299 The Diagnostic Value of D-dimer Levels for the Exclusion of Aortic Dissection

Abstract

Acute aortic dissection is a highly fatal disorder if not diagnosed promptly. The current gold standard for diagnosis of aortic dissection is with CT angiogram, MRI, or transesophageal echocardiogram. These advanced imaging techniques are expensive, time-consuming, and carry risks, so they are not ideal, especially for low risk patients. A universal, rapid-screening biomarker could reduce the need for advanced imaging tests to exclude the diagnosis of acute aortic dissection. Some prior studies have suggested that serum d-dimer levels are elevated in patients with aortic dissection, but the diagnostic value of d-dimer levels for ruling out aortic dissection remains uncertain. Thus, we sought to further assess the diagnostic usefulness of d-dimer levels as a means to rule out aortic dissection in emergency department patients. This was a retrospective analysis of emergency department patients from the Hospital Corporation of America (HCA) database. We queried the database to find patients who had a diagnosis of acute aortic dissection and who had a d-dimer test result within 24 hours of arrival to the emergency department. We also identified patients who had a diagnosis of (nonspecific) chest pain and who had a d-dimer test result within 24 hours of arrival to the emergency department. The median d-dimer value was compared in the aortic dissection group and the chest pain group using the Wilcoxon Rank Sum Test. We determined the sensitivity and specificity of d-dimer for aortic dissection using the standard cutoff value of 500 ng/mL, and we formulated a receiver operating characteristic (ROC) curve. Finally, we sought to determine if there is a different d-dimer cutoff value with better test characteristics than the standard 500 ng/mL. D-dimer values are reported in fibrinogen equivalent units (FEUs). A total of 48,902 patients’ records were analyzed for this study. This sample included 572 aortic dissection patients and 48,330 chest pain patients. The median d-dimer value for the aortic dissection group was 2455 ng/mL, and for the chest pain group, it was 385 ng/mL. The difference between the median values of each group of 2070 ng/mL was statistically significant (p-value <0.0001). Using a standard cutoff of 500 ng/mL, we found the sensitivity of the d-dimer to be 91.1% and the specificity to be 71.4%. To achieve a sensitivity greater than 95% for aortic dissection, the d-dimer cutoff had to be reduced to 375 ng/mL, at which point the sensitivity was 95.1% and the specificity was 48.5%. The Youden index maximized at 835 ng/mL, at which point the sensitivity was 82.9% and the specificity was 86.3%. The area under the ROC curve was 0.91 (95% CI 0.89 to 0.92). In general, serum d-dimer values are higher in patients with an aortic dissection than in those with nonspecific chest pain. At the standard cutoff of 500 ng/mL, the serum d-dimer seems to have a high sensitivity for aortic dissection, but it is likely not high enough such that d-dimer levels alone can be used to exclude aortic dissection.