The aims of this study were to assess radiofrequency (RF) shielding and susceptibility-induced imaging artifacts of venous stents with different designs at 1.5 T and to analyze the relationship between stent designs, that is, cell geometry and RF shielding.
Twelve dedicated venous stents and 1 stent used for venous pathologies with 8 different designs from 5 different manufacturers were tested: Blueflow (plus medica, Düsseldorf, Germany), Sinus Obliquus, Sinus Venous, Sinus XL (Optimed, Ettlingen, Germany), Vici (Veniti, St. Louis, MO), Zilver Vena (Cook, Bjaeverskov, Denmark), and Venovo (Bard, Tempe, AZ). Two versions with different lengths were available from all stents except the Venovo. For each stent, B1 and frequency mapping was performed using the double angle method and gradient multiecho imaging. Each stent was measured in 3 different orientations: parallel, orthogonal, and at 45 degrees to B0. A correlation analysis was performed between the induced B1 field strength inside the stents and the geometries of the cells.
Radiofrequency shielding was found to be strongly varying between different stent designs. The 120-mm-long Vici stent showed the lowest mean relative B1 amplitude of (38% ± 16%) when oriented parallel to B0. The highest mean B1 amplitude was measured inside the 100-mm-long Blueflow stent with diagonal orientation (90% ± 20%). Averaged over all stents, the shielding was 18% stronger when the stents are oriented orthogonal to B0 compared with a parallel orientation and the between-stent variation was lower for the orthogonal orientation (11%) compared with the parallel orientation (20%). For laser-cut stents, a linear correlation was found between the amount of RF shielding and the length of individual cells measured perpendicular to the stents’ long axes. The woven stents showed a strongly inhomogeneous intraluminal RF shielding pattern, whereas the laser-cut stents provided a more homogeneous shielding pattern. No substantial susceptibility-induced frequency shifts were measured near all stents with a maximum shift of ∆f = 96 Hz measured in the vicinity of the 150-mm-long Sinus Obliquus stent.
Magnetic resonance imaging in the vicinity of commercially available venous stents is feasible at 1.5 T with no substantial susceptibility-induced artifacts but reduced transmit and receive B1 field strengths inside the stents. The strength and homogeneity of the intraluminal B1 depend on the stents’ fabrication (woven or laser-cut) and cell geometry.