(124) Anticoagulation Use for Hospitalized COVID-19 Patients in the United States: Evolving Patterns from March 2020 to January 2022

Saturday, August 27, 2022

8:00 AM – 6:00 PM CEST

Location: Hall C

Publication Number: 798

Presenting Author(s)

Eileen Lee
– Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, United States

Co-Author(s)

Benjamin Bates
– Instructor of Medicine, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, United States
Nathaniel Kuhrt
– Student, Rutgers University, Ocean City, NJ, United States
Chang Xu, MS
– Rutgers The State University of New Jersey, NJ, United States
Aayush Visaria
– Rutgers University, United States
Soko Setoguchi, MD, DrPH, FISPE
– Professor of Medicine and Epidemiology, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, United States

Background: COVID-19 infection can predispose patients to a hypercoagulable state. While various guidelines have emerged to aid decision making on anticoagulation (AC) use for COVID-19 infection, they are based on limited and inconsistent evidence. With the emergence of COVID-19 variants, their effect on hypercoagulability is also unknown.

Objectives: To describe anticoagulation utilization over time and its predictors among hospitalized COVID-19 patients.

Methods: We conducted a retrospective cohort study using the National COVID Cohort Collaborative (N3C), a multi-institutional electronic health record (EHR) repository in the United States. We identified hospitalized COVID-19 patients between 4/1/2020 - 1/31/2022. We categorized AC (enoxaparin, heparin, or direct oral anticoagulant) use into 3 groups - no use, standard prophylaxis (SP), or treatment dose (TD) - and assessed use over calendar time. We also evaluated utilization patterns over time among subgroups defined by low (D-dimer < 1 mcg/mL), medium (D-dimer ≥ 1 mcg/mL and platelet < 50,000/mcL), and high (D-dimer ≥ 1 mcg/mL and platelet ≥ 50,000/mcL) thromboembolic risk.

Results: Among 141,661 COVID-19 patients (median age 57, 49% male, and median BMI 30) from 25 institutions, 66% received some level of AC, with 42% and 24% on SP and TD, respectively. From 4/2020 to 11/2021, the rate of AC was stable over time at around 65%; from 12/2021 to 1/2022, there was a significant decrease in AC use to 47% (p < 0.001). A length of stay (LOS) > 7 days was associated with a 56% increase in AC utilization compared to LOS ≤ 7 days (p < 0.001). Prior to 12/2021, average LOS was 8.6 days compared to 5.6 days after. Out of 53,927 patients stratified by thromboembolic risk, 16%, 26%, and 43% with low, medium, and high risk received TD, respectively. In the high risk group, 84% received any type of AC, and those with a longer LOS (> 7 days) had a higher AC utilization rate (p < 0.001). Compared to 25% of non-high risk patients receiving TD, 43% of high risk patients received TD (p < 0.001). TD rate dropped to 35% after 12/2021 from 44% prior while the percentage of high risk patients remained unchanged.

Conclusions: Two thirds of COVID patients received some level of AC. While trends in AC utilization remained stable over the first 1.5 years since the pandemic, after 12/2021 during the Omicron variant surge in the United States, AC use reduced by 35%, which was accompanied by decreased LOS. High thromboembolic risk status and longer LOS predicted higher AC use at therapeutic dose. Real-world evidence on safety and effectiveness is needed to optimize AC use as COVID-19 evolves.