Rick Schreurs

74 Chapter 4 dyssynchrony (various ventricular pacing sites). Calculation of the mean effective AV- delay, normalized for heart rate, can estimate the optimal AV-delay in a manner that is relatively independent of IAD or ventricular pacing site and might therefore be an interesting parameter for future pacemaker optimization. The influence of interatrial delay on the optimal AV-delay The difference between atrial pacing and atrial sensing in 2-chamber pacemakers has been studied since the late 1980s. Wish et al showed that invasively measured stroke volume decreased when pacing mode was changed from DVI (atrial pacing) to VDD (atrial sensing) without adapting the programmed AV-delay. They suggested that this change was the result of an increase in LA to LV depolarization time due to a decrease in IAD [11]. Comparable results were seen in a study by Janosik et al in which Doppler-derived cardiac output was used to examine the optimal AV-delay during DVI and VDD pacing [10]. Kyriacou et al have previously shown that IAD is longer during A-P mode and that the programmed AV-delay should therefore be longer compared to A-S to prevent truncation of ventricular filling in CRT patients [12]. Along with these previous studies, we found that there was a tendency for optimal AV-delay to be shorter during A-S compared to A-P. One of the possible reasons for the small difference in optimal AV-delay between A-S and A-P was that the optimal AV-delay was based on only 5 different AV-delays with a relatively large interval of 50ms, while the optimal AV-delay could well be somewhere between two AV-delays. Using second degree polynomial expression the optimal AV-delay could be calculated more accurately, leading to greater differences between A-P and A-S mode (~40- 50ms). In this study the maximal increases in cardiac output, stroke work and mean arterial pressure were similar for A-S and A-P, indicating that similar hemodynamic conditions can be achieved by both pacing modes as long as AV-delay is properly optimized. Like in other studies, the time difference between RA and LA activation was longer during A-P compared to A-S, evidenced by an increased P-wave width on the ECG and a larger IAD (difference between time of RA and LA dP/dt max ). This can be explained by delayed activation of the LA because the pacing electrode is generally positioned in the RA appendage, more remote from the LA, and besides that electrical conduction does not follow the Bachman’s bundle. These findings are in line with a clinical study by Dabrowska et al in which P-wave duration was shown to correlate with atrial electromechanical delay measured using tissue Doppler during different atrial pacing modalities [13]. In our study, the P-wave width increased by ~40ms when pacing mode was switched from A-S to A-P, while the IAD only increased by 10-15ms. This relatively small increase in IAD may be explained by the fact that P-wave width is the difference between the very earliest and latest activated atrial tissue, whereas IAD is determined by the mechanical activity of large parts of both atria. The 10-15ms values of IAD in our study are smaller than in human studies, in which a delay in left atrial contraction of ~60ms was found when comparing A-P to A-S [12, 14]. Possible explanations might be species differences and the different