BEST-2
Research type
Research Study
Full title
The Bedside Evaluation of Sensitive Troponin 2 Study
IRAS ID
303373
Contact name
Richard Body
Contact email
Sponsor organisation
Manchester University NHS Foundation Trust
ISRCTN Number
ISRCTN48377837
Duration of Study in the UK
1 years, 6 months, 1 days
Research summary
When patients are admitted to hospital with suspected heart attacks they must often have blood tests over a number of hours before the diagnosis of a heart attack (acute myocardial infarction) can be excluded. The tests measure cardiac troponin, a marker of heart injury. Over the last decade, high-sensitivity cardiac troponin assays have been widely implemented and, by precisely detecting small concentrations of cardiac troponin, allow much faster diagnosis of heart attacks than was previously possible.
Point of care high-sensitivity cardiac troponin assays, which sensitively measure cardiac troponin levels within minutes at the patient's side, are now available. In this study, we will test the accuracy of new point of care high-sensitivity cardiac troponin assays for the diagnosis of acute myocardial infarction in the Emergency Department (ED).
In this study, we will ask consenting patients presenting to the ED with chest pain or similar symptoms if they would donate blood samples for analysis in this research. We will draw blood on arrival and 1 hour later. The blood will be tested for levels of cardiac troponin using the new point of care assays. We will also collect data about patient's symptoms and past history to see if using established risk scores and decision aids alongside the new troponin assays will improve diagnosis. Patients will undergo standard clinical assessment as part of the routine care and we will check if they have had any further cardiac events over the next 30 days by checking medical records +/- contacting the patient or their general practitioner, and an expert panel will determine whether the final diagnosis was acute myocardial infarction.
The study will enable us to understand whether the new tests can be used in practice, enabling faster diagnosis and more efficient use of NHS resources.
Summary of results
: Thank you We would like to thank everyone who took part in BEST-2. We are also grateful to the clinical and research teams at the participating NHS hospitals.
Their contribution has helped us learn more about how patients with possible heart attacks might be assessed more quickly in emergency departments.
Why was the research needed?
Many people attend emergency departments with chest pain or related symptoms. Doctors must consider whether these symptoms could have been caused by a heart attack.
Most of these patients have not had a heart attack. However, it is important to identify the smaller number who have.
Doctors usually use an electrocardiogram, known as an ECG, and blood tests to help make this decision. The blood tests measure a substance called cardiac troponin. Troponin is released into the blood when heart muscle is damaged.
Troponin tests are usually carried out in a hospital laboratory. Waiting for laboratory results can contribute to the time patients spend in the emergency department.
Newer “point-of-care” troponin tests can be performed close to the patient. They may provide results more quickly. However, their accuracy must be carefully studied before they are used to guide decisions about patient care.
What did the researchers want to find out?
The researchers studied a point-of-care high-sensitivity troponin test called the Siemens Atellica VTLi test.
They wanted to find out how accurately one blood test, taken soon after arrival, could identify patients with a low probability of heart attack.
They also examined whether accuracy could be improved by combining the troponin result with existing clinical decision aids. These decision aids use information such as symptoms, medical history, age and the ECG. The three decision aids studied were T-MACS, the modified HEART score and EDACS.
This summary describes the first report of the BEST-2 findings. It focuses on the accuracy of a single blood test taken soon after arrival. Other planned work, including evaluation of tests taken one hour apart and an economic analysis, will be reported separately.
Who took part?
Adults were invited to participate if they attended an emergency department with chest pain or related discomfort that required investigation for a possible heart attack.
People were not included if they had a type of heart attack requiring immediate specialist treatment, another condition requiring hospital admission, symptoms that had peaked more than 12 hours earlier, or an inability to give informed consent.
The study recruited patients between August 2022 and September 2023. It took place in seven emergency departments in England and Scotland. A total of 656 participants had sufficient information to be included in at least one of the main analyses.
Of these 656 participants, 53 were diagnosed with a type of heart attack called a non-ST-elevation myocardial infarction. This is often shortened to NSTEMI.
What happened during the study?
A small blood sample was taken soon after the participant arrived in the emergency department. Whenever possible, this was taken at the same time as blood required for ordinary clinical care.
