Comparison of published guidelines for management of coagulopathy and thrombosis in critically ill patients with COVID 19: implications for clinical practice and future investigations

COVID-19-associated coagulopathy [1] is well established in COVID-19 patients and is marked by a state of profound inflammation with endothelial dysfunction, abnormal flow dynamics, platelet activation, and hypercoagulability resulting in higher rates of thrombotic complications such as deep venous thrombosis (DVT), pulmonary embolism (PE), and microvasculature thrombosis [2‐5]. Given the evidence that COVID-19 patients are at increased risk for thrombotic complications, a number of consensus guidelines and recommendations are now available to facilitate clinical decision-making. However, these guidelines are limited by their lack of data from randomized controlled clinical trials. This review provides an overview of major societal guidelines and recommendations that focus on prevention, monitoring, and treatment of COVID-19-associated coagulopathy (CAC). In addition, the similarities and differences between the guidelines are highlighted, and their implications for practice are explored. Lastly, proposed topics for future consideration and study are offered.

We review seven major societal recommendations and guidelines addressing the management of coagulopathy in COVID-19 patients: the Centers for Disease Control and Prevention (CDC) [6], International Society on Thrombosis and Haemostasis interim guidance (ISTH-IG) [7], American Society of Hematology (ASH) [8, 9], American College of Chest Physicians (ACCP) [10], Scientific and Standardization Committee of ISTH (SCC-ISTH) [11], Anticoagulation Forum (ACF) [12], and American College of Cardiology (ACC) [13]. These recommendations and guidelines were specifically selected for review because of their focus on anticoagulation in critically ill patients. Moreover, each sponsoring organization has prior expertise in developing evidence-based guidelines and providing health-related recommendations in the arena of venous thromboembolism (VTE). The aims of the major societal recommendations and guidelines vary. The ISTH interim guidance is unique in that it provides guidance on the risk stratification for COVID-19 patients on admission and the management of coagulopathy based on laboratory parameters. Alternatively, ACF, ASH, CDC, ACCP, ACC, and SCC-ISTH offer guidance on in-hospital and post-discharge VTE prevention, treatment, and monitoring.

Tables 1, 2, 3, 4, 5, 6, and 7 highlights six major societal recommendations and guidelines on the management of CAC focusing on several critical care issues: (1) laboratory testing for risk stratification and triage, (2) use of biomarkers to guide anticoagulation, (3) proposals for alterations of standard prophylactic VTE anticoagulation regimens for the prevention of thrombotic complications, (4) examination of available medications preferred for anticoagulation, (5) considerations for initiation of therapeutic anticoagulation, (6) indications for thrombolytic therapy, (7) decision-making regarding withholding anticoagulation treatment, (8) use of mechanical thromboprophylaxis, (9) monitoring of anticoagulation, (10) duration of therapeutic anticoagulation, (11) necessity of anticoagulation at discharge, and (12) treatment of active bleeding. We specifically focus on the recommendations and guidelines for critical care patients if mentioned in the major societal guidelines and recommendations; otherwise, we include the highest acuity patient population (i.e., hospitalized patients). A comparison of the similarities and differences between the guidelines and recommendations on each of the critical issues is provided.

Both ISTH-IG and ASH are the only societies that recommend monitoring D-dimer, PTT, platelet count, and fibrinogen levels for risk stratification of critical care treatment and consideration of experimental COVID-19 therapies based on changes in these parameters. ISTH-IG also recommends obtaining D-dimer, PTT, platelet count, and fibrinogen for all patients who present with COVID-19 infection to help guide which patients may require admission. Specifically, they provide an algorithm and the following thresholds for admission to the hospital: D-dimer markedly raised three- to fourfold, prothrombin time prolonged, platelet count < 100 × 10⁹, and fibrinogen < 2.0 g/L. Alternatively, the CDC states that there is a lack of prospective data demonstrating laboratory measures of coagulopathy for risk stratification in patients who are asymptomatic or with mild infection, and insufficient data to recommend for or against using such laboratory values to guide management. The SCC-ISTH mentions that the risk stratification algorithm for hospitalized patients requires further study while acknowledging the use of very elevated D-dimer (> 6 times upper limit of normal (ULN)) appears to be a consistent predictor of thrombotic events and poor overall prognosis in this patient population. ACC recommends that risk stratification of COVID-19 patients should be similar to other acutely ill medical patients without COVID-19 while stating that regular monitoring of platelet count, PT, D-dimer, and fibrinogen is important to diagnosing worsening coagulopathy and treatment of underlying conditions of DIC and bacterial superinfections is important.

None of the societies recommend daily monitoring of biomarkers as a means to guide intensity of anticoagulation management. The ACF states that biomarker thresholds such as D-dimer for guiding anticoagulation management should not be done outside the setting of a clinical trial. Similarly, the SCC-ISTH remarks that D-dimer levels should not be used to solely to guide anticoagulation regimens. Only ASH and ACC address the issue of anticoagulation in patients with lupus anticoagulant inhibitors. ASH recommends that those patients be treated with prophylactic anticoagulation regimens similar to other hospitalized patients, and ACC mentions that further investigation is required to determine the role of antiphospholipid antibodies in pathophysiology of CAC. ASH also mentions that thromboelastography (TEG) and rotational thromboelastometry (ROTEM) are currently under investigation for CAC and disseminated intravascular coagulopathy (DIC) and should not be used routinely to guide management. The ACC mentions that D-dimer > 2 times ULN may suggest that patient is at high risk for VTE and to consider extended prophylaxis (up to 45 days) in patients with low risk of bleeding.

None of the major societies recommend therapeutic anticoagulation for prevention of thrombotic complications. Only the ACF recommends increasing the intensity of anticoagulation in critically ill patients (enoxaparin 40 mg twice daily or heparin 7500 unit/dose three times daily) while they recommend standard dose anticoagulant prophylaxis for all other hospitalized non-critically ill patients. The SCC-ISTH mentions that intermediate LMWH dosing can be considered in high risk critically ill patients (50% of respondents) and may be considered in non-critically ill hospitalized patients (30% of respondents). The SCC-ISTH also specifically mentions that the anticoagulation regimens may be modified based on extremes of body weight (50% increase in dose if obese), severe thrombocytopenia, or worsening renal function. The ACC mentions that while there is insufficient data to consider routine intermediate or therapeutic dosing, a minority of their panel considers an intermediate intensity (31.6%; i.e., enoxaparin 1 mg/kg/day, enoxaparin 40 mg BID, UFH with target PTT 50–70) or therapeutic anticoagulation (5.2%) reasonable. The CDC recommends that any changes in anticoagulation regimens that are different than the current standard of care for patients without COVID-19 should only be administered in the setting of a clinical trial.

Many of the societies recommend using once daily prophylactic anticoagulants to reduce healthcare worker exposure. The ISTH-IG, ASH, and ACCP recommend using LMWH in critically ill patients. ASH particularly mentions that LMWH is preferred over UFH to reduce exposure unless the risk of bleeding outweighs the risk of thrombosis. The CDC, ACF, and SCC-ISTH recommend LMWH or UFH with the rationale provided by the CDC being due to their short half-lives, versatility in administration (IV or subcutaneously), and less drug-drug interactions compared to oral anticoagulants. The SCC does particularly mention that there are several advantages of LMWH over UFH including once vs twice or more injections and less heparin-induced thrombocytopenia. Similarly, the ACC mentions that enoxaparin 40 mg daily or similar LMWH regimen (i.e., dalteparin 5000 u daily) can be administered with consideration of subcutaneous heparin (5000 u twice to three times per day) in patients with renal dysfunction (i.e., creatinine clearance < 30 mL/min). They mention that once daily regimens of LMWH may be advantageous over UFH to reduce missed doses associated with worse outcomes, reduce healthcare worker exposure, and conserve personal protective equipment. None of the societies recommend routine use of direct oral anticoagulants because of potential drug-drug interactions, renal and hepatic function abnormalities that have implications on pharmacodynamics, longer half-lives, cost implications, and/or lack of reversal agents in some hospitals (specifically mentioned in CDC, SCC-ISTH, Anticoagulation Forum, and ACCP guidelines).

