Anesth Analg 2004;98:123-127
© 2004 International Anesthesia Research Society
TECHNOLOGY, COMPUTING, AND SIMULATION
Alaris AEPTM Monitor’s "Click Detection" Does Not Help to Detect Inadvertent Disconnection of Headphones During Anesthesia
Gunter N. Schmidt, MD,
Petra Bischoff, MD,
Thomas Standl, MD,
Andreas Gerhardt, MD,
Gunnar Lankenau, MD, and
Jochen Schulte am Esch, MD Section Editor
Department of Anesthesiology, University Hospital Eppendorf, Hamburg, Germany
Address correspondence and reprint requests to Gunter N. Schmidt, MD, Department of Anesthesiology, University Hospital Eppendorf, Martinistr. 52, 20246 Hamburg, Germany. Address e-mail to guschmid{at}uke.uni-hamburg.de
 |
Abstract
|
|---|
Auditory evoked potentials (AEP) can be suppressed by anesthetics dose dependently, but may fail to be registered because of the absence of adequate auditory stimuli. The Alaris AEPTM monitor includes the "Click Detection" (CD) (generating the message "NO AEP" or "LOW AEP") to detect the loss of auditory stimuli. We investigated the accuracy of the CD in 17 patients awake (AWAKE) and during anesthesia (ANESTHESIA) with accurately placed headphones (HP) and after disconnected HP (No HP) over 5 min each, respectively. Alaris AEPTM ARX index, CD, and Bispectral IndexTM were recorded each minute. Changes were evaluated with the Friedman and Wilcoxon test. Sensitivity (SEN) and specificity (SPE) and receiver operating characteristic curve were analyzed for the accuracy of the CD. During AWAKE after disconnection of the HP, Alaris AEPTM ARX index decreased significantly (P < 0.05). The CD was able to detect No HP after 2 min with a SEN of 88% and a SPE of 97%. During ANESTHESIA, no changes were found after HP disconnection. CD detected No HP with a SEN of 100% and a SPE of 20%. The CD of the Alaris AEPTM monitor is not able to detect unnoticed disconnection of HP during ANESTHESIA.
IMPLICATIONS: Signal transmission of auditory evoked potentials can be suppressed by anesthetics, but also by disconnection of headphones. In the present study, we demonstrate that even the Alaris AEPTM monitor with the very new feature "Click Detection" was not able to detect the loss of headphones during general anesthesia with propofol and remifentanil.
|
Introduction
|
|---|
Auditory evoked potentials (AEP) have been reported to be superior to the spontaneous electroencephalogram (EEG) for discriminating between consciousness and unconsciousness (1). Increases in sedation result in decreases of the amplitudes associated with increases of the peak latencies of the evoked potential (2). In clinical practice, especially for intraoperative monitoring, no commercial on-line monitoring was available until now. For this reason, a new commercial AEP monitor (Alaris AEPTM monitor) was developed (3). It works with an algorithm to calculate automatically an index (Alaris AEPTM ARX index, AAI) using the changes in the amplitudes and latencies of the AEP (4,5).
An unsolved problem of the first version of the Alaris AEPTM monitor was a missing artifact detection concerning adequate auditory stimulation via headphones (HP). An absence of auditory stimuli by disconnected HP resulted in loss of AEP amplitudes and low AAI values similar to AEP signal suppression resulting from deep anesthesia. An "AAI controlled" reduction of anesthetic drugs during low AAI values in a constellation with unnoticed disconnection of the HP may result in reduction of anesthesia and awareness. For this reason, the new version (1.5) of the Alaris AEPTM monitor includes the "Click Detection" (CD) to detect HP disconnection.
The aim of the study was to investigate the accuracy of the CD to detect the disconnected HP. The evaluation was performed in awake patients (AWAKE) as well as in patients under general anesthesia with propofol and remifentanil (ANESTHESIA).
|
Methods
|
|---|
After IRB approval and written informed consent, 17 patients scheduled for elective surgery were included in the study. Selection criteria were patients aged 18–75 yr and ASA risk classification I–II. Only patients with no known hearing deficits and without drug intake, which could interact with the central nervous or the cardiopulmonary system, were included in the study to avoid influences on the AEP and EEG.
After premedication with 7.5 mg of midazolam per os (30 min before induction), anesthesia was induced and maintained with a target controlled infusion of 3.0 µg/mL propofol ("Diprifusor," Graseby 3500; Graseby Medical Ltd., Watford, UK) followed by a remifentanil infusion of 0.3 µg · kg-1 · min-1. Rocuronium bromide 0.6 mg/kg was administered to facilitate tracheal intubation.
