Patient safety and economics require the most optimized monitoring during anesthesia. Several modern electroencephalographic (EEG) monovariables have been developed to evaluate changes in cerebral function and to minimize the chance of awareness and overdosing. However, a further goal is shortening emergence and recovery from anesthesia and thus reducing perioperative costs (1,2). During the last decade, the bispectral index (BIS; Aspect Medical Systems, Natick, MA), calculated by a phase-coupling algorithm, has been addressed as an indicator to assess the depth of sedation with propofol, isoflurane, and sevoflurane anesthesia (3,4). A new variable, the so-called “Narcotrend” (NT; Monitor Technik, Germany), automatically classifies the spontaneous EEG into stages defined by Kugler (5) during the 1980s. Narcotic stages deescalate from the awake state (Stage A) step-by-step to deeper depths of anesthesia during stages B, C, D, E, and, finally, F. The added substages (up to 3) result in at least 14 NT stages. Adequate depth of anesthesia during surgery has been described for D0, D1, D2, E0, and E1, followed by F0 and F1, which indicate “burst suppression” and “isoelectric EEG,” respectively. Initial results have shown a good correlation of automatically detected stages with those evaluated by visual inspection, whereas no validation with clinical variables (e.g., drug concentrations, hemodynamics, or BIS values) is available.
Modern techniques with total IV anesthesia (TIVA) by computer-controlled infusion systems, such as target-controlled infusion (TCI), provide reliable target plasma concentrations in patients. TCI propofol anesthesia with the Diprifusor (Graseby 3500, SIMS) system has turned out to be especially useful with this issue (6). Combinations with remifentanil, a short-acting opioid with a context-sensitive half-time of 3–4 min, facilitate control of anesthesia (7).
The goal of this study was to investigate analgesic and hypnotic components in the absence of surgery-related stimulation during emergence from remifentanil/propofol anesthesia by using NT, BIS, classical EEG variables, and hemodynamics. We hypothesized that NT and BIS would reflect the hypnotic-sedative effects of propofol, but not the analgesic power of remifentanil.
After IRB approval and written informed consent, 25 elective neurosurgical patients were included in the study. Selection criteria were age 18–75 yr, ASA risk classification I–II, and spinal surgery (Table 1). No patient with any medication interacting with the central nervous or cardiopulmonary system was included in the study, to avoid influences on the EEG and hemodynamic variables.
After premedication with 7.5 mg of midazolam (30 min before induction), anesthesia was induced with remifentanil 0.7 μg · kg−1 · min−1 and TCI of propofol 5.0 μg/mL, followed by 0.6 mg/kg of rocuronium bromide to facilitate endotracheal intubation. Anesthesia was maintained with 0.3 μg · kg−1 · min−1 of remifentanil and 3.0 μg/mL of propofol (Diprifusor). Decreases in heart rate (HR) to <45 bpm or mean arterial blood pressure (MAP) to <60 mm Hg were treated with atropine or theodrenaline. Because of the possibility of interaction with the EEG and hemodynamic variables, these patients were excluded from the study.
Study evaluation started after the end of surgery in the postoperative anesthetic care unit, without any surgical stimulation. Remifentanil 0.3 μg · kg−1 · min−1 and propofol 3.0 μg/mL (TCI) were defined as baseline values over 3 min, followed by further evaluation time points at 1, 3, 5, 6, 7, and 9 min after the end of remifentanil infusion. Propofol (3.0 μg/mL) was reduced step-by-step (0.2 μg/mL) every 3 min until spontaneous eye opening 10 min after the end of remifentanil infusion. Extubation criteria were spontaneous breathing and eye opening. Data collection for all variables was performed 1 min after each calculated target propofol concentration was reached. Because of a frequent artifact rate in response to extubation, procedure data from 1 min before extubation were excluded.
EEG was registered by seven adhesive silver/silver chloride gel-filled electrocardiogram electrodes (Blue-Sensor; Medicotest, Denmark) on carefully prepared skin (Arbo-Prep; Tyco Healthcare, Germany). Electrode placement was performed according to the instructions of the manufacturer of Aspect (BIS, 2-channel reference; At1-Fpz and At2-Fpz, ground Fp2) and NT (1 channel bipolar at the hairless skin of the forehead). Electrode impedance was kept at <5 kΩ. BIS (Version 3.1); relative δ (%δ: 0.5–3.75 Hz), θ (%θ: 4.0–7.75 Hz), α (%α: 8.0–13.5 Hz), and β (%β: 13.75–30.0 Hz); spectral edge frequency (SEF); and median frequency (Median) were recorded by an A-1000 EEG monitor (Aspect Medical Systems). NT stages (Version 2.0 AF/F) were registered by the EEG system NT (MonitorTechnik). All data were stored on disk.
Hemodynamic variables such as HR, MAP, and oxygen saturation (Sao2) were registered at every point of measurement (Marquette-Hellige Medical Systems). End-expiratory CO2 concentrations were maintained between 35 and 40 mm Hg during the entire observation time.
Because of the ordinal structure of NT variables, data evaluation from remifentanil and propofol applications was investigated by a Friedman test (analysis of variance) for repeated measurements. In case of significant “overall effects” in the analysis of variance, changes were evaluated in detail a posteriori by Wil-coxon’s test. Bonferroni’s correction was performed to account for the multiple testing. Moreover, correlations between TCI plasma concentrations of propofol, the EEG, and hemodynamic variables were evaluated by nonparametric Spearman’s rank correlation coefficients. The probability to predict the propofol target concentration (Pk), as described by Smith et al. (8), was calculated for every variable. A value of Pk = 0.5 means that the variable predicts the propofol target concentration no better than a 50:50 chance. A value of Pk = 1.0 means that the variable predicts the propofol target concentration correctly 100% of the time. A value <0.5 means that discordance is more likely than concordance. To enable comparison of Pk, we used 1 −Pk when the Pk value was <0.5 (8). P < 0.05 was considered to be significant. Statistical analysis was performed with the SPSS package (SPSS Inc., Chicago, IL) and PKMACRO (8).
Data evaluation was performed in 21 patients (Table 1) with almost artifact-free signal registration. None of these patients was treated with drugs for hypotension or bradycardia. Four patients were excluded from the study because of changes in the study protocol or artifact contamination. In the recovery room, all patients were tracheally extubated without any complications and had sufficient spontaneous breathing and eye opening.
The stop of the remifentanil infusion during constant TCI of propofol (3.0 μg/mL) resulted in significant increases of %α (from the sixth minute), %β (sixth minute), SEF (seventh minute), MAP (seventh minute), Median (ninth minute), and %θ (ninth minute). This was associated with decreases in %δ (sixth minute; all of them Bonferroni corrected, P < 0.05;Table 2). BIS, NT, HR, and Sao2 did not change during this observation time (Fig. 1).
Changes in target-controlled propofol concentrations were associated with changes in electrophysiological variables and hemodynamics (Table 2, Fig. 2, Fig. 3). Decreases from 3.0 μg/mL (TCI) of propofol resulted in statistically significant increases in NT, BIS, MAP, HR, Median, SEF, %θ, %α, and %β and decreases in %δ (all P < 0.05). Statistically significant changes were observed for NT, %δ, and %α (Fig. 3). These data indicate prediction (Pk) of the propofol target concentration (TCI) by NT, BIS, and classical EEG and hemodynamic variables. The probability to predict was best for MAP and worst for Sao2 (Table 3). Statistically significant correlations between decreasing TCI propofol concentrations and increases in MAP, %θ, BIS, SEF, NT, %β, MF, and HR or decreases in %δ were observed (P < 0.05;Table 3).
The assessment of depth of anesthesia is still an unsolved problem because anesthesiologists are not even able to define anesthetic depth accurately. Alternatively, a number of different terms are used to describe adequate anesthesia. Thus, “mental blockade” is associated with loss of consciousness and memory, and the lack of defense responses to noxious or surgical stimuli is often described as “motor blockade.” “Sensory blockade” describes the absence of nociception, and “autonomic vegetative blockade” is characterized by cardiocirculatory stability. However, all these markers are hard to quantify, and it remains difficult to assess when depth of anesthesia is adequate. In this study, we tried to evaluate how modern and classical EEG and hemodynamic variables reflect remifentanil-mediated “sensory” and propofol-related “mental blockade” in the absence of surgical stimulation.
