PEERLESS: Interventional therapies for pulmonary embolism

Cite this article as:
Morgenstern, J. PEERLESS: Interventional therapies for pulmonary embolism, First10EM, February 17, 2025. Available at:
https://doi.org/10.51684/FIRS.140888

The decision between mechanical thrombectomy and catheter directed thrombolysis for pulmonary embolism is not one that most emergency physicians are making. Most (like me) probably don’t even have access to such therapies, which might be a good thing, as both therapies remain completely unproven. However, I thought I would share the PEERLESS trial as a sneak peak into a much much longer discussion of the management of high risk pulmonary embolism coming soon, both on First10EM and EMCases.

The paper

Jaber WA, Gonsalves CF, Stortecky S, Horr S, Pappas O, Gandhi RT, Pereira K, Giri J, Khandhar SJ, Ammar KA, Lasorda DM, Stegman B, Busch L, Dexter DJ 2nd, Azene EM, Daga N, Elmasri F, Kunavarapu CR, Rea ME, Rossi JS, Campbell J, Lindquist J, Raskin A, Smith JC, Tamlyn TM, Hernandez GA, Rali P, Schmidt TR, Bruckel JT, Camacho JC, Li J, Selim S, Toma C, Basra SS, Bergmark BA, Khalsa B, Zlotnick DM, Castle J, O’Connor DJ, Gibson CM; 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 Feb 4;151(5):260-273. doi: 10.1161/CIRCULATIONAHA.124.072364. Epub 2024 Oct 29. PMID: 39470698 NCT05111613

The Methods

PEERLESS is a prospective, multicenter, open-label, randomized controlled trial.

Patients

Adult patients with intermediate risk PE based on the European Society of Cardiology guidelines.

Exclusions: patient could not receive therapeutic anticoagulation, clot in transit in right heart, life expectancy less than 30 days, intraprocedural systolic pulmonary artery pressure was ≥70 mm Hg.

Intervention 1

Aspiration/mechanical thrombectomy using the FlowTriever System.

Intervention 2

Catheter directed thrombolysis treatment per local standard for device selection and thrombolytic dosing.

Outcome

“The primary end point was a hierarchal win ratio composite of the following clinical outcomes assessed at the sooner of discharge or 7 days after the procedure: (1) all-cause mortality, (2) ICH, (3) major bleeding per International Society for Thrombosis and Haemostasis definition, (4) clinical deterioration and/or escalation to bailout therapy, and (5) postprocedural intensive care unit (ICU) admission and length of stay. Postprocedural ICU use was characterized hierarchically as follows: (1) no ICU admission, (2) admission lasting between 0 and 24 hours, and (3) admission lasting >24 hours.”

The Results

599 patients were randomized, of which 550 were included in the trial. (35 patients were excluded after the invasive procedure was started based on high pulmonary artery pressures, which is discussed below.) The mean age was about 62, with approximately 50% male/female mix. It is hard to tell exactly how sick these patients were. The vital signs were essentially normal, with just mild tachycardia (110) and tachypnea (22). They present the sPESI score, but it is mostly useless for this purpose. About 40% had saddle emboli and 95% had positive troponins.

Median dose of tPA was 16 mg over about 12 hours. Ultrasound ‘facilitated’ tPA was used in 60% of patients.

The overall composite primary outcome favoured mechanical thrombectomy, but this was clearly a negative trial. (This is the problem with composite outcomes that combine important clinical outcomes with pointless outcomes.)

There was no difference in mortality (0.0% versus 0.4%; P=1.00).

There was no difference in intracranial hemorrhage (0.7% versus 0.4%; P=0.62).

There was no difference in major bleeding (6.9% versus 6.9%; P=1.00).

There was less clinical deterioration and/or escalation to bailout with mechanical thrombectomy (1.8% versus 5.4%; P=0.04).

There was less postprocedural ICU use with thrombectomy (admissions 41.6% versus 98.6%; p<0.001).

My thoughts

Despite being statistically significant, I think this trial should be interpreted as showing that catheter directed thrombolysis is clinically indistinguishable from mechanical thrombectomy. There were no differences in any of the patient important clinical outcomes. Mortality and bleeding rates were identical. Both of the outcomes that were changed are highly biased in this unblinded trial. The progression to bailout therapy is obviously going to be higher in a therapy that is given over 12-24 hours than it is in a one time procedure because the patient has 12-24 hours for the treating team to decide that something different is needed. That is not a fair comparison in an unblinded trial. Similarly, ICU admission rate is completely subjective and mostly driven by hospital rules rather than patient need, and is not an appropriate outcome in this unblinded trial. The thrombolysis group was being given thrombolytics for at least 12-24 hours, so it is not surprising that essentially 100% were admitted to the ICU. The mechanical thrombectomy group had their procedure done, and therefore might be considered safer for the floor. That tells you nothing about real patient outcomes, and shouldn’t be used to guide clinical decisions. In fact, they did a secondary analysis that looked at the composite outcome without ICU admission, and there was no statistically significant difference between the groups. For everything that matters, within the limits of this trial, this data suggests equivalence between the two therapies.

That being said, this trial is far too small to assess rare outcomes. For example, there was one fatal bleed in the thrombolysis group as compared to zero in the thrombectomy group. These rare outcomes could go in either direction, and warrant larger trials, but before larger trials comparing two invasive approaches, we need to take a huge step back and confirm that either of these approaches is actually better than standard care.

The major flaw in this trial is the lack of a proper comparison group. We have no idea if either catheter directed thrombolysis or mechanical thrombectomy improves outcomes as compared to standard care. Comparing them to each other makes absolutely no sense at this point without a comparison to systemic thrombolysis and/or just anticoagulation. The only reason one might skip that step is if one was trying to sell devices (and the conflict of interest section for this paper is very long).

There are a number of other issues worth discussing, but all are minor as compared to the lack of a proper comparison group. We should not be using expensive, invasive treatments until they have been proven to help, and neither of these approaches has any evidence of benefit.

They justify their decision to not include a proper control group by stating that “the decision to use catheter-directed intervention is individualized and complex.” However, they should have added “and completely unproven at this time”, which is exactly the reason that they should have included a control group. That being said, they have a point that these patients are complex, and we might expect to see many negative studies while sorting out which subset of patients actually benefit (if any). This research is going to be difficult to get done. A single negative trial would not negate this entire modality. In fact, I expect negative trials. But the trials absolutely must be done.

I wonder whether this trial identified an appropriate subset of patients. Given that the mortality rate was 0%, and that there are no known long term benefits from these therapies, it seems like this was way too low risk a population to warrant invasive management. Would these patients have been better off if we had just left them alone?

The very low risk population is probably partly explained by their use of signs of RV strain on CT in their inclusion criteria. (They accepted signs of RV dysfunction on either echocardiogram or CT.) The test characteristics of CT are abysmal, especially over-calling RV dysfunction, so this population is likely to be lower risk than you would expect in an “intermediate risk” group.

I also think there is a clear bias in the trial design to downplay the risk and adverse events from these procedures. They excluded patients with significantly elevated pulmonary artery pressures. This criteria was meant to exclude patients with chronic PE, because they cannot possibly benefit. Unfortunately, all of these patients were still exposed to the harms of an invasive procedure, and therefore their outcomes need to be considered when assessing the harms and benefits of these techniques. Excluding them is inappropriate and biases the data away from demonstrating harm. (Although, again, without a proper control group this trial was never going to be able to demonstrate either benefit or harm.)

Finally, a thought on statistics: as compared to traditional composite outcomes, the hierarchical win ratio approach makes sense, because it allows you to rank clinical outcomes that are more important. That being said, I would love any nerdy stats readers to comment on the approach, because I am not sure I fully understand the statistics that are used, and making multiple comparisons is a very good way to increase the rate of false positive findings. Furthermore, if you are going to use it like these authors and still just make your conclusions based on the composite, despite the important outcomes all being negative, you sort of lose all value in the approach.

Bottom line

Despite the conclusion in the abstract, this open label RCT shows no real difference between mechanical thrombectomy and catheter directed thrombolysis in intermediate risk PE patients.

At this point there is absolutely no evidence that mechanical thrombectomy or catheter directed thrombolysis help patients. There might be times where they are reasonable, even without evidence, such as in a very sick patient with contraindications to systemic thrombolysis. However, at this point neither of these procedures should be used routinely until we see real evidence.