Identification and quantitative evaluation of symptomatic cerebral venous thrombosis using 3D variable flip angle turbo spin echo

Abstract

Purpose: To explore the feasibility of MR 3D T1w Sampling Perfection with Application optimized Contrasts by using different flip angle Evolutions (SPACE) sequence imaging in symptomatic CVT diagnose, extracting the imaging features with quantitative analysis.

Methods: Fifty-nine patients with suspected CVT with neurological symptoms were retrospectively included in this study. Of them, 35 patients were enrolled in the comparation of diagnostic accuracy between the contrast-enhanced magnetic resonance venograms (CE-MRV) and 3D T1w SPACE imaging. Forty-five patients with 101 involved segments were identified for the quantitative analysis. All MR images were acquired on a 3.0 T MR scanner. The reference standard used in this study was a comprehensive combination of the imaging techniques and clinical information. CVT patients were grouped as acute (≤48 h), subacute (>48 h and ≤30d), and chronic (>30d) clinical phase. CVT segments were grouped based on pre-contrast T1WI, as type A: hypo intense signal; B: heterogeneously hyper intense signal; C: iso intense signal. The feasibility of 3D T1w SPACE imaging for diagnosing CVT was explored. Diagnostic accuracy of T1w SPACE imaging was analyzed and compared with the CE-MRV. The signal intensity of pre-contrast images (SpreCE), signal intensity of post-contrast images (SpostCE), and contrast enhancement (CE) rate, CE rate relative to that of pituitary gland (PG), white matter (WM), gray matter (GM), and normal vein vessel wall (nVVW) were compared based on both patients and segments. The CE rate grade of CVT segments of different imaging types was compared.

Results: The MR 3D T1w SPACE imaging achieved a higher sensitivity and specificity (100%/94.1% and 100%/100% based on patients/segments separately) than that of the CE-MRV (73.9%/56.9% and 83.3%/98.9% based on patients/segments separately). No statistical correlation was found between the imaging types of CVT segments and onset time of clinical symptoms (χ² = 6.649, P = 0.171). Quantitative analysis showed that the CE rate relative to PG and that to WM were higher in the chronic CVT patients than that in the other two groups (H = 10.330 and P = 0.006, H = 9.898 and P = 0.007, separately). CE rate relative to GM in the chronic group was higher than that in the subacute group (H = 7.143 and P = 0.028). All of the quantitative parameters were statistically different across CVT segments of three imaging types (all P≤0.001).

Conclusion: MR 3D T1w SPACE imaging has the advantage to accurately diagnose CVT of different clinical stages, and identify the involved thrombus segments.

Keywords: Cerebral veins; Magnetic resonance imaging; Sinus thrombosis.