Pain in babies

Up until the last thirty years medicine seems to have been unable to make up its mind about whether babies feel pain. Doctors followed whatever the prevailing expert opinion was, and before the late nineteenth century it was generally considered that babies hurt more easily than adults. Then the opinion changed, and until the mid 1970s doctors “knew” that babies do not feel much pain. For much of the history of modern scientific medicine children and babies were therefore denied the benefits of the steadily improving medical techniques of treating pain. Fortunately for the linguistically challenged young human, having only a frowning face, fierce cry and angry movements with which to convey hurt, science has in the last quarter of the 20th century established beyond reasonable doubt that neonates and babies definitely do experience pain – probably more than adults – and has developed reliable means of measuring and of treating it.

This article sets out to review the main difficulties involved in assessing and treating pain in the newborn and the pre-verbal toddler. It aims to dispell those myths and fallacies which have caused whole generations of children to suffer unnecessarily, and to inform caregivers about the possibilities for helping the baby in pain.

However could we think babies don’t hurt?

Before the late nineteenth century babies were considered to be more sensitive to pain than adults. In the pre-scientific medicine era, this was patently obvious, by virtue of everyday observations and logic. Doris Cope quotes from The Children's Book, by Felix Wurtz, published in 1656, to illustrate this logic: "If a new skin in old people be tender, what is it you think in a newborn Babe? Doth a small thing pain you so much on a finger, how painful is it then to a Child, which is tormented all the body over, which hath but a tender new grown flesh?" The ease with which an apparently trifling hurt can set off a piteous crying spell in the young child had made its impression on Mr Wurtz, and set him a-wondering, as it has turned out, on the right path.

But then the whole idea did a turn-about. Suddenly, in the late nineteenth and first half of the twentieth century, doctors were being taught that babies did not experience pain, and they were treating their young patients according to this unproven idea. From needle sticks to tonsillectomies to heart operations were done with no analgesia or even anaesthesia, other than muscle relaxation for the surgery.

When one realises that a horse lover is prepared to put his beloved mount “out of its misery” when it sustains an incurable injury, this callous attitude of humans towards their young seems inexplicable. It is possible that this was a result of a scientific misinterpretation of the findings of the new science of embryology – which considered that the non-myelinisation of much of a baby’s nervous system indicated that it did not yet function – taken together with the writings of Charles Darwin in his book The Expression of Emotions in Man and Animal (1872), that babies (as well as “animals, savages and the insane”) were incapable of experiencing pain. Whatever the real cause for the myth was, doctors were taught that in babies the expression of “pain is merely reflexive and … owing to the immaturity of the infant brain, the pain could not really matter.” At the same time there was the unscientific belief that use of opiates would lead to addiction, that babies would not remember any pain that they happened to feel, and that lack of conscious memory meant lack of long-term harm. Scientific studies on animals with various brain lesions were interpreted as supporting the idea that the reponses seen in babies were merely spinal reflexes. Furthermore, the whole effort of relieving pain was considered futile since it was thought to be impossible to measure the child's pain. No doubt the perceived risk of opiates, and the time and effort needed to provide adequate analgesia to the newborn, contributed to the doctor clinging to the “sensible and safe” doctrine of “carry on regardless of the crying”, rather than worrying about the ethics of not providing pain relief – and the practical difficulties of providing it..

While there were always doctors who treated young patients’ pain at face value, the revolution towards studying and treating pain in babies gained momentum in the 1980s. Parents and doctors became aware of the then common practice of performing surgery on newborns with no anaesthesia other than drugs which prevent movement. Publications on the hormonal and metabolic responses of babies to noxious stimuli began to appear, in the face of which the arguments about an “inadequate” cerebral response to pain could not be sustained. Studies on the measurement of pain in young children, as well as ways of reducing the injury response began to be done. The medical opinion about the significance of pain in the neonate has come a full circle. We now know that the very young respond more extensively to pain than the adult does, and that exposure to severe pain, without adequate treatment, can have long-term consequences. Recent writers have even suggested that inadequate treatment of painful events in the newborn may be related to violence and self-mutilation, even suicide, in later life.

In spite of the difficulty of assessing how much pain a baby has, and the problem of finding the correct dosage or technique for treating a feeling which can be gauged but indirectly, modern medicine is firmly committed to improving the quality of pain relief for the very young. Science is catching up with what the caring mother has always known about her child, by the tone of her offspring’s cry.

The developmental neurobiology of pain

Pain can only be assessed by self-reporting of the person experiencing it. Lacking that, it may be inferred from activation of nociceptor nerve pathways, by biochemical changes, by changes seen during functional neuro-imaging, or by physiological and behavioural assessments. In deciding whether and when babies become capable of experiencing pain, one cannot take biopsies of babies’ nerves or brain tissue in order to study the neuronal anatomy and chemistry. If it were to become practical to turn the powerful tools of functional neuro-imaging to use in the study of neonates, it may well mean a second revolution in our understanding of pain in the newborn, but this has not yet been done. The reader has therefore to interpret this section in the light of the fact that beliefs about the development of the nociceptive component of the human nervous system frequently are deduced from findings in animal models of neonatal nociception.

One of the critical scientific events which lead to the present improved understanding of pain in the newborn was the realisation that the fetal and newborn unmyelinated nerve fibres are quite capable of relaying information, albeit slower than would be the case with myelinated fibres. The dictum that “babies don’t feel pain because their nerve pathways are not yet completely myelinated”, originating in the nineteenth century, was eventually, almost a century later, shown by experiment in animal models to be false.

At birth a human has developed the neural pathways for nociception and for experiencing pain, but the pain responses of a newborn baby is not simply a miniaturised or immature version of that of an adult. There are a number of differences in both nerve structure and in the quality and extent of nerve responses which are considered to be pertinent to understanding neonatal pain.

For one, the nerves of young babies respond more readily to noxious stimuli (lower threshold to stimulation), than those of adults. Secondly, their threshold for sensitisation is also decreased. Sensitisation refers to to the phenomenon that a noxious event increases the sensitivity of the nociceptor system to another such event – or to a much milder or even entirely non-noxious event. The experience of severe pain to a mildly painful stimulus is called hyperalgesia, and pain in response to a stimulus which would normally not be painful is called allodynia. This happens in adults and babies, but in babies this occurs more quickly and the sensitised area is larger than in an adult, for the same initial stimulus. A third factor which deserves mention is the fact that the pathways that descend from the brain to the spinal cord are not well developed in the newborn, so that the ability to reduce or inhibit nociception via central brain mechanisms is limited. For all these known reasons, and probably many which still need to be discovered, a noxious event which appears minor to adults (for instance, an intramuscular injection) can have unexpectedly widespread effects in the baby’s nervous system; it is sensed more intensely and potentially more diffusely than it would be in the adult.

As important as the basic increased sensitivity, it is important that the neonate's nervous system seems much more active than that of an adult in transforming its connections and central nerve pathways in response to stimuli. This is intuitively obvious, if one considers that the nervous system is actively growing and maturing. This reshaping - also called plasticity - involves both structural and chemical changes of the nervous system. It has been shown to occur in response to noxious events in the short term, and there are indications that such changes, once established, can persist until adult life. What precisely this implies for later childhood and adult life is as yet unclear, but the present feeling is that this potential for long term harm is yet another reason for working towards more effective management of neonatal pain.

Why treat pain?

If it made no difference to a person whether he or she experienced repeated and severe pain as a baby, then it would be a senseless exercise for the care-givers to look for ways of relieving such pain. What is known about the pain response and the effects of pain on the baby, both immediate and long term, argues strongly for the view that the effective treatment of pain benefits the baby immediate, reduces some medium-term negative consequences, and likely prevents a number of adult psycho-physiological problems.

Some of the immediate effects of pain are obvious to the observer, requiring no special equipment or training. The baby is crying and irritable when awake, develops a disturbed sleep pattern, feeds poorly, and shows a fearful, distrustful reaction towards care-givers. These all have implications for the well-being of the child, not least because the relationship between care-givers and neonate may suffer, with the necessary caring and tenderness being replaced by mutual irritation.

What the observer may not notice, however, is the number of metabolic and homeostatic changes that are caused by untreated pain. The neonate in pain develops an increased requirement for oxygen, while at the same time the gas exchange between air and blood in the lungs becomes less efficient. The combination can lead to inadequate oxygen supply to the cells. The damage that hypoxemia does to cells is well known, and may be permanent, but in the neonate it may be impossible to quantify. Another unseen disadvantage is the rise in stomach acidity that accompanies the stress reaction precipitated by the pain. The sick neonate, in pain, is at risk of aspirating this acid into the lungs, further endangering lung integrity and tissue oxygenation.

