GRAND RAPIDS, MICH—Now that the enzyme-linked immunosorbent assay (ELISA) D-dimer test for pulmonary embolism (PE) is becoming widely available, physicians are asking, “How good is it?” Although many studies have been published on the subject, they are of varying size, quality, and often use different parameters. Thus, Michael D. Brown, MD, and colleagues performed a meta-analysis to determine the accuracy of the D-dimer test in diagnosing PE in the emergency department (ED).

Dr. Brown, Associate Professor of Emergency Medicine at Michigan State University in Grand Rapids, found that the ELISA D-dimer test was 94% sensitive but only 45% specific for PE.[1] “So, the test is most useful for ruling out PE,” he said in an interview.

A MORE FOCUSED ANALYSIS

The meta-analysis was more focused than earlier meta-analyses of D-dimer testing, which mixed inpatients and outpatients who may have had deep vein thrombosis or PE and which combined findings for the different D-dimer tests (ELISA, latex agglutination, and whole-blood agglutination). “I wanted to focus on the ELISA D-dimer in adult outpatients with PE,” related Dr. Brown. “This more closely reflects a typical ED patient population,” he commented.

The investigators included 11 studies with a combined 2,126 patients in their analysis. To be included in the meta-analysis, a study had to be a prospective investigation primarily involving patients who presented to the ED with signs and symptoms suggesting PE. Each study also had to pass a rigorous quality review.

CONFOUNDING FACTORS

Dr. Brown and colleagues attempted to perform subgroup analyses by age, comorbidity, or PE symptom duration—these covariates can significantly influence the accuracy of the ELISA D-dimer test. Because of limitations in the data, however, only patients age 70 or older could be adequately analyzed using age as a subgroup; in this setting, the test’s specificity dropped to 14%.

Test sensitivity decreased in the presence of comorbidity, defined as surgery, trauma, myocardial infarction, stroke, disseminated intravascular coagulation, acute infection, pregnancy, a postpartum state, or active cancer within 10 days of evaluation for PE. Sensitivity and specificity both declined when PE symptom duration exceeded three days.

Two of the largest studies in the meta-analysis examined the rapid ELISA D-dimer test, and both studies showed an impressive 100% sensitivity for PE. However, each of the studies was prone to bias, which may have caused a slight overestimation of test sensitivity.

Based on his findings, Dr. Brown recommended additional PE tests (computed tomography, ventilation/perfusion scanning, or pulmonary angiography) when an ELISA D-dimer test result is positive. “If it is negative and the pretest probability of PE is low, you can feel fairly confident about discharging the patient home without further tests,” he added.

—Timothy Begany

Reference
1. Brown MD, Rowe BH, Reeves MJ, et al. The accuracy of the enzyme-linked immunosorbent assay D-dimer test in the diagnosis of pulmonary embolism: a meta-analysis. Ann Emerg Med. 2002;40:133-144.

Return to table of contents