Prothrombin time, international normalized rate and in-hospital mortality in COVID-19

BACKGROUND: Blood coagulation alterations are frequent in patients with Coronavirus disease 2019 (COVID-19), particularly in those with severe forms. We investigated the association between standard parameters of coagulation and in-hospital mortality in COVID-19. METHODS: Demographic, clinical and laboratory data at hospital admission, including prothrombin time (PT), international normalized rate (INR), activated thromboplastin time (aPTT), and D-dimer were retrospectively collected in a consecutive series of 309 COVID-19 hospitalized patients. The associations between parameters of coagulation and in-hospital mortality were investigated with receiver operating characteristics (ROC), multiple regression and Kaplan-Meyer analyses. RESULTS: In the overall population, 220 (71.2%) patients were discharged alive, whereas the remaining 89 (28.8%) died. Non-survivors had significantly higher INR (median: 1.20; IQR: 1.03-1.32 vs. 1.06; IQR: 1.02-1.11, P<0.001), PT (median: 12.0 sec; IQR: 11.1-14.0 vs. 11.4 sec; IQR: 11.0-11.9, P<0.001), aPTT (median: 25.1 sec; IQR: 22.7-29.6 vs. 23.4 sec; IQR: 21.4-25.1, P<0.001) and D-dimer (median: 1.36 mu g/mL; IQR: 0.87-4.11 vs. 0.77 mu g/mL; IQR: 0.43-1.58, P<0.001). In multivariate Cox regression analysis, both the INR (HR=1.8459; 95% CI: 1.0713-3.1806, P=0.027) and PT (HR=1.071; 95% CI: 1.0132-1.1303, P=0.015), but not the aPTT and D-dimer, remained independently associated with survival. CONCLUSIONS: Both the PT and INR are independently associated with in-hospital mortality in COVID-19. The clinical utility of these parameters for risk stratification warrants further investigations.