ARDS
===

[00:00:00] All right. I'm John McClellan. I'm here with Dennis Kim. Before we say hello, a quick one. Post-op day two after a trauma laparotomy for a splenic injury and bowel repair. He got a few units of blood in the OR. Now 36 hours out and the FiO2 is climbing. The PEEP is climbing and the chest film was clear yesterday, now has bilateral fluffy opacities.

The overnight note says fluid overload and diuresis. You walk in, what's the n- one number you're gonna calculate before you accept that?

So here, John, we're gonna start with the PF ratio. That's our P little a O2 divided by the FiO2. But we're not gonna stop there. Bilateral opacities plus hypoxemia can and should certainly make you think about ARDS.

But it could also be hydrostatic pulmonary edema or it could be both. So the point is to recognize and consider the syndrome early enough so that we can actually apply lung protective ventilation while we're getting things sorted out in terms of cause.

Okay. What if the PDF [00:01:00] comes back at 140?

That sounds pretty consistent with the moderate oxygenation range by the Berlin definition or criteria.

But the ratio alone doesn't diagnose ARDS. If the timing, imaging, and non-cardiogenic edema criteria fit, then the ventilator strategy changes immediately. And diuresis may still be appropriate later if the patient's resuscitated and adequately stable hemodynamically. I think one key point is that, again, ARDS and fluid overload can coexist.

All right. Welcome back. This is the surgical critical care curriculum from Behind the Knife. This is our walk through the topics you have to own as a surgical intensivist or as a new attending running the ICU. Today, it's acute respiratory distress syndrome, what it actually is, how to stop missing it, and the ladder of therapies from tidal-- low tidal volume basics all the way to ECMO.

And it's also from the seat of the person managing the patient, not the person reading about it before boards. [00:02:00]

So ARDS is common in surgical critical care. Multiple potential etiologies, pneumonia, aspiration, sepsis, pancreatitis, major trauma, pulmonary contusion, uh, as well as massive transfusion can all be triggers.

And I think the, the key fact here is that mortality runs anywhere from twenty-five percent in mild disease up to almost fifty percent in severe ARDS. And the key thing here is that most patients actually don't die of the one thing that we all get fixated on, refractory hypoxemia. They typically die of multiple organ failure.

Yeah. So we'll spend most of our time during this podcast talking about the physiology and the treatments with the strongest evidence for ARDS. And then we'll give you like a true practical rescue ladder for a patient that's failing that you can use clinically and also on the boards later on. All right.

So the problem, why [00:03:00] ARDS owns your unit and why it's so common. All right, we'll start with the definition. Most of us learn Berlin, but there's now even newer global definition. Give me the bedside version, Dennis.

So four key criteria. Number one is timing, John. So within seven days of a known clinical insult or new or worsening respiratory symptoms.

Number two, imaging, typically bilateral pulmonary opacities either on X-ray or CT, not fully explained by effusions, atelectasis, or nodules. Number three is the origin of the edema, so it's typically respiratory failure not fully explained by cardiac or LV failure or fluid overload. And if the cause is unclear, we're gonna use objective assessments and in our units that would typically be bedside echocardiography or POCUS, and then the clinical hemodynamic picture.

And finally, oxygenation measured on a PEEP of at least five. [00:04:00]

All right. So how do we grade the oxygenation? Because the grade really drives everything downstream.

Yeah. So on a PEEP greater than or equal to five, mild ARDS is a PF between 200 up to 300. And like I said earlier, mortality around 25%. Moderate is 100 to 200 with a mortality around 30%.

Severe is 100 or less, and here's where mortality climbs to about 45%. So severity is clinically useful, but treatment thresholds are therapy-specific, which is why it's important to know at least the PF ratios and what they mean. For example, Pronovost-style proning is triggered around a PF under 150 with substantial oxygen requirement, while the 2024 ATS neuromuscular blockade or paralysis recommendation is really limited to early severe ARDS, roughly a PF of 100 or less.

Okay, so what changed with [00:05:00] the global definition?

Yeah. So with the global definition, it really broadens our recognition that patients with ARDS, not all of them are intubated. It includes patients on high-flow nasal oxygen who are getting at least 30 liters per minute. And here we can look at the SF or saturation to FiO2 ratio, and the number we're looking at is 315 or less.

And that's typically when your SpO2 is 97% or lower. And we can also incorporate some ultrasound or lung ultrasound as an imaging modality versus just chest X-ray or CT. For an intubated patient, Berlin remains really familiar. I think it's a great practical framework. But the newer definition reflects how we actually support hypoxemic patients in 2026.

Okay. So that was a lot that we already know about ARDS and some updates that maybe somebody who studied ARDS 10 years ago. What's [00:06:00] like the true practical takeaway before we move on and get into the physiology?

Yeah, John, I think like so many things in clinical medicine, no single number confirms the diagnosis of ARDS.

So we certainly want to screen and think about it in our differential, calculate the PF or the SF ratio, and then apply the whole definition and think about the precipitating etiology or insult. I guess also remember that hydrostatic edema doesn't make ARDS impossible because mixed pulmonary edema is so common in the surgical ICU

All right.

So now we're gonna move into the first principles, really talking about ARDS physiology and the concept of the baby lung. So Dennis, how about we whiteboard the pathophysiology? Because rational ventilator management really falls out of ARDS, and this is actually really heavily tested on the ABS exam, so our fellows really need to know, understand it.

