Zhongming Huang1, Lili Chu, Risheng Zhao, Hui Wang. 1. Department of Colorectal Surgery, The Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou 510655, China. wanghuislh@hotmail.com.
Abstract
OBJECTIVE: To estimate the diagnostic accuracy of magnetic resonance(MR) in restaging of rectal cancer after preoperative chemoradiotherapy(CRT). METHODS: Comprehensive search of literature concerning the diagnosis of MR for rectal cancer after preoperative CRT was performed from databases of PubMed, EMbase, OVID and WOK. Sensitivity and specificity of MR on restaging of rectal cancer after preoperative CRT were investigated by SAS and MetaDiSc software. RESULTS: Thirteen articles including 749 patients were enrolled in this meta-analysis. For T3-T4 stage, sensitivity of MR was 82.1%(95%CI:67.9%-90.9%), specificity was 53.5%(95%CI:39.3%-67.3%), and diagnostic odds ratio(DOR) was 5.34(2.73, 6.59). For lymph node involvement, sensitivity of MR was 61.8%(95%CI:50.7%-71.8%), specificity was 72.0%(95%CI:61.3%-80.7%), and DOR was 4.33(95%CI:2.84-6.59). For circumferential resection margin(CRM) by MR, pooled sensitivity was 85.4%(95%CI:60.5%-95.7%), specificity was 80.0%(95%CI:57.4%-92.3%), and DOR was 27.62(95%CI:13.03-58.55). CONCLUSIONS: Restaging accuracy of T3-T4 and lymph nodes involvement of rectal cancer after preoperative CRT by MR is not high. MR may be a good method to make reassessment of CRM. To avoid overtreatment for T0-T2, negative lymph node and circumferential resection of rectal cancer, restaging by MR after preoperative CRT is important.
OBJECTIVE: To estimate the diagnostic accuracy of magnetic resonance(MR) in restaging of rectal cancer after preoperative chemoradiotherapy(CRT). METHODS: Comprehensive search of literature concerning the diagnosis of MR for rectal cancer after preoperative CRT was performed from databases of PubMed, EMbase, OVID and WOK. Sensitivity and specificity of MR on restaging of rectal cancer after preoperative CRT were investigated by SAS and MetaDiSc software. RESULTS: Thirteen articles including 749 patients were enrolled in this meta-analysis. For T3-T4 stage, sensitivity of MR was 82.1%(95%CI:67.9%-90.9%), specificity was 53.5%(95%CI:39.3%-67.3%), and diagnostic odds ratio(DOR) was 5.34(2.73, 6.59). For lymph node involvement, sensitivity of MR was 61.8%(95%CI:50.7%-71.8%), specificity was 72.0%(95%CI:61.3%-80.7%), and DOR was 4.33(95%CI:2.84-6.59). For circumferential resection margin(CRM) by MR, pooled sensitivity was 85.4%(95%CI:60.5%-95.7%), specificity was 80.0%(95%CI:57.4%-92.3%), and DOR was 27.62(95%CI:13.03-58.55). CONCLUSIONS: Restaging accuracy of T3-T4 and lymph nodes involvement of rectal cancer after preoperative CRT by MR is not high. MR may be a good method to make reassessment of CRM. To avoid overtreatment for T0-T2, negative lymph node and circumferential resection of rectal cancer, restaging by MR after preoperative CRT is important.