D-Dimer/Fibrinogen Ratio and Radiological Severity Scores in Acute Pulmonary Embolism: Is There Room for a New Thrombus-Burden Marker?

Le résumé fourni par la source

Background/Objectives: The D-dimer/fibrinogen ratio (D-d/F) has been proposed as a composite marker of fibrinolysis–coagulation balance. Whether D-d/F reflects CT-quantified thrombus burden and right ventricular dysfunction (RVD) in acute pulmonary embolism (PE) remains uncertain. Methods: Single-center retrospective cohort of consecutive adults with CTPA-confirmed PE (January 2022–October 2024). D-d/F = D-dimer (µg/mL)/fibrinogen (mg/dL). Thrombus burden: Qanadli and Mastora indices. RVD: RV/LV ratio, septal bowing, and IVC reflux. Associations: Spearman’s ρ with Steiger’s Z for between-marker comparisons. Discrimination for Qanadli ≥ 40% and RV/LV ≥ 1.0 by ROC. Two exploratory logistic models predicted Qanadli ≥ 40%: Model-1 (age, sex, D-d/F) and Model-2 adding RV/LV. Results: Among 112 patients (mean age 65.4 ± 15.6; 60% men), D-d/F correlated modestly with Qanadli (ρ = 0.233, p = 0.013) and Mastora (ρ = 0.274, p = 0.0034); strengths were similar to D-dimer (no between-marker difference: Steiger’s Z both p > 0.5). D-d/F correlated with RV/LV (ρ = 0.335, p < 0.001) and with IVC reflux (ρ = 0.247, p = 0.0085). CT indices related more strongly to hemodynamic markers (e.g., Qanadli with RV/LV ρ = 0.571, p < 0.0001; Mastora with RV/LV ρ = 0.620, p < 0.0001). Patients with septal bowing had higher D-dimer (median 4.65 vs. 2.74 µg/mL, p = 0.0037), higher D-d/F (1.04 vs. 0.61, p = 0.0018), and higher clot-burden scores (both p < 0.0001). For Qanadli ≥ 40%, AUCs were 0.621 for D-d/F (cut-off > 0.795; sens 58.8%, spec 62.3%) and 0.618 for D-dimer (>1.894 µg/mL; 84.3%, 37.7%); AUCs did not differ (p = 0.93). For RV/LV ≥ 1.0, AUCs were 0.693 for D-d/F (>0.607; 83.8%, 52.0%) and 0.684 for D-dimer (>2.849 µg/mL; 75.7%, 54.7%); p = 0.72. In Model-1, D-d/F predicted Qanadli ≥ 40% (OR = 1.43 per unit, p = 0.043; AUC = 0.64). After adding RV/LV (Model-2), discrimination improved (AUC = 0.796), RV/LV remained a strong predictor (p < 0.0001), and D-d/F was not retained (p = 0.287). Conclusions: In acute PE, D-d/F tracks thrombus burden and RVD to a degree comparable to D-dimer, but effects are modest. CT-based markers—particularly RV/LV—better reflect disease severity and are more predictive of high clot burden. Risk prediction and incremental utility of D-d/F were not assessed and warrant prospective evaluation.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.