Imaging the ruled-out chest pain does not change what happens next

Design

multicentre, randomised, controlled trial (TARGET-CTCA)

Population

3170 adults at 14 UK emergency departments with suspected acute coronary syndrome, myocardial infarction ruled out, maximum high-sensitivity troponin above 5 ng/L

Primary outcome

composite of myocardial infarction or death from a cardiac cause

Effect

7.1% (112/1587) with CT coronary angiography vs 7.3% (116/1583) with standard care, adjusted HR 0.95 (95% CI 0.73 to 1.23), P=0.71, over median 3.0 years

TARGET-CTCA enrolled 3170 patients at 14 UK emergency departments who had presented with suspected acute coronary syndrome, had myocardial infarction ruled out, and still had a maximum high-sensitivity troponin above 5 ng/L - the group everyone recognises as not quite reassured. They were randomised to outpatient CT coronary angiography-guided care or standard care. Median age 61 years, 30.2% women.

The intervention was delivered: 92.1% in the imaging arm had the scan within 90 days against 2.2% in standard care, and only 0.4% had a scan-related adverse event. After a median 3.0 years, myocardial infarction or death from a cardiac cause had occurred in 112 patients (7.1%) with imaging-guided care and 116 (7.3%) with standard care - adjusted hazard ratio 0.95 (95% CI 0.73 to 1.23, P=0.71).

The result is clean and the separation between arms was near-complete, so this is not a trial that failed to test its question. Finding coronary disease on a scan and starting a statin does not, over three years, prevent the events. The more useful number is the 7% event rate in both arms: a troponin above 5 ng/L with infarction excluded marks a genuinely raised-risk patient. The response to that should be the ordinary work of secondary prevention and risk factor control, done properly, rather than an imaging pathway that leaves the risk where it was.

• Stop treating outpatient CT coronary angiography as the default next step after a ruled-out troponin.
• Record the maximum troponin value, not just 'MI excluded' - above 5 ng/L identifies the group at about 7% three-year risk.
• Use the discharge conversation for blood pressure, lipids, smoking and diabetes, which is where the modifiable risk sits.
• Reserve CT coronary angiography for patients with ongoing symptoms suggesting stable angina, which is a different question.
• Where CT capacity is scarce, this trial is a reason to protect it for the indications that do change management.

The statistics, in plain English

This is a null result with enough events behind it to mean something: 228 primary outcomes, and a confidence interval of 0.73 to 1.23 that excludes anything like a large benefit. It does not prove exact equivalence - a 27% relative reduction still sits inside the interval - but the trial ran until the prespecified event count was passed, which is the design's way of guaranteeing it could detect a difference of the size it was looking for. Separation between arms was 92.1% versus 2.2%, so a null cannot be explained by the control group getting the intervention anyway.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.