A Clinical Study On MRI Volumetric Measurement Of Hippocampus And Entorhinal Cortex In Mild Cognitive Impairment

Obiective: To investigate if the volume of hippocampus , entorhinal cortex (ERC) reduce in mild cognitive impairment, to evaluate the value of using hippocampus and ERC to classify MCI and NC, to investigate the factors to influence the volume of hippocampus and ERC. Method: According to the clinical diagnostic criteria of MCI which made by Petersen and his colleagues, subjects were divided into two groups: MCI(21 subjects) and normal cognitive(18 subjects). All admitted subjects were evaluated with MMSE GDS CMS ADL. Quantitative volumes of the hippocampus and ERC were obtained by magnetic resonance imaging, the volumes of the hippocapums and ERC were normalized to individual total intracranial volume. General physical examinations, neurological examinations, laboratory studies and brain imaging (MRI) were done to all subjects. Statistical analysis was guided by a statistician and was performed using SPSS (11.5). Result:1. Total ERC volume of patients with MCI was2. 661 (SD 0. 173)cm~3, total hippocampal volume of patients with MCI was 6. 186(SD 0.740)cm~3 . The ERC and hippocampal volume was significantly reduced in MCI compared with normal cognitive(13% reduction of ERC and 10% reduction of hippocampal). 2. There was no significant group by side interaction neither for ERC volume changes nor for hippocampal volume changes. There was a significant positive correlation between hippocampal and ERC volumes(P<0.001), and hippocampus and Clinical Memory Scale(CMS) score(P<0. 001),there was a significant negative correlation between hippocampus and systolic blood pressure(P=0.007), hippocampus and diastolic blood pressure(P=0.039), and hippocampus and 2 hours postprandial blood glucose (P=0. 007). There was a significant negative correlation between ERC and systolic blood pressure (P=0. 022), and ERC and diastolic blood pressure(P=0. 024). 3. To compare the ERC and hippocampus volumes for their power to distinguish NC and MCI, uncondition logistic regression analysis was used to predict group memberships either with the ERC or hippocampus as independent variables. MCI and NC were classified with 66.7% specificity, and 76.2% sensitivity, and an overall classification of 71.8% using hippocampus alone, MCI and NC were classified with 88. 9% specificity, and 90. 5% sensitivity, and an overall classification of 89. 7% using ERC alone, and classification with the ERC was better than with hippocampus . 4. To investigate the factors that affect the volume of hippocampus and ERC, multivariate was used either with the ERC or hippocampus as independent variables. 2 hours postprandial blood glucose, systolic blood pressure, total cholesterol, diastolic blood pressure, CMS scores and MMSE scores entered the model with the hippocampus as independent variables. Only CMS scores entered the model with ERC as independent variables. Conclusions: 1. The ERC and hippocampal volume was significantly reduced in MCI compared with normal cognitive. 2. It is meaning that MCI and NC were classified with the volume of ERC and hippocampus obtained by MRI, classification with ERC was better than with hippocampus. ERC and hippocampus atrophy is valuable for the diagnosis and prognostic evaluation of MCI. 3. The higher blood glucose, blood pressure and blood lipid are, the smaller hippocampus is and the more seriously cognitive impairs. Effective control of blood glocuse, blood pressure and blood lipid is important, to prevent the occurrence and development of AD. 4. The volume of ERC and hippocampus obtained by MRI can not be a perfect indicator to diagnose MCI. It is difficult to distinguish MCI individual and NC individual exactly using ERC and hippocampus, but the volume of ERC and hippocampus obtained by MRI has important forecast value for MCI converting to AD.