Abstract
Sliding-window reconstruction (SWR) has been recently introduced for rapid imaging that improves temporal resolution while maintaining signal-to-noise ratio (SNR) and spatial resolution.
We assessed the quantity of 2-dimensional contrast-enhanced magnetic resonance digital subtraction angiography (2D CE-MRDSA) with non-Cartesian radial SWR in phantom and clinical studies. In phantoms, we compared the quantitative properties of time-intensity curves (TIC) obtained with dynamic 2D CE-MRDSA using SWR in a radial acquisition with those obtained by Cartesian acquisition. We calculated the mean variance and standard deviation among signal intensities in TICs and used SWR to study 2D CE-MRDSA in 5 patients with angiographically proven arteriovenous malformations. Using a 3-point grading scale, we individually scored vascular visualization capability and calculated time delay (TD) from the TIC in the feeding artery (FA), nidus, and draining vein (DV).
The maximum signal intensity variance in Cartesian SWR was 6.58±2.27% among time-intensity curves and was 0.87±0.77% radial SWR. Signal intensity in radial SWR decreased significantly (P<0.001) compared with the Cartesian SWR. In clinical study, the mean rating on 2D CE-MRA of the feeding artery was 2.3, nidus, 2.6, and draining vein, 2.6. Mean delay time between DV and FA was 1.8 s.
The radial SWR technique is useful for demonstrating the hemodynamic features of vascular malformations in the head with 2D CE-MRDSA.
