Microglial activation and blood-brain barrier permeability in cerebral small vessel disease

Abstract

Cerebral small vessel disease (SVD) is a major cause of stroke and dementia. The underlying pathogenesis is poorly understood, but both neuroinflammation and increased blood-brain barrier permeability have been hypothesized to play a role, and preclinical studies suggest the two processes may be linked. We used PET magnetic resonance to simultaneously measure microglial activation using the translocator protein radioligand 11C-PK11195, and blood-brain barrier permeability using dynamic contrast enhanced MRI. A case control design was used with two disease groups with sporadic SVD (n = 20), monogenic SVD (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy, CADASIL), and normal controls (n = 20) were studied. Hotspots of increased glial activation and blood-brain barrier permeability were identified as values greater than the 95th percentile of the distribution in controls. In sporadic SVD there was an increase in the volume of hotspots of both 11C-PK11195 binding (P = 0.003) and blood-brain barrier permeability (P = 0.007) in the normal appearing white matter, in addition to increased mean blood-brain barrier permeability (P < 0.001). In CADASIL no increase in blood-brain barrier permeability was detected; there was a non-significant trend to increased 11C-PK11195 binding (P = 0.073). Hotspots of 11C-PK11195 binding and blood-brain barrier permeability were not spatially related. A panel of 93 blood biomarkers relating to cardiovascular disease, inflammation and endothelial activation were measured in each participant; principal component analysis was performed and the first component related to blood-brain barrier permeability and microglial activation. Within the sporadic SVD group both hotspot and mean volume blood-brain barrier permeability values in the normal appearing white matter were associated with dimension 1 (β = 0.829, P = 0.017, and β = 0.976, P = 0.003, respectively). There was no association with 11C-PK11195 binding. No associations with blood markers were found in the CADASIL group. In conclusion, in sporadic SVD both microglial activation and increased blood-brain barrier permeability occur, but these are spatially distinct processes. No evidence of increased blood-brain barrier permeability was found in CADASIL.

Keywords: CADASIL; blood–brain barrier; cerebral small vessel disease; lacunar stroke; neuroinflammation.

Figure 1

BBB permeability comparisons between the groups. (A) Mean BBB permeability (K_i). Values were higher in sporadic SVD compared to control for NAWM (P < 0.001) and WMH (WML) (P = 0.003). Values were not significantly different between CADASIL and control in the NAWM (P = 0.34) or WMH (P = 0.51). (B) Volume of BBB permeability hotspots. Values were significantly higher in sporadic SVD compared to control in NAWM (P = 0.007) and WMH (P = 0.004). Values were not significantly different between CADASIL and control in the NAWM (P = 0.87) or WMH (P = 0.51). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2

¹¹C-PK11195 binding comparisons between the groups. (A) Mean ¹¹C-PK11195 BP_ND. Values were not different between sporadic SVD and control in the NAWM (P = 0.59) or WMH (WML) (P = 0.63). Similarly, values were not different between CADASIL and control in the NAWM (P = 0.84) or WMH (P = 1.00). (B) Volume of ¹¹C-PK11195 binding hotspots. Values were significantly higher in the sporadic SVD group in the NAWM (P = 0.003) and WMH (P = 0.004). Values were not different between CADASIL and control in the NAWM (P = 0.073) but were significantly higher in the CADASIL group in the WMH (P < 0.001). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3

Hotspots of increased BBB permeability and ¹¹C-PK11195 binding. Example images from four different SVD subjects (A–D) showing hotspots of increased BBB permeability (green) and ¹¹C-PK11195 binding (yellow) overlaid onto T₂ FLAIR images.