ECMO Lets Surgeons Operate on Patients Too Unstable for the Operating Room

Extracorporeal membrane oxygenation, or ECMO, has long been the machine of last resort in intensive care units: a heart-lung bypass circuit wheeled to the bedside when a patient’s lungs or heart fail so completely that conventional treatment cannot keep them alive. A new structured narrative review published in Frontiers in Anesthesiology suggests that this rescue technology is quietly transforming a very different arena, the operating room, where it is now being used to let surgeons operate on adult patients who would otherwise be considered far too unstable to survive an operation.

The review, led by Thomas F. O’Shea and colleagues at the University of Colorado Anschutz Medical Center, systematically assembled the published experience with ECMO-assisted adult non-cardiac surgery across airway, trauma, burn, vascular, and obstetric care. The authors searched PubMed/MEDLINE, PubMed Central, and Google Scholar from database inception through July 14, 2026, and drew on case reports, case series, retrospective cohorts, database analyses, and prior reviews. Crucially, they distinguished between two clinically different scenarios: planned or enabling ECMO, initiated specifically to permit an operation that would otherwise be technically or physiologically prohibitive, and ECMO-supported surgery, in which a necessary non-cardiac operation is performed while a patient is already on ECMO for respiratory or circulatory failure.

The most mature application, by far, is airway and tracheobronchial surgery. Here, ECMO is often deployed electively, before the first incision, because the operation itself would make conventional ventilation impossible. When surgeons resect the trachea or carina, the very structure that carries air to the lungs, there is no safe way to ventilate the patient through the surgical field. ECMO solves this by taking over gas exchange entirely, allowing the surgical team to work in an unobstructed field without an endotracheal tube crossing the lesion or the anastomosis. One narrative review cited in the paper summarized 168 published cases of extracorporeal life support during airway surgery, and institutional series have reported safe completion of tracheal resections with VV-ECMO support lasting as little as 10 to 31 minutes, with no in-hospital deaths in one series of seven oncologic tracheal operations.

The choice between the two main ECMO configurations follows a simple physiological logic. Venovenous ECMO, or VV-ECMO, drains deoxygenated blood from a vein, passes it through an external membrane oxygenator, and returns it to the venous system, providing gas exchange only. It is the preferred mode when the lungs, but not the heart, are the problem. Venoarterial ECMO, or VA-ECMO, returns oxygenated blood to the arterial system, providing both circulatory and respiratory support, and is chosen when hemodynamic compromise, anticipated major bleeding, or the demands of the procedure require more complete cardiopulmonary support. In the airway literature, VV-ECMO was commonly sufficient; in vascular and trauma settings, VA-ECMO predominates.

Trauma represents a different paradigm. Rather than enabling planned operations, ECMO in trauma patients is principally a stabilizing bridge that buys time for time-sensitive surgical injuries to be addressed. The review highlights striking examples: a multicenter analysis of the Trauma Registry-DGU found that 1.1 percent of patients with neurologic trauma received ECMO, with 61.8 percent surviving to discharge, challenging the historical view that brain injury is a contraindication to the therapy. One reported patient with severe respiratory distress syndrome, a large intracerebral hematoma, and multiple injuries after a fall underwent craniotomy while on VV-ECMO and was later discharged with full neurologic recovery. Small series describe femoral fracture fixation, pelvic ring fixation, and even posterior spinal fixation performed during ECMO support, with all reported femoral fixation patients surviving to discharge.

Abdominal trauma adds another layer of complexity. Damage-control laparotomy, the rapid staged approach to hemorrhage and bowel ischemia, has traditionally been viewed with suspicion in anticoagulated patients. Yet a single-center retrospective study of 52 patients supported with VV-ECMO before or after damage-control laparotomy reported an overall survival rate of 58 percent, and after adjustment for confounders, patients who could tolerate anticoagulation had improved survival. The authors concluded that damage-control laparotomy should not be considered a contraindication to VV-ECMO, or vice versa, a finding that could reshape how trauma teams think about the interplay of bleeding risk and extracorporeal support.

