After numerous observational studies demonstrated increased risk for thrombotic events in hospitalized patients with COVID-19, there was a surge of initial speculation for optimal anticoagulation strategies. Many providers when faced with limited but concerning data adopted nonstandard, higher intensity prophylactic anticoagulation regimens in an attempt to limit thrombosis and improve patient outcomes. To better address these multiple hypotheses, many randomized controlled trials were launched (Table), and initial speculation is now finally transitioning to high-quality, evidence-based medicine. The number of randomized clinical trials related to anticoagulation published in the past year has been truly remarkable; the challenge now is applying data from the positive trials while minimizing the potential risks of escalated doses of anticoagulation.
Selected Approved Regimens for Early Relapse
| Study . | Intervention . | Inclusion (hospitalized, COVID-19+) . | Primary Outcome . | Major Bleeding (ISTH) . | VTE/ATE . | Mortality . |
|---|---|---|---|---|---|---|
| ATTACC, ACTIV-4a, REMAP-CAP2 | Therapeutic UFH or LMWH up to 14 days vs. standard AC prophylaxis | Supplemental oxygen by nasal canula | Organ support–free days (≤ 21) or 1.27* 98.6% probability of superiority* | 1.9% vs. 0.9% | 1.1% vs. 2.1% | Death in hospital 7.3% vs. 8.2% |
| HEP-COVID3 | Therapeutic LMWH until discharge vs. standard AC prophylaxis | Supplemental oxygen by nasal canula and D-dimer > 4 × ULN or SIC ≥ 4 | Non-ICU subgroup VTE, ATE, death at 30 days 16.7 vs. 36.1% | Non-ICU Subgroup 2.4% vs. 2.3% | All patients (ultrasound at day 10) 10.9% vs. 29.0% | All patients 19.4% vs. 25.0% |
| ACTION7 | Therapeutic rivaroxaban or LMWH for 30 days vs. standard AC prophylaxis | D-dimer >ULN | Death, duration hospitalization, supplemental oxygen use (wins) 34.8% vs. 41.3% OR, 0.86 | 3% vs. 1 % | 7% vs. 10% | 11% vs. 8% |
| RAPID4 | Therapeutic LMWH or UFH up to 28 days or discharge vs. standard AC prophylaxis | Non-ICU and D-dimer >ULN and oxygen saturation < 94% on room air or D-dimer >2 × ULN | ICU admission, ventilation, death 16.2% vs. 21.9% OR, 0.69 | 0.9% vs. 1.7% | 0.9% vs. 3.4% | 1.8% vs. 7.6% OR, 0.22 (95% CI, 0.07-0.65)* |
| Study . | Intervention . | Inclusion (hospitalized, COVID-19+) . | Primary Outcome . | Major Bleeding (ISTH) . | VTE/ATE . | Mortality . |
|---|---|---|---|---|---|---|
| ATTACC, ACTIV-4a, REMAP-CAP2 | Therapeutic UFH or LMWH up to 14 days vs. standard AC prophylaxis | Supplemental oxygen by nasal canula | Organ support–free days (≤ 21) or 1.27* 98.6% probability of superiority* | 1.9% vs. 0.9% | 1.1% vs. 2.1% | Death in hospital 7.3% vs. 8.2% |
| HEP-COVID3 | Therapeutic LMWH until discharge vs. standard AC prophylaxis | Supplemental oxygen by nasal canula and D-dimer > 4 × ULN or SIC ≥ 4 | Non-ICU subgroup VTE, ATE, death at 30 days 16.7 vs. 36.1% | Non-ICU Subgroup 2.4% vs. 2.3% | All patients (ultrasound at day 10) 10.9% vs. 29.0% | All patients 19.4% vs. 25.0% |
| ACTION7 | Therapeutic rivaroxaban or LMWH for 30 days vs. standard AC prophylaxis | D-dimer >ULN | Death, duration hospitalization, supplemental oxygen use (wins) 34.8% vs. 41.3% OR, 0.86 | 3% vs. 1 % | 7% vs. 10% | 11% vs. 8% |
| RAPID4 | Therapeutic LMWH or UFH up to 28 days or discharge vs. standard AC prophylaxis | Non-ICU and D-dimer >ULN and oxygen saturation < 94% on room air or D-dimer >2 × ULN | ICU admission, ventilation, death 16.2% vs. 21.9% OR, 0.69 | 0.9% vs. 1.7% | 0.9% vs. 3.4% | 1.8% vs. 7.6% OR, 0.22 (95% CI, 0.07-0.65)* |
Abbreviations: AC, anticoagulation; ATE, arterial thromboembolism), ICU, intensive care unit; LMWH, low molecular weight heparin; NC, nasal canula; OR, odds ratio; SIC, sepsis-induced coagulopathy score; UFH, unfractionated heparin; ULN, upper limit of normal; VTE, venous thromboembolism.
Statistically significant results.
