BACKGROUND The aim of this study was to evaluate the usefulness of spectral frequency index (SFx) monitoring to assess the depth of anaesthesia during propofol-opioid total intravenous anaesthesia (TIVA). METHODS Thirty-three patients scheduled for laparoscopic cholecystectomy under propofol TIVA were prospectively and randomly allocated to receive either remifentanil (bolus of 1.0 microg/kg, followed by continuous infusion from 0.25 to 0.05 microg/kg/min) [n = 18] or alfentanil (bolus of 10 microg/kg, followed by continuous infusion from 2.0 to 0.5 microg/kg/min) [n = 15]. EEG activity was monitored to achieve the desired depth of anaesthesia, and intravenous propofol was titrated to keep the SFx at 70-80%. The remifentanil and alfentanil groups were compared in relation to the plasma propofol concentration required for an adequate level of hypnosis during maintenance of anaesthesia, Pearson correlation coefficient for the relationship between the plasma propofol concentration and SFx values, recovery parameters, and recall of events during anaesthesia. RESULTS The study groups were comparable with regard to demographic characteristics, type and duration of surgery, and time to resumption of spontaneous ventilation. No evidence of explicit recall was noted. The mean plasma propofol concentration required for an adequate level of hypnosis during maintenance of anaesthesia was significantly higher in the alfentanil group (3.20 microg/mL) compared with the remifentanil group (2.17 microg/L) [p < 0.05]. In both groups, the Pearson correlation coefficient test showed a linear correlation between SFx values and propofol concentration in the studied propofol concentration range. The mean time to orientation for name and place was significantly shorter (p < 0.05), the mean propofol concentration at orientation for name and place and at the resumption of spontaneous ventilation timepoints was significantly lower (both p < 0.01), and the mean SFx value at resumption of spontaneous ventilation timepoint was significantly higher (p < 0.01) in the remifentanil group. CONCLUSIONS As SFx is linearly related to plasma propofol concentration, this index may be used to measure anaesthetic effect during propofol anaesthesia. The results of this clinical trial are consistent with a previous computer-simulated opioid-propofol model with regard to intraoperative and recovery variables, although the recovery occurred at different propofol concentration and SFx values.