Reconsidering the default: should surgical pulmonary embolectomy have a larger role in high-risk pulmonary embolism at experienced centres?

High-risk pulmonary embolism (PE) remains one of the most time-critical emergencies in cardiovascular medicine, with mortality rates approaching 20-50% despite advances in reperfusion therapy.Although systemic thrombolysis (TL) has occupied the default firstline position in contemporary guidelines for decades, the evidence supporting its superiority over surgical pulmonary embolectomy (SPE) in genuinely high-risk PE remains limited and largely derived from studies including intermediate-risk populations (1).In this context, the study by Ishida and colleagues provides important contemporary data challenging long-standing assumptions regarding the optimal reperfusion strategy in high-risk PE.Ishida and colleagues take a meaningful step toward addressing this (2).Using the Japanese nationwide Diagnosis Procedure Combination (DPC) database covering approximately 70% of acute-care hospital beds across Japan, they identified 2,813 patients with high-risk PE who received either SPE or systemic TL within 48 hours of admission between July 2010 and March 2023.After overlap weighting, SPE was associated with significantly lower in-hospital mortality [22.2% vs. 30.1%;risk difference -8.0%, 95% confidence interval (CI): -13.1% to -2.9%] and a higher rate of favourable neurological outcomes at discharge (72.5% vs. 66.7%;risk difference 5.8%, 95% CI: 0.3% to 11.3%).These results were consistent across propensity score matching, multivariable regression, E-value analysis, and instrumental variable approaches.An absolute mortality reduction of eight percentage points is not trivial.In a condition where in-hospital mortality ranges from 20% to 52% (3,4), a reduction of that magnitude, if real and generalisable, would represent one of the most impactful treatment decisions in acute cardiovascular medicine.The neurological outcome advantage is, if anything, even more compelling.Favourable neurological status at discharge, defined here as a Japan Coma Scale of 0 or 1-3, broadly equivalent to a Cerebral Performance Category of 1 or 2, was meaningfully more common after SPE and persisted to an adjusted risk difference of 12.1% in the instrumental variable analysis.Mechanistically, the observed differences are biologically plausible.TL achieves reperfusion through systemic fibrinolysis but does so at the cost of increased bleeding risk, including intracranial haemorrhage.In contrast, SPE achieves direct mechanical clot extraction and immediate right ventricular decompression without inducing a systemic fibrinolytic state.Meta-analyses of randomised trials have reported intracranial haemorrhage rates of approximately