ECMO can involve mental health costs

Use of extracorporeal membrane oxygenation (ECMO) is associated with a modestly increased risk of new mental health diagnoses in patients treated for critical illness in an intensive care unit (ICU), a retrospective cohort study in Canada has shown.

Among adult survivors who received ECMO, the rate of new mental health diagnosis was 22.1 per 100 person-years compared with 14.5 per 100 person-years in ICU survivors who did not receive ECMO, reported Shannon M. Fernando, MD, MSc, of the University of Ottawa in Ontario, Canada, and colleagues at JAMA.

After propensity weighting, ECMO use for critical illness was significantly associated with an increased risk of new mental health diagnoses (HR 1.24, 95% CI 1.01-1.52), primarily depression, anxiety, and mood-related disorders trauma.

“The use of ECMO has increased worldwide, particularly in the context of the COVID-19 pandemic, where ECMO is being used for refractory respiratory failure in cases of severe COVID-19,” the researchers wrote.

“Given the severity of disease found among patients receiving ECMO, its invasive nature, and the prolonged duration of therapy and recovery often required among those who survive … ECMO survivors may be exposed to more greater risk of downstream mental health morbidity compared to other critical illness survivors,” they added.

Overall, the results of ECMO in the context of the pandemic are mixed, ranging from “reasonable” to “exceptional”.

Of the secondary mental health outcomes assessed by Fernando and team, ECMO and non-ECMO survivors had comparable rates per 100 person-years of substance abuse (1.6 vs 1.4; HR 0.86, 95% CI 0 .48-1.53) and deliberate self-harm ( 0.4 vs 0.3; HR 0.68, 95% CI 0.21-2.23).

In addition, “there were fewer than five total deaths by suicide in the entire cohort,” the group noted.

“As care providers, we can tell our patients that it’s common to struggle with your mental health after an ICU admission,” co-author Peter Tanuseputro, MD, also of the University of Ottawa, said in a statement. “ICU survivors need to realize that they often face months or years of recovery, and families and health care providers need to support them.”

In an accompanying editorial, Marieke Zegers, MD, of the Radboud University Medical Center in Nijmegen, the Netherlands, and colleagues noted that the study design only makes it “possible to hypothesize causal factors between ECMO and new mental health problems “reiterating biological mechanisms proposed by the study’s authors, including potentially increased sensitivity “to hypoxemia, shock, and reduced oxygen delivery, which may predispose neurons to apoptosis and a higher risk of neurological complications such as ischemic stroke or delirium.”

The editors also indicated that only an existing mental health diagnosis (HR 2.39, 95% CI 1.78-3.20) and outpatient psychiatric visit in the year before ICU admission (HR 1.82, 95% CI 1.25-2.65) were significantly associated with an increased risk of a new mental health diagnosis.

“This finding confirms the results of previous studies showing that pre-ICU health status is the most important prognostic factor for post-ICU mental health outcomes and even suicide,” they wrote. “This suggests that ECMO treatment may be an additional stimulus for psychiatric problems in patients who are already vulnerable to psychiatric illness.”

That “almost half of the study population, including patients who received ECMO and those who did not, had at least one visit to primary care or mental health psychiatry in the previous 5 years … raises the question of whether the primary outcome truly reflects new mental health problems,” Zegers and team added, noting that the primary finding held after excluding patients with prior mental health diagnoses in a sensitivity analysis.

“We really need more research and investment in post-critical illness,” Fernando said. “Patients will need help long after they leave the ICU.”

For this study, Fernando and colleagues followed 4,462 adults who were admitted to the intensive care unit and survived to hospital discharge from April 2010 to March 2020. Of these survivors, 642 received ECMO (mean age 50.7 years, 40.7 % women) and were followed for an average of 730 days; they were matched with 3820 ICU survivors who did not receive ECMO (mean age 51.0 years, 40.0% female) who were followed for a mean of 1390 days.

Groups were matched on characteristics including age, gender, mental health history, critical illness severity, and length of hospital stay.

Of the 642 ECMO survivors, 37% were diagnosed with a new psychiatric illness.

The primary outcome consisted of a composite of mood disorders, anxiety disorders, PTSD, schizophrenia, other psychotic disorders, other mental disorders, and social problems. The group also looked at eight secondary outcomes, including incidents of substance abuse, deliberate self-harm, death by suicide, and individual components of the composite primary outcome.

Fernando and colleagues acknowledge that their study was observational and therefore cannot confirm a causal relationship between ECMO and downstream mental health effects.

The team is now preparing to test a virtual program to treat post-ICU syndrome, a collection of physical, mental and emotional symptoms that persist after admission to an intensive care unit.

  • Kate Kneisel is a freelance medical journalist based in Belleville, Ontario.


This study was funded by the Institut du Savoir Montfort, Hôpital Montfort in Ottawa and supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health.

Fernando and Tanuseputro reported no disclosures. The co-authors report numerous relationships with governmental and foundational organizations and pharmaceutical companies.

The editors report no disclosures.

Leave a Comment