磁共振成像和腔内超声联合血清癌胚抗原检查对直肠癌术前精准分期的意义

Clinical significance of MRI and transrectal ultrasound combined with carcinoembryonic antigen examination on accurate preoperative staging of rectal carcinoma

目的研究高分辨率磁共振成像（MRI）加直肠腔内超声（TRUS）联合血清癌胚抗原（CEA）检查对直肠癌术前TNM分期和术后病理pTNM分期诊断的一致性。方法经手术治疗后病理证实的156例直肠癌患者按随机数字表法分为术前MRI组、TRUS组、MRI加TRUS组、MRI加TRUS联合血清CEA组各39例，回顾性分析4组术前T、N分期与术后病理T、N分期的诊断一致性。结果4组直肠癌患者术前T分期与病理T分期诊断一致性差异有统计学意义（T分期：Kappa=0．685，P=0．000；N分期：Kappa=0．544，P=0．000），MRI组、TRU组、MRI加TRUS组术前N分期和病理N分期诊断一致性差异无统计学意义（Kappa=0．142，P=0．329；Kappa=0．154，P=0．645；Kappa=0．154，P=0．229）；而MRI加TRUS联合血清CEA组术前N分期和病理N分期诊断一致性差异有统计学意义（Kappa=0．544，P=0．000）。MRI加TRUS联合CEA组与MRI加TRUS组、MRI组和TRUS组术前T分期的准确度比较，差异均无统计学意义（χ^2=0．326，P=0．574；χ^2=0．562，P=0．719；χ^2=0．287，P=0．986），N分期准确度之间的差异有统计学意义（χ^2=4．643，P：0．026；χ^2=6．643，P：0．026：χ^2=5．243，P=0．019）。结论MRI加TRUS联合CEA检查可提高直肠癌术前分期的准确性，为手术决策提供更可靠的依据，提高手术方案预测的符合率，同时也为直肠癌术前精确诊断和个体化治疗提供可靠依据。

Objective To study the consistency in the diagnosis of preoperative TNM rectal cancer staging using high resolution magnetic resonance imaging （MR1） and transrectal ultrasound （TRUS） combined with carcinoembryonic antigen （CEA） and postoperative pathological TNM. Methods 156 cases pathologically proven were retrospectively analyzed and divided into 4 groups including preoperative MRI group （39 cases）, TRUS group （39 cases）, MRI and TRUS group （39 cases）, MRI and TRUS combined with CEA group （39 cases）. The differences between preoperative T, N staging and postoperative pathologic T, N staging were analyzed. Results There were statistically significant differences in the diagnosis of preoperative T and postoperative pathological T in 4 groups （T： Kappa = 0.685, P = 0.000; N： Kappa = 0.544, P = 0.000）, but there were no significant differences in preoperative N and postoperative pathological N staging in preoperative MRI group, TRUS group, MRI and TRUS group （Kappa = 0.142, P = 0.329; Kappa = 0.154, P = 0.645; Kappa = 0.154, P= 0.229）, and significant difference was observed in MRI and TRUS combined with CEA group （Kappa = 0.544, P = 0.000）. There were no significant differences in the accuracy of T staging among the 4 groups （Xa = 0.326, P = 0.574; χ^2 = 0.562, P = 0.719; χ^2 = 0.287, P = 0.986）, but significant difference in the accuracy of N staging were showed among the 4 groups （χ^2 = 4.643, P = 0.026; χ^2 = 6.643, P = 0.026; χ^2 = 5.243, P = 0.019）. Conclusion Preoperative evaluation by the MRI add TRUS combined with CEA can improve the accuracy of preoperative staging, which can provide more reliable basis for decision-making and improve the coincidence rate of operative procedures in line with the estimate. It also provides the basis for the accurate preoperative diagnosis and individualized treatment.