Publication

• Title: Catheter-Directed Thrombolysis in Intermediate-High–Risk Pulmonary Embolism
• Acronym: PRAGUE-26
• Year: 2026
• Journal published in: New England Journal of Medicine
• Citation: Kroupa J, Radvan M, Mrozek J, et al. Catheter-directed thrombolysis in intermediate-high–risk pulmonary embolism. N Engl J Med. Published online August 31, 2026.

Context & Rationale

• Background

  • Patients with acute pulmonary embolism who remain haemodynamically stable but have right ventricular dysfunction and myocardial injury are at increased risk of early deterioration despite anticoagulation.
  • Full-dose systemic fibrinolysis reduced haemodynamic decompensation in PEITHO but did not reduce mortality and caused substantially more extracranial major bleeding and haemorrhagic stroke, preventing routine use in intermediate-risk pulmonary embolism.¹
  • Catheter-directed thrombolysis aims to deliver a much smaller alteplase dose directly into the pulmonary arteries, potentially preserving rapid reperfusion while reducing systemic bleeding.
  • Earlier studies mainly evaluated right ventricular size, pulmonary pressures or clot burden rather than clinical outcomes. A preceding Czech pilot trial found more rapid improvement in right ventricular measurements with conventional catheter-directed thrombolysis, without an evident safety signal, but was too small for clinical events.²
  • Before PRAGUE-26, no adequately sized randomised trial had shown that conventional catheter-directed thrombolysis without ultrasound facilitation improved clinical outcomes compared with anticoagulation alone.
• Research Question/Hypothesis

  • In haemodynamically stable adults with intermediate-high–risk acute pulmonary embolism, would conventional catheter-directed low-dose alteplase plus anticoagulation reduce death, recurrent pulmonary embolism, or cardiorespiratory decompensation or collapse within seven days compared with anticoagulation alone?
• Why This Matters

  • Routine anticoagulation with rescue reperfusion after deterioration may expose patients to avoidable shock, respiratory failure, cardiac arrest or complex rescue treatment.
  • A beneficial low-dose catheter strategy could shift treatment earlier, before haemodynamic collapse, provided that any reduction in deterioration outweighs intracranial, access-site and other bleeding risks.
  • Conventional infusion catheters are simpler and less expensive than proprietary ultrasound-facilitated systems, so a clinical benefit could be relevant to a wider range of interventional centres.

Design & Methods

• Research Question: Whether conventional catheter-directed thrombolysis with low-dose alteplase plus anticoagulation was superior to anticoagulation alone for preventing a seven-day composite of all-cause death, recurrent pulmonary embolism, or cardiorespiratory decompensation or collapse.
• Study Type: Investigator-initiated, phase 4, multicentre, open-label, active-controlled, parallel-group, randomised superiority trial conducted at 11 tertiary cardiovascular centres in the Czech Republic between October 2022 and March 2026.
• Population:
  • Adults aged 18–80 years with computed tomography-confirmed proximal pulmonary embolism involving at least one main or lobar pulmonary artery and symptom onset less than 14 days before randomisation.
  • Participants were haemodynamically stable, had a simplified Pulmonary Embolism Severity Index score of at least 1, right ventricular dysfunction defined by a right-to-left ventricular end-diastolic diameter ratio of at least 0.9 on echocardiography or computed tomography, and an elevated cardiac troponin or natriuretic peptide level.
  • Patients could be admitted directly to a trial centre or transferred from a referring hospital; 49.5% were transferred.
  • Important exclusions were active clinically significant bleeding; previous haemorrhagic stroke; ischaemic stroke or transient ischaemic attack within six months; cranial trauma within three months; major surgery within seven days; active cancer or another severe illness with expected survival below two years; haemoglobin below 80 g/L; platelet count at or below 100 × 10⁹/L; international normalised ratio above 2.0; creatinine above 200 μmol/L; pregnancy or breastfeeding; and floating thrombus traversing a patent foramen ovale.
• Intervention:
  • Conventional catheter-directed thrombolysis using standard 4-French valved infusion catheters with 10-cm infusion segments, generally introduced through femoral venous access.
  • Alteplase 1 mg was given as a bolus through each catheter, followed by 1 mg/hour per catheter for nine hours: a total of 10 mg for unilateral and 20 mg for bilateral pulmonary embolism.
  • Reduced-intensity intravenous unfractionated heparin was continued during alteplase infusion, targeting an activated partial-thromboplastin time of 50–60 seconds; after catheter removal, full therapeutic anticoagulation was resumed without interruption.
  • The procedure was intended to start within three hours of randomisation. Transition to oral anticoagulation was permitted no earlier than 24 hours after randomisation.
• Comparison:
  • Therapeutic unfractionated heparin or low-molecular-weight heparin without planned reperfusion.
  • Transition to oral anticoagulation was permitted after 24 hours at the treating physician’s discretion.
  • Rescue systemic or catheter-directed reperfusion and other escalation therapies remained available if clinical deterioration occurred.
• Blinding: Patients, treating clinicians and investigators were unblinded. The clinical events committee adjudicated prespecified outcomes but was also aware of treatment assignment; its members were site investigators. This is important because initiation of non-invasive ventilation, intubation and rescue therapy included clinical judgement.
• Statistics:
  • A total of 558 patients was required to detect a 4.5-percentage-point absolute reduction in the seven-day primary outcome, from 6.0% with standard care to 1.5% with catheter-directed thrombolysis, with 80% power and a two-sided alpha of 0.05.
  • O’Brien–Fleming alpha spending was applied for interim monitoring; after two completed interim analyses and the final analysis, P<0.047 was the threshold for the primary outcome.
  • The primary analysis followed the intention-to-treat principle and included all 558 randomised patients. A prespecified per-protocol analysis was supportive.
  • Fisher’s exact test was used for the primary and other categorical outcomes, and the Mann–Whitney test for continuous outcomes. Secondary-outcome confidence intervals were not adjusted for multiplicity.
• Follow-Up Period: The primary outcome was assessed through seven days; bleeding, stroke, serious adverse events, echocardiography, functional status and hospital use were reported through 30 days. Follow-up to 12 and 24 months is ongoing.

Key Results

This trial was not stopped early. It enrolled the planned 558 participants. The data and safety monitoring board recommended continuation after the 25% and 50% interim reviews; the planned 75% analysis was not performed because recruitment finished before monitoring and outcome adjudication for that milestone were completed.

• PRAGUE-26 met its primary superiority criterion: the seven-day composite fell from 6.8% to 0.7% (RR 0.10; 95% CI 0.02 to 0.44; P<0.001), almost entirely because cardiorespiratory decompensation or collapse was less frequent.
• The trial did not establish a mortality benefit. Four control patients died within seven days, but overall event counts were small and one thrombolysis recipient later died following intracranial haemorrhage.
• Clinically relevant and major bleeding rates were similar, but two intracranial haemorrhages occurred with catheter-directed thrombolysis and none with standard care; this uncommon but severe harm remains insufficiently quantified.

Internal Validity

• Randomisation and Allocation: Assignment was performed centrally through a web-based system and balanced by age, sex, unilateral or bilateral pulmonary embolism, time from diagnostic computed tomography and direct admission versus transfer. The statistical analysis plan describes dynamic minimisation: assignment was random when a stratum was balanced and otherwise went to the underrepresented group. Central implementation probably protected concealment, although the method is not fully captured by the report’s description of “simple stratified randomisation”.
• Completeness of the Primary Analysis: All 558 randomised patients were included, and seven-day follow-up was complete in both groups. There were no post-randomisation exclusions from the primary intention-to-treat analysis.
• Protocol Deviations: Twelve patients were excluded from the supportive per-protocol analysis: 6 in each group. Deviations involved biomarker eligibility, platelet count, international normalised ratio or creatinine. The per-protocol primary result was virtually unchanged.
• Thirty-Day Missing Data: Follow-up at 30 days was available for 278/280 (99.3%) thrombolysis recipients and 263/278 (94.6%) controls. This does not affect the complete seven-day primary outcome but weakens comparisons of 30-day mortality, bleeding and functional status.
• Performance and Detection Bias: The trial was fully open-label, and the clinical events committee was also unblinded. Death, cardiac arrest, shock and ECMO are comparatively objective, but initiation of non-invasive ventilation, intubation, rescue reperfusion and hospital discharge can be influenced by knowledge of treatment.
• Intervention Delivery: Catheter placement and alteplase infusion were successful in 279/280 patients (99.6%). Median time from randomisation to treatment was 76 minutes (IQR 49–126), and 250/280 (89.3%) began within three hours.
• Dose: Delivered alteplase closely matched the intended regimen: 19.9±1.2 mg for bilateral and 10.0±0.0 mg for unilateral pulmonary embolism, infused over 9.3±0.95 hours. The trial therefore evaluated a reproducible low-dose strategy, although it did not compare alternative doses or shorter infusions.
• Separation of the Variable of Interest: Nearly all intervention patients received catheter-directed alteplase, whereas controls initially received anticoagulation alone. Before randomisation, unfractionated heparin use was 52.0% vs 51.1% and low-molecular-weight heparin use 48.0% vs 48.6%. At 24 hours, unfractionated heparin use was 40.1% vs 31.9% and low-molecular-weight heparin use 56.3% vs 65.2%, reflecting the procedural anticoagulation strategy.
• Crossover and Rescue Therapy: One intervention procedure failed because pulmonary-artery access could not be obtained. In the control group, 12 patients (4.3%) received rescue reperfusion for persistent symptoms, hypoxaemia or right ventricular dysfunction without meeting the primary outcome, in addition to rescue treatment after some primary events. Such rescue treatment would tend to reduce later between-group differences, while the decision to escalate remained susceptible to open-label practice.
• Baseline Characteristics: Median age was 64 years in both groups, women comprised 41.1% vs 40.6%, median right-to-left ventricular ratio was 1.13 vs 1.14, and median Pulmonary Embolism Severity Index was 94 vs 93. Previous venous thromboembolism was more common with standard care (32.4% vs 22.5%; standardised difference 22.3%), and controls had modestly more chronic heart failure, chronic kidney disease and very-high-risk Pulmonary Embolism Severity Index class V disease.
