B-po02-040 correlation between av synchrony and device collected am-vp sequence counter in a leadless ventricular pacing using accelerometer based atrial sensing

Accelerometer based AV synchronous pacing by tracking atrial activity has been demonstrated using a leadless ventricular pacemaker. To date, AV synchrony (AVS) has been assessed by validating P-waves from a surface ECG. The device collected AM (atrial mechanical detect) to VP (ventricular pace) sequence counter reports the percentage of ventricular beats preceded by a detected atrial mechanical event and may be a useful surrogate for AVS. To characterize the correlation between AM-VP and AVS percentage at rest among patients with complete AV block in the MARVEL 2 study. AVS was assessed during a 20 minute resting period via blinded validation of P-waves on the surface ECG and was compared to the device generated AM-VP sequence count. Linear regression was used to quantify the association between AM-VP and AVS and patients with >10% difference were analyzed. Overall, 40 of 77 patients exhibited complete AV block with median %VP of 99.9% (range: 88.3% - 100%). Median AVS was 94.3% (range: 33.4% - 99.9%), while median AM-VP was 87.1% (range: 27.0% - 99.4%). AM-VP was highly correlated with true AVS (R2 = 0.819, P<0.001; Figure). AVS was underestimated in 11 patients, while 0 patients were overestimated. The rate smoothing algorithm, which maintains AVS during periods of intermittent atrial mechanical undersensing, accounted for the 7.2% increase in median AVS over median AM-VP. The AM-VP sequence counter provides a reasonable estimation of the level of AVS provided by a leadless pacemaker at rest in patients with complete AV block. Whether this correlation is maintained over time and during activity warrants further investigation.