Three-dimensional high-resolution dynamic contrast-enhanced MR angiographyof the pelvis and lower extremities with use of a phased array coil and subtraction: Diagnostic accuracy

Citation
Y. Yamashita et al., Three-dimensional high-resolution dynamic contrast-enhanced MR angiographyof the pelvis and lower extremities with use of a phased array coil and subtraction: Diagnostic accuracy, J MAGN R I, 8(5), 1998, pp. 1066-1072
Citations number
23
Categorie Soggetti
Radiology ,Nuclear Medicine & Imaging
Journal title
JMRI-JOURNAL OF MAGNETIC RESONANCE IMAGING
ISSN journal
10531807 → ACNP
Volume
8
Issue
5
Year of publication
1998
Pages
1066 - 1072
Database
ISI
SICI code
1053-1807(199809/10)8:5<1066:THDCMA>2.0.ZU;2-Q
Abstract
In this study, our purpose was to compare the high-resolution contrast-mate rial-enhanced three-dimensional subtraction MR angiography with conventiona l angiography for occlusive disease in the pelvic and lower extremity arter ies. A three-dimensional fast-imaging with steady precession (FISP) sequenc e with a 256 x 512 matrix was obtained on 1.5T MR unit using a phased array coil. Twenty patients with arteriosclerotic obstructive disease underwent subtraction dynamic contrast-enhanced MR angiography. In 15 patients, three regions (pelvis, upper knee, and lower knee) were sequentially obtained af ter repeated injection of gadolinium-diethylenetriamine pentaacetic acid (G d-DTPA), In the other five patients, one region was imaged (total of 50 exa minations); a maximum-intensity projection (MIP) algorithm was used for sub tracted images. All patients also underwent conventional angiography, Angio graphic images were divided into several anatomical segments. Three blinded radiologists independently graded a total of 50 anatomic segments with ste notic or obstructive diseases and 90 segments without disease. Subtracted i mages allowed resolution of small branch vessels in all examinations, altho ugh misregistration was seen in eight examinations of five patients. AU art eries larger than 1 mm in diameter were visualized on subtracted images. Fo r detection of significant stenosis (>50%), MR angiography had 96% sensitiv ity and 83% specificity. The correlation coefficient of degree of agreement between MR angiography and conventional angiography was .92. Stenotic vess els tended to be overestimated. We conclude that high-resolution dynamic co ntrast-enhanced three-dimensional MR angiography is capable of depicting sm all vessel anatomy of the pelvis and lower extremities. Sequential MR angio graphy of different regions was feasible by repeated injection of Gd-DTPA a nd subtraction, This technique is highly sensitive in detecting lesions, bu t stenosis tended to be overestimated.