183. Cerebral hypoperfusion precedes cognitive decline in hypertensive patients with ischemic white matter lesions
K. Kitagawa1, N. Oku2, Y. Kimura3, Y. Yagita1, M. Sakaguchi4, J. Hatazawa3 and S. Sakoda1
1Department of Neurology, Osaka University Graduate School of Medicine, Suita; 2Hyogo College of Medicine Hospital, Nishinomiya; 3Department of Tracer Kinetics and Nuclear Medicine; 4Stroke Center, Osaka University Graduate School of Medicine, Suita, Japan
Background and aims: Vascular risk factors have been believed important for cognitive decline and dementia. Our purpose was to determine if cerebral hypoperfusion contriubutes to cognitive decline in hypertensive patients with deep white matter lesions.
Methods: We have enrolled 1065 non-demented patients with at least one cardiovascular risk factor or any history of cardiovascular events, who had underwent carotid ultrasonography to evaluate the severity of carotid atheroscleroisis from 2001 to 2005 in a hospital-based cohort study. Among them, twenty-seven hypertensive patients (age, 64.8±7.5 yr) with an evidence of cerebral small vessel disease in MRI underwent PET study with an 15[O]-labeled water injection for measurement of baseline cerebral blood flow and mini-mental state examination for cognitive function. Patients without significant arterial stenosis in MR angiography and carotid duplex ultrasonography were enrolled to focus on cerebral hemodynamics with cerebral small vessel disease. At an interval of three years, these patients underwent second MMSE test. Cognitive decline was defined as a drop of four points in 3 year.
Results: Baseline MMSE score was 27.5±1.9 (mean±s.d.). Carotid IMT was 0.94±0.18 mm. Six patients exhibited cognitive decline, and three of them were diagnosed as dementia. Baseline CBF in six patients with cognitive decline was significantly lower than that in the rest 21 patients (31.2±2.4 vs. 42.5±5.9 ml/100g/min, P<0.001). Among two groups with and without cognitive decline in the follow-up period, no difference was observed in the baseline data of age, sex, education period, history of stroke, cardiovascular risk factor, the carotid IMT, MMSE score, severity of brain atrophy and white matter lesions.
Conclusions: Our prospective study shows that cerebral hypoperfusion is a significant predictor for future cognitive decline. Our results strongly support a hypothesis that cerebral hypoperfusion and vascular risk factors contribute to cognitive decline and incident dementia.
775. [18F]Feta PET—a new method for hot spot imaging of the ischemic penumbra in acute stroke
H. Barthel1,2, U. Großmann2, V. Zeisig3, M. Patt1, D. Wagner3, J. Patt1, M. Kluge1, H. Franke4, D. Sorger1, J. Luthardt1, B. Nitzsche3, A. Dreyer3, P. Brust5, J. Steinbach5,6, J. Boltze2,3, F. Emmrich2,3 and O. Sabri1
1Department of Nuclear Medicine, University of Leipzig; 2Translational Centre for Regenerative Medicine, University of Leipzig; 3Fraunhofer Institute for Cell Therapy and Immunology; 4Rudolph-Boehm-Institute of Pharmacology and Toxicology; 5Institute of Interdisciplinary Isotope Research, University of Leipzig, Leipzig; 6Institute of Radiopharmacy, Forschungszentrum Dresden-Rossendorf, Dresden-Rossendorf, Germany
Objectives: In acute ischemic stroke, an imaging method to directly visualize the ischemic penumbra – the salvageable part of the affected brain—in positive image contrast would potentially improve therapy stratification and monitoring. This study aimed to test [18F]fluoroetanidazole ([18F]FETA), a second-generation radiolabeled 2-nitroimidazole, for the first time with respect to its suitability to image brain hypoxia with positron emission tomography (PET).
