Phase-contrast MRI and 3D-CISS versus contrast-enhanced MR cisternography for the detection of spontaneous third ventriculostomy.

ABSTRACT

PURPOSE: To compare the diagnostic efficacies of phase-contrast MRI (PC-MRI) and three-dimensional constructive interference in steady-state (3D-CISS) sequence for the detection of spontaneous third ventriculostomy (STV) on the basis of contrast-enhanced MR cisternography (MRC). PATIENTS AND METHODS: Eleven obstructive hydrocephalus patients with clinically-radiologically suspected STV and ten controls were examined by PC-MRI, 3D-CISS and MRC. PC-MRI and 3D-CISS sequence were applied to view the third ventricle and basal cisterns. Following injection of 0.5-1ml intrathecal Gd-DTPA injection, postcontrast MRC images were obtained in three planes. Presence of STV was scored as follows grade 0, no existence of STV; grade 1, STV present. Results of PC-MRI and 3D-CISS were compared with the MRC findings. RESULTS: In PC-MRI, five patients were assessed as grade 0 and six cases grade 1. As a result of 3D-CISS sequence, eight cases were evaluated as grade 0 and three cases grade 1. Based on MRC, nine cases were assessed as grade 0 and two cases grade 1. False positivity was found in four cases by PC-MRI and in one case by 3D-CISS. The sensitivity, specificity and accuracy of PC-MRI and 3D-CISS sequence regarding demonstration of STV, were 100, 100, 56, 89, 64, and 91% respectively. DISCUSSION: PC-MRI and 3D-CISS are helpful in confirming the STV. PC-MRI and 3D-CISS should be the first preference. If PC-MRI and 3D-CISS give negative results, then MRC is not required. MRC should be performed on patients who demonstrate suspected STV findings on PC-MRI and 3D-CISS sequences. MRC may prevent false positive results.