- Design
- systematic review and individual patient data meta-analysis of 6 prospective studies
- Population
- 5,239 patients with suspected NSTE-ACS assessed by emergency medical services
- Primary outcome
- sensitivity and negative predictive value of low-risk classification at 30 days
- Effect
- mortality sensitivity 93.2% (83.5-98.1), NPV 99.8% (99.5-99.9); MACE sensitivity 92.8% (88.7-95.5), NPV 97.2% (92.1-99.0)
Six prospective studies of prehospital risk stratification in suspected non-ST-elevation acute coronary syndrome were pooled at individual patient level, 5,239 patients, all using a clinical risk score derived from HEART with troponin measured at the point of care by the ambulance crew.
Among patients classified low risk, sensitivity for 30-day all-cause mortality was 93.2% (83.5 to 98.1) with negative predictive value 99.8% (99.5 to 99.9); for death or myocardial infarction, 91.8% and 97.3%; for major adverse cardiac events, 92.8% (88.7 to 95.5) and 97.2% (92.1 to 99.0). Lowering the threshold improved both figures while shrinking the proportion classified low risk.
For a department on the receiving end this matters in two directions. It supports a pathway in which some low-risk chest pain never arrives, which is the point. It also sets the expectation for what walks through the door: if the ambulance service adopts this, the chest pain that is conveyed is enriched for risk, and the department's own rule-out thresholds and staffing should reflect that rather than continuing to assume the previous case mix. The authors' caveat is that these figures come from services with organised follow-up for non-conveyed patients; where that does not exist, the score's performance and the pathway's safety are different things.
- Ask what the ambulance service does with a low-risk patient before agreeing to a prehospital rule-out pathway
- Expect the conveyed chest pain population to become higher-risk if such a pathway starts
- Check the point-of-care assay against your laboratory troponin - the numbers are not interchangeable
- Audit 30-day outcomes for non-conveyed patients locally rather than relying on the pooled estimate
- Keep the ECG central: the score depends on one being acquired and read properly in the field
Why it matters
If the rule-out moves to the ambulance, the department's own case mix changes - which is a staffing and threshold question, not only an ambulance service one.
Don't overread it
Six selected, well-resourced services: this performance does not transfer to a system without organised follow-up.
The statistics, in plain English
Negative predictive value looks reassuring here mainly because death within 30 days is uncommon in this group; with a low event rate, almost any rule scores well on NPV. Sensitivity is the harder test, and at 93.2% with a lower bound of 83.5%, the pooled data are consistent with the score missing up to roughly one in six deaths. That is the figure to quote when the pathway is being discussed, not the 99.8%.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for emergency & critical care, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free