High-Sensitivity Troponin Implementation: HiSTORIC in Practice

John M. Mandrola, MD; Nicholas L. Mills, MBBS


September 09, 2019

This transcript has been edited for clarity.

John M. Mandrola, MD: Hi, everyone. This is John Mandrola from|Medscape Cardiology, and I'm here in Paris at the European Society of Cardiology Congress. I'm pleased to have Nick Mills from the University of Edinburgh with me today to discuss the HiSTORIC trial. Welcome, Nick. Tell us about the top-line results of HiSTORIC.


Nicholas Mills, MBBS: HiSTORIC was a stepped-wedge cluster randomized trial funded by the British Heart Foundation in the United Kingdom to evaluate the impact of introducing a rapid rule-out pathway with high-sensitivity cardiac troponin I (hs-cTnI). We randomized seven hospitals across Scotland to implement the pathway and looked at two co-primary endpoints: effectiveness and safety. Does the use of a single high-sensitivity troponin test at separate risk-stratification and diagnostic thresholds, to identify low-risk patients and allow them to be discharged, directly improve the effectiveness of care? And, critically, was it safe for patients?

The HiSTORIC trial reduced the length of stay in the emergency department by 3 hours across just over 31,000 patients. Furthermore, we increased the proportion of patients sent home directly from the emergency department by 57%, such that overall in our trial population, three quarters of all patients attending with suspected acute coronary syndrome were triaged directly to outpatient evaluation.

Now, was it safe? The primary safety outcome was a composite of myocardial infarction or cardiac death following discharge from hospital. We evaluated that at 30 days and at 1 year. We found that the event rates were 2.7% at 1 year in the standard-care arm versus 1.8% in the early rule-out pathway, demonstrating that this acceleration of care and discharge approach was not associated with any risk to patients or adverse outcomes.

Trial Design

Mandrola: This is great news. Tell us about this pathway.

Mills: It's designed to be simple, pragmatic, and easy for clinicians to follow. We have two separate thresholds: one to risk-assess patients, and one to rule-in the diagnosis. If your troponin concentrations are below the risk-stratification threshold, then you're triaged to the outpatient pathway. If you are above the sex-specific 99th centile, we recommend admission for serial testing and peak troponin concentrations. If you're in the middle, then we recognize that small increasing troponin concentrations in that reference range might be important. Those patients with rising concentrations within the reference range get a second test and are admitted for further evaluation. As I said, three quarters get evaluated for outpatient evaluation; one quarter require hospital admission.

Mandrola: What is a cluster randomized trial? You don't really randomize patients.

Mills: That is correct. There are four key features of this trial. First of all, the enrollment screening into the trial was performed by usual-care doctors at the time that they ordered cardiac troponin. Second, we randomized hospitals rather than individual patients because we felt that it was really important to evaluate the use of this pathway in all patients without selection bias. Third, it meant that no researchers were standing over the clinician observing what they did and causing excellent adherence from the Hawthorne effect. This study was done by the doctors themselves. Finally, it was a data-enabled trial. All of the insights—both the characterization of the patient population and outcome—were collected using routinely collected electronic data across these hospitals in Scotland.

Mandrola: I suspect that this pathway came from the previous High-STEACS trial.

Mills: High-STEACS was our first stepped-wedge randomized trial. Here, we evaluated the universal definition recommendations that we use sex-specific 99th centile diagnostic thresholds.

We learned a lot from the conduct of that trial. Principally, we learned that hs-cTnI is a more valuable test for ruling out myocardial infarction and identifying low-risk patients than it is for ruling in. In High-STEACS, we could not improve survival in those patients we classified by the hs-cTnI assay. And I think we're still trying to understand how to get the optimal use out of the rule-in part of any pathway for acute chest pain.

Potential Risk for Overdiagnosis?

Mandrola: This seems to be a problem in the United States. We are less experienced with the high-sensitivity troponins. We're seeing more and more of these troponin leaks, so to speak, and we don't know what to do with them. I'm concerned about overdiagnosis and overtesting.

Mills: That was one of the key concerns we had, and that is why when we introduced the test into practice across Scotland, we did it as part of a stepped-wedge cluster randomized trial. The benefit of improved sensitivity is that you might improve the targeting of treatments in patients with coronary artery disease and improve outcome. Of course, reductions in diagnostic thresholds inevitably impact specificity; therefore, it is plausible that you could see an increase in diagnosis, inappropriate investigation, and overtreatment that could cause harm.

