Clinical Anastomotic Leakage After Rectal Cancer Resection Can Be Predicted by Pelvic Anatomic Features on Preoperative MRI Scans: A Secondary Analysis of a Randomized Controlled Trial
Clinical Anastomotic Leakage After Rectal Cancer Resection Can Be Predicted by Pelvic Anatomic Features on Preoperative MRI Scans: A Secondary Analysis of a Randomized Controlled Trial
复制标题
直肠癌切除后的临床吻合口漏可以通过术前 MRI 扫描的盆腔解剖特征来预测：随机对照试验的二次分析
DOI：
10.1097/dcr.0000000000001481
复制
发表时间：
2019-11-01
影响因子：
3.9
通讯作者：
Wang, Lei
中科院分区：
文献类型：
--
作者：
Ma, Tenghui;Zhong, Qinghua;Wang, Lei
背景：我们先前已经证明，术前放疗引起的放射性直肠炎是直肠癌切除术后临床吻合口漏的易感因素。需要对放射性直肠炎进行定量测量。目的：量化直肠癌手术前放射治疗对解剖结构的影响，评价其对漏口的预测能力。设计：随机对照试验(NCT01211210)的二次分析。对MRI变量进行了回顾性评估。环境：这项研究是在试验的领先中心进行的，这是一家三级GI医院。患者：包括行术前放化疗和保括约肌手术的患者。主要观察指标：放射治疗前和放射治疗后的MRI测量解剖特征。采用单变量分析确定影响预后的因素。构建受试者操作特征曲线以确定MRI变量变化在预测渗漏中的临界值。结果：125例患者中有18例(14.4%)发生临床吻合口瘘。渗漏组和非渗漏组的基线特征相似。骶前间隙宽度、直肠壁厚度和乙状结肠远端的相对增量比随机机会更好地区分两组。相对增加的骶前间隙宽度预测效果最好，曲线下面积为0.722，敏感性为66.7%，特异性为72.0%，阳性预测值为28.6%，阴性预测值为92.8%。局限性：这项研究受到样本量小和回溯性设计的限制。结论：增加骶前间隙、直肠壁厚度和乙状结肠远端的厚度有助于识别直肠癌切除术后临床吻合口漏的无风险个体。第一个变量是最强的预测者。在评估去功能性造口的应用时，应考虑这些变量的变化。请参阅http://links.lww.com/DCR/B23.上的视频摘要
BACKGROUND: We demonstrated previously that radiation proctitis induced by preoperative radiotherapy is a predisposing factor for clinical anastomotic leakage in patients undergoing rectal cancer resection. Quantitative measurement of radiation proctitis is needed. OBJECTIVE: This study aimed to quantitate the changes of anatomic features caused by preoperative radiotherapy for rectal cancer and evaluate its ability to predict leakage. DESIGN: It was a secondary analysis of a randomized controlled trial (NCT01211210). MRI variables were retrospectively assessed. SETTINGS: The study was conducted in the leading center of the trial, which is a tertiary GI hospital. PATIENTS: Patients undergoing preoperative chemoradiation with sphincter-preserving surgery were included. MAIN OUTCOME MEASURES: Anatomic features were measured by preradiotherapy and postradiotherapy MRI. Univariate analyses were used to identify prognostic factors. Receiver operating characteristic curves were constructed to determine the cutoff value of the changes of MRI variables in predicting leakage. RESULTS: Eighteen (14.4%) of the 125 included patients developed clinical anastomotic leakage. Baseline characteristics were comparable between leakage group and nonleakage group. Relative increments of width of presacral space, thickness of rectal wall, and distal end of sigmoid colon discriminate between the 2 groups better than random chance. Relative increments of width of presacral space was the best performing predictor, with area under the curve of 0.722, sensitivity of 66.7%, specificity of 72.0%, and positive and negative predictive value of 28.6% and 92.8%. LIMITATIONS: The study was limited by its small sample size and retrospective design. CONCLUSIONS: Increments of the width of the presacral space, thickness of rectal wall, and distal part of the sigmoid colon helps to identify individuals not at risk for clinical anastomotic leakage after rectal cancer resection. The first variable is the strongest predictor. Changes of these variables should be taken into consideration when evaluating the application of defunctioning stoma. See Video Abstract at http://links.lww.com/DCR/B23.