Elevating AMI diagnosis: A deep dive into the global TSIX study

Guest Speakers:

Prof. Evangelos Giannitsis — Cardiologist working at the University Hospital of Heidelberg, Germany

Prof. Frank Peacock — Emergency physician, Professor of emergency medicine, Baylor College of Medicine, USA

Prof. Christian Müller — Professor of cardiology and director of the Cardiovascular Research Institute in Basel, Switzerland

Interviewer : Prof. Mamas Mamas — Professor of cardiology at Keele University in the United Kingdom

Dr. Mamas: So we often think about in clinical practice about the need and the importance of measuring troponin. Why is it so important? Why is it important that we can measure it precisely? 

Dr. Giannitsis: It’s not only about the precision. It's also about the time. The timing of making a diagnosis, make a disposition. So, rule out and rule in is very, very important. And precision means also having a better precision at the lower end and facilitating rule out of patients. And this translates into many, many benefits for all stakeholders.

Dr. Mamas: Having what we define as a normal troponin is important. Why do you think it is important to define what normal is? 

Dr. Peacock: So there's the cardiology 'normal' and there's the emergency 'normal', and I think that's the first part of the discussion. As cardiologists, you know, they have long-term management. And so that's a different normal. That's, I can say, in the 99th percentile. Get a thousand people and find out what the level is, and then there's the emergency doc normal. And that's the normal I'll throw you out of my ER and ensure you'll be alive for more than 30 days. Those are not the same numbers. So, my normal is usually around single digits, 2 or 7 or something like that, in that range, for most troponin assays. And the 99th percentile is in the 20s. So, big difference. What exactly the TSIX study involved? So, the TSIX  is probably the largest hs Troponin study ever done. All said and told, about 13,000 patients in the reference part of the trial and in the all-ACS-comers part. 

Dr. Mamas: Why do you think (TSIX) it is such an important study? 

Dr. Giannitsis: Because the databases from which reference populations were taken (in the past) to estimate, to construct the 99th percentile are very few. They were heavily selected, and some are biased. The definition of health is problematic. And, then there are some needs to be fulfilled, requirements that are set by, let's say IFCC criteria on the number of individuals who have to be included. I think that the Roche REF-TSIX study is the largest ever done in a very diverse population, covering the whole age spectrum, the genders, ethnicities. So this is a very, very important let's say that the most important reference study ever conducted. How do you think...now that we understand that, you know, normal is different in men and women, how may that impact on the disparities in heart attack care, that we see between men and women? Yeah. So the first is totally correct. And we know already that there are biological differences between genders. And we know that the heart and the function between men and women differ. And then we expect differences in the normal distribution. And of course, only a high-sensitivity assay can meet this. So if you do not have a high-sensitivity assay, you cannot disclose differences in genders. So how does this translate into clinical practice? I would say that this is very different and depends strongly on the assay you use. If you have a very high difference between the genders, then the numbers.... the chance to have a reclassification, based on baseline values, but also delta changes is much higher than if you have a smaller difference. 

Dr. Mamas: You talk about the fact that this is a multi-center study. This is global. When conceiving this study, why was it so important to make this into a global study rather than focus on 1 or 2 centers, but actually more patients from those centers? 

Dr. Müller: Very good point. So, early on, of course, we started, with a small number of centers, but then, we actually, realized that for the research question that we tried to address, we must have a broader, let's say, distribution of sites, so, that because we want to address to the research question: how well do these diagnostic tests work in the early diagnosis? and want to make sure that, if something works well at your site, it also works out in Basel, in Spain, in Italy or in any other country. And of course, it's a network of sites that have very strong, local investigators that have been able to contribute to the study. And how are we using this data to inform the European Society of Cardiology guidelines of 0/1-hour or 0/2-hour, to our protocols? So, the data that APACE can provide is, of course, only one step in this whole series of evaluation that we need, but, kind of having this high-quality reference standard is really the best methodology to assess both diagnostic accuracy, but also to define use optimized cutoffs. Because this is something that has been developed in the last decade, so, the manufacturers, they determined an upper reference limit for this task a healthy reference population is used. However, these upper reference limits are not necessarily the optimal decision values for what we have to do in the emergency department. We have a patient with an acute chest pain and we want to know as rapidly as possible: “Does he or she have acute myocardial infarction or possibly it's a completely benign cause?” And for this specific purpose, the APACE data was perfect to derive this use optimized cutoff.

