期刊文献+

基于倾向性评分探讨混合征预测血肿扩大的价值 被引量:6

Value of blend sign in predicting the hematoma expansion based on propensity score matching
下载PDF
导出
摘要 目的:采用倾向性评分匹配(PSM)研究CT混合征预测自发性脑出血(sICH)患者早期血肿扩大(HE)的价值。方法:筛选2010年9月-2019年12月本院收治的sICH患者,对符合纳入标准的273例患者的临床及影像学资料进行回顾性分析。所有患者均于发病后6h内行首次CT平扫及24h内复查CT平扫。根据有无HE把患者分为HE组(63例)和非HE组(210例)。使用PSM方法将两组患者进行匹配,匹配后得到55对患者。分析PSM前后混合征对早期HE的预测价值。结果:273例sICH患者中,57例出现混合征,HE组24例,非HE组33例。HE组和非HE组高血压史分别为23例和44例,合并破入脑室分别为32例和53例,HE组和非HE组入院收缩压、初始血肿体积及入院格拉斯哥昏迷评分(GCS)分别为164.64 mmHg和158.65 mmHg、28.47 mL和20.48 mL、8.40分和10.74分,单因素分析显示以上因素在两组间差异均有统计学意义(P<0.05)。Logistic回归分析显示,合并破入脑室、入院GCS评分、初始血肿体积及混合征是预测早期HE的独立危险因素。经PSM进行1:1匹配后的55对患者中,HE组及非HE组混合征分别为22例和12例,结果显示混合征仍是预测早期HE的危险因素。结论:对急诊CTA检查禁忌症的患者或无法行急诊CTA检查的医疗机构,根据CT混合征预测早期HE对指导临床采取及时有效的治疗措施、减轻患者的不良预后具有重要的意义。 Objective:To explore the value of blend sign on CT in predicting the early hematoma expansion(HE)in patients with spontaneous intracerebral hemorrhage(sICH)using propensity score matching(PSM).Methods:sICH patients admitted to our hospital from September 2010 to December 2019 were screened.Two hundred and seventy-three sICH patients met the inclusion criteria and their clinical and imaging data were retrospectively analyzed.All patients underwent the baseline CT scan within 6 hours and the follow-up CT scan within 24 hours after the onset of the disease.The sICH patients were divided into HE group(63 cases)and non-HE group(210 cases)according to the presence of early HE.The patients of the two groups were subjected to PSM,after which 55 pairs of patients were obtained.The predictive value of blend sign of early HE on CT was analyzed before and after PSM.Results:Of the 273 sICH patients,57 cases had blend sign who belong to HE group in 24 cases and non-HE group in 33 cases.Patients with hypertension and sICH combined with ventricular invasion 23 cases and 32 cases in HE group,and 44 cases and 53 cased in non-HE group respectively.The admission systolic blood pressure,the initial hematoma volume and the admission Glasgow coma score(GCS)were in 164.64 mmHg,28.47 mL,and 8.40 score in HE group and 158.65 mmHg,20.48 mL,and 10.74 score in non-HE group respectively.Univariate analysis showed that the above factors were significantly different between HE and non-HE groups.Logistic regression analysis showed that sICH combined with ventricular invasion,the admission GCS,the initial hematoma volume,and the blend sign were independent risk factors for predicting early HE.The patients after 1:1 matching by PSM,there were 22 cases in HE group and 12 cases in non-HE group respectively,and univariate analysis showed that the blend sign remained an independent risk factor for predicting early HE.Conclusion:For the sICH patients who can’t perform CTA,blend sign on CT is an valuable marker to predict early HE,which was important for clinicians to timely make effective treatment for improving the prognosis of sICH.
作者 时代 徐亮 范国华 沈钧康 王业青 金丹 SHI Dai;XU Liang;FAN Guo-hua(Department of Radiology,the Second Affiliated Hospital of Soochow University,Jiangsu 215000,China)
出处 《放射学实践》 CSCD 北大核心 2021年第7期843-847,共5页 Radiologic Practice
基金 苏州大学附属第二医院青年职工预研项目(SDFEYQN1816)。
关键词 体层摄影术 X线计算机 混合征 倾向性评分匹配 血肿扩大 Tomography,X-ray computed Blend sigh Propensity score matching Hematoma expansion
  • 相关文献

参考文献5

二级参考文献33

  • 1Austin PC.A critical appraisal of propensity-score matching in the medical literature between 1996 and 2003.Stat Med,2008,27:2037-2049.
  • 2Austin PC,Grootendorst P,Anderson GM.A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects:a Monte Carlostudy.Stat Med,2007,26:734-753.
  • 3Rosenbaum PR,Rubin DB.The central role of the propensity score in observational studies for causal effects.Biometrika,1983,70:41-55.
  • 4D' Agostino RB.Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group.Stat Med,1998,17:2265-2281.
  • 5Dehejia RH,Wahba S.Propensity score-matching methods for nonexperimental causal studies.Rev Eco Stat,2002,84:151-161.
  • 6Austin PC.Some methods of propensity-score matching had superior performance to others:results of an empirical investigation and Monte Carlo simulations.Biomet J,2009,51:171-184.
  • 7Rosenbaum PR.Optimal matching for observational studies.J Am Stat Asso,1989,84:1024-1032.
  • 8Ming K,Rosenbaum PR.A note on optimal matching with variable controls using the assignment algorithm.J Computat Graphic Stat,2001,3:455-463.
  • 9Smith JA,Todd PE.Does matching overcome LaLonde's critique of nonexperimental estimators? J Economet,2005,125:305-353.
  • 10Austin PC,Mamdanil MM.A comparison of propensity score methods:a case-study estimating the effectiveness of post-AMI statin use.Stat Med,2006,25:2084-2106.

共引文献114

同被引文献45

引证文献6

二级引证文献6

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部