Does ST-Elevation MI Need a Rebrand to ‘Occlusion MI’?

This transcript has been edited for clarity.

Robert A. Harrington, MD: Hi. This is Bob Harrington on theheart.org | Medscape Cardiology. I always like to do a show from the American Heart Association (AHA) meetings, the American College of Cardiology (ACC) meetings, and the European meetings because it gives me a chance to grab some great guests and talk about timely topics. Sometimes it’s what has been presented in the meetings, but this time I want to talk about a recent paper.

The STEMI Paradigm

The recent paper was a state-of-the-art review in JACC: Advances where they called into question what I’ll call the ST-segment elevation myocardial infarction (STEMI) paradigm. What I mean by that is it’s using the elevation on the EKG as a mechanism by which we make decisions, such as going to the cath lab or not going to the cath lab, going to the cath lab immediately, or going to the cath lab in a more delayed fashion.

The authors call that into question and say that there’s more to it than just ST elevation. What you’re really trying to figure out is whether there is an occluded artery. We’re going to walk through some of the history of how we got there, what people think of this premise, and then with my guests, we’ll talk about the future.

We have two great guests today who are good friends. Mike Gibson is from the Baim Institute, where he’s the CEO, and he’s a professor of medicine at Harvard and an interventional cardiologist at Beth Israel Lahey Health. Next to him is Manesh Patel, interventional cardiologist at Duke University, where he’s also the chief of the cardiovascular medicine division.

Mike and Manesh, thanks for joining us.

Manesh R. Patel, MD: Thanks for having us.

Harrington: Let me start with the easy one. Mike, do you think this is a worthwhile topic to dive into, and do you think it’s timely?

C. Michael Gibson, MD: It’s definitely worthwhile diving into. We’ve come a long way since Q waves at discharge as a metric. We realized that time is muscle, so we’ve moved up to STEMI, but it really is time to move beyond that.

We published findings that if you have ST depression in the anterior pericardium, one third to half the time, the artery is closed. It’s a posterior infarct. It took 29 hours in the TRITON-TIMI study for people to get to the cath lab. No one was there quicker than 9 hours. We’re sitting on many closed arteries with this ST elevation focus.

Harrington: That’s what these authors brought up as the central premise, that if you just go by the classic ST elevation as the diagnostic tool, you’re going to miss occluded arteries and you’re going to miss a fair bit of them. The posteriors — you’re going to miss ones where maybe the ST was up, but now it’s down. You and I have talked over the years about cyclic flow variation.

Patel: Large branches [and] ramus infarcts are often seen that way, maybe sometimes depending on the right coronary territory. Again, it’s really hard because we’re using the ECG to try to figure out what’s going on in the angiogram.

Harrington: When I first read the article, the other thing I thought about was that we actually didn’t just develop the paradigm, if you will, to diagnose ST elevation. We really tried to give a diagnosis to the others, which used to be called non–Q-wave infarction and unstable angina.

I always would ask, why are we giving a diagnosis three days after the event happened? How about more acute? That’s in the early 1990s, when we had acute coronary syndromes with and without ST elevation.

Gibson: It did us a bit of a disservice, though. We did do trials on non-STEMIs (NSTEMIs), treating them with primary angioplasty. We didn’t see much benefit, but NSTEMI is a bit of a wastebasket. You have closed arteries, you have open arteries, you have normal arteries, and you have normal arteries with supply and demand mismatch. We have to do better in terms of identifying the occluded arteries that really warrant timely intervention.

Patel: I would say that’s right. I would say the other thing clinically is it’s the occluded artery, but what we’re really trying to figure out is if somebody had a plaque rupture. As you guys and others have said, it’s occluded sometimes and it’s open sometimes, so it’s a dynamic process, too.

In this day and age, we have to move away from ST elevation, just like we have to move away from a single troponin value to say, oh, somebody’s having a heart attack, and instead say it’s a clinical issue. Yes, of course we want a threshold to do something, but what is the information you need to go do something?

Gibson: It’s not just plaque rupture; it’s also plaque erosion. It drives me insane in cath conference because everyone’s always looking for the tightest artery to go and intervene. Well, what about that circumflex, that clot over there? And they say, “It’s open or it’s not that tight.”

Harrington: Remember, we knew from studies a long time ago that the artery that appears to be the infarct artery is often not the only one with disrupted plaque. People have looked at the angiogram and they say, oh, you have a left anterior descending infarct, but wait a minute, there’s hot-looking plaque over here in the circumflex as well.

Patel: It’s quite complicated, right? We’ve also evolved from saying when you have cardiogenic shock to open everything, to only open the artery that you think is causing it, to where now when you have STEMI or acute MI, we’re saying treat the infarct-related artery but maybe also treat the others, I believe patients do benefit from getting all the arteries treated.

False Alarms or Aborted MI

Harrington: Manesh, let me go back because you have run many important committees for the ACC on appropriate use criteria, and you’ve been on guidelines. One of the things that ST elevation has really done is helped us on the quality front. How long did it take you to get from door to balloon? How long did it take you to get from home to diagnosis?

And for those of you who have run a cath lab, what about calling in your colleagues in the middle of the night for “false alarms”?

Patel: False alarms, right? I have to say, the work done by many others in this space has been amazing. The ST elevation paradigm has helped us take better care of acute MI (AMI). The AMI management strategy — the lifeline strategy of time, door to balloon — reduced a threshold that I have to be quite frank was much higher when I was a fellow. As you know, it would take a lot to bring some people to the cath lab.

Harrington: I’m also smiling because I’m thinking of our colleagues.

Patel: That’s right. There might have been some colleagues who had to be persuaded to bring people into the cath lab even when they had ST elevation. We know time is muscle. I think we’ve changed that paradigm; that’s important. Even in changing that paradigm, sometimes you swing too much one way or the other.

The threshold to go to the cath lab shouldn’t be very significant. You still need a clinical diagnosis. I do say this to our interventional fellows: If you call me in, give me one of the five reasons to come in. "Is the artery closed” is the first one. If you think the artery’s closed, it doesn’t matter if there are clinical symptoms or not. Call me because we don’t want to have the 9- to 25-hour wait for somebody who has a circumflex or ramus infarction.

You might be wrong sometimes, but that’s why we’re learning. That’s what the whole fellowship or training life is about. That’s one of the key steps, I think, for all of us.

Harrington: Tell us the other four.

Patel: The other ones are, does somebody need a pericardial effusion tapped? Is there hypotension? That’s important to do. Sometimes it’s a device, a balloon pump. The last one is, there’s a donation of the heart or something else. Those are usually the five clinical ones.

After that, if you don’t tell me one of those five, but you say, “I don’t know what’s going on,” that slows me down. Now I’m thinking. Otherwise, if you tell me one of those five, it sounds good, you have a plan, I’m coming in. I don’t have to hear all the risk factors. I just want to know the patient’s age and blood pressure.

Harrington: That’s good practical stuff in the middle of the night.

Mike, how about up in Boston? Is the false alarm issue still a big deal, or are people willing to come in based on these other clinical criteria?

Gibson: It depends what you mean by a big deal. It’s still prevalent. There’s a fairly high 30%-40% “false alarm” rate. That term makes me nervous, having looked at all those angiograms in the 1990s.

Harrington: I don’t know if anybody’s looked at more angiograms than you have.

Gibson: That’s right. We’d look at 45 minutes, 60 minutes, 75 minutes, waiting until that beautiful 90-minute endpoint, and the artery is winking and blinking. It’s opening and closing. When someone says, oh yeah, the artery is open, you’re like, well, 5 minutes earlier it was probably closed. I’m just uncomfortable with saying, oh, another false-positive. I don’t think that’s a false-positive.

Harrington: As Manesh is saying, and as this review article says, you have to put the whole context together.

Gibson: Correct. We have made tremendous strides, major public health victories on the door-to-balloon time. It’s probably one of the bigger success stories in cardiology. On the other hand, we have failed completely on symptom-to-door time. In the ’80s, the ’90s, and now in our current era, it’s 2.25 hours.

Harrington: Do you know what it was in GUSTO-I back in 1990-1993? It was about 2.25 hours.

Patel: It’s been around 2 hours for many years, and I would say that the big value proposition we’ve seen is we’ve reduced some metrics and statewide and regional mortality by going door to balloon, but that time of symptom onset is still bad. I also think that, as an interventional cardiologist for many years, we have said can it be door to sheath, because you want to take the pressure off to let them make the decision once they see the vessels. Instead of just going for the tightest, which one has the clot, right?

There is a little bit of tension. Of course, you can be fast. The metrics got us there faster. But it should be door to think. Now I have a moment to think once I’ve seen the patient.

At-Home MI Diagnosis

Gibson: We need to move the therapy into the home, and we need to move the diagnosis upstream, into the home. That’s where some of our efforts are focused. For instance, I published a paper last year looking at single-lead ECG.

Harrington: Obtained from a device?

Gibson: Yep. The truth standard was an occluded artery on the angiogram. This included hundreds to thousands of patients in Brazil, and then we had about 90% accuracy in detecting a closed artery.

There are other technologies coming along that don’t use traditional ST elevation. They use vector cardiography, and that’s a big advantage because it shows you is there a posterior infarct using HeartBeam technology.

We’re getting there. I think it’s going to take home-based monitoring. The other thing is continuous monitoring with an implantable device; 17 years studying this. We did get people to the hospital in 30 minutes if this thing went off.

