Use of global atrial fibrillation organization to optimize the success of burst pace termination
Thomas H. Everett, IV, PhD*,
Joseph G. Akar, MD ,
Lai-Chow Kok, MBBS,
J. Randall Moorman, MD and
David E. Haines, MD, FACC,*
* Krannert Institute of Cardiology, Indiana University School of Medicine, Indianapolis, Indiana, USA
Cardiovascular Division, Department of Internal Medicine, The University of Virginia Health System, Charlottesville, Virginia, USA
View larger version (37K):
[in a new window]
|
Figure 1 (A) A 4-s acquisition of an unfiltered wide bipolar electrogram from the coil electrodes placed in the left atrium and the right atrium. This signal was obtained just before a successful attempt to terminate atrial fibrillation (AF) through rapid burst atrial pacing. The high-amplitude activity represents the far-field ventricular activation, and the low-amplitude signal represents global bipolar interatrial activity. In this example, the AF signal is transiently very well organized ("flutterlike"), as evidenced by the regular 2:1 ventricular response to an atrial rhythm with a cycle length of 120 to 130 ms. (B) An unfiltered wide bipolar electrogram obtained from the same animal immediately before an unsuccessful attempt to terminate AF through rapid burst atrial pacing. The atrial activity and the ventricular response are grossly irregular, suggesting a more disorganized rhythm.
|
|
View larger version (17K):
[in a new window]
|
Figure 2 Results from a fast Fourier transform performed on the signals from Figure 1 after QRS subtraction and digital filtering. Panel A shows a dominant peak with discrete harmonics and little magnitude between the peaks. This resulted in a high (organized) organization index (OI) number. Panel B shows a high magnitude of spectral power between the harmonic peaks. This resulted in a low (disorganized) OI number.
|
|
View larger version (17K):
[in a new window]
|
Figure 3 Calculated organization index (OI) values for a 1-min atrial fibrillation signal acquisition. A sliding 2-s window was used to calculate an OI every 0.5 s. A large variance in the OI is seen over time.
|
|
View larger version (40K):
[in a new window]
|
Figure 4 The distribution of organization indexes (OIs) calculated over a 60-s AF electrogram acquired from all nine animals is displayed in panel A. These data are normally distributed and are representative of the OI values at the time of randomly delivered rapid burst. Panel B shows the distribution of OI values recorded immediately before the burst pace termination attempts synchronized to OI values >0.5. By definition, the distribution curve is truncated at the 0.5 threshold value.
|
|
View larger version (19K):
[in a new window]
|
Figure 5 Success rates for the termination of atrial fibrillation (AF) with rapid burst pacing. Shown are the point estimates and the 95% confidence intervals for predicting the probability of a successful termination of AF. The results from each rapid pacing duration (1 to 4 s) are shown for pacing from the right atrium (RA), left atrium (LA), and biatrial pacing (BiA). A significant increase in the efficacy of the rapid atrial pacing for the termination of AF is seen with biatrial pacing synchronized to an organization index >0.5. With this configuration, a 3-s rapid pacing window had the highest success rate (24.7%).
|
|
View larger version (67K):
[in a new window]
|
Figure 6 Electrograms from the decapolar catheters that were placed in the left atrium (LA) and the right atrium (RA). Biatrial rapid burst pacing for 2 s resulted in the termination of atrial fibrillation. During the pacing, both atria are captured. The channels labeled RA3 and LA3 were selected for pacing (pacing artifact during the pacing train is not uniformly displayed in the RA3 and LA3 channels because of limitations of the physiologic recorder software).
|
|
View larger version (58K):
[in a new window]
|
Figure 7 Electrograms from the decapolar catheters that were placed in the left atrium (LA) and the right atrium (RA). Right atrium rapid burst atrial pacing for 2 s did not terminate the atrial fibrillation. The LA electrograms showed no alterations in pattern or frequency in response to attempted burst pace termination from the RA. The channel labeled RA3 was selected for pacing (pacing artifact during the pacing train is not uniformly displayed in the RA3 channel because of limitations of the physiologic recorder software).
|
|
|
