Objective
It is unclear if stopping treatment with dabigatran, a new oral anticoagulant (NOAC), induces a paradoxical rebound prothrombotic state. We investigated if short‐term (1‐3 days) dabigatran cessation is associated with a higher thrombus volume than expected from a simple reversal of the anticoagulant effect.

Methods
Ten‐week‐old C57Bl/6 mice (n=338) received one of the following oral treatments: phosphate‐buffered saline (PBS), dabigatran for 7 days with or without 1‐4 day cessation, and aspirin in either a single dose or daily for 7 days. Some of the animals that ceased dabigatran for 1‐3 days received single‐dose aspirin. Thereafter, we induced FeCl3‐mediated carotid thrombosis in 130 mice, after which we performed microCT thrombus imaging. The other 208 mice underwent coagulation assays or platelet function tests. As an explorative pilot study, we reviewed the medical records of 18 consecutive patients with NOAC cessation‐related cerebral infarction in a large acute stroke cohort.

Results
We observed a higher volume of carotid thrombus after dabigatran cessation at 1‐3 days than after vehicle treatment and showed that this effect could be prevented by single‐dose aspirin pretreatment. Dabigatran cessation unduly increased platelet aggregability for 2 days after drug cessation, an effect mediated through arachidonic acid or thrombin, which effect was significantly attenuated by single‐dose aspirin pretreatment. In patients, short‐term (≤3 day) cessation of NOAC therapy, compared with longer‐term (≥5 days) cessation, tended to be associated with relatively high stroke severity.

Interpretation
We provide the first preclinical evidence that a rebound prothrombotic state follows short‐term cessation of dabigatran therapy.