General Purpose Pharmacokinetic-Pharmacodynamic Models for Target-Controlled Infusion of Anaesthetic Drugs: A Narrative Review.

Target controlled infusion (TCI) is a clinically-available and widely-used computer-controlled method of drug administration, adjusting the drug titration towards user selected plasma- or effect-site concentrations, calculated according to pharmacokinetic-pharmacodynamic (PKPD) models. Although this technology is clinically available for several anaesthetic drugs, the contemporary commercialised PKPD models suffer from multiple limitations. First, PKPD models for anaesthetic drugs are developed using deliberately selected patient populations, often excluding the more challenging populations, such as children, obese or elderly patients, of whom the body composition or elimination mechanisms may be structurally different compared to the lean adult patient population. Separate PKPD models have been developed for some of these subcategories, but the availability of multiple PKPD models for a single drug increases the risk for invalid model selection by the user. Second, some models are restricted to the prediction of plasma-concentration without enabling effect-site controlled TCI or they identify the effect-site equilibration rate constant using methods other than PKPD modelling. Advances in computing and the emergence of globally collected databases has allowed the development of new "general purpose" PKPD models. These take on the challenging task of identifying the relationships between patient covariates (age, weight, sex, etc) and the volumes and clearances of multi-compartmental pharmacokinetic models applicable across broad populations from neonates to the elderly, from the underweight to the obese. These models address the issues of allometric scaling of body weight and size, body composition, sex differences, changes with advanced age, and for young children, changes with maturation and growth. General purpose models for propofol, remifentanil and dexmedetomidine have appeared and these greatly reduce the risk of invalid model selection. In this narrative review, we discuss the development, characteristics and validation of several described general purpose PKPD models for anaesthetic drugs.

Utility of the SmartPilot® View advisory screen to improve anaesthetic drug titration and postoperative outcomes in clinical practice: a two-centre prospective observational trial.

BACKGROUND: The advisory system SmartPilot® View (Drägerwerk AG, Lübeck, Germany) provides real-time, demographically adjusted pharmacodynamic information throughout anaesthesia, including time course of effect-site concentrations of administered drugs and a measure of potency of the combined drug effect termed the "'Noxious Stimulation Response Index' (NSRI). This dual-centre, prospective, observational study assesses whether the availability of SmartPilot® View alters the behaviour of anaesthetic drug titration of anaesthetists and improves the Anaesthesia Quality Score (AQS; percentage of time spent with MAP 60-80 mm Hg and Bispectral Index [BIS] 40-60 [blinded]). METHODS: We recruited 493 patients scheduled for elective surgery in two university centres. A control group (CONTROL; n=170) was enrolled to observe drug titration in current practice. Thereafter, an intervention group was enrolled, for which SmartPilot® View was made available to optimise drug titration (SPV; n=188). The AQS, haemodynamic and hypnotic effects, recovery times, pain scores, and other parameters were compared between groups. RESULTS: There were 358 patients eligible for analysis. Anaesthesia quality score was similar between CONTROL and SPV (median AQS [Q1-Q3]) 25.3% [7.4-41.5%] and 22.2% [8.0-44.4%], respectively; P=0.898). Compared with CONTROL, SPV patients had less severe hypotension and hypertension, less BIS <40, faster tracheal extubation, and lower early postoperative pain scores. CONCLUSIONS: Adding SmartPilot® View information did not affect average drug titration behaviour. However, small improvements in control of MAP and BIS and early recovery suggest improved titration for some patients without increasing the risk of overdosing or underdosing. CLINICAL TRIAL REGISTRATION: NCT01467167.

Population Pharmacodynamics of Propofol and Sevoflurane in Healthy Volunteers Using a Clinical Score and the Patient State Index: A Crossover Study.

