The RECOVER IV trial is a two-arm study that seeks to analyze whether percutaneous coronary intervention (PCI) with Impella® heart pumps used prior to PCI is superior to PCI without Impella support.
In this video interview, Drs. Nick Smith (Johns Hopkins Hospital) and Gurleen Kaur (Brigham and Women’s Hospital) of the CardioNerds spoke with Dr. Bill O’Neill (Henry Ford Hospital), one of RECOVER IV’s lead investigators, to discuss this highly-anticipated study.
This interview was conducted as part of a collaboration between CardioNerds and SCAI SHOCK 2022, led by Dr. Julie Power, Dr. Dan Ambinder, and Dr. Amit Goyal with mentorship from Dr. Alex Truesdell.
Dr. Nick Smith: Hello everyone. My name is Nick Smith. I’m currently a fourth-year cardiology and critical care medicine fellow, and I’m chief fellow at the Johns Hopkins Hospital. We’re so excited for this collaboration between CardioNerds and SCAI SHOCK 22 with mentorship from Dr. Truesdell, and I’ll let Gurleen introduce herself.
Dr. Gurleen Kaur: Hi everyone, my name is Gurleen Kaur. I’m a second-year internal medicine resident at Brigham and Women’s Hospital, and I’m so excited to have Dr. Bill O’Neill here with us today. I’m also part of the Cardio Nerds Academy and director of the Cardio Nerds Internship. So, I’ll start off by introducing Dr. Bill O’Neill. Dr. Bill O’Neill is an interventional cardiologist and medical director for the Center of Structural Heart Disease at Henry Ford Hospital in Detroit. He has been at the forefront of many major advancements in interventional cardiology and is a national leader in cardiogenic shock by virtue of launching the National Card Shock Initiative in 2016. Welcome, Dr. O’Neill.
Dr. Bill O’Neill: Thanks very much, Gurleen and Nick, it’s great to be with both of you.
Dr. Gurleen Kaur: RECOVER IV form is a very anticipated trial that will assess whether Impella pre-PCI is superior to PCI without Impella in patients with acute MI cardiogenic shock. So, Dr. O’Neill, when I first heard of this concept, it seemed radical, which is not unusual for landmark analyses. There’s shock and awe, no pun intended, simply because of the novelty. Can you talk a little bit about how we got to this point where we’re randomizing the patient with a completely occluded coronary artery in cardiogenic shock to get Impella pre-PCI versus opening the vessel immediately?
Dr. Bill O’Neill: Yeah, just very briefly, I’ve been involved with acute MI cardiogenic shock since January of 1983. That was the first time a patient presented to me in a cath lab. We started the angioplasty protocol for acute MI therapy in December of 1982 and a month later this patient came in from the emergency room, blood pressure 70, cold, cyanotic, peripherally vasoconstricted, huge anterior ST elevation with actual elevation in two, three, and AVF as well. We were barely able to palpate a pulse. We got an access and then we shot an angiogram, and his LAD was occluded. At that time, he was on Levophed with a blood pressure of 70, and we put a wire across the LAD, across the occlusion and that opened up the vessel and also to collateralize flow to the distal right. And immediately, his blood pressure went from 70 to 110 and it was like my eyes lit up. I went, “Holy crap, this is absolutely amazing.”
Dr. Bill O’Neill: That was the first patient I treated within that I realized that if you could do this on a regular basis, you would improve the survival. Remember, before we started doing any of this, there was a 90% mortality for these patients, and there still is a very large mortality. So, we published our first work in circulation in 1987 and at that time we found that angioplasty improved the survival to 50%, and it’s remained at 50% for 30 years. And so, that’s been kind of the struggle and frustration. But, we haven’t really changed very much. We take the patient to the cath lab. We identify the culprit. We open up the vessel, whether it’s with a balloon or with a stent. You still open up the artery. But the frustrating thing has been that about half of the time the patients just sort of dwindle and their death is really very morbid.
