Object: Endoscopic third ventriculostomy is the treatment of choice in patients with obstructive hydrocephalus caused by aqueductal stenosis. The authors examined the clinical course and results of surgical treatment for obstructive hydrocephalus with pre- and postoperative refined constructive interference in steady-state (CISS) MR imaging.
Methods: Forty patients with obstructive hydrocephalus underwent pre- and postoperative 3D-CISS imaging and clinical evaluation. Radiological findings were correlated with intraoperative observations of the thickness and transparency of the floor of the third ventricle and the patient's postoperative clinical course.
Results: Three-dimensional CISS MR imaging provides precise visualization of the basilar/posterior cerebral artery, its distance to the clivus, the diameter of the foramen of Monro, and the extension of and thickness of the floor of the third ventricle. In 71% of patients a flow void was detectable postoperatively on the ventriculostomy. In this group 81.5% had strong and 14.8% moderate clinical benefit, and 3.7% required secondary shunt placement. In the remaining 29% of the patients without a visible flow void, strong improvement was seen in 54.5%, moderate improvement in 18.2%, and stoma failure occurred in 27.3% (p = 0.094). Radiological measurements of the thickness of the third ventricle floor correlated with intraoperative findings (r = 0.35, p = 0.029). Comparison of outcomes showed a statistically significant tendency for a better outcome in patients with thin and easily perforated third ventricle floors (p = 0.04).
Conclusions: Endoscopic ventriculostomy in patients with obstructive hydrocephalus is safe and mostly successful, and 3D-CISS MR imaging seems to be a valuable diagnostic method for precisely identifying the anatomy of relevant structures. Furthermore, 3D-CISS MR imaging allows judgment of the thickness of the third ventricle floor and display of the ventriculostomy/flow void, which are predictive for intraoperative course and clinical outcome.