4 clinical decision rules plus D-dimer testing each had high sensitivity and low specificity for excluding acute pulmonary embolism

Prospective cohort study comparing 4 previously developed clinical decision rules (CDRs) (WR, SWR, RGS, and SRGS), each combined with a high-sensitivity D-dimer test.

7 academic and nonacademic hospitals in the Netherlands.

807 patients ≥ 18 years of age (mean age 53 y, 60% women, 80% outpatients) with clinically suspected acute PE (sudden onset of dyspnea, increase in existing dyspnea, or sudden onset of pleuritic chest pain). Exclusion criteria included previous PE, treatment with therapeutic heparin initiated ≥ 24 hours before eligibility assessment, treatment with vitamin K antagonists, pregnancy, and contraindication to computed tomography (CT).

The WR/SWR and RGS/SRGS were summations of 7 and 8 patient variables, respectively. PE was considered unlikely if the D-dimer result was normal and scores were ≤ 4, ≤ 1, ≤ 5, and ≤ 2 for the WR, SWR, RGS, and SRGS, respectively. All other patients had CT, which led to exclusion of PE, further testing, or diagnosis and treatment of PE.

PE identified by CT or venous thromboembolism (VTE; PE or deep venous thromboembolism) during 3 months of follow-up.

23% of patients had prevalent PE. The operating characteristics of each CDR combined with D-dimer testing are in the Table. PE was excluded by CDR and a normal D-dimer result in 22% to 24% of patients.

In patients with suspected acute pulmonary embolism, the Wells rule, simplified Wells rule, revised Geneva score, and simplified revised Geneva score, when combined with D-dimer testing, are similar for ruling out PE, with high sensitivity and low specificity.