基于脑CT的胼胝体压部角鉴别正常压力脑积水与阿尔茨海默病及帕金森病的诊断效能与可靠性研究
本回顾性研究纳入325例受试者（87例NPH、71例阿尔茨海默病、66例帕金森病、101例健康对照），在脑CT上测量Evans指数、胼胝体角及胼胝体压部角。胼胝体压部角对NPH的诊断准确度最高（AUC=0.999，阈值49.5°，敏感度0.996，特异度0.989），且观察者内与观察者间重复性最佳，可作为CT辅助鉴别NPH与神经退行性疾病的可靠影像标志物。
要点
- 胼胝体压部角鉴别NPH诊断效能极高（AUC=0.999，阈值49.5°），可作为CT辅助参数。
- 胼胝体角诊断效能亦强（AUC=0.993，阈值87.5°）；Evans指数准确度高但特异度较低。
- 在阿尔茨海默病与帕金森病组中，上述测量指标的鉴别能力有限。
核心结论
- 胼胝体压部角可作为基于CT的可靠标志物，帮助鉴别正常压力脑积水与神经退行性疾病，作为临床评估的辅助影像参数。
研究设计与样本
- 回顾性研究，纳入325例受试者：87例NPH、71例阿尔茨海默病、66例帕金森病、101例健康对照。
- 在脑CT图像上测量Evans指数、胼胝体角与胼胝体压部角，由两名放射科医师评估观察者内与观察者间重复性。
- 连续变量采用ANOVA分析，分类变量采用卡方检验，以ROC分析确定诊断效能。
关键影像表现/征象
- NPH组的Evans指数、胼胝体角与胼胝体压部角与其他组相比差异均有统计学意义（p < 0.001）。
诊断标准与关键数值
- 胼胝体压部角诊断NPH：AUC=0.999，阈值49.5°，敏感度0.996，特异度0.989，阳性似然比86.63，阴性似然比0.004。
- 胼胝体角诊断NPH：AUC=0.993，阈值87.5°。
- Evans指数诊断NPH：AUC=0.928，准确度高但特异度较低。
- 在阿尔茨海默病与帕金森病组中，上述测量指标的鉴别能力有限。
- 三种参数的观察者内与观察者间一致性均高，其中胼胝体压部角的重复性最佳。
临床应用与阅片要点
- 在脑CT阅片鉴别NPH与神经退行性疾病时，可优先测量胼胝体压部角作为辅助诊断参数。
- 胼胝体压部角测量重复性最佳，便于在常规CT报告流程中标准化应用。
- Evans指数虽准确度高但特异度较低，单独使用需注意假阳性。
局限性
- 本研究为回顾性设计。
以上解读依据论文摘要（全文未开放获取或许可不允许翻译转载）。
英文原文（摘要）· 点击展开
BACKGROUND: Although MRI and diffusion tensor imaging (DTI) based studies have evaluated the Splenial Angle (SA) in distinguishing normal pressure hydrocephalus (NPH), its reproducibility and diagnostic utility on CT remain untested.
OBJECTIVE: To assess the diagnostic performance and intra/inter-observer repeatability of CT-derived SA measurements for differentiating NPH from Alzheimer's and Parkinson's diseases.
MATERIALS AND METHODS: This retrospective study included 325 individuals: 87 with NPH, 71 with Alzheimer's disease, 66 with Parkinson's disease, and 101 healthy controls. Evans Index (EI), Callosal Angle (CA), and SA were measured on brain CT images. Two radiologists assessed intra- and inter-observer repeatability. Continuous variables were analyzed using ANOVA, categorical variables using Chi-square tests, and Receiver Operating Characteristic (ROC) analysis determined diagnostic performance.
RESULTS: EI, CA, and SA differed significantly between the NPH group and the other groups (p < 0.001). The SA showed the highest diagnostic accuracy for NPH (AUC = 0.999; cut-off = 49.5°; sensitivity = 0.996; specificity = 0.989; LR + = 86.63; LR- = 0.004). The CA also demonstrated strong performance (AUC = 0.993; cut-off = 87.5°). The EI achieved high accuracy (AUC = 0.928) but was less specific. In the Alzheimer's and Parkinson's groups, these measurements had limited discriminatory ability. Intra- and inter-observer agreement was high for all three parameters, with the SA showing the greatest repeatability.
CONCLUSION: The SA can be used as a reliable CT-based marker to help distinguish NPH from neurodegenerative diseases, serving as a supportive imaging parameter in clinical assessment.
引用
- Kaya M, Konukoglu O, Genç H, et al. Differentiation of normal pressure hydrocephalus from alzheimer's and Parkinson's Diseases using the splenial angle measured on brain CT: diagnostic performance and reliability study. Neuroradiology. 2026. PMID: 42105113. doi:10.1007/s00234-026-04018-4
供执业医师工作参考，不替代临床判断；数值与分级以指南原文为准。