Electroacupuncture interferes with bispectral index monitoring.

As we were interested in the use of bispectral index (BIS) monitoring during the application of acupuncture, we conducted a PubMed search on 31 July 2012 using the keywords ‘acupuncture’ and ‘bispectral index’. This yielded 23 articles reporting the use of BIS during evaluation of the effects of acupuncture. Except for the paper by Zheng et al recently published in this journal, none of these articles mentioned the possible interaction of the acupuncture point stimulation procedure with EEG, which is the primary signal used in the BIS algorithm for monitoring the depth of anaesthesia. We present a clinical case which supports the concerns of Zheng et al on the influence of electroacupuncture (EA) on EEG. After giving informed consent, a 73-year-old woman received auricular EA during elective extracranial skin surgery. Total intravenous anaesthesia was induced and maintained with propofol (4–8 mg/kg/h), and the opioid analgesic sufentanil was given in a total dose of 50 mg during 140 min of anaesthesia to blunt noxious responses and keep the heart rate and blood pressure within 20% of the baseline level. The patient did not receive any other drugs which might have influenced the EEG signal during EA and BIS monitoring. The airway was secured with an endotracheal tube and lung ventilation was mechanically controlled to keep endtidal carbon dioxide at 4.5–5.3 kPa. The depth of anaesthesia was measured using a BIS IntelliVue MP70 monitor (Philips Medical Systems, Germany). The body temperature was maintained within 1°C of the baseline level. Two steel needles 0.22×13 mm (Vinco, China) were inserted 1.5 cm apart at the auricular concha ipsilateral to the attached BIS sensor. Stimulation was applied using the PuntoBravo device (MTR GmbH, Germany) yielding biphasic rectangular impulses with a length of 200 ms, intensity of 5 mA and varying frequencies (figure 1). As a result, EA caused clear changes in the raw EEG curve which is used for BIS calculation. The signal quality index (SQ on figure 1) decreased and electromyographic activity (EMG) and burst suppression index increased during EA compared with no stimulation (upper curve on figure 1). It is known that even smaller artefacts introduced by EMG or electric devices into raw EEG may be misinterpreted by the BIS algorithm as cerebral activity and thus lead to falsely high BIS values. We observed EEG artefacts under EA with 5 mA applied to the concha of the ear—that is, remote from the BIS sensor where acquisition of the EEG signal takes place. We presume that the stimulation of acupuncture points VG24 and Yintang (vicinity to BIS sensor) with