Risk of Ischemic Stroke Associated with DOAC Discontinuation on Patients with Non-valvular Atrial Fibrillation

Background: Although Direct Oral Anticoagulants (DOACs) have experienced immense growth in the number of prescriptions in the last decade, the proportion of discontinuers is still common thereby probably increasing the risk of ischemic stroke onset.

Objectives: We aimed to estimate the association between DOACs discontinuation and risk of IS among patients with non-valvular atrial fibrillation (NVAF).

Methods: We used data from a validated cohort of new users of DOACs from BIFAP database aged ≥18 years within the study period Jan 2008-Dec 2015, with at least 1 years' registration with the primary care practitioner (PCP). We followed up patients from the first DOAC prescription date up to the occurrence of IS and conducted a nested control analysis using conditional logistic regression. Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were calculated for discontinuation DOACs (past use (those stopping treatment between 31 and 180 days) compared with current use (those with use within the 7 days up to index date)). Past users were subdivided as stopping treatment within months 3 and 2 prior to index and between months 6 and 3 before index date. Also, as a sensitivity analysis, the effect of naïve current users against IS onset was evaluated compared with non-naïve current users.

Results: DOACs´ discontinuation for 1 to 6 months showed an OR of IS of 2.21 (95% CI: 1.56-3.13), corresponding estimates for individual DOACs were: 1.98 (95% CI: 1.25-3.12) for apixaban, 1.98 (95% CI: 1.24-2.65) for dabigatran, 2.13 (95% CI: 1.34-3.39) for rivaroxaban and 1.38 (95% CI: 0.40-4.72) for edoxaban. When subdividing the past users, we observed how the risk cumulated closer to the index date (corresponding estimates 2.51 (95%CI: 1.84-3.42) and 1.43 (95%CI: 0.96-2.13) for stopping within months 3 and 1 prior to index or between months 6 and 3 before). Naïve past users showed an estimate OR of 2.08 (95% CI: 1.43.-3.03) similar to non-naïve, 2.21 (95%CI: 1.30-3,78).

Conclusions: DOACs discontinuation is associated with a higher risk of ischemic stroke, especially in the second and third months following the interruption. Further research into reasons for DOAC discontinuation as well as strategies for increasing therapy DOAC adherence is warranted.