NEONATAL SOCIETY ABSTRACTS
Secondary energy failure in a piglet model of hypoxic ischaemic brain injury assessed by serial phosphorous magnetic resonance spectroscopy, water apparent diffusion and electrophysiology: a pilot study
Presented at the Neonatal Society 2004 Autumn Meeting (programme).
West DA1, Iwata O1, De Vita E2, Bainbridge A2, Iwata S1, Cheong JL1, Cady EB2, Ordidge RJ3, Wyatt JS1, Robertson NJ1
1 Paediatrics and Child Health, University College London, London, UK
2 Medical Physics and Bioengineering, University College Hospitals NHS Trust, London, UK
3 Medical Physics and Bioengineering, University College London, London, UK
Background: The electroencephalogram (EEG) provides a sensitive means of predicting outcome early after perinatal hypoxia-ischaemia (HI). The amplitude integrated EEG (aEEG), a compressed, rectified and filtered form of 1-channel EEG, is increasingly used clinically to define infants who may benefit from neuroprotective intervention following perinatal HI. The precise relationship between electrophysiology and brain energetics remains unclear however. Phosphorous-31 magnetic resonance spectroscopy (31P MRS) and magnetic resonance imaging measurements of the apparent diffusion coefficient of water (ADC) provide useful information about local tissue energetics following HI.
Objective: (i) To compare electro-cortical activity to cerebral energetics and ADC during the evolution of secondary energy failure (SEF); (ii) To assess the correlation between aEEG and standard EEG during HI and SEF.
Design/Methods: A newborn piglet was studied under general anaesthesia (isoflurane & fentanyl) before, during and after acute HI (carotid occlusion and reduction of FiO2 to 12% for 45min). The standard EEG (8 surface electrodes, 10-20 system) and aEEG were recorded for 30min before, immediately following and approximately every 4hrs after HI. Wholebrain 31P MRS spectra and ADC maps from the central axial slice were also obtained.
Results: Changes in the EEG are summarised in Table 1. The corresponding aEEG recordings are shown in Fig.1. Abnormal burst (4hrs post-HI) and seizure activity (11hrs post-HI) were transformed into similar appearing traces on aEEG. The density of high voltage activity was highest during seizures. Energetic status assessed by phosphocreatine to inorganic phosphate ratio ([PCr]/[Pi]) and ATP to total exchangeable phosphate pool ([ATP]/[EPP]) is shown in Table 1. ADC was reduced as early as 4hrs after the insult, initially in the deep grey matter and watershed deep white matter, extending throughout the brain by 15hrs (Table 1 & Fig.1).
Table 1: Serial EEG, 31P MRS and ADC data collected at baseline, during and following resuscitation from acute HI.
Figure 1: Serial aEEG traces and ADC parameter images collected at baseline and following resuscitation from acute HI.
Conclusion: Changes in electro-cortical activity, brain energy metabolism and water diffusion were observed during the evolution of SEF. The aEEG trace was sufficiently sensitive to capture globally abnormal cortical activity in this pilot study. We recorded abnormal burst activity many hours before PCr/Pi decreased. Future work aims to fully characterise the relationship between electrophysiology and brain energetics during SEF. Such information will facilitate the selection of infants who may benefit from neuroprotective strategies.