How delayed or missed doses influence efficacy of amoxicillin in outpatients with community-acquired pneumonia: A pharmacokinetic/pharmacodynamic simulation analysis
Suarez E(1) , Carral N(1), Estrade O(1), Jauregizar N(1), Lukas JC(1,2).
(1) Department of Pharmacology, School of Medicine, University of the Basque Country, Spain (2)Current affiliation: Drug Modeling & Consulting, Dynakin SL, Derio, Spain
Objective: The purpose of this study was to quantify the impact on efficacy of non adherence to treatment by amoxicillin in outpatients with community-acquired pneumonia.
Methods: A dose dependent (absorption) amoxicillin population PK model parameters [extracted from Sjövall published data (1)] was used to simulate the pharmacokinetic profiles of amoxicillin after an empirical dosing protocol (1000 mg/8h) on pneumonia produced by S Pneumoniae. Virtual outpatients (weight 70 kg) were divided in subgroups according to age (young: 18-25 years; adults: 25-65 years; and elderly: 65-80 years) and interindividual variability in creatinine clearance calculated according the Cockcroft-Gault formula, based on demographic characteristics and serum creatinine (Crs) (0.7-1.3 mg/dl). Treatment control and non-compliance scenarios were applied in simulation across age and Crs levels, for a missed or delayed dose (1, 2, 3, 4, 6, 8h delays), and the impact on drug exposure calculated for patient proportions (Monte Carlo simulations). The probability of target attainment (fT>MIC90 less than 50% as PK/PD predictor of antimicrobial efficacy), was calculated in each scenario, with a 90% clinical efficacy threshold (2).
Results: The medium PK parameters estimated from the public domain model extraction were for absorption rate constant=0.63 , apparent volume of distribution = 23.3 L, and different range of apparent clearance were established depending on Crs= 13.2-24.6 L/hr (young adults), 10.8-20.1 L/hr (adults) and 7.2-13.4 L/hr (elderly), similar to those reported elsewhere. Monte Carlo simulations of amoxicillin plasma concentrations in 10k virtual patients were done using the model above. In patients with Crs= 0.7 mg/dL (elderly) the probability of target attainment was less than 90% for a delay in drug intake > 2 hours, but in young and adults no delay can be allowed. In patients with Crs between 0.8 and 0.9 mg/dL, efficacy impacting delay times are >1h (young); >2h (adults); >6h (elderly). In patients with Crs of 0.9-1.3 mg/dL the corresponding prohibitive delays are 2-3h, 4-6h and 8h in young, adult and elderly, respectively.
Conclusion: In non-adherent dosing scenarios for virtual outpatients of different age and physiological interindividual variability on creatinine clearance, amoxicillin dose 1000 mg/8h was not always able to achieve the PK/PD index guaranteeing clinical efficacy in community-acquired pneumonia, especially in young patients with a Crs inferior to 0.9 mg/dL.
References:
[1] Sjövall J, Alvan G, Westerlund D. Dose-dependent absorption of amoxycillin and bacampicillin. Clin Pharmacol Ther 1985; 38:241-250
[2] Asín-Prieto E, Rodriguez-Gascon A, Isla A. Applications of the pharmacokinetic/pharmacodynamics (PK/PD) analysis of antimicrobial agents. J Infect Chemother 2015; 21:319-29
