training with NIVS.
Interestingly, a similar result was observed with supplementary oxygen during training; the greater the acute effect of oxygen on exercise tolerance, the greater the gain in exercise training intensity in patients using oxygen 7. Obviously, supplementary oxygen is a much easier and cheaper adjunct to training compared to ventilatory assistance. However, it should be noted that in three out of the four randomised controlled trials on the effects of training with supplementary oxygen in COPD, no additional effects were found. In fact, in these studies, training on oxygen resulted in smaller, albeit nonsignificant improvements, compared to those in the patients trained on air 8.
Finally, noninvasive ventilatory support and supplemental oxygen relate to the treatment of different causes of respiratory failure. The application of oxygen is mainly involved in the treatment of hypoxaemia, whereas noninvasive ventilatory support aims to reduce hypercapnia. In this context, future research is needed that is directed at identifying patients responding to noninvasive ventilatory support and those responding to supplementary oxygen. This will further reveal fundamentals for tailoring programs to optimise the benefits.
- © ERS Journals Ltd