Gadofosveset: parameter optimization for steady-state imaging of the thoracic and abdominal vasculature.

PURPOSE

Comparison of 3 optimized pulse sequences for thoracoabdominal contrast-enhanced magnetic-resonance angiography by signal-to-noise measurements and time-dependent T1 mapping in the steady state after injection of 0.03 mmol/kg BW gadofosveset.

MATERIALS AND METHODS

After institutional review board approval, 15 healthy volunteers (19-46 years, mean age: 31.5 years) were included in this prospective, intraindividual comparison study. All examinations were performed at 1.5 T. Three pulse sequences: volume interpolated breath-hold examination (VIBE) sequences as VIBESEMI (echo time [TE]: 1.64 milliseconds, repetition time [TR]: 3.77 milliseconds, FA: 15 degrees, voxel size: 1.2 × 1.2 × 1.2 mm) with short TR, VIBEOPT (TE: 2.2 milliseconds, TR: 5.2, FA: 15 degree, voxel size: 1.2 × 1.2 × 1.2 mm) with long TR, and a typical 3-dimensional fast low angle shot (FLASH) sequence (TE: 1.39 milliseconds, TR: 3.77 milliseconds, FA: 25 degree, voxel size: 1.0 × 0.8 × 1.0 mm) were repeated 10, 20, 30, and 40 minutes after the injection of 0.03 mmol/kg BW gadofosveset (mean dose: 9.7 mL). Mean signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were computed for the aorta and the inferior vena cava (IVC). Three-dimensional gradient echo sequences with variable flip angles were performed for T1 mapping 0 to 50 minutes postinjection (p.i.). Additional phantom measurements were performed to compare the sequences.

RESULTS

Significantly higher SNR values of the FLASH were found at every point compared with VIBEOPT (P = 0.002-P = 0.004), but only 10, respectively, 20 minutes p.i. to VIBESEMI. No significant differences of SNR were obtained between VIBESEMI and VIBEOPT. In the aorta, the maximal percentage gain of SNR was 29.2% for 3D-FLASH compared with VIBESEMI. Similar, but mostly not significant, results were obtained regarding the SNR in the IVC with the 3D-FLASH sequence yielding higher SNR versus both comparators (P = 0.007-P = 0.466). Except 10 minutes p.i., CNR analysis yielded higher values for the VIBESEMI versus both comparators in the aorta as well as in the IVC. No statistical significant difference was found for the VIBESEMI versus the 3D-FLASH sequence in all comparisons. Regarding the phantom measurements, statistically significant higher SNR was found for the VIBESEMI versus the 3D-FLASH. The T1 time in the aorta decreased p.i. from 1227 ± 383 milliseconds to 141 ± 27 milliseconds and showed over the time a slow reincrease to 175 ± 29 milliseconds at 50 minutes p.i.

CONCLUSION

Ten to 30 minutes after injection of gadofosveset, a relatively constant longitudinal relaxation is given. In this steady state, no additional improvements were obtained by theoretically optimized sequence parameters in the VIBEOPT with a longer TR.