Impact of apixaban treatment discontinuation on the risk of hospitalization among patients with nonvalvular atrial fibrillation and COVID-19.

INTRODUCTION: This study evaluated the risk of hospitalization among nonvalvular atrial fibrillation (NVAF) patients with an outpatient COVID-19 diagnosis who discontinued vs continued apixaban treatment. METHODS: Adult patients with NVAF with an apixaban prescription prior to an outpatient COVID-19 diagnosis were identified from Optum Clinformatics claims database (1 April 2020-31 March 2021). Continuers were those who continued apixaban as of the index date (date of initial outpatient COVID-19 diagnosis) and discontinuers were those who had the last day of apixaban supply on or before index. Patients were followed from COVID-19 diagnosis to change of continuation/discontinuation status, switch, death, end of continuous coverage or study end, whichever occurred first. Inverse probability treatment weighting (IPTW) was performed to balance cohorts. Cox proportional hazard models were used to compare the risk of all-cause hospitalization and hospitalization for ischemic stroke (IS), venous thromboembolism (VTE), myocardial infarction (MI), bleeding and mortality. RESULTS: A total of 7869 apixaban patients with COVID-19 were included: 6676 continuers (84.8%) and 1193 discontinuers (15.2%). Compared with continuers, discontinuers had a higher risk of all-cause hospitalization (hazard ratio [HR]: 1.23; 95% confidence interval [CI]: 1.08-1.40), IS (HR: 2.00; 95% CI: 1.03-3.87), VTE (HR: 2.37; 95% CI: 1.06-5.27) and mortality (HR: 2.28; 95% CI: 1.85-2.80). There were no significant differences in the risk of MI (HR: 1.01; 95% CI: 0.54-1.90) or bleeding-related hospitalization (HR: 1.13; 95% CI: 0.73-1.76). CONCLUSION: NVAF patients with COVID-19 who discontinued apixaban had a higher risk of hospitalization and thrombotic events vs those who continued apixaban, with no significant difference in bleeding-related hospitalization.

Thrombotic and bleeding events, mortality, and anticoagulant use among 546,656 hospitalized patients with COVID-19 in the United States: a retrospective cohort study.

This study describes demographics, thrombotic and bleeding events, mortality, and anticoagulant use among hospitalized patients with COVID-19 in the United States. Premier Healthcare Database data were analyzed to identify inpatients with a discharge diagnosis for COVID-19 (ICD-10-CM code: U07.1) from April 1, 2020 to March 31, 2021, and matched historical controls without COVID-19 (inpatients discharged between April 1, 2018 and March 31, 2019). Thrombotic [including venous thromboembolism (VTE)] and bleeding events were based on ICD-10-CM discharge diagnosis codes. Of the 546,656 patients hospitalized with COVID-19, 20.1% were admitted to the ICU, 62.8% were aged ≥ 60 years, 51.5% were male, and 31.0% were non-white. Any thrombotic event was diagnosed in 10.0% of hospitalized and 20.8% of ICU patients with COVID-19 versus (vs) 11.5% and 24.4% for historical controls, respectively. More VTE events were observed in hospitalized and ICU patients with COVID-19 than historical controls (hospitalized: 4.4% vs 2.7%, respectively; ICU: 8.3% vs 5.2%, respectively; both P < 0.0001). Bleeding events were diagnosed in 10.2% of hospitalized and 21.8% of ICU patients with COVID-19 vs 16.0% and 33.2% for historical controls, respectively. Mortality among hospitalized (12.4%) and ICU (38.5%) patients with COVID-19 was higher vs historical controls (2.4%, P < 0.0001 and 9.4%, P < 0.0001, respectively) and higher in hospitalized patients with COVID-19 who had thrombotic events (29.4%) vs those without thrombotic events (10.8%, P < 0.0001). VTE and mortality were higher in hospitalized and ICU patients with COVID-19 vs historical controls. The presence of thrombotic events was associated with worse outcomes.