D-dimer

Quantum mechanics

Clinical prediction rule

Pulmonary embolism

Statistics

Odds

Surgery

Receiver operating characteristic

Physics

Mathematics

Internal medicine

Geometry

Odds ratio

Medicine

Decision rule

Incidence (geometry)

Cutoff

Logistic regression

Replicate

Thrombosis and Haemostasis

Summary The Wells rule is a widely applied clinical decision rule in the diagnostic work-up of patients with suspected pulmonary embolism (PE).The objective of this study was to replicate, validate and possibly simplify this rule. We used data collected in 3,306 consecutive patients with clinically suspected PE to recalculate the odds ratios for the variables in the rule, to calculate the proportion of patients with PE in the probability categories, the area under the ROC curve and the incidence of venous thromboembolism during follow-up. We compared these measures with those for a modified and a simplified version of the decision rule. In the replication, the odds ratios in the logistic regression model were found to be lower for each of the seven individual variables (p=0.02) but the proportion of patients with PE in the probability categories in our study group were comparable to those in the original derivation and validation groups. The area under the ROC of the original, modified and simplified decision rule was similar: 0.74 (p=0.99; p=0.07).The venous thromboembolism incidence at three months in the group of patients with a Wells score ≤ 4 and a normal D-dimer was 0.5%, versus 0.3% with a modified rule and 0.5% with a simplified rule. The proportion of patients safely excluded for PE was 32%, versus 31% and 30%, respectively. This study further validates the diagnostic utility of theWells rule and indicates that the scoring system can be simplified to one point for each variable.

Further validation and simplification of the Wells clinical decision rule in pulmonary embolism