Proton magnetic resonance spectroscopy (MRS) of metastatic brain tumors: variations of metabolic profile.
Proton magnetic resonance spectroscopy (MRS) of metastatic brain tumors: variations of metabolic profile.
复制标题
DOI：
10.1007/s10147-006-0589-y
复制
发表时间：
2006-10-01
影响因子：
3.3
通讯作者：
Hori, Tomokatsu
中科院分区：
文献类型：
--
作者：
Chernov, Mikhail F;Hayashi, Motohiro;Hori, Tomokatsu
背景：光谱成像有助于脑实质肿瘤的无创性鉴别。本研究的目的是基于质子磁共振波谱（MRS）的颅内转移瘤的代谢特征。（TR，2000 ms; TE，136 ms）单体素体积选择质子MRS。在83例患者中，MRS体素内的肿瘤分数占50%以上。与正常脑组织相比，肿瘤组织中N-乙酰天冬氨酸含量显著降低，（P < 0.0001），肌酸（P < 0.0001），[NAA]/含胆碱化合物比值（P < 0.0001），[Cho]增加（P < 0.0001）和移动的脂质/[Cr]比值（P < 0.0001）和乳酸/[Cr]比值（P < 0.05），以及[Lip]（P < 0.0001）和[Lac]（P < 0.0001）共振的更频繁存在。然而，当单独分析MRS体素内肿瘤分数小于50%的患者的数据时，这些差异中的大多数丢失。通过对MR波谱上主要代谢物峰的测定[NAA、Cho、Lip]，我们确定了脑转移瘤的三种一般代谢模式，这些模式显示了与肿瘤大小的统计学显著相关性（P < 0.001），增强类型（P < 0.01），病灶周围水肿程度（P < 0.05）。结论：质子MRS可以定义转移性脑肿瘤的代谢上不同的子集，在脑实质病变的鉴别诊断中应考虑这些特征。
BACKGROUND: Spectroscopic imaging can be helpful for the noninvasive identification of parenchymal brain tumors. The objective of the present study was the characterization of the metabolic profile of intracranial metastases, based on proton magnetic resonance spectroscopy (MRS).METHODS: One hundred and four metastatic brain tumors were evaluated by long-echo (TR, 2000 ms; TE, 136 ms) single-voxel volume-selected proton MRS. In 83 patients the tumor fraction within the MRS voxel constituted more than 50%.RESULTS: Compared to normal brain, the tumors showed statistically significant decreases of N-acetylaspartate (P < 0.0001), creatine (P < 0.0001), and the [NAA]/choline-containing compounds ratio (P < 0.0001), increases of [Cho] (P < 0.0001) and the mobile lipids/[Cr] ratio (P < 0.0001) and the lactate/[Cr] ratio (P < 0.05), and the more frequent presence of [Lip] (P < 0.0001) and [Lac] (P < 0.0001) resonances. However, the majority of these differences were lost when data for patients whose tumor fraction within the MRS voxel constituted less than 50% were analyzed separately. Determination of the predominant metabolite peak on the MR spectrum [NAA, Cho, Lip] permitted us to define three general metabolic patterns of brain metastases, which, showed statistically significant associations with the size of the neoplasm (P < 0.001), type of its contrast enhancement (P < 0.01), and the extent of perilesional edema (P < 0.05).CONCLUSION: Proton MRS can define metabolically different subsets of metastatic brain tumors, and these characteristics should be taken into consideration during the differential diagnosis of parenchymal brain lesions.