As many as four out of five people without known heart disease will have detectable levels of cardiac troponins with the new high-sensitivity tests, according to a new meta-analysis. Even very low levels within the normal range indicated increased cardiovascular risk, but there is broad agreement that the practical role for these tests outside the emergency department remains extremely unclear.
The new high-sensitivity troponin tests are vastly more sensitive than previous troponin assays. , there is a vigorous debate taking place over the best way to use the new tests in people with suspected acute coronary syndromes. On the other hand, there has been no general discussion about potential roles for the new tests in the large number of people without known cardiovascular disease.
A Useful Adjunct?
In a paper published in the Journal of the American College of Cardiology, Peter Willeit (Innsbruck, Austria) and colleagues analyzed data from more than 150,000 people without known cardiovascular disease who were participants in 28 previous studies. They found that 80% of their subjects had detectable levels of troponins, though most of these were within the normal range. People with troponin levels in the highest third had a 43% increased risk for cardiovascular disease (CVD) compared to people in the bottom third.
"In the general population, high cardiac troponin concentration within the normal range is associated with increased CVD risk," the authors concluded. "This association is independent of conventional risk factors, strongest for fatal CVD, and applies to both [coronary heart disease] and stroke."
The authors said that their results suggest "that cardiac troponin assessment could be a useful adjunct to conventional risk factors in predicting CVD risk."
In an accompanying editorial, Jennifer Ho (Massachusetts General Hospital) wrote that the study raises several important questions. "How do we define a new normal in the era when most asymptomatic adults free of overt CVD have detectable values? More importantly, what are the clinical implications of a newly 'abnormal value'?"
"In my view, this paper highlights the true advantage of the high sensitivity assays, in that they can characterize a disease process -- chronic myocardial injury -- that is not readily detectable with standard assays in use now in the U.S.," said James de Lemos (UT Southwestern). "The opportunities highlighted here, which include robust prediction of first cardiovascular events, represent a re-thinking of the role of troponin testing. Whereas we have considered this as a test to diagnose MI in the emergency room and hospital, these new outpatient indications are very promising. The big next step is to determine how to integrate outpatient troponin testing with other risk prediction strategies, including existing risk models, as well as NT-proBNP, calcium scanning, and maybe hs-CRP [high-sensitivity C-reactive protein]. Also, we need to learn how to modify the risk identified here."
Clinical Concerns
But cardiologists are concerned that expanded usage of the test in this population might lead to all sorts of unintended consequences, perhaps leading to overtreatment and unnecessary procedures. de Lemos said he "definitely has concerns about additional downstream testing/costs. I don't think at all that this is ready for widespread testing. We need to work out how to interpret tests, what to do with results."
Sanjay Kaul (Cedars-Sinai) said that "identifying patients at risk without modifying their risk is simply an academic exercise with little to no practical significance. The analytical performance of the 'next-generation' troponin assay will only be as good as the individual clinician's thought process."
Ethan Weiss (UCSF) said:
"I am not a buyer. I do think the increased sensitivity and the fact that 80% of ALL people have detectable troponin in the serum means we are going to be dealing with how to interpret positive troponin in the ER or the office a lot. And while this study did not look at it, I am sure there will be effects of exercise, infection, etc. We have no idea how to begin to understand this and it has the potential to be incredibly disruptive. I guess I would say that I am not entirely sure I understand the 'unmet' need for a new troponin assay other than earlier diagnosis and, theoretically, less time in the ED. But in terms of utilization, one might be able to build a model where the amount of time, money and energy we spend goes UP relative to status quo but, perhaps, with no improvement in actual outcomes ... "
John Brush (Eastern Virginia Medical School) offered a detailed response to the paper:
"This is the kind of study that tends to drive a clinician like me crazy. This study seems to hint that I should be doing something (based on random troponin testing in presumably healthy people) but it doesn't tell me what to do. It seems to say 'don't just sit there - do something.' But what am I supposed to do?
"The relative risk of CV events is higher in patients at the high end of the troponin distribution than in patients at the low end of the troponin distribution. But the absolute baseline risk is pretty low (not 0, but about 5% suggesting that the population isn't really a random sample of normal healthy people). This risk gradient would probably be seen with many biomarkers, such as temperature, age, weight, hemoglobin, creatinine, coronary calcium score. The authors point out that other investigators have shown similar findings using BNP. A statistical correlation between some factor and risk doesn't tell me what to do about an individual patient, particularly in a population where the absolute baseline risk is so low.
"The study is more about screening than troponin. Screening is hypothesis-free testing. The farther you get from a starting hypothesis, the more you are likely to find an association without causation. It starts to seem more like reading the tea-leaves or palm-reading. Screening with troponin doesn't lead to a mechanistic understanding of a patient that could lead to a decision. The editorialist did a good job of pointing this out.
"Testing should be used to help us think about our patients. In the old days, when I was a student, I used to have teachers who asked me, 'How will that test change what you do'? According to the pragmatic maxim, we think to do. How does this study change our thinking and what does this study tell us to do? Maybe further epidemiology studies can determine independent factors, including troponin, that can be used for a regression formula that can drive a risk calculator, but this study doesn't provide this. It is only of epidemiological interest, and it unfortunately doesn't do much for the clinician, other than lead to muddled thinking. I'm afraid that this study will lead people to say that random troponin testing has prognostic significance, without thinking about the lack of practical significance.
"The troponin assay is an amazing test. It is a technological marvel that is very precise and analytically sensitive. But it seems to be the latest shiny object that has everyone's attention, as if it will be the magic key to unlocking the understanding of cardiac patients. The test needs to be used properly. Ideally, clinicians should evaluate a patient to determine what tests should be preformed to confirm or reject a hypothesis (a plausible diagnostic or prognostic possibility). Sure, we should screen for some things (taking blood pressures), but troponin testing is really non-specific as a screening tool and should be used in a more targeted fashion. I hope people don't misinterpret this study and think that we should be drawing troponin on everyone now."