Diagnostic accuracy of magnifying chromoendoscopy in the assessment of tumor invasion depth in early colorectal cancer: a systematic review and meta-analysis

PurposeThe aim of the study was to evaluate the ability of magnifying chromoendoscopy (MCE) to correctly differentiate early colorectal cancer (CRC) lesions with massively invasive submucosal cancer (SMm) from lesions without submucosal massive invasion (polyp, adenoma, dysplasia, intramucosal cancer, slightly invasive submucosal cancer (SMs)).MethodsWe searched PubMed, Embase, the Cochrane Library from the time of the establishment of each database to 5 April 2023. Stata 15 software was used to perform the meta-analysis for sensitivity, specificity, positive likelihood ratio (LR), and negative LR, diagnostic odds ratio, and 95% CI. A summary receiver-operating characteristic (SROC) curve was constructed, the area under the curve (AUC) was calculated, and the diagnostic value was evaluated. Furthermore, to explore the potential sources of heterogeneity, we used meta-regression to estimate the influencing factors of these studies and their impact on the diagnostic accuracy. MCE was used to evaluate the diagnostic accuracy in differentiating CRC lesions with SMm from lesions without submucosal massive invasion (polyp, adenoma, dysplasia, intramucosal cancer, SMs). Subgroup analysis was conducted as well. Deeks' funnel plots were also used to assess publication bias.ResultsA total of 11,387 colorectal lesions were included in 19 articles, including polyp, adenoma, dysplasia, and early cancer (intramucosal cancer, SMs, and SMm). The aggregate sensitivity, specificity, positive LR, negative LR, and diagnostic advantage scores of MCE in the diagnosis of differentiating CRC lesions with SMm from lesions without submucosal massive invasion (polyp, adenoma, dysplasia, intramucosal cancer, SMs) were 0.78 (95% CI 0.72-0.83), 0.95 (0.95% CI 0.91-0.97), 15.4 (0.95% CI 8.7-27.4), 0.23 (0.95% CI 0.18-0.30), and 66 (0.95% CI 32-136), respectively. The AUC of the SROC curve was 0.91 (0.95% CI 0.88-0.93). No significant publication bias was found with Deeks' funnel plot. The results showed significant heterogeneity due to the different objects included.ConclusionMCE can differentiate CRC lesions with SMm from lesions without submucosal massive invasion (polyp, adenoma, dysplasia, intramucosal cancer, SMs) with high accuracy and it can guide assessment of invasion depth of SMm in T1 early CRCs to help us select the most appropriate treatment.