Hello. I'm John Marini from the University of Minnesota. I'm going to be speaking to you about apathophysiology primer with regard to acute respiratory failure. I have no disclosure to reveal except perhaps that many of the figures are taken from my recently published book with Dave Dries. Now, I will try to cover a few important principles in the time available. General principles talk a little bit about hypoxemia, ARDS, and hypercapnia. In this talk, I am trying to cram, as I call it, three pounds of flour into a two-pound bag or overstuff a piece of luggage. I'll talk about general principles first and later more specifically about hypoxemia and hypercapnia. The first important principle is that the airway pressure that we apply with a ventilator, the positive pressure, in this case 20 centimeters of water, pushes out against the lung and the chest wall together. And the sum of those two forces must be satisfied so that if I apply 20 and I have a normal chest wall, I have a fairly large lung. If I apply 20 and I have a fairly stiff lung or chest wall, in this illustration, then I have less of a transpulmonary pressure, and I'll get to that in a second, and a smaller lung. The lung itself is passive and therefore transalveolar pressure and compliance determine its volume. If I have a high pleural pressure, let's say 5 centimeters of water, the same airway pressure or alveolar pressure of 15 generates 10 centimeters of water of distending pressure, the lung is flexible and I have a certain volume. If I have a normal intrapleural pressure of minus 5, then the transpulmonary pressure is 20, and the lung dimension, the lung being passive, is bigger. And if the lung is stiff, then the same fairly high transpulmonary pressure of 20 does not result in the same distension. Realize that low compliance can happen either because the lung units themselves are abnormally stiff, as in some chronic restrictive diseases, or abnormally few units are available to take whatever tidal volume you are pushing in, as in ARDS. And we'll talk about that in a moment. In every case, whether you're talking regionally or you're talking globally, transpulmonary pressure, including PEEP, plateau, and driving pressure are what matters. Don't be fooled by specific numbers, even under passive conditions that are said to be safe. For example, a plateau pressure of 25 is safe, or a plateau pressure, excuse me, a driving pressure of 15 is safe. It depends on the trans-lung pressure, and that we often don't know. We take our best guess with airway pressure alone, but we really don't know it unless we measure transpulmonary pressure with an esophageal balloon. Another important concept that we must be aware of is that mean airway pressure is important under passive conditions, but not under active conditions. In other words, if the respiratory system is totally passive, then an airway pressure that we measure externally is actually the same as the mean airway pressure at the alveolus. And I've tried to illustrate that here, but the basic message here is that the mean pressures averaged over the entire respiratory cycle should be equivalent at the airway opening and within the alveolus, only under passive conditions. So, mean airway pressure is associated with lung dimension, average lung dimension, and also average chest wall dimension under passive conditions, and this influences hemodynamics as well. Mean airway pressure is important to track under passive conditions. Also, remember that regional compliances vary, so they vary with the transpulmonary pressure in those regions. For example, here we have an increase in pleural pressure as we proceed from cephalocaudily, as the blue arrow shows. We have dependent lung units which are relatively small because they have smaller transpulmonary pressures at that level than do non-dependent units. If we measure just at the airway opening, we will get a mixture of such units, and what the configuration looks like will be dependent on how much the units of either type we're looking at. Transpulmonary pressure, compliance, effort, and volume are illustrated here. So, in summary, if we have a stiff chest wall, the same airway pressure of 30 may produce the same lung volume as if we use only 20 with a normal chest wall compliance or 10 if we are actively inspiring. The same transpulmonary pressure results in every case. Now, the cardiovascular effects of inflation with positive pressure are significant and different than during spontaneous conditions. On the left, you have, let's say, an alveolar pressure of zero at end exhalation, and with the next inspiration, there is blood drawn back into the the chest cavity, and the lung is well filled. Under positive pressure conditions, the pleural pressure goes up. There's a tendency for venous return to be impeded and for the alveoli to over-descend. This is quite important in the setting of passively ventilated patients, especially with those with baby lungs. When I say baby lungs, I mean small air spaces. The breathing mode, as you've already understood, is important in that it influences the transvascular pressure and edema formation, as well as the size of the lung unit from the airspace side. Let's say we have a capillary pressure of 15 centimeters of water, and we have three different conditions. One passive on the left, gentle and forceful breathing. What you'll see is that the transvascular pressures, which are conditioned by the difference between the intraplural, I mean the intracavitary vascular pressure and the intrapleural pressure, in this case, you're dealing with 15 minus 10, or a transvascular pressure of 5. Here, you're looking at 15, the same 15 that we measure externally, surrounded by a minus 10, and a