Contributions of PK/PD modeling to intravenous anesthesia

Hierarchical Deep Reinforcement Learning-Based Propofol Infusion Assistant Framework in Anesthesia

This article aims to provide a hierarchical reinforcement learning (RL)-based solution to the automated drug infusion field. The learning policy is divided into the tasks of: 1) learning trajectory generative model and 2) planning policy model. The proposed deep infusion assistant policy gradient (DIAPG) model draws inspiration from adversarial autoencoders (AAEs) and learns latent representations of hypnotic depth trajectories. Given the trajectories drawn from the generative model, the planning policy infers a dose of propofol for stable sedation of a patient under total intravenous anesthesia (TIVA) using propofol and remifentanil. Through extensive evaluation, the DIAPG model can effectively stabilize bispectral index (BIS) and effect site concentration given a potentially time-varying target sequence. The proposed DIAPG shows an increased performance of 530% and 15% when a human expert and a standard reinforcement algorithm are used to infuse drugs, respectively.

A Transformer-Based Prediction Method for Depth of Anesthesia During Target-Controlled Infusion of Propofol and Remifentanil

Accurately predicting anesthetic effects is essential for target-controlled infusion systems. The traditional (PK-PD) models for Bispectral index (BIS) prediction require manual selection of model parameters, which can be challenging in clinical settings. Recently proposed deep learning methods can only capture general trends and may not predict abrupt changes in BIS. To address these issues, we propose a transformer-based method for predicting the depth of anesthesia (DOA) using drug infusions of propofol and remifentanil. Our method employs long short-term memory (LSTM) and gate residual network (GRN) networks to improve the efficiency of feature fusion and applies an attention mechanism to discover the interactions between the drugs. We also use label distribution smoothing and reweighting losses to address data imbalance. Experimental results show that our proposed method outperforms traditional PK-PD models and previous deep learning methods, effectively predicting anesthetic depth under sudden and deep anesthesia conditions.

Deep reinforcement learning-based propofol infusion control for anesthesia: A feasibility study with a 3000-subject dataset

In this work, we present a deep reinforcement learning-based approach as a baseline system for autonomous propofol infusion control. Specifically, design an environment for simulating the possible conditions of a target patient based on input demographic data and design our reinforcement learning model-based system so that it effectively makes predictions on the proper level of propofol infusion to maintain stable anesthesia even under dynamic conditions that can affect the decision-making process, such as the manual control of remifentanil by anesthesiologists and the varying patient conditions under anesthesia. Through an extensive set of evaluations using patient data from 3000 subjects, we show that the proposed method results in stabilization in the anesthesia state, by managing the bispectral index (BIS) and effect-site concentration for a patient showing varying conditions.

Comparison of Remimazolam-Flumazenil versus Propofol for Rigid Bronchoscopy: A Prospective Randomized Controlled Trial

Background: Remimazolam is a novel ultrashort-acting intravenous benzodiazepine sedative−hypnotic that significantly reduces the times to sedation onset and recovery. This trial was conducted to confirm the recovery time from anesthesia of remimazolam-flumazenil versus propofol in patients undergoing endotracheal surgery under rigid bronchoscopy. Methods: Patients undergoing endotracheal tumor resection or stent implantation were randomly allocated into a remimazolam group (Group R) or a propofol group (Group P). The primary outcome was the recovery time from general anesthesia. The secondary outcomes were the time to loss of consciousness (LoC), hemodynamic fluctuations, and adverse events. Results: A total of 34 patients were screened, and 30 patients were enrolled in the study. The recovery time was significantly shorter for Group R (140 ± 52 s) than for Group P (374 ± 195 s) (p < 0.001). The times to LoC were 76 ± 40 s in Group R and 75 ± 25 s in Group P and were not significantly different. There were also no significant differences in hemodynamic fluctuations or adverse events between the two groups. Conclusions: The recovery time from general anesthesia in rigid bronchoscopy patients was shorter using remimazolam-flumazenil than with propofol, with no dramatic hemodynamic fluctuations and adverse events or differences between the agents. Remimazolam-flumazenil allows for faster recovery from anesthesia than propofol.

A "Smart" Biosensor-Enabled Intravascular Catheter and Platform for Dynamic Delivery of Propofol to "Close the Loop" for Total Intravenous Anesthesia

BACKGROUND

Target-controlled infusion anesthesia is used worldwide to provide user-defined, stable, blood concentrations of propofol for sedation and anesthesia. The drug infusion is controlled by a microprocessor that uses population-based pharmacokinetic data and patient biometrics to estimate the required infusion rate to replace losses from the blood compartment due to drug distribution and metabolism. The objective of the research was to develop and validate a method to detect and quantify propofol levels in the blood, to improve the safety of propofol use, and to demonstrate a pathway for regulatory approval for its use in the USA.

METHODS

We conceptualized and prototyped a novel "smart" biosensor-enabled intravenous catheter capable of quantifying propofol at physiologic levels in the blood, in real time. The clinical embodiment of the platform is comprised of a "smart" biosensor-enabled catheter prototype, a signal generation/detection readout display, and a driving electronics software. The biosensor was validated in vitro using a variety of electrochemical methods in both static and flow systems with biofluids, including blood.

RESULTS

We present data demonstrating the experimental detection and quantification of propofol at sub-micromolar concentrations using this biosensor and method. Detection of the drug is rapid and stable with negligible biofouling due to the sensor coating. It shows a linear correlation with mass spectroscopy methods. An intuitive graphical user interface was developed to: (1) detect and quantify the propofol sensor signal, (2) determine the difference between targeted and actual propofol concentration, (3) communicate the variance in real time, and (4) use the output of the controller to drive drug delivery from an in-line syringe pump. The automated delivery and maintenance of propofol levels was demonstrated in a modeled benchtop "patient" applying the known pharmacokinetics of the drug using published algorithms.

