Presented at the Neonatal Society 2008 Spring Meeting.

Durighel G, Xue H, Allsop J, Counsell S, Srinivasan L, Rutherford M

Imaging Sciences Department, Division of Clinical Sciences, Hammersmith Hospital Imperial College, London, UK

Background: During evolution individual brain structures significantly expand in relationship to one another leading to increased functional organisation1. The thalamus plays an important role in relaying and modulating the flow of information to the cerebral cortex and both probably develop in a coordinated manner. We know from deformation based morphometry and segmentation that preterm infants at term equivalent age have reduced thalamic and lentiform volume in spite of preserved total brain volume compared to term born neonates2&3. However, the relationship between the thalamic, cortical and white matter growth has not been studied in these infants.

Aim: Our aim was to gain a better insight into the growth pattern of the thalamus in relation to cortical and white matter development in preterm infants at term equivalent age. 

Methods: 37 preterm neonates (born at 24 to 32 wk of gestational age (GA)) were recruited from the NICU at Hammersmith Hospital with approval from the Research Ethics Committee. These infants were scanned between 38 and 44 weeks postmenstrual age (PMA). Images were acquired on a 3 Tesla Phillips Medical System (Best, Holland) using an 8 channel head coil. T1 and T2 volume scans were acquired with voxel dimensions of 0.86cm3. Thalami were manually segmented using T1 weighted images and Image J.  Total brain volume, cortical and white matter (WM) volumes were segmented using an automated EM/Markov/MLPV technique using the T2 weighted scans. Statistical analysis was performed using SPSS. Regression analysis was carried out to test the relationship of GA and PMA with volumes, whereas the relationships between the individual volumes were analysed using partial correlation coefficients.  

Results: Of the 37 thalamic data, 3 volume measurements were excluded due to the presence of underlying brain pathology. The median thalamic volume was 15.6 cm3 (13.04-19.47cm3) and the median cortical volume was 181.9cm3 (132.2-286.4cm3). Postmenstrual age influenced the total brain, cortical and thalamic volumes and had no effect on WM volumes, whereas GA at birth was not significantly related to any volumes. There was a positive correlation between thalamic and cortical volume (figure 1), however partial correlations showed that when controlling for total brain volume both cortical volume and thalamic growth were inversely related to white matter volume (table1).

Figure 1: The relationship between the thalamus and cortical volume of the
preterm at term brain.

Table 1: gives the partial correlations with 2-sided significance of the different
brain structure after controlling for total brain volumes.  

Conclusion: This study of preterm infants provides new insights into the developmental nature of the thalamus in relation to the cortex, white matter and total brain at term equivalent age. Of interest, increases in grey matter volume, both central and cortical, are associated with a proportional decrease in WM volume, perhaps as a result of normal pruning of WM tracts. However, in preterm infants with decreased thalamic and cortical volumes, suggestive of injury, there was a proportional increase in WM volume. A lack of pruning may explain the apparent “preservation” of total brain volumes at term equivalent age, as seen in previous studies.  It may also give rise to the increased diffusivity4 (ADC values) commonly seen in areas of WM injury (DEHSI) that is known to be associated with cognitive and behavioural problems in later life.   

References
1. R Barton et al. Nature 2000;405: 1055-1058
2. Boardman et al Neuroimage 2007
3. L Srinivasan et al. Paediatrics 2007;19:759-765
4. Counsell et al Paediatrics 2003.