Introduction Endoscopic mucosal resection (EMR) is now widely used for treatment of sessile and flat colonic adenomas. En-bloc resection is believed to have many theoretical advantages including better histological assessment and lower recurrence rates. There are limited data directly comparing the polyp recurrence rates following endoscopically complete resection by either piecemeal or en-bloc resection techniques.

Methods The aim of the study was to evaluate the incidence density (polyp recurrence) after piecemeal or en-bloc resection and to calculate the adjusted incidence rate ratio between these two methods. All patients undergoing EMR of flat or sessile colonic polyps >10 mm were prospectively audited in a single tertiary centre hospital. Duration of follow-up was calculated from the time of EMR to the date of the last follow-up colonoscopy. Follow-up colonoscopies were determined by the British Society of Gastroenterology guidelines. Incidence density of polyp recurrence was calculated in patients undergoing either piecemeal or en-bloc resection and a Poisson distribution was assumed for the purposes of calculation. Analysis for incidence rate ratio was adjusted for the site and size of lesion, use of argon plasma coagulation (APC), morphology of the lesion and histology of lesion. Stata V.9.1 was used for the analysis.

Results EMR was performed on 154 polyps >10 mm in size in 130 patients (69 males, mean age 68). 79 polyps with a mean size of 22.6±9 mm were resected piecemeal and 75 polyps with a mean size of 14.6±5.5 mm were resected en-bloc. There were nine recurrences in the piecemeal group and one recurrence in the en-bloc group. The incidence density (ID) of polyp recurrence in the piecemeal group was 80.2 per 1000 person years of follow-up (95% CI 36.7 to 152.2). The ID of polyp recurrence in the en-bloc group was 10.7 per 1000 person years of follow-up (95% CI 0.3 to 59.6). The adjusted incidence rate ratio (IRR) for recurrence in the en-bloc compared to piecemeal group was 0.42 (95% CI 0.22 to 0.78). Among patients undergoing piecemeal resection, the adjusted IRR for additional use of APC was 0.7 (95% CI 0.38 to 1.27), for histologically less advanced lesions was 0.3 (95% CI 0.14 to 0.61) and for sessile but not flat lesions was 0.13 (95% CI 0.19 to 0.75).

Conclusion The adjusted recurrence rates after en-bloc resection of colonic polyps are 60% less than that after piecemeal resection. The use of APC did not significantly change the recurrence rates after piecemeal resection, but flat lesions and histologically advanced lesions made recurrences more likely.