Estimation of prevalence and incidence rates in a disease is crucial to estimate the risk of the natural evolution of the disease. We have reviewed published data on intracranial AVM and have noted that reported estimated risks were quite variable mainly due to the lack of population-based prospective data. Nonetheless, estimation of these parameters can be made from selected populations. We will note also that there is no published data confirming a genetic origin for the most frequent group of sporadic AVM.