Letter: Incidence and Predictors of Major Gastrointestinal Bleeding in Patients on Aspirin, Low-Dose Rivaroxaban or the Combination: Secondary Analysis of the COMPASS Randomised Controlled Trial

Abstract

Editors,

Forbes et al. have explored gastrointestinal bleeding (GIB) risks with aspirin, low-dose rivaroxaban and their combination [1]. Although the study offers valuable insights, we identify several critical limitations.

First, the study failed to address the long-term consequences of GIB. Although the COMPASS trial had extended follow-up, the analysis focuses primarily on acute bleeding events, neglecting long-term outcomes such as recurrent bleeding, anaemia or need for invasive interventions. This omission limits the clinical relevance for managing high-risk patients [2, 3].

Second, the study population's heterogeneity impacts the generalisability of results. The inclusion of patients with varying gastrointestinal risks, from those with a history of peptic ulcer disease to those without, raises concerns. Detailed subgroup analyses, especially for high-risk patients with prior GIB or concurrent NSAID or corticosteroid use, would provide clearer safety profiles. Without these distinctions, the safety conclusions may be misleading for certain patient groups [4].

A significant limitation is the inadequate evaluation of proton pump inhibitor (PPI) use. Although the study mentioned PPI use, it did not analyse its impact on bleeding outcomes compared with non-users. Given PPIs' role in reducing bleeding risk associated with antithrombotic therapy, a detailed examination of this factor is missing. Future research should address PPI-anticoagulant interactions more rigorously [5].

Additionally, the fixed dosing regimens used do not reflect real-world practices, where doses are adjusted based on individual patient factors such as renal function and body weight. This difference may exaggerate observed bleeding risks, as personalised dosing could mitigate these risks [6].

Finally, the study's analysis of bleeding risk predictors was overly simplistic. It included factors like age and medication use but neglected other relevant variables such as lifestyle factors (e.g., alcohol consumption), socioeconomic status or genetic predisposition. A broader range of predictive factors would enhance risk stratification and clinical utility [7].

In summary, although the study has provided important insights into bleeding risks with aspirin and low-dose rivaroxaban, it has limitations in long-term outcome analysis, subgroup analysis, PPI use, dosing adjustments and risk predictors. Addressing these gaps would strengthen the study's conclusions and improve clinical guidance.

Yu Yang: conceptualization, writing – original draft, writing – review and editing, formal analysis. Yaling Li: conceptualization, formal analysis, writing – review and editing. Jun Li: conceptualization, formal analysis, writing – review and editing.

The authors declare no conflicts of interest.

This article is linked to Forbes et al papers. To view these articles, visit https://doi.org/10.1111/apt.18139 and https://doi.org/10.1111/apt.18317.