History and principles of the "open lung approach"

  • An acknowledgement that much of VILI is due to atelectrauma, and most of hypoxia is due to shunt. Both could be managed if we can "open the lung and keep it open."

  • Various interpretations of this over the years, including the use of aggressive recruitment maneuvers and then titrated PEEP scales to identify the ideal maintenance pressure.

  • However, two idiosyncratic ventilator modes emerged as the ultimate manifestation: HFOV and APRV.

  • HFOV used relatively high airway pressure ("PEEP") which was never exhaled. High mean airway pressure meant optimal recruitment, never exhaling meant no chance for collapse. Also no chance for ventilation, but this was achieved with micro-tidal volumes (essentially a vibrating loudspeaker at the vent outlet to vibrate the column of air), which were sufficient to create convection. Unfortunately, two large RCTs in the 2010s (OSCAR, OSCILLATE) showed either no benefit or harm, leading to HFOV's retirement from adult practice.

  • APRV used an alternate strategy, as below.

  • Ultimately the oscillator's history has been a theme, as there has not been much success demonstrating positive outcomes from any of these strategies. It has therefore survived more as a mindset as well as a somewhat niche approach to ventilation, with many clinicians favoring less aggressive, more protocolized approaches, and familiarity with specialty modes waned.

APRV background

  • Known on some ventilators as BiLevel, BiVent, DuoPap.

  • Most associated with Nader Habashi of Shock Trauma in Baltimore, who developed the modern approach often called TCAV (Time Controlled Adaptive Ventilation). Many APRV fans trained in Maryland, and this may be why APRV has somewhat more uptake in surgical, trauma, and burn settings.

  • The concept: since recruitment occurs with higher mean airway pressure and decruitment occurs at exhalation, try not to exhale. Hold your breath.

  • Oxygenation continues to occur even during apnea, as long as some oxygenation remains in the alveoli. The limitation becomes ventilation, which quickly stops (once PACO2 rises to equal PaCO2) and does not resume until breath is exchanged.

  • APRV therefore resembles a series of breath holds, interspersed with brief "releases" allowing a little ventilation, followed by rapid reinflation and return to the breath hold. The releases can be viewed as "breaths," but spontaneous breathing by the patient can also occur at any time, so it is more accurately seen as a CPAP mode with intermittent pressure drops. (They are really the opposite of breaths, as they occur with expiration, not inspiration.)

  • Most users split into two camps, APRV zealots (affectionately termed) and everyone else.

  • Everyone else's view: maximal mean airway pressure is a great way to achieve rescue oxygenation. They might consider this once "conventional" ventilation and rescue ARDS tools have failed, such as paralysis, proning, inverse ratio ventilation, inhaled vasodilators, etc. However, many are also skeptical whether it is truly lung protective, as the airway pressures and release volumes may be high compared to an ARDSnet strategy.

  • Zealot view: it is the most physiologic and minimally invasive way to ventilate. The key is not just the high MAP per se, but the allowance of spontaneous breathing, which allows for more normal lung function/recruitment, minimizing sedation, even allowing fairly simple equipment (in theory). For them, it is appropriate for almost any patient.

How to use APRV

  • Three phases: induction, maintenance, weaning

  • Induction

    • Start on a conventional mode like VC/AC, titrate to relatively standard targets (PEEP, Pplat, etc).

    • Flip to APRV, with settings:

      • Phigh: approximately equal to plateau pressure on conventional mode. This scales the inflation pressure to the current compliance and should help guarantee reasonable release volumes, though you will need to check. It should at least be somewhat reasonable (>200-250ml) and ideally similar to previous Vt, though should be less if you are air trapping correctly.

      • Plow: set to zero. This allows briskest exhalation and least fuss. PEEP is determined by the Tlow.

      • Tlow: Titrate to expiratory flow decay. Flow during releases should not approach anywhere near zero (total exhalation); it should be interrupted by the next inflation to intentionally trap air. Duration is set as a period of time, but is determined by checking flow pattern; this usually requires freezing the screen and pulling over a cursor on the flow scalar (a few vents either report this value in a subscreen, or allow you to set the desired flow decay, avoiding the fuss). Goal is trapping at ~75% peak expiratory flow; usually set around 0.3-0.6s and check. This is the "TCAV" concept and is thought to auto-titrate the "PEEP" to the physiology.

      • Thigh: Adjust to determine total number of releases; can think of this like the respiratory rate. Usually something like 2s initially; too long upfront is a common error. Divide Thigh into 60 seconds to get a "rate" (Tlow usually short enough to be negligible). In general, the more often you release the more you ventilate, but the less you oxygenate. However, with a well-set Tlow you should maintain pretty good recruitment, and ventilation is typically initially a challenge with these poorly-recruited ARDS lungs, so usually you need a fairly short Thigh and frequent releases to start. Most patients are not breathing much spontaneously at this point, so this amounts to the total minute ventilation. Check the total MV (vent will calculate) and compare to the MV on spontaneous mode; if significantly less you will likely have an acidosis problem.

      • Pressure support: usually zero at this point

      • FiO2: usually start 100%

    • Evaluate response:

      • Oxygenation and blood gas. Tolerate some acidosis but should be manageable, otherwise increase "rate" by lowering Thigh, or increase Phigh if there is room. Some would allow Phigh to go high (well above 30 cmH2O); I am reluctant for lung protection reasons, except in the obese.

