Purpose: The purpose of this study was to determine the dose-response relationships for edrophonium antagonism of mivacurium-induced neuromuscular block. Method: Seventy-five ASA physical status I or II adults were given mivacurium 0.15 mg . kg(-1) followed by an infusion (7 mu g . kg(-1) . min(-1)) during alfentanil-propofol-N2O anaesthesia. Train-of-four stimulation (TOF) was applied to the ulnar nerve every 20 sec and the response of the adductor digiti minimi was recorded (Relaxograph NMT-100, DATEX, Helsinki, Finland). Mivacurium infusion was adjusted at five minutes intervals in order to keep the height of the first twitch in TOF (T-1) at 5% of its control value. At the end of surgery the mivacurium infusion was stopped and edrophonium 0.0, 0.05, 0.1, 0.5 or 1.0 mg . kg(-1) combined respectively with glycopyrrolate 0.0, 0.0005, 0.001, 0.005 or 0.01 mg . kg(-1) were administered by random allocation. Results: All four edrophonium doses tested were statistically different from placebo with regard to time to attain a TOF ratio (fourth twitch in TOF/T-1) = 0.7 (0.05:780 +/- 179, 0.1:727 +/- 216, 0.5:547 +/- 287 and 1.0:640 +/- 236 vs 0.0 mg . kg(-1):1089 +/- 323 sec P < 0.05). Doses of 0.1, 0.5 and 1.0 mg . kg(-1) permitted faster recovery time of T-1 from 10 to 95% (T-10-95) (567 +/- 236, 419 +/- 166, 555 +/- 288 vs 861 +/- 224 sec P < 0.05) and from 25 to 75% (T-25-75) (253 +/- 121, 147 +/- 92, 217 +/- 175 vs 429 +/- 154 sec P < 0.05) than did placebo. However, data showed considerable variability for all neuromuscular indices, no matter the dose of edrophonium used. Conclusion: Edrophonium in doses of 0.1 mg . kg(-1) and higher permitted faster recovery of all indices from a mivacurium-induced block during alfentanil-propofol-N2O anaesthesia than did placebo.