For sure, John. So classically three [00:07:00] phases, and again, there's overlap with no hard timeline or cutoffs. But generally, first is the exudative phase, day zero to seven, and the insult activates alveolar macrophages that dump pro-inflammatory cytokines that recruit neutrophils, monocytes, and it's that immune response that really causes damage and chews through the alveolar epithelial endothelial barrier.

And then when that barrier fails, what actually happens within the alveolus?

Yeah, that's when it floods. You get this protein-rich edema that fills the interstitium as well as the alveolar space. And in addition to that, we lose surfactant, both in terms of dysfunction and loss of production, so our alveoli collapse, and we'll talk a little bit more about atelectoma later.

There's also deposition of hyaline membranes, and then we can also lose epithelial sodium channels that normally [00:08:00] absorb or reabsorb edema, so that fluid now can't clear. So ultimately, the endothelium is injured, and then there's a whole host of microvascular downstream effects as well. And ultimately, those mechanisms can result in a loss of hypoxic pulmonary vasoconstriction

Yeah, and that all really adds up to the hypoxemia we see clinically.

And explain more about why we actually see that hypoxemia.

Yeah, I think the two major etiologies are really severe VQ mismatch and then shunt, specifically intrapulmonary shunt. Like I said earlier, the vascular injury also matters. And as pulmonary vascular resistance rises, the afterload on the right ventricle increases.

And that's really something important to consider as we're adding PEEP and when we talk about tolerating things like hypercapnia.

Yep. And also, now all the damage is done, and the lung now tries to fix itself, [00:09:00] moving us into another phase.

Yeah, so that's the proliferative phase, typically day seven through 21.

And like you said, John, the repair phase. So type II alveolar cells are proliferating. They're differentiating and looking to reestablish the barrier. Edema starting to clear. And we have neutrophils, apoptotic neutrophils that get cleared as well. By this time, vasomotor tone starts to normalize, and we start to see clearance of some of that microthrombi at the vascular level.

Yeah, then finally, that phase that we've all seen but we don't want to reach.

Yeah. A- And it's hard to predict. In fact, it's probably impossible to predict. But in some patients, a fibroproliferative or fibrotic phase develops. And, and that's really marked by fibroblast activation, deposition of extracellular matrix.

There's this ongoing persistent low lung compliance. And [00:10:00] oftentimes, these patients have prolonged ventilator dependence. As we mentioned earlier, these phases overlap, and there's no clear-cut time limits or restrictions.

Yeah. Now, this is the concept where we talk about the baby lung. It reorganizes how you're gonna set your ventilator.

And it brings a little bit of a perspective on why we do the things within ARDS for clinical management. So let's go into that.

Yeah, I think the key point, and most people who have dealt with ARDS or managed it know this, is that ARDS injury is heterogeneous. It's not a uniformly stiff lung. So the dependent regions, and again, most patients are supine, the ones down against the bed are consolidated as well as atelectatic.

And then the non-dependent regions stay relatively aerated. So the lung that's actually available to ventilate is dramatically shrunk, often down to the size of a pediatric lung, hence the famous [00:11:00] term coined by Gaudenoni, the baby lung

And then, like, the trap that follows directly after that, what do you, how can you explain that?

Yeah. When we think of a, quote-unquote, "normal tidal volume," 10, 12 ccs per kg delivered into a baby lung, that's not normal at all. That's a huge volume going into a tiny aerated compartment and can massively overextend exactly the regions that are still working and that aren't as of yet injured. So respiratory system compliance reflects the size of that baby lung, and that's the whole physiologic rationale for low tidal volume ventilation.

We're sizing the breath to the functional lung, not to the patient's body.

Which sets us up to the four ways the ventilator actually ends- injures itself. Again, VILI, right? Ventilator-induced lung injury.

Yeah, for sure. Those four mechanisms can all be happening simultaneously in the heterogeneous lung. So one, [00:12:00] volume and barrel trauma, we typically put those into the same bucket.

That's really over-distension of aerated units from excessive tidal volume and transpulmonary pressure. Two is atelectrauma that we've already mentioned. That's that repetitive cyclical opening and closing of the unstable alveoli right at the border between the aerated and collapsed lung. And then three, which is probably the most important one, is biotrauma.

That's the activation of pro-inflammatory signals and releasing of cytokines, not just into and around the alveoli, but into the systemic circulation. And that's what feeds that multiple organ dysfunction syndrome that ultimately kills our patients. And then four, a lot more attention paid to this across various disease processes, severe TBI, and of course, in ARDS would be oxygen toxicity.

So prolonged high FiO2 makes reactive [00:13:00] oxygen species that can injure the epithelium further.

All right, and biotrauma is really the one that I want the fellows to sit with because it really connects the vent to the whole patient.

Yeah. So VILI is not just pneumothorax risk. It really is the excessive stress and strain that can amplify the pulmonary and systemic inflammation, which is a major reason why lung protective ventilation matters to the entire patient.

All right, now moving into the second critical point of this section is the driving pressure. And it's really what we talk a lot about in during ARDS and plateau pressures, everything which we'll go into as well. So after tidal volume and plateau pressure, the most useful mechanic number that you're gonna look at is your driving pressure.

Let's talk about that.

Yeah. I think when we look at management of ARDS, one major component of that is keeping our plateau pressures less than or equal to 30 centimeters of water, and that's typically measured with [00:14:00] an inspiratory breath hold. The driving pressure is the delta P. That's our plateau pressure minus total PEEP in a passive patient.

So you can also think about it as tidal volume divided by the static respiratory system compliance. So two patients both can be receiving six cc’s per kg of predicted body weight and yet still have very different stress on the available lung because their compliance is so different.

