Chinese Data Back Anticoagulants to Prevent COVID-19 Thromboembolism
To move forward, treatment strategies should have set protocols and be prospectively tested, one expert stressed.
COVID-19 ups the risk of thromboembolic events, bleeding, and mortality, confirm data on more than 1,000 hospitalized Chinese patients. The researchers also found that anticoagulants—particularly when given systemically—can help protect against these adverse outcomes.
“More and more evidence demonstrates that COVID-19 is associated with a high risk of thrombotic complications, which could raise the mortality and the morbidity,” said lead investigator Yutao Guo, MD, PhD (Chinese PLA General Hospital, Beijing, and Liverpool Heart & Chest Hospital, England), presenting the results during a late-breaking science session of the virtual European Society of Cardiology Congress 2020.
Less is known about systemic thromboembolism in particular, she continued, and how best to balance the possibly beneficial effects of anticoagulation against the potential for major bleeding. “Indeed, there are quite complex, interactive factors which would impact both thromboembolic and bleeding risk and thus [should] influence decision-making,” Guo observed.
She told TCTMD that, in their study, venous thromboembolism (VTE) was the most common type of thromboembolic event (3.7%), followed closely by ischemic stroke (3.3%). This suggests “that any thromboprophylaxis approach should aim to prevent both systemic and venous thromboembolic events, even for mild or moderately ill COVID-19 patients,” she wrote in an email.
There are quite complex, interactive factors which would impact both thromboembolic and bleeding risk and thus [should] influence decision-making. Yutao Guo
Stavros V. Konstantinides, MD, PhD (Center for Thrombosis and Hemostasis, Johannes Gutenberg University Mainz, Germany), who commented on the findings for TCTMD, said that many studies from China, Europe, and the United States also have indicated COVID-19 has thrombogenic effects. “What is surprising, in fact, is that the thrombosis rate was rather ‘low’ here, only 7.3%. In previous studies, it was as high as 20%, 40%, or even 50% or higher,” he pointed out. “These huge differences show that the studies are not well standardized yet.”
For this report in particular, there are several unknowns, he said via email. “Who was tested and why for pulmonary embolism or stroke among all these patients? Who decided and with what criteria? We simply don’t know. The patients’ data were mostly collected in retrospect, so they were ‘uncontrolled’ and missing information can no longer be recovered,” noted Konstantinides, suggesting the comparatively low event rates may be an indicator of less-severe COVID-19.
Still, said Konstantinides, “this study confirms that there is a serious problem of thrombosis in patients with COVID-19. Right now, thromboprophylaxis, mainly with a parenteral anticoagulant, is recommended by international societies for patients with COVID-19 who need to be treated in-hospital.”
To move knowledge forward, it will be important to test strategies prospectively using clear protocols for specific anticoagulants and dosing, he advised. “Moreover, hospitals must establish protocols to standardize which will be the symptoms and signs that will prompt a search for a thrombotic event in a patient admitted with COVID-19. This sounds easier than it is: most of these patients have dyspnea (shortness of breath), which is common to COVID itself and to pulmonary embolism.
“When standard criteria and protocols are put in place and are followed in ALL patients, then we will be able to do high-quality research,” Konstantinides emphasized.
Guo et al looked at 1,125 patients (mean age 58 years; 50% men) with COVID-19 admitted to Union Hospital in Wuhan, China, between January 8 and April 7, 2020. Among them, 7.3% experienced thromboembolic events, 10.0% developed major bleeding, and 2.2% had both. One in eight patients died (8.1%) died during their hospital stay. Overall, 22.1% were given some sort of anticoagulation, most commonly parenteral heparin (9.2%) or low-molecular-weight heparin (9.9%) at an “intermediate” dose of 40 to 60 mg. Only 7.7% received oral anticoagulants.
The researchers did multivariate analyses that adjusted for underlying factors that might be associated with thromboembolic and bleeding risks: comorbidities; COVID-19 severity; use of anti-COVID-19 drugs, including antivirals, immunomodulators, and Chinese herbs; use of antithrombotic drugs; and whether patients had undergone invasive surgery.
Parenteral anticoagulation was associated with a lower risk of the primary endpoint, a composite of in-hospital thromboembolism, major bleeding, and death (HR 0.70; 95% CI 0.51-0.95), as was the use of antiviral drugs (HR 0.66; 95% CI 0.47-0.92). A favorable balance for the combined thromboembolic and bleeding risks was seen among patients taking parenteral anticoagulants (HR 0.36; 95% CI 0.13-1.01).
Atrial fibrillation/irregular heart rhythm, seen in 5.5% of patients, stood out as a predictor for systemic thromboembolism (HR 3.16; 95% CI 1.06-9.46). For VTE, predictors included antivirals (HR 0.37; 95% CI 0.17-0.81), oral anticoagulants (HR 0.17; 95% CI 0.08-0.36), and parenteral anticoagulants (HR 0.32; 95% CI 0.14-0.75). Major bleeding risk was higher in patients who underwent surgery (HR 2.80; 95% CI 1.08-7.29) and received parenteral anticoagulants (HR 1.56; 95% CI 1.01-2.42). Liver dysfunction, observed in 5.8% of patients, increased the likelihood of both thromboembolic and bleeding events.
Older age, decreased lymphocytes, comorbid CAD, and obstructive sleep apnea each were associated with all-cause death, while the use of oral anticoagulants was related to decreased mortality (HR 0.15; 95%CI 0.05-0.49).
Giving no anticoagulation at all is in most cases the worst solution. Stavros V. Konstantinides
Based on their results, Guo said anticoagulation “could be considered as an option to reduce clinical events.” She emphasized, though, that clinicians should keep an eye out for atrial fibrillation in COVID-19 patients and be aware of its associated risks.
Konstantinides, asked whether these findings can inform who should get anticoagulation and how, said these observational data can’t provide that type of insight. “When all, unselected patients receive the same treatment, then we can tell in the end who benefited from it or not. But with so few patients anticoagulated as in the present study, we do not know what the reasons were to give or not give heparin, so we cannot tell if it had a protective effect and how big that was. This is a typical case of ‘bias’ in clinical research,” he commented.
He did, however, have some universal messages stemming from the fact that hospitalized COVID-19 patients are at increased thrombotic risk.
“As a general rule, prophylactic anticoagulation is needed for ALL of them. However, before starting, check first if the bleeding risk is high. This can be the case, for example, in patients who were recently operated on or had a serious injury, or in those with low platelets or fibrinogen in the blood (which can be because of the infection itself), or in those with severe failure of kidney and/or liver function,” Konstantinides advised. “Also, check if any of the other drugs that the patient receives (some of which may be new, experimental therapies) increases the bleeding risk or interacts with the action of anticoagulant drugs. If you suspect any problems, consult a thrombosis specialist.”
It’s possible to determine “the right drug and dose for prophylaxis,” he concluded. “Giving no anticoagulation at all is in most cases the worst solution.”
Guo Y, Lip GYH, Li W, Xiong J. Risk factors for systemic and venous thromboembolism, mortality and bleeding risks in 1125 patients with COVID-19: relationship to anticoagulation status. Presented at: ESC 2020. August 30, 2020.
- Guo reports no relevant conflicts of interest.
- Konstantinides reports lecture and advisory fees from Bayer AG, Pfizer - Bristol-Myers Squibb, Daiichi-Sankyo, and MSD, as well as institutional research support from Bayer AG and Daiichi-Sankyo.