November 15th, 2009

On PACE to Maintain LV Function

and

CardioExchange Editor:  Were the authors surprised by the magnitude of deterioration in EF seen with RVA pacing over the one year follow-up period of the study?  Could this possibly be related to the method of measurement (i.e., 3D echo) or could it be due to inappropriate pacing in patients with sinus node dysfunction?
 
Yu: The magnitude of deterioration in EF was slightly greater than we had initially estimated. As we are looking at small changes in EF and LV end-systolic volume, the use of real-time 3D echo is helpful which is far more accurate than 2D echo.  This might have helped to elucidate the changes in cardiac size and function.  These changes are genuine and occurred equally in both patient groups with high-grade AV block and sinus node dysfunction.    
 
CardioExchange Editor:  We would expect that individuals with baseline lower EF or abnormal LV volumes are more likely to see changes in these same parameters in the setting of a potentially detrimental exposure.  Did you also observe this trend in your study?  In other words, although these individuals may have been fewer in number for a formal analysis, were patients with an EF ~45% more likely to do worse?
 
Yu: This is an interesting question.  However, the primary analysis of PACE was unable to address this hypothesis.  Although it is tempting to suggest that a baseline lower EF might translate to greater a reduction of EF after RVA pacing, this needs to be substantiated by further detailed analyses. 
 
CardioExchange Editor:  This is an important study that builds on prior work in the area of pacing and also makes use of advanced techniques in echocardiography.  Many would agree that ejection fraction is widely assessed in practice, but is still a relatively crude measure.  On the other hand, LV volumes are a nice metric but difficult to measure in practice.  Given their reproducibility in practice, was there a particular reason why LV dimensions (end-diastolic and end-systolic) were not included as endpoints in this analysis? 
 
Yu: Although the use of EF is not a perfect measure of systolic function, it has been the most widely accepted method of analysis.  With the use of 3D echo, the changes in EF can accurately reflect changes in cardiac function as a result of different pacing modalities.  Furthermore, we have captured additional quantitative echocardiographic parameters, such as tissue Doppler imaging for assessment of myocardial systolic velocity, which will be analyzed in due course.  The LV diameter was not used as it is an oversimplified measure of LV size which will not be accurate enough to reflect the actual volumetric changes.  Furthermore, when patients developed systolic dyssynchrony as a result of RVA pacing, the paradoxical septal movement render the measurement of LV diameter even more difficult to interpret.

3 Responses to “On PACE to Maintain LV Function”

  1. Interesting discussion thus far on this important trial…a major issue of this trial seems to be the near constant chronic pacing, even among patients with sinus node dysfunction. In terms of generalizing these data to what is actually done in clinical practice, do you think that the results would have been similar if the protocol did not include such a high frequency of pacing in both arms? Is the thought that, in such a scenario, it would have taken longer follow-up to observe a difference in the same endpoints?”

  2. Thanks for the comment on pacing rate. A constantly high pacing rate is certainly needed for advance AV block. For patients with sinus node dysfunction, it was reported previously that ventricular pacing >40% will be associated with increased cardiovascular events, with a plateau effect on further pacing %. We believe this range is already deleterious. Though you might be right if patient need less than this range, it will take a longer time for cardiac dysfunction to develop.

  3. Was it a fair comparison?

    RV pacing allegedly resulted in decreased LV ejection fraction, whereas biventricular pacing did not.

    Was this actually a result of adverse LV remodeling, or just an artifact of trying to assess the EF in an LV that is not contracting synchronously because it has an “iatrogenic LBBB” due to the RV pacing?

    Would it not have been more appropriate to compare ejection fractions in the RV-paced group and the biventricular paced group at the 12 month follow-up with both groups paced similarly (i.e., biventricular pacing)?