Dr. Bill O’Neill: They’ll have persistent ischemic dysfunction and then something will break. Their liver will go. Their kidneys will go. Their bowel will go. Their brain. So, some other thing is ultimately what’s going to cause them to die. That’s really been the frustrating part. Half of the patients do really well, and the problem is how do you improve that outcome? So, in 2007, the FDA improved Abiomed for use in mechanical circulatory support. There wasn’t a disease indication, it was just if a doctor thought a patient needed support, just like with balloon pump, and they offered that support. So, people started using Impella, but there was no protocol. So, between 2007 and 2016, about 16,000 patients in the U.S. were treated with acute MI shock with Impella and the survival was 50%. So, we weren’t really doing any better. At that point we said, “Wait a minute, we’ve got to figure this out.” And, the device definitely provides better power and better support, but if you don’t organize and protocolize this, you’re not going to be able to improve outcomes.
Dr. Bill O’Neill: So, we did that in the National Cardiogenic Shock Initiative, which started in 2016. We enrolled 406 patients, and we were able to improve the survival to 70%. That’s by a systematic protocol, early identification of the patient, putting the Impella in rapidly, putting the Impella in before you do the PCI, and then using hemodynamic measures to try to figure out whether or not the patients are out of shock or not. That program really has resulted in improved outcome.
Dr. Bill O’Neill: I know that all the Nerds out there are saying, “Wait a minute, there’s no RCT.” And you’re absolutely right. There is no RCT, but you can’t do an RCT until you know you got the therapy right. So, for five years we tried to figure… What is the program? What is the therapy to optimize care? And I think that, that has matured to the point now where it can be tested. And that’s really where RECOVER IV comes in. Because we’re using optimal Impella Swan guided support and then doing usual care. And usual care is whatever the institution is usually doing, not using Impella. And unfortunately, it has to be done because without it, there’s always going to be a continued argument with people saying, “Well, you guys, there’s no randomized trial. You guys really don’t know what you’re doing. You’re cherry-picking,” whatever. But until we get this, there won’t be peace in the realm.
Dr. Nick Smith: That’s great. It is so nice to hear the entire arc of where we were and where we are now and actually how little progress has been made even over the course of some of your career in some respects. But in others, major leaps. So, I guess my next question will be that a lot of us have learned and been really excited about the door to unload, the DTU trial. Which is Impella, 30 minutes pre-PCI. I was wondering if you could talk a little bit about that trial, the overlap, and what makes this trial a little bit different?
Dr. Bill O’Neill: Yeah, if you take a look just broadly at PCI for STEMI, we are routinely getting, if a patient comes into hospital that’s not in shock, you should anticipate a mortality less than 2%, in-hospital mortality, less than 2%. So, we’ve done an incredible job of systematizing across the country. If you’re in Boston or in Baltimore or Detroit, wherever, you come into a hospital with a STEMI and you’re not in shock in a STEMI capable, PCI capable facility, you’ll get pretty much exactly the same care with very good outcomes. So, that whole therapy has matured, and it’s been an incredible testament to the advances of acute MI care. But we’re not totally there in terms of the long-term outcome. We know that hospital survival is extremely high, but now we want to try to take a look at the trajectory. And the patients where we’re not doing as great a job are patients that are coming in sort of after two to three hours with anterior MI.
Dr. Bill O’Neill: In those patients, we open up the artery, they survive, but they have a very poor ejection fraction with an anterior wall that is basically severely damaged. Then, that’s going to kind of give the trajectory of care for the patients for the rest of their lives. So, we’ve been looking for ways of decreasing infarct size, really again for the last 30 years. The best thing you can do is open the artery early. So, early opening within an hour or two of onset of symptoms is absolutely the best therapy. But a lot of patients are coming in later and those are the patients that we’re trying to treatment right now. So, this is what the door to unload trial is, is patients coming in with an anterior MI that are presenting between one and five hours after symptom onset. So, they’re not the hyper, hyperacute. But they’re a little bit later in duration, and we’re taking the patients and putting Impella in, and we were really concerned that people just weren’t going to be able to sit there.