Okay, so what's the evidence behind all that?

Yeah, I think there's a few good papers by Amato and colleagues and individual patient analysis showing that driving pressure has the strongest association with survival among the ventilatory variables that group examined. A delta P or driving pressure around 15 centimeters of water had the strongest association with survival among the various ventilatory variables examined.

I [00:15:00] think one key thing to remember here is that it's not a prospectively validated target and treat threshold. If it's high, first we're gonna ask whether we can reduce our tidal volume or at the same time improve recruitability without causing hemodynamic harm.

Yeah, so the concept is simple. We use our predicted body weight to set the initial tidal volume.

Then we use our plateau pressure and our respiratory system mechanics to see how that breath is being tolerated. All right, so now we're gonna move on to recognition of ARDS and really catching it in the RCUs and not mislabeling it as something else. All right, so Dennis, recognition. Yeah. When does ARDS light up for you at the bedside and you think this is a top of your list of what's going on?

Totally. I think when you've got that patient with acute hypoxemic respiratory failure that's developing relatively rapidly, that can be over hours or days after a known risk factor, of which we've already stated there are many, especially in our [00:16:00] surgical ICUs. Rising FiO2 and PEEP requirements, falling compliance, which at the bedside we use plateau pressure as the adjunct.

And if that's creeping up at the same time as the tidal volume, I'm getting really concerned. Of course, there's radiographic criteria, and so when we're seeing bilateral infiltrates that are progressing rapidly on the film, that should also raise our awareness

Yeah, that compliance clue is really underused.

You know, we used to-- I remember being as a med student in the ICU and talking about compliance and everybody saying, RTs are saying the compliance is bad. But I think it's a really interesting part of what we need to be watching in the ICU.

Yeah, and I think it's not just an absolute value in time. And like many things in critical care, it can be a useful trend, but not a diagnosis in and of itself.

So if plateau pressure is climbing at the same time as tidal volume, overall system compliance is [00:17:00] worsening, then we're gonna ask why. Is there a progressive lung injury, edema, atelectasis, other reversible causes like a pneumothorax? Perhaps there's worsening secretions. A lot of our patients post massive resuscitation, damage control surgery, or an XLAP in the setting of an EGS emergency also may have some intra-abdominal hypertension.

There's also chest wall mechanics and then, of course, dyssynchrony as well

Yeah, bilateral opacities really aren't automatically ARDS either. It's something that we have to be concerned about. But what's our differential for that?

Yeah. I think commonly cardiogenic pulmonary edema, pneumonia, aspiration is a big one, especially in our patients coming in with a low GCS and for whatever reason they've all got full stomachs.

And then there's less common things like diffuse alveolar hemorrhage, acute eosinophilic pneumonia, and then other causes of [00:18:00] diffuse infiltrates. So we really want to get a global picture. We're gonna look at the history, our fluid balance over the course of the first couple of days in the ICU, liberal use of bedside ultrasound as well as echocardiography and other invasive hemodynamics when needed.

One key thing here is that a B and P alone is not reliable enough to separate ARDS from cardiogenic edema in a critically ill patient.

Yeah. What's the most common trap or pitfall regarding the diagnosis of ARDS?

Yeah. I think it's that one point we keep hammering on, John. It's just not entertaining the diagnosis and forgetting to calculate a PF ratio on every ventilated patient with bilateral infiltrates.

If we build that reflex, we'll stop missing it.

So Dennis, why does missing ARDS matter?

Well, because recognition should trigger proven life-saving lung protective care, low tidal volume, [00:19:00] limitation of inspiratory pressures, appropriate titration of PEEP, and then prone positioning when severe. I think conservative fluid management is probably still part of the treatment.

The mistake is treating fluid balance while ignoring an injurious ventilator strategy.

Moving into the foundational therapy of ARDS, what is the single most important intervention for ARDS?

Yeah. Without a doubt, lung protective ventilation. It's the cornerstone. If we do one thing right in ARDS, it's this, and it's a strong recommendation.

All right. So give me the actual settings, the way you'd set them if you're the attending.

Yeah. So we typically want to use four to eight ccs per kg of predicted body weight, and six is a nice middle number there to start at. Predicted body weight is based on sex and height, not actual weight. So for a patient with a predicted body weight of 70 kilograms, 6 ccs per [00:20:00] kg is about 420 ccs.

If plateau pressure is high, then we'll start to slowly titrate down or work down towards maybe five and then four ccs per kg. If severe dyssynchrony or acidemia, particularly respiratory acidosis, persists despite optimizing the rate and sedation, then we can individualize within the four to eight range while protecting plateau pressure.

Let's highlight the predicted body weight or ideal body weight point.

Yeah, I think because lung size tracks height, not weight. And so a five foot four patient and a six foot two patient at the same 90 kilos have very different lungs. If we're dosing tidal volume off of actual weight, we'll massively over-ventilate the shorter patient, which in a baby lung is exactly the over-distension we're trying to avoid.

So predicted or ideal body weight is calculated from height for that reason.

All right, keep going. [00:21:00] We already talked about our plateau pressure. Let's talk about our pressure ceilings that we're comfortable dealing with.

Yeah. So again, plateau pressure at or below 30 centimeters of water. We're gonna measure that with an inspiratory hold when the patient is passive.

What that also means is that if patients have a high respiratory drive, we're not gonna be able to measure it in the absence of sedation. We're gonna watch driving pressure, often aiming around 15 or less. We're gonna increase respiratory rate as needed up to about 35 in an ARDS-style protocol while checking for auto-PEEP.