Study evaluation started with the AWAKE patients over 10 min. The first 5 min were performed with accurately placed HP followed by a 5-min disconnection of the HP (No HP). Another condition was defined 15 min after tracheal intubations under steady-state anesthetic conditions before surgical incision (ANESTHESIA) with a drug-dosing regime with propofol target controlled infusion 3.0 µg/mL and remifentanil 0.3 µg · kg-1 · min-1 over 5 min with accurately placed HP and 5 min after disconnection of the HP (No HP). Mea-surements were performed for every minute during AWAKE and ANESTHESIA, respectively.
For the AEP recordings, two silver/silver-chloride electrodes (Alaris AEPTM electrodes; Alaris Medical Systems, Inc., San Diego, CA) were placed on the forehead and one behind the ear. Auditory stimuli were applied by earphones providing an intermittent bilateral click (9 Hz, 2-ms duration, 65-dB sound pressure level). The information of AEP was calculated automatically to AAI from 100 (awake) to 0 (Alaris AEPTM monitor, Alaris Medical Systems). Processing time for the AAI is 30 s for the first signal and a total update delay of 6 s. Electrode impedance was tested automatically and was kept <5 k
. CD works by estimating the signal-to-noise ratio (SNR). The basic principle is that synchronized averaging will produce larger peaks as compared with asynchronized averaging if an AEP is present in each of the individual sweeps. However, if no AEP is present, the SNR will converge to one. In this case, the CD is activated and generates the message "NO AEP" or "LOW AEP" to the display of the monitor. Results of the CD were registered (separately for "LOW AEP" and "NO AEP") at every time of measurement.
EEG was registered by Bispectral IndexTM (BISTM)-Sensor electrodes (At-Fpzt; Aspect Medical Systems, Newton, MA) and calculated by the A-2000 BISTM monitor (version XP; Aspect Medical Systems, Newton, MA) Electrode impedance was kept <5 k. Bispectral smoothing rates were 30 s. For artifact detection "slow rate, suppression, motion, and height frequency" were enabled (Aspect Medical Systems, Newton, MA).
Changes after loss of HP for AWAKE and ANESTHESIA were investigated using the Friedman test for repeated measures. In case of significant "overall" effects, changes were evaluated in detail a posteriori by the Wilcoxon test. Bonferroni correction was performed to account for the multiple testing. Sensitivity (SEN) and specificity (SPE) were calculated to analyze the accuracy of the CD ("LOW AEP," "NO AEP") to detect the disconnected HP with missing acoustical stimuli. Calculations were performed for the time points separately when "LOW AEP" or "NO AEP" were displayed. Moreover, calculations were performed when even one of them was displayed ("NO/LOW AEP"). The area under the receiver operating characteristic (ROC) curve was also used to summarize the results of SEN and SPE. The ROC curve plots SEN against 1-SPE and reflects the discriminating power of each parameter. A value of ROC = 0.5 means that the parameter predicts the condition not better than a 50:50 chance. A value of ROC = 1.0 means that the parameter predicts the condition correctly 100% of the time. Statistical analysis was performed using the SPSS package (version 9; SPSS Inc., Chicago, IL). Data are presented as mean and standard deviation (SD).
|
Results
|
|---|
Data evaluation was performed for 17 patients (9 women, 8 men, aged 49 ± 16 [SD] yr, weight 77 ± 14 [SD] kg, height 173 ± 7 [SD] cm) with almost artifact-free signal registration.
During AWAKE, auditory stimuli resulted in typically waveform of middle latency AEP (MAEP, Fig. 1a) resulting in AAI values of 76 ± 21 [SD]. After disconnection of the HP, MAEP and AAI decreased immediately (Figs. 1b and 2b). Maximal decreases were observed 2 min after disconnection (28 ± 19 [SD], Fig. 2a). BISTM values were unaffected after disconnection of the HP (Fig. 2a).
View larger version (27K):
[in this window]
[in a new window]
|
Figure 1. Auditory evoked potentials (AEP) displayed on the Alaris AEPTM monitor. Example of a 41-yr-old man during the conditions (a) awake with headphones (HP), (b) awake after disconnection of the HP, (c) during anesthesia with propofol and remifentanil with HP, and (d) during anesthesia with disconnected HP.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 2. Changes after disconnection of the headphones (HP). Shown are mean and standard deviation for Alaris auditory evoked potentialTM ARX index (AAI) (black) and Bispectral IndexTM (BISTM) (gray) 1 min before (-1) and immediately (0) to 5 min after disconnection of the HP for the conditions (a) awake, and (b) during anesthesia. The asterisks indicate statistically significant changes to values 1 min before disconnection of the HP (P < 0.05).
|
|
ANESTHESIA with propofol and remifentanil resulted in suppression of the MAEP associated with low AAI values (Fig. 1c). The MAEP signals were similar and unchanged after disconnection of the HP (Fig. 1d). During ANESTHESIA there were no significant differences in the results for AAI and BISTM between connected (AAI = 12 ± 8 [SD], BISTM = 32 ± 10 [SD]) and disconnected (AAI = 9 ± 5 [SD], BISTM = 32 ± 10 [SD]) HP (Fig. 2b).
