Should Patients With COVID-19 Receive Post-Discharge Thromboprophylaxis? Short Answers to Frequently Asked Questions

Thrombosis is an important complication in acute COVID-19 patients, because of a high level of inflammation and coagulation abnormalities with an underlying hypercoagulable state [1,2]. The viral infection injuries promote vascular inflammation and endothelial dysfunction. As a result, the level of the von Willebrand factor is increased and, in conjunction with the activation of the receptors and tissue factors involved in the coagulation pathway, leads to the formation of fibrin clots. The levels of D-dimers are also increased and contribute to platelet activation and thrombin generation. Therefore, venous or arterial thrombosis are common features of COVID-19 infection [3,4,5,6]. A high incidence of VTE events was reported in up to 8% (without screening ultrasound) and 45.6% (with screening ultrasound) of the patients, often despite parenteral prophylactic or therapeutic anticoagulation [7,8]. Those with severe COVID-19 disease and supplementary risk factors (e.g., age, gender, cancer, history of VTE, obesity, comorbidities, intensive care unit stay), have the highest risk of VTE [7,8]. The PE was the most frequent thrombotic complication, as part of VTE [9,10]. The risk of COVID-19-associated VTE continues after hospital discharge, but the cumulative incidence of such events has not been clearly determined through the available retrospective and observational studies [11,12,13,14,15,16,17,18,19,20,21]. Current guidelines do not recommend extended anticoagulation for post-discharge patients who do not have suspected or confirmed VTE or other indications for anticoagulation (Table 1) [22,23,24,25,26,27]. The efficacy of this recommendation, however, which is based on small and medium sample size (predominantly retrospective studies) and not on randomized controlled trials (RCT), is low. The purpose of our commentary is to briefly discuss the dilemmas and the options for extended thromboprophylaxis in COVID-19 patients. We do not intend to provide a systematic review, and readers seeking more comprehensive information should consult the updated current guidelines on the use of anticoagulation for thromboprophylaxis in patients with COVID-19.

The belief that COVID-19 patients had a higher incidence of post-discharge VTE events than other acutely ill patients has not been confirmed by the available data [12,13,14,15,16,17,18,19]. The rate of VTE events was 2.0% within 30 days after discharge in a recent retrospective study on 447 hospitalized patients, and 0.7% for both DVT and PE in the only prospective study that systematically screened all patients for DVT and PE [18,28]. A meta-analysis of 11 studies (18,949 COVID-19 patients) reporting the incidence of post-discharge VTE (symptomatic and asymptomatic at a maximum of 180 days post discharge) showed a pooled VTE incidence of 1.8%, a DVT incidence of 0.9%, and an incidence of 1.5% for PE [29]. The incidence of post-discharge VTE events in COVID-19 patients was reported at 30 to 42 days in most of the studies and it is similar to the 1–4% incidence risk found for other medical patients [19,30]. Ongoing prospective studies such as COVID-PREVENT (NCT04416048) and PREVENT-HD (NCT04508023), however, will help us to a better understanding of the VTE and bleeding outcomes in COVID-19 patients after hospital discharge [31,32]. Table 1 summarizes the data about incidence of post-discharge VTE.

Interesting data have also come from the ACTIV–4b trial (effect of antithrombotic therapy on clinical outcomes in outpatients with clinically stable symptomatic COVID-19) which was an RCT that compared aspirin, or prophylactic-dose apixaban (2.5 mg orally twice daily) or therapeutic-dose apixaban (5 mg orally twice daily), versus placebo in clinically stable outpatients with COVID-19, who were older than 40 years. The primary composite objective (the reduction at 45 days of major adverse thrombotic or thromboembolic cardiopulmonary events) was not modified by the aspirin or apixaban treatment. The study was stopped after enrollment of only 9% of participants because of a lower event rate than anticipated [33]. Therefore, the incidence of thrombosis seems to be low in stable patients with COVID-19 and no supplementary risk factors. Different clinical and biological risk factors (increased levels of D-dimers, fibrinogen, or platelets) have been found to be associated with VTE after COVID-19 in high-risk patients or those with comorbidities such as those with advanced age, male sex, intensive care unit (ICU) hospitalization in case of respiratory distress, cardiovascular and chronic kidney diseases, recent history of VTE, trauma, immobilization, or surgery [7,18,19,28,29,34,35,36]. Hence, thromboprophylaxis after discharge should be discussed with those COVID-19 patients. In one prospective study of 146 patients, a post-discharge thromboprophylaxis with LWMH was prescribed in 28% of patients but was used more frequently after an ICU stay (61%) and in those with the highest levels of D-dimer and C-reactive protein during hospitalization [18].

