Embryological Consideration of Dural AVF

The distribution of intracranial dural AVFs (DAVFs) may be affected by the embryological bony structures that consist of membranous bone and endochondral bone.

We retrospectively analyzed the distribution of the shunt points in 58 consecutive cases of DAVFs. Shunt points were identified with selective digital subtraction angiography, high-resolution cone beam computed tomography (CT), or three-dimensional rotation angiography. All the shunt points were plotted on the map of the skull base in relation to the topography of the endochondral bone and the membranous bone. If the shunt point was localized on the surface of endochondral bone, this was categorized as the endochondral bone group. If it was located on membranous bone, this was categorized as the membranous bone group. If the shunt point was independent from both bony structures, this was categorized as the independent group.

In 55 of 58 cases, shunt points were identified angiographically. Three cases had multiple shunts. There were 33 shunt points (60 %) belonging to endochondral bone. In this group, 16 cases of sigmoid, 11 of carotid cavernous, 3 of petrosal apex, and 3 of sigmoid DAVF were observed. There were 12 shunt points (22 %) localized on membranous bone; in this group, there were nine cases of transverse sinus, two of superior sagittal sinus, and one case of confluence DAVF. There were ten shunt points (18 %) independent from these two bony structures: four cases of olfactory groove, four . of middle fossa, and two of hypoglossal canal DAVF.

There were correlations between the localization of shunt points of DAVFs and the topography of endochondral bone and the membranous bone. The histological difference of endochondral bone and membranous bone at the level of epidural space might cause the formation of DAVFs.