Correlation between manual and automatic measurements of the atrial signal in VDD leadless pacemakers (Micra AV)

Abstract Introduction The Micra AV leadless pacemaker (model MC1AVR1, Medtronic, MN, US) provides atrio-ventricular (AV) synchronous pacing thanks to accelerometer-based mechanical sensing of the atrial contraction (A4 signal). The device includes an automatic atrial sensing feature with auto-adjustment of the A4 threshold. To date, the correlation between automatic and manual A4 measurements has not been studied in a real-world setting. Therefore, we aimed to validate the accuracy and clinical relevance of the automatic measurements. Purpose To characterize the correlation between automatic measurements (Auto A4) and manual measurements (manual A4) and to report potential implications for AV synchrony. Methods All patients (n = 113) who underwent a Micra AV implantation at the University Hospitals between May 2020 and March 2024 were analyzed. Automatic sensing of the A4 signal occurs only during operating VDD pacing mode. The A4 amplitude is then measured as the maximal acceleration during the programmed A4 window. For manual measurement, the A4 signal was identified using its relation to the P wave. Measurements were repeated on 3 consecutive beats and the average is reported. For this analysis, only patients with paired data (auto and manual A4 measurements) on the same day were included. Discordance between both measurements was defined as at least 50% amplitude difference. Ventricular pacing percentage (VP%) and AV synchrony index were also collected. Results Sixty-nine patients (mean age: 77.2 ± 12.9 years, male: 66.7%) had an auto and manual A4 data pair. Median auto A4 amplitude was 1.9 [1.1-5.1] m/s², manual A4 amplitude 3.2 [2.0-4.7] m/s² (P <0.001). The distribution of measurements differed between both approaches (Figure 1A), mainly due to the limitation of 5.1 m/s² for automatic measurements. We also observed a relative sparsity of auto A4 values in the 3.0-5.0 m/s² range (n = 2 compared to manual A4 n = 23, P <0.001). Auto A4 amplitude was correlated with true/manual A4 measurements (R = 0.67, P <0.001). In 20 (29.0%) patients, auto and manual A4 measurements were considered discordant, with lower auto A4 in 17 patients and higher auto A4 in 3 patients. In patients with ventricular pacing >90%, the mean AV synchrony index was 80.4%. AV synchrony was significantly lower in patients with A4 discordance (73.8% vs. 83.4%, P = 0.02) (Figure 1B). Conclusions Automatic A4 measurements provide a reasonable estimation of the true A4 amplitude. Nevertheless, a discordance of >50% between manual and automatic measurements was associated with lower AV synchrony. Manual measurement of A4 remains the most accurate and should be the standard way to measure A4 signals for research purposes.