颅内动脉瘤破裂伴血肿夹闭术预后模型构建

Construction of a prognostic model for intracranial aneurysm rupture with hematoma clipping surgery

目的探究颅内动脉瘤破裂伴血肿患者夹闭术后预后不良的影响因素并构建临床预测模型。方法选取2017年9月至2023年10月东阳市人民医院收治的151例动脉瘤性颅内血肿患者。术后3个月,采用改良Rankin量表(modified Rankin scale,mRS)进行评分分组,预后良好组93例,预后不良组58例。采用单因素分析和多因素Logistic回归分析探究影响患者术后预后不良的风险因素,构建颅内动脉瘤破裂伴血肿患者夹闭术后预后不良预测模型,采用受试者操作特征曲线下面积(area under the curve,AUC)反映模型的鉴别效度,采取Hosmer-Lemeshow检验建立模型的拟合度。结果Hunt-Hess分级Ⅳ~Ⅴ级(OR=5.339)、改良Fisher分级Ⅲ~Ⅳ级(OR=5.145)、血肿量≥50ml(OR=7.426)、血肿清除率≤50%(OR=8.381)、责任动脉瘤直径>5mm(OR=3.053)、手术时间窗>5h(OR=2.659)、术中血管区操作时间>3h(OR=2.305)为颅内破裂动脉瘤伴血肿患者夹闭术后预后不良的独立风险因素(P<0.05)。本研究构建的夹闭术后预后不良预测模型的AUC为0.863(95%CI:0.781~0.946,P<0.001),模型预测效能较好,特异性79.6%,敏感度86.2%,预测正确率为82.1%。Hosmer-Lemeshow检验χ^(2)=5.778,P=0.679,模型预测值与实际观测值无显著差异。结论Hunt-Hess分级Ⅳ~Ⅴ级、改良Fisher分级Ⅲ~Ⅳ级、血肿量≥50ml、血肿清除率≤50%、责任动脉瘤直径>5mm、手术时间窗>5h、术中血管区操作时间>3h为颅内动脉瘤破裂伴血肿患者夹闭术后预后不良的风险因素。本研究构建的模型预测效能较高,可为临床开颅夹闭术患者治疗及术后康复提供指导。

Objective To explore the influencing factors of poor prognosis after clipping in patients with ruptured intracranial aneurysm and hematoma,and to construct a clinical prediction model.Methods A total of 151 patients with aneurysmal intracranial hematoma in Dongyang People’s Hospital were selected from September 2017 to October 2023.3 months after operation,the patients were grouped by modified Rankin scale(mRS),with 93 cases in good prognosis group and 58 cases in poor prognosis group.Univariate analysis and multivariate Logistic regression analysis were used to explore the risk factors affecting the poor prognosis of patients with postoperative prognosis,and a poor prognostic prediction model for patients with intracranial aneurysm rupture and hematoma clamping was constructed,the discriminant validity of the area under the curve(AUC)was evaluated,and the fit of the model was established using the Hosmer-Lemeshow test.Results The Hunt-Hess gradeⅣ-Ⅴ(OR=5.339),modified Fisher gradeⅢ-Ⅳ(OR=5.145),hematoma volume≥50ml(OR=7.426),hematoma clearance rate was≤50%(OR=8.381),size of the responsible aneurysm>5mm(OR=3.053),operation time window>5h(OR=2.659),and intraoperative vascular operation time>3h(OR=2.305)were independent risk factors for poor prognosis after clipping in patients with intracranial aneurysm ruptured and intracranial hematoma(P<0.05).The AUC of the poor prognosis prediction model of patients with intracranial aneurysm rupture with hematoma after clipping was 0.863(95%CI:0.781-0.946,P<0.001),the specificity was 79.6%,the sensitivity was 86.2%,and the prediction accuracy was 82.1%.Hosmer-Lemeshow testχ^(2)=5.778,P=0.679,and there was no significant difference between the predicted value of the model and the actual observed value.Conclusion Hunt-Hess gradeⅣ-Ⅴ,the modified Fisher gradeⅢ-Ⅳ,hematoma volume≥50ml,hematoma clearance rate≤50%,responsible aneurysm size>5mm,operation time window>5h,and intraoperative vascular operation time>3h were independent risk factors for poor prognosis after clipping in patients with intracranial aneurysm ruptured.The model constructed in this study has high predictive performance and can provide guidance for the treatment and postoperative recovery of patients undergoing craniotomy and clipping surgery in clinical practice.