The ACF and ACCP generally prefer LMWH over UFH to decreases staff exposure and laboratory monitoring. The ACCP also prefers fondaparinux over UFH and mentions that UFH may be preferred in patients at high bleeding risk, with renal failure or needing imminent procedures. Similarly, the ACF recommends UFH over LMWH in patients with AKI or creatinine clearance < 15–30 mL/min. ASH mentions that LMWH or UFH is preferred over oral anticoagulants due to potential drug-drug interactions and their shorter half-lives. The SCC-ISTH and ISTH-IG do not mention preferred medication regimens for therapeutic anticoagulation. The CDC mentions that standard regimens for patients without COVID-19 patients should be used. The ACC mentions that the medication regimen is likely to change depending on comorbidities (i.e., renal or hepatic dysfunction, gastrointestinal function, thrombocytopenia). They give no particular preference to an agent but do mention that (1) parenteral anticoagulation (i.e., UFH) may be preferred in many ill patients given it may be withheld temporarily and has no known drug-drug interactions with COVID-19 therapies, (2) LMWH may be preferred in patients who are unlikely to need procedures as there are concerns with UFH regarding the time to achieve therapeutic PTT and increased exposure to healthcare workers, and (3) direct oral anticoagulants (DOAC) have advantages including lack of monitoring that are ideal for outpatient management but have risks in settings of organ dysfunction related to clinical deterioration and lack of timely reversal at some centers.

Most of the major societal guidelines and recommendations (ISTH-IG, ACF, CDC, and ASH) advise holding anticoagulation in patients who are actively bleeding or severely thrombocytopenic. ISTH-IG recommends holding LMWH prophylactic anticoagulation for patients with platelet count less than 25 × 10⁹/L. ASH mentions that therapeutic anticoagulation should be held if platelet count < 30–50 × 10⁹/L or fibrinogen < 1.0 g/L, and prophylactic anticoagulation should be held only if platelet count < 25 × 10⁹/L or fibrinogen < 0.5 g/L. The SCC-ISTH does not recommend a particular platelet level to hold anticoagulation but does report that 50% of their respondents report holding for a platelet count < 25 × 10⁹/L. While the CDC and ACF recommend holding prophylactic anticoagulation in patients who are severely thrombocytopenic, they also do not recommend a particular platelet count threshold. The ACCP does not make any specific recommendations regarding the holding of anticoagulation. The ACC mentions that in patients with moderate or severe COVID-19 on chronic therapeutic anticoagulation who develop suspected or confirmed DIC with overt bleeding, it is reasonable to consider the indication of anticoagulation and risk of bleeding for adjusting dose or discontinuation of anticoagulation. The majority of their authors recommend reducing the intensity of anticoagulation unless there is an exceedingly high risk of thrombosis. Of note, the ISTH-IG, ACF, and ASH mention that abnormal PT or PTT is not a contraindication to thromboprophylaxis.

The ACF, ASH, ACCP, ACC, and SCC-ISTH recommend or suggest mechanical thromboprophylaxis when pharmacological thromboprophylaxis is contraindicated. The ACF, ASH, ACC, and SCC-ISTH particularly mention pneumatic compression devices as examples of mechanical thromboprophylaxis. The ACF mentions that it is reasonable to consider both mechanical and pharmacological thromboprophylaxis in critically ill patients if no contraindication exists for each modality. The SCC-ISTH mentions that multimodal thromboprophylaxis with mechanical methods (i.e., intermittent pneumonic compression devices) should be considered (60% of respondents). Similarly, the majority of ACC panel members (55%) considered the use of both pharmacological thromboprophylaxis and intermittent pneumatic compression reasonable while acknowledging a lack of high-quality evidence. The ACCP suggests against the additional use of mechanical thromboprophylaxis in critically ill patients receiving pharmacological prophylaxis while mentioning that its addition is likely not to cause harm. The CDC and ISTH-IG do not mention any recommendations or suggestions regarding the use of mechanical thromboprophylaxis.

