Podcast – Understanding Troponin – Part 1

Welcome to the St. Emlyn’s podcast. I’m Iain Beardsell and I’m Rick Body.

Now many of you will recognize Rick from his excellent blog post he’s been writing for many years on us in Emlyn’s site and we’ve managed to persuade him to come along to the podcast on what I hope with the first of many recordings. To ease him in gently we thought we’d start with a subject with which he is undoubtedly a world expert and internationally renowned. So, we’re going to spend a couple of episodes talking about everybody’s favourite cardiological topic, Troponins.

So, Rick, I hope you don’t mind if we go back to the very beginning of Troponin and talk about some of the basics and we’re going to take us all the way through to the most up-to-date thinking to do with high-sensitive troponins and how we can use them in ED. So, I’ll warm me up gently. Let’s start off with something pretty straightforward. Just remind us what is Troponin?

Well, Troponins are protein, first of all, that might come as a surprise to some people who think it’s a cardiac enzyme because we often talk about measuring cardiac enzymes but actually Troponins are protein and it’s involved in muscle contraction. If you remember from your anatomy and physiology days it’s basically part of the contractile apparatus within the muscle cells and Troponin occurs as part of a complex together with Troponins in. There are three types of Troponin – the Troponin eye, Troponin teeth and Troponin teeth. Together with Troponins in their wrapped around the “thin” filaments or “actin” in muscle. When your muscles are at rest, the Troponin Troponins in complex blocks the binding sites on the “thin” filaments so that “Myocin” – that’s the thick filament – can’t bind to it and there’s basically no movement. Then along comes an action potential and it stimulates some calcium release. The calcium binds to Troponin C which makes Troponin eye change shape and exposes the binding site on acting. “Myocin” binds to that, changes shape and it propels the thick filament along the thin filaments meaning you get a muscle contraction. So just to confirm, Troponins present in all muscle, it’s not just heart muscle.

That’s right, it’s present in both skeletal muscle and cardiac muscle. So there must be a difference in the type of Troponin as they’re if we’re able to pick up a difference that’s cardiac specific. That’s a really clever thing about Troponin is that they identify specific cardiac isopomes of Troponin “T” and Troponin “I” that they can target for detection and run a blood test to see if there’s any “Myocardial” damage. And that’s what we pick up when we talk about Troponin as a blood test, we’re looking at cardiac Troponin “I” and cardiac Troponin “T” which is very cardiac specific. So when are these Troponin proteins released into the blood?

Well, the release when there is “Myocardial injury” and it’s really important to recognize that it’s myocardial injury and not specifically myocardial infarction because any cause of myocardial injury will lead to Troponin “I” into the blood. So that could be, for example, a patient who has an acute myocardial infarction or a patient who has a particular stress on their system that causes them to damage the myocardium because of the demands that are placed on the body at the time.

So any injury at all to the myocardium, even if that’s a direct blow to the chest or pericarditis perhaps, will that cause Troponin release?

Absolutely, any myocardial injury whatsoever, regardless of the etiology. And I’m sure that that’s going to come back to us later in the episode where we think a bit more about how we pick it up and what the clinical significance is.

Absolutely, and I think this is a key point because we’ll undoubtedly start touching on the specificity of Troponins wouldn’t we? And when we talk about the specificity, we’ll be talking about the specificity for the diagnosis of an acute myocardial infarction. We’re not necessarily talking about the cardio specificity of the protein as a marker of myocardial injury. We know that cardiac troponin is very cardiac specific. That’s how it’s different to the previous generations of markers that we used to diagnose an acute myocardial infarction, like CK, LDH and ALT, if you’re old enough to remember those three enzymes, they were enzymes that we used to diagnose myocardial infarction with. We used to take three days for those substances to rise in the blood and based on the pattern of release, we’d decide whether a patient had an acute myocardial infarction. But of course, they weren’t cardiac specific at all. LDH and ALT, are liver function tests and CK, we now use more as a marker of rhabdomyolysis than acute myocardial infarction. So they weren’t cardiac specific markers. The beauty of Troponin is that it’s very cardiac specific and is very little cross-talk with skeletal muscle. When we think specifically about those cardiac parts of Troponin as opposed to skeletal Troponin, I think that’s one of the things I’ve learnt is that Troponin actually comes from all muscle. Of course, physiology lectures are a long time ago for me and even that bit you have just then, I was getting the hot sweats and a bit of PTSD about sitting in a medical school, the lecture theatre listening to those things about Troponin and “myocin” and “actin”. But it is an all muscle, but the cardiac Troponin that’s the thing that we’re measuring. So we find that in the blood when there’s any myocardial damage, these cardiac Troponin is not just when there’s myocardial infarction. When does that peak in the blood occur after the point of injury?