Dr. Bill O’Neill: Imagine you got a patient with a big infarct, and they’re writhing in chest pain, and you’re waiting for half an hour. So, we thought that patients were going to break the blinds. So, we did a pilot trial and the pilot actually showed that we could do it, the operators. And, what happened is that when the Impella was placed, the chest pain started to go away and the STs actually started to come down. So, the patients actually became quite stable, and the artery was able to be opened in half an hour. And it was a 50-patient pilot that was published in circulation. There was no safety problem, and the patients tolerated it. And there was actually a trend towards a decrease in infarct sizes, is what we’re looking for. Dr Navin Kapur from Tufts, and Dr. Bart Meyns from Europe have both demonstrated that if you unload the ventricle you actually decrease infarct size in an animal model.
Dr. Bill O’Neill: Now, we’re trying to demonstrate in humans, and there are 50 centers around the country that have recruited as of today 200 patients out of a planned over 400. So, we’re actually rapidly recruiting, and I think within about a year or a year and a half we’ll have an answer and that really could change the whole paradigm for care of anterior MI. Now, we’ve piggybacked a little bit of that onto the RECOVER IV trial because we’re kind of assuming that, that study is going to be positive. So, we’re putting in the Impella before the PCI. We’re not waiting for… It usually takes about 10 or 15 minutes to get the artery open after the Impella is placed. But we think that there is going to be some benefit and obviously if the DTU trial becomes positive, we would modify the trial, the recovery trial to wait for an obligatory half an hour before you open up the artery.
Dr. Bill O’Neill: But we’re not there with recovery yet because the science. But I think it’s… I’ve been working on infarct size reduction trials for almost 20 years. We started using PercuSurge, which is a thrombectomy device to see if that would work. We tried balloon pump in the early nineties and pexelizumab, adenosine, cooling. The only thing that’s worked besides… The only thing that’s actually been pretty systematically studied is super saturated oxygen with TherOx and that does decrease infarct size. And that is FDA approved. That’s the only thing that decreases is infarct size. So, we’re hoping that putting in the Impella will decrease infarct size further.
Dr. Gurleen Kaur: Yeah, thank you for explaining how the RECOVER IV trial kind of fits into the larger perspective of other trials like DTU. I know earlier you mentioned that some of the work for RECOVER IV came from the NCSI study and for the NCSI study institutions were told to adopt NCSI’s algorithm, which introduced a protocolized way to ensure timely recognition of shock as well as the utilization of early MCS. So, what are some of the challenges that you’ve noticed in institutions adopting the algorithm, and how do you think that will affect the RECOVER IV trial that’s going to use the NCSI algorithm in the treatment arm?
Dr. Bill O’Neill: Well, initially it was… As you guys said, it was kind of radical to try to get this done so quickly and identify these patients and get them to the cath lab. What’s happened in the U.S. is that operators actually were using the Impella kind of as a last resort. So, the patient comes into the hospital in shock to get the PCI maybe with a balloon pump and then a few hours later they’re still in persistent shock and then they will escalate to try to put an Impella in. And we realized when we looked at the registry numbers that the longer you delay, the worse the outcome in terms of institution of mechanical circulatory support. Again, remember what’s happening to these patients is as they remain in persistent shock for six to 12 hours, there’s a lot of end-organ damage that’s going on. You get renal dysfunction, you get a bowel ischemia, and ultimately you end up with something called SIRS, systemic inflammatory response.
Dr. Bill O’Neill: So, the periphery is so vasoconstricted with high doses of alpha agonists that the tissue integrity really goes away. And so, if they survive the initial phase, then they go into this sort of high output state where the SVR really drops, and there’s nothing you can do. It’s really kind of a refractory shock. The way to prevent this is to try to get the patients out of shock as soon as possible, to get them supported so that they don’t wind up with those secondary complications. I think that’s really what we really try to do with this. In terms of the obstacles, again, I think is inertia, no randomized trial. But people actually like it because when they start instituting it, they see a dramatic improvement in outcome.
