We investigated the effect of midazolam pre-medication on rocuronium-induced neuromuscular blockade during sevoflurane anaesthesia. Twenty-two patients scheduled for elective surgery were randomly divided to receive either no pre-medication (control group) or pre-medication with 0.1 mg/kg midazolam intramuscularly (midazolam group). Anaesthesia was induced with fentanyl and propofol, and maintained with sevoflurane and nitrous oxide in oxygen. Neuromuscular responses were monitored using acceleromyography. The onset and clinical duration of action, time to recovery of first twitch of train-of-four (TOF) response to 75% of control, recovery index and time for TOF recovery to 25% and 50% were recorded. Patient-related data were similar in both groups. The parameters recorded were not significantly different between the groups. Midazolam pre-medication does not influence the time-course of action of rocuronium during sevoflurane anaesthesia.