Erythrocytosis after SGLT2 inhibitor initiation and anticoagulated outcomes in atrial fibrillation: a real-world analysis with counterfactual modeling
Main finding · Incident erythrocytosisResult — go to the resultAdjusted incidence rate ratio 3.85 95% CI 2.78–5.31Authors' conclusion
SGLT2 inhibitor-associated erythrocytosis is a frequent hematological response in patients with AF receiving anticoagulant therapy and marks a phenotype predisposed to thromboembolism rather than bleeding.
Study context
- Population
- Adults with AF receiving oral anticoagulants, 2011–2024, Tianjin Health and Medical Data Platform; mean age 72.5 ± 9.7 years, 48.1% female47,401 eligible
- Participants
- 39,190 adults
- Exposure
- SGLT2i initiation vs non-use; erythrocytosis vs none among SGLT2i users
- SGLT2 inhibitor usersn = 6,697
- Outcome
- Incident erythrocytosis (Part 1); thromboembolic and bleeding outcomes by erythrocytosis among SGLT2i users (Part 2)
Methodological detailsnew-user design, stratified pseudo-time strategy, active comparator analyses, propensity score matching, counterfactual mediation and decision modeling
- New-user design
- Stratified pseudo-time strategy
- Active comparator analyses
- Propensity score matching
- Counterfactual mediation and decision modeling
Main finding
Incident erythrocytosis · within 6 ± 2 months
Adjusted incidence rate ratio
3.85
95% CI 2.78–5.31
By group
SGLT2i usersvs Non-users
adjusted IRR 3.85 (95% CI 2.78–5.31)
As reported in the abstract
Observational data: associations, not effects.
Further findings
| Outcome | aHR (Model 5) (95% CI) | P value | |
|---|---|---|---|
| Thromboembolic events, erythrocytosisvs none among SGLT2i users · full follow-up | 2.34 (1.39–3.94) | P = 0.001 | |
| 0.20.5125Adjusted hazard ratio (Model 5) (log scale) |
Subgroup and sensitivity analyses
- Consistent results across age, sex, kidney function and anticoagulant type
- Hemoglobin rise acted primarily as a risk marker rather than a causal mediator
Safety
| Outcome | aHR (95% CI) | P value | |
|---|---|---|---|
| Bleeding, erythrocytosisvs none among SGLT2i users · full follow-up | 1.42 (0.81–2.49) | P = 0.22 | |
| 0.512Adjusted hazard ratio (log scale) |
Original publication
Erythrocytosis after SGLT2 inhibitor initiation and anticoagulated outcomes in atrial fibrillation: a real-world analysis with counterfactual modeling. Cardiovasc Diabetol. Published 22 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →