| Background and purpose Diabetes mellitus is a well-recognized risk factor for acute stroke. It has been well documented that diabetes is associated with an increased risk of ischemic stroke with relative risk ranging from1.8to6. Diabetes also doubles the risk of stroke recurrence. And stroke outcomes are significantly worse among patients with diabetes compared with patients without diabetes.Traditional neuroimaging techniques, such as magnetic resonance imaging (MRI) and computed tomography, have provided data on structural changes such as location, size in the brain, whereas little information could be obtained about regional neurobiochemistry the severity of the illness. Magnetic resonance spectroscopy(MRS) combines functional examination and imaging examination. In vivo it not only can discriminate normal and pathological tissues non-invasively but also may provide neurochemical information on subtle and apparent brain parenchymal changes, giving biological process of human body under normal and under the states of various kinds of diseases. Therefore, it is increasingly being used to diagnose and manage various cerebral diseases. However, few studies investigated brain metabolite changes using MRS technique in patients with diabetes mellitus and cerebral infarction. This study was designed to contribute to a better understanding of cerebral metabolism on MRS in patients with type2diabetes mellitus and cerebral infarction to guide the clinical diagnosis and treatment.Subjects and Methods24patients with acute cerebral infarction and type2diabetes (DMCI group),23patients with acute cerebral infarction (CI group) and26healthy controls (HC group) were scanned on a1.5-T MRI/MRS imager. Voxels were placed to the basal ganglion for the detection of the levels of N-acetylaspartate (NAA), Choline (Cho) and lactate (Lac). The ratios of NAA/Cr, Cho/Cr and Lac/Cr were calculated. Glucose levels were measured via hexokinase method. HbAlc was measured by high-performance liquid chromatography method. The MRI examination consisted of routine imaging and Multi-voxel spectroscopy. MRI was performed on a1.5-T system (Siemens, Germany). T1-weighted images (TR:408ms, TE:11ms) were obtained in the axial and sagittal planes (with5-mm-thick slices). T2-weighted images (TR:4650ms, TE106ms) were obtained in the axial and coronal planes. Spectroscopy was performed in all patients using a point-resolved spectroscopy sequence (PRESS; TR:2000ms, TE:135ms) with128averages. Voxels were placed in the lesion of infarction and contralateral areas (10×10×10mm) in the basal ganglion of the brain tissue. Prior to MRS, shimming was performed to optimized field homogeneity, and water suppression was optimized with automated routines provided by a chemical-shift selective saturation pulse. All data postprocessing were performed with software provided by the manufacturer. Resonances peaks were assigned as follows:NAA2.02ppm; creatine3.04ppm; cho3.24ppm; Lac1.33ppm. Peak area metabolite ratios (NAA/Cr, Cho/Cr and Lac/Cr) were calculated.Results1、clinical characteristics of the DMCI group, CI group and HC group. No significant differences in gender and age distribution were observed. The glucose levels in the DMCI group were higher than those in the CI and HC group.2、Comparisons of the NAA/Cr, Cho/Cr and Lac/Cr ratios in the basal ganglion among DMCI group, CI group and-HC group. Cerebral NAA/Cr ratios in the infracted side of CI group and DMCI group were lower than in the HC group. There was a significant decrease in the level of NAA/Cr in the infarcted side of the DMCI group as compared to CI group. NAA/Cr ratios in the contralateral side of the DMCI group tended to be lower than those in the HC group. Lac/Cr ratios in the DMCI group and CI group were increased in the infracted side than those in the HC group, and the levels of Lac/Cr in the infracted side of DMCI group were higher than those in the CI group.3、Comparisons of the NAA/Cr, Cho/Cr and Lac/Cr ratios between infarcted side and contralateral side of DMCI group. Cerebral NAA/Cr ratios in the infarcted side were lower than those in the contralateral side of the DMCI group. There was no significant difference in Cho/Cr ratios between infarcted side and contralateral side. There was no Lac peak in the contralateral side of the DMCI group.4、Correlation between cerebral parameters and clinical characteristics in the DMCI group. The duration of diabetes, gender and age were not correlated with the cerebral parameters in the DMCI. HbAlc and FBG levels were inversely correlated with NAA/Cr ratios in the normal side and infracted side.(r=-0.498,p=0.026and r=-0.616,p=0.011). No significant correlation was observed in DMCI group between Lac/Cr ratios and clinical characteristics.Conclusion1. Type2diabetes may be associated with more serious neuron damage in cerebral infarction, which was in accordance with the levels of FBG and HbAlc.2. Type2diabetes mellitus may cause the cerebral damage before the occurrence of infarction.3.’H-MRS is promising to be useful in exploring the pathologic physiology research of the patients with diabetes mellitus and acute cerebral infarction. |
PDF Full Text Request