BACKGROUND: The population pharmacodynamics of propofol and sevoflurane with or without opioids were compared using the endpoints no response to calling the person by name, tolerance to shake and shout, tolerance to tetanic stimulus, and two versions of a processed electroencephalographic measure, the Patient State Index (Patient State Index-1 and Patient State Index-2). METHODS: This is a reanalysis of previously published data. Volunteers received four anesthesia sessions, each with different drug combinations of propofol or sevoflurane, with or without remifentanil. Nonlinear mixed effects modeling was used to study the relationship between drug concentrations, clinical endpoints, and Patient State Index-1 and Patient State Index-2. RESULTS: The C50 values for no response to calling the person by name, tolerance to shake and shout, and tolerance to tetanic stimulation for propofol (µg · ml) and sevoflurane (vol %; relative standard error [%]) were 1.62 (7.00)/0.64 (4.20), 1.85 (6.20)/0.90 (5.00), and 2.82 (15.5)/0.91 (10.0), respectively. The C50 values for Patient State Index-1 and Patient State Index-2 were 1.63 µg · ml (3.7) and 1.22 vol % (3.1) for propofol and sevoflurane. Only for sevoflurane was a significant difference found in the pharmacodynamic model for Patient State Index-2 compared with Patient State Index-1. The pharmacodynamic models for Patient State Index-1 and Patient State Index-2 as a predictor for no response to calling the person by name, tolerance to shake and shout, and tetanic stimulation were indistinguishable, with Patient State Index50 values for propofol and sevoflurane of 46.7 (5.1)/68 (3.0), 41.5 (4.1)/59.2 (3.6), and 29.5 (12.9)/61.1 (8.1), respectively. Post hoc C50 values for propofol and sevoflurane were perfectly correlated (correlation coefficient = 1) for no response to calling the person by name and tolerance to shake and shout. Post hoc C50 and Patient State Index50 values for propofol and sevoflurane for tolerance to tetanic stimulation were independent within an individual (correlation coefficient = 0). CONCLUSIONS: The pharmacodynamics of propofol and sevoflurane were described on both population and individual levels using a clinical score and the Patient State Index. Patient State Index-2 has an improved performance at higher sevoflurane concentrations, and the relationship to probability of responsiveness depends on the drug used but is unaffected for Patient State Index-1 and Patient State Index-2.

Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers.

BACKGROUND: Sedation indicators based on a single quantitative EEG (QEEG) feature have been criticised for their limited performance. We hypothesised that integration of multiple QEEG features into a single sedation-level estimator using a machine learning algorithm could reliably predict levels of sedation, independent of the sedative drug used. METHODS: In total, 102 subjects receiving propofol (N=36; 16 male/20 female), sevoflurane (N=36; 16 male/20 female), or dexmedetomidine (N=30; 15 male/15 female) were included in this study of healthy volunteers. Sedation level was assessed using the Modified Observer's Assessment of Alertness/Sedation (MOAA/S) score. We used 44 QEEG features estimated from the EEG data in a logistic regression algorithm, and an elastic-net regularisation method was used for feature selection. The area under the receiver operator characteristic curve (AUC) was used to assess the performance of the logistic regression model. RESULTS: The performances obtained when the system was trained and tested as drug-dependent mode to distinguish between awake and sedated states (mean AUC [standard deviation]) were propofol=0.97 (0.03), sevoflurane=0.74 (0.25), and dexmedetomidine=0.77 (0.10). The drug-independent system resulted in mean AUC=0.83 (0.17) to discriminate between the awake and sedated states. CONCLUSIONS: The incorporation of large numbers of QEEG features and machine learning algorithms is feasible for next-generation monitors of sedation level. Different QEEG features were selected for propofol, sevoflurane, and dexmedetomidine groups, but the sedation-level estimator maintained a high performance for predicting MOAA/S independent of the drug used. CLINICAL TRIAL REGISTRATION: NCT02043938; NCT03143972.

Tracking Electroencephalographic Changes Using Distributions of Linear Models: Application to Propofol-Based Depth of Anesthesia Monitoring.

