Everyone uses the phrase. Fewer apply the criteria. Intermediate-high-risk pulmonary embolism requires four things at once, and it is the conjunction that matters: the patient is haemodynamically stable; the simplified Pulmonary Embolism Severity Index score is at least 1; there is right ventricular dysfunction on imaging; and a cardiac biomarker is raised — troponin or a natriuretic peptide.
The common error is treating the last two as alternatives. Right ventricular dilatation on the CT pulmonary angiogram alone, or a raised troponin alone, puts the patient in the intermediate-low group, where anticoagulation is the answer and intervention is not. Both together is what defines the group in which early deterioration is likely enough to justify considering thrombolysis. The other frequent slip is forgetting the first criterion — a hypotensive patient is high-risk, not intermediate-high, and belongs in a different algorithm entirely.
The practical habit is to complete all four at the point of diagnosis, before deciding on disposition. The CT is already done. Add a troponin and a natriuretic peptide to the same blood draw, calculate the sPESI from data you already have, and read the right ventricle to left ventricle ratio off the scan rather than waiting for a formal echocardiogram. That takes minutes and determines whether this is a patient for the ward or a patient about whom someone senior needs a conversation now.
- Require all four: stable haemodynamics, sPESI at least 1, right ventricular dysfunction, AND a raised biomarker.
- Right ventricular dilatation alone, or a raised troponin alone, is intermediate-low risk — anticoagulate and observe.
- Send troponin and a natriuretic peptide on the same draw as the initial bloods, before the question arises.
- Read the RV:LV ratio off the CT pulmonary angiogram; do not wait for a formal echocardiogram to classify.
- Hypotension moves the patient to high risk and out of this pathway altogether.
The statistics, in plain English
This category exists because risk is not binary. The point of requiring both imaging and biochemical evidence of right ventricular strain is that either one alone is a weak signal, while the two together identify a group whose short-term risk of deterioration is high enough — around 7% within a week in recent trial data — to justify considering an intervention that carries its own hazard. Loosening the definition moves lower-risk patients into a treatment whose benefit was never demonstrated in them, while its bleeding risk applies to everyone.
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 top clinical updates, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free