Tanaka Yoji, Nariai Tadashi, Momose Toshiya, Aoyagi Masaru, Maehara Taketoshi, Tomori Toshiki, Yoshino Yoshikazu, Nagaoka Tsukasa, Ishiwata Kiichi, Ishii Kenji, Ohno Kikuo
Department of Neurosurgery, Graduate School, Tokyo Medical and Dental University, Bunkyo-ku, Japan.
J Neurosurg. 2009 Jan;110(1):163-72. doi: 10.3171/2008.4.17569.
A multimodal neuronavigation system using metabolic images with PET and anatomical images from MR images is described here for glioma surgery. The efficacy of the multimodal neuronavigation system was evaluated by comparing the results with that of the conventional navigation system, which routinely uses anatomical images from MR and CT imaging as guides.
Thirty-three patients with cerebral glioma underwent 36 operations with the aid of either a multimodal or conventional navigation system. All of the patients were preliminarily examined using PET with l-methyl-[11C] methionine (MET) for surgical planning. Seventeen of the operations were performed with the multimodal navigation system by integrating the MET-PET images with anatomical MR images. The other 19 operations were performed using a conventional navigation system based solely on MR imaging.
The multimodal navigation system proved to be more useful than the conventional navigation system in determining the area to be resected by providing a clearer tumor boundary, especially in cases of recurrent tumor that had lost a normal gyral pattern. The multimodal navigation system was therefore more effective than the conventional navigation system in decreasing the mass of the tumor remnant in the resectable portion. A multivariate regression analysis revealed that the multimodal navigation system-guided surgery benefited patient survival significantly more than the conventional navigation-guided surgery (p = 0.016, odds ratio 0.52 [95% confidence interval 0.29-0.88]).
The authors' preliminary intrainstitutional comparison between the 2 navigation systems suggested the possible premise of multimodal navigation. The multimodal navigation system using MET-PET fusion imaging is an interesting technique that may prove to be valuable in the future.
本文介绍一种用于胶质瘤手术的多模态神经导航系统，该系统结合了PET代谢图像和MR图像的解剖图像。通过将多模态神经导航系统的结果与传统导航系统（常规使用MR和CT成像的解剖图像作为引导）的结果进行比较，评估多模态神经导航系统的疗效。
33例脑胶质瘤患者借助多模态或传统导航系统进行了36次手术。所有患者术前均使用L-甲基-[11C]蛋氨酸（MET）PET进行手术规划。其中17例手术使用多模态导航系统，将MET-PET图像与解剖MR图像整合。另外19例手术使用仅基于MR成像的传统导航系统。
多模态导航系统在确定切除区域方面比传统导航系统更有用，它能提供更清晰的肿瘤边界，尤其是在失去正常脑回模式的复发性肿瘤病例中。因此，多模态导航系统在减少可切除部分肿瘤残余量方面比传统导航系统更有效。多因素回归分析显示，多模态导航系统引导的手术比传统导航系统引导的手术对患者生存的益处显著更大（p = 0.016，优势比0.52 [95%置信区间0.29 - 0.88]）。
作者在机构内部对这两种导航系统进行的初步比较表明了多模态导航的可能前提。使用MET-PET融合成像的多模态导航系统是一种有趣的技术，未来可能被证明具有价值。