Methods: Primary embryonal corticoencephalic cells (Wistar rats) and necortical brain slices (Sprague Dawley rats) were ex vivo exposed to nitrogen or air. The cells and brain slices were incubated with 5MBq [18F]FETA up to 120 min, respectively. The activities of three nitroreductases—enzymes which mediate the intracellular [18F]FETA accumulation—were determined in the corticoencephalic cells. Further, organ distribution was determined in Sprague Dawley rats up to 2 h after i.v. injection of 20MBq [18F]FETA, and ex vivo brain autoradiography was performed up to 24 h after permanent middle cerebral artery occlusion (pMCAO). Target-to-background image contrast of [18F]FETA autoradiograms at 3 h after pMCAO was compared with that of corresponding [18F]fluoromisonidazole ([18F]FMISO) autoradiograms. At 24 h after pMCAO, animals were additionally i.v. injected with 1MBq [14C]iodoantipyrine to determine the local cerebral blood flow (lCBF). Nissl staining of brain slices as well as stroke-specific MRI were carried out at 24 h after pMCAO to confirm the existence and localization of ischemic brain tissue damage.
Results:In vitro, the oxygen concentration in the cell suspension was <1 mm Hg and ∼70 mm Hg under nitrogen and air, respectively. The normoxic [18F]FETA uptake by the cells and the brain slices was low and constant over time (0.3±0.08 %ID.mio cells−1 and 0.04±0.01 %ID.g tissue−1). In contrast, under hypoxia a time-dependent linear increase of the [18F]FETA uptake was found which was 2.0- and 2.5-fold by the cells and 2.0- and 2.4-fold by the brain slices at 60 min and 120 min (P<0.05), respectively. The analyses of nitroreductases activities showed that cell oxygenation does not affect the enzyme activities. The biodistribution studies revealed a fast blood clearance, a rapid urinary excretion and a constantly low uptake in unaffected brain tissue (0.1±0.02 %ID.g−1). Ex vivo brain autoradiography in the pMCAO rats showed a relevant time-dependent [18F]FETA uptake in ipsilateral brain regions which reached maximum target-to-background ratios of 3.3±0.2 at 3 h. The corresponding [18F]FMISO uptake ratios were only 1.5±0.3 (P<0.05). Furthermore, at 24 h after pMCAO the lCBF was reduced in the infarction core (as determined by Nissl staining and MRI) and surrounding brain areas by 25% and 10%, respectively.
Conclusions: These results demonstrate that [18F]FETA has a better potential than [18F]FMISO to serve as a brain hypoxia marker. Further testing of this promising new stroke PET marker is warranted. First results employing a new sheep stroke model developed recently by our group [1] are encouraging.
This research was supported by the Translational Centre for Regenerative Medicine Leipzig/BMBF (PtJ-Bio 0313909).
First and second author contributed equally to this study.
131. Longitudinal μPET study on a transgenic RAT model for huntington disease reveals progressive metabolic and dopaminergic changes
N. Van Camp1, R. Boisgard1, K. Siquier1, J. Benoit1, I. Blockx2, F. Chauveau1, F. Hinnen1, B. Kuhnast1, S. Von Hörsten3, H. Nguyen4, O. Riess4, F. Dollé1, A. Van Der Linden2 and B. Tavitian1
Objectives: Huntington′s disease (HD) is an autosomal dominant neurological disorder caused by an expansion of CAG repeats within the coding region of the HD gene (IT15). It is characterized by striatal degeneration, motor symptoms and complex neuropsychiatric alterations. Recently, von Hörsten et al (2003) described the first transgenic rat model for HD with a slow onset and disease progression. We present here the first longitudinal μPET study on this model to characterize in vivo glucose metabolic changes and dopamine-2 receptor binding evolution.
