Population pharmacokinetic and pharmacodynamic model of propofol externally validated in Korean elderly subjects

• Published in: Journal of pharmacokinetics and pharmacodynamics Vol. 50; no. 2; pp. 97 - 109
• Main Authors: Kim, Kyung Mi, Choi, Byung-Moon, Noh, Gyu-Jeong
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.04.2023; Springer Nature B.V
• Subjects: Aged; Anesthetics, Intravenous - pharmacokinetics; Biochemistry; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Humans; Models, Biological; Original Paper; Pharmacodynamics; Pharmacokinetics; Pharmacology/Toxicology; Pharmacy; Propofol; Propofol - pharmacokinetics; Republic of Korea; Veterinary Medicine/Veterinary Science; Republic of Korea; Model; Pharmacodynamics; Aged; Propofol; Pharmacokinetics; Performance
• ISSN: 1567-567X; 1573-8744; 1573-8744
• DOI: 10.1007/s10928-022-09836-6

The Eleveld propofol pharmacokinetic (PK) model, which was developed based on a broad range of populations, showed greater bias (− 27%) in elderly subjects in a previous validation study conducted by Vellinga and colleagues. We aimed to develop and externally validate a new PK-pharmacodynamic (PK-PD) model of propofol for elderly subjects. A population PK-PD model was constructed using propofol plasma concentrations and bispectral index (BIS) values that were obtained from 31 subjects aged 65 years older in previously published phase I studies. The predictive performance of the newly-developed PK-PD model (Choi model) was assessed in a separate Korean elderly population and compared with that of the Eleveld model. A three-compartment mammillary model using an allometric expression and a sigmoid E max model well-described the time courses of propofol concentrations and BIS values. The V 1 , V 2 , V 3 , Cl , Q 1 , Q 2 , E 0 , E max , Ce 50 , γ , and k e0 of a 60-kg subject were 8.36, 58.0, 650 L, 1.26, 0.917, 0.669 L/min, 92.1, 18.7, 2.21 μg/mL, 2.89, and 0.138 /min, respectively. In the Choi model and Eleveld model, pooled biases (95% CI) of the propofol concentration were 7.78 ( 3.09–12.49) and 16.70 (9.46–23.93) and pooled inaccuracies were 22.84 (18.87–26.81) and 24.85 (18.07–31.63), respectively. The Choi PK model was less biased than the Eleveld PK model in Korean elderly subjects (age range: 65.0–79.0 yr; weight range: 45.0–75.3 kg). Our results suggest that the Choi PK model, particularly, is applicable to target-controlled infusion in non-obese Korean elderly subjects.