Muhammad W. Ashraf (1), Panu Uusalo (1,2), Mika Scheinin (3) and Teijo I. Saari (1,2)
(1) Department of Anesthesiology and Intensive Care, University of Turku, Turku, Finland. (2) Division of Perioperative Services, Intensive Care and Pain Medicine, Turku University Hospital, Turku, Finland (3) Institute of Biomedicine, University of Turku, and Unit of Clinical Pharmacology, Turku University Hospital, Turku, Finland
Objectives:
Dexmedetomidine is an α2-adrenoceptor agonist [1] that has been shown to be extremely useful in the management of intractable anxiety, stress and pain in palliative end-of-life care patients [2]. It provides an adequate therapeutic benefit of pharmacological dose-dependent sedation without any risk of ventilatory depression [3], while also avoiding the inconvenient side effects associated with opioids, benzodiazepines, antidepressants or antipsychotics [4]. Dexmedetomidine has also been reported to attenuate hemodynamic and endocrinal stress responses in humans [5]. The objectives of our study included: 1. The development of a population pharmacokinetic model to characterize the kinetic profile of intravenous (IV) and subcutaneously (SC) administered dexmedetomidine, 2. The development of a population pharmacokinetic-pharmacodynamic model to explain the inhibition of norepinephrine (NE) and epinephrine (E) production in vivo and, 3. The development of a population model to predict the effect of diminished NE release on heart rate (HR), mean arterial pressure (MAP), vigilance (VIG) and performance (PER).
Methods:
Data was gathered from an open two-period, crossover study with balanced randomization published previously [6]. Ten volunteers (aged 18 to 30 yrs) were given in 10 minutes 1 μg/kg dexmedetomidine either intravenously (IV) or subcutaneously (SC) on two occasions. Plasma dexmedetomidine concentrations were measured for 10 h after drug administration. In addition, the effects of dexmedetomidine on plasma catecholamine levels, vital signs and sedation were recorded. Nonlinear mixed effects modelling was performed with NONMEM software (version 7.4.1). First, IV and SC administration data was used to construct and validate a PK model for explaining dexmedetomidine disposition kinetics. One, two and three compartmental mammillary models were evaluated during the model development. Different absorption models were
tested to capture the trends in dexmedetomidine absorption after SC dosing. PK model was used further to develop a PD model to explain the inhibition of NE and E production caused by dexmedetomidine-induced sympatholysis. Finally, a direct effect model or an effect compartment model together with a sigmoidal Emax function was used to predict the effect of NE inhibition on other PD parameters.
Results:
A semi-mechanistic structural model was developed. A three-compartment (CMT) mammillary model was better than a two CMT model in explaining dexmedetomidine disposition kinetics [ΔOFV = -331 and -346, respectively]. Plausible parameter estimates [CL = 39 L/HR, V = 0.32 L, CL = 104 L, V = 76 L, CL = 312 L, V = 13.7 L] and visual
predictive checks (VPC) described model adequateness for the data. In the second stage, the absorption of dexmedetomidine after SC administration was captured by the addition of a fat CMT alongside depot. This assumption is supported by the high lipid solubility of dexmedetomidine (logP = 2.4). Rate constants for drug movement from the depot to SC fat layer (K,FAT), from the depot to central CMT (Ka,FAST) and from SC fat to central CMT (Ka,SLOW) were estimated. The model produced an adequate parameter estimates for the data [K,FAT = 1.9, Ka,FAST = 0.61 and Ka,SLOW = 0.083]. Next, an indirect response model was employed to explain dexmedetomidine-induced decrease in NE and E production. In the final stage, all other PD parameters that are dependent on the circulatory levels of NE and E were modelled using effect compartment model with sigmoidal EMAX function. The model parameter estimates were biologically plausible [C[50,NE] = 0.29, E[BASE,NE] = 0.80, K[OUT,SYN,NE] = 9.3, K[OUT,PER,NE] = 11, C[50,E] = 0.52, E[BASE,E] = 0.23, K[OUT,SYN,E] = 3.6] and the results were further evaluated with VPCs (Figure) and bootstrap resampling.
Conclusions:
The pharmacokinetic-pharmacodynamic model developed in this study adequately describes the pharmacokinetic profile of intravenous and subcutaneously administered dexmedetomidine in healthy human volunteers, along with accurately predicting the inhibition of norepinephrine and epinephrine production in vivo due to dexmedetomidine challenge, as well the inhibitory effect of reduced catecholamine production on heart rate, mean arterial pressure, vigilance and performance of the study subjects.
References:
[1] Wujtewicz M, Maciejewski D, Misiołek H, Fijałkowska A, Gaszyński T, Knapik P, Lango R (2013) Use of dexmedetomidine in the adult intensive care unit. Anaesthesiol Intensive Ther 45(4): 235–240
[2] Aantaa R, Tonner P, Conti G, Longrois D, Mantz J, Mulier JP (2015) Sedation options for the morbidly obese intensive care unit patient: a concise survey and an agenda for development. Multidiscip Respir Med 10(1):8
[3] Lodenius Å, Ebberyd A, Hårdemark Cedborg A, Hagel E, Mkrtchian S, Christensson E, Ullman J, Scheinin M, Eriksson LI, Jonsson Fagerlund M (2016) Sedation with dexmedetomidine or propofol impairs hypoxic control of breathing in healthy male volunteers: a nonblinded, randomized crossover study. Anesthesiology 125:700–715
[4] Devlin JW, Roberts RJ (2011) Pharmacology of commonly used analgesics and sedatives in the ICU: benzodiazepines, propofol, and opioids. Anesthesiol Clin 29(4):567–585
[5] Scheinin B, Lindgren L, Randell T, Scheinin H, Scheinin M (1992) Dexmedetomidine attenuates sympathoadrenal responses to tracheal intubation and reduces the need for thiopentone and perioperative fentanyl. Br J Anaesth 68(2):126–131
[6] Uusalo, P., Al-Ramahi, D., Tilli, I., Aantaa, R. A., Scheinin, M., & Saari, T. I. (2018). Subcutaneously administered dexmedetomidine is efficiently absorbed and is associated with attenuated cardiovascular effects in healthy volunteers. European Journal of Clinical Pharmacology, 74(8), 1047-1054. doi:10.1007/s00228-018-2461-1
Reference: PAGE 28 (2019) Abstr 9197 [www.page-meeting.org/?abstract=9197]
Poster: Drug/Disease Modelling - CNS