Guest Speakers:

Prof. Evangelos Giannitsis — Cardiologist working at the University Hospital of Heidelberg, Germany

Prof. Frank Peacock — Emergency physician, Professor of emergency medicine, Baylor College of Medicine, USA

Prof. Christian Müller — Professor of cardiology and director of the Cardiovascular Research Institute in Basel, Switzerland

Interviewer: Prof. Mamas Mamas — Professor of cardiology at Keele University in the United Kingdom

Dr. Mamas: How do you think that the Gen 6 built upon the success of the Gen 5 assay to provide this good performance at the really low levels that you need for these new 0/1 and 0/2-hour algorithms? 

Dr. Müller: It’s a very good question. The Gen 6 studies have been running for a while, but we are only now seeing the first data. Perhaps to start briefly, Gen 6 has been developed to further optimize performance of the previous generation assay, to increase the analytical sensitivity in the lower range, as you have mentioned, but also to improve further the issue of interference. That's something where our laboratory colleagues are very focused on and we as clinicians, to be honest, are quite ignorant. But, to simplify things, there are certain settings in which the assay results may fool us if there are other components in blood like hemolysis or autoantibodies, marker troponin and all this stuff. And so what our clinical chemistry colleagues explain to us is so that this new assay now is very strong to avoid any form of interference and has some improved analytical sensitivity in the lower range. Now, the clinical studies like APACE or PERFORM-TSIX have the task to tease out: Does this translate in identical or possibly even better performance than the current generation? 

Dr. Mamas: In all of cardiology, you have great science like APACE, but then it's a question of implementation. How are we going to successfully implement the work that you and the team have done into ED departments up and down the country and through the world? 

Dr. Müller: We have some data from early adopting sites that the implementation works well and that patient care delivered is extremely safe. And so, the most commonly used effectiveness measures for all of these triages is the length of stay in the emergency department and the lengths of stay could be substantially reduced, again, in these implementation studies in early adopters. But, one of the really important, remaining question is, most of the world is late adopting, because there's not the same power and enthusiasm in this interdisciplinary field as in many other aspects in cardiology, and therefore, we have built a large network, “PRECISE-MI” to provide these missing data. 

Dr. Mamas: Was is the additive value of the Gen 6 assay? 

Dr. Giannitsis: I think this is a very difficult question, but I think the “whole package” is very important because Roche responded to all critiques that they received from customers. They addressed the 99th percentile issue. They addressed the high-sensitivity designation. They addressed the lot-to-lot differences. They addressed the needs to have a very low limit of detection or limit of blank for effective rule-out. And also, of course, the reference, the “missed” reference is also now incorporated to the new assay. So this assay fulfills again, for the first time, all requirements that are asked for a high-sensitivity assay. So this is a comprehensive new assay. If this is really a revolution compared to the fifth generation is questionable, but it is a major step forward. 

Dr. Mamas: And from the TSIX data, how has that informed through the ED department? 

Dr. Peacock: So the potential is to send more people home faster. You know, any old troponin, contemporary troponin will eventually rule-in an MI. But it takes 3 or 6, or when I started it was nine hours. It is absurd. And with the new troponin, it's one and done. Alive and well in 30 days. And this is something that was not even possible the prior generation.

Guest Speakers:

Prof. Evangelos Giannitsis — Cardiologist working at the University Hospital of Heidelberg, Germany

Prof. Frank Peacock — Emergency physician, Professor of emergency medicine, Baylor College of Medicine, USA

Prof. Christian Müller — Professor of cardiology and director of the Cardiovascular Research Institute in Basel, Switzerland

Interviewer : Prof. Mamas Mamas — Professor of cardiology at Keele University in the United Kingdom

Dr. Mamas: Where do you think the challenges are in implementing this new data into clinical practice? 

