Circulation December 11, 2018 issue

Dr Carolyn Lam:                Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr Carolyn Lam, associate editor from the National Heart Center and Duke National University of Singapore. What are the long-term effects of oxygen therapy in patients with suspected acute myocardial infarction? Well, to find out, stay tuned for our discussion of our feature paper this week, coming right up after these summaries.

                                                The first two original papers demonstrate that, similar to neonatal mice, one day old and two-day old neonatal pigs are capable of mounting a cardiac regenerative response following myocardial infarction, which is characterized by restoration of contractile function, cardiomyocyte replenishment, and minimal fibrosis. Now, interestingly, this regenerative capacity is lost after the first two days of life.

                                                The first paper is from co-corresponding authors, Drs Yeh and Cook from National Heart Research Institute of Singapore and National Heart Center, Singapore, and the second from co-corresponding, authors Drs Zhang and Zhu from the University of Alabama at Birmingham.

                                                These authors report collectively that proliferation of preexisting cardiomyocytes appear to be the primary source of cardiomyocyte replenishment in neonatal pigs with markers of cardiomyocyte mitosis, sarcomere disassembly, and cytokinesis elevated following injury in the one day and two-day old hearts, but not at later time points.

                                                Furthermore, cardiomyocyte DNA synthesis was increased following neonatal pig myocardial infarction. Cardiomyocyte proliferation significantly decreased after this two-day window, which was associated with a marked reduction in telomerase activity.

                                                Heart failure with preserved ejection fraction may look different in the young compared to that in the elderly. First author, Dr Jasper Tromp, corresponding author, myself, Carolyn Lam from the National Heart Center, Singapore and Duke National University of Singapore, and our colleagues from the Asian Heart Failure Registry studied more than 1,200 patients with HEF PEF from 11 Asian regions and found that 37% of our Asian HEF PEF population was under 65 years of age. Younger age was associated with male preponderance, a higher prevalence of obesity, and less renal impairment, atrial fibrillation, and hypertension. Left ventricular filling pressures and the prevalence of left ventricular hypertrophy was similar in the very young of less than 55 years and elderly HEF PEF of more than 75 years of age.

                                                Compared to age matched controls from the community without heart failure, the very young HEF PEF patients had a three-fold higher death rate and twice the prevalence of left ventricular hypertrophy. Thus, young and very young patients with HEF PEF display similar adverse cardiac remodeling as their older counterparts, but very poor outcomes compared to controls without heart failure.

                                                Obesity may be a major driver of HEF PEF in a high proportion of HEF PEF in the young and very young.

                                                How important is hospitalization for heart failure as a complication of diabetes? In the next paper from first and corresponding author, Dr McAllister from University of Glasgow, the authors examined the incidents and case fatality of heart failure hospitalizations in the entire population age 30 years and older resident in Scotland during 2004 to 2013.

                                                Over the 10-year period of study, among 3.25 million people, the coot incidence rates of heart failure hospitalization were 2.4 per thousand-person years for those without diabetes, 12.4 for those with type two diabetes, and 5.6 for those with type one diabetes. Heart failure incidents had fallen over time for people with and without diabetes, but remained around two times higher in people with diabetes than those without diabetes. Heart failure case fatality was higher in people with type one diabetes. Duration of diabetes and glycated hemoglobin was associated with increased risk of heart failure in type one and type two diabetes. Thus, clinicians should be aware of the importance of heart failure and diabetes, especially in type one diabetes where this is under appreciated.

                                                What are epigenetic mechanisms contributing to ischemia reperfusion injury? Co-first authors Dr Yu, Yang, and Zhang, co-corresponding authors, Dr Xu from Nanjing Medical University, Dr Sun from Fudan University, and Dr Ge from Fudan University, and their colleagues evaluated the potential role of megakaryocytic leukemia one, or MKL 1, as a bridge linking epigenetic activation of NAD pH oxidases, or NOX, to reactive oxygen species production and cardiac ischemia reperfusion injury in mice. They found that genetic deletion of pharmaceutical inhibition of MKL 1 attenuated cardiac ischemia reperfusion injury in mice. MKL 1 levels were elevated in macrophages, but not in cardiomyocytes in vivo, following cardiac ischemia reperfusion injury.

                                                MKL 1 recruited the histone acetyltransferase, MOF, to activate NOX transcription in macrophages. Pharmaceutical inhibition of MOF attenuated cardiac ischemia reperfusion injury in mice, and pharmaceutical inhibition of NOX one or four attenuated cardiac ischemia reperfusion injury as well.