OBJECTIVE: Tracking brain states with electrophysiological measurements often relies on short-term averages of extracted features and this may not adequately capture the variability of brain dynamics. The objective is to assess the hypotheses that this can be overcome by tracking distributions of linear models using anesthesia data, and that anesthetic brain state tracking performance of linear models is comparable to that of a high performing depth of anesthesia monitoring feature. METHODS: Individuals' brain states are classified by comparing the distribution of linear (auto-regressive moving average-ARMA) model parameters estimated from electroencephalographic (EEG) data obtained with a sliding window to distributions of linear model parameters for each brain state. The method is applied to frontal EEG data from 15 subjects undergoing propofol anesthesia and classified by the observers assessment of alertness/sedation (OAA/S) scale. Classification of the OAA/S score was performed using distributions of either ARMA parameters or the benchmark feature, Higuchi fractal dimension. RESULTS: The highest average testing sensitivity of 59% (chance sensitivity: 17%) was found for ARMA (2,1) models and Higuchi fractal dimension achieved 52%, however, no statistical difference was observed. For the same ARMA case, there was no statistical difference if medians are used instead of distributions (sensitivity: 56%). CONCLUSION: The model-based distribution approach is not necessarily more effective than a median/short-term average approach, however, it performs well compared with a distribution approach based on a high performing anesthesia monitoring measure. SIGNIFICANCE: These techniques hold potential for anesthesia monitoring and may be generally applicable for tracking brain states.

Model-based drug administration: current status of target-controlled infusion and closed-loop control.

PURPOSE OF REVIEW: Drug administration might be optimized by incorporating pharmacokinetic-dynamic (PK/PD) principles and control engineering theories. This review gives an update of the actual status of target-controlled infusion (TCI) and closed-loop computer-controlled drug administration and the ongoing research in the field. RECENT FINDINGS: TCI is becoming mature technology clinically used in many countries nowadays with proven safety. Nevertheless, changing populations might require adapting the established PK/PD models. As TCI requires accurate PK/PD models, new models have been developed which should now be incorporated into the pumps to allow more general use of this technology. Closed-loop administration of hypnotic drugs using an electro-encephalographic-derived-controlled variable has been well studied and has been shown to outperform manual administration. Computer administration for other drugs and fluids have been studied recently. Feasibility has been shown for systems controlling multiple components of anaesthesia, but more work is required to show clinical safety and efficiency. SUMMARY: Evidence in the literature is increasing that TCI and closed-loop technology could assist the anaesthetists to optimize drug administration during anaesthesia.

Neural mass model-based tracking of anesthetic brain states.

Neural mass model-based tracking of brain states from electroencephalographic signals holds the promise of simultaneously tracking brain states while inferring underlying physiological changes in various neuroscientific and clinical applications. Here, neural mass model-based tracking of brain states using the unscented Kalman filter applied to estimate parameters of the Jansen-Rit cortical population model is evaluated through the application of propofol-based anesthetic state monitoring. In particular, 15 subjects underwent propofol anesthesia induction from awake to anesthetised while behavioral responsiveness was monitored and frontal electroencephalographic signals were recorded. The unscented Kalman filter Jansen-Rit model approach applied to frontal electroencephalography achieved reasonable testing performance for classification of the anesthetic brain state (sensitivity: 0.51; chance sensitivity: 0.17; nearest neighbor sensitivity 0.75) when compared to approaches based on linear (autoregressive moving average) modeling (sensitivity 0.58; nearest neighbor sensitivity: 0.91) and a high performing standard depth of anesthesia monitoring measure, Higuchi Fractal Dimension (sensitivity: 0.50; nearest neighbor sensitivity: 0.88). Moreover, it was found that the unscented Kalman filter based parameter estimates of the inhibitory postsynaptic potential amplitude varied in the physiologically expected direction with increases in propofol concentration, while the estimates of the inhibitory postsynaptic potential rate constant did not. These results combined with analysis of monotonicity of parameter estimates, error analysis of parameter estimates, and observability analysis of the Jansen-Rit model, along with considerations of extensions of the Jansen-Rit model, suggests that the Jansen-Rit model combined with unscented Kalman filtering provides a valuable reference point for future real-time brain state tracking studies. This is especially true for studies of more complex, but still computationally efficient, neural models of anesthesia that can more accurately track the anesthetic brain state, while simultaneously inferring underlying physiological changes that can potentially provide useful clinical information.

Propofol Breath Monitoring as a Potential Tool to Improve the Prediction of Intraoperative Plasma Concentrations.

