An Age-Adjusted D-dimer Threshold for Emergency Department Patients With Suspected Pulmonary Embolus: Accuracy and Clinical Implications

Abstract

We determine the accuracy of an age-adjusted D-dimer threshold to detect pulmonary embolism in emergency department (ED) patients older than 50 years and describe current ED practices when evaluating possible pulmonary embolism. This was a retrospective study of ED encounters for suspected pulmonary embolism from 2008 to 2013. We used structured data to calculate the sensitivity, specificity, negative predictive value, and positive predictive value of different D-dimer thresholds. We describe the incidence of pulmonary embolism, the proportion of patients receiving imaging concordant with D-dimer levels, and the number of "missed" pulmonary embolisms. These findings were used to estimate patient outcomes based on different D-dimer thresholds. Among 31,094 encounters for suspected pulmonary embolism, there were 507 pulmonary embolism diagnoses. The age-adjusted D-dimer threshold was more specific (64% versus 54%) but less sensitive (93% versus 98%) than the standard threshold of 500 ng/dL; 11,999 imaging studies identified 507 pulmonary embolisms (4.2%); of these, 1,323 (10.6%) were performed with a D-dimer result below the standard threshold. Among patient encounters without imaging, 17.6% had D-dimer values above the threshold, including 5 missed pulmonary embolisms. Among patients who received imaging, 10.6% had a negative D-dimer result. Applying an age-adjusted D-dimer threshold to our sample would avert 2,924 low-value imaging tests while resulting in 26 additional cases of missed pulmonary embolism. An age-adjusted D-dimer limit has the potential to reduce chest imaging among older ED patients and is more accurate than a standard threshold of 500 ng/dL. Our findings support the adoption of an age-adjusted D-dimer cutoff in community EDs.