Article Text


Disentangling PK-PD in neonates
  1. B Anderson
  1. Correspondence to:
    B Anderson
    Department of Anaesthesiology, University of Auckland, Auckland, New Zealand;

Statistics from

A commentary on the paper by Allegaert et al

Neonatal analgesia is a humane consideration, and paracetamol offers a reasonable alternative for mild to moderate pain. The investigation of propacetamol, an intravenous prodrug of paracetamol (acetaminophen), in premature neonates after intravenous administration has received scant attention. We might expect propacetamol to offer advantages clinically because a plasma target concentration can be rapidly attained and variability due to absorption kinetics and relative bioavailability is reduced. In addition, an intravenous drug offers a cleaner pharmacokinetic study that can further our knowledge of paracetamol pharmacokinetics in premature and term neonates.1–5 Allegaert et al are to be congratulated for investigating propacetamol use in premature neonates. Neonatal pharmacokinetic studies are fraught with pitfalls.

Clearance often increases rapidly after birth (postnatal age) no matter what the gestational age at birth. This study gets around this by studying all infants in the first day of life and stratifies by gestational age. It is difficult to enrol neonates of different gestational ages into a longitudinal study. Population modelling—for example, NONMEM software—offers great potential here. Population parameter estimates and their variability from rich or scant data involving a large number of individuals can be made.6 Missing data are tolerated. Infusion rather than instantaneous input can be examined. Covariate analyses can then be used to investigate the effects of age, size, sex, and concomitant drug therapy. Clearance changes with age are part of a continuum, rather than a cut off at 37 weeks gestation. This covariate analysis is difficult using the standard two stage approach of this study. It also ignores imprecision in the estimate of an individual’s structural parameter, and, if the response in one individual is much more variable than in another, some form of weighting is required.

The per kilogram and surface area models may be inappropriate for scaling small children to adults. It is possible to show that in almost all species including humans, the log of basal metabolic rate plotted against the log of body weight produces a straight line with a slope of ¾. West et al7,8 have used fractional geometry to explain this phenomenon mathematically. A great many physiological, structural, and time related variables scale predictably within and between species with weight exponents of 0.75, 1, and 0.25 respectively. These allometric “¼ power” models can be applied to pharmacokinetic parameter estimates in children—for example, clearance (0.75), half life (0.25), volume of distribution (1).9 If parameter values were standardised for a body weight of 70 kg using allometric “¼ power models”, then we can compare neonatal estimates with adult estimates and track the developmental changes with age. For example, a clearance of 0.2 litre/h/kg, when standardised to a 70 kg person (14 litres/h/70 kg), is reduced in a 1.5 kg intrauterine growth retarded term neonate (5.4 litres/h/70 kg) when compared with a 4 kg term neonate (6.9 litres/h/70 kg). Clearance is not the same in the two neonates. Clearance rather than half life dictates maintenance dosing. Half life is confounded by volume of distribution, which changes with age in early life.

Paracetamol concentration does not correlate directly with analgesia: several authors have shown effect compartment delays (Teq ≈ 1 h) for analgesia and antipyresis in children,10,11 but data in neonates are lacking. Assessment of effect in neonates is difficult, and attempts to show concentration-response relations have been unfruitful,12 as in the present study. Assessment tools may measure discomfort rather than pain, neonatal wind up may exist, pain is not constant, placebo effects contribute, and large interindividual variability has been reported for pharmacodynamic parameter estimates. Further, both the responses to a painful stimuli and pain pathway development may vary with age.

The present study is useful. Suggestions on dose regimens are practical. It does, however, raise some unanswered questions. For example, is the rate and extent of hydrolysis of propacetamol to paracetamol in neonates the same as in adults? What is the likelihood of hepatic damage if propacetamol is used in sick premature neonates? At what rate do clearance enzyme pathways mature? What is the target concentration in neonates? How effective is paracetamol in neonates and what effect end points should be used?

A commentary on the paper by Allegaert et al


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.