Not All Endoscopic Submucosal Dissections Are Created Equal: Rethinking Bleeding Risk Stratification During Antiplatelet Therapy

Abstract

Endoscopic submucosal dissection (ESD) has revolutionized the management of superficial gastrointestinal neoplasia by providing organ-preserving treatment with excellent oncological outcomes. At the same time, the aging of the population and the increasing prevalence of cardiovascular disease have led to a growing number of patients undergoing ESD while receiving antiplatelet therapy. Consequently, endoscopists are increasingly confronted with one of the most challenging aspects of therapeutic endoscopy: balancing the risk of post-procedural bleeding against the potentially serious consequences of thromboembolic events.

Current international guidelines provide an invaluable framework by classifying ESD among high-risk endoscopic procedures and recommending peri-procedural management of antithrombotic therapy accordingly [1, 2]. These recommendations have undoubtedly improved patient safety and standardized clinical practice. Nevertheless, guideline classifications inevitably represent a simplification of a far more heterogeneous clinical reality. While guidelines classify procedures, clinicians ultimately treat individual patients. Bridging these two perspectives remains one of the greatest challenges in contemporary therapeutic endoscopy.

Perhaps the question is no longer simply how antiplatelet therapy should be managed during ESD, but whether all ESD procedures truly share the same intrinsic bleeding risk.

In this issue of Digestive Endoscopy, Hirai et al. [3] report the largest multicenter experience to date evaluating peri-procedural antiplatelet management in patients undergoing ESD for esophageal squamous cell carcinoma. Importantly, this study does not attempt to demonstrate the superiority of one antiplatelet strategy over another. Rather, it provides carefully collected real-world data in an area where evidence has long remained remarkably limited.

The authors observed low rates of delayed bleeding irrespective of whether thienopyridine therapy was managed according to guideline-based discontinuation or continued throughout the peri-procedural period. Owing to the retrospective design and the very limited number of bleeding events, these findings cannot be interpreted as evidence supporting continued therapy, nor do they justify changes in current recommendations. Instead, they should be viewed as an important contribution to an evolving clinical discussion [3].

Indeed, the greatest value of this study may not reside in the absolute bleeding rate itself, but in the clinical question that naturally emerges from these observations.

Are all ESD procedures really equal? Therapeutic endoscopy has traditionally approached ESD as a single procedural entity when considering peri-procedural bleeding risk. Such an approach is understandable and has facilitated the development of practical recommendations. However, accumulating evidence suggests that delayed bleeding following ESD is not uniformly distributed across the gastrointestinal tract. Previous studies have consistently reported remarkably low rates of delayed bleeding after esophageal ESD for squamous cell carcinoma, and the present multicenter experience further reinforces this observation in patients receiving antiplatelet therapy [4-6].

This distinction deserves careful consideration. Indeed, several predictive models and systematic reviews have demonstrated that bleeding risk after ESD depends on the interaction of multiple patient- and procedure-related factors rather than on the procedure itself [7, 8]. Risk stratification is rarely determined by a single variable. Instead, it reflects the interaction between patient characteristics, thromboembolic burden, procedural complexity, operator expertise, and the biological features of the organ involved. Consequently, considering all ESD procedures as carrying an identical bleeding risk may represent an oversimplification of an increasingly complex clinical landscape.

Importantly, this concept should not be interpreted as a challenge to current guidelines. Rather, it reflects the natural evolution of evidence-based medicine. Guidelines necessarily rely on the best available evidence at a given time, whereas clinical research progressively identifies situations in which more refined stratification may become appropriate.

Another particularly interesting aspect of the present study is that it captures contemporary practice across multiple expert Japanese centers. A substantial proportion of patients underwent esophageal ESD without interruption of antiplatelet therapy, illustrating that individualized decision-making is already occurring in carefully selected clinical scenarios.

Future research should move beyond simply asking whether antiplatelet therapy can be continued during ESD. A more fundamental question may be whether different ESD procedures deserve distinct models of procedural risk stratification. Organ-specific evidence, together with patient-related thromboembolic risk and procedural complexity, may ultimately allow a more individualized approach to peri-procedural antiplatelet management.

The study by Hirai and colleagues does not redefine current clinical practice, nor was it designed to do so. However, it successfully challenges an assumption that has largely remained unquestioned: that all ESD procedures should necessarily be regarded as carrying the same intrinsic bleeding risk.

Whether organ-specific bleeding risk should eventually influence peri-procedural antiplatelet management remains unanswered. Nevertheless, the present study provides an important step toward addressing this question and reminds us that the future of therapeutic endoscopy will likely depend not only on improving procedural techniques but also on refining how procedural risk itself is defined.

Antonio Capogreco: conceptualization; writing – review and editing. Francesco Minini: conceptualization; writing – original draft and editing. Both authors approved the final version of the manuscript and agree to be accountable for all aspects of the work.

The authors have nothing to report.

The authors declare no conflicts of interest.

This article is linked to Hirai et al. To view this article, visit https://doi.org/10.1111/den.70206.

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.