CONCLUSIONS

We present a proof-of-concept and in vitro validation of accurate electrochemical quantification of propofol directly from the blood and the design and prototyping of a "smart," indwelling, biosensor-enabled catheter and demonstrate feedback hardware and software architecture permitting accurate measurement of propofol in blood in real time. The controller platform is shown to permit autonomous, "closed-loop" delivery of the drug and maintenance of user-defined propofol levels in a dynamic flow model.

Explaining anaesthetic hysteresis with effect-site equilibration

BACKGROUND

Anaesthetic induction occurs at higher plasma drug concentrations than emergence in animal studies. Some studies find evidence for such anaesthetic hysteresis in humans, whereas others do not. Traditional thinking attributes hysteresis to drug equilibration between plasma and the effect site. Indeed, a key difference between human studies showing anaesthetic hysteresis and those that do not is in how effect-site equilibration was modelled. However, the effect-site is a theoretical compartment in which drug concentration cannot be measured experimentally. Thus, it is not clear whether drug equilibration models with experimentally intractable compartments are sufficiently constrained to unequivocally establish evidence for the presence or absence of anaesthetic hysteresis.

METHODS

We constructed several models. One lacked hysteresis beyond effect-site equilibration. In another, neuronal dynamics contributed to hysteresis. We attempted to distinguish between these two systems using drug equilibration models.

RESULTS

Our modelling studies showed that one can always construct an effect-site equilibration model such that hysteresis collapses. So long as the concentration in the effect-site cannot be measured directly, the correct effect-site equilibration model and the one that erroneously collapses hysteresis are experimentally indistinguishable. We also found that hysteresis can naturally arise even in a simple network of neurones independently of drug equilibration.

CONCLUSIONS

Effect-site equilibration models can readily collapse hysteresis. However, this does not imply that hysteresis is solely attributable to the kinetics of drug equilibration.

Resistance to state transitions in responsiveness is differentially modulated by different volatile anaesthetics in male mice

BACKGROUND

Recent studies point to a fundamental distinction between population-based and individual-based anaesthetic pharmacology. At the population level, anaesthetic potency is defined as the relationship between drug concentration and the likelihood of response to a stimulus. At the individual level, even when the anaesthetic concentration is held constant, fluctuations between the responsive and unresponsive states are observed. Notably, these spontaneous fluctuations exhibit resistance to state transitions R_st. Therefore, the response probability in each individual depends not just upon the drug concentration, but also upon responses to previous stimuli. Here, we hypothesise that R_st is distinct from drug potency and is differentially modulated by different anaesthetics.

METHODS

Adult (14-24 weeks old) C57BL/6J male mice (n=60) were subjected to repeated righting reflex (RR) assays at equipotent steady-state concentrations of isoflurane (0.6 vol%), sevoflurane (1.0 vol%), and halothane (0.4 vol%).

RESULTS

Fluctuations in RR were observed for all tested anaesthetics. Analysis of these fluctuations revealed that R_st was differentially modulated by different anaesthetics (F[2, 56.01]=49.59; P<0.0001). Fluctuations in RR were modelled using a stochastic dynamical system. This analysis confirmed that the amount of noise that drives behavioural state transitions depends on the anaesthetic agent (F[2, 42.86]=16.72; P<0.0001).

CONCLUSIONS

Whilst equipotent doses of distinct anaesthetics produce comparable population response probabilities, they engage dramatically different dynamics in each individual animal. This manifests as a differential aggregate propensity to exhibit state transitions. Thus, resistance to state transitions is a fundamentally distinct, novel measure of individualised anaesthetic pharmacology.

Operative Gynecological Laparoscopy Under Conscious Sedation

BACKGROUND AND OBJECTIVES

Operative laparoscopy is generally performed under general anesthesia. Local anesthesia and conscious sedation may be useful in select short procedures. In the present study, we evaluated safety and efficacy of operative laparoscopy under conscious sedation.

METHODS

Retrospective observational study evaluating patients undergoing gynecologic laparoscopy. Laparoscopy under conscious sedation was performed for each patient with umbilical direct insertion of a 12-mm port, followed by 2 ancillary ports at 1 cm medially to the anterior superior iliac spine. Conversion to conventional laparoscopy or laparotomy was recorded. Conscious sedation was obtained using Remifentanil and Propofol, administered by an infusion system based on pharmacokinetic and pharmacodynamic models. Local anesthesia was administered at port insertion sites and for paracervical block. Pain intensity was evaluated with the Visual Analog Scale (VAS). Adverse events and drug concentrations throughout the procedure were retrieved.

RESULTS

Our study population included 166 patients. They underwent laparoscopic unilateral versus bilateral salpingo-oophorectomy, ovarian cystectomy, bilateral salpingo-oophorectomy and omentectomy for a borderline ovarian tumor, myomectomy; or underwent surgery for unexplained infertility evaluation, pelvic pain, staging of ovarian cancer. Mean duration of pneumoperitoneum was 22.3 ± 7.2 min. Rate of conversion to laparoscopy under general anesthesia was 17/166 (10.2%) and there were only 3 cases of patients with low tolerability to the procedure. No severe adverse events occurred. Hospital discharge occurred in all unconverted cases after 6 to 18 h.

