Factors influencing flow-induced signal loss in MR angiography: an in vitro study.

Abstract

Signal loss due to poststenotic turbulence is one remaining limitation in the clinical use of MRA. The objective of this study was to determine the factors influencing poststenotic signal loss (PSL) in a stenotic tube system. With use of a two-dimensional gradient echo sequence (FLASH), the influence of (a) the degree of stenosis (50, 75, and 91% cross-sectional area reduction), (b) flow velocity (12, 47, 71, and 94 cm/s), (c) TE (3, 6, 13, and 20 ms), and (d) flip angle (5, 10, 20, ... , 90 degrees) on the length of PSL was measured in a nonpulsatile stenotic tube system. To quantify the effect of first-order gradient motion refocusing (GMR), the signal intensity ratio from flow-compensated and -uncompensated images was calculated. For statistical analysis, multiple regression analysis and unpaired Student t test were used. In the flow model used, the length of PSL increased significantly with the degree of stenosis (beta ST' = 0.483 mm/%, beta ST" = 0.447 mm/%, p < 0.0001), flow velocity (beta v' = 0.297 s, beta v" = 0.213 s, p < 0.0001), and TE (beta TE = 1.88 mm/ms, p < 0.0001), respectively, whereas no correlation emerged for varying flip angles (beta alpha = 0.0214, p = 0.29). First-order GMR reduced significantly PSL in through-plane measurements (p < 0.0001). Maximal signal-enhancing effects of first-order GMR were observed 1-3 cm distal to the stenosis. Flow model parameters (i.e., degree of stenosis, flow velocity) markedly influenced the length of PSL that could be compensate for by use orf shortened TEs and first-order GMR. Varying flip angles had no significant influence on PSL.