Circulation August 25, 2020 Issue

This week’s episode of Circulation on the Run has 2 Feature Discussions. Associate Editor Ntobeko Ntusi discusses the article "Prevalence of Infectove Encocarditis in Streptococcal Bloodstream Infections is Dependent on Streptococcal Species." Then, author Anumpam B. Jena and Associate Editor Sandeep Das discuss trends in new diagnoses of atrial fibrillation after the release of an ECG-capable smartwatch.

TRANSCRIPT:

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.

Dr Greg Hundley: And I'm Dr Greg Hundley, associate editor from the Pauley Heart Center at VCU Health in Richmond, Virginia. Well, Carolyn, our feature this week involves infective endocarditis and looking at that in streptococcal bloodstream infections. Are they dependent on the different species of streptococci? But before we get to that, how about we grab a cup of coffee and start off and discuss other papers in the issue. You want to go first?

Dr Carolyn Lam: Sure, Greg. I got my coffee, but I want to start with a quick question. So do you think it's safe and efficacious to prescribe SGLT2 inhibitors to patients with Type 2 Diabetes and concomitant peripheral artery disease?

Dr Greg Hundley: Ah, Carolyn, very good question. I don't know if it's right or wrong. A little bit of controversy here.

Dr Carolyn Lam: Let's explain that controversy. So patients with peripheral artery disease are at heightened risk of cardiovascular complications. However, there's also an increased risk of amputation that was observed with canagliflozin in one prior trial. Now, the SGLT2 inhibitor dapagliflozin was shown to reduce the risk for hospitalization for heart failure and kidney events in patients with Type 2 Diabetes in the DECLARE–TIMI 58 trial. So authors Dr Bonaca from University of Colorado School of Medicine and colleagues examined the cardiovascular and kidney effects and the risk of limb related events in patients with and without peripheral artery disease in the huge DECLARE–TIMI 58 trial, including more than a thousand patients with peripheral artery disease.

Dr Greg Hundley: So this could be really helpful to answer this question. What did they find, Carolyn?

Dr Carolyn Lam: Well, patients with versus without peripheral artery disease were indeed at higher risk of major adverse cardiovascular events, cardiovascular death, or heart failure, hospitalization and kidney events. They also had consistent benefits for the outcomes of cardiovascular death or heart failure hospitalization as well as progression of kidney disease with dapagliflozin. Now ,patients with peripheral artery disease also had a higher risk of limb events, but there was no consistent pattern of incremental risk observed with dapagliflozin. So the take home, diabetes patients with peripheral artery disease are at risk of heart failure and kidney events and dapagliflozin is beneficial with no patterns of increase limb risk.

Dr Greg Hundley: Very nice, Carolyn. Boy, that is very helpful and a nice take home message for all of us administering these agents. Well, Carolyn, my first study comes from Professor Magnus Bäck from the Koralinska Institute. And the aim of this study was to identify the role of omega−3 polyunsaturated fatty acids, derived specialized pro resolving mediators, or SPMs, in relation to the development of aortic valve stenosis. The synthesis of specialized pro resolving mediators are potent beneficial anti-inflammatory agents. They have pro resolving and tissue modifying properties that are useful in managing patients with cardiovascular disease.

Dr Carolyn Lam: Interesting. So what did these authors find?

Dr Greg Hundley: This study showed that human stenotic aortic valves contain decreased levels of n-3 PUFA, and that n-3 PUFA treatment decreased aortic valve calcification, and aortic valve leaflet area in murine models’ concomitant with improved aortic valve hemodynamics. The pro resolving lipid mediator, resolvin E1, which is derived from the n-3 PUFA enoic acid, or EPA, exerted protective effects on valvular interstitial cell calcification and valvular inflammation through its receptor, chemR23. And therefore Carolyn, further clinical evaluation of n-3 PUFA treatment may open up novel therapeutic opportunities for preventing the progression of aortic valve stenosis.

Dr Carolyn Lam: Wow, really interesting, Greg. Just more and more data on these omega−3 PUFAs, huh? So cool. Well, the next paper is kind of related in the lipid world. Do you think treating to a low LDL cholesterol target of less than 70 milligrams per deciliter may impact carotid plaque evolution?

Dr Greg Hundley: I would think so. It seems like that target is beneficial in many ways.

Dr Carolyn Lam: Well, good guess, Greg. But I'm going to tell you about a study that actually looked at that. First of all, recall that the treat stroke to target, or TST trial, showed the benefit of targeting an LDL cholesterol concentration of less than 70 in terms of reducing the risk of major cardiovascular events in 2,860 patients with ischemic stroke with atherosclerotic stenosis of the cerebral vasculature. Now, today's paper describes results of the parallel TST plus study, which included 201 patients assigned to an LDL cholesterol concentration of less than 70 versus 212 patients assigned to a target of a hundred. To achieve these goals, investigators led by corresponding author, Dr Amarenco from Bichat Hospital in Paris, France, allowed investigators use the statin and dosage of their choice, and added ezetimibe as needed. After certification of ultra-sonographers, carotid ultrasound examinations were performed at baseline and at two, three, and five years, and blindly analyzed at a central core laboratory.

Dr Greg Hundley: Very nice, Carolyn. So what did they find?

Dr Carolyn Lam: After a median follow-up of 3.1 years, patients in the lower target group had a similar incidence of newly diagnosed carotid plaque compared to the higher target group, but significantly greater regression of carotid atherosclerosis as measured by the common carotid intima media thickness. So this really further strengthens the concept that the lower the LDL cholesterol, the better the clinical and atherosclerosis outcomes.

Dr Greg Hundley: Very nice, Carolyn. Another study emphasizing that very important point. Well, Carolyn, my next study comes from Professor Abdelkarim Sabri from the Temple University School of Medicine. So studies suggest that cardiac rupture can be accelerated by thrombolytic therapy, but the relevance of this risk factor remains controversial. In this study, the authors analyzed protease activated receptor or PAR4 expression in mouse hearts with myocardial infarction, and investigated the effects of PAR4 deletion on cardiac remodeling and function post demise by echocardiography, quantitative immunohistochemistry, and flow cytometry.

Dr Carolyn Lam: Oh, okay. What did they find, Greg?

Dr Greg Hundley: Three things. First, PAR4 deficiency leads to cardiac hemorrhage and increases the rates of cardiac rupture following chronic myocardial infarction. PAR4 deficiency in neutrophils, but not in platelets, impairs inflammation resolution and myocardial healing after myocardial infarction. And finally, adoptive transfer of neutrophils can be used as a novel therapy to modulate the inflammatory response and improve cardiac remodeling and function following myocardial infarction.

Dr Carolyn Lam: Okay, what's the take home message?

Dr Greg Hundley: It's a little tricky. So acute transient administration of par four inhibitors may provide a new approach to prevent early inflammation and myocyte loss immediately after ischemic injury. The keyword is immediately. But importantly, prolonged P4 inhibition strategies could impair myocardial healing and increased cardiac hemorrhage and the rates of myocardial rupture following infarction. PAR4 inhibition therapies should be limited to the acute phases of the ischemic and fault, and it should be avoided for the chronic treatment post myocardial infarction. Really intriguing, Carolyn.

Dr Carolyn Lam: Very nice and elegant results, kind of like yin and yang, huh?

Dr Greg Hundley: You bet.

Dr Carolyn Lam: All right. Let's sum up with the other papers in the issue. There's a white paper by Dr Psotka on challenges and potential improvements to patient access to pharmaceuticals with examples from cardiology. There's a perspective by Dr Armstrong comparing the benefit of novel therapies across clinical trials, and that provides important insights from the VICTORIA trial. There's an ECG challenge by Dr Chu regarding a young male with incessantly alternating tachyarrhythmias.

Dr Greg Hundley: Very nice, Carolyn. Well, in my mail bag, I have an on my mind piece from Dr Jamil Tajik relating to, our favorite, the art and science of occultation. Well, Carolyn, how about we get on to that feature article and talk a little bit more about those little devils, the streptococci?

Dr Carolyn Lam: Yeah, let's go, Greg. Infective endocarditis is the topic of our feature paper today. Now, we all know it's a life threatening disease, and despite its relative rarity, it's still consumes a disproportionate share of healthcare resources, and its annual mortality is still really high, exceeding 20%. Now, I think we all recognize that improved outcomes are critically dependent on a timely diagnosis and early investigation. The problem is the clinical presentation is less and less likely that classical textbook presentation, and the clinical suspicion is often triggered after microbiological identification of potential causative organisms in the blood. And while we as a medical community are usually aware of the dangerous of staphylococcal bacteremia, I think there's a lot less appreciation of the propensity of different streptococcal species to cause infective endocarditis. Greg, my dear partner, greatest friend, and colleague, do you agree?

