Society: ASGE

Introduction: Recent data suggested that cold snare EMR (C-EMR) offers equal efficacy, yet superior safety, compared with traditional hot EMR (H-EMR). We performed a systematic review and meta-analysis to assess the safety of C-EMR compared with H-EMR.

Methods: We performed a comprehensive literature search of MEDLINE (Ovid), Web of Science, Embase, Cochrane Library and CENTRAL, and World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) from inception to September 19, 2022. We included studies of endoscopic mucosal resection of colorectal polyps and/or polypectomy of polyps ≥10 mm and reported rates of adverse events including bleeding, perforation, and post polypectomy syndrome. The primary outcome was the adverse event rate for C-EMR vs. H-EMR for colorectal polyps ≥10mm. For comparative data, we reported odds ratios with 95% confidence intervals (CI’s). For cohort studies, we reported proportions with CI’s. We assessed publication bias by funnel plots with the classic fail-safe test. We used forest plots to report pooled effect estimates and assessed heterogeneity using I² and p-values.

Results: Our systematic review identified 1,215 unique citations, 19 of which met our inclusion criteria. In 4 comparative studies (2 randomized control trials and 2 retrospective studies), 413 patients underwent C-EMR, only two of whom suffered delayed bleeding. 35 of the 658 patients undergoing H-EMR suffered delayed bleeding. On random effect modelling, C-EMR carried a significantly lower risk of delayed bleeding compared to H-EMR (OR 0.2 [CI: 0.05 - 0.88], p=0.033, I² = 0%, Fig. 1). In the 10 cohort studies, the pooled rate of delayed bleeding was only 2% (CI: 1.3 – 3.2%), p<0.001, Fig. 2.

The rate of early bleeding was similar between both groups (OR 1.67, [CI: 0.77 - 3.6], p=0.1916, I² = 26%). On random effect modeling, the pooled rate of early bleeding for C-EMR was 1.9% (CI: 1.1% - 3.2%), p<0.001, I² = 12%.

Only one study by Rex et al. reported residual polyps. Margins were positive in 1 of 82 polyps resected by C-EMR and 4 of 65 resected by H-EMR. In the two comparative studies assessing polyp recurrence, there was a trend towards higher recurrence rates in the C-EMR cohort, but it did not reach statistical significance (OR 0.55 [CI: 0.29 - 1.03], p=0.0631, I² = 0).

There were no cases of perforation from C-EMR vs. 16 perforations from H-EMR. This difference did not reach statistical significance. (OR 0.02 [CI: 0.0 – 2.03], p=0.0995).

Conclusions: C-EMR is associated with equal efficacy compared with H-EMR, with similarly low rates of residual polyp tissue and polyp recurrence. The safety of C-EMR, however, is superior, with significantly lower rates of delayed bleeding than H-EMR. Rates of early bleeding are similar between the two methods.

Figure 1. Delayed bleeding risk for C-EMR vs H-EMR

Figure 1. Delayed bleeding risk for C-EMR vs H-EMR

Figure 2. Pooled rate of delayed bleeding from C-EMR

Figure 2. Pooled rate of delayed bleeding from C-EMR

William King

Appearances