Background: Pulmonary embolism (PE) is common, and diagnosis is often difficult. Investigation has traditionally required expensive imaging procedures that are frequently nondiagnostic. Consequently, current practice favors noninvasive diagnosis of PE using algorithms combining risk assessment and d-dimer. Despite the proven safety of this approach, concern persists about such strategies to exclude PE, largely due to variable d-dimer sensitivity. The aim of this study was to prospectively assess the safety of a new algorithm combining a novel, rapid d-dimer test (IL Test; Instrumentation Laboratory; Lexington, MA), Pao2 measurement, and risk factor assessment in excluding PE in subjects presenting to an acute care setting.
Methods: All patients aged 18 to 60 years presenting to the emergency department of Christchurch Hospital with suspected PE underwent measurement of d-dimer (IL Test latex-enhanced immunoassay) and Pao2, and were assessed for the presence of major clinical risk factors. Those with no risk factors, normal d-dimer findings, and Pao2 ≥ 80 mm Hg (study arm A) were discharged and followed up by telephone questionnaires over 12 months. Those with elevated d-dimer levels, Pao2 < 80 mm Hg, or one or more risk factors (study arm B) were managed as per hospital guidelines. Outcome data were collected on these patients. Our primary outcome was incidence of PE in group A during the first 3-month follow-up period.
Results: Three hundred twenty-eight patients were enrolled, of whom 149 were assigned to group A and 179 were assigned to group B. In none of the group A patients was PE diagnosed over the subsequent 3-month period (0%; 95% confidence interval, 0 to 2.1%). PE was diagnosed in 37 group A patients (21%).
Conclusions: The latex-enhanced immunoassay d-dimer and normal arterial oxygen pressure levels can safely exclude PE in a low-risk population presenting to an acute care setting. While these results cannot be extrapolated to all patients with suspected PE, they do confirm the safety of this approach in a population with a low underlying risk of PE.