You just intubated a 35-year-old woman (BMI 40) with COVID-19 and severe ARDS. You set the vent to VC/AC (VC-CMV) with FiO2 60%, Vt of 360 (6 mL/kg IBW), RR 32, and PEEP of 10. The PPlat is measured at 29. ABG is 7.20/55/60
What would you do?
The basics of lung-protective ventilation are to prevent
The above forms of ventilator-induced lung injury (VILI) besides physically injuring the lungs can also cause biotrauma (release of inflammatory mediators and leukocyte recruitment)
Low tidal volume (Vt) ventilation is the cornerstone of lung-protective ventilation in acute respiratory distress syndrome (ARDS). Low Vt is synonymous with <6 mL/kg ideal body weight as based on the ARMA trial (2). Also, the goal of lung-protective ventilation is also to maintain a plateau pressure less than 30 cmH20.
As the amount of lung participating in gas exchange is reduced in ARDS- the functional lung or the “baby lung” could suffer ventilator-induced lung injury (VILI) even with a 6 mL/kg IBW Vt. Amato et al. performed various statistical analyses in a cohort of 9 previous randomized controlled trials (RCT) of patients with ARDS (3). The study showed that the driving pressure was most strongly associated with survival among the ventilation variables.
We will try to understand the basics of driving pressure, how to use it at the bedside, and work through a few scenarios in this post.
What is driving pressure (ΔP)?
Starting at the basics
Compliance = ΔV/ ΔP
At the end of inspiration, the static compliance of the respiratory system (Crs) is:
Crs = Vt / ΔP
Rearranging the equation,
ΔP = Vt/Crs
Thus, the driving pressure is the ratio of Vt and the static compliance of the respiratory system (Crs). In other words, ΔP represents Vt scaled for a patient’s Crs.
Physiologically, this makes a lot of sense as depending on the severity of ARDS and size of the baby lung, even a Vt of 6 ml/Kg IBW could be higher than the “lung-protective” range. Delta P helps us determine the appropriate Vt and PEEP in patients with ARDS.
How to calculate ΔP at the bedside?
ΔP = PPlat – PEEP = Vt/Crs (in a passive patient)
Where PPlat is the plateau pressure and PEEP is the positive end-expiratory pressure.
Taking it a step further
For a given value of PEEP and compliance, PPlat will be higher if you increase Vt.
However, increasing the PEEP will not always lead to increasing TPP. Let’s see why:
As discussed above:
PPlat – PEEP = Vt/Crs
PPlat = PEEP + (Vt/Crs)
If you recruit the lung by increasing the PEEP, the Crs will increase, and thus the PPlat could increase, decrease, or remain the same (depending on how effective the recruitment is).
At the bedside, one way to determine if you have recruited more lung (improved Crs) by increasing the PEEP is through driving pressure.
Remember, ΔP = Vt/Crs
So for a given Vt,
- If by increasing the PEEP, you successfully recruit (i.e improve Crs), the ΔP will decrease
- If by increasing the PEEP, you cause over-distention of the functional lungs (i.e worsen Crs), the ΔP will increase.
- If ΔP remains the same, it reflects the absence of both recruitment and tidal over-distension
The most significant evidence on effect ΔP on outcomes comes from statistical analyses (including mediation analysis) performed by Amato et al. on ~ 3500 patients enrolled in previous RCTs on ARDS (3). The analysis showed that among the ventilation variables, ΔP had the strongest association with survival. An increase in ΔP was associated with an increasing mortality, with an inflection point around ΔP of 15 cmH20. Furthermore, Vt or PPlat was not associated with mortality when ΔP remained the same. See the classic figure (Fig 1) in the original paper here.
It is very important to note that even though this was an exhaustive analysis of patients enrolled in RCTs, the quality of evidence does not equal that of a prospective RCT. Attention should be paid at the bedside to optimize all the lung protection variables during mechanical ventilation.
Now let us work through a couple of scenarios to apply ΔP in making clinical decisions.
35-year-old woman (BMI 40) with COVID-19 ARDS- you just intubated and placed the patient on VC/AC (VC-CMV) with FiO2 60%, Vt of 360 (6 mL/kg IBW), RR 32, and PEEP of 10. The PPlat is measured at 29
ABG is 7.20/55/60
What would you do?
- Increase Vt or RR to correct acidosis -> let’s get to a pH of 7.4
- Increase FiO2 to 80% -> more oxygen is good
- Increase PEEP to 16 -> but wait, my repeat PPlat is now 31
- Do nothing! The gas is acceptable for a patient with ARDS
In most cases in critical care, D would be the correct option, however, in this case, you have the possibility to recruit the lungs by increasing the PEEP.
The driving pressure at PEEP of 10: ΔP = Pplat – PEEP = 29 – 10 = 19
The driving pressure at PEEP of 16: ΔP = PPlat – PEEP = 31-16 = 15
Even though the PPlat is higher than the magic cutoff of 30, the ΔP (PPlat – PEEP) decreased (from 19 to 15)
ΔP = Vt/Crs
Since Vt is unchanged, the reduction in ΔP shows that you improved Crs by recruitment. Another important point to note is that patients with obesity such as in this case may have mass loading of the chest wall and higher pleural pressures.
Thus for the same PPlat, the TPP is lower in obese patients (TPP = PPlat – Ppl). So, a cutoff of 30 for PPlat in those with poor chest wall compliance might not be feasible in order to achieve adequate recruitment and prevent atelectrauma. In situations like these, monitoring the ΔP and targeting it to <15 cmH20 might be a reasonable approach. Another alternative is to measure PPl via an esophageal balloon but this practice is limited by experience in accurate positioning and interpretation.