I think one thing we don't want to reflexively do is just think if we increase the resp rate at a set tidal volume, that's gonna result in improved CO2s. If we're not giving patients enough time to exhale, they will trap and auto-PEEP, and that certainly can make the situation worse in terms of a respiratory acidosis.

And then when it comes to the PaCO2, [00:22:00] we're not chasing a normal number at the expense of lung protection. ARDSNet recommends targeting a pH of 7.3 to 7.45. And in patients who have really sick lungs, we're gonna be even more tolerant of a severe acidemia. Once we get to below 7.15 though, that kind of forces a bit of a reassessment of our entire strategy.

All right. So yeah, and this isn't just expert opinion either. What's the evidence behind all this? It's pretty well-cited at this point.

Yeah. The ARMA trial, and we talk about landmark trials in trauma and critical care all the time, but this is really one of the key papers that we all need to know. So ARDSNet group, New England Journal of Medicine, 2000, 6 cc's versus 12 cc's per kilogram of predicted body weight.

And what they demonstrated was a 22% relative reduction in mortality. Absolute risk difference of 8.8%, 31 versus [00:23:00] 39.8%. And that's the landmark trial of the whole field.

All right, and so let's talk about this idea of permissive hypercapnia you mentioned. It makes people nervous, right? Especially our nurses and RTs are looking at the numbers when they come back on the ABGs.

How do we reassure everybody?

Yeah. I think the focus here is the principle is permissive hypercapnia, not permissive severe acidemia. And we know the sequelae of severe acidosis in our sick patients. So again, here we're focusing on the patient, not just the single value. But if the pH is drifting down, we can optimize the respiratory rate, think about contributors to dead space, whether or not patients are synchronous with the vent, and any other underlying metabolic problems that might be contributing.

So we're not gonna just automatically increase tidal volume just to make the numbers on our blood gas look better.

Is there any patient populations or specifically surgical patients that we need to worry [00:24:00] about?

Yeah, I think one of the big ones that comes to mind is a patient with intracranial hypertension.

That's sort of the classic example. We don't want their CO2s to be too low, which is gonna result in severe vasoconstriction and potential hypoxemia. But if we let the CO2 drift too high, then they're at risk for cerebral vasodilation. And in a patient with high ICPs, that can be deleterious. Also, severe right ventricular failure and patients with marked pulmonary hypertension can also make profound hypercapnia undesirable.

Those patients still need lung protection, but the gas targets start to become a little bit more individualized.

All right, so continuing our foundational therapy, our part two, PEEP and fluids. PEEP.

Yeah.

What's it doing for you and what's the strategy in ARDS?

Yeah, I think current ATS or American Thoracic Society guidance conditionally favors a higher PEEP strategy [00:25:00] without prolonged recruitment maneuvers, which we used to do quite commonly back when I was a fellow, and that's in moderate to severe ARDS.

If you look at the ARDS Net PEEP/FiO2 tables, I think that's a reasonable and very good starting framework. I know in our units we actually have that table just photocopied and pasted to the ventilator so that it's a good reminder of where we want to start. Then we're gonna individualize using oxygenation, calculate and follow trends in terms of compliance and driving pressure.

Obviously consider hemodynamics and the right ventricular response.

Yeah, so it's not just more is better PEEP. We think of ARDS, we don't just give more PEEP, right?

That's right. And personally, I'm a fan of a higher PEEP strategy. But excessive PEEP can over-distend already open lungs and that can potentially increase dead space.

We've already talked about the impact in terms of reducing venous return and it can also increase the afterload [00:26:00] on the RV. If PEEP rises and driving pressure or compliance improves, that supports recruitment. If mechanics and hemodynamics worsen, we're probably over-distending. So it's a physiologic clue.

It's not a perfect recruitment test.

Yeah, so moving into recruitment maneuvers is like a critical point to understand, right? This is a popular maneuver. Anybody training in the last decade is something they've done in the ICU and it probably needs to be retired.

Yeah, I think it does, John. And I'm also a big fan of these.

But I think the ARD trial showed harm with an aggressive recruitment and titrated PEEP strategy. And we've already referred to the 2024 ATS guidelines, which now strongly recommend against prolonged recruitment maneuvers in moderate to severe ARDS. So higher PEEP without a prolonged recruitment maneuver is the current evidence-based direction.

Yeah, I think it still surprises people who trained a decade ago too. This is more of a modern [00:27:00] push.

Yeah, for sure. And I think the reason, again, comes back to the heterogeneity of ARDS. The pressure required to recruit one region may drastically over-distend another. And that's why these sustained high-pressure maneuvers, 30 for 30 or 40 for 40, can look physiologically attractive and still produce harm.

Yeah. And so let's talk about fluids following PEEP here. So our surgical instinct and the evidence don't always line up. Like how should we manage our fluids in these sick ARDS patients?

Yeah, so I think after initial shock and resuscitation have resolved, we are gonna shift from resuscitation to fluid stewardship.

We're gonna avoid any unnecessary fluid creep, that classic four-two-one maintenance fluid that the vast majority of sick patients do not need. And de-resuscitation is something we're gonna institute early. So when the patient's perfused and once they can tolerate it in our assessment, we're gonna actively [00:28:00] de-resuscitate.

So the practical meaning here or takeaway is we want to be conservative in terms of our fluid strategy.

All right. Like before, what's the trial supporting this?

