Diagnostic Accuracy of Multiple D-Dimer Cutoff Thresholds and Other Clinically Applicable Biomarkers for the Detection and Radiographic Evaluation of Pulmonary Embolism

HIGHLIGHTS: According to recent guidelines for the diagnosis and management of pulmonary embolism (PE), the conventional Dd cutoff (0.5 mg/L) has excellent sensitivity for the diagnosis of PE, although there is a marked decrease in specificity. Age-adjusted Dd cutoff thresholds for ruling out diagnoses of PE in emergency departments have significantly improved the performance of Dd testing, especially for the elderly, although little is known about the efficacy to reduce the number of CTPAs that could have been performed. The most interesting finding in our study was that the novel proposed Dd cutoff (0.74 mg/L) emerged with increased specificity and equal sensitivity compared with 0.5 mg/L and could have reduced the number of CTPAs that were performed in our institution by 17.2%. Ancillary CTPA insights of consolidation/atelectasis and unilateral pleural effusion were significantly associated with PE. ABSTRACT: Background: Diagnostic work-up of pulmonary embolism (PE) remains a challenge. Methods: We retrospectively studied all patients referred for computed tomography pulmonary angiography (CTPA) with suspicion of PE during a 12-month period (2018). The diagnostic accuracy of different D-dimer (Dd) cutoff thresholds for ruling out PE was evaluated. Furthermore, the association of Dd and red cell distribution width (RDW) with embolus location, CTPA findings, and patient outcome was recorded. Results: One thousand seventeen (n = 1017) patients were finally analyzed (mean age: 64.6 years (SD = 11.8), males: 549 (54%)). PE incidence was 18.7%. Central and bilateral embolism was present in 44.7% and 59.5%, respectively. Sensitivity and specificity for conventional and age-adjusted Dd cutoff was 98.2%, 7.9%, and 98.2%, 13.1%, respectively. A cutoff threshold (2.1 mg/L) with the best (64.4%) specificity was identified based on Receiver Operating Characteristics analysis. Moreover, a novel proposed Dd cutoff (0.74 mg/L) emerged with increased specificity (20.5%) and equal sensitivity (97%) compared to 0.5 mg/L, characterized by concurrent reduction (17.2%) in the number of performed CTPAs. Consolidation/atelectasis and unilateral pleural effusion were significantly associated with PE (p < 0.05, respectively). Patients with consolidation/atelectasis or intrapulmonary nodule(s)/mass on CTPA exhibited significantly greater median Dd values compared to patients without the aforementioned findings (2.34, (IQR 1.29–4.22) vs. 1.59, (IQR 0.81–2.96), and 2.39, (IQR 1.45–4.45) vs. 1.66, (IQR 0.84–3.12), p < 0.001, respectively). RDW was significantly greater in patients who died during hospitalization (p = 0.012). Conclusions: Age-adjusted Dd increased diagnostic accuracy of Dd testing without significantly decreasing the need for imaging. The proposed Dd value (0.74 mg/L) showed promise towards reducing considerably the need of CTPA. Multiple radiographic findings have been associated with increased Dd values in our study.