A 50-year-old man (BMI 28) with ARDS transferred from OSH for evaluation for ECMO for severe ARDS. Currently on VC/AC (VC-CMV) with FiO2 70%, Vt of 460 (6 mL/kg IBW), RR 30, and PEEP of 24. PPlat measured at 39
What would you do next?
- Increase FiO2 and PEEP to correct hypoxemia
- Increase FiO2 and keep PEEP at 24 (since ΔP is acceptable at 15)
- Increase FiO2 and titrate PEEP to find the optimal range of PEEP
- Cannulate ASAP
In this patient, the PEEP was titrated from as high as 26 to as low as 8 while maintaining Vt at 460 mL. As shown in the figures, the ΔP for various levels of PEEP were as follows.
VC/AC (VC-CMV) demonstrating inspiratory hold maneuver: PPlat = 43 and PEEP = 26; thus ΔP is 17 (an expiratory hold showed minimal intrinsic PEEP, as the measured total PEEP is 27. In cases where the intrinsic PEEP is significant, ΔP should be calculated by PPlat – measured PEEP).
Based on the above data,
- The higher PEEP levels caused over-distention as evident by higher ΔP
- At a PEEP below 12, the ΔP started increasing again suggesting decreasing compliance (loss of recruitment)
In this case, the optimal PEEP for the best respiratory system compliance seems to be around 12-16 and thus PEEP was set at 16 as it resulted in the most optimal oxygenation.
Tips and tricks
- Increase the PEEP to a fairly high level (consider following the ARDSnet PEEP:FiO2 table) and titrate down by increments of 2. Wait a few (1-5) minutes after every titration to check Pplat (4).
- If the lowest ΔP seems to be unchanged over a certain range of PEEP (PEEP 12-16 in Scenario 2), set the PEEP at the value with the most optimal oxygenation.
- Therapies that change Crs will change ΔP for a given PEEP and Vt (e.g. proning, therapeutic thoracentesis, and natural course of ARDS).
- Not useful if the patient is very active on the ventilator and unable to successfully perform an inspiratory hold.
- Changing positions will change functional residual capacity (FRC) and Crs (for example supine vs head of bed 45 degrees in a patient with central obesity) (5).
- Intrinsic PEEP, if significant will alter the values. Ideally, measure ΔP with PPlat – total PEEP (and not applied PEEP)
- Driving pressure is calculated as PPlat – PEEP in a passive patient
ΔP = Vt/Crs
Some evidence suggests that lower ΔP is associated with better outcomes in ARDS, however, no RCTs are available
ΔP is a useful parameter at the bedside to determine optimal PEEP to minimize ventilator-induced lung injury in ARDS.
ΔP may also be a better target for ventilator changes than PF ratio! (6)
- Curley GF, Laffey JG, Zhang H, Slutsky AS. Biotrauma and Ventilator-Induced Lung Injury: Clinical Implications. Chest. 2016 Nov;150(5):1109-1117. doi: 10.1016/j.chest.2016.07.019. Epub 2016 Jul 29. PMID: 27477213.
- Acute Respiratory Distress Syndrome Network, Brower RG, Matthay MA, Morris A, Schoenfeld D, Thompson BT, Wheeler A. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000 May 4;342(18):1301-8. doi: 10.1056/NEJM200005043421801. PMID: 10793162.
- Amato MB, Meade MO, Slutsky AS, Brochard L, Costa EL, Schoenfeld DA, Stewart TE, Briel M, Talmor D, Mercat A, Richard JC, Carvalho CR, Brower RG. Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med. 2015 Feb 19;372(8):747-55. doi: 10.1056/NEJMsa1410639. PMID: 25693014.
- Sahetya SK, Hager DN, Stephens RS, Needham DM, Brower RG. PEEP Titration to Minimize Driving Pressure in Subjects With ARDS: A Prospective Physiological Study. Respir Care. 2020 May;65(5):583-589. doi: 10.4187/respcare.07102. Epub 2019 Nov 26. PMID: 31772068.
- Benedik PS, Baun MM, Keus L, Jimenez C, Morice R, Bidani A, Meininger JC. Effects of body position on resting lung volume in overweight and mildly to moderately obese subjects. Respir Care. 2009 Mar;54(3):334-9. PMID: 19245726.
- Yehya N, Hodgson CL, Amato MBP, Richard JC, Brochard LJ, Mercat A, Goligher EC. Response to Ventilator Adjustments for Predicting ARDS Mortality: Driving Pressure versus Oxygenation. Ann Am Thorac Soc. 2020 Oct 28. doi: 10.1513/AnnalsATS.202007-862OC. Epub ahead of print. PMID: 3311264
Thank you for this post.
In the table included in the post, there are six discreet measures of driving pressure made at various PEEP settings.
Are you waiting a minimal time period between each measure of driving pressure?
Interested to see if there’s any consensus on this.
Reference 4 by Sahetya is helpful here; between 1 and 5 minutes is probably adequate for driving pressure change. Gas exchange (particularly oxygenation) will take longer, easily an hour.
Thanks for this post. Looking at the scenario 2 ventilator waveforms I saw that you are using a decelerated flow. Is there any evidence of this for being superior to the square flow? I know that this question is out of this topic but I couldn’t dismiss the opportunity for asking.
Thanks for your comment! Not superior exactly, though there is a commonly-held belief that decelerating flow is more comfortable for the patient compared to constant flow (square waveform).