Aspirin remains the most widely used drug for prevention of vascular events. Recent observational epidemiological evidence has raised the concern that aspirin withdrawal for treatment non-compliance, surgery or side effects can carry an increased thrombotic risk. The delay to the thrombotic event was between 7 to 30 days in most reports and most frequently 7 to 10 days. The mechanism underlying this effect remains poorly understood. Using an in vivo model of laser-induced thrombosis, aspirin injected in one single dose of 100 mg/kg bw has also shown a prothrombotic activity in the rat 8 to 10 days after injection in the normal rat. The hypothesis was made that minimal concentrations of aspirin or ultra-low dose aspirin (ULDA) could induce this effect. ULDA showed prothrombotic properties in the same model of induced thrombosis that were very similar to those described after aspirin withdrawal, but the effect was observed only one hour after aspirin administration. This prothrombotic effect of ULDA is very similar to the effect observed after COX 2 selective inhibition with NS 398. The administration of both the selective COX 2 inhibitor and ULDA did not produce further changes. High-dose ASA counterbalances the lack of COX 2 with an antithrombotic effect. No effect of residual ASA was observed in COX 2 -/- mice, thus confirming the existence of a COX 2 inhibition pathway. COX 2 inhibition produced by residual ASA is the probable cause of ischaemic accidents and drug-eluting stents thrombosis a few days after ASA withdrawal.