Activity Transcript

Nicholas L Mills. MD, PhD: Hello, welcome. I'm Nick Mills, Professor of Cardiology at the University of Edinborough in Scotland. Welcome to this program titled, "The Use of High-Sensitivity Cardiac Troponin Testing in Cardiovascular Primary Risk Assessment: Assessing the Evidence."

Joining me today are 2 phenomenal researchers from across the world. We have Stefan Blankenberg, Professor of Medicine at University Heart Center in Hamburg, Germany. Welcome, Stefan.

Stefan Blankenberg, MD: Hello.

Dr Mills: And joining him, Rob Christenson, Professor of Pathology at the University of Maryland School of Medicine in Baltimore, Maryland. Welcome to you both, guys.

Robert H. Christenson, PhD: Great. Well, great to be here.

Dr Mills: We are all familiar with high-sensitive troponins. They are vital to everyday practice for the diagnosis of acute myocardial infarction in the emergency setting. We've known for a long time that they're also extremely powerful predictors of risk, which has raised this tantalizing prospect that we may be able to use this biomarker to guide risk, particularly in a primary prevention setting.

So today's program is going to take a deep dive into the evidence that underpins the analytical and clinical performance of high-sensitivity cardiac troponin as a risk assessment for a physician. So, if we are to expand the role of cardiac troponin into risk assessments, we first need to understand what current tools are available in practice. So around the world, we use different tools, and I wonder if I can invite Stefan to start by talking us through the widely used Systematic COronary Risk Evaluation (SCORE) in Hamburg, Germany, and in Europe, and how you use that in your day-to-day practice.

Dr Blankenberg: Thanks, Nick. Perfect. I think it is absolutely important to outline the current standard of care before we discuss the advantages of introducing a biomarker like high-sensitivity troponin I or T into our clinical routine. So far, we benefit from a much-improved SCORE evaluation here in Europe, which we are using on a daily routine, because the European SCORE tool substantially improved compared with the recent SCORE we applied 2, 3 years ago. Because recently, we only, at least in Europe, had the option to predict risk according to family events. And now we are able based on the work which had been led by the European Society of Cardiology to implement the new SCORE2, we introduce 5 or 6 risk factors like blood pressure, cholesterol, smoking, and others, and they are calibrated according to different European risk regions and provide very accurate risk estimates over the next 10 years to suffer from any fatal or nonfatal cardiovascular event. So this had been a huge improvement.

On the other side, we definitely know that the risk estimates we have only cover in male and female about 55 to 60 percentage of risk. So, 2 aspects are absolutely vital. First, we still have room left to improve the risk prediction for future cardiovascular events. This is a first lesson, and the second lesson is that we are still missing a cardiac-specific instrument to mainly mirror more periods or more recent risk, but nevertheless, I want to stress that the instruments of the risk prediction we have in our hands, which is at least here in Hamburg, in UK, in Europe, which is SCORE2 is a very powerful one, but nevertheless there is room left for improving risk prediction.

Dr Mills: So Stefan, I completely agree with you, and although United Kingdom has done their best to track themselves from Europe, I think the SCORE tool still can be widely applied in the UK and has been calibrated in [the] UK population. We also have tried in the UK to introduce other factors in the QRISK and assigned score which adjusts for social deprivation and other things, which are possible to do on a local scale but not on a global scale. Can I ask this question to Rob? Give us your perspective from the United States, what is [the] standard care currently in Baltimore?

Dr Christenson: So, the standard care for use of cardiac troponin of course is in acutely ill patients. So, I think right now the risk assessment in the United States has to do with many of the same variables that Dr Blankenberg talked about. So, it'd be gender, it would be cholesterol, it would be smoking, it would be age, that would go into a calculation and determine an estimate of the next 10-year risk. Given that, though, just like what was mentioned before, that's only about half the risk that patients are up against. So having another variable that might be able to lessen that amount of uncertainty would be great.

Dr Mills: I'm going to pose you a challenge here. Is the opportunity here for cardio biomarkers to tell us something more immediate about direct injury to my heart today? And could we use that information to, I think, change people's thinking and perspective?

Dr Christenson: So I think that it would still be very important, and the reason is that, just because you're feeling well there may be smoldering myocardial injury under the waterline that, you feel fine, but there may be something that's modifiable that you could do about that high troponin level. We know that I think the evidence is very strong that those subjects are at increased risk, and I guess the question is, what can we do to mitigate that risk? It seems like changes in lifestyle can do that, although we all know that having patients make changes in lifestyle is the most difficult thing to accomplish. But I think just recognizing that you're at increased risk would be a valuable thing to subjects.