transvascular pressure of 25. And that may mean that edema forms more readily, especially in injured lungs. Now let's talk about hypoxemia, ARDS, and the relationship to VILI. The mechanisms of arterial hypoxemia, and of course hypoxemia is what we're focused on at the bedside in many cases, can be addressed by low inspired FiO2 hypoventilation, impaired diffusion as causes, or ventilation perfusion mismatching or shunting, as well as desaturation of mixed venous blood. It is these measures that are most important in the setting of ARDS, and that includes desaturated mixed venous blood. Why is that true? Well desaturated mixed venous blood plays into what is pushed past the shunt units. The venous blood mixes with well-exchanged blood to give us our SaO2 that we're so interested in. And Sv bar O2, this mixed venous oxygen saturation, is proportional to the oxygen saturation of the arterial blood, of course, but subtracted from that is the oxygen consumption to oxygen delivery ratio. This is cardiac output, this is hemoglobin, this is oxygen consumption. So the mixed venous oxygen saturation is determined by those things, and that is what mixes with the good well-oxygenated blood in the open capillaries. These things will influence, therefore, mixed venous oxygen saturation and the hypoxemia, or correction of hypoxemia, that we produce. What's the implication here is, for the same cardiac output and hemoglobin or oxygen delivery circumstance, changing the oxygen demand is an important piece of this. Oxygen demand is quite important in determining what the mixed venous oxygen saturation is. So the techniques to improve tissue oxygenation include increasing FiO2, of course, especially when the lung is not terribly shunted, to increase mean lung volume and mean alveolar pressure by PEEP and other measures, upright or prone positioning, bronchodilation, improving oxygen delivery to consumption ratios we've just covered by such things as reducing agitation, reducing fever, increasing cardiac output, and correcting of severe anemia. Those all influence a tissue oxygenation more so when the lung is severely ill than when it is not. Removing systemic pulmonary vasodilators and considering adjunctive support or other measures that may come into play in certain patients. Now, remember that I said that the mixed venous oxygen saturation and its effect on arterial oxygen saturation depends on the lungs condition. As the lung becomes more severely affected, mixed venous oxygen saturation has a more profound effect and those measures that we talked about are more influential. Something I see all the time, which irritates me a great deal, is that people turn oxygen inspired oxygen values up to very high levels from 85 to 1.0, hoping that the arterial oxygen saturation will improve a lot. But when you have a high shunt fraction, this QSQT, it doesn't make much of a difference. And it can actually add to the inflammatory state of the lung. So once you get above 80-85, test whether higher FiO2s will work. But if they don't, then drop back to the lowest value that seems to preserve the arterial saturation that you want. Now, here we go again with the transpulmonary pressure and its regional effects. And those have an influence on compliance. If we're operating with a lot of unrecruited lung units, we have a certain compliance. If we look at a tidal volume being pushed in and the pressure change that results from that, we may see on the global external pressure volume relationship or the tracing that we're looking at under passive conditions, a certain tidal compliance. When there's a lot of recruitment and very little overdistension, we may see a fairly linear appearance. And not only a linear appearance, of course, because we're only looking at two points, but we're looking at a steeper slope and a better compliance. And as we approach the end here at the upper zone, we have low compliance again. So tidal compliance reflects the over and under distension balance of the lung, and it can be misleading if we assume that it's only at the upper end that it deteriorates. The compliance measured at the central airway may be misleading. The total lung may look fairly well open and not overdistended and not terribly recruiting, but in different zones, in the lower zone or the upper zone, for example here, you can have under recruitment or overdistension, and those have an impact on the micromechanical forces. At the top of the lung or in the lung units that are near overdistension, you have high levels of stress and strain. And at the junctions of open and closed lung units, you have amplified stresses and strains, which may actually be more intense than they are in a more open part of the lung. So that when we look at the effect of PEEP on lung protection, some folks like to just open everything up. Certainly that's good if you reopen vascular bed, have better oxygenation, less hypercapnia, because you've produced a lot of stable recruitment. But at the same time, in every patient, you're going to be overdistending some areas of the lung, and if you don't recruit much, you're going to increase and amplify the stresses on recruited tissues. PEEP is not your friend in many cases. It creates dead space the higher it goes. The amount of alveolar recruitment tends to tail off at higher levels. Oxygen delivery may be impaired because you're increasing mean airway pressure and impeding venous return. So you have to be a little bit concerned that you're titrating PEEP in the right way and to the lowest effective value. When we try to avoid excessive mechanical stresses that are associated with a ventilator-induced lung injury, I think we've progressed in our thinking from low tidal volumes to reduced cumulative injurious strain. It's not just driving pressure. It's not just plateau pressure. It's not just power or even driving power. It's reduced cumulative injurious strain. I don't have time to go into the details here, but the frequency of what you're doing as well as the nature of the individual cycle is important. Not only that, but the capacity of the lung that you're ventilating is important. If you have a normal lung capacity, moderately severe reduction in lung capacity or severe reduction in lung capacity, you're going to do some interesting things. The externally measured airway pressure may be the same in each condition. The global power that we're focused on right now may be the same in all conditions, but the specific power is concentrated the smaller the lung that you have. Also, notice that when you ventilate a lot and you have very little perfusion going through those ventilated alveoli, you're going to have a ventilation-perfusion ratio, which is higher than it would otherwise be and may be misleading for you. Dead space will track the progression of lung injury, the size of the baby lung. I don't have time to get into that either, but it is an important concept. The size of the lung that you're dealing with is important. And vigorous breathing increases the transpulmonary pressure, which should be obvious to you, and violates the objectives of lung protection that are provided by a given airway pressure. If you have an airway pressure of 20, one might become concerned under no effort conditions. 10 should also produce the same level of concern if you're having active inspiration. Position powerfully affects resting lung volume, an FRC, in normals and in patients with ARDS. In fact, the physiological effects of proning have become obvious with COVID. We've been talking about this for a long time. This image taken in my hospital from 1994, we were using it and finding its value. It not only alters the conformation of the lung, it reduces the gradient of trans-lung pressure we've been talking about. It recruits and stabilizes atelectatic units, encourages mouthward migration of secretions, attenuates villi risk, and may do other beneficial things as well. So the primary determinants of villi risk in 2021 involve power, frequency, damaging energy per cycle, and probably some things we haven't considered yet. Let's talk very briefly about hypercapnia and airflow obstruction. I think we know that contributors to ventilatory insufficiency include impaired muscle strength and endurance, impaired ventilatory drive, excessive ventilatory requirement, increased impedance to ventilation. All of those things we're familiar with as intensive care unit physicians. We're also familiar with AutoPEEP, which we described in 1980 and published in 1982. Basically, it's unintended extra pressure at end exhalation, which may produce hyperinflation and can only be detected if you stop flow. Notice that it can be caused by any number of things, only some of which are associated with hyperinflation when you compare different patients. In the same patient, AutoPEEP tends to parallel hyperinflation, but across different patients hyperinflation may not be as intense in some and more intense than others. And hyperinflation is the culprit. It increases the workload and decreases the ability to perform it. Applied PEEP may reduce the breathing workload if tidal flow limitation is present. Externally titrating in PEEP can help counterbalance AutoPEEP only if there is flow limitation. AutoPEEP is not uniformly distributed. Just like in ARDS, the transpulmonary pressure may be quite different in different zones. And airways will close earlier in the expiratory cycle in some regions than others, trapping gas at much higher pressures in some than others. And you may only be able to measure those that are in the open airways, which can be misleading, especially in asthma. The concept of dead space has already been introduced. Remember that PaCO2 is related to oxygen consumption and minute ventilation minus the dead space. And as the alveolar ventilation falls, the PaCO2 rises because the dead space is driving the reduction in effective ventilation. You can have the dead space increase because of vascular occlusion, shunting of CO2 containing blood as I've alluded to earlier, hyperperfusion of the lung even without a total occlusion of the pulmonary vasculature, excessive PEEP and mean airway pressure, and excessive demands on the baby lung. This CO2 production to alveolar ventilation ratio is what determines its effect on PaCO2. Let me just end by saying that measuring expired CO2 is very valuable for a number of reasons. It can help you determine how much dead space you're dealing with and actually how much CO2 is being produced. The ventilation ratio is something that's coming into parlance right now and it's being used increasingly. It's a good, convenient, but imperfect reflection of dead space. And as I mentioned, dead space is important. The ventilatory ratio is minute ventilation times PaCO2 divided by predicted body weight times 100 times 37.5. That influences the CO2 production. So there are consequences and cautions related to hypercapnia. I don't have time to go into this in detail, but they're covered on this slide. There are consequences and benefits of allowing permissive hypercapnia. Lastly, I want you to remember that a little physiology goes a long way at the bedside, even in 2022. These three gentlemen inspired me in my applied respiratory physiology research, etc. Otis Finn and Ron, they were instrumental and I still use their principles at the bedside. Thank you for your attention.

acute respiratory failure

hypoxemia

ARDS

hypercapnia

ventilator

transpulmonary pressure