From an ongoing stroke registry, including all patients with a first cerebral infarct, we studied 41 borderzone small deep infarcts: (SDIs) in comparison with 123 remaining SDIs. We noted vascular risk factors, presence of a carotid occlusion or Stenosis > 50%, and the presence of asymptomatic lesions on CT. By univariate analysis and subsequent multivariate logistic regression analysis, we found that carotid stenosis/occlusion was more frequent among the borderzone group, but this difference was not statistically significant. The vascular risk factor profile was similar in both groups. Asymptomatic lesions, but especially asymptomatic borderzone small deep lesions were significantly more frequent among the symptomatic borderzone group. We conclude that carotid stenosis is not a numerically important contributor to the occurrence of borderzone SDI; generalised small-vessel disease may be more important in this respect. Therefore borderzone SDIs are most likely due to haemodynamic compromise in the distal supply areas of the perforating arteries.