Pharmacodynamic studies have shown that, during anesthesia, analgesia is adequate with remifentanil 0.3 μg · kg−1 · min−1(9). According to intraoperative hemodynamic stability, a range of 0.25–4.0 μg · kg−1 · min−1 of remifentanil is recommended when combined with propofol for TIVA (10). From these data, we concluded that baseline values with remifentanil 0.3 μg · kg−1 · min−1 and propofol 3.0 μg/mL may mediate adequate sensory blockade in this study. Remifentanil is a synthesized opioid with an ester structure which is susceptible to hydrolysis by nonspecific ubiquitous distributed esterases, resulting in rapid metabolism to essentially inactive compounds. Remifentanil has a context-sensitive half-time of three to four minutes (7); thus, after this time, decreasing sensory blockade can be expected. In this study, remifentanil-related changes in electrophysiological and hemodynamic variables were evaluated within nine minutes after the end of infusion. Statistically significant changes were observed in the classical frequency bands, the SEF, and the Median beginning at the sixth minute after the infusion was stopped. EEG changes were indicated by increases in α, β, SEF, and Median and decreases in δ activity, often described as EEG fastening or classic electrophysiological arousal, during decreasing depth of anesthesia (11). In contrast, the investigated modern EEG monovariables NT and BIS remained unaffected. We conclude from these results that during propofol infusion and without surgical simulation, the analgesic component provided by remifentanil is reflected neither by NT nor by BIS.
Guignard et al. (12) have demonstrated unchanged BIS values during propofol TCI at 4.0 μg/mL combined with 0, 2, 4, 8, and 16 ng/mL remifentanil plasma concentrations. During propofol anesthesia and without surgical stimuli, the increasing sensory blockade mediated by increasing doses of remifentanil was not reflected by BIS. However, quantitative BIS responses to noxious stimuli (laryngoscopy and intubation) were attenuated in relation to increasing remifentanil applications. These findings indicate that BIS is not a reliable indicator for the prediction of responses to painful events, even when BIS responses are quantitatively related to analgesic components. As an explanation of the unsatisfying indication of the analgesic or sensory blockade by the BIS, Guignard et al. (12) hypothesized that these components are mediated in subcortical brain structures and at the level of spinal cord, which cannot be detected by EEG registration from the surface of the scalp. In our study, it was the first time that the BIS, NT, and classical frequency bands were registered simultaneously. According to our data, classical EEG variables indicated an increased sensitivity of decreasing remifentanil effects compared with the modern variables NT and BIS. We hypothesize that the absence of changes in NT and BIS is mostly mediated by the calculation of these monovariables and not by anatomic or pharmacodynamic factors.
Interestingly, during smaller plasma concentrations of propofol (2 μg/mL), increasing remifentanil dosages (0.01 to 0.1 μg · kg−1 · min−1) do correlate with decreasing BIS values during conditions without noxious stimulation (13). Besides a pronounced analgesic effect, all opioids also provide dosage-related sedative effects. Without propofol, remifentanil infusion resulted in high-amplitude EEG slowing and δ waves (14), whereas no findings from NT and BIS were available until now. BIS responses to remifentanil might not be simply dosage or stimulus related (12). Moreover, BIS responses also depend quantitatively on the existence of the “mental blockade” from propofol. During propofol-mediated unconsciousness and in the absence of surgical stimulation, remifentanil infusion does not alter NT or BIS.
In this study, it was possible to predict the propofol target concentration with NT and BIS. Moreover, correlation coefficients between propofol TCI and the electrophysiological and hemodynamic variables reached a significant level. Variations in propofol-induced “mental blockade” have been investigated by BIS and classical EEG variables (15). High correlation coefficients (0.68–0.78) between BIS and propofol plasma concentrations (3,16,17) and clinical criteria of sedation (18) have been observed. This would explain the successful development of closed-loop feedback systems that use median frequency (19) and BIS (20) for the infusion rate of propofol. In this issue, judgment of the usability of NT variables cannot be assessed because no NT data are available.
