Rick Schreurs

97 SonR for AV-optimization changes in aortic velocity time integral (VTI) as a measure of cardiac function [7]. Later the SonR algorithm was further optimized by Dupuis et al . to better predict the optimal AV-delay, which was qualified as the shortest AV-delay where the mitral valve closes in physiological fashion without truncation of the A-wave of the flow velocity curve [36, 37]. This moment appeared to correspond with the shortest AV-delay without significant rise in SonR1 amplitude, the so-called inflection point of the sigmoid SonR curve. We did not systematically determine the inflection point. However, figure 2 shows that during BiV and LV pacing the highest values of LVP max and LV dP/dt max occur at AV-delays slightly longer than the one where SonR1 rises. Interventricular dyssynchrony The present data indicate that interventricular dyssynchrony is not a major determinant of SonR1. This conclusion may be drawn from the high degree of similarity of SonR1 values during BiV and LV pacing in Resp animals. This finding seems to be in contrast with the advised use of the current SonR algorithm. This algorithm prescribes to first apply an AV- delay scan at five different VV-delays and then use the VV-delay with the highest average SonR1 value for finding the optimal AV-delay, as quantified by the ‘PEA area ’ [38]. In the latter study the optimal AV-delays defined by LV dP/dt max­ and SonR1 (PEA called at that time) were concordant in 9 out of 12 patients. On the other hand, PEA area was not much different between the various VV-delays, which seems to support the data from the present study. Position of SonR accelerometer The current experiment was performed with SonR leads located in the RA and RV. Initially the accelerometer was placed in the RV-lead, however in the clinical setting the switch has been made to the RA [15-17, 39]. Previous research using a temporary accelerometer in patients showed a higher SonR signal in the RV compared to the RA during sinus rhythm and atrial fibrillation [40]. In our study absolute values of RV SonR1 were indeed higher than RA SonR1 signals, but relative increases were more pronounced in RA SonR1 signals, especially during short AV-delays with early mitral flow reversal. This might possibly lead to more turbulence in the atria leading to greater vibrations and higher RA SonR1 values. We hypothesize that the placement of the SonR accelerometer in the RA gives a better overview of the overall cardiac function, including that of the left ventricle, while placing the SonR1 sensor in the RV might lead to a greater contribution of RV contractility. Potential clinical applications Currently the SonR algorithm is being used in a biventricular implantable cardiac defibrillator (CRT-D) system with the SonR sensor embedded in the atrial pacing lead (PARADYM TM RF SonR CRT-D). A randomized pilot study, The Clinical Evaluation of Advanced Resynchronization (CLEAR) trial, has shown that RA SonR1 based optimization of CRT in heart failure (HF) patients significantly improved on New York Heart Association (NYHA) functional class, with fewer deaths and fewer hospitalizations compared to 5