Background and objective: Patients having spinal anaesthesia with hyperbaric bupivacaine may become sensitive to sedative drugs but no data exists about any dose-related effect of the local anaesthetic on the sedative requirement. We aimed to investigate whether hyperbaric bupivacaine dose in spinal anaesthesia has any effect on midazolam requirements. Methods: Sixty unpremedicated patients were allocated to three equal groups. Patients in Groups I and II received hyperbaric bupivacaine 0.5% 10 and 17.5 mg respectively for spinal anaesthesia and Group III was a control group without spinal anaesthesia. In Groups I and II, after the evaluation of sensory block, patients received intravenous midazolam I mg per 30 s until the Ramsay sedation score reached 3 (drowsy but responsive to command). In Group III, general anaesthesia was induced after sedation score had reached 3 using midazolam. The total dose of midazolam (mg kg(-1)) given to each patient, the level of sensory block and complications were recorded. Results: The level of sensory block was higher in Group II (T7) than Group I (T9) (P < 0.01). The doses of midazolam were 0.063 mg kg(-1) in Group I, 0.065 mg kg(-1) in Group II and 0.101 mg kg(-1) in Group III (P < 0.001). There was no correlation between level of sensory block and dose of midazolam in Group I (r = -0.293, P = 0.21) and Group II (r = 0.204, P = 0.39). Conclusions: Different doses of hyperbaric bupivacaine for spinal anaesthesia do not affect the midazolam requirements for sedation. However, spinal anaesthesia with hyperbaric bupivacaine with a maximum spread in the middle thoracic dermatomes may be associated with sedative effects and thus a reduced need for further sedation with midazolam.