Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
THE Bispectral Index (BIS) is increasingly used to monitor the level of (un)consciousness during surgical anesthesia and conscious sedation. 1Generally, an intraoperative BIS of 40–60 is considered sufficient to maintain adequate hypnosis for surgery. 2,3Recently, a new version of the BIS® monitor has been introduced: the BIS-XP® (Aspect Medical Systems, Newton, MA). The BIS-XP® is said to exhibit improved resistance to artifacts from electrocautery devices and to detect and filter interference from electromyographic activity and other conditions commonly encountered during monitored anesthesia care sedation that may cause artifacts. #
We report three cases in which volunteers receiving combinations of propofol and midazolam as part of a pharmacokinetic-dynamic interaction study remained responsive to verbal command, although the BIS-XP® values were at, or just above, 40. A Web enhancement is provided with an MPEG1 digital video file displaying the responsiveness of one of the volunteers in relation to the BIS-XP® values.
With the approval of the Leiden University Medical Center ethics committee and informed consent from the subjects, a study on the pharmacokinetic-dynamic interaction between propofol and midazolam at varying concentration combinations was performed. For each subject, the electroencephalogram was recorded continuously using the BIS® Quatro sensor (Aspect Medical Systems), placed as prescribed on the left side of the skull and connected to the BIS-XP®. For all subjects, the impedance was low (on the order of 2–4 kΩ), and the signal quality index was high (0–100; well above 50) at the times of sedation assessments. The processed electroencephalogram variables were stored on a disk for off-line analysis. In addition, the electrocardiogram, transcutaneous arterial oxygen saturation, end-tidal carbon dioxide concentration, respiratory rate, and arterial blood pressure were monitored continuously throughout the study.
The volunteers breathed spontaneously through a mask with an inspiratory oxygen fraction of 40%. All three volunteers maintained adequate spontaneous respiration and were hemodynamically stable throughout the study. After a 10-min baseline recording period, a target-controlled infusion of propofol was started using the Diprifusor® (AstraZeneca, Macclesfield, United Kingdom) with maintenance of a constant target propofol concentration for 435 min. Fifteen minutes after the start of the target-controlled infusion of propofol, midazolam was given as a rapid infusion for 1 min, followed by a slower continuous infusion for 59 min. At regular intervals, when blood samples were taken from the arterial line for analysis of blood midazolam and propofol concentrations, the level of sedation was assessed by verbal command and/or mild prodding.
The first subject was a 27-yr-old man who weighed 85 kg and was 185 cm tall. The target propofol concentration for this subject was 0.6 μg/ml, and the initial and secondary midazolam infusion rates were 0.05 mg · kg−1· min−1and 0.05 mg · kg−1· h−1, respectively (total midazolam dose in 60 min, 8.5 mg). For the awake volunteer, the BIS exceeded 95 in the absence of any medication. Then, with the target propofol concentration of 0.6 μg/ml, the BIS was maintained at 97 after blood-effect site equilibration (fig. 1). Thereafter, during the first 40 min after the start of the midazolam administration, the BIS decreased to ≈ 60 during unstimulated periods and increased to 98 after verbal stimulation. Forty minutes after the start of the midazolam infusion, the BIS gradually decreased further to 40 at the end of the midazolam infusion. Unexpectedly, throughout the study period the volunteer remained responsive to verbal commands and/or mild prodding at the shoulder to a degree equivalent to an Observer’s Assessment of Alertness/Sedation score between 2 and 4, even at BIS levels of 40–45.
The second subject was a 25-yr-old man who weighed 100 kg and was 195 cm tall. The target propofol concentration for this subject was also 0.6 μg/ml, and the initial and secondary midazolam infusion rates were 0.05 mg · kg−1· min−1and 0.05 mg · kg−1· h−1, respectively (total midazolam dose in 60 min, 10 mg). For the awake volunteer, the BIS exceeded 95 in the absence of any medication. Then, with propofol at a target concentration of 0.6 μg/ml, the BIS was maintained at 97 after blood-effect site equilibration (fig. 2). Thereafter, within 3 min after the start of the midazolam administration, the BIS decreased to 67 and gradually decreased further to as low as 40 at the end of, and just after termination of, the midazolam infusion. Again, throughout the study period the volunteer remained responsive to verbal commands and/or mild prodding at the shoulder to a degree equivalent to an Observer’s Assessment of Alertness/Sedation score between 2 and 4, even at BIS levels of 40–45. We then decided to monitor the next volunteer even more closely and to record the next session on videotape.