The undesirable effects of pain go further. For the duration of the period of persistent pain (for instance after an operation), or of repeated episodes of acute pain (for instance from the procedures in an intensive care unit), the metabolism becomes predominantly catabolic, that is, the baby is not building up its own protein and tissues, but breaking them down through the action of the stress hormones. This same pain and stress response reduces the efficiency of the immune system. Taken together with the reduced oxygen available, it would not be surprising if healing of damaged or infected tissue is impaired, and morbidity and mortality increased. This has been shown to be true in trials of analgesia in newborns with pain.

The neuropsychological effect of untreated pain on the bonding between mother and child, on later contact with health professionals, and on personal and social psychological well-being is difficult to quantify, but research indicates that babies exposed to pain in the neonatal period have more difficulties in all these areas of function, to their possible detriment as older children and adults. Professionals working in the field of neonatal pain have even speculated that adolescent aggression and self-destructive behaviour – including suicide – may in some cases be caused by untreated neonatal pain and stress.

Diagnosis and assessment of a baby's pain

The classical IASP definition of pain as a subjective, emotional experience that is “described in terms of tissue damage”, a definition which depends on the sufferer being able to self-report pain. This is of little use to a person faced with decisions about whether a baby is suffering from pain, how severe it is, and whether treatment is necessary. Care-givers have therefore had to look to non-verbal responses which may be used to identify the experience of pain. The two kinds of responses used are gross physical movements and measurement of physiological responses to stress. The former entails simple direct observation, while the latter depends on measurements which generally require specific equipment, including blood pressure and blood stress hormone levels.

To all but the deaf, the most familiar consequence of hurting a baby is the baby’s cry. Mothers are able to interpret their babies’ cries, while researchers are able to differentiate between different kinds of cry, roughly classed as “hungry”, “angry”, and “fearful or in pain”. Unfortunately a specific baby’s cry is difficult to quantify. It depends on a certain sensitivity in the listener, and the interpretation varies between observers. Quite apart from difficulties of measuring the degree of pain from the type of cry, there is the problem of whether the pain is caused by physical hurt, by fear or by separation. Crying on its own, therefore, is not sufficient a measure of a baby’s pain - clearly it would be inappropriate to treat separation anxiety with morphine, or the pain of acute otitis media by coddling, clucking and cooing.

Additional visible and easily definable evidence of pain have been sought. Combinations of crying with facial expressions, posture and movements, as well as with physiological measurements, have been tested and found to be reliable indicators of pain, and of successful treatment of pain. A number of such measurements and observational scales have been published and verified, with some differences in the ease of use and applicability to different clinical situations (a parameter such as the baby’s blood pressure is of little use to a mother at home, while a person in pain cannot wait for the result of a measurement of plasma cortisol level, before being treated). Behavioral observations include changes in crying, facial expressions, body posture, limb movements and muscle tone. Physiological measurements related to stress, distress or pain include the heart rate, blood pressure, breathing, transcutaneous oxygen saturation, sweating (palms of hands), vagal tone and stress hormone levels (cortisol and norepinephrine).

As an illustration of the pain scales commonly used, we can consider one that may be used at home or during hospitalisation. The CHIPPS (Children’s and Infants’ Postoperative Pain Scale) does not use special measurements to obtain a score, so that these same sort of observations could be made by any care-giver, more or less anywhere. (Note that this example is not an endorsement that this scale is any better or worse than other scales mentioned in the reference section) It was described as recently as 2000, and uses a scale of five items, each rated as 0, 1, or 2: Item “Crying” could be none, moaning, or screaming. Item “Facial expression” could be relaxed smiling, wry mouth, or grimacing about face and mouth. Item “Posture of the trunk” could be neutral, variable, or rear up. Item “Posture of the legs” could be neutral, kicking about, or tightened legs. Item “Motor restlessness” could be none, moderate, or restless. The scores are totalled and the baby managed according to the rule that 0 to 3 indicates no requirement for treating pain, and 4 to 10 indicates progessively greater need for analgesia.

Two aspects of the neonate’s observed pain response deserve special mention. One is that all observations – both movements and physiological parameters – tend to decrease when the neonate is exposed to persistent pain, becoming less reliable as indicators of ongoing pain. This may be the same as the older person’s “strict bedrest” – a response to the increased metabolic requirements of stress and injury. The second, which may seem to be working against the forementioned, is that hyperalgesia, as well as the occurrence of allodynia, happens more quickly and more extensively than in adults. Day to day changes in the response to a specifc injury may therefore become somewhat unpredictable and variable.