Paediatric dosing
A Socratic walk-through of paediatric dosing — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why is a child's dose not simply an adult's dose scaled down by body weight?
There is an arithmetic that feels unarguable. A 70 kg adult takes a certain dose; a 14 kg toddler is one fifth the size; give one fifth of the dose. Milligrams per kilogram. It is the calculation every drug chart seems built around, and it has the tidy feel of a physical law.
But push on what that calculation quietly assumes. It assumes the body is a container, that the drug fills it, and that filling a smaller container to the same depth requires proportionally less. Is that what a dose is doing? A dose is not really filling a volume — it is holding a concentration against a body that is continuously removing the drug. And removal is a rate, produced by organs, not by mass. Once you say it that way, the question changes: does that rate really fall in step with body weight?
Reasoning it through
REASONING #Take the question to the whole animal kingdom first, where the effect is large enough to see plainly. If metabolic rate simply tracked mass, a mouse and an elephant would burn fuel per gram at the same speed. They do not: small animals run hot and fast per gram, and metabolic rate across species scales roughly with mass to the power of about three quarters rather than to the power of one. Drug clearance follows the same kind of curve, because it is produced by the same machinery of liver blood flow, enzyme mass and kidney filtration.
Now do the arithmetic for our toddler. Weight scaling says her clearance should be a fifth of the adult's. Allometric scaling — weight to the power of 0.75 — says roughly a third. Which means she clears the drug faster per kilogram than the adult does. Strict milligram-per-kilogram dosing therefore tends to leave a small child under-treated, not over-treated. That is the first surprise, and it runs opposite to the direction most people's caution points.
So should we simply give small children more per kilogram? Here the story doubles back, and the reason is that a newborn is not merely a small child.
Ask what the clearance machinery is actually made of at birth. The kidney is filtering at a fraction of the adult rate corrected for size, and takes months to catch up. The liver's enzyme systems mature on their own separate schedules, some of them over weeks, some over the first year or two. So in the newborn, immaturity dominates the allometric effect and reverses it: the neonate clears many drugs far slower, and the same milligram-per-kilogram dose accumulates. This is not a theoretical worry. Grey baby syndrome, described in newborns given chloramphenicol in the late 1950s, was the consequence of exactly this — an immature conjugation pathway and a dose worked out on adults.
Put the two forces on the same axis and you get a shape rather than a rule. Through the first weeks, immaturity holds clearance below what size alone would predict. As the enzymes and the kidney come up, the allometric effect takes over and per-kilogram clearance rises above adult values through infancy and early childhood, then settles back down toward adult values through adolescence. The dose per kilogram that keeps concentration steady traces that same rise and fall.
And there is more than clearance in play. A newborn is proportionally much wetter than an adult — a considerably larger share of body weight is water — so a water-soluble drug spreads through a bigger relative volume and needs a larger loading dose per kilogram to reach the same concentration. Plasma protein binding is lower too, leaving more of the drug in its free, active form, which is why sulfonamides in neonates could displace bilirubin and risk kernicterus.
One honest caution before we tidy this up. Much paediatric dosing rests on far less trial evidence than adult dosing does, because trials in children are hard to run and were historically avoided; a great deal of prescribing in neonatal and paediatric units is off-label. So some of these numbers are extrapolations that have been checked in practice rather than measured from the start.
The analogy
THE ANALOGY #Think of the drug level as the water level in a bathtub with the plug out. Your dose is the tap; the child's own body is the drain. Sizing the tap to the size of the tub only works if the drain shrank in the same proportion — and in a child it did not. The small tub has a proportionally wider drain, so it needs a relatively stronger flow to sit at the same level, except in the first weeks of life, when the drain is still partly blocked.
a bathtub's drain is fixed while you are filling it, whereas a growing child's drain is being rebuilt continuously through infancy, so the right tap setting keeps moving even for the same patient.
Clarifying the model
THE MODEL #The misconception to retire is that a child is a scale model of an adult. Almost nothing that matters here scales linearly: not clearance, not body water, not protein binding, not the maturity of any given enzyme. Milligram-per-kilogram dosing is not wrong so much as it is a first-order correction that happens to be accurate enough in the middle of childhood and least accurate at the two ends — the newborn, and the large adolescent where the same formula will overshoot the adult ceiling.
It also helps to keep two questions apart. How much drug to give is pharmacokinetics — what the body does to the drug, which is the argument we have just made. What the drug does at a given concentration is pharmacodynamics, and it is not guaranteed to be the same in a child either, since receptor numbers, the blood-brain barrier and the target physiology are all developing too. Getting the concentration right is necessary, not sufficient.
Finally, notice that the allometric exponent is a description of a broad regularity, not a law derived from first principles — its exact value and the theory behind it are still argued over. It is used because it fits the data across a wide size range better than linear scaling does, which is a modest and honest reason to prefer it.
A picture of it
THE PICTURE #How to readRead top to bottom as one patient ageing, each box a condition the body occupies for a while rather than a step in a procedure. The transition labels say what physically changes to end each state. Notice the direction reverses between the first and second transitions: in the newborn box immaturity is holding clearance down, and from the next box onward size is pushing it up. The note attached to the first state is where linear weight scaling does its damage.
What became clearer
WHAT CLEARED #A dose is not a quantity of substance matched to a quantity of body; it is a rate matched to a rate of removal. Removal is done by organs whose capacity scales with roughly the three-quarter power of mass, so a small child clears drugs faster per kilogram than an adult — unless the organs are not finished yet, which is the newborn's situation and the opposite error. Paediatric dosing is the resolution of those two forces, and it is a curve over age rather than a proportion of a number.
Where to go next
ONWARD #- Why oncology often doses by body surface area, and how well that proxy actually holds up.
- How therapeutic drug monitoring closes the loop when the prediction is unreliable.
- Why obesity breaks weight-based dosing in adults for much the same reason it breaks in children.
Key terms
TERMS #| Term | What it means |
|---|---|
| Allometric scaling | describing a biological rate as a power of body mass, with an exponent near 0.75 commonly used for clearance. |
| Clearance | the volume of blood effectively cleared of a drug per unit time, the quantity that sets the maintenance dose. |
| Volume of distribution | the apparent volume the drug spreads into, which sets the loading dose. |
| Grey baby syndrome | neonatal collapse from chloramphenicol accumulation due to immature conjugation, reported in the late 1950s. |
Every term the collection defines is gathered in the glossary.