Wells score calculator (DVT and PE)

Two checklists, one name

“Wells score” can mean either of two different rules, and mixing them up is the most frequent error. Both aim at the same problem: a clinician sees a person with a swollen leg or with chest symptoms and must decide how much testing is reasonable. A rule that sorts patients into “clot unlikely” and “clot likely” lets the team send those in the first group for a D-dimer blood test alone, and the second group straight to imaging. The sections below give the exact items, thresholds and the published evidence, as quoted in the NICE guideline NG158.

What are the Wells criteria?

The Wells criteria are two separate checklists that estimate how likely it is that a person with symptoms has a blood clot. One is for deep vein thrombosis (DVT) in the leg, from a 2003 paper in the New England Journal of Medicine. The other is for pulmonary embolism (PE), a clot in the lungs, from a 2000 paper in Thrombosis and Haemostasis. Both were written by Philip Wells and colleagues in Ottawa. Each item has a point value, and the total puts a person in a probability group.

How is the DVT score read?

There are 9 items worth one point each and one worth minus two points, if another diagnosis is at least as likely as DVT. The two-level reading used in the 2003 trial and in the NICE guideline is DVT likely at 2 points or more and DVT unlikely at 1 point or fewer. The older three-level reading is low at 0 or less, moderate at 1 to 2, and high at 3 or more. In the 1997 Lancet study the share of patients with a DVT was 3 percent, 17 percent and 75 percent in those three groups.

How is the PE score read?

The seven items are worth between 1 and 3 points each (up to 12.5 in total). The two-level reading is PE likely when the score is more than 4 and unlikely at 4 or less. The three-level reading is low below 2, moderate from 2 to 6 and high above 6. In the derivation study, patients scoring 4 or less with a negative SimpliRED D-dimer (an older point-of-care test) had PE in 2.2 percent of cases (1.7 percent in the validation set).

What do the guidelines say to do with the score?

The NICE guideline (NG158, updated 2023) says to use the two-level score. For a likely DVT it recommends a proximal leg vein ultrasound scan, and for an unlikely DVT a D-dimer test first. For a likely PE it recommends a CT pulmonary angiogram, and for an unlikely PE a D-dimer test. It also says to consider an age-adjusted D-dimer threshold for people over 50. The 2019 European Society of Cardiology PE guideline describes the same age-adjusted idea, with a cut-off of age × 10 µg/L above age 50.

What is the role of D-dimer?

D-dimer is a blood test for fibrin breakdown products. The 2000 and 2003 Wells studies were built to answer when a negative D-dimer is safe enough to avoid imaging. In the 2003 trial, a person judged unlikely to have DVT and with a negative D-dimer could skip ultrasound: of 566 patients in the D-dimer group, 218 (39 percent) needed no scan. D-dimer rises in many conditions, so a positive result alone proves nothing.

Does the US guideline use the Wells score?

The 2018 clinical policy of the American College of Emergency Physicians (Ann Emerg Med 2018;71:e59) addresses risk stratification in the emergency department. As summarized by REBEL EM, it says that PERC (the PE rule-out criteria) can exclude PE without further testing in low-risk patients, and that a negative age-adjusted D-dimer can exclude PE in patients over 50 who are at low or intermediate risk. Both are Level B recommendations. We did not find the Wells items themselves in that summary.

Why can two clinicians get different scores?

Two items are judgments: whether an alternative diagnosis is as likely as DVT, and whether it is less likely than PE. They carry a lot of weight (minus 2 for DVT, plus 3 for PE) and depend on who is judging. This page cannot make that call for you.

Worked examples

DVT: a person with localized tenderness along the deep veins, a swollen entire leg and pitting edema confined to that leg gets 3 points: likely on the two-level score and high on the three-level score. A person with only a calf 3 cm larger, but in whom another diagnosis is at least as likely, scores 1 − 2 = -1: unlikely, low. PE: a heart rate above 100 with surgery in the previous 4 weeks gives 3 points, which is unlikely on the two-level score and moderate on the three-level one. Signs of DVT, PE judged more likely than the alternatives and a heart rate above 100 give 7.5 points: likely and high.

Limits of this method

The 2003 trial enrolled outpatients with suspected leg DVT, and the 2000 PE model came from a prospective cohort of patients with suspected PE; the abstracts we read say nothing about hospitalized or critically ill patients, so this page makes no claim for them. The PE score cannot be used on its own to rule out PE; the derivation study only showed safety when combined with a negative D-dimer. The NICE document also reminds readers that the PERC rule has not been validated in people with COVID-19.

Common mistakes

A common error is mixing the two scores: a PE score of 3.0 is a very different thing from a DVT score of 3. Another is ticking the DVT-sign item in the PE list without the required findings: “Clinical signs and symptoms of DVT” in the PE score needs at least leg swelling and pain on palpation of the deep veins (NICE wording). A third is treating 4.5 as “unlikely” because it is close to 4: the PE two-level cut-off is strictly more than 4 points.

Sources: Wells PS, Anderson DR, Rodger M, et al., “Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis,” N Engl J Med 2003;349:1227–1235; Wells PS, Anderson DR, Bormanis J, et al., “Value of assessment of pretest probability of deep-vein thrombosis in clinical management,” Lancet 1997;350:1795–1798; Wells PS, Anderson DR, Rodger M, et al., “Derivation of a simple clinical model to categorize patients’ probability of pulmonary embolism,” Thromb Haemost 2000;83:416–420; National Institute for Health and Care Excellence, Venous thromboembolic diseases: diagnosis, management and thrombophilia testing (NG158), recommendations 1.1.2 to 1.1.21 and Tables 1 and 2; Konstantinides SV, et al., “2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism,” Eur Heart J 2020;41:543–603; ACEP Clinical Policies Subcommittee, Ann Emerg Med 2018;71:e59–e109 (known to us through a summary; the policy PDF was not opened). Full texts of the Wells papers were not opened; items and thresholds are as printed in NICE. Related: CHA₂DS₂-VASc, CURB-65, mean arterial pressure.

For general information only — not medical advice, and not a tool for patient care decisions. Talk with a doctor or call emergency services about anyone you are worried about. How we check our calculators.