Discussion
Amid the COVID-19 pandemic, treatment guidelines for physicians remain inconsistent. An important aspect of COVID-19 progression is the development of coagulopathy, which necessitates the adoption of proper antithrombosis treatment plans that could be administered to patients suffering from mild and moderate to severe diseases. With current guidelines emphasising the importance of thromboprophylaxis in both high-risk and low-risk patients with COVID-19, it has become imperative to assess the efficacy and proper prescription of such treatments.33–35 Overall, the use of anticoagulants has shown an increased survival and reduction in thromboembolic events in patients undergoing treatment when compared with untreated controls. Heparin, whether unfractionated heparin (UFH) or LMWH, was the most common choice of anticoagulant for the management of COVID-19-associated thrombotic events. On the whole, combined antiplatelet and anticoagulant therapy increased patient survival and alleviated respiratory symptoms secondary to PE, which indicates an advantage for combination therapy over antiplatelet or anticoagulant therapy alone.25
Prophylactic versus therapeutic anticoagulants
Clinically, when prescribing anticoagulant regimens, physicians attempt to achieve a balance between healthy blood coagulation and bleeding risk. This balance is contested in COVID-19 as multiple patients develop a hypercoagulable state despite anticoagulation use. The available data indicate that anticoagulation presents patients with a clear advantage over absence of anticoagulation, regardless of minor bleeding risk that accompanies the treatment.20–24 27 28 Thus far, early results of randomised controlled clinical trials on ideal anticoagulant regimen across disease severity in COVID-19 19 have been generated. According to press releases and preprint publications by the multiplatform collaborative clinical trial (ACTIV-4a (NCT04505774), REMAP-CAP (NCT02735707), ATTACC (NCT04372589)), the preliminary results seem to be conflicting across diseases severity.36 37 In moderately ill patients, therapeutic anticoagulation seems to be the preferred regimen.36. Although there was no difference in hospital survival, standard prophylactic dosing is preferred due to more organ support-free days defined as a composite of death, the number of days free of respiratory organ support and cardiovascular organ support, as well as less major bleeding.37 Similar findings in critically ill patients have also been published in the INSPIRATION trial that showed no benefit of intermediate doses of anticoagulation compared with prophylactic doses in terms of the composite endpoint of venous or arterial thrombosis, treatment with extracorporeal membrane oxygenation or mortality within 30 days. Less major bleeding was also noted in the prophylactic dose arm.38 Moreover, therapeutic dose anticoagulation seems to reduce endothelial cell lesion (p=0.02), which could also reduce the thromboembolic risk of COVID-19, suggesting another therapeutic target for anticoagulants.39
Bleeding events
Current evidence on the risk of increased bleeding events in patients with COVID-19 receiving therapeutic or prophylactic anticoagulant treatment remains inconclusive. In this context, some data suggest that bleeding events increase in patients placed on antithrombotic therapy especially when comparing high-dose therapies to lower doses (p<0.003).20 22 40 However, other studies indicated lower bleeding incidents in patients treated with therapeutic or prophylactic dose anticoagulant when compared with non-users.24 27 This necessitates further assessment to ascertain whether bleeding risk due to anticoagulant use outweighs the benefits.
C reactive protein
Knowing that C reactive proteins play an important role in activating the blood complement system and that the latter in turn can mediate coagulation increased C reactive protein levels reported in patients with COVID-19 might reflect increased levels of systemic coagulopathy, along with increased inflammatory responses that could be attributed to endothelial dysfunction.20 28 41 42 Complement activation has been well reported in patients with COVID-19, but its root causes remain to be properly delineated.43–45
Heparin resistance
A possible explanation for the failure of heparin to properly inhibit or reduce coagulation in patients with COVID-19 could be the development of heparin resistance in a select group of patients suffering from aggravated disease status. Patients prone to heparin resistance commonly present with a deficiency in antithrombin III, increased fibrinogen and D-dimer level.46 47 Both fibrinogen levels and D-dimer levels have been shown to be elevated in patients with COVID-19, especially in severe cases that develop thromboembolic complications and only respond to higher doses of antithrombotic treatment.9 22–24 29 30 48 49 Additionally, heparin resistance has been reported in patients with COVID-19.50 51 and could therefore explain the failure of antithrombotic therapies in some patients in reducing coagulopathy. The administration of antithrombin III proved useful in abolishing heparin resistance in patients who underwent cardiac surgery and could therefore possibly benefit patients with COVID-19.47 The impact of heparin resistance should be assessed in patients with COVID-19 to identify whether it only affects a few isolated cases or it is actually a key player in the COVID-19-induced coagulopathy.