• Population Heterogeneity: Eligibility created a recognisable right-heart-strain phenotype, but clinical severity ranged from Pulmonary Embolism Severity Index class I to V. Prespecified subgroup event counts were extremely small, no interaction P values were reported, and the apparent consistency should not be interpreted as proof of equal benefit across age, sex, body-mass index or right ventricular severity.
• Timing: Median time from hospital admission to randomisation was 17.8 hours (IQR 6.3–22.5) vs 16.9 hours (IQR 5.8–22.4), and computed tomography had been performed more than 24 hours before randomisation in 18.2% vs 18.3%. Rapid post-randomisation treatment was excellent, but the enrolled population necessarily survived diagnostic assessment, consent and sometimes interhospital transfer without collapsing.
• Outcome Assessment: The primary composite and deterioration criteria were prespecified and clinically relevant. The control group’s 15 deterioration events included 6 episodes of shock and 7 requiring invasive or non-invasive ventilation, so the result was not driven solely by the three NEWS ≥9 events. Nevertheless, components differed markedly in importance and several depended on clinician action.
• Statistical Rigour: The planned sample was reached, the observed control event rate approximated the assumed rate, the result crossed the interim-adjusted threshold by a wide margin, and intention-to-treat and per-protocol analyses agreed. Secondary results were exploratory, confidence intervals were not multiplicity-adjusted, and sparse events make safety and subgroup estimates imprecise.

Conclusion on Internal Validity: Overall internal validity is moderate. Complete follow-up, central randomisation, excellent treatment fidelity, a large primary effect and concordant per-protocol analysis support a real reduction in early deterioration, but open-label management, unblinded adjudication, a non-hierarchical composite and limited precision for death and intracranial haemorrhage prevent a definitive claim of overall net clinical benefit.

External Validity

• Population Representativeness: Median age was 64 years, 40.9% were women, median body-mass index was approximately 31, 98.2% had bilateral pulmonary embolism and 49.5% were transferred from another hospital. These patients resemble a clinically important European intermediate-risk population, but screening logs were not maintained, so the proportion assessed and reasons for non-enrolment are unknown.
• Clinical Severity: Approximately 65% had a simplified Pulmonary Embolism Severity Index score of 1. By the full Pulmonary Embolism Severity Index, 34.2% were class I or II, 39.8% class III and 26.0% class IV or V. The result applies to stable patients with objective right-heart strain, not to patients already in shock.
• Eligibility Definition: Elevated troponin was not mandatory if natriuretic peptide was elevated; 39 participants were troponin-negative and natriuretic-peptide-positive. The population therefore extends beyond the strictest troponin-positive intermediate-high–risk phenotype used in some trials and guidelines.
• Excluded High-Risk Groups: Patients older than 80 years, those with major bleeding risk, recent surgery or cerebrovascular events, thrombocytopenia, severe renal dysfunction, pregnancy, and serious illness limiting two-year survival were excluded. The benefit–harm balance may be less favourable in these common high-risk groups.
• Setting: All sites were Czech tertiary cardiovascular centres with catheter expertise, immediate access to rescue reperfusion and established referral networks. The two largest centres recruited 177 (31.7%) and 168 (30.1%) participants, respectively, making the results particularly dependent on experienced high-volume systems.
• Procedural Applicability: The use of standard, non-ultrasound 4-French infusion catheters improves technological and economic accessibility, but implementation still requires rapid transfer, an angiography suite, skilled operators, monitored thrombolytic infusion and careful anticoagulation and blood-pressure management.
• Treatment Scope: PRAGUE-26 does not compare conventional catheter-directed thrombolysis with ultrasound-facilitated thrombolysis, reduced-dose systemic fibrinolysis or mechanical thrombectomy, and it does not establish benefit for lower-risk pulmonary embolism.
• Duration of Benefit: Only early clinical and 30-day recovery data are available. Whether prevention of transient deterioration improves long-term functional capacity, quality of life, recurrent events, chronic thromboembolic pulmonary hypertension or survival remains unknown.

