COVID-19: Two Different Lung Pathologies, Two Different Ventilator Strategies

Barbara A. McLean, MN, RN, CCRN


April 13, 2020

Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

This transcript has been edited for clarity.

Good day. I'm Barbara McLean. I'm a clinical nurse specialist in critical care at Grady Health System in Atlanta, Georgia. I'm also a hemodynamic and oxygenation specialist. Today I have the pleasure of discussing some of the new strategic methodologies that are being recommended for treatment of patients in COVID-19 intensive care units with mechanical ventilation and proning. The subject is receiving a lot of attention across the critical care community and was inspired by an editorial by noted pulmonologist Dr Luciano Gattinoni. Like Gattinoni, I believe that our practice should be based on an understanding of pathophysiology in designing interventions to be used at the bedside.

Basically, Gattinoni has proposed that there are two subsets of patients with COVID-19 lung disease. One subset of patients has a loss of compliance and will be responsive to primary acute respiratory distress syndrome (ARDS) Clinical Network Mechanical Ventilation Protocol (with ARDS categorized by the most recent Berlin definition) for resuscitation, alveolar recruitment, and improvement in gas exchange. That means a step-up in fraction of inspired oxygen (FiO2), followed by a sideways step in positive end-expiratory pressure (PEEP) and initiation of early proning. Most of us around the country have discovered that we have a group of patients who are very responsive to this strategy.

A second group of patients, however, do not respond. Gattinoni's research suggests that there may be two specific phenotypes in the evolution of acute respiratory failure in the COVID-19 patient population. Gattinoni's sentinel research examining physiology of ARDS has been profound and has had a major effect on what we do for ARDS patients. It is important now that we understand his most recent research and apply it in our clinical approach to COVID-19 patients.

I'd like to share what we are seeing in my institution and what our provider teams are doing for these patients in order to, hopefully, improve outcome.

COVID-19 Lung Disease: Two Very Different Phenotypes

Patients who do not respond to nonventilation strategies. In this population of COVID-19 patients, we have one subset that has large lung volumes and evidence that the lungs are compliant and elastic, or at least that they have some modified elasticity. These patients may have a very severe hypoxemic hypoxia. They have been likened in presentation—though not in physiology, which is a very important concept to understand—to patients who have high-altitude pulmonary edema. In that subset of patients, one of our considerations will be to determine whether there is pulmonary edema, interstitial damage, and loss of compliance. In that scenario, first of all, we limit fluid resuscitation and apply alveolar recruitment strategies: higher PEEP, airway pressure release ventilation (APRV), ventilation, or high-frequency oscillatory ventilation. Extracorporeal membrane oxygenation (ECMO) should also be considered early on if you have the ability; in conjunction with that, we would prone-position.

My institution does not have capability for ECMO at this time, but with the input of our excellent respiratory therapy department, we utilize strategies for alveolar recruitment. So when we see that a patient is refractory to oxygen delivered via nasal cannula, even up to 7 L, we generally move to a non-rebreather mask. If the patient is still refractory with hypoxemic hypoxia, we add strategies to recruit the alveoli. In my experience, these patients typically have some heart rate variability, although some have noted that these patients may not have tachycardia. That has not been my experience.

It is important to note that, as we implement these alveolar recruitment strategies, we further obstruct right ventricular ejection. We actually put the patients into what I would call a nosocomial or iatrogenic or idiopathic state of right ventricular failure. When we see that a patient is failing to respond to nasal cannula, including non-rebreather, we quickly move to intubation and early ventilation strategies. Most of the time, we just start at or rapidly move to a peak of 14-cm PEEP or above along with high-dose FiO2, skipping some of that stepwise progression that is part of the ARDSNet protocol. We also consider the probability, not just the possibility, that we will need to add some form of pulmonary vasodilation and as pure an inotrope as possible. We consider epoprostenol (Flolan) and nitric oxide. We also consider milrinone primarily as a vasodilator and not necessarily as an inotrope, in which case we would still have to consider some very low-dose inotrope if necessary. Of course, we add prone positioning if we are able.