                                                These findings provide a novel link between MKL 1-mediated epigenetic regulation of gene expression in macrophages and ischemic heart disease. This opens the door to small molecule compounds targeting the MKL 1 MOF NOX access as a novel therapeutic strategy against ischemic heart disease.

                                                Is the time from last hospitalization for heart failure to placement of a primary prevention ICD associated with patient outcomes? First and corresponding author Dr Ambrosy from the Permanente Medical Group in San Francisco performed a post hoc analysis of Medicare beneficiaries enrolled in the national Cardiovascular Data Registries implantable cardioverter defibrillator, or ICD registry, all with a known diagnosis of heart failure and an ejection fraction of less than 35%, undergoing a new ICD placement for primary prevention.

                                                They found that older patients, currently or recently hospitalized for heart failure, undergoing initial ICD placement for primary prevention, experienced a higher rate of periprocedural complications and were at increased risk of death compared to those receiving an ICD without recent heart failure hospitalization. Additional prospective real world pragmatic comparative effectiveness studies should be conducted to define the optimal timing of ICD placement.

                                                The final original paper presents result of the VERDICT trial, a large scale randomized controlled trial evaluating the value of very early invasive strategy conducted within 12 hours of diagnosis on long term clinical outcomes in patients with non-SD elevation acute coronary syndrome. First and corresponding author Dr Kofoed from University of Copenhagen and colleagues studied 2,147 patients who were randomized and found that an invasive strategy performed within 4.7 hours after diagnosis was not associated with improved outcomes, compared to an invasive strategy conducted within two to three days.

                                                However, in the pre-specified subgroup of patients with a GRACE risk score of more than 140, a very early invasive treatment strategy did appear to improve outcomes, compared to a standard invasive treatment strategy. And that wraps it up for our summaries. Now, for our feature discussion.

                                                For our feature discussion today, we are talking about oxygen therapy for patients with suspected acute myocardial infarction. Something that seems so benign, something we've taken for granted, and yet now we now question since the Detox AMI trial. Well, for today's feature paper, we have a follow-up of this trial, and I'm so pleased to have actually our associate editor, but also author of this paper, Dr Stefan James from Uppsala Clinical Research Center, and the guest editor for this paper, Dr David Morrow, who's from Brigham Women's Hospital and Harvard Medical School. So, thank you both for being here.

                                                Stefan, could I just ask you to start by taking us back. How was Detox AMI first conceived? What made you even question oxygen therapy? And then, perhaps then, tell us about what this new paper adds.

Dr Stefan James:              I think that's so interesting because I think we all learned in medical school that for myocardial infarction, you should always deliver oxygen. That's sort of the first choice. And the other sort of first choice that we learned was morphine. Some of the other important things that we learned was to give not only oxygen but morphine, and nitroglycerin, and perhaps aspirin. And by those four, only aspirin is really the agent that has been proven beneficial to patients.

                                                But we thought for many years actually about this oxygen hypothesis, or we were interested in trying to understand, is it really helpful to give patients oxygen? Or are we in fact harming patients? Because there is, as you may know, there is a metanalysis performed long ago with small trials on the fibrinolysis era that showed actually a threefold increased risk of dying in those patients who had received oxygen in randomized various small trials, and their animal experience actually suggesting that oxygen is also hazardous. You don't think about that so often, but it's really an agent that constricts arteries, and so as the arteries close by a clot in myocardial infarction, there is no way the oxygen that you breathe in your nose can reach the suffering myocardium. It actually contracts the arteries, and may make the infarct larger than it would be otherwise.

Dr Carolyn Lam:                I love that explanation. Alright, so what did you find in the current analysis of longer term results?

Dr Stefan James:              So, we performed this, the main oxygen trial that we call Detox. We built it upon our national registries, and so we decided to include not only MI patients, but patients who were suspected of MI, in order to be able to enroll patients before the diagnosis was clear. We didn't want to wait for troponins, so we enrolled patients in the ambulances, in the emergency departments, in the cath labs, or in the wards, patients who had suspected myocardial infarction.

                                                Most of them, eventually, did have myocardial infarction, but a proportion did not have myocardial infarction. They had other diseases that resembles MI and have breathing problems. And we selected the cut point of 90%. We said if they are below 90%, they're hypoxic, and it would be unethical to withdraw oxygen, if you were hypoxic. So, we sort of arbitrarily selected the cut point of 90%. And then, we randomized patients to receive oxygen or do not receive oxygen.