In a multinational platform collaboration, the ATTACC (Antithrombotic Therapy to Ameliorate Complications of COVID-19), ACTIV-4a (A Multicenter, Adaptive, Randomized Controlled Platform Trial of the Safety and Efficacy of Antithrombotic Strategies in Hospitalized Adults with COVID-19), and REMAP-CAP (Randomized, Embedded, Multifactorial Adaptive Platform Trial for Community-Acquired Pneumonia) groups performed a randomized controlled trial enrolling hospitalized patients with COVID-19.1,2 Patients were stratified into “severe” and “moderate” illness based on need for organ support. Moderately ill patients were those not requiring respiratory or cardiovascular support, defined as use of high-flow nasal canula or noninvasive or invasive mechanical ventilation. Researchers randomized 2,219 participants to therapeutic anticoagulation with low molecular weight or unfractionated heparin or to usual care/standard anticoagulant prophylaxis for 14 days or until recovery.2 The study used a unique primary outcome of organ-support-free days on an ordinal scale through day 21. The trial was ended after meeting prespecified stopping criteria for superiority of the therapeutic anticoagulation group. The bottom line: There is a 98.6 percent chance that therapeutic anticoagulation increased organ-support-free days (median adjusted odds ratio [OR], 1.27; 95% CI, 1.03-1.58). Survival until hospital discharge as a secondary outcome was not different (92.7% vs. 91.8%; OR 1.21; 95% CI, 0.87-1.68). A numerical reduction in major thrombotic outcomes was seen with therapeutic anticoagulation (1.1% vs. 2.1%) but was mirrored by a similar absolute increase in major bleeding (1.9% vs. 0.9%).
The HEP-COVID study also demonstrated positive results, with a significant reduction in the outcome of venous and arterial thrombosis and death with therapeutic anticoagulation. This study selected a high-risk group of hospitalized patients with d-dimer levels four times the upper limit of normal who required supplemental oxygen but not more advanced respiratory support. Thrombotic events (10.9 vs. 29.0%; p<0.001) and death (19.4 vs. 25.0%; p=0.28) occurred more frequently in the HEP-COVID study in therapeutic versus standard prophylactic anticoagulation, respectively. Data from the RAPID trial also supports the use of therapeutic anticoagulation in 228 moderately ill hospitalized patients with elevated d-dimers. Although a reduction in its primary composite outcome was not observed, the data did demonstrate a reduction in mortality (1.8% vs. 7.6%; OR, 0.22; 95% CI, 0.07-0.65) at 28 days. When outcome data from RAPID were meta-analyzed with the ATTACC, ACTIV-4a, REMAP-CAP study of moderately ill patients, combined outcomes of death and invasive mechanical ventilation (OR, 0.77; 95% CI, 0.62-0.95) and death and organ support (OR, 0.77; 95% CI, 0.63-0.93), as well as major thrombotic events (OR, 0.47; 95% CI, 0.09-2.88), all favored therapeutic anticoagulation in moderately ill patients.
Contrary to initial suspicions, the sickest patients with COVID-19 do not seem to benefit from escalation of their anticoagulation. The INSPIRATION trial studied 562 adults with COVID-19 infection who were admitted to the intensive care unit (ICU) and randomized them to enoxaparin 40 mg once daily or 1 mg/kg once daily.5 The primary outcome of venous or arterial thrombosis at 30 days was not lower in those randomized to the higher intensity regimen (HR, 1.06; 95% CI, 0.83-1.36). Another smaller randomized controlled trial was also unable to demonstrate a benefit for intermediate compared to standard enoxaparin for the primary outcome of overall mortality in a cohort of 176 high-risk patients admitted to the ICU or with evidence of coagulopathy.6 Investigators from the ATTACC, ACTIV-4a, REMAP-CAP trial also evaluated the role of therapeutic dose heparin compared to standard anticoagulant prophylaxis in critically ill patients and similarly could not demonstrate a reduction in the primary outcome of organ-support-free days (OR, 0.83; 95% CI, 0.77-1.03) or overall mortality (OR, 0.84; 95% CI, 0.64-1.11). The AntiCoagulaTlon cOroNavirus (ACTION) trial also failed to demonstrate improvement with therapeutic anticoagulation; however, these results have not been stratified by definitions of “moderately ill” patients that have proven successful in other studies.7 Hospitalized patients in this study with an elevated d-dimer were randomized to therapeutic anticoagulation (rivaroxaban if stable or heparins if/when unstable) for 30 days or standard prophylactic anticoagulation and showed no difference in net clinical benefit (composite of death, thrombotic events, or major or clinically relevant non-major bleeding; RR, 1.17; 95% CI, 0.82-1.66).
Data from the randomized clinical trials of various antithrombotic regimens have made it clear that not all patients hospitalized with COVID-19 infections will benefit from higher intensity anticoagulation strategies. No study has been able to demonstrate an improvement with intermediate doses over prophylactic doses. With therapeutic anticoagulation the results are mixed and indicate that a nuanced and careful application may benefit some hospitalized patients. Most trials have not stratified their results by age, but there is at least some suggestion in the ACTION trial that patients older than 60 years might benefit most from the therapeutic anticoagulation strategy, and that would fit with epidemiology studies that have demonstrated a jump in both mortality and thrombotic events in those older than 60 years.8 It is less clear how helpful d-dimer stratification will be, but based on the RAPID, ACTION, and HEP-COVID studies, it does seem to select for a population with higher event rates. Successful studies used numerous exclusion criteria that should be evaluated thoroughly to weigh the risk and benefits; additionally, some degree of increased bleeding risk should be anticipated. While awaiting expert review and consensus guidance statements from ASH on these new data, we are now tasked with how and when to apply this evidence to our patients in clinical practice. In doing so clinicians must determine whether to apply more broad inclusion criteria (as with the ATTACC, ACTIV-4a, REMAP-CAP trial) or whether to target a higher risk population (as in the HEP-COVID and RAPID trials). These trials also need to be interpreted alongside trials evaluating outpatient8 and post-hospital discharge anticoagulation therapies to complete the full story of anticoagulation for this disease.
Competing Interests
Dr. Houghton indicated no relevant conflicts of interest.