CONCLUSIONS

Operative laparoscopy under conscious sedation and local anesthesia appears to be a feasible technique in gynecologic surgery with no adverse patient outcomes.

Age progression from vicenarians (20-29 year) to nonagenarians (90-99 year) among a population pharmacokinetic/pharmacodynamic (PopPk-PD) covariate analysis of propofol-bispectral index (BIS) electroencephalography

BACKGROUND

Pharmacokinetic/pharmacodynamic (PK/PD) modeling has made an enormous contribution to intravenous anesthesia. Because of their altered physiological, pharmacological and pathological aspects, titrating general anesthesia in the elderly is a challenging task.

METHODS

Eighty patients were consecutively enrolled divided by decades from vicenarians (20-29 year) to nonagenarians (90-99 year) into eight groups. Using target controlled infusion (TCI) and electroencephalographic (EEG)-derived bispectral index (BIS) we set propofol plasma concentration (C_p) to gradually reach 3.5 μg mL^-1 over 3.5-min. In each patient, we constructed a PK/PD model and conducted a population PK/PD (PopPK-PD) covariate analysis.

RESULTS

Age was significant covariate for baseline BIS effect (E₀), inhibitory propofol concentration at 50% BIS decline (IC₅₀) and maximum BIS decline (E_max). First-order rate constant K_e0 of 0.47 min^-1 in vicenarians (20-29 year) gradually increased with age-progression to 1.85 min^-1 in nonagenarians (90-99 year). Simulation modelling showed that clinically recommended C_p of 3.5 μg mL^-1 for 20-29 year BIS 50 should be reduced to 3.0 for 30-49 year, 2.5 for 50-69 year and 2.0 for 80-89 year.

CONCLUSION

We quantified and graded EEG-BIS age-progression among different age groups divided by decades. We demonstrated deeper BIS values with decades' age progression. Our data has important implications for propofol dosing. The practical information for physicians in their daily clinical practice is using propofol C_p of 3.5 μg mL^-1 might not yield BIS value of 50 in elderly patients. Our simulations showed that the recommended regimen of C_p 3.5 μg mL^-1 for 20-29 year should be gradually decreased to 2.0 μg mL^-1 for 80-89 year.

CLINICAL TRIAL REGISTRY NUMBERS

European Community Clinical Trials Database EudraCT (http://eudract.emea.eu) initial trial registration number: 2011-002847-81, and subsequently registered at www.clinicaltrials.gov; trial registration number: NCT02585284. Xijing Hospital of Fourth Military Medical University ethics committee approval number 20110707-4.

Effect of Hemodynamic Changes in Plasma Propofol Concentrations Associated with Knee-Chest Position in Spinal Surgery: A Prospective Study

Background

Anesthesia induction and maintenance with propofol can be guided by target-controlled infusion (TCI) systems using pharmacokinetic (Pk) models. Physiological variables, such as changes in cardiac output (CO), can influence propofol pharmacokinetics. Knee-chest (KC) surgical positioning can result in CO changes.

Objectives

This study aimed to evaluate the relationship between propofol plasma concentration prediction and CO changes after induction and KC positioning.

Methods

This two-phase prospective cohort study included 20 patients scheduled for spinal surgery. Two different TCI anesthesia protocols were administered after induction. In phase I (n = 9), the loss of consciousness (LOC) concentration was set as the propofol target concentration and CO changes following induction and KC positioning were quantified. In phase II (n = 11), based on data from phase I, two reductions in the propofol target concentration on the pump were applied after LOC and before KC positioning. Propofol plasma concentrations were measured at different moments in both phases: after induction and after KC positioning.

Results

Schnider Pk model showed a good performance in predicting propofol concentration after induction; however, after KC positioning, when a significant drop in CO occurred, the measured propofol concentrations were markedly underestimated. Intended reductions in the propofol target concentration did not attenuate HD changes. In the KC position, there was no correlation between the propofol concentration estimated by the Pk model and the measured concentration in plasma, as the latter was much higher (P = 0.013) while CO and BIS decreased significantly (P < 0.001 and P = 0.004, respectively).

Conclusions

Our study showed that the measured propofol plasma concentrations during the KC position were significantly underestimated by the Schnider Pk model and were associated with significant CO decrease. When placing patients in the KC position, anesthesiologists must be aware of pharmacokinetic changes and, in addition to standard monitoring, the use of depth of anesthesia and cardiac output monitors may be considered in high-risk patients.

Are opioids indispensable for general anaesthesia?

The drug-induced, reversible coma of anaesthesia requires three clinical outcomes: unconsciousness, immobility, and the control of autonomic nervous system (ANS) responses to surgical stimulation. Producing the anaesthetised state with a single anaesthetic agent, such as an inhaled vapour or propofol, is challenging, primarily because suppressing ANS responses requires very high anaesthetic concentrations, resulting in haemodynamic depression and prolonged recovery. The antinociceptive effects of opioids (i.e. minimum alveolar concentration reduction) are thus central to the well-entrenched 'balanced anaesthesia' concept. In recent years, the notion of 'multimodal general anaesthesia' has extended the concept of balanced anaesthesia to include more drugs that target different neuroanatomical circuits and multiple neurophysiologic mechanisms. The opioid epidemic has provided some of the motivation to move away from opioids toward other adjunct drugs. Persistent opioid use after surgery is a component of the opioid epidemic and is a major concern for perioperative physicians. Potential solutions to the problem of persistent opioid use after surgery have focused on proper 'opioid stewardship' after operation, wherein opioids are used conservatively in combination with other analgesic adjuncts, and excessive opioid prescribing for home use is avoided. But there is a paucity of data on how intraoperative opioid usage patterns may be contributing to persistent opioid use after surgery. There are cogent reasons to moderate perioperative opioid use, including intraoperative opioids, but whether these changes in practice integral to the multimodal general anaesthesia concept will improve anaesthesia outcomes, including persistent opioid use after surgery, is unknown. Studies investigating these issues are an important research priority.

