Acute hypoxemic respiratory failure (AHRF) and its more severe form, acute respiratory distress syndrome (ARDS), imply a decrease in the ability to exchange oxygen and carbon dioxide by the lungs due to different causes (1). This induces an extra activation of central respiratory drive by a multiplicity of factors (2). In turn, excessive respiratory drive increases diaphragm effort and transpulmonary pressure, potentially causing self-inflicted diaphragm and lung injury (3). In recent years, huge efforts have been made to test noninvasive respiratory support able to attenuate respiratory drive (e.g., high flow nasal cannula and helmet noninvasive ventilation) (4), protecting the lungs and the diaphragm while the initial insult (e.g., pneumonia) heals. In fact, when respiratory drive remains within physiological limits, spontaneous breathing is associated with many physiological benefits as compared to controlled mechanical ventilation. Sedation and paralysis can be avoided or limited, secretion clearance and airway mucosal activity are preserved limiting the risk of infections, tidal volume distribution within the lungs is more homogenous, and venous return is maintained favoring hemodynamic stability (5). In addition to the unphysiological shortcomings of mechanical ventilation, the intubation procedure itself carries potentially catastrophic consequences for hypoxemic patients (6).

Still, in recent trials, around 30–40% of patients admitted with a diagnosis of AHRF or ARDS have progressive respiratory failure despite noninvasive support and end up requiring intubation (4). These patients are target candidates for an alternative (and costly) treatment, namely awake extracorporeal membrane oxygenation (ECMO).

Awake ECMO has been studied since the 1970s as a means to control central respiratory drive. The study by Gattinoni et al (7) first described the linear decrease of minute ventilation in healthy sheep obtained by increasing CO2 removal through awake ECMO. The mechanism being simple and straightforward: when the main determinant of respiratory drive is blood CO2 tension and the resulting pH, and an external system clears a substantial part of a patient’s CO2 production (VCO2), less minute ventilation is needed to maintain stable levels (8). Modern ECMO systems can potentially clear all the patient’s VCO2 (9) and the increase in the patient’s mixed venous and arterial oxygen saturation induced by ECMO carries additional physiological benefits for extrapulmonary organs’ function (Table 1). Additionally, awake ECMO has been shown to be an effective and safe modality to promote physical therapy and maintain conditioning in patients with advanced respiratory failure (10). Awake ECMO could be seen as the perfect modulator of respiratory drive with additional distal organ protection, thus representing the ideal noninvasive respiratory support to promote lung healing and avoid intubation. This, in fact, is not entirely true for the following reasons: 1) there is evidence that determinants of respiratory drive in hypoxemic patients are multiple and independent from CO2 clearance (11), 2) awake ECMO is associated with a number of complications such as bleeding, thrombosis, infections, despite many technological advances, and 3) awake ECMO is resource intensive, given the high costs for the materials and medications, and the highly specialized staff available only in few centers (12).

TABLE 1. - Pathophysiology of Awake Extracorporeal Membrane Oxygenation as Compared With Intubation Awake Extracorporeal Membrane Oxygenation vs Intubation Effects Benefits Risks Transpulmonary pressure ↓ Through ↓ respiratory drive ↓ Risk of PSILI Difficult to monitor in nonintubated patients Inspiratory effort ↓ Through ↓ respiratory drive ↓ Risk of PSILI and diaphragm injury Difficult to monitor in nonintubated patients Right heart function ↑ Cardiac output by ↑ venous return ↑ Oxygen transport May be outweighed by ↑ afterload Sedation ↓ Need for sedative drugs ↓ Weakness and delirium Could be a stressful experience Coagulation Need for full anticoagulation Prevention of deep vein thrombosis and pulmonary embolism Bleeding Inflammation Activated by the circuit ↓ Risk of additional inflammation by PSILI Systemic activation (thrombosis, increased patient’s CO2 production, etc.) Infections No tube, more intravascular catheters ↓ Risk of ventilator-associated tracheobronchitis/ventilator-associated pneumonia ↑ Risk of catheter-related bloodstream infections Physical therapy Increased opportunity for early mobilization Maintain conditioning, ↓ weakness, and delirium ↑ Risk for cannula dislodgement and complications

PSILI = patient self-inflicted lung injury.

