Objectives This systematic review and meta-analysis aimed to assess the diagnostic accuracy of CT in differentiating high-risk from low-risk colon cancer, with a focus on staging parameters and the impact of CT slice thickness. Materials and methods A systematic search of Ovid MEDLINE and Embase.com was conducted from January 1, 2015, to September 24, 2024, to identify studies evaluating CT-based staging accuracy using histopathology as the reference standard. The QUADAS-2 tool assessed the risk of bias. Pooled sensitivity, specificity, and diagnostic odds ratio (DOR) were calculated using a bivariate random-effects model. Subgroup analyses explored the influence of CT techniques, slice thickness, and study design on diagnostic performance. Results The meta-analysis included forty-four studies. CT demonstrated 83% sensitivity (95% CI, 79–86%) and 70% specificity (95% CI, 66–74%) for detecting pT3-T4 tumors (DOR: 10.0). For pT3cd-T4 (> 5 mm muscularis propria invasion), sensitivity was 67% (61–73%), specificity 88% (83–92%) and DOR 13.7 (9.0–21.0). Detection of pN+ yielded 64% sensitivity (60–68%), 67% specificity (62–72%) and DOR of 3.5 (3.0–4.2). Sensitivity for extramural venous invasion (EMVI+) was 49% (41–56%), with 77% specificity (67–84%) and DOR 3.0 (2.0–4.4). Studies with < 5 mm slice thickness showed higher sensitivity but lower specificity. High I² values indicated substantial heterogeneity across studies. Conclusion CT demonstrates high sensitivity for detecting T3-T4 colon cancer but moderate sensitivity for nodal involvement and EMVI+. Diagnostic performance varies with technical factors, emphasizing the need for standardized imaging protocols and supplementary diagnostic tools to improve colon cancer staging. Registration PROSPERO (International Prospective Register of Systematic Reviews) CRD42022374615. Key Points <Emphasis Type="BoldItalic">Question</Emphasis> Accurate CT staging is crucial for guiding neoadjuvant therapy in colon cancer, but its ability to distinguish high-risk from low-risk cases remains uncertain. <Emphasis Type="BoldItalic">Findings</Emphasis> CT showed high sensitivity for distinguishing pT3-T4 tumors but only moderate sensitivity for pT3cd-T4, nodal involvement, and extramural venous invasion. <Emphasis Type="BoldItalic">Clinical relevance</Emphasis> This systematic review critically evaluates CT diagnostic accuracy in colon cancer staging, revealing its strengths and limitations. The findings highlight the need for optimized imaging protocols and complementary tools to enhance risk stratification and guide clinical decisions.

Diagnostic accuracy of CT for identifying high-risk colon cancer: a systematic review and meta-analysis

期刊论文
出版模式：

开放获取
发表日期：

2025年8月16日

查看全文

Jona Shkurti,
Kim van den Berg,
Renaud L. M. Tissier,
Stevie van der Mierden,
Max J. Lahaye,
Regina G. H. Beets-Tan,
Joost Nederend

摘要

Objectives

This systematic review and meta-analysis aimed to assess the diagnostic accuracy of CT in differentiating high-risk from low-risk colon cancer, with a focus on staging parameters and the impact of CT slice thickness.

Materials and methods

A systematic search of Ovid MEDLINE and Embase.com was conducted from January 1, 2015, to September 24, 2024, to identify studies evaluating CT-based staging accuracy using histopathology as the reference standard. The QUADAS-2 tool assessed the risk of bias. Pooled sensitivity, specificity, and diagnostic odds ratio (DOR) were calculated using a bivariate random-effects model. Subgroup analyses explored the influence of CT techniques, slice thickness, and study design on diagnostic performance.

Results

The meta-analysis included forty-four studies. CT demonstrated 83% sensitivity (95% CI, 79–86%) and 70% specificity (95% CI, 66–74%) for detecting pT3-T4 tumors (DOR: 10.0). For pT3cd-T4 (> 5 mm muscularis propria invasion), sensitivity was 67% (61–73%), specificity 88% (83–92%) and DOR 13.7 (9.0–21.0). Detection of pN+ yielded 64% sensitivity (60–68%), 67% specificity (62–72%) and DOR of 3.5 (3.0–4.2). Sensitivity for extramural venous invasion (EMVI+) was 49% (41–56%), with 77% specificity (67–84%) and DOR 3.0 (2.0–4.4). Studies with < 5 mm slice thickness showed higher sensitivity but lower specificity. High I² values indicated substantial heterogeneity across studies.

Conclusion

CT demonstrates high sensitivity for detecting T3-T4 colon cancer but moderate sensitivity for nodal involvement and EMVI+. Diagnostic performance varies with technical factors, emphasizing the need for standardized imaging protocols and supplementary diagnostic tools to improve colon cancer staging.

Registration

PROSPERO (International Prospective Register of Systematic Reviews) CRD42022374615.

Key Points

Question Accurate CT staging is crucial for guiding neoadjuvant therapy in colon cancer, but its ability to distinguish high-risk from low-risk cases remains uncertain.

Findings CT showed high sensitivity for distinguishing pT3-T4 tumors but only moderate sensitivity for pT3cd-T4, nodal involvement, and extramural venous invasion.

Clinical relevance This systematic review critically evaluates CT diagnostic accuracy in colon cancer staging, revealing its strengths and limitations. The findings highlight the need for optimized imaging protocols and complementary tools to enhance risk stratification and guide clinical decisions.