Skip to main content
  • Paper Report
  • Published:

Differing the gas density for NIPSV


The crossover design of this study was used well to answer the question set by the authors. One of the authors of the study, who was responsible for setting up the NIPSV, had an admirable success rate for establishing the patients on this mode of ventilation. Such success may suggest that this technique can be clinically useful but may require a learning curve for the physicians involved in establishing the individual patient settings for NIPSV.


The rate of intubation in patients with pre-existing chronic obstructive pulmonary disease (COPD) and having acute respiratory failure has been reduced by the use of noninvasive pressure support ventilation (NIPSV) via a facemask. This reduction in intubation rate has been associated with a decrease in mortality and morbidity. However, many patients require intubation because they develop respiratory muscle fatigue. A helium : oxygen mix may be useful as it reduces density of inspired gas and so the work of breathing. It may also be beneficial because a helium : oxygen mix would increase the likelihood of laminar flow and so reduce the driving pressure needed for a given flow.


To assess whether, in decompensated COPD, NIPSV using 70:30 helium : oxygen instead of 70:30 air : oxygen reduces dyspnea and improves ventilatory variables, gas exchange and hemodynamic tolerance.


The study was a prospective, randomized, crossover trial involving 19 medical intensive care patients who had severe COPD. These patients were admitted for NIPSV and stabilized with air : oxygen NIPSV for less than 24 h after intensive care unit (ICU) admission. This was to accustom the patients to NIPSV, so that when the trial period began there was no learning curve. The study protocol follows: 1, baseline measurements 2 h after last use of NIPSV; 2, 45 min of NIPSV with air : oxygen or helium : oxygen; 3, no ventilation for 45 min; 4, 45 min of NIPSV with air : oxygen or helium : oxygen. The same ventilators were used for all patients, and the settings determined before the study period by the physician in charge of the patients. The same physician performed all the ventilatory trials. Any patient requiring more than 30% inspired oxygen was excluded from the study.


Of the original 20 patients, one patient was excluded due to lack of technical information. No patient required intubation because of failure of stabilization. Most patients had severe obstructive airways disease, with a forced expiratory volume of less than 1 litre. NIPSV decreased respiratory rate and increased tidal and minute volumes. There was no difference in these variables when the helium : oxygen mix was compared with the air : oxygen mix. Peak inspiratory flow was higher with helium : oxygen. PaO2 was increased with NIPSV, while gas mixture made no further difference. PaCO2 decreased during NIPSV and was lower with helium : oxygen than with air : oxygen. Both gas mixtures decreased dyspnea, as measured by the Borg scale, but the decrease was greater with helium : oxygen. All but two patients showed improvement in dyspnea score with helium : oxygen mix. Heart rate was also reduced with NIPSV by both gas mixtures. Mean arterial pressure was reduced with air : oxygen but not helium : oxygen mix. NIPSV appeared to be well tolerated, as no ventilatory protocol had to be discontinued.


In decompensated COPD patients, NIPSV with helium : oxygen improved dyspnea and PaCO2 more than air : oxygen mix. The authors felt that the use of helium : oxygen with NIPSV may be beneficial to COPD patients and may reduce the need for intubation.


  1. Jolliet P, Tassaux D, Thouret JM, Chevrolet JC: Beneficial effects of helium : oxygen versus air : oxygen noninvasive pressure support in patients with decompensated chronic obstructive pulmonary disease. Crit Care Med. 1999, 27: 2422-2429.

    Article  PubMed  CAS  Google Scholar 

Download references


Rights and permissions

Reprints and permissions

About this article

Cite this article

Robertshaw, H. Differing the gas density for NIPSV . Crit Care 2, 4031 (2000).

Download citation

  • Published:

  • DOI: