Construction of a postoperative pulmonary infection risk prediction score model for craniocerebral tumor patients based on meta-analysis
基于Meta分析构建颅脑肿瘤患者术后肺部感染风险预测评分模型

摘要

目的 基于Meta分析及外部验证构建颅脑肿瘤患者术后肺部感染风险预测模型评分模型，为早期识别颅脑肿瘤患者术后肺部感染高危人群提供借鉴。方法 对颅脑肿瘤患者术后肺部感染的风险因素进行Meta分析，以各风险因素OR值的自然对数为模型的β系数，以术后肺部感染发生率与未发生率比值的自然对数为模型的α系数，建立预测模型。结果 颅脑肿瘤患者术后肺部感染风险预测评分模型验证结果显示，得分为0~201分，评分＞41.5分为高危人群，约登指数为0.586，敏感度为0.794，特异度为0.792。ROC曲线（Receiver Operating Characteristic curve）下面积为0.807，95%CI（0.733-0.881）。结论 基于Meta分析的颅脑肿瘤患者术后肺部感染风险预测评分模型有很好的预测性能及实用价值，可作为临床医护人员发现颅脑肿瘤患者术后肺部感染实施预防性护理措施的依据。

Abstract

Objective To construct a predictive scoring model for postoperative pulmonary infection risk in craniocerebral tumor patients based on meta-analysis and external validation, providing a reference for early identification of high-risk groups.
Methods A meta-analysis was conducted on risk factors for postoperative pulmonary infection in craniocerebral tumor patients. The natural logarithm of the odds ratio (OR) for each risk factor was used as the β coefficient of the model, while the natural logarithm of the ratio of infection to non-infection rates served as the α coefficient, establishing the predictive model.
Results Validation results of the postoperative pulmonary infection risk prediction scoring model for craniocerebral tumor patients showed scores ranging from 0 to 201, with scores >41.5 indicating high-risk groups. The Youden index was 0.586, sensitivity was 0.794, and specificity was 0.792. The area under the ROC curve was 0.807 (95% CI: 0.733-0.881).
Conclusion   The risk prediction scoring model for postoperative pulmonary infection in patients with craniocerebral tumor based on Meta-analysis has good predictive performance and practical value, which can be used as a basis for clinical medical staff to detect postoperative pulmonary infection in patients with craniocerebral tumor and implement preventive nursing measures.