作者：郭雅琴，许群峰，江丽华

单位：

单位（英文）：

Department of Neurosurgery, Maanshan Shiqiye Hospital, Maanshan 243000, China Corresponding author: GUO Yaqin, E-mail: guoyaqin1926@126.com

关键词：

颅脑损伤; 脑梗死; 开颅血肿清除术; 影响因素; 列线图;

关键词（英文）：

Craniocerebral injury; Brain infarction; Craniotomy hematoma removal; Influencing factors; Nomogram

中图分类号：

DOI：

10.12114/j.issn.1008-5971.2022.00.101

基金项目：

摘要：

目的 探讨颅脑损伤患者开颅血肿清除术后并发脑梗死的影响因素并构建其风险预测列线图模型。方法选取2018年1月至2021年6月于马鞍山十七冶医院行开颅血肿清除术的颅脑损伤患者191例。收集患者的临床资料,采用单因素分析和多因素Logistic回归分析探讨颅脑损伤患者开颅血肿清除术后并发脑梗死的影响因素,采用R 3.5.3软件包和rms程序包建立颅脑损伤患者开颅血肿清除术后并发脑梗死的风险预测列线图模型。结果 191例患者中,有52例发生脑梗死,发生率为27.2%。根据脑梗死发生情况将患者分为脑梗死组(n=52)和非脑梗死组(n=139)。多因素Logistic回归分析结果显示,入院时GCS评分、脑疝、术后低血压、蛛网膜下腔出血和受伤至开颅时间是颅脑损伤患者开颅血肿清除术后并发脑梗死的影响因素(P<0.05)。基于上述因素建立颅脑损伤患者开颅血肿清除术后并发脑梗死的风险预测列线图模型。模型验证结果显示,该列线图模型的一致性指数为0.778;校正曲线趋近于理想曲线;ROC曲线分析结果显示,该列线图模型预测颅脑损伤患者开颅血肿清除术后并发脑梗死的AUC为0.792〔95%CI(0.765,0.821)〕。结论 入院时GCS评分、脑疝、术后低血压、蛛网膜下腔出血和受伤至开颅时间是颅脑损伤患者开颅血肿清除术后并发脑梗死的影响因素,本研究基于上述因素构建的列线图模型对颅脑损伤患者开颅血肿清除术后并发脑梗死有一定预测效能,具有一定的临床应用价值。

英文摘要：

【Abstract】　Objective To explore the influencing factors of cerebral infarction after craniotomy hematoma removal inpatients with craniocerebral injury, and to construct its risk prediction nomogram model. Methods A total of 191 patients withcraniocerebral injury who underwent craniotomy hematoma removal in Maanshan Shiqiye Hospital from January 2018 to June2021 were selected, and the clinical data of the patients were collected. Univariate analysis and multivariate Logistic regressionanalysis were used to investigate the influencing factors of cerebral infarction after craniotomy hematoma removal in patients withcraniocerebral injury. R 3.5.3 software package and rms software package were used to establish a nomogram model of the riskof cerebral infarction after craniotomy hematoma removal in patients with craniocerebral injury. Results Fifty two of the 191patients had cerebral infarction, and the incidence was 27.2%. According to the occurrence of cerebral infarction, the patientswere divided into cerebral infarction group (n=52) and non-cerebral infarction group (n=139) . The results of multivariate Logisticregression analysis showed that, GCS score at admission, brain herniation, postoperative hypotension, subarachnoid hemorrhageand time from injury to craniotomy were the influencing factors of cerebral infarction after craniotomy hematoma removal inpatients with craniocerebral injury (P < 0.05) . Based on the above factors, a nomogram model for predicting cerebral infarctionafter craniotomy hematoma removal in patients with craniocerebral injury was established. The model validation results showedthat the consistency index was 0.778. The calibration curve was close to the ideal curve. The results of ROC curve analysisshowed that the AUC of nomogram model for predicting cerebral infarction after craniotomy hematoma removal in patients withcraniocerebral injury was 0.792 [95%CI (0.765, 0.821) ] . Conclusion GCS score at admission, brain herniation, postoperativehypotension, subarachnoid hemorrhage and time from injury to craniotomy are the influencing factors of cerebral infarction aftercraniotomy hematoma removal in patients with craniocerebral injury. The nomogram model constructed based on the above factorshas certain predictive efficiency for cerebral infarction after craniotomy hematoma removal in patients with craniocerebral injury,and has certain clinical application value.

参考文献：