So the best days we have on the necessary in the context of an acute myocardial infarction and we know that it takes 12 to 24 hours for the levels to peak in the blood. And conversely after that, how long does it take for the Troponin levels to fall? I’m assuming that it has to get metabolised somehow, does it go through the renal system and get excreted that way?

Yeah, it’s excreted renally and the length of time that it takes you to clear the Troponin really depends on how good your renal function is, whether there’s any ongoing release of Troponin into the blood and how much damage there was in the first place. So if you’ve had a substantial myocardial infarction, for example, if you had a STEMI and you measure the Troponin it’d be really high, it might take a patient weeks to clear that Troponin from their bloodstream. Whereas if you have a very, very small NSTEMI, for example, it could take a lot less time to clear that from your blood. It could be even cleared by the next day, for example.

And I guess if we flip that round, if you have a small NSTEMI and you have renal failure, that may take longer to clear the Troponin from your blood than if you have a larger infarction, but normally functioning kidneys.

Yeah, absolutely. And you need to take all of those things into account. How big might the infarction have been, and how could you expect them to clear that level of Troponin? How good is the renal function? And is there any ongoing myocardial injury that would cause ongoing Troponin release?

This brings us really nicely onto one of those, I don’t know if it’s a myth, maybe you can tell us this idea about patients with renal failure and Troponin not being a useful test in those patients. From what you’re saying, I assume that’s just that patients with renal failure won’t be able to clear the Troponin as effectively, but surely the test can still be useful.

Yeah, absolutely. It’s still the reference standard biomarker for diagnosing an acute MI, even in renal failure. I have a story that sticks in my mind when I witnessed a doctor telling an earth off for testing a Troponin in a patient with renal failure, which was a really embarrassing conversation to witness because he was totally wrong. We should be using Troponin in patients with renal failure, but we should be understanding that those patients are likely to have a high level at baseline, and when we know that, we can understand how to interpret that level, and we’ll be more interested in the rise and or fall of Troponin when we do serial testing than in an absolute level.

So these patients who have renal function that isn’t as perfect as we’d like it, they may just have a baseline level of Troponin that’s permanently not being able to be cleared from their blood? Doesn’t always represent “New Myocardial injury” it must, I suppose, be Myocardial injury at some point, but they’re taking longer to clear it, but the test itself is still worthwhile.

That’s right, it’s still very worthwhile, so at baseline these patients can have a very high Troponin can be surprisingly high Troponin levels, actually, at baseline in apparently healthy people with chronic kidney disease. Of course, that’s because they’re not clearing Troponin, but it’s abnormal for them to get that level of Troponin in the first place, and it has a prognostic significance for the patients, they have a worse long-term outcome. Whether it’s an acute Myocardial infection or not, well, we don’t know at baseline, just based on a single level. If the level appears to be what we might expect for somebody with a renal failure in that degree, then we’ll need serial sampling before we can differentiate those with acute MI from those who just have a chronic elevation of Troponin secondary to renal failure.

Now, when we talk about a lot of diseases, I always find it easier to go back and think about the basic physiology if possible to then relate it to the illness itself. I wondered if we could do that the same, especially for the geeks out there, and I’m a bit geeky. Could you just tell us how one of these analysers works? Because perhaps maybe if we understand how the analyser works, that’ll help us understand a little bit where the tests come from and how we use the test.

Yeah, so there’s some really funky technology involved in doing a Troponin test, and as emergency physicians, I guess we put the tubes in the shoes off, they go to the lab, they work some magic, and we get a result. But actually, it’s quite nice to understand how they go about getting that result. So, Troponin tests are immuno assays. What that means is the clever people have identified parts of the Troponin molecule that they can target with an antibody. Secure the antibody onto that part of the Troponin molecule, and then attach some kind of signal to that antibody, maybe to emit some light. That’s for example how we use the Troponin essay. We attach the antibody, we attach the “Rothinium” to the antibody, and then we pass electricity through the complex, and that causes any Troponin or Troponin and “Rothinium” complex to light up. We can measure how much light is emitted to quantify the amount of Troponin that’s in that sample.

That sounds like quite a big piece of kit, yet some of the Troponin machines we have are sort of point-of-care testing. Do you lose some of the accuracy of your testing if you reduce down the complexity of your measuring equipment?

Yes, totally. They all work on the same principle, they’re all immunoscipes. But if you look at the lab-based technology, it’s huge, it fills the whole lab, you’ve got these massive analyzes. You get your sample, you plug it in at the beginning, you tell a computer what you want to do, it senses a few, just your sample, takes off the serum, whizzes it round on the robotics? All these robotic arms spin around and add antibodies, and this is reaction, and then on the large computer at the end you get your results. Trying to replicate that with a very small piece of kit that you can have at the bedside is, of course, going to be really, really challenging. And because of that, the technology, the point-of-care testing isn’t quite as good as the lab technology, and you don’t get quite a sensitive or precise results.