Dr Greg Hundley: Yes, Carolyn. So this paper and this feature addresses bacterial endocarditis and focuses on streptococcal infections just as you've described. Now, streptococci I frequently cause infective endocarditis. Yet, the prevalence of infective endocarditis in patients with bloodstream infections caused by different streptococcal species is unknown. So in this study, Dr Shamat and associates aim to investigate the prevalence of infective endocarditis at species level in patients with streptococcal bloodstream infections.

Dr Carolyn Lam: Now, we're taking a lot of pains to describe this paper, Greg and I, because we didn't manage to get hold of the authors this time. We certainly have our editor who managed the paper and that's Ntobeko Ntusi from University of Cape Town. So welcome Ntobeko, and thank you so much for discussing this paper with us. But before we get to that, I think Greg and I are going to try to, in our usual fashion, get to the bottom of describing. Here's something I learned that I didn't know before. That the ESC guidelines for example, are primarily based on the modified Duke criteria, which include blood cultures, mentioning viridians streptococci. Remember those? Viridians streptococcus, or strep bovis, as a diagnostic major criterion for streptococcal infective endocarditis.

And yet, the term viridians is based on bacterial culture using green hemolysis on blood agar plates, which is outdated. So I didn't realize that. It's outdated and inconsistent because some of the included streptococcal species do not even cause hemolysis, and other species are able to produce different kinds of hemolysis. And thus, we really need more details on specific streptococcal types that are much more current. And when we face these different streptococcal species, they're not mentioned in the guidelines and so we need more data. So that's why this paper is so important. So Greg, now over to you. Do you mind to describe what the authors did?

Dr Greg Hundley: Sure, Carolyn. So these investigators identified and assessed all patients with streptococcal bloodstream infections from the period of time of 2008 to 2017 in the capital region of Denmark. And data were cross-linked with Danish nationwide registries for identification of concomitant hospitalization with infective endocarditis. In multi-variable logistic regression analyses, the authors investigated the risk of infective endocarditis according to some of those species that you just mentioned. And they adjusted their analyses for age, sex, greater than or equal to three, positive blood culture bottles, native valve disease, prosthetic valves, prior episodes of infective endocarditis, and whether or not they may have had an implanted cardiac device.

Dr Carolyn Lam: Great, great. So tell us the results, Greg.

Dr Greg Hundley: So Carolyn, they had 6,500 plus cases with streptococcal bloodstream infections. And the average age of the patients was 68 years, and a little more than half. So 52% to 53% were men. The prevalence of infective endocarditis overall was 7%. Now, the lowest infective endocarditis prevalence was found with strep pneumoniae and strep pyogenes, ranging from 1.2% to 1.9%. the highest infective endocarditis prevalence, and that was found with strep mitis or oralis at 19%. Streptococcus gallolyticus, formerly known as strep bovis, at 30%. Strep sanguinis at 35%, and strep mutans at nearly 50%, at 48% overall. So in multi-variable analysis, using the strep pneumonia at 1.2% as a reference, all species except strep pyogenes were associated with a significantly higher infective endocarditis risk. Again, the highest with the odds ratio of strep gallolyticus is at an odds ratio of 31 ranging up to straight mutans with an odds ratio of 81.3.

Dr Carolyn Lam: Whoa, Greg, that was beautifully summarized, and frankly, beautifully pronounced. I don't think I could have done that with all the species. That is so cool. And in case everyone didn't get it, I strongly suggest you refer to figure three of this beautiful paper. It shows the prevalence of infective endocarditis in bloodstream infections with the different streptococcal species, all in one figure. And Ntobeko, with that introduction, if you may, by both Greg and I, could you please let us behind the scenes? Tell us what you first thought when this paper came across your desk and perhaps what the editors’ thought was so important about this paper.

Dr Ntobeko Ntusi: Thank you very much, Carolyn and Greg. So this is an important paper coming out of circulation. And while infective endocarditis may not be so harmonic in North America and western Europe, in many parts of the world, including where I come from in Sub-Saharan Africa, with a high prevalence of rheumatic heart disease, it remains an important cause of morbidity and mortality. And we forget that it's a disease with very high inpatient mortality of up to 50% in many countries. And of course in those with rheumatic heart disease, streptococcal bloodstream infections remain an important cause of infective endocarditis.

So when I saw this paper, I very much enjoyed reading it. I thought it was well written and beautifully illustrated. In some ways, even though I say it's an original paper, it has a feel of a review because the discussion, as well as the figures and tables, are quite instructive. And the comments from the reviewers were very much aligned with my own thinking that there were a number of important new learnings coming out of this paper. The first important message for me was that the distribution of streptococcal infections in the population was not uniform and I had assumed infections to be broadly the same across the population. And contrary to what I thought, the risk of infective endocarditis was inversely related to the frequency of streptococcal bloodstream infections in the population.

