Thrombo-embolic events associated with Covid-19 ARDS - Epidemiology and risk factors

Introduction: Covid-19 severe ARDS is frequently associated with venous thrombosis and pulmonary emboli The first events we diagnosed lead us to recommend a high heparin prophylaxis (4000 IU enoxaparin bid or 6000 IU bid if weight > 120 kg;targeted anti-Xa activity 0 3-0 5 UI/mL) or an Immediate use of curative anticoagulant therapy (enoxaparin 100 IU/kg bid or unfractionated heparin 500 IU/ kg/d;targeted anti-Xa 0 5-0 7 UI/mL) Objectives: To evaluate the risk factors and epidemiology of deep venous thrombosis (DVT) and pulmonary emboli (PE) when this strategy is applied to all the Covid-19 patients with ARDS Methods:All the consecutive patients with severe ARDS due to SARSCov2 were systematically followed until hospital discharge or day60 Doppler US of the vein was routinely performed to all patients IV CT scan was performed as clinically indicated We recorded daily clinical and biological covariates as well as antimicrobial, anticoagulant and anti-inflammatory therapies from ICU admission to ICU discharge Patients were followed until hospital discharge or day 60 No patient was lost to follow-up We use a competing cause-specific risk model to evaluate the risk of DVT/PE during the follow-up (SAS 9 X, SAS system NC, USA) We tested variables at ICU admission and variables collected daily as time-dependent covariates Variables collected at time t-1 were used to model even occurring at time t A p value of 0 05 or less was considered significant The impact of time-dependent DVT/PE on patients' survival was tested using a Cox model Results: Of the 134 patients with Sars-Cov2 ARDS admitted in our ICU (age 59 5 [51;69];SAPSII: SOFA 5 [4;7]), 47 (35 1%) died at day 28 21 developed a DVT/PE (10/21 ultimately died), 45 died without DVT/PE and 68 were censored alive without DVT/PE Main characteristics of the patients and anticoagulant therapy used at ICUadmission are on Table The cause specific model identified high SOFA score, use of central vein catheter, hypothermia less than 36 °C, high BUN, high ALAT, high LDH, low hematocrit, high neutrophils count as risk factors of DVT/ PE Neither anticoagulant therapy (curative vs prophylaxis) given the day before nor anti-Xa activity were related with DVT/PE risk Platelet count, prothrombin time, fibrinogen level, fibrin monomers, D-dimer, troponin levels were not related to DVT/PE risk DVT/PE was not related to survival even after adjusting on comorbidities and SOFA score at ICU admission HR = 1 679;95%CI [0 802;3 512], p = 0 17 Conclusion: Even when using a high-risk prophylaxis, curative anticoagulant therapy, DVT/PE occurred in 16% of cases The risk is higher in patients with high SOFA score, renal and hepatic failure, high neutrophil count suggesting that other mechanisms than coagulation (i e endothelial activation and/or hypofibrinolysis) may concur to thromboembolic events (Table Presented)