Preoperative localization of colorectal cancer: a systematic review and meta-analysis
Supplementary Materials
Supplementary Figure 1. Funnel plots for localization error rates
Supplementary Figure 2. Sensitivity analysis excluding studies with high risk of bias.
Supplementary Figure 3. Leave-one-out sensitivity analysis.
Supplementary Figure 4. Tumor localization error of conventional colonoscopy - Subgroup meta-analysis for rectal tumors.
Supplementary Figure 5. Tumor localization error of colonoscopic tattooing - Subgroup meta-analysis for circumferential versus conventional tattoo placement.
Supplementary Figure 6. Tumor localization error of colonoscopic tattooing - Subgroup meta-analysis for saline injection bleb technique versus conventional submucosal tattooing technique.
Supplementary Figure 7. Tumor localization error of colonoscopic tattooing - Subgroup meta-analysis for type of dye.
Supplementary Figure 1. Funnel plots for localization error rates for (A) conventional colonoscopy and (B) colonoscopic tattooing with sample size as vertical axis.
A
B
Supplementary Figure 2. Sensitivity analysis excluding studies with high risk of bias. Tumor localization error of (A) conventional colonoscopy and (B) colonoscopic tattooing. Point estimates are provided with 95% confidence intervals. The overall estimate was pooled using an inverse-variance weighted random-effects model following the conventional method of DerSimonian and Laird. The studies are arranged chronologically.
A
B
Supplementary Figure 3. Leave-one-out sensitivity analysis. Tumor localization error of (A) conventional colonoscopy and (B) colonoscopic tattooing after leaving out each one of the included studies.
Supplementary Figure 4. Tumor localization error of conventional colonoscopy - Subgroup meta-analysis for rectal tumors. Tumor localization error of conventional colonoscopy for studies including and excluding patients with rectal cancer. Point estimates are provided with 95% confidence intervals. The overall estimate was pooled using an inverse-variance weighted random-effects model following the conventional method of DerSimonian and Laird. The studies are arranged chronologically.
Supplementary Figure 5. Tumor localization error of colonoscopic tattooing - Subgroup meta-analysis for circumferential versus conventional tattoo placement. Point estimates are provided with 95% confidence intervals. The overall estimate was pooled using an inverse-variance weighted random-effects model following the conventional method of DerSimonian and Laird. The studies are arranged chronologically.
Supplementary Figure 6. Tumor localization error of colonoscopic tattooing - Subgroup meta-analysis for saline injection bleb technique versus conventional submucosal tattooing technique. Point estimates are provided with 95% confidence intervals. The overall estimate was pooled using an inverse-variance weighted random-effects model following the conventional method of DerSimonian and Laird. The studies are arranged chronologically.
Supplementary Figure 7. Tumor localization error of colonoscopic tattooing - Subgroup meta-analysis for type of dye. Point estimates are provided with 95% confidence intervals. The overall estimate was pooled using an inverse-variance weighted random-effects model following the conventional method of DerSimonian and Laird. The studies are arranged chronologically.