Presented at the Neonatal Society 2014 Spring Meeting.
Lally PJ1, Zhang H2, Pauliah SS1, Price DL3, Bainbridge A3, Cady EB3, Shankaran S4, Thayyil S1
1 Imperial College London, London, UK
2 University College London, London, UK
3 University College Hospital, London, UK
4 Wayne State University, Detroit, USA
Background: Magnetic resonance (MR) based measurements of cerebral diffusivity, such as fractional anisotropy (FA), are influenced by microstructural change but cannot distinguish between pathological changes in, for example, neuronal density or neuronal reorganisation. We hypothesised that the neurite orientation dispersion and density imaging (NODDI) model (1) would allow us to better elucidate these changes associated with neonatal encephalopathy, and relate these to neurodevelopmental outcomes in early childhood.
Methods: We recruited consecutive term or near-term neonates in Government Medical College, Kozhikode, with evidence of perinatal hypoxic-ischaemia, using a <6h Thompson encephalopathy score (2) of ≥6. MRI was performed at 1.5 Tesla within three weeks after birth as outlined previously (3). Using two models of cerebral water diffusion (DTI and simplified NODDI) we generated maps of diffusion-based variables and compared these between subjects based on the outcome of neurological assessments at 3½ years of age (Bayley III scoring) (4), using tract based spatial statistics (5). To support the fitting of the NODDI model with single-shell DTI data, we excluded the isotropic component (which models CSF partial volume) and fitted this simplified model to voxels in the posterior limb of the internal capsule, where CSF contamination is minimal. The Government Medical College Institutional Review Board and University College London Ethics Committee approved the study.
Results: Fifty-four neonates were recruited, of which 26 had both MR and outcome data for analysis. An abnormal neurological outcome at 3½ years was associated with a significant decrease in fractional anisotropy (FA) in the posterior limb of the internal capsule, driven by a reduced neurite density as modelled with a simplified NODDI model (see Figure: mean cohort neurite density index map, with PLIC colour coded green where p≥0.05, yellow-red where p<0.01-0.05). No difference in orientation dispersion index (ODI) was seen in this region between groups.
Conclusion: The reduced FA seen in neonates with poor neurological outcomes in early childhood3 is driven by a decreased neurite density, as fitted with a simplified NODDI model, and not with a change in the neurite orientation dispersion. When performing group-wise comparisons of cerebral diffusion properties, NODDI fitting provides further insight into biophysical mechanisms of disease progression.
Corresponding author: email@example.com
1. Zhang H et al. NeuroImage 2012;61:1000-1016
2. Thompson CM et al. Acta paediatrica 1997;86:757-761
3. Lally PJ, Price DL et al. PLoS ONE 2014;9(2): e87874
4. Bayley N. Bayley Scales of Infant and Toddler Development® 3rd Edition (Bayley-III)
5. Smith SM et al Neuroimage 2006;31(4):1487-505