Are there differences in the different analyzes we use, whether you send off the bottle to the lab as you describe or whether you use a point-of-care kit, there’s some even portable kits that people can use by the roadside now, aren’t they?

Yeah, well, I think some of those might be qualitative tests that simply you read like a, like you would for a home pregnancy test, is it positive or is it negative above a certain threshold? There are handheld devices that you can use like the Abitai Statsemunzultro, you can get a quantitative result using a device that you can take to the patient’s bedside, potentially you could even use a finger-prick to get that result within 10 minutes or so. Those are fantastic pieces of kit, the sensitivity is not quite as good as lab technology though, so whether we can actually use it to reassure people that they don’t have an acute myocardial infarction, we don’t yet know, and it’s unlikely that the current carchets we could rely on them as well as we can with the lab technology.

You used a really important word there where you’re talking about things and a word that we often perhaps use incorrectly with different circumstances, and that’s one that’s going to be important in our next episode, and that sensitivity, sensitivity I understand from talking to you a bit more, is that not just the diagnostic sensitivity in the idea of sensitivity, specificity, positive predictive values, but in this sense, especially when we come to talk about high sensitivity, troponin, and why are you talking about analytical sensitivity? So we’ve got to be a bit careful on how the words we use. So one of the most important things to recognize is it’s not simply a question of tropelling T versus tropelling I, it’s a question of the different assays because they all have different analytical characteristics. There’s a really important point to be made here about the difference between analytical sensitivity and diagnostic sensitivity. Analytical sensitivity is about how low are the concentrations that we can reliably detect using this troponin assay. And that’s very different from diagnostic sensitivity, which is of course how many people with the condition that we’re trying to diagnose will get a positive test. In terms of diagnostic sensitivity, a contemporary assay that isn’t high sensitivity will, on average, have a sensitivity of about 80, 85% when patients first arrive in the emergency department with a suspected acute amide, and the specificity of those contemporary assays will be somewhere in the 90s. So quite high. So doing what we used to call on, still call a stat troponin can be very helpful for ruling in myocardial infarction, and so you get that positive test at the beginning, helps you with your decision, making bit like an ECG is highly specific for myocardial infarction, but it wasn’t sensitive enough to rule out. Now that must be why we then have to wait for that blood peak to occur, to then do our rule out, more sensitive, diagnostically sensitive test further down the line.

Yeah, absolutely. So you might respond to a positive troponin using those contemporary assays, because the specificity is pretty good, kind of similar to the specificity of an ST elevation on an ECG for STEMI, for example. When we use those older assays or contemporary assays, there aren’t high sensitivity. Seeing a positive troponin was a significant event, and we’d tend to label those patients with an acute myocardial infarction, unless there was a good reason to do otherwise. But the sensitivity isn’t there, you know, you can’t rule out as a sensitivity of 85%. But we know that it’ll increase later, because the troponin levels will peak, so if we do serial sampling, later on we’ll reach a sensitivity of 100%, at which point we can safely rule out the diagnosis of an acute MI.

So we often use two troponin values. We do one when the patient first arrives, and then we do another one at a time interval further down the line. I think looking at the literature, you’ll obviously know a lot more about this than me, because you’ve read it all, well, the bits you haven’t written, you’ve read. Some people seem to use a six-hour window. Other people seem to use 10-hour, 12-hour windows. Is this just because they’ve chosen to make a compromise with how sensitive the test is? Or is it to do with when they’re judging with pain and when the patient arrives? Why are the different values?

You know, it’s really interesting to see how we arrived at these different values. In the UK, we start the clock when the patient symptoms started, and I think that’s different to the rest of the world, where they start the clock when the patient arrives in the emergency department, regardless of the time of symptom onset. And we could argue, well, they about whether patient’s story is reliable, and whether you can actually use that time of symptom onset to time your troponin test. But that’s what we’ve been doing for years. When you look at the evidence behind those things, actually, it’s relatively scant. When I looked for this, I found a paper in academic emergency medicine from 1997 by Tukka, which did a great, they did a great study. It’s serial, deposition in sampling for, I think, 24 or 36 hours in patients with suspected acute MI. And it wasn’t until about 12 hours after symptom onset that they found 100% or very close to 100% sensitivity. It’s brilliant piece of work. Actually, where the 12-hour value came from for us, was parts of the negotiation in terms of why we would implement a troponin test. It was probably a bit more expensive than the cardiac enzymes that we were using before. But by bringing the rule out to 12 hours, patients could go home a lot sooner than they would have been able to go home using the traditional cardiac enzyme panel of CK, LDH and ALT, for example. So it’s a little bit more arbitrary than you might think. And we’re using that value related to pain onset, but in other countries, they may use a six-hour value, but this is assuming that the patient’s presenting a few hours after the pain started.