Also, people thought, oh my gosh, you’re going to dramatically increase the flow through the cath lab. Actually, no. One of six people had a true heart attack, and many of those just by symptoms; they were on the rocket sled to the cath lab, but this actually changed things. Finally, I’ll say 5% of heart attacks were asymptomatic that we detected.

Patel: That you picked up.

Gibson: That we picked up. Of 1000 patients, we had about 45 people who had a silent MI. The device was going off and they were like, what is happening? They brought them to the lab and they had a closed artery.

Harrington: Let me ask you, Manesh, because you’ve been involved along with both of us in many of the trials of NSTEMI. There have been randomized trials. We’ve taken people earlier and later.

Patel: We’ll just say that the sweet spot for us is that going earlier is generally better, but you don’t have to go as early as we do with a STEMI. I think that’s partly because of the heterogeneity. Whether it’s the TIME trial or others, we’ve shown — and I like to say this, but it’s not fully evidence based — that generally, not a large amount of good stuff happens from waiting.

You don’t want to take a bunch of people to the cath lab until you clinically know what’s going on, but for people where there’s ongoing chest pain and a true feeling that there’s an MI, then going sooner is better.

In the NSTEMI trials, it looks like going sooner, meaning within the first 24 hours, vs later, meaning 3-5 days, was better. Then there have been a few trials where they tried to go in the first few hours. There, we didn’t see a huge difference.

Harrington: Except in that group where the GRACE score was really high, right? In the trial out of McMaster University, they stratified people based on their GRACE score, and they did benefit going early — maybe not that quickly.

Patel: You get benefit going earlier and we risk-stratify whether with the GRACE score, TIMI risk score, or other scores, you can go sooner.

One of the things we think about all the time is that this is going to get much more personalized. We’re going to get more information for that patient. As Mike is saying, whether the diagnosis is upstream because we’ve got some monitors saying this person’s having an event, to that person now in the hospital is going to have a variety of other data streams, I suspect, to say, they’re much higher-risk than you think.

One of the things we often see is risk miscalculation by the people. Whether they’re coming to the lab saying, oh, this person’s not going to be too sick because they’re younger, or they’re actually having a big event vs other features.

AI-Based Tools and Wearables

Harrington: Let’s talk about something I saw yesterday at AHA. There was a group from Europe that took a bunch of angiograms with occluded arteries and used that as the central truth, had the associated EKGs, fed that all into the machine using large language models, and came up with an ability to predict whether or not you’d have occluded artery.

All sorts of stuff was being missed looking at just ST elevation. Is it practical to think about starting to incorporate these into our ambulances and our emergency departments?

Patel: I think so. We had a piece we recently wrote, called, “Translating AI for the Clinician.” I think the thing I would say about this is, what’s the indication? Doctors are going to ignore new AI tools unless you say, here’s the indication, we randomized people, and show them how to do it, just like any other diagnostic thing.

For this tool, for example, I would randomize people to turning that on or off to see if we identify more people with an occluded artery. I suspect it will, but let’s do the study. My biggest fear is — and I say this about AI tools in general — doctors that know how to use it would be better than doctors that don’t. But you have to get people to use it.

Gibson: There are also tremendous advances in biomarkers. There’s going to be a troponin [test] where you spit.

Harrington: There is troponin that gets picked up in sweat.

Patel: Oh my gosh, this is like a nightmare for me in the ER. Sweat troponins?

Harrington: They put a device on the skin.

Gibson: Binah.ai has a ring that detects A1c, glucose, and potassium, and I’m sure we’re going to get to a troponin here. You’re going to couple a wearable with an alarm with a biomarker in the home that triggers some home therapy or gets them to the hospital.

Patel: It’s awesome if it works well, and I think it will if we can get the right testing. That’s what I think will happen. The technology is coming where we’re going to know a lot more about people or we’re going to be able to predict based on what we know about what’s going on.

Gibson: Here’s the paradox. The better you get this diagnosis in the home, and the better you get a prehospital treatment, they show up to the cath lab and they have an open artery. Then you say, oh, false positive. That’s why you need the studies, because otherwise you’re going to have a bias.

Harrington: You have to test the strategy to show if doing this is better than the strategy of doing what we usually do.

Gibson: We’re trying to abort the MI, and if you have aborted the MI, you don’t get credit for it. That’s the paradox.

Harrington: Mike, you’re doing a trial of the injectable IIb/IIIa inhibitor. What triggers the patient or the EMT to use the injectable IIb/IIIa?

Gibson: They get a 12-lead ECG in the field.

Harrington: It’s not the patient doing it at home?

Gibson: No. When the ambulance shows up, they consent them. If they have ST elevation, they go ahead and give the drug subcutaneously.

Harrington: They get randomized.

Gibson: They get randomized and give the drug subcutaneously, then take them to the hospital.

Harrington: It’s a big trial.

Gibson: It will be 2500 patients. We’re 2000 patients in, so we’ll know the answer soon.

