Response to: The (Mechanical) Power of (Automated) Ventilation

Dear Editor,

We would like to thank Dr Bruitman-Kruizinga and Dr Schultz for their interest in our study ¹ and their important comments. The authors postulate that measuring mechanical power from continuous recordings in our study yielded a higher mechanical power when compared to previous studies, and they propose using the more widely applied simplified formula: ²

$$\text{Mechanical}\mathrm{power}\left(\text{J}/\text{min}\right)=0.098\times {\text{V}}_{\mathrm{T}}\times \text{f}\times \left(\text{Ppeak}–½\Delta \text{P}\right)$$

We would like to emphasize that this mathematical approximation has only been validated during volume controlled mechanical ventilation, ^{3 , 4} and some authors have suggested alternative formulas. ^{3 -6} Most importantly, the above formula has not been validated in patients receiving adaptive support ventilation (ASV) or other pressure-regulated modes of ventilation such as adaptive pressure ventilation (APV) (which were the 2 ventilator modes utilized in the initial study). In response to the comments by Dr Bruitman-Kruizinga and Dr Schultz, we calculated mechanical power using this formula, which was 21.8 [interquartile range [IQR] 18.9–31.8] J/min in ASV and 22.3 [IQR 19.2–29.8] J/min in APV in comparison to 28.2 [22.2–36.4] J/min and 26.9 [23.8–37.9] J/min for ASV and APV, respectively, calculated in our manuscript. We would like to take this opportunity to discuss the use of simplified formulae when estimating mechanical power.

Any simplified formula makes assumptions about the shape of the pressure-volume curve during inspiration. During volume controlled ventilation, the pressure and volume change will occur in a roughly linear manner (when in a square flow pattern), approximating a triangular shape for the driving pressure component of mechanical power (which is favorable for the proposed formula). By contrast, with a pressure controlled breath, the driving pressure component of mechanical power on the pressure-volume curve approximates a rectangular shape. ⁶ Therefore, a separate formula has been proposed for pressure controlled ventilation, which might be more appropriate when using ASV and APV. ^{4 - 6}

$$\text{Power}\left(\mathrm{J}/\min \right)=0.098\times {\mathrm{V}}_{\mathrm{T}}\times \mathrm{f}\times \left(\Delta {\text{P}}_{\text{insp}}+\text{PEEP}\right).$$

(Where P_insp is the change in P_aw during inspiration, the pressure control level while in pressure control or the inspiratory applied pressure in ASV or APV.)

Utilizing this surrogate formula for pressure control ventilation instead of the initial proposed formula by Dr Bruitman-Kruizinga and Dr Schultz resulted in a higher estimated mechanical power for the patients while on ASV (28.5 [IQR 21.5–37.7] J/min) and on APV (28.2 [IQR 24.7–33.4] J/min). Notably, this mechanical power estimate provided values very similar to those measured in our study directly via waveforms. Due to the limitations and approximations implied in these formulas, calculation of mechanical power from continuous recordings has been the accepted standard to which abbreviated formulas have been compared. ⁷ Whereas these formulas can be very helpful in epidemiological research, we believe in utilizing the highest available data granularity when continuous recordings are available.

Other factors may have contributed to a higher mechanical power in our study. One important and much-discussed factor included into mechanical power calculations is PEEP. In our study, PEEP was titrated using esophageal manometry as per routine clinical care at our institution. ⁸ This often results in higher PEEP levels ⁹ compared to standard PEEP tables. ¹⁰ Indeed, approximately 50% of the measured mechanical power in our cohort was due to the PEEP component. There is extensive debate regarding inclusion of the PEEP component into mechanical power, with many advocating for removal of this from the measurement. ^{11 - 14} Suffice to say, with widely different PEEP strategies between studies, it is challenging to compare mechanical power between different studies.

Another limitation of the commonly used calculation of mechanical power is the inclusion of the chest wall component. Due to the availability of esophageal pressure recordings, we calculated “lung-directed power” (total mechanical power delivered to the lung and excluding the chest wall). These values in ASV were 12.3 [9.8–17.6] J/min and 12.5 [9.6–17.0] J/min in APV. Currently there is not an agreed upon method for calculation of lung directed power, ¹² nor a simplified formula to estimate this value. As such, we did not include it into the initial manuscript.

We agree with Dr Bruitman-Kruizinga and Dr Schultz about modulation of mechanical power by permissive hypercapnia, which can be an effective way to decrease the energy delivered to the respiratory system. In our study, the protocol dictated that minute ventilation during day 1 crossover measurements would remain unchanged from the baseline values established by the clinical team, so we cannot analyze the impact of clinical variation in breathing frequency as well as permissive hypercapnia on mechanical power in this study. Using permissive hypercapnia to lower mechanical power, however, will be an important tool for future investigations.

In conclusion, the letter by Dr Bruitman-Kruizinga and Dr Schultz, as well as the discussion above, highlights the urgent need for standardization of the methods of calculating mechanical power before this concept can become clinically useful at the bedside.