Skip to main navigation Skip to search Skip to main content

Diagnostic accuracy of 3D contrast-enhanced T1 high resolution isotropic volume excitation MRI for non-cavernous intracranial dural arteriovenous fistula

  • Ramathibodi Hospital

Research output: Contribution to journal › Article › peer-review

Abstract

Objective: To determine the accuracy of 3D contrast-enhanced THRIVE MRI in diagnosis of non-cavernous intracranial dural AVF compared with DSA. Materials and Methods: Thirty-three patients including fourteen dural AVF cases and nineteen control subjects, were included in the present study. They underwent 3D contrast-enhanced THRIVE MRI by 3T, contrast-enhanced MRA, and DSA. Two independent readers reviewed 3D contrast-enhanced THRIVE images for the presence of transosseous arterial feeders, low signal intensity curvilinear structures in the dural venous sinus, shaggy dural sinus, and cortical venous dilatation. Diagnostic performance values were calculated for 3D contrast-enhanced THRIVE MRI. Results: The 3D contrast-enhanced THRIVE MRI identified 23 from 29 locations of dural AVFs. The overall accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were 91.4%, 79.3%, 93.5%, 67.6% and 96.3%, respectively. Shaggy dural sinus and dilated cortical veins were found in 78.6% and 85.7%, respectively. Conclusion: The 3D contrast-enhanced THRIVE MRI is a valuable tool for diagnosis of dural AVF as well as exclusion of dural AVF due to high sensitivity and specificity.

Original languageEnglish
Pages (from-to)1769-1776
Number of pages8
JournalJournal of the Medical Association of Thailand
Volume104
Issue number11
DOIs
Publication statusPublished - Nov 2021

Keywords

  • 3D contrast-enhanced THRIVE MRI
  • Dural AVF
  • Dural arteriovenous fistula

Fingerprint

Dive into the research topics of 'Diagnostic accuracy of 3D contrast-enhanced T1 high resolution isotropic volume excitation MRI for non-cavernous intracranial dural arteriovenous fistula'. Together they form a unique fingerprint.

Cite this