INTRODUCTION: Monitoring of drug concentrations in breathing gas is routinely being used to individualize drug dosing for the inhalation anesthetics. For intravenous anesthetics however, no decisive evidence in favor of breath concentration monitoring has been presented up until now. At the same time, questions remain with respect to the performance of currently used plasma pharmacokinetic models implemented in target-controlled infusion systems. In this study, we investigate whether breath monitoring of propofol could improve the predictive performance of currently applied, target-controlled infusion models. METHODS: Based on data from a healthy volunteer study, we developed an addition to the current state-of-the-art pharmacokinetic model for propofol, to accommodate breath concentration measurements. The potential of using this pharmacokinetic (PK) model in a Bayesian forecasting setting was studied using a simulation study. Finally, by introducing bispectral index monitor (BIS) measurements and the accompanying BIS models into our PK model, we investigated the relationship between BIS and predicted breath concentrations. RESULTS AND DISCUSSION: We show that the current state-of-the-art pharmacokinetic model is easily extended to reliably describe propofol kinetics in exhaled breath. Furthermore, we show that the predictive performance of the a priori model is improved by Bayesian adaptation based on the measured breath concentrations, thereby allowing further treatment individualization and a more stringent control on the targeted plasma concentrations during general anesthesia. Finally, we demonstrated concordance between currently advocated BIS models, relying on predicted effect-site concentrations, and our new approach in which BIS measurements are derived from predicted breath concentrations.

The History of Target-Controlled Infusion.

Target-controlled infusion (TCI) is a technique of infusing IV drugs to achieve a user-defined predicted ("target") drug concentration in a specific body compartment or tissue of interest. In this review, we describe the pharmacokinetic principles of TCI, the development of TCI systems, and technical and regulatory issues addressed in prototype development. We also describe the launch of the current clinically available systems.

Comparisons of Electroencephalographically Derived Measures of Hypnosis and Antinociception in Response to Standardized Stimuli During Target-Controlled Propofol-Remifentanil Anesthesia.

BACKGROUND: Current electroencephalogram (EEG)-derived measures provide information on cortical activity and hypnosis but are less accurate regarding subcortical activity, which is expected to vary with the degree of antinociception. Recently, the neurophysiologically based EEG measures of cortical input (CI) and cortical state (CS) have been shown to be prospective indicators of analgesia/antinociception and hypnosis, respectively. In this study, we compared CI and an alternate measure of CS, the composite cortical state (CCS), with the Bispectral Index (BIS) and another recently developed measure of antinociception, the composite variability index (CVI). CVI is an EEG-derived measure based on a weighted combination of BIS and estimated electromyographic activity. By assessing the relationship between these indices for equivalent levels of hypnosis (as quantified using the BIS) and the nociceptive-antinociceptive balance (as determined by the predicted effect-site concentration of remifentanil), we sought to evaluate whether combining hypnotic and analgesic measures could better predict movement in response to a noxious stimulus than when used alone. METHODS: Time series of BIS and CVI indices and the raw EEG from a previously published study were reanalyzed. In our current study, the data from 80 patients, each randomly allocated to a target hypnotic level (BIS 50 or BIS 70) and a target remifentanil level (Remi-0, -2, -4 or -6 ng/mL), were included in the analysis. CCS, CI, BIS, and CVI were calculated or quantified at baseline and at a number of intervals after the application of the Observer's Assessment of Alertness/Sedation scale and a subsequent tetanic stimulus. The dependency of the putative measures of antinociception CI and CVI on effect-site concentration of remifentanil was then quantified, together with their relationship to the hypnotic measures CCS and BIS. Finally, statistical clustering methods were used to evaluate the extent to which simple combinations of antinociceptive and hypnotic measures could better detect and predict response to stimulation. RESULTS: Before stimulation, both CI and CVI differentiated patients who received remifentanil from those who were randomly allocated to the Remi-0 group (CI: Cohen's d = 0.65, 95% confidence interval, 0.48-0.83; CVI: Cohen's d = 0.72, 95% confidence interval, 0.56-0.88). Strong correlations between BIS and CCS were found (at different periods: 0.55 < R2 < 0.68, P < 0.001). Application of the Observer's Assessment of Alertness/Sedation stimulus was associated with changes in CI and CCS, whereas, subsequent to the application of both stimuli, changes in all measures were seen. Pairwise combinations of CI and CCS showed higher sensitivity in detecting response to stimulation than CVI and BIS combined (sensitivity [99% confidence interval], 75.8% [52.7%-98.8%] vs 42% [15.4%-68.5%], P = 0.006), with specificity for CI and CCS approaching significance (52% [34.7%-69.3%] vs 24% [9.1%-38.9%], P = 0.0159). CONCLUSIONS: Combining electroencephalographically derived hypnotic and analgesic quantifiers may enable better prediction of patients who are likely to respond to tetanic stimulation.