The CDC, ACCP, and ACF do not recommend therapeutic anticoagulation except in cases where there is a thromboembolic event or a high suspicion for a thromboembolic event when imaging is not possible. ASH recommends therapeutic anticoagulation only if there is a documented clinical indication (e.g., VTE, atrial fibrillation, or mechanical valve). ACC mentions that therapeutic anticoagulation is the key to VTE treatment but does not make a distinction between confirmed or suspected VTE. They do mention that hemodynamically stable patients with intermediate-low to intermediate-high risk PE should be managed initially with anticoagulation and close monitoring rather than fibrinolytics. While the SCC-ISTH recommends that therapeutic anticoagulation should not be considered for primary prevention until randomized trials are available, they do mention that therapeutic anticoagulation (i.e., changing from standard or intermediate intensity to therapeutic intensity) can be considered in patients without confirmed VTE but who have deteriorating pulmonary or ARDS. ASH also states that it is reasonable to consider increasing the intensity of the patient’s anticoagulation regimen (i.e., from standard to intermediate intensity, from intermediate to therapeutic intensity) or change anticoagulants in patients who have recurrent thrombosis of catheters and extracorporeal circuits (i.e., ECMO, CRRT) on prophylactic anticoagulation regimens. Similarly, the ACCP recommends increasing the dose of LMWH by 25–30% in patients with recurrent VTE despite being on therapeutic LMWH anticoagulation. The CDC specifically mentions that patients who have thrombosis of catheters or extracorporeal filters should be treated accordingly to standard institutional protocols for patients without COVID-19.

The ACF recommends against the use of thrombolytics outside of a clinical trial unless there is a clinical indication for which it is already approved such as ST elevation myocardial infarction, acute ischemic stroke, or massive PE with hemodynamic instability. The CDC also mentions that there is insufficient data to recommend for or against thrombolytic therapy outside the context of a clinical trial. However, the CDC specifically mentions that in pregnant patients in particular, thrombolytic therapy should only be used for acute PE with life-threatening hemodynamic instability due to risk for maternal hemorrhage. The ISTH-IG, ASH, and SCC-ISTH do not mention any recommendations or suggestions regarding the use of thrombolytic treatment.

The ACCP suggests the use of thrombolytic therapy in patients with massive PE (i.e., hypotension with systolic blood pressure < 90 mmHg or when there are signs of obstructive shock) and specifically recommends peripheral thrombolysis using a peripheral vein over catheter-directed thrombolysis. In addition, ACCP suggests that in patients in whom therapeutic anticoagulation fails and who continue to have evidence of cardiopulmonary compromise, thrombolytic therapy may be beneficial. Similarly, the ACC mentions that systemic fibrinolysis is indicated in patients with significant hemodynamically unstable high risk PE and catheter-based therapies be reserved for situations that are not amenable to systemic fibrinolysis. Furthermore, hemodynamically stable patients with intermediate-low or intermediate-high risk PE should receive anticoagulation and fibrinolysis or catheter-directed therapies should be considered in cases of further deterioration. They also report that a multidisciplinary PE response team (PERT) may be helpful for intermediate and high risk patients with VTE and that catheter-directed therapies should be limited to most critical situations given minimal data demonstrating a mortality benefit.

The ACF recommends at least a 3-month course of anticoagulation for patients who are started on anticoagulation for a presumed provoked thrombus from the inflammatory state of CAC but did not have imaging available for confirmation. The ACF also recommends that standard anticoagulation guidelines be used to determine length of anticoagulation beyond the initial 3-month period. The ACF mentions that an exception to the minimum 3-month course is in patients with recent bleeding or high risk of bleeding. In addition, the ACF comments that if confirmatory imaging cannot be obtained within a couple days of presumed diagnosis, it is possible that delayed imaging may show an absence of thrombus when one was present initially. Therefore, they recommend continuing empiric treatment for the minimum 3-month course. Similarly, the ACCP and SCC-ISTH recommend a minimum of 3 months of anticoagulation in those patients with confirmed PE or proximal DVT. The ISTH-IG, ASH, ACC, and CDC do not mention any recommendations or suggestions regarding the duration of therapeutic anticoagulation.

The CDC and ACF recommend against the routine use of prophylactic anticoagulation for patients discharged from the hospital. However, the CDC and ACF as well as ASH and SCC-ISTH mention that FDA-approved post-discharge prophylactic anticoagulation (PDPA) regimens (rivaroxaban and betrixaban) may be considered in patients with high risk for VTE and low risk for bleeding. Similarly, the ACCP mentions that PDPA can be considered in patients who are at low risk of bleeding if emerging data suggests that the risk of VTE outweighs the risk of bleeding. The ACF and SCC-ISTH both mention that enoxaparin in addition to betrixaban or rivaroxaban can be considered if PDPA is thought to be reasonable. The duration of anticoagulation recommended by ACF is based on the timing used in clinical trials which is 31–39 days for rivaroxaban, 35–42 days for betrixaban, and 6–14 days for enoxaparin. The SCC-ISTH recommends a duration of a minimum of 14 days and up to 30 days. ACC mentions that it is reasonable to consider extended prophylaxis with LMWH or DOACs for up to 45 days in patients at high risk for VTE (i.e., D-dimer > 2 times ULN, reduced mobility, active cancer) and low risk of bleeding. As an alternative, ASH mentions that aspirin can also be considered based on studies for VTE prophylaxis in low risk patients after orthopedic surgery. Aspirin is not mentioned in any of the other guidelines and recommendations with the exception of SCC-ISTH which reports that none of their respondents recommended aspirin for post-discharge thromboprophylaxis.