The evaluated accuracy of the CD to detect disconnection of HP with SEN and SPE is shown in Table 1. ROC analysis (Fig. 3) involved the information of SEN and SPE. An accurate detection of disconnected HP with ROC values >0.9 were only observed during AWAKE with a delay of 2 min and only if both "LOW AEP" and "NO AEP" indications were considered ("NO/LOW AEP," Fig. 3a). For ANESTHESIA, the CD indicated no accurate detection for the disconnected HP (Fig. 3b).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 3. Accuracy of the Alaris AEPTM monitor to detect the disconnection of the headphones. Presented are the results of the area under the receiver operating characteristic (ROC) curve with standard deviation for (a) awake, and (b) during anesthesia. Filled circles indicate "NO AEP," open triangles indicate "LOW AEP," and filled triangles indicate when "NO AEP" or "LOW AEP" ("NO/LOW AEP") was displayed by the "Click Detection." The dashed line marked ROC values >0.9.
|
|
|
Discussion
|
|---|
The Alaris AEPTM monitor is the only commercially available AEP monitor evaluated for on-line measurements of depth of hypnosis. The respective index AAI is a new variable which has been investigated in very recent studies. AAI is reliable to assess different degrees of hypnosis during propofol (4–7), sevoflurane (4,8), and midazolam applications (7). In comparison with the BISTM, AAI indicates less correlations to propofol plasma concentrations (BISTM r = -0.9, AAI r = -0.8) (6) but better correlations to observer’s assessment of alertness/sedation scale under propofol (BISTM r = 0.95, AAI r = 0.97) and midazolam (BISTM r = 0.70, AAI r = 0.95) (7). Ge et al. (7) supposed that the AAI may be more valuable in monitoring depth of hypnosis than BISTM.
In the present study, we demonstrated that during anesthesia with propofol and remifentanil, the Alaris AEPTM monitor (version 1.5) does not discriminate between measurements with connected and disconnected HP. Both conditions result in similar low AAI values. In case of inadequate auditory stimulation, an "AAI value controlled" reduction of anesthetics implicates a risk of reduction of anesthesia and awareness. In none of the recent studies was this pitfall of the Alaris AEPTM monitor described (4–8).
We used the version 1.5 of the Alaris AEPTM monitor. This version includes the CD for detecting a disconnection of the HP. The CD was developed by estimating the SNR. The basic principle is that synchronized averaging will produce larger peaks as compared with asynchronized averaging if an AEP is present in each of the individual sweeps. However, in the absence of AEP signals (if no AEP is present) the SNR will converge to one. In this case, the CD is activated and generates the message "NO AEP" or "LOW AEP" to the monitor display. Therefore, the CD can only be an accurate tool if the signal involves amplitudes. In awake patients, AEP are normally present with amplitudes after 9.6 (N0), 12.5 (P0), 18.1 (Na), 29.4 (Pa), and 38.5 (Nb) (9). In this case, the SNR is useful to detect missing AEP, as we demonstrated in the present study for the awake patients. When the signal involves no amplitudes, the SNR is even one. Under general anesthesia, the AEP is normally suppressed (2), resulting in an SNR of nearly one. This would explain why the CD was not able to distinguish between HP on versus off in the anesthetized patients. Drug-related AEP suppressions during anesthesia could not be differentiated by CD from loss of AEP because of disconnection of the HP resulting in a very low SPE (20%) and a very high SEN (100%). In the present study, the CD was generating the message "LOW AEP" or "NO AEP" throughout most of the ANESTHESIA period with connected HP. The false alarm of the CD to indicate the loss of auditory stimulation indicates the limitation of the CD in clinical practice.
However, tools for monitoring depth of hypnosis using AEP need a reliable artifact detection which guarantees adequate auditory stimuli to evoke respective responses from the brain. This system must be independent from anesthetic drug effects. The recording of the brainstem AEP (BAEP) may help to detect disconnection of HP. BAEP provide very stable signal quality and they are almost independent from anesthetic drug effects. That is the reason why BAEP in contrast to MAEP are not helpful for the monitoring of depth of hypnosis (10). BAEP are useful to assess brainstem function and to verify acoustical stimulation. BAEP are generated by the same stimuli as the MAEP. Analyzing techniques of both components needs a respective time window including resolution segments of <10 ms. For the calculation of an AEP index (e.g., AAI), the evaluation of signal quality by BAEP may be useful. No BAEP identification should result in any index calculation—"NO AEP."
