Assessment of pain and its treatment in neonates

Abstract

Babies on the neonatal unit are often exposed to a number of noxious stimuli. Procedures such as venepuncture, insertion of cannulae, heel prick and endotracheal ventilation can occur on a daily basis, and the majority of these babies do not receive any pain relief treatment. Current methods of assessing pain in neonates, comprising behavioural cues and physiological markers, can be difficult to relate directly to pain perception, especially in preterm infants, as they can also be indicative of other states such as stress, anxiety and hunger. Accurate recognition of pain in a neonate could enable more appropriate use of analgesics, thereby reducing morbidity and limiting potential side-effects.

We investigated 48 babies using techniques that may evaluate the function of nociceptor-specific nerve fibres and central pain pathways, in comparison with currently used behavioural and physiological indicators. These techniques included “objective” physiological responses to procedural pain, such as skin axon-reflex vasodilatation (flare) responses, and novel contact cerebral evoked potentials in response to warm (non-painful) stimuli.

We recorded physiological responses (i.e. sweat rate, changes in blood flow, sympathetic skin response) during clinically required heel prick procedures in babies on the neonatal unit. Palmar sweat rate was higher in babies ≥ 36 weeks, and significantly correlated with gestational age. We did not see any relationship between the painful heel prick procedure and palmar sweat response. We were unable to demonstrate any increase in the sympathetic skin response waveforms post-heel prick procedure in the one baby we studied. Changes in local skin blood flow around the heel prick (ipsi-lateral limb) and in the opposite foot (contralateral limb) were very variable, although many babies showed an increase in the blood flow post-heel prick. We also performed a behavioural analysis (using PIPP) during heel prick procedures in babies whilst simultaneously recording their palmar sweat levels. We could not demonstrate any definite association between the palmar sweat response to pain and the PIPP scores.

The feasibility of recording warm temperature induced cerebral evoked potentials in neonates was investigated, using scalp EEG electrodes, in response to a cutaneous stimulus with a thermode destination temperature 37°C. However, the EEG waveforms were obscured by ocular artefact, and thus warm temperature evoked potentials could not be demonstrated in neonates, unlike adults.

In conclusion, current methods of behavioural and physiological assessment of pain in neonates have an important role, but are not specific and unlikely to be helpful in monitoring the effects of new analgesics. Warm stimuli evoked cerebral potentials are technically difficult to record in neonates, and heat pain stimuli are unjustified, but other techniques such as functional MRI may provide a useful measure and deserve study. The assessment and treatment of pain in neonates remains a major unmet clinical need.