3D-visualization of neurovascular conflict in patients with classical trigeminal neuralgia

Abstract

Introduction. MRI is well-recognized as a leading method of neurovascular conflict (NVC) evaluation in patients with classical trigeminal neuralgia (CTN). NVC is regarded as the cause of the development of deformity, atrophy, or dislocation of the trigeminal nerve root. Currently, the number of worldwide publications concerning the analysis of NVC 3D-reconstruction for the diagnosis of cranial nerve hyperfunction syndromes is still small. The aim of the study was to evaluate the efficacy of NVC 3D-visualization for CTN diagnosis compared to conventional examination protocols. Materials and methods. From 2018 to 2020, 30 patients with CTN who underwent 3D-visualization of the NVC as a preoperative examination were involved in prospective nonrandomized study. T-2 3D (FIESTA), 3D-TOF and T1 FRSGR were utilized for 3D rendering (Fusion protocol). The obtained data were compared only FIESTA-based «prediction» of NVC and with intraoperative findings. Results. FIESTA-only prediction was effective in 93 %, and Fusion protocol — in 97 % of arterial type of NVC cases. Fusion protocol showed its supremacy in detecting venous type of compression. Average efficacy of FIESTA-only prediction was 84 %, whereas Fusion protocol showed 97 % correlation with intraoperative findings. Conclusion. Fusion protocol allows fast detection of NVC, its type and localization (not only at the REZ of the fifth nerve but also at its cisternal portion). It makes possible tracing the course of the compressing vessel and assess the «convenience» of decompressing the TN root, taking into account the spatial relationship of the brainstem, vessels of the cerebellopontine angle and cranial nerves. Unlike FIESTA-only scanning, 3D-rendering also offers a possibility of preoperative simulation of microvascular decompression by recreating intraoperative view with precise details. Patient examination with FIESTA-only scanning is inferior to the «extended» algorithm for NVC diagnosis and does not have these capabilities.