Maite Garraza-Obaldia (1,2), Sebastián Jaramillo (3), Zinnia P Parra-Guillén(1,2), José Fernando Valencia (4), Pedro L Gambús (3), Iñaki F Troconiz (1,2,5)
(1) Department of Pharmaceutical Sciences, School of Pharmacy and Nutrition, University of Navarra, Pamplona, Spain. (2) Navarra Institute for Health Research (IdiSNA), Pamplona, Spain. (3) Anesthesiology Department, Hospital Clinic de Barcelona, Barcelona, Spain. (4) Biomedical Engineering Program, Universidad de San Buenaventura, Cali, Colombia. (5) Institute of Data Science and Artificial Intelligence, DATAI, University of Navarra, Pamplona, Spain.
Introduction and Objectives:
Pulmonary ventilation is significantly affected during sedation, often leading to respiratory depression and the subsequent hypoxia. Changes in pulmonary ventilation during sedation are primarily driven by the effect of anaesthetics and surgical stimuli. However, the quantitative relationship between these variables and pulmonary ventilation remains unclear. Understanding the relationship between anaesthetics and pulmonary ventilation is essential to develop dosing strategies to avoid respiratory depression and its complications.
Previous pharmacodynamic models of pulmonary ventilation have focused on describing changes in partial pressure of CO2 (pCO2) [1]. Although pCO2 is clinically relevant, it does not adequately reflect pulmonary ventilation (minute-volume, or MV).
The aim of this study is to develop a population pharmacodynamic model to characterize the relationship between propofol and remifentanil concentrations and ventilation responses (MV and pCO2) in patients undergoing sedation with laryngeal mask placement, including the impact of the surgical stimulus within the quantitative framework.
Methods: Twenty female patients undergoing gynaecological surgical interventions were analysed. pCO2 was measured from the beginning, and after laryngeal mask placement, patient spontaneous ventilation was measured as MV. 4077 observations of MV and 5537 of pCO2 were recorded, with a granularity of ten seconds.
Patients received intravenous administrations of propofol and remifentanil achieving different effect-site concentration levels through controlled infusions according to Schnider [2,3] and Minto [4,5] pharmacokinetic models, respectively. Maximal effect-site concentrations achieved for propofol and remifentanil were 6 ng/mL and 3 ng/mL respectively. Ten patients received standardized tetanic stimulation (60 mA for 5 seconds); hence, the associated nociceptive effect on the respiratory parameters was evaluated during model building as well as the covariates of weight and age.
Both respiratory responses were analysed simultaneously using the population approach with nonlinear mixed effects modelling and the software NONMEM 7.4. R Version 4.0.2, with RStudio interface, was used for data formatting and exploration.
Results: Indirect response models were developed for both responses. In the final model, a maximum inhibitory model was implemented to describe remifentanil drug effect on MV input rate. In addition, a rebound term was included at this level to acknowledge the presence of compensatory mechanisms that would allow for a fast recovery of MV upon perturbation. This rebound, dependent on an intermediate modulator with a fast turnover (kMOD=0.03 min-1) and also influenced by remifentanil concentrations, significantly improved model predictions during the recovery phase, when drug is being cleared from the system. Both remifentanil drug effects were described with a common estimate for the inhibitory concentration triggering 50% of the maximum effect (EC50 = 1.5 ng/mL and associated interindividual variability of 41 %), although different shape parameters were assumed. Finally, a significant impact of the tetanic stimulation was identified at MV level, and included in the model as a bolus with a mono-exponential decay (kSTIM), driven by a dissipation rate constant of 1.93 min-1. Regarding pCO2 response, as baseline observations were available, the B2 method [6] was implemented. Degradation rate for pCO2 was estimated, being the typical value 0.2 min-1. Through the physiological relationship between MV and pCO2, remifentanil exerts its effect over pCO2 response.
Overall, the model exhibited a satisfactory performance, as shown by the different diagnostics techniques, and adequate parameter precision (relative standard errors < 50 %) for all parameters except for kSTIM, associated to a larger uncertainty due the nature of the procedure.
Conclusions: We have developed a fully identifiable semi-mechanistic model handling simultaneously two relevant variables of the respiratory response in a variety of scenarios during routine anaesthesia practice, including drug combinations and the absence/presence of noxious stimulus.
This model provides valuable insights into the risk factors and mechanisms of respiratory depression in general anaesthesia. It is intended to facilitate anaesthesiologist to personalize drug optimization protocols and prevent respiratory depression in patients undergoing sedation.
References:
[1] Hannam JA, Borrat X, Trocóniz IF, et al. Modeling Respiratory Depression Induced by Remifentanil and Propofol during Sedation and Analgesia Using a Continuous Noninvasive Measurement of pCO2. J Pharmacol Exp Ther. 2016;356(3):563-573. doi:10.1124/jpet.115.226977
[2] Schnider TW, Minto CF, Gambus PL, et al. The influence of method of administration and covariates on the pharmacokinetics of propofol in adult volunteers. Anesthesiology. 1998;88(5):1170-1182. doi:10.1097/00000542-199805000-00006
[3] Schnider TW, Minto CF, Shafer SL, et al. The influence of age on propofol pharmacodynamics. Anesthesiology. 1999;90(6):1502-1516. doi:10.1097/00000542-199906000-00003
[4] Minto CF, Schnider TW, Egan TD, et al. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology. 1997;86(1):10-23. doi:10.1097/00000542-199701000-00004
[5] Minto CF, Schnider TW, Shafer SL. Pharmacokinetics and pharmacodynamics of remifentanil. II. Model application. Anesthesiology. 1997;86(1):24-33. doi:10.1097/00000542-199701000-00005
[6] Dansirikul C, Silber HE, Karlsson MO. Approaches to handling pharmacodynamic baseline responses. J Pharmacokinet Pharmacodyn. 2008;35(3):269-283. doi:10.1007/s10928-008-9088-2
Reference: PAGE 32 (2024) Abstr 10880 [www.page-meeting.org/?abstract=10880]
Poster: Drug/Disease Modelling - Other Topics