D-dimer — Pulmonary Embolism
Test Characteristics
| Metric | Value |
|---|---|
| False-negative rate | 3% (sensitivity 97%) |
| False-positive rate | 59% (specificity 41%) |
| Bayes factor (positive test) | 1.5× |
| Bayes factor (negative test) | 1/10× |
| Base rate | 5% of ED patients with suspected PE (Wells low/moderate probability) |
Interpreting Results
| Scenario | Prior | + Result | − Result |
|---|---|---|---|
| Low-probability Wells score | 5% | 1.5 × 5% = 7.5%5%× 1.57.5% | 1/10 × 5% = 0.5%5%÷ 100.5% |
| Moderate-probability Wells score | 20% | 1.5 × 20% = 30%20%× 1.530% | 1/10 × 20% = 2%20%÷ 102% |
The paradigmatic rule-out test. A positive D-dimer is nearly worthless (1.5×). But a negative D-dimer (1/10×) drops PE probability 10-fold — in a low-probability patient, this avoids CT angiography. Always combine with Wells criteria.
- + result: nearly worthless
- − result: risk drops from vegetarian (5%) to identical twin (0.5%) — that's the whole point
Sources:
- Systematic review and meta-analysis, Blood Advances, 2020. ASH Publications.
- NEJM, 2019. D-dimer adjusted to clinical probability.