      • Comfort and oxygenation, although recruitment is likely to take some time. Waiting (and maybe sedation) are expected early; if settings are well-chosen it should improve.

  • Maintenance:

    • Ideally, recruitment occurs over time, improving oxygenation and perceived compliance.

    • Wean FiO2 as usual

    • Do NOT rush to make changes initially or derecruitment can happen easily.

    • Monitor flow trapping and adjust Tlow if needed, mostly to ensure adequate trapping. Once locked in this should not need much fuss but may need adjustment early.

    • Adjust Thigh for ventilation. As recruitment occurs, release volumes will increase, and can often start to increase Thigh (drop the "rate").

    • Don't mess with Phigh much; it is normal for volumes to increase at this point. (Except maybe if you had to go very high to start and weren't comfortable with it.)

  • Weaning:

    • As recruitment continues, you will see release volumes get higher and higher, necessitating longer Thigh to avoid respiratory alkalosis. Continue to increase the Thigh to avoid this.

    • At the same time, once stability and recruitment have been found, wean sedation aggressively. It will now be comfortable for them to breathe, so allow it. You will now see a larger and larger portion of the total minute ventilation comprised by spontaneous breathing. (Many vents report each MV portion separately, which is helpful.) A nice general goal is about 50/50. This spontaneous breathing will also help ventilate. Breathing should look comfortable and easy (not sharp peaks, high rates, low volumes); if not, they are not recruited yet.

    • Traditionally weaning was referred to as "drop and stretch," i.e. stretch the Thigh and drop the Phigh. We would suggest doing more stretching than dropping, as aggressively reducing the Phigh can derecruit. Rather than automatically dropping Phigh, do it in response to your perception of the physiology, eg are they getting better and underlying compliance improving, or just recruiting due to the APRV? Weaning Phigh will usually not occur until a lot of spontaneous breathing is happening; if you drop it and breathing looks worse, you derecruited.

    • Don't overrely on oxygenation as a marker of improvement. Spontaneous work of breathing and efficient ventilation are more sensitive. Persistent respiratory alkalosis on APRV is a sign of failure; they are not recruiting.

    • The amount of air trapping may increase as they improve, as a more compliant lung will not exhale as fast. At this point, you generally don't need to mess with the Tlow to chase 75%; it's okay for them to retain more, as more and more of their ventilation is from spontaneous breathing now.

    • In the weaning process, if they are breathing comfortable but volumes are low, this is a reasonable time to introduce pressure support. However, reflect on why this is happening, as if they are well-recruited but weak, it may denote true respiratory muscular weakness (sedation, myopathy, etc).

    • How to liberate depends on your comfort with APRV. If you'd rather not be on it, once they are making good progress (FiO2 weaned, release volumes improving, ventilating better), you can consider going to a conventional mode with higher PEEP, although it can be tricky to know how much. Friends of APRV will usually wean all the way to extubation readiness or nearly so. Either way, once they are largely breathing on their own (releases are rare), you are not doing much more than CPAP.

    • In most cases people would still perform a traditional SBT (e.g. with PSV), just because they're familiar with it.

Practicalities and subtleties

  • Is APRV lung protective?

    • This is mostly a philosophical question. Some would argue that allowing higher pressures and volumes creates VILI risk, while APRV fans might say the counterbalanced reduction in atelectrauma, and increase in homogeneity from FRC, avoids this.

    • Not much data. The only real positive trial was by Zhou et al, showing some benefit, but (like most vent literature) most has been equivocal. The variations in how to use APRV probably don't help.

  • My take: at least early in the sick patient, try to limit the Phigh and release volumes to something you'd consider lung protective in conventional modes. This gets the best of both worlds.

  • Does APRV "require" spontaneous breathing?

    • Not initially. This is viewed by some as a branch point in severe ARDS, eg paralyze/take over/high PEEPs or APRV and encourage breathing. But initially, APRV probably will not/should not occur in APRV only. Your releases will constitute the ventilation.

    • Later, yes, if no spontaneous breathing emerges, you're doing something wrong.

  • Are there contraindications?

    • Most commonly cited are hypotension and pneumothorax, due to the high airway pressures.

    • You can avoid in these cases, but concern may be overblown. Most shock does best with the lungs at FRC (neither atelectatic or overdistended), although it's true that initial applications of higher pressures may transiently overdistend until recruitment occurs (overdistention happens immediately, recruitment takes time), so a cautious onramp is appropriate. The most sensitive state will be hypovolemia.

    • Barotrauma like pneumothorax is probably most sensitive to the regional stress at the site of leak; the better recruited the overall lung, the more evenly and hence lower pressure can be applied. Also, many times you can ultimately use lower or at least the same pressures than you might use on a conventional mode; the difference is how long it's sustained, which is probably not injurious.

  • Should you use it?

    • Most would agree that oxygenation tends to improve in most cases. However, this doesn't mean it's good for outcomes.

    • In practice it tends to depend on local comfort, as real APRV savviness is somewhat niche. A real consideration is that it's not enough for you to be good at it; your colleagues need to be as well. Or they'll be unable to manage and wean the mode you selected (unless you just use it transiently and get them off it ASAP). Many "occasional users" fail due to subtleties, such as using too low an effective rate or encouraging too much spontaneous breathing early.

    • In the end, the most important function may be the encouragement of higher airway pressures and the mindset that restoring FRC is important. In many cases this could be achieved with a conventional mode with high PEEP, but many people are too PEEP-phobic.