Although the High-STEACS trial did not improve our primary outcome of subsequent myocardial infarction or cardiac death, we did not find evidence of harm or overtreatment. No excess bleeding, missed diagnosis, or alternative death. The adoption of high-sensitivity troponin, although the change may generate anxiety, did not have any detrimental effect on patient care. So I don't believe that it is leading to overdiagnosis. I think clinicians are having to think more carefully about the interpretation of the test, but that is a good thing. I think what HiSTORIC tells us is that we need to move away from this being some sort of binary test for a heart attack. If the troponin is positive, call the cardiologist. This is a continuous marker of injury to the heart that informs on a wide range of diagnoses and ultimately is a very powerful test of cardiac outcome.

The challenge for a cardiologist and the emergency room physician is to understand what pathway that patient needs to go into. A lot of patients with heart failure and other comorbidities have myocardial injury. That is real and it's because their heart is not healthy. But they don't need the cath lab.

Only a Single Troponin Test?

Mandrola: I heard you say that 75% of patients in HiSTORIC were discharged with a single troponin and no inpatient testing. Did they get outpatient testing? Do you have those data?

Mills: Good question. Just to clarify, 75% overall were discharged. Remember, there was an intermediate-risk group that had two tests. Although the majority went home with a single test, some did have two tests in the emergency department. One other important clarification: In the High-STEACS pathway which we used in the HiSTORIC trial, we always recommended two tests in early presenters. So, for patients who come in within an hour or two of onset of symptoms, it's really important to do two tests. Not everyone has a single test.

On to your question about downstream investigation. This varies all across the world. In the United States, everyone gets a stress test, a nuclear test, and a number of other tests. Probably a CT thrown in now. In New Zealand and Australia, they treadmill everybody. But the evidence base for any of these approaches is completely lacking. We have no randomized trial with any particular investigative strategy. In fact, our next trial, TARGET-CTCA, is a randomized trial of the intermediate-risk group identified between these thresholds, to try to provide some evidence-based guidance on what investigations are indicated in this setting. We know from our article published in Circulation that the event rate at 1 year was 0.5% for those below the risk-stratification threshold (< 5 ng/L).[1]

Mandrola: Which is the same as the general population.

Mills: In terms of concentrating our efforts to investigate patients by using separate risk-stratification diagnostic thresholds, I think we will become better at targeting these very valuable investigations to the right patients. But to directly answer your question: Do I have the data as to what tests to order in the trial population? Not yet.

Mandrola: You also mentioned sex-specific thresholds? Speak on that.

Mills: This has been an area of some controversy over the past couple of years. I don't think it is controversial at all. There was a systematic review of over 30 different reference range studies done in populations across the world for every troponin assay available. And consistently, the upper limit of normal in women is substantially lower than it is in men for all tests.[2] If we are going to follow the universal definitions recommendation, that we use the upper reference limit of normal to diagnose myocardial infarction, we must take into account that there are differences by sex. I think it's fairly clear.

Mandrola: But also more complicated.

Mills: Having implemented it in 10 hospitals in Scotland, it was not complicated in the slightest. Within 2 days, people got it. We have different reference ranges for hemoglobin for men and women, and that does not cause too many challenges in the clinical practice. This is easy to incorporate into your reporting system in any electronic patient record system.

High-Sensitivity Troponin as a Biomarker

Mandrola: Another question that comes up with high-sensitivity troponin that never came up with older troponins is the use of it as a potential risk biomarker. Do you have any feelings about that?

Mills: I think that people are learning. One of the reasons why it's so powerful at identifying low-risk patients in the emergency department is because it is a barometer of heart health. It has very little to do with the diagnosis of heart attacks at that level. It's telling us how healthy your heart is and whether you have active coronary artery disease, underlying structural heart disease, or comorbidities that might influence your heart health.

It's a very powerful tool. It reflects a range of pathologies, which is our challenge. But in terms of how we might use it in the future more broadly with the emergency department, there is a range of different applications. There is a lot of interest in using it as a primary prevention or screening tool. We need more evidence that reporting that evidence to an individual would in some way impact the way they consider their risk and have some relevance to the clinician managing their care. What would they do differently as a consequence of this information? There is a lot of enthusiasm and a great opportunity here, but the research community needs to design the right trials to demonstrate the value of this test more broadly. We're really learning how to use it in the acute setting, but we've got some way to go to understand how to use it in the chronic disease setting.

Mandrola: I love that word of caution. On that note, I'll say thank you for joining us. It was great talking to you.

Mills: My pleasure.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: