Clinical Question
Among hemodynamically stable patients with intermediate-high–risk acute pulmonary embolism, does conventional catheter-directed thrombolysis with low-dose alteplase reduce early clinical deterioration compared with anticoagulation alone?
TL;DR
Catheter-directed thrombolysis reduced the 7-day composite of death, recurrent pulmonary embolism, or cardiorespiratory decompensation or collapse from 6.8% to 0.7% (RR 0.10, 95% CI 0.02–0.44). The absolute risk reduction was 6.1 percentage points, which translates to a number needed to treat of roughly 17.
The result was driven mainly by fewer episodes of cardiorespiratory decompensation, not a proven reduction in mortality. The trial was open-label, the adjudicators knew treatment assignment, and part of the decompensation definition relied on a National Early Warning Score threshold.
Bleeding rates were similar overall, but two patients in the thrombolysis group had an intracranial hemorrhage. That is the tension at the center of this trial: catheter-directed alteplase appears to prevent early deterioration, but the benefit has to be weighed against a rare and potentially devastating harm.
For carefully selected patients with intermediate-high–risk PE at experienced centers, PRAGUE-26 lowers the threshold for an early PERT discussion. It does not justify routine catheter-directed thrombolysis for everyone with RV strain and a positive troponin.
The Deets
Intermediate-high–risk pulmonary embolism is one of the most uncomfortable categories in acute care. The patient is normotensive, so systemic thrombolysis is difficult to justify. But the RV is strained, biomarkers are positive, and we know a meaningful minority will deteriorate despite anticoagulation.
Our current strategy is usually close monitoring plus rescue reperfusion if the patient worsens. The problem is that rescue therapy often arrives after the physiology has already started to go downhill (the slippery slope).
PRAGUE-26 tested whether intervening earlier could prevent that slide.
The investigators enrolled 558 adults across 11 tertiary cardiovascular centers in the Czech Republic. Patients were 18 to 80 years old, had proximal acute PE, remained hemodynamically stable, had an sPESI score of at least 1, and had both RV dysfunction and an elevated troponin or natriuretic peptide level.
Patients were randomized to anticoagulation alone or anticoagulation plus catheter-directed alteplase. The intervention was deliberately simple: a standard 4-French valved infusion catheter with a 10-cm infusion segment, a 1-mg alteplase bolus through each catheter, and then 1 mg per hour per catheter. Most patients with bilateral PE received about 20 mg over nine hours.
This is one of the most interesting parts of the trial. The investigators did not use an ultrasound-facilitated proprietary catheter. They achieved a clinical result that looks broadly similar to HI-PEITHO using conventional infusion catheters at a similar alteplase dose.
That does not prove catheter delivery is superior to peripheral low-dose alteplase, and PRAGUE-26 did not directly compare conventional with ultrasound-assisted thrombolysis. But it makes the incremental value of expensive ultrasound augmentation even harder to defend without better head-to-head evidence.
The primary outcome was a 7-day composite of death from any cause, recurrent PE, or cardiorespiratory decompensation or collapse. Decompensation included cardiac arrest, shock with end-organ hypoperfusion, ECMO, intubation, new noninvasive ventilation, or a persistent National Early Warning Score of at least 9.
The primary outcome occurred in 2 of 280 patients receiving thrombolysis and 19 of 278 receiving standard care. Cardiorespiratory decompensation or collapse occurred in 2 versus 15 patients and drove most of the difference.
There were no deaths by day 7 in the thrombolysis group and four in the standard-care group. That signal is clinically important, but with only four early deaths, the trial cannot establish a mortality benefit.
