Diagnostic Performance of D-Dimer Compared to Wells and Geneva Scores for Pulmonary Embolism in COVID-19: A Retrospective Cohort Study

Introduction: Classical clinical prediction rules (Wells and Geneva) are widely used to assess the risk of Pulmonary Embolism (PE). Still, their performance in Coronavirus Disease 2019 (COVID-19) is uncertain, given frequent D-Dimer (DD) elevations unrelated to thrombosis. Aim: To compare the diagnostic performance of an optimised DD cut-off against Wells and Geneva scores in Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) confirmed COVID-19 patients with suspected PE. Materials and Methods: A retrospective, single-centre cohort study was conducted at Department of Internal Medicine, Consorci Sanitari de Terrassa, Barcelona, Spain, from Consorci Sanitari de Terrassa (Barcelona, Spain) from January 2021 to June 2021. Adult COVID-19 patients undergoing Computed Tomography Pulmonary Angiography (CTPA) for suspected PE were analysed. Receiver Operating Characteristic (ROC) analysis was used to assess DD, Wells, and Geneva; the optimal DD threshold was identified by Youden’s index. Group comparisons used t-test or Mann-Whitney U for continuous variables and Chi-square or Fisher’s-exact test for categorical variables; p-value <0.05 was considered significant. Results: CTPA was performed in 586 patients, of whom 148 had confirmed COVID-19 infection. PE was diagnosed in 13.5% (20/148) of cases. Patients with PE had significantly higher DD levels than those without PE (p-value <0.05). A DD threshold of 3126 ng/mL yielded 80% sensitivity and 68.5% specificity, potentially avoiding 87 CTPAs while missing 4 PE diagnoses. In comparison, the Wells score showed poor performance {Area Under Curve (AUC) 0.60, 95% Confidence Interval (CI) 0.48-0.72; sensitivity 55%, specificity 56%}, whereas the revised Geneva score was even less accurate (AUC 0.41, 95% CI 0.26-0.55; sensitivity 45%, specificity 55%). Conclusion: In the present cohort of COVID-19 patients with suspected PE, an optimised DD cut-off demonstrated superior discriminatory performance compared to conventional clinical prediction rules. Therefore, in COVID-19 settings with suspected PE, a DD threshold of approximately 3,126 ng/mL may help reduce unnecessary Computed Tomography (CT) pulmonary angiographies when interpreted in conjunction with clinical probability and physician judgment.