In contrast, findings from Høymork et al. (21) did not indicate any interrelation between TCI propofol and BIS values during standardized surgical procedures. Rapid variations and BIS-related smoothing (moving average), combined with delayed display indications of almost 30 seconds, have been assumed to be causal. That is why, in this study, the TCI of propofol was maintained over one minute each. Further reasons described by Høymork et al. (21) for low interrelations between TCI propofol and BIS are probably related to the chosen deep levels of anesthesia. BIS values between 30 and 50, indicating deep levels of hypnosis, seemed not to be sensitive enough to discriminate different stages. In this regard, BIS is much more sensitive in a range of 75 to 95 (18). Thus, interrelations between TCI propofol concentrations and BIS probably depend on the range of propofol concentrations. Data for high correlations between TCI propofol concentrations and EEG variables have resulted from studies with plasma concentrations from 0 to 8 μg/mL and 0 to 5 μg/mL (3,16,17). However, the small data range from the Høymork study (4.62 μg/mL; sd, 0.98 μg/mL) may also explain the missing interrelations from TCI propofol and BIS. Propofol plasma levels of 2.6 to 2.8 μg/mL have been addressed as the optimum when combined with remifentanil for anesthesia (22). That is why we were interested in investigating decreasing plasma concentrations beginning at 3 μg/mL of propofol until awakening (1.8 μg/mL of propofol; sem, 0.1 μg/mL). In this small range, low but significant interrelations between decreasing TCI propofol and decreasing depth of anesthesia indicated by NT, BIS, and classical and hemodynamic variables were observed. However, the issue of correlation of electrophysiological variables versus clinical end-points or propofol concentrations will require further investigation. In this regard, it is hard to speculate whether variables with higher coefficients would be better in clinical practice for assessing depth of anesthesia. In this study with healthy patients, MAP was the most suitable variable for indicating decreasing propofol effects. However, in case of cardiovascular diseases or respective drug medications (e.g., β blockers), MAP responses are interfered with and become less reliable. Thus, especially with limited hemodynamic variables, electrophysiological evaluation provides more evidence to asses propofol concentrations. Modern monitor systems provide simple electrode placement and signal interpretation. In this study, the BIS was more reliable than the NT concerning the interrelations with TCI propofol concentrations.
Finally, it should be mentioned that we did not measure propofol blood/plasma concentrations; we used a calculated professional pharmacokinetic model, the Diprifusor system (6). This computerized program takes a patient’s age and weight into account to estimate the target propofol blood/plasma concentration. Previous studies with the Diprifusor system have shown variations between measured and calculated concentrations. The bias of systemic over- and underprediction of the measured plasma concentrations (median performance error) was 7%(23), 16%(24), or 49%(21). The calculated errors were greater during infusion than during emergence from anesthesia (25). However, because the real plasma concentrations were not analyzed in this study, potential relevant differences from the TCI data remain unclear.
In summary, our data show that, during conditions without surgical stimulation, neither NT nor BIS indicates decreasing remifentanil effects during propofol anesthesia. In contrast, changes of propofol target concentrations were reflected by both variables. We conclude that in the absence of noxious stimulation, e.g., before skin incision, neither NT nor BIS provides an acceptable depth of anesthesia monitor when remifentanil infusion is used.
1. Gan TJ, Glass PS, Windsor A, et al. Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanil, and nitrous oxide anesthesia. Anesthesiology 1997; 97: 808–48.
2. Song D, van Vlymen J, White PF. Is the bispectral index useful in the predicting fast-track eligibility after ambulatory anesthesia with propofol and desflurane? Anesth Analg 1998; 87: 1245–8.
3. Glass PS, Bloom M, Kearse L, et al. Bispectral index measures sedation and memory effects of propofol, midazolam, isoflurane and alfentanil in healthy volunteers. Anesthesiology 1997; 86: 836–47.
4. Blake DW, Hogg MN, Hackman CH, et al. Induction of anaesthesia with sevoflurane, preprogrammed propofol infusion or combined sevoflurane/propofol for laryngeal mask insertion: cardiovascular, movement and EEG bispectral index response. Anaesth Intensive Care 1998; 26: 360–5.
5. Kugler J. Elektroenzephalographie in Klinik und Praxis. Stuttgart, Germany: Thieme, 1981.
6. Gray JM, Kenny GNC. Development of the technology for “Diprifusor” TCI systems. Anaesthesia 1998; 53 (Suppl): 22–7.
7. Kapila A, Glass PSA, Jacobs JR, et al. Measured context-sensitive half-time of remifentanil and alfentanil. Anesthesiology 1995; 83: 968–75.
