Faculty Bibliography

PURPOSE: To examine the sensitivity of proton MR spectroscopy for detecting early central nervous system abnormalities in neonates born to human immunodeficiency virus (HIV)-positive mothers. METHODS: Asleep, unsedated, and continuously monitored by electrocardiography, 10 newborns, 5 with HIV-positive and 5 with HIV-negative mothers, were studied within the first 10 days of life in a 1.5-T scanner. After T1- and T2-weighted images were obtained, proton spectra were performed using voxels of interest (3.4 cm3) in the deep parietooccipital white matter. Peaks were identified as N-acetyl-aspartate (2.0 ppm), creatine and phosphocreatine (3.0 ppm), choline (3.2 ppm), and inositol (3.5 ppm). Peak areas were used to calculate metabolic ratios: N-acetyl-aspartate to creatine, inositol to creatine, and creatine to choline. RESULTS: All newborns of HIV-positive mothers had abnormal proton spectra compared with control infants; a nonspecific amino acid peak in the 2.1- to 2.6-ppm area was elevated, broad, and overlapping the N-acetyl-aspartate peak in all the HIV-exposed newborns and in only 1 of the 5 control newborns. The choline-to-creatine ratio was higher in HIV-exposed newborns at 2.3 +/- 0.4 (normal term, 0.9 +/- 0.3), as was the N-acetyl-aspartate-to-creatine ratio at 2.6 +/- 0.9 (for control subjects, 1.2 +/- 0.4). MR images from these brain regions were all considered normal. Because acquired immunodeficiency syndrome develops in only a small fraction of neonates born to HIV-seropositive mothers, the above spectral abnormalities found in all our subjects may result from indirect effects of HIV, such as intrauterine growth retardation. CONCLUSIONS: These findings indicate that proton MR spectroscopy might play an important role in detecting early central nervous system complications in newborns of HIV-seropositive mothers

PURPOSE: To assess the prevalence of MR evidence for diffuse axonal injury at 1.5 T in patients with normal head CT findings after mild head injury. METHODS: Twenty consecutive patients with mild head injury (Glasgow Coma Scale, 13 to 15; no subsequent deterioration, loss of consciousness < 20 minutes) and normal head CT findings were examined with MR at 1.5 T. Pulse sequences included a conventional T2-weighted spin-echo sequence (2500-3000/30,80/1[repetition time/echo time/excitations]) and a T2*-weighted gradient-echo sequence (750/40/2, 10 degrees flip angle). Each sequence was read independently by two blinded readers. RESULTS: The readers agreed that abnormalities compatible with diffuse axonal injury were present in the white matter of 6 (30%) of 20 patients (95% confidence interval, 12% to 54%). Both readers agreed that foci of high signal intensity were present on the T2-weighted spin-echo sequence in 3 (15%) of the 20 cases (95% confidence interval, 3% to 38%) and that foci of hypointensity compatible with hemorrhagic shear injury were present on the T2*-weighted sequence in 4 (20%) of the 20 patients (95% confidence interval, 6% to 44%). Both types of abnormality were noted by the readers in one patient. CONCLUSIONS: MR shows evidence of diffuse axonal injury in some patients with normal head CT findings after mild head injury. These lesions may represent the pathologic substrate underlying the postconcussion syndrome that occurs in many patients with moderate to severe head injury

Magnetization transfer imaging (MT) and localized proton spectroscopy (1H-MRS) were utilized in the evaluation of lesions (high signal abnormalities on T2-weighted images) and normal-appearing white matter (NAWM) in multiple sclerosis (MS). Eleven patients with a clinical diagnosis of MS were independently evaluated with both 1H-MRS and MT. The magnetization transfer ratio (MTR) of lesions was compared with the relative concentration of N-acetyl-aspartate (NAA) and a composite peak at 2.1 to 2.6 ppm termed 'marker peaks.' The MTR of white matter lesions in the MS patients was markedly decreased (6-34%; normal approximately 42%), and correlated well with increase in the marker peaks region (0.94-3.89). There was no correlation between the relative concentration of NAA and MTR. Increased resonance peaks in the 2.1 to 2.6 ppm range and marked decreases in MTR may be a relatively specific indicators of demyelination