To help shed light on the current role of awake ECMO in the management of hypoxemic patients, in this issue of Critical Care Medicine, the study by Belletti et al (13) presents an updated systematic review of clinical studies performed thus far on this topic. After an extensive and rigorous search, 57 articles (28 case reports, 29 case series, no randomized controlled trial) with 467 awake ECMO patients were included in the analysis. The primary outcome was failure of awake ECMO as defined by the original authors or, if not originally defined, as need for intubation. Secondary outcomes included mortality, successful weaning from ECMO, and complications. Statistical meta-analyses to calculate estimates for dichotomous outcomes were performed only considering series with greater than one patient, in the whole population (n = 261) and by a priori separation of studies reporting indications for awake ECMO (bridge-to-lung transplant for terminal chronic disease [9 studies, 62 patients] vs bridge to lung healing for ARDS [8 studies, 89 patients]).

Pooled estimates for awake ECMO failure were rather promising: 40.1% for all patients, 23.6% in bridge to lung transplant, and 41.9% in ARDS. Overall mortality was, surprisingly, even lower: 27.5% for all, 29.7% for bridge, and 20.2% for ARDS. However, mortality in patients failing awake ECMO and proceeding to intubation was very high: 58.1% overall, 85.7% in bridge, and 53.8% in ARDS. Interestingly, Supplementary Table 11 in (13) reports reasons for awake ECMO failure, which almost entirely coincides with either uncontrolled respiratory drive (agitation, worsening respiratory failure) and/or ECMO-related complications (bleeding, thrombosis, surgical infections, technical failure). Finally, pre-planned sensitivity analysis considering only a few studies including greater than five patients, reassured on the magnitude and direction of the results.

The study surely has multiple strengths: search terms together with rather conservative inclusion and exclusion criteria are listed at the beginning of online supplement in (13); statistical methodology is rigorous and up to date, and from a group that already published a number of meta-analyses and rigorous large clinical studies in high impact journals (14); a priori separation of indications for awake ECMO based on physiology; extensive reporting of complications and reasons for awake ECMO failure; list of excluded studies with reference and detailed reason. Nonetheless, awake ECMO still represents a niche experimental practice and the low number of patients included in the analysis, the observational nature and the heterogeneity of the original studies, the lack of prospective randomized controlled trials, the variability of ECMO setup and of the respiratory support used (Supplementary table 5 in [13]), the difficulty to objectivate novel respiratory super-infections in nonintubated patients, all represent significant limitations. The low overall mortality rates also likely reflect a highly curated patient population, reporting likely from primarily experienced ECMO centers, and publication bias, so the generalizability of the findings must be interpreted cautiously.

The study is useful to confirm the feasibility of performing awake ECMO in selected patients admitted to specialized centers. Additionally, this study serves to describe the possible complications that healthcare providers will need to manage, report the very high success rate for bridge-to-lung transplants at experienced centers, and provide a warning regarding the high mortality of patients started on awake ECMO and subsequently intubated. This last point warrants particular attention in selecting appropriate patients for a trial of awake ECMO given the significant consequences of a failed attempt. Furthermore, the study offers solid risk estimates to design adequately powered prospective randomized clinical trials, which are needed.

Although we wait for such studies, results from the study by Belletti et al (13) may foster physiological and clinical reasoning and guide more accurate selection of candidates for awake ECMO. Indeed, two main reasons could explain better outcomes for bridge-to-lung transplant versus ARDS patients: more effective control of respiratory drive by awake ECMO and availability of an early effective treatment for respiratory failure. Severe chronic disease generates hypoxemia by very different mechanisms than ARDS. Ventilation-perfusion mismatch is mainly due to airway constriction and occlusion by secretions, parenchymal remodeling (e.g., emphysematous changes and bullae), and muscular exhaustion leading to hypoventilation (15). Thus, increased respiratory drive in these patients is mainly driven by altered gas exchange rather than by alternative factors such as inflammation, acidosis, alveolar collapse, and agitation, so that awake ECMO will more easily control it (2). Furthermore, at least in the short-term perspective, lung transplant represents a more effective treatment than those for causes of ARDS, which sometimes don’t even have therapies (e.g., viral pneumonia or trauma).

