Safety and efficacy of two drug regimens used for sedation and analgesia during lower-extremity angiography were studied. Midazolam (loading dose 14.5 micrograms/kg; maintenance dose, 7.2 micrograms/kg) with or without fentanyl (loading dose, 0.725 micrograms/kg; maintenance dose, 0.362 micrograms/kg) was administered in a blind fashion as an IV bolus to 50 patients. Hemodynamic responses were monitored, and a standardized questionnaire was given to the patient before, immediately after, and 8-24 hr after the procedure. Physicians also were asked questions after the procedure. Heart rate and oxygen saturation were not significantly different before and 2 min after administration or between groups. A small but significant decrease in blood pressure occurred 2 min after the loading dose was given in both groups. Both drug regimens had statistically significant anxiolytic effects; however, anxiolytic effects were superior in the midazolam/fentanyl group (p less than .02). Patients rated the overall effectiveness of anesthesia with midazolam/fentanyl as superior (p less than .02). Analysis of physicians' judgments of the effectiveness of sedation and analgesia showed a statistically significant advantage for the midazolam/fentanyl group (p less than .01). We conclude that midazolam/fentanyl appears to be as safe as and more efficacious than midazolam alone for sedation and analgesia during lower-extremity angiography.