In vivo concentrations of cerebral metabolites were obtained by means of 52 single-voxel, localized proton magnetic resonance (MR) spectroscopic examinations of different regions of the brain performed in 26 healthy adults aged 21-32 years. The study was performed at 2.0 T with use of a circularly polarized head coil to ensure homogeneous radio-frequency excitation and signal reception. Proton MR spectra were obtained in the stimulated-echo acquisition mode under fully relaxed conditions (repetition time > or = 6,000 msec) and at short echo times (20 msec) to minimize corrections due to T1 and T2 attenuation and depict the spectra of metabolites with strongly coupled resonances. Absolute concentrations were obtained by means of calibration of resonance signal areas with those of pertinent metabolite solutions from separate studies and correction for coil loading and partial volume effects (eg, with perfused capillary networks and cerebrospinal fluid). The results provide a quantitative basis for studies of both normal human neurochemistry in vivo and metabolic alterations in diseases of the brain.

Water-suppressed localized proton NMR spectroscopy using stimulated echoes has been successfully applied to detect metabolites in the human brain in vivo. The STEAM spectroscopy sequence allows single-step localization by exciting three intersecting slices. Water suppression is achieved by preceding chemical-shift-selective (CHESS) rf pulses. High-resolution (0.05 ppm) proton NMR spectra of healthy volunteers have been High-resolution (0.05 ppm) proton NMR spectra of healthy volunteers have been obtained on a conventional 1.5-T whole-body MRI system (Siemens Magnetom). Volumes-of-interest (VOI) of 64 ml (4 x 4 x 4 cm3) were localized in the occipital area of the brain and spectra were recorded within measuring times ranging from 1 s (single scan) to about 10 min. The experimental procedure is described in detail. Resonance assignments include acetate, N-acetyl aspartate, gamma-amino butyrate, glutamine, glutamate, aspartate, creatine and phosphocreatine, choline-containing compounds, taurine, and inositols. Cerebral lactate was found to be at a maximum concentration of 0.5 mM when assuming N-acetyl aspartate in white matter to be 6 mM.

Thermotherapy using magnetic nanoparticles is a new technique for interstitial hyperthermia and thermoablation based on magnetic field-induced excitation of biocompatible superparamagnetic nanoparticles. To evaluate the potential of this technique for minimally invasive treatment, we carried out a systematic analysis of its effects on experimental glioblastoma multiforme in a rat tumor model. Tumors were induced by implantation of RG-2-cells into the brains of 120 male Fisher rats. Animals were randomly allocated to 10 groups of 12 rats each, including controls. Animals received two thermotherapy treatments following a single intratumoral injection of two different magnetic fluids (dextran- or aminosilane-coated iron-oxide nanoparticles). Treatment was carried out on days four and six after tumor induction using an alternating magnetic field applicator system operating at a frequency of 100 kHz and variable field strength of 0-18 kA/m. The effectiveness of treatment was determined by the survival time of the animals and histopathological examinations of the brain and the tumor.Thermotherapy with aminosilane-coated nanoparticles led up to 4.5-fold prolongation of survival over controls, while the dextran-coated particles did not indicate any advantage. Intratumoral deposition of the aminosilane-coated particles was found to be stable, allowing for serial thermotherapy treatments without repeated injection. Histological and immunohistochemical examinations after treatment revealed large necrotic areas close to particle deposits, a decreased proliferation rate and a reactive astrogliosis adjacent to the tumor.Thus, localized interstitial thermotherapy with magnetic nanoparticles has an antitumoral effect on malignant brain tumors. This method is suitable for clinical use and may be a novel strategy for treating malignant glioma, which cannot be treated successfully today. The optimal treatment schedules and potential combinations with other therapies need to be defined in further studies.

