The efficacy and safety of apixaban use in chronic liver disease: a secondary analysis of the ARISTOTLE trial
B Suleiman, S S Patel, D M Wojdyla, R D Lopes, L Wallentin, E A Harris, W M Frandah, R Zuzul, J H Alexander, M C Bahit, D Vinereanu, C B Granger, S M Al-KhatibAbstract
Introduction
Atrial fibrillation (AF) and chronic liver disease (CLD) frequently coexist; however, high-quality data on the use of apixaban for stroke or systemic embolism prevention in this population remain limited. We aimed to evaluate the safety and efficacy of apixaban compared with warfarin in patients with AF and CLD1.
Methods
While patients with AST or ALT > 2X or total bilirubin ≥ 1.5X the upper limit of normal were excluded from enrollment in ARISTOTLE, 511 patients had CLD as defined by site investigators, 2,314 had elevated hepatic enzymes at enrollment defined as AST > 41 U/L or ALT > 39 U/L or total bilirubin > 1.5 mg/dL, and 15,376 had neither CLD nor elevated hepatic enzymes.2 We conducted a secondary analysis to evaluate efficacy and safety outcomes by liver disease status and to assess treatment effects within these baseline subgroups.
Results
After multivariable adjustment, patients with CLD, compared with those without CLD, had no significant difference in stroke or systemic embolism (HR 0.92, 95% CI 0.46–1.85; p=0.96) or ISTH major bleeding (HR 0.56, 95% CI 0.25–1.24; p=0.20), but had a higher risk of any bleeding (HR 1.35, 95% CI 1.11–1.64; p=0.001). Patients with elevated hepatic enzymes did not differ from those without CLD in stroke or systemic embolism (HR 1.13, 95% CI 0.81–1.58; p=0.65), ISTH major bleeding (HR 0.89, 95% CI 0.66–1.19; p=0.58), or any bleeding (HR 1.03, 95% CI 0.93–1.14, p=0.81).
When comparing the treatment effect of apixaban versus warfarin across liver disease subgroups, there was no statistically significant interaction for any efficacy or safety endpoints (all p ≥ 0.08; Figure 1).
Conclusion
In this secondary analysis of the ARISTOTLE trial, CLD was associated with an increased risk of any bleeding events, but this was not observed in patients with elevated hepatic enzymes. The treatment effect of apixaban versus warfarin on clinical outcomes did not differ by presence of CLD or elevated hepatic enzymes. These findings provide important reassurance regarding the use of apixaban in patients with mild CLD or mildly elevated hepatic enzymes.