The signal averaged P wave in atrioventricular block--evidence for abnormal atrial conduction?

The incidence of atrial fibrillation in patients with atrioventricular block (AVB) appears increased over that for the unpaced population even if atrioventricular synchrony is maintained by dual chamber pacing. To assess whether atrial fibrillation in these patients might be due to concurrent abnormality in atrial activation we performed signal averaged P wave (SAPW) recordings in 15 patients with dual chamber pacemakers implanted for AVB and compared the results to those from 21 unpaced controls. The median (range) age was 69 (53-89) years for patients and 60 (51-78) years for controls. Eleven patients and 12 controls were male. All patients were pacing in VDD mode at the time of study. SAPW recordings were obtained using our previously reported selective P wave averaging system. We measured P wave duration after high pass filtering at 40 Hz, the rate of change of P wave voltage over time (spatial velocity) and low and high frequency spectral energy after Fourier transformation of the P wave signal. We found increased P wave duration, mean spatial velocity and lower frequency energy in patients with AVB compared to controls [duration, 144 (5) vs. 134 (2) ms, p<0.05; mean spatial velocity, 5.6 (0.4) vs. 4.6 (0.1) mV/s, p<0.05; energy 20-150 Hz, 57.4 (8.2) vs. 36.3 (2.8) muV2.s, p<0.01. All values mean (SEM)]. These results suggest that the increased incidence of atrial fibrillation in patients paced for AVB may be related to intrinsic abnormalities of atrial activation and not solely to the pacing mode. Ensuring sequential atrioventricular pacing in these patients may not completely abolish the increased incidence of atrial fibrillation.