Cryptogenic stroke is a type of stroke in which the cause of the blood clot cannot be identified, leaving many patients without a clear treatment plan and at high risk for another stroke. Current medical guidelines recommend blood-thinning medication (anticoagulation) only when atrial fibrillation (AF) -an irregular heart rhythm- can be documented. However, AF may occur silently and remain undetected. Long term implantable (placed invasively under the skin) devices may be needed to capture these episodes.

AF is known to develop from disease of the left atrium, the upper chamber of the heart that receives blood from the lungs. When the left atrium does not contract normally, blood flow may slow down, increasing the risk of clot formation. Nowadays, the left atrial (LA) function can be quantified precisely using a noninvasive ultrasound technique called strain imaging.

This study aims to determine whether reduced LA function is associated with cryptogenic stroke and its recurrence even when AF is not observed. If such an association is confirmed, LA strain could serve as a new biomarker to identify patients at risk, earlier than the development of overt AF, enhance preventive measures to reduce recurrent strokes. Because echocardiographic strain imaging is safe, cost-effective, and widely available, it may become an important tool for improving care in this high-risk population.

Cryptogenic stroke accounts for nearly one-third of ischemic strokes, yet its underlying mechanism often remains unidentified, leaving patients without targeted therapy. Atrial fibrillation (AF) is detected up to one third of cryptogenic strokes, but its detection frequently requires prolonged monitoring due to subclinical paroxysmal episodes. Increasing evidence suggests that structural and functional abnormalities of the left atrium (LA) precede AF and contribute to thromboembolic risk independently of rhythm disturbances.

This multicenter retrospective study aims to determine whether impaired LA strain, a sensitive echocardiographic measure of atrial function, is associated with cryptogenic stroke and can predict recurrent events. Echocardiographic images of the patients who were referred to the echocardiography laboratory as part of stroke work up will be analyzed offline, using standardized speckle-tracking software. De-identified clinical and imaging data will be collected in the central core laboratory at the University of Pittsburgh for uniform analysis and validation.

By establishing LA strain as a biomarker of atrial cardiopathy and cryptogenic stroke risk, this study seeks to bridge the current gap between stroke classification and management. Demonstrating the independent association between LA dysfunction and cryptogenic stroke and the recurrence of cryptogenic stroke independently of AF could support the use of LA strain as a risk marker to for monitoring patient and developing preventive strategies such as anticoagulation without documented AF. Hence, our findings may improve preventive care and reduce the burden of disabilities due to cryptogenic stroke.

Inclusion Criteria:

• Adult patients (age > 18 years and < 80 years) who had a complete transthoracic echocardiography (TTE) exam at the University of Pittsburgh Medical Center (UPMC) Presbyterian Echocardiography Lab with a referral diagnosis of suspected stroke.

  2- Patients with an established diagnosis of stroke by stroke neurologists.

  3- Patients in sinus rhythm documented at the time of referral.

  4- Availability of adequate echocardiographic images for strain quantification.

  5- Clinical follow-up data available for evaluation of study endpoints, including stroke recurrence, atrial fibrillation development, and mortality.

  6- Patients from collaborating centers will be included with de-identified echocardiographic images and clinical data meeting the above criteria.

Exclusion Criteria:

Incomplete or technically difficult transthoracic echocardiography (TTE) studies and studies from subjects younger than 18 years of age will be excluded. Additional exclusion criteria include:

1- Pre-existing atrial fibrillation (AF): Patients with a documented history AF prior to the index echocardiography study will be excluded because it is well established that AF justifies cryptogenic stroke requiring anticoagulation.

2 -Structural heart disease: Patients with significant structural abnormalities, such as moderate or severe mitral stenosis, moderate or severe mitral regurgitation, prosthetic heart valves, devices in the left atrium, atrial septal defect or patent foramen ovale will be excluded as these conditions are known causes of stroke and abnormal left atrial function.

3- Coagulopathy: Patient having disease leading to hypercoagulable (thrombus formation) state will be excluded (presence of cancer, antiphospholipid antibodies, hematological diseases).

4- Inadequate clinical follow-up: Patients with insufficient follow-up data or missing medical records that preclude evaluation of study endpoints (e.g., stroke recurrence, AF development) will be excluded unless death is documented.

5- Poor image quality for strain analysis: Patients having echo exams with inadequate image quality that prevent reliable strain quantification will be excluded.

-

Participants will be selected from adult patients evaluated for stroke at the University of Pittsburgh Medical Center (UPMC) and collaborating university hospitals that serve as tertiary care centers with established cardiovascular imaging programs.

Only subjects who have already undergone transthoracic echocardiography as part of their stroke workup, and who have clinical and imaging data available for retrospective analysis will be included.

The UPMC will provide both the data collection platform -REDCap- and Echocardiography Core Laboratory for centralized image analysis.

De-identified individual participant data (IPD) collected in this study, including baseline characteristics, clinical outcomes, and left atrial strain measurements, will be shared. Data will be available upon reasonable request for researchers conducting methodologically sound studies that align with the objectives of this research.

IPD and supporting documents will be available starting 6 months after publication of the primary results and will remain accessible for 5 years thereafter.

Access to the dataset will be granted to researchers upon submission of a formal request, outlining the study objectives and analysis plan. Approval will be contingent upon compliance with ethical guidelines, data privacy regulations, and institutional review board (IRB) requirements. Researchers will be required to sign a data-sharing agreement (DSA) before receiving access with the University of Pittsburgh..