Abstract 16174: Electrical Features Associated With the Occurrence of Ventricular Fibrillation in STEMI Patients
Background: More than 300,000 people develop sudden cardiac arrest (SCA) each year in the United States, and the majority (>80%) occur in the setting of coronary artery disease. Efforts to improve long-term prediction of individuals at risk for SCA have made slow progress and in this context, alternative approaches aimed at pre-emptive risk stratification and prevention are needed.
Objective: To identify sub-acute static and dynamic electrocardiographic (ECG) features associated with the development of ventricular fibrillation (VF) among patients with acute STEMI.
Methods: We developed a systematic approach to process digital ECG data recorded by LIFEPAK defibrillators using PhysioControl software and MATLAB®. From Emergency Medical Services in Portland, Oregon, ECG (12-lead as well as continuous) tracings of 5 cases (STEMI with VF occurrence prior hospital admission) and 10 controls (STEMI patients without VF occurrence) were analyzed (mean age 65±17 years, males 47%, anterior STEMI 40%). Analysis included heart rate, ventricular excitability, durations of different intervals (PR, QRS, QT, QTc), amplitude at J-point, ST segment elevation, as well as ST integrals (defined as the ST wave segment area in relationship to the isoelectric line). Analyses were carried out at baseline (time of EMS arrival) as well as over a 10-minute period.
Results: The majority of the duration-based measurements of different ECG segments showed similar results among cases and controls. By contrast, amplitude measurements, especially the ST integral was observed to be larger among cases compared to controls (604±218 vs. 390±133 mV*ms, P=0.07). In addition, the number of premature ventricular beats was higher among cases compared to controls (P=0.049). Dynamic analyses found that only changes in ST integral were associated with VF occurrence -- ST integral becoming significantly higher among cases compared to controls (P=0.028). The MATLAB analysis program was able to reproduce the results of the manual analysis with significantly increased efficiency and only minor discrepancies.
Conclusion: Our preliminary findings suggest the feasibility of developing automatic algorithms that predict VF in STEMI patients.
Author Disclosures: S. Tang: None. E. Marijon: None. A. Lou: None. H. Chugh: None. R. Mariani: None. A. Uy-Evanado: None. C. Teodorescu: None. K. Reinier: None. J. Jui: None. S. Chugh: Research Grant; Modest; NIH>100k. Other; Modest; Fellowship funding (Boston Scientific and Medtronic) 50-100K.
- © 2014 by American Heart Association, Inc.