Neurophysiological effects of multiple mood episodes in bipolar disorder.

OBJECTIVES: Bipolar disorder is marked by progressive symptomatic changes, which have been linked with episode-related structural findings-particularly in the prefrontal cortex. However, few studies have examined neurofunctional and neurochemical effects of disease burden. In this study, we compared first- and multi-episode bipolar individuals. We hypothesized that the latter would demonstrate evidence of neurophysiological differences consistent with a model of progressive functional degradation of these networks. METHODS: First- and multi-episode manic bipolar subjects participated in functional magnetic resonance imaging (fMRI) including a continuous performance task with emotional distractors, and in single-voxel (1 H) magnetic resonance spectroscopy (MRS). A priori fMRI regions-of-interest (ROI) included structures comprising prefrontal-striatal-amygdala networks; (1 H)MRS voxels were placed within bilateral ventrolateral prefrontal (VLPFC) and anterior cingulate cortex (ACC). Both ROI and voxel-based brain activation in response to emotional stimuli, and neurochemical concentrations derived from (1 H)MRS were compared across bipolar groups. RESULTS: Multi-episode bipolar subjects showed relatively lower regional activation across prefrontal-striatal-amygdala networks, including bilateral VLPFC, orbitofrontal cortex, ACC, putamen, caudate, and amygdala. Exploratory whole-brain, voxel-based analysis suggested additional areas of lower activation extending into Brodmann area 22, posterior parietal regions, and right thalamus. Glutamate and N-acetylaspartate (NAA) concentrations were also relatively lower in the ACC of multi-episode subjects. CONCLUSIONS: Disease burden, exemplified by multiple affective episodes is associated with evidence of widespread decrements in affective network activity. Lower ACC NAA concentration is similarly consistent with a model of progressive functional deficits. These findings support the functional significance of previously observed progressive structural changes throughout these regions.

31P NMR spectroscopy of phospholipid metabolites in postmortem schizophrenic brain.

Evidence has been accumulating that schizophrenia involves abnormalities in the composition and metabolism of cell membrane phospholipids (PLs) in the brain. In vivo 31P MRS has been used to measure the metabolic precursors and degradation products of PL metabolism in schizophrenia. Because in vivo line widths are substantially broader than in solution, only the broad phosphomonoester (PME) and phosphodiester bands, or partly resolved resonances of individual metabolites, are typically measured in vivo in the 31P spectrum. In addition to poor resolution, the relatively low signal-to-noise ratio (SNR) makes precise quantitation difficult. An alternative with substantially better resolution and precision for quantitation is high-resolution NMR spectroscopy of extracts of samples from postmortem brain. Here we determine absolute concentrations of the individual PL metabolites phosphocholine (pc), phosphoethanolamine (pe), glycerophosphocholine (gpc), and glycerophosphoethanolamine in aqueous extracts of tissue from frontal, temporal, and occipital cortex of postmortem brain for schizophrenics, controls, and patients with other mental illnesses (psychiatric controls [PC]) using high-resolution 31P NMR spectroscopy. For the complete groups, which included both males and females, there were no statistically significant differences for schizophrenics vs. controls for any of the four PL metabolites in any of the three brain regions. Trends (0.05 < P < 0.10) were noted for increased gpc in schizophrenia in all three regions. PC differed from both controls and schizophrenics in several measures. When only males were considered, gpc was significantly (P < 0.05) elevated in all three brain regions in schizophrenia.

Brain metabolite concentration ratios in vivo: multisite reproducibility by single-voxel 1H MR spectroscopy.

Ten normal subjects were scanned identically at three separate sites (Little Rock, Houston, and New Orleans) to evaluate the reproducibility of brain metabolite ratios in single-voxel (1)H point-resolved spectroscopy sequence (PRESS) magnetic resonance (MR) spectroscopy in vivo. All scans were processed by a single individual at a single site. Coefficients of variation of the measured metabolite ratios generally were in the range found for previous single-voxel, single-site reproducibility studies. No differences were found among the sites for ratios of N-acetylaspartate to creatine (NAA/Cr) or choline to Cr (Cho/Cr) in left thalamus by multivariate ANOVA. Metabolite ratios of Cr or Cho relative to local brain H(2)O did not vary among the sites. However, by multivariate ANOVA, NAA/H(2)O differed between Little Rock and New Orleans, but not between those sites and Houston. Intraclass correlation coefficients suggested reasonable reproducibility between Little Rock and New Orleans, but not between those sites and Houston.