Semiautomated Segmentation for Volumetric Analysis of Intratumoral Ethiodol Uptake and Subsequent Tumor Necrosis After Chemoembolization
Semiautomated Segmentation for Volumetric Analysis of Intratumoral Ethiodol Uptake and Subsequent Tumor Necrosis After Chemoembolization
复制标题
DOI：
10.2214/ajr.09.3964
复制
发表时间：
2010-11-01
影响因子：
5
通讯作者：
Badawi, Ramsey D.
中科院分区：
文献类型：
--
作者：
Monsky, Wayne L.;Kim, Isaac;Badawi, Ramsey D.
目标。线性测量，如实体肿瘤反应评价标准（RECIST）所描述的标准，在评估经动脉化疗栓塞（TACE）后的反应时可能受到限制。这项初步研究的目的是展示在TACE治疗后24小时内肿瘤内以及治疗后1个月坏死和活的肿瘤内观察到的硫代油（硫代油）体积测量的再现性。将体积测量与线性测量和生存结果进行比较。材料和方法。在2006年至2009年期间，27例肝脏恶性肿瘤患者连续接受了37例TACE手术。回顾性分析术后24小时和1个月的CT图像。三个观察者测量了两次体积。使用Wilcoxon's有符号秩检验来确定操作者内部的可重复性，以评估每次体积测量的差异是否接近于零。采用类内相关系数（ICC）和Bland-Altman图确定操作符间的可重复性。生存数据回顾性地从电子病历中获得。对Ethiodol、整个肿瘤和坏死肿瘤体积均显示出良好的观察者内再现性和观察者间再现性（p < 0.05, ICC < 0.9）。乙醇体积与随后坏死肿瘤体积（p = 0.009）、整个肿瘤体积减少（p = 0.004）和患者生存（p = 0.029）相关。Kaplan-Meier曲线显示，50%以上的肿瘤中硫噻酚的蓄积和坏死肿瘤体积增加10%及以上与生存率相关（p分别为0.028和0.047）。半自动化容量分析可以进行良好的内部和观察者之间的再现性。TACE术后肿瘤内积累的硫代酚的体积与随后的坏死有关。这些早期测量可以预测生存结果。
OBJECTIVE. Linear measurements, such as those described by the Response Evaluation Criteria in Solid Tumors (RECIST) criteria, may be limited for assessment of response after transarterial chemoembolization (TACE). The purpose of this pilot study was to show intra-and interobserver reproducibility of volumetric measurements of Ethiodol (ethiodized oil) seen within tumor 24 hours after TACE and of necrotic and viable tumor 1 month after treatment. Volumetric measurements are compared with linear measurements and survival outcomes.MATERIALS AND METHODS. Between 2006 and 2009, 37 consecutive TACE procedures were performed in 27 patients with hepatic malignancies. CT images obtained 24 hours and 1 month after TACE were retrospectively analyzed. Three observers measured volumes twice. Intraoperator reproducibility was determined using Wilcoxon's signed rank test to assess whether the difference in each volumetric measurement approaches zero. The intraclass correlation coefficient (ICC) and Bland-Altman plots were used to determine interoperator reproducibility. Survival data were retrospectively obtained from the electronic medical record.RESULTS. Good intraobserver reproducibility and interobserver reproducibility (p > 0.05, ICC > 0.9, respectively) were shown for Ethiodol, whole tumor, and necrotic tumor volumes. The volume of Ethiodol correlated with subsequent necrotic tumor volume (p = 0.009), reduction in whole tumor volume (p = 0.004), and patient survival (p = 0.029). Kaplan-Meier curves suggest that Ethiodol accumulation in more than 50% of the tumor and a 10% or greater increase in the volume of necrotic tumor correlated with survival (p = 0.028 and 0.047, respectively).CONCLUSION. Semiautomated volumetric analysis can be performed with good intra-and interobserver reproducibility. The volume of Ethiodol accumulated in the tumor after TACE correlates with subsequent necrosis. These early measurements may predict survival outcomes.