Background. We studied the relationship between the clinical endpoints of unresponsiveness to verbal commands (UVC), loss of eyelash reflex (LER), and inhibition of body movement response and/or sneezing to mechanical nasal membrane stimulation (INBMR) vs. the processed EEG variables. Methods. A total of 48 patients were randomized to receive either propofol infusion, 30 mg . kg(-1). h(-1) (group P) or ketamine bolus, 0.25, 0.5 or 0.75 mg i.v., followed by propofol infusion, 30 mg . kg(-1). h(-1) and variable dose ketamine infusion, 0.25, 0.5 or 0.75 mg . kg(-1). h(-1) (groups PK0.25, PK0.5 and PK0.75) until UVC, LER and INBMR. Results. Propofol infusion, 30 mg . kg(-1). h(-1), induced UVC, LER and INBMR at bispectral index (BIS): 96.7 +/- 1.7, 63.1 +/- 9.1 and 33.3 +/- 7.2; 95% spectral edge frequency (SEF): 17.4 +/- 3.3, 16.8 +/- 3.9 and 13.9 +/- 2.7; and median frequency (MF) values of 4.8 +/- 2.4, 4.8 +/- 2.5 and 3.2 +/- 1.6 (mean +/- SD), respectively. Ketamine supplementation in groups PK0.5 and PK0.75 reduced the propofol dose requirement and achieved the clinical endpoints at higher BIS and 95% SEF values with lower propofol concentrations as compared to groups P and PK0.25 (9.9 +/- 5.8 and 9.4 +/- 3.4 vs. 13.4 +/- 4.5 and 14 +/- 5.8 mu g . ml(-1)). Conclusions. Our results suggest additive interaction between propofol and ketamine for achieving the clinical (hypnotic) endpoints. The paradoxically higher BIS and 95% SEF values at clinical endpoints with ketamine supplementation of propofol may be due to either lower plasma propofol concentrations and/or excitatory effects of ketamine.