Why this article matters
Veno-venous extracorporeal membrane oxygenation (VV-ECMO) is routinely initiated in patients with severe ARDS after intubation, deep sedation, and often neuromuscular blockade. During the COVID-19 pandemic, however, some centers explored an alternative strategy: awake ECMO cannulation, in which patients are cannulated while spontaneously breathing and supported without invasive mechanical ventilation.
The theoretical appeal is clear. Avoiding intubation may reduce ventilator-induced lung injury, prevent ventilator-associated pneumonia, preserve diaphragmatic function, and allow early mobilization. Yet robust comparative outcome data in ARDS have been limited, particularly in a homogeneous disease population.
This multicenter study from the Israeli ECMO registry provides one of the most comprehensive comparisons to date between awake and sedated VV-ECMO cannulation in patients with COVID-19-induced ARDS.
Study design and population
Investigators conducted a retrospective, multicenter cohort study across eight ECMO centers in Israel, using registry data collected between April 2020 and December 2022.
- Study group: 24 patients cannulated for VV-ECMO while awake and spontaneously breathing
- Control group: 96 patients cannulated after sedation and invasive mechanical ventilation
- Groups were matched 1:4 by age, sex, and body mass index
All patients met Berlin criteria for ARDS and were treated according to local clinical judgment, as no formal selection criteria for awake ECMO exist.
The primary outcome was six-month survival. Secondary outcomes included ECMO duration, ICU length of stay, duration of invasive mechanical ventilation, and complication rates.
Baseline differences worth noting
Patients cannulated while awake differed from controls in several clinically meaningful ways:
- Lower SOFA scores at ECMO initiation (5.2 vs. 9.1), largely reflecting the absence of mechanical ventilation and sedation
- Lower pre-ECMO PaCO₂ and higher pH, consistent with preserved spontaneous ventilation
- Half of the awake group was supported with HFNC, and ~42% with NIV, prior to cannulation
Importantly, many conventional severity metrics (SOFA, PaO₂/FiO₂) are inherently influenced by ventilatory status, making direct severity comparisons challenging.
Key results
Several findings stood out:
Survival
- Six-month survival was higher in the awake group (75% vs. 49%) in unadjusted analyses
- After adjustment for PaO₂/FiO₂ and COVID-19 variant, the association was attenuated and no longer statistically significant, though the hazard ratio remained below 1
Mechanical ventilation exposure
- 63% of awake-cannulated patients were eventually intubated, typically due to uncontrolled respiratory drive or clinical deterioration
- Despite this, the awake group accrued a mean of ~12 mechanical ventilation-free days during ECMO
Infectious complications
- No cases of pneumonia occurred in patients who remained non-intubated throughout their ECMO course
- Overall infectious complications were numerically lower in the awake group
ECMO and ICU duration
- ECMO run time and ICU length of stay were similar between groups, arguing against simple disease-severity differences
How should we interpret these findings?
The results suggest that awake VV-ECMO is feasible in carefully selected patients with COVID-19 ARDS and may reduce exposure to invasive ventilation and ventilator-associated complications.
However, the authors appropriately emphasize caution:
- The study is retrospective and susceptible to selection bias
- Awake cannulation was not randomized and depended heavily on clinician judgment and patient cooperation
- A phenomenon akin to confounding by indication or depletion of susceptibility may be present, where patients with more favorable trajectories were preferentially selected for awake ECMO
- Adjustment for key covariates reduced the apparent survival advantage
As such, the findings should be considered hypothesis-generating, not definitive evidence of superiority.
Clinical implications
This study does not argue that awake ECMO should replace conventional practice. Rather, it highlights that:
- Awake ECMO can be safely attempted in highly selected ARDS patients
- Avoiding intubation may reduce infectious complications and preserve ventilator-free time
- Careful monitoring of respiratory drive is critical, as loss of control remains the most common reason for intubation
- Awake ECMO requires substantial institutional experience, staffing, and multidisciplinary coordination
Conclusion
Awake VV-ECMO cannulation in COVID-19-related ARDS is feasible and associated with encouraging unadjusted outcomes, including fewer infections and longer periods without invasive ventilation. However, once key confounders are accounted for, the apparent survival benefit becomes less certain.
For now, awake ECMO should be viewed as a selective strategy rather than a new standard of care—one that warrants further investigation through larger observational studies and target-trial emulation approaches.
