August Felix, Rudi Zurbuchen, Joshua Ferreri, Emma Huebner, Ron Pauldine, Raquel Bartz
Extracorporeal membrane oxygenation (ECMO) is increasingly used to support patients with cardiac or respiratory failure as a bridge to recovery, transplantation, or durable mechanical circulatory support. When these goals cannot be achieved, withdrawal of ECMO presents complex clinical and ethical challenges, yet standardized guidance remains limited. We performed a narrative review of the literature from ECMO inception through June 2026 using PubMed, Embase, Scopus, and JSTOR. Studies, guidelines, registry analyses, and commentaries addressing ECMO withdrawal, transplant eligibility, communication, and end-of-life care in adults were included. Withdrawal of ECMO is common, particularly among patients receiving VA ECMO, but practices vary considerably across institutions. Decisions are typically made by surrogate decision-makers after determination that recovery, transplantation, or durable mechanical support is no longer feasible. Current evidence emphasizes the importance of multidisciplinary decision-making, longitudinal goals-of-care discussions, proactive symptom management, and structured approaches to compassionate ECMO discontinuation. We also identify substantial gaps in evidence regarding consent, withdrawal protocols, and clinician support. ECMO withdrawal remains an understudied area with institutional variation and without standardized best practices. We propose reframing ECMO consent as an ongoing goals-of-care process and present an ECMO-specific consent template to improve communication, align expectations, and support compassionate, patient-centered end-of-life care. Further research is needed to inform evidence-based guidelines.
Thomas F. O’Shea, Elizabeth J. Bashian, Sarah Y. Park, Rachel Carpenter, Jordan R. H. Hoffman, Michael T. Cain, Markus Kowalsky, Benjamin Abrams, Nathan Clendenen, Nicholas R. Teman, Barbara Wilkey
Extracorporeal membrane oxygenation (ECMO) has traditionally been used as rescue support for severe respiratory failure, cardiopulmonary failure, or perioperative cardiac surgical care, but its expanding perioperative role has enabled selected non-cardiac operations in patients otherwise considered too unstable for surgery. We performed a structured narrative review of reported ECMO-assisted adult non-cardiac surgery, focusing on airway, trauma, burn, vascular, and obstetric applications. Published case reports, case series, retrospective cohorts, database analyses, and relevant reviews were identified through structured searches of PubMed/MEDLINE, PubMed Central, and Google Scholar, supplemented by manual reference review. For this review, ECMO-assisted surgery included both planned or enabling ECMO initiated specifically to permit an operation and ECMO-supported surgery performed while a patient was already receiving ECMO for respiratory or circulatory failure. Airway and tracheobronchial surgery represented the most developed application, whereas trauma, burn, vascular, and obstetric reports more commonly described ECMO as rescue or stabilizing support that permitted necessary surgery to proceed. Across specialties, the published experience demonstrates technical feasibility in highly selected reported patients but is dominated by case reports and small observational series, with substantial publication and selection bias and few comparator populations. Multidisciplinary planning, individualized cannulation and anticoagulation strategies, hemorrhage preparedness, anesthetic planning, and postoperative critical care are central to safe implementation. The current literature should not be interpreted as establishing a survival or treatment-effectiveness benefit.