That strategy fits one of the subsets of patients that Gattinoni talks about.

Patients who do respond to nonventilation strategies. The other subset of patients are those who have a primary hypoxia that is responsive to oxygen therapy. In that situation, we might progress from nasal cannula to high-flow oxygen. We might consider continuous positive airway pressure (CPAP) or bilevel positive airway pressure (BiPAP). These patients actually can respond quite beautifully to noninvasive strategies. In the world of COVID, there have been some significant concerns regarding aerosolization of particles with all but the nasal cannula in these instances.

I think this is one of the most essential points in evaluating our current population of patients. We really do see two different phenotypes.

One has loss of alveolar compliance and recruitment, and therefore has a ventilation-perfusion mismatch because they have a failure of alveolar distension.

The second phenotype manifests with a loss of hypoxic vasoconstriction, which occurs at a microvascular level. Because these patients have lost hypoxic vasoconstriction, they have a high blood flow through their lung field. So, again, you see a ventilation-perfusion mismatch. But in this situation, it is not necessarily from a failure of ventilation but from an abundance of perfusion.

Clearly, we should handle these patients quite differently.

A New Paradigm for Critically Ill COVID-19 Patients

Currently, in my institution, we begin treatment with nasal cannula and then, as necessary, move to non-rebreather mask, early intubation, alveolar recruitment, high FiO2, and proning.

In my interpretation of the current data, I think it is clear that no one is saying that the ARDS protocol is not correct for some subset of patients. Rather, the current data illustrate that protocols are not necessarily designed based on an individual patient's physiology or pathophysiology. Instead, they are designed to help us treat the majority, or the mean, of the patients who present with a problem. And in COVID-19 patients who may quickly develop profound hypoxia, we don't want to apply protocols blindly and assume that a failure to respond is the result of the patient. Instead, when our protocol methodology has failed, we need to step back and ask if we are using the wrong strategy for this patient's phenotype.

I think that is the most important lesson from Gattinoni's research. I've been in practice for 45 years; I appreciate the evolution that critical care has made during that time, progressing from a mindset of trying everything to instead developing strategies that work well for the majority of patients. We have transitioned from a time when we used subjective data, not really evidence, in patient care with a lot of give-and-take. Now in critical care, we have really strong, evidence-based strategies that we apply almost religiously to all patients. And in general, that works.

However, the current population of COVID-19 patients has forced us to rethink that protocolized approach.

Gattinoni is telling us that we need to determine what the specific problem is in these COVID-19 patients. Early recognition of what is occurring in an individual patient may prevent their evolution into the second phenotype, what he has termed a type H phenotype, fitting criteria for severe ARDS and very resistant to therapy.

The first presentation in COVID-19 patients, which Gattinoni has termed type L, may be the only presentation that a patient ever has if we are aware and apply therapies correctly. In these patients, those therapies include larger volumes and higher FiO2 than we are used to giving, allowing patients strategically to mobilize their volumes and then reproducing those volumes if we are giving them therapy. We may then be able to prevent the evolution from type L (in which a patient has compliant lungs and moves volumes) hypoxia to type H noncompliant lungs resistant to therapy.

Patient Assessment

In the United States we don't, in general, have the capability of looking at esophageal pressures as a reflection of airway pressures and a means to evaluate thoracic pressure and intrathoracic strain, which is a tool Gattinoni used in his research. What we do have, however, is the capability of looking at dynamic and static compliance, which is the pressure generated in the delivery of tidal volume and the response of the lung to captured volume (closing the exhalation valve). Those calculations and measures give us substantial information about how we should be supporting our patients.

If we embrace this conceptualization of two phenotypes, then our thinking must evolve. Most of us are ready to embrace a new idea because we have seen so much failure with ARDS protocolization in the COVID patient. Early recognition that the patient does have a compliant lung, that they are able to move tidal volume, that we might need to consider larger tidal volumes for those patients, will allow us to move away from a rigid application of a very well-defined and excellent protocol from ARDSNet that absolutely has a place in the patient with noncompliant lungs.

