To the Editor:
The disappointing results of the randomised trials on the use of noninvasive ventilation (NIV) in stable hypercapnic chronic obstructive pulmonary disease (COPD) patients [1] necessitates a search for the reasons for NIV inefficiency in long term treatment. Windisch et al. [2] suggested that NIV aimed at increasing maximum-tolerated inspiratory positive airway pressure and taking over patients breathing rhythm by a ventilator is a key factor in the success of NIV. The new idea proposed by Windisch et al. [3], called high intensity noninvasive positive pressure ventilation (Hi-NPPV), seems to be attractive, but requires verification. Therefore, we would like to congratulate Lukácsovits et al. [4] on their excellent physiological study of the short-term effects of Hi-NPPV published in the April 2012 issue of the European Respiratory Journal. However, we believe that some aspects of this study need to be critically discussed.
Firstly, Lukácsovits et al. [4] proved that Hi-NPPV decreases both arterial carbon dioxide tension (Pa,CO2) values and the work of breathing, even if it is obtained at the cost of a significant reduction in cardiac output. The rate of decrease in cardiac output under Hi-NPPV was 28% versus spontaneous breathing and 13% versus low-intensity noninvasive positive pressure ventilation (Li-NPPV). A similar reduction was noted in oxygen delivery capacity under Hi-NPPV and it consisted of 250 mL of oxygen per min (versus spontaneous breathing). These findings raise a very interesting issue concerning adequate tissue oxygen delivery during Hi-NPPV. The authors claimed that this finding did not necessarily have an adverse effect. This statement is not supported by scientific results but rather by the intuition of researchers, and we cannot share their optimism about it. Above all, we doubt if it is worth decreasing Pa,CO2 by 10 mmHg at the expense of 250 ml of oxygen delivered to tissues every minute. If oxygen delivery becomes insufficient to meet peripheral oxygen consumption, the tissues have to extract more oxygen from the haemoglobin. However, a substantial reduction in oxygen delivery cannot be compensated for by increased oxygen extraction and results in anaerobic metabolism with lactic acidosis. Unfortunately, the lack of data dealing with changes in saturation of mixed venous blood under Hi-NPPV does not give us clear information about tissue oxygen extraction.
Secondly, the detrimental effects of significant falls in cardiac output are likely to have a clinical impact (e.g. insufficient perfusion of vital organs) in stable patients on home NPPV without known heart disease under those conditions which are associated with increased ventilatory demands. This is particularly true given the fact that high pressure level-induced cardiac output fall is more evident in COPD patients as compared to those with acute lung injury [5]. Moreover, the negative haemodynamic effects of Hi-NPPV are based on the multiple interferences between the type of mask, the level of inspiratory pressure and the external positive end expiratory pressure (PEEP) applied [5].
Thirdly, we are not convinced by the hypothesis given by the authors that considers that the cardiac output reduction reflects, at least in part, the lower metabolic demand of respiratory muscles which undergo a full rest under Hi-NPPV. Unfortunately, this speculation is not supported by the measurement of biochemical indexes of muscle energy consumption (lactate) or oxygen uptake (V′O2max).
Fourthly, Lukácsovits et al. [4] used transthoracic echocardiography to exclude patients with existing heart disease on the basis of an ejection fraction less than 40%. However, a normal ejection fraction does not exclude the presence of diastolic left ventricular dysfunction and pulmonary arterial hypertension with/without right heart failure, both cardiovascular conditions that are able to influence the response to NPPV in the acute setting [5].
Fifthly, in our opinion the Hi-NPPV-induced worsening pulmonary hyperinflation should be emphasised. The authors could not calculate the amount of dynamic intrinsic PEEP from the tracing of oesophageal pressure due to the full abolishment of inspiratory muscle activity. However, due to the physiological alterations occurring in COPD, causing an increased constant of time, the greater the delivered tidal volume during mechanical ventilation the higher the chance of an incomplete expiratory emptying of the lungs. A worse dynamic hyperinflation during Hi-NPPV may increase the risk of asynchrony between patient and ventilator (i.e. wasted efforts), barotrauma and cardiovascular interferences.
In conclusion, we would like to underline that Lukácsovits et al. [4] showed that Hi-NPPV is very efficient in improving gas exchange in the lung; however, it has a negative impact on oxygen delivery and therefore has to be used with caution in COPD patients with severe hypoxaemia and conditions vulnerable to ischaemia. If “lungs” are likely to benefit from the implementation of the Hi-NPPV philosophy, the “heart” seems to suffer these “strong pressures”, which could have negative impact on oxygen delivery.
Footnotes
Statement of Interest
None declared.
- ©ERS 2013