Ketamine is a frequently employed anesthetic that has been shown to increase the risk of new psychiatric diagnoses within 30 days postoperatively. Midazolam is a benzodiazepine that is often utilized by providers to attenuate emergence phenomena when ketamine is used. The presence of emergence phenomena is known to be associated with the development of subsequent 30-day psychiatric complications. In this investigation, we explored whether the use of midazolam with ketamine in patients undergoing vascular surgery is associated with reduced odds of developing new psychiatric events within 30 days postoperatively. This study aimed to determine whether the midazolam-ketamine combination is associated with reduced risk for development of long-term psychiatric events compared to ketamine alone in patients undergoing vascular surgery. We conducted a retrospective cohort study using TriNetX, a national database that collects deidentified patient data from electronic health records. Patients were included if they were over 18 years of age and underwent vascular surgery procedures. Patients with outcomes of interest were excluded since preexisting psychiatric conditions are known risk factors for the development of postoperative psychiatric events. Patients were categorized into the following three cohorts: ketamine alone, ketamine and midazolam, and controls receiving neither. Our primary outcome was any new psychiatric disorder diagnosis occurring within 30 days postoperatively. The ketamine cohort was associated with an increased odds of composite [odds ratio (OR) 2.25, 95% confidence interval (CI): 1.81–2.79, p < 0.0001], mood (OR 1.69, 95% CI: 1.12–2.54, p = 0.0113), anxiety (OR 2.21, 95% CI: 1.57–3.12, p < 0.0001), and behavioral (OR 2.79, 95% CI: 2.13–3.65, p < 0.0001) 30-day psychiatric diagnoses compared to controls. The ketamine and midazolam cohort was associated with an increased odds of composite (OR 2.00, CI: 1.82–2.20, p < 0.0001), mood (OR 1.53, CI: 1.29–1.82, p < 0.0001), anxiety (OR 2.63, CI: 2.26–3.06, p < 0.0001), and behavioral (OR 1.85, CI: 1.65–2.07, p < 0.0001) 30-day psychiatric diagnoses compared to controls. The Ketamine cohort was associated with increased odds of composite (OR 1.22, CI: 1.01–1.48, p = 0.044) and behavioral (OR 1.44, CI: 1.14–1.81, p = 0.0016) 30-day psychiatric diagnoses compared with the ketamine and midazolam cohort. Midazolam and ketamine coadministration was associated with decreased 30-day behavior-related psychiatric diagnoses in patients undergoing vascular surgery compared to patients receiving ketamine alone.
Maeve F. Givens, Edward Rodriguez Caceres, Rohan Jotwani, John E. Rubin, Peter A. Goldstein, Robert S. White
Artificial intelligence (AI) is expanding rapidly across anesthesiology, with growing applications in obstetric anesthesia at a time when maternal mortality in the United States remains high and marked racial and ethnic disparities persist. This scoping review aimed to characterize the current uses, emerging applications, and future directions of AI across the field of obstetric anesthesiology. We searched PubMed and Embase, combining clinical terms for obstetric anesthesia and delivery with methodological terms for AI and machine learning (ML). English language original studies, systematic reviews, clinical trials, and meta-analyses published between January 2015 and December 2025 describing the development, validation, or clinical application of AI/ML tools relevant to obstetric anesthesia were eligible. Records were screened by two reviewers with consensus adjudication by two board-certified anesthesiologists. Included studies were categorized by AI domain, ML class, intended application, validation approach, and aspect of care. Of 3,135 records identified, 172 met inclusion criteria for review, comprising 154 original research studies, 17 systematic or narrative reviews, and 1 case report. 12 other records encompassing 9 registered clinical trials or protocols and 3 workforce-focused studies were also identified and reported upon separately. 69% of publications appeared after 2022. The most common domains were maternal risk and complications prediction (n = 61), delivery mode prediction (n = 43), and patient education and conversational AI (n = 15). Reported model discrimination was frequently strong, with areas under the curve (AUC) commonly between 0.75 and 0.95. However, only 14% of the 154 original research studies reported external validation. Prospective evaluations were rare, and almost none described integration into clinical workflow. The use of AI in obstetric anesthesia has accelerated rapidly, spanning maternal risk prediction, delivery planning, fetal monitoring, patient education, and provider training. Although discrimination metrics are generally strong, the literature remains largely retrospective, with limited external validation and little evidence of real-world clinical impact. Future efforts must prioritize externally validated, prospectively evaluated, and workflow-integrated studies if AI is to meaningfully reduce preventable maternal morbidity and mortality.
