Adaptive drug interaction model to predict depth of anesthesia in the operating room. (May 2020)

• Record Type: Journal Article
• Authors: Gonzalez-Cava, Jose M.; Reboso, José A.; Calvo-Rolle, José Luis; Mendez-Perez, Juan Albino
• Is Part Of: Biomedical signal processing and control. Volume 59(2020)
• Publication Date: 2020-05
• Subjects: Bispectral index -- Depth of anesthesia -- Drug interactions -- Interpatient variability -- Intrapatient variability -- PK–PD model; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2020.101931
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: An optimization-based algorithm is proposed to synthesize the adaptive model. A PK-PD structure for additive drug interaction is considered. The parametric model deals with patient and process variabilities. Simulation results evidence the benefits of using adaptive models in anesthesia. Abstract: The availability of accurate models for predicting the drug effect in patients undergoing general anesthesia is an important factor in producing a personalized drug infusion. These models should consider different clinical factors to provide realistic predictions. This paper proposes a new methodology for modeling the depth of hypnosis (DOH) during anesthesia. The model, which is based on a pharmacokinetic–pharmacodynamic structure, explicitly takes into account the interaction between the hypnotic and opioid drugs delivered during surgery. Patients undergoing general surgery with intravenous propofol–remifentanil anesthesia were considered. The bispectral index (BIS) was used for monitoring the DOH. In contrast with previous research, the uniqueness of this study lies in the proposal of an adaptive model to deal simultaneously with the variabilities in the clinical response of the patients, the drug interactions, and the variable time delay introduced by the BIS monitor. The proposed method was validated using data from 17 patients undergoing general anesthesia. Successful results were obtained for predicting the evolution of BIS during the induction and maintenance phases of propofol–remifentanil anesthesia. Specifically, the convenience of an adaptive model that included all the factors likely to affect the anesthetic process was demonstrated. The proposed methodology can be used for the development of new models to be employed in model predictive control strategies for closed-loop anesthesia.