Characteristics of Cardiac Memory in Patients with Implanted Cardioverter Defibrillator: the CAMI study

Objective The goal of this study was to determine factors associated with cardiac memory (CM) in patients with implantable cardioverter-defibrillators (ICD). Methods Patients with structural heart disease (n=20; mean age 72.6±11.6 y; 80% male; mean left ventricular ejection fraction (LVEF) 31.7±7.6%; history of myocardial infarction (MI) in 75%, ventricular tachycardia (VT) in 85%) and preserved atrioventricular (AV) conduction received primary (80%) or secondary (20%) prevention dual-chamber ICD. Standard 12-lead ECG was recorded in AAI and DDD mode, before and after 7 days of right ventricular (RV) pacing in DDD mode with short AV delay. Direction (azimuth and elevation) and magnitude of spatial QRS, T, and ventricular gradient (SVG) vectors were measured before and after 7 days of RV pacing. CM was quantified as the degree of alignment between QRS DDD-7 and T AAI-7 vectors (QRS DDD-7 -T AAI-7 angle). Circular statistics and mixed models with a random slope and intercept were adjusted for days 1-7 change in cardiac activation, LVEF, known risk factors, and use of medications known to affect CM. Results QRS DDD-7 -T AAI-7 angle strongly correlated (circular r = - 0.972; P<0.0001) with T AAI-7 -T DDD-7 angle. In the mixed models, history of MI (−180°(95%CI −320° to −40°); P=0.011) and female sex (−162°(95%CI −268° to −55°); P=0.003) counteracted CM-T azimuth changes (+132°(95%CI 80°-184°); P<0.0001). History of VT (+27(95%CI 4-46) mV*ms; P=0.007) amplified CM-T area increase (+15(95%CI 6–24) mV*ms; P<0.0001). Conclusions Existing cardiac remodeling affects CM in response to RV pacing. Women develop less CM than men. Activation memory is another manifestation of CM.