In burn care, the review found that ECMO has been used to support excision, debridement, and wound coverage in patients with severe inhalation injury or burn-associated respiratory distress syndrome. One burn center proposed a standardized perioperative protocol centered on multidisciplinary planning, individualized anticoagulation, and point-of-care viscoelastic coagulation monitoring, targeting a low-dose heparin regimen with an activated partial thromboplastin time of approximately 40 seconds during surgery. The dominant challenges are surgical blood loss, hypothermia, infection, and the resource intensity of repeated wound care, all compounded by the need to keep the ECMO circuit free of clots while the patient bleeds.

Vascular surgery illustrates both the promise and the risks of the technology. VA-ECMO has been used as planned support for high-risk thoracoabdominal aortic aneurysm repair and as a bridge during emergent pulmonary embolectomy, and some teams propose advantages over conventional cardiopulmonary bypass, including lower anticoagulation targets, reduced blood-air interface, and avoidance of systemic cooling. But peripheral cannulation carries its own vascular toll: a systematic review of 47 studies involving 6,583 patients reported an overall pooled vascular complication rate of 29.5 percent, with limb ischemia in 12.6 percent and significant bleeding in 15.4 percent. Preventive strategies include ultrasound-guided access, the smallest adequate arterial cannula, prophylactic distal perfusion cannulation, and serial perfusion checks by examination, Doppler, and near-infrared spectroscopy.

Obstetric applications are among the most dramatic. The literature consists primarily of cesarean delivery performed during ongoing ECMO support for severe maternal cardiopulmonary failure, with early reports dating back more than a decade and the COVID-19 pandemic expanding the experience considerably. A ten-year case series of 54 pregnant and postpartum patients requiring ECMO identified acute respiratory distress syndrome as the most common indication, and a systematic review and meta-analysis identified 1,460 reported cases of pregnant patients receiving VA- or VV-ECMO, with survival rates of 62.8 percent for VA-ECMO and 83.6 percent for VV-ECMO. These figures, however, come from heterogeneous cohorts not restricted to surgical patients and cannot be read as estimates of the effect of ECMO-facilitated delivery itself.

The review’s most important message may be its caution. The evidence base is dominated by single-patient case reports and small observational series from experienced centers, with substantial publication and selection bias: successful, unusual operations are far more likely to be reported than unsuccessful attempts or patients who were considered but not selected. No randomized studies define thresholds for ECMO use, optimal cannulation strategy, or standardized anticoagulation targets in this setting, and comparator populations are generally absent. The authors are explicit that the current literature should not be interpreted as establishing a survival or treatment-effectiveness benefit. What the review does establish is a practical framework: multidisciplinary planning, individualized cannulation and anticoagulation, hemorrhage preparedness, careful anesthetic and ventilatory management, and coordinated postoperative critical care are the recurring pillars of every successful reported case. Until robust multicenter and comparative data arrive, ECMO-assisted non-cardiac surgery remains a selective strategy for experienced teams, deployed when the patient’s physiology, the operative requirement, and the likelihood of a meaningful surgical objective all align.

Subject of Research: Use of extracorporeal membrane oxygenation to facilitate adult non-cardiac surgery

Article Title: ECMO-facilitated adult non-cardiac surgery: a structured narrative review

Article References: O’Shea, T. F., Bashian, E. J., Park, S. Y., Carpenter, R., Hoffman, J. R. H., Cain, M. T., Kowalsky, M., Abrams, B., Clendenen, N., Teman, N. R., & Wilkey, B. (2026). ECMO-facilitated adult non-cardiac surgery: a structured narrative review. Frontiers in Anesthesiology, 5, Article 1944652. https://doi.org/10.3389/fanes.2026.1944652

Image Credits: AI Generated

DOI: 10.3389/fanes.2026.1944652

Keywords: ECMO, non-cardiac surgery, airway surgery, trauma surgery, burn surgery, vascular surgery, obstetrics, anticoagulation, VV-ECMO, VA-ECMO, perioperative care, critical care