The shifting sands of chest pain assessment…

This post refers to

Serial Changes in Highly Sensitive Troponin I Assay and Early Diagnosis of Myocardial Infarction by Till Keller et al. which was published in the Journal of the American Medical Association on 28th December (Vol 36, p268) http://jama.ama-assn.org/content/306/24/2684.short

You will hear a lot about chest pain on this blog. No apologies…it is my bias.

Assessment, diagnostics and management of patients who present acutely with chest pain encapsulates all that is good and bad about general cardiology. It raises all sorts of issues ranging from basic clinical practice and pathology through to health system design (and redesign) and everything else in between.

The other reason that is of particular interest at the moment is for the amount of research that is currently emerging in relation to the diagnostic processes and the organisation of protocols and systems that occur around these clinical and scientific developments. Not to mention the political imperatives around timely management of patients presenting to EDs. Remember that chest pain is the most common single reason for ED presentation.

So this recent paper from Germany by Keller and colleagues will be very important. It grows the general trend of increasingly rapid protocols for risk stratification of patients presenting to the emergency department with acute chest pain, an area that TheLastGeneralCardiologist has been interested in for some time. Most of this revolves around better understanding of the role of biomarkers, essentially cardiac troponin, and their use in risk models.

The paper concerns a large cohort studied with robust methodology by a team with a strong track record in the area (see New England Journal of Medicine 2009, Vol 361, p868 http://www.nejm.org/doi/full/10.1056/NEJMoa0900428).

What particularly catches the eye about these results is the very limited advantage offered by the so called ‘high sensitivity’ biomarker assay over the existing, or contemporary, assay. In fact, the sensitivity and negative predictive value to rule out myocardial infarction were essentially the same for both assays when used at three hours after presentation. Neither test was sufficiently sensitive to adequately rule out an MI at presentation.

I like Keller’s papers because they use a robust and pragmatic definition of unstable angina (the third component of acute coronary syndromes after ST elevation and non-ST elevation MIs) which can be a real problem in these studies.

A couple of words of caution though. Firstly, this appears to be the exact same population used in the groups previous major publication. It’s a little surprising that this doesn’t even get a mention. The text stresses that the adjudication of the clinical end points was performed blind to the results of the investigational biomarker assays described in this more recent paper – one would certainly hope so given that the adjudication must have been performed a few years ago! In fact, it would be more important to ensure that the collation of the newer biomarker results was performed independently to the clinical end point results – which must have been known ahead of time.

Finally, the biggest flaw in this article, shared by the groups previous work, is the rate of acute coronary syndrome of ~20%. Anyone involved in studying these type of populations in the ED will tell you that this is much higher than would be expected in a truly ‘unselected’ sample of patients with chest pain where you would expect a rate of 10-15% – The difficulty then arises when the described strategy is applied in a much less selected way as we know biomarker testing is out there in the real world.

1 comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: