Liu Yu, Xu Xiaoyu, Zhou Yanlong, Du Bo, Cheng Yanbo, Feng Yu
Department of Neurology, Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China.
Xuzhou Cancer Hospital, Xuzhou, Jiangsu, China.
Eur J Med Res. 2025 Sep 26;30(1):880. doi: 10.1186/s40001-025-03171-5.
Progressive ischemic stroke (PIS) is a severe adverse cerebrovascular event that can occur shortly after an acute ischemic stroke (AIS).The clinical factors that predict PIS remain poorly understood. This study aims to develop a nomogram for predicting PIS following AIS.
This study retrospectively analyzed clinical data from patients diagnosed with AIS at the Affiliated Hospital of Xuzhou Medical University between 2018 and 2021 who subsequently developed PIS. Risk factors associated with PIS were identified using univariate logistic regression, followed by stepwise multivariate logistic regression to construct a predictive model. The resulting model was then transformed into a nomogram, providing neurologists with a clinically practical tool for rapidly assessing the risk of PIS following AIS.
Among 580 patients with AIS, 14.31% developed progressive stroke within 14 days. The data set was split into a training set (70%) and a test set (30%). Univariate analysis identified ten indicators associated with progressive stroke, and multivariate logistic regression in the training set revealed four independent risk factors. A nomogram was developed using R software (version 4.3.2) to predict progressive stroke risk. The Model demonstrated strong performance, with ROC curve AUCs of 0.849 (training set) and 0.829 (test set). The DeLong test showed no significant difference between the data sets (P > 0.05), confirming robustness. The overall AUC was 0.974, and the Hosmer-Lemeshow test indicated good calibration (P = 0.887). The calibration plot's mean absolute error was 0.012, and decision curve analysis confirmed the nomogram's clinical utility. Internal validation showed close agreement between the training and test sets.
The nomogram model appears to enhance the prediction of progressive stroke risk in patients with AIS, potentially supporting neurologists in making more informed and timely clinical decisions.
进展性缺血性卒中（PIS）是一种严重的不良脑血管事件，可在急性缺血性卒中（AIS）后不久发生。预测PIS的临床因素仍知之甚少。本研究旨在开发一种用于预测AIS后PIS的列线图。
本研究回顾性分析了2018年至2021年在徐州医科大学附属医院诊断为AIS且随后发生PIS的患者的临床数据。使用单因素逻辑回归确定与PIS相关的危险因素，随后进行逐步多因素逻辑回归以构建预测模型。然后将所得模型转换为列线图，为神经科医生提供一种临床实用工具，用于快速评估AIS后PIS的风险。
在580例AIS患者中，14.31%在14天内发生了进展性卒中。数据集被分为训练集（70%）和测试集（30%）。单因素分析确定了10个与进展性卒中相关的指标，训练集中的多因素逻辑回归显示了4个独立危险因素。使用R软件（版本4.3.2）开发了列线图以预测进展性卒中风险。该模型表现出强大的性能，训练集的ROC曲线AUC为0.849，测试集为0.829。DeLong检验显示数据集之间无显著差异（P>0.05），证实了稳健性。总体AUC为0.974，Hosmer-Lemeshow检验表明校准良好（P=0.887）。校准图的平均绝对误差为0.012，决策曲线分析证实了列线图的临床实用性。内部验证显示训练集和测试集之间具有密切一致性。
列线图模型似乎增强了对AIS患者进展性卒中风险的预测，可能有助于神经科医生做出更明智、更及时的临床决策。
