So yesterday's episode, EMCrit – Ghali Grills 1 – When to Use APRV, got some immediate and excellent feedback. I want to address two of the comments in this wee.
Comment From Joe Shiber
Over the last 5 years, every consult I have had for V-V ECMO due to ARDS has been on ARDSnet regimen, most paralyzed, deeply sedated, and requiring pressors. We have used APRV to avoid ECMO in almost all cases when not already dying or in extremis from hypoxemia (pO2 in 30s) so that we have only cannulated 1 out of every 5 consults (~22%) and have an 80% survival rate (this data includes COVID-19 as well as trauma patients whose families withdrew care due to severe TBI despite excellent lung recovery). Our actual “failure rate” of APRV – meaning it was applied when already meeting criteria for severe ARDS and now ECMO (P/F<60, pCO2 elevated causing pH <7.20), but at least not actively dying, so we had 12-24 hrs to attempt to rescue using APRV instead of ECMO, is <10%. So if we are given a chance to use APRV (and what comes with it is less sedation since more comfortable and supported spontaneous breathing, no paralytics, improved cardiovascular function so less pressors, better secretion clearance) then we are able to avoid ECMO.
If I can use it when already critical ARDS to rescue these patients, then it should be used to avoid the need to rescue them by applying it early on EVERYONE. Yes it takes more effort to take of patients using APRV but when is that an excuse to not use a better therapy. We spend time evaluating volume status and dynamic response to volume challenge, we don’t just keep giving IVF boluses for shock. Yes, I trained under Nader too but I have been using APRV for >20 years and understand why it would be personally unethical to conduct an RCT of APRV using a control group of ARDSnet regimen.
Comment from Rory Spiegel
Great new segment on EMCRIT!
Really enjoyed the back and forth, and of course was very excited to hear that the topic on your inaugural episode was APRV. A couple points if you will permit me regarding the topic at hand.
1. The data
As you said, the data on APRV is lacking. But to be fair so is the data on any mode of ventilation. Probably the most frequent and strenuously used argument against the use of APRV, is that it is not in line with the ARMA trial and rather we should be using more traditional modes of ventilation because that was the mode used in the ARMA protocol. This of course is a misunderstanding of the ARMA trial, as it was a trial that examined two drastically different strategies of tidal volume. It did not in fact examine whether a specific mode of ventilation was optimal to any other mode of ventilation. Given the reality of ARMA, its methodology and its findings, it is nonsensical to extrapolate its findings as “proof” that a volume-cycled low-tidal volume strategy is superior to APRV.
A more genuine debate exists. APRV, at times, provides tidal volumes outside the safe volumes prescribed in the ARMA trial. While release volumes may occasionally exceed 6 cc/kg, looking at tidal volumes in isolation lacks the nuanced understanding of APRV mechanics. The size of the release volume is determined by the P-high, T-low and compliance of the patient’s lung. In patients with poor compliance, due to severe ARDS, the size of the lung that is participating in ventilation is very small. In these patients, if the T-low is set to 75% of PEFR, then the release volume will also be low (often less than 6 cc/kg). As the patient’s lung compliance improves, the release volumes will naturally increase without any change to either the P-high or the T-low. Essentially, the release volumes will be automatically tailored to the patient’s individual, increasing functional lung size.
For patients with a severe reduction in lung volumes, who will truly benefit from LTV ventilation, APRV will deliver desired volumes. In patients with larger lung volumes, capable of accepting larger volumes, APRV will allow for these increased volumes. Thus, the debate should not be whether APRV is capable of delivering a LTV strategy in patients with ARDS, but rather, in patients with improving compliance, is tolerating the natural increases in release volumes safe? While the data is not definitive, studies comparing APRV to traditional modes of ventilation have failed to identify any signals of harm due to the variable tidal volumes associated with the use of APRV. In fact, most of the pre-clinical data demonstrates decreased strain on the individual lung units due to APRV’s superior recruitment capabilities. At least in its limited capacity the Zhou et al. trial confirmed these findings. There was no statistical difference in the rate of pneumothoraces between the two groups, and the point estimate noticeably favored the APRV arm (4.2% vs 10.4% p-value of 0.199).
2. Is APRV the superior mode of ventilation?
I suppose it is all in how you define superior. It is hard to say any one mode of ventilation is superior to any other in all situations. Each mode of ventilation offers you some advantages and disadvantages that may or may not be important to consider depending on the specific clinical situation. APRV is a fairly unique mode of ventilation that, like you said, is most similar to CPAP. For this reason it allows you to manage people with far less sedatives on board, have them actively breathing throughout their disease course, thus limiting the iatrogenesis we typically expose our patients to in the ICU. Conversely, it is a mode of ventilation where you have very little precision in determining respiratory mechanics, and so in patients where precise control is key, it can be a difficult mode to deploy.