On admission, all hospitalized patients should get a complete blood count, prothrombin time (PT), active partial thromboplastin time (aPTT), fibrinogen, and D-dimer dosage. A normal or prolonged PT and aPTT, normal or elevated platelets, elevated von Willebrand factor antigen, and especially elevated D-dimer levels are characteristic for the COVID-19 disease [37]. Different reports have demonstrated that high D-dimer levels are markers of disease severity, thrombotic predisposition, or a pejorative prognosis [38]. However, the D-dimer value alone, without considering the patient's clinical status, is not sufficient for deciding on extended thromboprophylaxis. The modified International Medical Prevention Registry on Venous Thromboembolism (IMPROVE VTE) and elevated D-dimer level score (IMPROVE DD) have been empirically used to select the COVID-19 patients with an increased risk for VTE: age >60 years, thrombophilia, immobilized ≥7 days immediately prior to and during hospital admission, active cancer, intensive care hospitalization, lower-limb paralysis, plus D-dimer values of 2x the upper limit of normal [39]. The results of only two RCTs on post-discharge thromboprophylaxis have been published until now. The first was the ACTION trial (therapeutic versus prophylactic anticoagulation for patients admitted to hospital with COVID-19 and elevated D-dimer concentration), which did not show any benefits for using a DOAC (rivaroxaban 20 mg daily for 30 days post discharge), compared to placebo [40]. The second was the more recently published MICHELLE trial (rivaroxaban versus no anticoagulation for post-discharge thromboprophylaxis after hospitalization for COVID-19: an open-label, multicenter, RCT), which compared rivaroxaban (10 mg daily) versus placebo in a population of high risk patients for VTE, with elevated IMPROVE DD and IMPROVE VTE scores (≥4 points independent of the D-dimer level at discharge, or 2–3 points with a D-dimer >500 ng/mL). At 35 days, the composite endpoint of thrombotic events (arterial and VTE) and cardiovascular death was significantly lower in the rivaroxaban group (RR 33%, 95% CI 0.12–0.90; p = 0.02) [41]. The favorable results of this trial came out after the main guidelines published recommendations against the routine post-discharge thromboprophylaxis for patients who do not have suspected or confirmed VTE or another indication for anticoagulation. However, the guidelines recommend an individual evaluation of the thrombotic and bleeding risks to determine who should have extended thromboprophylaxis. The COVID-19 patients with risk factors for thrombosis (obesity, previous VTE or arterial events, reduced mobility, advanced age, ICU admission, cancer, thrombophilia) or those with IMPROVE-DD scores merit continued thromboprophylaxis Table 1 [22,23,24,25,26,27]. The ongoing trials have the potential to add valuable data about the safety and effectiveness of post-discharge thromboprophylaxis in COVID-19 patients [42,43,44,45].

The anticoagulants of choice for patients with COVID-19 during hospitalization are unfractionated heparin (UFH) or LMWH [6,46,47]. Until now we haven’t had specific evidence to make recommendations about the duration and the type of post-discharge anticoagulation to be used. Some patients (those with mechanical valve prostheses, advanced chronic kidney disease, and antiphospholipid syndrome), however, must remain on vitamin K antagonists (VKA) [48]. Other patients could take LWMH or DOACs. These options were empirically recommended by the current guidelines and include an LMWH (enoxaparin, tinzaparin, dalteparin) or a DOAC (apixaban, rivaroxaban, betrixaban) for 14 to 45 days, for patients with high risk of thrombosis and low risk of bleeding (Table 2) [22,23,24,25,26,27]. Prophylactic LMWH (adjusted for weight and renal function) or low doses of DOACs (rivaroxaban 10 mg daily or apixaban 2.5 mg twice per day), already used in published or ongoing trials, may be given with the aim of minimizing the bleeding risk [22,23,24,25,26,27,40,41,42,43,44,45]. In fact, the open-label design of the MICHELLE trial, the single positive published RCT that tested extended thromboprophylaxis with a DOAC (rivaroxaban, 10 mg daily) against placebo for 35 days after hospital discharge, has a potential risk of bias and needs confirmation by the ongoing trials [42–43]. Overall, the existing studies and recommendations about medication type and dosing of thromboprophylaxis in COVID-19 after hospital discharge share the same limitations, such as the retrospective nature of the study and the limited population size. Thus, the choice of whether to use anticoagulation and duration of treatment depends on clinical judgment and should be decided after a discussion between the patient and the clinician. DOACs should be prescribed carefully and only in the absence of interactions with antiviral therapies, which are common among patients with severe COVID-19 [49,50]. Also, the use of the DOACs for thromboprophylaxis in COVID-19 patients after hospital discharge is off label, and therefore the patient's consent needs to be obtained.

The COVID-19 disease is associated with higher rates of thrombosis during hospital stays, and patients must be evaluated for ongoing risk of VTE before discharge. The incidence of post-discharge VTE seems to be less than 2%, and therefore the guidelines advise against routine extended thromboprophylaxis for patients who do not have suspected or confirmed VTE or another indication for anticoagulation. The residual risk for VTE events, however, even if not high, is present, and in selected cases the anticoagulation must continue for a period of time after hospital discharge. Clinical judgment with an individual assessment of the VTE and bleeding risks and use of a score like IMPROVE-DD helps to identify those patients who could benefit from post-discharge prophylactic anticoagulation. The type of agent and the duration of thromboprophylaxis is still being debated, while prophylactic LMVH or DOACs at low doses have been used in most patients. Until the guidelines can provide strong recommendations for use of the extended prophylactic anticoagulation after COVID-19, it is mandatory to inform the patients and their families about the benefits and risks of different types of extended thromboprophylaxis.