                                                We considered to do double blind, but in order to do a double blind, you need to provide air on a mask. And air is not available in ambulances or in the emergency department. We cannot put a mask without anything in it because then it will feel more difficult to breathe. So, we had actually oxygen versus nothing, and we enrolled all patients coming to the cath labs, and emergency departments, and ambulances in Sweden. And thanks to the infrastructure that we have built on the national registries, we were able to enroll these to conduct this large trial, larger than any other trial, 6,600 patients.

                                                In the main study, we found no benefit, and fortunately, no harm of providing oxygen for our primary end point, which was all caused death. But we realized that we were little bit underpowered actually to really clearly rule out that there was any benefit on the primary endpoints. And so, we said, we probably need a longer follow-up, and we probably also need other important measures such as heart failure. Because we thought that oxygen may, if it works, it may reduce the infarct size and may result in a lower risk of heart failure in the long-term. We don't believe that we will reduce the risk of re-MI because we're not interfering with atherosclerosis or plaque ruptures, but we may interfere with the development of heart failure.

                                                So, in this particular paper, we said, longer follow up in order for patients to possibly develop heart failure and increase their risk of heart failure hospitalizations. So, in this paper, we used as a primary endpoint of this analysis, death or hospitalization for heart failure, post MI. And with this way of calculating events, we are more sure that we are not underpowered for this evaluation.

Dr Carolyn Lam:                Right. And the results?

Dr Stefan James:              The results were completely neutral. There was no benefit at all in any sub group. It doesn't matter if you were ST elevation MI, or no ST elevation MI, or no MI, or high risk prior MI, prior heart failure, respiratory disease, there is no benefits and no harm, which is good. And those results are supported by our findings on troponin levels. So, we checked troponins repeatedly. I shouldn't say top troponin, but the highest measured, we did not find any difference between the two groups in Troponin elevations. And we did not find any difference in LVEF and in Echo performed during the initial hospitalization.

                                                So, I think both of those results support the primary endpoint of death and repeat hospitalization for heart failure.

Dr Carolyn Lam:                So David, you've thought a lot about this, and also framed it so nicely when we were just talking a little bit earlier. What do you think is the real significance of this paper on so many levels?

Dr David Morrow:            Yeah, I think there are many levels. I think it's such important work because it takes something that we are still doing in many hospitals every day for patients and is difficult to study because it's become part of standard of care, as Dr James pointed out, and so the authors are to be congratulated for being able to study this intervention. And I think in additionally because it is a therapy that's not associated with high cost, has been part of our care for so long, it's not one where there is the support for a large type of randomized trials. So, the ability to perform this with relatively low costs by nesting it in a registry is important, not only for this particular test, but also as a model for future research of so many interventions that we make right now where they started in a time where our threshold for a need for data was much less.

Dr Carolyn Lam:                Yeah. Indeed. That's wonderfully put. I am also really struck. It's the importance of the message, but also especially about how you do a pragmatic registry-based randomized trial. The ability of Sweden to do this, it's just rock the world, right? Because we really need solutions like that for our clinical trial world, which has to be sustainable somehow. Could you maybe take us behind the scenes a little bit? I mean you did already in your description. I didn't realize there were so many considerations when you're planning this, but how easy or difficult is it to do a trial like this?

Dr Stefan James:              We call the entity RRCT. We call it registry based randomized trial, but being aware that there is no strict definition of what is a registry based randomized trial. So, sometimes for some simple interventions like strategies, we can use only the registry for collection of baseline variables, procedure variables, and also outcomes. The registry can really do everything. The only thing we need to add is a randomization, so then we just program into the registry, which is used live in front of the patients.

                                                So, when I enter a patient in the registry, the personal identification number collects me to the population registry that supports directly back to me name and gender of the patient, and then I enter all the baseline characteristics anyway in the registry. And then, there is a question that comes up that screens my patients. So, the system proposes to me to randomize patients who are eligible because I programmed the inclusion/exclusion criteria. So, it proposes to every doctor in the country, this is a patient that is eligible potentially for this trial and just click randomize, and that's the trial. Everything is completed by that. No extra tests, no visits, no follow up, no telephone calls.

                                                That's the basic, very simple format that can only be used for a strategy, like a device or a strategy. But many of the questions we have in medicine are really regarding strategies. How long should you treat? How often do you need to come back? Sort of strategies. Then, when we've tried to expand this to pharmaceutical agents, and oxygen was the first pharmaceutical agent that we wanted to try. You may not consider oxygen as a pharmaceutical agent, but it is in fact. But it's not manufactured by any companies, and we are still, in this trial, wanted to keep all-cause mortality as the primary end point because that's very reliable. That's indisputable, and in our country it's absolutely 100% correct. If they registered dead, they are dead. There's no question.

