Population pharmacokinetics of dexamethasone in critically ill COVID-19 patients: does inflammation play a role?
Letao Li1, Sebastiaan Sassen1, Nicole Hunfeld2, Tim Smeets1,2, Tim Ewoldt1,2, Sjoerd A.A. van den Berg3, Birgit Koch1, Henrik Endeman2
1. Department of Hospital Pharmacy, Erasmus MC-University Medical Center; 2 Department of Intensive Care, Erasmus MC-University Medical Center; 3 Department of Clinical Chemistry, Erasmus MC-University Medical Cente
Objectives: One of the common causes of COVID-19 related death is acute respiratory distress syndrome (C-ARDS). Dexamethasone is the cornerstone in the treatment of C-ARDS and reduces mortality probably by suppressing inflammatory levels in ICU patients. Its anti-inflammatory effects may be concentration-dependent. It is still unclear whether a "one dose fits all" strategy for COVID ICU patient is appropriate, and the optimal dose and therapeutic target of dexamethasone remains uncertain. However, no pharmacokinetic studies of dexamethasone have been conducted in ICU COVID patients. Therefore, we designed a population pharmacokinetic study to gain a deeper understanding of the pharmacokinetics of dexamethasone in critically ill patients in order to identify relevant covariates that can be used to personalize dosing regimens and improve clinical outcomes.
Methods: A retrospective, single center study was performed in critically ill patients admitted at the department of intensive care at the Erasmus University Medical Center. Blood samples were collected in adults at the ICU with COVID-ARDS who were treated with a fixed dose of intravenous dexamethasone (6mg per day). The data were analyzed using Nonlinear Mixed Effects Modelling (NONMEM) software for population pharmacokinetic analysis. Physiological and biochemical indicators, and medication variables age, gender, length, weight, BMI, BSA, albumin, ASAT, ALAT, GGT, LDH, eGFR, IL-6, CRP, PCT, WBC and potential interaction drugs including tocilizumab, erythromycin, voriconazole and fluconazole were tested as covariates. To show an illustration of the covariate effect in the final model on the plasma concentrations of dexamethasone, Monte Carlo simulations were performed by using NONMEM.
Results: A total of 51 dexamethasone samples were analyzed in 18 patients. A two-compartment model with first-order kinetics best fitted the data. The mean population estimates were 2.85 L/h (IIV 62.9%) for clearance, 15.4 L for the central volume of distribution, 12.3 L for the peripheral volume of distribution and 2.11 L/h for the inter-compartmental distribution clearance. Model evaluation (GOF, VPCs, bootstrap) showed that the model fit the data well. he covariate analysis showed a significant correlation between dexamethasone clearance and CRP. Dexamethasone clearance decreased significantly with increasing CRP in the range of 0-50 mg/L and a correlation was observed with CRP up to 100 mg/L. The simulation results showed that increased dexamethasone exposure in patients with elevated CRP.
Conclusions: Our study showed that the pharmacokinetics of dexamethasone could be adequately described by a two-compartment model. The dexamethasone PK parameters of ICU COVID patients were quite different from those come from healthy populations. Inflammation might play an important role in dexamethasone clearance, suggesting that several days of fixed-dose dexamethasone affect patients differently on the first day than on subsequent days. This is a total new way of thinking on immunomodulation and might explain the rebound effect sometimes seen in immunomodulation by dexamethasone.
