Grignon Y, Duyckaerts C, Bennecib M, Hauw J J
Laboratoire de Neuropathologie R. Escourolle, INSERM U 360, Hôpital de La Salpêtrière, Paris, France.
Acta Neuropathol. 1998 Apr;95(4):395-406. doi: 10.1007/s004010050816.
The intellectual status of 28 women of over 75 years of age had been prospectively assessed by the Blessed test score. It ranged from nearly normal to deeply altered by dementia. After autopsy, the supramarginal gyrus was marked at the surface of the brain. Sections, 1 cm thick, were cut with a specially devised macrotome. The volume of the parietal lobe was measured by a point counting method, using Cavalieri principle. A sample from the supramarginal gyrus was taken from the previously marked area and the shrinkage due to the histological procedures was measured (it averaged 12%). More than 500 nucleolated neuronal profiles per case were mapped with a semi-automatic system. Density maps of the neuronal profiles were drawn and mean density was calculated using Dirichlet tessellation. The thickness of the cortical ribbon was standardized on the maps. The density of the neurons per unit volume was calculated, taking into account the section thickness measured for each sample with a length gauge fastened to the Z axis of the microscope. Statistical correlations were sought between the mean and laminar densities of the neurons on one hand, and Blessed test score, the densities of neurofibrillary tangles (NFT) and of senile plaques profiles, on the other hand. Finally, the total number of neurons present in the parietal lobe was estimated in each case. Neuronal loss appeared to be linked with the density of the NFT (r = -0.52; P < 0.004). The correlation was mainly due to a severe drop in neuronal number observed in the cases with more than 5 NFT/mm2. An average difference of 98 x 10(6) neurons per parietal lobe was found between the cases with less than 5 NFT/mm2 and those with more. The neuronal loss predominated in layers II and III (upper part). A multivariate analysis showed that the intellectual status was better correlated with the density of the tangles than with the neuronal loss.
通过Blessed测试评分对28名75岁以上女性的智力状况进行了前瞻性评估。其范围从接近正常到因痴呆而严重改变。尸检后，缘上回在脑表面被标记出来。用专门设计的大切片机切出1厘米厚的切片。使用卡瓦列里原理，通过点计数法测量顶叶的体积。从先前标记的区域获取缘上回的样本，并测量由于组织学程序导致的收缩（平均为12%）。每例用半自动系统绘制500多个有核仁的神经元轮廓图。绘制神经元轮廓的密度图，并使用狄利克雷镶嵌法计算平均密度。在图上对皮质带的厚度进行标准化。计算每单位体积的神经元密度，同时考虑到用固定在显微镜Z轴上的长度测量仪为每个样本测量的切片厚度。一方面寻找神经元的平均密度和分层密度与Blessed测试评分、神经原纤维缠结（NFT）密度和老年斑轮廓密度之间的统计相关性。最后，估计每例顶叶中存在的神经元总数。神经元丢失似乎与NFT密度相关（r = -0.52；P < 0.004）。这种相关性主要是由于在NFT/mm2超过5的病例中观察到神经元数量严重下降。在NFT/mm2小于5的病例和大于5的病例之间，每顶叶平均相差98×10⁶个神经元。神经元丢失在II层和III层（上部）占主导。多变量分析表明，智力状况与缠结密度的相关性比与神经元丢失的相关性更好。
