Abstract:
The practice of anaesthesia involves creating profound drug effects (i.e. lack of awareness and sensation) and then reversing these rapidly. Endpoints are achieved by combining drugs that collectively contribute to desired effects. This requires an understanding of the pharmacological profile of drugs when given alone and alongside others. Pharmacokinetic (PK) and pharmacodynamic (PD) models summarise complex drug relationships so they may be better understood and capitalised on. I report my investigations into using PKPD models to describe combined drug responses for applications in anaesthesia. I investigated analgesia in children using paracetamol with two non-steroidal anti-inflammatory drugs. Diclofenac dosing in children is currently extrapolated from adult data. I modelled paracetamol and diclofenac analgesia using a modified EMAX model with placebo effects and interval censored dropout. Diclofenac and paracetamol effects were additive. Diclofenac analgesia had a maximal effect (EMAX) of 4.8 cm (visual analogue pain scale 0-10), an equilibration half-life 0.23 h and Ce50 of 1.20 mg.l-1 in children after adeno-tonsillectomy. The hazard of dropping out was 1.36 for each unit in pain. This interaction model can be used to guide combination diclofenac and paracetamol dosing in children, and future study design of analgesics in similar populations. I then developed a model describing processed electroencephalograph (EEG) response, using Bispectral Index (BIS), for multiple drug classes. I aimed to establish a model that could be used to simulate EEG response, as a cue of anaesthetic depth, in a human patient simulator (HPS). I developed a model for propofol-alfentanil BIS effects using response surface methodology, and established generalisability of this model to remifentanil. I investigated the use of a combined propofol-remifentanil PKPD model for BIS targeted infusion in surgical patients. Finally, I developed a model describing BIS effects for: propofol (an intravenous anaesthetic); midazolam (a sedative), alfentanil, fentanyl, remifentanil (phenyl-piperidine opioids); desflurane, isoflurane and sevoflurane (inhalation anaesthetics). This is the first study to describe propofol-midazolam BIS effects, which were additive. I discuss how the model might be used in the HPS with attention to creating realistic variability in response, and correlating BIS to manikin eye closure during induction. I anticipate that this work will improve our currently limited simulation of depth of anaesthesia, and enhance our anaesthetic-based simulations for students and experienced clinicians alike.