Abstract
BACKGROUND: Advances in in vivo MRI techniques enable cerebral barrier transfer rates (K (trans) ) measurement in patients with vascular cognitive impairment and dementia (VCID). However, a consensus has not been reached on the dynamic contribution and importance of cerebral barrier abnormalities to the differential diagnosis of dementia subtypes. Our goal was to investigate the dynamics of blood-brain barrier (BBB) and blood-CSF barrier (BCSFB) K (trans) in patients with VCID longitudinally and determine the effect of aging. METHODS: We studied subjects at two time points over two years; they were 65.5 years of age (SD = 15.94, M/F = 24/14) at the first visit. We studied 38 patients, 18 of whom had two visits. We calculated the BBB and BCSFB K (trans) with dynamic contrast-enhanced T1 MR, and we used (1)H-MR spectroscopy to measure N-acetylaspartate (NAA) levels in the white matter as a marker of injury. In addition, we measured CSF levels of active-matrix metalloproteinase-3 (MMP3) as an inflammatory biomarker to aid in patient clustering. RESULTS: Longitudinal BBB measurements revealed variable dynamic behavior: after two years, the BBB K (trans) increased in 55% of patients and decreased in the remaining 45% unpredictably. We did not find a significant linear model of BBB K (trans) versus age for VCID. For healthy controls, the model was K (trans) = 0.0014 + 0.0002 × age, which was significant (p = 0.046). VCID patients showed a reduction in BCSFB K (trans) compared to healthy controls (p = 0.01). Combining NAA, CSF MMP3, and K (trans) in a clustering analysis separated patients into groups. CONCLUSION: These results suggest that BBB K (trans) in VCID is dynamic and BCSFB K (trans) reduced by age. By combining inflammatory biomarkers with BBB K (trans) data, it is possible to separate VCID patients into distinct groups with different underlying pathologies.