Ohashi Toshio, Momma Tetsuo, Yamashita Shoji, Nagatsuma Katsumi, Kanai Kunimitsu, Kitagawa Kazuhito, Takahashi Shinichi, Hanada Takashi, Yorozu Atsunori, Shigematsu Naoyuki
Department of Radiology, Keio University School of Medicine, Tokyo, Japan.
Brachytherapy. 2012 Nov-Dec;11(6):468-75. doi: 10.1016/j.brachy.2011.12.010. Epub 2012 Feb 11.
To compare contrast-enhanced T1-weighted (CE-T1WI) magnetic resonance imaging (MRI) with computed tomography (CT) for postimplant dosimetry and seed recognition in prostate brachytherapy.
A total of 245 patients who received (125)I prostate brachytherapy with or without external beam radiotherapy were enrolled. For postimplant analysis, CT and MRI scans were obtained at 1 month after seed implantation. For MRI-based dosimetry, T2-weighted images were fused with the CE-T1WI; the prostate was delineated on the T2-weighted images, and the seed detection was performed manually on the CE-T1WI. In CT-based dosimetry, the seed detection was essentially performed automatically. The dosimetric results obtained by MRI-based and CT-based dosimetry were compared.
The mean prostate D(90) (the minimum dose received by 90% of the prostate volume) estimated by MRI-based and CT-based dosimetry were 113% and 115%, respectively, with no significant difference. The mean prostate V(100) (the percent volume of the postimplant prostate receiving 100% of the prescribed dose) estimated by MRI-based and CT-based dosimetry were 95.2% and 95.8%, respectively, again with no significant difference. The mean prostate V(150) (the percent volume of the postimplant prostate receiving 150% of the prescribed dose) estimated by MRI-based and CT-based dosimetry were 52.8% and 57.0%, respectively (p<0.01). In all of the 35 patients (14%) in whom the MRI-based V(150) were at least 10% lower than the CT-based results, the seed detection by CT-based dosimetry was overestimated in highly seed-clustered areas or in the areas close to calcifications because of reconstruction artifacts in CT images.
MRI-based dosimetry using CE-T1WI appears to be acceptable. Our results suggest that MRI-based dosimetry is a practical method for estimation of the higher dose distribution, especially if seeds are clustered together or when they are close to calcifications.
比较对比增强T1加权（CE-T1WI）磁共振成像（MRI）与计算机断层扫描（CT）在前列腺近距离放射治疗中植入后剂量测定及籽源识别方面的应用。
共纳入245例接受（125）I前列腺近距离放射治疗（无论是否联合外照射放疗）的患者。对于植入后分析，在籽源植入后1个月进行CT和MRI扫描。对于基于MRI的剂量测定，将T2加权图像与CE-T1WI融合；在T2加权图像上勾画前列腺，在CE-T1WI上手动进行籽源检测。在基于CT的剂量测定中，籽源检测基本自动进行。比较基于MRI和基于CT的剂量测定所获得的剂量学结果。
基于MRI和基于CT的剂量测定所估计的平均前列腺D（90）（前列腺体积的90%所接受的最小剂量）分别为113%和115%，无显著差异。基于MRI和基于CT的剂量测定所估计的平均前列腺V（100）（植入后前列腺接受100%处方剂量的体积百分比）分别为95.2%和95.8%，同样无显著差异。基于MRI和基于CT的剂量测定所估计的平均前列腺V（150）（植入后前列腺接受150%处方剂量的体积百分比）分别为52.8%和57.0%（p<0.01）。在所有35例（14%）基于MRI的V（150）比基于CT的结果至少低10%的患者中，由于CT图像中的重建伪影，基于CT的剂量测定在籽源高度聚集区域或靠近钙化的区域对籽源的检测被高估。
使用CE-T1WI的基于MRI的剂量测定似乎是可接受的。我们的结果表明，基于MRI的剂量测定是估计较高剂量分布的一种实用方法，特别是当籽源聚集在一起或靠近钙化时。
