慢性硬膜下血肿钻孔引流术后近期预后的影响因素及Nomogram预测模型的构建

Influencing factors of short-term prognosis after drilling and drainage surgery for chronic subdural hematoma and establishing a Nomogram prediction mode

目的分析慢性硬膜下血肿钻孔引流术后近期预后的影响因素并构建Nomogram预测模型。方法回顾性选取2015至2022年上海交通大学医学院附属瑞金医院神经外科收治的慢性硬膜下血肿患者336例为研究对象。根据术后1个月内随访时是否出现预后不良,将患者分为预后良好组(n=241)与预后不良组(n=95)。分析对比与慢性硬膜下血肿术后复发可能相关的13项因素。将差异有统计学意义的变量纳入多因素Logistic回归分析并构建风险预测模型,采用Rstudio建立预测慢性硬膜下血肿钻孔引流术后近期预后不良的Nomogram模型,进行可视化展示。绘制受试者工作特征(ROC)曲线对慢性硬膜下血肿钻孔引流术后近期预后不良风险预测模型进行预测效能评价。结果共收集慢性硬膜下血肿钻孔引流术患者336例,其中发生预后不良的患者95例,占比为28.27%。多因素Logistic回归分析结果显示,年龄、脑萎缩、脑复张、术前血肿宽度、术前中线移位、术后硬膜下间隙、出院时中线偏移程度是慢性硬膜下血肿钻孔引流术后近期预后的危险因素(P<0.05)。依据多因素Logistic回归分析结果构建慢性硬膜下血肿钻孔引流术后近期不良预后风险的Nomogram预测模型,所得列线图一致性指数为0.796(0.693~0.867),显示此列线图预测与实际观察均有较好的一致性。ROC曲线下面积为0.942,95%CI:0.887~0.982,Youden指数为0.586,灵敏度为92.46%,特异度为78.25%,显示模型具有较好的区分度。结论年龄、脑萎缩、脑复张、术前血肿宽度、术前中线移位、术后硬膜下间隙、出院时中线偏移程度是慢性硬膜下血肿钻孔引流术后近期预后的危险因素,据此构建的Nomogram预测模型具有良好的区分度和校准度,能够有效地辅助医护人员有效识别慢性硬膜下血肿钻孔引流术后近期预后不良的高危群体,以采取针对性的措施,降低慢性硬膜下血肿钻孔引流术后近期不良预后的发生风险。

Objective To analyze the factors influencing the short-term prognosis of chronic subdural hematoma after trepanation and drainage and establish a Nomogram prediction model.Methods A total of 336 patients with chronic subdural hematoma admitted to the Department of Neurosurgery,Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from 2015 to 2022 were retrospectively selected as the study objects.The patients were divided into good prognosis group(n=241) and poor prognosis group(n=95) according to whether there was poor prognosis at follow-up within 1 month after surgery.Thirteen factors that may be related to the recurrence of chronic subdural hematoma after operation were analyzed and compared.Variables with statistically significant differences were included in multiple Logistic regression analysis,and a risk prediction model was constructed.A Nomogram model for predicting poor short-term prognosis after drill-drainage of chronic subdural hematoma was established by Rstudio for visualization.The receiver operating characteristic(ROC) curve of subjects were drawn to evaluate the predictive efficacy of the risk prediction model for poor prognosis after trepanation and drainage of chronic subdural hematoma.Results In this study,a total of 336 patients with chronic subdural hematoma were collected,and 95 patients had adverse reactions,accounting for 28.27%.Multiple Logistic regression analysis showed that age,cerebral atrophy,cerebral reexpansion,preoperative hematoma width,preoperative midline shift,postoperative subdural space,and midline shift degree at discharge were risk factors for the short-term prognosis of chronic subdural hematoma after trepanation and drainage(P<0.05).Based on the results of multiple Logistic regression analysis,a Nomogram model was established to predict the risk of short-term adverse prognosis after trepulation and drainage of chronic subdural hematoma.The consistency index of the nomogram was 0.796(0.693-0.867),indicating that the prediction of this nomogram was in good agreement with the actual observation.The area of ROC curve was 0.942,95%CI:0.887-0.982,Youden index was 0.586,sensitivity was 92.46%,specificity was 78.25%,indicating that the model had good differentiation.Conclusion Age,cerebral atrophy,cerebral reexpansion,preoperative hematoma width,preoperative midline displacement,postoperative subdural space,midline deviation degree at discharge are risk factors for the short-term prognosis of chronic subdural hematoma after trepanation and drainage.The established Nomogram prediction model has a good degree of differentiation and calibration,which can effectively assist the medical staff to effectively identify the high-risk groups with poor prognosis after trepanation and drainage of chronic subdural hematoma,and take targeted measures to reduce the risk of poor prognosis after trepanation and drainage.