Yeah. So the FACTT Trial, conservative fluid management really increased ventilator-free days, 14.6 versus 12.1. It also reduced ICU length of stay without an increased risk for renal failure or the need for renal replacement therapy.

And I think the last part matters, and that's where a lot of our earlier fears were, is that if we start to diurese them too early, we might be contributing to AKI or acute kidney injury. And the FACTT Trial says that fear's not really something we need to worry about.

All right. And how would we bring this into our clinical practice?

Yeah. If you look at the FACTT Trial, they use some complex or somewhat complex protocolized hemodynamic targets and held de-resuscitation during shock. I think at the bedside, I wouldn't [00:29:00] teach a universal daily net negative target, but it is so important to pay attention to 24-hour balance and how that fluid is accumulating.

We're gonna make liberal use of diuretics when appropriate, and obviously we have an ongoing reassessment of perfusion and kidney function. I think we're gonna aim to avoid ongoing fluid balance positivity once resuscitation is done.

All right. And with the FACTT Trial, what's, like, our key guardrail that y- everybody should know?

Yeah. I think the issue here, John, is going into the complete opposite direction where instead of being conservative after a really aggressive resuscitation, we now make that patient intravascularly deplete or hypovolemic. So if perfusion's marginal based on clinical assessments and your overall sort of evaluation or the patient's obviously in shock or a fluid-responsive problem is still being treated, we're gonna pause and hold [00:30:00] off on that de-resuscitation.

FACTT improved ventilator-free days, and it doesn't really force us or mandate a negative balance regardless of physiology.

Okay. So moving into part three of our foundational therapy for ARDS, corticosteroids. That is the steroids in ARDS. Where does the guidance actually sit now after much back and forth?

Yeah. So both the 2024 ATS as well as the SCCM guidelines conditionally suggest corticosteroids for adult ARDS. In the past, this is something that's been quite debatable, particularly in terms of the impact of steroids on the development of that fibrotic phase of ARDS. But the evidence here, the certainty is moderate.

And I think the important nuance is that neither guideline endorses one mandatory drug dose or duration. And the fact of the matter is, John, a lot of our patients may have septic shock, and these patients [00:31:00] oftentimes are already on corticosteroid replacement therapy with hydrocortisone. Whether we're using dexamethasone, hydrocortisone, methylprednisolone probably doesn't matter as much.

That's what always comes up on rounds, right? So what's the regimen we're actually talking about? It'll be really clean just to kinda lay that out for everybody.

Yeah, no, fair enough, John. I think we probably have to refer to the DEXA ARDS trial. That gives us one well-studied and accepted regimen, and that's dexamethasone, 20 milligrams IV daily for five days.

Then we're gonna taper down to 10 milligrams daily for five days. But we've got to remember the population in that trial. This is established moderate to severe ARDS that persisted 24 hours after onset, PF 200 or less on a PEEP of at least 10 and FiO2 at least 0.5. So that's trial evidence, not a guideline-mandated recipe for every [00:32:00] ARDS patient.

And how about the timing for steroids, early versus late?

Yeah. Relatively early ARDS. That's not to say that at day 14 it's a universal hard no. But late initiation, I think there is much less certainty. And older persistent ARDS data raises concerns when methylpred was started after two weeks. So the later the disease, probably the more careful we wanna be defining exactly what we're treating.

Yeah. And so where-- what's the population that obviously we gotta be thinking twice before just starting steroids?

Yeah. I think there's a trade-off that we're all familiar with, especially when it comes to things like hyperglycemia or if there's ongoing concerns with infection. And again, I think the combination of steroids with neuromuscular blockade and the concerns with PICS is something to consider as well.

I think in trauma [00:33:00] patients, when we look at a lot of these trials, they're not the dominant population included. So some of the evidence here for our trauma-specific patient populations is a little more indirect

Yeah. Would you change anything? What are your other concerns thinking about trauma ARDS patients?

Yeah, I don't know that I'd create a sort of separate bucket for trauma ARDS. If the patient has ARDS and there's no major patient-specific reason to avoid steroids. And in this particular scenario, I'm thinking of severe TBI, where if you look at the BTF guidelines, if there's one well-supported recommendation, it's that we're not using steroids in severe TBI.

But, um, yeah, I think, uh, generally we're gonna treat trauma ARDS similar to other medical patient populations.

Yeah, that rounds out really the foundational therapy for ARDS. Let's move into when patients [00:34:00] aren't responding to that foundational therapy. Let's go into rescue therapy. So proning and paralysis, part one.

So like I mentioned, the patient is responding to foundational therapy. We want to start rescuing. What's the best evidence for rescue therapies?

Yeah, so prone positioning strongly recommended for severe ARDS for more than 12 hours per day. So they've got a PF under 150, FiO2 of at least .6, PEEP at least five, and we're gonna prone those patients for at least 16 hours a day.

All right. What's the trial?

Yeah. So ProCIVA reported 28-day mortality of 16% with prone positioning, and that's versus 33% supine. Again, this was a highly selected severe ARDS population with early prolonged proning, and they also had really experienced teams and a well-defined protocol. And I think it really demonstrates the importance of having protocolized [00:35:00] therapy and not something that we occasionally do as a rescue maneuver.

Yeah. Why does this actually work? What's the physiology behind it?

Yeah. I think the big thing, it's gonna improve our V/Q matching by redistributing ventilation into lung regions that were dependent and are now non-dependent. It also reduces over-distension of the ventral lung, improves chest wall mechanics, and can also help drain secretions.