This study investigated effects of age and sex on regional brain structure in humans, focusing on the frontal and temporal lobes. Hemispheric volumes were obtained from magnetic resonance images (MRIs) of 96 young (53 men, 43 women; aged 18-40 years) and 34 older (17 men, 17 women; aged 41-80) healthy volunteers. Images (5 mm axial spin-echo, repetition time of 3000 msec and echo times of 30 and 80 msec) were resliced along the anterior commissure-posterior commissure (AC-PC) axis to standardize for difference in head tilt, and imported into a computer program where borders of the frontal and temporal lobes were delineated. The program calculated regional brain volumes based on slice data from which CSF was segmented out. An age x sex x hemisphere x region interaction indicated that age-related reductions in brain volume were sexually dimorphic, lateralized, and region specific. Greater decrements in brain volume occurred with age in the frontal lobe than in the temporal lobe. Age-related reductions in both regions were greater in men than in women, demonstrating that sexual dimorphisms in human neuroanatomy are not fixed, but continue to change throughout adulthood. The possibility that gonadal hormones play a role in the promotion and/or prevention of neural atrophy with aging is discussed

PURPOSE: To study solitary metastatic lesions of the brain with routine spin-echo (SE) and magnetization transfer (MT) magnetic resonance (MR) imaging. MATERIALS AND METHODS: In 15 patients with such lesions, the MT ratio (MTR) was calculated in the center and at the periphery of the metastatic lesion, and distally in adjacent white matter, from the periphery of the lesion radially to the most distant cortex. It also was calculated for mirror-image locations in the opposite hemisphere to provide control values. RESULTS: MTRs were decreased in and immediately around the site of the metastatic focus. MTRs also were lower than control values far distal to the metastatic focus, even when no abnormality was seen on SE MR images obtained before and those obtained after administration of gadopentetate dimeglumine. CONCLUSION: MT enables demonstration of white matter abnormalities in patients with metastatic lesions not seen on SE MR images. These changes can be found in white matter far distal to the lesion and surrounding areas of edema

PURPOSE: To evaluate the role of MR in the diagnosis and treatment of patients with neurosarcoidosis. METHODS: The MR studies of 24 patients who satisfied stringent criteria for the diagnosis of sarcoid were retrospectively reviewed. All patients had signs and symptoms referable to the head and/or spine. The majority, 17 patients (71% of the total), were examined at least once with gadolinium enhancement. Fifteen of 24 patients (63%) underwent serial examinations during steroid therapy. RESULTS: A wide spectrum of findings was noted: white matter and periventricular high signal intensity on long-repetition-time/long-echo-time sequences, mimicking multiple sclerosis (11 patients); leptomeningeal enhancement (11 patients); brain parenchymal mass (seven patients)--six demonstrated enhancement, one did not receive contrast; lacrimal gland mass (three patients); hydrocephalus (three patients); enlarged ventricles, apparently atrophic (one patient); periventricular enhancement (three patients); extraaxial mass, mimicking meningioma (two patients); chiasmal enhancement or swelling (one patient); enhancing nerve roots (two patients); enlarged pituitary stalk (two patients); pontine infarct (one patient); and enhancing parenchymal spinal cord mass (three patients). Partial or complete resolution of the radiographic abnormality occurred in 13 of 15 cases (87%), which paralleled clinical improvement. No response was detected in the remaining two. Abnormal enhancement was the finding that was most responsive to steroid therapy, with response seen in nine of 10 patients with leptomeningeal enhancement, in six of six patients with enhancing brain parenchymal masses, in three of three patients with enhancing cord masses, and in all three patients with periventricular enhancement. CONCLUSIONS: 1) MR shows a spectrum of protean central nervous system abnormalities associated with neurosarcoidosis. 2) This high sensitivity for associated abnormalities aids in differentiating central nervous system sarcoid from the many diseases that it can mimic. In particular, enhancement was a useful clue to the diagnosis in 15 of 17 cases in which it was used (88%). 3) MR demonstrates regression of central nervous system abnormalities during steroid therapy, in particular abnormal meningeal, periventricular, and parenchymal enhancement