The research sample was tested using the point-of-care troponin device. Participants continued to receive the usual tests and care provided by the hospital. The research test did not replace the hospital’s standard troponin testing.
The researchers also collected information about symptoms, medical history, examination findings, ECG results and routine blood tests. This information allowed them to calculate the three clinical decision aids.
At least two cardiologists independently assessed the available clinical information to decide whether each participant had experienced an NSTEMI. The cardiologists did not know the research troponin result or the results of the research decision aids when making this assessment.
Participants were followed up after 30 days. The researchers looked for serious events including death, heart attack or a procedure to restore blood flow to the heart.
What were the main results?
The single troponin test
The researchers first examined a troponin level below 4 nanograms per litre. This threshold had been identified in earlier research. Using this threshold, around one in five patients could be told that they don’t have NSTEMI without needing further tests. Using the test in this way would have had around the same accuracy as the laboratory tests that are used in practice. The chance of a missed NSTEMI would have been less than 1% in this group.
However, there remains some uncertainty around these estimates. The study included fewer heart attacks than expected, which reduced the precision of the results. The findings therefore cannot be used by themselves to bring about a change in practice. More research will be needed.
Combining the test with clinical decision aids The T-MACS decision aid and the EDACS score both placed around 4 out of 10 patients into its very-low-risk group. If T-MACS was used in practice alongside the new test, these patients would not have any further tests for NSTEMI. The test also had a high accuracy rate, though around 4% of patients with NSTEMI would not be detected. However, EDACS would have missed more NSTEMIs: around 1 in 7 NSTEMIs would not be detected by the test.
No additional serious cardiac events were observed during 30-day follow-up among the T-MACS very-low-risk participants with available follow-up information.
The modified HEART score placed around 1 in 5 patients in its low-risk group. It had similar accuracy to T-MACS.
Overall, T-MACS identified a larger proportion of participants as very low risk than the troponin threshold alone or the modified HEART score. However, the study was not large enough to make a reliable direct comparison between the accuracy of the different approaches.
Were there any medical problems caused by taking part?
BEST-2 was an observational study. It did not allocate participants to a treatment or change their routine clinical care.
The additional procedure was the collection of small blood samples. The study protocol identified temporary discomfort and occasional bruising as possible effects of blood sampling. Research blood was taken alongside routine blood tests whenever possible.
The first findings report does not identify any unexpected medical problems caused by the research procedures.
What are the limitations of the study?
The number of participants who had an NSTEMI was lower than expected. This means there is more uncertainty around the estimates of test accuracy.
The researchers could only recruit when trained research nurses were available. The participants may therefore not fully represent all patients attending emergency departments with possible heart attacks.
Some participants could not be included because they did not undergo all the standard hospital tests required for the main analysis. The research tests were also carried out by research nurses rather than by routine clinical staff.
The manufacturer has since discontinued the particular troponin test studied for business reasons. The findings cannot therefore be introduced directly using that device unless the technology is supported by another manufacturer. Nevertheless, the results provide useful evidence about the potential of rapid, high-sensitivity troponin testing close to the patient.
What do the findings mean?
The findings suggest that a rapid troponin test performed soon after arrival can identify a group of patients with a low probability of heart attack.
Combining the test with T-MACS identified more than one third of participants as very low risk. This approach could potentially reduce the need for repeat blood tests and prolonged assessment for some patients.
However, the research did not test what would happen if these approaches were used to make real discharge decisions. The findings should not be taken to mean that the test can be used on its own to decide whether an individual patient should leave hospital.
Larger studies and combined analyses of several studies are needed to provide more precise estimates. Further research should also examine whether point-of-care testing reduces time in hospital, improves patient care and represents good value for the NHS.
Patient and public involvement
There was no formal patient or public involvement specifically in the design, conduct or analysis of BEST-2.
The research team’s broader programme had previously received advice from local heart patient groups on similar studies. This earlier experience informed aspects of the general research approach, but those groups were not formally consulted about BEST-2 itself.
Who carried out and funded the research?
BEST-2 was carried out by researchers and NHS clinical teams across seven emergency departments in England and Scotland. The study was sponsored by Manchester University NHS Foundation Trust.REC name
East Midlands - Derby Research Ethics Committee
REC reference
22/EM/0005
Date of REC Opinion
16 Feb 2022
REC opinion
Further Information Favourable Opinion