Methods: Male homozygous transgenic (Tg) and wild type (WT) animals (12/12) were scanned at the age of 5, 10 and 15 months with MRI (9T, Biospec, Bruker, Ettlingen) and μPET (Concorde Focus 220) using [18F]FDG (65±9MBq) and [18F]Fallypride (1.6±0.7 nmol/kg) to study brain glucose metabolism and D2-receptor binding. Tracers were injected one hour before a 30-mins PET scan. For [18F]FDG imaging, animals were fastened 24 h prior to the experiment and blood glycaemia was determined. During imaging, animals were anesthetised using isoflurane (2%) and body temperature was kept constant at 37 °C. After image reconstruction, PET images were coregistered to the corresponding high-resolution MRI images using brainvisa/anatomist software. [18F]FDG images were proportionally scaled while [18F]Fallypride images were corrected for non-specific binding using the cerebellum as a reference region. Finally, SPM student t-tests between Tg and WT were applied for each age. Statistical maps (P<0.01, minimum cluster size of 20 pixels) were then displayed on the template images and further used to define volumes of interest on the PET images.
Results: At 5months of age we did not detect any differences in glucose metabolism, though D2-receptor binding was significantly increased in Tg in the ventral striatum and parts of the olfactory bulb. At 10 months, glucose metabolism was increased in Tg in the ventral striatum whilst D2-receptor binding was significantly decreased in the ventral striatum and dramatically decreased in the olfactory bulb. Finally, at 15months of age, brain glucose metabolism was decreased in Tg animals, mostly in striatum and cortex, and D2-receptor binding was decreased in the entire striatum and olfactory bulb.
Conclusions: The first neurodegenerative changes were observed at 10 months of age for D2-receptor binding, and brain glucose metabolism was decreased at 15 months. Surprisingly, this neurodegeneration was preceded by increased D2-receptor binding and brain metabolism starting from 5 months, suggesting the existence of compensatory mechanisms before the appearance of clinical signs. Temporary compensation during the presymptomatic stage of HD has been observed previously (Nguyen et al, 2006). Here, imaging studies revealed a high level of differences between Tg and WT in the ventral striatum starting as early as 5months, a region involved in the control of emotions, which might correspond to emotional disturbance patterns described in this model (Bode et al, 2008). We also found significant changes at the level of the olfactory bulb, which corresponds to the observation that olfactory deficits occur at a very early stage of the disease in presymptomatic HD patients. These hypotheses will be further correlated with behavioral and histological findings on the Huntington rat model.
Acknowledgements: This study was funded in part by the EC-FP6-project DiMI (LSHB-CT-2005-512146), EC-FP6-project EMIL(LSHC-CT-2004-503569), RATstream (LSHM-CT-2007-037846), FWO.
174. Spatial patterns of longitudinal changes in amyloid deposition in nondemented older adults
J. Sojkova1,2, Y. Zhou2, M. Kraut2, J. Brasic2, L. Ferrucci1, D. Wong2 and S. Resnick2
1National Institute on Aging, National Institutes of Health; 2Russell H. Morgan Department of Radiology and Radiological Sciences, Johns Hopkins Medical Institutions, Baltimore, Maryland, USA
Objectives: Cross-sectional studies of nondemented older adults suggest that amyloid deposition in the precuneus represents an early finding in the progression of Alzheimer′s disease. In this longitudinal study, we investigate whether changes in retention of [11C]PIB, a PET tracer that binds to β-amyloid, are detectable on a voxel-wise basis. We hypothesize that the spatial distribution of longitudinal changes in amyloid deposition will localize to regions of increased amyloid deposition observed in cross-sectional studies of nondemented older adults.
Methods: 24 nondemented participants (5 CDR = 0.5; baseline age 79.2(8.1) years) in the Neuroimaging Substudy of the Baltimore Longitudinal Study of Aging had [11C]PIB PET and volumetric MRI at baseline and again after a mean (SD) of 1.5 (0.5) years. Parametric images reflecting the distribution volume ratio (DVR), an index of [11C]PIB specific binding to β-amyloid, were generated from [11C]PIB dynamic PET using a simplified reference tissue model and a linear regression with spatial constraint algorithm (SRTM-LRSC) (Zhou et al, 2003, Neuroimage). The longitudinal changes in gray matter [11C]PIB retention over time were analyzed using SPM5 (P<0.005, spatial extent 100 voxels), adjusting for baseline age.
