Abstract
Background: Early detection and prevention of colorectal cancer (CRC) are key goals of population-based screening. Several diagnostic tests have been proposed for CRC screening. This study compares the diagnostic accuracy of colon capsule endoscopy (CCE), colonoscopy (COL) and computed tomographic colonography (CTC), focusing on risk factors such as polyps. Methods: We conducted a systematic review with meta-analyses and network meta-analysis. Pooled estimates of sensitivity (SE), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV) were calculated using a random-effects model. Diagnostic performance was assessed for first- and second-level screening based on effect size estimates. Results: For first-level screening, sensitivity was 0.79 (95% CI: 0.60-0.91) and specificity 0.95 (95% CI: 0.88-0.98); PPV and NPV were 0.89 and 0.97, respectively. In second-level screening, sensitivity was 0.75 (95% CI: 0.65-0.83), specificity 0.95 (95% CI: 0.92-0.97), PPV 0.76 and NPV 0.95. The indirect sensitivity estimate of CCE vs. COL (SMD = 0.30; 95% CI: 0.12-0.47) was lower than the direct estimate for CTC (SMD = 0.44; 95% CI: 0.29-0.59). CCE showed better comparative performance than CTC relative to COL (CCE SMD = -0.18; 95% CI: -0.29 to -0.06 vs. CTC SMD = -0.98; 95% CI: -1.07 to -0.90). However, both CCE and CTC had lower specificity than COL. Conclusions: CCE represents a valuable tool for early CRC detection. Test selection should be guided by clinical and epidemiological settings to optimize screening strategies.