| 周晓飞,彭金燕,彭一鹏.动脉瘤术后颅内细菌感染的影响因素分析及预测模型构建.[J].中南医学科学杂志.,2023,(6):904-907. |
| 动脉瘤术后颅内细菌感染的影响因素分析及预测模型构建 |
| Analysis of influencing factors of intracranial bacterial infection after aneurysm surgery and construction of predictive model |
| 投稿时间：2023-04-11 修订日期：2023-07-08 |
| DOI：10.15972/j.cnki.43-1509/r.2023.06.025 |
| 中文关键词: 颅内动脉瘤 细菌感染 影响因素 预测模型 [ |
| 英文关键词:intracranial aneurysm bacterial infection influencing factors prediction model |
| 基金项目:武汉市卫生健康委员会(WX21D43) |
|
| 摘要点击次数: 195 |
| 全文下载次数: 124 |
| 中文摘要: |
| 目的分析颅内动脉瘤术后颅内细菌感染的影响因素,并构建其预测模型。 方法回顾性分析收治的332例颅内动脉瘤患者的资料,分析患者颅内细菌感染的影响因素,并构建列线图预测模型,患者的资料利用Bootstrap法验证预测模型的区分度,采用ROC曲线分析预测模型的效能。 结果患者术后颅内细菌感染发生率为24.40%；合并糖尿病、开放术式、手术时间、术中颅内动脉瘤破裂、术后脑脊液漏、术后留置引流时间、气管切开、应用糖皮质激素均是颅内细菌感染的危险因素(P＜0.05),基于此建立风险预测列线图模型,且经验证风险预测模型的C-index值为0.851,校正曲线与标准曲线拟合度良好。列线图模型预测颅内细菌感染的曲线下面积为0.866(P＜0.001),灵敏度为90.12%,特异度为93.63%。 结论合并糖尿病、开放术式、手术时间、术中颅内动脉瘤破裂、术后脑脊液漏、术后留置引流时间、气管切开、应用糖皮质激素均是颅内细菌感染的危险因素,基于危险因素分析可构建颅内动脉瘤术后颅内细菌感染的风险预测模型。 |
| 英文摘要: |
| AimTo analyze the influencing factors of intracranial bacterial infection after intracranial aneurysm surgery, and build a predictive model. MethodsRetrospective analysis was made on the data of 332 patients with intracranial aneurysms admitted to the hospital. The influencing factors of intracranial bacterial infection for the patients were analyzed, and a nomogram prediction model was established. The data of patients were used to verify the discrimination of the prediction model with Bootstrap method. The ROC curve was used to analyze and predict the effectiveness of the model. ResultsThe incidence rate of intracranial bacterial infection was 24.40%. Complication with diabetes, open operation, operation time, intracranial aneurysm rupture during operation, cerebrospinal fluid leakage after operation, retention and drainage time after operation, tracheotomy, and glucocorticoid application were all risk factors for intracranial bacterial infection (P＜0.05). Based on this, the risk prediction nomograph model was established. The C-index value of the verified risk prediction model was 0.851, and the fitting degree between the calibration curve and the standard curve was good. The area under curve of nomogram model for predicting intracranial bacterial infection was 0.866 (P＜0.001), with sensitivity of 90.12% and specificity of 93.63%, respectively. ConclusionComplication with diabetes, open operation, operation time, intracranial aneurysm rupture during operation, cerebrospinal fluid leakage after operation, retention and drainage time after operation, tracheotomy, and glucocorticoid application were all risk factors for intracranial bacterial infection. Based on risk factor analysis, a risk prediction model for intracranial bacterial infection after intracranial aneurysm surgery can be constructed. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|