Rick Schreurs

72 Chapter 4 Differences in timing of cardiac events during various pacing modes The timings and intervals of various parameters at AV opt are presented in table 3 . P-wave duration was significantly shorter in A-S compared to A-P during BiV, LV and RV pacing (75-77ms vs. 113-116ms, respectively). This indicates a prolonged total activation time of the atria and larger interatrial dyssynchrony, which is supported by a longer interval between RA dP/dt max and LA dP/dt max in A-P than in A-S mode (ranging from 14-20ms in A-S compared to 26-30ms in A-P). Right-eAVD, left-eAVD and mean-eAVD were larger during A-S, however only significant for left-eAVD (214±24ms vs 147±11ms) and mean- eAVD (215±25ms vs 156±12ms) during RV pacing (see table 3 ). Where D right-left-eAVD became as large as 45ms for LV pacing in A-P mode, it was shorter in A-S, however only significantly for RV pacing (2±4ms). In A-P mode the interventricular dyssynchrony (interval between times of LV and RV dP/ dt max ) was significantly different between LV pacing (19±5ms) and BiV (-4±5ms) or RV pacing (-9±6ms, table 3 ) and QRS duration was significantly longer during LV pacing (109±4ms) than during BiV (86±5ms) and RV pacing (85±5ms). Comparable results were observed in A-S mode. The interval between LV and RV dP/dt­ max differed between all three different ventricular pacing sites (see table 3 ). Optimization of AV-delay using effective mechanical AV-delay Figure 6 shows the average AV-delay optimization curves based on cardiac output for six pacing protocols. When cardiac output (% of maximal change) was plotted against the absolute programmed AV-delay, the optimal AV-delay was shorter for the A-S than for the A-P curves. The opposite was true when cardiac output change was plotted against right- eAVD and left-eAVD (middle and right upper panels of figure 6 ). Mean-eAVD showed less difference between the six optimization plots and the optima were even closer together when mean-eAVD was expressed as a percentage of RR-interval. For this RR-normalized mean-eAVD the curves of the different pacing modes fell well on top of each other with a common optimum with a mean-eAVD of ~25% of the RR-interval. Stroke work and mean arterial pressure showed comparable patterns when plotted against the various effective AV-delays (data not shown). Mean optimal AV-delays calculated from all 39 pacing settings in all seven experiments are presented in table 4 for cardiac output, stroke work and mean arterial pressure. For all three hemodynamic parameters optimal right-eAVD, left-eAVD and mean-eAVD were significantly longer compared to the optimal programmed AV-delay. The coefficient of variance for the optimal AV-delays of the cardiac output and stroke work curves was lowest for mean-eAVD corrected for HR.