Dr. Peacock: The biggest problem is the existing inertia. People like what they have. I get a lot of pushback from cardiologists all the time saying this test is too sensitive and that they'll be consulted for everything. And the answer is: “Yes, you are a cardiologist. Sorry.” Because, you know, if troponin is elevated, part of the heart muscle has died. And if part of your heart dies, you should probably still see a cardiologist. And maybe it's not due to an MI (myocardial infarction), and frequently it isn’t. But if it's heart failure, your heart failure needs to be treated better. It was renal failure and heart is dying from that renal failure. Yeah, you need better renal failure care. But at the end of the day, your heart is still dying. So, cardiologist are going to be involved in this. They have to be. And from an institutional perspective, upgrading from Gen 5 to Gen 6, is that a big problem? I think is pretty easy if you already have Gen 5, you just drop in a new business. Now, my lab people aren’t  going to just say yes. And I lived through the last few of these generations and the lab people have to validate it. But we've got data from 13,000 people. The validation is maybe 100 patients to make sure they know how to do the lab and it correlates. It's not a big study. It doesn't take a long time. It's not expensive. So, it's pretty easy if you're already in the Roche environment. If you're not in the Roche environment, then that's a different animal. You have to swap out the big box in the basement with a new big box in the basement, and there's a lot more shenanigans institutionally to do that. But, if you have the Gen 5, it’s pretty simple. 

Dr. Mamas: In this field, when you have something that works well, like the Gen 5, many watching this interview will say: “Well, okay, in terms of implementation of this new assay, should we implement it and how easy, how is the best way to implement it into our practice?” Do you have any guidance? 

Dr. Giannitsis: I think we were the second hospital in the world, after a very small hospital in Austria, that started with the Gen 5 of troponin, and we had no transition time. We started right away, have no transition between the fourth generation of troponin and the fifth generation, and we did fairly well. We didn't struggle. We already had anticipated possible issues with interpretation of more sensitive assays. We were well prepared. To start the next assay generation you need to have some communication with customers. You need some education. You need some familiarization with new values of the assay. And I think that could be very good accompanied by a clinical decision tool that helps to overcome all staffing issues. And of course, not start with an average hospital, but with an experienced hospital that has already implemented the previous generations or has a large patient turnover. 

Dr. Mamas: My last question, what are the remaining challenges in your mind that we need to address? 

Dr. Müller: I think implementation, definitely. It's a number one and, educational sessions like this one, and many other efforts. I think contribute a lot. If I may, just to select one of the remaining issues to me is the Observe Zone. These algorithms are fantastic to identify a large proportion of patients who do not have MI and to also have an extremely low risk of dying or having type one MI in the next the weeks or months. So, I think we might not get so much better because for this task the algorithms are very good. Nearly 60% of all comers, get into this triage pathway. Regarding rule-in, I also think, we are pretty good, with a very high specificity, good positive predictive value. We still can do better, but no game changer. For me, where we should focus is the Observe Zone, with all of these algorithms, about 20 to 25% of patients remain in the Observe Zone. And here, work needs to focus on how can we get better at standardize the workup. Currently, the ESC recommendation is that these patients need very detailed clinical assessment in general, need a troponin measurement at three hours and usually do need some form of non-invasive imaging. But, because this is a highly heterogeneous population there are not specific guidelines and fortunately there's a very good research ongoing, hopefully to report at the next ESC from the UK, about the value of CT angiography in this field. And our work within APACE is to tease out on the biomarker level. So what in addition to the 0/1-hour algorithm can biomarkers help to have additional patients come to the rule-out zone and additional ones that rule-in, for example, with a three hour delta of high-sensitivity troponin, in addition to what the 0/1 already has achieved.