March 13, 2019

expert reaction to study looking at effectiveness of troponin heart attack blood test

Reactions to research published in the BMJ into the effectiveness of the troponin heart attack blood test.

Prof Hugh Montgomery, Professor of Intensive Care Medicine, UCL, said:

“Troponin isn’t the ‘blood test to diagnose heart attacks’.  It is a test which contributes to the diagnosis.  It is applied when patients present with a worrying clinical history/ECG changes which raise suspicion.

“It has long been known that levels are elevated in those sick with other conditions (including intense exercise).  This study makes the point forcefully, and is helpful in reminding clinicians that hs-Troponin assay (and, indeed, all tests) should be ordered, and results interpreted, based on clinical history and examination.”

Prof Steve Goodacre, Professor of Emergency Medicine, University of Sheffield, said:

“This study provides doctors with a helpful reminder that a heart attack should not be diagnosed on the basis of a blood test alone.  Troponin is a very useful blood test because it identifies patients with a heart attack who will benefit from treatment, but doctors need to be aware that many other conditions can increase troponin levels.  We would therefore expect a hospital population to have higher troponin levels than a healthy population, even in the absence of a heart attack.”

Prof Charalambos Antoniades, Professor of Cardiovascular Medicine, Radcliffe Department of Medicine at the University of Oxford, said:

“This is a large study of 20,000 participants from a single centre, showing that the absolute cut-off for troponin test provided by a specific manufacturer, does not correspond to the “99th percentile” of troponin in that centre, which is the recommended cut-off above which the test is abnormal.  This is generally well known among the practicing cardiologists, hence why the concept of the “99th percentile” cut-off is proposed, rather than relying on the indicative cut-off provided by any manufacturer (typically determined in healthy volunteers).  Also, troponin measurement gives a continuous number, and we need to take into account also its actual value in clinical interpretation, rather than the simple binary “above/below a cut-off”, which may not be always reliable for borderline values.  Even variations of troponin levels within the “normal range” have prognostic significance for future cardiovascular events.

“In any case, troponin remains our best tool to identify the patient with a heart attack, and its value should not be questioned.

“The test is not flawed; as with any test used in clinical practice, it needs to be used wisely, and studies like this remind us how important it is to look at the patient as a whole, rather than giving a medical diagnosis based a single laboratory test.”

Prof Sir Nilesh Samani, Medical Director at the British Heart Foundation, said:

“Currently, troponin measurements are widely used in patients suspected of having a heart attack because troponin is released from the heart when it is damaged.  However, the heart can release troponin when a patient is ill from other conditions and not necessarily due to a heart attack.

“As this study emphasises, a positive test should not always be interpreted as being due to a heart attack, and other information should be taken into account – for example, a patients symptoms and the results of other tests including an ECG.”

‘True 99th centile of high sensitivity cardiac troponin for hospital patients: prospective, observational cohort study’ by Mark Mariathas et al. was published in the BMJ at 23:30 UK time on Wednesday 13 March 2019.

Declared interests

Prof Charalambos Antoniades: “My conflict of interest is indirect, as Founder and director of Caristo Diagnostics, a cardiac CT Image analysis company.”

Prof Steve Goodacre: “I have a potential CoI in that I have published research into high sensitivity troponin tests, in particular a cost-effectiveness analysis that formed the basis of NICE recommending high sensitivity troponin to diagnose heart attack.”

None others received.