In other words, the most common streptococcal blood infections that are very low prevalence of infective endocarditis. And that leads to the second important learning from this paper, which is that if you are a clinician evaluating a patient with suspected infective endocarditis, the risk of infective endocarditis should be evaluated on a species level, as the species from Greg's description is probably the most important determinant of the likelihood of developing infection. And then the other important, I think, take home message from this paper, I'm looking at the results of multivariate regression analysis, is that it confirms much of what we know from studies with older patients studies focusing primarily on staphylococcal bloodstream infections as well as studies focusing on device therapies.

And that message is that the risk of developing infective endocarditis, even with streptococcal blood infections is related down to presence of native or valve disease. Those with trust that tech devices, intracardiac devices, and of course, the number of fat blood culture bottles that are positive, which is something that is well accepted and established in clinical practice. Thank you, Carolyn.

Dr Carolyn Lam: I just love those three take home messages. So beautiful. And I really also love that you invited this fantastic editorial, Dr Prendergast, Dr Allen, and Dr Klein. I thought it was wonderful the way they summarized the paper, put it into context, and perhaps, I could borrow their words and also explaining that they pointed out that we do need to be cautious about interpreting this being a retrospective series classified by diagnostic coding. And of course, by design, because of that, we can't fully attribute causal associations.

They also pointed out that this paper, I believe did not include echocardiographic data as one of the variables. And so of course, that could be something that could be further explored in future studies. And I'd love your thoughts on what they also said about before we extrapolate regional data, I mean, it's all from Denmark after all, to other regions, this work should probably be considered something that should inspire extension of further studies and prospective evaluations in other areas, and yet never losing sight of the fact that this informative paper will of course be of considerable interest to not just cardiologists, but also infectious disease specialists, because we're often called to sort of assess, "All right, how much do you need to work up for infective endocarditis if you see this specific streptococcal bacteremia?" And this paper definitely puts us strides ahead in this area. Would you agree with that, Ntobeko, and anything to add?

Dr Ntobeko Ntusi: Indeed, Carolyn. So I was delighted that the review, the editorial, was co-authored by cardiologists and infectious disease specialists. We might view as they balanced in as one of the modifications to the original submission. We actually posed that question to the office that how can we be certain in the absence of echocardiographic data that the diagnosis was in fact infective endocarditis in patients? And so they went back and performed a sensitivity analysis and triangulated the principle diagnosis of infective endocarditis with the treatment of patients in the duration in hospital. And that sensitivity analysis in fact strengthened our believe that this wasn't fit the diagnosis of infective endocarditis. And I think your point is well taken that similar studies need to be conducted in different regions of the world and this field will be strengthened by large amount of prospective studies on top of the plethora of retrospect data that we have.

Dr Greg Hundley: Well listeners, we have a second feature discussion this week. A very interesting article pertaining to the use of smartwatches and detection of atrial fibrillation. And to present this work, we have Dr Bapu Jena from the Harvard Medical School, and our own associate editor from University of Texas Southwestern Medical Center in Dallas, Dr Sandeep Das. Welcome gentlemen. And Bapu, perhaps, could you tell us a little bit about the hypothesis and the background of this material?

Dr Anupam B. Jena: Sure. So you probably are aware that in December of 2018, Apple released this new watch and this watch made a lot of buzz in part because it featured this single lead EKG that the company said would be able to detect atrial fibrillation. And there was this question among many, at least certainly among clinicians, as to whether or not we would see a large increase in atrial fibrillation diagnoses after this watch was released onto the market. And then the second question was, would we be picking up worrisome cases of atrial fibrillation that we needed to act on, or would we be picking up cases that patients probably would have been okay living with and would never have known that they were living with?

We were lucky to be able to work with a company called Athenahealth. Athenahealth is a nationwide cloud-based healthcare information technology company. What that allowed us to do was to do a very early analysis of the change in atrial fibrillation diagnosis after the Apple Watch was approved on the market. So that's the data we use is from small physician practices. It's not nationally representative, but these practices are all across the US. The data from Athenahealth have been used in other studies, including some by myself and other colleagues.