Dr Mills: Okay, so I think we'll maybe come back to what recommendations we would make and people who are increased risk using cardiac arm markers later in the discussion, but before we get to that point, there are some important analytical issues to consider before someone was to adopt this into practice. So Rob, I want just to, if you could just talk us through from your perspective, what are the key sort of analytical issues if we were to try and adopt this approach of practice?

Dr Christenson: So I have to say as someone for whom laboratory medicine is their chosen profession, we have not done a very good job with cardiac troponin as far as standardization goes, as far as having all the values from hospital to hospital, from assay to assay be the same. So there are pretty wide differences between cardiac troponin assays. There's cardiac troponin T, there's cardiac troponin I, and they vary even more. So we have analytical challenges that lay ahead of us I think. It's been quite an innovation, though, to have high-sensitivity cardiac troponin assays available because, they not only measure, they can measure down to the normal range. We can see what a normal troponin might be for individuals, but also the precision, the uncertainty surrounding those measurements. The 95% confidence interval is much smaller with high-sensitivity troponin assays. I think 1 thing we have to remember, too, is we're not measuring a different anolyte. This is still the same troponin that's been around for the last 30 years, it's just that the innovation has been just much better assays to do these measurements.

Dr Mills: Okay. I cannot stop the opportunity to ask you both. So troponin I and troponin T, in our acute care practice we consider them to be much the same. They provide the same insights into the diagnosed acute myocardial infarction, and there is some evidence suggesting that they might differ in their ability to predict future risk within particular cardiac troponin I predicting coronary events and cardiovascular death, whereas troponin T seems to provide insights more into all-cause mortality. This is just my observation. Do you think this is important? If so, why? I mean, one would be better or worse for this particular task.

Dr Christenson: So I do think that there is a difference between cardiac troponin T and cardiac troponin I. This is one of the caveats that moving a cardiac troponin or even a high-sensitivity cardiac troponin into this risk profile will be important. And again, all-cause mortality. Gee, you wonder why, how come one is able to do better than the other? Well it may be that cardiac proponent T might not be quite as cardiac specific. There may be other causes for cardiac troponin increases besides just issues with the heart. If it's re-expressed, for example, with other kinds of sort of adverse events. But anyhow, this is one of the differences that we're going to have to deal with if we go this path.

Dr Mills: Stefan, I was just going to say, what's your perspective? Are these 2 biomarkers telling us the same thing in this setting or not? If so, what mechanisms is as informing as that?

Dr Blankenberg: Yeah, I think it is a little bit too premature to differ between troponin T and troponin I in terms of very long-term perspective and prediction in the apparently healthy population. What we observe in our very large assessment of mainly European population, but also from others from the United States and so on and we just recently measured first quarter in Africa as well . . . we observe that both are very strong predictors for short-, intermediate-, and very long-term events. So that's a first very important message, I believe. Second, they have very strong predictors for overall mortality both, and a little bit less strong for cardiac specific mortality or nonfatal cardiovascular disease events, but to distinguish between troponin and troponin I in terms that one is stronger in that or in that direction, I personally feel that this is a little bit too premature, but nevertheless, they are both very strong predictors.

I see 1 particular difference because, some of the high-sensitive troponin  I analytes are able to detect very, very, very low specific concentration in the healthy population, which is in high-sensitive troponin T, a little less possible at that stage in our evidence, and therefore my impression is that in the very low ranges of troponin concentration which are reliably detected, we can distinguish a little bit more at the lower ends with high-sensitivity troponin I. But having said that, I'm very cautious overall at the current stage, I would clearly state both are very strong predictors for future cardiovascular and overall mortality events.

Dr Mills: Yeah, yeah. Could not disagree. I think the other really critical question for people considering this, once you move beyond which assays, so this is a high-sensitivity assay with this sort of precision performance that the raw highlights is how do we use it? And Stefan, you've done some very nice work across a consortium in Europe defining threshold. Can you just talk us through how you would propose to use this in this setting, to define risk?