3. Can you just put them on CPAP?
I would say if you can simply put the patient on CPAP then they do not require APRV. Most of our patients were breathing comfortably with normal respiratory mechanics just a few days prior to this presentation. Whatever respiratory insult that caused this presentation was simply due to worsening compliance of the lung parenchyma resulting in reduced resting lung volumes below FRC. The patient’s respiratory muscles are working just as well as they did previously, only now they have to work harder to manage the same volume of CO2 production. If one could restore the patient back to their previous lung volumes with the use of PEEP, then potentially the patient could more comfortably breathe on their own. The classic example of this, is a patient presenting to the ED with pulmonary edema due to congestive heart failure. These patients typically have relatively normal respiratory muscles, and have respiratory distress due to the increased extravascular lung water leading to a heavy lung resulting in reduced lung volumes. The use of non-invasive positive pressure ventilation is so effective in this patient population because it allows for the clinician to re-establish normal lung volumes using a non-invasive form of PEEP. Rapid restoration of normal lung volumes is readily obtainable due to the relatively swift recruitability of patients with cardiogenic pulmonary edema. This is why most of these patients are managed non-invasively, not requiring invasive mechanical ventilation.
Unfortunately, most forms of respiratory distress are due to underlying disease processes that are not as easily recruitable as cardiogenic pulmonary edema. And while the application of PEEP will often restore normal lung volumes, it does so in a slow fashion. Given the reluctant nature of lung recruitment of an injured lung, it is not feasible or safe to expect a patient to manage their own ventilation. During this period of recruitment, until the patient can safely manage their own ventilatory needs with spontaneous breathing, APRV uses release volume to manage a patient's CO2. As the patient recruits and lung volumes get closer to FRC it is safe to allow them to breath and you can now stretch your APRV to CPAP.
4. Is 70% of APRV just picking the correct PEEP?
Maybe, though I would say how do you decide what the correct PEEP is? The lung is a viscoelastic organ, thus changing over time. The ideal PEEP at any given moment does not necessarily represent the ideal PEEP to re-establish FRC. When using APRV you don't actually set a PEEP. You set the pressure that you think will get you to FRC. The “PEEP” (the trapping pressure in the case of APRV) will increase naturally over time. You never actually reach your P-high . In my opinion, this means you do not have to be exact in the P-high that you select. A range of pressures will have the same effect on the pt's recruitment as long as you have properly set your T-low. This is not the case in traditional modes of ventilation, where you are trying to identify an exact PEEP which will get you to FRC using tools that are incapable of predicting the recruitability of the lung. In addition, APRV offers a number of tools to assess whether you are getting closer to FRC, and if you are not, allows you to correct your course. Standard modes of ventilation for the most part do not. Every PEEP titration strategy I'm aware of uses instantaneous PEEP titration scales which tell you very little about the true recruitability of the lung or what is the ideal PEEP to get you to FRC.
I would add, you mentioned the traditional drop and stretch method of weaning APRV. Or decreasing the P-high (dropping) and increasing the T-high (stretching) in tandem. I would say many of us no longer marry these maneuvers. Rather the T-high is increased as a patient's minute ventilation needs allow it. The P-high should be decreased as the pressure dependence improves. In my experience the latter occurs in a much delayed fashion. For the most part I have to stretch a patient quite extensively before I start to drop their P-high, which typically only occurs after I have woken the patient up, diuresed them extensively and allowed the initial pulmonary insult to run its course.
5. APRV vs. Bilevel, etc
Just a brief note on doing APRV on different ventilators. As Scott said the ability to allow the pt to breathe in as close to an open circuit as possible is key. The sensitivity of the inspiratory valve is what allows them to do this. You can improve things by lowering the %PIFR that closes the valve, but an insensate valve really makes it difficult. The other important factor requires a ventilator that precisely controls the T-low. Many ventilators allow the patient to kick out the T-low and you end up with larger release volumes than you would like. In cases like this it becomes near impossible to stabilize the lung. For me this is a non-negotiable. If you cannot control the T-low then you cannot effectively do APRV.
6. Finally, who should you use APRV on?
Everyone and no one. What I mean by this is in the ICU I pretty much use APRV or CPAP on most patients I manage. The exception to this are asthmatic patients, for whom I do not see the point. They are going to break and once they do, they come off the vent fairly easily. It is possible to manage them on APRV, but it is difficult and the consequences of doing it wrong are dire. Outside of asthmatics I pretty much use APRV/CPAP on everyone. What determines whether I use APRV or CPAP is whether the patient is close to FRC or not.
Conversely in the ED, I rarely put anyone on APRV. APRV is a long-term vent mode. Its major advantages are primarily how it allows you to progress your patients through their disease course more rapidly. If it is not going to be carried over upstairs I'm not sure it has many benefits in the short term. In addition as you mentioned it is fairly complex and requires a great deal of attention. If the patient does not have a primary lung pathology that would benefit from APRV, using APRV during active resuscitations can distract you from more important parts of the resuscitation. Pretty much the only patients for whom I utilize APRV in the ED are those with refractory hypoxemia with double lung physiology.
Thanks for listening. Great podcast, I look forward to hearing more!