Development of an Optimized Pharmacokinetic Model of Dexmedetomidine Using Target-controlled Infusion in Healthy Volunteers.

BACKGROUND: Several pharmacokinetic models are available for dexmedetomidine, but these have been shown to underestimate plasma concentrations. Most were developed with data from patients during the postoperative phase and/or in intensive care, making them susceptible to errors due to drug interactions. The aim of this study is to improve on existing models using data from healthy volunteers. METHODS: After local ethics committee approval, the authors recruited 18 volunteers, who received a dexmedetomidine target-controlled infusion with increasing target concentrations: 1, 2, 3, 4, 6, and 8 ng/ml, repeated in two sessions, at least 1 week apart. Each level was maintained for 30 min. If one of the predefined safety criteria was breached, the infusion was terminated and the recovery period began. Arterial blood samples were collected at preset times, and NONMEM (Icon plc, Ireland) was used for model development. RESULTS: The age, weight, and body mass index ranges of the 18 volunteers (9 male and 9 female) were 20 to 70 yr, 51 to 110 kg, and 20.6 to 29.3 kg/m, respectively. A three-compartment allometric model was developed, with the following estimated parameters for an individual of 70 kg: V1 = 1.78 l, V2 = 30.3 l, V3 = 52.0 l, CL = 0.686 l/min, Q2 = 2.98 l/min, and Q3 = 0.602 l/min. The predictive performance as calculated by the median absolute performance error and median performance error was better than that of existing models. CONCLUSIONS: Using target-controlled infusion in healthy volunteers, the pharmacokinetics of dexmedetomidine were best described by a three-compartment allometric model. Apart from weight, no other covariates were identified.

Differential effects of phenylephrine and norepinephrine on peripheral tissue oxygenation during general anaesthesia: A randomised controlled trial.

BACKGROUND: Phenylephrine and norepinephrine are two vasopressors commonly used to counteract anaesthesia-induced hypotension. Their dissimilar working mechanisms may differentially affect the macro and microcirculation, and ultimately tissue oxygenation. OBJECTIVES: We investigated the differential effect of phenylephrine and norepinephrine on the heart rate (HR), stroke volume (SV), cardiac index (CI), cerebral tissue oxygenation (SctO2) and peripheral tissue oxygenation (SptO2), and rate-pressure product (RPP). DESIGN: A randomised controlled study. SETTING: Single-centre, University Medical Center Groningen, The Netherlands. PATIENTS: Sixty normovolaemic patients under balanced propofol/remifentanil anaesthesia. INTERVENTIONS: If the mean arterial pressure (MAP) dropped below 80% of the awake state value, phenylephrine (100 µg + 0.5 µg kg(-1) min(-1)) or norepinephrine (10 µg + 0.05 µg kg(-1) min(-1)) was administered in a randomised fashion. MAIN OUTCOME MEASURES: MAP, HR, SV, CI, SctO2, SptO2 and rate-pressure product (RPP) analysed from 30 s before drug administration until 240 s thereafter. RESULTS: Phenylephrine and norepinephrine caused an equivalent increase in MAP [Δ = 13 (8 to 22) and Δ = 13 (9 to 19) mmHg, respectively] and SV [Δ = 6 ± 6 and Δ = 5 ± 7 ml, respectively], combined with a significant equivalent decrease in HR (both Δ = -8 ± 6 bpm), CI (both Δ = -0.2 ± 0.3 l min(-1) m(-2)) and SctO2 and an unchanged RPP (Δ = 345 ± 876 and Δ = 537 ± 1076 mmHg min(-1)). However, SptO2 was slightly but statistically significantly (P < 0.05) decreased after norepinephrine [Δ  = -3 (-6 to 0)%] but not after phenylephrine administration [Δ = 0 (-1 to 1)%]. In both groups, SptO2 after vasopressor was still higher than the awake value. CONCLUSION: In normovolaemic patients under balanced propofol/remifentanil anaesthesia, phenylephrine and norepinephrine produced similar clinical effects when used to counteract anaesthesia-induced hypotension. After norepinephrine, a fall in peripheral tissue oxygenation was statistically significant, but its magnitude was not clinically relevant.

