Zhongguo cuzhong zazhi (Jul 2023)

BAT评分联合CTA点征对幕上自发性脑出血患者早期血肿扩大的预测价值研究 The Predictive Value of BAT Score Combined with CTA Spot Sign for Hematoma Expansion in Patients with Spontaneous Supratentorial Intracerebral Hemorrhage

  • 李娜,姬泽强,文心瑜,吴蕾,赵性泉

DOI
https://doi.org/10.3969/j.issn.1673-5765.2023.07.006
Journal volume & issue
Vol. 18, no. 7
pp. 780 – 786

Abstract

Read online

目的 研究BAT评分联合CTA点征对幕上自发性脑出血患者早期血肿扩大的预测价值。 方法 回顾性分析2021年9月—2022年4月连续于首都医科大学附属北京天坛医院急诊就诊且发病至一站式CT检查时间在6 h内的幕上自发性脑出血患者的资料,根据发病后24~48 h复查头颅CT与基线CT的血肿体积对比,分为血肿扩大组和非血肿扩大组。比较两组患者的基本临床资料、CTA点征及非增强CT(non-contrast computed tomography sign,NCCT)特点(低密度征、混合征等)影像资料、BAT评分(blend sign,any hypodensity,timing of NCCT score,BAT)。采用单因素分析和多因素logistic回归分析血肿扩大的危险因素,并绘制ROC曲线分析BAT评分、CTA点征及两者结合对自发性脑出血患者早期血肿扩大的预测价值。 结果 共纳入97例患者,血肿扩大组28例,非血肿扩大组69例。血肿扩大组患者CTA点征,NCCT低密度征、混合征的发生率均高于非血肿扩大组,BAT评分中位数和BAT评分≥3分的患者比例均高于非血肿扩大组,发病至基线影像时间短于非血肿扩大组,基线出血体积和基线随机血糖水平高于非血肿扩大组,上述差异均有统计学意义。多因素logistic回归分析结果显示,CTA点征阳性(OR 31.828,95%CI 5.350~189.337,P<0.01)、BAT评分≥3分(OR 71.976,95%CI 5.391~960.899,P<0.01)、基线出血体积(OR 1.029,95%CI 1.003~1.055,P=0.03)和基线随机血糖(OR 1.355,95%CI 1.070~1.714,P=0.01)可独立预测血肿扩大。ROC曲线分析显示,基线出血体积预测血肿扩大的AUC为0.762,BAT评分≥3分预测血肿扩大的AUC为0.716,CTA点征阳性预测血肿扩大的AUC为0.756,BAT评分≥3分联合CTA点征阳性预测血肿扩大的AUC为0.833。De Long检验提示两者联合预测效力优于单独预测效力(P<0.05)。 结论 幕上自发性脑出血患者发病6 h内,BAT评分≥3分及CTA点征阳性都可有效预测血肿扩大,两者联合预测的效力更佳。 Abstract: Objective To study the predictive value of the blend sign, any hypodensity, timing of NCCT (BAT) score combined with CTA spot sign for early hematoma expansion (HE) in patients with spontaneous supratentorial intracerebral hemorrhage (sICH). Methods A retrospective analysis was conducted on sICH patients who were admitted to the emergency department of Beijing Tiantan Hospital, Capital Medical University from September 2021 to April 2022 with a time from onset to one-stop CT within 6 hours. The patients were divided into HE group and non-HE group based on the comparison of hematoma volume between baseline CT and the follow-up CT done within 24-48 hours after onset. The clinical characteristics, CTA spot sign, non-contrast computed tomography sign (NCCT) (including the hypodensity sign, blend sign, etc) , and BAT score were identified and compared between 2 groups. Univariate analysis and multivariate logistic regression analysis was used to analyze the risk factors affecting the HE, and receiver operating characteristic was drawn to analyze the predictive value of BAT score, CTA spot sign and their combination on hematoma expansion in patients with sICH. Results A total of 97 patients, including 28 in the HE group and 69 in the non-HE group. The incidence of CTA spot sign, NCCT hypodensity sign, and blend sign in patients with HE group was higher than that in non-HE group. The median BAT score and the proportion of patients with BAT score≥3 were higher than those in non-HE group. The time from onset to baseline imaging was shorter than that in non-HE group. Baseline hematoma volume and baseline random blood glucose levels were higher than those in non-HE group. The above differences were statistically significant. The results of multivariate logistic regression analysis showed that positive CTA spot sign (OR 31.828, 95% CI 5.350-189.337, P<0.01), BAT score≥3 (OR 71.976, 95%CI 5.391-96.899, P<0.01), baseline hematoma volume (OR 1.029, 95%CI 1.003-1.055, P=0.03), and baseline random blood glucose (OR 1.355, 95%CI 1.070-1.714, P=0.01) can independently predict hematoma enlargement. ROC analysis showed that the AUC of baseline hematoma volume predicting hematoma enlargement was 0.762, the AUC of BAT score≥3 predicting hematoma enlargement was 0.716, the AUC of CTA spot sign predicting hematoma enlargement was 0.756, and the AUC of BAT score≥3 combining with CTA spot sign predicting HE was 0.833. The De Long test suggests that the combined predictive power of the two is superior to the individuals (P<0.05). Conclusions Within 6 hours of onset, both a BAT score≥3 and CTA spot sign can independently predict HE in patients with sICH, and the combined prediction is more effective.

Keywords