Faculty Bibliography

There is compelling evidence for the early involvement of the hippocampal formation in the natural history of Alzheimer's disease (AD). The evidence comes from recent neuropathology, neuropsychology, and neuroimaging studies. AD-type histopathologic changes limited to the hippocampus have been described and may be seen in normal aging subjects. The sites of maximal neuronal loss in the hippocampal formation are in the CA1, subiculum, and entorhinal cortex. Minimally cognitively impaired (MCI) individuals (defined by ratings of functional capacity and psychiatric symptomatology) exhibit a neuropsychological profile that is distinct from that of the unimpaired elderly. Pathologic evidence suggests that most of these cases already have AD brain changes accentuated in the hippocampal region, and our own longitudinal studies reveal that 70% of this group develop dementia within a 4-year period. We have developed a negative-angle axial view designed to cut parallel to the anterior-posterior plane of the hippocampus. Using this modified axial plane of section in conjunction with computed tomography (CT) and magnetic resonance imaging (MRI), we estimated the prevalence of hippocampal atrophy in normal aging and across severity levels of cognitively impaired elderly patients. Longitudinal study shows that hippocampal atrophy is a sensitive and specific predictor of future AD for patients with MCI. MRI volume study of AD patients, controls, and MCI patients shows specific hippocampal volume loss in MCI. We conclude that the atrophic changes associated with early AD can be visualized using qualitative techniques and are readily quantifiable with volumetry. This article is not intended to be comprehensive, but to provide an overview of some of the structural neuroimaging data from our laboratory

The magnetic field distortion due to magnetization tolerances, fabrication imperfections, and design alterations can be reduced by inserting into a permanent magnet structure thin straps of high permeability ferromagnetic material that lie along equipotential surfaces of the ideal field of a theoretical design. Numerical results are given for a simple model showing quantitatively the beneficial effect of strapping in improving the quality of the field

PURPOSE: To assess the safety of high-dose nonionic contrast media (CM) during a single radiologic procedure. MATERIALS AND METHODS: From November 1991 to August 1995, 255 high-dose angiographic procedures were performed in 228 patients with normal serum creatinine (SCr) levels (< or = 1.6 mg/dL [141 mumol/L]). All patients received 250-800 mL low-osmolarity CM (300 mg iodine per milliliter). Pre- and postprocedure SCr levels were assessed. Urine output was measured daily in the 75 patients who received more than 400 mL CM. With linear regression analysis, a dose-related elevation in SCr levels was calculated. RESULTS: No patient developed abnormal SCr levels (> 1.6 mg/dL [141 mumol/L]) as a result of the CM. Among the patients who received more than 400 mL, none developed oliguria over the first 36 hours. With follow-up up to 3 years, no patient experienced delayed clinical renal failure. In 11 (4.3%) patients, the SCr levels increased more than 25%, but all increases were within expected limits (chi 2 analysis). Linear regression analysis revealed a 0.015 mg/dL (1 mumol/L) increase in SCr levels per 100 mL CM. CONCLUSION: Intravenous administration of high-dose low-osmolarity iodinated CM appears safe in patients without renal dysfunction or other underlying risk factors, in doses as large as 800 mL (300 mg iodine per milliliter)

Positron emission tomography and the fluorodeoxyglucose method were used to measure regional brain metabolism before and 2 h after haloperidol (5 mg, i.m.) in 11 young normal men. These data were compared with measures obtained from nine previously studied normal men who had received no drug intervention. Although a previously published study had demonstrated significantly decreased metabolism in whole brain, neocortex, limbic cortex, thalamus, and caudate nucleus 12 h after a 5-mg dose of haloperidol, the present 2-h study did not show significant metabolic changes despite the fact that significant extrapyramidal effects occurred. Taken together, these studies demonstrate differences in the temporal organization of behavioral and metabolic responses to haloperidol challenge

Population trends indicate that in the near future the size of the elderly population will increase. This will result in a large increment in the numbers of persons suffering mild to severe levels of cognitive impairment. While considerable efforts continue to be made to explain brain changes associated with Alzheimer disease (AD), little is known of the brain changes in aging without dementia or so-called normal aging. Pathologic studies suggest that the medial temporal lobe is informative in the examination of the early brain changes related to AD. However, pathologic studies only offer a single observation and considerable uncertainty exists regarding the likelihood of progression of disease and the development of dementia. Several structural neuroimaging studies have recently investigated this anatomy and recent reports are encouraging for a medial temporal lobe based diagnosis for age-related cognitive impairments. We will present our findings on the MRI anatomy of the hippocampal formation as well as data bearing on the use of hippocampal formation imaging in the diagnosis of AD and as a predictive marker for future dementia. Our findings suggest an anatomically specific relationship between hippocampal volume and secondary memory performance. Because these observations apply to nondemented and normal elderly subjects, we are encouraged that the anatomy of age-related cognitive impairments can be reliably recognized and possibly put to use in therapeutic studies

Designing gradient coils for permanent magnets requires modeling the interaction between the currents and the ferromagnetic material. We have developed and tested an iterative design approach based on the boundary element method. The power dissipated by the gradient coil is treated as a constraint of the minimization problem. The technique is illustrated on a magnet with a rectangular prismatic cavity and two ferromagnetic plates adjacent to the coil. The procedure improved the linearity of the field from 10.5% to 3.4% and reduced the power dissipation to 63% of the power of the unoptimized coil. The optimized distribution of currents has been transformed to variable locations of coil windings without loss of gradient uniformity. The new approach significantly reduces computation time and memory storage and yields practical designs of gradient coils for permanent magnets