Background Wavelet index (WLI) is a new parameter for monitoring depth of anesthesia based on Wavelet analysis. We observed the change of WLI and bispectral index (BIS) in patients regain of consciousness (ROC) in the absence of frontalis electromyographic (EMG) interference. The aim of this study was to compare the ability of WLI and BIS for reflecting ROC in patients undergone surgery. Methods Twelve patients who were scheduled for maxillofacial surgeries were enrolled in the study. BIS and WLI values were monitored continually. After being given midazolam and remifentanil, patients were asked to squeeze the investigator's hand ever 30 seconds. Patients were continuously given propofol until loss of consciousness (LOCI). Tunstall's isolated forearm technique was used to test the surgical consciousness in patients. After total muscle relaxation, endotracheal intubations were performed, and the patients were connected to a ventilator. Then, propofol was withdrawn until the patients showed regain of consciousness (ROC1) and an awareness reaction. After the command test, patients were readministered with propofol until loss of consciousness (LOC2). After surgery, all of the sedatives were withdrawn, and the patients were let to regain consciousness (ROC2). Results The BIS values of twelve patients at ROC after using muscle relaxant were much lower than those at LOCI and ROC2 without using muscle relaxant, showing statistical significance (P<0.05). Meanwhile, the WLI values of twelve patients at ROC1 after using muscle relaxant were much higher than those at LOCI and equal to ROC2 without using muscle relaxant, showing statistical significances (P <0.05) between ROC1 and LOCI. Conclusions This study showed that under muscle relaxation or facial paralysis, when there is no EMG signal, BIS can not accurately reflect regain of consciousness in surgical patients, but WLI can reflect it accurately. So WLI may have advantages for reflecting state of consciousness in surgical patients. Chin Med J 2010;123(12):1520-1523