In proton MR spectroscopy (MRS) of the brain, the application of long echo times (TEs) (for example, 135 ms) allows an easy spectral quantitation of the few visible major metabolites. In short-TE spectra, several additional metabolites are visible that may provide useful information. Their spectral evaluation, however, is complicated by the complex spectral pattern of the metabolites and the severe overlap of resonances. The use of prior knowledge can greatly facilitate the evaluation of short-TE brain spectra. Some advanced fitting procedures include prior knowledge of the spectral pattern of single metabolites for the evaluation of in vivo spectra. This prior knowledge can be obtained from simulations of model spectra (1), or by measurement of model spectra from aqueous solutions, whereas the obtained model spectra can be either parameterized (2-4) or directly used as prior knowledge (5).Besides the metabolites, broad resonances from macromolecules contribute to short-TE brain spectra. These resonances have shorter T 1 relaxation times than metabolites, and therefore can be separated by the use of a preceding inversion pulse for metabolite suppression (6,7), as shown in Fig.

In vivo longitudinal relaxation times of N-acetyl compounds (NA), choline-containing substances (Cho), creatine (Cr), myoinositol (mI), and tissue water were measured at 1.5 and 3 T using a point-resolved spectroscopy (PRESS) sequence with short echo time (TE). T 1 values were determined in six different brain regions: the occipital gray matter (GM), occipital white matter (WM), motor cortex, frontoparietal WM, thalamus, and cerebellum. The T 1 relaxation times of water protons were 26 -38% longer at 3 T than at 1.5 T. Significantly longer metabolite In vivo 1 H magnetic resonance spectroscopy (MRS) offers researchers the opportunity to investigate changes in metabolite composition that occur in a variety of brain diseases. Because of time constraints, spectroscopic measurements in clinical studies are often performed at repetition times (TRs) that do not allow full relaxation of the metabolite MRS signals. Therefore, it is necessary to obtain accurate values of metabolite T 1 relaxation times in order to perform an adequate correction of the resulting T 1 losses in absolute quantification. Since the relaxation times of metabolite and water protons are greatly influenced by the microenvironment, and reflect mobility at the molecular level, they can also be used to gain a better understanding of the molecular organization of brain structure.T 1 relaxation times of metabolites have been determined for different regions of the human brain in several studies at lower field strengths, such as 1.5 T (1-8) and 2.0 T (9); however, longitudinal relaxation times at 3 T have been ascertained only for occipital gray matter (GM) and occipital white matter (WM) (10). The published reference values for 1.5 T and 3 T are presented in Table 1. Assuming their behavior is analogous to that of aqueous protons in vivo, the longitudinal relaxation times of brain metabolites are expected to increase with the field strength (11). Surprisingly, the T 1 values measured in occipital regions at 3 T (10) were found to be similar to published reference values for 1.5 T. However, it is difficult to compare data from different studies because of possible differences in scanner performance, sequence design, and strategies for acquiring and evaluating the spectra. To date, no published studies on metabolite T 1 relaxation times have included measurements at two different field strengths to provide a direct comparison of the results.The aim of this study was to determine the longitudinal relaxation times of N-acetyl compounds (NA ϭ N-acetylaspartate ϩ N-acetylaspartylglutamate), creatine (Cr), choline-containing substances (Cho), myo-inositol (mI), and water in six different regions of the brain at both 1.5 T and 3 T, and to evaluate whether the amount of GM and WM within the examined voxel influences metabolite T 1 relaxation. Furthermore, T 1 relaxation times of N-acetyl aspartate (NAA), glycerophosphocholine (GPC), phosphocholine (PCh), Cr, and mI were measured in vitro to determine whether the dependence of metabolite T 1 values on field stre...

After successful liver transplantation, renormalization of HE-specific brain metabolite changes is detected at (1)H spectroscopy and precedes the disappearance of BGH. The neuropsychologic signs of subclinical or overt HE follow the changes seen at (1)H spectroscopy rather than those seen at MR imaging.

Purpose: To quantify the macromolecular content in different anatomic brain regions and to evaluate an age dependency of the macromolecular concentrations. Material and Methods:A short echo time Stimulated Echo Acquisition Mode (STEAM) sequence was used without and with inversion recovery metabolite nulling in 8 -12 healthy volunteers. Quantitation was achieved by an extended LCModel, and macromolecular resonances at 0.9, 1.4, 2.1, and 3.0 ppm were evaluated. Results:In the cerebellum, the 1.4, 2.1, and 3.0 ppm resonances were highest compared to all other regions (P Ͻ 0.02); the 0.9 ppm resonance was significantly higher than that of pons (P Ͻ 0.01). In the motor cortex, the 0.9, 1.4, and 2.1 ppm resonances were higher than those of white matter and pons (P Ͻ 0.02). Pons and white matter did not differ significantly from each other. A significant correlation of the macromolecular concentrations with the age could not be found. Conclusion:There were higher macromolecular concentrations in the cerebellum and motor cortex than in pons or white matter. These were probably due to the higher portions of gray matter in these volumes of interest (VOIs) than in the other regions. SHORT ECHO TIME PROTON MAGNETIC resonance spectroscopy (MRS) of the brain allows for the detection of so-called macromolecules with a molecular weight of above 3500 Da, forming a broad substructure below the signal from metabolites with lower molecular weights (1-6). Several approaches have been made to evaluate the macromolecular contributions in the baseline of the spectra to achieve a reliable quantitation of the metabolites (6 -10). Other studies pointed out a potential pathologic relevance of increased macromolecular resonances, e.g., in stroke, brain tumors or multiple sclerosis (3,(11)(12)(13). Although a correlation of metabolite concentrations with anatomic regions is well known (14 -19), there is only one work concerning regional alterations of the macromolecules with respect to supratentorial gray and white matter (6). More knowledge about variations of the macromolecules within different supra-and infratentorial anatomic regions of the brain is essential to avoid misinterpretations of macromolecular signals in healthy and pathological conditions. This study was focused on the macromolecular content of different anatomic brain regions, such as the cerebellum, motor cortex, pons, and parietal white matter in healthy subjects. Correlations between the macromolecular concentration and age were analyzed and compared with the work of Hofmann et al (6). The metabolite concentrations obtained from the different regions were compared to the literature. Possible factors influencing the measurements of the macromolecules such as relaxation effects and partial volume problems were considered. To evaluate the quality of the quantification method, the results for the macromolecular concentrations of the metabolite-nulled and not-nulled spectra were compared. MATERIALS AND METHODS SubjectsTwelve healthy male volunteers, aged 28 -62 years (median a...

No abstract