Predictors of oral cavity bleeding and clinical outcome after dental procedures in patients on vitamin K antagonists. A cohort study.

Patients on vitamin K antagonists (VKA) often undergo invasive dental procedures. International guidelines consider all dental procedures as low-risk procedures, while bleeding risk may differ between standard low-risk (e. g. extraction 1-3 elements) and extensive high-risk (e. g. extraction of >3 elements) procedures. Therefore current guidelines may need refinement. In this cohort study, we identified predictors of oral cavity bleeding (OCB) and evaluated clinical outcome after low-risk and high-risk dental procedures in patients on VKA. Perioperative management strategy, procedure risk, and 30-day outcomes were assessed for each procedure. We identified 1845 patients undergoing 2004 low-risk and 325 high-risk procedures between 2013 and 2015. OCB occurred after 67/2004 (3.3 %) low-risk and 21/325 (6.5 %) high-risk procedures (p=0.006). In low-risk procedures, VKA continuation with tranexamic acid mouthwash was associated with a lower OCB risk compared to continuation without mouthwash [OR=0.41, 95 %CI 0.23-0.73] or interruption with bridging [OR=0.49, 95 %CI 0.24-1.00], and a similar risk as interruption without bridging [OR=1.44, 95 %CI 0.62-3.64]. In high-risk procedures, VKA continuation was associated with an increased OCB risk compared to interruption [OR=3.08, 95 %CI 1.05-9.04]. Multivariate analyses revealed bridging, antiplatelet therapy, and a supratherapeutic or unobjectified INR before the procedure as strongest predictors of OCB. Non-oral cavity bleeding (NOCB) and thromboembolic event (TE) rates were 2.1 % and 0.2 %. Bridging therapy was associated with a two-fold increased risk of NOCB [OR=1.93, 95 %CI 1.03-3.60], but not with lower TE rates. In conclusion, predictors of OCB were mostly related to perioperative management and differed between low-risk and high-risk procedures. Perioperative management should be differentiated accordingly.