The third subject was a 25-yr-old man who weighed 87 kg and was 187 cm tall. The target propofol concentration for this subject was 1 μg/ml, and the initial and secondary midazolam infusion rates were 0.035 mg · kg−1· min−1and 0.035 mg · kg−1· h−1, respectively (total midazolam dose in 60 min, 6 mg). With written informed consent and Leiden University Medical Center Ethics Committee approval, we gathered digital video data from this session until the termination of the midazolam infusion. For the awake volunteer, the average BIS was 96 in the absence of any medication. With a target-controlled infusion of propofol of 1 μg/ml, the BIS decreased to a mean level of 92 after blood-effect site equilibration (fig. 3). Then, within 3 min after the start of midazolam administration, the BIS decreased to values as low as 44. During midazolam administration, the BIS varied between 40 and 60. Again, throughout the study period the volunteer remained responsive to verbal commands and/or mild prodding to a degree equivalent to an Observer’s Assessment of Alertness/Sedation score between 2 and 4, even with BIS levels of 40–45. The videotape, displayed as a Web enhancement, furthermore provides data on the responsiveness of this volunteer at low BIS levels. Additional information regarding this case is available on the Anesthesiology Web site at http://www.anesthesiology.org.
We describe three cases in which the BIS-XP® provided BIS values of 40–50 for volunteers who were responsive to verbal commands while receiving a combination of propofol and midazolam. In our hospital, we tend to administer propofol infusion regimens during propofol-opioid anesthesia on the basis of the BIS level. Based on the current literature, we advise our residents to maintain the BIS level between 40 and 60 intraoperatively. 2Most of our patients receive midazolam for premedication. The observations described herein therefore raise various questions that are relevant to our daily clinical practice.
Two issues must be considered when interpreting our observations in relation to data from the current literature. First, most data in the literature were determined using earlier versions of the BIS® monitor. 4Second, few data exist from careful evaluation of the effect of combinations of agents on the BIS.
Regarding the first issue, it may well be that the BIS-XP® provides lower BIS values than previous versions at similar hypnotic levels in similar subjects. As stated earlier, the BIS-XP® is claimed to be less sensitive to artifacts of the electromyographic activity than earlier versions of the BIS® monitor. Previously, it was reported that electromyographic activity falsely elevates the BIS. 5Introducing a version that is less sensitive to this may thus result in lower BIS levels in the absence of full muscle relaxation (as occurs in most patients).
Regarding the second issue, we note that the electroencephalographic activation induced by both propofol and midazolam has been difficult to interpret and model in the past. 6,7It may well be that the particular combination of propofol and midazolam at these low concentrations is not part of the BIS-behavioral database on which the BIS calculation is based. As a result, the electroencephalographic pattern induced by this combination may well be misinterpreted by the BIS® monitor as an electroencephalographic pattern associated with a patient experiencing a surgical hypnotic sedation level instead of actually being responsive to verbal commands. However, to our knowledge, no controlled studies have been done to examine hypnotic drug interactions and their effect on BIS, especially not using the BIS-XP®.
In conclusion, we report on the responsiveness of three volunteers with BIS-XP® values of 40–50 while receiving a combination of propofol and midazolam. The case reports draw attention to the relationship between the BIS and the responsiveness of patients as derived by the BIS-XP® in the presence of a combination of two hypnotic agents. The BIS user should be aware that the BIS is a measure of drug effect, not an independent measure of brain function. Consequently, the clinical anesthesiologist has no guarantee that a particular BIS will relate to the desired effect when a particular drug, or combination of drugs, is not part of the data file used to train the algorithm of the BIS calculation. As such, the case reports stress the need for further investigation of both the BIS-XP® itself and the effect of combinations of hypnotic agents on the BIS-XP®. Furthermore, the case reports stress the need for careful interpretation by the anesthesiologist of the BIS-XP® values in the clinical setting as long as the scientific basis for the clinical application of the BIS-XP® is not yet completely clear.