Evaluating outcomes of patients with suspected pulmonary embolism using an age-adjusted cutoff with a D-dimer unit-based assay

Introduction

Studies have suggested that an age-adjusted D-dimer (AADD) could improve specificity while maintaining sensitivity for the diagnosis of pulmonary embolism (PE). Most laboratories and published data report D-dimer in fibrinogen equivalent units, but some laboratory assays report using D-dimer units (DDU). However, only two small studies have examined AADD with a DDU-based assay. Our study aims to assess the sensitivity and specificity of AADD cutoffs, compared to the conventional cutoffs, in a large cohort of patients who underwent D-dimer testing using a DDU-based assay, for suspected PE.

Methods

We included 1831 patients who presented to the emergency department and underwent a diagnostic workup that included a D-dimer test followed by computed tomography pulmonary angiography (CTA). The laboratory utilized the HemosIL D-Dimer HS reagent, reporting D-dimer values in DDU (ng/mL). We retrospectively adjusted D-dimer levels for patients using the formula: age (years) x 5 ng/mL. This adjustment created a comparison group to evaluate the test characteristics of the AADD cutoff compared to the conventional cutoff.

Results

The conventional D-dimer cutoff of 230 ng/mL had a sensitivity of 97.4 % (95 % Confidence Interval [CI]: 95.3–99.4 %), specificity of 7.3 % (CI: 6.0–8.6 %), positive predictive value (PPV) of 13 % (CI: 11.4–14.6 %), and negative predictive value (NPV) of 95.1 % (CI: 91.3–98.9 %). The AADD cutoff had a sensitivity of 96.1 % (CI: 93.5–98.6 %), specificity of 16.6 % (CI: 14.8–18.4 %), PPV of 14.1 % (CI: 12.3–15.8 %), and NPV of 96.7 % (CI: 94.6–98.8 %).

Conclusion

In this study using a DDU-based assay, the specificity of the AADD cutoff was greater while the sensitivity was similar to that of the unadjusted cutoff, with no statistically significant difference observed.