Incorporating ventilatory support parameters into the PaO₂/FiO₂ ratio in ARDS patients

To the Editor,

We read with great interest the study by Palanidurai et al. in which the authors compare the predictive validity for hospital mortality of a new oxygenation index (P/FP) (P/FP = 10 × PaO₂/(FiO₂ × PEEP) versus the classical P/F ratio and evaluate changes in severity classification of ARDS from the use of the P/FP rather than the P/F ratio [1]. We totally concur with the notion that the P/F ratio can be significantly improved for superior reflection of ARDS severity and prediction of intensive care unit survival/mortality by incorporating mechanical ventilatory support variables into the P/F ratio. The authors argued that “in general for the same P/F ratio, a patient on a higher PEEP has more severe ARDS than a patient on a lower PEEP” which makes perfect sense. For that reason, they elected to incorporate PEEP to the P/F ratio. However, we can further argue that for the same P/F ratio, a patient on a higher mean airway pressure ($\overline{{{\text{Paw}}}}$) has more severe ARDS than a patient on a lower $\overline{{{\text{Paw}}}}$ even when PEEP levels are the same or even irrespective of the PEEP levels. $\overline{{{\text{Paw}}}}$ not only reflects the applied PEEP during mechanical ventilation but also takes into consideration other important mechanical ventilatory support variables such as inspiration:expiration ratio (I:E) and peak inspiratory and alveolar pressures [2].

Previously, we have described a new oxygenation index, termed oxygenation factor (OF) that incorporates $\overline{{{\text{Paw}}}}$ and is expressed as:

$${\text{OF}} = \frac{{{\text{P/F}}}}{{\overline{{{\text{Paw}}}} }} = \frac{{{\text{P}}_{{\text{a}}} {\text{O}}_{2} }}{{{\text{F}}_{{\text{i}}} {\text{O}}_{2} \times \overline{{{\text{Paw}}}} }}.$$

As such, the oxygenation factor (OF) will normalize the P/F ratio to the mean airway pressure. Our study showed that the OF is more reliable than the P/F ratio in reflecting intrapulmonary shunt in patients undergoing coronary artery bypass grafting and with no underlying lung diseases [3]. Recently, we published a study in which we compared our oxygenation (OF) to the P/F ratio in 50 ARDS patients with P/F ≤ 100 mm Hg, 50 ARDS patients with 100 mm Hg < P/F ≤ 200 mm Hg, and 50 ARDS patients with 200 < OR ≤ 300 mm Hg. Our results showed that the OF is superior to the P/F ratio in reflecting oxygenation in ARDS and results in a different patients’ classification for ARDS severity [3].

The study by Palanidurai et al. incorporated only PEEP in the P/F ratio and overlooked other important mechanical ventilatory support parameters such as I:E ratio, and peak inspiratory and alveolar pressures and subsequently the driving pressure. The advantage of incorporating the mean airway pressure ($\overline{{{\text{Paw}}}}$) rather than PEEP into the P/F ratio such as in our OF index is that $\overline{{{\text{Paw}}}}$ can be considered a more appropriate modifier for the P/F ratio since it better reflects the mean alveolar pressure and as such reflects not only the applied PEEP but rather most of the mechanical ventilation parameters and settings such as I:E ratio, tidal volume, and/or peak alveolar and driving pressures [4]. Another concern with the Palanidurai et al. P/FP ratio is the use of a correction factor of 10. Although the authors provided several justifications for this approach, it remains unclear why the authors did not keep their P/FP index as is [i.e., P/FP = PaO₂/(FiO₂ × PEEP)] and proceeded to assess and evaluate their index for superior reflection of ARDS severity and prediction of survival/mortality without the use of the correction factor. Furthermore, Palanidurai et al. used the Berlin definition’s thresholds of ≤ 100, 101–200, and 201–300 to differentiate severe, moderate and mild ARDS, respectively, for both the P/F (mmHg) and the P/FP (mmHg/cmH₂O) ratios. This makes sense only when PEEP is 10 cmH₂O because in such case the P/FP ratio will always be equal to P/F ratio due to the presence of the correction factor 10. What about ARDS patients with different levels of PEEP (i.e., PEEP is not 10 cmH₂O) and will the same thresholds applied in the Berlin definition remain applicable for the new P/FP ratio.

Nevertheless, Palanidurai et al. deserve to be commended for conducting such a valuable study and providing data on a highly needed subject [1]. Both their results and our recently reported results clearly support the concept that multifactorial oxygenation indexes such as the P/FP and OF ratios that incorporate mechanical ventilatory support parameters into the classical P/F ratio are superior indexes to use in mechanically ventilated ARDS patients, particularly for assessing the oxygenation status and the classification of ARDS severity and that these indexes have a greater predictive validity for hospital mortality in ARDS than the P/F ratio [1, 4]. In ARDS patients, oxygenation indexes should be multifactorial and should reflect not only the true illness severity, but also the applied mechanical ventilation strategies.