Objectives: The hypothesis was that the tandem measurement of D-dimer and myeloperoxidase (MPO) or C-reactive protein (CRP) could significantly decrease unnecessary pulmonary vascular imaging in emergency department (ED) patients evaluated for pulmonary embolism (PE) compared to D-dimer alone.Methods: The authors measured the sequential combinations of D-dimer and MPO and D-dimer and CRP in a prospective sample of ED patients evaluated for PE at two centers. Patients were followed for 90 days for venous thromboembolism (VTE, either PE or deep venous thrombosis [DVT]), which required the consensus of two of three blinded physician reviewers.
Results:The authors enrolled 304 patients, 22 with VTE (7%; 95% confidence interval [CI] = 5% to 10%). The sensitivity and specificity of a D-dimer alone (cutoff ‡ 500 ng ⁄ mL) were 100% (95% CI = 85% to 100%) and 59% (95% CI = 53% to 65%), respectively, and was followed by pulmonary vascular imaging negative for PE in 38% (115 ⁄ 304; 95% CI = 32% to 44%). The combination of either a negative D-dimer, or MPO < 22 mg ⁄ dL, had a sensitivity of 100% and specificity of 73% (95% CI = 67% to 78%). Thus, tandem measurement of D-dimer and MPO would have decreased the frequency of subsequent negative pulmonary vascular imaging from 38% to 25% (95% CI of the difference of )13% = )5% to )20%). The combination of CRP and D-dimer would not have significantly improved the rate of negative imaging.
Conclusions:The tandem measurement of D-dimer and MPO would have significantly decreased negative pulmonary vascular imaging compared with D-dimer alone and should be validated prospectively.
ACADEMIC EMERGENCY MEDICINE 2008; 15:800-805 ª 2008 by the Society for Academic Emergency MedicineKeywords: biological markers, pulmonary embolism E mergency physicians' awareness of pulmonary embolism (PE) as an important and potentially life-threatening illness is increasing. Currently, emergency physicians employ a diagnostic strategy that involves an assessment of pretest probability based on clinical evaluation followed by D-dimer testing in low-and moderate-risk patients. Patients with a positive D-dimer, or those at high risk, generally defined as a pretest probability for venous thromboembolism (VTE) of about 40% or greater, go on to pulmonary vascular imaging.1 While the D-dimer has proven to be a sensitive marker for PE, a low specificity leads to frequent negative pulmonary vascular imaging.Pulmonary vascular imaging is a significant source of the overall increase in imaging studies performed in emergency departments (EDs). Currently, approximately 10% of ED patients receive either a computed tomography (CT) scan or magnetic resonance imaging as a part of their ED care, and more than 1% receive