High-resolution proton NMR spectra of normal human brain in vivo have been obtained from selected 27- and 64-ml volumes-of-interest (VOI) localized in the insular area, the occipital area, the thalamus, and the cerebellum of normal volunteers. Localization was achieved by stimulated echo (STEAM) sequences using a conventional 1.5-T whole-body MRI system (Siemens Magnetom). The proton NMR spectra show resonances from lipids, lactate, acetate, N-acetylaspartate (NAA), gamma-aminobutyrate, glutamine, glutamate, aspartate, creatine and phosphocreatine, choline-containing compounds, taurine, and inositols. While T1 relaxation times of most of these metabolites were about 1100-1700 ms without significant regional differences, their T2 relaxation times varied between 100 and 500 ms. The longest T2 values of about (500 +/- 50) ms were observed for the methyl protons of NAA in the white matter of the occipital lobe compared to (320 +/- 30) ms in the other parts of the brain. No significant regional T2 differences were found for choline and creatine methyl resonances. The relative concentrations of NAA in gray and white matter were found to be 35% higher than those in the thalamus and cerebellum. Assuming a concentration of 10 mM for total creatine the resulting NAA concentrations of 13-18 mM are by a factor of 2-3 higher than previously reported using analytical techniques. Cerebral lactate reached a maximum concentration of about 1.0 mM.

Dynamic FLASH (fast low-angle shot) magnetic resonance (MR) imaging was used to monitor changes in brain oxygenation in the human visual cortex during photic stimulation. The approach exploits the sensitivity of the gradient-echo signal to susceptibility changes induced by varying concentrations of paramagnetic deoxyhemoglobin in the cerebral blood pool. After the onset of binocular photic stimulation (10 Hz, red light, checker-board), there was a distinct increase in the MR signal in the calcarine cortex within 6-9 seconds, indicating a decrease in the total deoxyhemoglobin concentration. After the stimulation was switched off, the MR signal returned to a basal value within a similar period of time. Assuming enhanced blood flow and only a minor increase in oxygen consumption (production of deoxyhemoglobin) during physiologic activation, the results reflect an enhanced supply of diamagnetic oxyhemoglobin and an increase in the partial oxygen pressure in the capillary and venous blood pools. In addition, a decrease in the basal MR signal in the calcarine cortex was observed during the first 60-90 seconds of persistent activation, which may be understood as an autoregulatory adaptation to increased overall brain activity associated with information processing due to continuous perception of visual stimuli.

A recently developed method for image-selected localized hydrogen-1 magnetic resonance (MR) spectroscopy was assessed in the differential diagnosis of nine primary and secondary cerebral tumors, including four gliomas, two meningiomas, one neurilemoma, one arachnoid cyst, and one metastasis of breast cancer. Well-resolved H-1 MR spectra of these tumors were obtained in vivo with a conventional 1.5-T whole-body MR imaging system. All tumor spectra were remarkably different from spectra from normal brain tissue. Spectra obtained from different tumors exhibited reproducible differences, while histologically similar tumors yielded characteristic spectra with only minor differences. The observed spectral alterations reflect variations in concentrations and relaxation times of the H-1 MR sensitive pool of free (mobile) metabolites within the tissues. In most cases, the concentrations of N-acetyl-aspartate and creatine/phosphocreatine are reduced below detectability, whereas choline-containing compounds are generally enhanced. The spectral differences between the tumors are mainly due to the differing concentrations of lipids, lactic acid, and carbohydrates. Localized H-1 MR spectroscopy may become an important clinical tool for the differentiation of tumors as well as for therapeutic control.

Localized proton NMR spectroscopy at 1.5 T using stimulated echoes has been applied to study metabolic alterations in the postischemic phase of patients with acute cerebral infarction. A complete depletion of N-acetyl aspartate in the area of infarction has been observed in a patient studied 4 days after stroke. This finding was paralleled by a dramatic increase in the concentration of lactic acid to about 16 mM within the lesion, indicating continued anaerobic glycolysis. The diluting effect of the edema has been estimated to reduce average metabolite concentrations by about a factor of 3.

No abstract