Dr Blankenberg: As a clinician, and potentially in future or even right now, applying this in my ambulatory practice, we definitely want to catch something where we can say: Oh yes, like low-density lipoprotein cholesterol --below 100, above 100, below 70, above 70, below 55, above 55 --we need to have something in our hands where we can classify, where action is maybe needed at that point. So this is something which supports, but defining a cutoff or a threshold, we definitely on the one hand miss information; on the other hand, we have clearly the option to tell our patients, "Look. We have something, let's say above 6, for instance, which we can, in the area under the curve distinguish as in need of risk reduction, as a potential cutoff. You are a little bit at risk, change your behavior, you should maybe intake statin therapy, you should maybe intake aspirin therapy” and so on. So this is something helpful, but I'm not quite 100% confident if we should really apply thresholds or cutoffs at the current stage. I rather take this high-sensitivity troponin in the current practice, apart from the diagnosis of acute myocardial inflammation, by mirroring a longer time period and observing whether there's a change in the troponin over time. So this is very useful information.

Dr Mills: Yeah, couldn't agree more. And can I ask, Stefan, on your perspective on some of the research that's come out of the UK recently by Dorian Kimani and  colleagues has looked at serial measurements across the life course and demonstrated that your trajectory of troponin over a 20-year period is highly predictive of your risk of cardiovascular death, and the trajectory, although troponin tends to increase with age in everyone, it increases more steeply if they've got cardiovascular disease.

And another really important insight from that work is that trajectory differs in men and women. And I'd like just to for you to comment a little bit on your thoughts as to how we apply that insight, the troponin changes over time, and it differs critically between men and women, how you would apply that in adopting a kind of threshold-based approach for risk stratification?

Dr Blankenberg: This is a very challenging question. First of all, it's a very interesting observation, and it's clearly understandable that risk markers, and in particular cardio-associated markers, increase over age, and they certainly differ between male and female. On the other hand, how do we introduce this in our clinical practice, and therefore, it is probably not appropriate to define 1 clinical threshold for any age and for male and female at the same level. We probably need to adapt, because if I, or an individual at an age of let's say, 40 [years], has troponin concentration of, let's say 14 pg/nL, at an age of 75 [years] is going to be completely different.

And so, I think the future solution might be that you introduce this information into a bigger surrounding included in your algorithm: age, sex, and maybe 2 or 3 other variables, such as  troponin concentration as a key variable, and then you have the adapted information. But again, this is a music of the future, and it is very important to understand that there is development of this cardiac biomarker over time.

Dr Mills: Can I bring in Rob here to comment? You've done some very recent important work, I think, in the National Health and Nutrition Examination Survey (NHANES) population, looking not just at age and sex, although they are important variables, but also in this simulation, could you comment on whether we need to take a different approach from different ethnicities or what does the that NHANES study add to this area?

Dr Christenson: Yeah, so NHANES data definitely show that there is a race difference and there is it certainly an ethnic [difference], and I guess one of the things that we really have to be careful of is the comorbidities that may impact these different populations. When a population's been underserved for a long period of time, their hypertension, they're lipids, they're body mass, etc, are all impacted, and I think these are things that we have to be careful that we take into account comorbidities as well as race and ethnicity so, we're learning a lot, but I think there's still more to come.

Dr Mills: Okay. So can we move on to the application, clinical application of this? What are the clinical consequences for a patient, Stefan, who has a value that puts them above a high-risk threshold? What are the practical steps that you would take in that setting in order to try and mitigate that risk?

Dr Blankenberg: This is a very important question. What do I do if I have a patient with a severe 3-vessel disease, coming back from the catheterization (cath) lab, 3 months later, stable and I have a clearly elevated troponin, but the sustainable, I'm not concerned because I know it's a very severe coronary artery disease with a continuous leakage of very low troponin concentrations, and it definitely gives me a small hint whether or not I should treat, in terms of dual antiplatelet therapy, in terms of introducing other therapy like COMPASS scheme for instance and so on and so on.

A completely different story is if I have a patient or individuals, actually a currently healthy, with a little bit higher measurable troponin concentrations, I'm concerned, because I don't find an immediate reason. Then I start the workup with imaging -- echocardiogram (echo) for sure -- in some circumstances magnetic resonance imaging (MRI), I'm very much interested about the musculature of the heart, the MRI, and then I will definitely treat the hypertensive disease, treat other diseases much more rigorous compared to someone who is a little bit hypertensive to the completely normal or completely lower by sensitive troponin I. In summary, very clear, I don't have a very clear immediate action plan, but it gives me a very valuable additional hint whether or not I should intensify my therapy or add different diagnostic procedures. So that's the way.

