So, last year Amato et al published a paper on Driving Pressure:
Driving Pressure and Survival in the Acute Respiratory Distress Syndrome
It contained an incredibly interesting theory–driving pressure may be a much better measure of lung protection than Plateau Pressure. Driving pressure is Vt/Crs (Vt being tidal volume based on ideal body weight and Crs is the compliance of the respiratory system. In patients who are not spontaneously breathing, driving pressure could also be calculated by Plateau Pressure minus PEEP. These graphs explain how it can be both:


Based on the study, a driving pressure < 15 seems associated with lower mortality, even in patients with elevated plateau pressures. Conversely, patients with Plats < 30, but driving pressures >=15 still seem to be at risk. The study also offers a possible explanation for why the ARDSnet higher PEEP trials did not show benefit–higher PEEP only seems to help if it leads to reductions rather than elevations of the driving pressure (due to lung recruitment).
In this episode, I dicuss the logistics of Driving Pressure with Dr. Roy Brower, senior author of the study. Dr. Brower is a professor of Medicine and Medical Director of the MICU at the Johns Hopkins Hospital. He has over 150 publications, numerous national teaching awards and is the most recent recipient of the ATS Life Time Achievement Award (intro cribbed from MCCP).
After listening to the show, if you want to hear more of the underpinnings of driving pressure, listen to Dr. Brower's lecture on the amazing Maryland CC Project:
Additional Reading
Additional New Information
From a fantastic review of driving pressure:
A recent secondary analysis of five randomized trials showed that the mortality benefit of lowering Vt in patients with ARDS was only observed in those patients with high elastance of the respiratory system, suggesting that lung stress, rather than lung inflation per se, was a determinant of ventilator-induced lung injury (VILI) [Am J Respir Crit Care Med 203:1378–1385].
The frequency of tidal inflation may also contribute to VILI. However, recent data have shown that the impact of decreasing ΔP on reducing mortality was four times stronger compared with the effect of decreasing respiratory rate (RR) [10. 1164/ rccm. 202009- 3467OC].
Paper also includes methods for measuring ΔP in spont breathing pts
More on EMCrit
Additional Resources
You Need an EMCrit Membership to see this content. Login here if you already have one.
- EMCrit 394 – CV-EMCrit – Inotrope Basics Part 2 – Specific Scenarios - February 7, 2025
- EMCrit RACC-Lit – January 2025 - February 4, 2025
- EMCrit 393 – CV-EMCrit – Inotrope Basics Part 1 - January 25, 2025
Scott, This is a great topic. Thanks for the post and covering it! One point of feedback: The Amato paper was a great study because it took 9 trials of ARDS patients, derived the effects of driving pressure from 4 trials, validated it in 1, then validated it again in 4 other trials, then applied it across the enter group of trials. The paper, by itself, wasn’t a trial. Twice above your referred to the study as a trial. It would be even more amazing if it were a clinical trial. I thought the word trial could potentially be misleading… Read more »
Chris, great point-I need to be more careful to avoid using study and trial as synonyms. Edited above.
Excellent Podcast and accompanying Lecture . . . looking forward to a second Podcast on this. My quick takeaway is that we are still using Vt that are too high.
Scott, great podcast as always. Two comments: Low Vt ventilation is undoubtedly the standard of care for lung injured patients, but I worry about its wholesale application to all mechanically ventilated patients. Breathing at 6 ml/kg IBW is impressively uncomfortable. Patients ventilated at low Vt generally (again, generally) require deeper sedation, and there’s no question that deeper sedation worsens outcomes in ventilated patients. I don’t think there’s any question what to do when significant lung injury or ARDS is present, but I wonder whether we rob Peter to pay Paul when we have to sedate patients with little or no… Read more »
Yes and Yes! I’ve used and recently recommended in a piece in Annals of EM that 8 ml/kg should be starting point for non-lung injury patients. There was much outcry and many letters to the editor, but I just don’t see venting every pt at 6 ml/kg as supported or good practice. Just as I don’t think 10-12ml/kg is clever. As to transpulm, makes total sense. Would love to see better outcome studies, but more importantly an easy to use device that would substitute for our gastric drainage tubes and one that had esoph. temp monitoring. Imagine an OGT/NGT that… Read more »
We also fail to discuss how much misery vent dyssynchrony causes; more sedation, higher risk of VILI and exhaustion. I’ll take a synchronous and awake patient with slightly suboptimal Vt over a dyssynchronous patient with “perfect” Vt. I’m actually increasingly using good old fashioned PSV or APRV for a fair number of my patients. They’re more comfortable… and they recruit their bases all by their lonesomes. Everything old is new again. Our Avea balloons are decent OGTs and manometers, and they’re almost as easy to insert as your typical Salem sump. No temp probe function, but we get that with… Read more »
Completely agree with the 8cc/kg starting point then decrease as tolerating. I think there’s some ‘thought provoking’ reviews that show a lower Vt may be associated with lower progression to ARDS/potential preention- but I think (& is prob true with anything in medicine/critical care) there’s probably a subgroup that benefits & one that gets harmed. Simply have zero idea which is which at this point. I do the same for dyssynchrony -esp post-acute phase ARDS when almost 100% of patients have air hunger/dyssynchrony– APRV or PSV. I don’t have esophageal balloons – how much do they cost?
almost all of those reviews compare 6 to 10. The range in between is still who knows. In the largest Meta-Analysis I can find, 7 ml/kg was actually lumped into the low tidal volume group.
For readers with an interest in reading more about esophageal pressure transduction, an excellent review was recently published in Intensive Care Medicine
One rhing that I have not seen addressed is flow rate. Do we have any evidence that the rate at which a breath goes in, has an effect? In Gattinoni’s recent lecture for the World Sepsis Congress he seemed prety enamored by driving pressure x respiratory rate as determinants of VILI? Thoughts?
Mike
flow rate will affect the peak pressure, shouldn’t have any effect on the plat or the driving pressure. The rr thing is interesting, did he posit a reason?
Basically, he propsed that driving pressure x rr = work done on the lung.. My interpretation was that this was essentially the two factors driving VILI. Further proposed that decreasing either driving pressure or rr would both be effective in decreasing VILI. Not sure if they were equivalent though, That got me thinking that the pressure and cycle rate might not be the only factors. H nce the flow rate issue. I wonder if the rate at whoch the driving pressure is applied may have some role in VILI? That peak pressure, however transient, is a stressor of the lung… Read more »
now that is fascinating–it would in some ways be an argument against maximally low Vt as it requires a higher RR when you are reaching the limits of permissive capnea. did he present any evidence?
No evidence that i recall. Talked about future publication. I.e. Soon. He did point out exactly that issue of the max low Vt with high rr as I recall. I’ve asked the World Sepsis Congress to break their podcast up. You have to wade through several hours of podcast to get to the Gattinoni part
Hello Scott, and Mike, I have been wonder about Mike’s point as well: Do high flow rates have an impact on lung injury? Tonight, I am caring for an intubated and mechanically ventilated (lung protective) ARDS patient. The rate is high (30), volumes low (350) with short inspiration time (.8 seconds). The delta P is 15. Deep down, after listening to this podcast, I feel that flow rates must fit into the equation somehow. How can a pressure-cycled breath delivered over .8 seconds be equivalent to one delivered over a longer period of time? With a more gentle flow rate?… Read more »