Contact us / Members' area    

Home > Abstracts > Index > 2007 > Anjari M, et al


Diffusion tensor imaging with tract-based spatial statistics reveals local white matter abnormalities in preterm infants

Presented at the Neonatal Society 2007 Spring Meeting (programme).

Anjari M1, Srinivasan L1, Allsop JM1, Hajnal JV1, Rutherford MA1, Edwards AD1,2, Counsell SJ1

1 Imaging Sciences Department, MRC Clinical Sciences Centre, London, UK
2 Department of Paediatrics, Imperial College London, Hammersmith Campus, Du Cane Road, London, W12 0HS, UK

Background: Infants born preterm have a high incidence of neurodevelopmental impairment in later childhood, often associated with poorly defined cerebral white matter abnormalities. Diffusion tensor imaging (DTI) quantifies the diffusion of water within tissues. By calculating the eigenvalues of the diffusion tensor, diffusion parallel and perpendicular to the white matter tracts can be measured, which provide non-subjective measurements that reflect tissue microstructure and can be used to assess microstructural abnormalities in the developing preterm brain. Tract based spatial statistics (TBSS) (1) is an automated observer-independent method of aligning fractional anisotropy (FA) images from multiple subjects to allow groupwise comparisons of DTI data.

Aim: The aim of this study was to determine if TBSS could be implemented in the preterm population, and to test the hypothesis that preterm infants have microstructural differences in cerebral white matter compared to term born control infants in the absence of focal abnormalities such as cystic periventricular leukomalacia (cPVL) or haemorrhagic parenchymal infarction (HPI) on conventional magnetic resonance imaging (MRI).

Subjects: Ethical permission for this study was granted by the Hammersmith Hospital Research Ethics Committee and written parental consent was obtained prior to imaging for each subject. We studied 26 preterm infants (11 females) with no evidence of focal lesions on conventional MRI at term equivalent age. The median (range) gestational age of the infants at birth was 28.9 (25.732.6) weeks, and the median post-menstrual age at the time of imaging was 41.3 (38.145.3) weeks. This group contained a subset of 11 infants (4 females) who were born at 28 weeks gestation or less, with a median (range) gestational age at birth of 26.7 (25.728.0) weeks, and median post-menstrual age at the time of imaging of 41.0 (38.144.0) weeks. These images were compared with DTI data obtained from 6 healthy term-born control infants (2 females), with a median (range) gestational age of 39.7 (39.040.6) weeks, and median post-menstrual age at time of imaging of 41.7 (41.046.0) weeks. There were no significant differences in age at scanning (p = 0.24) or in gender (p = 0.53) between the preterm group and the term-born controls. There were no significant differences in age at scanning (p = 0.14) or in gender (p = 0.47) between the subset of preterm infants born ≤ 28 weeks gestational age and term-born control infants.

Methods: Voxelwise statistical analysis of the FA data was carried out using TBSS implemented in FSL (1). All subjects' FA data were aligned into a common space using a non-linear registration algorithm ( The mean FA image (thresholded to FA ≥0.20) was then created and thinned to generate a mean FA skeleton which represented the centres of all tracts common to the group. Each subject's aligned FA data was then projected onto this skeleton and the resulting data was fed into voxelwise cross-subject statistics (t > 3, p < 0.05, uncorrected).

Results: We found that the centrum semiovale, frontal white matter and the genu of the corpus callosum showed significantly lower FA in the preterm group (t > 3, p < 0.05). Infants born at less than or equal to 28 weeks gestational age (n = 11) displayed additional reductions in FA in the external capsule, the posterior aspect of the posterior limb of the internal capsule and the isthmus and middle portion of the body of the corpus callosum. We analysed the three eigenvectors of the diffusion tensor and found that regions which exhibited decreased FA showed elevated intermediate (λ2) and/or minor (λ3) eigenvalues.

Discussion: TBSS is a powerful observer independent method that was able to survey the whole brain to identify sites of white matter abnormality in preterm infants at term equivalent age. This observed decrease in FA is consistent with delayed maturation or oligodendrocyte and/or axonal damage.

1. Smith SM, Jenkinson M, Johansen-Berg H et al. NeuroImage 2006; 31:1487-1505.

Home / The Neonatal Society, founded 1959 / 2001 - 2012