Prevalence and Risk Factors of Venous Thromboembolism in Critically Ill Patients with Severe COVID-19 and the Association between the Dose of Anticoagulants and Outcomes

Severe Coronavirus Disease 2019 (COVID-19) is characterized by hyperinflammation accompanied by a procoagulant state [1, 2, 3]. In addition to this state, critically ill patients with COVID-19 frequently have other factors, such as antecedent comorbidities, immobility, sedation, central venous lines and mechanical ventilation, that increase the risk of venous thromboembolism (VTE) and other thrombotic events. The studies that evaluated the epidemiology of VTE in COVID-19 patients have showed variable results, likely due to differences in methodology and case mix [4, 5, 6, 7, 8, 9]. A systematic review and meta-analysis in which 15 standard sources and COVID-19-specific sources were searched between January 1, 2020, and July 31, 2020 found that the pooled incidence was 17.0% (95% confidence interval [CI], 13.4-20.9%) for VTE, 12.1% (95% CI, 8.4-16.4%) for deep vein thrombosis (DVT) and 7.1% (95% CI, 5.3-9.1%) for pulmonary embolism (PE) [10]. The VTE incidence was higher among patients in the intensive care unit (ICU) (27.9% versus 7.1% in the ward) [10]. Whether COVID-19 has higher VTE rates than other severe infections is nevertheless unclear. In critically ill patients with sepsis, VTE incidence was as high as 37% in one prospective study where 80.5% of patients received pharmacologic prophylaxis and 19.5% sequential compression devices because of a contraindication for anticoagulants [11]. Predictors of thrombosis in COVID-19 patients include age, prolonged prothrombin time and partial thromboplastin time (PTT), higher D-dimer, and central venous lines [5, 7, 9].

Anticoagulation at different intensities (standard prophylactic, intermediate or therapeutic doses) have been advocated for patients with COVID-19. Higher intensity anticoagulation was suggested in patients with increased D-dimers and worsening hepatic, renal or respiratory function [6, 12, 13, 14] This suggestion was based on the high prevalence of hypercoagulability and high rates of VTE in patients with COVID- 19 with early observational studies showing benefits associated with the use of anticoagulants [15]. A retrospective study from China found lower 28‐day mortality with heparin use (mostly prophylactic enoxaparin) than non-use in patients with sepsis-induced coagulopathy score ≥ 4 (40.0% versus 64.2%, p=0.03), and when D‐dimer was > 6 fold the upper limit of normal (32.8% versus 52.4%, p=0.017) [15]. Another study found lower 30-day mortality among the 3627 patients who received prophylactic anticoagulation (14.3%; 95% CI, 13.115.5%) compared with the 697 patients who did not (18.7%; 95% CI, 15.1-22.9%) [16]. Higher enoxaparin doses (0.62±0.16 mg/kg) has been associated with a better thromboprophylactic action (hazard ratio, 0.2; p=0.04) compared with lower doses [17]. In another retrospective study, a multivariable regression analysis showed that intermediate- compared to prophylactic-dose anticoagulation was associated with a significantly lower cumulative incidence of in-hospital death (hazard ratio, 0.518; 95% CI, 0.308-0.872) among propensity score-matched patients (N=382) [18]. However, a large multicenter observational cohort study of 2809 critically ill patients with COVID-19 in the US found no benefit of anticoagulation at therapeutic doses initiated within 2 days of ICU admission compared with standard-dose thromboprophylaxis [19]. Multiple randomized trials have addressed different anticoagulants at different doses in COVID-19. So far, the published ones have not demonstrated significant benefits from anticoagulation at intermediate or therapeutic doses compared with standard doses in critical COVID-19 [20, 21, 22].

The objectives of this study were to determine the prevalence of clinically suspected and confirmed VTE in critically ill patients with COVID-19, describe the thromboprophylaxis and anticoagulation practices and outcomes, and identify predictors of symptomatic VTE and mortality.

