OBJECTIVES: The purpose of this study was to evaluate and compare Bispectral index (BIS) and A-line auditory evoked potential index (AAI) for monitoring depth of low-dose midazolam and propofol sedation during colonoscopy. METHODS: A total of 115 consecutive patients (ASA I-IV), receiving low-dose midazolam and propofol sedation for colonoscopy, were evaluated. BIS and AAI levels, Observer's Assessment of Alertness/Sedation (OAA/S) scores, blood pressure, heart rate, oxygen saturation, as well as the presence or absence of eyelash reflex, patient reaction to an external noxious stimulus and to procedure-related pain were recorded every 1-3 min by a single trained observer. RESULTS: There was a positive correlation between BIS and OAA/S scores (correlation coefficient=0.77) and to a lesser extent AAI and OAA/S scores (correlation coefficient=0.47). BIS and AAI showed significant differences between subsequent levels of sedation (P<0.001). The clustered receiver operating characteristic curve estimate of BIS for the detection of deep sedation was significantly better than that of AAI (P<0.001). Regarding the presence or absence of eyelash reflex and patient reaction to an external noxious stimulus and to procedure-related pain, significant different levels were found for BIS as well as AAI, respectively. Only small changes were observed in hemodynamic variables and oxygen saturation. Overall, our data suggest target BIS levels of slightly above 73 for moderate sedation (defined as OAA/S scores 2 and 3). CONCLUSIONS: BIS and AAI correlated with the level of sedation. Hemodynamic variables were poor indicators of the hypnotic-anesthetic status of the patient. BIS discriminated best between moderate and deep sedation and could complement clinical observation for guidance of moderate sedation.