超急性期脑出血血肿增大的CT预测指标  

Predictor measures on CT for hematoma expansion following acute intracerebral hemorrhage

作　　者：刘蓉[1] 龚建平[1] 朱江涛[1] 付彤[1] 张伟[1] 蔡武[1] 乔方[1] 沈钧康[1] 

机构地区：[1]苏州大学附属第二医院影像科,215004

出　　处：《中华医学杂志》2016年第9期720-723,共4页National Medical Journal of China

摘　　要：目的评估超急性期脑出血血肿增大的预测指标的价值。方法回顾性分析2012年1月-2015年2月苏州大学附属第二医院的105例超急性期脑出血患者，症状发生在6h以内，均行首诊头颅CT平扫、CT血管成像（CTA）及24h内复查头颅CT平扫。影像资料由两名有经验的放射科医师进行分析，以血肿有无增大为因变量，以血肿位置、形态、初始血肿体积、斑点征及UHG速度分组为自变量做二元Logistic回归分析。结果105例患者中，血肿增大组30例，无血肿增大组75例。单因素分析显示血肿位置、血肿形态、斑点征、UHG速度分组及初始血肿体积在两组间比较（χ2值：13．125、23．987、25．846、20．328及t=-3．183，均P〈0．05）；二元Logistic回归分析显示血肿形态（不规则及分裂型）、斑点征及UHG速度分组在两组间差异有统计学意义（P=0．033、0．009、0．000及0．040）；ROC曲线显示血肿形态、斑点征及UHG速度分组曲线下面积分别为0．776、0．740及0．720，其95％C1分别为0．682～0．870、0．625—0．855及0．604～0．836，敏感度均为60％，特异度分别为84％、88％及84％。结论血肿形态、斑点征及UHG速度分组都是急性脑出血血肿增大的独立预测因素，斑点征特异度更高，而血肿形态及UHG速度分组应用更简便。Objective To evaluate the worth of solid predictors in acute intracerebral hematoma （ICH） expansions in computer tomography images. Methods A total of 105 patients with acute ICH in The Second Affiliated Hospital of Sooehow University during January 2012 to February 2015 were enrolled. CT plain scan, CTA within 6 hours since the symptoms and CT plain scan recheck within 24 hours were executed. Hematoma location, initial volume of hematoma, shape of hematoma, ＂spot sign＂, UHG speed were analyzed with single factor and binary Logistic regression between the patients with and without hematoma expansion. Results There were 30 cases with hematoma expansion and 75 cases with no hematoma expansion in 105 patients. In single factor comparisons, hematoma location（ χ2 = 13. 125, P 〈 0.05）, hematoma shape （ χ2 = 23.987, P 〈 0. 05）, spot sign （ χ2 = 25. 846, P 〈 0. 05 ）, UHG speed （ χ2 = 20. 328, P 〈 0. 05 ） and the initial hematoma volume （ t = - 3. 183, P 〈 0. 05 ） between the hematoma expansions and the non-hematoma expansions made significant differences. In binary Logistic regression, hematoma shape（irregular （P =0. 033） and cleavage（P =0. 009） ）, spot sign（P =0. 000） and UHG speed （P =0. 040） had significant differences between the two groups. ROC curve areas of hematoma shape, spot sign and UHG speed were 0. 776 （ 95% CI 0. 682 - 0. 870 ）, 0. 740 （ 95% CI 0. 625 - 0. 855 ） and 0. 720 （95% C0. 604 -0. 836）. The high specificities of hematoma shape（84% ） , spot sign （88%）and UHG speed（84% ）revealed their great reliabilities with equal sensitivity （60%）. Conclusion Hematoma shape, spot sign and UHG speed are solid predictors of hematoma expansion among which spot sign has promising specificity, hematoma shape and UHG speed are more convenient to be observed.

关 键 词：脑出血 体层摄影术 X线计算机 血管造影术 血肿增大 

分 类 号：R816.1[医药卫生—放射医学] R743.34[医药卫生—临床医学]