Development of an individualized model to predict the risk of depression after acute ischemic stroke in the elderly
- VernacularTitle:个体化预测老年急性缺血性脑卒中后发生抑郁的风险列线图模型的建立
- Author:
Wenke LIU
1
;
Chao XU
1
;
Shuyan ZHANG
1
Author Information
- Publication Type:Journal Article
- Keywords: Individualized prediction; Acute ischemic stroke; Depression；Risk profile; Elderly patients
- From: Journal of Apoplexy and Nervous Diseases 2022;39(1):59-63
- CountryChina
- Language:Chinese
- Abstract: To explore the main influencing factors of acute ischemic stroke complicated with depression in elderly patients，and to establish an individualized prediction of depression risk histogram model. Methods A total of 330 elderly patients with acute ischemic disease admitted to our hospital from March 2018 to March 2021 were selected as the model group，and 120 patients with depression treated from June 2018 to December 2020 were selected as the validation group. Clinical data of the two groups were collated. Univariate and Logistic multivariate regression models were used to explore the influencing factors of depression in AIS patients. And R software is used to verify the individualized prediction risk histogram model. Results There were significant differences in age，diabetes mellitus，hypertension，chronic kidney disease，serum creatinine，EGFR，BNP，hs-CRP and cardiac function Ⅳ grade between the two groups （P<0.05），and was determined to be related to depression in AIS patients. R software was used to establish a histogram model for predicting the risk of depression in AIS patients. According to the histogram score，the score of cardiac function Ⅳ was 40.30 points，and the score of associated diabetes was 24.00 points. With the increase of age，the decrease of EGFR and the increase of hs-CRP，the corresponding score of the histogram model increased，and the risk of complicated depression also increased. The practicability and feasibility of the model were verified （internal and external）. The C-index value of the model was 0.857 and 0.823 in the model group and the validation group，respectively. When ROC curve was used to analyze the predictive risk efficiency of this model for patients with depression in the model group and the validation group，the AUC of the model group was 0.841，while the AUC of the validation group was 0.817.Conclusion This study showed that the patient’s age，diabetes mellitus，cardiac function grade Ⅳ，EGFR and hs-CRP were independent influencing factors for depression. The individualized predictive histogram model based on this model has good resolution and accuracy，and can be widely used in the prediction of other clinical diseases.
- Full text:2024072722451368126个体化预测老年急性缺血性脑卒中后发生抑郁的风险列线图模型的建立.pdf