D-dimers
When a blood clot is dissolved, D-dimers are disseminated in the blood as a biproduct of coagulated platelet breakdown.52 Given that coagulopathy is not uncommon in patients with COVID-19, it is expected that D-dimer levels might play an important role in diagnosis or monitoring of patients. Current studies observed elevated D-dimer levels in patients with COVID-19, ranging from a 2.5-fold to a 6.0-fold increase, with these increased levels predicting mortality and the development of thromboembolic events in patients.22–24 29 30 48 Yet, there is no consensus on a specific cut-off level which can accurately predict disease course, mortality or response to antithrombotic therapy. Mouhat et al suggested the use of 2590 ng/mL as threshold level which corresponds to a ≈5-fold increase in normal D-dimer level.53 This 5-fold increase was associated with a 17-fold increase in the incidence of PE among patients (OR 16.9, 95% CI 6.3 to 45.0, p<0.001).53 Similarly, Ventura-Diaz et al reported that a 5.5-fold increase in D-dimer level could predict the occurrence of PE with 81% sensitivity and 59% specificity (p<0.001),54 while Ooi et al examined a larger sample of 974 patients and reported that a 4.5-fold increase in D-dimer level would predict the occurrence of PE with a 72% sensitivity and a 74% specificity.48 Moreover, patients presenting with increased D-dimer levels equivalent to more than fourfolds the normal level exhibit a higher mortality risk when compared with their counterparts, who present with lower levels (OR 10.17, 95% CI 1.10 to 94.38, p=0.041).55 These findings were corroborated by Zhang et al, who had similar findings and showed that D-dimer levels ≥4 times the normal level could predict mortality with 92.3% sensitivity and 83.3%. specificity (HR 51.5, 95% CI 12.9 to 206.7, p<0.001).56 Further research on possible explanations for anticoagulant treatment failure in some patient subgroups is warranted.24 Collectively, these preliminary findings indicate the possible use of D-dimers to guide treatment regimens and to establish different patient subgroups that should receive personalised treatment. However, the use of D-dimer in clinical settings as a stand-alone marker for COVID-19 thrombosis does not seem practical, as a clear cut-off value cannot be easily established since sensitivity of the test is compromised with age.57 58
Arterial thrombotic events
Apart from the venous thrombotic and thromboembolic events, there are also reports of arterial thrombosis including stroke, myocardial infarction, acute limb ischemia, aortic thrombosis and splenic infarcts.59–62 Microvascular thrombosis also is present in COVID-19 disease. Autopsies done on patients who died from COVID-19 infection demonstrated microvascular thrombosis in the lungs. The mechanism of development of this entity is unclear and is thought to be multifactorial.63–66
Guideline recommendations
Guidance on the treatment of COVID-19-associated hypercoagulability has been provided by several international organisations.34 67–69 For outpatients with COVID-19 infection, it is not recommended to initiate anticoagulant and antiplatelet therapy for prevention of venous thromboembolism or arterial thrombosis unless other indications are present.70 However, there is general agreement that thromboprophylaxis should be administered to all patients hospitalised with COVID-19 infection with the prophylactic dose anticoagulation being the backbone of the antithrombotic therapy.34 67–70 Nevertheless, there exist some variations concerning the recommended dosages, duration of treatment and optimal agent. Some societies recommend intermediate dose anticoagulation (eg, enoxaparin 0.5 mg/kg two times per day or 40 mg two times per day or intravenous UFH to achieve antifactor Xa level of 0.3–0.7 units/mL) in a specific subgroup of patients considered at high risk of thrombotic events, whereas others recommend against it.34 67 71 In addition, extended thromboprophylaxis even after patient discharge can be considered in patients at high thrombotic and low bleeding risk.34 67 70 The optimal anticoagulant agent is yet to be determined. Currently, most societies recommend the use of LMWH and UFH as first-line agents for inpatient management of these patients, whereas others recommend the use fondaparinux in addition to the previously mentioned agents.34 71 However, the use of antiplatelet alone for thromboprophylaxis is not recommended.34 When pharmacological anticoagulation is contraindicated, it is recommended to use mechanical thromboprophylaxis.34
Commentary