But timing matters. We want to start with it early, within 12 to 24 hours of meeting criteria, not as a last-ditch move on day seven or 10

Yeah, and we're not just talking about medical ARDS patients specifically for this, well, this podcast and this segment because we're all surgeons, right? So what are the surgical-specific concerns we have to think about?

Yeah, I think the surgical or trauma patient definitely adds some complications or concerns. One is an unstable spinal injury in which proning may simply be [00:36:00] unsafe. Patients may come out after damage control with an open abdomen or fresh laparotomy incision with a challenging body habitus. There's also the drains, the stomas, external fixators, and sometimes tenuous hemodynamics that do require some planning.

But they're not automatic exclusions. I think the, the major preventable complications are pressure injuries and, of course, the one big thing we're always concerned about in the process of proning is losing the airway, lines, drains, and tubes. So a trained team, there's lots of good videos out there, and a checklist.

Yeah. One thing that always comes up in our ICU, too, is proning a fresh laparotomy in open belly. And that always sends red flags up throughout the unit. But it's not an automatic no, right?

No, definitely not. I think the question is whether the surgical constraint can be managed safely enough to deliver a therapy that has [00:37:00] strong evidence in severe ARDS.

So that decision has to be patient-specific. A lot of it will also do with the team comfort as well as experience, and really has to be made at the bedside with all team members present.

Okay. So moving on to really a next kind of facet of rescue therapy, neuromuscular blockade. Where does it sit? And what's the really the evidence showing for it?

Yeah. So the 2024 ATS guideline conditionally suggest neuromuscular blockade in early severe ARDS. And that typically means within 48 hours and with a PF around 100 or less. The practical use case is a patient with severe dyssynchrony or they've got injurious respiratory effort despite ventilator adjustment and deep or adequate sedation, or a patient who already requires deep sedation for proning.

All right, Dennis, let's talk about the [00:38:00] zebras in ARDS. The myths out there, the traps, the things that harm. All right, so this is a rapid segment. We're gonna go through a lot of different ones. So let's just kinda knock 'em out. You tell me your thoughts. Sure. So inhaled pulmonary vasodilators. Inhaled nitric oxide, inhaled EPO.

What's the verdict on those?

I think when it comes to therapies for ARDS, it's nice to think about them or ask the question, two questions. One, does it improve oxygenation? And two, does it improve survival? When it comes to things like inhaled nitric oxide or inhaled EPO, they can definitely transiently improve oxygenation, and they do that by redistributing pulmonary blood flow towards ventilated lung.

But they haven't shown a mortality benefit. So use them, I think, maybe as a short bridge while we're optimizing proning and PEEP or arranging for potential ECMO evaluation. And then nitric oxide also carries a [00:39:00] theoretical or potential signal for renal harm.

Yep. So just like I say in the ICU, a better number, but not a better outcome.

Yeah, exactly. And, and I think there are times when an improved saturation can be useful. It can buy you time, but it's not a physiologic bridge, and it's definitely not a disease-modifying therapy.

All right, next one. High-frequency oscillatory ventilation. Where are we at?

We used this all the time when I was a fellow, and this was before the widespread availability of extracorporeal life support.

But routine high-frequency oscillatory ventilation, or HFOV, is no longer recommended in adults with moderate to severe ARDS. The pediatric literature may be different. But the earlier ATS/ESICM as well as SCCM guideline recommend against routine use. And I think the, the big trial OSCILLATE showed increased mortality.

So definitely not gonna be part of our usual [00:40:00] rescue ladder, John.

All right, and going back to it one more time, recruitment maneuvers.

So that's a no. I think ARRT showed harm, and current ATS guidelines recommend against them.

All right, and the biggest myth of all of how these patients die

Yeah, it's not that blue number on the screen.

It's not refractory hypoxemia. That is not why patients die. Sepsis, shock, multiple organ dysfunction, those are the major reasons that these patients pass away. And that's why source control, optimization of hemodynamics, and avoidance of VILI really belong in the same conversation, even more so than the SpO2.

All right. We're at the bedside dealing with a really sick ARDS patient. And let's try to incorporate our ECMO, the ladder, and really a stepwise approach to dealing with one of these sick people. First, top of the rescue ladder, VV ECMO. When do we do this?

The guidelines would conditionally suggest VV ECMO in selected patients with severe [00:41:00] ARDS, and we have a whole topic dedicated to this in the Critical Care series.

I think the EOLIA trial gives us really useful escalation thresholds. So a PF less than fifty for more than three hours, below eighty for more than six hours, or a pH below 7.25 with a PaCO2 of at least sixty for more than six hours despite optimized conventional ventilation. I'm not at an ECMO center so I kinda tend to use those as referral triggers.

I don't think they're universal automatic cannulation rules.

Yeah. It's impossible to talk about ARDS without really going into ECMO. But I think we'll reiterate some of the stuff from the VV ECMO critical care topic as well. But the evidence for ARDS, it is contested a bit, or am I wrong?

No, absolutely. I think the EOLIA trial itself didn't meet conventional statistical significance for its primary mortality endpoint, and in [00:42:00] part because there was substantial crossover that occurred from controls to ECMO.

In an individual patient data meta-analysis of the CESAR trial, an EOLIA ECMO was favored in terms of ninety-day mortality. There is a bit of a trade-off, though, major bleeding and, of course, resource intensity of ECMO, which matters even more in post-op as well as trauma patients

Yeah, that being said, in ARDS, when you're at your wit's end of all the things you've tried in your rescue therapies, the first call you're thinking, or at least the first thing you're thinking about is, how can I get these people to ECMO or an ECMO center, right?

