Rick Schreurs

66 Chapter 4 left (left-eAVD) effective AV-delay were calculated as the interval between the times of RA dP/dt max and RV dP/dt max and the times of LA dP/dt­ max and LV dP/dt max , respectively. The mean effective AV-delay (mean-eAVD) was defined as the interval between the mean moment of atrial dP/dt max and the mean moment of ventricular dP/dt max . Subsequently, mean-eAVD was presented as a percentage of the RR-interval to correct for differences in heart rate (HR). The difference between right-eAVD and left-eAVD was calculated (D right- left-eAVD), a positive value indicating that left-eAVD is shorter than right-eAVD. Figure 2 shows examples in two different situations. Timing A-P, LV, AV-delay 150ms A-S, BiV, AV-delay 150ms Atrial sensing RA dP/dt max LV dP/dt max LA dP/dt max RV dP/dt max IAD Right eAVD Left eAVD IVD △ Right- left eAVD Mean eAVD Atrial pacing RA dP/dt max LV dP/dt max LA dP/dt max RV dP/dt max 30ms 183ms 138ms 15ms 45ms 160ms Atrial sensing RA dP/dt max LV dP/dt max LA dP/dt max RV dP/dt max 18ms 211ms 199ms 6ms 12ms 205ms Figure 2. Examples of calculation of the effective right, left and mean AV-delay as well as the inter- atrial and interventricular dyssynchrony, presented for atrial sensing (A-S) BiV pacing (middle) and atrial pacing (A-P) LV pacing (right), both at an AV-delay 150ms. During A-S BiV pacing the interatrial dyssynchrony (18ms) and difference between right-eAVD (211ms) and left-eAVD (199ms) was small. A-P increased interatrial dyssynchrony (30ms) leading to a decrease in right-eAVD (183ms), caused by the fact that atrial electrical activation was not immediately sensed. Left-eAVD (138ms) shortening was even more pronounced, caused by the delayed activation of the LA compared to the RA while the ventricles were simultaneously paced in case of BiV pacing. As a result right-eAVD was 45ms longer than left-eAVD. The arrows indicate the direction of activation. RA: right atrium, LA: left atrium, LV: left ventricle, RV: right ventricle. Statistical analysis Continuous data are presented as mean ± standard error of the mean (SEM). Paired sampled t-tests were used for comparing dependent continuous variables. A two-sided probability value of <0.05 was considered statistically significant. Repeated measures ANOVA with Bonferroni correction were applied when different ventricular pacing sites and various AV-delays were compared. Statistics were performed using Statistical Package for Social Sciences version 26.0 (SPSS Inc., Chicago, Illinois, USA). RESULTS Studies were performed in 7 animals. In 1 animal the A-S protocol was not performed. Baseline hemodynamic data are presented in table 1 . Cardiac output was significantly lower during A-S than during A-P in BiV, LV and RV modes, partly related to a significantly lower heart rate. In general, LV and RV end diastolic pressures were significantly higher during A-S than during A-P. RV dP/dt max was significantly lower in A-S mode compared to