Rick Schreurs

131 General discussion AV-coupling in first degree AV-block It was against this background that we performed the studies described in chapter 3 . A first interesting and important finding was the similarity in hemodynamic effects of restoring AV-coupling in healthy pigs and heart failure patients. Moreover, these data were quite well replicated by the CircAdapt computer model. The implication of these findings is that they strongly support the physical basis of ventricular filling and that these principles are (largely) the same in normal and failing hearts: passive filling due to AV-pressure gradient, active filling due to atrial contraction and diastolic MR in case there is a reverse AV-pressure gradient. All experimental and simulation data show that LV inflow is reduced when there is fusion of E- and A-wave, as is the case when the PR-interval is prolonged. Moreover, in all models, PR-prolongation leads to MR, even in healthy hearts. On the other hand, shortening the AV-delay too much, led to a truncation of the A-wave caused by too early ventricular activation leading to a reduction in ventricular filling, indicating that and intermediate, optimal AV-delay represents the best cardiac function. Moreover, also the translation between improved filling pressures and pump function, through the Frank-Starling mechanism, is comparable between healthy and failing hearts. In the clinical study we showed that restoration of AV-coupling by BiV pacing resulted in a significant increase in stroke work and stroke volume that was most pronounced at an AV-delay of 50% of the intrinsic PR-interval (137±30ms). We found that the increase in stroke work and cardiac output was obtained by more efficient ventricular filling due to separation of the E- and A-waves (compared to E-A fusion for long AV-delay), resulting in a greater attribution of the atrial kick. Additionally, the amount of diastolic MR decreased, attributing to better ventricular filling. Optimal ventricular filling leads to improvements in length-dependent activation of the myocardium, which is the cellular basis of the Frank- Starling mechanism [12]. The computer simulations and patient studies also showed that the benefits of AV-coupling may be abrogated by pacing-induced ventricular dyssynchrony. Normalizing the AV-delay using RV pacing failed to show improved ventricular filling in patients, possibly explained by prolonged isovolumic contraction and dyssynchronous contraction of the papillary muscles, increasing diastolic MR. In the computer simulations, LV end diastolic volume did increase at optimal AV-delays during RV pacing, but still no improvements in cardiac output or mean arterial pressure were observed, presumably caused by poorer ventricular pump function due to ventricular dyssynchrony. It is therefore important to maintain synchronous activation of the ventricles. In patients with heart failure a concern may be that restoring AV-coupling leads to an increase in LV end diastolic pressure, possibly giving rise to an increase in pulmonary congestion. However, the data in the porcine hearts and computer simulations show that this increase in LV end diastolic pressure does not coincide with an increase in mean left atrial (LA) pressure, a variable that is probably more related to pulmonary congestion. 7