Brain damage in experimental neonatal hydrocephalus: correlations between diffusion tensor imaging and cytopathology
Authors: James P. (Pat) McAllister, Kelley E. Deren, Ahmed Shereen, Weihong Yuan, Diana M. Lindquist, Scott K. Holland, Francesco T. Mangano
Background
Diffusion tensor imaging (DTI) is an advanced non-invasive
magnetic resonance imaging (MRI) technique used clinically to
quantify white matter (WM) abnormalities in various pathologic
conditions, and thus could be beneficial in detecting progressive
brain damage in hydrocephalus. This study was designed to correlate
DTI and cytopathology in a rat model of neonatal hydrocephalus.
Materials and Methods
Obstructive hydrocephalus was induced by intracisternal
injection of kaolin on postnatal day 1 (P1); age-matched control
animals were either intact or received intracisternal injections of
saline. Six animals (3 hydrocephalics and 3 intact controls) were
imaged in vivo at 8-9 days of age (P8-9) and then sacrificed by
cardiac perfusion of paraformaldehyde. Four animals (2
hydrocephalics and 2 saline controls) were sacrificed at postnatal
day 21 (P21) and paraformaldehyde-fixed brains were imaged ex vivo
with a Bruker 7T MRI scanner. DTI was acquired in 6 directions to
obtain measurements of the directionality of water diffusion
through tissue (Fractional Anisotropy, FA; 0 = isotropic where
water can move freely in any direction; 1 = anisotropic where water
can move in only one direction) and the magnitude of diffusivity of
water in tissue (Mean Diffusivity, MD). Values were computed
bilaterally in the genu of the corpus callosum (gCC), external
capsule (EC), internal capsule (IC), and cortical gray matter
(Ctx).
Results
At P8-9 ventriculomegaly was severe and by P21 had progressed to
where the periventricular white matter and internal capsule were so
thin that DTI could only be reliably performed ex vivo. At both
times FA values were significantly reduced in the gCC (p<0.001)
but not the EC, IC or Ctx. There was a trend toward higher FA
values at P21 than at P8/9 for both hydrocephalic rats and normal
controls. In contrast, MD values increased only in the EC.
Conspicuous gliosis was prevalent in all structures examined but
differed depending on cell type. Younger animals showed a more
robust reaction of microglial cells compared to 21 day old
hydrocephalics. Astrocytes exhibited the opposite pattern with a
more robust reaction in the older animals.
Conclusions
These results demonstrate the feasibility of applying DTI to
experimental hydrocephalus in neonatal rats, and reveal impairments
in the corpus callosum and external capsule. The reduction in FA
with hydrocephalus suggests that structural and perhaps
physiological impairments exist in cortical connectivity.
Whether the associated gliosis in these structures is causative or
a response to axonal and myelin damage requires further study.
Department of Neurosurgery, Primary Children's Medical
Center, Salt Lake City, Utah
Email: pat.mcallister@hsc.utah.edu