Magnetization transfer, a new technique for improving image contrast in magnetic resonance (MR) imaging, is based on application of off-resonance radio-frequency pulses and observing their effects on MR images, as well as measuring the signal intensity with and without application of the pulses (ie, magnetization transfer ratio [MTR]). MTRs can be used to detect changes in the structural status of brain parenchyma that may or may not be visible with standard MR techniques. Use of MTRs may allow subcategorization of multiple sclerosis lesions into those with very low MTR (demyelinated lesions) and slightly decreased MTR (edematous lesions). In cases of wallerian degeneration, use of MTRs appears to allow reliable detection of changes undetectable with MR imaging or even light microscopy. In cases of infection with human immunodeficiency virus, MTRs seem to indicate that the macromolecular structure of white matter remains intact until relatively late in the course of disease. In cases of metastatic disease, MTRs of brain lesions indicate structural changes beyond the extent of the lesions seen on standard MR images. These findings may be due to chronic edema, myelin loss, and perhaps previous undetected tumor. In addition to being a new method of providing contrast, the magnetization transfer technique enables semi-quantitative, reproducible characterization of tissue and pathologic entities, which could substantially improve the specificity of MR imaging

PURPOSE: To examine the utility of measuring magnetization transfer ratio for for delineating the dynamic changes of wallerian degeneration which occur after controlled injury in a feline model in which anatomic pathways are well understood. METHODS: Using standard neurosurgical techniques, discrete lesions were made to ablate the visual cortex. Gradient imaging was performed serially at 1.5 T, with and without a saturation pulse to create a magnetization transfer effect. At varying intervals, the animals were killed for histologic analysis. RESULTS: Within the first 2 weeks there is a statistically significant increase in magnetization transfer ratio relative to the control hemisphere within the white matter connections between the lateral geniculate nucleus and the visual cortex at a time when no effects are visually detectable on spin-echo images. Between 16 and 28 days, this reverses to a decrease in magnetization transfer ratio in both the lateral geniculate nucleus itself and the adjacent superolateral white matter. More remote white matter tracts remained stable, without significant change. CONCLUSIONS: Magnetization transfer ratio seems to be more sensitive for early detection of degeneration than conventional spin-echo imaging. Moreover, temporal changes in magnetization transfer ratio seem to correspond well with known histologic phases of wallerian degeneration

Magnetization transfer (MT) is a technique that has the potential for detecting changes in myelin. The rigid macromolecular structure of myelin is the physical basis of MT. By using off-resonance irradiation, macromolecular protons can be saturated. These protons then exchange with free-water protons and produce a decrease in signal intensity of the free-water protons. This can be quantitated by using a magnetization transfer ratio (MTR) of signal intensities (Mo-Ms/Mo. Mo represents the signal intensity without off-resonance irradiation, and Ms represents the signal intensity with off-resonance irradiation). This method has been sensitive to changes in a spectrum of white-matter lesions, including an edema model (EAE), and regions of apparent myelin loss in patients with multiple sclerosis (MS). Furthermore, MTRs may be abnormal in patients with normal-appearing white matter, demonstrated by standard MR imaging. MTRs may enable subcategorization of MS lesions into lesions with low MTRs (presumed to be demyelinating lesions) and lesions with higher values (primarily edematous lesions). Another important use of MT is, in conjunction with gadolinium, to increase the number and extent of enhancing MS lesions. This can improve the detection of blood-brain barrier abnormalities in patients with MS