Imaging-Driven Risk Stratification and Endovascular Decision Pathways in Acute Pulmonary Embolism

Acute pulmonary embolism (PE) is increasingly managed as a dynamic risk continuum in which imaging findings guide therapeutic escalation rather than merely confirm diagnosis. The principal challenge still remains normotensive patients with intermediate–high-risk features, where early right ventricular (RV) dysfunction may precede overt hemodynamic collapse. New trends focus on a trajectory-based model by integrating clinical, laboratory, and standardized imaging parameters into severity categorization. This review critically examines how imaging-derived markers influence risk stratification, escalation timing, and endovascular decision pathways in contemporary PE management. A structured narrative review was conducted focusing on the literature published between January 2020 and January 2026. PubMed/MEDLINE, Scopus, and Web of Science were searched for studies addressing imaging-based risk assessment, catheter-based reperfusion strategies, randomized trials, prospective registries, and guideline documents. Contemporary data consistently demonstrate that catheter directed therapies (CDTs) lead to rapid improvement in RV imaging surrogates and hemodynamic parameters. However, short-term mortality differences are uncommon in predominantly normotensive cohorts. Clinically meaningful signals instead emerge in the reduction in early clinical deterioration, the need for rescue escalation, bleeding optimization, and healthcare resource utilization. Imaging, as standardized reporting of RV strain on computed tomography pulmonary angiography and echocardiography, should be further embedded into escalation algorithms. In modern PE care, imaging functions as a trigger for escalation within multidisciplinary pathways rather than as a passive prognostic marker. CDTs should be interpreted as tools for trajectory modulation in selected intermediate-risk patients rather than mortality-reduction strategies. Future research should integrate imaging phenotyping, dynamic reassessment models, and organizational variables to refine patient selection and optimize outcome-relevant endpoints.