Results: In nondemented older adults, voxel-wise changes in amyloid deposition can be detected over a 1.5-year follow-up. Significant increases in amyloid deposition are seen in left precuneus (BA 19), right fusiform (BA 19) and right lingual gyri (BA 19); left superior frontal gyrus (BA 8), left orbitofrontal gyrus (BA11), and right medial frontal gyrus (BA 10) extending to right anterior cingulate gyrus.
Conclusions: In nondemented older adults, longitudinal changes in amyloid deposition occur in regions showing increased amyloid deposition in cross-sectional studies, specifically the precuneus region and the prefrontal cortex. Longitudinal changes in amyloid retention are detectable at the voxel level using the DVR images generated from [11C]PIB dynamic images with SRTM-LRSC parametric imaging algorithm.
Support: This research was supported by the Intramural Research Program of the NIH, National Institute on Aging and N01-AG-3-2124.
320. MP4A PET in mild cognitive impairment and Alzheimer's disease
W.-D. Heiss1, C. Haense1, C. Hohmann1, E. Kalbe2, R. Krais1, B. Bauer1, B. Neumaier1 and K. Herholz3
1Max Planck Institute for Neurological Research with Klaus-Joachim-Zuelch-Laboratories of the Max Planck Society and the Faculty of Medicine of the University of Cologne; 2Department of Neurology, University of Cologne, Cologne, Germany; 3Wolfson Molecular Imaging Centre, University of Manchester, Manchester, UK
Background and aims: The radiolabeled acetylcholine analogue [11C]N-methyl-4-piperidyl acetate (MP4A) is a tracer for in vivo imaging of cerebral acetylcholine esterase (AChE) activity with positron emission tomography (PET). MP4A is hydrolyzed specifically by AChE whose activity is a marker of the integrity of cholinergic axons and cholinoceptive neurons in human cerebral cortex.1 Previous preliminary studies indicate a general reduction of AChE activity in Alzheimer dementia (AD) and Mild Cognitive Impairment (MCI).2 We used [11C]-MP4A PET to investigate the functionality of the cholinergic system in healthy controls, MCI patients and AD patients.
Methods: MP4A PET was performed in 21 healthy controls (age 64.6±8.7 years), 20 patients with MCI (age 67.1±7.9 years, MMSE 27.2±1.6) and 15 patients with AD (age 66.1±7.9 years, MMSE 21.1±2.7). Subjects were free of cholinesterase inhibitors. A comprehensive neuropsychological battery was administered. Voxel based activity curves were obtained from Gauss filtered normalized frames and evaluated against putamen as reference. Afterwards parametric images of hydrolysis rate constant k3 were generated by non-linear least squares fits and analysed using standard atlas regions.
Results: All groups were comparable regarding age (Controls vs. MCI, P = 0.884; Controls vs. AD, P = 1.00; MCI vs. AD, P = 1.00). Regions with the highest values of k3 in all three groups were amygdala, entorhinal cortex, hippocampus and precentral gyrus whereas in cuneus, precuneus, inferior parietal lobule and occipital cortex in all groups and additionally in temporal regions in MCI and AD patients the cortical AChE activity was lowest. The mean cortical k3 of controls was 0.084±0.008 min−1 which was significantly higher than in MCI patients (0.076±0.011 min−1, P<0.05, reduction 9.58%) and in AD patients (0.066±0.009 min−1, P<0.01, reduction 20.71%). The reduction of mean cortical k3 in AD compared to MCI (12.32%) was also significant (P<0.05). Reduction of k3 in temporal regions was most pronounced. Hippocampus, amygdala and entorhinal also showed large k3 reductions in patients, but also high variance of values. Lower but still significant reductions were found for frontal regions and the cuneus (Figure).
Parametric images of [11C[-MP4A hydrolysis rates.