Conclusion on External Validity: Generalisability is moderate for haemodynamically stable adults aged 18–80 years with proximal pulmonary embolism, right ventricular dysfunction and biomarker elevation treated in experienced tertiary networks. It is limited for older or bleeding-prone patients, non-tertiary hospitals, overt shock, lower-risk pulmonary embolism and systems unable to deliver catheter therapy rapidly and safely.

Strengths & Limitations

• Strengths:
  • Investigator-initiated and publicly funded, without device-industry sponsorship.
  • Largest randomised clinical-outcome evaluation of conventional, non-ultrasound catheter-directed thrombolysis against anticoagulation alone.
  • Multicentre recruitment across 11 tertiary cardiovascular centres.
  • Clinically meaningful primary outcome rather than a purely radiological or haemodynamic surrogate.
  • Complete seven-day follow-up and inclusion of every randomised participant in the primary analysis.
  • Excellent procedural success, rapid intervention delivery and close adherence to the intended alteplase dose.
  • Consistent intention-to-treat and per-protocol primary analyses.
  • Detailed reporting of bleeding using BARC, GUSTO and ISTH definitions.
• Limitations:
  • Open-label treatment with an unblinded clinical events committee.
  • The primary effect was driven by cardiorespiratory deterioration rather than established mortality or recurrent-pulmonary-embolism benefit.
  • The non-hierarchical composite gave equal statistical weight to death, recurrence, shock, ventilation and NEWS ≥9.
  • Initiation of non-invasive ventilation, rescue reperfusion and discharge were susceptible to clinician behaviour.
  • No screening logs were retained, preventing assessment of selection and recruitment fractions.
  • Single-country, predominantly White population; the two largest centres recruited 177 and 168 participants, respectively.
  • Patients older than 80 years and many with greatest bleeding risk were excluded.
  • Two intracranial haemorrhages occurred after catheter-directed thrombolysis; the trial was underpowered to quantify rare serious harms.
  • Thirty-day follow-up was less complete in the control group.
  • Long-term functional, quality-of-life, chronic thromboembolic pulmonary hypertension and survival outcomes remain pending.
  • The trial does not determine which reperfusion modality is preferable.

Interpretation & Why It Matters

• Primary Clinical Meaning

  In selected haemodynamically stable patients with proximal pulmonary embolism, right ventricular dysfunction and biomarker elevation, early conventional catheter-directed low-dose alteplase substantially reduced the seven-day risk of a composite dominated by cardiorespiratory deterioration.
• Prevention, Not Rescue

  The treatment strategy acts before overt collapse. Its clinical value is avoidance of shock, respiratory support and complex rescue escalation rather than treatment of established high-risk pulmonary embolism.
• Not a Mortality Trial

  Four control patients and no thrombolysis recipients died within seven days, but the small number of deaths and the later fatal intracranial haemorrhage in the thrombolysis group mean that a survival benefit has not been established.
• Safety Trade-Off

  Aggregate clinically relevant and major bleeding did not increase, but two intracranial haemorrhages occurred after catheter-directed thrombolysis. For an intervention applied prophylactically to stable patients, even an uncommon catastrophic harm has major weight.
• Independent Replication

  The direction of effect closely matches HI-PEITHO, despite use of conventional rather than ultrasound-facilitated catheters and a somewhat broader risk phenotype. Together, the trials make an early deterioration-prevention effect increasingly credible.
• Practice Implication

  PRAGUE-26 supports discussion of early catheter-directed thrombolysis for carefully selected intermediate-high–risk patients in experienced pulmonary embolism networks, but does not justify routine treatment of every stable patient with right ventricular strain.