Event-Based control of depth of hypnosis in anesthesia

BACKGROUND AND OBJECTIVE

In this paper, we propose the use of an event-based control strategy for the closed-loop control of the depth of hypnosis in anesthesia by using propofol administration and the bispectral index as a controlled variable.

METHODS

A new event generator with high noise-filtering properties is employed in addition to a PIDPlus controller. The tuning of the parameters is performed off-line by using genetic algorithms by considering a given data set of patients.

RESULTS

The effectiveness and robustness of the method is verified in simulation by implementing a Monte Carlo method to address the intra-patient and inter-patient variability. A comparison with a standard PID control structure shows that the event-based control system achieves a reduction of the total variation of the manipulated variable of 93% in the induction phase and of 95% in the maintenance phase.

CONCLUSIONS

The use of event based automatic control in anesthesia yields a fast induction phase with bounded overshoot and an acceptable disturbance rejection. A comparison with a standard PID control structure shows that the technique effectively mimics the behavior of the anesthesiologist by providing a significant decrement of the total variation of the manipulated variable.

Optimized PID control of depth of hypnosis in anesthesia

BACKGROUND AND OBJECTIVE

This paper addresses the use of proportional-integral-derivative controllers for regulating the depth of hypnosis in anesthesia by using propofol administration and the bispectral index as a controlled variable. In fact, introducing an automatic control system might provide significant benefits for the patient in reducing the risk for under- and over-dosing.

METHODS

In this study, the controller parameters are obtained through genetic algorithms by solving a min-max optimization problem. A set of 12 patient models representative of a large population variance is used to test controller robustness. The worst-case performance in the considered population is minimized considering two different scenarios: the induction case and the maintenance case.

RESULTS

Our results indicate that including a gain scheduling strategy enables optimal performance for induction and maintenance phases, separately. Using a single tuning to address both tasks may results in a loss of performance up to 102% in the induction phase and up to 31% in the maintenance phase. Further on, it is shown that a suitably designed low-pass filter on the controller output can handle the trade-off between the performance and the noise effect in the control variable.

CONCLUSIONS

Optimally tuned PID controllers provide a fast induction time with an acceptable overshoot and a satisfactory disturbance rejection performance during maintenance. These features make them a very good tool for comparison when other control algorithms are developed.

Altered pharmacodynamics and pharmacokinetics of cisatracurium in patients with severe mitral valve regurgitation during anaesthetic induction period

AIM

The aim of the current study was to characterize the pharmacokinetics (PK) and pharmacodynamics (PD) of cisatracurium in patients with severe mitral valve regurgitation (MR) during the anaesthetic induction period.

METHODS

Thirty patients in the clinical trial were divided into two groups: the MR group (n = 15) and the control group (n = 15). Arterial blood samples were obtained before (time 0) and at 1, 2, 4, 6, 8, 10, 15 and 20 min after intravenous injection of 0.15 mg kg^-1 cisatracurium. The degree of neuromuscular block was measured by train of four (TOF) testing. The concentration of cisatracurium in the plasma was determined by high-performance liquid chromatography. A conventional two-compartment model and integrated PK/PD model were applied to PK and PD data analysis, respectively.

RESULTS

The results of PK model fitting demonstrated that severe MR reduced the distribution rate of cisatracurium from the central to peripheral compartment, resulting in a higher concentration of the drug in the plasma. The time to the maximal neuromuscular blocking effect of cisatracurium was delayed in the MR group (2.08 min in the control group vs. 4.12 min in the MR group). The PK/PD model indicated that the distribution rate of cisatracurium from the blood to the effect compartment was decreased in the MR group.

CONCLUSIONS

The present study suggested that the PK and PD of cisatracurium were significantly altered in patients with severe MR. The study has the potential to improve the safety of anaesthetic induction in patients with severe MR through accurate prediction of the PD responses of cisatracurium using the established PK/PD model.

Patient-controlled Analgesia with Target-controlled Infusion of Hydromorphone in Postoperative Pain Therapy

Background

Patient-controlled analgesia (PCA) is a common method for postoperative pain therapy, but it is characterized by large variation of plasma concentrations. PCA with target-controlled infusion (TCI-PCA) may be an alternative. In a previous analysis, the authors developed a pharmacokinetic model for hydromorphone. In this secondary analysis, the authors investigated the feasibility and efficacy of TCI-PCA for postoperative pain therapy with hydromorphone.

Methods

Fifty adult patients undergoing cardiac surgery were enrolled in this study. Postoperatively, hydromorphone was applied intravenously during three sequential periods: (1) as TCI with plasma target concentrations of 1 to 2 ng/ml until extubation; (2) as TCI-PCA with plasma target concentrations between 0.8 and 10 ng/ml during the following 6 to 8 h; and (3) thereafter as PCA with a bolus dose of 0.2 mg until the next morning. During TCI-PCA, pain was regularly assessed using the 11-point numerical rating scale (NRS). A pharmacokinetic/pharmacodynamic model was developed using ordinal logistic regression based on measured plasma concentrations.

