The goal is to review patients with known atrial fibrillation who suffered a stroke or transient ischemic attack (TIA) to incorporate any structural heart abnormalities into the overall clinical picture. A better understanding of the additional risk of a stroke or TIA in patients with specific structural abnormalities may be beneficial in guiding future treatment decisions.

Atrial fibrillation is the most common heart rhythm disorder seen in clinical practice. Patients with atrial fibrillation are at increased risk for stroke. This risk is determined by several factors. Congestive heart failure, hypertension, age, diabetes, female sex, previous history of heart attack or vascular disease, and previous history of stroke all increase a patients' risk for clot formation and stroke. These risk factors have been formulated into a scoring system called the CHA2DS2-VASc score. A score of greater than 2 on this scale indicates a moderate to high risk for clot formation and stroke. Anticoagulation medications are recommended in these patients. Non-traditional risk factors such as left atrial size and left ventricular hypertrophy have also been associated with increased stroke risk. These factors have not yet been routinely incorporated into treatment decisions regarding anticoagulation and stroke prevention.

This study is an expansion of an initial pilot conducted at Pen Bay Medical Center of patient records from 2014-2017. This pilot found 19 patients with a recorded stroke or TIA, history of atrial fibrillation, and CHA2DS2-VASc score of 2 or lower within that time period. This study will be an expansion of that pilot to include patient data from Maine Medical Center.

This study aims to look at these non-traditional risk factors in patients who have an otherwise low stroke risk based on the CHA2DS2-VASc score.

Our specific aims are:

1. Assess left atrial volume using volumetric measurements and left ventricular septal and posterior wall thickness from 2-D transthoracic echocardiography studies in patients with known atrial fibrillation who suffered a stroke or TIA. The investigator would then compare these data to control patients with atrial fibrillation with matched CHA2DS2-VASc scores, age (+/- 5 years), and gender who have no history of stroke or TIA. The goal is to identify higher risk patients with alternate anatomy who otherwise would be considered lower risk using traditional risk scoring methods. Identifying these individuals may lead to more aggressive interventions and subsequently may help in reducing the risk of stroke.
2. Assess stroke outcome as measured by NIH Stroke Scale and/or Modified Rankin Scale score, in the population of patients with atrial fibrillation who suffered a stroke to identify any potential differences associated with structural changes. If structural changes are associated with stroke severity, echocardiogram findings again may aid in guiding treatment decisions for patients with atrial fibrillation.

Hypothesis: Patients with atrial fibrillation and a history of stroke or TIA will have alternate anatomy when compared to patients with atrial fibrillation with matched CHA2DS2-VASc scores who have not had a stroke or TIA.

This is a retrospective matched case-control study of data collected at Pen Bay Medical Center (PBMC) and Maine Medical Center (MMC) from 2014-2017. Patient records that met the criteria as a case from PBMC have already been collected from existing data housed in the American Heart Association's "Get with the Guidelines" registry as a pilot project. These data will be included in this study and combined with controls identified from PBMC, and cases and controls identified from existing EPIC data for MMC patients.

Inclusion Criteria Group 1:

• Diagnosis of stroke or TIA
• Atrial fibrillation
• Transthoracic echocardiogram
• CHA2DS2-VASc score of 0, 1, or 2

Inclusion Criteria Group 2:

• Patients with matched age (+/- 5 years), gender, and CHA2DS2-VASc score
• Atrial fibrillation
• No stroke or TIA

Exclusion Criteria:

- Patients with a CHA2DS2-VASc score greater than 2

Patients with known atrial fibrillation who suffered a stroke or transient ischemic attack (TIA) and matched controls with matched age (+/- 5 years), gender, and CHA2DS2-VASc score who had atrial fibrillation and had not had a documented stroke or TIA in EPIC or the "Get with the Guidelines" registry maintained at PBMC. Controls must have an echocardiogram on record and have a diagnosis of atrial fibrillation at the time of the echocardiogram.