The purpose of this study was to compare the frequency of psychomimetic reactions after 24 h and 3 months following total intravenous anaesthesia with propofol and neurolept anaesthesia. Forty otherwise healthy female patients were randomly divided into two groups. All were undergoing elective gynaecological laparotomy for non-malignant disease. Nineteen patients were anaesthetized with droperidol, fentanyl, pancuronium, N2O/O2. Twenty patients received total intravenous anaesthesia with propofol, fentanyl and pancuronium. Twenty-four hours after the anaesthesia the patients were interviewed about their subjective experiences of anaesthesia and recovery. Three months after the operation the patients were sent a questionnaire concerning ability to work, sleep and memory disorders. After 24 h the anaesthesia was judged as good by 18 patients receiving propofol and 13 patients receiving NLA (n.s.). The recovery was judged as good by 16 patients in the propofol group and six patients in the NLA group (P < 0.05). Locked-in feelings were reported by one patient in the propofol group and ten patients in the NLA group (P < 0.01). Impairment of memory was reported by one patient in the propofol group and seven patients in the NLA group (P < 0.01). A questionnaire used after 3 months was answered by 18 patients in the propofol group and 17 patients in the NLA group. There were few complaints, and no differences were found between the two groups. In conclusion, total intravenous anaesthesia with propofol seems more acceptable than anaesthesia with neurolept as judged by the patients 24 h after anaesthesia. There were no differences between the two groups concerning psychomimetic reactions 3 months after anaesthesia.