Healthcare disparities among anticoagulation therapies for severe COVID-19 patients in the multi-site VIRUS registry

March 26,2021

March 5, 2021

Abstract: COVID-19 patients are at an increased risk of thrombosis and various anticoagulants are being used in patient management without an established standard-of-care. Here, we analyze hospitalized and ICU patient outcomes from the Viral Infection and Respiratory illness Universal Study (VIRUS) registry. We find that severe COVID patients administered unfractionated heparin but not enoxaparin have a higher mortality-rate (311 deceased patients out of 760 total patients = 41%) compared to patients administered enoxaparin but not unfractionated heparin (214 deceased patients out of 1,432 total patients = 15%), presenting a risk ratio of 2.74 (95% C.I.: [2.35, 3.18]; p-value: 1.4e-41). This difference persists even after balancing on a number of covariates including: demographics, comorbidities, admission diagnoses, and method of oxygenation, with an amplified mortality rate of 39% (215 of 555) for unfractionated heparin vs. 23% (119 of 522) for enoxaparin, presenting a risk ratio of 1.70 (95% C.I.: [1.40, 2.05]; p-value: 2.5e-7). In these balanced cohorts, a number of complications occurred at an elevated rate for patients administered unfractionated heparin compared to those administered enoxaparin, including acute kidney injury (227 of 642 [35%] vs. 156 of 608 [26%] respectively, adjusted p-value 0.0019), acute cardiac injury (40 of 642 [6.2%] vs. 15 of 608 [2.5%] respectively, adjusted p-value 0.01), septic shock (118 of 642 [18%] vs. 73 of 608 [12%] respectively, adjusted p-value 0.01), and anemia (81 of 642 [13%] vs. 46 of 608 [7.6%] respectively, adjusted p-value 0.02). Furthermore, a higher percentage of Black/African American COVID patients (375 of 1,203 [31%]) were noted to receive unfractionated heparin compared to White/Caucasian COVID patients (595 of 2,488 [24%]), for a risk ratio of 1.3 (95% C.I.: [1.17, 1.45], adjusted p-value: 1.6e-5). After balancing upon available clinical covariates, this difference in anticoagulant use remained statistically significant (272 of 959 [28%] for Black/African American vs. 213 of 959 [22%] for White/Caucasian, adjusted p-value: 0.01, relative risk: 1.28, 95% C.I.: [1.09, 1.49]). While retrospective studies cannot suggest any causality, these findings motivate the need for follow-up prospective research in order to elucidate potential socioeconomic, racial, or other disparities underlying the use of anticoagulants to treat severe COVID patients. 

Christian Kirkup1+,Colin Pawlowski1+,Arjun Puranik1,Ian Conrad1,John C. O’Horo2,Dina Gomaa3,Valerie M. Banner-Goodspeed4,Jarrod M Mosier5,Igor Borisovich Zabolotskikh6,Steven K. Daugherty7,Michael A. Bernstein8,Howard A. Zaren9,Vikas Bansal2Brian Pickering2,Andrew D. Badley2,Rahul Kashyap2,AJ Venkatakrishnan1*,Venky Soundararajan1
1nference, Cambridge, MA 02142, USA
2Mayo Clinic, Rochester, MN 55905, USA
3University of Cincinnati, Cincinnati, OH, USA
4Beth Israel Deaconess Medical Center, Boston, MA, USA
5Banner University Medical Center, Tucson, AZ, USA
6Kuban State Medical University, Krasnodar, Russia
7Cox Medical Center Springfield, IL, USA
8Stamford Health, Stamford, CT, USA
9St. Joseph's Candler Health System, Savannah, GA, USA
Correspondence: Venky Soundararajan ( 
Media coverage:
The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.