首页|个体化预测重型颅脑损伤患者术后并发脑积水的风险列线图模型的建立

个体化预测重型颅脑损伤患者术后并发脑积水的风险列线图模型的建立

Establishment of an individualized risk nomogram model for predicting postoperative hydro-cephalus in patients with severe craniocerebral injury

扫码查看
目的 构建重型颅脑损伤患者术后并发脑积水的列线图模型.方法 选取2019年2月至2022年7月进入医院进行手术治疗的148例重型颅脑损伤患者作为研究对象.采用Logistic回归筛选重型颅脑损伤患者术后并发脑积水的危险因素,采用R软件构建重型颅脑损伤患者术后并发脑积水的列线图模型,并对列线图模型进行内部验证.结果 148例重型颅脑损伤患者中有39例患者术后并发脑积水,脑积水的发生率为26.35%(39/148).Logistic回归分析结果显示,脑室出血、蛛网膜下腔出血、开放性损伤、去骨瓣减压术及骨窗面积较大(≥120 cm2)是重型颅脑损伤患者术后并发脑积水的危险因素.基于危险因素建立了重型颅脑损伤患者术后并发脑积水的列线图模型,重型颅脑损伤患者术后并发脑积水的列线图模型校正曲线显示预测概率和实际概率基本一致,模型的ROC曲线下面积是0.876(95%CI:0.814~0.938).结论 脑室出血、蛛网膜下腔出血、开放性损伤、去骨瓣减压术及骨窗面积较大是重型颅脑损伤患者术后并发脑积水的危险因素,构建的重型颅脑损伤患者术后并发脑积水的列线图模型的准确性较好.
Objective To construct the nomogram model of postoperative hydrocephalus in patients with severe craniocerebral injury.Methods A total of 148 patients with severe cranioce-rebral injury who entered the hospital for surgical treatment from February 2019 to July 2022 were selected as the study objects.Logistic regression was used to screen the risk factors of post-operative hydrocephalus in patients with severe craniocerebral injury.R software was used to build the nomogram model of postoperative hydrocephalus in patients with severe craniocerebral injury,and the nomogram model was internally verified.Results Among 148 patients with severe craniocerebral injury,39 cases were complicated with hydrocephalus after operation,and the inci-dence of hydrocephalus was 26.35%(39/148).Logistic regression analysis showed that ventricu-lar hemorrhage,subarachnoid hemorrhage,open injury,decompressive craniectomy and large bone window area(≥120 cm2)were the risk factors for postoperative hydrocephalus in patients with severe head injury.Based on the risk factors,a nomogram model was established for patients with severe craniocerebral injury complicated with hydrocephalus after surgery.The correction curve of the nomogram model for patients with severe craniocerebral injury complicated with hydroceph-alus after surgery showed that the predicted probability was basically consistent with the actual probability,and the area under the ROC curve of the model was 0.876(95%CI:0.814~0.938).Conclusion Ventricular hemorrhage,subarachnoid hemorrhage,open injury,decompressive crani-ectomy and large bone window area are risk factors for postoperative hydrocephalus in patients with severe craniocerebral injury.The accuracy of the nomographic model of postoperative hydro-cephalus in patients with severe craniocerebral injury is good.

Severe craniocerebral injurySurgeryHydrocephalusRisk factorsNomogram model

魏曙岚、陈炜钰、花胜松

展开 >

230011 合肥 合肥市第八人民医院神经外科

重型颅脑损伤 手术 脑积水 危险因素 列线图模型

2024

立体定向和功能性神经外科杂志
安徽省脑立体定向神经外科研究所

立体定向和功能性神经外科杂志

CSTPCD
影响因子:0.488
ISSN:1008-2425
年,卷(期):2024.37(3)