Really, the differentiation for us is: Does the patient have a compliant lung or not? If the patient does have a compliant lung, we may need to consider the theory that the hypoxemia is related to alterations in blood flow dynamic and consider ways to better oxygenate blood flow with minimal damage to the lungs.

Optimization of blood flow in this current environment has been very difficult. Initially, we were treating patients as if they were septic and doing aggressive fluid resuscitation. Then there was discussion about dehydrating the patients so as to not exacerbate interstitial edema. That created an additional blood flow problem, making it difficult to find equipoise and euvolemia, defined as volume sufficient to allow for organ perfusion, but not enough to cause profound pulmonary edema.

Applying these new strategies, it becomes important to diagnose where a patient is on the continuum of COVID-19 lung dysfunction, recognizing that there is more than one pathway and that not all patients will progress straight into ARDS. I think that is what Gattinoni is telling us.

I want us to be able to apply this new information simply and not necessarily introduce new methodologies and strategies at such a time of crisis when our healthcare staff is already so overburdened with trying to prevent death every single day. Applying this new strategy requires that we assess where on the ARDS continuum a patient is and correlate that with blood flow dynamics.

Most patients who are admitted to an intensive care unit, especially those with hypoxemia, will receive an arterial line and a central line. I want to remind us all to look at central venous pressure—not as a volume indicator, but as a right heart compliance indicator and systolic blood pressure as an expression of left ventricular ejection. If you give your patient fluid and the central venous pressure rises, but the systolic pressure does not, the patient clearly has right ventricular dysfunction. If you give fluids to a patient and the central venous pressure stays the same, or perhaps goes up or down just a little, and systolic pressure goes up, then you have done the right thing.

We have to combine strategies of blood flow equipoise to pulmonary management criteria and utilize alveolar recruitment strategies that use flow to produce pressure. That, of course, is what PEEP, APRV, and pressure control inverse ratio do. I always want to remind myself to consider the effect on the perfusion component of ventilation perfusion. Again, Gattinoni reminds us that we have to consider the ventilation perfusion ratio.

Don't Forget Hyperthermia

Last but not least, I would like to add my own particular concern in this situation, and that is that many of these patients exhibit effervescent hyperthermia—meaning, a temperature that rises and does not respond, or only poorly responds, to acetaminophen. This also puts an extraordinary burden on blood flow, myocardium, and pulmonary, or respiratory, work. It's an incredible burden to be hyperthermic. One of the things that I really have tried to integrate into our daily care—admittedly, without 100% effectiveness—is use of a cooling blanket. I'm not talking about therapeutic temperature management but rather simply cooling the patient. I put that blanket on top of the patient, especially in patients who are awake, so that the patient can move the blanket off themselves if they feel they've gotten too cool.

I can tell you that in every one of my patients for whom I have used a cooling blanket strategy, I actually do see that the heart rate comes down. In patients with tachycardia and tachypnea, use of the cooling blanket actually improves the data that I'm monitoring.

Looking Ahead

It will be a long time before we know whether anything we do has actually improved outcomes. We will need to create an abundance of data to determine whether the things we have tried have improved survival with reduced morbidity and a reduction in the very significant frequency of death in patients with COVID viral disease who have been intubated in our ICUs.

That moral and ethical challenge is accompanied by the intellectual challenge of understanding physiology and applying protocols in an appropriate manner.

We will need to accept that sometimes, no matter what we've done, we cannot overcome this dark, dangerous, ever-changing enemy. We also have to keep changing. We can't stay static with our protocols. I think that's really the message from Gattinoni.

Thank you so very much.

Barbara McLean, MN, RN, CCRN, is a nurse intensivist and clinical specialist in the Division of Critical Care at the Grady Health System in Atlanta, Georgia. She has published and lectured extensively, both nationally and internationally, on a range of critical care topics, many of which can be accessed online.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.