急性缺血性卒中后并发血管性认知障碍的相关因素及列线图模型的构建

卢楠楠; 范慧; 李伟

急性缺血性卒中后并发血管性认知障碍的相关因素及列线图模型的构建

Analysis of the related factors of vascular cognitive impairment after acute ischemic stroke and construction of nomogram model

摘要

摘要:
目的： 探讨急性缺血性卒中（AIS）后并发血管性认知障碍（VCI）的相关因素及构建列线图预测模型，筛选出与VCI发病相关的影响因素，为VCI的早期防治及预测提供数据支持和理论依据。
方法： 选取242例AIS病人为研究对象，根据6个月后随访病人是否并发VCI分为病例组（85例）与对照组（157例）。SPSS 26.0分析得出AIS并发VCI相关因素。R 4.5.2构建列线图预测模型，ROC曲线分析模型的预测价值，Hosmer–Lemeshow拟合优度检校准曲线评价模型的拟合程度，决策曲线分析（DCA）评估临床价值。
结果： 2组在年龄、病程、睡眠障碍、NIHSS评分、mRS评分、低蛋白血症、糖尿病与出院后6个月内接受延续护理等因素上差异均有统计学意义（P ＜ 0.05）。多因素logistic回归分析得出，病程长（OR = 2.112）、NIHSS评分高（OR = 1.254）、mRS评分高（OR = 7.896）、低蛋白血症（OR = 2.735）、糖尿病（OR = 2.526）是AIS并发VCI的影响危险因素（P ＜ 0.05）,而延续护理是AIS并发VCI的保护因素（OR = 0.239，P ＜ 0.05）。列线图模型预测AIS并发VCI的曲线下面积（AUC）为0.906（95%CI：0.862 ~ 0.951）；Hosmer–Lemeshow拟合优度检验显示，C–index = 0.573，χ² = 0.818，P = 0.366，模型拟合度较好；DCA分析显示该模型在较广的阈值概率范围内（0.16～0.59）净收益水平较高。
结论： AIS并发VCI与多种因素相关，构建的列线图预测AIS并发VCI模型的预测价值良好，临床实用性较强。

Abstract:
Objective To explore the related factors of vascular cognitive impairment (VCI) after acute ischemic stroke (AIS) and construct a nomogram prediction model for screening out the influencing factors related to the onset of VCI to provide data support and theoretical basis for the early prevention, treatment and prediction of VCI.
Methods A total of 242 patients with AIS were selected as the research subjects, and divided into the case group (85 cases) and control group (157 cases) based on whether the patients were complicated with VCI after 6 months of follow-up. The related factors of AIS complicated with VCI were analyzed by SPSS 26.0. The nomogram prediction model was constructed using R 4.5.2, the predictive value of model was analyzed by ROC curve, the fitting degree of model was evaluated by Hosmer-Lemeshow goodness-of-fit calibration curve, and the clinical value was assessed by decision curve analysis (DCA).
Results The differences of the age, disease duration, sleep disorders, NIHSS score, mRS Score, hypoproteinemia, diabetes and receiving continuous care within 6 months after discharge were statistically significant between two groups (P ＜ 0.05). The results of multivariate logistic regression analysis showed that the long disease course (OR = 2.112), high NIHSS score (OR = 1.254), high mRS Score (OR = 7.896), hypoproteinemia (OR = 2.735) and diabetes (OR = 2.526) were the influencing risk factors of AIS complicated with VCI (P ＜ 0.05), while the continuous nursing was a protective factor of AIS complicated with VCI (OR = 0.239, P ＜ 0.05). The area under the curve (AUC) predicted by the nomogram model for AIS complicated with VCI was 0.906 (95%CI: 0.862-0.951). The Hosmer-Lemeshow goodness-of-fit test showed that the model had a good fit at the C-index = 0.573, χ² = 0.818, P = 0.366. The results of DCA analysis showed that the model had a relatively high net level within a wide threshold probability range (0.16 to 0.59).
Conclusions AIS complicated with VCI is related to multiple factors. The constructed nomogram model for predicting AIS complicated with VCI has good predictive value and strong clinical practicability.

HTML全文

参考文献(32)

施引文献

资源附件(0)