Study Snapshot
Investigator-initiated, multicenter, open-label randomized trial
N=558 across 11 tertiary cardiovascular centers in the Czech Republic
Adults 18–80 years old with proximal acute PE and symptom onset within 14 days
Intermediate-high risk required all four criteria:
Criterion | PRAGUE-26 Definition | Bedside Assessment |
|---|---|---|
Hemodynamic status | Hemodynamically stable | No shock or persistent hypotension requiring vasopressors |
Clinical risk | sPESI score ≥1 | One point each for age >80, cancer, chronic cardiopulmonary disease, heart rate ≥110, systolic BP <100 mm Hg, or oxygen saturation <90% |
RV dysfunction | RV-to-LV end-diastolic diameter ratio ≥0.9 | Present on transthoracic echocardiography or CT angiography |
Cardiac biomarker elevation | Elevated troponin or natriuretic peptide | hs-troponin I >53 ng/L in males or >34 ng/L in females; hs-troponin T >14 ng/L; NT-proBNP >600 pg/mL; or BNP >100 pg/mL |
Conventional catheter-directed alteplase plus anticoagulation versus anticoagulation alone
Typical bilateral-PE alteplase dose: approximately 20 mg over nine hours
Primary outcome: death, recurrent PE, or cardiorespiratory decompensation or collapse through day 7
Investigator initiated and supported by Czech public and academic funding; no industry sponsor reported
The population was selected. Patients older than 80 years were excluded, as were those with recent major surgery, substantial renal dysfunction, active clinically significant bleeding, or severe illness with an expected survival under two years. Nearly every patient had bilateral PE, and all were treated in a tertiary cardiovascular network with rapid access to rescue therapy.
What They Found
Primary Outcome
Death, recurrent PE, or cardiorespiratory decompensation or collapse by day 7
Events: 2/280 vs. 19/278
Event rate: 0.7% vs. 6.8%
RR 0.10 (95% CI 0.02–0.44; P<0.001)
Absolute risk reduction: 6.1 percentage points
Number needed to treat: approximately 17
Secondary Outcomes
Outcome | Catheter-Directed Thrombolysis | Standard Care | Effect Estimate |
|---|---|---|---|
Cardiorespiratory decompensation or collapse by day 7 | 2 (0.7%) | 15 (5.4%) | RR 0.13 (95% CI 0.03–0.57) |
Recurrent PE by day 7 | 2 (0.7%) | 1 (0.4%) | RR 1.99 (95% CI 0.18–21.77) |
All-cause death by day 7 | 0 | 4 (1.4%) | Too few events for a reliable estimate |
WHO functional class I at day 30 | 58.2% | 44.6% | Exploratory secondary outcome |
Median hospital stay | 4.3 days | 5.1 days | Exploratory secondary outcome |
The physiology moved in the expected direction. At 24 hours, the median RV-to-LV ratio was 0.94 with thrombolysis and 1.07 with standard care. ICU and hospital stays were also shorter, and more patients were in WHO functional class I at 30 days.
Those results support faster early recovery. They remain secondary outcomes without multiplicity adjustment, so they should not carry the same weight as the primary endpoint.
Safety
Safety Outcome | Catheter-Directed Thrombolysis | Standard Care |
|---|---|---|
Clinically relevant bleeding by day 7 | 13 (4.6%) | 14 (5.0%) |
GUSTO major bleeding by day 7 | 4 (1.4%) | 6 (2.2%) |
ISTH major bleeding by day 7 | 5 (1.8%) | 7 (2.5%) |
Intracranial hemorrhage by day 7 | 2 (0.7%) | 0 |
The overall bleeding numbers are reassuring. They do not prove that catheter-directed thrombolysis is equally safe, because the trial was not powered to exclude clinically meaningful differences in rare events.
Pay attention to the two intracranial hemorrhages.
One followed a hypertensive episode shortly after alteplase.
The other occurred on day 5 in the setting of a documented low-molecular-weight heparin overdose. The second event may not be attributable to alteplase alone, but it still occurred in a patient exposed to thrombolysis and highlights how fragile the safety margin can become when anticoagulation protocols break down.
What Bothers Me
The primary result was not a mortality result. Four patients died by day 7 with standard care and none with thrombolysis, but the trial had too few deaths to establish a reliable survival benefit. Most of the treatment effect came from preventing decompensation.
Some decompensation events were softer than the headline suggests. The composite included a sustained NEWS score of at least 9. That is clinically relevant, but it is not equivalent to cardiac arrest, shock, intubation, or ECMO. The trial does not clearly show us how many primary events represented truly catastrophic deterioration versus worsening physiology captured by a score.
The trial was open-label, and adjudication was not blinded. Treating teams and the clinical-events committee knew which therapy each patient received. That creates room for differential monitoring, escalation, and event classification, particularly when a composite includes physiologic thresholds and decisions to initiate respiratory support.