Absolutely, and this is a really key point, because when you talk about doing a six-hour troponin, if you did it six hours after symptom onset, that’s very different from doing it six hours after arrival. Because patients will tend to arrive on average about four hours after the onset of the pain, in which case a six-hour troponin is actually 10 hours after symptom onset. If you do a six-hour troponin after symptom onset, you might lack the sensitivity and actually be missing some, some of the patients who’d go on to develop a late rise.

And that really explains for me why we have some of these different values. I worked in a place before we moved to what we’ll talk about in our next episode, the high sensitivity troponin, where we would do a 10-hour troponin. And I was thinking, “But how come we’re doing 10 hours and other people are doing six hours?” When in actual fact, we were all doing 10 hours after symptom onset, bearing in mind it was four hours after the onset of pain that the patient arrived. It was just when they decided to start the clock. So actually, everybody was doing the same thing. The key was not to then go pack for the means of expediting patient journeys to six hours after symptom onset. In my head, I’d always wondered why we hadn’t done that, but you’ve really cleared that up.

So Rick, that’s really a beginner’s guide, I think, to the troponin, the molecule where it comes from how we measure it, why it’s raised in renal failure. Some of those other bits and pieces that people might have been asking themselves as they’re doing these tests. Just the last thought, I guess, is that we seem to be doing troponin on an awful lot of people. I’m assuming that we can’t just ignore the pre-test probability. We have to still take into account the clinical picture.

Absolutely, and that’s a really key point, isn’t it? When we’re under pressure in emergency department to meet the processed targets, tests are often requested on arrival perhaps by a nurse or support worker who hasn’t taken a thorough history, or maybe even hasn’t got the expertise to take a thorough history. Some of those patients who have troponin tests may have a very, very low pre-test probability of disease, so low that we’d never even have considered the diagnosis to execute myocardial infarction. It’s really important to be a Bayesian thinker about this. So if you see a positive troponin and a patient with a very, very low pre-test probability of disease, the post-test probability will also be low. For example, we take just the patient who has suspected cardiac chest pain. So these are people who have an appropriate troponin request, and we look at the positive predictive value of our positive troponin using, we’re going to talk for now about a high sensitivity assay, but let’s just for a moment, I know we’re going to talk about that in more detail later on, a positive high sensitivity troponin. The post-test probability of an acute myocardial infarction is only 50%, and that’s in patients who have suspected cardiac chest pain. The post-test probability is going to be far, far lower than that in patients who have pneumonia, chest infection, transient loss of consciousness. So it’s really important to bear that in mind. We’ll really dilute the value of this test if we use it indiscriminately.

And taking the other way about sensitivity, if you have a higher pre-test probability, even with a sensitivity that’s approaching 100% at 10, 12 hours, a negative test by itself at 10 hours, 12 hours, even with a high sensitivity, doesn’t rule out disease. So we can’t just do troponins on everybody, hoping that that clears us and tells us that they don’t have illness. We still need to be clinicians, we need to be thinkers, we need to form that pre-test probability.

Absolutely, and that’s another really important point for several reasons. One is that the patient may develop a late troponin rise, and if they’ve got a high pre-test probability for ischemic ECG changes, high risk history, then we’re going to be a bit more worried about it. Similarly, it’s really important, I think that we touch on the diagnosis of unstable angina, because that’s a troponin negative state. These patients will never develop a rise and fall of troponin because they don’t have an acute myocardial infarction, but they still have ACS. Troponin can’t rule that out, you’ve got to think, you’ve got to use your clinical judgment, and the other clinical information is presented to you in order to establish that diagnosis.

That’s really, we’ll just reiterate that patients with unstable angina, they’re not getting myocardial injury, myocardial infarction, so troponin isn’t released into the blood, but it’s still representing a disease state that we need to do something about.

Absolutely, and those patients could well go on to develop a major adverse cardiac event in the near future, unless we give them appropriate treatment.

Rick, it’s been great getting the basis of the troponin test from you in this episode. I know we’re going to go on and talk more in our next episode about your perhaps greatest love, apart from your family, of high sensitivity troponin, and we’re going to delve even deeper into that. But for now, we hope you’ve all got something from listening to this podcast. Please, obviously we have Rick available 24 hours a day, seven days a week to answer any questions you might have about troponin. Feel free to write comments on the blog post, email us, Rick will be only too happy to answer them. We hope this has been useful, we can’t wait to speak to you again very soon, where we’ll be exploring an even more depth, the troponin molecule, and high sensitivity troponin. So from both of us for now, take care.