The main findings of this study were the following: anticoagulant thromboprophylaxis was provided in 97% of patients mostly as low-molecular weight heparin (LMWH); clinically suspected and confirmed VTE cases were common at 13.2% while arterial thrombotic events were less common at 6.1%; vasopressor therapy was associated with increased risk for VTE while intermediate-dose of enoxaparin versus standard-dose UFH or enoxaparin with decreased risk of the composite outcome of VTE or hospital mortality.

Thrombosis is highly prevalent in COVID-19. However, studies in the ICU setting showed variable results. For instance, one study of 184 ICU patients with COVID-19 pneumonia in three Dutch hospitals found that the cumulative incidence of a composite outcome of symptomatic acute PE, DVT, ischemic stroke, myocardial infarction or systemic arterial embolism was 49% (95% CI, 41-57%) [5]. On the other hand, VTE rates were much lower in a study from Milan (8.3% of 48 patients in the ICU) [7]. In the randomized controlled trial that compared intermediate-dose (N=276) with standard-dose enoxaparin (N=286) in ICU patients with COVID-19, symptomatic VTE occurred in 3.3% and 3.5% of patients, respectively [23]. In a meta-analysis, the VTE incidence was higher among patients in the ICU (27.9% versus 7.1% in the ward) [10]. We found that symptomatic VTE was diagnosed in 13.2% and arterial thrombosis in 6.1%. Other symptomatic VTE may have been overlooked as patients may have been too sick to have diagnostic workup or they were empirically treated with therapeutic anticoagulation. One meta-analysis showed that VTE incidence was higher when assessed according to VTE screening was performed (33.1% versus 9.8% by clinical diagnosis) [10].

Predictors of thrombosis in COVID-19 include age, prolonged prothrombin time > 3 seconds or PTT > 5 seconds [5], high D-dimerC [9] and central lines [6, 7]. We found that vasopressor therapy was associated with increased VTE risk on multivariable regression analysis. Admission D-dimer and D-dimer/fibrinogen ratio had only fair ability to discriminate VTE from non-VTE cases (area under the ROC curve of approximately 0.7 for both). One study found that D-dimer > 2,600 ng/mL predicted VTE (area under the ROC curve, 0.760; 95% CI, 0.661-0.858) with a sensitivity of 89.7%, and specificity of 59.5%.(9) In a meta-analysis, the pooled sensitivity of D-dimer for VTE in COVID-19 was 90% (95% CI, 90-90%) with a specificity of 60% (95% CI, 60-60%).

