How reliable is atrial sensing in single-lead VDD pacing: comparison of three systems

This study evaluated the reliability of atrial sensing, expressed as AV synchronous stimulation, in three VDD systems with the atrial sensitivity (AS) programmed to a conventional value with a 2:1 safety margin compared to most-sensitive values. We studied 34 sex- and age-matched patients with 3 VDD systems: 14 with Unity 292-07, 10 with Saphir 600, and 10 with Thera VDD (5 model 8948 and 5 model 8968i). Two 24-hour Holters were performed on consecutive days. The AS was programmed in a randomized order to its most-sensitive value or to a 2:1 safety margin. All other parameters were programmed identically. The patients underwent a myopotential oversensing test and a daily life activity protocol. A beat-to-beat analysis of the Holters was performed to determine AV synchrony. For the entire group AV synchrony with conventional AS was 98.63% +/- 2.57%, compared to 99.80% +/- 0.43% with most-sensitive values (p = 0.002). There was no difference between the three systems with conventional AS. With the most-sensitive AS, AV synchrony was: Unity 99.99% +/- 0.03%, Saphir 99.42% +/- 0.60% (P = 0.002), Thera 99.81% +/- 0.35% (ns). In the Saphir system with an atrial blanking period of 150 ms, ventricular far-field sensing could be demonstrated in 5 of 10 patients. This reduced the percentage of AV synchrony due to an unwanted mode-switch to a nontracking mode. Myopotential oversensing was not detected in any patient.

The VDD systems tested under identical conditions showed reliable P wave sensing at the most-sensitive atrial sensing setting without myopotential oversensing. Ventricular far-field sensing reduced AV synchrony and must be avoided by appropriate refractory periods.