There is growing evidence linking microbleeds to macrobleeding in selected patient populations. Yet, prospective data in unselected populations are scarce. The high prevalence of microbleeds and relatively low incidence of symptomatic intracerebral haemorrhage suggest that it may be a specific sub-group of persons with microbleeds developing macrobleeding. This may be related to the severity and nature of the underling vascular pathology. Future studies should therefore use a longitudinal design and not only focus on microbleed presence, but also on location and number of microbleeds.