In terms of the method that we applied; it was pretty straight forward. So what you basically want to see is before and after December of 2018, do we see a market spike in the diagnosis for atrial fibrillation? Again, these are diagnoses that physicians who are in these offices would be making and that we would be seeing in the electronic health records of their practices. Now, the problem is, is suppose over time, atrial fibrillation diagnoses are going up. So if we looked at the first six months of 2019 compared to the last six months of 2018, and we saw an increase, we obviously wouldn't want to attribute that solely to the Apple Watch because atrial fibrillation diagnoses may be going up for a lot of other reasons. What we said as well, let's at least try to account for the possibility that there are seasonal trends at play in the diagnosis of atrial fibrillation.

Dr Greg Hundley: Very nice. What did you find?

Dr Anupam B. Jena: So the basic finding is as follows that we didn't really see a differential increase in atrial fibrillation diagnoses. So for example, in the months before the watch was released, in this population, about 0.4% of all visits had an atrial fibrillation diagnosis. 0.4%. If you look at the year after the app release, about 0.4% of visits have the diagnosis. If you look one year back, you also find that one year before in the pre period, there is about 0.36% of visits were for atrial fibrillation, and that increased the 0.39%. So the difference in difference change is basically about zero. Meaning, we found that there was no increase in atrial fibrillation diagnoses. The second thing that we did is we looked at high income zip codes, and we looked at zip codes with the population of individuals would be such that we might expect an increase. And we found no changes in either one of those two subgroups.

Dr Greg Hundley: Very good. Well, Sandeep, help us put this in the context of using these watches and then also using these watches to identify patients with atrial fibrillation.

Dr Sandeep Das: Absolutely. So let me first just add a comment that Bapu was a little too modest to blow his own horn, but he's really built a lovely research program of identifying and answering really interesting questions in large datasets. So the fundamental question here about whether the use of the Apple Watch and these detection algorithms would be associated with a big spike in diagnosis of atrial fibrillation was extremely important or is extremely important. So that was really the hook. I knew it was going to be something interesting. In our heart of hearts, we're all a little worried that people are going to have wearables. There's going to be an 8 million people presenting with abnormal findings on their watch and it's going to break medicine. So that was really kind of the context and the key for what made it interesting to us.

Dr Greg Hundley: Very nice. And so what do you think, maybe start with Babu, and then come back to Sandeep. What do you think will be the next study in this area using these devices as they pertain to patients with atrial fibrillation?

Dr Anupam B. Jena: Yeah. Great question. So I think there's two things that come to mind. So first is this was really an early analysis. I do expect that as the Apple Watch grows in popularity and as other similar such devices get introduced to the market, that we probably will pick up patients with atrial fibrillation who otherwise wouldn't have been diagnosed. I think that's less likely, but would be diagnosed earlier than they otherwise would have. So I think that if we look a couple of years out, we probably will find different answers that we found here. But as Sandeep said, at least in the first year after the watch was introduced to the market, we didn't break the bank. So I think this first natural question is to look longer out. And I think the second thing, which is particularly interesting to me at least, is that this potentially gives us a nice natural experiment to understand whether or not all patients with atrial fibrillation or what are the other factors that we should be thinking about in terms of benefits?

Dr Greg Hundley: Very good. Sandeep, do you have anything to add?

Dr Sandeep Das: So I agree with that answer entirely. I think that one of the things that really is going to be the $64,000 question is what does wearable diagnosed atrial fibrillation mean? So we were pretty good. As Bapu said, we have well established history of understanding what to do with patients that have atrial fibrillation that we diagnose in conventional ways. People either present with symptoms or are diagnosed externally in the hospital. But what does it mean if you're perfectly fine and you just have a watch that gives you an alarm? What are the implications of that? Because there are therapies generally revolve around anticoagulation in a large fraction of patients with atrial fibrillation. And that's the big deal.

If we're going to commit people to lifelong anticoagulation, based on what they're watch told us, is that the right thing to do? So I think that to me, the most important question going is what does a diagnosis of a-fib by wearable mean for downstream treatment implications and outcomes? And then the larger scope is also how can we then incorporate the data that we're going to continue to get from wearables into practice? So studies on sort of the practical downstream implications of wearable technology on utilization are also going to be super important and interesting.

Dr Greg Hundley: Very good. Well, I want to thank you Bapu, and also thanks Sandeep for your time today. And just sharing this new research using Apple smartwatches and trying to diagnose atrial fibrillation and compared it at least in this first year to, I guess, historical controls, not an overabundance or mass increase in the diagnosis of atrial fibrillation. And as you both have identified more to come with research on using these watches in the future for management, and then how we might improve therapeutic interventions through the use of these devices. Well, listeners, we hope that you have a great week and we look forward to catching you on the run next week. Take care. This program is copyright of the American Heart Association, 2020.