F_IO2, P_aO2, or Else – What Matters in Noninvasive Ventilation in Stable COPD?

To the Editor:

We read with great interest the article by Cardinale et al¹ on delivered by noninvasive ventilation (NIV) compared with long-term oxygen therapy (LTOT) at the same flow. We would like to add some comments regarding this study.

The article offers evidence that in NIV with home-care ventilators, the addition of oxygen does not deliver the same as the same flow of oxygen applied via nasal cannula (0.25 vs 0.30). This is not surprising, given the oxygen dilution in higher NIV flows. However, does not directly translate into when different modalities of oxygen delivery are used. Therefore, the comparison of nasal cannula and NIV is inappropriate, and some key points have to be addressed.

First, the authors have not provided data on achieved under different oxygen delivery modalities. Oxygenation data would provide crucial information on the clinical relevance of the difference in observed in the study. It has been established that NIV alone influences gas exchange. Ambrosino et al² compared spontaneous breathing with pressure-support ventilation without oxygen supplementation and reported a significant improvement in pH, , and under NIV. Therefore, the combined effect of NIV and oxygen likely compensates for lower delivery by NIV. The effect on arterial gases cannot be assumed from but should be determined directly through arterial gas measurement. In our clinical practice, oxygen supplementation is titrated during NIV to reach target oxygen saturation and regardless of and is not simply translated from LTOT flows. In Table 1 we provide data on 3 patients with COPD on NIV at follow-up in recent months to show the comparison of blood gases on spontaneous and pressure-support ventilation, measured within hours. All 3 subjects signed an informed consent, the study was approved by the ethics council. The analysis demonstrates higher on NIV compared to LTOT on the same or lower oxygen flow. Cardinale et al¹ reported that a 2.5-fold increase of LTOT oxygen flow was required during NIV to achieve the same . This is inconsistent with our clinical experience with NIV oxygen flows (Table 1) and is a further sign that during NIV is not the sole determinant of . Additionally, published data,³ our clinical experience (Table 1), and the present study itself show that patients with COPD require relatively low flows of oxygen during NIV. Therefore we believe the real clinical target in oxygenation is and not .

Second, it is encouraging that the study demonstrated on average similar during the daytime and during the nighttime NIV trials (0.24 vs 0.25), although the same mask type was utilized for all subjects. Potential poor fit of the mask could cause higher unintentional leakage (and further decrease) overnight, but the study found low levels of leakage. Nevertheless, results reported by Storre et al⁴ indicate that major leaks influence both and , but, more importantly, and were measured and compared within different NIV circuit setups. In our opinion, the issue of leaks influencing should be addressed with optimal mask selection and with oxygen titration for the specific NIV circuit setup used for a patient.

Finally, the authors speculate that lower during NIV might explain the lack of benefits of NIV across some earlier studies performed in patients with COPD. Even if we assume that significant arterial hypoxemia does occur during NIV (which the present study did not demonstrate), we believe that the lack of benefit in these studies is better explained by lower inspiratory pressures used in earlier studies and failure to significantly influence the level of .⁵ Furthermore, nocturnal oxygen desaturation itself is frequent in patients with COPD (ie, present in up to 48%).⁶ Several studies have reported that the addition of nocturnal oxygen and correction of nocturnal oxygen desaturation do not influence the survival of subjects with COPD with nocturnal oxygen desaturation per se.⁶ However, the 2 studies of NIV in COPD demonstrating the survival benefit of NIV used higher inspiratory pressures and achieved a significant reduction of .⁷ Higher inspiratory pressures lead to higher leaks, which in turn will decrease . Considering all these facts, it is more likely that the survival benefit of NIV in COPD is linked to increased alveolar ventilation when higher pressures are used, and that the potential of lower does not seem to be crucial.

To conclude, the results of the study by Cardinale et al¹ demonstrate lower at the same oxygen flows in NIV compared to spontaneous breathing, but further studies are needed to determine the impact of this finding on oxygenation or relevant clinical outcomes.

• Copyright © 2021 by Daedalus Enterprises