Ruling out ACS at the desk: Marburg Heart Score plus point-of-care troponin

Design

Cluster-randomised diagnostic trial (2:1), Dutch primary care

Population

827 patients with acute chest pain (740 intervention)

Primary outcome

Diagnostic accuracy for ACS; referral rate vs usual care

Effect

Sensitivity 98.3% (90.8–100), NPV 99.7%; referrals 54.6% vs 47.1% (p=0.39)

POB-HELP, a cluster-randomised diagnostic trial in the British Journal of General Practice (10 September), assigned Dutch practices 2:1 to use a decision rule or usual care for patients with acute chest pain. The rule combined the Marburg Heart Score with a point-of-care high-sensitivity troponin I test, applied by the GP. In total 827 patients took part: 740 with the rule and 87 controls.

The rule missed one acute coronary syndrome, giving sensitivity of 98.3% (95% CI 90.8–100%) and negative predictive value of 99.7% (98.0–100%). It missed no myocardial infarction (sensitivity and NPV 100%). Specificity was about 49%, so roughly half of patients without ACS were still flagged. Referral rates did not differ significantly from usual care (54.6% vs 47.1%).

So the rule is safe for ruling out, but in this trial it did not reduce referrals. Its value is confidence: a GP who applies it can keep a low-risk patient in primary care knowing the miss rate is very low. The size of the control group limits the referral comparison.

• Apply the Marburg Heart Score to every patient with acute chest pain
• If a point-of-care hs-troponin device is available, combine it with the score as in the trial
• Refer urgently if either the score or the troponin is not low
• Keep ECG and clinical judgement — the rule is for ruling out, not replacing assessment
• Point-of-care hs-troponin is not widely available in Indian primary care; the score alone is a starting point

Why it matters

It gives GPs a tested way to keep low-risk chest pain in the surgery rather than referring by default.

Don't overread it

The rule was safe but did not reduce referrals in this trial, and the control group was small.

The statistics, in plain English

Negative predictive value of 99.7% means that among patients the rule called low risk, about 3 in 1,000 had ACS. The lower limit of sensitivity (90.8%) is wide because there were only about 60 ACS cases. Low specificity means many patients are still flagged, which is why referrals did not fall.

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.