Presented at the Neonatal Society 2008 Autumn Meeting.
Arichi T1,4, Melendez A3, Moraux A3, Doria V1, Burdet E3, Beckmann CF2, Edwards AD1,4
1 Division of Clinical Sciences and MRC Clinical Sciences Centre Hammersmith, Imperial College Comprehensive Biomedical Research Centre, DuCane Road, London, W12 0HS, UK
2 Division of Neurosciences and Mental Health, Imperial College Comprehensive Biomedical Research Centre, DuCane Road, London, W12 0HS, UK
3 Department of Bioengineering, Imperial College Comprehensive Biomedical Research Centre, DuCane Road, London, W12 0HS, UK
4 Division of Neonatology, Imperial College Comprehensive Biomedical Research Centre, DuCane Road, London, W12 0HS, UK
Aims/Background: Activation in the primary somatosensory cortex can be identified in the newborn brain using electrophysiological techniques and near infrared spectroscopy, but these methods provide only poorly localised data that do not facilitate detailed analysis of brain development. Functional MRI (fMRI) has been used to detect activation in term infants (1), but not preterm due to difficulties identifying positive Blood Oxygen Level Dependent (BOLD) signals, and the lack of a reliable and reproducible MR-safe stimulus. We hypothesised that well localised cortical activation could be identified in the preterm brain using an appropriate synchronised tactile and proprioceptive stimulus.
Methods: 3 preterm infants of gestational age: 30, 33, and 34 weeks, and 4 term control infants (2 born preterm at 34 and 36 weeks) were imaged using a 3T Philips MRI scanner (Best, Netherlands) located within the Neonatal Intensive Care Unit. Written consent was taken from parents following discussion with one of the investigators and the provision of written information. The somatosensory stimulus was elicited by a programmable hand interface custom-designed by the Department of Bioengineering and tested for MR compatibility. An inflatable latex balloon in the right hand was intermittently inflated via a remote device synchronised to the MR image acquisition. Inflation resulted in the passive extension of the fingers, while deflation allowed flexion. Before use with newborn infants, it was confirmed that the device produced characteristic cortical activation in adult volunteers. MRI was performed with a phased array head coil using an echo-planar imaging (EPI) sequence (TR/TE/FA = 1.5 s/45 ms/90°, 2.5´2.5 mm2, 22 slices) lasting 6 minutes and 30 seconds (total 256 volumes). An “on-off” block paradigm was used with 24 second periods of activity, with balloon inflation/deflation at a frequency of 0.3 Hz. Data analysis was performed using FEAT v5.98 (part of FSL www.fmrib.ox.ac.uk/fsl )2.
Results: Well-localised activation was identified in the term infant group, as described peviously1. In all 3 preterm neonates, passive stimulation of the right hand resulted in contralateral positive BOLD signal activation which was well localised to the primary somatosensory cortex in the left hemisphere, an example is shown.
Conclusion: This preliminary data shows that it is possible to identify well localised functional activation in the primary somatosensory cortex of preterm infants.
References
1. Erberich SG et al. Neuroimage, 29: 155-61, 2006
2. Smith SM et al. Neuroimage, 23(S1):208-219, 2004