目的:蛛网膜下腔出血(SAH)与高死亡率和永久性残疾有关。目前,目前尚无明确的临床工具来预测SAH患者的预后。本研究旨在开发和评估预测列线图模型,用于估计非创伤性或创伤后SAH患者的28天死亡风险。
方法:搜索MIMIC-III数据库以根据ICD-9代码选择SAH患者。患者分为非创伤性和创伤后SAH组。使用LASSO回归分析,我们确定了与28日死亡率相关的独立危险因素,并将其纳入列线图模型.通过计算各种指标来评估每个列线图的性能,包括曲线下面积(AUC),净重新分类改进(NRI),综合歧视改进(IDI),和决策曲线分析(DCA)。
结果:该研究包括999例SAH患者,非创伤组631人,创伤后组368人。Logistic回归分析显示非创伤性SAH患者28天死亡的关键独立危险因素。包括性别,年龄,葡萄糖,血小板,钠,BUN,WBC,PTT,尿量,血氧饱和度、心率和年龄,葡萄糖,PTT,尿量,创伤后SAH患者的体温。预后列线图优于常用的SAPSII和APSIII系统,如优越的AUC所示,NRI,IDI,和DCA结果。
结论:该研究确定了与28天死亡风险相关的独立危险因素,并为非创伤性和创伤后SAH患者建立了预测列线图模型。列线图有望指导SAH患者的预后改善策略。
OBJECTIVE: Subarachnoid hemorrhage (SAH) is associated with high rates of mortality and permanent disability. At present, there are few definite clinical tools to predict prognosis in SAH patients. The current study aims to develop and assess a predictive nomogram model for estimating the 28-day mortality risk in both non-traumatic or post-traumatic SAH patients.
METHODS: The MIMIC-III database was searched to select patients with SAH based on ICD-9 codes. Patients were separated into non-traumatic and post-traumatic SAH groups. Using LASSO regression analysis, we identified independent risk factors associated with 28-day mortality and incorporated them into nomogram models. The performance of each nomogram was assessed by calculating various metrics, including the area under the curve (AUC), net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA).
RESULTS: The study included 999 patients with SAH, with 631 in the non-traumatic group and 368 in the post-traumatic group. Logistic regression analysis revealed critical independent risk factors for 28-day mortality in non-traumatic SAH patients, including gender, age, glucose, platelet, sodium, BUN, WBC, PTT, urine output, SpO2, and heart rate and age, glucose, PTT, urine output, and body temperature for post-traumatic SAH patients. The prognostic nomograms outperformed the commonly used SAPSII and APSIII systems, as evidenced by superior AUC, NRI, IDI, and DCA results.
CONCLUSIONS: The study identified independent risk factors associated with the 28-day mortality risk and developed predictive nomogram models for both non-traumatic and post-traumatic SAH patients. The nomogram holds promise in guiding prognosis improvement strategies for patients with SAH.