Magnetic Resonance Spectroscopy

Chemists have relied on nuclear magnetic resonance (NMR) spectroscopy for 50 years for molecular structure elucidation; magnetic resonance spectroscopy (MRS) is an in vivo extension of NMR. More than for structure determination, however, MRS is applied in medicine to determine the concentrations of a relatively few metabolites that are altered in disease. In MRS, the high-resolution morphologic imaging capabilities of MR are sacrificed to provide metabolic data that, in many cases, precede structural abnormality.3

Proton MRS is the most commonly applied technique for brain tumors because of the high natural abundance of protons in tissue. For brain tumor proton spectroscopy, the metabolites of interest include N-acetylaspartate (NAA), choline (Cho), creatine (Cr), lactate, lipids, and certain amino acids, such as alanine and succinate.37 MRS imaging (MRSI) and chemical shift imaging (CSI) provide phase encoding of spatial information and generate metabolite maps. Multislice MRSI competes with single-voxel MRS, in which a small portion of the lesion is interrogated rather than the whole tumor volume, which is more easily implemented, with brief imaging times (less than 10min/volume element, or voxel) and commercially available software. Nevertheless MRSI, with its smaller voxel size (less than 1 cm3) and superior brain coverage, is necessary for complete characterization of heterogeneous brain tumors.

The key metabolite in brain tumor MRSI is choline (Cho), the underlying causes for the alteration of which remain controversial. The majority of choline in the brain is, in normal figure 27.6. (A) T2-weighted image of glioblastoma multiforme showing necrotic changes with well-defined areas of abnormal signal intensity, compatible with tumor, anterior, medial and posterior to the cystic component as well as in the right centrum semiovale. (B) Corresponding regional cerebral blood volume (rCBV) maps show marked increased values in the posterior component (arrows) and in the wall of the necrotic tumor (small arrowheads), suggestive of a higher-grade component of the tumor.

figure 27.6. (A) T2-weighted image of glioblastoma multiforme showing necrotic changes with well-defined areas of abnormal signal intensity, compatible with tumor, anterior, medial and posterior to the cystic component as well as in the right centrum semiovale. (B) Corresponding regional cerebral blood volume (rCBV) maps show marked increased values in the posterior component (arrows) and in the wall of the necrotic tumor (small arrowheads), suggestive of a higher-grade component of the tumor.

figure 27.7. Patient with right temporal lobe glioblastoma multiforme. Metabolite maps on the right side are notable for increased choline (Cho) in the lesion (circle). Note decreased N-acetylaspar-tate (NAA) peak and marked increased Cho peak in the right temporal lobe (upper spectrum) when compared to the normal contralateral temporal lobe (lower spectrum).

figure 27.7. Patient with right temporal lobe glioblastoma multiforme. Metabolite maps on the right side are notable for increased choline (Cho) in the lesion (circle). Note decreased N-acetylaspar-tate (NAA) peak and marked increased Cho peak in the right temporal lobe (upper spectrum) when compared to the normal contralateral temporal lobe (lower spectrum).

conditions, bound to cell membranes, myelin, and complex brain lipids. In pathologic conditions, Cho is thought to reflect cell membrane, myelin, and lipid turnover, leading to release of MRS-visible Cho.38 Cho is present primarily within glia.39 Because malignant brain tumors are glial neoplasms, it seems reasonable that Cho would be elevated within them, as is usually reported (Figure 27.7). In fact, in one study by Gupta et al., a statistically significant linear correlation between tumor to contralateral normalized Cho signal ratio (nCho) and cell density was found, although nCho did not significantly correlate with proliferative index.40

Low NAA levels in brain tumors are believed to be the result of the lack of neurons in what are essentially glial neoplasms (Figure 27.7). Increased lactate and lipids are found in brain tumors, the former believed to be associated with high tumor glycolytic rates and the latter caused by cellular breakdown and necrosis.3

Originally, MRSI was evaluated in the differentiation of normal from neoplastic tissue and its impact on decision making and surgical planning. Rand et al. found that the prospective accuracy of MRSI in the nonblinded and retrospective accuracy in the blinded discrimination of neoplastic from nonneoplastic disease were 0.96 and 0.83, respectively.41 However, preoperative diagnosis and grading remain the goals of brain tumor MRSI. Despite the stunning results obtained in one study in which Preul et al. accurately graded 90 of 91 brain tumors,42 most studies report such overlap as to make spectroscopy of marginal utility for tumor grading. In a study by Bulakbasi et al., MR spectroscopy could differentiate benign from malignant tumors but was not useful in grading malignant tumors. However, in the same study, ADC values were effective for grading malignant tumors (P less than 0.001) but not for distinguishing different tumor types with the same grade, as previously mentioned. The authors concluded that the two modalities can have a complementary effect in the differentiation and grading of brain tumors.24

Law et al. proved that the combination of rCBV, Cho/Cr, and Cho/NAA resulted in increased specificity for the detection of high-grade gliomas from 57.5% with rCBV alone to 60.0% with both modalities. No increased sensitivity, posi tive predictive value (PPV), or negative predictive value (NPV) was achieved, however, when the MRS findings were combined with perfusion.31 In a series of 176 patients, however, Moller-Hartmann et al. were able to establish that the additive information of proton MRSI led to a 15.4% higher number of correct diagnoses, 6.2% fewer incorrect, and 16% fewer equivocal diagnoses than with structural MRI data alone, in a multitude of pathologies, including brain tumors.43

Perhaps more clinically useful than tumor grading is the role of MRS in the planning of guided biopsies. In 29 patients in whom the preoperative metabolite levels were correlated with the histologic findings, it was found that with abnormally increased Cho and decreased NAA, biopsy invariably was positive for tumor.44 Similar results were reached by Martin et al.45

In the evaluation of tumor borders in 31 patients with diffusely infiltrating gliomas, Croteau et al. tried to determine a correlation between different proton MRS/I metabolic ratios and the degree of tumor infiltration. They correlated the metabolite ratios with the histopathologic analyses of biopsies obtained at the same location and found that the Cho to normal contralateral Cr and Cho to normal contralateral Cho ratios (Cho/nCr and Cho/nCho) were positively correlated with the degree of tumor infiltration, whereas the NAA to normal contralateral Cr ratio (NAA/nCr) was negatively correlated, for all tumor grades combined.46 The strength of this study resides in the coregistration of the biopsy sites with the metabolite values obtained from the same voxels. The potential of MRSI in the accurate delineation of tumor borders requires further investigation.

Another use for MRSI is in the differentiation of primary from metastatic brain tumors. In a recent study, a significant increase in Cho concentration was found in both the peritu-moral and tumoral regions of malignant gliomas, compared with metastases. Similarly, a prominent difference in the Cho/Cr ratio between gliomas and metastases (P less than 0.05) and elevated myo-inositol levels (MI/Cr) within the enhancing foci of gliomas but not in the metastases were also found (P less than 0.05). That study, however, was limited by the small number of patients (22).47

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment