An introduction to predictive modelling of drug concentration in anaesthesia monitors

Abstract

Summary A significant amount of anaesthetists’ work involves the prediction of drug effects and interactions to produce a smooth general anaesthetic that minimises drug side effects and promotes rapid emergence. Successfully managing this process requires a basic understanding of drug effects, experience and inevitably some guesswork, since it is difficult (and in some cases impossible) to anticipate all relevant patient and surgical factors. Although data are generally available to allow calculation of plasma drug and effect site concentrations, this is often difficult to apply in complex clinical contexts, particularly when multiple drug types are used. In recent years, manufacturers have developed and incorporated into anaesthetic workstations technologies that use drug pharmacodynamic and pharmacokinetic data to predict drug effects and interactions. Such systems can predict the duration and effects of drugs during anaesthesia and assist the anaesthetist to understand complex drug interactions. With this information available, different drug types, doses and combinations may be tailored in a scientific way to maximise useful effects whilst minimising overdose and side‐effects, particularly in high‐risk patients. Examples are used to illustrate how such systems can be used in practice, and how drug effects and interactions can be simulated to “rehearse” an anaesthetic before any drugs are actually administered. At present only a small number of anaesthetic workstations use this technology, and as yet they are not able to manage all drugs used in anaesthetic practice. However, such systems have the potential to help anaesthetists manage the complexity of their work, and to provide information on predicted drug effects in a way that is useful and relevant to both experienced anaesthetists and trainees.

Big Data Analytics in Healthcare: A Review of Opportunities and Challenges

“…Although many prediction models have been created and verified on information of cohort researches, little attention has been paid to guarantee the reliability of a forecast for an individual, which is crucial for point of care judgments. Since the objective of prediction models is to forecast results in new individuals, a significant problem in prognostic investigation would be to determine what facts beyond validation are required before physicians can confidently use a model on their patient [72,73]. There are several main different arguments in applying ML in the biomedical and other fields.…”

Investigation of data-driven optical neuromonitoring approach during general anesthesia with sevoflurane

“…MAC can be a good indicator of the effect of anesthetic drugs during steady-state conditions, but during dynamic times the values may have significant lag. 1 Furthermore, the effect of other drugs such as opioids is not accounted in this number and each combination of drugs will have specific effects on analgesia. 1 In recent years, however, predictive modeling of drug concentrations in anesthesia monitors have been implemented to account for the effect of multiple drugs and these algorithms could be used to improve anesthesia monitoring in combination with fNIRS and EEG.…”

“…1 Furthermore, the effect of other drugs such as opioids is not accounted in this number and each combination of drugs will have specific effects on analgesia. 1 In recent years, however, predictive modeling of drug concentrations in anesthesia monitors have been implemented to account for the effect of multiple drugs and these algorithms could be used to improve anesthesia monitoring in combination with fNIRS and EEG. 1 BIS and MAC also have the advantage of providing a continuous index instead of binary classification provided by fNIRS-SVM.…”

“…Even though MAC is a widely used method, it is known that at MAC 1.00 only 50% of the population will not respond to painful stimuli and during nonsteady-state conditions the end-tidal estimates may lag. 1 Furthermore, the MAC value does not account for nonvolatile agents administered during surgery that could have an effect on pain and consciousness. 1 Since the anesthesiologist tool box does not contain information on the effects of the anesthetics on the brain, a method that directly measures brain activity would provide important data to help clinicians deliver individualized anesthetic care.…”

Influence of an “Electroencephalogram-Based” Monitor Choice on the Delay Between the Predicted Propofol Effect-Site Concentration and the Measured Drug Effect

“…This action can be quantified by a first-order process defined as hysteresis and characterized by effect-compartment equilibrium rate constant (k eo ). During total intravenous anesthesia (TIVA), drug Ce can be predicted using target-controlled infusion (TCI) technology [1][2][3] or anesthesia drug displays, 4 while achieved drug effect can be measured by various processed electroencephalographic (EEG) measures. 5 The concept of Ce takes into account this hysteresis, and as such, the time course of the Ce should match the time course of the drug effect.…”