Risk Factors for Thromboischemic Complications After Andexanet Alfa Reversal of Factor Xa Inhibitors

Introduction: Despite andexanet alfa’s (AA) novel mechanism of action for factor Xa inhibitor (FXaI) reversal and demonstrated superiority in decreasing intracranial hematoma expansion, global adoption has been limited by the concern for higher rates of thrombo-ischemic events (TE) compared to usual care. This study aims to evaluate safety outcomes in patients reversed with AA and identify independent risk factors associated with TE after AA reversal.


Methods: This single health-system, retrospective cohort study included patients who received AA for reversal of FxaI-related major or critical site bleeding. The primary outcome was 30-day TE after AA administration, which included acute venous thromboembolism (VTE), pulmonary embolism (PE), ischemic stroke (AIS), and myocardial infarction (MI). Patients were divided into cohorts of TE versus no TE and evaluated for characteristics associated with the development of TE post-reversal. A multivariate logistic regression was performed to identify independent risk factors for TE after AA administration.


Results: A total of 146 patients were included with a majority male population (89 [60.9%]) and a median age of 74 years (IQR 64-81 years). There were no significant differences in patient demographics between cohorts. Most patients were prescribed apixaban (97 (66.4%) for atrial fibrillation (95 [65.1%] and received low-dose AA (127 [86.9%]) for an intracranial bleed (106 [72.6%]). A total of 27 (18.5%) of patients experienced a TE within 30 days of AA administration (VTE, 9 (6.2%); PE, 2 (1.4%); VTE+PE, 4 (2.7%); AIS, 7 (4.8%); MI, 5 (3.4%)). The median time from AA administration to the TE was 4.7 days (IQR 0.9-15.5 days). A majority of patients (115 [78.8%]) were minimally restarted on prophylactic anticoagulation within 44.6 hours (IQR 34.1-65.3 hours) of reversal. A pre-reversal anti-Xa level < 1.8 units/mL, history of malignancy, and a BMI > 40 kg/m2 were associated with higher rates of TE on univariate analysis. The only identified independent risk factor for development of a TE post-AA was a BMI > 40 kg/m2 (OR 6.99 [1.5-33.6]; p=0.015).


Conclusions: Patients with a BMI > 40 kg/m2 are at a higher risk for TE post-reversal with AA. Further research is warranted to identify ideal populations for reversal with AA while minimizing risk of TE.