Results

Data of 43 patients aged 40 to 81 yr were analyzed. The hydromorphone dose during TCI-PCA was 0.26 mg/h (0.07 to 0.93 mg/h). The maximum plasma target concentration during TCI-PCA was 2.3 ng/ml (0.9 to 7.0 ng/ml). The NRS score under deep inspiration was less than 5 in 83% of the ratings. Nausea was present in 30%, vomiting in 9%, and respiratory insufficiency in 5% of the patients. The EC50 of hydromorphone for NRS of 4 or less was 4.1 ng/ml (0.6 to 12.8 ng/ml).

Conclusion

TCI-PCA with hydromorphone offered satisfactory postoperative pain therapy with moderate side effects.

A Pharmacokinetics-Neural Mass Model (PK-NMM) for the Simulation of EEG Activity during Propofol Anesthesia

Modeling the effects of anesthetic drugs on brain activity is very helpful in understanding anesthesia mechanisms. The aim of this study was to set up a combined model to relate actual drug levels to EEG dynamics and behavioral states during propofol-induced anesthesia. We proposed a new combined theoretical model based on a pharmacokinetics (PK) model and a neural mass model (NMM), which we termed PK-NMM--with the aim of simulating electroencephalogram (EEG) activity during propofol-induced general anesthesia. The PK model was used to derive propofol effect-site drug concentrations (C(eff)) based on the actual drug infusion regimen. The NMM model took C(eff) as the control parameter to produce simulated EEG-like (sEEG) data. For comparison, we used real prefrontal EEG (rEEG) data of nine volunteers undergoing propofol anesthesia from a previous experiment. To see how well the sEEG could describe the dynamic changes of neural activity during anesthesia, the rEEG data and the sEEG data were compared with respect to: power-frequency plots; nonlinear exponent (permutation entropy (PE)); and bispectral SynchFastSlow (SFS) parameters. We found that the PK-NMM model was able to reproduce anesthesia EEG-like signals based on the estimated drug concentration and patients' condition. The frequency spectrum indicated that the frequency power peak of the sEEG moved towards the low frequency band as anesthesia deepened. Different anesthetic states could be differentiated by the PE index. The correlation coefficient of PE was 0.80 ± 0.13 (mean ± standard deviation) between rEEG and sEEG for all subjects. Additionally, SFS could track the depth of anesthesia and the SFS of rEEG and sEEG were highly correlated with a correlation coefficient of 0.77 ± 0.13. The PK-NMM model could simulate EEG activity and might be a useful tool for understanding the action of propofol on brain activity.

The optimal effect-site concentration of sufentanil for laryngeal mask insertion during induction with target-controlled propofol infusion at 4.0 μg/mL

Objective:

The objective of this study is to determine the optimal effect-site concentration (Ce) of sufentanil for satisfactory insertion of laryngeal mask airway (LMA) when administered with a target-controlled infusion (TCI) of propofol at 4.0 μg/mL.

Materials and Methods:

A total of 25 adult patients scheduled for minor elective surgery were enrolled in this study. All patients received induction with a combination of propofol and sufentanil TCI. The TCI of sufentanil was started at a target Ce of 0.1 ng/mL. After equilibrium with the plasma concentration, the TCI of propofol was initiated, targeting a preset Ce of 4.0 μg/mL. After the loss of consciousness, LMA was inserted and assessed by an experienced Anesthesiologist. The Ce of sufentanil for the next patient was guided by modified Dixon's up-and-down method using 0.05 ng/mL as a step size. The Ce of sufentanil required for successful LMA insertion in 50% of adults (EC50) was determined by calculating the midpoint concentration of all independent pairs of patients after at least seven crossover points.

Results:

The optimal Ce (EC50) of sufentanil for LMA insertion during propofol induction using target Ce of 4 μg/mL was 0.16 ng/mL (95% confidence interval [CI] = 0.12-0.20). There was a significant reduction in propofol induced pain score P = 0.0275 and insignificant hemodynamic changes.

Conclusion:

Ce of sufentanil required for successful LMA insertion in 50% of patients (EC50) using propofol target Ce of 4.0 μg/mL was 0.16 ng/mL (95% CI = 0.12-0.20) with a significant reduction in the propofol induced pain and hemodynamic stability.

Population Pharmacokinetic Modeling of Hydromorphone in Cardiac Surgery Patients during Postoperative Pain Therapy

Background:

Hydromorphone is a µ-selective opioid agonist used in postoperative pain therapy. This study aimed to evaluate the pharmacokinetics of hydromorphone in cardiac surgery patients during postoperative analgesia with target-controlled infusion and patient-controlled analgesia.

Methods:

In this study, 50 adult patients were enrolled to receive intravenous hydromorphone during postoperative pain therapy. Arterial plasma samples were collected for measurements of drug concentration. Population pharmacokinetic parameters were estimated using nonlinear mixed-effects modeling. Results were validated and simulations were carried out to evaluate results.

Results:

Data from 49 patients (age range, 40–81 yr) were analyzed. The pharmacokinetics of hydromorphone were best described by a three-compartment model. Age was incorporated as a significant covariate for elimination clearance and central volume of distribution. Scaling all parameters with body weight improved the model significantly. The final estimates of the model parameters for the typical adult patient (67 yr old, weighing 70 kg) undergoing cardiac surgery were as follows: CL1 = 1.01 l/min, V1 = 3.35 l, CL2 = 1.47 l/min, V2 = 13.9 l, CL3 = 1.41 l/min, and V3 = 145 l. The elimination clearance decreased by 43% between the age of 40 and 80 yr, and simulations demonstrated that context-sensitive half-time increased from 26 to 84 min in 40- and 80-yr-old subjects, respectively.

