Objective: To evaluate the overall diagnostic value of magnetic resonance imaging (MRI) in restaging of rectal cancer after preoperative chemoradiotherapy based on qualified studies.
Methods: PubMed, Cochrane, and EMBASE database were searched by the index words to identify the qualified studies, and relevant literature sources were also searched. The latest research was done in April 2019. Heterogeneity of the included studies was tested, which was used to select proper effect model to calculate pooled weighted sensitivity, specificity, and diagnostic odds ratio (DOR). Summary receiver operating characteristic (SROC) analyses were also performed.
Result: Nineteen studies with 1262 patients were involved in the meta-analysis exploring the diagnostic accuracy of MRI for rectal cancer. The diagnostic accuracy of MRI in T3-T4 rectal cancer was as follows: sensitivity, 81% (95% confidence interval [CI], 67%-90%); specificity, 67% (95% CI, 51%-80%); positive likelihood ratio, 2.48 (95% CI, 1.57-3.91); negative likelihood ratio, 0.28 (95% CI, 0.15-0.52); global DOR, 6.86 (95% CI, 3.07-15.30); the area under the SROC was high (0.81; 95% CI, 0.78-0.84). The diagnostic accuracy of MRI in lymphatic metastasis of rectal cancer was as follows: sensitivity, 77% (95% CI, 65%-86%); specificity, 77% (95% CI, 63%-87%); positive likelihood ratio, 3.40 (95% CI, 2.07-5.59); negative likelihood ratio, 0.30 (95% CI, 0.20-0.45); DOR, 10.81 (95% CI, 4.99-23.39); area under the SROC was high (0.84; 95% CI, 0.80-0.87).
Conclusions: This study provides a systematic review and meta-analysis of diagnostic accuracy studies of MRI for rectal cancer. The results indicate that MRI is a highly accurate diagnostic tool for rectal cancer T3-T4 staging and N staging but sensitivity and specificity are not high.