Hong Life CalculatorsUnderstanding the Wells score for DVT — concept and principle
The Wells score for deep vein thrombosis (DVT) estimates how likely it is that a painful or swollen leg is caused by a blood clot in the deep veins. It is used at the start of the work-up to decide what comes next: a D-dimer blood test, compression ultrasound, or both.
How it is calculated
Add 1 point for each of the following: active cancer; paralysis, paresis or recent plaster cast of the leg; bedridden for more than 3 days or major surgery within 12 weeks; localised tenderness along the deep veins; entire leg swollen; calf swelling more than 3 cm compared with the other leg; pitting oedema confined to the symptomatic leg; collateral (non-varicose) superficial veins; previously documented DVT. Then subtract 2 points if an alternative diagnosis is at least as likely as DVT. The total runs from −2 to 9.
In the two-level model used here, a score of 2 or more is "DVT likely" and 1 or less is "DVT unlikely".
The principle
No single sign reliably proves or excludes DVT. The Wells score combines risk factors (cancer, immobility, previous clot) with examination findings into a pretest probability. Diagnostic tests are then interpreted against that probability, as Bayes' theorem implies: a negative D-dimer is very reassuring when the starting probability is low, but much less so when it is high. The −2 item reflects the fact that many leg complaints, such as a muscle strain, cellulitis or a Baker's cyst, mimic DVT.
Worked example
- Defaults: all ten items No.
- Score = 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 = 0.
- 0 is 1 or less, so the result is DVT unlikely.
- For comparison, a patient with active cancer, a whole swollen leg and calf swelling over 3 cm scores 3, which is DVT likely.
How to read the result
"DVT unlikely" usually leads to a D-dimer test; if it is negative, DVT can generally be ruled out without imaging. "DVT likely" usually leads directly to ultrasound, with D-dimer and repeat scanning used in certain pathways. The score chooses the order of tests; it does not diagnose DVT.
Common mistakes
- Skipping the −2 item, or applying it without considering a real alternative.
- Using the score in pregnancy or in patients already on anticoagulation, settings in which it was not well validated.
- Assuming "unlikely" means DVT is excluded without any test.
Frequently asked questions
What is the difference between the original and the two-level Wells score?
The 1997 version used three groups (low, moderate, high). The 2003 version added previous DVT and split patients into likely (2 or more) and unlikely (1 or less).
Why is D-dimer not enough on its own?
D-dimer rises in many conditions, including infection, surgery, cancer, pregnancy and older age, so a positive result is not specific. Its value is mainly in ruling out.
Sources: Wells P. S. et al., New England Journal of Medicine, 2003 (D-dimer and the two-level DVT score) · Wells P. S. et al., Lancet, 1997 (original DVT pretest probability model)