Time is muscle—or is it? Challenging reperfusion dogma in the STEMI-DTU era

Le résumé fourni par la source

The ST-segment elevation myocardial infarction Door-To-Unload (STEMI-DTU) randomized trial tested whether left ventricular unloading with a trans-valvular microaxial flow pump before reperfusion, combined with a protocolized 30-min delay to percutaneous coronary intervention (PCI), could reduce infarct size in patients with anterior STEMI without cardiogenic shock (CS).1 In 527 patients, this strategy did not reduce infarct size compared with immediate PCI alone, while bleeding and vascular complications were more frequent in the unloading arm. These findings challenge the clinical translation of a mechanistic and preclinical concept, raising important questions about patient selection, device-related risk, and the future role of mechanical circulatory support (MCS) in acute myocardial infarction. To contextualize these results, we asked Navin K. Kapur, principal investigator of STEMI-DTU, and Holger Thiele to discuss the implications of the trial from complementary pro and con perspectives. To both: How do you interpret this trial results: do they reflect limitations of the unloading concept itself or the challenges of translating it into clinical practice? Prof. Kapur: Infarct size correlates directly with mortality and heart failure development. The STEMI-DTU pivotal trial illustrated that nearly one-third of the left ventricle is infarcted even with optimal door-to-balloon and total ischaemic times. A sobering observation that identified the need for continuous improvement in STEMI management. The trial is the first RCT testing the effect of a rotational trans-valvular pump combined with an intentional delay to reperfusion on myocardial infarct size. The intentional delay of primary PCI is provocative and based on extensive preclinical data. My gratitude to the DTU patients and operators who participated in this protocol as it represents a landmark in cardiovascular medicine that will inform future study designs. Several reasons may explain why the trial did not meet its primary endpoint. First, most patients were hypertensive, which is a challenge for any rotational support pump since their function is negatively affected by high afterload. As a result, patients may not have been consistently unloaded, and a heterogeneous effect was observed. Notably, all preclinical studies demonstrated significant infarct size reduction in the setting of relative hypotension, representing an important translational gap for patients without CS. Second, without consistent unloading, the 30-min delay period limits ‘mechanical conditioning’ of the heart and may have confounded any beneficial effect of the protocol. Collectively, these observations identify challenges in translating science from the preclinical to clinical environment and opportunities to advance our understanding of load conditions with circulatory support in anterior STEMI without shock. Prof. Thiele: This is not a surprising result. In my scientific career I have seen many concepts including drugs and devices or conditioning strategies that have worked in animal models but did not translate in humans. For decades now, this specific concept has never translated from animal models to humans. We did the same in the previous CRISP-AMI trial with the intra-aortic balloon pump (IABP) which even in a trend showed larger infarct size with the IABP vs. control in large anterior infarctions.2 Some argued the concept worked since infarct size did not increase in the microaxial flow pump group despite a 47-min reperfusion delay.3 However, this requires assessing myocardial salvage in relation to ischaemic time: with an average ischaemic time of nearly 3 h, reperfusion benefit is diminished and the salvage effect has already plateaued. To both: Given that current STEMI care is built around the principle that time is muscle, how should clinicians interpret a strategy that introduces a delay to reperfusion without clear clinical benefit (and even with a higher risk of complications)? Do you see any specific scenarios where this approach might still be justified, or should immediate PCI remain the standard? Prof. Kapur: The trial does not impact current clinical practice. STEMI-DTU is an important mechanistic study that tested a hypothesis using a protocol that included delayed reperfusion. To be clear, clinicians should not attempt to unload and intentionally delay PCI in patients with anterior STEMI. Primary PCI remains the gold standard, for now. I add that last phrase because the science of unloading is sound. We see it every day in patients with crippling angina who are referred for left ventricular assist device (LVAD) implantation and then remain angina-free as long as the LVAD remains functional. Also, it is for this reason that we do not perform coronary artery bypass surgery at the same time as an LVAD. Effective unloading terminates ischaemia. Consistent with this observation, operators from multiple nations observed a reduction in ischaemic symptoms as soon as a patient received the microaxial flow pump. We also observed a significant increase in coronary perfusion pressure during the unloading period and despite an over 40 min of delay to reperfusion, no increase in infarct size was observed. While more studies are required, DTU supports the concept that pre-PCI haemodynamic support with a trans-valvular pump does not increase infarct size. These observations will inform future studies exploring the utility of trans-valvular pumps to reduce infarct size and the burden of post-infarction heart failure. Regarding the bleeding observation, DTU shows how bleeding is a combination of technology and technique. Same technology, but 22% bleeding rates in the US and 48% outside the USA. The primary difference appears related to closure methods, with pre-closure being associated with the lowest bleeding rates. Importantly, the lack of association between bleeding and mortality has now been observed in DTU and DanGer Shock.4 This likely reflects the inappropriate application of Bleeding Academic Research Consortium (BARC) definitions in the setting of MCS. For example, all MCS devices are preload dependent and many patients receive volume resuscitation, which causes haemodilution and a drop in haemoglobin, thereby meeting BARC 3 criteria. Nearly all ‘bleeding’ events in DTU were BARC 3 and the majority did not require a blood transfusion. More analyses of the bleeding events are forthcoming. Going forward, technologies need to improve, techniques need to be protocolized based on data, and we need to re-define bleeding in the setting of acute MCS. Prof. Thiele: For me, this is relatively clear. Immediate percutaneous coronary intervention (PCI) by radial access remains the reference standard in this population. There is no convincing evidence for any subgroup that unloading may have a benefit. We must also keep in mind that STEMI-DTU had an unusual design. It did not mandate or suggest a radial first approach which is recommended in all guidelines with a Class I, Level A recommendation. This led to only 16% of patients being treated according to guidelines in the control group. Furthermore, it had a performance goal to compare safety outcomes and not a comparison to the control group—which is the scientific standard. This performance goal was set very high to have the chance that the active unloading arm at least show safety. Despite this very high-performance goal and the not state-of-the-art control group, the microaxial flow pump group had a 30.8% bleeding rate. With such a bleeding rate, probably, nobody will consider using a microaxial flow pump for a haemodynamically stable STEMI patient. To Prof. Kapur: If you were designing the next study, would you change anything? Prof. Kapur: The STEMI-DTU community remains motivated to explore approaches to reduce infarct size with left ventricular unloading. One approach may be to capitalize on the therapeutic window created during delay to PCI with unloading an

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.