In the present study, the BISTM was measured with AAI simultaneously. There are many studies indicating that the BISTM was a reliable monitor for depth of hypnosis (11,12). In the present study, the BISTM was used to control stable EEG conditions. No changes of BISTM were obtained for the awake patients and the patients under anesthesia attributed to disconnection of the HP. Thus, changes of AAI after disconnection of the HP were probably not caused by changes of the state of hypnosis or anesthesia. Moreover, the results of the present study underline the findings from Struys et al. (6) indicating (in contrast to AAI) no significant differences in BISTM values between the "on" and "off" stimulation periods. For simultaneously monitoring of the BISTM and AAI, no significant interaction of the auditory clicks on the BISTM can be assumed.
In conclusion, the CD of the Alaris AEPTM monitor is not able to detect disconnected HP during anesthesia accurately. Low values attributed to disconnection of HP and missing auditory stimulation could be misinterpreted as a deep stage of anesthesia. An "AAI controlled" reduction of anesthetics by the anesthesiologist implicates a risk of decreasing anesthesia and awareness. A more valid and sensitive method to identify disconnected HP and the absence of auditory signal transmission to the brain may be the recording of BAEP. BAEP reflect auditory signal processing to the brainstem. There are no BAEP without auditory stimuli, thus BAEP may be a reliable signal quality variable for MAEP-related indices such as AAI.
|
References
|
|---|
- Doi M, Gajraj RJ Mantzarídis, Kenny GNC. Relationship between calculated blood concentration of propofol and electrophysiological variables during emergence from anesthesia: comparison of bispectral index, spectral edge frequency, median frequency and auditory evoked potential index. Br J Anaesth 1997; 78: 180–4.[Abstract/Free Full Text]
- Iselin-Chaves IA, El Moalem HE, Gan TJ, et al. Changes in the auditory evoked potentials and the bispectral index following propofol or propofol and alfentanil. Anesthesiology 2000; 92: 1300–10.[ISI][Medline]
- Jensen EW, Lindholm P, Henneberg S. Autoregressive modeling with exogenous input of middle-latency auditory-evoked potentials to measure rapid changes in depth of anesthesia. Methods Inf Med 1996; 35: 256–60.[ISI][Medline]
- Litvan H, Jensen EW, Revuelta M, et al. Comparison of auditory evoked potentials and the A-line ARX Index for monitoring the hypnotic level during sevoflurane and propofol induction. Acta Anaesthesiol Scand 2002; 46: 245–51.[ISI][Medline]
- Litvan H, Jensen EW, Galan J, et al. Comparison of conventional averaged and rapid averaged, autoregressive-based extracted auditory evoked potentials for monitoring the hypnotic level during propofol induction. Anesthesiology 2002; 97: 351–7.[ISI][Medline]
- Struys MM, Jensen EW, Smith W, et al. Performance of the ARX-derived auditory evoked potential index as an indicator of anesthetic depth: a comparison with bispectral index and hemodynamic measures during propofol administration. Anesthesiology 2002; 96: 803–16.[ISI][Medline]
- Ge SJ, Zhuang XL, Wang YT, et al. Changes in the rapidly extracted auditory evoked potentials index and the bispectral index during sedation induced by propofol or midazolam under epidural block. Br J Anaesth 2002; 89: 260–4.[Abstract/Free Full Text]
- Alpiger S, Helbo-Hansen HS, Jensen EW. Effect of sevoflurane on the mid-latency auditory evoked potentials measured by a new fast extracting monitor. Acta Anaesthesiol Scand 2002; 46: 252–6.[ISI][Medline]
- Buettner UW, Drost E. Akustisch evozierte Potentiale mittlerer Latenz. Z EEG-EMG 1985; 16: 145–7.
- Thornton C, Newton D. The auditory evoked response: a mea-surement of depth of anaesthesia? Ballieres Clin Anaesth 1989; 3: 559–85.
- Katoh T, Suzuki A, Ikeda K. Electroencephalographic derivatives as a tool for predicting the depth of sedation and anesthesia induced by sevoflurane. Anesthesiology 1998; 88: 642–50.[ISI][Medline]
- Rampil IJ. A primer for EEG signal processing in anesthesia. Anesthesiology 1998; 89: 980–1002.[ISI][Medline]
Accepted for publication August 13, 2003.
This article has been cited by other articles:
|
|
|
|
 
F. Weber, M. Zimmermann, and T. Bein
The Impact of Acoustic Stimulation on the AEP Monitor/2 Derived Composite Auditory Evoked Potential Index Under Awake and Anesthetized Conditions
Anesth. Analg.,
August 1, 2005;
101(2):
435 - 439.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. T. V. Chan, S. S. Ho, T. Gin, G. N. Schmidt, T. Standl, and P. Bischoff
AEP "Click Detection" Failure: May Be, May Be Not! * Response
Anesth. Analg.,
September 1, 2004;
99(3):
948 - 950.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
