The purpose of this study was to determine the onset times of vecuronium neuromuscular block administered into either the central circulation or a peripheral vein. One hundred and twenty adult patients with a pulmonary artery (PA) catheter were randomly divided into one of three groups with respect to the routes of vecuronium administration (n = 40 in each group). Anaesthesia was induced with midazolam 2.5 mg iv and fentanyl 10-50 micrograms.kg-1 iv and maintained with intermittent doses of fentanyl 50 micrograms iv and nitrous oxide 60-70% in oxygen. After immobilization of the forearm in a splint, the ulnar nerve was stimulated supramaximally every 12 sec. The resulting force of the evoked thumb twitch was recorded (Myograph 2000, Biometer, Denmark). The times from the injection to the first depression of twitch response (latent onset) in patients given vecuronium 0.08 mg.kg-1 into the pulmonary artery, the right atrium, and a peripheral vein on the hand were 58.0 +/- 19.5, 71.5 +/- 17.1, and 82.4 +/- 18.0 sec (mean +/- SD), respectively. The latent onset of neuromuscular block occurred sooner in patients given vecuronium into the central vein than when administered into a vein on the hand (P