Nancy L. Sicotte

Automated image registration : II. Intersubject validation of Linear and nonlinear models

PURPOSE Our goal was to validate linear and nonlinear intersubject image registration using an automated method (AIR 3.0) based on voxel intensity. METHOD PET and MRI data from 22 normal subjects were registered to corresponding averaged PET or MRI brain atlases using several specific linear and nonlinear spatial transformation models with an automated algorithm. Validation was based on anatomically defined landmarks. RESULTS Automated registration produced results that were superior to a manual nine parameter variant of the Talairach registration method. Increasing the degrees of freedom in the spatial transformation model improved the accuracy of automated intersubject registration. CONCLUSION Linear or nonlinear automated intersubject registration based on voxel intensities is computationally practical and produces more accurate alignment of homologous landmarks than manual nine parameter Talairach registration. Nonlinear models provide better registration than linear models but are slower.

Treatment of multiple sclerosis with the pregnancy hormone estriol.

Multiple sclerosis patients who become pregnant experience a significant decrease in relapses that may be mediated by a shift in immune responses from T helper 1 to T helper 2. Animal models of multiple sclerosis have shown that the pregnancy hormone, estriol, can ameliorate disease and can cause an immune shift. We treated nonpregnant female multiple sclerosis patients with the pregnancy hormone estriol in an attempt to recapitulate the beneficial effect of pregnancy. As compared with pretreatment baseline, relapsing remitting patients treated with oral estriol (8mg/day) demonstrated significant decreases in delayed type hypersensitivity responses to tetanus, interferon‐γ levels in peripheral blood mononuclear cells, and gadolinium enhancing lesion numbers and volumes on monthly cerebral magnetic resonance images. When estriol treatment was stopped, enhancing lesions increased to pretreatment levels. When estriol treatment was reinstituted, enhancing lesions again were significantly decreased. Based on these results, a larger, placebo‐controlled trial of estriol is warranted in women with relapsing remitting multiple sclerosis. This novel treatment strategy of using pregnancy doses of estriol in multiple sclerosis has relevance to other autoimmune diseases that also improve during pregnancy.

Regional hippocampal atrophy in multiple sclerosis

Gray matter brain structures, including deep nuclei and the cerebral cortex, are affected significantly and early in the course of multiple sclerosis and these changes may not be directly related to demyelinating white matter lesions. The hippocampus is an archicortical structure that is critical for memory functions and is especially sensitive to multiple insults including inflammation. We used high-resolution MR imaging at 3.0 T to measure hippocampal volumes in relapsing remitting MS (RRMS) and secondary progressive MS (SPMS) patients and controls. We found that both groups of MS patients had hippocampal atrophy and that this volume loss was in excess of global brain atrophy. Subregional analysis revealed selective volume loss in the cornu ammonis (CA) 1 region of the hippocampus in RRMS with further worsening of CA1 loss and extension into other CA regions in SPMS. Hippocampal atrophy was not correlated with T2-lesion volumes, and right and left hippocampi were affected equally. Volume loss in the hippocampus and subregions was correlated with worsening performance on word-list learning, a task requiring memory encoding, but not with performance on the Paced Auditory Serial Addition Task (PASAT), a test of information processing speed. Our findings provide evidence for selective and progressive hippocampal atrophy in MS localized initially to the CA1 subregion that is associated with deficits in memory encoding and retrieval. The underlying histopathological substrate for this selective, symmetric and disproportionate regional hippocampal vulnerability remains speculative at this time. Further understanding of this process could provide targets for therapeutic interventions including neuroprotective treatments.

Onset of multiple sclerosis associated with anti-TNF therapy

Therapies aimed at inhibiting tumor necrosis factor (TNF), a proinflammatory cytokine implicated in autoimmune disease are effective, especially for rheumatoid arthritis. We report a patient with new onset MS closely associated with the initiation of anti-TNF therapy for juvenile rheumatoid arthritis. It is possible that the inhibition of TNF triggered MS in this individual.

Immune Modulation in Multiple Sclerosis Patients Treated with the Pregnancy Hormone Estriol

The protective effect of pregnancy on putative Th1-mediated autoimmune diseases, such as multiple sclerosis and rheumatoid arthritis, is associated with a Th1 to Th2 immune shift during pregnancy. The hormone estriol increases during pregnancy and has been shown to ameliorate experimental autoimmune encephalomyelitis and collagen-induced arthritis. In addition, estrogens induce cytokine changes consistent with a Th1 to Th2 shift when administered in vitro to human immune cells and in vivo to mice. In a pilot trial, oral estriol treatment of relapsing remitting multiple sclerosis patients caused significant decreases in enhancing lesions on brain magnetic resonance imaging. Here, the immunomodulatory effects of oral estriol therapy were assessed. PBMCs collected longitudinally during the trial were stimulated with mitogens, recall Ags, and glatiramer acetate. Cytokine profiles of stimulated PBMCs were determined by intracellular cytokine staining (IL-5, IL-10, IL-12 p40, TNF-α, and IFN-γ) and cytometric bead array (IL-2, IL-4, IL-5, IL-10, TNF-α, and IFN-γ). Significantly increased levels of IL-5 and IL-10 and decreased TNF-α were observed in stimulated PBMC isolated during estriol treatment. These changes in cytokines correlated with reductions of enhancing lesions on magnetic resonance imaging in relapsing remitting multiple sclerosis. The increase in IL-5 was primarily due to an increase in CD4+ and CD8+ T cells, the increase in IL-10 was primarily due to an increase in CD64+ monocytes/macrophages with some effect in T cells, while the decrease in TNF-α was primarily due to a decrease in CD8+ T cells. Further study of oral estriol therapy is warranted in Th1-mediated autoimmune diseases with known improvement during pregnancy.