Taken together, this reasoning may suggest that, to improve the outcome of awake ECMO, physiology-based selection of patients is crucial. For nonintubated ARDS, ideal candidates could be those with large tidal volumes indicating preserved respiratory system and minimal collapse, no metabolic acidosis and stable hemodynamics, lower systemic inflammation, minimal agitation or confusion, and proved treatable etiology. For intubated ARDS patients, direct screening for optimal respiratory mechanics could facilitate recognition, while all other points broadly coincide. For bridge-to-lung transplant, instead, exclusion of severe infection and higher likelihood of receiving early transplantation may be crucial.

For healthcare workers in the ICU, caring for patients on awake ECMO can be a more rewarding experience than caring for intubated patients as these patients speak, laugh, eat, interact with family, and perform active physiotherapy despite the severity of their disease. This, together with clear and relevant physiological and clinical research questions, represent major drivers to continue research in such a forward-looking and fascinating topic.

1. Matthay MA, Arabi Y, Arroliga AC, et al.: A new global definition of acute respiratory distress syndrome. Am J Respir Crit Care Med. 2023 Jul 24. [online ahead of print] 2. Spinelli E, Pesenti A, Slobod D, et al.: Clinical risk factors for increased respiratory drive in intubated hypoxemic patients. Crit Care. 2023; 27:138 3. Goligher EC, Dres M, Patel BK, et al.: Lung- and diaphragm-protective ventilation. Am J Respir Crit Care Med. 2020; 202:950–961 4. Grieco DL, Maggiore SM, Roca O, et al.: Non-invasive ventilatory support and high-flow nasal oxygen as first-line treatment of acute hypoxemic respiratory failure and ARDS. Intensive Care Med. 2021; 47:851–866 5. Spinelli E, Scaramuzzo G, Slobod D, et al.: Understanding cardiopulmonary interactions through esophageal pressure monitoring. Front Physiol. 2023; 14:1221829 6. Russotto V, Tassistro E, Myatra SN, et al.: Peri-intubation cardiovascular collapse in patients who are critically ill: Insights from the INTUBE study. Am J Respir Crit Care Med. 2022; 206:449–458 7. Kolobow T, Gattinoni L, Tomlinson TA, et al.: Control of breathing using an extracorporeal membrane lung. Anesthesiology. 1977; 46:138–141 8. Mauri T, Grasselli G, Suriano G, et al.: Control of respiratory drive and effort in extracorporeal membrane oxygenation patients recovering from severe acute respiratory distress syndrome. Anesthesiology. 2016; 125:159–167 9. Spinelli E, Mauri T, Lissoni A, et al.: Spontaneous breathing patterns during maximum extracorporeal CO2 removal in subjects with early severe ARDS. Respir Care. 2020; 65:911–919 10. Salna M, Abrams D, Brodie D: Physical rehabilitation in the awake patient receiving extracorporeal circulatory or gas exchange support. Ann Transl Med. 2020; 8:834 11. Spinelli E, Mauri T, Beitler JR, et al.: Respiratory drive in the acute respiratory distress syndrome: Pathophysiology, monitoring, and therapeutic interventions. Intensive Care Med. 2020; 46:606–618 12. Tonna JE, Abrams D, Brodie D, et al.: Management of adult patients supported with venovenous extracorporeal membrane oxygenation (VV ECMO): Guideline from the Extracorporeal Life Support Organization (ELSO). ASAIO J. 2021; 67:601–610 13. Belletti A, Sofia R, Cicero P, et al.: Extracorporeal Membrane Oxygenation Without Invasive Ventilation for Respiratory Failure in Adults: A Systematic Review. Crit Care Med. 2023; 51:1790–1801 14. Belletti A, Todaro G, Valsecchi G, et al.: Barotrauma in coronavirus disease 2019 patients undergoing invasive mechanical ventilation: A systematic literature review. Crit Care Med. 2022; 50:491–500 15. Demoule A, Brochard L, Dres M, et al.: How to ventilate obstructive and asthmatic patients. Intensive Care Med. 2020; 46:2436–2449