Development of a Morphine Pharmacokinetic and Pharmacodynamic Model for the Neonatal Population

Critically ill immature preterm infants experience multiple noxious stimuli while receiving care in the Neonatal Intensive Care Unit (NICU). These noxious stimuli include, but are not limited to: venipuncture; insertion of intravenous and arterial catheters; suctioning of the nose, mouth and oropharynx; endotracheal intubation for mechanical ventilation; insertion of chest drains; repositioning and other types of patient manipulation. The delivery of optimal doses of analgesics for these noxious stimuli is a major challenge due to the lack of knowledge about drug disposition and its effects in this patient population.

Morphine is the commonest analgesic used in the NICU. The Premature Infant Pain Profile (PIPP) is used to quantify pain in the NICU1. This objective score, which combines physiological and behavioural variables defining levels of discomfort, is used as a guide for the use of morphine in newborn infants. Multidimensional pain assessment tools, such as PIPP, can easily identify behaviour in healthy infants undergoing painful events, however, its efficiency is questionable when applicable to critically ill premature infants with neurological impairment, where the pain processing and modulation may be altered. Pharmacokinetics/Pharmacodynamics (PKPD) models can be used to quantitatively describe and predict drug disposition in the blood and the target organ (e.g., brain) in relation to doses and patient characteristics. Although there has been a global effort to describe morphine plasma levels in this population using a pharmacokinetic modelling approach2-5, PKPD model development has not been reported. The study of morphine pharmacokinetics to determine the optimal dose for balancing analgesia/sedation together with the design of pharmacodynamic model for morphine may provide a better understanding of nociception/pain profile based on the physiological variables of immature infants. Moreover, the PKPD model may be used to achieve optimal therapeutic effects through individualised model-based dose selection. Objective: This study is composed by four main objectives:

1. First: Define target morphine plasma and brain concentrations. To this end, we will develop a morphine PKPD model based on population PK characteristics and morphine effects captured by functional readout of central nervous system function;
2. Second: Develop an opportunistic sampling method for fragile populations based on saliva sampling;
3. Third: Compare pharmacokinetic parameters calculated from saliva with plasma sample;
4. Fourth: Application of the morphine PKPD model for prediction of optimal morphine dosing in individual patients using the Bayesian framework of model refinement.