Harrington: It’s an interesting way to think about it.

Gibson: Another fascinating thing is the FDA finally agreed that infarct size is clinically relevant.

Patel: We’ve been spending years trying to get to that, whether it’s the concerted MRI data or spec data.

Gibson: It’s death, MI, stroke, stent thrombosis, hospitalization for heart failure, and then the bottom of the win ratio is, do you have a 10-fold elevation in your troponin?

Harrington: Really trying to incorporate everything that’s going on, including a biologic marker.

Gibson: Correct.

Patel: In addition to this, we’re also doing a study with an injectable P2Y12 inhibitor that patients can use at home if they’re having chest pain. So we’re moving therapies closer.

Gibson: Using the same win ratio.

Patel: Yes. The idea is that if you can identify sooner, you’ll have therapies at home sooner, which will be really important.

Harrington: First, we have to figure out if we can identify it.

Patel: That’s right. This closed artery is a great way of thinking about it for the cath lab, but it’s also going to be a great way we think about it as we have patients with more information from continuous monitoring.

Harrington: I think all three of us agree that this article got us thinking about this topic. It draws on many thoughts that have evolved over the past 30 years, to really try to put at the forefront how we use all available information to make the best decisions, and then how do we empower patients to get involved in their own therapy while they’re making their way to the hospital. Is that a reasonable summary?

Patel: I totally agree. I’d say back to basics for us in practice right now is making sure you’re still making a clinical decision, you still have good EKG reading skills, and you’re still thinking about the things, such as posterior infarcts and vectors. Additionally, one can imagine that the article is really pushing us to say, how do we change the practice and get it into the home?

Harrington: Anything else to add, Mike?

Gibson: Well, we’ve always said, oh, NSTEMIs have similarly bad outcomes as STEMIs. That, again, is the heterogeneity of the NSTEMI coming through. If you look at the NSTEMIs that have a closed artery, there’s an 8.9% rate of adverse events. If they have an open artery, there’s a 2.6% rate. They of course have more comorbidities, but I think we’ve got to maybe refocus our classification of these NSTEMIs to this nonoccluded vessel paradigm.

Harrington: Maybe this helps us think about that.

Gibson: It helps us when we’re doing trials to say not all NSTEMIs are the same. We need NSTEMIs that have an occluded vessel to really enrich.

Harrington: Manesh, I’m sure, has heard me say over the years that I have always tried to teach residents and fellows that it’s the NSTEMI cohort that, in many ways, is more interesting because it’s harder to figure out, because you’ve got everything from really bad disease to normal arteries. They’re not the same.

Then there’s the issue that while their acute risk may not be as high, their delayed risk is really high. Maybe that’s a whole other area of identifying the NSTEMI patient so that you can treat them down the road.

Patel: I think the NSTEMI is much more of a disease of aging. [People with] STEMIs are often younger with single-vessel [disease] — not to say [that it] can’t happen later and be more complicated. There’s huge overlap, but when you get a large NSTEMI population, there is much more comorbidity and they’re much older.

Gibson: We have changed the biology of STEMI. When you and I were young — you’re younger than us, Manesh — every day, Bob, you knew there’d be three or four STEMIs coming in.

That’s not happening now. With statins and aspirin, we may see one every other day. It’s really changed.

Patel: We see a lot more NSTEMIs and much older patients. In our lab, we are stenting and taking care of patients that are in their 80s routinely. Multivessel, proximal left main, not just in the acute coronary syndrome scenario, burden of disease, complex coronary disease where surgery is really complicated.

Gibson: Maybe a fair amount of plaque erosion that we’re not seeing as much on the angiogram and less plaque rupture with visible thrombus. More burden.

Harrington: All right, I’m going to have you both back to talk about the diffuseness of the disease, the changing paradigms of using things like CT angiography to look at plaque before you go to the lab, better characterization of plaque so that you can decide upon both the medical interventional strategies when you get into the lab, and making the lab a more high-yield environment as opposed to going in there when you don’t know what’s going on so you’ll take them to the cath lab.

Patel: As I often say, if you don’t know, let’s go figure it out.

Harrington: Thanks for joining us. It’s been a fabulous discussion.

I want to thank you, the audience, for joining us. Let us know if you like the show. Give us a like, write us a comment, or subscribe. It’s really what you tell us as an audience that really defines the kind of topics we’re going to address in the future. Thanks for joining us, and thank you both for joining us.

Gibson: Thanks for having us, Bob.

Robert A. Harrington, MD, is the Stephen and Suzanne Weiss Dean of Weill Cornell Medicine and provost for medical affairs of Cornell University, as well as a former president of the American Heart Association. He cares deeply about the generation of evidence to guide clinical practice. When not focusing on medicine, Harrington dreams of being a radio commentator for the Boston Red Sox fan.