High-risk pulmonary embolism (PE) has a low incidence but progresses rapidly, often causing obstructive shock, cardiac arrest, and extremely high mortality [1, 2]. Early identification and timely intervention are crucial for improving patient outcomes. According to guidelines from European Society of Cardiology (ESC), pulmonary revascularization should be initiated promptly in suspected high-risk PE patients who are unable to undergo computed tomography pulmonary angiography (CTPA) due to unstable hemodynamics [3, 4].

However, clinical experience shows that although guideline-directed approaches benefit some patients, they pose risks of misdiagnosis and inappropriate treatment. Furthermore, in specific patient populations, such as those contraindicated for thrombolysis, the absence of CTPA results can substantially increase the complexity of surgical thrombectomy and may lead to missed opportunities for timely intervention.

ECMO provides robust cardiopulmonary support. Case series have shown that ECMO is an effective circulatory support strategy for high-risk PE, with some patients achieving favorable outcomes [5, 6, 7]. Nevertheless, large-scale studies report treatment success rate of 38–50% [8, 9, 10]. Improving survival in high-risk PE remains a focus of ongoing research.

ECMO for managing high-risk PE was first implemented in the Pearl River Delta region in China. For suspected high-risk PE patients with hemodynamic instability who are unable to undergo CTPA due to unstable hemodynamics. ECMO is initiated as a priority. Subsequently, CTPA is performed under ECMO to confirm the diagnosis, followed by tailored revascularization. For patients whose families decline ECMO due to personal considerations, revascularization is pursued. To date, no studies have evaluated whether this ECMO-prioritized strategy benefits such patients.

We retrospectively analyzed the clinical data of patients with suspected high-risk PE at Four high-volume ECMO centers in the Pearl River Delta region. This study aimed to: (1) determine whether prioritizing ECMO improves diagnostic accuracy and survival in suspected high-risk PE; (2) assess the correlations between pulmonary artery thrombus burden and clinical severity. Our objective is to improve clinical outcomes in high-risk PE.

Previous studies have demonstrated that thrombus burden, obstruction location, and extent of vascular occlusion are associated with the clinical severity of patients with PE [12, 13] and the risk of complications [3, 4]. Contrary to these reports, our study found no correlation between the Qanadli score and the APACHE II score at the time of ECMO initiation in patients with high-risk PE. This finding suggests that in this specific population, the extent of pulmonary arterial obstruction may not directly reflect the clinical severity. These results should be interpreted with caution and warrant validation through larger, multicenter studies.

The pathophysiology of pulmonary hypertension in acute PE involves not only mechanical obstruction by thrombi but also functional contributors such as hypoxemia, hypercapnia, and acidosis [14, 15]. Moreover, acute right ventricular failure in these patients is multifactorial, influenced by the severity of pulmonary hypertension, baseline right heart function, and volume status [3, 16]. Therefore, mechanical obstruction is only one component contributing to hemodynamic compromise. In high-risk PE, a comprehensive management strategy addressing all reversible factors is essential to improve outcomes.

ECMO provides substantial cardiopulmonary support, improving gas exchange and circulatory function while allowing time for diagnostic evaluation and targeted treatment [17–19]. In this study, 79 patients with suspected high-risk PE underwent CTPA during ECMO support. Of these, 63 were confirmed to have high-risk PE, while the remaining 16 had alternative diagnoses. This underscores that ECMO facilitates hemodynamic stabilization prior to definitive imaging, enabling accurate etiological assessment in cases of undifferentiated shock. Furthermore, ECMO contributed to rapid cardiopulmonary stabilization, ameliorating functional contributors to pulmonary hypertension such as hypoxemia, acidosis, and hypercapnia.

In the Pearl River Delta region of China, VA-ECMO is promptly initiated in cases of suspected high-risk PE with persistent hypotension (SBP < 90 mmHg for > 30 minutes despite maximal vasopressor support) or cardiac arrest. When ECMO is declined, urgent pulmonary revascularization strategies-such as systemic thrombolysis or anticoagulation-are implemented (Supplementary Fig. 2). To ensure comparability between groups, patients were excluded if they maintained SBP > 90 mmHg with vasopressors, thereby balancing baseline illness severity. Although missing data occurred in some critically ill patients (e.g., cardiac arrest), appropriate statistical imputation was applied. Subsequent analyses showed no significant differences in baseline characteristics or key severity indicators between the two groups before initial intervention (Table 1 and Table 2), supporting adequate group matching.

