Gastrointestinal Bleeding Risk Associated With Pulse Pressure in Patients With Atrial Fibrillation: A Retrospective Cohort Study (Preprint)

Q: What is the clinical evidence from this study?

Study design: Cohort. Population: Atrial fibrillation (n=4935). Intervention: Pulse pressure vs. Lower pulse pressure. Primary outcome: Gastrointestinal bleeding (OR 1.013 per mmHg, p=0.028).

Q: What does this research mean for the field?

Elevated pulse pressure independently predicts gastrointestinal bleeding in anticoagulated atrial fibrillation patients, with each 1 mmHg increase raising the risk (OR 1.013). Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Michel Abou Khalil; Carlo Khoury; Maximilian Moersdorf; Christian Massad; A Darzi; Yara Menassa; Mohammad Montaser Atasi; Ghassan Bidaoui; Hadi Younes; Ala Assaf; Mario Mekhael; Eli Tsakiris; Chanho Lim; Charbel Noujaim; Abboud Hassan; Yingshuo Liu; Yishi Jia; Mary Margot M Maleckar; Rodolfo A Montiel Quintero; Omar Kreidieh; Amitabh C. Pandey; Feng Han; Qussay Marashly; Nassir F. Marrouche

doi:10.2196/93347

Key result

Elevated pulse pressure predicts ~1% greater gastrointestinal bleeding risk per mmHg in anticoagulated AF.

OR 1.013 per mmHg
P=0.028
n=4,935

Figures

Why the study?

Anticoagulated patients with AF face significant bleeding risks that current risk scores inadequately predict, and pulse pressure may offer additional prognostic value.

Does elevated pulse pressure predict major bleeding events in anticoagulated patients with atrial fibrillation?

Key Points

The study evaluates whether elevated pulse pressure independently predicts major bleeding events in atrial fibrillation patients on anticoagulation.
Retrospective cohort study of 4,935 AF patients on oral anticoagulation from 2010 to 2019.
Pulse pressure analyzed from outpatient blood pressure readings in tertiles and as a continuous variable.
Kaplan-Meier curve and Cox regression models assessed the association between pulse pressure and bleeding events.
13.7% of patients experienced major bleeding over 5 years, with significant GI bleeding in the highest pulse pressure tertile (p = 0.007).
Each 10 mmHg increase in pulse pressure correlated with a 15% higher risk of GI bleeding (HR: 1.014; p = 0.042).
Pulse pressure was not significantly associated with intracranial or other bleeding types.

Study Design

Type

Cohort (n=4,935)

Multicenter

Yes

Structured PICO

Does elevated pulse pressure predict major bleeding events in anticoagulated patients with atrial fibrillation?

P

Population

4,935 patients with atrial fibrillation on oral anticoagulation (2010–2019) from the REACHnet network

I

Intervention

Elevated pulse pressure (analyzed in tertiles and as a continuous variable)

C

Comparator

Lower pulse pressure

O

Outcome

Major bleeding events (including gastrointestinal, intracranial, and other bleeding types)safety

Main Result

Effect estimate: OR 1.013 per mmHg

p-value: p=0.028

Pulse pressure independently predicts gastrointestinal bleeding in anticoagulated patients with atrial fibrillation, suggesting it could improve current bleeding risk stratification models.

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Should not yet change practice for bleeding risk assessment in anticoagulated patients; leaves open whether pulse pressure is a modifiable predictor.

Abstract

Background: Anticoagulated patients with atrial fibrillation (AF) face significant bleeding risks, which current risk scores inadequately predict. Pulse pressure (PP), a marker of arterial stiffness, may offer additional prognostic value. Objective: This study aimed to evaluate whether elevated PP independently predicts major bleeding events. Methods: We conducted a retrospective cohort study using electronic health records from 4,935 AF patients on oral anticoagulation (2010–2019) in the REACHnet network. PP was calculated from outpatient blood pressure readings and analyzed in tertiles and as a continuous variable. Kaplan-Meier curve and log-rank test were conducted to assess the association between PP and clinical outcomes. Cox regression models further adjusted for demographics, comorbidities, systolic blood pressure, medications, and the ORBIT bleeding score. Results: Over a median 5-year follow-up, 677 patients (13.7%) experienced major bleeding. GI bleeding was significantly more frequent in the highest PP tertile (p = 0.007), while intracranial and other bleeding types showed no significant differences. Each 10 mmHg increase in PP was associated with a 15% higher risk of GI bleeding (HR: 1.014; p = 0.042), and this association remained significant after adjusting for systolic blood pressure and the ORBIT score (OR: 1.013 per mmHg; p = 0.028). PP was not significantly associated with intracranial, other, or overall bleeding. Conclusions: Pulse pressure independently predicts gastrointestinal bleeding in anticoagulated AF patients, even after accounting for traditional bleeding risk factors. These findings support the inclusion of PP in future risk stratification models and clinical monitoring strategies. Clinical Trial: N/A

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Khalil et al. (2026) conducted a cohort in Atrial fibrillation (n=4,935). Pulse pressure vs. Lower pulse pressure was evaluated on Gastrointestinal bleeding (OR 1.013 per mmHg, p=0.028). Elevated pulse pressure independently predicted gastrointestinal bleeding in anticoagulated atrial fibrillation patients, with each 1 mmHg increase raising risk (OR 1.013; p=0.028).

Gastrointestinal Bleeding Risk Associated With Pulse Pressure in Patients With Atrial Fibrillation: A Retrospective Cohort Study (Preprint)

Figures

Key Points

Study Design

Structured PICO

Main Result

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Gastrointestinal Bleeding Risk Associated With Pulse Pressure in Patients With Atrial Fibrillation: A Retrospective Cohort Study (Preprint)

Figures

Key Points

Study Design

Structured PICO

Main Result

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading