Evaluation of focal cerebral perfusion using arterial spin labeling (ASL) technique.

点击翻译按钮获取中文摘要

During the first year, imaging parameters of the ASL technique affecting the image qualities were evaluated by using a single-slice 2D ASTAR technique in order to optimize the sequence. Because multi-slice imaging is essential for wide clinical applications of ASL techniques, we have evaluated the multi-slice 3D ASTAR technique after the second year. In multi-slice techniques, the influence of magnetization transfer contrast originating from a tagging pulse can't be totally collected with a control pulse. In this study, quality of the 3D ASTAR image was improved by using multiple non-slice selective inversion pulses which suppress the signal intensity of stationary tissues. Signal-to-noise ratio (SNR) of ASTAR images was significantly affected by arterial transit time (ATT), and image contrast changed according to the variation of TI. In the evaluation of the single-slice technique in young normal volunteers, SNR in the posterior watershed areas was lower than other areas probably beca … More use of the longer ATT in these areas. To achieve a proper perfusion map, TI long enough for areas with a long ATT should be selected (TI of 1000 to 1200 ms in normal young subjects). In multi-slice techniques, gap between the tag pulses and imaging slices was larger in more cephalad slices. So, influence of the ATT is more significant in multi-slice techniques than in single-slice ones, especially in cephalad slices. In normal volunteers, TI of longer than 1400 ms was needed to achieve proper signal in the most cephalad slice. In elder subjects, ASL signal of the brain parenchyma tended to be lower and contamination of intravascular signal was more pronounced, compared with younger subjects. These findings suggested the longer ATT in elder subjects due to decreased arterial flow rate. In patients with steno-occlusive vascular disease, elongation of ATT resulted in two errors ; 1) underestimation of focal CBF, and 2) overestimation of focal CBF due to intravascular signal. Influence of ATT was more pronounced in cephalad slices in the 3D ASTAR technique. Though longer TI is necessary to reduce the influence of ATT, longer TI resulted in decreased signal secondary to the longitudinal relaxation and resulted in longer examination time. It seemed to be necessary to decide the proper TI according to the predicted ATT. It, however, is difficult to predict ATT of each area in each subject. And elongation of TI was limited by the longitudinal relaxation and examination time. So, proper evaluation of focal CBF using the ASL technique could be difficult in patients with extremely long ATT, such as patients with progressed moyamoya disease. A saturation pulse could be applied in the same position of a tagging pulse to suppress intravascular signal. However, suppression of signal obtained with this technique was not selective for intravascular signal and could result in lower SNR. Less