The pointwise encoding time reduction with radial acquisition (PETRA) is the ultrashort TE sequence (1), which provides good image quality with less susceptibility artifacts and less acoustic noise (2).PETRA-MRA, which is the subtraction image created from PETRA with/without saturation slab, showed promising results without flow dephasing artifacts (3). There are few reports that intracranial TOF-MRA and PETRA-MRA were compared (3), therefore, we evaluated the differences of the visualization of intracranial arteries between TOF-MRA and PETRA-MRA. We also hypothesized that the facial nerve canal would be visualized better in PETRA than the other 3D sequences due to less susceptibility artifact from skull and mastoid air cell.

Subjects

Fifteen volunteers (mean age 41, range 31-57 years) were enrolled in this under approval of institutional review board with written informed consent. Scan was conducted using a 3T-MR system (Magnetom Skyra, Siemens, Erlangen, Germany) with a 32-channel head coil for 3D TOF-MRA (TR/TE 20/3.69 ms, FA18°, resolution 0.47 × 0.82 mm, 0.7 mm-thickness, matrix 384 × 267), PETRA (axial, TR/TE 4/0.07 ms, FA4°, resolution 0.67 × 0.67 mm, 0.67 mm-thickness, matrix 320 × 320), T2-SPACE (sagittal, TR/TE 1800/273 ms, FA120°, resolution 0.65 × 0.98 mm, 0.8 mm-thickness, matrix 320 × 260) , and MPRAGE (sagittal, TR/TE 1900/2.58 ms, TI 900ms, FA9°, resolution 0.9 × 0.9 mm, 0.9 mm-thickness, matrix 256 × 256).

MR angiography

In PETRA-MRA, subtraction images are created from PETRA with/without saturation slab at the neck. In the evaluation of angiography, the MIP images of the TOF-MRA and PETRA-MRA were reviewed (Figure1), and the visualization of middle cerebral artery (M1, M2, M3, M4), anterior cerebral artery (A1, A2, A3), posterior cerebral artery (P1, P2, P3) and internal carotid artery (C) was subjectively evaluated on a 4-point scale as follows: 1=not visible (almost no signal), 2=poor (structures are discontinuous), 3=good (structures are continuous, but blurring), or 4=excellent (structures are continuous and sharply defined). In addition, homogeneity of the signals in C was evaluated on a 3-point scale as follows: 1=poor, 2=good, 3=excellent. The scores of M1, M2, M3, M4, A1, A2, A3, P1, P2, P3, P4, C and homogeneity were compared using Wilcoxon test and the total scores of all vessels were compared by using a paired t test.

The facial nerve canal

The image volumes of PETRA, T2-SPACE and MPRAGE were reconstructed in axial image of 0.9 mm thickness by using syngo FUSION, and the visualization of the facial nerve canal was evaluated (Figure2). The facial nerve canals were divided into the four segments (labyrinthine, geniculate ganglion, tympanic, and mastoid). For each of the left and right facial nerve canal, a three-point scale was used: 0 = not visible, 1= partially visible, or 2= excellently visible. The scores were compared among PETRA, T2-SPACE and MPRAGE using Friedman test. A P value less than 0.05 was considered statistically significant.

MR angiography

In the comparison of the MRAs, the scores of A3 and P3 and total scores of TOF-MRA were statistically better than PETRA-MRA (P=0.0049, 0.0425, 0.0040), and others didn’t have statistically differences (Figure3). PETRA-MRA showed significantly higher homogeneity of the signal at C (P=0.0010).

The facial nerve canal

In the comparison of the facial nerve canals, at labyrinthine, the scores of PETRA were best and there were significant differences between MPRAGE and PETRA. At geniculate ganglion and tympanic and the sum of four segments, the scores of PETRA and T2-SPACE were statistically better than MPRAGE. At mastoid, the scores of T2-SPACE were best and there were significant differences between MPRAGE and T2-SPACE (Figure4).

PETRA-MRA was less visualized at the peripheral artery (A3, P3), but PETRA-MRA visualized intracranial arteries as good as TOF-MRA at the other parts. In the visualization of facial nerve canal, PETRA was better than MPRAGE at all segments and best at labyrinthine among all 3D sequences.PETRA and PETRA-MRA were useful for the visualization of intracranial artery and facial nerve canal.