Conclusion: These results confirm impairment of the cholinergic system in AD which is even present to a lower degree in MCI. Since [11C]-MP4A PET allows a clear differentiation between groups it might be a valuable tool for early diagnosis of degenerative dementia.
Acknowledgements: This study was funded by the Marga and Walter Boll Foundation and the WDH Foundation.
748. Cerebral haemodynamics changes after 20 mL lumbar drainage in patients with good grade subarachnoid haemorrhage. A pet study
E. Schmidt1, J.F. Al Bucher2, S. Silva Sifontes3, P. Payoux4, A. Luzi3, I. Loubinoux2, C. Cognard5 and F. Chollet2
1Service de Neurochirurgie; 2Service de Neurologie; 3Service d′Anesthésie; 4Service de Médecine Nuclaire, 5Service de Neuroradiologie, Toulouse, France
Background: After subarachnoid haemorrhage (SAH), vasospasm -i.e. delayed ischemia- and hydrocephalus greatly contribute in a context of cerebral energy crisis to unfavourable outcome. The link between hydrocephalus and vasospasm is rather unclear, but it has been shown that after SAH the ventricular size seems correlated with vasospasm1 and a CSF lumbar drainage protocol reduces the risk of vasospasm2. In clinical practice, lumbar punctures to withdraw CSF are often performed to release headaches. However there are no clear evidences of the benefit of this procedure in terms of headache, but also no data on the cerebral haemodynamics consequences of a drop in intracranial pressure (ICP). We hypothesize that in patients with good grade SAH, a 20 ml CSF lumbar drainage significantly reduces the intensity of headache and improves cerebral blood flow measured with O15 PET scan.
Methods: 7 patients were included in our study (mean age 47, sex ratio 4/3) after signed consent. They were all good grade SAH (WFNS 1 to 2); 5 patients had an anterior communicating artery aneurysm and 2 a right middle cerebral artery aneurysm. All aneurysms have been successfully coiled. Every patient was complaining of severe headache. A 2O ml lumbar CSF withdrawal was performed in the PET scanner. ICP and arterial blood pressure (ABP) were continuously monitored throughout the procedure. Regional CBF changes was measured with O15 PET scan before and after CSF drainage. We applied masks (Pickatlas software) on the vascular territory related to the aneurysm. Global differences in CBF were covaried for all voxels and comparisons across conditions were made using t statistics with appropriate linear contrasts and then converted to Z scores. Only regions which exceeded a threshold of Puncorrected<0.01 (Z score >3.2) were considered significant.
Results: A 20 ml CSF drainage yielded a significant:
drop in ICP from 23.5±11.3 to 6.9±3.8 mm Hg,
reduction in visual pain scale (-5), but
no change in ABP (105.2±22.6 vs. 103.8±10.6 mm Hg).
O15 PET study demonstrated in all patients highly significant changes in cerebral haemodynamics before and after the CSF withdrawal (z>3.2). The augmentation of CBF was prominent in the vascular territory of the aneurysm and the reduction of CBF took place in the rest of the vasculature.
Conclusions: In good grade SAH, a 20 ml CSF withdrawal significantly reduces headaches and lowers ICP. This drop in ICP seems to yield a redistribution of the whole CBF at the benefit of the vascular territory of the aneurysm, prone to vasopasm. Our study supports the use of lumbar drainage after SAH not only to reduce headaches but also to improve CBF in potentially jeopardized vascular regions.
References
1.
Boltze. J Cereb Blood F Metab2008.
2.
MesulamMMGeulaC. Acetylcholinesterase-rich neurons of the human cerebral cortex: Cytoarchitecture and ontogenetic patterns of distribution. J Comp Neurol1991;306:193–220.
3.
HerholzKWeisenbachSZuendorfGLenzOSchroederHBauerBKalbeEHeissWD. In vivo study of acetylcholine esterase in basal forebrain, amygdala, and cortex in mild to moderate Alzheimer disease. Neuroimage2004;21:136–143.