In the setting of bleeding, the ISTH-IG recommends transfusion to keep platelet count > 50 × 10⁹/L, fibrinogen > 1.5 g/L, and PT ratio < 1.5. ACC recommends platelet transfusion to maintain platelet count > 50 × 10⁹/L in DIC and active bleeding (> 20 × 10⁹/L in patients at high risk of bleeding or requiring invasive procedures), fresh frozen plasma (FFP) (15 to 25 mL/kg) in patients with active bleeding with either prolonged PT or PTT ratios (> 1.5 times normal) or decreased fibrinogen (< 1.5 g/L), and fibrinogen concentrate or cryoprecipitate in patients with persisting severe hypofibrinogenemia (< 1.5 g/L). Similarly, ASH recommends transfusion if platelets < 50 × 10⁹/L, INR < 1.8, and fibrinogen < 1.5 g/L. ASH also recommends that in the setting of severe coagulopathy and bleeding, 4F-PCC be considered instead of plasma so as to prevent significant volume overload. ACC mentions that prothrombin complex concentrate is recommended if FFP is not possible and tranexamic acid should not be used routinely in patients with COVID-19-associated DIC given the existing data. Moreover, ASH mentions that transfusions to simply to correct laboratory abnormalities in the absence of bleeding may worsen disseminated thrombosis and deplete blood product resources without any evidence of improving clinical outcomes. Similarly, ACC mentions that correction of coagulopathy in unselected patients without significant bleeding is not recommended.

These societal guidelines are based on consensus of expert opinion with limited evidence available in the form of robust clinical trials. Consequently, it is not surprising there are clinical issues that are either not addressed by the guidelines or have varying opinions. For instance, practitioners who deem a critically ill COVID-19 patient at high risk for thrombotic complication (e.g., trauma, cancer, antiphospholipid antibody) may decide to use a higher intensity prophylactic anticoagulation regimen consistent with the ACF guidelines rather than routine anticoagulation prophylactic regimen recommended by other societies. The major societal guidelines and recommendations share significant similarities, but also have some discrepancies. We recommend that providers manage their patients in the framework of these major societal guidelines, and where discrepancies do exist, decisions be made based on the practitioner’s experience and their understanding of a patient’s medical history, clinical course, and perceived risk.

Based on this review and comparison of the societal guidelines, there are a number of unaddressed issues that warrant further investigation and incorporation into updated guidelines. Further studies of COVID-19 pathophysiology are required to elucidate the mechanisms of coagulation disruption in order to identify potential pharmacologic targets and new biomarkers for risk stratification. Large randomized control trials are needed to investigate the optimal prophylactic anticoagulation regimen, the utility of intermediate and therapeutic anticoagulation, and the use of fibrinolytic therapy in COVID-19 patients. There are several clinical trials currently in the ClinicalTrials.gov database focused on optimizing prophylactic anticoagulation regimens (8 total) (Table 8), therapeutic anticoagulation (5 total) (Table 9), and fibrinolytic therapy (2 total) (Table 10) in COVID-19 patients [14], and we eagerly await those results. Further guidance from prospective studies is required to guide the clinical significance of biomarkers (e.g., D-dimer, antiphospholipid, and lupus anticoagulant antibodies) in risk stratification and treatment of CAC.

There are significant similarities and differences surrounding the major societal recommendations and guidelines regarding risk stratification and management of COVID-19-associated coagulopathy in regard to use of laboratory makers, prevention, treatment, and monitoring of VTE, and post-discharge thromboprophylaxis. In addition, there are a number of unanswered questions that warrant further investigation. It is incumbent on the clinicians to remain updated on emerging literature in this area in order to optimize the care of critically ill patients with COVID-19.