Patients who develop COVID-19 are at risk of venous and arterial thrombosis, so multiple randomized trials have assessed the benefit of anticoagulant and antiplatelet treatment. Results have been mixed, and the optimal thromboprophylaxis strategy remains uncertain, particularly for critically ill patients.
In COVID-PACT, a 2×2 factorial, randomized, controlled trial that enrolled critically ill patients with COVID-19, researchers at Brigham and Women’s Hospital determined full-dose anticoagulation was superior to standard-dose prophylaxis in reducing the proportion of patients experiencing a thrombotic event. The addition of clopidogrel in eligible patients was neither beneficial nor harmful.
Erin A. Bohula, MD, DPhil, cardiovascular medicine and critical care specialist, and staff investigator for the Thrombolysis in Myocardial Infarction Study Group at the Brigham; Marc S. Sabatine, MD, MPH, chair of the TIMI Study Group and Lewis Dexter, MD Distinguished Chair in Cardiovascular Medicine; David A. Morrow, MD, MPH, director of the Levine Cardiac Intensive Care Unit; and colleagues detail the results in Circulation.
Study Population and Procedures
From August 5, 2020, to March 2, 2022, COVID-PACT investigators at 34 U.S. sites enrolled adults with COVID-19 who were requiring ICU-level care, had been at that level of care for ≤96 hours before randomization, and did not have another indication for full-dose anticoagulation:
- 390 patients were randomly assigned 1:1 to receive full-dose or standard-dose prophylactic anticoagulation until the end of the follow-up period, defined as death, hospital discharge, or day 28, whichever came first
- 292 (75%) of all patients had no ongoing or planned antiplatelet therapy and were also randomized 1:1 to clopidogrel (300 mg on the day of randomization, then 75 mg/day) or no antiplatelet therapy until the end of the follow-up period
Patients had a screening lower-extremity venous ultrasound 10 to 14 days after randomization.
Efficacy Endpoints
The primary efficacy outcome was a hierarchical composite of thrombotic events “on-treatment” (during therapy or within 72 hours of the last dose). The events were defined in the following order: death due to venous or arterial thrombosis, pulmonary embolism, clinically evident deep venous thrombosis (DVT), type 1 myocardial infarction, ischemic stroke, systemic embolic event or acute limb ischemia, and clinically silent DVT (identified on the ultrasound).
The key secondary outcome was a hierarchical composite of clinically evident events that contained the same elements, except for clinically silent DVT.
The efficacy of each intervention was analyzed as an unmatched pair win ratio, a newer method of evaluating composite endpoints:
- Every patient in the full-dose anticoagulation or antiplatelet arm was compared with every patient in the relevant control group
- A better outcome for the patient on the treatment of interest, according to the order of the endpoint components, was called a “win”
- A win:lose ratio >1 favored the full-dose anticoagulation or antiplatelet arm
Efficacy of Anticoagulation
Full-dose anticoagulation, compared with standard-dose anticoagulation, substantially reduced the incidence of thrombotic events:
- Primary endpoint—wins in 12.3% vs. 6.4% of patients (win ratio, 1.95; P=0.028)
- Key secondary endpoint—10.0% vs. 5.5% (win ratio, 1.79; P=0.087)
For the primary endpoint, there was no heterogeneity in treatment benefit across any of the nine prespecified subgroups, including patients randomized to antiplatelet therapy and those with elevated D-dimer.
Results were consistent in a time-to-event analysis, where full-dose anticoagulation led to a 44% reduction in thrombotic events compared to standard dose prophylaxis (HR 0.56, 95% CI of 0.32-0.99.
Safety of Anticoagulation
There was no apparent excess in mortality with full-dose anticoagulation:
- On-treatment fatal or life-threatening bleeding (primary safety endpoint)—2.1% of patients with full dose vs. 0.5% with standard dose (P=0.19); no fatal bleeding events
- On-treatment GUSTO moderate or severe bleeding—7.9% vs. 0.5%, driven by higher rates of moderate bleeding
There was no difference between groups in rates of any adverse events, adverse events leading to discontinuation of antithrombotic therapy, or serious adverse events.
Efficacy and Safety of Antiplatelet Therapy
No treatment effect of clopidogrel was observed:
- Primary endpoint—wins in 9.8% of patients with clopidogrel vs. 9.5% wins with no clopidogrel (win ratio, 1.04; P=0.90)
- Key secondary endpoint—6.9% vs. 9.0% (win ratio, 0.79; P=0.53)
Likewise, there were no between-group differences in the primary or secondary safety endpoints.
Results Should Be Considered When Revisiting Practice Guidelines
For critically ill COVID-19 patients, including those who receive advanced non-invasive respiratory support, higher-intensity prophylactic anticoagulation should be considered based on an individualized risk/benefit analysis of bleeding risk. Clinical guidelines should consider these results when revisiting about using antithrombotic therapy in COVID-19.