Scholarlink

Background <p>Spontaneous intracranial hypotension (SIH) is a disabling neurological disorder. Currently, no validated prediction model is available, limiting effective risk stratification and early intervention. The purpose of our study is to analyze the risk factors for subdural hematoma (SDH) in patients with SIH and to construct a clinical prediction model for early identification and clinical decision-making.</p> Methods <p>A total of 575 SIH patients hospitalized at Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, from January 1, 2023 to December 30, 2024, were included. Patients were divided into SDH and non-SDH (NSDH) groups based on the presence of subdural hematoma. Demographic, clinical, and imaging data were collected and analyzed. 80% of the cohort was randomly selected as the training set for model development, and the remaining 20% was used as the test set. In the training set, univariate logistic regression was performed to identify potential factors associated with SDH occurrence, followed by Lasso regression for variable selection. Selected variables were incorporated into a multivariate logistic regression model using stepwise regression. Model validation was conducted using 500 bootstrap resampling iterations. Model performance was assessed by receiver operating characteristic (ROC) curves for discrimination, calibration curves for calibration, and decision curve analysis (DCA) for clinical utility.</p> Results <p>Among 575 patients, 101 (17.6%) developed SDH (14 patients with unilateral SDH and 87 patients with bilateral SDH). Multivariate analysis identified history of prior treatment for intracranial hypotension, pre-admission subdural effusion and presence of dural enhancement as predictors of SDH occurrence, and these variables were also independent risk factors (<i>P</i> &lt; 0.05). The model demonstrated strong discrimination, with AUCs of 0.839 (training set), 0.837 (test set), and 0.790 (overall). Sensitivity and specificity were 0.438/0.965 (training), 0.421/0.965 (test), and 0.434/0.963 (overall). Calibration curves showed minimal deviation (mean absolute error &lt; 0.05), and DCA indicated favorable clinical utility when the high-risk threshold was below 0.73.</p> Conclusion <p>History of prior treatment for intracranial hypotension, pre-admission subdural effusion and presence of dural enhancement are independent risk factors for SDH in SIH patients. The prediction model shows strong discrimination, calibration, and clinical utility, supporting early risk assessment and personalized clinical management.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction for occurrence of subdural hematoma following spontaneous intracranial hypotension: a large cohort of 575 adults in a single center from 2023 to 2024

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月26日

查看全文
  • Kun Wang,
  • Keng Chen,
  • Zundao Shan,
  • Yinxin Zhu,
  • Bo Li,
  • Jingyang Hong,
  • Xinwei Li,
  • Feifang He,
  • Yirong Wang

摘要

Background

Spontaneous intracranial hypotension (SIH) is a disabling neurological disorder. Currently, no validated prediction model is available, limiting effective risk stratification and early intervention. The purpose of our study is to analyze the risk factors for subdural hematoma (SDH) in patients with SIH and to construct a clinical prediction model for early identification and clinical decision-making.

Methods

A total of 575 SIH patients hospitalized at Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, from January 1, 2023 to December 30, 2024, were included. Patients were divided into SDH and non-SDH (NSDH) groups based on the presence of subdural hematoma. Demographic, clinical, and imaging data were collected and analyzed. 80% of the cohort was randomly selected as the training set for model development, and the remaining 20% was used as the test set. In the training set, univariate logistic regression was performed to identify potential factors associated with SDH occurrence, followed by Lasso regression for variable selection. Selected variables were incorporated into a multivariate logistic regression model using stepwise regression. Model validation was conducted using 500 bootstrap resampling iterations. Model performance was assessed by receiver operating characteristic (ROC) curves for discrimination, calibration curves for calibration, and decision curve analysis (DCA) for clinical utility.

Results

Among 575 patients, 101 (17.6%) developed SDH (14 patients with unilateral SDH and 87 patients with bilateral SDH). Multivariate analysis identified history of prior treatment for intracranial hypotension, pre-admission subdural effusion and presence of dural enhancement as predictors of SDH occurrence, and these variables were also independent risk factors (P < 0.05). The model demonstrated strong discrimination, with AUCs of 0.839 (training set), 0.837 (test set), and 0.790 (overall). Sensitivity and specificity were 0.438/0.965 (training), 0.421/0.965 (test), and 0.434/0.963 (overall). Calibration curves showed minimal deviation (mean absolute error < 0.05), and DCA indicated favorable clinical utility when the high-risk threshold was below 0.73.

Conclusion

History of prior treatment for intracranial hypotension, pre-admission subdural effusion and presence of dural enhancement are independent risk factors for SDH in SIH patients. The prediction model shows strong discrimination, calibration, and clinical utility, supporting early risk assessment and personalized clinical management.

Graphical Abstract