Do Regions of Increased Inflammation Progress to New White Matter Hyperintensities?: A Longitudinal Positron Emission Tomography-Magnetic Resonance Imaging Study

Abstract

Background: Recent studies have demonstrated increased microglial activation using 11 C-PK11195 positron emission tomography imaging, indicating central nervous system inflammation, in cerebral small vessel disease. However, whether such areas of neuroinflammation progress to tissue damage is uncertain. We determined whether white matter destined to become white matter hyperintensities (WMH) at 1 year had evidence of altered inflammation at baseline. Methods: Forty subjects with small vessel disease (20 sporadic and 20 cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy) and 20 controls were recruited to this case-control observational study from in- and out-patient clinics at Addenbrooke’s Hospital, Cambridge, UK and imaged at baseline with both 11 C-PK11195 positron emission tomography and magnetic resonance imaging; and magnetic resonance imaging including diffusion tensor imaging was repeated at 1 year. WMH were segmented at baseline and 1 year, and areas of new lesion identified. Baseline 11 C-PK11195 binding potential and diffusion tensor imaging parameters in these voxels, and normal appearing white matter, was measured. Results: Complete positron emission tomography-magnetic resonance imaging data was available for 17 controls, 16 sporadic small vessel disease, and 14 cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy participants. 11 C-PK11195 binding in voxels destined to become new WMH was lower than in normal appearing white matter, which did not progress to WMH (−0.133[±0.081] versus −0.045 [±0.044]; P <0.001). Mean diffusivity was higher and mean fractional anisotropy lower in new WMH voxels than in normal appearing white matter (900 [±80]×10 −6 versus 1045 [±149]×10 −6 mm 2 /s and 0.37±0.05 versus 0.29±0.06, both P <0.001) consistent with new WMH showing tissue damage on diffusion tensor imaging a year prior to developing into new WMH; similar results were seen across the 3 groups. Conclusions: White matter tissue destined to develop into new WMH over the subsequent year is associated with both lower neuroinflammation, and white matter ultrastructural damage at baseline. Our results suggest that this tissue is already damaged 1 year prior to lesion formation. This may reflect that the role of neuroinflammation in the lesion development process occurs at an early stage, although more studies over a longer period would be needed to investigate this further.