Controversies & Other Evidence

• Composite Outcome: Death, recurrence, shock, ECMO, ventilation and NEWS ≥9 were treated equally, although their clinical importance differs. The result was driven by cardiorespiratory decompensation or collapse: 2 vs 15 patients. Reassuringly, the control events included 6 episodes of shock and 7 requiring invasive or non-invasive ventilation, so the treatment effect cannot be dismissed as a NEWS-only signal.
• Open-Label Ascertainment: Treating teams knew allocation, and the clinical events committee comprised unblinded site investigators. This is a larger concern than in a conventional PROBE design because decisions to initiate non-invasive ventilation, intubate or escalate treatment contributed directly or indirectly to important outcomes.
• Randomisation Description: The trial report describes a computer-generated, simple stratified process, whereas the statistical analysis plan describes dynamic minimisation with deterministic assignment to the underrepresented group whenever a randomisation stratum was imbalanced. Central web allocation probably preserved concealment, but the discrepancy should have been explained explicitly.
• Definition of Intermediate-High Risk: Troponin elevation was not compulsory when natriuretic peptide was elevated; 39 participants were troponin-negative and natriuretic-peptide-positive. This broadens applicability but complicates direct comparison with trials requiring both right ventricular dysfunction and myocardial injury.
• Rescue Reperfusion Endpoint: The definition of first-line treatment failure was amended in March 2025 to include catheter-directed as well as systemic thrombolysis, because local rescue had become common. The change occurred before database lock and outcome analysis and did not alter the primary outcome, but the 0% vs 4.3% result for this secondary endpoint should be interpreted as supportive rather than confirmatory.
• Intracranial Haemorrhage: Two intracranial haemorrhages occurred after catheter-directed thrombolysis and none with standard care. One followed a hypertensive episode soon after alteplase; the other occurred on day 5 with documented low-molecular-weight heparin overdose and was ultimately fatal. Their different mechanisms do not remove the safety signal: low-dose local thrombolysis still occurs within a complex anticoagulation pathway.
• Comparison with Systemic Fibrinolysis: PEITHO showed the same broad efficacy pattern with systemic tenecteplase—less early haemodynamic collapse without proven mortality benefit—but extracranial major bleeding increased from 1.2% to 6.3% and stroke from 0.2% to 2.4%.¹ PRAGUE-26 suggests that lowering the alteplase dose and delivering it locally may improve this benefit–harm balance, but the intracranial events show that risk is not abolished.
• HI-PEITHO: The contemporaneous HI-PEITHO trial enrolled a more clinically distressed population and found a similar reduction in seven-day pulmonary-embolism-related death, cardiorespiratory deterioration/collapse or recurrent pulmonary embolism: 4.0% vs 10.3%; RR 0.39; 95% CI 0.20 to 0.77; P=0.005.³ Its effect was also dominated by deterioration rather than death or recurrence, and major bleeding was numerically higher with intervention, although no intracranial haemorrhage occurred.
• What PRAGUE-26 Adds to HI-PEITHO: PRAGUE-26 independently reproduces the early clinical signal using ordinary infusion catheters, without proprietary ultrasound facilitation, in patients defined largely by standard risk markers rather than requiring two additional features of distress. This strengthens biological and clinical plausibility but also raises the question of whether the broader population exposes more patients to bleeding for a benefit concentrated in an unidentified higher-risk subset.
• Editorial Caution: The HI-PEITHO editorial emphasised that reducing a deterioration composite is not equivalent to improving survival or long-term health, and that low event counts cannot establish bleeding neutrality.⁴ Those cautions apply equally to PRAGUE-26 and are sharpened by its two intracranial haemorrhages.
• Long-Term Benefit: Long-term PEITHO follow-up found no reduction in mortality, persistent dyspnoea, right ventricular dysfunction or chronic thromboembolic pulmonary hypertension after systemic fibrinolysis.⁵ Whether the faster recovery seen with catheter-directed treatment translates into durable patient-centred benefit is therefore a major unanswered question.
• Guidelines: The 2026 multisociety American guideline preceded both HI-PEITHO and PRAGUE-26 and did not recommend routine catheter intervention for most intermediate-risk pulmonary embolism; it allowed selective use in higher-risk patients with transient hypotension or malperfusion.⁶ No guideline published after PRAGUE-26 has yet incorporated its findings.
• Mechanical Thrombectomy: STORM-PE showed greater 48-hour right ventricular recovery with computer-assisted vacuum thrombectomy than anticoagulation alone but enrolled only 100 patients and was not powered for clinical events.⁷ PEERLESS favoured large-bore mechanical thrombectomy over catheter-directed thrombolysis on a hierarchical composite driven by clinical deterioration, bailout treatment and ICU use, without lower mortality or bleeding; it did not compare either intervention with anticoagulation alone.⁸
• Subsequent Evidence: PRAGUE-26 was published on August 31, 2026. No subsequent randomised trial, meta-analysis or guideline incorporating its results is yet available; its two-year follow-up will be important.