Creation and use of a Talairach-compatible atlas for accurate, automated, nonlinear intersubject registration, and analysis of functional imaging data.

Spatial normalization in functional imaging can encompass various processes, including nonlinear warping to correct for intersubject differences, linear transformations to correct for identifiable head movements, and data detrending to remove residual motion correlated artifacts. We describe the use of AIR to create a custom, site‐specific, normal averaged brain atlas that can be used to map T2 weighted echo‐planar images and coplanar functional images directly into a Talairach‐compatible space. We also discuss extraction of characteristic descriptors from sets of linear transformation matrices describing head movements in a functional imaging series. Scores for these descriptors, derived using principal components analysis with singular value decomposition, can be treated as confounds associated with each individual image in the series and systematically removed prior to voxel‐by‐voxel statistical analysis. Hum. Brain Mapping 8:73–79, 1999.

Comparison of Multiple Sclerosis Lesions at 1.5 and 3.0 Tesla

ObjectiveTo evaluate the relative sensitivity of MR scanning for multiple sclerosis (MS) at 1.5 Tesla (T) and 3.0 T using identical acquisition conditions, as is typical of multicenter clinical trials. MethodsTwenty-five subjects with MS were scanned at 1.5 T and 3.0 T using fast spin echo, and T1-weighted SPGR with and without gadolinium contrast injections. Image data, blinded to field strength, were analyzed using automated segmentation and lesion counting. ResultsRelative to scanning at 1.5 T, the 3.0 T scans showed a 21% increase in the number of detected contrast enhancing lesions, a 30% increase in enhancing lesion volume and a 10% increase in total lesion volume. DiscussionThe improved detection ability using high-field MR imaging is prominent even when sequence parameters are optimized around the midfield units. Multicenter trials using both 1.5 T and 3.0 T instruments may be affected by these sensitivity differences.

Smaller Cornu Ammonis 2-3/Dentate Gyrus Volumes and Elevated Cortisol in Multiple Sclerosis Patients with Depressive Symptoms

BACKGROUND The hippocampus is likely involved in mood disorders, but in vivo evidence for the role of anatomically distinct hippocampal subregions is lacking. Multiple sclerosis, an inflammatory disease of the central nervous system, is linked to a high prevalence of depression as well as hippocampal damage and may thus provide important insight into the pathologic correlates of medical depression. We examined the role of subregional hippocampal volume for depression in relapsing-remitting multiple sclerosis. METHODS Anatomically defined hippocampal subregional volumes (cornu ammonis 1-3 [CA1-CA3] and the dentate gyrus [CA23DG], subiculum, entorhinal cortex) were measured using a high-resolution T2-weighted magnetic resonance imaging sequence in 29 relapsing-remitting multiple sclerosis patients and 20 matched healthy control subjects. Diurnal salivary cortisol was assessed at awakening, 4 pm, and 9 pm on 2 consecutive days. Subjects also completed the Beck Depression Inventory. RESULTS Multiple sclerosis patients showed smaller hippocampal volumes compared with control subjects, particularly in the CA1 and subiculum subregions. In addition, multiple sclerosis patients with depressive symptoms (Beck Depression Inventory score >13) also showed smaller CA23DG volumes and higher cortisol levels. Within the multiple sclerosis group, CA23DG volume was correlated with depressive symptoms and cortisol levels. There were no associations with number of previous steroid treatments, global atrophy, or disease duration. CONCLUSIONS This report provides in vivo evidence for selective association of smaller CA23DG subregional volumes in the hippocampus with cortisol hypersecretion and depressive symptoms in multiple sclerosis.

Handedness in Twins: A Meta-analysis

In the largest meta-analysis of twins and singletons conducted to date we have found a higher incidence of left-handedness in twins compared to singletons. Our analysis revealed no difference in the frequency of left-handedness among monozygotic versus dizygotic twins. However, identical twins were more likely to be concordant for hand preference than non-identical twins, which is consistent with a genetic model of handedness. Prior analyses have not revealed these findings consistently, and this has led to a number of conflicting models of handedness.

Temporal and Topographical Characterization of Language Cortices Using Intraoperative Optical Intrinsic Signals

We used intraoperative optical imaging of intrinsic signals (iOIS) and electrocortical stimulation mapping (ESM) to compare functionally active brain regions in 10 awake patients undergoing neurosurgical resection. Patients performed two to four tasks, including visual and auditory naming, word discrimination, and/or orofacial movements. All iOIS maps included areas identified by ESM mapping. However, iOIS also revealed topographical specificity dependent on language task. In Brocas area, naming paradigms activated both anterior and posterior inferior frontal gyrus (IFG), while the word discrimination paradigm activated only posterior IFG. In Wernickes area, object naming produced activations localizing over the inferior and anterior/posterior regions, while the word discrimination task activated superior and anterior cortices. These results may suggest more posterior phonological activation and more anterior semantic activations in Brocas area, and more anterior/superior phonological activation and more posterior/inferior semantic activations in Wernickes area. Although similar response onset was observed in Brocas and Wernickes areas, temporal differences were revealed during block paradigm (20-s) activations. In Brocas area, block paradigms yielded a boxcar temporal activation profile (in all tasks) that resembled response profiles observed in motor cortex (with orofacial movements). In contrast, activations in Wernickes area responded with a more dynamic profile (including early and late peaks) which varied with paradigm performance. Wernickes area profiles were very similar to response profiles observed in sensory and visual cortex. The differing temporal patterns may therefore reflect unique processing performed by receptive (Wernickes) and productive (Brocas) language centers. This study is consistent with task-specific semantic and phonologic regions within Brocas and Wernickes areas and also is the first report of response profile differences dependent on cortical region and language task.