Presented at the Neonatal Society 2006 Spring Meeting.
Raafat RM, Dresner MA, Counsell SJ, Srinivasan L, Hajnal JV, Edwards AD
MRC Clinical Sciences Centre and Clinical Sciences Division, Hammersmith Hospital, Imperial College London, London, United Kingdom.
Background: The neuroanatomical substrate for cognitive and neuropsychiatric impairment following premature delivery is incompletely understood, hampering efforts to develop neuroprotective strategies for these infants. Combining diffusion tractography and functional magnetic resonance imaging (fMRI) allows cerebral structure-function relationship to be assessed non-invasively, but to date whole-brain fMRI responses of young children to visual stimuli using 3.0T field strength have not been studied. We performed experiments to determine if cortical connectivity in the optic radiation and elsewhere develops symmetrically in preterm infants, and to test the hypothesis that brain activation due to visual stimulus will result in a reproducible pattern. We then asked whether an infant with a significant brain lesion would show asymmetrical connectivity and cerebrocortical activation.
Methods: Approval was granted by the Research Ethics Committee and written parental consent was obtained prior to scanning. To ascertain typical values for (i) connectivity and (ii) activation in the preterm population: (i) Diffusion Tensor Imaging (DTI) at 3T was performed on 11 infants who were born preterm (median (range) gestational age at birth = 31 (25.1 – 34.4) weeks) and imaged at a median corrected age of 24.6 months (range 21 – 26.4 months). Connectivity distributions in the optic radiations were investigated using the FMRIB diffusion toolkit (FDT) (1). (ii) Brain activations due to a diffuse full-field, 1 Hz light stimulus were assessed in a group of 12 sedated children born preterm (median (range) gestational age at birth = 28.8 (25.7 – 33.7) weeks) and imaged at a median corrected age of 14.5 months (range 6 – 28.6 months) using fMRI at 3T. All subjects also had structural imaging using T1- and T2-weighted protocols, and the structural images were reviewed prior to analysis. Data from children with normal structural scans were registered to the T2-weighted scan of one subject and assessed in a random-effects (RFX) model in SPM2. We then examined in detail the connectivity and activation in one preterm infant with focal brain damage detected on structural imaging.
Results: In infants without focal brain lesions: (i) diffusion tractography visualised the optic radiations in all infants studied, showing equal connectivity bilaterally (ii) a consistent negative BOLD effect was detected; RFX analysis showed symmetrical activation around the calcarine fissure, consistent with bilateral activation of V1-3 visual areas. In the infant with focal injury, connectivity was reduced in the optic radiation ipsilateral to the lesion and fMRI showed less activation on that side.
Discussion: fMRI responses to light in sedated children result in a characteristic negative BOLD response and the increased signal to noise ratio afforded by imaging at 3 Tesla provides robust measures of connectivity and functional activation. Combining fMRI and diffusion tractography has the potential to be an effective approach to understanding abnormalities in the brains of preterm infants.
References
1. Behrens et al. Nature Neuroscience 2003; 6: 750-7.