Analysis of remifentanil with liquid chromatography-tandem mass spectrometry and an extensive stability investigation in EDTA whole blood and acidified EDTA plasma.

BACKGROUND: Remifentanil is a µ-opioid receptor agonist that was developed as a synthetic opioid for use in anesthesia and intensive care medicine. Remifentanil is rapidly metabolized in both blood and tissues, which results in a very short duration of action. Even after blood sampling, remifentanil is unstable in whole blood and plasma through endogenous esterases and chemical hydrolysis. The instability of remifentanil in these matrices makes sample collection and processing a critical phase in the bioanalysis of remifentanil. METHODS: We have developed a fast and simple sample preparation method using protein precipitation followed by liquid chromatography-tandem mass spectrometry analysis. To improve the stability of remifentanil, citric acid, ascorbic acid, and formic acid were investigated for acidification of EDTA plasma. The stability of remifentanil was investigated in stock solution, EDTA whole blood, EDTA plasma, and acidified EDTA plasma at ambient temperature, 4 °C, 0 °C, and at -20 °C. RESULTS: The analytical method was fully validated based on the Food and Drug Administration guidelines for bioanalytical method validation with a large linear range of 0.20 to 250 ng/mL remifentanil in EDTA plasma acidified with formic acid. The stability results of remifentanil in EDTA tubes, containing whole blood placed in ice water, showed a decrease of approximately 2% in 2 hours. EDTA plasma acidified with citric acid, formic acid, and ascorbic acid showed 0.5%, 4.2%, and 7.2% remifentanil degradation, respectively, after 19 hours at ambient temperature. Formic acid was chosen because of its volatility and thus liquid chromatography-tandem mass spectrometry compatibility. The use of formic acid added to EDTA plasma improved the stability of remifentanil, which was stable for 2 days at ambient temperature, 14 days at 4 °C, and 103 days at -20 °C. CONCLUSIONS: The analytical method we developed uses a simple protein precipitation and maximal throughput by a 2-point calibration curve and short run times of 2.6 minutes. Best sample stability is obtained by placing tubes containing EDTA whole blood in ice water directly after sampling, followed by centrifugation and transfer of the EDTA plasma to tubes with formic acid. The stability of remifentanil in EDTA plasma was significantly improved by the addition of 1.5 µL formic acid per milliliter of EDTA plasma. This analytical method and sample pretreatment are suitable for remifentanil pharmacokinetic studies.

Accuracy of the composite variability index as a measure of the balance between nociception and antinociception during anesthesia.