But you have to find the right candidate. So who actually benefits?

So I think the best candidate, number one, has a potentially reversible cause, very severe gas exchange failure despite optimization of evidence-based conventional care, and they also have to have a reasonable likelihood of recovery.

Longer duration of high-intensity mechanical ventilation, [00:43:00] the more number of organs that are failing, and major bleeding risk, I think all worsen candidacy, but they're not captured by one universal bedside cutoff. So ECMO's not a shortcut past lung protection and proning, John.

Yeah. And Dennis would probably give perspective on this coming from a non-ECMO center, but what's, like, every fellow that's dealing with ARDS in the ICU, what's the ECMO logistics that you have to overcome?

So if you're somewhere like I am where you can't provide ECMO, we're gonna make that call early, John, and understand what our regional transfer agreements and criteria are. I think we want to do this when the trajectory suggests the patient may meet criteria. I think referral is easier before refractory shock, transport risk, and prolonged injurious ventilation narrow down the options.

Then going back to the surgical ICU, are, does trauma and recent surgery [00:44:00] exclude somebody for VV ECMO?

Not automatic exclusions. Certainly, there is a bleeding risk, and we have to think a little bit more about the anticoagulation strategy, but definitely not an absolute contraindication.

All right, so w- now we're putting it, the whole thing back together as an algorithm.

We gave the overview of ECMO. We have the deteriorating ARDS patient. Let's go step-by-step of how we're gonna do it.

So step one, we're gonna confirm the diagnosis, and then we're gonna identify the potential trigger or etiology. Number two, we're gonna make the ventilator truly lung protective. We're gonna start at 6cc per kg ideal body weight.

We're gonna target a plateau at or below 30 and monitor driving pressure. Three, we're gonna use an appropriate PEEP strategy. That can be a low or a high strategy, and we're also gonna remember to be conservative in terms of our fluid management once the shock is resolved. Four, we're considering [00:45:00] corticosteroids, recognizing that recommendation is conditional and the regimen isn't one size fits all.

Five, we're gonna prone early when severe, and the perceiver-type threshold is a PF under 150 with substantial O2 requirements. Six, consider short-course neuromuscular blockade for early severe ARDS, especially if patients are dyssynchronous or have obvious injurious ventilation. And then seven, for persistent life-threatening gas exchange failure, we're gonna use an inhaled vasodilator only as a bridge and then contact an ECMO center early.

Yeah, and just like our trauma algorithms, every time we're potentially escalating care in ARDS, what are we rechecking?

Yeah, the basics, tube position and patency, secretions, ruling out complications of barotrauma like an unrecognized pneumothorax. Check the [00:46:00] circuit, look for auto-PEEP, check your plateau pressures, make sure we're sticking with the ARDSNet protocol.

Figure out when patients aren't adequately sedated or receiving adequate analgesia manifested as dyssynchrony. Look at their fluid balance and obviously the impact of our strategies on their hemodynamics in the right ventricle. A patient can look refractory because one of these fundamentals have been overlooked, John.

Yeah. Going back to the surgery side, too, it's easy to get pigeonholed. Bed 13 has ARDS. We're sur- surgeons as intensivists. We're co-managing both pathologies. What don't we miss?

Yeah. I think the big thing comes down to source control, which hopefully as surgical intensivists, that's something that always comes to mind.

Patients may have an anastomotic leak. They may have developed ischemic bowel or worsening sources of sepsis, and so these sorts of things, if not recognized and [00:47:00] treated, are gonna keep that inflammatory insult going. So lung protective strategy is great for ARDS but does nothing for source control.

All right, so we've gotten down through the weeds of ARDS.

We're managing these patients. Some take five days, some take multiple weeks, especially if you're looking at a couple weeks of ECMO run. Let's talk about some of the complications of ARDS. Because of being some of the critically ill patients in the hospital, they do show up. So let's talk about the acute complications first.

What do we have?

Yeah. I think we've already mentioned barotrauma, so we can skip over that one. Certainly patients, their ARDS may have started with a pneumonia, and certainly they're gonna be at risk for hospital-acquired or ventilator-acquired pneumonia as well. We've talked about the widespread systemic effects of ARDS.

And so, not surprisingly, a lot of these patients may develop an AKI. But I think the one thing that we always wanna think about is the impact on [00:48:00] hemodynamics because the physiologic link between our right ventricle and lung injury really impacts our ventilator choices. And so hypoxemia, hypercapnia, high plateau pressures, and excessive PEEP can all increase RV afterload and RV load in general.

So if the patient becomes hypotensive, we wanna make liberal use of echocardiography and then reassess not just our fluid status, but our ventilator strategy as well.

All right. Yeah. So the RV is super important in ARDS, and we have to always kinda keep an eye on it throughout just doing bedside maneuvers or routine echocardiograms.

Going back to complications, so some of these complications also declare late, and what isn't-- what makes us concerned during all this?

Yeah. I think if a ventilated ARDS patient suddenly desaturates or becomes hypotensive, we're not just thinking that the ARDS [00:49:00] is worse. And again, we're surgeons, so we're good at running things like mechanical differentials, and we've alluded to tube displacement or obstruction, pneumothorax, issues with the circuit.

I think the slower failure to improve should trigger a fresh search for infection, overzealous fluid accumulation. Potentially, these are patients who are moving into that fibrotic phase. But other things like PE and, again, uncontrolled surgical sources need to be thought about

Yeah, and going forward, we got this ARDS patient maybe extubated at this point.

