Prof Nicolas Van Mieghem:
Welcome to our ACC.26 late‑breaking clinical trial wrap‑up. I’m joined today by my dear friend and colleague, Dr Azeem Latib, an expert interventional cardiologist from Montefiore Medical Center in New York. We’re going to reflect on some of the most important late‑breaking trials from ACC.26 and what they mean for day‑to‑day practice.

Let’s start with pulmonary embolism and the HI‑PEITHO trial. This study focused on patients with intermediate–high‑risk pulmonary embolism. It builds on the original PEITHO trial, which compared anticoagulation alone with full‑dose systemic thrombolysis. In PEITHO, thrombolysis reduced death and haemodynamic collapse but at the cost of more intracranial haemorrhage, so systemic lysis never really made it into routine practice for this group.

HI‑PEITHO revisited the same risk category, but this time patients were randomised to anticoagulation alone or ultrasound‑facilitated, catheter‑directed thrombolysis using the EKOS system plus anticoagulation. Only a small proportion of screened patients were actually enrolled, so this was a very selected population. The mean age was around 58, with a roughly equal sex split, and most patients had bilateral pulmonary embolism.

The primary endpoint at 7 days – a composite of death, recurrent PE and haemodynamic collapse – was significantly reduced with EKOS‑guided thrombolysis compared with anticoagulation alone. Rescue therapy was also less common in the interventional arm. However, by 30 days there was no difference in mortality or recurrent PE. Personally, I had expected a negative trial, but the early benefit is clear. At the same time, I still believe large‑bore thrombectomy devices and catheter‑based thrombectomy will prove more attractive in routine practice.

Dr Azeem Latib:
I agree it’s an important trial, even if it may not immediately change how most centres manage PE. HI‑PEITHO is the first randomised study in this space to show that doing something interventional for intermediate–high‑risk PE can improve hard clinical endpoints, not just RV/LV ratios or echo surrogates.

Catheter‑directed thrombolysis is one interventional option; it was the one studied here and may have advantages in specific patients, such as those with older thrombus. But many of us have already moved towards large‑bore thrombectomy. For that strategy we still need randomised data versus medical therapy, similar to what HI‑PEITHO has now provided for ultrasound‑facilitated thrombolysis.

One practical point is that pushing very large‑bore catheters into the pulmonary arteries is not trivial. It requires experienced operators and carries access and vascular risks. EKOS, by contrast, is relatively easy to position. So, while practice may not flip overnight, we now have proof that an interventional approach can favourably affect hard outcomes in this risk group.

Prof Nicolas Van Mieghem:
Let’s move on to mechanical circulatory support and STEMI. For years we’ve discussed the concept of “door‑to‑unload” – using LV unloading with a device like Impella before opening the infarct‑related artery in STEMI, with the goal of reducing infarct size. STEMI‑DTU has finally tested this in a randomised fashion.

Patients with large anterior STEMI were randomised to standard primary PCI as quickly as possible versus LV unloading with Impella followed by delayed reperfusion. The primary endpoint was infarct size measured by cardiac MRI several days after the event.

The bottom line is that STEMI‑DTU was negative. There was no reduction in infarct size with the unload‑then‑reperfuse strategy. In fact, there was a signal towards slightly larger infarcts, although that needs cautious interpretation. What the trial clearly showed is that unloading delayed reperfusion by roughly 30–40 minutes.

Clinically, this reinforces the message we’ve had from primary PCI for decades: time is muscle. In haemodynamically stable STEMI, you should not delay opening the artery to insert a mechanical support device. That doesn’t mean you shouldn’t use support in cardiogenic shock – that’s a different scenario – but in stable patients LV unloading before PCI does not appear beneficial and may be harmful.

Dr Azeem Latib:
Exactly. The trial also highlights the real‑world downside of large‑bore devices: bleeding and vascular complications. You’re inserting a large‑profile catheter in patients who are fully anticoagulated and receiving potent antiplatelet therapy. That price shows up in every Impella trial we’ve seen.