BACKGROUND: The Composite Variability Index (CVI), derived from the electroencephalogram, was developed to assess the antinociception-nociception balance, whereas the Bispectral Index (BIS) was developed to assess the hypnotic state during anesthesia. We studied the relationships between these indices, level of hypnosis (BIS level), and antinociception (predicted remifentanil effect-site concentrations, CeREMI) before and after stimulation. Also, we measured their association with movement in response to a noxious stimulus. METHODS: We randomized 120 patients to one of 12 groups targeting different hypnotic levels (BIS 70, 50, and 30) and various CeREMI (0, 2, 4, or 6 ng/mL). At pseudo-steady state, baseline values were observed, and a series of stimuli were applied. Changes in BIS, CVI, heart rate (HR), and mean arterial blood pressure (MAP) between baseline and response period were analyzed in relation to level of hypnosis, antinociception, and somatic response to the stimuli. RESULTS: CVI and BIS more accurately correlate with somatic response to an Observer Assessment of Alertness and Sedation-noxious stimulation than HR, MAP, CeREMI, and propofol effect-site concentration (Tukey post hoc tests P < 0.01). Change in CVI is more adequate to monitor response to stimulation than changes in BIS, HR, or MAP (as described by the Mathews Correlation Coefficient with significance level set at P < 0.001). In contrast, none of the candidate analgesic state indices was uniquely related to a specific opioid concentration and is extensively influenced by the hypnotic state as measured by BIS. CONCLUSIONS: CVI appears to correlate with somatic responses to noxious stimuli. However, unstimulated CVI depends more on hypnotic drug effect than on opioid concentration.

A response surface model approach for continuous measures of hypnotic and analgesic effect during sevoflurane-remifentanil interaction: quantifying the pharmacodynamic shift evoked by stimulation.

BACKGROUND: The authors studied the interaction between sevoflurane and remifentanil on bispectral index (BIS), state entropy (SE), response entropy (RE), Composite Variability Index, and Surgical Pleth Index, by using a response surface methodology. The authors also studied the influence of stimulation on this interaction. METHODS: Forty patients received combined concentrations of remifentanil (0 to 12 ng/ml) and sevoflurane (0.5 to 3.5 vol%) according to a crisscross design (160 concentration pairs). During pseudo-steady-state anesthesia, the pharmacodynamic measures were obtained before and after a series of noxious and nonnoxious stimulations. For the "prestimulation" and "poststimulation" BIS, SE, RE, Composite Variability Index, and Surgical Pleth Index, interaction models were applied to find the best fit, by using NONMEM 7.2.0. (Icon Development Solutions, Hanover, MD). RESULTS: The authors found an additive interaction between sevoflurane and remifentanil on BIS, SE, and RE. For Composite Variability Index, a moderate synergism was found. The comparison of pre- and poststimulation data revealed a shift of C50SEVO for BIS, SE, and RE, with a consistent increase of 0.3 vol%. The Surgical Pleth Index data did not result in plausible parameter estimates, neither before nor after stimulation. CONCLUSIONS: By combining pre- and poststimulation data, interaction models for BIS, SE, and RE demonstrate a consistent influence of "stimulation" on the pharmacodynamic relationship between sevoflurane and remifentanil. Significant population variability exists for Composite Variability Index and Surgical Pleth Index.

Interaction between nitrous oxide, sevoflurane, and opioids: a response surface approach.

BACKGROUND: The interaction of sevoflurane and opioids can be described by response surface modeling using the hierarchical model. We expanded this for combined administration of sevoflurane, opioids, and 66 vol.% nitrous oxide (N2O), using historical data on the motor and hemodynamic responsiveness to incision, the minimal alveolar concentration, and minimal alveolar concentration to block autonomic reflexes to nociceptive stimuli, respectively. METHODS: Four potential actions of 66 vol.% N2O were postulated: (1) N2O is equivalent to A ng/ml of fentanyl (additive); (2) N2O reduces C50 of fentanyl by factor B; (3) N2O is equivalent to X vol.% of sevoflurane (additive); (4) N2O reduces C50 of sevoflurane by factor Y. These four actions, and all combinations, were fitted on the data using NONMEM (version VI, Icon Development Solutions, Ellicott City, MD), assuming identical interaction parameters (A, B, X, Y) for movement and sympathetic responses. RESULTS: Sixty-six volume percentage nitrous oxide evokes an additive effect corresponding to 0.27 ng/ml fentanyl (A) with an additive effect corresponding to 0.54 vol.% sevoflurane (X). Parameters B and Y did not improve the fit. CONCLUSION: The effect of nitrous oxide can be incorporated into the hierarchical interaction model with a simple extension. The model can be used to predict the probability of movement and sympathetic responses during sevoflurane anesthesia taking into account interactions with opioids and 66 vol.% N2O.