Dr Mills: I mean, I think you've described it perfectly. You're using clinical judgment. This is an additional tool that helps you understand whether we're managing current risk factors adequately or not and whether we should intensify or not care. I wonder if it's worth for us discussing the modification of troponin. We've done a little bit of work here with WOSCOPS, which demonstrates that if your troponin concentrations are elevated, yes, you're at high risk of events in the future, but interestingly, pravastatin in the WOSCOPS study reduced troponin by a quarter, and those that had the greatest fall in cardiac troponin on single measurements, entirely independent of the effect that statins had on LDL cholesterol, had led by far the greatest risk reduction, suggesting that it was giving some insight into the effectiveness of treatment. I don't fully understand the mechanism of that, but it's fascinating that the findings were very consistent in JUPITER, which was a primary prevention population, a broader population in WOSCOPS. I wonder if you could just comment a little on your interpretation of that observation and what that means for practice.

Dr Blankenberg: First of all, it makes it very confident that in 2 studies, 2 important studies like WOSCOPS, as you described, and Jupiter, what we confirmed is clearly associated with an improved risk reduction in those individuals having a higher troponin concentration, so the first lesson is, we have in those individuals with higher-sensitivity troponin concentrations, improved risk reduction, or an enhanced risk reduction, and it is worthwhile to treat those individuals even more carefully. Secondly, the observation that troponin concentration decreases and the decreases associated with an decreased risk in future, so the magnitude of effect mirrored by troponin is associated with outcome is a very important one, because troponin mirrors perfectly an entire cardiac specific risk, and therefore it might be very worthwhile to monitor it and to see whether it is stable, decreasing or increasing and reacting accordingly.

I also do not fully understand the pathophysiological background of that, but a perfect theory is that this mirror of the heart prompts me to guide therapy more intensive or less intensive. So therefore it is very useful.

Dr Mills: This has been a fabulous discussion. I want you to now give us some insights before we finish into the future. When do you think we will be using cardiac biomarkers, high-sensitivity cardiac troponin in routine like we do in cholesterol, for the primary prevention and risk stratification? And how do you think we're going to be doing? So, I'll start with Rob.

Dr Christenson: So, I don't think we're done yet with high-sensitivity troponin. I think there will be other generations that even measure troponin down lower that will improve our ability to risk stratify hopefully, but at least we'll have better insight as to what's going on, because we talked about high-sensitivity troponins like they're just 1 animal, and we know there's differences even in their quality between the assays. I think Dr Blankenberg mentioned that with some of the assays it can measure very precisely very low. The other is, I wonder if another biomarker, another cardiac-specific biomarker, and I'll talk about either N-terminal pro b-type natriuretic peptide (NT-proBNP) or BNP, might help when Dr Blankenberg was giving his articulate discussion on how he might treat patients, whether that other biomarker might give insight into hemodynamic stress, for example. Is it heart failure that may be especially in patients with hypertrophy or some other? So I think there's, we're really only scratching the surface here, and it's really a very exciting area to think about the future of this and sort of asymptomatic individuals.

Dr Mills: Great point, Rob. Stephan, are we ready?

Dr Blankenberg: Yeah, I think I want to summarize in 3 aspects. First, the classic core estimates leave a lot of room for additional risk prediction. So that's the first message. The second message is, with the cardiac-specific biomarkers, we might have tools in our hands to close this gap a little bit more. And the third message is, yes, we are using this currently, cautiously to understand the overall risk, cardiac-specific risk a little bit more. But the real breakthrough will be once we perfectly understand the immediate action which results out of specific increased troponin concentration, once this step is closed, we will use it routinely. Nick, whether this is in 1 year or 10 years, I'll leave it to you.

Dr Mills: Very good summary. And I agree with both of you. We are learning more and more every year about how we can use these biomarkers to help patients. I think we all firmly believe that Stella is something really important. There's ongoing research to help us define what the best care pathways and strategies are for patients, but can they improve care in the future? I believe firmly they can. So Stefan, Rob, thank you for what's been a great discussion. Really appreciate you joining us today. Thank you also to our audience for participating in this activity. Please continue to answer the questions that follow to complete the evaluation. Thank you.

This transcript has not been copyedited.