Does Atrial Fibrillation Cause Pulmonary Embolism?
Atrial fibrillation does NOT directly cause pulmonary embolism, and the observed association between these conditions is explained by shared risk factors rather than a causal relationship.
The Evidence Against a Causal Relationship
The most definitive evidence comes from a large Swedish registry study of 1.4 million residents that directly addressed this question 1. After adjusting for age and comorbidities, atrial fibrillation without anticoagulation was not associated with increased risk for pulmonary embolism (HR 1.03,95% CI 0.94-1.13) 1. This finding remained consistent even after accounting for the competing risk of death 1.
Why the Confusion Exists
The apparent association between AF and PE in unadjusted analyses is entirely explained by:
- Age differences: Patients with AF are on average >25 years older than those without AF 1
- Shared comorbidities: Both conditions share common risk factors including advanced age, cardiac disease, and immobility 1, 2
- Confounding by indication: The crude incidence of PE appears higher in AF patients (2.91 vs 1.09 per 1000 person-years), but this disappears after proper adjustment 1
Mechanistic Considerations
Left vs. Right Atrial Thrombosis
Atrial fibrillation causes left atrial thrombus formation leading to systemic arterial emboli (stroke), not right-sided thrombi that would cause PE 1. The pathophysiology is fundamentally different:
- Stroke mechanism: Left atrial stasis → left atrial appendage thrombus → systemic arterial embolization 3
- PE mechanism: Venous thrombosis (typically lower extremity DVT) → right heart → pulmonary arterial embolization 4
Supporting Evidence for Separate Mechanisms
Patients with PE and concurrent AF actually have significantly lower rates of deep venous thrombosis (21% vs 44%, P<0.001) compared to PE patients without AF 4. This suggests that when PE occurs in AF patients, it follows the traditional DVT pathway rather than originating from right atrial thrombi 4.
Clinical Implications
Risk Stratification
While one study suggested AF patients with PE have worse outcomes (higher mortality, longer hospital stays) 5, this reflects the burden of comorbidities rather than a causal PE risk from AF itself 1. The CHA₂DS₂-VASc score, designed for stroke risk in AF, does not predict PE risk 4.
Anticoagulation Decisions
- For stroke prevention in AF: Use standard CHA₂DS₂-VASc scoring and guideline-directed anticoagulation 3
- For PE prevention: Base decisions on traditional VTE risk factors (immobility, surgery, malignancy, prior VTE), not on AF diagnosis 1
- The presence of AF alone does not justify prophylactic anticoagulation specifically for PE prevention 1
Important Caveat
One older study in stroke patients suggested AF was an independent risk factor for DVT/PE 2, but this has been contradicted by more robust, contemporary evidence with better adjustment for confounders 1. The modern understanding, based on a population of 1.4 million with rigorous statistical adjustment, supersedes this earlier finding 1.
Bottom Line for Clinical Practice
Do not consider atrial fibrillation as a risk factor when assessing PE probability or making decisions about VTE prophylaxis. The higher crude rates of PE observed in AF patients reflect their older age and greater burden of comorbid conditions, not a direct thrombogenic effect of AF on the venous/right heart circulation 1.