Therapeutic anticoagulation ups mortality risk among patients hospitalized with COVID-19
Patients hospitalized with COVID-19 who received high-dose therapeutic anticoagulation appeared to be at increased risk for mortality compared with those on prophylactic anticoagulation, according to a research letter in Thrombosis Research.
Researchers also observed an increased risk for thrombocytopenia among patients who received therapeutic vs. prophylactic anticoagulation.
“At the beginning of this pandemic, we had limited understanding of COVID-19 infection. A few early studies reported increased incidence of venous thromboembolism in patients with severe COVID-19 pneumonia, raising concerns as a possible contributor of mortality,” Lei D. Lynn, MD, researcher in the division of hospital medicine at The George Washington University School of Medicine and Health Sciences, told Healio. “In light of these concerns and the absence of clear guidance regarding anticoagulation use, some patients were empirically treated with therapeutic anticoagulation based on either ICU status or an elevated D-dimer level threshold we identified in our hospital as being associated with statistically significant odds for death and reported in the literature out of China [as associated with] an increased probability of developing clots, [whereas] others were on prophylactic anticoagulation per standard of care.
Lynn and colleagues hoped to gain insight into how to better treat patients with COVID-19 by retrospectively analyzing the effects of prophylactic vs. therapeutic anticoagulation on outcomes, with a focus on mortality and subsequent stratification by severity of illness requiring critical care and D-dimer threshold levels, she said.
The analysis included 402 patients (53.7% men; 57.2% aged > 60 years) hospitalized with COVID-19 between March 15 and May 31. Overall, 26.9% of patients required ICU admission and 15.7% were intubated.
Researchers compared clinical outcomes of patients treated with therapeutic anticoagulation (n = 152; 55.7% men; 69.8% Black) with those of patients on prophylactic anticoagulation (n = 250; 52.6% men; 69.6% Black). They used elastic net logistic regression to identify key variables affecting mortality, which they included as covariates to anticoagulation in standard multivariate logistic regression models.
Results showed patients treated with therapeutic anticoagulation had an increased risk for mortality compared with those on prophylactic anticoagulation (unadjusted hospital mortality, 34.8% vs. 15.2%; OR = 3.42; 95% CI, 2.06-5.67).
Researchers also found that patients with COVID-19 coagulopathy who received therapeutic anticoagulation had a higher mortality rate than patients with atrial fibrillation, atrial flutter or prior VTE (38.3% vs. 26.4%), although the difference did not reach statistical significance.
Results of a subset analysis that adjusted for disease severity showed similar survival curves between the two groups. When the researchers stratified the analysis based on D-dimer levels less than or greater than 3 g/mL, the log-rank test favoring prophylactic anticoagulation among non-ICU patients disappeared.
The overall rate of bleeding was 7.2%. Compared with only 3% of patients on prophylactic anticoagulation, 9% of those receiving therapeutic anticoagulation experienced clinically significant bleeding or thrombocytopenia that resulted in treatment discontinuation.
“Our analysis did not show clear benefit with the use of therapeutic anticoagulation empirically in patients with severe COVID-19 pneumonia. On the contrary, we saw an increased incidence of adverse effects, such as clinically significant anemia or thrombocytopenia,” Lynn said. “Thus, we cautioned against empiric use of therapeutic anticoagulation without clear clinical indications, until we gain better understanding of the role of anticoagulation. We understand how challenging this is, especially with recent data suggesting that much of the VTE burden may be seen in the ICU setting, despite patients already on prophylactic anticoagulation.”
Lynn acknowledged limitations of the study, including the fact that it was not powered to detect a difference when analyzing the effect solely among patients in the ICU setting.
“We look forward to perhaps having other questions answered at a granular level from the randomized controlled trials that are ongoing,” Lynn said. “Apart from the ongoing randomized trials, future directions of the study could include examining the role of biomarkers of inflammation and coagulopathy in guiding therapeutic decisions, especially when diagnostic modalities to diagnose VTE may be limited or prohibitive in some institutions due to isolation requirements and patients’ critical illness.”
For more information:
Lei D. Lynn, MD, can be reached at The George Washington University School of Medicine and Health Sciences, 900 23rd St. NW, Washington, D.C., 20037; email: email@example.com.