Platelet aggregation was studied on 670 samples of 20 healthy controls and 279 patients in the late period after cardiovascular surgery. Aggregating agents used were 0.5 and 2.0μM ADP as well as 2mM arachidonic acid (AA). Among the healthy controls, a maximum aggregation rate over 18% (ADP 0.5μM) was considered hyperaggregability, those under 47% (ADP 2μM) or under 78% (AA 2mM) consid-ered hypoaggregability. Maximun aggregation rate in patients not given antiplatelt agents widely varied from hyperaggregability to hypoaggregability. There was a positive correlation botween the first sample and the second one gained from each patient. This suggested that individual variation existed on platelet aggregation. Maximum aggregation rate significatly decreased by ticlopidine for ADP and by aspirin for AA. Both ADP and AA should be used for platelet aggregation. Comparison among three groups of different ticlopidine doses revealed that the doses should be decided on the basis of platelet aggregation for each patient. Thrombotic or hemorrhagic complications were observed in only 9 cases. Four of 7 cases of thrombotic complicatiion showed hyperaggregability and only 1 case showed hypoaggregability. Both cases of hemorrhagic complication showed hypoaggregability. These findings suggest that indication for antiplatelet therapy and doses of antiplatelet agent should be decided in consideration of individual variation on platelet aggregation.