This chapter has no correlating entry in the CICM First Part Exam syllabus document, and this topic has never appeared in the past papers. It was somewhat unclear where to put it, or indeed whether it belongs here at all. Certainly it is a matter advanced enough for senior ICU trainees to consider, but on the other hand it is sufficiently esoteric and academic that people can go through their entire career without thinking about it (which makes it similar to a lot of the Primary material).
Definition and normal values
- Power is the transmission or conversion of energy
- Defined as work over time
- In mechanical ventilation, the power of each breath is the area under the pressure/volume loop
- Measured in joules per minute
- Normal = 2-3 J/min
- Normal in mechanical ventilation: 15-20 J/min
Rationale for the measurement of the power of mechanical ventilation
- Energy transmitted to the lung, spent on deforming lung tissue, can produce injury
- Measuring this variable and titrating ventilation to achieve the lowest possible power may have a positive effect on clinical outcomes
- All other lung-protective strategies (low tidal volumes, low driving pressures) are ultimately all aspects of the same equation, i.e. all these strategies converge in their aim to reduce the power of mechanical ventilation
- Adds an additional element of complexity
- Harder to measure than driving pressure, plateau pressure or weight-adjusted tidal volume
- Difficult to calculate in patients ventilated with volume-controlled modes
- The relationship with lung injury remains an association, and no RCTs have been done to support this as a target
- There is disagreement as to what value we should regard as a safe threshold (12 vs 17 vs 22)
- Impossible to guess whether the mechnical power of spontaneous breathing has the same effect as mechanical ventilation, and whether to calculate this variable for patients on NIV
- Animal and human data suggests that ventilation with a power lower than around 12 J/min should be safe from VILI
- Analysis of data from many trial sets (Amato, 2015) suggests that power of ventilation provides similar predictive information to driving pressure
Power can be broadly defined as the rate of transfer of energy, or the rate of transformation of energy. The official physics definition of power is
Power is the time derivative of work, dW / dt
Work is the product of force and distance,
Work is the product of pressure × volume
for each specific breath cycle. Which means
Power = respiratory rate × work per breath,
Work per breath = area under the pressure- volume loop =∫0 VTPaw dV
But because we do not usually record high-fidelity measurements of pressure-volume loops or integrate their area,
Power = respiratory rate × (VT × (PEEP + ΔPinsp))
(at least for pressure-control ventilation, which is the easiest to calculate because the pressure is constant).
Power is usually measured in watts, but because everything in respiratory physiology has to be different, by convention the power of breathing and mechanical ventilation is measured in joules per minute. The end product of the equation above (where volume is in litres and pressure is in cm H2O) would have to be multiplied by 0.098 in order to be expressed in joules per minute. For ventilator settings resembling a reasonably normal ventilated patient (resp rate 14, PEEP 8, ΔPinsp 14, VT around 550ml ),
Power = (14 × (0.55 × (8 + 14)) × 0.098 = 16.6 J/min
Respiratory rate (breaths/min) Tidal volume (L) PEEP (cm H2O) ΔPinsp (cm H2O) Power of ventilation
That's all fine and good, but with the academic definitions calculations and normal values now behind us, the question we must ask is, why do we care?
Even without calculus or equations, as soon as one accepts the concept that power is the transmission or transformation of energy, one is able to appreciate that transmitting large amounts of energy to your lungs may not be a perfectly benign process. Consider the components of the work of breathing. A lot of this work is "elastic work", particularly where lung compliance is low, i.e. work done to deform the parenchyma (chest wall and airway resistance being the other energy sinks). The mechanical (kinetic) energy generated by the ventilator is therefore transmitted to alveolar tissue, and the more injured and inelastic this tissue becoms, the more energy is transmitted to it, as more work is required per unit of minute volume. Each breath becomes a punch in the lung.
A more professional way to put this would be to say that there is a relationship between the power of ventilation and lung injury. Certainly that was the take by Gattinoni et al (2016), who proposed that all of our lung-protective strategies (tidal volumes, driving pressures) converge on this one concept:
"it is worth considering that all the ventilator-related causes of VILI, although investigated separately, are components of a unique physical variable (i.e. the mechanical power)"
It is therefore not surprising that, when Tonna et al (2020) analysed the same data as Amato et al (2015), they found that power and driving pressure produced a roughly similar change in the hazard ratios for mortality, suggesting that both variables were measuring the same basic thing.
From the above, it follows that if there is at all times energy being transmitted to lung tissue, and that lung tissue is not being smashed to pieces in the course of normal breathing, then there must surely be some sort of safe threshold which it can tolerate, below which injury does not happen. That sounds like something that would be good to know. Unfortunately, we do not know what value we should be aiming for, at least not in the setting of ARDS (where we assume this would have the greatest relevance). Animal data would suggest that 3J/min is probably safe, and 7 or 12 J/min is probably not, but these were a) pigs, b) prone and c) ventilated with PEEP of 4. Paudel et al (2021) went through an entire list of variously ventilated porcine material and concluded that something around 12 J/min is probably enough to cause VILI. Serpa Neto et al (2018) would have called it at around 17J/min, and Zhang et al (2019) would have us index it to body size (as this seemed to be more predictive of lung injury than raw power values).