Clinicians treating hospitalized COVID-19 patients have remarked about how it differs from typical acute respiratory distress syndrome (ARDS), sparking debate about the best course of treatment.
Driving their concerns is the fact that many of these patients have severe hypoxemia, but their respiratory compliance -- the ability of the lung to expand when taking in air -- frequently remains normal.
"This remarkable combination is almost never seen in ARDS," Luciano Gattinoni, MD, of the University of Gottingen in Germany, who has emerged as a , wrote in an .
"It's different with this disease," said Sapna Kudchadkar, MD, PhD, a pediatric critical care physician and anesthesiologist at Johns Hopkins. "You can have low saturation without increased work of breathing. That's not the usual with ARDS. With ARDS, when you're hypoxic, you're working hard to breathe, and it's clear you need to be intubated."
"People are trying to make sense of what they're seeing," said pulmonologist Andrew Luks, MD, of the University of Washington in Seattle, which saw the earliest outbreak of COVID-19 in the U.S. "There's not a lot of data to guide what to do [in this situation]. We have plenty of data from good studies over many years about how to manage ARDS. The question is how well our standard practices apply in ARDS due to COVID-19."
Gattinoni and colleagues have published several commentaries, including an analysis of 150 patients in Northern Italy, describing their observations. Most recently, they've proposed that there are : one that looks less like ARDS, and one that's .
As a result, difficult questions have been raised: for instance, is intubation the best treatment for all COVID-19 patients with hypoxemia? It's also prompted some speculation about treatments that could lead down dangerous roads, some experts say.
ARDS in the Strictest Sense
Technically, most COVID-19 patients meet the Berlin definition for ARDS, says Hooman Poor, MD, a pulmonologist at Mount Sinai Hospital in New York City.
They have an acute syndrome, with bilateral opacities on lung imaging, gas exchange abnormalities, and their disease can't be explained by heart failure or volume overload, Poor said.
"Lung compliance isn't part of the definition," he told 鶹ý. "By the strictest sense, these patients meet ARDS criteria."
By comparison, hospitalized patients with severe pneumonia from influenza, for instance, don't have preserved compliance: "Influenza is a classic example. It causes ARDS. The worse the degree of gas exchange abnormalities, the worse the compliance."
There have also been higher rates of mortality than expected among COVID-19 patients on ventilators, prompting some to ask whether intubation protocols for ARDS patients are the right ones to be following for these patients.
Cameron Kyle-Sidell, MD, an emergency and critical care physician in New York City, posted a that went viral, calling for an urgent re-examination of ventilator protocols for COVID-19 patients.
Kyle-Sidell questioned whether the high pressures forced into patients' lungs by the ventilators may actually be doing damage.
"COVID patients need oxygen, they don't need pressure," Kyle-Sidell says in . In , he asks whether the "ARDS we see after they're on the ventilator -- not before -- is due to the natural evolution of the disease, or [ventilator-induced] lung injury?"
Some physicians have started to use other methods of providing oxygen, including non-rebreather masks, high-flow nasal cannula, and continuous positive airway pressure (CPAP), to stave off the need for ventilation. Proning patients earlier in their disease course has been another part of that strategy.
Gattinoni and colleagues have made similar recommendations, based on their observations of what they describe as , Type L and Type H (or, ). Patients typically present with Type L disease, which is characterized by normal lung compliance and gas volume in the presence of hypoxemia. These patients may improve, or they may worsen, the researchers wrote.
About 20% to 30% of patients had or evolved to Type H disease, characterized by worse lung compliance and increased edema and lung weight, they wrote.
"The transition from Type L to Type H may be due to the evolution of the COVID-19 pneumonia on one hand and the injury attributable to high-stress ventilation on the other," they wrote.
Type L patients can still be placed on the ventilator but with higher tidal volumes with lower positive end expiratory pressure (PEEP), while Type H patients "should be treated as severe ARDS, including higher PEEP, if compatible with hemodynamics, prone positioning and extracorporeal support."
Luks said the Gattinoni paper "highlights the fact that there is not one single presentation of this disease, so a one-size-fits-all approach is not going to work. There are people with high compliance, people with low compliance, so you have to adjust your therapies accordingly."
"In the end, some people are going to require invasive mechanical ventilation as their primary form of respiratory support to have the chance to survive," Luks said. "There are other patients who we can get them through the course of their care without a ventilator, using continuous positive airway pressure (CPAP), high-flow nasal cannula, and other non-invasive forms of support."
Dangerous Ideas?
While calling Gattinoni's comments reasonable, Luks said about other ideas that may appear to be valid but aren't vetted, and that get floated quickly with the amplification of social media.
One example, he says, involves discussion about how COVID-19 pneumonia looks more like high altitude pulmonary edema (HAPE). Luks has spent several weeks taking care of patients with COVID-19 at Harborview Medical Center in Seattle, but one of his long-standing academic interests has been high-altitude medicine and physiology.
that was particularly concerning, he said, discussed the use of medications used to treat HAPE that could be damaging to COVID-19 patients.
"Using acetazolamide or nifedipine for COVID patients is just wrong," Luks said. "First, while there are some similarities between HAPE and COVID ... the mechanisms by which these diseases develop are fundamentally different from each other."
"Nifedipine is used to treat HAPE because it lowers the blood pressure in the lungs, which is responsible for fluid build-up in the lungs," he said. "If you give that medication to someone with lung injury due to COVID-19, that's going to make their oxygenation worse. It's going to mess up the balance between ventilation and perfusion in the lungs."
Acetazolamide can also interfere with ventilation/perfusion matching as well as harming the diaphragm and affecting blood transportation of carbon dioxide, he added.
"We don't want people looking at these diseases as the same thing," Luks said. "We don't want them to be using these medications in the treatment of COVID-19."
Role of Endothelial Damage
But experts are still taking a hard look at the role that endothelial damage plays in COVID-19.
"Questions have been raised as to whether this is an early endothelial injury that causes thrombosis and later gets complicated by ARDS," Poor told 鶹ý. "If it's a driver, then perhaps we need treatment geared toward that: anticoagulation earlier in the disease, or thrombolysis for more severe disease may be beneficial."
Poor shared a case series he submitted to a preprint server of five COVID-19 patients at Mount Sinai who were in respiratory failure and shock, and had evidence that parts of their lung were being ventilated but not perfused. Some had been on anticoagulants but despite that were doing poorly.
So he and his team gave those patients systemic thrombolysis with tissue plasminogen activator (tPA) and all cases saw "dramatic physiologic improvement immediately," though not all survived.
"None of this is even remotely proven, but it does raise that question," Poor said. "Until trials are done to demonstrate whether [anticoagulation and thrombolysis] are beneficial, there needs to be lots of caution."
Critical care and pulmonary specialists agreed that clotting is prevalent in COVID-19 patients. The vast majority have elevated D-dimer, though it's unclear if D-dimer is a true marker of thrombosis in such an inflammatory setting, he noted.
Still, "our clinical observations are that extracorporeal membrane oxygenation (ECMO) circuits are clotting more, that dialysis circuits are clotting more," said Luks. "We can't explain why they're hypoxemic without a change in chest radiograph. We have seen overt pulmonary embolism or microemboli that are developing in the lungs. There's a plausible argument that maybe this is related to a pro-coagulant state."
He cautioned that studies of ARDS patients before COVID-19 do show documentation of microthrombi on autopsy, and critically ill patients do develop venous thromboembolism and pulmonary embolism and have had other issues like clotted dialysis circuits.
"How much is this what we've seen in the past, but we're seeing more now because we have more patients intubated, and how much is this truly unique to COVID?" Luks asked. "Time is going to tell as more observations and more data come out."
Hospitals treating COVID-19 patients are moving fast on providing some anticoagulant guidance even in the absence of hard evidence, as previously reported by 鶹ý. That includes moving from prophylactic to therapeutic doses of anticoagulants, as some patients have developed clots despite being on preventive doses.
The exact mechanism by which SARS-CoV-2 may cause endothelial damage isn't clear, but it does enter cells via the angiotensin converting enzyme 2 (ACE2) receptor, which is most commonly found in the alveolar epithelial cells, followed by endothelial cells. Bin Cao, MD, of the National Clinical Research Center for Respiratory Diseases in Beijing, who had proposed the mechanism that endothelial damage could lead to clotting, reported that when the virus binds to these cells, it may damage the blood vessel, especially the microcirculation of small blood vessels, and thus spur platelet aggregation.
As answers are being teased out, experts have urged caution about making treatment decisions without substantial evidence.
"The last thing you want to do," Poor said, "is administer a therapy that's not proven and cause a horrific complication when the patient would have improved on their own or with other therapies."