8. Smith WD, Dutton RC, Smith NT. Measuring the performance of anesthetic depth indicators. Anesthesiology 1996; 84: 38–51.
9. Fragen RJ. Interaction of remifentanil and propofol during intubation and skin incision/arthroscope for outpatient orthopedic surgery. Internal study report 1994: UCR/94/027, USA-226.
10. Hogue CW, Bowdle A, O’Leary C, et al. A multicenter evaluation of total intravenous anesthesia with remifentanil and propofol for elective inpatient surgery. Anesth Analg 1996; 83: 279–85.
11. Kochs E, Bischoff P, Pichelmeier U, Schulte am Esch J. Surgical stimulation induces changes in brain electrical activity during isoflurane/nitrous oxide anesthesia: a topographic EEG analysis. Anesthesiology 1994; 80: 1026–34.
12. Guignard B, Menigaux C, Dupont X, et al. The effect of remifentanil on the bispectral index change and hemodynamic responses after orotracheal intubation. Anesth Analg 2000; 90: 161–7.
13. Strachan AN, Edwards ND. Randomized placebo-controlled trial to assess the effect of remifentanil and propofol on bispectral index and sedation. Br J Anaesth 2000; 84: 489–90.
14. Egan TD, Muir KT, Stanski DR, Shafer SL. Using the electroencephalographic (EEG) fingerprint to define the clinical pharmacodynamics (PD) of a novel opioid: application to remifentanil (REMI). Clin Pharm 1996; 59: 129.
15. Schwender D, Daunderer M, Mulzer S, et al. Spectral edge frequency of the electroencephalogram to monitor “depth” of anaesthesia with isoflurane or propofol. Br J Anaesth 1996; 77: 179–84.
16. Leslie K, Sessler DI, Schroeder M, Walter K. Propofol blood concentration and the bispectral index predict suppression of learning during propofol/epidural anesthesia in volunteers. Anesth Analg 1995; 81: 1269–74.
17. Doi M, Gajraj RJ, Mantzaridis H, Kenny GNC. Relationship between calculated blood concentration of propofol and electrophysiological variables during emergence from anaesthesia: comparison of bispectral index, spectral edge frequency, median frequency and auditory evoked potential index. Br J Anaesth 1997; 78: 180–4.
18. Liu J, Harbhej S, White PF. Electroencephalographic bispectral index correlates with intraoperative recall and depth of propofol-induced sedation. Anesth Analg 1997; 84: 185–9.
19. Schwilden H, Stoeckel H, Schüttler J. Closed-loop feedback control of propofol anaesthesia by quantitative EEG analysis in humans. Br J Anaesth 1989; 62: 290–6.
20. Mortier E, Struys M, De ST, et al. Closed-loop controlled administration of propofol using bispectral analysis. Anaesthesia 1998; 53: 749–54.
21. Høymork SC, Ræder J, Grismo B, Stehen PA. Bispectral index, predicted and measured drug levels of target-controlled infusion of remifentanil and propofol during laparoscopic cholecystectomy and emergence. Acta Anaesthesiol Scand 2000; 44: 1138–44.
22. Vuyk J, Mertens MJ, Olofsen E, et al. Propofol anesthesia and rational opioid selection: determination of optimal EC50
propofol-opioid concentrations that assure adequate anesthesia and a rapid return of consciousness. Anesthesiology 1997; 87: 1549–62.
23. Coetze JF, Glen JB, Wium CA, Boshoff P. Pharmacokinetic model selection for target controlled infusion of propofol. Anesthesiology 1995; 82: 1328–45.
24. Swinhoe CF, Peacock JE, Glen JB, Reilly CS. Evaluation of predictive performance of a “Diprifusor” TCI system. Anaesthesia 1998; 53: 61–7.
© 2002 International Anesthesia Research Society
25. Davidson JA, Macleod AD, Howie JC, et al. Effective concentration 50 for propofol with and without 67% nitrous oxide. Acta Anaesthesiol Scand 1993; 37: 458–64.