Conclusions:

The final pharmacokinetic model gave a robust representation of hydromorphone pharmacokinetics. Inclusion of age and body weight to the model demonstrated a significant influence of these covariates on hydromorphone pharmacokinetics. The application of this patient-derived population model in individualized pain therapy should improve the dosing of hydromorphone in patients undergoing cardiac surgery.

Pharmacokinetic-pharmacodynamic modeling in acute and chronic pain: an overview of the recent literature

In acute and chronic pain, the objective of pharmacokinetic-pharmacodynamic (PKPD) modeling is the development and application of mathematical models to describe and/or predict the time course of the pharmacokinetics (PK) and pharmacodynamics (PD) of analgesic agents and link PK to PD. Performing population PKPD modeling using nonlinear mixed effects modeling allows, apart from the estimation of fixed effects (the PK and PD model estimates), the quantification of random effects as within- and between-subject variability. Effect-compartment models and mechanism-based biophase distribution models that incorporate drug-association and -dissociation kinetics are applied in PKPD modeling of pain treatment. Mechanism-based models enable the quantification of the rate-limiting factors in drug effect owing to drug distribution versus receptor kinetics (since receptor kinetics are nonlinear they are discernable from the linear effect-compartment kinetics). It is a helpful technique in understanding the complex behavior of specific analgesics, such as buprenorphine, but also morphine and its active metabolite morphine-6-glucuronide, especially with respect to the reversal of opioid-induced side effects, most importantly life-threatening respiratory depression. One approach in chronic pain studies is the application of mixture models. Mixture models do not necessarily need to take PK data into account and allow the objective differentiation of measured responses to analgesics into specific response subgroups, and as such, may play an important role in analyzing Phase I and II analgesia studies. Appropriate application of PKPD modeling leads to the improvement of current therapeutics with respect to dose design and outcome, understanding the interaction of analgesics within complex chronic pain disease processes and may play an important role in drug development. In the current article, novel observations using the aforementioned techniques on opioids, NSAIDs, epidural analgesia, ketamine and GABA-ergic drugs in acute and chronic pain are discussed.

Allometric or lean body mass scaling of propofol pharmacokinetics: towards simplifying parameter sets for target-controlled infusions

Uncertainty exists as to the most suitable pharmacokinetic parameter sets for propofol target-controlled infusions (TCI). The pharmacokinetic parameter sets currently employed are clearly not universally applicable, particularly when patient attributes differ from those of the subjects who participated in the original research from which the models were derived. Increasing evidence indicates that the pharmacokinetic parameters of propofol can be scaled allometrically as well as in direct proportion to lean body mass (LBM). Appraisal of hitherto published studies suggests that an allometrically scaled pharmacokinetic parameter set may be applicable to a wide range of patients ranging from children to obese adults. On the other hand, there is evidence that propofol pharmacokinetic parameters, scaled linearly to LBM, provide improved dosing in normal and obese adults. The 'Schnider' pharmacokinetic parameter set that has been programmed into commercially available TCI pumps cannot be employed at present for morbidly obese patients (body mass index >40 kg/m2), because of anomalous behaviour of the equation used to calculate LBM, resulting in administration of excessive amounts of propofol. Simulations of TCI using improved equations to calculate LBM indicate that the Schnider model delivers similar amounts of propofol to morbidly obese patients as do the allometrically scaled pharmacokinetic parameter sets. These hypotheses deserve further investigation. To facilitate further investigation, researchers are encouraged to make their data freely available to the WorldSIVA Open TCI Initiative (http://opentci.org).

Rational opioid dosing in the elderly: dose and dosing interval when initiating opioid therapy

Opioids are the mainstay of treatment for moderate to severe pain. However, opioid therapy in the elderly is often associated with significant morbidity because of excessive ventilatory depression. The large amount of interindividual variability in opioid dose-response relationships makes it difficult to individualize the dose and dosing interval to provide safe and effective analgesia. By examining how aging affects the pharmacokinetics (PK) and pharmacodynamics (PD) of opioids, it is possible to provide a rational basis for age adjustment in opioid dosing.

The optimal effect-site concentration of remifentanil for lightwand tracheal intubation during propofol induction without muscle relaxation

STUDY OBJECTIVE

To determine the most suitable effect-site concentration of remifentanil during lightwand intubation when administered with a target-controlled infusion (TCI) of propofol at 4.0 μg/mL without neuromuscular blockade.

DESIGN

Prospective study using a modified Dixon's up-and-down method.

SETTING

Operating room of an academic hospital.

PATIENTS

28 ASA physical status 1 and 2 patients, aged 18-65 years, scheduled for minor elective surgery.

INTERVENTIONS

Anesthesia was induced by TCI propofol effect-site concentration to 4.0 μg/mL, and the dose of remifentanil given to each patient was determined by the response of the previously tested patient using 0.2 ng/mL as a step size. The first patient was tested at a target effect-site concentration of 4.0 ng/mL of remifentanil. If intubation was successful, the remifentanil dose was decreased by 0.2 ng/mL; if it failed, the remifentanil dose was increased by 0.2 ng/mL. Successful intubation was defined as excellent or good intubating conditions.

MEASUREMENTS AND MAIN RESULTS

The remifentanil effect-site concentration was measured. The optimal effect-site concentration of remifentanil for lightwand tracheal intubation during propofol induction using 2% propofol target effect-site concentration to 4 μg/mL was 2.16 ± 0.19 ng/mL. From probit analysis, the effect-site concentration of remifentanil required for successful lightwand intubation in 50% (EC50) and 95% (EC95) of adults was 2.11 ng/mL (95% CI 1.16-2.37 ng/mL) and 2.44 ng/mL (95% CI 2.20-3.79 ng/mL), respectively.

