Alejandro Rodríguez-Vázquez, MD
Nowadays, Boston criteria are pillars for the diagnosis of cerebral amyloid angiopathy (CAA). According to those criteria, CAA diagnosis is excluded due to the presence of deep microbleeds independent of other neuroimaging findings and in absence of pathology. However, Thal et al. proposed that CAA pathology extends sequentially from cortical to cerebellar and finally basal ganglia and brainstem vessels. Therefore, deep microbleeds added to lobar both cerebral and cerebellar ones do not exclude CAA but are related to an advanced stage of the disease.
In this study, the authors tried to determine if the topographical pattern of microbleeds could help to establish the underlying pathophysiology comparing amyloid-β burden and cerebral small vessel disease markers according to the anatomic distribution of microbleeds. From 2333 patients who visited the authors’ memory clinic, they included 71 with suspected CAA markers on MRI neuroimaging, and they categorized them in 4 groups based on the distribution of microbleeds: strictly lobar (n=33); strictly lobar cerebral and strictly lobar cerebellar (n=13); lobar, cerebellar both lobar and dentate and deep (n=17); and lobar and deep (n=7). Brainstem microbleeds were categorized as deep. In addition, they performed an amyloid-β PET and a complete cognitive assessment. The prevalence of hypertension was slightly higher on the lobar and deep group than in the strictly lobar, but the rest of classic cardiovascular risk factors and dementia did not significantly differ between groups.
After the analysis, patients with strictly lobar microbleeds were more likely to have CAA-like imaging features, whereas any combination of concurrent lobar and deep microbleeds, regardless of cerebral or cerebellar compartment, were associated with hypertensive angiopathy (Figure 2).
In addition, the strictly lobar cerebral/cerebellar group had a higher amount of lobar microbleeds than the strictly lobar cerebral. Considering that the microbleed burden predicts more advanced CAA, the presence of lobar microbleeds in both cerebral and cerebellar tissues could represent a more evolved stage of the disease. Thus, these findings support Thal’s hypothesis of anatomic CAA propagation: first neocortical and leptomeningeal vessels and cerebellar implication in a second, more advanced stage.
The study, however, has some limitations. First of all, due to absence of pathology, it is impossible to make a definite CAA diagnosis. In this line, amyloid-β PET is not enough to differentiate vascular from parenchymal distribution, so its positivity could be related to concurrent dementia rather than to CAA, although there are some experts that support a continuum of CAA-Alzheimer disease. There is also a selection bias because of the participants being recruited from a memory clinic with at least some degree of cognitive impairment. Besides, mixed amyloid and hypertensive findings are common in older people, so the coexistence of hypertensive angiopathy could not be completely excluded even in the strictly lobar microbleeds group. Even considering its limitations, the results of this study represent another step towards further clarifying CAA evolution and its clinical implications.
Really it is time tested that cerebral lobar bleeds or microbleeds are suggestive of cerebral amyloid angiopathy. Of course the things need to be upgraded in future continuous studies and observations and the new techniques.