Objective：To examine the feasibility of using cerebral state index (CSI) for monitoring the sedation depth during target-controlled infusion (TCI) with propofol and remifentanil．Methods: Forty-four consenting ASA Ⅰ or Ⅱ patients (aged 18-60 years) undergoing elective surgery under general anesthesia were randomly divided into 4 groups (n=11 each) according to the target effect-site concentrations of remifentanil administered by TCI during induction of anesthesia.The target effect-site concentrations of remifentanil of R0,R2,R4,and R6 groups were 0,2 ng·ml-1,4 ng·ml-1,and 6 ng·ml-1,respectively.Anesthesia was induced by TCI with remifentanil and propofol.CSI and bispectral index (BIS) were used to measure the sedation depth.The initial effect-site propofol concentration(PCe) was 1.5 μg·ml-1,which was increased by 0.5 μg·ml-1 every 4 min．The modified OAA/S score (5=alert，1=does not respond to prodding)，loss of eyelash reflex (LOR eyelash) and loss of response to electric tetanic stimulation (LOR tetanic) were compared against CSI,BIS and PCe(calculated effect-site propofol concentration).Correlation coefficients were calculated between CSI and other parameters．Results: The 4 groups were comparable with respect to the ages and bodyweights．CSI and BIS values were higher but PCe value were lower at LOR eyelash and LOR tetanic in R2,R4,and R6 than those in the R0 group ( P<0.05 or 0.01)．CSI was positively correlated with modified OAA/S score and BIS,but negatively with PCe (All P<0.01).Conclusion: CSI can be used for monitoring the depth of sedation produced by TCI with propopo1 and remifentanil.