Routine continuation of high-dose antithrombotic therapy after cold snare polypectomy <10 mm: are the data sufficient for a practice recommendation?

Letter to: Delayed post-polypectomy bleeding following cold snare polypectomy of lesions <10 mm in patients on high-dose antithrombotic therapy: insights from a Dutch colonoscopy cohort Endoscopy 2026; 58(04): 384-396 DOI: 10.1055/a-2721-3151 10.1055/a-2721-3151 Van Bokhorst et al. report very low rates of delayed post-polypectomy bleeding (DPPB) after cold snare polypectomy of lesions <10 mm and suggest routine continuation of high-dose antithrombotic therapy (ATT) for colonoscopy indications with a low likelihood of advanced polyps [ 1 ]. We believe this practice-level conclusion exceeds the strength of the data. First, the recommendation rests on only 2 DPPB events among 343 routine diagnostic colonoscopies performed under continued high-dose ATT. The most clinically relevant comparison – continued versus discontinued high-dose ATT for routine diagnostic colonoscopies with only low-risk polypectomies – was 2/343 versus 0/180 ( P = 0.55), which is too sparse to support a practice change [ 1 ]. Second, continued high-dose ATT is inseparable from participating center. In routine diagnostic colonoscopy, continuation occurred in center A, whereas center B discontinued or downgraded high-dose ATT by protocol. Although center was included as a covariate, the authors also state that low DPPB counts precluded multivariable models with more than two variables; only three DPPB events occurred among all colonoscopies with polypectomy in continued high-dose ATT users [ 1 ]. Under such conditions, adjustment cannot neutralize site-level confounding. Third, the comparison groups were not clean. Patients who discontinued high-dose ATT were reassigned to the no-ATT group, and those downgraded were reassigned to the low-dose ATT group. The resulting contrasts therefore compare mixtures partly created by periprocedural management itself, not baseline treatment strategies [ 1 ]. We acknowledge the authors’ broader composite harm argument, which incorporates thromboembolic events and additional procedures. However, that trade-off rests on the same fragile base: only 2 thromboembolic events, none under continued high-dose ATT, and center-level, nonrandomized protocols that also shape the estimated repeat-procedure burden [ 1 ]. These data are reassuring and hypothesis-generating. They do not, however, justify routine continuation of high-dose ATT as a practice recommendation. Publication History Article published online: 22 June 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany Letter to this article: Reply to Qin et al. Endoscopy 2026; 58(07): 804-805 DOI: 10.1055/a-2863-8492