Duy Le, MD
Sporadic cerebral amyloid angiopathy (CAA) is a common entity that we see manifested as intracerebral hemorrhages in elderly patients. This article attempts to identify imaging markers which may be associated with patients who may develop CAA or cerebral superifical siderosis (CSS). The feeling is that MRI-visible centrum semiovale perivascular spaces (CSO-PVS) is such a potential biomarker for these two entities.
To achieve this, the Charidimou et al retrospectively evaluated 138 CAA patients based on possible/probably CAA according to the Boston Criteria. This was a multicenter study. Raters were blinded and patients were deemed to have degree of CSO-PVS (0=no PVS, 1=<10 PVS, 2=11-20 PVS, 3=21-40 PVS and 4=>40 PVS) based on 1.5T MRIs. This was evaluated for both the centrum semi ovale and the basal ganglia. The results show that High CSO-PVS degree was present in 61.2% of CAA cases. There was no correlation or increase with basal ganglia PVS. The prevalence of any CSS or disseminated CSS (involving > 3 sulci), was higher in patients with high CSO-PVS degree as compared to the low CSO-PVS degree. Using multivariable logistic regression analysis, CSS presence (OR 4.78; p 0.004) was an independent predictor of high CSO-PVS degree.
Ultimately, this study shows that a high degree of MRI-visible CSO-PVS are prevalent in patients with CAA. Although only correlation can be drawn from this study, and not causation, this supports the potential idea that patients with CAA may have underlying fluid drainage impairment within the perivascular spaces. Because of this, beta amyloid is trapped and deposited in the walls of small arteries. Although the results of this study cannot change current clinical practice, it opens up the opportunity for a prospective study which may be able to establish causation. And from there, if we can predict the chances of someone developing CAA, we can decide who to be wary of anti-coagulating.
