Oxidized low-density lipoprotein (LDL) is implicated in atherogenesis, but the mechanisms that oxidize LDL in the human artery wall have proven difficult to identify. A powerful investigative approach is mass spectrometric quantification of the oxidized amino acids that are left in proteins by specific oxidation reactions. Comparison of these molecular fingerprints in biological samples with those produced in proteins by various in vitro oxidation systems can indicate which biochemical pathway has created damage in vivo. For example, the pattern of oxidized amino acids in proteins isolated from atherosclerotic lesions implicates reactive intermediates generated by myeloperoxidase, a major phagocyte enzyme. These intermediates include hypochlorous acid, tyrosyl radical, and reactive nitrogen species, each of which generates a different pattern of stable end products. Despite this strong evidence that myeloperoxidase promotes LDL oxidation in vivo, the antioxidant that has been tested most extensively in clinical trials, vitamin E, fails to inhibit myeloperoxidase pathways in vitro. Because the utility of an antioxidant depends critically on the nature of the pathway that inflicts tissue damage, interventions that specifically inhibit myeloperoxidase or other physiologically relevant pathways would be more logical candidates for the prevention of cardiovascular disease. Moreover, levels of oxidized amino acids in urine and plasma might reflect those in tissues and therefore identify individuals with high levels of oxidative stress. Trials with such subjects would seem more likely to uncover effective antioxidant therapies than trials involving the general population.