Dr. Bill O’Neill: The best testament is from Inova Fairfax that I went and talked with them about this program, and I said, “Look, just look at your outcomes.” And they looked at the two years before and they had a 40% survival and then they instituted the program and then within a year they published an 80% survival. So, they doubled the survival within a year. I think that’s really the excitement that we’re generating.
Dr. Bill O’Neill: I think it’s really important to look at your outcomes and if you’re doing okay, then you don’t need to change things. But the problem with shock is that it’s a complete morass right now around the country. If you go around the country, a hospital-to-hospital, operator-to-operator, day or night, there’s a complete difference in the way the same patient is treated. Like I said, with STEMI you’re going to get exactly the same care around the country. With shock, it’s entirely different. You may not get anything. You may get a PCI. You may get PCI and a balloon pump. If you’re in an outlying hospital, you might get thrombolic therapy, you might get ECMO support, or there’s a lot of interest in that.
Dr. Bill O’Neill: So, it’s honestly a morass, and we have to come up with something systematic. I would say, if you think that this doesn’t work, then define what your outcomes are at your institution. If you’re getting a 50% survival, you’re doing pretty much what everybody else is. If you’re getting worse or better than, you have to… certainly if you’re getting worse outcomes. In the Abiomed experience there were centers that were getting zero survival with use of Impella in shock. So, something is really wrong with if you’re not getting anybody to survive in acute MI shock, and I think you really have to take a cold look in the mirror, what are you doing wrong or incorrectly and improve the outcomes. Because centers around the country and in Japan now are demonstrating a greater than 70% survival.
Dr. Nick Smith: So interesting. Pivoting a little bit, we know based on work from the cardiogenic shop working group and multiple studies you’ve coauthored that isolated or concomitant right-sided congestion and failure is not uncommon in AMI and pretends a worse prognosis. But obviously, we know an Impella is a left-sided pump, and there’s a delicate balance between selecting appropriate patients by phenotyping, taking the time and phenotyping them correctly. But then, also, you want results that are generalizable and can be broadly applied to a broad set of patients. So, could you talk a little bit about how RECOVER IV will aim to balance these competing interests and whether you’ll be studying this kind of breakdown specifically and just talk a little bit about the overall trial design in terms of patient population.
Dr. Bill O’Neill: Thanks so much Nick for prompting me because I think this is a message that I do want to get to your listeners and that is that if you take a look at acute MI shock, 40% of the patients have some component of RV dysfunction or RV shock. It’s not just the inferior MIs. If you have a huge inferior infarct, then look out for RV infarction. But anterior MIs also have a component of RV dysfunction because remember the septum is part of what is actually generating force for the right ventricle. So, the septum is akinetic or dyskinetic, a big part of RV is not working and that can become quite a problem. So, we’ve looked in the National Cardiogenic Shock Initiative, mortality for patients that had RV dysfunction was 10 percentage point higher, and it hasn’t been really well recognized. We’ve been dealing so much with left-sided support that we’re kind of not recognizing right ventricular dysfunction adequately.
Dr. Bill O’Neill: So, what happens is, the right ventricle is not working, is not sending flow to the left side. The left ventricle is actually hypovolemic and then the Impella’s then is trying to pull, and you get chatter and the Impella RP really starts to go down in revolutions because you’re getting too much wall suction. So then, you’re saying the device isn’t working. That’s not the problem. The problem is that you’re not getting flow from the right to the left. So, you have to recognize that. And there are right ventricular support devices that are available. So, it’s really important to recognize that there is a substantial amount of right ventricular dysfunction even in anterior MI. So, part of the RECOVER IV protocol is to… After the PCI is done, we’re going to be using the Swan-Ganz catheter for measuring hemodynamics. That’s really the only way, well you can do an echo, but quickly, hemodynamically, you can tell RV dysfunction. And if there is substantial RV dysfunction, then we’re trying to use a low threshold for inserting right-sided support devices.
