Jabbar · Cureus 2026 · retrospective observational study · n=350

Predictors of Ischemic Stroke in Patients With Non-valvular Atrial Fibrillation: A Retrospective Study.

Cited 0 times in the scientific literature.

Level 4 - case-series / case-control

Retrospective single-center observational medical record review

PubMed 42583511 · doi:10.7759/cureus.112522 · record verified 2026-08-28

What was done

This retrospective study analyzed medical records of 350 patients with non-valvular atrial fibrillation treated at a single tertiary care hospital in Rawalpindi, Pakistan. Researchers collected demographic, clinical, anticoagulation, CHA₂DS₂-VASc, and echocardiographic variables. Independent-samples t-tests, chi-square tests, and multivariable logistic regression were conducted to identify factors independently associated with ischemic stroke.

What was found

Ischemic stroke occurred in 71 of 350 patients (20.28%). Patients with stroke had higher mean CHA₂DS₂-VASc scores than those without stroke (4.50 ± 1.40 vs. 3.20 ± 1.20, p < 0.001). Multivariable logistic regression identified multiple independent predictors of ischemic stroke: age ≥65 years (adjusted odds ratio [AOR] = 2.59, p = 0.003), prior stroke/transient ischemic attack (AOR = 2.37, p = 0.012), vascular disease (AOR = 1.88, p = 0.045), smoking (AOR = 1.93, p = 0.039), chronic kidney disease (AOR = 2.12, p = 0.028), inadequate anticoagulation (AOR = 2.90, p = 0.001), left atrial enlargement (AOR = 2.32, p = 0.007), left ventricular ejection fraction <40% (AOR = 1.94, p = 0.042), and left atrial appendage thrombus or spontaneous echo contrast (AOR = 3.07, p = 0.004).

Why it matters

Echocardiographic findings and renal comorbidity provide predictive value for stroke risk alongside standard CHA₂DS₂-VASc scoring and emphasize the impact of inadequate anticoagulation in clinical practice.

Limits

The study is limited by its retrospective, single-center design, which introduces potential documentation and selection biases. The abstract does not provide 95% confidence intervals for the reported odds ratios, and longitudinal follow-up duration or timing of stroke relative to AF diagnosis was not reported.

Cited by