Describing the trajectories of age-related change for different brain structures has been of interest in many recent studies. However, our knowledge regarding these trajectories and their associations is still limited due to small sample sizes and low numbers of repeated measures. For the present study, we used a large longitudinal dataset (four measurements over 4 years) comprising anatomical data from a sample of healthy older adults (N = 231 at baseline). This dataset enables us to gain new insights about volumetric cortical and subcortical changes and their associations in the context of healthy aging. Brain structure volumes were derived from T1-weighted MRI scans using FreeSurfer segmentation tools. Brain structure trajectories were fitted using mixed models and latent growth curve models to gain information about the mean extent and variability of decline trajectories for different brain structures as well as the associations between individual trajectories. On the group level, our analyses indicate similar linear changes for frontal and parietal brain regions, while medial temporal regions showed an accelerated decline with advancing age. Regarding subcortical regions, some structures showed strong declines (e.g., hippocampus), others showed little decline (e.g., pallidum). Our data provide little evidence for sex differences regarding the aforementioned trajectories. Between-person variability of the person-specific slopes (random slopes) was largest in subcortical and medial temporal brain structures. When looking at the associations between the random slopes from each brain structure, we found that the decline is largely homogenous across the majority of cortical brain structures. In subcortical and medial temporal brain structures, however, more heterogeneity of the decline was observed, meaning that the extent of the decline in one structure is less predictive of the decline in another structure. Taken together, our study contributes to enhancing our understanding of structural brain aging by demonstrating (1) that average volumetric change differs across the brain and (2) that there are regional differences with respect to between-person variability in the slopes. Moreover, our data suggest (3) that random slopes are highly correlated across large parts of the cerebral cortex but (4) that some brain regions (i.e., medial temporal regions) deviate from this homogeneity.

中文翻译：

---

衰老过程中皮质和皮质下区域的变异性下降：纵向研究

描述不同大脑结构与年龄相关变化的轨迹在许多最近的研究中引起了人们的兴趣。然而，由于样本量小和重复测量次数少，我们对这些轨迹及其关联的了解仍然有限。在本研究中，我们使用了一个大型纵向数据集（4 年四次测量），其中包含来自健康老年人样本（基线 N = 231）的解剖数据。该数据集使我们能够在健康老龄化的背景下获得关于体积皮层和皮层下变化及其关联的新见解。脑结构体积来自使用 FreeSurfer 分割工具的 T1 加权 MRI 扫描。使用混合模型和潜在生长曲线模型拟合大脑结构轨迹，以获得关于不同大脑结构衰退轨迹的平均程度和可变性以及个体轨迹之间关联的信息。在组水平上，我们的分析表明额叶和顶叶脑区的线性变化相似，而内侧颞区则随着年龄的增长而加速下降。关于皮层下区域，一些结构显示出强烈的下降（例如海马体），其他的显示出很少的下降（例如苍白球）。我们的数据几乎没有提供关于上述轨迹的性别差异的证据。人特定斜率（随机斜率）的人际变异在皮层下和内侧颞叶脑结构中最大。当查看每个大脑结构的随机斜率之间的关联时，我们发现大多数皮质大脑结构的下降在很大程度上是同质的。然而，在皮质下和内侧颞叶脑结构中，观察到更多的衰退异质性，这意味着一个结构的衰退程度对另一种结构的衰退的预测性较低。综上所述，我们的研究通过证明 (1) 大脑中的平均体积变化不同以及 (2) 斜坡的人与人之间的变异性存在区域差异，有助于增强我们对结构性大脑衰老的理解。此外，我们的数据表明（3）随机斜率在大脑皮层的大部分区域高度相关，但（4）某些大脑区域（即，