Warfarin or DOAC use prior to gastrointestinal bleeding may reflect a more complex care pathway rather than intrinsically more severe bleeding, highlighting potential exposure misclassification.
We read with interest the research communication by Holtslander and Rockey examining outcomes of gastrointestinal bleeding (GIB) in patients receiving warfarin or direct oral anticoagulants (DOACs) 1. The study addresses an important clinical question, because anticoagulant-associated GIB is increasingly encountered in routine practice and is often perceived as intrinsically more severe. We agree that anticoagulated patients may experience worse outcomes. However, the reported findings seem better interpreted as associations between recent anticoagulant prescribing status and a more complex GIB care pathway, rather than as evidence that confirmed active anticoagulation directly causes intrinsically more severe bleeding. First, exposure ascertainment warrants caution. In the study, anticoagulated patients were defined by a prescription within 1 month before GIB, whereas non-anticoagulated patients had no prior warfarin or DOAC prescription 1. This approach is pragmatic, but recent prescribing does not necessarily equate to active anticoagulant effect at presentation. Without information on adherence, last-dose timing, reversal treatment, coagulation-related laboratory indices or contemporaneous renal function at presentation, clinically relevant exposure misclassification remains possible. The comparison may therefore be most robustly interpreted as one of recent prescribing status rather than confirmed on-treatment anticoagulation biology. Second, the signal for more endoscopy at 2–5 days should not be equated too readily with delayed endoscopy. Later endoscopy in anticoagulated patients may plausibly reflect pathway complexity, including early haemodynamic stabilisation, transfer, procedural caution, reversal strategies or survival to a later intervention window, rather than a simple delay in care 2, 3. In this context, endoscopy timing is itself partly downstream of early clinical decision-making. A time-dependent approach would therefore be important before linking later endoscopy to poorer outcomes. Third, transfusion, rebleeding and mortality are clinically important outcomes, but they are also shaped by management decisions and case-mix, not solely by intrinsic bleeding severity. Although the authors matched for a broad range of diagnostic comorbidities, residual confounding by indication, management, unmeasured differences in hematologic reserve and initial bleeding severity is likely to remain substantial in electronic health record-based analyses 4, 5. It would therefore be helpful if the authors clarified whether their findings should be framed more cautiously as pathway-associated rather than severity-specific, and whether outcomes differ across individual DOAC agents rather than the DOAC class as a whole. These clarifications would strengthen the clinical interpretation of this useful study and may help refine future analyses of anticoagulant-associated GIB. Jiaqi Shen: conceptualization, investigation, writing – original draft. Fangfang Huang: writing – review and editing, supervision. The authors have nothing to report. The authors declare no conflicts of interest. This article is linked to Holtslander et al paper. To view this article, visit https://doi.org/10.1111/apt.70329. The authors have nothing to report.
Shen et al. (Mon,) conducted a letter in Gastrointestinal bleeding. Warfarin or DOACs vs. No prior warfarin or DOAC prescription was evaluated. Warfarin or DOAC use prior to gastrointestinal bleeding may reflect a more complex care pathway rather than intrinsically more severe bleeding, highlighting potential exposure misclassification.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: