Physicians in the COVID-19 trenches are beginning to question whether standard respiratory therapy protocols for acute respiratory distress syndrome (ARDS) are the best approach for treating patients with COVID-19 pneumonia.
At issue is the standard use of ventilators for a virus whose presentation has not followed the standard for ARDS, but is looking more like high-altitude pulmonary edema (HAPE) in some patients.
In apublished in the American Journal of Respiratory and Critical Care Medicine on March 30, and in an accepted for publication in Intensive Care Medicine, , of the Medical University of Göttingen in Germany and colleagues make the case that protocol-driven ventilator use for patients with COVID-19 could be doing more harm than good.
Dr. Gattinoni noted that COVID-19 patients in ICUs in northern Italy had an atypical ARDS presentation with severe hypoxemia and well-preserved lung gas volume. He and colleagues suggested that instead of high positive end-expiratory pressure (), physicians should consider the lowest possible PEEP and gentle ventilation–practicing patience to “buy time with minimum additional damage.”
Similar observations were made by, a critical care physician working in New York City, who has been speaking out about this issue on Twitter and who shared his own experiences in this with WebMD chief medical officer .
The bottom line, as Dr. Kyle-Sidell and Dr. Gattinoni agree, is that protocol-driven ventilator use may be causing lung injury in COVID-19 patients.
Consider disease phenotype
In the editorial, Dr. Gattinoni and colleagues explained further that ventilator settings should be based on physiological findings – with different respiratory treatment based on disease phenotype rather than using standard protocols.
‘“This, of course, is a conceptual model, but based on the observations we have this far, I don’t know of any model which is better,” he said in an interview.
Anecdotal evidence has increasingly demonstrated that this proposed physiological approach is associated with much lower mortality rates among COVID-19 patients, he said.
While not willing to name the hospitals at this time, he said that one center in Europe has had a 0% mortality rate among COVID-19 patients in the ICU when using this approach, compared with a 60% mortality rate at a nearby hospital using a protocol-driven approach.
In his editorial, Dr. Gattinoni disputed the recently publishedfrom the Surviving Sepsis Campaign that “mechanically ventilated patients with COVID-19 should be managed similarly to other patients with acute respiratory failure in the ICU.”
“Yet, COVID-19 pneumonia, despite falling in most of the circumstances under the Berlin definition of ARDS, is a specific disease, whose distinctive features are severe hypoxemia often associated with near normal respiratory system compliance,” Dr. Gattinoni and colleagues wrote, noting that this was true for more than half of the 150 patients he and his colleagues had assessed, and that several other colleagues in northern Italy reported similar findings. “This remarkable combination is almost never seen in severe ARDS.”
Dr. Gattinoni and colleagues hypothesized that COVID-19 patterns at patient presentation depend on interaction between three sets of factors: 1) disease severity, host response, physiological reserve and comorbidities; 2) ventilatory responsiveness of the patient to hypoxemia; and 3) time elapsed between disease onset and hospitalization.
They identified two primary phenotypes based on the interaction of these factors: Type L, characterized by low elastance, low ventilator perfusion ratio, low lung weight, and low recruitability; and Type H, characterized by high elastance, high right-to-left shunt, high lung weight, and high recruitability.
“Given this conceptual model, it follows that the respiratory treatment offered to Type L and Type H patients must be different,” Dr. Gattinoni said.
Patients may transition between phenotypes as their disease evolves. “If you start with the wrong protocol, at the end they become similar,” he said.
Rather, it is important to identify the phenotype at presentation to understand the pathophysiology and treat accordingly, he advised.
The phenotypes are best identified by CT scan, but signs implicit in each of the phenotypes, including respiratory system elastance and recruitability, can be used as surrogates if CT is unavailable, he noted.
“This is a kind of disease in which you don’t have to follow the protocol – you have to follow the physiology,” he said. “Unfortunately, many, many doctors around the world cannot think outside the protocol.”
In his interview with Dr. Whyte, Dr. Kyle-Sidell stressed that doctors must begin to consider other approaches. “We are desperate now, in the sense that everything we are doing does not seem to be working,” Dr. Kyle-Sidell said, noting that the first step toward improving outcomes is admitting that “this is something new.”
“I think it all starts from there, and I think we have the kind of scientific technology and the human capital in this country to solve this or at least have a very good shot at it,” he said.
Proposed treatment model
Dr. Gattinoni and his colleagues offered a proposed treatment model based on their conceptualization:
- Reverse hypoxemia through an increase in FiO2 to a level at which the Type L patient responds well, particularly for Type L patients who are not experiencing dyspnea.
- In Type L patients with dyspnea, try noninvasive options such as high-flow nasal cannula, continuous positive airway pressure, or noninvasive ventilation, and be sure to measure inspiratory esophageal pressure using esophageal manometry or surrogate measures. In intubated patients, determine P0.1 and P occlusion. High PEEP may decrease pleural pressure swings “and stop the vicious cycle that exacerbates lung injury,” but may be associated with high failure rates and delayed intubation.
- Intubate as soon as possible for esophageal pressure swings that increase from 5-10 cm H2O to above 15 cm H2O, which marks a transition from Type L to Type H phenotype and represents the level at which lung injury risk increases.
- For intubated and deeply sedated Type L patients who are hypercapnic, ventilate with volumes greater than 6 mL/kg up to 8-9 mL/kg as this high compliance results in tolerable strain without risk of ventilator-associated lung injury. Prone positioning should be used only as a rescue maneuver. Reduce PEEP to 8-10 cm H2O, given that the recruitability is low and the risk of hemodynamic failure increases at higher levels. Early intubation may avert the transition to Type H phenotype.
- Treat Type H phenotype like severe ARDS, including with higher PEEP if compatible with hemodynamics, and with prone positioning and extracorporeal support.
Dr. Gattinoni reported having no financial disclosures.