Rick Schreurs

95 SonR for AV-optimization involved in the SonR1 signal, because the myocardial vibrations that are considered to be responsible for the heart sounds are produced by the RV valves as well. This phenomenon can be further understood from the resemblance between the SonR signal and the phonocardiogram. In 1963 Shah et al . showed that the second component of the first heart sound (S1) as derived from intracavitary and external phonocardiograms coincided with the peak of LV pressure rise (LV dP/dt max ) in dogs [18]. Subsequently, the amplitude of S1 seemed to show a strong linear correlation to LV dP/dt max over a wide range of pressures, volumes and cardiodynamics, suggesting that S1 amplitude is mainly determined by cardiac contractility under these conditions [19]. The origin and significance of the different components of the first heart sound are considered to come from a M1 and T1 component. Some studies seem to indicate that mitral valve closure has the largest contribution to S1, but in this study the sensor was positioned in the LV [20]. Instead, in our study, with the sensor positioned in the RA and RV, the dobutamine data show a slightly higher correlation of SonR1 with RV than with LV dP/dt max . AV-delay optimization during resynchronization To the best of our knowledge, the present study is the first to validate SonR1 against invasively measured LV pressure and LV dP/dt max during various AV-delays. At first sight, the link between SonR1 and LV dP/dt max during BiV and LV pacing in Resp animals seems comparable with that during the dobutamine stress test. However, several differences are present. First of all, the relative change in LV dP/dt max is much smaller during pacing (~10%) than during dobutamine (100-200%) and so are changes in SonR1 (up to ~80% for RA SonR1 and ~20% for RV SonR1 for pacing as compared to 100-200% during dobutamine). These numbers do point out that the SonR1 signal seems more sensitive to pacing-induced resynchronization than to dobutamine-induced increase in contractility. The latter is driven by better calcium handling of the myocardium, whereas resynchronization creates better mechanical coordination between ventricular wall segments. Secondly, while RA and RV SonR1 show similar relative increases during dobutamine stress test, during resynchronization (BiV and LV pacing during intermediate AV-delays) the relative RV SonR1 increase is only a quarter of that of RA SonR. A possible explanation for this observation is that the RV sensor primarily senses RV processes, whereas the RA sensor, although at larger distance of the sources of the acceleration (hence the lower amplitudes), senses changes at both ventricles. The change in RV SonR1 may be smaller because during resynchronization RV dP/dt max does not increase and has even the tendency to decrease. RV SonR1 performs well during dobutamine testing, presumably because dobutamine increases RV and LV contractility to a similar extent. 5