摘要：

目的 筛查急性基底动脉闭塞血管内支架取栓术后预后不良的危险因素，并评价4种影像学评分系统的预测价值。方法 2012年3月至2018年8月共57例急性基底动脉闭塞患者均行血管内支架取栓术，术前行DWI后循环Alberta脑卒中计划早期CT评分（pc-ASPECTS）、DWI脑干评分（BSS）、后循环CTA评分（pc-CTA）和基底动脉CTA评分（BATMAN），术后即刻采用脑梗死溶栓血流分级（TICI）评价血管再通，术后36 h内记录症状性颅内出血发生率，发病后3个月采用改良Rankin量表（mRS）评价预后。单因素和多因素Logistic回归分析筛查急性基底动脉闭塞血管内支架取栓术后预后不良的危险因素。绘制受试者工作特征曲线（ROC）并计算曲线下面积，评价DWI pc-ASPECTS、DWIBSS、pc-CTA和BATMAN评分的预测价值。结果 57例患者中53例（92.98%）血管完全再通，2例（3.51%）部分再通，2例（3.51%）未再通；3例（5.26%）症状性颅内出血；22例（38.60%）预后良好、35例（61.40%）预后不良；14例（24.56%）死亡。Logistic回归分析，入院时高NIHSS评分（OR=0.879，95% CI：0.783~0.986；P=0.028）、高DWI BSS评分（OR=0.348，95% CI：0.177~0.683；P=0.002）和低BATMAN评分（OR=1.549，95% CI：1.019~2.353；P=0.040）是急性基底动脉闭塞血管内治疗后预后不良的危险因素。ROC曲线，DWI pc-ASPECTS、DWI BSS、pc-CTA和BATMAN评分曲线下面积分别为0.787（95% CI：0.658~0.884，P=0.000）、0.861（95% CI：0.744~0.938，P=0.000）、0.634（95% CI：0.496~0.757，P=0.091）和0.698（95% CI：0.562~0.813，P=0.012）。结论 入院时高NIHSS评分、高DWI BSS评分和低BATMAN评分是急性基底动脉闭塞血管内治疗后预后不良的危险因素。

关键词: 卒中, 脑缺血, 动脉闭塞性疾病, 支架, 弥散磁共振成像, 体层摄影术, X 线计算机, 危险因素, Logistic 模型

Abstract:

Objective To screen the risk factors related to unfavorable prognosis and evaluate the predictive effect of four imaging scoring systems on the unfavorable prognosis in the patients with acute basilar artery occlusion (BAO) after endovascular thrombectomy. Methods From March 2012 to August 2018, a total of 57 patients with acute BAO were treated with endovascular thrombectomy, preoperative DWI Posterior Circulation Alberta Stroke Program Early CT Score (pc-ASPECTS), DWI Brain Stem Score (BSS), Posterior Circulation CTA (pc-CTA) and Basilar Artery CTA (BATMAN) were used. Thrombolysis in Cerebral Infarction (TICI) was used to evaluate the incidence of vascular recanalization. The incidence of symptomatic intracranial hemorrhage (sICH) was recorded within 36 h. The prognosis was evaluated by modified Rankin Scale (mRS) at 3 months. Univariate and multivariate backward Logistic regression analysis were used to screen the risk factors related to the unfavoriable prognosis in the patients with acute BAO after endovascular thrombectomy. Receiver operating characteristic curve (ROC) was used to calculate the area under the curve (AUC), and evaluate the predictive value of DWI pc-ASPECTS, DWI BSS, pc-CTA and BATMAN scores for prognosis. Results In 57 patients, 53 patients (92.98%) had successful recanalization, 2 (3.51%) had partial recanalization, 2 (3.51%) had no recanalization, 3 (5.26%) had sICH, 22 (38.60%) had good prognosis, 35 (61.40%) had poor prognosis, and 14 (24.56%) died. After univariate and multivariate backward Logistic regression analysis, higher NIHSS score on admission (OR=0.879, 95% CI:0.783-0.986; P=0.028), higher DWI BSS score (OR=0.348, 95% CI:0.177-0.683; P=0.002) and lower BATMAN score (OR=1.549, 95% CI:1.019-2.353; P=0.040) were risk factors for unfavorable prognosis of patients with acute BAO after endovascular thrombectomy. ROC analysis of the four radiographic scoring systems showed that the AUC of DWI pc-ASPECTS, DWI BSS, pc-CTA and BATMAN score were 0.787 (95% CI:0.658-0.884, P=0.000), 0.861 (95% CI:0.744-0.938, P=0.000), 0.634 (95% CI:0.496-0.757, P=0.091) and 0.698 (95% CI:0.562-0.813, P=0.012), respectively. Conclusions Higher NIHSS score on admission, higher DWI BSS score and lower BATMAN score are risk factors for unfavorable prognosis of acute BAO treated with endovascular thrombectomy.

Key words: Stroke, Brain ischemia, Arterial Occlusive diseases, Stents, Diffusion magnetic resonance imaging, Tomography, X-ray computed, Risk factors, Logistic models