Summary

• PRAGUE-26 randomised 558 haemodynamically stable adults with proximal pulmonary embolism, right ventricular dysfunction and biomarker elevation to conventional catheter-directed low-dose alteplase plus anticoagulation or anticoagulation alone.
• The seven-day composite of all-cause death, recurrent pulmonary embolism, or cardiorespiratory decompensation or collapse occurred in 0.7% vs 6.8%; RR 0.10; 95% CI 0.02 to 0.44; P<0.001.
• The benefit was driven principally by fewer episodes of cardiorespiratory deterioration: 0.7% vs 5.4%; RR 0.13; 95% CI 0.03 to 0.57. Mortality and recurrent pulmonary embolism were too uncommon for reliable conclusions.
• Major and clinically relevant bleeding rates were similar, but intracranial haemorrhage occurred in 2 patients (0.7%) after catheter-directed thrombolysis and none after standard care; one event was ultimately fatal.
• The trial supports early conventional catheter-directed thrombolysis as an option for carefully selected patients in experienced centres, but long-term benefit, precise safety, optimal selection and comparative effectiveness against other reperfusion strategies remain unresolved.

Overall Takeaway

PRAGUE-26 is an important practice-shaping trial because it independently shows that early, conventional catheter-directed low-dose thrombolysis can prevent short-term deterioration in selected intermediate-high–risk pulmonary embolism. It does not establish a mortality benefit or eliminate the risk of catastrophic bleeding; the appropriate conclusion is selective adoption within experienced pulmonary embolism networks, not routine intervention for every stable patient with right-heart strain.

Bibliography

• 1.Meyer G, Vicaut E, Danays T, et al; PEITHO Investigators. Fibrinolysis for patients with intermediate-risk pulmonary embolism. N Engl J Med. 2014;370(15):1402-1411.
• 2.Kroupa J, Buk M, Weichet J, et al. A pilot randomised trial of catheter-directed thrombolysis or standard anticoagulation for patients with intermediate-high risk acute pulmonary embolism. EuroIntervention. 2022;18(8):e639-e646.
• 3.Rosenfield K, Klok FA, Piazza G, et al; HI-PEITHO Investigators. Ultrasound-facilitated, catheter-directed fibrinolysis for acute pulmonary embolism. N Engl J Med. 2026;394(20):1979-1990.
• 4.Spyropoulos AC, Vedantham S. Advanced therapy for intermediate-risk pulmonary embolism. N Engl J Med. 2026;394(20):2052-2054.
• 5.Konstantinides SV, Vicaut E, Danays T, et al; PEITHO Investigators. Impact of thrombolytic therapy on the long-term outcome of intermediate-risk pulmonary embolism. J Am Coll Cardiol. 2017;69(12):1536-1544.
• 6.Creager MA, Barnes GD, Giri J, et al. 2026 AHA/ACC/ACCP/ACEP/CHEST/SCAI/SHM/SIR/SVM/SVN guideline for the evaluation and management of acute pulmonary embolism in adults: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2026;153(12):e977-e1051.
• 7.Lookstein RA, Konstantinides SV, Weinberg I, et al; STORM-PE Trial Investigators. Randomized controlled trial of mechanical thrombectomy with anticoagulation versus anticoagulation alone for acute intermediate-high risk pulmonary embolism: primary outcomes from the STORM-PE trial. Circulation. 2026;153(1):21-34.
• 8.Jaber WA, Gonsalves CF, Stortecky S, et al; PEERLESS Committees and Investigators. Large-bore mechanical thrombectomy versus catheter-directed thrombolysis in the management of intermediate-risk pulmonary embolism: primary results of the PEERLESS randomized controlled trial. Circulation. 2025;151(5):260-273.