OBJECTIVE The authors aimed to compare propofol and midazolam/fentanyl for reduction of anterior shoulder dislocations using the modified Kocher's maneuver. METHODS This was a multicenter, randomized, clinical trial of patients with anterior shoulder dislocation. Patients were randomized to either propofol or midazolam/fentanyl. The randomized drug was titrated to a clinical sedation end point (spontaneous eye closure). One physician sedated the patient. Another, blinded to the drug administered, reduced the shoulder and recorded details of muscle tone and ease of reduction. RESULTS Eighty-six patients were randomized to treatment with propofol (n = 48) or midazolam/fentanyl (n = 38). Patients in the propofol group had shorter mean times to first wakening (difference in means, 4.6 minutes; 95% confidence interval [CI] = 0.7 to 8.6; p = 0.097) and full consciousness (difference in means, 21.7 minutes; 95% CI = 14.7 to 28.7; p <0.001), had easier shoulder reduction (difference in mean rating, 0.5; 95% CI = 0.0 to 0.9; p = 0.047), and needed fewer reduction attempts (difference in means, 0.5; 95% CI = 0.1 to 1.0; p = 0.02). Patients in the propofol group also had less mean muscle tone at the first reduction attempt (p = 0.08) and needed fewer reduction maneuvers (p = 0.40) but had more respiratory depression (11 vs. six patients; difference in proportions, 7.1%; 95% CI = -11.8 to 26.1; p = 0.58) and had one patient who vomited. CONCLUSIONS Propofol appears to be as effective as midazolam/fentanyl for reduction of anterior shoulder dislocation using the modified Kocher's maneuver. However, the advantage of shorter wakening times associated with propofol should be weighed against the possibility of adverse events, particularly respiratory depression and vomiting.