Mobile ECGs in Detection of Subclinical Atrial Fibrillation in High Risk Outpatient Populations

Ajay Mittal, Susana Shih, Riken Nathu, Danelle Antelo, Ashley Abraham, Victoria Vailoces, Justin Encienzo, Mark Segal, MD/PhD




University of Florida


BACKGROUND Smartphone-based Mobile ECG screening has the potential to provide a cost effective, preventive means of detecting atrial fibrillation (AF) in outpatient non-cardiology specialty clinics. Patients with chronic kidney disease (CKD) are at greater risk for developing arrhythmias and a simple 30-second sinus rhythm strip from the KardiaMobile ECG can provide an effective screen for cardiac rhythm issues.


OBJECTIVES This study aims to assess the KardiaMobile device in recording ECGs among high risk patient groups as a screening tool to detect AF and other cardiac arrhythmias.


METHODS Over 200 enrolled participants were surveyed and had a 30-second ECG recording using the KardiaMobile. Participants that agree to partake in the study are provided with a consent form and questionnaire that aims to reveal any risk factors for cardiovascular disease. Their ECGs are then taken by the KardiaMobile ECG device. If patients are found to have newly diagnosed AF, the treating physician will be notified, and a 12-lead ECG or standard ECG equivalent will be ordered during the patient visit at the discretion of the treating physician. Furthermore, to assess the feasibility of incorporating a mobile ECG in outpatient clinics there will be a means to measure the time associated with collecting this data within clinic flow. This will be measured in comparison to patients not being seen during the study and will be collected either retrospectively or during clinic based off of UFHealth IT’s EPIC configurations.


RESULTS As of November 2020, our team of data collectors have enrolled a cohort of 212 participants. Of those study participants, the KardiaMobile rhythm strip reported 22 abnormal readings which are pending analysis from a cardiologist. 18 readings were labeled unclassified, 8 were poor quality, and 164 were considered normal by KardiaMobile. From those 22 participants with abnormal ECG strips, five are not currently seeing a cardiologist. Of note, the average age of participants was 64.7, and the average self-reported weight was 203.67 pounds. Additionally, 43.39% of participants reported a history of being diagnosed with hypertension; 17.45% reported having heart stents; 17% reported having a history of congestive heart failure, heart attack, and coronary artery disease; and 10.4% reported that they had already been diagnosed with AF. The group of participants includes 76.9% White, 20.3% African American, 6.6% Latino, 1.0% Asian. 51.4% of participants are females, and 48.5% are males. All patients surveyed are currently not on dialysis.


CONCLUSIONS While the data collection is currently in its initial steps and has yet to amass more statistically significant data, the preliminary data shows promise regarding the feasibility of using KardiaMobile ECGs for the prevention, treatment and diagnosis of cardiological disease.