PO-07 - Excluding pulmonary embolism in cancer patients using the Wells rule and age-adjusted D-dimer testing: an individual patient data meta-analysis.

INTRODUCTION: Among patients with clinically suspected pulmonary embolism (PE), imaging and anticoagulant treatment can be safely withheld in approximately one-third of patients based on the combination of a "PE unlikely" Wells score and a D-dimer below the age-adjusted threshold. The clinical utility of this diagnostic approach in cancer patients is less clear.

AIM: To evaluate the efficiency and failure rate of the original and simplified Wells rules in combination with age-adjusted D-dimer testing in patients with active cancer.

MATERIALS AND METHODS: Individual patient data were used from 6 large prospective studies in which the diagnostic management of PE was guided by the original Wells rule and D-dimer testing. Study physicians classified patients as having active cancer if they had new, recurrent, or progressive cancer (excluding basal-cell or squamous-cell skin carcinoma), or cancer requiring treatment in the last 6 months. We evaluated the dichotomous Wells rule and its simplified version (Table). The efficiency of the algorithm was defined as the proportion of patients with a "PE unlikely" Wells score and a negative age-adjusted D-dimer, defined by a D-dimer below the threshold of a patient's age times 10 μg/L in patients aged ≥51 years. A diagnostic failure was defined as a patient with a "PE unlikely" Wells score and negative age-adjusted D-dimer who had symptomatic venous thromboembolism during 3 months follow-up. A one-stage random effects meta-analysis was performed to estimate the efficiency and failure.

RESULTS: The dataset comprised 938 patients with active cancer with a mean age of 63 years. The most frequent cancer types were breast (13%), gastrointestinal tract (11%), and lung (8%). The type of cancer was not specified in 42%. The pooled PE prevalence was 29% (95% CI 25-32). PE could be excluded in 122 patients based on a "PE unlikely" Wells score and a negative age-adjusted D-dimer (efficiency 13%; 95% CI 11-15). Two of 122 patients were diagnosed with non-fatal symptomatic venous thromboembolism during follow-up (failure rate 1.5%; 95% CI 0.13-14.8). The simplified Wells score in combination with a negative age-adjusted D-dimer had an efficiency of 3.9% (95% CI 2.0-7.6) and a failure rate of 2.4% (95% CI 0.3-15).

CONCLUSIONS: Among cancer patients with clinically suspected PE, imaging and anticoagulant treatment can be withheld in 1 out of every 8 patients by the original Wells rule and age-adjusted D-dimer testing. The simplified Wells rule was neither efficient nor safe in this population.