CONCLUSION

A remifentanil effect-site concentration of 2.16 ± 0.19 ng/mL given before a propofol effect-site concentration of 4 μg/mL allowed lightwand intubation without muscle relaxant.

Designing simple PK-PD studies in children

Conducting clinical pharmacology research studies in pediatric patients is challenging because of ethical and practical constraints but necessary to ensure that drugs are used safely and effectively in this population. Developments in laboratory analytical techniques, such as improved assay sensitivity and the use of alternative sample matrices, can reduce blood loss and offer less invasive blood sampling, causing less trauma to the patient and fewer ethical concerns. Recent advances in data analysis techniques, which aim to extract the maximum amount of useful information from small sample numbers, should be considered when planning a clinical trial and incorporated into the study design. Using 'population' methodology allows a more flexible sampling strategy that enables valuable data to be collected in the course of routine clinical practice, rather than in a rigid, and potentially artificial, setting. Integration of pharmacokinetics and pharmacodynamics and the application of physiological approaches and simulation techniques to the analysis and interpretation of drug concentration and effect data offer new opportunities that have particular relevance to pharmacological research in the field of pediatric anesthetics.

Population pharmacokinetics and pharmacodynamics in anesthesia, intensive care and pain medicine

PURPOSE OF REVIEW

Population modeling is a relatively new pharmacological discipline, the development of which has largely been stimulated by the need for accurate models for the pharmacokinetics and dynamics of anesthetic agents.

RECENT FINDINGS

Population-based modeling is now considered superior to older, more traditional modeling methods. Nonlinear mixed-effect modeling - a commonly used population-based modeling approach - estimates intraindividual and interindividual variability, limits the influence of outlying samples and individuals through the use of Bayesian statistical analysis, and provides a potential means of optimizing drug delivery regimens, especially when used to define pharmacokinetic-dynamic models for target-controlled infusion systems. In addition to being used for pharmacokinetic modeling, in which the influence of factors such as age, weight and illness can be studied, it is a powerful tool for the study of the influence of multiple factors on drug pharmacodynamics.

SUMMARY

Nonlinear mixed-effect population-based modeling has become the gold standard method of pharmacokinetic and pharmacodynamic analysis during new drug development and during subsequent pharmacological studies. Population-based modeling techniques have been applied to numerous aspects of drug delivery in anesthesia, intensive care and pain medicine.

Pharmacokinetics and pharmacodynamics of the short-acting sedative CNS 7056 in sheep

BACKGROUND

CNS 7056 is a new short-acting esterase-metabolized benzodiazepine. We report the first pharmacokinetic (PK) and pharmacodynamic (PD) study of CNS 7056 and its inactive metabolite CNS 7054 in sheep.

METHODS

The stability of CNS 7056 in blood samples was examined ex vivo. Six sheep were prepared with physiological instrumentation, and were given doses of 0.37, 0.74, and 1.47 mg kg(-1) (2 min infusion) of CNS 7056 in alternating order on separate days.

RESULTS

CNS 7056 was degraded in warm whole sheep blood (23% over 2 h), but not in plasma or blood stored on ice. Using non-compartmental analysis (NCA), CNS 7056 had a mean (sd) clearance of 4.52 (0.96) litre min(-1) and a terminal half-life of 21.3 (10.9) min. There was a rapid conversion of CNS 7056 to its metabolite CNS 7054, which had a terminal half-life of 22.5 (3.4) min. The arterial kinetics of CNS 7056 could be described by a three-compartment model, with volumes of 1.9, 3.9, and 79 litre, a clearance of 4.2 litre min(-1), and inter-compartmental clearances of 2.85 and 1.44 litre min(-1), while the metabolite could be described by a two-compartment model. Cardiac output was an important covariate. Sedation as measured by the alpha power band of the EEG showed rapid onset and offset. The t(1/2,)(k)(e0) for sedation was 1.78 min, and the EC(50) was 0.10 µg ml(-1).

CONCLUSIONS

CNS 7056 has PK-PD properties compatible with its potential human use as a short-acting i.v. sedative.

Beyond Genetics-Stratified and Personalised Medicines Using Multiple Parameters

Prescribers have been practicing stratified medicine for many years. Patient characteristics, usually non-genetic, including age, comorbidities and concomitant medications are taken into account when deciding which drug to prescribe. In addition, the majority of drugs require dose adjustments across patient subgroups, usually determined by non-genetic differences between the subgroups. Whilst pharmacogenetics hold promise for enhancing treatment stratification and even treatment individualisation, non-genetic factors will continue to be very important. Both non-genetic and genetic factors must be considered to improve understanding and quantification of the variability in treatment outcomes and to guide stratification and targeting of patient subgroups to the right drug and also to the right range of doses within that subgroup. Development of stratified medicines must consider non-genetic as well as genetic factors and, where appropriate, include stratification through optimising the dose for each patient or subgroup as well as by choosing the drug most likely to deliver efficacy to that patient or group.