What's the next few weeks after that extubation look like?

Think with the increasing recognition of the post-intensive care syndrome and PICS, these patients can have long protracted post-ICU recovery. Weakness, reduced exercise capacity, cognitive impairment, mood disorder, anxiety, depression. We're recognizing that PTSD is not uncommon in survivors of the ICU, bundled with [00:50:00] voice or airway problems after prolonged intubation.

So lung function often improves substantially over the first year, but functional recovery may lag seriously behind.

All right, so we've gotten through our stepwise approach for ARDS. Let's talk about some special populations and de-escalation specifically. So if we're coming down off the ladder, we have a patient that's now improving.

As ARDS resolves, how do we really de-escalate?

As oxygenation and lung compliance improves, we're gonna reduce our FiO2, and we should be thinking about doing that early and aggressively, as well as PEEP in a controlled way. We're gonna lighten sedation. We definitely want to be stopping neuromuscular blockade as early as possible and then start, when appropriate, our daily liberation assessment.

We're gonna continue lung protective tidal volumes during recovery and then improving [00:51:00] oxygenation is not a reason to jump back to large, potentially injurious breaths.

Yeah, and the fluid piece after, as we're de-escalating?

Yeah. Again, this is where we're definitely gonna be focusing on de-resuscitation.

Once perfusion's secure, reducing excess fluid can certainly improve lung mechanics and help with our efforts to liberate that patient from the ventilator. So here we're using physiology, not just fixed daily negative balance prescriptions. But I do find it helpful to set some targets, one, two liters per day, and then checking in a couple times throughout the day to make sure that we're meeting those targets.

Yeah, and all right, Dennis, so we got through a pretty good review of ARDS there. Let's go through some quick hits, the things you must know, and these are the things you might get on rounds or on the board exam. So landmark trial of the field in ARDS.

Yeah, ARDSNet or the ARMA trial, 6 cc per kg ideal [00:52:00] body weight, plateau pressures under 30, mortality benefit.

The ventilatory variable that is most tied to survival.

Yeah, this one's driving pressure and lower is generally better. Around 15 centimeters of water or less is useful as a pragmatic goal, but it's not a validated treatment threshold.

Why do normal tidal volumes injure our ARDS lung?

Yeah, that's the baby lung.

Over-distension of the small aerated compartment.

Prone positioning, trial and dose.

ProCIVA, early, at least 16 hours a day for patients with a PF under 150. All

right. Conservative fluids, trial and the payoff from that.

That's the fact trial. More ventilator-free days, no more renal failure.

All right. Steroid regimen for ARDS patients.

Yeah, conditionally recommended. DEXA-ARDS used dexamethasone, 20 milligrams for five days, then 10 [00:53:00] milligrams for five days in established moderate to severe ARDS. Guidelines don't mandate that exact regimen.

All right, we've hit this one a few times. Recruitment maneuvers.

No prolonged recruitment maneuvers.

All right. High-frequency oscillatory ventilation.

Nope. Oscillate showed harm.

Preferred paralytic in ARDS and why?

Yeah. This one, I think we're all familiar with rocuronium, but generally it's gonna be cisatracurium. That's been used in the vast majority of major trials. It's got organ-independent Hoffman elimination.

So we want to keep paralysis selective and generally no longer than 48 hours.

All right, the ECMO evidence that we're gonna quote in ARDS in one line. Yeah.

CESAR and EOLIA meta-analysis. Lower 90-day mortality, relative risk of .75.

All right, we know we use permissive hypercapnia. What's an acceptable pH in ARDS?

Yeah, no single universal [00:54:00] pH cutoff. ARDSNet targeted 7.3 to 7.45, increased rate up to 35 first, and treated pH below 7.15 as severe acidemia requiring escalation.

What is new in that new global definition?

Yeah. It includes high-flow nasal oxygenation at 30 liters per minute or more, permits calculation of an SF ratio less than or equal to 315 when the SpO2 is less than or equal to 97%, and accepts lung ultrasound as an imaging modality.

All right. What do most ARDS patients die from?

Yeah. Multi-organ failure, not refractory hypoxemia.

And then finally, our reflex on every ventilated patient with bilateral infiltrates.

Yeah. We're gonna calculate the PF ratio, then apply the full ARDS definition rather than diagnosing from the ratio alone.

All right, Dennis, let's wrap this [00:55:00] up. Let's bring it home. Three things that you want everybody who listened to this podcast to take away.

Yeah. One, recognize ARDS early and protect the lung. Two, escalate in evidence-based order, appropriate PEEP, conservative fluids, conditional steroids, early prolonged proning for severe ARDS, and selective short courses of neuromuscular blockade.

Always consider early ECMO referral for truly refractory disease. We're gonna avoid prolonged recruitment maneuvers and routine HFOV. And then three, we're gonna treat the cause as well as the whole patient. So although we might become fixated on the ERDS, we always want to be thinking about source control, infection treatment, consider the impact on hemodynamics and the RV.

And then of course, rehabilitation matters just as much as the O2 sat.

All right. And also as surgical intensivists, other things we must consider.

[00:56:00] Yeah. The, the ventilator buys time while the lungs heal and do their thing. So it doesn't fix an anastomotic leak, an undrained collection, ischemic bowel. We're gonna recheck our diagnosis, recheck the fundamentals, and keep asking whether there's a surgical driver that still needs identification and control.

All right, that wraps up ARDS and our surgical critical care curriculum with Behind the Knife. The key trials and the current guidelines are in the show notes. Stay sharp and we'll see you next time