The procoagulant state in COVID-19 has been associated with worse outcomes. In 183 consecutive patients with severe COVID-19 pneumonia in China, the mortality rate was 11.5% with 71.4% of non-survivors having overt-DIC (≥ 5 points on the International Society on Thrombosis and Hemostasis diagnostic criteria) compared with 0.6% of survivors [31]. Hence, anticoagulants have been recommended in the management of COVID-19 to mitigate the micro- and macro-thrombotic complications before organ failure occurs [32], and to improve outcomes including survival. However, the counter argument is that in acute thrombotic microangiopathy, which results from hyperinflammation and endothelial damage as occurs in COVID-19 [3], treatment should be directed against the underlying disease and anticoagulation has a limited role [32, 34]. Multiple observational studies evaluated anticoagulants in COVID-19 with mixed results [5, 15, 35]. A recently published randomized controlled trial compared intermediate-dose and standard-dose enoxaparin in COVID-19 patients requiring ICU admission and found no difference in the primary efficacy outcome (a composite of venous or arterial thrombosis, treatment with extracorporeal membrane oxygenation, or mortality within 30 days of enrolment) between the two groups [22]. A collaboration between 3 independent international trial platforms (ATTACC, REMAP-CAP, and ACTIV-4) evaluating the safety and efficacy of therapeutic-dose versus standard dose thromboprophylaxis in hospitalized patients with COVID-19 recently reported their results [21]. Among patients requiring ICU admission for organ support, such as invasive medical ventilation or vasopressors, enrolment was halted for futility in December 2020 by the data and safety monitoring boards after 1123 patients had been enrolled [21]. In a multicenter randomized controlled trial, therapeuticdose LMWH reduced the composite outcome of VTE, arterial thromboembolism or death (relative risk, 0.68; 95% CI, 0.49-0.96; p=0.03) compared with standard heparin thromboprophylaxis among inpatients with COVID-19 and elevated D-dimer levels (> 4 fold the upper limit of normal) [21]. However, this treatment effect was not seen in ICU patients [20]. In the current study, different doses of anticoagulants were used; almost one third of patients received intermediate-dose enoxaparin; and patients with VTE had lower mortality. One explanation is that sicker patients or those who had Do-Not-Resuscitate orders, who typically have worse outcomes, did not have workup for VTE. In the multivariable regression analysis, intermediatedose LMWH was associated with lower risk or VTE or hospital mortality compared with standard-dose UFH or LMWH. Heparin use may have contributed to improved outcomes, through its anticoagulation effect or through other anti-inflammatory or antiviral mechanisms. In critically ill patients, standard-dose (40 mg) enoxaparin may yield subtherapeutic levels of anti-factor Xa [36], likely secondary to impaired absorption from vasopressor-mediated vasoconstriction, subcutaneous edema, and obesity. Hence, higher anticoagulant doses may be needed to have effective antithrombotic activity [36]. A recent meta-analysis of 24 trials found that anticoagulants had no significant effect on mortality in sepsis in general, but with significant reduction in mortality (risk ratio, 0.72; 95% CI, 0.62-0.85) in the subgroup of patients with sepsis‐induced DIC [37].

We did not observe increased risk of major bleeding with intermediate-dose LMWH versus prophylactic-dose heparin. Data about bleeding with anticoagulation in COVID-19 have shown different results. Prophylactic anticoagulation was not associated with increased risk of bleeding that required transfusion (hazard ratio, 0.87; 95% CI, 0.71-1.05) in one study [16]. Intermediate-dose enoxaparin did lead to higher rate of major bleeding in a trial [22]. A retrospective study which evaluated 105 patients hospitalized for COVID-19 pneumonia (63.8% severe) treated with various doses of subcutaneous enoxaparin (40-100 mg/ day) found uncommon hemorrhagic incidents (0% fatal hemorrhage and 1.9% major bleeding); and only one thrombotic event (PE) [38]. When compared to younger patients, patients older than 85 years did not have an increased risk of bleeding or need for blood transfusion [38]. However, a meta-analysis found an incidence of 7.8% (95% CI, 2.6-15.3) for bleeding, and 3.9% (95% CI, 1.2-7.9) for major bleeding [10]. The highest pooled incidence estimate of bleeding was reported for patients receiving intermediate- or full-dose anticoagulation (21.4%) [10]. The occurrence of major bleeding has been associated with high mortality [19].

The study findings should be interpreted considering its strengths and limitations. All patients in the current study were critically ill and our study reflected the practices of anticoagulant use early in the COVID-19 pandemic when there was paucity of good quality evidence with inconsistent recommendations and guidelines. We evaluated the composite outcome of VTE or hospital mortality to address ascertainment bias. The limitations include the retrospective analysis from a single center and the use of only admission variables. We also did not collect data about mechanical thromboprophylaxis. However, only 9 patients did not receive any form of heparin and adjunctive use of intermittent pneumatic compression has not been shown to reduce VTE risk [39]. Additionally, our analysis focused on cases with venous, but not arterial thrombosis. Confounders may have led to the observed associations as discussed earlier.

In conclusion, we found that anticoagulant thromboprophylaxis was used for most critically ill patients with COVID-19 using different heparin doses. Clinically suspected and confirmed VTE was common occurring in 13.2% of patients, but arterial thrombosis was less frequent and was diagnosed in 6.1%. Intermediate-dose of enoxaparin versus standard-dose UFH or enoxaparin was associated with decreased risk of VTE and of the composite outcome of VTE or hospital mortality in patients with severe COVID-19.