Nicolás Marco-Ariño (1,2), Itziar Irurzun-Arana (1,2), Sérgio Vide (3), Sebastian Jaramillo (4), Pedro L Gambús (4), Iñaki F. Trocóniz (1,2)
(1) Pharmacometrics and Systems Pharmacology, Departament of Pharmacy and Pharmaceutical Technology, School of Pharmacy, University of Navarra, Pamplona, Spain. (2) IdiSNA; Navarra Institute for Health Research, Pamplona, Spain (3) Center for Clinical Research in Anesthesia, Serviço de Anestesiologia, Centro Hospitalar do Porto, Porto, Portugal. (4) Systems Pharmacology Effect Control and Modeling Research Group, Department of Anesthesia, Hospital CLINIC de Barcelona, Barcelona, Spain.
Introduction and objectives:
Movement is an extensively characterised response to noxious stimulus. However, its applicability to assess intraoperative pain has limitations due to the effect of anaesthesia in the response. Previous research has shown that Pupillary Dilation Reflex (PDR), an indicator of anaesthetic depth, could also predict movement after noxious stimuli1,2. The objective of this project is to expand a previously developed model for the pupil effects of remifentanil and characterise the effect of propofol and remifentanil on movement response after nociceptive stimulus, under the hypothesis that both responses are governed by the same physiological mechanisms.
Data overview and methods:
Eighty-seven female patients undergoing gynaecological surgery were recruited for the study. Exclusion criteria included ocular diseases, prescription of drugs affecting the size or reflex of the pupil and morbid obesity (IMC> 35). Pupil diameter was measured multiple times before and after surgery using the AlgiScan (IDMED™) hand-held pupillometer which was also used to delivered a 60 mA tetanic stimulus during 5 seconds in the forearm of the patient. Movement response to the tetanic stimulus was evaluated in a categorical scale ranging from 0 (absence) to 3 (strong movement) by the physicians. Propofol and remifentanil concentrations in plasma were predicted using the Schnider3 (propofol) and Minto4 (remifentanil) validated models for target control infusion (TCI). Each patient received a median of three stimuli before surgery and one after surgery. The first stimulus was performed in the presence of propofol and the rest in the presence of propofol and remifentanil. Pupil size and movement were recorded for thirteen seconds in each stimulus, resulting in 3846 (pupil) and 3838 (movement) observations. No data from basal pupil size in the absence of anaesthesia was available. Data were analysed using NONMEM 7.4.
Results:
Pupil diameter and movement were modelled separately. A semi-mechanistic two-compartment indirect response model accurately described the time course of the pupil diameter. In this model, the administration of the tetanic stimulus modulates a theoretical nociceptive which subsequently controls the turnover of the pupil diameter. No effect of propofol, on pupil size could be detected, while remifentanil modulated the system by attenuating the perception of the nociceptive stimulus and increasing the turnover of the nociceptive compartment. Precision of model parameter was estimated accurately (RSE<45%) and evaluated with bootstrapping techniques. Medium to high inter-individual variability was observed. Model performance was based on goodness of fit plots and visual predictive checks. Covariates (age, weight and height) did not show correlation with model parameters and therefore were not included in the model. Following a hypothesis-driven exercise, the structure and empirical Bayesian estimates obtained from the pupil model were used to characterise the movement response. In this approach, the logit function defining the probability of each movement grade is proportional to the levels of the nociceptive compartment estimated in the pupil model. Logistic regression resulted in a poor description of the movement data and an over-prediction of the number of transitions between grades. Conversely, a discrete-time Markov with a proportional odds model provided a good description of the data with high parameter precision (RSE<20%) and a number of transitions that matched the observations. In absence of stimulus, probabilities for movement grades 0 to 3 were estimated in 94, 3, 2 and 1% respectively. Changes in the nociceptive compartment were incorporated into the logit via a scaling factor (4.56). Transition probabilities from movement grades 1, 2 and 3 to absence of movement were 4.8, 9.5 and 7.7% and provided a good description of the transitions between movement grades across the different stimulus.
Conclusions:
A semi-mechanistic pharmacodyamic model to describe pupil size and movement response after a noxious stimulus in the presence of propofol and remifentanil was successfully implemented. This model structure supports the hypothesis that both responses share a common physiological mechanism of control. Future work will focus on investigating at an individual level the correlation between the magnitude of pupil change and the probability of movement.
References:
[1] Leslie, K. et al. Prediction of movement during propofol/nitrous oxide anesthesia: Performance of concentration, electroencephalographic, pupillary, and hemodynamic indicators. Anesthesiology 84, 52–63 (1996).
[2] Funcke, S. et al. Validation of innovative techniques for monitoring nociception during general anesthesia: A clinical study using tetanic and intracutaneous electrical stimulation. Anesthesiology 127, 272–283 (2017).
[3] Schnider, T. W. et al. The influence of age on propofol pharmacodynamics. Anesthesiology 90, 1502–16 (1999).
[4] Minto, C. F., Schnider, T. W. & Shafer, S. L. Pharmacokinetics and pharmacodynamics of remifentanil. II. Model application. Anesthesiology 86, 24–33 (1997).
Reference: PAGE () Abstr 9480 [www.page-meeting.org/?abstract=9480]
Poster: Drug/Disease Modelling - CNS