We reviewed studies of the effects of different estrogens, progestins, and selective estrogen receptor modulators at the coronary and carotid arterial sites to help determine their likely effects on cardiovascular morbidity and mortality. All English-language studies published between 1997 and 2000 on MEDLINE, Current Contents, and Best Evidence were reviewed, including in vitro, other animal, human physiologic, and clinical trial studies. We synthesize, assess limitations, and integrate across systems with the in vivo experience in humans to evaluate the clinical context. Estrogens have favorable direct effects in most circumstances, progestins oppose these effects, and early studies suggest that selective estrogen receptor modulators are protective. In some systems the dosage, route of delivery, and type of progestin may be important and risk factors may modulate hormone effects. The evaluation of endothelial dysfunction gives a unique in vivo opportunity to assess the vascular properties of hormones, although the relationship between the in vivo physiologic effects of hormones and clinical outcomes remains to be determined.