Search

Pain assessment is critical to prevent suffering and harm in infants admitted to the neonatal care unit. As pain is a subjective experience, its assessment in nonverbal infants relies on surrogate measures. Current infant pain assessment tools that are based on behaviour and autonomic nervous system measurements lack face validity - they are unlikely to reflect pain in all its dimensions. In recent years, EEG-derived measures of pain have been developed in late preterm and term infants. Multimodal tools which include these cerebral measurements are conceptually more appropriate to measure pain. Yet, their use is still limited to specific research applications. This thesis focuses on outstanding questions that need to be addressed in order to advance the development of multimodal pain assessment tools that incorporate cerebral measurements. In the first part of this thesis, I focus on the characterisation of preterm infants' noxious-evoked responses and their development. Across several modalities, premature infants have dampened or altered responsiveness compared to term infants, and it is uncertain if these responses can be reliably discriminated from tactile-evoked responses. In particular, a discriminative pattern of noxious-evoked EEG activity that is present in term infants, is unlikely to be present in preterm infants. In addition, it is unclear how noxious-evoked responses, especially brainderived responses, change with age. In this thesis, I use a classification model to show that infants aged 28-40 weeks postmenstrual age display discriminable multimodal responses to a noxious clinical procedure and a tactile control procedure, and I provide examples of how a such a model could be used in clinical trials of analgesics. I show that noxious-evoked responses change magnitude and morphology across this age range, and that discriminative brain activity emerges in early prematurity. In the second part of this thesis, I focus on improving the neuroscientific validity of a noxious-evoked EEG response measured at the cot-side, as the spatial neural correlates of these responses are still poorly understood. I present an EEG-fMRI pilot study to investigate the spatial neural correlates of inter-individual differences in noxious-evoked EEG responses and provide recommendations for a larger follow-up study. Overall, this thesis provides a characterisation of infants' noxious-evoked responses and their development across multiple modalities, a crucial next step in improving multimodal neonatal pain assessment.