ECMO Has Become a Harder Sell in Severe COVID-19

HONOLULU — In hindsight, researchers saw more limitations to extracorporeal membrane oxygenation (ECMO) and proning, therapies initially touted as helpful for COVID patients in intensive care.

Nationwide hospitalization records showed that ECMO, a labor- and cost-intensive intervention for severe acute respiratory distress syndrome (ARDS), actually portended worse survival in critically ill COVID-19 patients. Use of ECMO was already on its way down by the second wave of the pandemic in the fall of 2020, another retrospective study showed.

Separately, proning did not offer additional help with breathing in a case series of people already on ECMO, unlike those on mechanical ventilation in prior literature.

The new findings were presented here at the CHEST annual meeting hosted by the American College of Chest Physicians.

More Mortality After ECMO

COVID-19 changed the survival equation for ARDS patients put on ECMO in an analysis of the National Inpatient Sample (NIS) database.

ECMO was associated with higher odds of inpatient mortality in COVID-19 patients (OR 1.428, 95% CI 1.303-1.566), but lower odds in non-COVID patients (OR 0.839, 95% CI 0.739-0.953), according to a presentation by Mustafa Al-Taei, MD, internal medicine resident at Montefiore Medical Center in New York City.

Despite lower unadjusted inpatient mortality on ECMO vs no ECMO in COVID-19 (47.2% vs 51.4%) multivariable adjustment led to the study’s negative finding for ECMO and survival.

Al-Taei acknowledged that the statistical adjustment was likely imperfect, as there were many patient variables lacking in the dataset. “We’re just bringing this to researchers to see if it is possible to perform [further studies],” he said.

“Despite the limitations of the database utilized, our data call the utility of ECMO in patients with COVID-19 and ARDS into question. Further studies are needed to evaluate the effectiveness and safety of ECMO in this population of patients and to identify optimal selection criteria and timing for this intervention,” Al-Taei concluded.

For the study, he and his collaborators analyzed hospitalizations from 2020 that had been recorded in NIS. Of the 135,760 people admitted with ARDS, nearly three in four were due to COVID-19.

ECMO was administered to 2.4% of people with COVID-19 and ARDS vs 4% of non-COVID ARDS patients. COVID patients who were selected for ECMO tended to be older, more likely men, and have generally fewer comorbidities — albeit more obesity and need for dialysis — compared with peers not receiving ECMO.

Hospitals Sour on Costly ECMO

Clinicians realized they needed less ECMO for COVID as early as late 2020, according to a separate study, also based on NIS data.

Nationwide, there was lower ECMO use between the first pandemic wave in the spring and the second wave in the fall — possibly related to the positive dexamethasone trial published that June, or because hospital systems had learned how to manage their patients over time, speculated study author Kam Sing Ho, MD, a fellow at the University of Maryland Medical Center in Baltimore.

Even so, centers racked up $908 million in total hospitalization costs that calendar year for COVID patients put on ECMO.

The mean hospital cost for survivors was $256,018 (mean hospital length of stay [LOS] 39 days) and for non-survivors $237,835 (LOS 28 days). Compare that to the price tag for COVID-19 hospitalizations without ECMO support — an average $17,933 for survivors (LOS 7.45 days) and $41,626 for non-survivors (LOS 11.43 days).

“This highlights the importance of effective risk-stratifying patients being considered for ECMO to help identify the most appropriate candidates for ECMO cannulation,” Ho told the audience at CHEST.

Like Al-Taei, Ho warned of the NIS database’s limited data quality.

Nevertheless, his group relied on it to find records for over 1.6 million adult COVID-19 hospitalizations in the year 2020. Of those, 208,595 individuals required mechanical ventilation, and 3,745 received ECMO support.

ECMO recipients were mostly people ages 40-60 (61.8%), men (68.8%), and either white (35.2%) or Hispanic (34.4%).

Prone Position Plus ECMO?

In a separate report, researchers from one center could not find evidence that a single proning cycle offers additional benefit for 12 COVID-19 patients already on ECMO for ARDS.

Sixty days after being placed belly-down on ECMO, four patients were alive in the hospital, two had been discharged home, one discharged to rehab, and five died, reported Xuan Han, MD, MS, a pulmonologist and critical care intensivist at Tufts Medical Center in Boston.

Han said she saw no difference on several breathing parameters before and soon after ECMO patients were laid face-down in prone position:

• Respiratory mechanics (e.g., tidal volume, plateau pressure, positive end-expiratory pressure, driving pressure, fraction of inspired oxygen)

• Gas exchange (e.g., pH, partial pressure of carbon dioxide, arterial oxygen pressure)

• ECMO support (e.g., flow rate and sweep)

At least no adverse events related to proning were observed, she noted.

Proning has been shown to improve survival in the ARDS literature, ostensibly by improving gas exchange and decreasing lung stress. Putting someone in prone position after intubation is associated with better clinical outcomes.

This has not been replicated in awake prone positioning, however, a setting in which proning has only reduced the need for intubation in some individuals with severe COVID-19 without helping hard outcomes.

At CHEST, Han presented a retrospective study that included 12 adults with COVID-19 pneumonia (mean age 51 years, 75% men) who were admitted to an urban academic tertiary center from January 2020 to February 2022 and placed in prone position while on venovenous ECMO.

It had taken an average 5.3 days from intubation to ECMO initiation. Proning was initiated an average 7.3 days after ECMO cannulation.

Data were collected pre-proning and again 1 hour after initiation of the first proning cycle. All 12 patients were deeply sedated and largely paralyzed during this time due to concerns about safety, the researcher noted.

Han acknowledged that benefits of proning during ECMO could have taken more than 1 hour to manifest. Patients had undergone a median 3.5 proning cycles, and data were not collected beyond the first one.

Please enable JavaScript to view the

comments powered by Disqus.