Notably, patients receiving only heparin anticoagulation-which is not considered an urgent revascularization strategy-were excluded. All 38 patients received thrombolytic therapy, which carries an inherently high risk of bleeding. In contrast, only 12 patients in the ECMO group received intravenous thrombolysis. Eight who received ECMO after failed thrombolysis remained in the thrombolysis group, as thrombolysis was their first-line therapy. Therefore, the incidence of fatal bleeding was significantly higher in the thrombolysis group (34.2%) compared to the ECMO group. Our findings provide valuable clinical insights: achieving early hemodynamic stabilization through ECMO, followed by precise diagnosis and individualized revascularization strategies, may contribute to reducing in-hospital mortality rates among these patients. Notably, our study revealed no statistically significant difference in patients’ in-hospital mortality when ECMO was employed as the prioritized initial treatment modality.

Several limitations must be acknowledged. First, this was a retrospective study with a small sample size. Second, excluding patients who received only heparin monotherapy and reclassifying patients who required rescue ECMO may introduce selection bias. Nonetheless, all participating centers employed consistent criteria for ECMO implantation and revascularization, supporting the internal validity of our comparisons. Despite these limitations, key points deserve emphasis: 1) High-risk PE is rare but highly lethal, and randomized controlled trials are unfeasible for it. 2) To our knowledge, this multicenter analysis is the largest reported cohort of suspected high-risk PE patients treated with ECMO. While the study has inherent limitations, its core findings remain clinically relevant for high-risk PE management.

In conclusion, for patients with suspected high-risk PE who are unable to undergo CTPA due to hemodynamic instability, although ECMO is considered to have potential benefits, our study found that when ECMO is used as the prioritized initial treatment strategy, there is no statistical difference in the in-hospital mortality of the patients. Given the retrospective nature and limited sample size of this study, further validation through larger cohorts is warranted.

The data supporting the findings of this study were obtained from the clinical case management systems of four ECMO medical centers in the Pearl River Delta region. Due to the retrospective nature of this study, some data may be incomplete. To protect patient privacy and comply with ethical regulations, the original data are not publicly available. However, de-identified datasets may be made available to qualified researchers upon reasonable request for legitimate medical research purposes, subject to review and approval by the relevant institutional review boards (IRBs) and data custodians. Requests for data access should be directed to the corresponding author, who will coordinate with the participating medical centers to facilitate the permission process. Use of the requested data will require adherence to applicable ethical standards and patient privacy laws, and may necessitate the signing of a data use agreement.​

All authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.​

This study was approved by the Ethics Committee on Clinical Scientific Research and Laboratory Animal Care of Zhongshan People’s Hospital (Approval Number: 2025-031). Given the retrospective design of the study and the use of de-identified patient data, the requirement for written informed consent from individual patients was waived by the above-mentioned IRB. This study was conducted in strict accordance with the ethical principles outlined in the Declaration of Helsinki and its subsequent amendments.​

A flow chart of the treatment strategies applied in the study population, with outcomes and causes of death. Heparin therapy does not qualify as an urgent pulmonary revascularization strategy for high-risk pulmonary embolism (PE). Therefore, 17 patients who received heparin monotherapy (15 died, 2 survived) were excluded from the analysis. 8 patients received ECMO following failed thrombolysis, however, since ECMO was not the first-line treatment for these individuals (thrombolysis was the initial intervention), they were retained in the non-ECMO group. Thrombolysis protocol: intravenous rt-PA (50 ~ 100mg) administered within 30 minutes. Thrombolysis failure, as judged by persistent clinical instability (systolic blood pressure < 90 mmHg or cardiac arrest) after 6h. 79 initially received ECMO treatment (ECMO group), and 63 of these 79 patients were diagnosed with high-risk PE via CTPA under ECMO, 46 patients (46/79, 58.2%) survived to discharged. Refractory shock during ECMO was the leading cause of mortality. Abbreviations: PE, Pulmonary Embolism; ECMO, Extracorporeal Membrane Oxygenation; CPH, Chronic Pulmonary Hypertension; HIE, Hypoxic-Ischemic Encephalopathy.

Linear regression analysis of the correlation between pulmonary artery thrombus burden and clinical severity. This scatter plot demonstrates the relationship between the Qanadli Score (used to quantify thrombus burden on CTPA) and the APACHE II Score (used to assess overall disease severity) in the ECMO-treated cohort (n = 63). Individual data points represent individual patients. The solid black line denotes the linear regression fit, with the gray shaded band indicating the 95% confidence interval of the regression. No significant correlation was observed (p = 0.39). Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; CTPA, computed tomographic pulmonary angiography; ECMO, Extracorporeal Membrane Oxygenation.

Diagnostic and Management Algorithms for Suspected High-Risk PE. Diagnostic pathway: Bedside TTE is used to assess RV dysfunction. Absence of RV dysfunction prompts investigation for alternative causes of shock in the patient. If RV dysfunction is present, the feasibility of CTPA should be evaluated. If CTPA is feasible, it is performed directly. If CTPA is not feasible (e.g., due to hemodynamic instability), ECMO is initiated to facilitate CTPA performance. A positive CTPA result confirms high-risk PE, while a negative CTPA result warrants further investigation for alternative etiologies.