P-Wave and QRS-Complex Durations Can be Measured Automatically from Far-Field Implantable Cardioverter Defibrillator Electrograms

Abstract

Background: ECG P-wave and QRS-complex durations characterize atrial and ventricular conduction and are used in clinical practice for diagnostics. QRS duration $120 ms is an indication for Cardiac Resynchronization Therapy (CRT). Studies suggest that ECG P-wave and QRS duration can be used to optimize CRT device programming. Objective: To determine whether automatically measured device intracardiac electrogram (EGM) conduction intervals can be used as a surrogate for ECG conduction intervals to identify patients for CRT and optimize CRT device programming. Methods: A 5-lead surface ECG and device telemetered signals, which included marker channel, Can-SVC (Superior Vena Cava coil) EGM, and Can-RAr (right atrial ring electrode) EGM, were digitally recorded from CRT patients (n585) during intrinsic ventricular activation with or without atrial pacing. A Wave Duration Measurement (WDM) algorithm was developed to automatically measure sensed (Ps) or paced P-waves (Pp) and sensed QRS conduction intervals (QRSs) from the EGMs as the interval from the device sensed or pacing marker to the end of the respective wave. The WDM results were compared to the corresponding P and QRS conduction intervals manually measured from the surface ECG using Pearson correlation analysis and mean absolute difference (MADIFF). Manual wave measurements were performed twice by the same expert to evaluate intra-observer variability. Results: Automatically determined wave conduction intervals correlated with manually measured values (Table). Intra-observer MADIFFs for the manually measured Ps, Pp and QRSs from surface ECGs were 1167 ms, 11614 ms and 967 ms respectively, comparable to the accuracy of the automatic measurements. Conclusion: The study demonstrated the feasibility of determining sensed or paced P and sensed QRS conduction intervals from far-field EGMs which may then be used to identify patients for CRT and optimize CRT parameters.