A two-compartment effect site model describes the bispectral index after different rates of propofol infusion

Different estimates of the rate constant for the effect site distribution (k(e0)) of propofol, depending on the rate and duration of administration, have been reported. This analysis aimed at finding a more general pharmacodynamic model that could be used when the rate of administration is changed during the treatment. In a cross-over study, 21 healthy volunteers were randomised to receive a 1 min infusion of 2 mg/kg of propofol at one occasion, and a 1 min infusion of 2 mg/kg of propofol immediately followed by a 29 min infusion of 12 mg kg(-1) h(-1) of propofol at another occasion. Arterial plasma concentrations of propofol were collected up to 4 h after dosing, and BIS was collected before start of infusion and until the subjects were fully awake. The population pharmacokinetic-pharmacodynamic analysis was performed using NONMEM VI. A four-compartment PK model with time-dependent elimination and distribution described the arterial propofol concentrations, and was used as input to the pharmacodynamic model. A standard effect compartment model could not accurately describe the delay in the effects of propofol for both regimens, whereas a two-compartment effect site model significantly improved the predictions. The two-compartment effect site model included a central and a peripheral effect site compartment, possibly representing a distribution within the brain, where the decrease in BIS was linked to the central effect site compartment concentrations through a sigmoidal E(max) model.

Pharmacokinetics and pharmacodynamics of a new reformulated microemulsion and the long-chain triglyceride emulsion of propofol in beagle dogs

BACKGROUND AND PURPOSE

Microemulsion propofol was developed to eliminate lipid solvent-related adverse events of long-chain triglyceride emulsion (LCT) propofol. We compared dose proportionality, pharmacokinetic and pharmacodynamic characteristics of both formulations.

EXPERIMENTAL APPROACH

The study was a randomized, two-period and crossover design with 7-day wash-out period. Microemulsion and LCT propofol were administered by zero-order infusion (0.75, 1.00 and 1.25 mg kg(-1) min(-1)) for 20 min in 30 beagle dogs (male/female = 5/5 for each rate). Arterial samples were collected at preset intervals. The electroencephalographic approximate entropy (ApEn) was used as a measure of propofol effect. Dose proportionality, pharmacokinetic and pharmacodynamic bioequivalence were evaluated by non-compartmental analyses. Population analysis was performed using nonlinear mixed effects modelling.

KEY RESULTS

Both formulations showed dose proportionality at the applied dose range. The ratios of geometric means of AUC(last) and AUC(inf) between both formulations were acceptable for bioequivalence, whereas that of C(max) was not. The pharmacodynamic bioequivalence was indicated by the arithmetic means of AAC (areas above the ApEn time curves) and E(0) (baseline ApEn)-E(max) (maximally decreased ApEn) between both formulations. The pharmacokinetics of both formulations were best described by three compartment models. Body weight was a significant covariate for V(1) of both formulations and sex for k(21) of microemulsion propofol. The blood-brain equilibration rate constants (k(e0), min(-1)) were 0.476 and 0.696 for microemulsion and LCT propofol respectively.

CONCLUSIONS AND IMPLICATIONS

Microemulsion propofol was pharmacodynamically bioequivalent to LCT propofol although pharmacokinetic bioequivalence was incomplete, and demonstrated linear pharmacokinetics at the applied dose ranges.

Merging PK/PD information in a minimally parameterized model of the neuromuscular blockade

A recursive system identification algorithm that merges PK/PD information in a minimally parameterized Wiener model for the NMB level is presented. The results show that the coupling between one parameter from the linear block and one from the static nonlinearity is advantageous, when evaluated on a database of 60 real collected NMB cases.

Pharmacokinetic-pharmacodynamic modeling in anesthesia, intensive care and pain medicine

PURPOSE OF REVIEW

Studies from the anesthesiology literature published in the last 2 years were selected to illustrate the most important developments in the field of pharmacokinetic-pharmacodynamic modeling.

RECENT FINDINGS

The pharmacokinetic models focused on incorporating covariate, especially age for pediatric-geriatric use, and altered physiological states. The pharmacodynamic models studied the effect of rate of anesthetic administration, age, experimental conditions, and delay within the monitor on estimation of drug concentration in the biophase. Models for the surrogate measure of the components of general anesthesia, hypnosis (bispectral index scale, entropy), immobility (limb tetanic stimulus-induced withdrawal reflex) and antinociception (surgical stress index, skin conductance algesimeter) were developed and validated. Response surface models were used to study drug interactions for important end-points during surgery and also to optimize dosing of anesthetic agents to maximize the desired/undesired effect ratio. The models for target-controlled infusions were improved by incorporating more covariates, and the closed-loop system was refined by using adaptive controllers that individualize the pharmacokinetic/pharmacodynamic parameters to the particular patient by using Bayesian, Kalman filters, fuzzy logic or neural networks.

SUMMARY

Progress was made by improving population pharmacokinetic/pharmacodynamic models, developing new indexes to measure drug effect and using them in an adaptive delivery system to the individual patient.

Effect of ketamine on the limb withdrawal reflex evoked by transcutaneous electrical stimulation in ponies anaesthetised with isoflurane

The purpose of this study was to evaluate the anti-nociceptive activity of ketamine and isoflurane in horses using a limb withdrawal reflex (WR) model. Single and repeated stimulations were applied to the digital nerve of the left forelimb in ponies anaesthetised with isoflurane before, during and after intravenous administration of racemic ketamine. Surface electromyographic activity was recorded from the deltoid muscle. Higher stimulation intensity was required to evoke a reflex during ketamine administration. Furthermore, the amplitudes of response to stimulations were significantly and dose-dependently depressed and a flattening of the stimulus-response curves was observed. The reflex activity recovered partially once the ketamine infusion finished. The results demonstrated that the limb WR can be used to quantify the temporal effect of ketamine on the sensory-motor processing in ponies anaesthetised with isoflurane.