High spatial-resolution CE-MRA of the carotid circulation with parallel imaging: comparison of image quality between 2 different acceleration factors at 3.0 Tesla.

PURPOSE

We sought to evaluate and compare the image quality and vessel delineation of the carotid arteries with high spatial-resolution contrast-enhanced MRA (CE-MRA) at 3.0 T using integrated parallel acquisition (iPAT) with acceleration factors of 2 and 4.

MATERIALS AND METHODS

Using an 8-channel neurovascular array coil, we performed prospective high-spatial resolution CE-MRA at 3.0 T of the head and neck on 24 patients (11 men, 13 women, ages 37-89) with suspected arterio-occlusive disease who were assigned randomly to 2 groups. Twelve patients (group A) were examined with a 3D-GRE sequence using iPAT with acceleration factor of 2. For the next 12 patients (group B) a near-identical sequence with an acceleration factor of 4 was applied. Higher iPAT factors were used to increase the spatial-resolution while keeping scan time unchanged. Two volunteers were scanned by both protocols. Phantom measurements were performed to assess the signal-to-noise ratio (SNR). The presence of artifact, noise, image quality of the arterial segments, and the presence and degree of arterial stenosis were evaluated independently by 2 radiologists. Statistical analysis of data was performed by using Wilcoxon rank sum test and 2-sample Student t test (P < 0.05 was indicative a statistically significant difference). The interobserver variability was tested by kappa coefficient.

RESULTS

SNR values were significantly lower when iPAT with acceleration factor of 4 was used (P < 0.001). There was no significant difference between 2 groups in regards to image noise (P = 0.67) and artifact (P = 0.8). Both readers visualized the majority of carotid circulation with good image quality in both groups. For smaller intracranial arteries, such as the second-division of anterior and middle cerebral artery, anterior communicating artery, and superior cerebellar artery, the image quality and vessel delineation was significantly better at an iPAT factor of 4 (P < 0.01). The overall interobserver agreement for both the vessel depiction, and detection of arterial stenoses was higher in group B compared with group A.

CONCLUSION

Use of parallel acquisition techniques with a high acceleration factor (iPAT-4) results in superior depiction of small intracranial arterial segments. Imaging at higher magnetic field strength, in addition to the use of an optimized 8-channel array coil, provides sufficient SNR to support faster parallel acquisition protocols, leading to improved spatial-resolution. More extensive clinical studies are warranted to establish the range of applications and confirm the accuracy of the technique.