Critical review · Mar 29, 2026
HI-PEITHO: ultrasound-facilitated catheter-directed fibrinolysis versus anticoagulation in intermediate-risk pulmonary embolism
Rosenfield K, Klok FA, Piazza G, Sharp ASP, Ní Áinle F, Jaff MR, et al. Ultrasound-Facilitated, Catheter-Directed Fibrinolysis for Acute Pulmonary Embolism. N Engl J Med. 2026;394(20):1979-1990.
doi:10.1056/NEJMoa2516567 · Open-label multinational adaptive randomised trial with blinded adjudication (NCT04790370; funded by Boston Scientific), n = 544
Summary
HI-PEITHO randomised 544 patients with intermediate-risk pulmonary embolism to ultrasound-facilitated catheter-directed fibrinolysis (USAT, alteplase) plus anticoagulation or anticoagulation alone. The primary endpoint was the composite of PE-related death, cardiorespiratory decompensation or collapse, and symptomatic PE recurrence within 7 days.
The primary endpoint occurred in 4.0% vs 10.3% (RR 0.39; 95% CI 0.20 to 0.77; P = 0.005).
| Component | USAT | Anticoagulation | RR |
|---|---|---|---|
| PE-related death | 1.1% | 0.4% | 3.0 |
| Cardiorespiratory decompensation | 3.7% | 10.3% | 0.4 |
| PE recurrence | 0.4% | 0.4% | 1.0 |
Major bleeding at 72 h was 3.7% vs 1.5% (RR 2.5). At 7 days it was 4.1% vs 2.2% (P = 0.32). All-cause death at 30 days was 1.8% vs 1.1% (RR 1.7).
Strengths
- Randomised, with blinded outcome adjudication.
- Answers whether low-dose local fibrinolysis reduces early decompensation against anticoagulation alone.
Methodological issues
- A result that depends on one component. PE-related death goes the wrong way (RR 3.0) and recurrence is equal (RR 1.0). The composite is positive because of cardiorespiratory decompensation.
- Performance bias. Of 28 events in the control group, 14 were NEWS ≥ 9 alone, a vital-sign score. The trial is open-label, and 71% of the intervention group went to the ICU against 50% of controls. With closer monitoring, deterioration is treated before NEWS reaches 9. On the general ward, controls accumulate high scores before escalation.
- Underpowered safety signal. Major bleeding at 72 h (RR 2.5) and 30-day death (RR 1.7) were not significant, but the trial was not powered for them. Both estimates point against the intervention.
- Comparator. Systemic lysis already prevents decompensation (PEITHO, tenecteplase, NEJM 2014). The clinical question is whether about 17 mg of catheter-delivered alteplase beats reduced-dose IV alteplase (about 50 mg). HI-PEITHO compared against anticoagulation alone and does not answer it.
- Ultrasound component not tested. In SUNSET sPE (Avgerinos, JACC Cardiovasc Interv 2021), 81 patients with submassive PE were randomised to USAT or standard catheter-directed thrombolysis. Thrombus reduction did not differ, and RV/LV recovery was better without ultrasound (P = 0.01). OPTALYSE showed that about 4 mg per catheter works as well as higher doses.
Opinion
What HI-PEITHO shows: low-dose catheter-delivered fibrinolysis reduces, at 7 days, a composite driven by decompensation, against anticoagulation alone. What it does not show: a mortality benefit, superiority over reduced-dose IV lysis, or a need for the ultrasound component. Given SUNSET sPE and OPTALYSE, the more likely reading is that any low-dose local alteplase beats anticoagulation alone, without the EkoSonic device.
Three ongoing trials should complete the answer: PEITHO-3 (reduced-dose IV lysis vs anticoagulation), PE-TRACT (platform-agnostic catheter therapy vs anticoagulation) and PEERLESS II (large-bore thrombectomy vs anticoagulation). Together they should tell us whether, how and with what to intervene.