Background and objective: Host defence mechanisms to counter virus infection include the activation of the broncho-alveolar haemostasis. Fibrin degradation products secondary to extravascular fibrin breakdown could contribute to the marked increase in D-Dimers during COVID-19. We sought to examine the prognostic value on lung injury of D-Dimers in non-critically ill COVID-19 patients without thrombotic events.
Methods: This study retrospectively analysed hospitalized COVID-19 patients classified according to a D-Dimers threshold following the COVID-19 associated haemostatic abnormalities (CAHA) classification at baseline and at peak (Stage 1: D-Dimers less than three-fold above normal; Stage 2: D-Dimers three- to six-fold above normal; Stage 3: D-Dimers six-fold above normal). The primary endpoint was the occurrence of critical lung injuries on chest computed tomography. The secondary outcome was the composite of in-hospital death or transfer to the intensive care unit (ICU).
Results: Among the 123 patients included, critical lung injuries were evidenced in 8 (11.9%) patients in Stage 1, 6 (20%) in Stage 2 and 15 (57.7%) in Stage 3 (p = 0.001). D-Dimers staging at peak was an independent predictor of critical lung injuries regardless of the inflammatory burden assessed by CRP levels (OR 2.70, 95% CI (1.50-4.86); p < 0.001) and was significantly associated with increased in-hospital death or ICU transfer (14.9 % in Stage 1, 50.0% in Stage 2 and 57.7% in Stage 3 (p < 0.001)). D-Dimers staging at peak was an independent predictor of in-hospital death or ICU transfer (OR 2.50, CI 95% (1.27-4.93); p = 0.008).
Conclusions: In the absence of overt thrombotic events, D-Dimers quantification is a relevant marker of critical lung injuries and dismal patient outcome.
Keywords: coronavirus; fibrinolysis; lung injury; respiratory distress syndrome; thrombosis.