Rare bleeding harm remains unresolved. Similar aggregate bleeding rates are reassuring, but two intracranial hemorrhages occurred after thrombolysis and none with standard care. A 558-patient trial cannot confidently define the risk of uncommon catastrophic events.
We do not know which part of the intervention mattered. PRAGUE-26 compared catheter-directed alteplase plus anticoagulation with anticoagulation alone. It did not compare conventional catheter delivery with peripheral reduced-dose systemic alteplase, ultrasound-assisted thrombolysis, or mechanical thrombectomy.
Generalizability is limited. The trial took place in one country, excluded patients older than 80 years, did not maintain screening logs, and relied on experienced tertiary cardiovascular centers. The procedural success rate of 99.6% and median treatment start 76 minutes after randomization may be difficult to reproduce broadly.
The 30-day follow-up was also less complete in the standard-care group than in the thrombolysis group—94.6% versus 99.3%. That does not undermine the fully observed 7-day primary endpoint, but it adds uncertainty to later exploratory comparisons.
Dashevsky's Dissection
For patients: Catheter-directed thrombolysis delivers a smaller dose of clot-dissolving medication through a catheter placed in the pulmonary arteries. In this trial, it substantially reduced the risk of early clinical worsening compared with anticoagulation alone.
The procedure has not been proven to improve survival, and it still carries a risk of serious bleeding in the brain. The decision should depend on the patient’s trajectory, bleeding risk, comorbidities, goals, and access to an experienced center rather than the presence of RV strain alone.
For pulmonary and critical care physicians: This trial should change the conversation more than it changes the protocol.
PRAGUE-26 and HI-PIETHO now point in the same direction: early low-dose catheter-directed fibrinolysis can reduce short-term deterioration in selected intermediate-high–risk PE. That makes it harder to dismiss catheter-directed thrombolysis as a therapy supported only by improvements in RV imaging.
Would you opt to treat every normotensive patient with RV strain and a positive troponin with CDT? Probably not. The absolute benefit depends on baseline risk, and the patients most likely to benefit are probably those showing meaningful clinical distress or a worsening trajectory despite anticoagulation (not those who happen to meet a biomarker definition while remaining comfortable and stable).
The practical change is a lower threshold for early PERT activation and transfer planning, which the updates PE guidelines already account for. If we wait for shock, we have already lost the preventive window the trial was designed to test. At the same time, we should explicitly weigh intracranial bleeding risk and avoid letting a favorable composite endpoint become a blanket procedural indication.
Let’s talk about the catheter choice, briefly. PRAGUE-26 used a conventional infusion catheter and produced results similar to an ultrasound-facilitated strategy. Ultrasound may still offer a technical advantage in some settings, but the burden is shifting toward proving that advantage rather than assuming it. Next up would be evaluating low dose alteplase administered peripherally!
This evidence also helps fill a gap left by the recent guidelines mentioned above, which recognized catheter-based therapies but could not draw firm lines between thrombolysis, mechanical thrombectomy, and other reperfusion strategies.
For the health system: The use of a standard infusion catheter (vs ultrasound) may matter economically. If conventional catheters can reproduce the outcomes of proprietary ultrasound-facilitated systems, hospitals should question whether the added device cost buys meaningful clinical benefit.
The larger financial question is whether preventing decompensation and shortening ICU or hospital stays offsets the cost and complexity of an invasive procedure. PRAGUE-26 suggests that it might, but the trial did not include a formal cost-effectiveness analysis and was not designed to compare device strategies (this should be next on the agenda!).
In summary, PRAGUE-26 provides strong randomized evidence that conventional catheter-directed low-dose alteplase can prevent early deterioration in carefully selected patients with intermediate-high–risk pulmonary embolism. The effect is clinically meaningful and consistent with HI-PEITHO, but it remains primarily a “decompensation story” not yet a “mortality story.” For now, the best application is earlier multidisciplinary evaluation, careful patient selection, and a healthy skepticism toward claims that more expensive catheter technology automatically produces better outcomes.
Kroupa J, et al. “Catheter-Directed Thrombolysis in Intermediate-High–Risk Pulmonary Embolism.” New England Journal of Medicine. Published August 31, 2026. doi:10.1056/NEJMoa2608012.
