BACKGROUND: Despite improvements in technique, severe adverse events (SAE), including post-procedure bleeding, remain a major concern following endoscopic resection of large colorectal polyps. We examined whether cold resection without the use of electrocautery reduces the risk of SAE and affords completeness of resection.

METHODS: We performed a randomized trial of patients with a large nonpedunculated colon polyp (≥20 mm) at 10 medical centers in North America from October 2019 through January 2023. Patients were randomly assigned to endoscopic mucosal resection without electrocautery (cold EMR group) or with electrocautery (hot EMR group) and were followed until their first surveillance colonoscopy. Hot EMR included margin treatment and defect closure as indicated. The primary outcome were SAEs in intention to treat analysis, defined as an event that required hospitalization, a blood transfusion, colonoscopy, surgery, or another invasive intervention within 30 days after completion of the colonoscopy. The secondary outcome was the rate of polyp recurrence at surveillance colonoscopy. Because crossover from cold to hot EMR was expected (assumed 10%), we also performed a per protocol analysis.

RESULTS: 660 patients were randomized, and 518 (78.5%) completed their first surveillance colonoscopy. Crossover occurred in 14.6% in the cold EMR group and in 13.4% in the hot EMR group. An SAE was observed in 2.1% of patients in the cold EMR group and in 4.3% in the hot EMR group (p=0.62) with postprocedure bleeding in 0.9% and 2.8%, respectively (p=0.52). When the analysis was restricted to patients who received the intervention as randomized (per protocol analysis), significantly fewer SAEs occurred in the cold EMR group as compared to the hot EMR group (1.4% vs 4.9%, p=0.017), with postprocedure bleeding in 1.1% and 2.5%, respectively (p=0.34). Polyp recurrence was detected in 28.0% in the cold EMR group and in 14.2% in the hot EMR group (p<0.001). In the per protocol analysis recurrence rate were 28.7% and 15.4% (p<0.001), respectively. In subgroup analysis SAE risk differences between groups remained unchanged. However, risk of recurrence was similar for the subgroup of 20-29 mm polyps (18.5% vs 15.7%) and for sessile serrated polyps (15.0% vs 14.5%). The greatest recurrence risk was noted for adenomas with high grade dysplasia (46.5% vs 18.4%, respectively, p=0.004).

CONCLUSIONS: This large multicenter trial showed no significant safety benefit of universal cold EMR for all large colorectal polyps. However, after accounting for a high crossover rate, cold EMR had a significantly lower SAE rate compared to hot EMR. This safety benefit of cold EMR is offset by a greater recurrence rate, particularly for advanced pathology. Our findings do not support the universal application of cold EMR for large non-pedunculated colorectal polyps.

<b>Table 1.</b> Patient characteristics and severe adverse events

Table 1. Patient characteristics and severe adverse events

<b>Table 2. </b>Polyp and EMR characteristics and polyp recurrence

Table 2. Polyp and EMR characteristics and polyp recurrence

Jeremy Barber

Appearances