基于CT诺模图预测高血压脑出血血肿增大的研究OA
Study on CT-Based Nomogram for Predicting Hematoma Expansion in Hypertensive Intracerebral Hemorrhage
目的:探讨基于临床特征和平扫 CT(NCCT)影像特征的诺模图预测高血压脑出血血肿增大(HE)的价值.方法:回顾性分析 2018 年 12 月—2024 年 6 月韶关市第一人民医院收治的 248 例急性早期高血压脑出血患者临床资料及 NCCT 扫描资料.患者发病 24 h 以内行首次颅脑NCCT 检查,并于 24 h 内行颅脑NCCT 复查,根据是否存在早期高血压脑出血血肿增大(HE)分为 HE组(84 例)和非 HE 组(164 例),对两组研究对象的临床特征及 NCCT 影像特征,基于单因素及多因素 logistic 回归分析所筛选的危险因素建立模型,建立了一个诺模图预测模型.利用 ROC 曲线对模型预测效能进行检验,绘制校准曲线估算预测概率及实际概率一致性,评估诺模图预测效能.结果:单因素及多因素结果显示饮酒史、血肿形状不规则、岛征和混杂征是 HE 的独立危险因素(P<0.05).基于这些因素构建的诺模图预测模型具有较高的临床应用价值,绘制其 ROC 曲线,预测高血压脑出血血肿增大风险的曲线下面积 AUC 为 0.858(95%CI:0.802～0.914),表明诺模图模型区分度良好.Hosmer-Lemeshow 拟合优度检验结果(字2=4.263,P=0.898),该模型的校准度和稳定性良好.结论:根据临床和NCCT 影像特征的个体化预测模型可预测高血压脑出血血肿增大.
Objective:To investigate the value of a nomogram based on clinical features and non-contrast computed tomography(NCCT)imaging features for hematoma expansion(HE)prediction in hypertensive intracerebral hemorrhage.Method:A retrospective analysis was conducted on the clinical data and non-contrast computed tomography(NCCT)scans of 248 patients with acute early-stage hypertensive intracerebral hemorrhage admitted to Shaoguan First People's Hospital from December 2018 to June 2024.All underwent initial cranial NCCT within 24 hours and follow-up NCCT within 24 hours of symptom onset,and were divided into HE group(n=84)and non-HE group(n=164).Univariate plus multivariate logistic regression tests were performed to screen out various risk factors,and a nomogram prediction model was established accordingly.The predictive performance of the model was estimated from ROC curves and area under curve(AUC).Calibration curves were plotted in order to obtain the agreement between predicted probabilities and actual probabilities to measure predictions.Result:Alcohol consumption history,irregular hematoma shape,island sign and mixed density sign were independent risk factors for HE(P<0.05).The nomogram had good discrimination(AUC=0.858,95%CI:0.802-0.914)and satisfactory calibration/stability(Hosmer-Lemeshow test:字2=4.263,P=0.898).Conclusion:The individualized nomogram based on clinical and NCCT features can effectively predict HE risk in hypertensive intracerebral hemorrhage.
陈龙;郭强;王康;林文聪;陈文坚;何宁波
韶关市第一人民医院 广东 韶关 512000韶关市第一人民医院 广东 韶关 512000韶关市第一人民医院 广东 韶关 512000韶关市第一人民医院 广东 韶关 512000韶关市第一人民医院 广东 韶关 512000韶关市第一人民医院 广东 韶关 512000
高血压脑出血血肿增大计算机体层摄影术诺模图
Hypertensive intracerebral hemorrhageHematoma expansionComputed tomography imagingNomograph
《中外医学研究》 2026 (10)
25-29,5
韶关市科技局高水平医院建设科研项目(220527084530221)
评论