Auditory evoked potentials have been used as an indicator of awareness. During combined local and general anesthesia clinical signs of adequate anesthesia are difficult to evaluate. In the present study we combined peridural analgesia with three techniques of general anesthesia. Intraoperative wakefulness was documented and correlated with cardiocirculatory parameters as well as with mid-latency auditory evoked potentials (MLAEP). METHODS. After institutional approval and informed consent 30 patients undergoing elective laparotomy were studied as follows: first, continuous peridural analgesia was instituted in all patients to block painful sensation of surgical stimuli and the anesthetic level was maintained at T5. Then general anesthesia was induced with propofol 2.5 mg/kg i.v. (group I, n = 10), thiopental 5 mg/kg i.v. (group II, n = 10), or etomidate 0.2 mg/kg i.v. (group III, n = 10) and maintained with propofol 3-5 mg/kg per hour i.v. (group I), isoflurane 0.4-0.8 vol.-% (group II), or flunitrazepam 0.005-0.01 mg/kg i.v. and fentanyl 0.0025-0.005 mg/kg i.v. bolus injections every 20-30 min (group III). Heart rate and arterial pressure were registered continuously. Purposeful movements of the limbs, eye-opening, or other movements as well as coughing were documented as signs of intraoperative wakefulness. AEP were recorded in the awake state, after induction, and during maintenance of general anesthesia. Latencies of the peaks V, Na, and Pa were measured. By fast-Fourier transformation corresponding power-spectra were calculated to analyze the energy content of the AEP frequency components. RESULTS. Intraoperative wakefulness occurred statistically significantly more often in the patients of group III than in those of groups I and II. There was no correlation between wakefulness and cardiocirculatory parameters. Latencies of peaks V, Na, and Pa in the awake patients were in the normal range; the corresponding power-spectra had their major energy content in the 30-40-Hz range. After induction of general anesthesia with propofol, thiopentone, and etomidate as well as during maintenance of general anesthesia with propofol and isoflurane peak latencies of Na and Pa increased, frequencies in the 30-40 Hz range became suppressed, and MLAEP energy maxima shifted to the low-frequency range. In contrast, during maintenance of general anesthesia with flunitrazepam/fentanyl peak latencies of Na and Pa returned to awake values and frequencies in the range of 30 Hz regained energy dominance in the corresponding power-spectra. CONCLUSIONS. The maintenance of MLAEP and the primary cortical complex Na/Pa correlates with the incidence of motor signs of wakefulness. During the combination of regional and general anesthesia, isoflurane and propofol seem to provide better suppression of intraoperative wakefulness than bolus injections of flunitrazepam/fentanyl.