Dr. Bill O’Neill: There is the RPM pellets available that’s actually now very easy to use and will provide right-sided support, and that’s part of the protocol. So, we’re not kind of competing, we’re trying to integrate the whole thing to try to improve the survival. It’s really kind of interesting. This whole thing is kind of just dawned on a lot of us as we’ve been watching the left-sided support and keep hearing, “Oh my god, it didn’t work. I put the patient on ECMO.” And the reason that it didn’t work was because the right side wasn’t working. And ECMO is a very good RV support. You take blood out of the right-side oxygen and you’re putting into the left. So, that does work for RV dysfunction, but you can probably do it with a little less morbidity with a bi-ventricular support.
Dr. Nick Smith: Thank you.
Dr. Gurleen Kaur: Yeah, it’s definitely very interesting thinking about the right ventricle, just when I’ve been in the CCU, as well, kind of. Sometimes, we forget about the RV, about how important it’s and how it affects what we determine about next steps for patients. Just to pivot off a little bit, I know you mentioned earlier that there’s a lot of variability in practice when it comes to cardiogenic shock and so far there haven’t been that many evidence-based treatments for acute MI cardiogenic shock because of the limitations of trials. They’ve been difficult to conduct, or they’ve been underpowered or neutral. So, RECOVER IV is definitely a very eagerly awaited, randomized controlled trial in this area. So, when do you anticipate the trial will start enrolling and seek completion, and do you have any predictions of what the trial would show us?
Dr. Bill O’Neill: I stopped predicting. I did a study back in 1985 with Eric Topol, he was a Hopkins resident by the way, so he’s a storied alumni of your institution. But we thought that angioplasty was going to work. We’ve done a lot of work with angioplasty and then we said, “Well, that’s good, but there’s a delay so why don’t we just give the patients Litex ahead of time and then do the angioplasty, best of both worlds. You can open up the artery early. If it’s not open, you can open it.” And, that study was negative, and the Europeans did a study, and the NIH did a study. So, you take three of those together, and it’s like, “Holy crap, we’re killing people with angioplasty.” So, sometimes stuff that seems intuitively obvious isn’t. So, I’ve given up predicting. I’m hoping that we’ve improved it because if RECOVER IV works, that means that we’ve advanced the field, and we’ve improved survival. So, that’s kind of my optimistic hope.
Dr. Bill O’Neill: But this trial has to be done. This is going to be the last mechanical support trial encouraging shock that’s done in the United States. One way or another, it has to be completed, and it doesn’t really matter what the answer is to be truthful. We have to come up with a clean answer. It can’t be blurred or smudgy. So, it’s got to be adequately powered. One of the things, and you mentioned, Gurleen, there’s a lack of randomized trial. There’s really been only one randomized trial in shock that was completed in 1997 by Judith Hochman in the United States. In Europe, it’s different. They’ve had four or five trials. The reason is that in Europe they’re allowed to enroll patients without consent and then retrospectively go back and get consent.
Dr. Bill O’Neill: So, there is actually a process in the United States called Exemption from Informed Consent. That’s part of what we’re doing with this trial, with RECOVER IV. We’ve worked with the FDA already. They’ve already approved this. So, we haven’t approved trial, and we’re just sort of coming to site selection and use this [inaudible] form, exemption from an informed consent process to try to more rapidly enroll the patients. So, we’re hoping that the study is going to start sometime early next year, and we’re actually in the site selection process right now.
Dr. Nick Smith: Amazing. Well, thank you so much for your generous time today, and this has been a fascinating discussion listening to the way this has all come about, and we’re just both very thankful.
Dr. Bill O’Neill: Well, I look forward to hearing more from you guys and seeing you up on the podium, and you might be talking about RV shock or acute MI shock, so love to meet you both.
Dr. Gurleen Kaur: Thank you.