Cyclooxygenases represent a major target of pharmaceutical therapy. Cyclooxygenase inhibitors are applied in order to reduce inflammation, to relieve from pain, or to prevent atherothrombotic complications in cardiovascular disease. Inhibition of platelet aggregation by aspirin, which is due to inhibition of platelet cyclooxygenase-dependent formation of thromboxane A2, is a cheap, safe and effective strategy to prevent myocardial infarction or stroke and is thus the most established strategy of secondary prevention of atherothrombotic disease. However, the existence of several isoforms of the cyclooxygenase enzyme, their tissue-specific expression patterns, their spatial and functional association with enzymes that further degrade the major cyclooxygenase products and the specific pharmacological properties of substances that have the potential of inhibiting cyclooxygenase make the ultimate physiological effects of a specific cyclooxygenase inhibitor a highly sophisticated pharmacological question. Specific inhibitors of the cyclooxygenase-2 isoform (COX-2) have at first promised to represent a major improvement of pharmaceutical therapy, but later have been suggested to enhance the risk of atherothrombotic events in vivo. This has led to the withdrawal of some of these substances from global markets and also initiated a discussion as to whether some non-selective cyclooxygenase inhibitors (non-steroidal anti-inflammatory drugs, NSAID), such as naproxen would also have an antithrombotic effect in vivo. This review summarizes current pathophysiological and clinical knowledge about the effects on atherothrombosis of drugs that target cyclooxygenases either as specific inhibitors of a cyclooxygenase isoform, or as non-specific cyclooxygenase inhibitors. It specifically discusses the question, whether all selective COX-2 inhibitors may have an intrinsic risk of enhancing the risk of atherothrombosis.