The key nuance is haemodynamic status. In shock, you often stabilise the patient with mechanical support first and then proceed to PCI in a more controlled environment. STEMI‑DTU specifically excluded those shock patients; these were haemodynamically stable STEMIs. For them, you should focus on rapid reperfusion and avoid unnecessary delays.

Prof Nicolas Van Mieghem:
That brings us to CHIP‑BCIS3, which addressed mechanical support in a different context: high‑risk PCI. This UK trial enrolled genuinely high‑risk patients with complex multivessel or left main disease and severely reduced LV function. Patients were randomised to high‑risk PCI with or without elective Impella CP support.

Again, the trial was negative for its hierarchical primary endpoint, which included death, stroke, MI, rehospitalisation and procedural complications. There were more “wins” in the control arm than in the Impella arm, and a worrying trend toward higher mortality with routine support. Complications, including vascular issues and bleeding, were more frequent in the Impella group.

In my view, CHIP‑BCIS3 sends a strong message: we should not be using mechanical support systematically in all high‑risk PCI patients. Instead, it should be reserved for a clearly defined subset – those with extreme LV dysfunction, complex left main or last‑remaining‑vessel interventions where even brief ischaemia would be poorly tolerated.

Dr Azeem Latib:
I think that’s the right interpretation. In my practice we already use a selective approach. We base decisions on angiographic complexity, LV ejection fraction, pulmonary pressures and, crucially, invasive haemodynamics from a right‑heart catheter.

You sometimes see patients with an EF of 25–30% but a perfectly acceptable cardiac index who can tolerate a complex PCI without mechanical support. Conversely, another patient with an EF of 35% may have very high filling pressures and a low cardiac index, telling you that they’re going to decompensate when you start long inflations in a left main. That’s the patient who truly needs Impella or another support device.

CHIP‑BCIS3 does leave some questions, for example why the mortality curves continue to diverge out to one year in a way that’s hard to explain mechanistically. Ongoing studies such as PROTECT IV may help clarify whether there are subgroups that genuinely benefit from planned support.

Prof Nicolas Van Mieghem:
We also discussed several key structural and heart failure trials, including CHAMPION‑AF in left atrial appendage occlusion, SPIRIT‑HF in device‑based heart failure therapy, PRO‑TAVI and PROTECT H2H in TAVI and embolic protection, and the tricuspid interventions TRILUMINATE, TRI‑FR and TRISCEND II. Each of these contributes to the broader question of which patients we should treat, with which device, and at what point in their disease trajectory.

For example, CHAMPION‑AF raises the possibility that left atrial appendage closure could be considered earlier in selected AF patients, not only in those who have “failed” anticoagulation. SPIRIT‑HF prompts us to think harder about which heart failure patients might benefit from interventional strategies beyond drugs and CRT. And the tricuspid trials show that transcatheter repair and replacement are now viable options, but we still need to refine patient selection and understand long‑term outcomes.

Dr Azeem Latib:
Overall, ACC.26 was a very rich meeting, with many large randomised trials and several publications in major journals. Some of these studies will change our practice immediately; others mostly raise new questions or confirm that we should *not* adopt certain strategies routinely.

For me, the key practice messages are:

- In intermediate–high‑risk PE, HI‑PEITHO shows that an interventional approach can improve hard endpoints, but choice of modality and patient selection remain crucial.
- In stable STEMI, do not delay primary PCI for prophylactic LV unloading.
- In high‑risk PCI, routine mechanical support with Impella does not improve outcomes and may increase harm; a selective, haemodynamics‑guided approach is essential.
- In structural and heart failure interventions, we’re seeing rapid evolution, particularly for LAAO, TAVI and tricuspid therapies, but careful interpretation and thoughtful adoption are needed.

Prof Nicolas Van Mieghem:
I fully agree. We will all need time to read the full papers and work through the supplementary analyses, but this ACC.26 has certainly moved the field forward.

Azeem, thank you very much for sharing your insights and for an excellent discussion. And thank you to everyone watching these chapters. We hope they help you integrate the latest ACC.26 data into your daily practice, and we look forward to continuing the conversation in future previews and wrap‑ups.

Dr Azeem Latib:
Thanks, Nick, and thanks to our audience. Take care and see you next time.