Risk factors for intra and post-procedural bleeding following endoscopic mucosal resection of nonpedunculated colorectal lesions

Aims Post-polypectomy bleeding is one the most common complications of endoscopic mucosal resection (EMR). We aimed to assess the incidence of intraprocedural bleeding (IPB) and post-procedural bleeding (PPB) following EMR of nonpedunculated colorectal lesions and identify potential risk factors.

Methods Retrospective single center cohort study. All consecutive patients that underwent EMR in a non-tertiary hospital from January 2020 to September 2023 were included. Demographic information, clinical variables, characteristics of colorectal lesions and procedural aspects were collected. Possible risk factors were assessed using logistic regression analysis.

Results 302 patients (61% male; mean age: 68.9±9.3 years) with a total of 410 colorectal lesions underwent EMR. The mean lesion size was 20.9±10.5mm, with over half (51%) exceeding 20mm, and the majority (57%) were located in the right colon. En bloc resection was achieved in 47% of cases. Regarding morphology, 58% (n=238) were classified as Paris 0-IIa and 30% (n=122) as lateral spread tumors.

IPB and PPB occurred in 8.3% and 3.9% of procedures, respectively, with none being clinically significant. Of these, 38% required endoscopic therapy with snare tip soft coagulation or clip placement. Although right colon location showed a higher incidence of IBP and PPB, these differences were not statistically significant (p=0.356 and p=0.569, respectively).

Larger lesions were associated with a higher incidence of IBP (p<0.05), with size>20mm harboring the highest rate (p=0.039. OR 2.159. IC 1.023-4.554). Higher SMSA (p=0.001) and SERT (p<0.005) scores were also associated with an increased incidence of IPB. Multivariate logistic regression analysis identified size as an independent risk factor for IPB (p=0.015).

Patients medicated with antiplatelet (p<0.05. OR: 7.924. IC 2.415-25.996) and anticoagulant (p<0.05. OR 10.697. IC 3.366-33.994) drugs exhibited higher rates of PPB, accounting for 69% of cases. Larger lesions were also associated with a higher incidence of PPB, though not statistically significant (p=0.337). Prophylactic soft coagulation of visible scar vessels was associated with a reduced risk of PPB (p=0.013. OR 2.420. IC 1.233-9.484). Mechanical prophylaxis with clip placement, performed in 19% of cases (71% in the right colon), did not reduce the risk of PPB (p=0.992). Multivariate logistic regression analysis confirmed anticoagulation as an independent risk factor for IPB (p=0.028).

Conclusions Lesion size, SMSA and SERT scores were associated with a higher risk of IPB. Patients on antiplatelet and anticoagulant drugs are at higher risk of PPB. Prophylactic endoscopic coagulation of nonbleeding visible vessels within the mucosal defect reduced the risk of PPB and may be a viable strategy to decrease PPB in patients taking anticoagulants.

 

MissingLinkText

Conflicts of interest

Authors do not have any conflict of interest to disclose.

Publication History

Article published online:
15 April 2024