Clinical suspicion of pulmonary embolism: diagnostic utility of prediction scores and D-dimer measurement

Lekerika N, Arana Arri E, García Montero A, García L, Gómez A, Carreras M

Objective: To assess the utility of the Wells and Geneva scoring systems and D-dimer

measurements when pulmonary embolism (PE) is suspected.

Methods: Prospective cohort study of the diagnostic performance of the Wells and

Geneva scores and D-dimer measurement. We analyzed data for 637 consecutive

patients suspected of having PE; the patients were being treated in the emergency

department of a tertiary care hospital. The following performance measures were

assessed: sensitivity, specificity, positive and negative predictive values (PPV and NPV),

and the positive and negative likelihood ratios (LR+ and LR–). PE was confirmed by

computed tomography and the patients were followed for the recommended 3 months.

Results: The mean (SD) age was 67.9 (16.3) years and 54.6% of the patients were

women. The prevalence of PE in the cohort was 15.1%. For the Wells score, the NPV was

85.3% (95% CI, 82.1%–88.0%), and the PPV was 62.5% (95% CI, 42.7%–78.8%).

Sensitivity was 15.6% (95% CI, 9.7%–24.2%) and specificity 98.1% (95% CI,

96.5%–99.0%). The LR+ was 8.3 (95% CI, 3.7–18.4), and the LR– was 0.86 (95% CI,

0.78–0.95). For the Geneva score, the NPV was 82.8% (95% CI, 78.9%–86.0%), and

the PPV was 58.8% (95% CI, 36.0%–78.4%). Sensitivity was 11.8% (95% CI,

6.5%–20.3%) and specificity 98.1% (95% CI, 96.1%–99.1%). The LR+ was 6.2 (95% CI,

2.42–15.74), and the LR– was 0.90 (95% CI, 0.81–1.0). For D-dimer level, the NPV was

99.2% (95% CI, 95.4%–99.9%), and the PPV was 21.7% (95% CI, 18.1%–25.9%).

Sensitivity was 98.9% (95% CI, 94.2%–99.8%) and specificity 26.4% (95%,

22.6%–30.7%). The LR+ was 1.34 (95% CI, 1.27–1.43), and the LR– was 0.04 (95% CI,

0.01–0.29).

Conclusions: The Wells and Geneva scoring systems are good useful in the emergency

department for establishing a preliminary clinical diagnosis of suspected PE. The Wells

score has better specificity. However, diagnosis cannot be based on any of these tools

because of their low sensitivity. Similarly, D-dimer measurement can allow PE to be ruled

out when the result is negative, but this criterion must be used in conjunction with the

scoring systems.

Arana Arri E, Carreras M, García L, García Montero A, Gómez A, Lekerika N