Coagulation is a highly well-organised procedure that involves the interaction of endothelial cells, platelets and coagulation factors.72 Under physiological conditions, platelets circulate without adhering to intact and inactive endothelia. COVID-19 infection was shown to be highly associated with endothelial dysfunction.73–75 Thus, when endothelial activation and dysfunction occur, disruption of vascular integrity and endothelial cell apoptosis results in exposure of the thrombogenic basement membrane and activation of the clotting cascade by displaying von Willebrand factor, P-selectin and fibrinogen, onto which activated platelets bind and play their primary role in thrombosis.22 76 77 Those activated platelets also produce VEGF, which induces endothelial cells to express the tissue factor, the main activator of the coagulation cascade. The coagulation pathway is also activated when the blood vessels are injured. The transfer of microthrombi into the systemic circulation increases the risk of development of DVT.78 Even though the underlying mechanisms of thrombosis in COVID-19 are incompletely understood, the major contributors of damage are caused by endothelial injury and hypercoagulability.79 80 So far, anticoagulation was associated with better outcomes in patients with COVID-19 with many societies recommending its use as part of the treatment in most patients with COVID-19.20–22 However, these results were not consistent among all studies, with some reporting no added value24 25 and others reporting the occurrence of thrombotic events on anticoagulation.50 51 Recently, increased interest in the use antiplatelets has surfaced for treating and preventing thrombotic events wherein there has been some benefit associated with aspirin use.31 The dual effect of antiplatelet and anticoagulant therapy on COVID-19-induced platelet thrombosis and hypercoagulability, respectively, may result in a synergistic and superior outcome than the use of either medication alone; especially considering that thrombotic manifestations of COVID-19 are heterogenous since they arise from several pathological mechanisms occurring simultaneously. Using a drug combination is not unheard of as it has already been studied in patients with acute coronary syndrome.81
High-quality data to guide the optimal strategy for antithrombotic therapy in patients with COVID-19 are unavailable; enrolment in randomised controlled trials (RCTs) is encouraged and needed to refine guidelines and optimise patient care.82 83 Studies are needed to define the best anticoagulation agent for patients with COVID-19 as so far there is no direct high-certainty evidence which compares different anticoagulants. In addition, examining the role of antiplatelets alone or the combination of anticoagulant and antiplatelet therapies in these patients is highly needed, especially since COVID-19 hypercoagulability is a result of endothelial dysfunction in addition to the activation of the coagulation cascade.22 76–80 To add, the optimal duration for anticoagulation should be determined, especially since COVID-19 has been associated with thrombotic events even after recovery.84 85 Furthermore, the ideal anticoagulation dosage is a matter of huge debate and is yet to be determined by well-conducted studies . Also, studies focusing on antithrombotic strategies in special populations like elderly, pregnant women, children and different ethnicities are also needed.
Limitations
Due to paucity of research on the use of antiplatelet and antithrombotic therapeutics in the treatment of patients with COVID-19, the recommendations made in this review are not conclusive. Moreover, the lack of comparable studies in terms of study population, clinical intervention and disease severity further restricts drawing generalisable conclusions regarding therapeutic efficacy and safety. RCTs remain the gold standard in clinical research in establishing causation; however, the current lack of such studies reduces this review’s ability in drawing conclusions regarding safety, tolerability and efficacy of the proposed treatments in patients with COVID-19. Evidence synthesised from retrospective and prospective cohorts should therefore be analysed with caution as validity and interpretation of such studies are less clear, lack temporality and are prone to some degree of bias. In addition to observational studies being subject to information bias and reverse causation, one major bias encountered was selection bias, since patients who were treated with antithrombotic agents most likely had a higher mortality risk, due to critical disease status, as compared with those who were left untreated. Additionally, another major issue could be confounding bias, since available studies lacked multivariate techniques such as restriction, matching and stratification, which could have potentially exposed the presence of confounders such as the effect of other administered treatments or COVID-19 variants. Furthermore, this rapid review could be subject to study selection bias as search terms were not tailored to target specific types of antiplatelet or antithrombotic therapies. It is worthwhile to mention that rapid reviews are perishable quickly as new evidence will emerge continuously and its synthesis will require updating.