If you want to tell whether your baby is in pain, looking at its face may not be enough, researchers have found.
Generations of mothers have depended on their baby's facial expressions to tell them what they are feeling. But a study has found that giving a baby a spoonful of sugar before an injection or blood test may alter its expression without lessening its pain.
The finding casts doubt on whether we can really know what a baby is feeling from observing its responses - and on the decade-old practice of using sugar as a pain reliever for infants.
Until the 1950s, doctors thought babies did not suffer pain because their consciousness was not sufficiently developed. The normal pain responses - grimacing and crying - were dismissed as reflexes. Babies subjected to surgery were given anaesthetics to put them to sleep but not analgesic drugs for the pain, as children and adults were.
In the 1970s, a definitive study showed babies did benefit from analgesia. But as it is difficult to test them on babies, few drugs are available.
Giving a teaspoonful of sugar solution to babies was thought to relieve pain based on the way it reduced grimacing and crying after a painful procedure. It is believed to stimulate the production of "endogenous opiates" - the body's own natural pain-relieving drugs - and has become standard practice before blood tests and similar procedures. Some doctors maintain the evidence is now so strong that it may be unethical not to use it.
But doctors at University College Hospital in London who measured brain activity in babies subjected to a painful procedure - having their heels pricked to obtain a drop of blood for testing - have found that even when they did not cry or grimace there was still a "pain response" in their brains.
The authors of the study, published in the Lancet, suggest that the sugar solution could inhibit the normal responses of grimacing and crying by giving the baby something else to focus on, even while brain activity suggests there is pain.
"Sucrose seems to blunt facial expression activity after painful procedures but our data suggest that it does not reduce direct [pain] activity in central sensory circuits, and therefore might not be an effective analgesic drug," they say.
Judith Meek, consultant neo-natologist at University College London Hospital and joint author of the study, said: "When you give the sugar solution to babies they do look dreamy. Most likely it is a distraction but it doesn't affect the pain. There are other ways to distract babies - cuddling or feeding them - which may be more effective.
"Some babies born prematurely may have a blood test or other procedure every day and there is concern the pain could lead to emotional or behavioural disorders later on. So it is important that when we think we are abolishing pain we really are. If we think we are and we're not, it may lead to adverse outcomes.
"The pain response in the brain and the facial responses are related but a number of the babies didn't grimace or cry but did show a response in their brains. The same baby on different occasions might have different responses. What our study shows is that we may underestimate what is going on in the babies [based on their facial responses]."
The study included 59 babies, half of whom received a spoonful of sugar and half a spoonful of water two minutes before the heel prick. The results showed that all the babies had a pain response in their brains but a third of those given the sugar solution did not grimace or change their facial expression.
The quantity of sugar, though small, can add up to a significant dose in ill babies subjected to many routine procedures.
According to one calculation, for a premature baby undergoing 10 painful procedures a day, it could be equivalent to giving half a can of cola to a one-year-old every day, on a weight-for-weight basis.
In a commentary on the paper, also published in the Lancet, two US scientists say the size of the study was too small to demonstrate whether sugar was an effective painkiller or not.
But they say it "adds an important and innovative measurement to evaluating pain management in newborn babies".