Although many drugs have inhibitory effects on platelet function, none of them inhibits all of the mechanisms that may be involved in the various forms of thrombosis. Choice of suitable drugs is hampered by lack of full knowledge concerning the reactions that make the major contributions to the formation of arterial thrombi at sites of repeated vessel wall injury or on atherosclerotic lesions. Drugs such as aspirin that inhibit the arachidonate pathway in platelets can only be expected to be effective against thromboembolic events in which the generation of thromboxane A2 plays a major part. If thrombin and fibrin formation are dominant, oral anticoagulant agents or heparin should be beneficial; thus, experimental evidence indicates that with repeated vessel wall injury, the formation of platelet fibrin thrombi on the vessel wall is probably influenced more by inhibitors of thrombin generation than by the subendothelial constituents such as collagen. Agents like prostacyclin that raise platelet cyclic adenosine monophosphate (AMP) levels in platelets by stimulating adenylate cyclase are potent inhibitors of the reaction of platelets to all aggregating and release-inducing stimuli, but these agents are not suitable for long-term administration. The effect of dipyridamole on platelet cyclic AMP levels is weak, and this drug may act through other effects on platelets or on other cells. Indeed, several of the drugs that have been tested in clinical trials may exert their effects through unrecognized mechanisms. Many combinations of drugs have been used to affect platelets or platelets and coagulation. This practice has been based on the theory that because several mechanisms may be involved in thrombus formation, combinations of drugs that inhibit different mechanisms may be beneficial.