                                                The next level we did in the validate was a true pharmaceutical agent manufactured by a company, [byobatterin seprin 00:18:31]. A little bit more complex because you need to be careful about making sure that the patients are receiving the pharmaceutical agent in the right manner, in the right time point. We need to be a little bit careful about collection of side effects, and complications, and so on, but it also worked very well in that trial. If they validated, we did actually adjudicate events because in the primary end point we had it where it was more complex primary endpoint, including myocardial infarction. If you include myocardial infarction or bleeding events, that needs to be defined in a certain way according to protocol. You need to adjudicate. If you really need to rely on the outcome assessment.

                                                We're not trying to take this type of study to the next level, to use it for typical oral pharmaceutical agents. Our largest trial now running is the spirit HFPF lactone versus no treatment in patients with HFPF. And again, this is a pharmaceutical agent that is a very inexpensive. There's no company that would sponsor such a trial, but we think it's a really important question. There's so many patients that suffer from HFPF, and in order to do that trial, it has to be simple and inexpensive.

                                                So, that's running. We hope to be successful. There are, of course, many challenges. Like any other trial, it's difficult to write a protocol. You have to be very dedicated and detailed for any trial. So backstage, this is not easier than any other trial, but for the investigator, it is much easier. That's the reason we have succeeded to reach out to every hospital in the country, and every physician seeing these patients are investigators. And many of them have never done any trials before. They have no experience with research, but still they should be able to randomize and do the trials because it seems to be so easy for them and for the patients. That's the whole idea.

Dr Carolyn Lam:                Yeah. I'm just enamored by the whole concept, and of course, a lot of people I think are wishing that we could institute that in all countries as well. Trust me, a lot of conversation has occurred about that in Singapore, for example, where population based capture is possible. But, as you said, it's not that easy. It's got to be well thought out. Protocols still have to well thought out. Investigators still need to be trained, and so on.

Dr Stefan James:              We want the investigators to feel that it's easy, that it's attractive to participate. Not for money, just because it's so easy and so interesting to be part of such an experiment.

Dr David Morrow:            I think testing some of those therapies that are commonplace that they're used to, and our nature of practice is this is the perfect type of setting than more complicated interventions where you may need to train the investigators more in order how to implement to them, and apply the therapy correctly. That's the new trend, is ... I think the key issue is that in order to reliably test things where mortality is not the acceptable outcome that you could power adequately for, it's really the endpoint collection in the safety collection, and because of the robust medical record systems you have, you're able to do that. And we're so far from being able to do that reliably in the United States right now that it's not possible to do that. Unless we have specific well-constructed registries, which we do in some areas. I think we're learning, and hopefully we'll get there, but we're far behind [crosstalk 00:21:55].

Dr Stefan James:              [crosstalk 00:21:55] Yeah, but even-

Dr David Morrow:            [crosstalk 00:21:57] Nationals-

Dr Stefan James:              Even if you're not able to do a registry based, I think we all should consider in all trials to do it as easy as possible and really try to ask ourselves, what is the most important reason we're doing this trial? Sometimes we need to collect a lot of extra information because we need to understand the mechanisms or the side effects. If that's the case, I don't think at this trial methodology is not suitable. You shouldn't perform it that way. It needs to be the more traditional, more conservative, more expensive and burdensome way, but for many therapies, a more simple approach, more pragmatic approach is preferable.

Dr Carolyn Lam:                Well, thanks again for diving into that because it gives us a real, to me at least, even greater appreciation for this paper when you understand the amount of work that's gone into it. But may I just end by saying, what do you think is the take home message for clinicians now? David, for example, you started by saying everyone's still doing it? I fully agree.

Dr David Morrow:            Yeah. I think it's a very simple message, and that we know that oxygen is not effective in patients who have an oxygen saturation above 90%. And there's really no rationale to use it.

Dr Carolyn Lam:                Perfect. Has this been put in practice in Sweden already?

Dr Stefan James:              It has been. One of the virtues of running these registries is that we can also check the adherence to the results, so we can check that this is not used anymore.

Dr David Morrow:            And since the investigators are your entire country, they all learned actually from participation in these trials.

Dr Stefan James:              Exactly. Exactly.

Dr David Morrow:            There's more of an investment in it already.

Dr Carolyn Lam:                That's amazing. So, thank you again for sharing. Thank you for publishing this in circulation and for helping us to do that.

                                                You've been listening to Circulation on the Run. Don't forget to tune in again next week.

                                                This program is copyright the American Heart Association in 2018.