Document Type

Conference Proceeding

Publication Date


Publication Title

Am J Hematol


Background: Clinical trials comparing direct oral anticoagulants (DOACs) to warfarin included only a small number of patients that weighed less than 60 kilograms (kg). The safety and efficacy of DOACs in low weight adult patients with atrial fibrillation (AF) is still unclear. Published data is not only sparse but have mixed outcomes. Therapy with DOACs may increase bleeding and/or clotting risk with uncertain antithrombotic benefit in low weight patients.

Objective: To assess bleeding and thrombotic event rates for patients with AF that are prescribed a DOAC and have a low body weight (less than 60 kg) versus patients that have a normal body weight (60 to 100 kg).

Methods: Within the Michigan Anticoagulation Quality Improvement Initiative (MAQI2), we analyzed data for patients with AF prescribed apixaban or rivaroxaban from 2017 through 2021 who had at least 12 months of follow-up. Patients were excluded if they were prescribed dosing different from package insert instructions. Patients were divided by weight into low (less than 60 kg) and normal (60 to 100 kg) cohorts. Assessments included rates of thrombotic events, major bleeding events (International Society on Thrombosis and Haemostasis [ISTH]), and non-major bleeding events requiring an Emergency Department (ED) visit. Patient characteristics were compared using Chi-square and t-test. Bleeding event rates were adjusted for age, gender, and diabetes mellitus and thrombotic event rates were adjusted by CHA2DS2-VASc score. Poisson regression was used to estimate adjusted adverse event rates to control for potentially confounding covariates (apixaban only due to few patients prescribed rivaroxaban).

Results: A total of 616 patients met the inclusion criteria: 83 (13.5%) low weight and 533 (86.5%) normal weight. Most patients were prescribed apixaban (88.5%) with the low weight cohort more often prescribed the lower dose of apixaban (55% versus 6.2%, p<0.0001). The low weight cohort had a higher mean age (78.9% versus 74.4%, p<0.0002), proportion of females (94% versus 54%, p<0.0001) and CHA2DS2-VASc score (4.4 (1.6) versus 3.9 (1.6)), but a lower proportion of patients with diabetes mellitus (9.6% versus 25.1%, p<0.0018) [Table 1]. In the unadjusted analysis of patients prescribed apixaban, non-major bleeding events requiring an ED visit (10.8 per 100 patient-years versus 7.4 per 100 patient-years, p<0.0001), occurred more often in the low versus normal weight patient cohort [Table 2]. However, adjusted analysis found no statistically significant difference in events in low and normal weight cohorts prescribed apixaban [Table 2]. Comparisons within patients prescribed rivaroxaban could not be made due to a small sample size of low weight patients.

Conclusions: Among low weight patients with AF the use of apixaban was not associated with bleeding (major and non-major) or thrombotic events after adjusting for potential confounding covariates. Larger studies may offer further insight into the overall safety and efficacy of DOAC therapy in these patients.



First Page